HomeMy WebLinkAboutAPA1203----------l> r --l> -Ul === " c.J -----l> o=== .!>.!!!!!!!!!!!!!!! ;o Ul=== m c.n= Ul o= 0 o-c o= ;o c.n = ()
-.l= m c.n=== Ul
.!>.!!!!!!!!!!!!!!! m= r
.1>. .....
(D
;o -----l> === ;o
-< -----ELECTR IC PO\-lER CONS~TION FOR THE RAILBELT:
A PROJECTION OF REQUIRE~lliNTS
TECHNICAL APPENDICES
by
Scott Goldsmith
Lee Huskey
Institute of Social and Economic Research
Anchorage ~~ Fairbanks :< Juneau
prepared jointly for
State of Alaska
House Pm,;rer Alternatives Study Co mm ittee
and
Alaska PoHer Authority
May 23, 1980 --
I
PORil.AND, Oti:{;fJ:l
(
ELECTRIC POWER CONSUMPTION FOR THE RAILBELT:
A PROJECTION OF REQUIREMENTS·
TECHNICAL APPENDICES
by
Scott Goldsmith
Lee Huskey
Institute of Social and Economic Research
Anchorage * Fairbanks * Juneau
prepared jointly for
State of Alaska
House Power Alternatives Study Committee
· and
Alaska Power Authority
May 23, 1980
l-ID
~es.'65"""
U.lo
C::J{.ps-
vI fJ_
c.-z...
ELECTRIC POWER REQUIREHENTS FOR THE RAILBELT
TECHNICAL APPENDICES
TABLE OF CONTENTS
A. METHODOLOGICAL REVIEW
Type of Forecasting Models. • • • • •••••••
Forecasting Alaskan Economic Activity • • • • • • • •
Forecasting Alaskan Electricity Requirements .•
Endnot.es. • • • • •
B. THE ECONOMIC PROJECTION HETHODOLOGY
Introduction. • • • • • • •
The MAP Economic Model. • • • • •
Household Formation Hodel • • • • •
Regional Allocation Model •
Housing Stock Model • • • • • • • •
C. ECONOMIC PROJECTIONS
What Are Projections.
The Approach of the Current
The Scenarios 1980-2000 • •
Post 2000 • • • • • • • • •
Study •
Projections of State Growth 1980-2000 • •
Regional Projections. • ••••
Housing Stock Projections • • • • • • •••••
D. COMPONENTS OF THE END USE MODEL
A-1
A-6
A-10
A-16
B-1
B-2
B-6
B-13
B-19
C-1
C-3
C-5
C-36
C-37
C-45
C-54
Households and Housing Stock. D-1
Residential Electric Space Heating. • D-20
Major Appliance Saturation Rates. • D-51
Major Appliance Fuel Type Mode Split. • • • • D-60
Electrical Appliance Average Annual Consumption • D-74
Small Appliance Electrical Use. • • • • • • D-81
Commercial-Industrial Requirements. • • • • D-82
Endnotes. . . . . . . . . . . . ·. . . . . . . . D-99
E. ELECTRIC UTILITY SALES PROJECTION METHODOLOGY
Residential Nonspace Heating Electricity
Requirements • • • • • • • ·• . • • • • • • • E-1
Residential Space Heating Electricity Requirements. E-7
Commercial-Industrial-Government Electricity
Requirements • • • • • • • • • • • • • • • • • • E-ll
TABLE OF CONTENTS
(cont.)
F. ELECTRIC UTILITY SALES PROJECTION MODEL PARAMETERS
Projection Model .Parameters for Base Case • •
Assumptions for the Price Induced Shift Toward
Electricity Consumption Case • • • • .
G. PROJECTING HILITARY AND SELF-SUPPLIED INDUSTRIAL
ELECTRICITY NET GENERATION
Military Requirements • • • . • • • • •
Self-Supplied Industrial Requirements .
H. HISTORICAL ELECTRICITY SALES AND ECONOHIC DATA
I. A REVIEW AND COHPARISON OF RAILBELT ELECTRIC POWER
REQUIREMENT PROJECTIONS
J. BIBLIOGRAPHY
. . ..
F-1
F-49
G-1
G-3
APPENDIX A. METHODOLOGICAL REVIEW
A.l. Types of Forecasting Models
There are three general types of forecasting techniques or models.
Time series or trend models include all mathematical techniques in
which the forecast is a function of time. The justification for this
approach is that past behavior is the best guide to forecasting the future
behavior of a variable. Past trends form the basis for forecasting~ and
the various techniques used in this form of modeling are concerned with
identifying the most significant past movements of the variables being
forecast.
A simple example of a trend technique is the classical (past, univer-
sally used) method of forecasting electricity consumption growth. During
every decade since the turn of the century until 1970, electricity con-
sumption in the United States has'doubled. This corresponds to an annual
growth rate of 7 percent. This information is enough to formulate a
trend model for projecting future electricity consumption growth. Con-
sumption next year is 7 percent greater than this year.
An obvious advantage of this type of model is that it is easy and
cheap to construct and use (although some time series models can be very
complicated and expensive to develop). If the actual process by which
the forecast variable is determined is stable over time, then this
technique will work well. In general, the shorter the forecast interval,
the more likely it is that the system will be stable. The historical
growth of electricity consumption could be interpreted as a reflection
of a very stable process. For many decades, a simple trend model worked
well to forecast future consumption.
Basically, a time series or trend "model is appropriate when
1) cost and time minimization involved in making the
forecast is important,
2) the process determining the forecast variable is stable,
3) the forecast interval is short, and/or
4) no information is available on what factors really
account for the level of the forecast variable.
Causal models include all mathematical techniques where a specific
set of factors is assumed to "cause11 or determine the value of the fore-
cast variable and the causal relationship is quantified. The idea behind
these techniques is that the identification of cause and effect relation-
ships facilitate our understanding of and forecasting of future events.
These modeling techniques thus try to pick out the most important causal
relationships and to measure them quantitatively. An example of a
simple causal model used to forecast electricity consumption would be
based on the idea that the level of population and income determines
or "causes11 the level of electricity consumption. Thus, as population
grows or income grows, electricity consumption is forecast to increase,
and vice versa.
A-2
The obvious advantage of this type model is that the forecast is
based upon the notion of cause and effect~ rather than the identification
of patterns in the past movement of the forecast variable (as in· the
time series technique). If there are past or forecasted changes in
the process by which the level of the forecast variable is determined,
this forecasting technique can accommodate them. Thus, a time series
model for forecasting electricity consumption would not be able to
respond to a sudden drop in population which would~ in all likelihood,
"cause" a reduction in consumption. A causal model with population as
a causal factor would be able to accommodate this structural change.
Another advantage of causal models for forecasting is that they
allow one to do "what if" experiments. Using the causal model of elec-
tricity consumption, one could determine what would happen to consumption
if population doubled, fell by half, etc.. Time series models do not
have this capability.
Obvious problems with causal models are the time and cost generally
associated with their construction and the requirement that the causal
variables must be forecast in order to use the model. For example,
forecasts of population and income are necessary to use the causal model
for the electric~ty consumption forecasting discussed here.
The causal model is generally appropriate when
1) the process by which the forecast variable is deter-
mined is not constant over time,
A-3
2) it is possible to quantify links between causal variables and
the forecast variable,
3) the forecasting model will be used repeatedly using varying
assumptions regarding the causal variables,
4) the time interval for the forecast is long, and/or
5) sufficient time and money are available to develop the model.
Judgmental models are all nonquantitative or nonmathematical tech-
niques for forecasting. The simplest example is the informed opinion
of an expert. A more complex and structured technique is the Delphi
Technique, in which a group of experts interact in a formal way to
develop a consensus forecast. The rationale for judgmental models is
that individuals with an understanding of the process by which the fore-
cast variable. is determined can directly interpret a large amount of
information relevant to projecting the future level of the forecast
variable and on that basis develop a forecast. No specific mathematical
o:i: formal techniques are used.
An example of judgmental techniques would be in the forecasting
of technological change affecting electricity consump,tion. It is not
possible to forecast by quantitative methods the types and timing of
technological change which would either increase or decrease electricity
consumption.
This technique is particularly appropriate where the number of
potential causal variables is large and they cannot be systematically
analyzed for lack of information, resources, etc. Obviously, the longer
A-4
r
the forecast interval, the more general are the variables which need to
be considered and, thus, the more appropriate judgmental forecasting
becomes. A possible dra'tvback of this method is that system interactions
(the effect of one variable upon another) may be overlooked in complex
systems.
This technique is appropriate when
1) resources for forecasting are limited,
2) the process determining the forecast variable is
not amenable to quantification, and/or
3) data are not available to develop a quantitative model.
Each of these general techniques is appropriate in certain situations
depending upon the process being.analyzed, the desired accuracy of the
forecast, the resources available to develop the forecasting tool, the
uses to which the forecasts will be put, and an evaluation of what method
actually forecasts best.
In any complex forecasting effort, elements of all three methods
will likely be present. The forecaster must determine what technique is
appropriate ~t each step in the process of developing the overall fore-
cast. ·For example, the forecaster may decide that a causal model is the
appropriate general approach. He may use time series or trend analysis
to forecast the value for an explanatory variable. Finally, he may
adjust the forecast derived from the causal model on the basis of infor-
mation he has which is nonquantifiable in the sense that it cannot be
incorporated into the causal model in a formal way.
A-5
A.2. Forecasting Alaskan Economic Activity
The important considerations in choosing a general technique for
forecasting Alaskan economic activity for the purpose of projecting
electricity requirements are as follows:
1) a thirty-year forecast is required;
2) the forecasting model will be used repeatedly and
will be utilized to do "what if" experiments;
3) the Alaskan economy is a complex system with many
factors interacting in important but non-obvious
ways; and
4) the future growth and development of the Alaskan economy
may be considerably different from the past because of
a qualitative and quantitative change in the factors
determining the state of the economy.
These considerations suggest a quantitative causal model which has
the advantages of
1) identifying structural relationships within the
economic system, quantifying them, and automatically
tracing the effects of these relationships through
the system;
2) allowing "what if" experiments to exci.mine alternate
futures under alternate assumptions about important
future events and variables; and
3) providing a framework within which all forecasting
assumptions are explicit.
Arguments against the causal approach are
1) the complexity of the model makes its development
and utilization relatively expensive;
2) the quality of the data may not warrant sophis-
ticated causal models;
A-6
3) the causal relationships may not be quantifiable; and
4) the causal relationships may not be understood.
In choosing a quantitative causal model, we feel that, although the
causal relationships in the Alaskan economy may be only partially under-
stood and the data of poor quality, the advantages of formalizing the
relationships that are considered relevant outweigh these problems.
Large elements of trend analysis and judgmental models enter into ·
the causal modeling approach in the development of the projections of
some of the most important variables in the economy, and these techniques
become more important the further into the future the forecasts are pushed.
Currently, twenty years is the limit for the quantitative causal model.
Beyond that time, we revert to,trend and judgment.
Having decided upon a quantitative causal modeling approach to
economic forecasting, there remains the question of choosing among the
many types of models available. In fact, there have been at least seven
different economic models of the Alas~an economy developed over the
last fifteen years, although only three econometric models are in current
use. (These are the Man in the Arctic Program [MAP]" model, the Alaska
Department of Labor model LABMOD, and the Alaska Department of Commerce
and Economic Development model AEIRS.)
In choosing a type of causal model, we would want one which is best
able to represent the important factors in the workings of the economy
A-7
itself. Also, since the projection is over a long-time period, we would
want a model capable of adequately forecasting the level of economic
activity in the long run.
These criteria eliminate both input-output and short-run forecasting
models. The first is rejected because it best represents interindustry
flows of intermediate goods in manufacturing. There is almost no manu-
facturing in Alaska and, thus, an interindustry transaction table would
be mostly zeros. Short-run forecasting models are designed specifically
for that purpose and as such can ignore many important long-run phenomena.
The choice is thus narrowed down to some type of long-run, econo-
metric or simulation model. One advantage of both these model types is
their flexibility in terms of the situations they can model and the
theories they can represent. The difference between them is that the
econometric model uses statistical techniques (such as regression analysis)
to identify.the quantitative relationship's among variables, while the
simulation model uses nonstatistical methods (averages, point estimates,
judgment, etc.).
Econometric determination of quantitative relationships has the
advantage of being able to interpret formally all historical information
concerning those relationships although the techniques used can be
expensive, time consuming, and on occasion too sophisticated for the
Alaskan data. Because of repeated model use, however, we feel the
econometric approach is appropriate.
A-8
Finally, we want a model that specifically incorporates several
important features of the Alaskan economy. These include:
1) the importance of state government activity in
determining the level of economic activity;
2) the importance of and potential variability in
development of Alaskan natural resources
(particularly petroleum) on the level of economic
activity;
·3) the process of maturation of the economy as it grows
in size; and
4) the links which exist between the Alaskan and national
economy.
The MAP econometric model includes features which address all of
t~ese important relationships but is criticized in the treatment of
some of them. Specifically, it is suggested that the "multiplier" is
too large and that the method used to determine population is inappro-
priate.
The "multiplier" is a quantitative measure of the support sector
response to changes in basic ~ector activity. The support sector
includes such industries as retail and wholesale trade and services,
while the basic sector is made up of those industries which "drive"
the economy, such as petroleum, portions of government and construe-
tion, etc. In a developing economy such as Alaska, the support sector
is growing rapidly, and this is reflected in a large "multiplier" value
on increments to basic sector activity.
A-9
It is also argued that the "multiplier" is too larg~ because it
includes a state government response to increases in economic activity
.in the private sector. We argue that this type of response has indeed
occurred historically (the level of the state government budget has
grown partially because private sector growth has increased the demand
for public goods and services) and, absent specific state policies to
severe the connection, will continue to occur in the future.
The method used in the MAP model to determine the level of net
migration to the state, and thus ultimately the level of population, uses
the relative Alaskan wage rate and the change in the employment level as
explanatory variables. These variables are generally accepted by econo-
mists and demographers as being important in the determination of migra-
tion patterns. Clearly, other factors are also important, although
they are not quantifiable. Neither is it clear how, taken together,
they would effect the results obtained using the two economic variables.
In the absence of such analysis, the present technique appears appro-
priate.
A.3. Forecasting Alaskan Electricity Requirements
As previously noted, the traditional method of forecasting elec-
tricity requirements was the use of trend analyses. In the aggregate,
it worked well until the early 1970s. At that time, however, there
occurred a sharp break with the past in terms of behavior in all
energy markets. Growth in consumption of electricity began to fall
A-10
short of growth projected by trend analysis. It was clear that the
structure of electricity markets was changing in ways that trend analysis
was unable t·o anticipate and adjust to. These changes included a re-
versal in the long-term trend of declining real prices for electricity,
the attainment of high saturation rates for many appliances, the
appearance of the conservation ethic, demographic changes, and other
factors. Figure A.l dramatically demonstrates this inadequacy of the
traditional forecasting method as applied to California.
It may be that a new long-term trend in growth of electricity con~
sumption may emerge after the electricity market again stabilizes, but
it is unlikely that this will occur for some time. In the interim, causal
forecasting techniques are necessary which explicitly attempt to identify
those factors which are causing change in the basic electricity con-
sumption patterns. In a9dition, causal models explicitly allow utilities
and policy analysts to examine the effects on electricity consumption
of policies which effect electricity prices and other causal variables.
Judgemental models have a role in this forecasting task for two
reasons. First, the data is often not available with which to develop
a quantitative model. Second, there are some relationships which may
be difficult or impossible to quantify. Future changes in technology
which will alter consumption patterns cannot be forecast quantitatively,
for example. Emerging important factors effecting load growth in a
particular market for which no historical information is available is
another. In all these cases, however, the judgement must come down to
A-ll
FIGURE A.l.
CALIFORNIA UTILITIES STATEHIDE ELECTRICITY SALES FORECASTS
----_Utility Forecast
·(])
-----CEC Adoptc1 1977 Forec_?st
235
:X: 220 ;:
~
z
c _, _,
a3
1980 1985
(D,{DG.O. 131 -Utility submissions to the Public Utilities Commission
G),@ ,@Utility s-ubmissions to the CEC
~CEC adopted forecast (1977)
-:::s--.-,... --------ource: c l"f
76
BI.LUO
J(Wtf ·
32
BILLION
KWH
a ~ ornia Energy Commission "Califor . E
A Preliminary Assessment," August 1979~~~. n{l.gy Demand 1978-2000:
l 10~
I
I ~
10:
a quantitative value which can be used in the analysis because the
electricity requirements forecasts are quantitative.
The primary forecasting tool is the quantitative causal model of
which there are two basic types: econometric and end use models.
Econometric models of electricity consumption use statistical tech-
niques to quantify relationships between electricity consumption or
electrical appliance ownership and price~ income~ and other "explanatory"
variables. They are generally quite aggregate in the sense that all
residential consumption for all purposes may be lumped together and
"explained" by the same price and income variables. These models have
been developed by economists who have primarily been interested in
identifying the exact strength of the relationship between electricity
consumption and the important economic variables price and income. (They
often refer to the concept of elasticity~ which is the percentage change
in electricity consumption in response to~a 1 percent change in price
or income. Thus~ a-price elasticity of -.5 would mean that when the
electricity price increased 1 percent electricity consumption declined
.5 percent.)
End use models disaggreate electricity consumption not only by
class of user but also by use. Total consumption is then the sum of
all electricity consuming appliances and the average amount of elec-
tricity consumed in each. The stock of electrical appliances will be
determined by demographic and economic variables and average consump-
tion by those same variables~ as well as technology~ use patterns~ and
other variables.
A-13
Because of the great diversity observable in electricity consump-
tion, it is necessary to collect large amounts of data in order to con-
struct an end use model. Its advantage is a capability of identifying
technological, regulatory, and other policy-related factors which effect
demand. Having done this, the model can be used to forecast consump-
tion pattern changes in response to changes in these f~ctors.
Each o~ these modeling approaches has advantages and drawbacks.
The econometric approach uses accepted st-atistical procedures to develop
quantitative relationships for· certain variables, but data limitations
restrict its ability to analyze specific electricity uses and non-
economic-demographic factors effecting consumption. There is concern
within the economics profession that the many "demand functions" which
have been estimated by researchers over the years really are measuring
a relationship between electricity consumption and price and income.
We feel there is substantial truth to these arguments. In addition,
there is suspicion that relationships estimated using historical elec-
tricity prices may no longer be appropriate because of recent and· rapid
price increases in contrast to the previous long-run secular decline in·
electricity price. Even if there was agreement that the approach of the
econometricians was valid, the substantial variation in results re-
ported by various studies invites caution in the utilization of ~ny
particular estimate. One review article cited long-run price elastici-
ties from different studies of between -1 and -2 and long-run income
elasticities of between .2 and 2.1
A-14
End use models are disaggregated and lend themselves to policy
analysis but generally are lacking in a strong statistical foundation
for the parameters (quantitative relationships) used in the modeling.
This is largely because of a lack of adequate data. In many instances,
a single data point may be all the researcher has available to· work with,
and this does not provide a firm basis for quantification of a rela-
tionship between two variables.
The best solution, suggested by a recent article in Public Utilities
Fortnightly, is to utilize the better features of both modeling approaches
to make the modeling sensitive to both the economic factors of price
and income as well as to regulatory and technological factors.2
In the model developed for this study, this was also the preferred
approach with a heavy emphasis on end use because of the anticipated
/
use of the model not only for forecasting but also policy analysis.
Relatively little emphasis was placed on econometrics becau~e of the
absence of data, the theoretical problems alluded to.previously, and
earlier unsuccessful attempts by one of the authors to develop econo-
metric models of electricity demand for Alaska.3
A-15
ENDNdTES: APPENDIX A
1. Lester Taylor, "The Demand. for Electricity: A Survey," Bell
Journal of Economics, Spring 1975,. Vol. 6, No. 1, p. 10~
2. Robert Shaw, Jr., "New Factors in Utility Load Forecasting,"
Public Utilities Fortnightly, July 19, 1979, p. 19.
3. Scott Goldsmith, "Future Electricity Requirements in Alaska,"
paper presented at Western Economics Association Annual Meetings,
San Francisco, California, June 1976.
APPENDIX B. THE ECONOHIC PROJECTION METHODOLOGY
Introduction
The projections of end-use energy requirements are based on pro-
jections of economic activity in the state and its railbelt region.
This eco?omic projection provides estimates of employment, population,
households, and housing stock for the state and the major regions in the
railbelt. These projections are provided for five-year periods through-
out the projection period; the projection period is through 2010.
The main component of the economic projection methodology is the
Man-in-the-Arctic Program (MAP) statewide econometric model, which is
used to project the employment, population, and fiscal variables. In
addition to this major component, three ne-.:..r components have been deve-
loped for this study: a household formation model, a regional allocation
model, and a housing stock model. The household formation model estimates
the number of households in the state based on the MAP model population
projection. The regional allocation model allocates the major variables--
population, employment, and households--to the study regions. Finally, ·
the housing stock model projects the distribution of housing by type for
each region of interest. This appendix provides a detailed description
of each of these components.
The MAP Economic Hodel
The statewide econometric model developed by the Han-in-the-Arctic
Program (MAP) at the University of Alaska's Institute of Social and
Economic Research is the principal model used in the projection of
economic activity for the end-use forecast. The model consists of three
interrelated components: an economic model, a fiscal model, and a demo-
graphic model. The basic structure of the model is shown in Figure B.l.
The economic model divides the economy into exogenous or basic sectors
and endogenous or nonbasic sectors. The level of output in the exogenous
sectors is determined outside th~ state's economy. The level of output
in the nonbasic sectors of the economy is determined within the Alaska
economy since their primary purpose is to serve local Alaska markets.
The basic industries in the model are mining, agriculture-forestry-
fisheries, manufacturing, federal government, and the export component
of construction and transportation. The nonbasic industries are whole-
sale and retail trade, finance-insurance-real estate, services, communi-
cations, utilities,·and the remainder of construction and transportation.
Incomes, output, and employment are simultaneously determined in
the model. The demand for local economic production is determined by
the level of real disposable income in the economy. The level of indus-
trial production determines the demand for labor; employment is that level
which is needed to produce the required output. The labor demand is
always satisfied since the Alaska labor market is assumed to be open to
B-2
..
'
I.~UTS~DE
·L ~.
PERSONAL
TAXES
IND'{JStz'lUAL
?RODUC'l'ION
. . 1
,.
· DIS1?0SN3LE
"'
. ·'
..
'· I'
••
. . .
..
..
. i
PERSON~L ~~~-------·~ .... . ... ,
. .
..
. ;
\v~GB
. R."a.TES
It
t'll\GES AND
Sl\LARIES
J?U:RSONAL.
..
-
. .
·, .. ·--.. - - - . ., li' . ~.-________ _,.,...f'-_.,.J-~--· .., . w ~. . .. .------'--.........__...K-t ·I FU~lO STATE R!Wt:::NOES ~O>A' ..... -•• • I
'1-~CCUXULATION ...... . .
&.-:------J\ /[ .............. OUTS ~DE FORCES ww
\!t ..
'I!
CONSU!I.ER
PRICES
. . . .
.
u.s.
?RICES
,~;
. .
.
IN CO HE
>%:1
H'
~
t:.d .
1-' .
migration from the rest of the United States. Because of this, both the
supply and price of labor (wage rate) are linked to outside activity;
wage rates in Alaska are determined in part by wages outside the state.
Once wage rates and employment are determined, aggregate wages and
salaries are known. The level of disposable income is estimated by
·adding an estimate of nonwage income to wages and salaries and reducing
this by the level of income taxes. Real disposable income is found by
deflating the level of disposable income by a relative price index; the
major determinants of Alaska prices are U.S. prices, the size of the
economy, and the growth rate of the economy.
The level of population projected is based on a projection of each
of its components: births, deaths, and migration. The model uses age-
sex-race specific survival rates and fertility rates to project the
births and deaths of the civilian population;·· the number of births net
of deaths determines the natural increase in the populatio~. Total
civilian population is found by adding net civilian migration to the
natural increase. Net migration is determined by the relative economic
opportunities in the state. Economic opportunities are described in the
model by the change in employment and the Alaska real per capita income
relative to the U.S. average. Total population is determined by adding
an exogenous estimate of military population.
The final component of the MAP model is the fiscal model. This
model describes the activity of the state and local governments. The
fiscal model calculates the level of personal tax payments which are
B-4
necessary for projecting disposable income. The fiscal model, based on
an assumed state spending rule, also calculates personnel expenditures,
the level of state government employment, and the amount spent on capital
improvements. The amount the state government spends on capital improve-
ments partly determines the level of employment in the construction
industry. All three model components are linked by their requirement
for information produced by the other compo?ents. A description of the
model can be found in Goldsmith, Man in the Arctic Program Hodel Docu-
mentation (1979).
The model has been updated in connection with the current study.
This updating included the reestimation of important economic component
equations with data series which ·included the most recent information.
In most cases, this consisted of including the 1978 data in the series;
however, some data series wer~ recalculated based on recent information.
These data changes had only marginal effects on the equations; however,
several equations were respecified. The primary reason for these speci-
fication changes was to capture the buoyancy of the economy in the post-
pipeline period. Equations describing wage rates in the exogenous
sector and output in the endogenous sector were respecified in an
attempt to improve the model's description of post-boom periods.
An additional change was mad0 to the fiscal model. The fiscal. ·
model was changed to incorporate recent changes in state tax laws which
eliminated the state income tax for residents with over three years in
the state. The recent permanent fund distribution program was also
reflected in model changes.
B-5
Household Formation Model
MODEL DESCRIPTION
The primary unit on which projections of residential energy con-
sumption are based is the household.· Households are living units. They
are distinct from families and can contain more than one family as well
as unrelated individuals. This section describes the model developed
for this study to project the number of households in the state.
The population projections determine the number of households in
the state. The number of households is a function of both the level of
population and its age-sex distribution. The age-sex distribution of
the population is important because the rate at which people form house-
holds differs across age-sex cohorts. The model described below accounts
for both of these influences of population on household formation.
The household formation model is an accounting model which depends
on a set of assumptions about the cohort specific rate of household forma-
tion and changes in those rates. The model is based on the assumption
that the social, economic, and life cycle factors which determine the
formation of households can be described by ~ set of household formation
rates. Household formation rates describe the probability that a person
in a particular cohort is a household head.
The model requires input from the MAP population model in the form
of the projected size and age-sex distribution of the population. The
B-6
total number of households in the state (HH)· is equal to the number of
households summed across age and sex cohorts.
HH = L~ HH •.
~J ij
(B.l)
The total number of households in sex cohort i and age cohort j
(HH .. ) describes the number of households with household hea4 or primary ~J
individual in the ith sex and jth age cohort. The total number of house-
holds in cohort ij equals the number of civilian non-Native households
in cohort ij (CHH .. ) plus the number of Native households in cohort ij
~J
(NHH .. ) plus the number of military households in cohort ij (MHH .. ) •
~J ~J
HH ..
~J
CHH .. + NHH •• + MHH ..
~J ~J ~J
The number of civilian and Native households in each cohort is a
(B.2)
function of the population and household formation rate for the cohort.
The number of households in any cohort equals the cohort specific house-
\ .
hold formation rate (HHR .. for civilian non-Natives and NHHR .. for Natives)
~J ~J
multiplied by the total population (CNNP .. for civilian non-Natives and
~J
NATP .. for Natives) net of the population in group quarters (CPGQ .. for
~J ~J
civilian non-Natives and NPGQ .. for Natives).
~J
CHH. ~ = (CNNP .. -CPGQ .. ) * HHR .. ~J ~J ~J ~J
(B.3)
NHH .. = (NATP .. -NPGQ .. ) * NHHR.'. ~J ~J ~J ~J
(B.4)
B-7
Household formation rates describe the probability that someone in
a particular cohort is the head or primary individual of a: household. ·
These rates change through the projection period. The initial rates
are those found in 1970 (HHR:~70 for civilian non-Natives and NHHR:~70
1] 1]
for Natives).
(B.S)
·(B. 6)
Household formation rates are assumed to change at a constant yearly
rate (CHHR .. for civilian non-Natives and NCHHR .. for Natives). So the
1] 1]
household formationrate inany year equals the ratein the previous
year times the rate of change.
HHR .. = HHR .. (-1) * CHHR ..
1] 1] 1]
(B. 7)
NHHR .. = NHHR .. (-1) * NCHHR .. 1] 1] 1]
Finally, the cohort distribution of military households is assumed
to rema.in constant throughout the projection period. The number of
military households (MHH .. ) equals the number in 1970 (MHH:~70 ) times
1] 1]
the percentage of 1970 military population in the state (MILPCT).
MHH .. = MHH~~70 * MILPCT
1] 1]
B-8
PARAMETER ASSUHPTIONS
The most important source of variation in the model results is the
assumed rates of household formation. These rates have been subject to
dramatic changes in recent history. The change in these rates can be
judged by examining recent changes in the average household size.
Between 1960 and 1970, the average household size in the United States
fell by 6 percent from 3.33 pe?ple per household to 3.14 people per
household. Between 1970 and 197&, the rate of decrease was almost twice
as fast as in the previous period; the average household size fell by
more than. 10 percent from 1970 to 2.81 (Bureau of the Census, Projections
of the Number of Households and Families: 1979 to 1995, 1979). Alaska
is experiencing a similar trend, the average household size fell by
almost 7 percent between 1970 and 1976 (Bureau of the Census, Demographic,
Social, and Economic.Profile of States: Spring 1976, 1979).
The important recent changes in household formation rates, illus-
trated by the recent changes in household size, and the lack of agreement
by population experts on future changes in these rates make the selection
of any particular set of parameters probabilistic. Two sets of para-
meter assumptions are required by the model, household formation rates
arid yearly changes _in those rates. Table B.l. presents the initial set
·of household formation rates. Thes.e rates are derived from the 1970
census and equals the number of household heads per population in
households in the cohort. These rates were adjusted in some cohorts
to insure consistency with U.S. rates in 1970.
B-9
TABLE B.l. 1970 ALASKA CIVILIAN POPULATION
HOUSEHOLD FORMATION RATES (HHR .. ) . l.J
NON-NATIVE NATIVE
Male Female Male Female
0 - 1 0 0 0 0
1 -5 0 0 0 0
5 -9 0 0 0 0
10 -14 .001 .001 .003 0
15 -19 .040 .018 .017 .006
20 -24 .583 .107 .238 .069
25 -29 .900 .109 .576 .082
30 -34 .933 .117 .746 .095
35 -39 .955 .126 .881 .119
40 -44 .962 .133 .894 .120
45 -49 .963 .148 .907 .139
50 -54 .964 .164 .922 .149
55 -59 .956 .207. .947 .296
60 -64 .956 .245 .926 .313
65 + .885 .320 .816 .385
SOURCE: Bureau of the Census, 1970 Census of Population Detailed
Characteristics: Alaska, 1972, Table 153.
B-10
Table B.2. illustrates the second set of required parameter assump-
tions, the assumed yearly change in the household formation rates. These
changes are based on the changes implicit in the Series B projections of
households by age and sex cohorts prepared by the Bureau of the Census
(Bureau of the Census, Projections of the Number of Households and Families:
1979 to 1995, 1979). ·The average yearly change in household formation
rates by age-sex cohort for the nation were assumed to hold for the
civilian non-Native population. It was assumed that Native household
fordation rates would not change as rapidly; Native household formation
rates were assumed to change at rates which would provide half the change
in average household size projected for the nation.
B-11
TABLE B.2. YEARLY PERCENT CHANGE IN HOUSEHOLD
FORMATION RATE (CHHR .. )
~J
NON-NATIVE NATIVE
Male Female Male Female
0 - 1 0 0 0 0
1 -5 0 0 0 0
5 -9 0 0 0 0
10 -14 1.002 1.045 1.001 1.028
15 -19 1.002 1.045 1.001 1.028
20 -24 1.002 1.045 1.001 1.028
25 -29 1.000 1.045 1.002 1.028
30 -34 1.001 1.040 1.001 1.024
. 35 -39 1.000 1.027 1.000 1.016
40 -44 1.000 1.027 1.000 1.016
45 -49 1.001 1.012 1.000 1.006
50 -54 1.001 1.012 1.000 1.006
55 -59 1.001 1.000. 1.000 1.000
60 -64 1.001 1.000 1.000 1.000
65 + 1.001 1.000 1.000 1.000
SOURCE: Bureau of the Census, Current Population Reports Series P-25,
No. 805, Projections of the Number of Households and Families,
1979 to 1995, May 1979.
B-12
Regional Allocation Model
MODEL DESCRIPTION
A method for making substate regional projections was required by
this study. The economic and household projections described above are
made at the state level. These projections serve as the basis for the
energy demand projections; however, the Susitna project will provide
energy for only a portion of the state, the railbelt region. This section
describes the regional allocation model used in this study to allocate
statewide projections to the region of interest.
Methods of projecting substate regional economic activity range
from the simple to the complex. The simplest methodology is to allocate
state economic activity to the region based on its historical share or
to allocate nonbasic activity based on the regional share of basic sector
activity. The most complex approach involves the estimation of complete
regional models. The first approac~ suffers from its simplicity; it
fails to recognize the importance of changes in the structure of the
regional economy over time as the economy grows. The·more complex
approach requires massive commitments of time and resources to develop.
It may also suffer from a lack of consistent data in the regions, par-
ticularly in areas like Alaska where many of the regions have small,
undeveloped economies.
In choosing a regional projection technique, we were interested in
three things. First~ we wanted a model which was simple and efficient
B-13
...
to use and could be used to project activity in a number of regions.
Secondly, we wanted a model which made maximum advantage of the short·.
data series available in the regions. Finally, we wanted a model which
provided results consistent with the state projections.
The method used in this study is a regional shares model. In this
model, the regional shares of st~te support sector employment, state and
local government employment, and population are estimated econometrically
as a function of basic sector activity as well as proxies for comparative
advantage and scale of the regional economy. A pooled time series cross-
section approach is used to estimate the model. This econometric approach
has two purposes. First, it allows us to make us.e of the short data
·Series in the census divisions. Secondly, it allows us to capture the
major variability in the regional shares of economic activity which is
across regions rather than over time.
Traditional explanations of regional economic growth explain growth
as a function of growth in the region's basic sector. The regional
allocation model recognizes that the local support sector response
depends not only on basic sector growth but also on the position of the
region in economic space •. Larger economies will provide a greater
support sector response since they offer economies of scale and produce
more goods and services locally. Regions may also respond differently
if they provide support sector services to. regions other than their
own. These trade centers have a comparative advantage in producing
these goods and services. The scale and comparative advantage effects
B-14
. .
are accounted for in this model by the use of lagged population and
regional dummies in each equation.
The model consists of four equations which estimate the regional
share of population, support sector employment (in two categories), and
state and local government employment. State and local government employ-
ment is assumed to be allocated across regions in the state primarily to
serve the population, so the share of state and local government employ-
· ment in region i (REGSL.) is primarily a function of lagged share of
~
population in the region (LRPOP.). Special characteristics such as the
~
capital in Juneau and administrative centers in larger regions are
accounted for through the use of regional dummy variables (D.).
~
REGSL. = F1 (LRPOP.,D.)
~ ~ ~
(B.lO)
Support sector economic activity is disaggregated into two distinct
sectors: direct support, which includes construction and transportation
employment, and other support which includes trade, services, finance,
utilities, and communication. This distinction is a function of the
assumed relation between these sectors and basic activity and assumed
differences in causes of growth in each sector. Direct support sector
employment depends not just on the size of the community or its basic
sector, but on the growth of the community. This is a result of con-
struction employment serving mainly an investment function. Because of
this, the share of direct support sector employment in region i (RESA.)
~
is a function not only of the size of the community (LRPOP.) but also
~
B-15
the change in the economy; this change is described by the lagged share
of the change in total employment in region i (L298.). Uniqueness of
~
regional economies is captured by the inclusion of a regional dummy (D.).
~
RESA. = F2 (LRPOP.,L298.,D.)
~ ~ ~ ~ (B.ll)
Employment in the other support sector is assumed to be a function
of the level of basic sector activity in the region. For our purpose,
basic sector is defined broadly to include traditional basic sector in-
dustries as well as local and state government, and direct support sector
employment in region i (RESB.) may differ from a direct relation to the
~
regional share of basic sector employment (RR3EB.) for two reasons.
~
First, the support sector may not expand immediately because of lags or
because the basic sector growth is temporary. Secondly, the region may
provide support sector services to a large region and be related to the
basic sector activity in those regions. To account for these effects,
both the lagged share of population (LRPOP.) and regional dummies (D.)
~ ~
were included.
RESB. = F3 (RR3EB. ,LRPOP. ,D.)
~ ~ ~ ~
Finally, the regional share of population (RPOP.) was assumed to be
. . ~
a function of employment in the region. In Alaska, workers often travel
to jobs in other regions; this is most important in basic sector activi
Because of this, the effects of basic and other support sector employment
or population w.ere separ.ated. · Population may also differ from employment
B-16
since a region may house the families of workers who work in other regions,
such as Anchorage providing the homes of families of Prudhoe Bay workers.
The regional share of population (RPOP.) was assumed to be a function of
~
the share of support sector employment (RESB.), the share of basic sector
~
employment (RR3EB.), and a regional dummy (D.) which reflects the fact
~ ~
that a region can serve as home to families of workers employed in other
regions.
Regional totals are found by multiplying state totals by the shares
estimated by the model.
PARAMETER DESCRIPTIONS
These equations were estimated using a pooled time series cross-
sectional technique. Data onAlaska labor divisions (similar inmost
cases to census divisions) from 1965 to 1976 were used in the estimation.
A linear form of the equations was estimated; the primary reason for this
choice of functional form was the ability to use the set of equations to
~project the growth of any region defined as an aggregate of census divisions.
Conopsak (1978) identifies both the advantages and problems with
using this technique. First, the use of pooled data increases the number
of degrees of freedom compared to either cross-section or time-series
regressions. Second, pooling techniques limit structural change biases
which may occur in time ·series. Thirdly, by using a time series of
cross-sections, it is possible to measure the effect of time and struc-
ture on the relationships. This is particularly important when estimating
B-17
a shares model since examining time series of pooled cross-section
provides for ample variation in the data.
Two types of problems may exist when using this approach. There
may be systematic bias of the disturbance term because of cross-sectional
influences, or the source of bias may be autocorrelation of the residuals.
The first problem may be reduced with the inclusion of the regional dum-
mies in the equation. The second problem may be reduced by adjusting for
autocorrelation in the regression; both of these corrections were made.
Equations B.l3 to B.l6 are the equations used in the model. Dummies
were included for all regions; however, only those dummies for the regions
in this study are shown (DA-Anchorage, DK-Kenai, DS-Seward, DM-Matanuska,
DF-Fairbanks, and DV-Valdez). The high R2 (uncorrected) result from two
factors, the inclusion of the regional dummies and the relatively small
variability of regional shares in the historical period.
REGSL = .246 * LRPOP + .224 * DA + .124 * DF + .025 * DK
(3.48)1 (6.99)1 (10.07)1 (3.56)1
+ .021 * DM + .009 * DS + .018 * DV
(3.15)1 (1.41) (2.78)1
RESA = 1.357 * L298 + .744 * LRPOP + .148 * DA + .045 * DF
(4.73)1 (3.44)1 (1.67)1 (1.27)
+ .025 * DK
(2.36)1
.008 * DM
(-1.07)
.003 * DS + .003 * DV
(-.45) (.45)
R2 = 987 .
R2 = 987 .
1 t statistic in parentheses significant at greater than 95 percent.
B-18
RESB = .269 * LRPOP + .086 * RR3EB + .409 * DA + .111 * DF
(3.26)1 (1.31) (11.46)1 (7.71)1
(B.l5)
+ .017 * DK + .007 * DM + .004 * DS + .006 * DV .997
(j.74)1 (1.95)1 (1.19) (1.94)1
RPOP = .290 * RESB + .157 * RR3EB + .213 * DA + .073 * DF
(4.70)1 (2.93)1 (5.78)1 (4.58)1
(B.l6)
+ .029 * DK + .022 * DM + .005 * DS + .012 f DV
(7.38)1 (6.81)1 (1.59) (3.45)
R2 = .995
Housing Stock Model
MODEL DESCRIPTION
Regional projections of households and housing stock are the basic
components of the residential energy demand projections. This model uses
the output of the three components described above to project both the
number of households by region and the distribution of.those households
by housing type. The housing types projected by this model include single-
family, duplex, multifamily, and mobile homes. The total housing stock by
type is found by adjusting for vacan·t housing.
Housing stock projections are influenced by three factors. First,
the number of households determines the aggregate demand for housing.
Secondly, the distribution of households by income, family size, and
tenure determines the effective demand for different types of housing.
Finally, housing has a long life; once a type of housing is built, it
1 t statistic in parentheses significant at greater than 95 percent.
B-19
exists for a long time and influences the actual distribution of hous-
ing by type. The model described in this section attempts to account
for each of these factors.
The number.of households in region i (THH.) is found by dividing
l.
the regional population (POP.) [from the regional allocation model] by
l. ;
a regional population per occupied dwelling units parameter (PPODU.).
l.
This provides an estimate of the total number of households in the
region. On-base households (BHH.), which are assumed to remain constant
l.
throughout the period, are subtracted from total households to find
total off-base households (HH.).
l.
THH. = POP./PPODU.
l. l. l.
HH. = THH. -BHH.
l. l. l.
(B.l7)
Once total off-base households (from hereon, total households) are
found, the demands for various housing types are projected through the
use of housing type demand coefficients. The demand for housing type T
(H:) equals the total housing units (HH.) times the demand coefficient
l. ~
for type T (HDT). The demand coefficients describe the distribution of
households by "preferred" housing type.
HH. * HD: (B.
l. l.
B-20
The initial stock of housing of type i in any period is equal to
last period's housing stock of that type cs:(-1)) minus the removals
~ .
from the stock since the previous period. Removals are due to demoli-
tions, accidental loss (fire, flood, etc.), and conversions to other
types of units or uses. The model finds the initial stock of housing
of type T (s:) by multiplying the stock from the previous period times
~
one minus the removal rate (rT), where the removal rate equals the
proportion of the previous period housing lost between the periods.
(B.20) ·
Construction of new housing of each type is determined by the net
T demand for that type (ND.). The net demand equals the demand for housing
~
T AT of type T (H.) minus the initial supply of that type (S.).
~ ~
T AT
=H. -S.
~ ~
(B.21)
If the net demand for all types of housing is positive, new construction
(ND:) equals net demand plus the equilibrium amount of vac?nt housing.
~ .
In this case, new construction equals net demand plus the vacancy rate
(VT) times the net demand plus the initial supply of housing type T.
(B.22)
If the net demand for a particular housing type is negative, an
adjustment is required. The adjustment recognizes implicitly the effect
B-21
of prices on demand. When net demand for a housing type is negative, this
excess supply is assumed to drive down the price of this type of housing
relative to others and switch demand. For this adjustment, single-family
and mobile homes and multifamily and duplexes are assumed to be close
substitutes; when one type has excess supply, it is filled by households
with the other type demand. If excess supply continues to exist and the
vacancy rate is greater than an assumed maximum rate, the units are filled
proportionally from the other types with excess demand. Once these adjust-
ments are made, new construction occurs to meet the excess demand.
PARAMETER DESCRIPTION
There are four sets of parameters which determine the housing stock
projections. The assum~d people per occupied dwelling unit (PPODU.)
~
determines the number of households in a given regional population. The
removal rates (rT) determine the proportion of last period's housing
which has been removed from the housing stock. The vacancy rates (VT)
determine the supply of vacant housing in any period. Finally, the
housing demand coefficients (HD:) determine the initial distribution
~
for demand by housing type.
The initial people per occupied dwelling rates were taken from the
most recent information found for e~ch region. Table B.3 shows the
initial rates used in each region. These rates were adjusted each
period to reflect projected changes in the population per household rate
on the state level; this adjustment assumed the changes in the state
rate were reflected proportionally in each region.
B-22·
TABLE B.3. INITIAL PEOPLE PER OCCUPIED
DWELLING UNITS
1 Greater Anchorage
3.03
Fairbanks2
3.01
3 Valdez
3.1
4 Rest of State
3.5
~eighted average of rates found in: Anchorage Municipality, 1978
Population Profile, 1978 (for Anchorage); Kenai Borough, Profile of Five
Kenai Peninsula Towns, 1977 (for Kenai and Seward); and Rivkin Associates,
Workbook on the Economic and Social Impacts of the Capital Move on Juneau
and the Mat-Su Borough, 1977 (for Matanuska-Susitna).
2Assumes Fairbanks people per occupied dwelling decreased at same
rate as the U.S. average between 1970 and 1977·; the 1977 rate was .91
less than in 1970 for the United States.
3 .
M. Baring-Gould, Valdez City· Census, 1978.
4weighted average for the nonrailbelt area of the state in the 1970
census (3. 7) assumed to d.ecline at one-half the rate of the decline for
the United States.
An average removal rate is assumed for this report and applied to
all types of housing in all regions. Table B.4 shows the."rates assumed
in this study.
TABLE B.4. ASSUMED HOUSING REMOVAL RATES
1975-1980 1980-1985 1985-1990 1990-1995 1995-2000
1.0% 1.25% 1.50% 1.75% 2.0%
B-23
Removal rates are a function of the age of the housing stock and the
growth of the region. Older housing may be subject to filtering, and
areas with more rapid growth may remove more older housing to make room
for new construction. Comparisons of dwelling unit estimates in 1970
and 1979 with building permits da,ta on units constructed during that
period show an approximate removal rate of 1 percent per five-year
period in Anchorage and Fairbanks. It was assumed that, as the existing
stock ages, the removal rate will grow toward the U.S. average which has
been estimated to be between two and four percent for a five-year period
(deLeeuw, 1974). We assumed the removal rate would reach the lower
bound of 2 percent by 2000.
Vacancy rate assumptions are based on U.S. averages and recent Alaska
-experience. Table B.5 shows the assumed normal and maximum vacancy rates
used in the housing stock projection. The normal vacancy rates are the
ten-year U.S. averages for owner and renter units (Bureau of the Census,
Housing Vacancies: Fourth Quarter 1979, 1980). Single-family and mobile
homes have the owner rate; multifamily, the rental rate; and duplexes, the
TABLE B. 5. VACANCY RATE ASSUMPTIONS
Single Family
Multifamily
Duplex
Mobile Home
Normal
1.1
5.4
3.3
1.1
B-24
Maximum
3.3
16.0
10.0
3.3
average of owner and renter. Maximum rates are based on recent Anchorage
. multifamily experience (Anchorage Real Estate Research Report, 1979);
single-family, mobile home, and duplexes are assumed to maintain the
normal relationship to multifamily vacancies.
The final parameter assumptions concern the housing demand coeffi-
cients; these are the most important parameters for determining the
housing stock distributions. The assumed housing type demand distribu-
tions used in this study are based on the analysis of existing survey
information from Anchorage. A 1978 survey of the Anchorage population
conducted by the Urban Observatory .of the University of Alaska -Anchorage
(Ender, 1979) provided information on housing type choice and demographic
variables. Regression analysis was used to analyze this information. A
linear probability model was estimated for each of three housing types:
single-family, multifamily, and mobile home. These regressions estimate
the probability that a person would occupy a particular housing type as
a function of the age of the household head and family size. Family
size and incomes have often been isolated as the major determinants of
housing type choice. .In reality, current income and wealth influence a
household's ability to purchase a home; the age of the household head is
a proxy for both wealth and income. Table B.6 shows the results of the
Anchorage regressions.
These equations were tested by estimating the 1970 housing type
distribution in the Anchorage, Fairbanks, and Valdez regions and with
more recent data in Fairbanks and Anchorage. The performance of the
B-25
TABLE B.6. HOUSING CHOICE REGRESSIONS
Single Family
SF = .461 -.303 * Sl -.175 * S2 + .08 * S4 + .182 * A2
(70.36)1 (20.52)1 (1.87) (12.24)1
+ .317 * A3 + .380 * A4
(47.33)1 (43.85)~
MultifamiU:_
MF = .383 + .225 * Sl + .086 * S2 -~09 * S4 -.203 * A2
(50.75)1 (6.46)1 (3.07) (19.84)1
-.280 * A3 -.352 * A4
(47.96)1 (49.02)1
Mobile Home
MH = .097 + .068 * Sl + .039 * S2 + .014 * S4 + .008 * A2
(7.01)1 (1.98) (.121) (.043)
.020 * A3 -.016 * A4
(.366) (.151)
Family Size Age of Household Head
Sl <2
S2 3
S4 5<
A2 25-30
A3 30-55
A4 55<
-2 R =
1F statistic in parentheses significant at greater than 95 percent.
B-26
models in these cases suggested that no specific regional adjustmen·t of
the equations was required in Anchorage and Fairbanks, and they were
used to project housing demand coefficients in each region.
In Valdez, comparison with 1978 housing stock distribution showed a
major difference. This was assumed to be a result of the recent rapid
growth in the region connected with construction of the Trans-Alaska
Pipeline Service (TAPS) pipeline and associated port facility. In
Valdez, demand coefficients were assumed to change linearly from the
1978 housing type distribution to the projected 2000 housing demand
distribution.
Use of these equations assumes that the existing relationships
.between housing type choice and non-included variables remains the same.
Most important in the case is the effect of housing price. We are
_implicitly assuming that the relation between housing prices and income
will remain constant throughout the period. The importance of housing
means there may be some adjustments such as two-income families. The
importance of the government in the housing and mortgage markets makes
any change impossible to forecast. This approach also ignores the
existence of land-use constraints which may prevent actual construction
of these units.
The housing type parameters were projected over the period based on
assumed changes in household head age distribution in each region and
the family size distribution. The state distribution of households
B-27.
by age of household head was used to estimate the regional distribution;
each region was assumed to mainta·in the same relation to the state dis-
tribution as in 1970. The family size distribution in each region was
assumed to follow a pattern of change which approached the 1977 Western
Regional distribution (Bureau of Census, Annual Housing Survey: 1977,
1979). The level of household size in the Western region of the United
States was similar to that projected for our regions in 2000. The
change from the 1970 distribution to this 1977 distribution was assumed
to be at the same rate as the rate of the change in population per
household projected by the regional allocation model.
B-28
APPENDIX C. ECONOHIC PROJECTIONS
The projections of electricity consumption in the Susitna Hydro-
electric Power Project service area are based on a set of projections of
economic activity in the state and the railbelt region. This appendix
describes those projections. The appendix is divided into three parts.
First will be a short general discussion of projections, the reasons
for doing them, and general problems with projections. A discussion of
the specific sets of assumptions which were used in conjunction with the
models described in Appendix B follows the first section. Finally, the
projections themselves will be reviewed.
What are Projections?
Projections provide a description of a future level of activity;
the economic projections described in tpis appendix describe possible
future levels of important economic variables. Projections cannot be
assumed to be an accurate description of what will happen but rather
a description of what could happen if the assumptions which determine
the projections come true. This means that projections are probabilistic.
The uncertainty of the .future, though it may increase the problems
associated with making projections, increases the importance of projec-
tions. Decision makers in both public and private sectors need informa-
.tion about the future in order to plan their actions since they must
make decisions which both are affected by and affect future events.
The more uncertain are the future events, the more important are some
projections of them in decision making.
All methods for making projections of future economic activity re~
quire assumptions about the future. The simplest projection technique
is simply to assume a certain growth for each of the major variables of
interest. More complex methodologies employ some form of model to
translate assumptions about specific events into projections. Models
describe the relationship between variables about which assumptions are
made (exogenous variables) and those of which projections are made (endo-
genous variables); an important assumption when a model is used is that
the relationship described by the model remains constant. The use of
models makes explicit the assumptions implicit with simpler projection
techniques, and it provides consistency between sets of projections.
The major problems with projections is the uncertainty attached to
the projections. The uncertainty results because of uncertainty about
the future levels of exogenous variables and uncertainty that the rela-
tionships will continue to hold as described by the model. There are
two major ways to limit the importance of this problem, although un-
certainty can never be eliminated from projections. The first measure
is to provide a clear, complete description of the assumptions on which
the projections are based; this allo"tvs users to know exactly what is
behind the projections. The second measure involves producing many
alternative projections instead of just one; these alternative projec-
tions provide an indication of effect of altering major assumptions.
C-2
These measures in themselves do not limit the uncertainty of any par-
ticular projection, but they allow the establishment of a range of
possible outcomes which the researcher expects to have a very high
probability of occurrence.
The Approach of the Current Study
In the present study, we present a series of projections which,
although of limited number, reflect the range of probable future levels
of economic activity. Because these projections may be used in the
actual design of the pr0ject, it is important to provide a range of
futures with a relatively high probability of occurrence. We do not
assume that the actual growth in future economic activity will match any
of our scenarios in level of activity or timing and magnitude of events.
What we assume is that the general level of economic activity described
by-these scenarios will occur with a high probability.
The projections in this study are for a thirty-year period from
1980 to 2010. Because of the long projection period, there is a large
potential for error in utilizing any single technique in developing the
projections. Two general techniques have been employed in these projec-
tions. The first is-scenario building in which the aggregate values are
developed by constructing scenarios composed of specific events. For
example, the level of total employment in the petroleum industry would
be estimated by .assuming the development of specific reserves with
' -
associated manpower requirements and timing. The scenario building
C-3
technique reduces the potential for error in the projection by dividing
the development of the assumptions into a series of small decisions.
This technique allows specific information about future developments to
be built into.the projection. The most important reason for using the
scenario building technique is that it allows for consistency in the
forecast. By assuming specific events, we can make sure the growth rate
projected is at least possible.
The major disadvantage of the scenario building technique occurs as
the projection period is extended in time. Because many of the possible
future events .are unknown, they may be ignored when developing the sce-
narios. This myopia may result in a downward bias in the projection as
the projection period lengthens.
To attempt to overcome this problem in these projections, a second
technique was used in the post-2000 period when information about speci-
fic events is undefined. This second technique is a judgmental approach
The judgmental technique projects directly the aggregate variables. The
judgmental projection is based on an analysis of both the historic
period growth and the growth in other regions. This technique fails to
provide reality or consistency checks on the assumptions but provides a
method of projection when the scenario building technique is impossible
to use.
The study combines the scenario building and judgmental techniques
to produce the required projections. Each technique is used to project
C-4
economic activity in the period when it is most appropriate; the scenario
building technique is used in the period between 1980 and 2000 when
reasonable information about possible economic events is available and
the judgmental approach is used after 2000 when there is only limited
information on the possibilities.
The Scenarios: 1980-2000
For the period between 1980 and 2000, specific scenarios are developed.
These scenarios consist of two major components, an economic scenario and
a state government fiscal scenario. The economic scenarios consist of
a set of assumptions which describe_the special projects and industrial
growth in the period. The state government fiscal scenario describes
the assumed level of state expenditures; these expenditures result in
the creation of jobs in both state government and the construction industry.
Each economic scenario describes the growth in the exogenous indus-
· tries: mining, manuf a·c turing, agricul ture-fores try-fisheries, federal
t
government; and in the.exogenous components of construction and transpor-
tation. Each state government scenario describes the growth of state
operations and capital expenditure which affects the level of employment
in state government and construction. The next two sections describe
the economic and state government scenarios.
C-5
THE ECONOMIC SCENARIOS
The economic scenarios consist of time series on employment and
output in certain export base or exogenous industries. These assump-
tions are organized into three separate scenarios which describe a high,
moderate, and low series of economic events which describe what we feel
to be a reasonable range of economic events. This does not mean that we
are predicting that all or any of these events will occur since there is
a highly variable degree of uncertainty with respect to the levels and
timing of
a lrtain
I
I
the events in these scenarios. What it does mean is that with
degree of probability, we expect the general level of economic
abtivity to follow these scenarios. We assume that there is a very high
\_
probability that the level of activity will be at least as great as that
described by the low scenario, a medium probability that the level of
activity will be at least as great as that described by the moderate
scenario, and a low probability that activity levels higher than those
described in the moderate case will.occur.
Primarily as a result of the uncertainty attached to the occurrence,
magnitude, and timing of any particular event, agreement about particular
scenarios is hard to achieve even among those most knowledgeable about
the Alaska economy. Emphasizing our concern mainly with general levels
of activity and the probabilistic nature of any specific scenario should
reduce the disagreement. In an attempt to reduce even further the dis-
agreement, the scenarios were developed based upon existing scenarios
which have attained some measure of consensus. The most important
source for these scenarios were the scenarios developed in the level B
Southcentral Water Study (Scott, 1979).
C-6
The individual scenarios are described in Tables C.l through C.3.
The assumptions are described below; these discussions are organized by
industry.
Mining
Currently, the mining sector in Alaska is dominated both in terms
of employment and output by the petroleum industry. This is assumed to
continue in the future in all scenarios.
All three scenarios include production at Prudhoe Bay and in the
Upper Cook Inlet. Production from the Sadlerochet formation at Prudhoe
is assumed to include both primary recovery and secondary recovery using
water flooding. The Kuparak formation is also assumed to be developed
with production rising to 120,000 barrels per day by 1984. Employment
associated with these developments peaks in the early 1980s "tvith the
development of Kuparak and the water flooding project. Upper Cook Inlet
employment is assumed to remain at its existing ievel throughout the
projection period. This assumes a rising level of exploration, develop-
ment, and production of gas in the Kenai fields which would replace
employment lost because of declining oil production~
The major new source of petroleum production assumed in these
scenarios is Alaska's Outer Continental Shelf (OCS). Alaska is the area
of primary importance to future OCS activity. Nearly 60 percent of oil
and 40 percent of gas resources which are expected to be found in the
United States OCS are expected to be in Alaska waters (Bureau of Land
C-7
(")
I
00
Special Projects
Trans-Alaska
Pipeline
Northwest Gasline
Prudhoe B.ay
Petroleum
Production
Upper Cook Inlet
Petroleum Pro-
duction
TABLE C.l. HIGH SCENARIO ECONOMIC ASSUMPTIONS
Description
The construction of
the TAPS was com-
pleted in 1977.
Additional construc-
tion of four pump
stations is assumed
as well as pipeline
operations.
Construction of natural
gas pipeline from
Prudhoe Bay which in-
cludes construction of
an associated gas
conditioning facility
o~ the North Slope.
Primary recovery from
Sadlerochit formation,
secondary recovery
using water flooding
of that formation and
development of the
Kuparuk formation.
Employment associated
with declining oil
production is assumed
to be replaced by
employment associated
with rising gas pro-
duction maintaining
current levels of
Dates & Employment
1979-1982 -Pump
station construction
of 90/year
1977-2000 -Operations
employment of 1000/yr.
1981-1985 -Construc-
tion peak employment
of 7,823 (1983)
1986-2000 -Operations
begin employing 400
petroleum and 200 trans-
port workers
1982-1984 -Construction
of water flooding pro-
ject peak employment of
2,917 (1983)
Railbelt Location
Operations employ-
ment allocated:
1/3 to Valdez
1/3 to Fairbanks
1/2 of construc-
tion and trans-
portation employ-
ment in Fairbanks
1980-2000 -Mining employ-
ment long-run average of
1,802/year
1980-2000 -Mining em-
ployment of 705/year
All in Anchorage
region
Source
E. Porter, Bering-Norton
Statewide-Regional
Economic and Demographic
Systems, Impact Analysis,
Alaska OCS Socioeconomic
Studies Program, Bureau
of Land Management, 1980.
E. Porter, 1980.
E. Porter, 1980.
E. Porter, 1980
n
I
\0
Special Projects
National Petro-
leum Reserve in
Alaska
State Capital
Move
Beluga Coal
Production
TABLE C.l. HIGH SCENARIO ECONOMlC ASSUMP~IONS (cont.) ·
Description
Petroleum production
in NPRA. Production
in five fields with a
total reserve of 2.5
billion bbls equiva-
lents of oil and gas.
Construction of 525
miles of pipeline.
Movement of the state
capital from Juneau
to Willow begins in
1983. A full move
involving 2,750 state
employees.
Major development of
Beluga coal reserves
for export.
Dates & Employment
Leases held between
1983-1990. Develop~
ment and .exploration
begins in 1985.
Average mining employ-
ment of 460/year.
' . '
1983-1996 -Construe-
Railbelt Location
· All in Anchorage
tion -peak employment region
in 1990 of 1,560/year. ·
Move completed in 1996~ ·
1985-1990; 1994 -Con-Located in
struction with peak Anchorage region
employment of 400
(1987)
1988-2000 -production
employment of· 370/yr.
for long-run average
Source
Based on mean scenar~o
under Management plan 2
in Office of Minerals
Policy and Research
Analysis, U.S. Department
of Interior, Final Report
of the 105(b) Economic
and Policy Analysis, 1979.
High Scenario in M. Scott,
Southcentral Alaska's
Economy and Population,
1965-2025: A Base Study
and Projections, Economics
Task Force, Alaska Water
Resources Study (Level B),
1979.
Pacific Northwest Labora-
tory, Beluga Coal Field
Development: Social
Effects and Management
Alternatives, 1979.
.,
{")
I
1-'
0
TABLE C.l. HIGH SCENARIO ECONOMIC ASSUMPTIONS (cont.)
·special Projects
Outer Continen-
tal Shelf
Petroleum Pro-
duction
Description
Production in eleven
OCS lease sale areas:
Beaufort 1 (1979)
Lower Cook (1981) .
Bering-Norton (1982)
St. George 1 (1982)
North Aleutian (1983)
Beaufort 2 (1983)
Navarian Basin (1984)
Hope Basin (1985)
Chukchi Basin 1 (1985)
Navarian Basin 2 (1989)
Chukchi Basin 2 (1994)
U.S. Borax Mining Development of mining
operation.
Alpetco Project Major petrochemical
facility developed
as originally pro-
posed by Alpetco.
Dates & Employment
Peak OCS employment
-mining -9,066/year
(2000)
-construction -5,300
/year (1992)
Exploration and devel-
opment begins in 1980.
Long-run mining em-·
ployment of 440/year
begins in 1993. ·
Railbelt Location
Lower lease· sale
(68) is in
Anchorage region.
Headquarters
employment averag-
ing 12 percent of
mining is in
Anchorage
1982-1986 -Construe-Located in Valdez
tion -peak employment region
of 3,500/year (1984-
1986)
1987-2000 -Operations
employment of 1,925/
year
Source
E. Porter, 1980 {for
Lower Cook and Bering-
Norton lease sales).
Employment scenarios for
remainder of sales esti-
mated based on N. Gulf
(Sale 55) high case ad-
justed to include LNG
plant (Huskey and Nebesky,
Northern Gulf Petroleum
Scenarios: Economic and
Demographic Systems
Impacts, Socioeconomic
Studies Program, Alaska
OCS Office, 1979).
Northern Gulf Scenario
was apjusted by difference
in resource estimates
to produce scenarios for
specific areas.
U.S.D.A. Forest Service,
E.I.S.: U.S. Borax Mining
Access Road for Quartz
Hill Proposal, 1977.
S. Goldsmith and L. Huskey,
The Alpetco Petrochemical
Proposal: An Economic
Impact Analysis, Institute
of Social and Economic
Research, 1978.
I
Special Projects
Pacific LNG
Project
Forestry/Pulp
and Paper
Manufacturing
n Other Manu-1
1-' facturing ....
Federal Govern-
ment
Fairbanks Petro-
chemical
TABLE C.l. HIGH SCENARIO ECONOMIC ASSUMPTIONS (cont.)
Description
.Const~uction of cur-
rent proposal by
Pacific LNG
Employment in these
industries expands to
accommodate an annual
cut of approximately
1.3 million board
feet by 2000.
Expansion of existing
manufacturing as well
as new local manu-
facturing of locally~
consumed goods.
A doubling of the re-
cent growth rate of
civilian federal
government. Military
government employment
assumed to remain con-
stant at 1978 level.
Moderate.petrochemical
facility using the
state's royalty gas
as feedstock.
Dates & Employment
1982-1985 -Construc-
tion peak employment
of 1,323/year (1984)
1986-2000 -Operations
employment of 100/yr.
Growth of output at
3% a year
Civilian federal
government employ-
ment grows at 1%
per year
1984-1986 -construc-
tion of 1,500/year
1987-2000 -operation
employment of 600/yr.
Railbelt Location
Located in
Anchorage region
Source
E. Porter, 1980.
Approximately 11% M. Scott, 1979
of this activity
occurs in th~
Fairbanks region
81% in Anchorage
region, 15% in
Fairbanks region
and .4% in Valdez
region
• 56% of civilian
employment in
Anchorage, 15%
in Fairbanks, .3%
Regional distribution based
on existing distribution
of employment
M. Scott, 1979.
in Valdez. Military
employment as in
1978.
:Located in
Fairbanks region
J. Kruse, Fairbanks
Petrochemical Study,
1978.
(")
I ....
N
7 Industry
Assumptions
Other Mining
Agriculture
Fisheries/
Food Processing
TABLlf C .1. JIIGH SCENARIO ECONOMIC ASSUMPTIONS (cont.)
Description
Assumed expansion of
hardrock and other
mining opportunities
in the state.
Major development of
agriculture in Alaska.
Reflects favorable
state and federal
policy and favorable
markets.
Continued level of em-
ployment in existing
fisheries. Major de-
velopment of bottom-
fishing, with 100%
replacement of foreign
fishing effort in 200
mile limit by 2000.
Dates & Employment
Growth of employment
at 1% a year. Start-
ing at existing level.
Employment in agri-
culture reaches about
4,600 by 2000.
Employment in fish-
eries increases to
1,350 by 2000. Fiah
hatchery and proces-
sing plant construc-
tion employment
averaging 150/year.
Appropriate expansion
of food processing
industry.
Railbelt Location
Growth is assumed
to be distributed
across regions as
existing mining
employment. (67%
in Anchorage and
2% in Fairbanks)
Source
Major emphasis in M. Scott, 1979.
Tanana Valley.
71% of growth in
Fairbanks and 18%
in Anchorage.
.14% of existing
fisheries and
8% of bottom-
fish development
in Anchorage;
.1% of exist-
ing fishery
in Valdez.
M. Scott, 1979. M. Scott,
"Prospects for a Bottom-
fish Industry in Alaska, 1'
Alaska Review of Social
and Economic Conditions,
1980.
("')
I ..... w
Special Projects
Trans-Alaska
Pipeline
Northwest Gasline
Prudhoe Bay
Petroleum
Production
Upper Cook Inlet
Petroleum Pro-
duction
,1'_ .-
TAaLE C.2. MODERATE SCENARIO ECONOMIC ASSUMPTIONS
Description
The construction of
the TAPS was com-
pleted in 1977.
Additional construc-
tion of four pump
stations is assumed
as well as pipeline
operations.
Construction of natural
gas pipeline from
Prudhoe Bay which in-
clude!') construction of
an associated gas
conditioning facility
on the North Slope.
Primary recovery from
Sadlerochit formation,
secondary recovery
using water flooding
of that formation and
development of the
Kuparuk formation.
Employment associated
with declining oil
production is assumed
to be replaced by
employment associated
with rising gas pro-
duction maintaining
current levels of
employment.
Dates & Employment
1979-1982 -Pump
station construction
of 90/year
1977-2000 -Operations
employment of 1000/yr.
1981-1985 -Construc-
tion peak employment
of 7,823 (1983)
1986-2000 -Operations
begin employing 400
petroleum and 200 trans-
port workers
1982-1984 -Construction
of water flooding pro-
ject peak employment of
2,917 (1983)
Railbelt Location
Operations employ-
ment allocated:
1/3 to Vq.ldez
1/3 to Fairbanks
1/2 of construc-
.tion and trans-
portation employ-
ment in Fairbanks
1980-2000 -Mining employ-
ment long-run average of.
1,802/year
1980-2000 -Mining em-
ployment of 705/year
All in Anchorage
region
Source
E. Porter, Bering-Norton
Statewide-Regional
Economic and Demographic
Systems, Impact Analysis,
Alaska OCS Socioeconomic
Studies Program, Bureau
of Land Management, 1980.
E. Porter, 1980.
E. Porter, 1980.
E. Porter, 1980
Special Projects
National Petro-
leum Reserve in
Alaska Petroleum
Production
Outer Continental
Shelf Pe troleum
Production
Beluga Coal Pro-
duction
TABLE C.2. MODERATE SCENARIO ECONOMIC ASSUMPTIONS (cont.)
Description
Petroleum production
in NPRA. Production
in two fields with
total reserves of 1.2
billion barrels equi-
valents of oil and gas.
Construction of 266
miles of pipeline.
Production in six OCS
lease sale areas:
Beaufort 1 (1979)
Lower Cook (1981)
Beaufort 2 (1983)
Navarian Basin 1 (1984)
Hope Basin (1985)
Chukchi Basin (1994)
Moderate development
of Beluga coal re-
source for export.
Dates & Employment
Leased between 1995
and 2013. Exploration
and development begins
in 1998. Average
mining employment of
286 (between 1998-2000).
Peak OCS employment
-mining -4,900 (1996)
-construction -3,300
(1992)
1985-1990 -construc-
tion -peak employment
of 400 (1987)
1988-2000 -operations
employm e nt of 210/ye ar
1on R -run av era~e
Railbelt Location
Lower Cook lease
sale in Anchorage
region. Head-
quart e rs employ-
ment averaging
12% of OCS mining
employment
Located in
Anchorage region
Source
Based on mean scenario under
Ma nagement Plan 4 in Office
of Miner a ls Policy and
Research Analysis, U.S.
Dept. of In t erior, Final
Re port of the 105(b) Econo-
mic and Policy Analysis,
1979.
E. Porter, 1980 (for
Lower Cook and Bering-
Norton lease sales).
Employment sc e narios for
remainder of s a l e s esti-
mated based on N. Gulf
(Sale 55) high case ad-
justed to includ e LNG
plant (Huskey and Nebesky,
Northern Gulf Petroleum
Sc e narios: Economic and
Demographic Systems
Impacts, Socioeconomic
Studies Program, Alaska
OCS Office, 1979).
Northern Gulf Scenario
was adjusted by difference
in resource estimates
to produce scenarios for
specific areas.
Pacific Northwest Labora-
tory, Beluga Coal Field .
Development: Social Effects
and Management Alternatives,
' 1979.
n
I ......
1.11
Special Projects
Alpetco Project
Pacific LNG
Project
Industry
Assumptions
Other Mining
Agriculture
Fisheries/
Food Processing
/
TABLE C. 2. MODERATE SCENARIO ECONOMIC ASSUMPTIONS (cont.)
Description
Development of modi-
fied Alpetco proposal;
configuration is pri-
marily as a refinery
rather than petro-
chemical operation.
Construction of cur-
rent proposal by
Pacific LNG
No expansion of exist-
ing non-special pro-
ject mining.
Assumes that a rela-
tively low priority is
given to agriculture
development because
of priorities for
recreation and wilder-
ness or the lack of
markets.
Maintenance of current
levels of employment
·in existing fishery.
Expansion of bottom-
fishery to replace
one-half of foreign
fishery in the 200
mile limit.
Dates & Employment
1982-1984 -Construc-
tion employment of
900/year
1985-2000 -operations
employment of 518/yr.
1982-1985 -Construc-
tion peak employment
of 1,323/year (1984)
1986-2000 -Operations
employment of 100/yr.
Employment constant at
1979 level, 2,350/yr.
Employment grqws to
1,037 by 2000.
Employment in .fisheries
increases to 1,228 by
2000. Construction of
hatchery and processing
facilities employs 75/
year. Approriate ex-
pansion of food pro-
cessing industry •.
Railbelt Location
~ocated in Valdez
region
Located in
Anchorage region
Regional allocation
~onl!>tant (67% in
Anchorage and 2%
in Fairbanks)
71% located in
Fairbanks region
and 18% in
Anchorage region
14% of existing
fisheries and
8% of bottom-
fishery in
Anchorage; .1%
of existing .
fishery in.Valdez
Source
E. Porter, 1980.
E. Porter, 1980.
M •. Scott, 1979.
~· Scott, 1979. M. Scott
"Prospects for a Bottom-
fish Industry in Alaska,"
Alaska Review of Social
and Economic Conditions,
1980 •.
Industry
Assumptions
Forestry/Pulp
and Paper
Manufacturing
Otner Manu-
facturing
Federal Govern-
ment
TABLE C. 2. MODERATE SCE~ARIO ECONOMIC ASSUMPTIONS (con t • )
Description
Employment expands to
accommodate 960 mil-
lion board feet of
lumber.
Expansion of existing
manufacturing of
locally consumed goods.
Civilian employment
assumed to grow at
recent historical rate.
Military constant at
current level.
Dates & Employment
Growth of output at
2% per year.
Civilian employment
grows at .05%/year
Railbelt Location Source
Approximately 11% M. Scott, 1979.
of activity in
Fairbanks region.
81% in Anchorage,
15% in Fairbanks,
.4% in Valdez
56% of civilian
employment in
Anchorage, 15% in
Fairbanks, .3% in
Valdez
Regional distribution based
on existing di$tribution
of employment.
M. Scott, 1979.
Special Projects
Trans-Alaska
Pipeline
Northwest Gasline ·
Prudhoe Bay
Petroleum
Production·
Upper Cook Inlet
Petroleum Pro-
duction
-~ ' ·~ .,. ': .I •
TABLE C.3 •. LOW SCENARIO ECONOMIC A~~UMPTIONS
Description·
The construction of
the TAPS was com-
pleted in 1977.
Additional construc-
tion of four pump
stations is assumed
as well as pipeline
operations.
Construction of natural
gas pipeli~e from
Prudhoe Bay which in-
cludes construction of
an associated gas
conditioning facility
on the North Slope.
Primary recovery from
Sadlerochit formation,
secondary recovery
using water flooding
of that formation and
development of the
Kuparuk formation.
Employment associated
with declining oil
production is assumed
to be replaced by
employment associated
with rising gas pro-
duction maintaining
current levels of em-
ployment.
Dates & Employment
1979-1982 -Pump
station construction
of 90/year
1977-2000 -Operations
employment of 1000/yr.
1981-1985 -Construc-
tion peak employment
of 7,823 (1983) · .
1986-2000 ~ Operations
begin employing 400
petroleum and 200 trans-
port workers
1982-1984 -Construction
of water flooding pro-
ject peak employment of
2,917 (1983)
Railbelt Location
Operations employ-
ment allocated:
1/3 to Valdez
1/3 to Fairbanks
1/2 of construc-
tion and trans-
portation employ-
ment in Fairbanks
1980-2000 -Mining employ-
ment long-run average of
1,802/year
1980-2000 -Mining em-
ployment of 705/year
All in Anchor~ge
region
Source
E. Porter, Bering-Norton
Statewide-Regional
Economic and Demographic
Systems, Impact Analysis,
Alaska OCS Socioeconomic
Studies Program, Bureau
of Land Management, 1980.
E. Porter, 1980.
E. Porter, 1980.
E. Porter, 1980
(')
I
1-'
00
Industry
Assumptions
Other Mining
Agriculture
Fisheries/Food
Processing
Forestry/Pulp
and Paper
Manufacturing
Other Manu-
facturing
Federal Govern-
ment
TABLE C.3. LOW SCENARIO ECONOMIC ASSUMPTIONS (cont.)
Description
Assumed reduction in
mining employment in
the state as a result
of land policy or
world market conditions.
Unfavorable conditions
for agricultural de-
velopment. These in-
clude land policies as
well as lack of markets.
Agriculture disappears
in Alaska.
Existing fishery is
maintained but no
bottomfish develop-
ment occurs.
Employment expands to
accommodate 960 mil-
lion board feet of
lumber.
Expansion of existing
production for local
markets.
Civilian employment
assumed to grow at
recent historical rate.
Military constant at
current level.
Dates & Employment
Mining employment de-
clines at 1% per year
from existing levels.
Employment in agricul-
ture declines to zero
by 1992.
Employment remains at
1000. Moderate growth
in food processing to
accommodate expanding
catch.
Growth in output at 1%
per year.
Civilian employment
grows at .05%/year
Railbelt Location
Decline distributed
across regions as
existing mining
employment.
14% in Anchorage
and .1% in Valdez
Source
M. Scott,· 1979.
M. Scott, 1979.
Approximately 11% M. Scott, 1979.
of activity in
Fairbanks region.
81% in Anchorage,
15% in Fairbanks,
• 4% in Valdez.
56% of civilian
employment in
Anchorage, 15% in
Fairbanks, .3% in
Valdez.
Regional distribution based
on the existing employ-
ment distribution •
M. Scott, 1979.
; _._
., -
Management, 1980). The present scenarios are constructed around the
present lease schedule, although the projected probability of finding
oil in each area is considered. For areas with large reserves; we
assumed more than one sale would be held. Table C.4.1 describes the lease
sales, their assumed lease date, the assumed level of resources developed,
the probability of finding oil or gas, and the scenarios in which they
are included. The low scenario assumes no OCS development in the period
prior to 2000; this is a result of assumed environmental and legal
challenges in the Beaufort Sea, lack of technology and market conditions
for the major resource areas in l-lestern Alaska, and only limited resource
finds in the Gulf of Alaska. In the moderate scenario, only the most
probable areas are developed. The high scenario includes both more
areas and a second round of sales in some areas in the moderate scenario.
Although five sales are scheduled for the Southcentral region of the
state, the probability of finding resources in all of them is extremely
low; only the second Lower Cook sale is assumed to be developed in the
high and moderate scenarios.
Both the moderate and high scenarios also include petroleum develop-
ment in the National Petroleum Reserve in Alaska. In the high scenario,
five fields are developed beginning in 1983 and extending through the
period. These fields contain reserves of 2.5 billion barrels of oil
equivalents in oil and gas. Pipelines are constructed to bring the
resources to the Trans-Alaska Pipeline Service (TAPS) pipeline and to
the Northwest gas pipeline. In the intermediate case, two fields with
approximately half the reserves of the high case are developed. This
development does not occur until near the end of the period in 1998.
C-19
TABLE C.4.1. FUTURE OCS ACTIVITY
Lease Sale Area Reserves 1 Risk Factor 1 Scenario
Oil Gas (Probability (H = high,
(billion (trillion of finding M = moderat
barrels) cu. ft.) no resources) L = low)
Beaufort Sea .04
Sale 1 (1979) .75 1.6 M,H
Sale 2 (1983) .75 1.6 M,H
Northern Gulf
(1980) .s 1.3 .95
Lower Cook (1981) .2 .5 .95 M,H
Bering-Norton
(1982) 1.4 2.3 .60 H
St. George .40
Sale 1 (1982) 1.4 5.2 H
Kodiak (1983) .2 5.4 .92
North Aleutian
Shelf (1983) .7 2.7 .29 H
Navarian Basin .33
Sale 1 (1984) 2.8 9.8 M,H
Sale 2 (1989) 2.8 9.8 H
Chuckchi Sea .30
Sale 1 (1985) 2.1 5.2 H
Sale 2 (1994) 2.1 5.2 M,H
Hope Basin (1985) .43 1.72 .35 M,H
1Alaska OCS-Office, BLM
C-20
In addition to the petroleum development, .some other mining is
assumed to take place. Development of the U.S. Borax mining operation
at Quartz Hill in Southeast Alaska is assumed to occur in the high
scenario. In addition, development of the Beluga coal resources is
assumed in both the moderate and high scenarios. In both scenarios,
coal is assumed to be produced for export.
The special projects described above do not exhaust the mining
employment in the state. Additional employment occurs in the explora-
tion, development, ·and production of nonpetroleum minerals, as well as a
major component of headquarters employment in Anchorage. Market forces
and governmental policies are assumed to be such that this component of
mining declines in the low case, remains constant in the moderate case,
and grows in~the high case.
Table C.4 describes the three separate mining scenarios used in
this study. In the low scenario, mining rises in connection with devel-
opment at Prudhoe, but falls after 1983. By 2000, mining employment is
almost 275 less than in 1980. Growth occurs in both the moderate and
high scenarios throughout the period. By 2000, mining employment is
9,500 greater than in 1980 in the high scenario and 2,900 greater in
the moderate.
Agriculture-Forestry-Fisheries
This industry is, in reality, three distinct subindustries which
represent Alaska's renewable resource industries. Of the three, the
C-21
TABLE C.4. MINING EMPLOYMENT
(thousands of employees)
Low Moderate High
Scenario Scenario Scenario
:1.980 ~::; .. '\ ""1f::· r.:-1. 3'7' r.:-:1.63 .,. t .. -l .... J .... J {o .. } •
:L!f81 ~:; + :1. 6~5 r.:-. ., ... ... 5 36-4 ,; • .. :a .L / •
1~->82 I • ~12~~ 7 -;·':i.3 ""? 784 • , . •
:1. 9~33 ~\ () ~~\ (~) !""!-+ ····-~ 8 6'?!3 t> • Ci • .\ // •
:I.9B-4 ..... i!):2:4 8 06:1 . B 410 I .. .. .,.
19B!5 1::· ,; • :i ::} l~. ~) • .4;~~5 6 .. O!:):i.
:1. (_:;t~ C) ·-r.~· 097 1:!' ;=·''?£{. 6 ~j~i9 ,J .;. ·..J • ..
1 '7lfj~7 !5 • ()~7~5 \.1 • 3<S6 7 .. .. -,r.::o ., ...... .; ...
1 ,-, .. ..,. ..... .. ,. (j(".) ;3 • ()5·4 6 .. .:>55 7 • 9Cl•:;
:1.989 ~=.i o) 05() I 6 .. , 4? 8 40:L + I • 19{1() 3 0:1. :L 6 , ........ -.. ·-:; 885 • • -~!.:>I •
1. ·:j 9 :f. . -:'t 9'?0 . l.i t::-.. ~.· C'l :1. () 6C1 :L ·> ·> .... J .. :i'-) •
;LSHJ::!. lt • {;i .:S 9 6 • 8-d.C.: ' ' :1. 1 • ,., ~( .. .:..• .. ) .. !.
1"?93 4 ; ~-~ /.{. s:· 7 • 2 \·:~'J. 1 1 .. 91 :1.
19';'4 ··4 • 9:::.~~3 ,.,
0 • ()89 :l =~ + 7-4rl
19";;.'5 l+ ·~/()8 •'''I 1. 51. 13 .. 2·43 • (.") ·>
1996 4 • t;:ti~3 8 .. 2·4·;:t :1.3 • 808
:i. 9'7~7 4 • .··-. .. ;·-:::"' Oz .. ) I...; 8 • :i. ~s ~:5 1 ., .;> • 966
J. '7 s=· tl ·4 ::;.<"J.a ~, O(i? :i. 4 2~)9 • 0 • •
:i.99(? 4 • 829 8 • ().4;2 14 + 650
:~~0()0 -4 ,. ......... 8 0~)-4 14 639 • (:).J.'.,..t • •
-.. ... -.;-.._·_, .
fishing industry is currently the largest in terms of both.employment
and value of product •. Agriculture is currently only a marginal industry
employing few people statewide (Scott, 1979). Current state
efforts to develop agriculture may lead to its increased importance in
the future. Forestry consists of only a small component; the future of
forestry is most appropriately discussed with the future of lumber and
wood products manufacturing.
The future of agricultural development in·the state depends impor-
tantly on governmental policies and actions. State and Federal land
policies, infrastructure development and loan programs, and marketing
programs will determine the future of this industry. In the low sce-
nario, it is assumed that government policies do not favor agriculture.
New land is not opened up; ·old agricultural areas suffer from competition
With other land uses (recreation, residential) and from competition for
markets from outside producers. In the low scenario, agriculture dis-
appears in Alaska. The high scenario ass~m~s a major positive government
effort in support of agriculture with a fifty-fold increase in land in
agricultural production by 2000. In the intermediate case, agriculture
is assumed to rise only slightly from its current levels of employment.
This assumes, as in the low case, that agriculture receives low pri-
orities from government.
Fisheries also hold promise for the future. The major determinant
of future increases in· fisheries employment will be the expansion of
the Alaska bottomfish industry. The creation of the 200 mile limit may
C-23
support increased Alaska bottomfish activity. In all cases, employment
in the existing fisheries is assumed to remain at its current level.
Increases in production are assumed to have no effect on employment
because of limited entry and labor-saving improvements in the fleet.
Employment increases occur in both the high and moderate cases as a
result of the development of an Alaska bottomfish industry. In the high
case, the Alaska fishing industry is assumed to replace all of the
existing foreign fishing·effort inside the 200 mile limit; while the
moderate case assumes only 50 percent replacement. No bottomfish industry
is. assumed to be developed in the low case.
Table C.S illustrates the three agriculture-forestry-fisheries sce-
narios used in this. study. In the low case, employment decreases by
about 170 over the period due to the reduction in agricultural employ;..
ment. The high case shows employment rising by almost 4,700 during the
projection period. In the moderate case, employment rises by almost
1,000 between 1980 and 2000.
Federal Government
Federal government employment has always been an important component
of Alaska's economy. In recent years, federal government employment has
been growing very little; increases in civilian employment have been
offset by decreases in military employment. Low rates of growth in
federal government employment are assumed to occur in all three scenarios.
In all scenarios, federal military employment is assumed to remain
constant at existing levels. In the low and moderate cases, federal
C-24
TABLE C.S. AGRICULTURE-FORESTRY-FISHERIES EMPLOYMENT
(thousands of employee~)
Low Moderate High
Scenario Scenario Scenario
.. 1'n10 1 + ;~03 1. + ~)0:3 j_ v304
1.981. l .204 1 • :3:1.-4 1 + 32~5
., r''"'"'·~ :i. ·' r"-J :i. .::n2 :i. ........ OJ'"'
J. "To..: • .L r .•. o} ..:}~-<t
1.983 l • :L83 1 (·· ~3 ~::. :L . ----·---1. v -4()3
:1.984 1. • J. 8:3 :i. • ~} <;= ~5 1. .,480
198!"5 1 • :1.55 1. A_.., .. \
+ ·'t Jt,·:. i' :l + !5-41
1.986 :L • :1.54 j_ • o\t~)9 :J. (p ~)8\)
1. <?G/' J l ::2_.-=~;. :1. + -4~38 :l ll'"\1"\ • t-C>L-7
1.9f38 ·r 1 (J!:j 1. + ~:5 ~:5 J. 1 / .. 32 -~ • •
19G9 -:l .087 :L • 612 ·! +863 ..
1. 9'70 :l (1..., ,.
• ,}/ 0 1 ~ 6·:;'~:; 2.069
1991 :1 .• 06~! :1. _,,_., ""'r
+I ,':J/ 2.309 1 (;)•;:) •")
, J -~ :1. +031 1 + 7i.Y·7 ;2-l-620
J q(">"Z ~ .1' 'l w :1. ·:-()32 :l. (""t ·" -'"'\ +C) C)~·-3.029 I
19S.'4 :L o··:t ... 't 1 s~~2~l -.r 450" ' • ...:> .. .:.: • .....
' 1995 :l + 03~) 1 .. 7'86 3 + 74-4
1996 :1. • o~}::s :~+04~1 4.054
1997 :l + (;~)3 2. llO 4.462
1.998 ·I .033 ~, :L 7:1. 4.896 .~ ~v
:l 9(?·~~) :L (pC•33 .·: .23:3 5.396 .... ~
2000 :1. • 0~5:~ ~~-):~95 5+996
C-25
civilian employment is assumed to continue to grow at its historical
rate of about .05 percent per year. In the high case, this rate of
growth is assumed to double to one percent per year. Table C.6 illus-
trates the three alternative federal government employment scenarios.
Manufacturing
The manufacturing industry in Alaska has four important components:
seafood processing, lumber-wood products-pulp, petrochemicals, and manu-
facturing for the local economy. (Assumptions are discussed in terms of
industry product since this is their form of input in the MAP model.)
Production of seafood processing is expected to continue to dominate
the food processing industry in Alaska. Growth of this industry is
dependent on the growth of the fisheries catch by Alaskans, so these
scenarios reflect the fisheries scenarios. In all scenarios, the output
of the food processing industry is affected by growth in the catch in
existing Alaska fisheries and growth in the bottomfishery. In the high
case, output in the food processing industry is assumed to expand by
100 percent between 1980 and 2000 due to increases in the catch of the
existing fishery and by an additional 57 percent because of the develop-
ment of a bottomfish industry. In the moderate case, output expands ~y
149 percent in existing fisheries and an additional 49 percent because
of the bottomfish development. In the low case, no bottomfish industry
is assumed to develop, so output expands only because of increased catch
in existing fisheries, and a 22 percent increase is assumed.
C-26
J
TABLE C. 6 • FEDERAL GOVERNMENT EMPLOYMENT
(thousands of employees)
Low Moderate High
Scenario Scenario Scenario
1 ~)8() 4~3 ·':) :'i ·"" A,.,_. . -· .-, r.,... • ~ ....... ~ ~)· .. :. ·> ... ~·: \}f.j i~ "=· + .-:.'J\J
19~31 "'':." "'t,;) • ·<10 () A1.:1 • ... ,., ....
.-;~. •,,/\,/ 43 • ·40(>
:1. '7"c1;.~ 4~i;. 600 4:::-> . , 600 4::> • l>0()
1 9FJ:3 ·43 • 7(){'
" .... .._1 --:13 • ?0() 43~ 80()
l'l'E::..:} -4:3 • ~~(){) A1 ::~: • ~:i 0() 4~) • 'iOO
198!5 43 + ·t~-)!J 4~5 ·> '-;: t.,} ~) 44 ·> lOG
·f 9B6 ·4.-;~ ()()() .\{ .• -:t (' " .. , 44 -.... .:•. '"'\
J. • • ,li\.}t.} • ~\}t,,:
1 ~y'B:;/ 4-4 t ()() .• .... :l. ........ lf4 ~)()() • ~t .!.~ ·> t..)' .. J •
1.S>ti Ej ,.~~4 ... , .... '" -44 ~:~o ()· 44 + :70() .,:.t,}'J ·>
:L 9f~ s:· -4~+ .300 4A·+~500 44 ., 900
:1.9':>'0 44 400 44 1: •••• , •• ·4~5 1.00 <· ·> ··f \) ... , •
1.99:1. 44 • ::'iOO ·4·4 • 5()0 ..ct5 + 300
:1.9?2 ··:f.~~} • <SO() ··J-<{. • ~~· ~) () 4!:) • 500
:l9<7'3 -4i~ (o -1' ()(). 44. '7()0 45 • BOO
:1.994 44 .. 80() 4·4 • f~ () () 46 + ooc,
l99~'i 44 900 4-4 9()() _, ·' 200 • • -4t0 •
:i.996 45 + 000 -4!:5? 000 46 .400
:l ~1 r:;· /' 4!5 • J. ()~ ·' t·--'·{ .::..• • :L 0() ·4C) • 60•)
j ,.,,..,, .. ,
• -;~ 7 i...1 -4~j • 20() .-:t~.) ·:r-"2(;() 46 • 800
:i. 99S·) .l~5 ~ 3()() -45+ 300 47'. 00()
2000 4!5 • 400 -45 • -40() ·4i"' + 300
C-27
The growth of the lumber-wood-paper-pulp sector of man~facturing in
the state is determined primarily by two factors. These are the Forest
Service allowable annual cut and the Japanese market conditions. In the
high case, these industries' growth reflect almost a doubling (over its
1970 level) of the annual cut by 2000. In the low and moderate cases~
growth in annual cut is only one-half this amount.
The petrochemical industry in Alaska currently consists of the
developments in Kenai. In the low case, there is no expansion of this
industry. In the moderate case, the petrochemical industry .expands with
the construction of the Pacific LNG facility as currently planned, the
development of LNG facilities associated in the OCS activity in Western
Alaska~ and the development of a fuels refinery as the ALPETCO project.
The high case contains two additions to these projects. A petrochemical
complex is assumed to be established in Fairbanks, using the state's
royalty gas, to produce ethylene or fuel-grade methanol. The Alpetco
project is assumed to be developed as a major petrochemical facility as
originally proposed.
The final component of the manufacturing industry consists of those
industries producing for local consumption and other diverse specialized
production. It was assumed that this sector would grow in all scenarios
because of increased market size, allowing scale economies which make
local production viable. This sector was assumed to grow at 1 percent
per year in the low case, 2 percent in the moderate, and 5 percent in
the high.
C-28
Table C.7 shows the three alternative manufacturing scenarios.
Manufacturing employment increases continually through all scenarios.
It increases by SO percent over the period in the low case, .83 percent
in the moderate, and 137 percent in the high case.
Transportation
I
The exogenous portion of the transportation industry is that which
serves special projects. In all scenarios, this industry includes the
.operations employment for TAPS and the Northwest gasline. The other
major source of transportation employment is.the· OCS petroleum development.
This employment is associated with both SUPPfY ships and helicopters used
in the OCS development. The difference in transportation employment
reflects the difference in the OCS lease sale areas assumed to be developed.
Table C.8 iliustrates the three transportation employment scenarios.
Construction
The final exogenous industry for which scenarios are required is
that portion of the construction industry where the level is determined
outside the economy. This sector includes construction employment
associated with the special projects described above. This sector does
not include capital improvement projects of any level of government or
construction activity which supports the local economy;. the remainder of
construction activity is determined endogenously in the MAP model. In
all scenarios, the major development of special projects occurs in the
early part of the projection period. The most important project during
this period is the construction of the Northwest gasline which is assumed
C-29
:1.'700
1'7B1
1982
:J.<Yt53
1. 98-4
1 9:3!.5
1. ~"'8<.1
:i.9B7
1 ';> 8f~
1913S)
1.99:l
1·9S)2
:J.l.?93
1.994
1995
17"?6
19?7
1998
:woo
TABLE C. 7 • MANUFACTURING EMPLOYMENT
(thousands of employees)
Low
Scenario
:1.2. ~):i. f:l
12 .,. s-~6f:3
l :3 i :3-.:f.-4
:L :7) f .s ~) .·ff
14~()0()
14+:~0~~
1. . .;:} , ... ~:; ';> '?
,1 ;::• ,••, .r '"' .l ,J "' t.j .L ~::.
:J. !5 ~-2 .(~. :L
1!5 , .. 48?
l !5 + /' .\l!)
:1.6. 0:1.2)
•f ·' ·-)(;\r''\ .i.O -o-..-..\.)7
1.6.570
16. a5a -
l/ + j_~)-4
1 j! (o -4!:5.:S
:l?.,. 771
:1. B ~ 09/'
:1.13. 4:34
C-30
Moderate
Scenario
:t~o):'?-41.
_, 1~:-··:rr.:·~
1 •... J ·:0 -.:> -...i ... .r
1 ~:; T ·::;. :~2 4
l (:) v ~~8~5
j, '7 + !~j-4:f.
·t ,., -~ -:-· r:~ .L C) • .1. ,J •.• J
l f~-o) ~)~~5
'"1E)-=-::;'-4-4
:i.'~i. :35S\
2:1. ~ 3!:)-4
21 + 8/'()
_.,,.., .. ,..'lC"
.,:....;:,y ~ ... :.0
High
Scenario
l~J + ~:)48
14.2/'!..:t
:!.5.1.76
l ~:; ~ 9C)~)
•f ·' ·-t I' ... , .L<:>+/07
:1.? ., 54J.
:i.El.317
~~1 + -46i7
:~2to :L4l
;.~3. 610
:23.341
24~350
2~5.461
::!<S •. :582
2:/' t 555
28.,.601
29.370
~}().139
31.180
• -:i 980
1981.
1982
1'7'83
:t 9<':)·4
.. ,.,, ... , ...
·.L ;' \:>\:>
:i.988
:L 99<)
:L993
1994
·I •'""•"":I!':" J. '17 .... 1
:i. ~~'?6
199·7
1'1'98
:i. c;~i\)-'
:2000
TABLE C.8. EXOG~NOUS TRANSPORTATION ENPLOYMENT
(thousands of employees)
Low
Scenario
..__ --· ----·.
.t , ... ~ ....
.1. • ;".)~.v
:L • ;:)00
:!..500
1.500
1.500
1 + ?0\)
:t f. ·?c~o
:L.?OO
:t-.;. ::.:-' () 0-
:l + 70(j
:I.+ 7()0
l (, '70CJ
1.700
1.700
:L .. /'00
:1..700
1 + 7<J·()
:! .• 700
1.700
:l <· 700
C-31
Moderate
Scenario
:L .,. ~50{)
.t l::>r,t::• .1. + .. J~~o-: ... _1
~~ •. ~5lti
2.,3:tOi
2 ~ ~549
2t6~~:L
2<-t.)6t3
'2-)'?0~5
.• ., •'''1''7 " ••
-<-•0<0
~-~ + t)():~
:~~~596
High
Scenario
:!. • 500
1. ·> !500
:i. t 52~5
1 + !:f82
:L .... 85~5
2~~4f~:l
2 ~ !:558
~~ y 92~)
,..,_ , ·" ,.,.
,. .. •• <-t:ic .. ~·.
,. ... -_,"" 4 .-:.~ + 1·"1-.. :..~
2 t 9'72
3.291
3.:397
~5 + 722
~s~;;.-76
3.901
3 .; 986
3.913
3.751
to begin in 1981. This is the only special activity assumed in the low
case. The high and moderate cases reflect completion of other projects.
Both cases assume Pacific LNG and Alpetco projects will begin in 1982,
although a more massive-scale Alpetco development is assumed in the high
case. Construction employment is also required in the development of
the OCS fields, NPRA, and Beluga. Additional sources of construction
employment in the high case are the construction of a new capital at
Willow and a petrochemical complex in Fairbanks.
Table C.9 illustrates the three exogenous construction scenarios.
In all cases, employment peaks in the early 1980s. This peak is pri-
marily a result of the construction of the Northwest gasline·which is a
major one-time project. The bunching of other large projects, as well
as the beginning of OCS development at this time, also leads to this
early peaking.
"THE STATE GOVERNMENT SCENARIOS
Past studies of the Alaska economy have indicated the key role
state government plays in the Alaska economy. State fiscal policy has
been a major determinant of state economic growth. State expenditures
determine not only direct government employment, but also through expen-
ditures on goods and services and capital improvements, they will affect
all endogenous sectors of the economy. The state government scenarios
described in this section attempt to define the most likely range of
state government activity.
C-32
TABLE C.9. EXOGENOUS CONSTRUCTION EMPLOYMENT
(thousands of employees)
Low Moderate High
Scenario Scenario Scenario
--.. I I
1~18() • O'iOj .<>s:·o~-.090 .J
.f (''of"":·! J. 7 (;) .1. () ~ ~)S:'C' o. 73.":4 _., .... n.-,
\) ~ /07
1982 2.,8a5 ·4 + 3:3~5 3t 960
198:3 I' 8,., .. ., + ,:;. . .:) ,., ·•n····,
7*C>07 U. "I(~ C) .; ~t..\.J,
:t 98-4 '7 ~ ()38 9 r.--···-· + .. :)\j..;} l4v488
. J. 98~:.:; 1 r·-, ·-+ ~(j..:} 2 ~ 35!.=5 8. 963
:L 9B.:; o .. ooo :L ~ (j ~:.~ '? (~ ~ :;:?2
1 ;;i<:5:? o.ooo 1 :i. ~:)·:;;-•;.· ·"' t::"' , .. , .. ..:>.,; "t,JO
:l9E~3 0 ·> 000 1 •I ''1'1-,. ~-. M''Y .. ,. • .I.~C) .:..; 'l..:i..:>
198S-~ o.ooo 1. • 08~) 2+7..;fS:t
1. 99(j o.ooo 0 + ·4("32 3 ., 88-4·
199:!. ()v_0()0 :i. + (j()g> 6o293
199;.~=: () y ()()()
,..) r::· ,, C) ' 7()1 ~--~ ... J.L.~ Ot
•! !'~ ..... ··~·· • 1. 'l ·;-· ... ::a 0 ~ ()0() :'"\ •··.• ,, __ ,
... ...:. y / ~} .1 . , Jj--.,1~
b+.:-17
199-•l· 0;. ()()() 1. ,:.0\::i-4 ·4 + 158
:l9i_;t5 Ot•JC'O o. ~5() ~:-· 3. 193
1.99r.~ 0 ->-()0() 0 + :2!)2 3 + ·<+36
1. ~19/' o.ooo o .. -r r::·.-, ..,. 101 .L..J . ..::. ,j +
19~tt~; i.).? 00() Ov3()~,j .-., .•, ,,
.t::. t 00~)
1999 o.ooo () ·> 40() o. 714
...-_ ............. ~\,'i .. }v o .. 00() C•,t54~:; 0 ~ 7~18
C-33
Two factors affect our ability to project the future course of
state expenditures. First, since the beginning of production at Prudhoe
Bay, state revenues have overtaken expenditures; revenues from this pro-
duction will continue to increase in the projection period. Secondly,
the establishment of the Permanent Fund and recent tax reduction and
' wealth-sharing prog;.am? place constraints on the use of certain petro-
leum revenues. These recent changes in the structure of state spending
constraints limit the usefulness of past fiscal policy for determining
projected future policy.
For this study, we will assume thr'ee separate directions for state
fiscal policy, each of which will be defined by the growth of real per
capita expenditures. Real per capita expenditures measure the effect
of increases in prices and population on state expenditures. Between
1970 and 1972, real per capita expenditures grew at almost 24 percent
per year; this was primarily a response to the lease sale bonus of
$900 million from Prudhoe Bay in 1969. After 1972, the rate of growth
dropped to .5 percent per year.
We will describe the growth of real per capita state expenditures
in terms of its relation to real per capita incomes. The relationship
between income and state expenditures will be described in terms of the
income elasticity of state government expenditures; this elasticity equals
the assumed proportionate increase in real per capita expenditures which
would result from a one percent increase in real per capita income. The
historical pattern of state expenditure growth shows real per capita
C-34
expenditures as an increasing proportion of ~eal per capita income
through most of the period., The proportion increased through 1971 with
a rapid expansion between 1969 and 1971 as a result of the Prudhoe lease
sale bonus. Between 1971 and 1977, the ratio of real per capita expen-
ditures to real per capita income remained constant (Goldsmith, 1977) • .
The state's present revenue situation makes it hard to forecast how this
ratio will change in the future.
Our three scenarios assume that real per capita expenditures consume
a growing, constant, and declining portion of real per capita income.
The low case assumes that the level of real per capita state expenditures
stays constant through the projection period and real per capita state
expenditures decline over the projection period as a proportion of real
per capita income. The moderate case assumes the real per capita state
expenditures proportion of personal income stays constant with real per
capita state expenditures increasing at the rate of real per capita
income~ Finally, in the high scenario, real per capita expenditures
increase at one and one-half the rate of real per capita income and
increase as a portion of real per capita income.
In combination, these three state expenditure scenarios and the
three economic scenarios produce nine growth scenarios for the period
between. 1980 and 2000.
C-35
POST-2000
For the period between 2000 and 2010, a judgmental approach to
projecting the level of economic activity was used. The approach used
for the post-2000 period was to assume a rate of growth which described
the possible continuation of the high, moderate, and low scenarios. In
each case, a similar rate of growth was assumed for all three scenarios
for the major variables--employment, population, and households. This
implicitly assumes changes in household formation and labor force par-
ticipation assumed between 1980 and 2000 do not continue after 2000.
The assumed growth rates describe three possible post-2000 growth paths
which are based on examination of growth in other similar areas as well
as the historical growth of the Alaska economy. The high case assumes
a continued expansion of the Alaska economy as a result of increasing
resource development, although a reduced role of state government. The
major economic variables are assumed to grow at 3.3 percent per year,
which is approximately the rate in the high economic-moderate government
scenario in the last part of the period. The moderate scenario assumes
slightly slower growth at 2 percent per year, which is slightly less
than in the moderate economic-low government scenario. This growth is
assumed to result from more moderate resource development and reduced
government activity. The low scenario provides only minimal growth at
one percent per year, which reflects a self-generated growth 'from govern-
ment expenditures.
C-36
Projections of State Growth 1980-2000
This section presents the statewide projections of future economic
and demographic activity. These projections are the basis for the
energy end-use projections. The projections presented in this section
are projections of the MAP model and the economic and government scenarios
described above. The combination of three economic and three government
scenarios produced the nine alternative projections presented here.
Table C.lO describes the projected growth of total employment in
each scenario. As would_be expected, the combination of high economic
and high government scenarios (HH) produces the greatest growth~ and the
low economic-low government scenario (LL) produces the lowest. In
scenario HH, total employment .grows by over 300,000 between 1980 and
2000, an average annual rate of growth of 4.5 percent per year. Total
employment grows by only 78~000 in the ~cenario LL, which is an average
annual rate of 1.6 percent per year. In all the scenarios, the bunching
of major construction projects in the early 1980s results in the most
rapid growth occurring in this period.
The effects of the alternative economic scenarios can be seen by
comparing three economic scenarios with the same government expenditure
assumption. We will examine those scenarios with a moderate level of
government expenditure. Total ~mployment grows at an annual average
rate of 2.3 percent per year in the low growth case, a growth in employ-
ment of 122,150. In the moderate case, total employment grows by 161~420
C-37
1 =:_;.:·=.;.:()
:1. =:? ·:) :.~~;
·' , •••••.••• ,!'".
l. '7 Ci iv'
·: ::::==::)!:-:
·'· J ,_,,...;
TABLE C.lO. TOTAL EMPLOYMENT, 1980-2000
(thousands of employees)
i_EE T CJL.
SCENARIO NAMES:
:.~:30 ~ '7' 16
:~::)B + 1. :?<)
210 .. 0 '·?';.~
304.(·:1."7
445 + 032::
/
• -. I , ·-.• ,• r •. •••t .~.-:: ·'·~ ,:) + (':) ··;' /
-2 '/ :i-~-\).I ~5
::~9A \\ 96::)
LES.GL -Low Economic/Low Government
LES.GM -Low Economic/Moderate Government
LES.GH -Low Economic/High Government
MES.GL -Moderate Economic/Low Government
MES.GM -Moderate Economic/Moderate Government
MES.GH -Moderate Economic/High Government
HES.GL -High Economic/Low Government
HES.GM -High Economic/Hoderate Government
HES.GH -High Economic/High Government
26<_;: ... ~):;B
::29~::; ~ j·'f3~5
3-49 ·) B;~~5
t·iES. GL
-,.,-1-'~ . -0· -;_;--9.-,
··-: ·_V: ~-~'-
2-45 ;098 .
::~5B. l 02
292.023
319+9'72
'I •---, ,..,,~ t··; 1:. :J + L"JI ;I
2:to~o~~r;
29:1..414
3 30 + 47"7
Note: Values in 1980 adjusted to be the same in all cases; adjusts
for minor differences in exogenous series.
C-38
between 1980 and 2000, which is 32 percent greater than in the low case
and an average annual growth of 2.9 percent per year. Total employment
in the high scenario grows by 244,550, which is an annual average rate
of 3.9 percent per year.
The effects of the alternative government expenditure scenarios on
economic growth can be examined by comparing three alternative projec-
tions 'tvith the same economic scenario. Examining the projections with
the moderate economic scenario and low, moderate, and high exp~nditure
scenarios shows that the effect of ·varying state expenditure scenarios
is similar to altering the economic scenarios. Under the moderate
economic growth scenario, total employment increases at an annual average
rate of 2 .1. percent per year·-Total employment increases at. an annual
average rate of 2.9 percent per year in the moderate expenditure case.
This is 38 percent faster than in the low case; when the government ex-
penditure assumptions are held constant at the moderate l~vel, the growth
rate in the moderate economic scenario is 2.6 percent greater than in the
low. The average annual rate of growth in the high government scenario
is 3.4 percent per year, which is 17 percent faster than in the moderate
scenario. This compares with the 34 percent difference in growth rates
between the moderate and high economic scenarios.
Examining the effects of altering the government expenditure scenarios
shows that in all cases state government expenditure is expected to play
an important role in projected future growth. 1 State government employ-
ment assumes a different role under each scenario, which reflects the
C-39
alternate assumption about state government expenditures as a proportion
of personal income. State government employment as a proportion of total
employment falls in the low scenario, increases slightly in the moderate
scenario, and increases in the high scenario. In 1980, state government
employment is 21 percent of the total. By 2000, this proportion has
fallen to 19 percent in the low scenario, risen to 23 percent in the
moderatescenario, and risen to 26 percent in the high scenario.
The importance of state government spending to the projections of
total state activity makes it necessary to examine the consistency of
these projections. It is necessary to ask whether the state can make
.1 ~1
.]
·~ :,J '1
. ;.!J
i ·~
-~
.J
~~
41 -~
-~
this level of expenditures without running out of money or requiring {~
.~
large increa.ses in taxes. One consi~tency check is to examine the state' sJ
fund balance in 2000. The fund balance is where the state accumulates · ::.1
excess revenues; it includes both the Permanent and General Funds. The
most important source of revenue for the state during the projection
period will be petroleum revenues. The revenue projections used in this
study are based on the most recent projections of the Alaska Department
of Revenue (Alaska Department of Revenue, Petroleum Production Revenue
Forecast: Quarterly Report, March 1980). Based on this assumed gro~~h
in revenues, the fund balance is positive and large in all scenarios in
2000. Only.in the high economic-high government expenditure scenario
has the fund blanace peaked. This scenario has the lowest level of fund
balance in 2000, $48.9 billion (in current dollars). Given the petroleum
revenue assumptions, all three of the assumed government expenditure
scenarios are possible.
C-40
Table C.ll describes the growth of the population in each scenario.
In the high economic-high government scenario, population more than
doubles over the period, growing by 487,000. In the low economic-low
government scenario, population is projected to increase by only 36 per-
cent~ In all scenarios, population growth follows the pattern of employ-
ment growth.
Examining the moderate government expenditure scenarios illustrates
the effect of the different economic scenarios on population growth. In
the low economic scenario (LM), population grows at an average annual
rate of 2.1 percent per year, reaching 635,578 by 2000. In the moderate
economic scenario, population grows slightly faster (a rate of 2.6 per-
cent per year); by 2000, population in this case is 10 percent greater
than in the low case. Population in the high case reflects the rapid
economic growth assumed in this case. · Population grows by 97 percent;
by 2000, population is 19 percent greater than in the moderate case~
The effects of the alternate government expenditure scenarios pro-
vide as great a variance as the economic scenarios. By 2000, population
in the moderate economic-moderate government scenario is 12 percent ·
greater than in the moderate economic-low government scenario. The
moderate economic-high government scenario projects population in 2000
which is 8 percent greater than in the moderate government scenario.
Population growth rates between 1980 and 2000 vary from an annual average
of 2.0 percent per year in scenario ML to 2.6 percent per year in MM and
2.9 percent per year in scenario MH.
C-41
J. ·~=-==Jo
1. 99~5
TABLE C.ll. POPULATION, 1980-2000
LE13. C!L
4 6 :-",' i-J. ~:; ... -~.
.!-~·:;.~ov ~3t6
• ,., ·t --· -· -·· , . . "+~ ..• /~/
5 () ::> y 9 .-:·i-~:-~
~) 4 :.:.• .; •:_; Cf' \;)
Hi:::;;;. C7H
55 0 + :5 ~}-4
64~:j ·i :.~~5 :i.
(thousands of peoplk)
··----··---:]
.. A2:1.. 737 !
-
4\::i J. v 34::s
~:5 :L i + c:.::)~5
HES.GH
-t21. 73"71
!.=5 :i. ~:i ~ ~~ :·; '7
~=j6 {r ·:· ::::6 ?
~".) ~~; E: -) ~~~ ~:.i ?
7 :~=; :J -:--4 -'1· ::)
LE3 .• GH
HES.GL
4;_:~1 . /3/·'
~:5 J. ::~ ·> ?' ::2 0
<S 6 0 .,. () _,·:;. 3
7~.3 -:L +!54~~
421.737
~l f~ ~i ~ -:~-6 3
51 s . -'t9 6
57~5 + 22 7
6.27 .1so
HES. C1f-i
421. 73 i
----·-......,..------------
SCENARIO NAMES:
LES.GL -Low Economic/Low Government
LES.GM -Low Economic /Hodera te Government
LES.GH -Low Economic/High Government
MES .GL -Moderate Economic/Low Gov ernment
MES.GM-Moderate Eco n omic/Moderate Government
MES.GH -Moderate Economic/High. Government
HES.GL -High Economic/Low Governmen t
HES.GM -High Economic /Hod e rat e Government
HES.GH -High Economi c /High Governme nt
Note: Values in 1980 adjusted to be the same in all cases; adjusts
for minor differences in exogenous series.
C-42
In all scenarios, population grows at rates slightly lower than
employment. This reflects, in part, the increased labor force participa-
tion of both Alaska Natives and women and the changing age structure of the
population. Total employment as a proportion of population is 49 percent
in 1980. By 2000, this proportion is 56 percent in scenario HH, 54 percent
in scenario MM, and 50 percent in scenario LL. The difference between
scenarios results from the importance of migration in each scenario.
Migration brings in fe>ver dependents per employee than in the existing
population. Migration is more important as a source of population
growth in the moderate and high scenarios. This is responsible for the
greater increase in employment as a proportion of population in these
scenarios.
Table C.l2 shows the growth of households in each scenario. House-
hold growth reflects two factors, the growth of the population and the
changing structure of households reflected in an increased probability
that certain sectors of the population will form households. All sce-
narios follow the same pattern of increasing proportion of households in
the population. The pattern of this change can be seen by examining the
low economic-low government (LL), moderate economic-moderate government
(MM), and high economic-high 'government (HH) scenarios. In scenario LL,
the number of households reaches 210,790 by 2000; this is a 58 percent
increase during the projection period. The number of households by 2000
is 24 percent greater in scenario MM than in LL; the number of house-
holds has increased by 96 percent over the projection period in HM. In
scenario HH, the number of households is 32 percent greater than in MM;
C-43
·' .... ,,..,r:·· .I. ~:.· () ,,;
.-.... , '· .•·,
.-:.: \,! '.) ~ . ..J
·::· .. ···.·'j.j
TABLE C.l2. HOUSEHOLDS, 1980-2000
(thousands of households)
/._Lb. Cl...
SCENARIO NAJ.'1ES:
. _:i. 3 3 :·.; .. ,~--:; 1
:1. ~:_:; ~-:) ~ 3 ~:; '?
:L <~i/> t 9:1. ;3
1.~3(, v ~:)'7B
2l·J" /90
-v -.,. -! L :) . .:>. 043 j
164 ,, ~?9.:i
:L87 .300
:l)'~?v:.347
I I''',-., _.,, ~ .. ·• J ..• J::. ;;) • l J 1'1
l .;~) 8 ~· :=5 ::j 4
1'7'~~ ~ ?:~:L
::::3-4 + 464
LES.GL -Lmv Economic/Lmv Government
LES .GM -Low Economic/Moderate Government
LES .GH -Lmv Economic/High Government
MES.GL -Moderate Economic/Low Government
MES.GM -Moderate Economic/Moderate Government
MES.GH -Moderate Economic/High Government
HES.GL High Economic/Low Government
HES.GM -High Economic/Moderate Government
HES.GH -High Economic/High Government
L 1·-,... . ... , I -::. \:) '\ lJ , ..
ol 0"1 .-, •'', ''• , .. ,
.l / )' • v ~ ........ :
~:2 <> 8 v E~ ::S 0
::2 ~=5 ;,:.: ~ 1 0 ()
,,, ... , .. , ... ,,
J"l t::. ~:> .,. \.:t 1...
·f .t. ·-;.r .. ~. C)".:: .L ~.J / 4 '-·' , ....J
:f.9 ~=j + 23'7
2 :; :~:; y :_:) ·4 /}
2/~} ·) ~:)07
.hEE;. GL
'133.043
15<:;.~;,.~8
176. ?:1.2
20 3 . 8 9 3
2~):l ~ 897
HES . Gt'l
:L33 . 0 43
:L75.0 79
.. , ' .. ., ·-'"1 0 .-_l\,'t i ,J t
26~2. 42 1
Note: Values in 1980 adjusted to be the same in all cases; adjusts
for minor differences in exogenous series.
.the number of households increased by 158 percent over the projection
period.
In all three scenarios, over 80 percent of the expansion of house-
holds results from the increase in the population. In scenario LL,
82 percent of the household expansion results from population growth,
85 percent in scenario ~ill, and 84 percent in scenario HR. These dif-
ferences reflect the different household age structures which result
from rapid growth. The average number of people per household drops
from 3.2 in 1980 to 2.7 in LL, 2 • .7 in MM, and 2.6 in HR. This approxi-
mate 20 percent drop in the average people per household is consistent
with the projected decline in the national level of number of persons
per household (Bureau of the Census, 1979).
REGIONAL PROJECTIONS
Anchorage Region
This section describes the projection of employment, population,
and households for the Anchorage ·region. These projections are for the
period 1980 to 2010; growth beyond 2000 is assumed to follow the state
patterns for each of the major variables. The Anchorage region includes
the Anchorage, Matanuska~Susitna, Kenai, and Seward Census Divisions.
Three state scenarios were chosen for the regional economic and end-use
projections. These scenarios are the high economic-moderate government,
moderate economic-mod erate government, a nd low economic-mod erate govern-
ment scenarios ; these scenarios were chosen since they reflect the most
likely range of future growth. Table C.l3 shows the grmvth in Anchorage ..
C-45
TABLE C.l3. ANCHORAGE ECONOMIC GROWTH, 1980-2000
Low Scenario 1 Moderate Scenario 2 High Scenario3
EmEloyment PoEulation Households 4 EmEloyment Poeulation Households 4 EmEloyment Poeulation Households 4
19805 102,529 219,303 68,224 102,529 219,303 68,224 102,529 219,303 68,224
1985 111,118 248,850 85,177 119,352 260,034 85,805 132,186 275,848 89,515
1990 116,939 265,539 94,528 128,267 282,766 97,827 148,498 314,247 108,048
1995 134,425 293,381 108,377 151,735 322,582 116,718 185,601 375,483 136,364
('") 2000 157,268 329,865 127,099 173,021 361,239 137,172 211,011 427,146 163,560
I .p.
0\ 2005 ·165,290 346,691 133,582 191,029 398,837 151,449 248,203 502,433 192,388
2010 173,722 364,376 140,396 210,912 440,348 167,212 291,950 590,989 226,278
1 Growth beyond 2000 at 1 percent per year.
2 Growth beyond 2000 at 2 percent per year.
3 Growth beyond 2000 at 3.3 percent per year.
4Households exclude 3,212 on-base housing not included in energy projections.
5 1980 has been adjusted to be consistent ·among scenarios.
The Anchorage region is of central importance to the Alaska economy.
Because it contains Anchorage--the state's administrative, distribution,
and finance center--much of the growth in the -state will be reflected
in this region. In the past, many of the events which have influenced
state growth have occurred in the region. Projected future growth will
continue to follow these patterns; however, the projected future contains
relatively more activity occurring out of this region than in the past.
The low scenario reflects limited growth in the state and Anchorage
region. Anchorage is assumed to grow at an annual average rate of
1.8 percent per year over the projection period (2.2 percent per year
between 1980 and 2000). This is approximately the rate of growth in the
state economy and reflects the fact that the growth of basic sector
activity which is assumed promotes the existing distribution of activity.
Population growth follows the pattern of employment. Population grows
slightly less rapidly than employment; population gro,vs at an annual
average rate of 1.7 percent per year between 1980 and 2010 (2.1 percent
per year between 1980 and 2000). Finally, household growth is determined
by the growth in population and the changing pattern of household com-
position assumed at the state level. The number of households in the
Anchorage region is projected to increase by 106 percent over the pro-
jection period; as at the state level, over 83 percent of this growth
results from population growth.
The moderate scenario illustrates the effect of the increased basic
sector activity outside of the Anchorage region; Anchorage growth, as
C-47
measured by employment and population, is slightly slower than the state
growth. Employment in this scenario grO\vS at an annual average rate of
2.43 percent (2.7 percent for the 1980-2000 period). Population grows
at an annual average rate of 2.35 percent (2.5 percent for the 1980-2000
period). As at the state level, population grmvs less rapidly than
employment as a result of increased labor force participation. The
number of households in this scenario is 19 percent more than in the low '
scenario. Households increase by 145 percent between 1980 and 2010;
84 percent of this growth results from the increase in population.
The growth of basic sector activity outside of Anchorage has a
more profound effect on the growth of Anchorage relative to state growth.
Employment in the Anchorage region grows at an annual average rate of
3.6 percent between 1980 and 2010 (3.7 percent between 1980 and 2000),
which is .1 of a percent slower than state growth of 3.7 percent.
Population in this scenario increases to 377,000 by 2010 and averages
a 3.4 percent rate of growth over the projection period (3.4 percent
between 1980 and 2000). As in the other scenarios, the change in the
number of households is a result of changes in the population and in
household size. Households increase by 221 percent in this scenario;
84 percent of this growth is a result of population growth.
Fairbanks Region
Table C.l4 presents the projections for the Fairbanks region for
the low economic-moderate government, moderate economic-moderate govern-
ment, and high economic-moderate government scenarios. The Fairbanks
C-48
n
I .s::-
\0
TABLE C.14. FAIRBANKS ECONOMIC GROWTH, 1980-2000
. 1
Low Scenario Moderate Scenario 2 High Scenario 3
EmJ2lO:¥:ment PoEulation Households 4 Em2loyment PoEul a tion Households 4 EmJ2lO:¥:ment PoEulation
19805 29,641 59,268 17,114 29,641 59,268 17 '114 29,641 59,268
1985 36,508 70,276 21,152 38 '813 73,072 22,ll8 43,223 78,354
1990 37,270 74,187 23,530 40,485 78 '911 25,330 47,638 88,555
1995 41,729 81,966 27,433 46,840 89,398 30,414 57,492 104,871
2000 48,326 92,159 32,712 53,068 100,111 35,843 65,852 ll8,836
2005 50,791 96,861 34,381 58,591 110,531 39,574 77,459 139,782
2010 53,382 101,802 36,134 64,690 122,035 43,692 91,111 164,41 9
1 beyond 2000 1 Growth at percent per year.
2 beyond 2000 2 Grm..rth at percent per year.
3 Growth beyond 2000 at 3.3 percent per year.
4 Households exclude 3,062 on-base households not included in energy projections. Energy projections
assume only 91 percent of households are served by electricity in 1980 (based on 1978 end-use inventory).
This rate grows to 95 percent by 2010.
5 1980 has been adjusted to be consistent among scenarios.
Hous eholds 4
17 '114
24,121
28 '711
36,287
43 '716
51,422
60,836
1 '
region contains the Fairbanks and Southeast Fairbanks Census Divisions.
The projection period is between 1980 and 2010; employment, population,
and households are assumed to grow at state rates after 2000.
Fairbanks is a regional center for the Interior and Arctic regions
of-Alaska. Its past growth has been connected with resource development
in the region; most recently, Fairbanks has acted as a center for develop-
ment of Prudhoe Bay and the trans-Alaska pipeline. Since it is a regional
center, Fairbanks' future growth will be affected by growth of state
government as well as resource development in the region.
The growth in the Fairbanks region is only slightly faster than for
the state; both major resource development and growth of state government
affect the growth of the region. In the low scenario, employment grows
at a rate of 1.9 percent per year between 1980 and 2010 (2.5 percent
between 1980 and 2000). Population in this scenario reaches almost
102,000 by 2010; the growth is at an annual average rate of 1.8 percent
per year between 1980 and 2010 (2.2 percent between 1980 and 2000). As
at the state ievel, the increased labor force participation accounts
for a slower rate of population growth. The number of households almost
doubles, growing by 94 percent over the period. Eighty-eight percent
of this growth results form population growth.
In the moderate scenario, the Fairbanks region grows at approximately
the same rate as the state; resource development is spread more evenly in
this scenario, with fisheries and OCS development occurring out of the
C-50
Fairbanks region. Employment grows at an average annual rate of 2.6 per-
cent per year between 1980 and 2010 (3.0 percent between 1980 and 2000).
Population in 2010 is 20 percent greater than in the low scenario;
growth during the projection period is faster, averaging 2.4 percent per
year (2.7 percent between 1980 and 2000). The number of households in
the Fairbanks region increases by 132 percent in the moderate scenario;
88 percent of this change is a result of population growth.
The high scenario has major developments--petrochemicals and agri-
culture--occurring in the region. Because of this, growth (particularly
in the 1980-2000 period) is faster than for the state. Employment in
this scenario grows at an annual rate of 3.8 percent (4.1 percent for
the 1980-2000 period). Population follows the typical pattern, growing
slightly less rapidly than employment. The growth rate of population
averages 3.5 percent per year between 1980 and 2010 (3.5 percent between
1980 and 2000). Households follow the same pattern; the number of house-
holds more than doubles, with the majority of the growth resulting from
population growth.
Valdez Region
The Valdez Region consists of the Valdez-Chitina-Whittier Census
Division. This region has experienced major growth recently as a result
of the construction of the trans-Alaska pipeline and tanker port in Valdez.
Future growth of this region may result from expansion of industrial
activity due to the location of the pipeline terminus. Table C.l5
shows the projected growth in Valdez.
c-sl
("")
I
Vl
N
TABLE C.l5. VALDEZ ECONOMIC GROWTH, 1980-2000
Low Scenario 1 Moderate Scenario2 High Scenario 3
Em£loyment Po£ulation Households 4 Em£loyment Po£ulation Households 4 Em£loyment Po£ulation Households
19805 2,146 5,821 1,878 2,146 5,821 1,878 2,146 5,821 1,878
1985 2,967 6,739 2,255 3,782 8,063 2,698 7,464 9,660 3,182
1990 ·3,328 7,163 2,491 4,241 8,768 3,059 7,323 11,080 3,830
1995 3,532 7,914 2,853 4, 713 10,003 3,628 7,358 12,467 4,522
2000 4,033 8,898 3,354 5,237 11,201 4,197 7,717 13' 296 5,060
2005 4,239 9,352 3,525 5,782 12,367 4,634 9,077 15,640 5,952
2010 4,455 9,82~ 3,705 6,384 13,654 5,116 10,677 18,396 7,001
1 beyond 2000 1 Growth at percent per year.
2 beyond 2000 2 Growth at percent per year.
3 Growth beyond 2000 at 3.3 percent per year.
4 Energy projections assume only 71 percent of households are served by electricity in 1980 (based on
1978 end-use inventory). This rate is assumed to grow to 75 percent by 2010 .
. 51980 has been adjusted to be consistent among scenarios.
6Because of the rapid growth assumed in the Valdez economy in this scenario (between 1980 and 1985,
employment more than triples), we assume that not all of the new employees bring families but that they live
in an enclave-type area and commute to a shift-work situation from other regions. We assume that in 1985,
this amounts to close to 40 percent of total employment, but this drops to 20 percent by the end of the period.
4
Valdez is projected to grow rapidly in all scenarios. The rapid
rate of growth results from the location of major projects in Valdez and
the small population and employment base at the beginning of the period.
In the lmv scenario, employment is projected to increase by 2,310 by
2010, at an annual rate of growth of 2.5 percent per year (3.2 percent
per year between 1980 and 2000). Population increases at a slower rate
of 1.7 percent per year (2.1 percent per year between 1980 and 2000).
Households follow the pattern of population, increasing by 97 percent
over the projection period.
In the moderate scenario, the construction of a fuels refinery in
Valdez results in a greater divergence from the state growth. Employ-
ment increases at an average rate of 3.7 percent per year, tripling
during the period. Population increases at a slower rate of 2.9 percent
per year (2. 7 percent per year bet~veen 1980 and 2000) . Finally, house-
holds follow the pattern of population and increase by 172 percent over
the period.
In the high scenario, a major petrochemical facility is developed
in Valdez. This results in major growth in the region; employment
almost triples between 1980 and 1985. It is assumed that, because of
this major growth, not all employees bring families to Valdez but com-
mute, on some shift basis, from other regions. We assume that 40 per-
cent of the employees commute in 1985; this proportion is assumed to
decrease to 20 percent by 2000 and remain at this level for the rest of
the period.
C-53
Employment in the high scenario increases by almost 300 percent
between 1980 and 2010; this is an annual rate of 5.5 percent per year
(6.6 percent per year between 1980 and 2000). Population, because of
our assumption, increases much less rapidly, increasing at an annual
average rate of 3.9 percent over the period (4.2 percent between 1980
and 2000). Households follow the pattern of population, increasing by
273 percent over the period; 85 percent of this is due to poulation.
HOUSING STOCK PROJECTIONS
The growth in population and households determines the growth of
the housing stock in the three regions. Tables C.l6 through C.l8 illus-
trate the projected growth in housing stock in each region. The growth
in the housing stock parallels the growth in the number of households.
Housing stock does not grow as rapidly as the number of households
because each region begins the projection period with excess housing.
In Anchorage and Fairbanks, minimal change in the housing distribu-
tion is projected. In Anchorage, single-family units go from 52 percent
to approximately 51 percent of the housing stock in all scenarios. In
Fairbanks, the reduction in the proportion of single-family housing is
somewhat greater, falling from 52 percent to 49 percent in each scenario.
The other important distributional shift involves a shift in the type
of multifamily housing from duplex to other multifamily units.
c-54
Single Family
Multifamily
Mobile Home
Duplex
Total
Single Family
Multifamily
~tobile Home
Duplex
Total
Single Family
Multifamily
Mobile Home
Duplex
Total
TABLE C.16. ANCHORAGE HOUSI NG STOCK 1
1980-2010
Low Scenario
1980 1990 2000
37,422 50,130 65,506
19,061 25,409 36,430
9,239 11,725 16,032
5,871 6,226 8,958
71,593 93,490 126,927
Moderate Scenario
1980 1990 2000
37,422 53,309 71,837
19,061 27,530 40,772
9,239 12,561 17,890
5,871 6, 770 10,060
71,593 100 ,"170 140,559
High Scenario
1980 1990 2000
37,422 57,894 85,160
19,061 31,090 49,132
9,239 13' 982 21,331
5,871 7,101 11,996
71,593 . 110,667 167,619
2010
72,346
40,239
17,666
9,955
140,206
2010
87,555
49,689
21,760
12,337
171,341
2010
117,802
67,945
29,450
16,696
231,893
1Housing served, only off-base housing. The distribution in 2000
is assumed to remain constant after 2000.
C-55
Single Family
Multifamily
Mobile Home
Duplex
Total
Single Family
Multifamily
Mobile Home
Duplex
Total
Single Family
Multifamily
Mobile Home
Duplex
Total
TABLE C.l7. FAIRBANKS HOUSING STOCKl
1980-2010
Low Scenario
1980 1990 2000
9,009 11,462 15,446
4,792 6,550 10,005
2,252 2,964 4,178
1,272 1,278 1, 772
17,325 22,254 31,401
Moderate Scenario
1980 1990 2000
9,009 12,244 17,026
4,792 7,245 10,162
2,252 3,192 4,545
1,272 1,279 1,906
17,325 23,960 34,439
High Scenario
1980 1990 2000
9,009 13,529 20,436
4,792 8,519 13,586
2,252 3,617 5,543
1,272 1,505 2,410
17,325 27,170 41,975
2010
17,321 .
11,230
4,682
1,971
35,204
2010
21,048
13,592
5,624
2,343
42,607
2010
28,873
19,209
7,826
3,379
59,287
1Housing served, an increasing proportion of offbase households.
The distribution in 2000 is assumed to remain constant after 2000.
C-56
Single Family
. Multifamily
Mobile Home
Duplex
Total
Single Family
Multifamily
Mobile Home
Duplex
Total
Single Family
Multifamily
Mobile Home
Duplex
Total
TABLE C.l8. VALDEZ HOUSING STOCK1
1980-2010
Low Scenario
1980 1990 2000
472 706 1,184
189 306 543
642 629 606
192 197 189
1,495 1,838 2,522
Moderate Scenario
1980 1990 2000
472 951 1,572
189 412 725
642 684 649
192 220 212
1,495 2,267 3,158
High Scenario
1980 1990 2000
472 1,231 1,902
189 533 892
642 792 763
192 259 250
1,495 2,815 3,807
2010
1,334
610
681
213
2,838
2010
1,950
902
808
263
3,923
2010
2,684
1,256
1,074
354
5,368
1Housing served, an increasing proportion of total households.
The distribution of housing stock is assumed to change. in a straight-
line manner from the 1978 distribution to that projected for 2000.
Distribution in 2000 is assumed to remain constant beyond 2000.
C-57
In Valdez, the change in housing stock distribution is somewhat
more pronounced. It was assumed that housing preferences, which were
projected to be much different than the beginning 1978 housing stock,
would only slowly change the distribution as removal and growth of the
population increased the demand for housing of different types. Because
of this assumption, the proportion of housing which is single-family
increases from 32 percent in 1980 to 47 percent in 2010. In all sce-
narios, the proportion of housing stock which is mobile homes decreases;
this reflects a stabilizing of the population over time.
C-58
APPENDIX D. COMPONENTS OF THE END USE HODEL
D.l. HOUSEHOLDS AND HOUSING STOCK
The basic consuming unit for residential electricity consumption is
the off-military-base household. Huch of the data available to analyze
energy consumption, hmvever, is more closely associated with the number
of housing units, which will generally be larger than the number of
households for a number of reasons.
Tables D.l and D.2 present information on housing units for the
years 1960 and 1970 and indicate the different types of housing. Using
the 1970 census definition as a guide, all housing units can be divided
into year-round and seasonal units. The latter are not designed for
year-round habitation. Of the year-round units, only a portion are
occupied; and of those not occupied, only a portion are vacant in the
sense of being for sale or rent. Second homes are not identified but
are a component of both the seasonal category and the year-round category.
For energy consumption purposes, there are three important housing
stock measures:
1. Occupied housing units. Each household will occupy a
housing unit, and this forms the basis for estimating the
appliance electricity demand in the residential sector.
t:l
I
N
TABLE D.l. HOUSING UNIT ANALYSIS: 1960
Vacant
Year-Round
Year-Round Occupied Housing Units Population
All Housing Housing Housing Not for Sale Per Occupied
Census Division Units Units a Units or Rentb Housing Unit
GREATER ANCHORAGE AREA
Anchorage 23,972 23,564 21,853 788 3.4
Matanuska-Susitna 2,593 2,346 1,501 775 3.4
Kenai-Cook Inlet 2,504 2,339 1,686 551 3.4
Seward 1,494 1,294 966 146 3.0
GREATER FAIRBANKS AREA
Fairbanks 12,598 11,928 11,056 550 3.3
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier 1,241 1,049 785 153 3.1
WESTERN REGION U.S. 3.2
aConstructed variable equal to occupied housing units plus vacant year-round housing units.
bConstructed variable equal to year-round housing units minus occupied housing units minus
vacant year-round housing units for sale or rent. This category thus includes 1) rented and sold
awaiting occupancy, 2) held for occasional use, 3) held for other reasons, and 4) dilapidated.
Notes: Second homes may be classified as either seasonal or year-round housing units.
Southeast Fairbanks \<las not a separate census division in 1'960.
SOURCES~ l960 Census of Rousing,.General Rousing Characteristics: Alaska. Tables 28, 29.
Median
Rooms
3.9
3.4
3.4
3.2
3.6
3.1
TABLE D.2. HOUSING UNIT ANALYSIS /: 1970
Vacant
Year-Round Occupied
Year-Round Occupied Housing Units Population Units
All Housing Housing Housing Not for Sale Per Occupied Median Which Own
Census Division Units Units Units or Rent Housing Unit Rooms Second Home
GREATER ANCHORAGE AREA
Anchorage 37,650 37,617 34,988 975 3.4 4.5 3,492
Matanuska-Susitna 4,214 3,355 1,826 1,073 3.4 4.0 177
Kenai-Cook Inlet 4,877 4,650 3,889 470 3.5 4.0 442
Seward 1,106 956 605 239 3.1 3.9 157
t:1
I GREATER FAIRBANKS AREA w
Fairbanks/
S.E. Fairbanks 13,895 13,729 12,644 596 3.4 4.3 986
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier 1,447 1,405 947 411 3.2 2.9 257
WESTERN REGION U.S. 12,031,802 11,938,658 11,171,550 302,970 3.0 4.7
SOURCES: 1970 Census of Housing, Detailed Housing Characteristics: Alaska. Tables 60, 63, 66.
1970 Census of Housing, Detailed Housing Characteristics: United States Summary. Tables 1, 3.
2. Occupied plus vacant (but available for rent or purchase)
housing units. Housing units which are occupied plus vacant
units form the basis for space heating requirements because
houses which are vacant, but available, must be heated in
winter (although to a lower temperature) to prevent damage.
3. Second homes. Vacation homes and homes used seasonally
will have different energy use characteristics than first
homes.
Before estimating the number of households and the housing stock
for 1978, those individuals housed in group quarters must be identified
and subtracted from total population since their consumption of elec-
tricity is not reflected in the utility residential load . Table D.3
shows that in 1970 the population in group quarters was large in both
Anchorage and Fairbanks. Although a large proportion of this population
is military and the military population has declined since 1970, we
assume the same number of individuals in group quarters in 1978 as in
1970. That is, the decline has affected military personnel not in group
quarters.
Table D.4 presents two estimates of railbelt households in 1978 and
compares them with year-end electric utility residential customers. At
least four factors contribute to the discrepancies between the household
and utility customer figures:
D-4
TABLE D.3. 1970 POPULATION LIVING IN GROUP QUARTERS
GREATER ANCHORAGE AREA
Anchorage
Matanuska-Susitna
Kenai-Cook Inlet
Seward
GREATER FAIRBAl~S AREA
Fairbank s
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier
Population
124,542
6,552
14,250
2,021
50,262
3,116
Population in Population in
Housing Units Group Quarters
118,809 5,733
6,208 344
13,719 531
1,870 151
42,682 7,580
3,023 93
Note: Group quarters are primarily institutions, boarding houses,
military barracks, college dormitories, hospitals, religious
centers, and ships.
SOURCE: 1970 Census of Housing, Detailed Housing Characteristics:
Alaska. Table 60.
D-5
TABLE D.4. 1978 HOUSEHOLD ESTIMATES
(thousand)
PoEulation (1 2 000) Electric d
Population Utility Ratio of a c Department b in Housing Population Residential Households
of Department Units per Occupied 1978 and Rural to
Commerce of Labor (1' 000) Housing Unit Households Customers Customers
Greater
Anchorage Area 215.7 226.3 209.0 219.6 61.6 64.7 77,000 .80 .84
Anchorage 179.0 185.5 173.3 179.8 3.4 51.0 52.9 57,916 .88 .91
Kenai-Cook Inlet 19.6 22.3 19.1 21.8 3.4 5.6 6.4 7,904 .71 .81
Matanuska-Susitna 14.2 15.4 13.9 15.1 3.4 4.1 4.4 10,152 .40 .43
Seward 2.9 3.1 2.7 2.9 3.0 .9 1.0 1,027 .90 1.00
Greater
t:1 Fairbanks Area 59.4 60.8 51.8 53.2 3.3 15.7 16.1 17,524 .90 .92 I
(j\
Fairbanks 54.1 55.5
Southeast Fairbanks 5.3 5.3
Glennallen-Valdez Area
Valdez-Chitina-
Whittier 5.9 5.0 5.8 4.9 3.1 1.9 1.6 1,539 1·.27 1.07
SOURCES: (a) State of Alaska Department of Commerce and Economic Development, Division of Economic
Enterprise, "Numbers:· Basic Economic Statistics of Alaska Census Division," 1979.
(b) State of Alaska, Department of Labor.
(c) 1970 Census of Housing, Detailed Housing Characteristics: Alaska.
(d) Federal Energy Regulatory Commission, Utility Power Systems Statements.
1. The household calculations and population estimates
may be incorrect.
2. Some residential electricity hookups are for second and
vacation homes.
3. Some residential electricity hookups are for units which
are vacant since a minimal amount of electricity is
necessary even in an unoccupied housing unit for such
things as the heat distribution pump.
4. Some households may not have access to the electric utility.
Table D.5 (compared to Table D.4) shows that in all areas of the
railbelt the ratio of residential customers-to-households has apparently
increased since 1970 unless the population and household size estimates
for 1978 are very low.
TABLE D.5. 1970 UTILITY HOOKUP RATES
Greater Anchorage Area
Greater Fairbanks Area
Glennallen-Valdez
A
1970 Occupied
Housing Units
41,233
12,612
1,017
D-7
B
1970 Residential
& Rural Utility
Customers
41,151
10,756
561
B/A
1.00
.85
.55
Part of the differences among the census divisions in the ratios of
households to utility customers arises from the fact that the utility·
boundaries do not correspond to the census division boundaries used -in
developing Table D.4. Specifically, Chugach Electric serves portions of
Anchorage, Kenai -Cook Inlet, Seward, and Valdez-Chi tina-lfui ttier.
Matanuska Electric serves a portion of Anchorage. A portion of Golden
Valley Electric Association customers reside in the Yukon-Koyukuk Census
Division.
The only identifiable household concentrations not having access to
utility service from the seven major railbelt utilities appear to be
portions of Valdez-Chitina-lfuittier (Chistochina, Mentasta Lake, Tatitlek ),
portions of Kenai-Cook Inlet (Tyonek, Seldovia), and portions of South-
east Fairbanks (Tok).
Vacancy rates for housing are not collected in a complete and
accurate nianner. ·The rental housing vacancy rate for Anchorage in 197 8
was estimated at 14 percent.1 As of June 1979, the unsold inventory o f
new houses was approximately 2 percent of the stock.2 It is not possib le
from this information to develop an overall vacancy rate, but realtors
generally agreed that the rate was higher in 1978 than in a normal
market. A recent housing study by the Fairbanks North Star Borough
estimated a vacancy rate of 14 percent overa11.3 Vacancy rates for
other areas are unknown but assumed to be low·er than for Anchorage and
Fairbanks.·
D-8
The number of secon? homes within the railbelt served by electric
u tilities is not known but is assumed to be concentrated in the Matanuska-
Sus itna Census Division. In the 1970 census, 3,492 households in the
. 4 Anc horage Census Division -indicated they owned second homes. Some are
l o cated in the Matanuska-Susitna Census Division, and some of these would
have appeared in the census as year-round housing units although there
is no accurate information on actual numbers.
We assume 2,000 second homes among the Greater Anchorage Area utility
5 c u stomers. Using this figure and the information from Table D.4 results
in an estimate of the overall Greater Anchorage Area vacancy rate of
be tween 14 and 18 percent, which would be about twice the normal rate.
This may be somewhat high based upon the rate reported for Fairbanks.
We have no vacancy information for Glennallen-Valdez. If we assume
25 percent of the population is not serviced by Copper Valley Electric
Association, the vacancy rates derived using Table D.4 data range from
zero to 20 percent, which probably brackets the actual value.
The estimate of the number of households could be obtained either
by estimating population and dividing by average household size or by
using utility hookup numbers and adjusting for vacancies. Neither method
is foolproof; but the latter involves only one estimate, while the former
requires two. .We choose the latter and assume a 13 percent vacancy rate .
The resulting household estimates are shown in Table D.6.
D-9
TABLE D.6. HOUSING UNIT AND HOUSEHOLD ESTIMATES
First Vacancy Rate
Housing Units (percent) Households
Greater Anchorage 75,000 13 65,250
Greater Fairpa?ks 17,524 13 15,245
Glennallen-Valdez 1,539 13 1,339
The housing stock may alternatively be calculated directly by a
count or estimate, independent of the number of electric hookups. This
method provides a check on the utility hookup method of housing stock
estimation as well as providing information on the geographic distribu-
tion of the stock (within the Greater Anchorage Area) and an estimate of
the distribution of the stock by type. Table D.7 tabulates the housing
stock analyses which have been done for the railbelt communities.
From these analyses, it is relatively easy to construct an estimate
of the 1978 housing stock for the Anchorage Census Division of 57,896,
which is included in the table. Estimates for other Census Divisions
must be developed more indirectly. Table D.8 shows the result of apply-
ing the 1978-to-1970 population ratios to the 1970 year-round housing
unit stock in each census division. These housing stock estimates can
be adjusted to arrive at final estimates.
D-10
TABLE D.7. HISTORICAL RAILBELT HOUSING STOCK
DISTRIBUTION BY TYPE
Single Hulti-Hobile
Family Duplex Family Home
GREATER _ANCHORAGE AREA
Anchorage Census Division
1950: 3,325 964 1,128 202
1960 13,435 1,427 7,625 1,485
1970~ 15,572 3,813 13' 368 4,864
1978 (off base) 28,530 4,581 18,196 6,589
Anchorage Bowl
1975e (off base) 23,227 5,324 14,754 6,246
1975e (on base) 34 0 4,122 0
1975e (total) 23,261 5,324 18,876 6,246
1979f 26,300 24,203 6, 960 -
Eagle River 1979f
Gird~·.TOod 1978g
Kenai-Cook Inlet
Census Division
1960b 2,117 19 182 186
1970c 2,627 108 594 1,321
Seldovia 1970~ 102 29 22
1976] 153 20 41
Soldotna 1970~ 159 95 143
1976] 311 110 180
Homer 1970~ 310 39 18
1976] 251 31 134
Kenai 1970~ 574 371 231
1976] 684 350 274
D-11
Other Total
0 5,619
0 23 '972
0 37,617
57,896
0 49,551
0 4,156
0 53,707
0 57,463
0 3,524
198
0 2,504
0 4,650
0 153
15 229
0 397
0 601
0 367
16 432
0 1,176
0 1,308
Table D.7. (continued)
Single Multi-Mobile
Family Duplex Family Home Other Total
Seward Census Division
1960b 1,307 86 77 24 o · 1,494
1970c 789 19 138 10 0 956
Seward 1976j 497 212 36 36 78 1
Matanuska-Susitna
Census Division
1960b 2,336 20 149 88 0 2,59 3
1970~ 2,947 41 159 208 0 3,35 5
1978 310 7,616
GREATER FAIRBANKS AREA
Fairbanks-Southeast Fairbanks
Census Division
1950: (urban) 1,295 166 352 2 0 1,815
1960 6,527 671 4,547 853 0 12,598
1970c . 5,335 1,068 6,072 1,254 0 13,729
Fairbanks Census Division
1970c 4, 775 1,017 5,603 1,129 0 12 ,5 24
Southeast Fairbanks
Census Division
1970c 560 51 469 125 0 1 ,205
Fairbanks North Star Borough
(off base)
19701 9 ,884
1975m 11 ,324
1976n 15,200
1976° 1 6 ,89 4
1978° 8,787 1,232 5,616 2,306 0 1 7 ,94 1
1979° 17 '684
Fairbanks North Star Borough
(off base)
1976p 49% 8% 25% 17 % 1%
1978q 52% 30% 12% 6%
D-12
Table D.7. (continued)
Single Multi-Mobile
Family Duplex Family Home Other Total
· GLENNALLEN-VA LDEZ AREA
Valdez-Chitina-Whittier
Census Division
1960b 803 31 392 15 0 1,241
1970c 881 34 278 212 0 1,405
Valdez 1970h 105 95 98 0 298
1978r 222 143 135 518 .12 1,030
1978s 314 171 521 16 1,022
Glennallen 1970 h 58 25 26 0 109
SOURCES: (a) U.S. Department of Commerce Census of Housing 1950: Alaska,
General Characteristics, Table 14. These are all dwelling
units.
(b) U.S. Department of Commerce Census of Housing 1960: Alaska,
Table 28. These are all housing units.
(c) U.S. Department of Commerce Census of Housing 1970: Alaska,
Table 62. These are all year-round housing units.
(d) Estimated by author by nett i ng out 1978 housing units
authorized for Anchorage Municipality from 1979 total
and adding Eagle River and Girdwood (latter assumed
all single-family units).
(e) Anchorage Urban Observatory, University of Alaska, 1975
Housing Survey, Appendix I, p. 2.
(f) Municipality of Anchorage, Planning Department.
(g) Municipality of Anchorage, Planning Department. These are
full-time residences only. Total residences weLe calculated
at 729.
(h) State of Alaska Department of Community and Regional Affairs,
Division of Community Planning, Selected 1970 Census Data
for Alaska Communities, 1974.
(j) Kenai Peninsula Borough, Profile of 5 Kenai Peninsula Towns,
1977, Table 130. These are year-round dwelling units (vacant
and occupied units designed for year-round living). This
includes housing within the city limits of these tmms only
and estimates 250 units outside Homer.
D-13
Table D.7. (continued)
(k) Matanuska-Susitna Borough Planning Department.
(1) 1970 Census of Housing as reported in Fairbanks North
Star Borough, FMATS Housing Study, draft, 1980.
(m) E. Allen Robinson. "Situation Report: Fairbanks, Alaska,"
HUD Anchorage 1975, as reported in Fairbanks North Star
Borough, F}~TS Housing Study, draft, 1980.
(n) Jack Kruse, "Fairbanks Community Survey/' .Institute of
Social and Economic Research, University of Alaska,
Fairbanks, 1976, as reported in Fairbanks North Star
Borough, ~TS Housing Study, draft, 1980.
(o) William Rose, Fairbanks North Star Borough Planning
Department, as , reported in Fairbanks North Star Borough,
FMATS Housing Study, draft, 1980.
(p) Jack Kruse, "Research Notes: Fairbanks Community Survey,"
Institute of Social and Economic Research, 1976, p. 2.1.
(q) Jack Kruse, "Fairbanks Petrochemical Study," Institute of
Social and Economic Res~arch, University of Alaska, Fair-
banks, 1978, as reported in Fairbanks North Star Borough,
FMATS Housing Study, draft, 1980.
(r) Michael Baring-Gould et al. Valdez City Census, 1978.
University of Alaska, Anchorage, Table 13.
(s) Northrim Associates, Inc., CCC/HOK from the Environmental
Impact Statement, Alaska Petrochemical Company Refining and
Petrochemical Facility, Valdez, Appendix , Vol. II, p. 93 .
This is total housing net of hotel-motel units and campers.
D-14
TABLE D.8. 1978 YEAR-ROUND HOUSING STOCK ESTI}~TE
BASED ON POPULATION RATIOS
1970
Year-Round 1978/1970a
Ce nsus Division Housing Units Population
Anchorage
Kenai-Cook Inlet 4, 650 1.56
Matanuska-Susitna 3,555 2.35
Seward 956 1.53
Fairbanks/Southeast Fairbanks 13,729 1.25
Valdez-Chitina-Whittier 1,405 1.61
aBased on Alaska Department of Labor estimates for 1978.
D-15
1978
Year-Round
Housing Unit
Estimate
7,277
8,356
1,463
17,207
2,268
In Seward, 40 percent of year-round housing units were unoccupied
in 1970 and 35 percent in 1960.6 If we assume that 25 percent were
unoccupi ed and did not require space heating in 1978, this results in an
estimate of about 1,100 "first" year-round housing units (year-round
housing units, as defined by the census, which are actually utilized on
a year-round basis).
Kenai-Cook Inlet had a 16 percent vacancy rate in 1970, dm.;rn from
33 percent in 1960. The growth in the number of units in the major com-
munities of the census division bet~.;reen 1970 and 1978 was much slower
7 than would be indicated by the 1978 estimate of Table D.8. Thus,
growth was more rapid outside these cities and may or may not have been
accounted for by second homes. We assume 500 second homes and thus
arrive at an estimate of the first home housing stock of 6,777.
In the Matanuska-Susitna Borough, 44 percent of year-round housing
units were unoccupied at the time of the 1970 census and 42 percent in
1960. If we assume that this rate has fallen since 1970 (Matanuska
Electric Association had 678 seasonal rate customers in 1978, but this
is not equivalent to second homes) to about 35 percent, it would be
consistent with a 15 percent vacancy rate and 1,500 second homes. The
Borough counted 7,616 dwelling units in 1978 but did not distinguish
vacation homes. Netting out second homes would produce a first housing
unit estimate of about 6,100 units.
D-16
These census division estimates can be aggregated to arrive at an
overall first h ousin g unit e stimate for the Greater Anchorage Area of
7 1,873, sho\vn in Table D.9. This is somewhat lower than the estimate
d erived b y counting the number of electric utility accounts but is more
reasonable as a basis for calculating ·electricity consumption on-an end _
u se basis. (For e x ample, there are some residences in Anchorage with
t wo electric meters, each of which were counted as a customer during
1 978.)8
For the Greater Fairbanks area, the Fairbanks North Star Borough
housing surveys closely correspond to the utility hook up data; hmvever,
researchers admit that deficiencies exist in at least some of the surveys,
\vhich could lead to an over count. 9 These house counts, however, would
not include utility customer s located outside the Borough in the South-
east Fairbanks a nd Yukon-Koyukuk Census Divisions. We assume that these
effects, as well as the presence of some vacant, non-market housing in
the North Star Borough and second homes outside the Borough, cancel one
another out so that the utility hookup figure becomes our housing stock
estimate for the Greater Fairbanks Area.
In the Valdez-Chitina-Whittier Census Division, the vacancy rate
was 33 percent in 1970, up from 25 percent in 1960. A large portion of
this increase could be the decline in population of Whittier. Without
additional information on the housing stock in the utility service area
in the census division, we must use the electric utility residential
hookup estimate of about 1,500.
D-17
TABLE D.9. FIRST HOME HOUSI NG STOCK ESTIMATES
FOR 1978 USED IN END USE CALCULATIONS
Greater Anchorage Area
Anchorage
Kenai-Cook Inlet
Matanuska-Susitna
Seward
Greater Fairbanks Area
Fairbanks/Southeast Fairbanks
Glennallen-Valdez
Valdez-Chitina-Whittier
71,873
57,896
6, 777
6,100
1,100
17,500
1,500
The first housing unit housing stock is divided into four housing
types which have very different space heating characteristics--single-
family detached, duplex, multifamily, and mobile home. Information on
the distribution of the housing stock by type comes primarily from the
housing stock surveys shown in Table D.7.
For Matanuska-Susitna, the planning department has estimated multi-
family units and mobile homes. We allocate the remaining units between
single-family and duplex units on the basis of the Anchorage proportions.
In Kenai-Cook Inlet, the proportion of single-family units in the
larger communities was representative of the census division as a whole.
Thus, the 1976 proportion of 54 percent found in these larger communit i es
D-18
is used for the 1978 estimate. For mobile homes, this was not the case
as the proportion in the whole census division in 1970 was 28 percent,
while it was only 20 percent in the larger cou@unities. In these larger
communities, it grew to 24 percent by 1976, so we assume the same type
of growth for the census division as a \vhole but that some of the rela-
tive growth in the utilization of mobile homes is in areas inaccessible
to the railbelt utilities. Thirty percent becomes our estimate. The
most recent estimate of the distribution bet\veen duplex and multifamily
units is the 1970 census. From the total data, a pattern toward single-
family living is evident • so \ve assume that a majority of the growth
since 1970 is in duplexes and that multifamily units are 600.
For Seward, we assume the same distribution for the utility service
area as indicated in the 1976 survey and that multifamily and duplex
units are equal in number.
For Fairbanks single-family units, we utilize information collected
in the 1978 survey for the Borough and assume the same distributions for
housing units outside the Borough. For trailers, \ve assume a downward
trend in the percentage since 1976 and assume that the "other" category
from the 1978 survey is not relevant for · our purposes. Thus, the 13 per-
cent figure from the Borough count is taken. We further assume 21 percent
of duplex and multifamily units are duplexes, which is an average of the
various surveys.
D-19
For Glennallen-Valdez, the data indicates a much higher proportion
of mobile homes in Valdez than in Glennallen. lve use the 1978 Valdez
City Census for Valdez and apply the 1970 Glennallen distribution to the
remainder of the service area. The results of this analysis are shown
in Table D.lO.
D.2. RESIDENTIAL ELECTRIC SPACE HEATING
Data on the proportion of housing units heating with electricity
and average consumption levels for various housing types in different
locations is fragmentary. None of, the electric utilities compile this
information at present; and although some had special all-electric rates
in the past, utility records of those customers have not been retained.
The space heating distribution is currently relatively stable
except in the outlying areas of the Greater Anchorage Area and in Fair-
banks. In t4e fo~er, use of electricity fpr space heating. is growing
relative to the primary alternative (fuel oil) because of the rising
price of fuel oil and the relatively stable price of natural gas-genera
electricity. In Fairbanks, there is a shift away from electric space
heat toward fuel oil as the price of oil increases since incremental
electricity isproduced by fuel oil. These shifts make it more difficult.
to estimate the actual space heating mode split in these areas.
Census data on fuels used for space heating presented in Table D.ll
encompass the whole railbelt but are not current because .of the rapid
D-20
~ ..
·k· ...
.t\1 ~
'lri ~ ·l-·•
i>·' .. ..
., • .j;
:1!!~ . . o···
¥r :w.
1\'l'
!-,•
~
~':
t,._.,;
G!!i :;,
~--
t;-·
,.
3···· .
-./•
t;.:,
~t·
TABLE D.lO. 1978 FIRST HOME HOUSING STOCK
DISTRIBUTION BY HOUSING TYPE
Single Multi-
Family Duplex Family
GREATER ANCHORAGE AREA 37,357 . 5, 930 19,254
Anchorage 28,530 4,581 18,196
Kenai-Cook Inlet 3,660 484 600
Matanuska-Susitna 4,463 717 310
Seward 704 148 148
GREATER FAIRBANKS AREA
Fairbanks/Southeast
Fairbanks 9,100 1,285 4,840
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier 472 197 189
D-21
Mobile
Home Total
9,332 71,873
6,589 57,896
7,033 6, 777
610 6,100
100 1,100
2,275 17,500
642 1,500
t:J
I
N
N
TABLE D.ll. PERCENT DISTRIBUTION OF SPACE HEATING FUELS
IN THE RESIDENTIAL SECTOR
Census Division Utility Gas Oil Electric Coal
GREATER ANCHORAGE AREA
Anchorage
1950 0 92 0 7
1960 0 82 0 15
1970 53 34 6 1
Matanuska-Susitna
1960 0 47 0 24
1970 0 62 1 14
Kenai-Cook Inlet
1960 0 69 0 8·
1970 31 52 4 3
Seward
1960 0 88 0 0
1970 0 92 4 0
SOURCES: 1950 Census of Housing, General Housing Characteristics: Alaska.
1960 Census of Housing, General Housing Characteristics : Alaska.
1970 Census of Housing, Detailed Housing Ch a racteristics: Alaska .
Wood Propane
1 0
0 0
0 2
28 0
17 6
23 0
5 3
12 0
4 0
Table 17.
Tables 29, 30.
Table 63 .
Other
0
3
4
1
0
0
2
0
0
Table D.ll. (continued)
Census Division Utility Gas Oil Electric Coal Wood Propane Other
GREATER FAIRBANKS AREA
Fairbanks and
Southeast Fairbanks
1950 0 30 0 54 16 0 0
1960 0 47 0 49 3 1 0
1970 3 61 7 20 2 1 6
GLENNALLEN-VALDEZ AREA
t;:j
I
N Valdez-Chitina-Whittier (,.)
1960 0 57 0 0 17 0 26
1970 0 72 0 0 21 7 0
growth in the housing stock since 1970. Fuel oil is the predominant
fuel except in those areas of Anchorage and the Kenai-Cook Inlet Cens~s
Division where natural gas is now available. Coal was historically very
important in Fairbanks, but it was surpassed by fuel oil in the 1960s.
The census data does indicate a significant proportion of the occupied
housing stock utilizing coal, wood, propane, and other fuels, even in
1970.
Information on four electric utilities is available from a Federal
Power Commission (now Federal Energy Regulatory Commission) report.
This information is shown in Table D.l2. Utility personnel are unable
to determine the source of this information but feel it is reasonable.
This data shows that all electric customer growth in Anchorage in the
early 1970s was more rapid than total customer growth. No such trend is
apparent for Fairbanks.
Additional published data on the residential space heating mode
split in the railbelt is shown.in Table D.l3. This data tends to con-
firm information gathered informally in conversations with utility and
real estate personnel as well as analyses of utility monthly load curves.
[Matanuska Electric Association (MEA) and Homer Electric Association
(HEA) analyzed monthly.bills in an attempt to identify the number and
average consumption levels for their electric space heating customers.
The HEA analysis requires further work! The MEA analysis yielded an
estimate of 2,685 space heating customers in 1978 and 18,172 kWh annual
consumption for space heating per customer in 1979.]
D-24
t::l
I
N
Vt
Customers
Total All-Electric Non-All-Electric
Chugach Electric
1963
1970
1971
1972
1973
1974
13,170
24., 682
25,761
28,687
29,077
31,779
120
1,280
1,475
1,756
2,010
2,605
Anchor a ge Municipal Light and Power
1963
1970
1971
1972
1973
1974
6,592
8,477
9,295
10,130
10,523
11,268
NA
381
700
700
928
1,123
Golden Valley Electric Association
1963
1970
1971
1972
1973
1974
NA
6,624
6,741
6,947
7,382
8,643
NA
802
850
850
1,448
900
Fairb a nks Municipal Utility System
1963
1970
1971
1972
1973
1974
4,120
4,532
4,443
4,540
4,443
NA
0
NA
NA
19
19
NA
13,050
23,402
24,286
26,931
27,067
29,174
NA
8,096
8,595
9,430
9,595
10,145
NA
5,822
5,891
6,097
5,934
7,743
4,120
NA
NA
4,521
4,424
NA
Aver age Annual Consumption (kWh)
Total All-Electric Non-All-Electric
6,137
8,057
9,194
9,386
9,887
9,621
4,681
6,431
6,782
7,080
7,855
7,982
NA
10,133
12,158
13,920
14,479
14,795
3,013 .
5,167
5,504
5,341
5,841
NA
37 .. 882
38,500
38,700
39,000
39,050
·39,100
NA
18,045
18,127
18,127
17,985
17,355
NA
42,516
43,000
43,000
40,000
43,000
NA
NA
46,316
46,316
NA
5,!345
6,392
7,402
7,455
7, 721
6,989
NA
5,884
5,858
6,260
6,875
6,944
NA
5,672
7,708
9,866
8,251
16,015
3,013
·NA
NA
5,169
5,667
NA
He a ting Only
For All-El e ctric
Custome rs
29,600
29,100
29,600
29,750
29,900
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
SOURCE: Compiled from Federal Power Commission, All Electric Homes, annu a l.
TABLE D.13. RAILBELT RESIDENTIAL SPACE HEAT
MODE SPLIT INFORMATION
(percent)
gas oil electric wood other-unknown
Seward 1977 a 96 1 3
Kenai-Cook Inlet 1977
Seldovia a 88.2 3.9 3.9 4
Soldotna a
City 70.4 25 2.8 .9 .9
Total 57.7 35.8 4.4 1.5 .7
Kenai 76.2 13.2 9.3 .7 .7
Total of
Three Cities 66 25 7 1 1
Anchorage 1975
b Anchorage· Bowl
house 68 17 11.5 .5 3
house new
since 1970 77 .5. 6 16 0 0
trailer 48.5 36.5 7.5 1 6.5
trailer new
since 1970 33.5 55.5 0 11 0
apartment 52 8.5 31 0 8.5
Fairbanks 1979c 0 70 11 12 7
SOURCES: (a) Kenai Peninsula Borough, Profile of 5 Kenai Peninsula
Towns, 1977, Table 72;
(b) Anchorage Urban Observatory, Anchorage Housing Survey,
1975, unpublished data.
(c) Fairbanks North Star Borough, Community Information
Center, Community Information Quarterly, Spring 1980,
Vol. III., No. 1, p. 81.
D-26
Additional information for drawing inferences about the number of
electric space heating customers and their annual average consumption
rates can be obtained from several sources. These.include the following:
1. natural gas consumption data,
2. "typical" home space· heating analysis, and
3. average annual and monthly residential consumption per
customer information of electric utilities.
In addition, information on housing unit sizes, in terms of average
square feet, and the relative sizes and heating efficiencies of different-
sized units can be helpful.
Using data obtained from Alaska Gas and Service Company, an analysis
of natural gas consumption for space heating was done. This analysis
indicates that for. residential gas utility customers (assumed to be
single-family, mobile home, and duplex units and all space heated with
gas) average annual consumption for space heating (that portion of the
load which varies over the course of the year with heating degree days)
is about 84 percent of total consumption or 187 mcf/customer/year for
the 1970s and 175 mcf/customer for 1978. Details of this analysis are
shown in Table D.l4.
The analysis further reveals no trend in consumption per customer
in the 1970s, which might be the result of either larger homes or better
insulation. Personnel at Alaska Gas and Service Company using national
data estimate the space heat load at 75 percent of the total. Recog-
nizing the imprecision of both the regression model and the national
D-27
TABLE D.l4.
DETERMINATION OF AVERAGE NATURAL GAS CONSUMPTION
IN THE RESIDENTIAL SECTOR
A. Alaska Gas and Service Company Residential Consumption Data
Consumption Customers Average Consumption
(mcf) (mcf)
Eagle Soldotna/
River Kenai Anchorage Total
1970 2,615,042 12,097 216.2
1971 3,406,227 15,233 223.8
. 1972 3,817,869 16,f31 235.2
1973 4,162,662 17,983 231.5
1974 4,312,701 20,135 214.2
1975 5,402,111 22 '779 257 237 236 237.2
1976 5,765,871 25,659 225 225 225 224.7
1977 5,848,812 27,901 262 243 206 209.6
1978 6,367,015 30,629 199 196 209 207.9
1979 6,730,022 33,229 202.5
Average 1970,....1978 222.3
Normalized by Heating Degree Days 226
Heating
Degree Days
(annual)
10,137
11,879
12,016
11,.665
10,683
11,308
.10,361
9,394
9,131
-
10,730
10,911
SOURCE: Annual financial report to APUC and internal company records.
D-28
>
TABLE D.l4.· (continued)
DETERMINATION OF AVERAGE NATURAL GAS CONSUMPTION
IN THE RESIDENTIAL SECTOR
Honthly Natural Gas Sales Data
Sales/Customer (mcf) Heating Degree Days (Anchorage)
Year Residential Commercial Industrial Current Lagged 1 Month
1974 26 100 NA 1,425 1,263
1975 37 140 872 1,643 1,425
1975 8 31 345 192 354
1975 7 29 360 252 192
1975 30 109 640 1,654 1,517
1976 36 136 889 1,485 1,654
1976 7 29 307 184 332
1976 7 25 310 262. 184
,~. '
D'ec: 1976. 25 92 706 1,294 1,028
u~ti 1977 40 142 606 1,017 1,294
~uly 1977 7. 26 306 75 208
~~g 1977 6 26 337 144 75
.rlec 1977 32 117 836 1,659 1,486
'Jan 1978 31 118 813 1,349 1,659
quly 1978 9 31 290 186 308
'!\ug 1978 7 23 281 160 186
C(~.
Dec 1978 25 90 708 1,344 1,153
Jan 1979 28 101 788 1,321 1,344
July 1979 7 24 285 NA 265
Aug 1979 6 19 179 NA
SOURCE: Alaska Gas and Service Company records.
D-29
TABLE D.l4. (continued)
DETERMINATION OF AVERAGE NATURAL GAS CONSill1PTION
IN THE RESIDENTIAL SECTOR
C. Regression Results
Equation: monthly consumption a + b * degree days
a is interpreted as monthly, nonclimate related gas
consumption in mcf or simply nonspace heat
related consumption
Dependent Variable a Value
Standard Error
of Equation Independent Variable
monthly residential
consumption
monthly residential
consumption
monthly residential
consumption
monthly commercial
consumption
monthly commercial
consumption
monthly commercial
consumption
monthly industrial
consumption
monthly industrial
consumption
monthly industrial
consumption
2.86 .93
2.87 .88
4.65 .86
10.81 .93
10.93 .89
17.70 .86
239.69 .91
242.94 .93
257.23 .92
D-30
3.60
4.49
5.06
12.58
16.18
18.15
74.09
63.36
68.49
heating degree days
(HDD) lagged
2 month average HDD
HDD
HDD lagged
2 month average HDD
HDD
HDD lagged
2 month average HDD
HDD
data, we average the estimates to obtain 80 percent as the space heat
load of residential natural gas sales.
Table D.l5 shows the total number of residential natural gas space
heat customers, including both gas utilities in the Greater Anchorage
Area. Average annual consumption in the residential sector of the Kenai
sy~tem was 90 percent of the Anchorage system in 1978.
TABLE D.l5. 1978 RESIDENTIAL NATURAL GAS CUSTOMERS
Anchorage
Kenai
Soldotna/North Kenai
Eagle River
Total
27,664
910
1,103
1,856
31,533
SOURCE: Alaska Public Utilities Commission records.
The University of Alaska Fairbanks, Cooperative Extension Service,
has developed a model which is capable of analyzing the fuel requirements
necessary to heat a typical house with design specifications chosen and
input by the model user. This model has calculated the annual fuel
requirements shown on Table D.l6.
D-31
·;:
TABLE D.l6.
"TYPICAL" HOUSE SPACE HEATING FUEL REQUIREMENTS
Single Family House
1. 2300 square feet (2 floors
with daylight basement)
natural gas (mcf)
electricity (kWh)
fuel oil (gallons)
2. 768 square feet (closed
crawl space)
electricity (kWh)
fuel oil (gallons)
3. 768 square feet (heated
crawl space)
electricity (kWh)
fuel oil (gallons)
Mobile Home
1. 768 square feet (closed
crawl space)
electricity (kWh)
fuel oil (gallons)
2. 768 square feet (heated
crawl space)
electricity (kWh)
fuel oil (gallons)
Anchorage
200
40,917
1,492
Fairbanks
52,392
1,910
29,042
1,059
26,620
970
34,873
1,272
32,761
1,194
SOURCE: Axel Carlson, Extension Engineer, Cooperative Extension Service,
University of Alaska, Fairbanks, as reported in Fairbanks North
Star Borough, Community Information Center, Special Report #2,
1978, and Special Report #4, 1976.
D-32
Recent trends in average annual residential consumption per customer
are depicted in Figure D.l. The rapid growth for Golden Valley Electric
Association (GVEA), Matanuska Electric Association (MEA), and Homer
Electric Association (HEA) is primarily due to space heating load.
Table D.l7 shows electric utility monthly residential loads per
customer in 1979 for.several utilities. The winter-to-summer month
ratios provide some indication of the space heating load. Copper Valley
Electric Association (CVEA), with no significant space heating load,
has a winter-to-summer ratio of 1.48. The other utilities with space
heating load have higher winter peaks. Unfortunately, this information
is not precise enough to allow one to draw inferences about the amount
of load devoted to space heat for the various utilities. An attempt was
made to calculate the space heat load for the average space heat customer;
but the results, shown in the final row, are implausible.
Information on the average size of units in the housing stock is
available for Fairbanks and is shown in Table D.l8 along with national
averages.
Anchorage retailers indicate that trailer dimensions have been
increasing over time, although it is difficult to use sizes of trailers
sold to estimate size of trailers in place. This is because people
tend to add to trailers in place. Newer trailers are 924 square feet
(14x66), while those sold in the early 1970s are 732 square feet (12x61).
D-33
Figure D.l.
Residential Electricity Consumption Per Customer
II
~mh/ .·:
year '1_
1
_
-i r
I
..... -,.-
,.-,..-
,.,. .. -,,
-------:---.. ____ ,
L ___ , --------1····-----~-----t-----·------·----··-----.. , _______ -· -
6S t,(> 67 (,i ~1 70 7/ 7:J.. 73 7•{
YEAR
D-34
\ :HEA ~~·
/" GVEA
""""-~ ~__.., CEA
*.AP&T
--I . --J---------·
71 7;?,
TABLE D.l7.
MONTHLY RESIDENTIAL ELECTRIC UTILlTY LOAD FOR 1979
CVEA CEA AMLP HEA MEA GVEA
January 620 1,179 1,131 1,418 2,017 1,308
February 646 1,324 762 1,501 1,936 1,495
March 562 1,127 1,062 1,407 1,691 969
April 525 856 783 1,183 1,396 803
May 466 779 678 1,004 1,079 637
June 432 741 568 909 903 613
July 371 726 563 740 850 562
August 426 583 482 737 771 592
September 432 779 611 720 834 671
October 434 783 410 849 962 743
November 571 953 666 1,002 1,245 887
December 549 1,279 917 1,216 1,590 1,258
Monthly Average 491 871 716 1,054 1,270 877
Winter-Summer R . a at~o 1.48 1.84 1.74 1. 73 2.20 2.30
Total 5-,892 10,452 8,592 12,648 15,240 10,524
Nonspace Heat Load b 5,892 9,828 7, 726 11,429 12,090 8,464
Total Minus
Nonspace Heat 0 624 866 1,219 3,150 2,060
Percent Space Heat
CustomersC 0 .14 15 30 33 6
Space Heat Average 4,457 5,907 4,063 9,545 34,333
SOURCES: Utility Monthly Reports to Rural Electrification Association
and internal utility records.
(a) (December + January+ February)/(June + July + August)
(b) Based upon the CVEA ratio of total annual sales to sales
in the summer months of June, July, and August (4.79).
(c) Author estimate.
D-35
TABLE D.l8. AVERAGE SQUARE FEET FOR VARIOUS HOUSING TYPES
FAIRBANKS AND NATIONAL DATA
Fairbanks a
'
Nationalb
Single-Family 1,384 1,570
Duplex 796 1,370
Apartment 847 900
Mobile Home 919 720
Total 1,116
aFairbanks North Star Borough, Community Information Center, "1978
Fairbanks Energy Inventory," Special Report No. 4, July 1979, p. 40.
b U.S. Department of Energy, Office of Buildings and Community
Systems, Comprehensive Community_Energy Planning: A Workbook, Vol. 1.
D-36
.New homes constructed in the Anchorage area tend to be in the range
of 1,600 square feet on average according to real estate personnel.
This includes a mixture of one-story ranch homes, split levels, and
other types. The average house also seems.to be getting larger. This
is consistent with the hypothesis that the Alaskan housing stock is
being upgraded toward the national average. In 1977, the average size
of new, single-family homes built in the United States wa~ 1,720 square
10 feet. Evidence that the average size of the Alaskan housing unit was
smaller than the national average in 1970 can be inferred from earlier
data from the 1970 census shown in Table D.2 on the median number of
rooms and population per housing unit for Alaska. In all railbelt
census divisions, median number of rooms was lower and population per
occupied housing unit was higher than the average for the western region
of the United States.
•
Real estate and electric utility personnel indicate that housing
units are smaller in the outlying areas of the railbelt such as Seward,
Homer, and Valdez.
Finally, Table D.l9 shows national average estimates of average
size and thermal requirements fo~ various types of structures. This
table demonstrates the apparent variation in thermal requirements of
different types of buildings. Also, the very high average electrical
requirements calculated by the author for Anchorage, based upon the
thermal requirements data, bring into question the use of nationally
determined formulas and ratios for the Alaskan railbelt.
D-37
TABLE D.l9. NATIONAL AVERAGES: RESIDENTIAL SPACE HEATING
Anchorage a
Thermal Electrical
Average Size Requirement
Type of Structure (square foot) (btu/sg.ft./HDD
Single Family Detached 1~570 11.3 56~716
Single Family Attached 1,370 6.2 27,154
Multifamily High Rise 900 4.5 12~947
Multifamily Low Rise 900 5.0 14,386
Mobile Home 720 15.0 34,526
aCalculated on the assumption of 10,911 heating degree days.
SOURCE: 11 Comprehensive Community Energy Planning: A Workbook,"
for the Office of Buildings and Co~~unity Systems, U.S.
ment of Energy, Volume 1, pp. 4-7.
D-38
To begin the actual determination of residential electric space
heating, based upon all this fragmentary information, it is possible to
first net out natural gas users in Anchorage and Kenai-Cook Inlet among
single-family, duplex, and mobile home residents. We assume all resi-
dential gas customers use gas for space heating. We can thus calculate
the non-natural gas c~stomers according to Table D.20. In order to
divide total gas customers among structural types, we note that trailers
are somewhat less likely to be heated by gas according to the 1975
Anchorage survey and use this information for both census divisions.
In 1970, fuels other than natural gas, electricity, and fuel oil
accounted for 7 percent of space heating units in Anchorage and 13 per-
cent in Kenai-Cook Inlet. If we assume no additions to the number of
such units and no conversions, these percentages fall to about 3 and
8 percent, respectively, by 1978. It seems reasonable to assume that
units burning these fuels would not be multifamily but might otherwise
be randomly distributed among different types of structures. Netting
these units out leaves only electric and fuel oil-heated units.
Information on electricity and fuel oil for Anchorage consists of
the census, the l975 Survey, and all-electric homes data; and these
sources are not consistent. According to the census, 6 percent of
residences were electrically space heated in 1970; while only 4 percent
were all-electric homes in the Anchorage utility service areas. Growth
in the proportions of all-electric homes was rapid in the early 1970s,
based on new all-electric homes as a proportion of all new homes (about
D-39
TABLE D.20. NON-NATURAL GAS CUSTOMER CALCULATION
Natural Gas Non-Natural Gas
Total Units Customers Customers
Anchorage 39,700 29,520 10,180
Single Family 28,530 22,437 6,093
Duplex 4,581 3,603 978
Mobile Home 6,589 3,480 3,109
Kenai-Cook Inlet 6,177 2,013 4,164
Single Family 3,660 1,363 2,297
Duplex 484 180 304
Mobile Home 2,033 470 1,563
SOURCE: See text.
D-40
17 percent). If that growth held through the mid-1970s, then using the
all-electric homes information as a base, about 5,800 units would be
all-electric presently, or about 10 percent of the total.
Discussions with realtors indicate a significant proportion of
multifamily units built in the middle 1970s were electrically heated.
The number of building permits issued for multifamily units between 1974
11 and 1977 was 6,855 (including duplexes until 1977) or about 44 percent
of total permits. If their electric space heat installation proportion
was double that of the historical trend, then another approximately
1,000 multifamily units were built that were electric. Apart from these
multifamily units (including condominiums), electric space heating is
allocated based upon total units of each type (not heated by propane,
wood, or coal) after an up•vard adjustment of the total by 3 percent to
account for the discrepancy between the census and the Federal Power
Commission's estimates of all-electric homes and the growth in housing
in areas not served by gas.
In Kenai-Cook Inlet, the census indicates 4 percent electric space
heat in 1970; while the proportion from the 1977 Survey for the three
cities of Kenai, Soldotna, and Seldovia. is 7 percent. Average residen-
tial consumption data for Homer Electric Association (HEA) is indicative
of a substantial growth in space heating load in the last few years. In
1977, a mail survey of their service area indicated a 33 percent electric
space.heat proportion for 7,171 active accounts.12 Based upon consump-
tion data, the 7 percent figure is obviously lm~, which is logical
D-41
because Kenai and Soldotna have access to natural gas; while Homer
itself does not. We scale down to 30 percent the REA figure on the
assumption that it might be somewhat of an over-estimate for 1977 for
first homes but that the electric heat load trend was upward throughout
1977 and 1978. The load is distributed among all struc.tures propor-
tionately after netting out propane, wood, coal, and other fuels.
For Matanuska-Susitna and Seward, natural gas is not an available
option. Seward Electric Association estimated 2 percent all-electric
13 homes, and the census reported 4 percent. We compromise on 3 percent
and allocate them all to single-family and mobile home units. For
Matanuska-Susitna, the census reports 1 percent electrically heated
homes, but much of the growth in the housing stock since then has been
electrically heated units. If all housing added since 1970 was elec-
trically space heated, the proportion could be over 50 percent even
without· retrofitting electric systems. A portion of the Matanuska
Electric Association (MEA) serVice area is in Eagle River where natural
gas is available. Of total MEA residential customers in 1978, about
3,500 were located in the Anchorage Census Division and 1,856 of these
were gas customers. Thus, when MEA calculates 2,685 to be the number of
all-electric customers on their system (26 percent), this converts to
about 40 percent in the Matanuska-Susitna Borough proper. On the basis
of the criterion used by MEA to identify electric space heat customers
(2,500 kWh or more on the December bill), they may have underestimated
their electric space heating load. With this in mind, as "to7ell as a com-
parison of the average residential bill with REA which is estimated at
D ... 42
percent electric, we raise our estimate of electric space heating to
percent. To distribute this among types of units, we assume that use
coal, wood, and propane has fallen from 37 to 15 perc~nt of the total
not utilized in multifamily units at all. Thus, the allocation
of this amount.
For Fairbanks, we have information from the census and from the ·
all-electric home data, both of which indicate about 7 percent electric
mode split in 1970 and the latter which indicates that the proportion
remains constant during the early 1970s. A recent mailback survey
reported 11 percent late in 1979.14 The Fairbanks utilities say that
growth was rapid ill. the mid-1970s but that now, because of the high
price of peak electric power, they are discouraging the use of electricity
for space heating. In April 1975, Golden Valley Electric Association
(GVEA) put a prohibition on further electric space heat installations.
In addition, they are assisting people to get off electric space heat
and, as a consequence, the average residential consumption fell from a
peak of 17,332 kWh in 1975 to 10,524 kWh in 1979. (Their average bill
in 1970 was 10,785 lUfu.)
On this basis, an electric space heat load of about 800 units (the
1970 number) would be reasonable for 1979 and a slightly higher number
for 1978. GVEA estimates that they currently have about 750 electric
space heat customers, which would be about 6 percent. ·we extrapolate
back to 1978 and estimate 1,000 electric space heat units in that year
whiah would be 8 percent of the total stock served by both utilities.
D-43
To allocate the electric heat among units, we estimate 20 percent of the
housing stock is now heated by coal, wood, propane, steam, and other but
that those fuels are not used in multifamily units. The electric space
heat is evenly distributed among all types of units net of these fuels.
The resulting electric space heat mode splits are shown in Table D.21.
Average consumption data is available for past years from the
Federal Power Commission All-Electric Homes reports, from engineering
analyses, and from inferences dra~vn from natural gas consumption. Since
this last source contains the most recent information and is from a
known source, it forms the basis for determining average consumption
which will vary according to:
1. location (heating degree days),
2. type of structure, and
3. size of structure (square feet of floor space).
The age of the structure and the habits of the occupants are impor-
tant sources of variation which cannot be formally addressed at present.
In addition, the heating load can vary considerably from year-to-year
because of variation in weather conditions.
Using the average annual heat load of 162 mcf calculated for
Anchorage Natural Gas customers in 1978, we convert to k\~ of elec-
tricity on the assumption of 65 percent efficiency for gas space heating
and 95 percent for electricity resulting in an electric equivalent of
approximately 32,400 k~. This is the average among three types of
D-44
TABLE D.21. 1978 ELECTRICAL SPACE HEATING PERCENTAGES
Single Multi-Mobile
Family Duplex Family Home Total
Greater Anchorage Area 18.8 18.4 19.9 18.4 19.0
Anchorage 12.9• 12.9 18.9 11.9 14.8
Kenai-Cook Inlet 30.0 30.0 32.7 29.4 30.0
Matanuska-Susitna ·49.5 49.5 58.7 49.5 50.0
Seward 4.0 0 0 5.0 3.0
Greater Fairbanks Area
Fairbanks 7.3 7.2 10.0 7.3 8.0
~
Glennallen-Valdez Area
Valdez-Chitina-Whittier 2 0 0 0 0
D-45
structures of different characteristics and sizes. There are estimated
to be 23,800 single-family units, 3,783 duplexes, and 3,950 mobile homes
using natural gas, for a total of 31,533 units.
Using this information as well as average structure size and space
heating efficiency estimates, the average electric heating load for
Anchorage by type of structure can be calculated. This calculation is
shown in Table D.22.
Floor space is the average of the values for Fairbanks and the
national average. Heating efficiency factors for duplexes are based on
the idea that the heating requirement is a function of wall and roof
surface area, which increases less than proportionally as floor space
increases. Specifically, two duplex units with 1,085 square feet each
have a floor area of 2,170 which is 1.47 the area of the average single-
family unit. The duplex wall and ceiling area, however, is about 1.38
times the single-family unit, indicating that heating requirements per
square foot of floor space will be less. The ratio of outside wall-to-
floor space for the duplex is about 1.85 and for the smaller single-
family house, 2.05. On this basis, one can calculate that the duplex
is about 10 percent more efficient to heat on a square foot basis.
Studies in Fairbanks indicate that a mobile home requires 20 percent
more energy to heat per square foot than a single-family unit of the same
size. Using this assumption ~nd the fact that the average mobile home
has a surface area-to-floor space ratio of about 2.35 (which would mean
D-46
,1.:,
Type of Unit
Single-Family
Duplex
Mobile Home
t::l
I .p.
.......
Total
TABLE D.22. CALCULATIO~ OF ANCHORAGE ELECTRIC SPACE HEAT
LOAD BY TYPE OF UNIT
Number
of Units
23,800
3,783
3,950
31,533
Average
Floor Space
(square foot)
1,480
1,085
820
1,350
Total
Floor Space
35,224,000
4,104,555
3,239,000
42,567,555
Average Heat
Requirement
Per Square Foot
Re~ative to Total
1.0
.9
1.38
1.02
Average
Space Heat
Load/Square Foot
(kWh)
23.53
21.18
32.47
24
Average
Space Heat
Load
(kWh)
34,823
·22,976
26,626
32,400
it consumes about 15 percent more energy per square foot of floor space
f•
than the average single-family home), the average heat ~equirement factor
for mobile homes relative to single-family units thus becomes 1.38.
The requirement for a multifamily unit is calculated similar~y to
that of a duplex assuming an av·erage unit size of 900 square feet and
that the average multifamily structure is 8 units. The ratio of surface
area to floor space is calculated as 1.48. Compared to the single-
family ratio of 2.05, the heating requirement on a square foot basis
would only be 72 percent as large. Since the floor space of the multi-
family unit is 61 percent that of the single-family unit, the overall
energy requirement is calculated at 44 percent of the single-family
unit.
Adjustments for other parts of the railbelt are made on the basis
of average size of units and average heating degree days. The former is
directly available only for Fairbanks, but a_general idea of average
size of unit is available from information on median rooms from the
census. Outlying parts of the Greater Anchorage Area h~ve somewhat
smaller units, and Glennallen-Valdez has considerably smaller units than
Anchorage.
We assume per-unit heating requirements outside Anchorage are
20 percent less for Kenai-Cook Inlet and Seward and 15 percent less for
Matanuska-Susitna than in Anchorage on the basis of smaller average unit
size. For Fairbanks, we assume the average size of units is 92 percent
D-48
of Anchorage, based on actual survey data for Fairbanks on single-family,
duplex, and mobile home units.
No difference in heat requirements, based upon variation in heating
degree days w~thin the Greater Anchorage Area, is assumed. The ratio of
heating degree days in Fairbanks-to-Anchorage for the average of the
1977 and 1978 seasons was 1.43. This information, in addition to aver-
age unit size i~formation, results in an estimate of Fairbanks unit
requirements at 132 percent of those in Anchorage.
For Glennallen-Valdez, we assume the average unit is 75 percent of
the size of an Anchorage unit and average the heating degree days in the
different parts of the census division to obtain a ratio to Anchorage of
122 percent. Combining these yields an estimate for Glennallen-Valdez
which is about 91 percent that of Anchorage. The results of these cal-
culations are presented in Table D.23.
For projection purposes, it is necessary to adjust the figures on
average annual electric space heating requirements to account for the
fact that 1978 was a warmer-than-normal year. Adjustment factors,
based on the ratio of normal to 1978 heating degree days, are used in
the projections and presented in Table D.24.
D-49
TABLE D.23.
1978 AVERAGE ANNUAL ELECTRIC SPACE HEATING REQUIREMENT
GREATER ANCHORAGE AREA
Anchorage
Kenai-Cook Inlet
Matanuska-Susitna
Seward
GREATER FAIRBANKS AREA
Fairbanks/Southeast
Fairbanks
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier
(kWh)
Single Family Duplex Multifamily
34,800 23,000 15,300
27,800 18,400. 12,200
29,600 19,600 13,000
27,800 18,400 12,200
45,900 30,400 20,200
31,700 20,900 13,900
D-50
,. ·.
Mobile Home ·
26,600
21,300
22,600
21,300
35,100
/ TABLE D.24. HEATING DEGREE DAY COMPARISONS
Ratio·of
Average of 1977-78 Normal
and 1978-79 Normal to.Recent
Anchorage 9,548 10,911 1.14
Fairbanks 13,719 14,344 1.05
Glennallen-Valdez 11,621 12,241 1.05
D.3. MAJOR APPLIANCE SATURATION RATES
Present appliance saturation rates (Table D.25) must be based upon
past Alaskan trends in saturation rates, trends in other states, and
national trends in the percentage of wired homes which have at least one
of a particular appliance. This is because current data on saturation
rates is not available. Because future saturation rate projections
utilize the same methodology used to develop the present estimate~,
they are also discussed in this section.
D.3.A. Ranges
The saturation rate is essentially 100 percent for ranges; it is
not assumed to change in the future. ,
D. 3.B. .Refrigerators
The penetration rate (number of households with at least one unit/
numbeh of households) for refrigerators is assumed to be approximately
100 percent. The saturation rate may exceed 100 percent, however,
D-51
TABLE D~25.
1978 CALCULATED APPLIANCE SATURATION RATES
Appliance Census Division
Matanuska-Kenai-Glennallen-
Anchorage Susitna Cook Inlet, Seward Fairbanks Valdez
Hot Water 100 91 94 93 97 91
Clothes Dryers 71 . 65 76 73 66 48
Freezers 42 62 56 66 42 43
Clothes Washers 76 82 85 83 74 65
1:::1
I
VI
N Television Sets 156 108 104 147 149 80
Dishwashers 50. 35 31 45 36 11
Ranges 100 100 100 100 100 100
Refrigerators 100 100 100 100 100 100
Room Air Conditioners 0 0 0 3 1 0
because of multiple ownership. The California Energy Commission, for
15 example, estimates a saturation rate of between 113 and 116 percent.
It can safely be assumed that the colder climate of Alaska reduces the
incidence of second refrigerators below that of California, but a satura-
tion rate in excess of 100 is still a possibility. Since no information
is currently available on this possibility, however, it is assumed that
the saturation rate remains at 100 percent.
D.3.C. Air Conditioners
The incidence of room air conditioners is rare in Alaska. The
1970 saturation rates are assumed to continue in all future years
(Table D;26).
D.3.D. Dishwashers
The saturation of dishwashers should be rel&ted to the rate of
growth of the housing stock since a large proportion of new housing is
built with dishwashers. It should also be related to the incidence of
households with two wage earners because it is a labor-saving device.
On this basis, we assume that the growth in the dishwasher saturation
rate since 1970 is slightly in excess of the national average. In 1970,
the various Alaskan census divisions were close to or slightly lower
than the national average and the Western Region U.S. saturation rates.
Nationally, dishwasher saturation rates have grown at 6 percent
annually since 1970, as calculated from data in Table D.27. For Alaska, ..
we assume 7 percent for A~chorage, Fairbanks, and Seward and 8 percent
D-53
TABLE D.26.
Hot Water Range
GREATER ANCHORAGE AREA
Anchorage
1960
1970
Matanuska-Susitna
1960
1970
Kenai-Cook Inlet
1960
1970
Seward
1960
1970
GREATER FAIRBANKS AREA
Fairbanks
1960
1970
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-ffi1ittier
95
99
64
82
53
88
79
87
88
94
1960 66
1970 57
WESTERN REGION U.S.
1960
1970 99
99
100
100
99
100
98
93
100
94
100
97
98
100
A~J>L.IANCE SATURATION __ RATESa _
Clothes
Dryer
28
59
15
47
18
55
12
53
21
48
6
35
45
Clothes Dish
Freezer Washer Washer
29
39
43
67
35
61
33
71
21
39
19
40
28
62
63
65
68
60
70
72
83
58
61
39
54
70
NA
29
NA
19
NA
17
NA
26
NA
21
NA
6
NA
27
. -----------·---------------------------------
Room Air
Television Conditioning
99
121
61
84
43
81
6
114
88
116
26
40
122
1
0
0
0
0
0
0
3
1
1
0
0
15
aCalculated as the number of appliances divided by occupied housing units. A response of twq or
more television sets or room air conditioners is counted as two sets.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Tables 29, 30.
1970 Census of Housing, Detailed Housing Characteristics: Alaska. Tables 62, 63.
1970 Census of Rousing, Detailed Rousing Characteristics: United States Summary Tables 23, 24.
TABLE D.27. PERCENTAGE OF ELECTRICALLY WIRED HOMES
WITH SELECTED APPLIANCES
1960 1965 1970 1973 1974 1975
Rooni
Conditioners 15.1 24.2 40.6 48.9 51.6 52.8
Dishwashers 7.1 13.5 26.5 34.3 36.6 38.3
Clothes Dryers
(include gas) 19.6 26.4 44.6 53.9 56.5 57.7
23.4 27.2 31.2 37.9 41.7 43.5
Televisions
(color) 9.5 42.5 67.1 71.5 74.4
Televisions
(black & white) 89.4 97.1 98.7 99.9 99.9 99.9
Clothes \vashers 68.3 57.4 62.1 67.8 68.4 69.9
Refrigerators 98.2 99.5 99.8 99.9 99.9 99.9
1976
54.4
39.6
58.6
44.4
77.7
99.9
72.5
99.8
SOURCE: U.S. Department of Commerce, Statistical Abstract, annual.
D-55
1977
55.3
40.9
59.3
44.8
81.3
99.9
73.3
99.9
for the other smaller census divisions with lower 1970 saturations.
Subsequent to 1980, the saturation rate (St) is determined by the
following equation:
= S l + (1-S 1 ) * t-t-.02
The parameter. .02 is based upon a rough estimate of the long-run national
trend.
D.3.E. Clothes Washers
With respect to clothes washers in 1970, most Alaskan railbelt
census divisions had saturation rates relatively close to the national
and Western Region U.S. rates. Thus, the same growth rate observed
nationally between 1970 and 1977 is applied to Alaskan census divisions
to generate 1978 saturation rates. This procedure is modified ,only in
the case of Seward, where it would result in a saturation rate in excess
of 100 percent. Th~ saturation rate for Seward is arbitrarily maintained
at its 1970 level of 83 percent.
Subsequent to 1978, the increase in saturation-rates is assumed to
moderate as a practical upper limit on clothes washer. saturation would
be less than 100 percent. This is based upon the assumption that most
apartment units will have washer and dryer units for tenants but at less
than a one-to-one ratio. Consequently, the growth of clothes washer
satu.rations are based upon the following equation:
D-56
D.3.F. Clothes Dryers
The 1970 Alaskan saturation rates for clothes dryers are close to
the national average except for Anchorage, where it is considerably
higher. Thus, all census division saturation rates except Anchorage are
assumed to grow at the national average rate between 1970 and 1978.
Anchorage already had reached the· average nation~l saturation rate for
1978 of 59 percent in 1970. Also, Anchorage figures reflect a higher-
than-average ratio between clothes dryers and clothes washers. We base
the Anchorage clothes dryer estimate on the clothes washer estimate and
on this 1970 ratio.
Subsequent to 1978, it is assumed that the ratio between clothes
dryers and washers will remain constant so that clothes dryer saturations
grow at the same rate as clothes washer saturations.
D.3.G. Freezers
Alaskan freezer saturation rates were consistently and considerably
above the national and Western Region rates for 1970. In addition, the
rural census divisions had considerably higher saturations than the
urban areas, suggesting that the high saturation might be partially
attributable to the unavailability of adequate grocery facilities. It
seems reasonable to assume that this will continue to be .an important
factor in determining future freezer saturation rates and, in addition,
that the strong hunting and fishing interests of the typical Alaskan
will contribute to high freezer saturation rates.
D-57
..
Nationally, the freezer saturation rate has grown about 5 percent
annually since 1970, compared to 3 percent annually in the previous
decade. These growth rates applied to 1970 Alaskan saturation rates
would give unreasonably high values for 1978 and subsequent s<~turation
levels.
We assume that for the Anchorage, Fairbanks, and Valdez-Chitina-
Whittier Census Divisions, the annual saturation rate growth rates are
one percent. For the other census divisions with unusually high 1970
saturation rates, we assume that they decline by one percent annually to
reflect growth in those areas close to adequate grocery facilities.
Subsequent-to 1978, those census divisions which have increasing
saturation rates continue to grow, based on the following equation:
while those census divisions with declining saturation rates are assumed
to maintain constant saturation rates in future years.
D.3.H. Water Heaters
Alaskan 1970 water heater saturation rates show surPrising varia-
tion among census divisions. Here national data cannot serve as a guide
because the national average is virtually 100 percent. This was the
case in Alaska only for Anchorage. One possible explanation for the
relatively low saturation rates might be the fact that a percentage of
D-58
the population in 1970 lived outside utility service areas and also
areas where petroleum products were readily available. This proportion
population has been greatly reduced since 1970 as population
grmv-th has centered in the urban areas.
One pattern that does emerge from examinatio.n of the historical
Alaskan saturation data is that between 1960 and 1970, the saturation
rates grew rapidly toward 100 percent (except in the inexplicable case
of Valdez-Chitina-Whittier). On this basis, it is reasonable to assume
~
a continuation of those growth trends in the 1970s. We assume that
between 1970 and 1978 nonsaturation (1-S) is reduced by 50 percent in
each instance, except for Valdez-Chitina-Whittier which is arbitrarily
set equal to Matanuska-Susitna. Subsequent saturation rate growth results
in 100 percent saturation by 1990.
D.3.I. Television Sets
The differences in television saturation rates between Alaska
census divisions and the nation likely reflect the possibility of the
household's receiving a television signal. Thus, Anchorage has a pat-
tern close to the national growth, while the saturation of television
sets in Seward grew dramatically between 1960 and 1970.
Nationally, growth in saturation in the 1970s has been the result
of increased ownership of color television sets, which nearly doubled
between.l970 and 1977. We assume the same growth rates in the 1970s
for Alaskan census divisions, except for Valdez-Chitina-Whittier which
D-59
is assumed to follo\v the pattern of Kenai-Cook Inlet with an approximate
ten-year lag. ..
Future growth of television saturations will be modest if recent
national trends can be used to gauge future trends. Alaskan growth will
continue to lead national trends, but Alaskan saturations will not reach
national levels. We assume a growth in saturation as follows:
st = s 1 + (2-s . 1 ) "' . 02 t-t-
D.4. MAJOR APPLIANCE FUEL TYPE MODE SPLIT
Four. appliances--water heaters, ranges, clothes dryers, and
refrigerators--may normally be designed to operate on fuels other than
electricity. The appliance mode split, defined as the proportion of ..
appliances using a particular energy fuel, is determined by current
relative prices of fuels and appliances and consumer tastes, as well as
the past values for these variables. That. is, the current mode split
will partially be a reflection of past patterns of relative prices which
are no longer relevant for the purchase of new appliances so that the
mode split observed at any point in time may not be an equilibrium
·split.
Because mode split is determined by these factors, national infor-
mation is not particularly applicable to Alaskan conditions. In order
D-60
to calculate mode split, we thus rely on Alaskan information. Because
information on consumption of other fuels in the~e appliances is rele-
. vant to the determination of what percentage of the appliance utilizes
electricity, it is utilized wherever appropriate. (It is basically used
to provide a check on the mode split by insuring that some alternative
fuel has not been allocated an unrealistically large or small percentage
of total appliances of a particular type.)
Information to develop the mode split estimates sho~vn in Table D.28
comes primarily from the census of housing (Table D.29) and conversations
with utility, horne construction, and real estate personnel.
D.4.A. Within the Natural Gas Service Area
The starting point of the analysis is to net out, where applicable,
natural gas utility sales. No natural gas refrigerator sales have been
identified, except for campers and recreational vehicles; so water
heaters, ranges, and clothes dryers are the appliances which may be gas
fired. From the 1970 census and more recent national data, one can
calculate ratios between gas space heating customers and gas water
heating and cooking customers. These ratios are shown in Table D.30.
They indicate that the choice of gas water heat is closely asso-
ciated with the gas space heat decision but that such may not be the
case for cooking fuel. In particular, Anchorage is far ·below the
national average for gas ranges among natural gas customers.
D-61
TABLE D.28. 1978 ELECTRICAL APPLIANCE MODE SPLITS
CENSUS DIVISION
Greater
Matanuska-Kenai-Anchorage
Anchorage Susitna Cook Inlet Seward Area Fairbanks Glennallen
Water Heater 33 43 41 35 34 43 40
Range 66 75 35 53 64 81 40
Clothes Dryer 91 96 78 70 90 98 75
t;:j Refrigerator 100 100 • 100 100 100 100 100 I
0'\
N
PROPORTION OF APPLIANCES USING VARIOUS FUELSa
Utility Gas Oil Electric Coal Wood Propane Other
GREATER ANCHORAGE AREA
Anchorage
1960 0 41 39 15 0 4 1
1970 45 13 35 1 0 4 2
Matanuska-Susitna
I=' 1960 0 19 50 20 4 I 7 0
Cl' 1970 0 28 38 11 UJ 12 11 0
Kenai-Cook Inlet
1960 0 60 21 7 0 12 0
1970 34 17 37 2 0 9 1
Seward
1960 33 57 0 0 5 5 0
1970 0 60 35 0 0 5 0
a Proportions sum to 100.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Tables 29' 30.
1970 Census of Housing, Detailed Housing Characteristics: Alaska. Table 63.
TABLE D.29. (continued)
( '
WATER HEAT MODE SPLIT INFORMATION
PROPORTION OF APPLIANCES USING VARIOUS FUELSa
GREATER FAIRBANKS AREA
Fairbanks and
Southeast Fairbanks
. '1960
1970
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier
a
1960
1970
Proportions sum to 100.
Utility Gas
0
3
0
• 0
Oil
23
33
56
41
Electric
19
37
4
36
Coal
48
15
0
0
SOURCES: 1960 Census of Housing, General ·Housing Characteristics: Alaska.
1970 Census of Housing, Detailed Housing Characteristics: Alaska.
. ·. . ' .: . . . ',
Wood
0·
0
0
0
Propane
9
8
0
23
Tables 29, 30.
Table 63.
'ttrt·trttt&'i#!tt#ittttatiS·Jrw·tt¥Atrtw''t&Jfifi&rf#ttSifftr'~;.~tfir±t&ttre¥wrew&&t¥ri¥!t*¥#ijAW®Mm,'i%~*+w&ttwer·B¥k-·:'w,·t.~:,r:v?+mi4~~~~~~~~1i~~-ik_~~~~,,J:~~"'~~J.:~~,~-~~~~:-i~~;\1A~-,"~: .. ·( ... ~·.}.·.
Other
1
4
40
0
PROPORTION OF APPLIANCES USING VARIOUS FUELS a
Utility Gas Oil Electric Coal Wood Propane Other
GREATER ANCHORAGE AREA
Anchorage
1960 2 8 64 0 0 26 0
1970 20 1 65 0 0 14 0
Matanuska-Susitna
t:;j
I 1960 6 7 38 1 15 33 0
~
\.J1 1970 0 1 44 0 16 38 1
Kenai-Cook Inlet
1960 21 17 23 8 0 31 0
1970 30 4 24 1 1 40 0
Seward
1960 0 29 35 0 4 32 0
1970 4 15 47 0 0 34 0
a Proportions sum to 100 •.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Tables 29' 30 •
1970 Census of Housi\T' Detailed Housing Characteristics: Alaska. Table 63.
TABLE D.29. (continued)
COOKING APPLIANCE MODE SPLIT INFORMATION
PROPORTION OF APPLIANCES USING VARIOUS. FUELSa
Utility Gas Oil Electric Coal Wood "Propane Other
GREATER FAIRBANKS AREA
Fairbanks and
Southeast Fairbanks
1960 3 2 62 2 2 29 0
1970 4 1 76 0 1 18 0
~
I
"' "'
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier
1960 3 19 29 0 14 35 0
1970 6 6 18 0 5 65 0
aProportions sum to 100.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Tables 29' 30.
1970 Census of Housing, Detailed Housing Characteristics: Alaska. Table 63.
·'· £." '~:; ._, A .... :. ;;. ;._ ~.. ',
TABLE D.29. (continued)
CLOTHES DRYING APPLIANCE MODE SPLIT INFORMATION
PROPORTION ..OF APPLIANCES USING VARIOUS FUELSa
Utility Gas Oil Electric
GREATER ANCHORAGE AREA
Anchorage
1960 1 0 99
1970 9 0 91
Matanuska-Susitna
t;;
I
0\ 1960 0 0 100 .....,
1970 7 0 93
Kenai
1960 15 0 85
1970 22 0 78
Seward
1960 18 0 82
1970 30 0 70
a Proportions sum to 100.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Tables 29, 30.
1970 Census of Housing, Detailed Housing Characteristics: Alaska. Table 63.
TABLE D.29. (continued)
CLOTHES DRYING APPLIANCE MODE SPLIT INFORMATION
PROPORTION OF APPLIANCES USING VARIOUS FUELSa
Utility Gas Oil Electric
GREATER FAIRBANKS AREA
Fairbanks and
Southeast Fairbanks
1960
1970
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier
1960
1970
a ,
Proportions sum to 100.
1
2
100
50
0
0
0
0
99
98
0
50
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska.
1970 Census of Housing, Detailed Housing Characteristics: Alaska.
Tables 29, 30.
Table 63.
TABLE D.30. PROPORTION OF GAS SPACE HEATING CUSTOMERS
WITH OTHER GAS APPLIANCES
Water Heater Range
Anchorage 1970 .84 .38
Kenai-Cook Inlet 1970 .96 .94
United States 1970 1.00 .89
United States 1975 .79
SOURCES: 1970 Census of Housing, Detailed Housing Characteristics:
Alaska.
American Gas Association, Gas Facts, annual.
We assume that since 1970 the ratios of gas water heating to gas
space heating have approached 100 percent and the gas range ratio has
increased in Anchorage and remained constant in Kenai-Cook Inlet.
Growth in the housing stock in each census division has been about
50 percent since 1970 so that the differences between the national and
local ratios could have declined by 30-to-50 percent. We assume 50 percent
for water heaters to arrive at 92 percent for Anchorage and 98 percent
for Kenai-Cook Inlet. Noting the declining ratio in gas ranges nation-
ally, we estimate 50 percent for Anchorage and 90 percent for Kenai-Cook
. Inlet.
D-69
I This yields the mode spli~ estimates for gas water heaters and
ranges for 1978, shown in Table D.31.
TABLE D.31. 1978 GAS APPLIANCE MODE SPLIT ESTIMATES
Gas Space•Heat Gas Hot Water Gas Range
Anchorage .69 .63 .35
Kenai-Cook Inlet .33 .32 .30
Comparing these estimates to the 1970 census information on mode
split indicates that the gas mode split has grown in Anchorage and has
remained constant in Kenai-Cook Inlet. This is consistent with the
observation that 'growth in Kenai-Cook Inlet has been balanced between
areas accessible to gas and those not accessible to gas.
To determine the electric mode split for these regions and appli-
ances, use is made of an incremental mode split calculation using the
1960 and 1970 censuses. These incremental mode splits are calculated as
the increments to the number of a particular appliance between 1960 and
1970 divided by the increment to total housing units over the same
period (see Table D.32). Applying these incremental mode splits for
electric water heating and cooking to the increase in housing units
between 1970 and 1978 (about 50 percent) results in average mode splits
which are reasonable when considered in relation to the natural gas mode
splits, except for ranges in Kenai-Cook Inlet. These values are, therefore;
D-70
TABLE D.32. INCREMENTAL MODE SPLIT CALCULATIONS
BASED UPON HOUSING CENSUS DATA
•
Electric Electric
Built-In Space Heating Water Heating Electric Electric·
Electric Units Fuel Fuel Cooking Fuel Clothes Dryer
GREATER ANCHORAGE AREA
Anchorage .1403 '.1695 .2962 .6706 .9841
Matanuska-Susitna .0525 .0676 .2770 .7669 1.8919
Kenai-Cook Inlet .0815 .0638 .4989 .2378 .6410
Seward 0.0 -0.0486 .1991 .1644 .2431
t;;;l
I
"-J ...... GREATER FAIRBANKS AREA
Fairbanks/Southeast
Fairbanks .8320 .5553 1.6292 2.0463 2.3406
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier .0427 o.o .7974 -0.1897 . 7672
SOURCES: Census of Housing: 1970 Detailed Housing Characteristics Final Report HC(l)-B3, Alaska.
U.S. Census of Housing: 1960 Vol. 1, States and Small Areas Final Report HC(l)-3, Alaska.
used as the estimates, except that the Kenai-Cook.Inlet range estimate
..
is increased based on Homer Electric Association survey data. For gas
clothes dryers, the 1970 mode split is used for 1978.
D.4.B. Outside the Natural Gas Service Area
For Matanuska-Susitna, the information from. the census; is difficult
to interpret because, bet>v-een 1960 and 1970, electric water heaters
declined as a percentage, while those of wood and propane increased.
Wood and propane also gained as cooking fuels, but electricity also
increased at the expense of oil. One explanation of this phenomenon
would be that the population increase was in relatively inaccessible
areas and, thus, the same pattern of growth might not be expected to
have occurred in the more recent decade.
If we can assume that 90 percent of the appliances used by new
residents since 1970 are oil or electric, then the proportions accounted
for by coal, wood, propane, and other fall to about 18 percent for water
heat and 30 percent for cookin& and 4 percent for clothes drying. This
is based upon the observation that the number of residences has essent
doubled since 1970. In addition, changeovers of existing appliances
toward electricity and oil probably occurred, thus further decreasing
the proportions of the other fuels. If this was about 25 percent, then
the remaining'percentages would be 14 percent for water heat and 23 per-
cent for cooking, with clothes drying assumed unchanged.
D-72
In the absence of better information, we assume an equal mode split
between oil and electricity for water heat and a continuation of the
1970 census ratio for cooking.
In Seward, electric appliances made imp~essive gains over propane
and oil in the 1960s as a proportion of the totals in spite of declines
in the total housing stock. At the same time, oil maintained its share
of the water heat market. In the absence of other information, we
assume a continuation of the same mode splits for water heat and clothes
drying. For cooking, we assume 75 percent of new households choose
electricity, which brings the electric mode split up to 53 percent.
Fairbanks electricity consumption for water heat increased in the
1970s, as did fuel oil consumption, both at the expense of coal which
was rapidly phased out. We assume that no new housing since 1970 uses
·coal and that half the existing coal units have been converted. This
leaves coal with about 4 percent mode split in 1978 since the housing
units in Fairbanks have nearly doubled in the 1970s. We assume that the
other fuels, except oil and electricity, maintain their 1960 mode splits
and that oil and electricity maintain their 1970 ratio, which gives a
43 percent split for electricity.
Electricity is preferred for cooking by the majority of Fairbanks
households. If 85 percent of new households since 1970 chose it, then
the electric cooking mode split would be about 81 percent. We assume
the ·clothes drying mode split is unchanged from 1970.
D-73
In the Glennallen-Valdez area, electric water heaters become much
more pop~lar in the 1960s. On the other hand, the electric mode split
for cooking fell. In the absence of other information, we assume the
mode splits for the Glennallen-Valdez area move toward those of similar
parts of the railbelt, as reflected in the 1970 census.
D.5. ELECTRICAL APPLIANCE AVERAGE ANNUAL CONSUMPTION
A reliable library of information on average electricity consump-
tion by various appliances is just now beginning to emerge based upon
manufacturer estimates, modeling efforts, and actual surveys which
monitor consumer behavior. Some of the more widely circulated estimates
appear as Table D.33. These estimates are all national averages.
Examination of the various estimates reveals several important
facts. First, only one--the Midwest Research Institute--is based upon
actual metering of appliances; and it shows a considerable variance from
other sources for the use of several appliances. Their results for
ranges, dishwashers, and water heaters are below the others and for
refrigerators, considerably above. This suggests actual use patterns
may be considerably different from those assumed by manufacturers and
other researchers.
Second, the consumption of electricity by appliances depends upon
the features of the appliance. In particular, larger refrigerators and
D-74
?:'A~L !;:__1;1_=-;i_~ AVERAGE ANNUAL ENERGY CONSUMPTION EST MATE
FOR MAJOR APPLIANCES: VAR IOU S S OURCES
(kWh /year )
1976-1977 1968 1976 19 7 8
Midwest Edison Stanford 1970 1973 California California
Research Electric ·Research Arthur D. Merchandising Energy Energy
Institutea Instituteb InstituteC Littled Weeke Commissionf Commissiong
Water Heater 4,046 4,811 4,490 5,626 4,515 3,025 4,876
Cooking 1,180 1,143 1,200
Range 782 700 2,071 778
Auto. Clothes Washer 88 103 98 88 90 70 76
including hot water 2,500 1,115
Clothes Dryer 1,032 993 990 937 993 950 1,212
Refrigerator (12 cu.ft.) 1,665 1,270 1,084 1,228 1,235 1,515
frostless (12 cu.ft.) 728
Refrig/Fr e ezer (12 cu.ft.) 1,217
t:l frostl e ss (12 cu.ft.) 1,500
I frostless (17 .5 cu.ft.) 2,250 ......
U1
Fr e ezer 1,342 1,348 1,480 1,294
regular (16 cu.ft.) 1,190
frostless (16.5 cu.ft.) 1,820
Di shwash e r 149 363 360 . 352 363 230 305
including hot water 2,100 925
Te levision -Black & White
tube 350 360
solid state 120
general 362 140 129
Television -Color
tube 660 490
solid state 440
g e ner a l 502 420 300
Television -General 439
Air Conditioner 978 860 1,389
SOURCES FOR TABLE D.33.
(a) Midwest Research Institute, Patterns of Energy Use by Electrical Appliances, EPRI EA 682,
January 1979, p. 5-3.
(b) Edison Electric Institute.
(c) Office of Science and Technology Executive Office of the President, Patterns of Energy
Consumption in the United States.
(d) Federal Energy Administration, Project Independence Blueprint Final Task Force Report,
Residential and Commercial Energy Use Patterns 1970-1990. Under the Direction of
Council on Environmental Quality, November 1974, Volume I, p. 98. (Presented in btu's
and converted by author.)
(e) Merchandizing Week as reported in Patterns of Energy Use by Electrical Appliances,
EPRI EA 682, p. 5-23.
(f) California Energy Commission, "Appendix A: Analysis of Residential Energy Uses,"
300-76-034, 1976.
(g) California Energy Commission, "California Energy Demand 1978-2000: A Preliminary Assessment,"
August 1979, Table D.5. These figures are for new appliances purchased in 1978.'
f re ezers, as well as frost-free features, add considera bly to annual
energy requirements. In contrast, solid state televisions use con-
siderably less electricity than tube models.
Third, the consumption of hot water, and thus of electricity by a
wa ter heater, is related to the ownership and utilization of dishwashers
and clothes •.;rashers.
Because of these complications, it is not a straightfon.;rard matter
t o apply national averages to Alaska. Furthermore, average annual
consumption of electricity in various appliances will vary as a function
of the following:
1. geographic location
2. size of household
3. income of household
4. age of appliance stock.
Geographic location is obviously a factor iri room air conditioning
consumption and appears also to be an important facet in refrigerator
d f . 1 h h h . f . . 1. . 16 an reezer consumpt1on, a t oug t e 1n ormat1on lS pre 1m1nary.
It is reasonable to assume that refrigerators need not work as hard in
colder climates and will not be opened and closed as often. Water
heating requirements may vary based upon the average annual inlet
temperature of the cold water being heated .
D-77
It is also reasonable to expect that use of electric ranges, clothes
'"ashers and dryers, water heaters, and perhaps dishwashers would b;
related to household size. There is some statistical information and
h h . h h . 17 researc to support t ~s ypot es~s. The relationship appears most
important for clothes washers where consumption might be estimated as
a simple multiple of household size; while for the others, there is a
significant "base load" independent of household size.
Common sense and economic theory suggest that households with
higher incomes will prefer appliances with features which may (frost-
.free and large refrigerators) or may not (solid state television sets)
be more energy using than the average. This idea is related to the age
of the appliance stock which is a reflection of paftt purchase decisions
based upon past income levels. As incomes grow over time, the existing
stock of appliances will not accurately reflect present income levels of
households but rate back present and past levels.
There is no direct information available on Alaskan-specific elec-
tricity consumption in various large appliances or how that consumption
may differ from national norms. On the basis of climate, one might
assume higher-than-average use for water heaters and ranges and lower-
than-average use for air conditioners and refrigerators. On the basis
of a slightly smaller average household size in Alaska, use of clothes
washers and dryers, water heaters~ and dishwashers might be less than
the average. On the basis of income and the appliance stock, Alaska
could be less than or greater than the national average. Contributing
D-78
a lower average for Alaska would be the fact ·that, historically,
•Alaskan personal income per capita has been less than the national
The higher-than-average per capita income in recent years in
combination with rapid growth in the appliance stock could more than
offset this, however.
Since there is no clear direction to adjust the national data, the
choice becomes essentially that of choosing among the various national
series available. The choice makes significant difference only in the
_cases of water heaters, ranges, refri-gerators, and dishwashers and
clothes washers.
We use the California Energy Commission's estimates of water heat
consumption because they separate out consumption related to clothes and
dish washing machines. The latter, we attribute directly to those
appliances ba,.sed upon the probability of an mmer of an electric washer
also owning an electric water heat·er. This could be an underestimate
because of not counting hot water consumption for dish washing by those
households without dishwashers. To adjust for this and the added heat-
ing load caused by a colder-than-average water inlet temperature, we
adjust the base figure upward by 15 percent·.
We define the appliance entitled "range" broadly to encompass all
heating for cooking purposes, including electric skillets, etc. This
probably accounts for the discrepancies among the national estimates • ...
D-79
Variation among estimates for refrigerator consumption is
and is a function of the average size and type of refrigerator ass~med for
the stock. The Midwest Research Institute estimate may be high because
the sample chosen for metering was 97 percent single-family homes and
duplexes.18 The other estimates fall within a much narrower band.
Variation in freezer use is again a function of the stock although
the various sources are in general agreement as to the average stock
characteristics and electricity consumption. Anchorage merchants indi-
cate most of their sales are in the manual defrost category, so we choose
a lower-range estimate.
Clothes dryer consumption may be nearer the high range of estimates
in Alaska because of the cold climate. Air conditioner consumption will
be considerably below the national average, and television consumption is
a broad average of all estimates. The values used in the model are sum-
marized in Table D.34.
TABLE D.34. MODEL VALUES FOR MAJOR APPLIANCE
AVERAGE CONSUMPTION FOR 1978.
Water Heater
Rc;tnge
Clothes Dryer
Clothes Washer
\ Clothes Washer Hot Water
Refrigerator
Freezer
Dishwasher
Dishwasher Hot Water
Television
Air Conditioner
D-80
Average kWh/Year
3,475
1,200
1,000
70
1,050
1,250
1,350
230
700
400
400
SMALL ELECTRIC APPLIANCE USE
The average annual electricity requirements of several common,
·smaller electrical appliances are listed in Table D.35. Individually,
the electricity used in each is not a large amount; but in the aggregate,
it can be substantial for a household. Netting out the calculated quan-
tities of electricity consumed in the residential sector for large
appliances and space heating results in the amounts attributable to.
small appliances in 1978 as shown in Table D.36.
SOURCE:
TABLE D.35. AVERAGE ANNuAL CONSUMPTION OF
VARIOUS SMALL APPLIANCES
(kWh)
Trash Compactor 50
Iron 144
Electric Blanket 147
Humidifier 163
Hair Dryer 14
Clock 17
Sewing Machine 11
Vacuum Cleaner 46
Hi-fi (tube) 120
Hi-fi (solid state) 30
Radio (tube) 100
Radio (solid state) 10
Headbolt Heater 480
Garbage Disposal 36
Lighting 720
Edison Electric Institute, Canadian Wind Energy Program Papers
by the National Research Council of-Canada, and Alaska Village
Electrical Cooperative "Light Lines," May 1978.
D-81
TABLE D.36. 1978 RESIDENTIAL ELECTRICITY CONSUMPTION
ATTRIBUTABLE TO SMALL APPLIANCES·
kWh/Customer
Greater Anchorage Area ~ 2,010
Greater Fairbanks Area 2,466
Glennallen-Valdez Area 2,333
It is not possible at present to identify the specific small appli-
ance mix in each region or to account for the interregional differences
in small appliance consumption. We assume the Alaskan lighting load is
considerably greater than the national average and arbitrarily set it
at 1,000 kWh annually. The remainder is attributed to all other small
appliances.
D.7. COMMERCIAL-INDUSTRIAL REQUIREMENTS
Commercial and industrial electricity requirements supplied by
utilities are combined because of the small industrial base in the
railbelt. Table D.37 presents the distribution of employment by category
and indicates the relatively small size of manufacturing employment in
the railbelt economy. Table D.38 further shows that the total number of
manufacturing establishments is small (244), that they average 16
and that a large number are food and kindred products (at ·least 47) or
D-82
t::l
I
00
... _______ ~ABLE D.37. 1978 WAGE AND SALARY EMPLOYMENT .. . . . . . .. ____ _ ·--·· ' --. . .... ··-.. -.. ······~·· .. , -·· . . .. ·····------"---·-·····--'----·-·---:·.,..
Government
Civ. Mil.* Trade
GREATER ANCHORAGE AREA
Anchorage
Kenai-
Cook Inlet
Matanuska-
Susitna
Seward
21,161 11,098 16,865
1,414 0 1,190
1,220 7 639
313 101
Finance-
Services Ins.-R.E.
15,526 5,081
853 197
363 124
164 16
Trans.-
Comm.-
Utilities
7,924
574
307
**
Manuf.
1,683
989
Const. Mining
6,431 1,874
485 805
235 **
12 **
Misc.
459
**
**
w GREATER FAIRBANKS AREA
Fairbanks
Southeast
Fairbanks
7,218
570
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-
Whittier 838
* Active duty
5,399 4,072
845
0 259
3,939 1,004 2,765 564 1,960 54
157 24 0
409 56 362 ** 89 **
)'cl'( Information withheld under regulations protecting confidentiality of data for individual firms.
86
**
SOURCE: State of Alaska Department of Commerce and Economic Development, Division of Economic Enterprise;
"Numbers: Basic Economic Statistics of Alaska Census Divisions," 1979.
Total
88,040
6,565
3,090
1,327
27,061
1,719
2,043
Food and Kindred
Textile Mills
Apparel and Other Textiles
Lumber and Wood
Furniture and Fixtures
Paper
Printing and Publishing
Chemicals
Petroleum and Coal
Rubber and Misc. Plastics
Leather and Products
Stone, Clay, Glass
Fabricated Metal
Nonelectrical Machinery
Electrical Equipment
Transportation Equipment
Instruments
Miscellaneous
Total Reporting Units
Average Second Quarter
1979 Employment
TABLE D.38. RAILBELT MANUFACTURING ACTIVITY
Number of Reporting Units in 1979
Kenai-Matanuska-Southeast Valdez-Chitina-
Anchorage Cook Inlet Susitna Seward Fairbanks Fairbanks Whittier
23
2
6
10
1
1
33
1
2
4
0
8
8
10
1
4
1
18
133
1,709
20
0
1
4
0
0
4
1
2
0
0
5
0
3
0
5
0
1
46
1,155
14 11
35 ,.,
4
0
2
4
2
0
7
0
1
1
0
3
3
3
1
0
0
3
34
682
2 4
* 15
* Information withheld under regulation protecting confidentiality of data.
SOURC~~ State of Alaska Department of Labor. Statistica1 Quarter1y, Second Quarter 1979.
printing and publishing (at least 44). Petroleum processing and seafood
processing are the most visible components of the manufacturing sector in
terms of employees.
Electricity and other energy end use in the commercial sector is
most conveniently tabulated on the basis of the energy requirement per
square foot of floor space. An accurate measure of this quantity--and
its disaggregation into types of commercial floor space--is not available
for any portion of the railbelt at the present time.19 The only fragment
~
of information available is an unpublished study conducted by the Muni-
cipality of Anchorage Planning Department· which classified all nonresiden-
tial floor space by use.20 An attempt was made, shown as Table D.39,
to modify and update the survey results to make them compatible with the
purposes of this study. The result_is an estimate of 37 million square
feet of floor space in Anchorage in 1978. This figure is about 35 percent
higher than would be calculated using Anchorage employment and national
square feet per employee ratios. The difference may be attributable to
a high vacancy rate in Anchorage office and retail space, a large pro-
portion of newer construction, or the mix of employment within industries.
This result is also considerably higher than a recent informal realtor
21 survey •.
Estimates of floor space in other regions of the railbelt can be
based either on national ratios of floor space per employee or on the
22 estimate developed for Anchorage~ The Anchorage ratio is preferred
because it is based, however roughly, on Alaskan data.
D-85
TABLE D.39.
CALCULATION OF ANCHORAGE
COMMERCIAL-INDUSTRIAL FLOOR SPACE
AMATS Survey (Anchorage Bowl, 1975)
Minus Non-energy Using (parking lots,
cemetaries, etc.)
Energy Using Floor Space
20 Percent Adjustment for Underreporting
Sectors not Included in Survey:
1. Girdwood/Indiana b
2. Eagle River/Chugiak
3. Hotels/Motelsc
4. Assorted Cultural Buildingsd
Retail Trade
Warehousing
Education
Wholesale Trade
Transport-Communication-
Public Utilities
Government
Manufacturing
Other
6,148
3,722
3,528
3,131
2,663
1,405
706
7",331
Growth Between 1975-1978e (approximately 25 percent)
f 1978 Estimated Commercial-Industrial Floor Space
General
Education
Warehousing
Hotels
Manufacturing
See following page for table notes.
D-86
25,120
5,000
4,520
1,500
860
42",067
18,918
23,149
4,630
27,779
53
300
1,000
500
29,632
7,400
37,000
TABLE D.39. (continued)
(a) Twenty-five businesses in 1975 according to telephone book.
Assume 2,500 square feet/business.
(b) Based on the ratio of the housing stock in 1978 between Eagle
River/Chugiak and Anchorage.
(c) Assumes 2,000 rooms at 500 square feet/room. Based on Oak Ridge
National Laboratory, "Commercial Energy Use: A Disaggregation by
Fuel, Building Type, and End Use," Oak Ridge, Tennessee, p. 40.
(d) Forty-six establishments identified in 1975 telephone book. Average
size assumed to be 10,000 square feet.
(e) This is based upon two indicators. The first is the growth in
employment between 1974-75 and 1978. Civilian employment WqS as
follows: 1974-58,700, 1975-69,650, and 1978-76,900. Employment
growth was 31 percent in the period 1974 to 1978 and 10 percent in
the period 1975 to i978. (State of Alaska, Department of Labor,
Alaska Labor Force Estimates by Industry and Area, various issues.)
·The second is the growth in the appraised value of buildings over
the period 1975 to 1978. After adjusting for inflation, the increase
was 48 percent. Based on the assumption that the rapid employment
increase in 1975 resulted in undersupply of floor space.in that
year, we assume a 25 percent growth in floor space between the
summer of 1975 and 1978.
(f) Independent estimates of floor space in 1978 in the educational
category and the hotel/motel category were available from the
Anchorage School District and Anchorage Chamber of Commerce,
respectively. The remaining growth was allocated proportionately
among the other categories.
D-87
On this basis, 1978 commercial-industrial floor space estimates for
* the railbelt have been developed and are presented in Table D.40, using
the Anchorage estimate of 480 square feet per employee.
TABLE D.40. 1978 COMMERCIAL-INDUSTRIAL
FLOOR SPACE ESTIMATES
GREATER ANCHORAGE AREA
Anchorage
Kenai-Cook Inlet
Matanuska~susitna
Seward
GREATER FAIRBANKS AREA
Fairbanks
Southeast Fairbanks
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-Whittier
Million Square Feet
42.3
37.0
3.2
1.5
.6
10.8
10.4
.4
1.0
Because different types of commercial establishments have different
energy use characteristics, it is useful to categorize the commercial-
industrial floor space by type of use. Unfortunately, it is not possible
to do this accurately with the data that is presently available beyond the
general categorization of the preceding Table D.39.
Several studies at the national level and for other regions have
broken the commercial sector into separated categories and have estimated
the annual energy requirements for each on a square foot basis or a per-
employee basis. Table D.41 reveals that no consistent pattern
D-88
Offices
Restaurants
Retail
Groceries
Warehouse
Elementary &
Secondary
Schools
Health
Hotel
Misc.
FEA (1974)b
Offices
Retail
Schools
Hospitals
Other
TABLE D.41.
CATEGORIZATION OF THE COMMERCIAL SECTOR
VARIOUS END-USE ANALYSES
FEA (1977) c
Retail
-food
-restaurants
-office
-other
Wholesale
-warehouse
-non-ref.
Finance-Insurance-
Real Estate
-office
Offices
Dispersed Retail
CBD Retail
Other
State Bldgs.
Health Care
Education
Local Gov't.
Office
PRNLf
Retail-Hhlse.
Educational
Public Admin.
Finance &
Office
Health
Hotel
Warehouse
Religious
Garage
Other
Forestry-Fishing-
Construction
Transportation &
Public Utilities
Wholesale Trade
Retail Trade
·Finance-Insurance-
Real Estate
Services & Gov't.
State Offices t::l -lodging
-laundry I
00
\0 -recreation
-automotive
-outdoor
activities
SOURCE: (a) California Energy Commission, "California Energy Demand 1978-2000: A Preliminary
Assessment," 1979.
(b) Federal Energy Administration, "Residential and Commercial Energy Use Patterns,
1970-1990," 1974.
(c) Federal Energy Administration, National, Energy Information Center, "~nergy Consumption
in Commercial Industries by Census Division-1974," March 1977.
(d) T. Owen Carroll et al. "The Planners Energy Workbook," Brookhaven National Laboratory
and State University of New York Land Use and Energy Utilization Project.
?
\0
0
TABLE D.41. (continued)
(e) New York State Energy Office, "Development of a Commercial Sector Energy Use Model for
New York State," 1979.
(f) Jerry R. Jackson and WilliamS. Johnson, "Commercial Energy Use: A Disaggregation by
Fuel, Building Type, and End Use," Oak Ridge National Labs., 1978.
(g) NEPOL Load Forecasting Task Force and Battelle Columbus, "Report on Model for Long-
Range Forecasting of Electric Energy and Demand to the New England Power Pool," 1977.
in terms of the categorical breakdown utilized by different researchers
either because of data restrictions, different objectives, or different
perceptions and data availability concerning 0nd-use patterns in different
sectors. Most researchers separate office, retail, warehousing, educational,
and health-related buildings as having obviously different energy end-use
characteristics.
The most useful national source of information on energy consumption
in the commercial sector for this study proved to be an analysis done by
the Oak Ridge National Laboratory. Based upon their analysis of end use,
Table D.42 was constructed, and it shows, for various components of the
·commercial sector, both proportion of energy consumed in various uses
and the space heating load relative to office space.
For Alaska, there is scattered information available on energy
and electricity use on a square foot basis, essentially all of it from
public buildings in Anchorage and Fairbanks. This information can be
summarized as follows:
1. Based upon an inventory of 4.3 million square feet of space
used by the Anchorage School District for the fiscal year 1978-1979,
average electricity consumption (none for space heat) was 11.9 kWh/square
foot/year. In addition, based upon 4 million square feet, the average
natural gas consumption for space heat and related uses during the same
period was .143 mcf/square foot/year.
D-91
Commercial Sector
Office
Retail-Wholesale
Garage & Service
-Station
Warehouse
Education
Hospital
Religious
Hotel
Other
TABLE D.42.
ENERGY USE IN THE COMMERCIAL SECTOR
NATIONAL PATTERN
Percent of Fuel Used
For Various Activitiesa
Space
Heat
Water
Heat Lighting Other
78 4 15 4
80 0 13 7
83 0 14 3
84 1 12 3
80 5 12 3
72 11 10 7
87 1 10 2
90 5 4 1
76 3 18 3
Space Heating
Requirement per
Square Foot Relati
to Office
1
1.02
.45
-.45
.84
1.58
.67
1.58
.88
nationally
weighted
average 1.01
aDoes not include fuel used for air conditioning.
SOURCE: Jerry R. Jackson a,nd William S. Johnson, "Commercial Energy
Use: A Disaggregation by Fuel, Building Type, and End Use,"
Oak Ridge National Laboratory, 1978.
D-92
2. Anchorage commercial real estate experts use 2-4¢/square foot/
a rule of thumb for electricity costs in the commercial sector.
between·8 and 16 kWh/square foot/year if electricity
'averages 3¢ kWh. Gas consumption for space heating is also estimated
to cost between .8 and 1.2¢/square foot/month which converts to between
.062 and .093 mcf/square foot/year at a gas price of about $1.56 mcf.
3. Consumption of electricity on the Anchorage Community College
campus in 1975 averaged 20.8 kWh/square foot/year; and for Kenai and
.~· . v
Matanuska-Susitna, it averaged 17.63 kWh/square foot/year.
4. Commercial consumption of electr~city in 1975 in the Fair-
banks Area averaged 12.3 kWh/square foot/year for 552 thousand square
feet of space surveyed. In addition, the 1.6 million square feet of
space occupied by the University of Alaska averaged 17.4 kWh/square
I
24 foot/year. Variation in use among consumers was dramatic, ranging
from about 4 kWh to 50 kWh.
5. A 1978 inventory of Fairbanks schools.indicates·electricity
consumption of 12.4 kWh/square foot/year based on 1.2 million square
feet. In addition, space heating plus water heating requirements for
the schools averaged 136 btu/square foot/year x 103 broken down by
fuel type as follows: oil -103 thousand btu, coal -185 thousand btu,
25 steam -70 thousand btu.
D-93
6. Another Fairbanks survey of public buildings done in 1978
-. indicates that the space heating requirement in two electrically heated
buildings in Fairbanks is 302 thousand btu/square foot/year (88 kWh);
and for four buildings heated by fuel oil, the average is 52 thousand.26
7. A sample of electricity and natural gas consumption of build-
ings used by the Anchorage Borough indicates an average consumption of
electricity of 21.8 kWh/square foot/year. The figure calculated for gas
seemed unreasonably high and so is not reported.
8. The Anchorage International Airport reports electricity use
at 51.8 kWh/square foot/year.
No information is available on the incidence of the use of electrici
as a fuel for space heating in the commercial-industrial sector,
it is generally agreed that it is not significant. Gas supplies the
majority of the load in the Anchorage area, supplemented by fuel oil;
and fuel oil, coal, and steam supply the Fairbanks market. Fuel oil
and propane serve Glennallen-Valdez.
On the basis of this small amount of data, it is not possible to
develop a plausible end use model of the commercial sector•
It is, however, possible to substantiate the assertion that there
is little electric space heating in the commercial sector in the Anchor-·
age region and to develop a rough suggestion of electricity end use in
I D-94
commercial sector. First, we calculate the use of natural gas in
commercial-industrial category. The total of such consumption in
1978 was 6,917,786 mcf~ including apartments which are defined in this
study to be in the residential sector. Netting out this component of
27 gas demand leaves 5,848,922 mcf. Using the school system
consumption figures as a rule of thumb, this amount of consump-
tion could provide space·heat to 40.9 million square feet of space, or
100 percent of the calculated space in the gas service area.
Adjusting the·school district figure upward by the ratio of general
· office-to-education building heat requirements taken from the Oak Ridge
National Laboratory Report (1.5) yields 27.3 million square feet gas
heated. (Natural gas use for process heat and air conditioning is very
limited in the Anchorage market.) Thus, at most, somewhat between zero
and one-third of commercial space heating requirements in the Anchorage
area and within the gas utility service district could have electric
space heat.
Second, we confirm the predominance of gas fo~ space heating by
allocating nonspace heat electricity consumption. Again using the Oak
Ridge National Laboratory (ORNL) information on proportions of fuel us~d
in various activities within a type of commercial building, we calculate
the lighting and miscellaneous electricity requirement in various cate-
gories of buildings relative to education. We then apply those ratios
to the observed electricity consumption figures for electricity in
Alaskan educational buildings. This results in Alaskan estimates of
nonspace heating and cooling electricity consumption, which are sho'vn
in Table D.43.
D-95
TABLE D.43. ESTIMATES OF CO}illERCIAL ELECTRICITY CONSUMPTION
Sector
Education
General
Warehouse
Hotel
Manufacturingc
Electricity Requirement
Relative toa
Education
1
1.5
.53
-~63
1.5
Calculated
Electricity in kWh b
Requiremerit/ft.2/year
13
19.5
7
8
19.5
aBased on Jerry Jackson and William Johnson, "Commercial Energy
Use: A Disaggregation by Fuel, Building Type and End Use," Oak Ridge
National Laboratory, 1978.
bBased on 13 kWh/ft.2 /year assumed for the education sector (a
rough-weighted average of the ayailable. information).
c Assumed to have the same requirement as general commercial space.
D-96
Using this data and the previously calculated information on square
footage, it is possible to estimate total electricity use for nonspace
in the commercial sector and compare it to the actual commercial
sales~ This is done in Table D.44, which a.~so includes comparable cal-
culations for other portions of the railbelt. By netting this use out
of total sales, the amount remaining must be allocated to either space
r
Vheating, water heating, air conditioning, or process electricity.
The 165,128 MWh calculated for Anchorage could, at 36 kWh annually
per-square foot for space heat (based again on the Anchorage School·
.District data converted from gas to electricity and adjusted to cover
general commercial space using the ORNL data) heat about 4.6 million
square feet of Greater Anchorage Area commercial space (about 11 percent).
The comparable figures for Fairbanks and Glennallen would be about 12
and 23, respectively, if the electric heating requirement for Anchorage
is converted on the basis of heating degree days in these other locations.
The electric space heating load is lower than indicated by this
procedure. There are some industrial and miscellaneous users of elec-
tricity which consume much larger than average amounts of energy. The
Anchorage International Airport, the petroleum-related facilities on the
Kenai Peninsula, and the pipeline pump stations in Valdez are examples.
D-97
TABLE D.44. CALCULATING THE POSSIBLE USE OF ELECTRICITY
FOR SPACE HEATING IN THE COMMERCIAL-
INDUSTRIAL-GOVERNMENT SECTOR
Unit
Electricity Total Actual 1978 Space Heat
Floor Consumption Electricity Utility or Process
Type of s3ace (non-space heat) Consumption Sales Electricity a
Area Building (10 ft.2) kWh/ft.2 /year (mWh) (mWh) (mWh)
GREATER ANCHORAGE AREA 721,620 886,748 165,128
Anchorage 618,270 67.7,021 58,751
education 5,000 13 65,000
general 25,120 19.5 489,840
hotel 1,500 7 10,500
warehouse 4,520 8 36,160
manufacturing 860 19.5 16,770
Kenai-Cook Inlet-
~ general 3,200 19.5 62,400 129,483 67,083
I
1.0 ,Matanuska-Susitna (.XI
general 1,500 19.5 29,250 67,835 38,585
Seward
general 600 19.5 11,700 12,409 709
GREATER FAIRBANKS AREA
Fairbanks and
Southeast Fairbanks
. general 10,800 19.5 210,600 27i,726 61,126
GLENNALLEN-VALDEZ AREA
Valdez-Chitina-Whittier
general 1,000 19.5 19,500 28,604 9,104
'
a Difference between total utility sales and calculated non-space heat sales.
ENDNOTES: APPENDIX D
1. Anchorage Real Estate Research Report Vol. III, Fall 1979, p. 44.
Anchorage Real Estate Research Committee.
2. Ibid. , p. 55.
3. FMATS Study of Housing (draft), p. 16.
4. 1970 U.S. Census of Housing, Detailed Housing Characteristics:
Alaska, Table 63.
5. In 1970, there were 2,757 year-round housing units unoccupied but
not for sale or rent. Not all of these are vacation homes. In
1978, the Municipality estimated 729 residences in Girdwood~ of
which 198 were full time.
~. Calculated from census data.
7. In 1970, Homer, Kenai, Seldovia, and Soldotna had 2,093 year-round
housing units according to the census; and in 1978, 2,570 according
to the Kenai Peninsula Borough, Profile of 5 Kenai Peninsula Towns.
Applying this growth rate to the whole census division would yield
only 5,710 un~ts in 1978.
. .
8. Discussions with AMLP personnel.
9. Fairbanks North Star Borough, FMATS Housing Study (draf.t), 1980,
p. 14.
10. U.S. Department of Commerce, Bureau of the Census, Privately Owned
One-Family Homes Completed, 1977, Table 15.
11. Anchorage Real Estate Research Report, Ibid, p. 12.
12. Homer Electric Association, Power Requirements Study 1978, p. C.4.
13. CH 2M Hill, "Feasibility Assessment f9r Hydroelectric Development
at Grant and Crescent Lakes," for Seward Electric Association 1979,
. p. 10.
14. Fairbanks North Star Borough, "Community Information Quarterly,"
Spring 1980, p. 81. This survey may have substantial bias, but
its direction is uncertain.
15. California Energy Commission, "Appendix A: Analysis of Residential
Energy Uses," 1976, Table A.7.2.
16. Midwest Research Institute, "Patterns of Energy Use by Electrical
Appliances," DPRI EA-682 project 576, January 1979, p. 5-3.
D-99
ENDNOTES: APPENDIX D (continued)
17. Midwest Research Institute, Ibid., pp. 5-12; and California Energy
Commission, Ibid.
18. Midwest Research Institute, Ibid., p. 3-2.
19. A partial tabulation is available for Fairbanks through the Municipality
Planning Office in conjunction with the Fairbanks Metropolitan Area
Transportation System (FMATS) Studies.
20. Anchorage Hetropolitan Area Transportation System (ANATS) Studies.
21. Alaska Center for Real Estate Education and Research and University
of Alaska, Anchorage, Department of Real Estate •.
22. National estimates are available in Ide Associates Inc., "Estimating
Land and Floor Area Implicit in Employment Projections," prepared
for U.S. Department of Transportation, Bureau of Public Roads, 1970.
23. Based on studies conducted by Mark Fryer and Associates.
24. Ibid.
25. Ibid.
26. Fairbanks North Star Borough, Community Information Center, "Special
Report 114," July 1979, p. 84.
27. Assume 90 percent of non-electrically heated units are gas (14,064).
Converting the average heat load to gas yields about 76 mcf per unit
or 1,068,864 mcf total apartment gas for space heating demand.
D-100
APPENDIX E. ELECTRIC UTILITY SALES PROJECTION METHODOLOGY
E.l. Residential Nonspace Heating Electricity Requirements
The following appliances are identified in this model:
1. water heater
2. range (cooking)
3. clothes dryer
4. refrigerator
5. freezer
6. dishwasher
7. clothes washer
8. television
9. air conditioner
10. small appliances
Time is measured in f1ve-year increments beginning in 1980, and there
are three separate regions: Greater Anchorage, Greater Fairbanks, and
Glennallen-Valdez.
The electricity requirement for appliance type j at time t for
region r (the r notation is dropped in all that follows) (REQ ) is j, t,r
the product of five components. It begins with the number of households
(HHt) multiplied.by the appliance saturation rate (sj,t). This yields
the total n~~ber of appliances of type j (A. t). The portion of those
J'
appliances which use electricity is determined by the fuel mode split
for electricity (FMS. t). The result is the number of electric appli-J,e,
ances which is multiplied by average annuai consumption (KWH. ) to yield
J, t .
preliminary totalfconsumption. This is finally adjusted by an average
household size adjustment (AHS. ) to yield total consumption. The ],t
equa~ion is as follows:
REQ.
J. t
HH * S * FMS . * KHH ;, AHS t j,t J,e,t j,t j,t (E.l)
Total electricity requirement is the sum over all appliances j.
L REQ. j -J, t (E. 2)
The number of households (HH ) is determined by the demographic model. t
The saturation rates (S. ) are exogenously entered. The average house-J,t
hold size adjustment (AHS. t) applies to the following appliances only:
J •
clothes washer, water heater, clothes dryer. For these appliances, the
adjustment is the ratio of average household size at t to average house-
hold size in the base year, 1980. For other appliances, it is one.
The fuel mode split (FMS. ) is the proportion of appliances of J,e,t
type j which use electricity.
FMS. J,e,t
A. ],e,t
A. t J.
(E. 3)
The number of appliances of type j in period t which use electricity is
a function of the previous stock of such appliances (A. 1), the pro-J , e, t-
portion of those appliances which are scrapped (d. 1 ), and the number J,e,t-
of new appliances of type j purchased at time t which utilize electricity
(NA. t). J,e,
At this point, it is useful to distinguish different appliance
"vintages," that is to identify the time period during which an appliance
E-2
was manufactured (and sold). This is important because appliances of
different vintages may have inherently different energy-use character-
istics. This identification is accomplished by defining the appliance
stock at any time t in terms of the initial stock and additions to the
stock during each subsequent time period. Each vintage or appliances (m)
may have a different scr~pping rate (d~ ) between time m and time t J,e,t
because of different characteristics. The scrapping rate of the initial
stock (d~980 ) will differ from that of subsequent vintages because the J,e,t
initial stock is composed of appliances of different ages and, thus,
vintages.
For the existing appliance stock, the scrapping rate is also a
function of the past grmv.th rate of the stock. Specifically, if gk is
the historical growth rate of the stock and ex.k is the average lifetime,
J
then the scrapping rate in any year y can be approximated by:
* (l+g.)y-1980
. J
(E.4)
This equation captures the phenomenon that the observed annual scrapping
rate for a growing appliance stock will be less than (1/average lifetime)
and that for a declining appliance stock the scrapping rate will be greater
than .(1/average lifetime).a
.,.aln practice, it was not possible to utilize this function. The
scrapping rates for the existing appliance stock were based upon a cal-
culation using the scrapping rate for new appliances extrapolated back-
wards and weighted by the post-1970 growth rate of the stock.
The stock of electrical appliances of type j at time t (A. ) is· J,e,t
thus as follows:
A. J,e,t A * (l-d:980) + NAJ.,e,l985 * (l-d:985) j,e,l980 J,e,t J,e,t
+ • • • + NA. 5 * (1-d :--5 ) + NA .. J,e,t-J,e,t J,e,t
The number of additions to the appliance stock j at time t using
electricity is the product of the total number of new appliances at
time j (NA. ) and the incremental (or marginal) electrical mode split J,t
(msi.' t) • This equation is simply, J,e,
NA. = NA. t * msi. J,e,t J, J,e,t (E.
The number of new appliances of type j at time t is the difference
betlveen the stock demanded at time t, represented by the number of house-
holds multiplied by the saturation rate, and the supply of those appli-
ances which is the stock from the previous period net of scrapping.
Thus, it is necessary to consider appliances using all types of fuels
in calculating new appliance type j requirements at t. This demand can
also be stated in terms of the initial appliance stock and all subsequent
additions to the stock as follows:
E-4
NA. =A. ],t ],t
E A * (1-dl980 )
k j,k,l980 j,k,t
E NA. k 1985 * (l-dl985 ) -
k J, ' j, k, t
-E NA. * (l-d:~5 ) k J,k,t-5 J,k,t (E. 7)
The average annual electricity requirement in kilowatt hours of
appliance type j at time t (KWH. t) is a function of the age distribution
J'
of the appliance stock and the electricity requirement for each vintage.
Four factors are involved in the determination of the electricity re-
,quirement for each vintage. First, there is an average electricity
requirement for the existing appliance stock (kwhj,1980). Second, there
is an average electricity requirement for new additions to the stock in
year t (kwh. t). Third, the average size of additions to the appliance J,
stock j may change. This is accounted for by a growth rate on average
appliance size subsequent to 1985 (l+kwhg.). Finally, mandatory improve-
]
ments in appliance efficiency may reduce average electricity requirements
for new vintages of appliances. These Federal conservation targets
(cs. t) are implemented beginning in the interval 1980 to 1985. The
J'
average consumption of appliance j at time t can thus be expressed as
the following:
E-5
KWH. t J, kwhJ.,l9.80 * [A * (l-d:980 )]/A. + j,e,l980 J,e,t J,e,t
0 (l+kwhgj) * (l-csj,1985 ) * kwhj,1985 *
[NAJ.,e,l985 * (l-d:985 )]/A. + .•• J,e,t J,e,t
(l+kwhg.)t-1985 * (1-cs. t) *kwh. 1985 *
·J .J, J,
[NA. ]/A. J,e,t J,e,t
A major portion of electricity use in dish and clothes washer opera-
tion is for water heating. We account for this by separately calculating·
this component of electricity use in these appliances. The estimate of
the use of electricity for water heating for these appliances is the
product of the water heater saturation rate, the electric mode split,
and the washer saturation rate.
Small appliances are not differentiated. Thus, their electricity
requirement is simply the product of the number of households, the ini
consumption level (~~1980 ), and a growth factor [nkwh *. (t-1980)].
*KWH. t * [nkwh * (t-1980)] J,
Electric light consumption is assumed to be a simple multiple of
the current consumption level per household and the number of households.
E-6
E.2. Residential Space Heating Electricity Requirements
Total residential space heating electricity requirements in region r
(not indicated in the algebraic notation for ease of exposition) at
time t (SHREQ ) is composed of the requirements calculated separately e,t
for four different types of structures. These are the following:
1. single family
2. duplex
3. multifamily
4. mobile home
The electricity requirement for each type of structure j (SHREQ. ) is J,e,t
based upon the required heat load measured in btu's 1TOTBTU. ), the J,e,t
conversion efficiency of electric devices for producing space heat
(EFF. t), and the conversion constant between btu and kHh. J,e,
Algebraically, it is
SHREQ. t J,e, TOTBTU. t/EFF. t/CONV J,e, J,e, e (E.lO)
In the present use of the model, the conversion efficiency is assumed
constant throughout (no utilization of heat pumps), so the energy require-
ment can be calculated as kHh throughout.
The space heating requirement for structu~e type j can be further
defined as the product of the number of housing units of type j (HT.),
J
the proportion utilizing electric space heat (FMS. ), the average J,e,t
level of consumption (KHH. t), and the utilization rate (UT. t) • . J, J,e,
E-7
SHREQ. = HT. * FMS . * KWH. * UT . J,e,t J,t J,e,t J,t J,e,t
The number of units of housing type j in year twill be a proportion.
(HMSj,t) of the first home housing stock in year t (FHUt).
The proportion of housing units of type j at year t using electric
space heat will be simply
FMS . = HT. t/HT. J,e,t J,e, J,t
The number of housing units of type j using electric space heat at
time t will be a function of the initial stock of such installations
(HTj,e,1980) plus additions to the stock over time (NHTj,e,t) and net
of scrapping from the stock over time (d~ . ) . The r.ate of scrapping J,e,t
is vintage specific for the reasons mentioned in the appliance stock
discussion. Thus, the electric space heat units for structure type j
at time t can be written as follows:
+ . . . + NHT. 5 * (1-d~-5 ) + NHT. J,e,t-J,e,t J,e,t
E-8
(E.
New electric heating units for structures of type j at time t are
determined by the electric incremental mode split (msi. t) applied to
J 'e,
all purchasers of space heating appliances for housing units of type j
at timet (NHT. ), J,t
NHT. = NHT. t * msi J,e,t J,. j,e,t
New space heating requirements for structures of type j at· time t
(E.l5)
are determined by the difference between the total number of housing units
type j required at time t (HT. ) and the existing stock w~th space ],t
heating appliances. We assume the scrapping rate for the housing stock
is zero but that there is a scrapping·rate for space heating appliances
using different fuel types k. Thus, new space heating appliance demand
in housing units type j at time t is as follows:
NHT. = HT. ],t J,t r HT. k 19.80 * (l-d:980 ) -1: NHT. k 1985 * (l-d:985 ) k J, , J,k,t k J, , J,k,t
t-5
1: NHT. k t-5 * (1-dJ.,k,t) k J, ,
Finally, the electricity requirement per unit (KWH. ) will be ],t
the weighted average of .the per unit requirement of space heating
appliances of all different vintages of the stock. The average per
unit electricity requirement for any vintage will be the product of
(E.l6)
the base year requirement (kwhj,1985 ), the growth rate in basic unit
consumption (b.rhg.), the mandated energy savings for the unit (1-cs. t), J J,
E-9
and the energy savings as a result of system retrofitting (1-ret~ ) J ,e, t
in the interval between m and t. As with appliances·, a difference between
average existing and new unit energy consumption is recognized so that
there is a separate and different unit consumption parameter for the
existing stock (kwh. 1980). The average consumption calculation is
. J ,
thus as follows:
KWH = (1 tl980 ) * k h * [HT * j,t -re j,e,t w j,l980 j,e,l980
(l-d~980 )]/HT. + (l-ret~985 ) * J,e,t J,e,t J,e,t
* kwhj,l985 * (l+kwhgj)o * [NHTj,e,l985 *
(l-d~985 ) ] /HT. + . . . + (1-cs. ) * kwhJ. , 1985 * J,e,t J,e,t J,t
(l+kwhg.)t-l985 * NHT. /HT. J J,e,t J,e,t
(E .17)
E-10
E.3. Commercial-Industrial Electricity Requirements
Commercial-industrial-government electricity requirements are aggre-
_gated into a single category because of data limitations. Total require-
ments at time t in region r (dropped from equation for ease of exposition)
for these combined sectors (CIREQt) is the product of nonagricultural wage
and salary employment (EMt) and average consumption per employee (CIKWHt).
(E.l8)
Average consumption per employee varies as a function of time and
the implementation of conservation standards. Specifically, new or
incremental electricity users (EMt-EMt_5 ), who represent new commercial-
industrial-government floor space, will have different electricity require-
ments than existing customers. Thus, existing customers at the beginning
of the projection will have an average consumption rate (kwhl980) different
from incremental customers added in subsequent five-year intervals. A
different consumption rate is assigned to incremental customers (kwh 1985 )
and this consumption rate grows over time linearly at a fixed amount
(nkwh). Efficiency standards at timet (l-est) are effective on incre-
ments to electricit~ requirements but not to the total.
The general equation for the commercial-industrial-government load
can then be written as follows:
E-ll
* (l-csl985) + (EM1990 -EM1985) * [kwhl985 + nkwh]
(E .19)
E.4 . Miscellaneous Electric Utility Sales
Miscellaneous sales consist of two very small components of total
sales: street lighting and recreational homes.
Street lighting sales (SLREQt) is a fixed percentage (sl) ·of the
total of residential and commercial-industrial-government sales.
sl * [REQ + SHREQ + CIREQ ] t t t
(E. 20)
Similarly, recreational home consumption (RECREQ ) is calculated as
t
a fixed level of consumption (kwh) multiplied by a fixed proportion of
households (rec).
RECREQ
t HHt * rec * kwh
E-12
(E. 21)
APPENDIX F. ELECTRIC UTILITY SALES
PROJECTION MODEL PARAMETERS
F.l. Projection Model Parameters for Base Case
F.l.A. RESIDENTIAL NON-SPACE HEATING ELECTRICITY REQUIREMENTS
Parameters used in this model are presented in Tables F.l and F.2.
F.l.A.l. Conservation
The Energy Po·licy and Conservation Act of 1975 and the National
Energy Conservation Policy Act of 1978 direct the Federal Energy Admin-
istration (Department of Energy) to promulgate energy efficiency targets
for electrical appliances sold beginning in 1980.
The targets are based upon an aggregate 20 percent increase in
the efficiency of energy use for 13 appliances using 1972 as a base
year. The targets for the various appliances differ because each is
determined on the basis of economic and technical feasibility.
For example, the efficiency improving changes in refrigerator
design include improved compressor motor efficiency, improved insula-
tion, improvement of door seals, provision of an on/off switch for anti-
' sweat heaters and elimination of the condenser fan motor. These changes
were estimated to increase the retail price of the refrigerator by half
the cost of electricity saved in the first year of operation.1
The efficiency targets are presented in Table F.3.
F.l. MODEL PARAMETERS: RESIDENTIAL NON-SPACE HEAT APPLIANCES
Parameter Region
Greater Greater Glennallen-
Anchorage Area Fairbanks Area Valdez Area
Saturation Rates (S. ) J,t
SWH 1980 .99 .97 .91
1985 ·1.00 .99 .94
1990+ 1.00 1.00 1.00
sc 1980+ 1.00 1.00 1.00
SCD 1980 .71 .67 .49
1985 .72 .69 .52
1990 .73 .71 .54
1995 .74 .72 .56
2000 .75 .73 .58
2005 • 76 .74 .60
2010 .77 .75 .62
SR 1980+ 1.00 1.00 1.00
SF 1980 .46 .45 .46
1985 .48 .48 .49
1990 .51 .51 .52
1995 .52 .53 .54
2000 .55 .55 .56
2005 .57 .57 .58
2010 .58 .59 .60
SDW 1980 .49 .38 .15
1985 .54 .44 .24
1990 .59 .50 .32
1995 .63 .55 .39
2000 .67 .60 .45
2005 .71 .64 .51
2010 .74 .68 .• 56
KEY: WH Water Heater DWW Dish\vasher \-later
c Cooking cw Clothes Washer
CD Clothes Dryer cww Clothes Washer Water
R Refrigerator TV Television
F Freezer AC Air Conditioner
DW Dishwasher
F-2
(Continued) '
Region
Greater Greater Glennallen-
Anchorage Area Fairbanks Area Valdez Area
Saturation Rates (S. t)
J'
sew 1980 .77 .75 .66
1985 .78 .76 .68
1990 .79 .77 .70
1995 .80 .78 .72
2000 .81 .79 T' • J
2005 .82 .80 .74
2010 .83 .81 .75
STV 1980 1.50 1.51 .85
1985 1.55 1.56 1.00
1990 1.60 1.60 1.10
1995 1.64 1.64 1.19
2000 1.68 1.68 1.27
2005 1.71 1.71 1.34
2010 1.74 1.74 1.41
SAC 1980+ 0 .01 0
Incremental Electrical
Appliance Hode Split (msi. ) J ,e, t
ms~ 1980+ .35 .5 .4
msic 1980+ .66 .85 .4
msiCD 1980+ .90 .98 .75
msi (other)
1980+ 1.0 1.0 1.0
F-3
F.l. (Continued)
Parameter
Greater
Anchorage Area
Average Annual Appliance
Consumption in 1980 (kwhj,1980)
WH
c
CD
R
F
DW
DWW
cw
cww
TV
AC
Average Annual New
Appliance Consumption (kwhj,1985 )
WH
c
CD
R
F
DW
DWW
cw
cww
TV
AC
F-4
Region
Greater
Fairbanks Area
3,475
1,.200
1,000
1,250
1,350
230
700
70
1,050
400
400
3,650
1,250
1,000
1,560
1,550
230
740
70
1,050
400
400
-(Continued)
Greater
· Anchorage Area
Conservation Target
for New Appliances (csj,1985)
WH
c
CD
R
F
DW
cw
TV
AC
Grmvth in
Appliance Size (kwhg.)
J
WH
c
CD
R
F
DW
cw
TV
AC
F-5
Region
Greater
Fairbanks Area
.14
.03
.06
.29
.21
.18
.29
.32
.21
.005
0
0
.01
.01
.005
0
0
0
Glennallen-
Valdez Area
F.l. (Continued)
Parameter
Average Lifetime
of Appliance (ex.)
J
WH c
CD
R
F
DW
CT.V
TV
AC
Historical Electric
AEEliance Stock
Growth Rates (gj)
WH
c
CD
R
F
DW
cw
TV
AC
Greater
Anchorage Area
.05
.05
.06
.05
.05
.09
.06
.07
0
F-6
Region
Greater
Fairbanks Area
10
10
15
15
20
10
10
10
10
10
.03
.03
.04
.03
.03
.04
.04
.04
.03
Glennallen-
Valdez Area
.15
.10
.14
.10
.11
.25
.12
.20
0
TABLE F.2. MODEL PARAMETERS: SMALL RESIDENTIAL APPLIANCES
Parameter Region
Greater Greater Glennallen-
Anchorage Area Fairbanks Area Valdez Area
Average Annual Con-
sum:etion Level (KWh
1980)
electric lights 1,000 1,000 1,000
assorted appliances 1,010 1,466 1,333
Annual Increment to
Small A:epliance Con-
sum:etion (nK\.fu) 50 70 70
F-7
TABLE F.3. FEDERALLY-MANDATED ELECTRICAL APPLIANCE EFFICIENCY
Electrical Appliance
1.
2.
3.
·4.
5.
6.
7.
8.
9.
10.
Refrigerator
Freezer
Dishwasher
Clothes Dryer
Water Heater
Air Conditioner
Television (black & white)
Television (color)
Range
Clothes Washer
Percent Reduction in Energy
Consumption of New Appliance
Using 1972 as a Base
32
23
20
7
15
23
65
35
3
32
SOURCE: Federal Energy Administration, Energy Conservation Program
for Appliances. Federal Register, Vol. 42, No. 136.
F-8
Independent studies of the potential for conservation of electricity
in appliance design conclude that much greater savings in energy is pos-
sible at modest cost. For example, a Danish study in 1979 concluded that
a 64 percent savings in el~ctricity consumption could be-obtained from
refrigerator design changes with a payback time of five years. The
design included increased insulation thickness, elimination of automatic
defroster, and increased condenser efficiency.2 More radical design
changes could further reduce the electricity consumption of refrigerators
and other appliances.
This study assumes that the Federal guidelines will be implemented
during the period 1981 to 1985.3 The target efficiencies are reduced
by 10 percent to take account of the fact that the base consumption
levels from which the targets are calculated are those of 1972; and in
the interim betw·een 1972 and the present, some efficiency improvements
have already appeared in new appliances. For example, the primary
design improvement for television sets is a conversion to solid-state
circuitry. This is already happening and, consequently, the applica-
tion of the target value to the 1980 stock would overestimate the
1 . 4 actua energy sav1ngs.
The rationale for assuming implementation of the Federal guidelines
but no additional changes in appliance efficiency is based on the
idea that these are improvements which will, in fact, occur, while
further improvements will require addition-specific programs at the
F-9
Federal or state level~ Should they occur, they can be subsequently
introduced into the model.
F.l.A.2. Appliance Lifetime
Appliance lifetime estimates are available from the Home Appliance
Manufacturers Association. Relevant appliance lifetimes are presented
· in Table F.4.
TABLE F.4. APPLIANCE LIFETIMES
Appliance Lifetime in Years
Freezer 20
Refrigerator 15
Electric Clothes Dryer 14
Electric Range 12
Television (color) 12
Dishwasher 11
Clothes Washer 11
Television (black & white) 11
Room Air Conditioner 10
Electric Water Heater 10
SOURCE: Richard B. Comerford, "PSE&G Method of Forecasting
Kilowatthour Consumption," and George L. Fitzpatrick, "Peak
Forecasting Methodology," in How Electric Utilities Forecast:
ERPI Symposium Proceedings, March 1979.
Other estimates of appliance lifetimes indicate some variation
these figures, although it is not substantial.5
F-10
In this analysis, since appliances of different vintages have differ-
ent energy-use characteristics,_it is important to identify not only the
-
average lifetime but also the probability distribution of lifetimes for
specific appliances. One recent study which has investigated the dur-
ability probability distribution for air conditioners concluded that it
is a Weibull distribution which has a mean of approximately ten years
with 50 percent of the population wearing out in the interval between
. 6
4.5 years and 13.75 years.
We assume the same probability distributions for the durability of
other electrical appliances and use a simplified distribution to dis-
tribute appliance· disposals around the mean lifetime. Since the model
calculates appliance stocks on a five-year interval, appliance disposals
are assumed to occur in the five-year interval preceeding and the ten-
year interval succeeding the average lifetime which is adjusted to be a
multiple of five years. The typical distribution is shown in Figure F.l.
FIGURE F.l. PROBABILITY DISTRIBUTION FOR TYPICAL
APPLIANCE DURABILITY
%
Scrapped
25%
Average
Lifetime
-5 Years
40%
Average
Lifetime
Years
F-11
25%
Average
Lifetime
+5 Years
10%
Average
Lifetime
+10 Years
F.l.A.3. Appliance Capacity Growth Rates
The capacity of the existing appliance stock and consequently the
average consumption level for those appliances is discussed in Appendix
D, Section 5.
The capacity of many major appliances has grown over the years.
This contributes to higher electricity consumption. For example, it
has been estimated that the capacity in cubic feet of the average refrig-
erator sold in the United States has increased from 8.5 in 1950 to 12.0
in 1960 and to 14.0 in 1970.7
Analysis of past Alaskan electricity consumption patterns indicates
that this phenomenon is occurring here and presumably will continue to
occur in the future.8
This growth is a function of both increasing incomes and technological
change. The latter factor is obvious in the example of refrigerators
because in 1950 there were no refrigerators sold with a capacity of
15 cubic feet, while by 1970 the average was that size.9
The following appliances are assumed to maintain a constant capacity
over time in terms of energy consumption:
1. Range
2. Clothes washer
3. Clothes dryer
4. Television sets
5. Room air conditioners
F-12
There is no direct information available on changes in the average
capacity of appliances in use in Alaska, and the information available
nationally is not necessarily applicable to Alaska because of differences
in existing stock as well as preferences. Some general assumptions may·
be made, however, for those appliances which may be subject to capacity
change.
Dishwashers may include two features which affect electricity con-
sumption for each load. These are the "pots and pans" feature which adds
about 1 kWh per load and a "no-heat dry" feature which reduces consumption
10 by about .4 kWh per load. There is no current information on the
utilization of these features, although the Department of Energy appli-
ance efficiency guidelines include the "no-heat dry" feature as one of the
components of their efficiency target. It is assumed here that the "pots
and pans" feature becomes more commonplace over time on new dishwashers
but is used relatively infrequently so that the average capacity growth
rate is .5 percent annually for new dishwasher purchases. The average
new dishwasher is assumed to use 1.05 times the electricity of the
existing stock based roughly on this growth rate and our estimate of
average appliance life.
Refrigerator·volumes have been increasing over time as indicated
above, and with increasing volumes has come increasing energy consumption.
The inclusion of a freezer and a "no-frost" feature also add to average
consumption rates. Elimination of these features is not suggested in
the Department of Energy appliance efficiency standard targets.
F-13
Nationally, the percentage of refrigerator sales by type for
selected years is as follows:
TABLE F.S. DISTRIBUTION OF DOMESTIC REFRIGERATOR
SALES FOR SELECTED YEARsll
(percentage)
Single-Door Two-Door Freezer
Manual Partial Non-Frost Side-By-Side
1950 100 0 0 0
1963 39 28 33 0
1966 23 20 so 7
1973 13 18 49 20
Using annual refrigerator sales volumes, the growth rate in the
size of the average volume of refrigerators sold was 3.6 percent in the
. 12 1950s, 2.3 percent in the 1960s, and 1 percent in the early 1970s.
From this information, it is apparent that the trend nationally is
for some moderation in the growth rate of refrigerator volumes with a
simultaneous shift toward refrigerators with more energy-using features.
Alaska is assumed to be experiencing the same trends with a resultant
increase in the energy requirements of new refrigerators.
The Californi~ Energy Commission's end-use model assumes that the
average electricity consumption of new refrigerators is 1,609 k\Vh annually
in 1974; while for the existing stock, it is 1,235 k\Vh annually, a ratio
F-14
of 130 percent. It also presents information indicating that the growth
rate in annual energy consumption of new refrigerators was 3.7 percent
13 between 1960 and 1974 and 2.2 percent between 1970· and 1974.
On this basis and the assumption that the average age of the
appliance stock is less in Alaska than nationally, it is assumed that
new refrigerators operate at 1.25 times the consumption rate of the
existing stock and that the growth rate in consumption of the new stock
is 1 percent annually.
Freezers vary in size, shape, and whether they have a "no-frost"
feature. The "no-frost" feature adds about 50 percent to average
1 . . . 14 e ectr~c~ty consumpt~on. As with refrigerators, elimination of this
optional feature is not a stated component of the appliance efficiency
standards targets of the Department of Energy. There is no historical
series nationally on the proportion of sales of freezers which are of
the "no-frost" type and no data in Alaska on the characteristics of the
existing stock.
Because of this lack of data, the most reasonable assumption would
be to assume the same growth characteristics for freezers as for refrig-
erators; that is, the new freezers will be more likely to have a larger
capacity and the 11 frost-free" feature than the existing stock. However, it
is assumed that the standard deviation around the mean of the size distri-
bution of existing freezers is smaller than for refrigerators so that the
ratio of the average consumption of the new-to-existing stock will be
smaller than in the case of refrigerators. It is assumed to be 1.15.
F-15
Water heater size has increased over time, as indicated by manufac-
turer shipments of domestic gas water heaters,15 which grew in capacity
by about 1 percent annually during the mid-1970s. Since energy require-
ments will not grow as fast as volume and because average per capita
residential hot water consumption is relatively constant except for use
in dishwashers and clothes dryers, continued growth in water heater
energy use should be small. However, because the existing stock may
.continue to be replaced by larger units, a small positive growth rate
is assumed for energy use of additions to the stock. The average con-
sumption of new units is .assumed to be 1.05 times the existing stock
based upon this assumption and on estimate of the age of the existing
stock.
Small appliances. Electricity consumption from the use of small
appliances will increase as additional appliances are purchased by the
average household. Individually, such appliances do not constitute a
large proportion of total demand, but the combined consumption of
electricity through such appliances could continue to increase as new
appliances, some not even now on the market, become available and are
purchased.
It is difficult to specify a growth rate for electricity consump-
tion through these appliances because there is no information available
on existing saturation rates and use patterns in Alaska and because of
the following considerations:
F-16
1. Use of electricity in some new appliances may substitute
for electricity use in existing appliances. For example,
increased use of microwave ovens may partially reduce
electricity use for conventional ranges.
2. Use patterns for smaller appliances may ch~nge significantly
in the future. For example, a more dispersed population
would result in greater use of electricity in water pumps
to bring well water to the surface.
3. It is not possible to anticipate all future uses of electricity
in the home. Humidifiers, large-screen televisions, and
trash compactors are examples of recent additions to small
appliances in use in the residential sector.
An annual increase of 5 percent of the 1978 consumption level for
small appliances is assumed for future growth. The base figure used in
this calculation varies between the regions because of different climate,
preferences, and other unidentified factors. These differences are
assumed to persist.
The average household use of electricity for lighting is assumed
to remain constant over time.
F.l.A.4. Appliance Saturation Rates
Deviation of appliance saturation rates is discussed in Appendix
D, Section 3.
F-17
F.l.A.5. Incremental Mode Split
To calculate incremental mode splits for water heaters, ranges, and
clothes dryers, we rely upon the same sources of information used in
the development of the 1978 end use inventory. We begin by comparing
the average mode split reported in the 1970 Census (Table D.29) to the
incremental mode split calculated for the period 1960 to 1970 (Table
D.32). When the average and incremental mode splits thus calculated
are approximately equal, the market for the appliance is in equilibrium
with respect to the fuel types used. Unfortunately, it was not gen-
erally the case that such an equilibrium could be identified.
Table F.6 shows the incremental mode splits used in the model.
They remain constant throughout the projection period on the assumptions
: '~
that the relative prices and availabilities of the various fuels will
not change and that preferences for various fuels does. not change.
The Anchorage splits have been relatively constant historically
with only electric water heaters losing ground to gas. The census~
calculated incremental mode splits are utilized.
The census-reported information for the Matanuska-Susitna Borough
is not useful because of the rapid subsequent growth. there which has
relied heavily on electricity. We calculate the water heater and
range incremental mode splits on the assumption that the price advan-
tage. enjoyed by electricity over fuel oil will persist and the majoritY
of purchasers will choose electricity. Electric ranges will be slightlYf'
F-18
TABLE F.6.
INCRE~ffiNTAL ELECTRICAL APPLIANCE MODE SPLIT
Water Heater Range Clothes Dryer Refrigerator
Anchorage ~30a .67a .98a 1.00
Matanuska-
.75d .sod .96e Susitna 1.00
Kenai-
.soa .4ob .90c Cook Inlet 1.00
Seward .35e .75£ .70e 1.00
Fairbanks .sog .85g .98g 1.00
Glennallen-.40h .40h .soe 1.00
Valdez
aCensus calculated incremental mode split·.
bEA . H survey est~mate.
cAssumes a shift away from gas toward the pattern observed for
Anchorage.
dBased on price advantage of electricity.
e 1970 Census.
£Assumes shift toward electricity.
gBased on growth since 1970.
hBased on shift toward electric range preference.
F-19
more preferred than electric water heaters. The electric clothes dryer
mode split is taken from the 1970 Census.
In Kenai-Cook Inlet, we utilize the incremental mode split calcu-
lated from the census for water heaters reflecting a shift in preference
toward electricity. The electric range split is taken from an end use
survey conducted by Homer Electric Association in 1977 because the
census figures appeared low based upon the general pattern of growth
since 1970. The clothes dryer mode split presumes a shift away from
natural gas toward electricity in a reflection of Anchorage preferences.
For Seward, the 1970 Census data was used to calculate water heater
and clothes dryer mode splits while a shift toward a preference for
electricity for cooking was assumed on the basis of cleanliness and con-
venience.
For Fairbanks, the 1970 water heater electric mode split was 37
percent and the end use inventory calculated a 43 percent split in 19~8.
The incremental split over the interval would thus be about 50 percent.
We assume this for future projections, although the recent electricity
price increases might result in a shift in preference back to fuel oil
in the future. Electricity is preferred for cooking in Fairbanks based
upon the census-calculated incremental mode split which,shows a substan-
tial electric range retrofit between 1960 and 1970. We assume a continua
tion of the trend toward electric ranges with an 85 percent incremental
mode split. For clothes dryers, we use the 1970 Census information.
F-20
The Glennallen-Valdez census data is out-dated because of the
rapid post-1970 growth, but subsequent information is not cl,lrrently
available. We assume a shift toward electrical appliances occurs in
reflection of trends observed elsewhere in the railbelt. The clothes
drying mode split is based on the 1970 Census.
F.l.A.6. Household Size Adjustment Factor
Clothes washers, clothes dryers, and water heaters are used more
intensively by larger households. A study conducted by the Midwest
Research Institute calculated average annual electricity requirements
for these appliances as a linear function of household size using
d 1 . 16 metere app ~ances. These equations are converted to use in this
model by:
1. annualizing them (they are based on daily consumption);
2. normalizing them by a 1980 average household size of three
persons; and ·
3. calculating a ratio by which to adjust calculated consump-
tion to account for changing household size.
The adjustment factor is a function of the ratio of average household
size in year t to 1980 (AHHt) and is formed from the equations of
Table F.7.
F.l.B. RESIDENTIAL SPACE HEATING ELECTRICITY REQUIREMENTS
Parameters used in the residential space heating model are pre-
sented in Table F.8.
F-21
TABLE F.7. EQUATIONS TO DETERMINE HOUSEHOLD SIZE
ELECTRICITY CONSUMPTION ADJUSTMENTS
Appliance
Clothes washer
Clothes washer water
Clothes dryer
Water heater
Equation
1 * AHH
.25 + .75 * AHH
• 25 + . 7 5 * AHH
.51 + .49 * AHH
F-22
F.8. MODEL PARAMETERS: RESIDENTIAL SPACE HEATING
Parameter
Greater
Anchorage Area
.·· Average Annual
Unit Consum:etion
(Existing Units) (kwh.
J' 1980)
SF
DP
MF
:MH
Average Annual
Unit Consumption
. (New Units) (kwhj ,1985)
SF
DP
MF
:MH
Growth in Unit Size (kwhg.)
J
SF
DP
MF.
:MH
Average Unit lifetime (ex.)
J
SF
DP
MF
MH
KEY: SF Single Family
DP Duplex
MF Multifamily
:MH Mobile Home
36,500
24,200 .
17,100
27,300
40,100
26,600
18,800
30,000
F-23
Region
Greater
Fairbanks Area
48,200
31,900
21,200
36,900
53,000
35,100.
23,300
40,600
.01
.01
.01
.01
20
20
20
20
Glennallen-
Valdez Area
33,300
21,900
14,600
25,400
36,600
24,100
16,100
"27,900
F.8. (Continued)
Parameter
Greater
Anchorage Area
Incremental Electrical
Applinnce Mode Split (msi. ) -J,e,t
msiSF,l980+
ms~P,l980+
ms~,l980+
ms\m,l980+
Conservation Target
for New Appliances (csj,1985 )
SF
DP
MF
MH
Utilization Rates (UT. ) J ,e, t
UTSF,l980+
UTDP,l980+
UTMF,l980+
UTMH,l980+
.19
.19
.19
.19
F-24
Region
Greater
Fairbanks Area
.01
0
0
.01
.05
.05
.05
0
1
1
1
1
F.8. (Continued)
Parameter Region
Greater Greater Glennallen-
Anchorage Area Fairbanks Area Valdez Area
Retrofitting Coefficients
m (ret. t) J ,e,
. 1980
retSF,l985 .02 .04 .03
1980
retDP,l985 0 0 0
1980
retMF,1985 0 0 0
1980.
retMH,l985 0 0 0
F-25
F.l.B.l. Conservation
Conservation enters into the determination of the residential
space heating load through assumptions about retrofitting of existing
units with energy saving improvements and the application of efficiency
standards to new housing units.
Retrofitting existing structures to reduce the required heat load
will be a function of the quality of the housing stock~ the expected
length of housing unit ownership, the amount of information available
to individuals interested in retrofitting, and the cost of fuel saved
compared to the investment in supplies and labor required to do the
retrofit.
Several federal and some current state programs are designed to
stimulate retrofitt~ng in the residential sector. Among the important
federal programs are a tax credit of 15 percent of the first $2,000
expended for conservation measures, a home improvement loan program
for energy conservation measures, and a weatherization grant program
f 1 . f .1. 17 or ow ~ncome am~ ~es.
The most important state program is a 10 percent residential fuel
conservation tax credit for capital improvements to reduce the heat
load of residential buildings. Several studies have attempted to
assess the impact of retrofitting on energy requirements for space
heating. In 1974 a study by Arthur D. Little estimated that nationally-
applied retrofitting measures based upon current reasonable technology
F-26
and cost could reduce the electric space heating load by 26 percent,
20 percent, and 17 percent for single family, multi-family, and mobile
h . . 1 18 orne type un1ts, respect1ve y. A 1977 study estimated at 20 percent
savings in energy consumption from retrofitting 20 million single family
units· bet"t:veen 1977 and 1990.19
. Unfortunately, these estimates are not based upon actual,observed
human behavior but, rather, are based upon simple engineering models. A
study reported by the California Energy Commission indicates that the
actual response to retrofit conservation programs and actual energy
savings may be only about 50 percent of what engineering models would
20 suggest.
The only information currently available concerning retrofitting
of the housing stock is available from state tax returns for 1977 and
1978. The number of returns, percentage claiming credits, average
credit claimed, and implied value of capital investment in retrofitting
equipment are shown in Table F.9.
Since this data is not regionally divided and specific fuel used
is not specified, it is not possible to accurately estimate the impact
of this program on electric space heat consumption.
If we assume an even distribution statewide, an even distribution
among all types of.fuels, and a 5-year payback for investments (with
no discounting), then in 1978 about $975 thousand was spent for railbelt
F-27
TABLE F.9. STATE OF ALASKA RESIDENTIAL FUEL
CONSERVATION CREDITS
1977
tax returns 195,394
percent claiming credit 5.6%
number claiming credit 10,942
average credit $74.10
;
implied capital investment
(@ 15% credit) $741
implied total capital investment $8,108,070
1978
183,725.
8.1%
14,882
$57.61
$576
$8,571,873
Source: Alaska Department of Revenue, "Fuel, Conservation, and
Industrial Credits Relative to the Individual Income
Tax," 1980.
F-28
electric space heat unit retrofits. This saved $195,000 in electric
bills. If the average cost of electricty was 5¢, then about 3,900 ~Vh
of' electricty were conserved by the retrofit program, or less than 1.
percent of residential electric space heat requirements. This is ob-
viously only· on order of magnitude estimate, but it suggests that the
impact of the existing state retrofit program on aggregate consumption
of electricity is probably modest. The impact could quite possibly
be much smaller with a longer payback period or if a smaller percentage
of credits were taken by electric space heaters than assumed.
A further problem with using national estimates of the potential
savings from retrofitting is that the thermal integrity of the typical
Alaskan house may be much better than the national average. It is
clearly a younger than average stock, so few homes would be without in-
sulati9n as in the lower 48. The harsh winters would suggest more con-
cern during construction for thermal integrity, but this may not have
been the case in fact.
On the basis of this spotty information, it is not·possible to
assume a substantial impact on electric space heating of the existing
federal and state retrofitting programs. Obviously, some of the impact
has already occurred, and to project an estimate of the full impact of
these programs into the future would involve some double counting of
conservation.
F-29
We assume that retrofitting is confined to single family residences
and occurs on the existing housing stock during the period 1980 to 1985.
It is twice as important in Fairbanks as in Anchorage because the higher
price of electricity in Fairbanks creates an extra incentive there. The
impact on Glennallen-Valdez falls midway between the 4 percent saving
for Fairbanks and 2 percent saving for Anchorage.
The application of mandatory construction or performance standards
to new housing in order to improve their thermal integrity has been .
under consideration for several years by the federal government. The
sets of standards which may be implemented are either the American
Society of Heating, Refrigerating, and Air Conditioning Engineers
90-75 standards or standards developed by the U.S. Department of Hous~ng
and Urban Development (HUD). These standards now are supposed to become
effective in 1980.
National studies have estimated the impact of these mandatory stan-
dards on energy consumption. The Arthur D. Little study estimated po-
tential savings of 35 percent, 45 percent, and 40 percent in mobile homes
single family units, and multi-family units, respectively. A 1977 study
estimated savings of 11 percent for single family units and 46 ~ercent
for multi-family units under the ASHRAE 90-75 standards and 20 percent
and 51 percent savings under the HUD standards.22 Substantial savings
are thus apparently possible, but there are no precise estimates of
what the savings would be from standards.
F-30
An attempt has been made to estimate the impact of the two afore-
mentioned standards on Alaskan energy consumption, but the conclusions
of the study were qualitative rather than quantitative and suggested
1 h b . 1 . ld b .bl 23 on y t at su stant~a sav~ngs wou e poss1 e.
We assume-that a program of mandatory standards is implemented in
1981 which has the effect of reducing the h~at load in new construction
(except for mobile homes) b~ 5 percent independent of other factors.
This percentage takes into account the assumption that Alaskan housing
is already more thermally efficient than the national avetage, the fact
that actual savings observed will be less than savings in theory, and
the idea that it will take some time to actually implement the program.
No conservation is assumed for mobile home units.
F.l.B.2. Heating Appliance Lifetimes
For ease of calculation, the demolition rate for the existing housing
stock is set at zero. This is not significantly different from actual
ratios as indicated from building permit information. The assumption
~lso applies to the heat distribution system for the home. The heat
generating system (boiler, furnace, etc.) is assumed to have an average
lifetime of 20 years, independent of type or fuel used. This is based
on Home Appliance Manufacturers Association data, and, as with other
appliances, the actual time of scrapping of an appliance is determined
by a probability_distribution centered at the average lifetime.
F-31
F.l.B.3. Average Housing Unit Size and Consumption
The average housing unit size was estimated in Appendix D, Section
2, and electricity consumption for space heating was assumed to be a
function of the average unit square feet of floor space. (An adjustment
factor was calculated to account for the fact that 1978 was a warmer
than normal winter. See Table D.24.)
We assume that the average electricity requirement for new units
constructed after 1980 (independent of conservation) is 10 percent
greater than existing units because new construction is assumed to con-
sist of larger units on average than the existing stock of housing
Two sources of information confirm the observation of an
average size of the housing stock. In Table F.lO, the average
tion of natural gas per heating degree day is presented
The consumption growth in the 1970s of between 2 and 3 percent
annually can be attributed to growth in the average size of
stock (or to deterioration in the thermal integrity of the housing
stock). Second, as noted in Appendix D, Section 2, the average size
new single family units nationally is larger than the average for the
existing stock by about 10 percent (1720/1570).
We assume new housing of all types constructed after 1980 will be
on average 10 percent larger than the existing stock based upon this
national ratio.
F-32
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
TABLE F.lO. RESIDENTIAL CONSUMER NATURAL GAS CONSUMPTION
FOR SPACE HEATING PER HEATING DEGREE DAY
Average Annual Consumption/
Consumption for Heating Heating Degree
Space Heating (mcf) Degree Days Day
172 10,137 .0170
180 11,879 .0152
191 12,016 • .0159
188 11,665 .0161
170 10,683 .0159
193 11,308 .0171
181 10,361 .0175
166 9,394 .0177
164 9,131 .0180
159 9,430 .0169
Source: Alaska Gas and Service Company annual financial reports and
internal company records.
F-33
In subsequent years, new additions to the housing stock are larger
by 1 percent annually. This is a balance of several factors which can
be identified but not quantified. Increasing disposable incomes will
increase the demand for larger housing units, but the increasing cost
of hou.sing will partially offset this. The role of the federal and
state governments in stimulating home ownership through various programs
could increase the size of new additions to the housing stock or reduce
it, depending on the particulars of the program. The declining average
household size in future years should reduce the demand for larger
units somewhat. The Alaskan climate which requires that people spend
a large proportion of their time indoors during the winter months sug-
gests a strong preference for larger housing units.
The Arthur D. Little study of 1975 assumed that these various
factors would cancel one another out so that the size of new housing
was projected to remain constant in future years and that only replace-
ment of demolitions would increase the average size of the stock by
4 percent between 1975 and 1990. We assume the disposable
the climate-related preference, and the presence of s_tate intervention
into the housing market predominate and r.esult in an increasing size
for increments to the housing stock.
F.l.B.4. Incremental Mode Split*
We assume that the Greater Anchorage Area space heat mode split
is in equilibrium. Thus, the incremental mode split will be equal to
*See also Section F.2 for a discussion of the space heating mode
choice decision.
F-34
the average mode split for electricity. The Greater Anchorage Area
average of 19 percent electric residential space heat~ng is assumed for
all housing types. This is a slight decline from the existing multi-
family stock of 19.9 percent (see Table D.21) but a slight increase
for the other type of units. This assumption presumes no shift in the·
geographic distribtuion of new housing units either toward or away from
areas where natural gas is available or the extension of natural gas
service into new areas. Growth in the mid-1970s in these outlying areas
has been relatively rapid, but it is not clear whether this is a tem-
porary phenomenon or represents the emergence of a long-term trend.
Growth has decelerated in the last year, but that could be a reflection
of the gene~al softening of the Alaskan economy.
In Fairbanks_, Golden Valley Electric Association (GVEA) has put a
ban on new electric_space heat hookups. This is assumed to be permanent
in the absence of new generating facilities powered by fuels other then
fuel oil because of the high incremental cost of power from this source.
Nevertheless, we assume that 1 percent of new and replacement single
family and mobile home units are heated by electricity. This represents
a gradual decline in the electric space heat load occurring over a
period of about 20 years as existing units wear out and are replaced.
GVEA in their own load growth estimates assumes that all of their resi-
dential space heat customers will be shifted off of electrici~y by
1982.
F-35
There is very little electricity used for space heating in the
Glennallen-Valdez service area because of its relative price. ·We
assume the same incremental mode split for electric space heating as
the present average.
F.l.B.5. Utilization Rates
We assume utilization rates are unchanged from current levels.
That is, people do not manually set back their thermostats at night,
etc.
F.l.C. CO~rnERCIAL-INDUSTRIAL-GOVERNMENT ELECTRICITY REQUIREMENTS
Model parameters for the component are shown in Table F.ll.
F.l.C.l. Conservation
Conservation measures in the industrial sector consist of effi-
ciency standards for new appliances, construction or performance stan-
dards for new construction, and retrofitting of existing structures to
increase the thermal efficiency and reduce the electricity load in the
various building systems such as the heating, ventilating, and air con-
ditioning systems (HVAC). Because detailed end use information is not
currently available, it is not possible to identify in detail the impact
on electricity consumption of specific conservation measures. Because
federal conservation programs are and will be in effect, however, it is
necessary to try to quantify their impact.
The major conservation programs of the federal government specifi-
cally targeted to the commercial, industrial, and government sectors
F-36
TABLE F.ll. MODEL PARAMETERS:
COMMERCIAL-INDUSTRIAL-GOVERNMENT SALES
Parameter Region
Greater Greater GlennalLen-
Anchorage Area Fairbanks Area Valdez
Average Consumption Rate
for Existing Stock (KWhl980) 10.675 10.983 9.178
Average Consumption Rate
for Increments to
Stock (KI.fu 1985 ) 15.156 18.537 12.979.
'!•. fi·.
·~-;.-;. Subseguent Increases to -
-~~.:-_ Incremental ConsumEtion
~[;· Rate (nKvlh) · 3.020 3.707 2.596 :{.
'>:;
·f
~(;
Design and Performance
Efficiency Targets (est)
1985 0 0 0
1990 .05. • 05 .• 05
1995+ .1 .1 .1
F-37
include grants to schools and hospitals for improving energy efficiency,
a local public building energy audit program, conservation requirements
for federal buildings, energy efficiency labeling of industrial equip-·
ment, stimulation of cogeneration, business energy tax credits, and per.;..
formance standards for new commercial buildings similar to the residen
24 sector.
National studies have attempted to measure the potential impact of
these federal programs. A 1979 analysis of the National Energy Plan
estimated the growth rate of energy use in the commercial sector could
be reduced from 4 to 3.2 percent annually as a result of.implementation
25 of the plan.
The Arthur D. Little study previously mentioned estimated potential
energy conservation factors for several types of commercial buildings.
These factors, shown in Table F.l2, are the proportion of energy
savings possible using "practical methods and existing materials in
addition to allowing for some technological improvements in selected
HVAC and electrical components between now and 1990."26
These calculations are based upon a technical analysis of possi-
bilities, but the study also includes a discussion of the institutional
setting within which energy conservation in the commerciai sector would
be addressed and provides some insight into the problems which imple-
mentation of energy conservation would entail. Specifically, the
relative complexity of the typical commercial structure makes it
F-38
TABLE F.l2. ENERGY CONSERVATION FACTORS FOR
COMMERCIAL BUILDINGS
(1970 = 1.0)
Existing
Buildings
Office Buildings
Lighting .80
Auxilliary equipment .95
Space heating .78
Cooling .82
Hot water heating .95
Retail Establishments
Lighting .70
Auxilliary equipment .95
Space heating .76
Cooling .77
Hot water heating .95
Schools, Educational
Lighting .. .80
Auxilliary equipment .95
Space heating .79
Cooling .81
Hot water heating .95
New
Construction
.50
.90
.60
.53
.90
.50
.90
.50
.54
.90
.50
.90
.50
.59
.90
Source: Arthur D. Little, "Residential· and Commercial Energy Use
Patterns: 1970 to 1990," for Federal Energy Administration,
197'•, p. 156.
F-39
difficult to calculate actual energy consumption of the various systems
in the building and to determine potential savings from design changes.
(For example, the lighting system waste heat provides some space heat.)
In addition, the design of a building normally involves the attempt
to meet a large number of objectives, only-one of which would be energy
efficiency. The implementation of this objective requires the close
interaction of client, architect, and engineers designing the various
building systems. It is-clear from the discussion in the Arthur D.
Little report that energy conservation was not a major concern in
design and maintenance in the early 1970s. This was reflected in the
fact that architects consulted were sensitive to conservation issues
but lacked "the detailed know·ledge to apply conservation measures with
d f h . . . ,.27 any egree o sop ~st~cat~on.
The heterogeneity of this component of electricity consumption
is a further problem, making it difficult to analyze electricity use
potential savings. Finally, consumption is dramatically effected by
building use patterns. The same building, from a design standpoint,
can have energy and electricity consumption differences of over 100
d d . h t f of the bu-lld-lng.28 perce~t, epen ~ng upon t e pat erns o . use ~ ~
We assume that the majority of electricity used in the commercial~
industrial-government sector is for lighting, in conjunction with space
conditioning systems, and for auxilliary electrical equipment. The
new construction conservation potential in these areas is significant,
but we assume that the impact of currently-planned federal programs,
F-40
including design or performace standards, will be more modest and will
take considerable time in implementation due to institutional constraints
to development of the standards and immediate implementation when they
become available. We assume a 5 percent reduction in electricity re-
quirements for new construction during the period 1985 to 1990 (indepen-
dent of other growth factors), i~creased to 10 percent in the following
decade. This suggests a higher conservation potential in this sector
than the.residential space heat sector but a longer time for implementa-
tion.
At the same time, because of the absence, particularly in the Greater
Anchorage Area, of a strong price incentive, retrofitting measures in the
commercial sector are assumed to have no impact on. electricity consump-
tion. (See next section concerning assumptions regarding growth of
consumption by existing customers.) In other words, the _new construction
standards program is the measure which results in conservation in this
sector.
F.l.C.2. Commercial-Industrial-Government Utility Sales Per Customer
·Historical annu~l utility sales per customer data for the major
railbelt regions and the U.S. as a whole are compared in Table F.l3.
The average Alaskan customer consumes about 70 percent more electricity
in a year than those in the U.S. as a whole, and over the long run, the
growth rate in average sales has been realtively equal for-Alaska and
the U.S. In the period of the 1970s, the Alaskan growth rates have been
more rapid, but this has been offset by apparently slower growth rates
in the late 1960s.
F-41
1950
1955
1960
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
TABLE F.l3
C0~1ERCIAL/INDUSTRIAL UTILITY SALES PER CUSTOMER
(MWh)
Greater Greater
Anchorage Area Fairbanks Area·
47.2 41.2
52.0 40.5
55.2
57.7 53.0
61.7 57.8
62.7 62.8
69.1 69.0
73.7 67.9
80.6 73.4
79.8 72.1
85.1 90.0
87.6 86.6
85.8 84.8
86.5 85.0
aEdison Electric Institute, Statistical Year Book, annual.
F-42
. a U.S .A.
9.3
12.7
17.0
27.4
40.0
41.9
44.5
47.6
46.6
49.0
50.7
52.9
It would be premature to identify the period of the mid-1970s as
a per customer peak for commercial sales in Alaska, but there is a
noticeable deceleration of growth rates in more recent years. This
could partially be the result of more rapid than normal growth during
the Alyeska oil pipeline boom years or a succession of abnormally warm
winters (in Anchorage) in the late 1970s or both. These are temporary
phenomena which should not form the basis for analysis of underlying
trends.
Examination of the year-to-year growth rates of commercial sales
nationally reveals a ve~y rapid growth rate in consumption historically
·and also the possibility of a new long-term trend after the watershed
years of 1973-1974 (the time of the great .recession and oil embargo).
The average annual growth rate in the 1970s before the embargo was 6
percent, while afterwards it was 4.3 percent. Again, it is premature
to emphatically call this a shift in long-term trend, but ft is sug-
gestive of that. )
In order to facilitate analysis of various conservation programs
and trends in new commercial structures, we calculate sales estimates
for the existing stock of commercial-industrial-government buildings
and for increments to the stock. For ease of calculation, we assume a
zero rate of demolition of the stock. For the existing stock, the
average consumption rates in 1978 are util~zed in the projections.
This assumes, therefore, that all growth in consumption is the result
of additional hookups at higher consumption rates and that commercial-
F-43
industrial-government consumption is not sensitive to heating degree
days. (1978 was a warmer than normal year.)
The annual consumption per hookup for incremental customers is
assumed to grow at a rate consistent with the period between 1973 and
1978. Specifically, the following equation was solved in Greater
Anchorage and Greater Fairbanks to obtain the average sales per cus-
tamer of customers added to the systems after 1973.
average sales per customer
for customers added after =
1973
total 1978 sales -total 1973 sales
customers added after 1973
This value was then compared to the 1973 average to arrive at an esti-
mate of the growth in average sales to incremental customers. These
results are summarized in Table F.l4.
Greater
Greater
TABLE F.l4. CALCULATIONS OF ELECTRICITY SALES TO
INCREMENTAL COMMERCIAL-INDUSTRIAL-GOVERNMENT CUSTOMERS
Average Sales Average Incremental
1973 Sales (1973-1978) Ratio
(MWh) (MWh)
Anchorage 80.557 97.903 1.22
Fairbanks 73.429 114.806 1.56
Thus, assuming that existing customer sales remained constant, new
tamers between 1973 and 1978 on average purchased 22 percent more
electricity inAnchorage and 56 percent more electricity in Fairbanks.
F-44
Based upon the longer trend for the two areas from TAble F.l3
and the slower growth for Fairbanks in the early 1970s, we assume the
same growth characteristics for the two regions and also for Glennallen-
Valdez. Specifically, Table F.l5 shows the steps in calculating
average and incremental rates of consumption. Consumption figures have
been converted to a per employee basis (discussed in the next section).
1978 actual average consumption figues are converted to 1980 estimates
on the basis of growth rates calculated from recent trends on average
sales growth rates. This is multiplied by the ratio of incremental to
average sales to obtain a base for calculating incremental consumption
rates for the period 1981 to 1985. These rates are 25 percent above
the 1980 base incremental rates. In subsequent five-year periods, the
same amount is added to incremental sales per employee.
F.l.C.3. Commercial-Industrial-Government Hookups
The number of electric utility hookups is related to employment.
Table F.l6 indicates the stability of that relationship historically
except during the pipeline years of 1975 through 1977.
Because of this stability between the level of employment and the
number of hookups, it is possible to utilize sales per employee as the
forecasting variable in this section of the model and link it directly
to the employment forecast generated by the economic model.
F-45
I
I'Zj
I
~
"'
TABLE F.l5. COMMERCIAL-INDUSTRIAL-GOVERNMENT
CONSUMPTION PER EMPLOYEE DATA
Rate of
1978 1978-1980. 1980 19n-1978
average sales growth in average sales incremental to
/emElo:lee c /emEloyee 1978 av. sales sales/emEl•
(MWh) (%) (MWh)
Greater Anchorage 10.071 6% 10.675 1.13
Greater Fairbanks 10.768 2% 10.983 1.35
Glennallen-Valdez 10.085a -9% 9.178 1.13b
a 8,000 MWh of pipeline pump station sales netted out.
b Assume the same relationship as Anchorage.
cBased on recent trends in average sales growth.
Av. sales/ Succeeding 5
employee for yr. increments
1981-1985 to sa.les to
incremental incremental
customers customers
(MWh) (Mivh)
15.156 3.020
18.537 3.707
12.979 2.596
' TABLE F .16.. RATIO OF NON-AGRICULTURAL HAGE AND EMPLOYMENT
(NET OF MILITARY) TO COMMERCIAL-INDUSTRIAL-GOVERNMENT HOOKUPS
Greater Anchorage Greater Fairbanks Glennallen-
Area Area Valdez
1965 8.55 8.73 5.37
1966 8.63 8.02
1967 8.79
1968 8.34 8.43
1969 8.50 8.80
1970 7.79 7.61 3.58
1971 8.37 7.59 4.37
1972 8.46 7.78
1973 8.54 7.54 2.66
1974 9.12 8.81 4.32
1975 10.12 13.14 11.18
1976 9.45 12.15 17.69
1977 8.94 9.06 8.55
1978 8.59 7.90 4.12
Average 8. 72 8.89 6.87
Average net of 1975-1977 8.52 8.12 4.07
F-47
F.l.D. MISCELLANEOUS ELECTRICITY REQUIREMENTS
This very small category consists of street light and second home
sales. Street light sales are assumed to be 1 percent of the sum of
all other components of sales.
The electricity requirements for second homes is difficult to
identify for several reasons. First, as discuss.::-d in Appendix D,
Section 1, it is difficult to identify from the existing housing stock
studies just what is a second home or a vacation home. Specifically,
what the census defines as a year-round housing unit may actually be a
second or vacation home. It is possible from the census to determine
the number of households within a census division which own a second
home, but not its location. Most utilities do not have separate rate
schedules for second homes, and if they did, the utility definition
would not necessarily be the appropriate one since it might cover
seasonal units rather than units used year-round but infrequently.
It is also difficult to estimate average electricity requirements for
second homes because of this lack of data.
We make the following very rough estimates to calculate second
home electric utility sales:
1. 25 percent of households have second homes, based on census
information;
I 2. 50 percent of the second homes are located within the
railbelt;
F-48
3. 50 percent of the second homes in the railbelt are serviced
by electric utilities.
The average annual consumption per second home is 500 ~~, based
upon conversations with utility personnel.
F.2. Assumptions for the Price Induced Shift
Toward Electricity Consumption Case
F.2.A. FACTORS INVOLVED IN APPLIANCE CHOICE FOR SPACE HEATING
The most important variable in a model of appliance choice for space
heating or any other function is the system cost, including both the
initial purchase price and lifetime fuel costs for system operation.
Other characteristics are important and will affect the choice but will
not be explicitly considered here. Some of these other considerations
are as follows:
Heating System
1. Heat distribution within the building
2. Amount of space occupied by the heating system
3. Multiple controls capability
4. Integration with other appliances (hot water,
humidifier, air conditioner)
5. Comfort factor (annoyance of hot air, for example)
6. Reliability
7. Compatability with auxilliary heating systems
Fuel
8. Fuel availability
9. Haintenance cost
10. Safety of fuel
11. Cleanliness
12. Convenience
F-49
Because of these considerations, not all households will make the same
appliance type and fuel choice even when faced with identical prices.
In considering the least-cost space heating system, it is necessary
to recognize that for the new residence there are at least two decisions
after having decided in favor of a centralized heating system. The first
is the type of heat distribution system and the second is the type of
fuel to produce the heat. The choice of distribution system--hydronic
(hot water or steam), hot air, electric resistance, etc.--affects not
only the initial cost but also the cost of a subsequent retrofit to ,._n
alternative system. Having once chosen a heating distribution system,
the fuel to provide the heat is determined, basically on the basis of
. 29 prJ.ce. -
For a given set of desired heating characteristics, the distribution
system and fuel type chosen will, in theory, be that which minimizes the
present value of total future system costs. In practice, several factors
tend to distort the decision in favor of the system with the least initial
cost or the least cost over a shorter period than the system lifetime.
1. Homebuilders who build for others are concerned with
minimizing construction costs and will opt for the
system with the least initial cost if a related higher
total operating cost is not reflected in a reduced
market price for the house.
2. The same analysis applies for landlords to the extent
that they are able to pass system operating costs on
to tenants.
F-50
3. For individuals who own their homes for only a short
time before moving, the lifetime operating costs of
the heating equipment in those homes is less important
than initial system cost.30
4. Lack of information about the least-cost system may
prevent people from switching to it.
5. Individuals may not act consistently with the actual
opportunity cost of money. In other words, a system
choice with a high initial cost may have a rate of
return in terms of money saved (compared to the next
best alternative which has a lower initial cost but a
higher operating cost) which exceeds the return the
purchaser could receive investing the same amount of
money alternatively. Yet, for some reason, the con-
sumer chooses the system with the lower initial cost
but higher lifetime cost. In other words, the ob~
served discount rate used by the consumer exceeds his
opportunity·cost.31
6. Promotional activities of utilities.
In general, electric baseboard heating is the cheapest system to
install followed by hydronic and then warm air systems. Both hydronic
and warm air systems require a flue and a distribution system. The
initial cost of oil relative to natural gas depends upon the cost of
connecting the residence to the gas main compared to the cost of oil
storage tanks and the somewhat higher cost of an oil burner. This may
vary with location.
Based on this discussion, it can be seen that the proportion of a
particular type of heating system in place at any time may exceed what
would appear to be economically justified based upon total system
lifetime cost.
F-51
Table F.l7 shows the prices of various fuels which equalize the
operating cost of space heating. For example, if ele c tricity is 3~/kWh,
then heating with fuel oil will be less expensive if it is available
for under $.84/gallon.
Table F.l8 shows the actual relative prices presently encountered
in various parts of the railbelt. It is clear that natural gas is the
cheapest fuel wherever it is available. Outside of the gas utility
service area in the Anchorage region, the prices of fuel oil and
electricity are comparable after adjusting for the higher conversion
efficiency of electricity. This is because electricity is produced
using cheap natural gas.
In the Fairbanks area, fuel oil and electricity are comparable for
one utility, while the price of electricity from the other exceeds that
of fuel oil. This situation results from the fact that the lower cost
electricity is produced by coal, while the higher cost electricity is ·
generated by a combination of fuel oil and coal.
In Glennallen-Valdez, fuel oil is clearly cheaper than electricity.
This pattern is confirmed by looking at historical data on relative
fuel prices. In Anchorage, natural gas and electricity prices are largely
determined by the long-term purchase contracts for Cook Inlet gas. The
pricing clauses in these contracts have resulted in fairly stable prices
during the 1970s. In contrast, fuel oil prices have risen with rising
crude oil prices.
F-52
$/106 btu
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
TABLE F.l7.
EQUIVALENT DELIVERED PRICE FOR SPACE HEATING
USING VARIOUS FUELS
Electricity Fuel Oil Crude Oil
¢/kWh $/gallon $/barrel
.5 .14 5.88
1 .28 11.77
1.5 .42 17.64
2 .56 23.54
2.5 .70 29.40
3 .84 35.28
3.5 .98 41.16
4 1.12 47.07
4~5 1.26 52.92
5 1.40 58.80
5.5 1.54 64.68
6 1.68 70.56
6.5 1.82 76.44
7 1.96 82.32
7.5 2.10 88.20
Notes: 1) Furnace conversion efficiencies:
electric .95
gas and oil .65
2) BTU content of fuels:
1 kWh = 3,413 btu
1 gallon fuel oil = 138,000 btu
1 mcf gas = 1,005,000 btu
1 barrel crude oil = 5,800,000 btu
3) No r.efinery margin netted out of crude oil price
F-53
Natural Gas
$/me£
1.53
3.06
4.59
6.12
7.65
9.18
10.71
12.24
13.77
15.30
16.83
18.36
19.89
21.42
22.95
TABLE F .18.
PRICES OF ALTERNATIVE FUELS FOR RESIDENTIAL SPACE HEATING
Natural Gas a Fuel Oilb Electricityc
$/mcf $10 6btu $/gallon $10 6btu ¢/kWh $10 6btu
GREATER ANCHORAGE AREA
Anchorage
1980 .98 7.10
1979 2.98/d. 8.73/d
2.49/ 7.30/
4.52 13.25
1978 1.89 1.88
Kenai-Cook Inlet
1980 .94 6.80
1979 3.52l 10.32/
3.75/ 10.99/
4.23 12.40
1978 2.01e 2.00
:Hatanuska-Susitna
1980 unavailable 1.07 7.75
1979 unavailable 4.52 13.25
GREATER FAIRBANKS AREA
Fairbanks
1980 unavailable .84 6.09
1979 unavailable 3.50/g 10.26/
7.97 23.37
GLENNALLEN-VALDEZ AREA
1980 unavailable .98 7.10
1979 unavailable 12.1/h 35.47/
13.82 40.52
See following page for table notes and sources.
F-54
Table F .18 (Continued)
Table Notes: (a) 10 mcf monthly bill
(b) 500 gallons delivered
(c) 1,500 kWh monthly bill
(d) CEA/AMLP/MEA
(e) ANG
(f) HEA/SES/HEA (Kenai)
(g) FHUS/GVEA
(h) CVEA (Valdez)/CVEA (Glennallen)
SOURCES: Electricity: State of Alaska, Division of Energy and Power
Development, "1980 Alaska Power Development Plan."
Gas: Alaska Public Utilities Commission, Annual Report, 1978.
Fuel Oil: Survey by authors and Fairbanks North Star Borough,
Community Information Quarterly, Vol. III, No. 1, Spring 1980.
F-55
In Fairbanks, fuel oil and electricity generated by fuel oil have
·.
been most susceptible to pric.e increases in the 1970s. The price of
electricity generated by coal has been less susceptible to increases in
the 1970s.
In Glennallen-Valdez, electricity is produced by fuel oil and, thus,
the two prices move together.
This review suggests that a substantial change in the existing fuel
mode split for space heating would require a large change in the rela-·
tive price of electricity. Specifically, for electricity to become
the least expensive space heating fuel, the following price changes
would be necessary;
Anchorage -the relative price of natural gas would need to
increase at least 3 times;
Fairbanks -the relative price of fuel oil would need to increase
at least two and one-half times; and
Glennallen-Valdez -the relative price of fuel oil would need
to increase at least three and one-half times.
On the other hand, it is possible that a large increase in the
price of all fuel will result in a shift away from the "conventional"
fuels-oil, gas, and electricity--toward more conservation or auxilliary
systems such as efficient fireplaces and wood stoves. This phenomenon
may be beginning to occur ,.in the outlying parts of the Greater Anchorage
area and in Fairbanks.
F-56
This introduces the second group selecting a space heating fuel--
the retrofit market. This consists of households whose existing system
has worn out as well as those whose systems are still functiorting but
because of changed operating costs decide that a system replacement is
cost effective.
Fuel retrofits are relatively ·common when the switch is between
oil and gas in hot air or hydronic systems. For example, a large
portion of the Anchorage residential market has been retrofit from
oil to gas. This required only switching the burner and connecting
the unit to the gas main. Switches to an electric resistance furnace
could similarly be relatively inexpensively accomplished.
System retrofits in which one type of heating system is replaced
with another are far less common, and the feasibility of such a switch
will depend upon the construction of the building.
For example, in a house built on a concrete slab, it would be
virtually impossible to retrofit a hydronic or hot air system because
there would be nowhere for the placement of th~ pipes or ductwork. In
replacing an electric resistance system with a hydronic or hot air system,
it is necessary both to locate a place for the furnace and to install a
flue. Generally, the cheapest system retrofit is to electric resistance
heating since the installation of the required wiring is less com-
plicated than that of pipes or ductwork.
F-57
Table F.l9 summarizes the conversion costs (prices as of the early
1970s, although the relative costs are unchanged) of v arious fuel and
system retrofits. The table c.onfirms the discussion that retrofits to
electric baseboard are relatively inexpensive, but retrofits from elec-
tricity to oil or gas conventional systems are relatively expensive.
Basically, it is easier to switch into the electric mode than out of it.
Obviously, for the household contemplating a retrofit, the cost of
the switch plus the present discounted cost of the fuel in the new system
is the relevant variable against which the cost of using the current
system must be compared. The higher the cost of retrofit, the higher
the relative price of the existing fuel used for space heating can become
and still be less costly than switching.
Table F.20 shows what the actual railbelt heating s y stems have been
historically. The pattern has been one of shifting from heaters to
central space heating systems over time (which consume more energy).
Hydronic systems were more common in Anchorage, Fairbanks, and Glennallen-
Valdez in 1970, while hot air systems were more common elsewhere. This
suggests that shifts to electric heat could involve either resistance
systems or electric furnaces using the existing heat distribution systems
(or possibly heat pumps).
There is a historical example of a retrofitting phenomenon in the
railbelt. In the early 1960s, natural gas from the Cook Inlet fields
became available to the Kenai Peninsula and Anchorage. The natural gas
F-58
'j1
V1
1.0
CONVERSION COSTS OF RESIDENTIAL HEATING SYSTEMS
(dollars)
~ Gas Electricity
Fuel System From Warm Air Hot Water Baseboard Warm Air
Gas Warm Air X 3,300 1,500 1,500
Hot Water 2,600 X 1,500 3,100
Electricity Baseboard 2,600 3,300 X 2,700
Warm Air 1,000 3,300 1,100 X
Warni Air 400 3,300 1,500 1,500 Oil Hot Water 2,600 400 1,500 3,100
SOURCE: Arthur D. Little, "Project Independence: Residential and Conunercial
Energy Use Patterns 1970-1990," for Federal Energy Administration,
1974, p. 175.
Oil
Warm Air Hot Water
65o· 3,400
2,700 650
2,700 3,400
1,100 3,400
X 3,400
2,700 X
TABLE F.20.
HOME HEATING EQUIPMENT
PROPORTION OF TOTAL
Built-in Floor, Hall,
Steam or Warm Air Electric or Pipeless Heaters,
Hot Water Furnaces Units Furnaces FireElaces None
GREATER ANCHORAGE AREA
Anchorage
1960 46 20 0 8 26 0
1970 56 26 5 3 9 0
Matanuska-Susitna
1960 12 13 0 7 67 1
1970 16 30 1 1 53 0
Kenai-Cook Inlet
1960 9 14 0 2 72 4
1970 . 25 32 4 5 33 1
Seward
>tj
I 1960 18 12 0 14 56 1
~
0 1970 27 28 0 0 44 1
GREATER FAIRBANKS AREA
Fairbanks
1960 50 25 0 3 22 1
1970 59 22 7 1 11 0
GLENNALLEN-VALDEZ AREA
Va1dez-Chitina-Hhittier
1960 34 11 0 3 51 1
1970 28 20 1 1 50 1
WESTERN REGION U.S.
1970 5 40 9 22 22 2
Note: Based on all year-round housing units.
SOURCES: 1960 Census of Housing, General Housing Characteristics: Alaska. Table 29.
1970 Census of Housing, Det~iled Housing Characteristics: Alaska. Table 62.
1970 Census of Housing, Detailed Housing Characteristics: United States Summarv.
was cheaper than fuel oil; but in response to competition, the fuel oil
distributors lowered their price. This, in combination with the fact
that residents of short tenure would not recover their capital costs .
of conversion from an oil to gas boiler in spite of the relatively low
cost, kept the rate of conversions low. The limiting factor does not
appear to have been the speed at which gas mains could be laid to the
various neighborhoods.32
Over a period of 15-to-20 years, as indicated by Table F.21, there
was a substantial shift toward natural gas space heating as a result of
both new units choosing gas and retrofitting of gas burners where fuel
oil had previously been used. The retrofitting of gas burners continues
today but is asymptotically approaching zero. Unfortunately, because
of a lack of intercensal housing stock data for the 1960s for Anchorage,
it is neither possible to trace the exact time pattern of the retro-
fitting of gas burners nor possible to correlat·e the rate of retrofitting
with·the relative prices of fuels or other variables in an attempt to
develop a model for predicting possible future responses to relative
fuel price changes.
F. 2. B. ASSUMPTIONS FOR A HIGH ELECTRIC SPACE HEAT SCENARIO
Given the present structure of relative fuel prices in the rail-
belt, the ·electric portion of the residential space heating load should
remain relatively constant over time. This is predicated on the follow-
ing assumptions:
F-61
Year
1960
1965
1970
1975
1978
TABLE F.21.
GROWTH OF USE OF NATURAL GAS FOR
ANCHORAGE RESIDENTIAL SPACE HEATING
Dwelling Units
(not including Residential
multifamily) Gas Customers
16,313 0
19,876 5,922
24,216 12,097
33,894 22' 779
39,702 30,629
Percentage of
Residences
U~ing Gas
0
30
50
67
77
SOURCES: 1960 Census of Housing, General Housing Characteristics:
Alaska.
1970 Census of Housing, Detailed Housing Characteristics:
Alaska.
Anchorage Urban Observatory, University of Alaska. 1975
Housing Study.
Alaska Gas and Service Company Annual Reports.
F-62
1. Retrofitting of fuel oil for electric space heating
in Fairbanks will continue, but more slowly than in
the years since 1975 when a ban on new residential
electric space heating was imposed by Golden Valley
Electric Association (GVEA)~ This is because of the
high cost of conversion from resistance electric
heating to a fuel oil boiler. Some switching will
be partial conversions not to an alternative central
heating system but rather to room-by-room heating
units.
2. Retrofitting of natural gas for electric space heating
in Anchorage will continue but also at a slow pace
because not only is the conversion cost high but also
because a large portion of the electric space heating
load is in multifamily units where conversion may be
relatively more expensive and the incentive to the
owner less if the tenant absorbs the cost of the fuel.
The conversion of such units to condominiums might
speed the conversion process but not guarantee it.
3.. The existing natural gas utility service areas do not
expand significantly and thus capture a larger market
share from the electric utilities. Such expansion
might occur north into the Matanuska Valley, south
into the Homer area, or into thinly populated portions
of the Anchorage Borough such as the Hillside area and
Girdwood. Such expansions would be a question of regu-
latory policy and the economics of laying new gas mains.
This relates to the fourth assumption.
4. The distribution of the population within the regions
of the railbelt (particularly Anchorage) does not change
significantly. If all future population increase asso-
ciated with economic growth in the Municipality of
Anchorage settled in the Matanuska Valley, then the use
of electric space heating would expand relatively rapidly.
However, this growth, particularly if it were accompanied
by an increase in the density of settlement~ would be a
stimulus to the extension of gas service into the Matanuska
Valley, thus reducing the electric space heating proportion
in favor of gas.
5. New natural gas utilities do not make natural gas an avail-·
able alternative in either the Fairbanks or Glennallen-
Valdez areas.
6. Presently available fuels, particularly natural gas, will
continue to be available for residential space heating
purposes.
F-63
The amount of recoverable reserves of natural gas remaining in Cook
Inlet is the subject of some controversy. This is understandable since
exploration is still proceeding and new reserves may be discovered. The
total is obviously finite, but the proportion of ~xisting reserves
annually utilized for Anchorage space heating is relatively small. In
1979, for example, of 266 million mcf utilized, only 14 million was
marketed to consumers for space heating and other uses.33 Thus, to the
extent space heating is a priority use of natural gas, relatively modest
reserves and reserve additions would be sufficient to satisfy even a
rapidly growing market. To a certain extent, the space heating priority
is automatically built into the gas distribution structure. Gas sales
from the gas utility to the electric utilities are under interruptible
contract· so that if supply constraints develop, the shortfall will occur
in the availability of ga~ for electricity generation (at least in the·
34 short run).
A reliable published estimate of currently proven recoverable
of natural gas in Cook Inlet does not exist. Recent past estimates are
as follows:
1. 7.044 trillion cubic feet as of January 1, 1977.35
2. 6.413 trillion cubic feet as of January 1, 1977.36
Estimates of undedicated reserves are as follows:
1. 3.919 trillion cubic feet as of January 1, 1977.37
F-64
2. 4.236 trillion cubic feet as of January 1, 1978,
from the currently producing fields.38
3. 5.422 trillion cubic feet as of January 1, 1978.39
Estimates of potential additional resources are from 6 to 29 trillion
b . f 40 cu ~c eet.
On the basis of existing reserves, the supply of gas in Cook Inlet
is sufficient to meet demand growth through 2000 if a large proposed
LNG export .facility were not built. If it is built to preliminary de-
sign capacity and consumes 3.2 trillion cubic feet of gas over a twenty-
year lifetime, then the existi-ng supply will not be able to meet all
41 expected demands.
If the availability of alternative fuels (including natural gas,
fuel oil, but also wood) is reduced or if the relative prices of alter-
native fuels rise and thus make electric space heating mo+e economically
attractive, then the proportion of space heating needs met electrically
might increase.
It is clearly impo~sible to identify all of the conditions under
·which such a change might actually take place. Thus, it is also
impossible to quantify the impacts on electricity use of all possible
scenarios of changing relative electricity prices.
The factors determining relative prices could be divided into four
categories as follows: Alaskan market conditions (supply and demand),
F-65
national and international market conditions, national energy regula-
tions and policies, and Alaskan energy policies. Both Federal and state
policies will undoubtedly alter the relative prices and availabilities
of fuels which would result from normal market forces over the next
thirty years.
The state is in a central position in terms of being able to affect
fuel prices and availabilities because of both its substantial surplus
revenue position and its ownership of significant energy resources.
The state government could, in the short run, easily subsidize the total
energy requirements of the entire population out of excess revenues.
It can provide fuel resources such as coal, oil, and natural gas to
local markets at below market prices. Whether it will choose to do
these things is a political question.
Federal regulations and policies may act to make particular fuels
more expensive or unavailable for specific uses. Environmental regula-
tions on coal are one example of the former, while possible prohibition
on the use of gas in the generation of electricity is another.
Absent such government induced changes in energy fuel markets, the
long-run trend in prices will likely be towards comparability and
with world oil prices. In particular, it is plausible to assume that
fuel oil prices and domestic oil prices will gradually approach the
level of world oil prices as decontrol of prices is phased in. Prices
of other fuels can thus be examined in relation to the fuel oil price.
Examining region-by-region, the following price scenario .is reasonable:
F-66
~-
Anchorage. Natural gas prices will rise relative to fuel oil as
existing contracts expire and as pricing provisions of existing contracts
are activated which require that all purchasers pay the same price as
the highest priced purchaser from a field.42 The effect of these trends
will be to raise the delivered price to the consumer of both natural gas
and electricity since electricity is gas generated. Because of different
contracts, the exact relative effects cannot be estimated accurately.
The advent of the Pacific LNG facility has been estimated to have a
larger price impact on electricity users rather than gas users, but
11 . . "1 . d 43 I overa pr~ce ~ncreases are not so eas~ y est~mate • n any event, as
long as gas is used to generate electricity and current space heating
practices are maintained, gas will be the less expensive space heating
fuel. It is possible, but not likely, that gas prices will reach parity
with fuel oil. Working against such parity is the cost of transportation
of gas·to its alternative market on the West Coast. Working toward price
parity is the fact that the market for gas on the West Coast may value
the gas at its peak rather than baseload value. In this case, gas in
Anchorage would lose its attractive price relative to fuel oil, but not
electricity.
In order for electricity to be the cheapest fuel of the three, it
must be produced by a means other than natural gas or fuel oil such as
coal or hydropower. Electricity thus generated has the potential for
being least cost, although it is by no means assured.
Fairbanks. As fuel oil prices rise, electricity prices will also
increase because the present generating mode in Fairbanks utilizes fuel
F-67
oil for incremental electricity generation. To the extent that electri-
city consumption also grows, the cost of electricity will continue to
exceed the fuel oil cost for space heating purposes. This link will
broken, and the cost of electricity made independent of the fuel oil
price if an alternative fuel such as coal or hydropower is used to
generate electricity. In such a case, electricity may become a less
costly fuel for space heating than fuel oil.
Glennallen-Valdez. ·The cost of electricity is presently tied to
that of fuel oil because fuel oil is used to generate electricity.
the·future, this will no longer be the case since a hydroelectric
facility is currently under construction in the area. In the short
the integration of the electricity from hydropower is not anticipated
reduce the price of electricity. If the price is stabilized at its
current level, the price of fuel oil would have to increase by three
times before electricity would be priced equivalent to fuel oil for
space heating. This would occur in twenty years at a 6 percent rate
fuel price increase, or ten years at a 13 percent rate of fuel price
increase.
Under these assumptions, a significant shif·t toward use of elec~
tricity could occur under the following conditions:
Anchorage. Decreased availability and/or increased price of gas
result in the addition of coal and/or hydroelectric baseload electric
F-68
generating facilities by 1990. The electricity price does not fall
relative to that of gas until 2000, however, because:
1. The initial cost of those capital-intensive facilities
is large.
2. The main source of electricity will continue to be gas-
fired turbines which, since the gas will have become
expensive, will keep the average price of electricity
high.
Fairbanks. Increased demand.or very rapid increase in the price
of fuel oil makes a coal plant attractive. If it is in place by 1990,
it could immediately "back out" high priced oil, but the electricity
price would not immediately fall relative to that of alternatives
because:
1. The initial cost of the capital-intensive coal plant
will be large.
2. The cost of the fuel oil-fired generation facilities
will still be a part of the price of electricity.
Glennallen-Valdez. By 1990, the fuel oil price may have increased
sufficiently to make electricity from hydropower relatively attractive.
For this to happen, any additional generating capacity requirements must
also be met by low cost modes of generation.
This study cannot predict these outcomes since they are obviously
dependent upon not only the level of demand but also upon costs of supply,
taking into account not only the cost of additions to systems but the
impact on system cost of those additions.
F-69
We can, however, analyze the case presented above where electricity
becomes relatively inexpensive as a fuel. Based upon the foregoing, we
assume a shift towards electricity for space heating beginning in the
period 1995-2000 for Anchorage and Fairbanks and 1990-1995 for
Valdez.
The price advantage for electricity is assumed to be real and las
but not of a substantial magnitude. Thus, the shift to electric space
heating follows the pattern observed·in Fairbanks in the early 1970s,
rather than the pattern in Anchorage in the 1960s during the shift to
natural gas. That is, new installations are predominantly electric,
existing units are not retrofitted to electric space heat.
marily because of the cost of retrofitting to electric space
pared to switching from an oil to a gas burner, for example) combined
with the relatively short average tenure by an owner in a home.
In addition, electric appliances became more attractive relative
to natural gas and fuel oil. The electric incremental mode splits
for water heaters, ranges, and clothes dryers increase at the same
time that the shift to electric space heating occurs. The commercial-
industrial-government sector projections are similar to those of the
base case.
The parameter changes for this case are summarized in Table F.22.
F-70
TABLE F.22. PARAMETER VALUES: THE PRICE INDUCED SHIFT
TOWAP~ ELECTRICITY CONSUMPTION IN THE RESIDENTIAL SECTOR CASE
Parameter Region
Greater Greater Glennallen-
Anchorage Fairbanks Valdez
SPACE HEAT
incremental mode split (msi. ) J,t
SF
1985 .19 .01 .02
1990 .19 .01 .02
1995 .19 .01 .9
2000+ .9 .9 .9
DP
1985 .19 0 0
1990 .19 0 0
1995 .19 0 .9
2000+ .9 .9 .9
MF
1985 .19 0 0
1990 .19 0 0
1995 .19 0 .9
2000+ .9 .9 .9
NH
1985 .19 .01 0
1990 .19 .01 0
1995 .19 .01 .9
2000+ .9 .9 .9
APPLIANCES
incremental mode split (msi. t) J,e,
WH
1985 .35 .5 .4
1990 .35 .5 .4
1995 .35 .5 .9
2000+ .9 .9 .9
c
1985 .66 .85 .4
1990 .66 .85 .4
1995 .66 .85 .9
2000+ .9 .85 .9
SF = single family MH = mobile home
DP = duplex WH = water heating
MF = multifamily C = cooking
F-71
ENDNOTES: APPENDIX F
1. Eric Hirst and Janet Carney, "Residential Energy Conservation:
Analysis of U.S. Federal Programs," Energy Policy, September 1977,
pp. 211-222.
2. Jorgen S. Norgard, "Improved Efficiency in Domestic Electricity
Use," Energy Policy, March 1979, pp. 43-56.
3. Battelle Columbus and Nepool Load Forecasting Task Force, "Report
on Model for Long-Range Forecasting of Electric Energy and Demand
to the New England Power Pool," June 30, 1977, p. 624.
4. California Energy Commission, "Appendix A: Analysis of Residential
Energy Uses," Section A9.
5. Eric Hirst, William Lin, and Jane ·Cope; i'A Residential Energy-Use
Model Sensitive to Demographic Economic and Technological Factors,"
Quarterly Review of Economics and Business, Vol. 17, No. ·2, p. 13.
6. Jerry A. Hausman, "Individual Discount Rates and the Purchase
and Utilization of Energy Using Durables," Bell Journal of Economics,·
Vol. 10, No. 1, p. 45.
7. Data from California Energy Commission, "Appendix A: Analysis
of Residential Energy UseB," Appendix 7, Table A7.6.
8. Present average electricity consumption figures applied to 1960
saturation rate data results in a substantial overestimate of
actual electricity consumption for that year.
9. California Energy Commission, Ibid.
10. California Energy Commission, Ibid., Section 12.
11. California Energy Commission, Ibid., Section 7, Table A7.5.
12. Compiled from California Energy Commission, Ibid., Section 7,
Table A7.6.
13. California Energy Commission, Ibid., Section 7, Table A7.10.
14. California Energy Commission, Ibid., Section 10, Table Al0.2.
15. American Gas Association, Gas Facts, annual.
16. Midwest Research Institute, Patterns of Energy Use by Electrical
Appliance, for Electric Power Research Institute, EPRIEA 682,
1979, p. 512.
F-72
17. U.S. Department of Energy, Office of Public Affairs, "The Natural
Energy Act," 1978.
18. Arthur D. Little, "Project Independence Residential and Commercial
Energy Use Patterns 1970-1990," for Federal Energy Administration,
1974, p. 21-23.
19. Hirst and Carney, Ibid., p. 217.
20. California Energy Commission, Ibid., Section A3.
21. Arthur D. Little, Ibid.
22. Hirst and Carney, Ibid.
23. Larry Breeding. "Phase III Evaluation: ASHRAE 90-75 Energy Con-
servation in New Building Design, HUD Minimum Property Standards,"
for State of Alaska, Energy Conservation Program, 1976.
24. U.S. Department of Energy, Ibid.
25. Eric Hirst and Bruce Hannon, "Effects of Energy Conservation in
Residential and Commercial Buildings," Science, Vol. 205, August 17,
1979, p. 656-661.
26. Arthur D·. Little, Ibid., .p. 157.
27. Ibid., p. 137.
28. Ibid., p. 139 and conversations with local engineers.
29. Charles Rivers Associates, "Analysis of Household Appliance Choice,"
report for Electric Power Research Institute, 1979, Chapter 2 .•
30. In Alaska, this is definitely a factor. A 1976 survey in Fairbanks
found the median length of residence in owner-occupied housing to
be three years. John Kruse, Fairbanks Community Survey, Institute
of Social and Economic Research, 1977, p. 5. A 1975 Survey of
Anchorage found that 61 percent of households had lived at their
present address less than three years. Diddy Hitchens et al,
"Anchorage Municipal Housing Study," Anchorage Urban Observatory,
1976.
31. A recent study estimated the discount rate used by consumers in
making appliance choice decisions to be Z5 percent, which is con-
siderably above the opportunity cost of capital for many consumers.
This suggests a bias in favor of systems with the least initial cost.
Jerry Hausman, "Individual Discount Rates and the Purchase and Utili-
zation of Energy Using Durables," The Bell Journal of Economics,
Vol. 10, No. 1, p. 33.
F-73
32. Based on conversations with Harold Schmidt of Anchorage Natural Gas
Company.
33. Scott Goldsmith and Kristina O'Connor, "Alaska Historic and Projected
Oil and Gas Consumption," prepared for the Alaska Royalty Oil and Gas
Development Advisory Board and the Alaska State Legislature, 1980,
p. 5.
34. Harold Schmidt, Ibid.
35. Stanford Research Institute, "Natural Gas Demand and Supply to the
Year 2000 in the Cook Inlet Basin of Southcentral Alaska," for
Pacific Alaska LNG Company, 1977, p. 37.
36. Alaska Department of Natural Resources estimate noted in Scott
Goldsmith and Tom Lane, "Oil and Gas Consumption in Alaska 1976-
2000," for the Alaska Royalty Oil and Gas Development Advisory
Board and the Alaska State Legislature, p. III.5.
37. W. H. Swift et al, "Alaskan North Slope Royalty Natural Gas: An
Analysis of Needs and Opportunities for In-State Use," for State
of Alaska, Division of Energy and Power Development, p. V.5.
38. Scott Goldsmith and Tom Lane, Ibid., p. III.7.
39. ·Ibid., p. IV.5.
40. Stanford Research Institute, Ibid., p. 36.
41. Scott Goldsmith and Tom Lane, Ibid., p. IV.5.
42. For a recent analysis of this, see Jack Kreinheder, State·of
House Research Agency Report -Request No. 30, February 1980.
43. Ibid.
F-74
APPENDIX G: PROJECTING MILITARY AND SELF-SUPPLIED
INDUSTRIAL ELECTRICITY NET GENERATION
G.l. Military Requirements
Six major military installations, all of which produce and consume
their own electricity, are located within the railbelt region of Alaska.
The level of activity at these bases is a function of national defense
policy and, as such, it is difficult to project. Historically , there
has been considerable variation in the number of personnel stationed
at these facilities with peaks occurring during 1\Torld \\Tar II and the
Korean \\Tar. Table G.l sho\vs the variation in the level of military
personnel for the whole state over the historical period. (Currently,
about 80 percent of statewide military personnel are located in the
railbelt.) The general trend since the late 1950s has clearly been for
a gradual decline, but this may not be a reasonable guide to future
personnel levels.
We have not attempted to correlate military electricity consump-
tion with the level of personnel or other factors for a number of
reasons. A complete historical data series on military consumption
would be difficult to obtain. A large and varying proportion of per-
sonnel live off base, and, consequently, their residential consumption
is satisfied by utility-supplied power. Finally, the difficulty in
projecting military personnel nullifies the value of utilizing a
functional relationship between personnel and electricity consumption.
TABLE G.l. AVERAGE ANNUAL HILITARY
PERSONNEL IN ALASKA
(thousand)
1940 1 1960 33
1941 8 1961 33
1942 60 1962 33
1943 152 1963 34
1944 104 1964 35
1945 60 1965 33
1946 19 1966 33
1947 25 1967 34
1948 27 1968 33
1949 30 1969 32
1950 26 1970 31
1951 38 1971 30
1952 50 1972 31
1953 50 1973 27
1954 . 49 1974 26
1955 50 1975 25
1956 45 1976 24
1957 48 1977 25
1958 35 1978 23
1959 34 1979 23
Source: 1940-1959 -George Rogers, The Future of Alaska: The
Consequences of Statehood, Resources for the Future,
1960-1965 -George Rogers, "Alaska Regional Population and
Employment," ISER, 1967, p. 42.
1960-1969-MAP model.data.
1970-1979 -Alaska Department of Labor, "Alaska Population
Overview, December 1979, p. 50.
G-2
The Air Force conservation goal is to reduce their total energy
requirements by 20 percent by 1985, according to the Alaskan Air
Command. The Army may have a similar conservation goal. It is not
clear what impact this policy will have on electricity consumption.
Because of these difficulties which make detailed projections of
military electricity requirements questionable, we assume the current
level of net generation in all future years. Current requirements are
shown in Table G.2~
G.2. Self-Supplied Industrial Requirements
The largest industrial users of self-generated electricity in the
railbelt are, with one exception, in the category of petroleum production,
processing, and transportation. The University of Alaska, Fairbanks
campus, is the only large public, non-utility generator of electricity.
Table G.3 shows that most of the self-supplied electricity is
centered in the Greater Anchorage Area. Offshore and onshore drilling
and producing petroleum rigs contribute a major portion of the total
load, along with the pipelines and other facilities for transporting
and transshiping the petroleum. The major industrial facilities at
North Kenai, consisting of two refineries, an LNG plant, and a fertilizer
plant, complete the list of major consumers.
In Valdez, the oil pipeline Pump Station 12 and the facilities at
the pipeline terminal are the major consumers.
G-3
TABLE G.2. RAILBELT MILITARY ELECTRICITY NET GENERATION
(FY 1979)
(103 MWh)
Greater Anchorage
Fort Richardson Army Base
Elemdorf Air Force Base
Total
Greater Fairbanks
Fort Greely Army Base
Fort Wainwright Army Base
Eilson Air Force Base
Clear Air Force Base
Total
Total Railbelt
56.7
9?.8
155.5
14.4
36.8
47.0
80.0
178.2
333.7
Source: Military and Alaska Power Authority records.
G-4
. ~. '
TABLE G.3. RAILBELT SELF-SUPPLIED INDUSTRIAL
NET GENERATION
(103 MWh)
Area 1977 1978 1979
North Kenai 69.5 94.6 94.6
Valdez 39.4 54.8 54.8
Cook Inlet 208.9 _226.7 226.7
Interior Alaska 25.7 37.9 37.9
Total 343.5 414.0 414.0
Source: Alaska Power Authority worksheets •
G-5
In Fairbanks, the Univer$ity and· pipeline Pump Stations 8 and 9
are the major consumers.
In some cases, an industrial facility will both generate its own
electricity and purchase power from the local utility. For example,
Alyeska Pump Station 12 uses electricity provided by Copper Valley
Electric Association for all its needs except the pumps themselves,
which are powered by self-supplied generation. In other instances,
the facility may purchase power but maintain its own backup generation
capability.
The difficult-ies in attempting to project self-supplied industrial
electricity are that additions over time have been "lumpy" (large
but infrequent) and that there is not always a clear criterion to deter-
mine whether a particular consumer will choose to provide his electricity
from self generation rather than from utility purchased power. In some
instances, such as the case of offshore drilling and production plat-
forms,_ self-supplied electricity is the only practical method of ob-
taining power. In other cases, however, the industrial facility faces
a choice, and the decision will depend upon the cost of self-generated
electricity vs. the price of purchased electricity. Each instance will
be different depending upon, among other things, the type of load, the
capacity and load characteristics of the utility, and the resources
available to the facility for generating electricity compared to those
which the utility has available.
G-6
The utilities are presently supplying a portion of the large indus-
trial consumer load, even though in total the load is relatively small.
This is the case in Fairbanks for the refinery and a portion of oil
pipeline requirements, in.Valdez for a portion of oil pipeline require-
ments, and in the Greater Anchorage Area for the refineries and the LNG
plant as well as some petroleum production and transportation facilities.
Thus, a portion of increments to industrial requirements is already
included in the utility load projections. Self-supplied industrial
_requirements should be limited to new industrial uses that would not
normally be picked up by the utilities given the same general market
conditions in the future as in the past.
Having thus narrowed the definition of self-supplied industrial
requirements, it is still possible to identify two types of industrial
facilities. The first is any facility ~.;rhich chooses to locate in the
railbelt independent of the price of electricity, while the second is
any facility which chooses to locate in the railbelt because of price
of electricity or the availability of electricity. We address ourselves
to and consider only the first type of facility because the determina-
tion of whether such "electricity intensive" industries will locate in
Alaska in the future is a function of, among other things, future
electricity price, which is beyond the scope of the present study.
Table G.4 presents the self-supplied electricity requirements for
. those facilities identified in the economic scenarios. Three have
relatively small requirements which we assume to be incorporated in
G-7
TABLE G.4. PROJECTED ADDITIONS TO RAILBELT SELF-SUPPLIED
INDUSTRIAL ELECTRICITY REQUIREMENTS
(103 MWh annual)
Facility Economic Scenario
Minimum Likely Maximum
,Consumption Start
. (lo3 Ml;.fu) Date
Consumption Start
(103 Ml;.fu) Date
Consumption Start
(103 Ml;.fu) ~
Pacific Alaska
LNG a 0 127 1985 127
Alpetco Refinery b 0 30 1985 306
Cook Inlet Oil 0 0 46 c Development
Fairbanks aetro-
chemicals 0 0 88
Northwest Alaska (
Gas Pipeline Incorporated in utility sales forecast.
State Capital Move Incorporated in utility sales forecast.
Beluga Coal
Development Incorporated in utility sales forecast.
~omer Electric Association, Power Requirements Study, 1977.
bCalculated from Alpetco Refin~ry Environmental Impact Statement,
Volume II, p. 343 and, 413. Assuming 51.5 MW operating at. 80 percent
capacity, 85 percent of the year for high case. Authors' estimate of
basic refinery requirement for medium case.
1985
cAssumes a 20 percent increase from present requirements.
dWith a capacity for processing 200 million cubic feet/day of
royalty gas and if 8 percent of input is used for processin~ and if
25 percent of processing is steam and electricity, 4.0 x 10 btu daily
would be used. (Based upon same proportions as the Alpetco refinery.)
This can be converted to electricity by assuming a 25 percent conversion
efficiency and 80 percent load factor.
G-8
the utility projections. Of the other facilities, they all could have
their electricity requirements met by utilities if it were available,
except perhaps in the case of additional oil development in the Cook
Inlet. Their requirements are large enough, however, that they should
be treated, for projection purpose, as additions to, rather than com-
ponents of, the utility electricity requirement.
G-9
APPENDIX H
HISTORICAL ELECTRICITY SALES AND ECONOMIC DATA
::X::
I
N
RAILBELT TOTAL ELECTRICITY SALES
(GWh)
GREATER ANCHORAGE AREA
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979p
Chugach Anchorage Matanuska
Electric Municipal Electric.
Association Light & Power Association
165 133 34
190 143 36
207 159 39
235 170 39
262 190 44
309 222 50
368 254 61
435 288 72
485 326 81
517 350 92
596 405 118
668 468 147
727 492 172
781 498 223
803 523 232
PPre1iminary, based on data from various sourcesf
* Approximate.
TO FINAL CONSUMERS
Homer Seward
Electric Electric
Association System Total
31 6 369
39 7 415
49 7 461
67 8 519
82 9 587
93 10 684
103 11 797
99 13 906
105 14 1,010
112 14 1,086
133 17 1,270
161 18 1,463
194 17 1,603
224,.~ 20 1,747
247
TABLE H.l. (continued)
RAILBELT TOTAL ELECTRICITY SALES TO FINAL CONSUMERS (Continued)
(GWh)
GLENNALLEN-
GREATER FAIRBANKS AREA VALDEZ AREA RAILBELT TOTAL
Fairbanks Alaska Power
Golden Valley Municipal and Copper Valley
Electric Utilities Telephone Electric
Association System Tok Total Association
1965 50 47 1 98 6 473
1966 59 49 ** 108 "'* 523
1967 66 *~" ** 66 'lc'l; 527
::d 1968 84 58 *-1• 141 ~""~• 661
I 1969 104 66 w ~"* 170 '"* 758
1970 136 75 2 213 9 907
1971 175 . 76 ** 251 10 1,059
.1972 190 70 2 262 6**," 1,174
1973 206 81 3' 290 ll 1,311
1974 231 88 3 322 14 1,422
1975 300 110 3 413 24 1,707
1976 306 114 3 423 33 . 1,920
1977. 324 118 5 447 42 2,092
1978 310 116 5 432 38 2,217
1979p 302 37
PPreliminary, based on data from various sources.
TABLE H.2. HISTORICAL RESIDENTIAL UTILITY SALES
H-4
~-
..
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Chugach Electric Association (CEA)
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
1965 111,587 15,446 7,224
1966 128,484 16,487 7,793
1967 134,985 17,037 7,923
1968 148,591 19,893 7,470
1969 . 166~146 22,036 7,540
1970 198,856 24,682 8,057
1971 236,857 25,761 9,194
1972 269,252 28,687 9,386
1973 287,484 29,077 9,887
1974 305,739 31,779 9,.621
19.75 359,922 34,031 10,576
1976 397,846 35,960 ll,064
1977 432,070 41,025 10,532
1978 472,040 • 43,542 10,841
1979p 477,189 42,761 l1,161
PPreliminary, based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-5
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979p
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Anchorage Municipal Light and Power (AMLP)
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(H\Vh) (kWh)
34,656 6,664 5,201
35,o56 6-,516 5,380
38,426 6,894 5,574
42,825 7,544 5,677
47,781 8,043 5,941
54,518 8,477 6,431
63,038 9,295 6,782
72,993 10,130 -7,206
82,663 10,838 7,627
89,946 11,674 7,705
105,214 11,803 8,914
119,475 12,353 9,672
117,986 13,605 8, 672
115,638 14,374 8,045
116,211 13,517 8,597
PFreliminary, based on data from various sources.
NOTE: Year-end customer data overstated in 1977 and 1978 due to a
computer error within Municipality of Anchorage.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-6
-·:·,
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Matanuska Electric Associati~n (MEA)a
Year-End Energy Delivered b
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
1965 16,628 2,688 6,186
1966 18,012 2,707 6 ,'554
1967 19,623 3,022 6,493
1968 20,760 3,174 6,541
1969 24,861 3,611 6,885
1970 29,416 3,975 7;400
1971 37,791 4,281 8,828
1972 44,147 4,669 9,455
1973 51,026 5,045 10,114
1974 59,764 6,153 9, 713
1975 77,592 6,834 11,354
1976 96,280 7,681 12,535
1977 112,662 8,321 13,539
1978 152,133 10,152 • 14,986
1979p 157,889 10,362 15,237
aPalmer and Talkeetna stations.
bFarm (including irrigation) and nonfarm residential.
PPreliminary, based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-7
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Homer Electric Association (HEA)a
Year-End Energy Delivered b
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
1965 7,526 1,569 4,797
1966 9,809 1,753 5,596
1967 12,402 2,441 5,081
1968 17,673 3,182 5,554
1969 20,200 3,296 6,129
1970 22,768 3,312 6,874
. 1971 27,267 3,431 7,947
1972 28,299 3,491 8,106
1973 30,849 3,708 8,320
1974 33,752 4,215 8,008
1975 44,008 4, 773 9,220
1976 55,859 5,508 10,141
1977 70,742 7,346 9,630
1978 94,846 7,904 12,000
1979p 108,973 8,764 12,434
aHomer and Kenai stations.
bFarm and nonfarm residential customers.
PPreliminary, based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-8
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
. 1976
1977
1978
1979
*
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Seward Electric System (SES)
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(Mllli) (kWh)
3,169 649 4,883
3,073 656 5,439
2,987 634 4,711
3,179 650 4,891
3,481 667 5,219
3,771 705 5,349
4,101 718 5,712
4,535 730 6,212
4,711 765 6,158
4,664 785 5,941
5,120 885 5,785
5,632 911 6,182
6,020 978 6,155
6,807 1,027 6,628
* *" *
Not available
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-9
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
·1978
GREATER ANCHORAGE AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Total Anchorage Area System
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
173,566 27,016 6,425
194,434 28,028 6,937
208,423 30,028 6,941
233,028 34,443 6,766
262,469 37,653 6,971
309,329 41,151 7,517
369,054 43,486 8,487
419,226 47,707 8,788
456,733 49,433 9,239
493,865 54 ·606 9,044
591,856 58,326 10,147
675,092 62,413 10,817
739,480 71,275 10,375
841,464 76,999 10,928
H-10
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979p
GREATER FAIRBANKS AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Golden Valley Electric Association (GVEA)
Edergy Delivered Year-End
To .Final· Customers Customers Consumption/Customer
(MWh) (kWh)
23,142 4,036 5,734
29,184 4,213 6,927
33,444 4,402 7,597
41,917 . 4, 95'7 8,456
54,569 5,459 9,996
67,123 6,224 10,785
81,960 6,741 12,158
96,702 6,947 13,920
. 106,882 7,382 14,479
127,873 8,643 14,795
160,199 9,243 17,332
162,369 10,680 15,203
168,275 12,443 p,524
150,804 13,030 11,574
142,960 13,711 10,427
PPreliminary, based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-11
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
*
GREATER FAIRBANKS AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Fairbanks Municipal Utility System (FMUS)
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
16,172 4,147 3,900
17,485 3,957 4,419
* * *
19,461 4,387 4,436
22,327 4,564 4,892
23,419 4,532 5,167
24,456 4,443 5,504
24,248 4,540 5,341
25,952 4,443 5,841
25,909 4,618 5,610
30,181 4,634 6,513
31,302 4,739 6,605
29,497 4,754 6,205
27 ,109· 4,494 6,032
* * *
Not available
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-12
GREA~ER FAIRBANKS AREA
RESIDENTIAL ELECTRICITY CONSUtWTION
Alaska Power and Telephone, Tok
Energy Delivered • y;ear-End
To Final Customers Customers Consumption/Customer
(MHh) (kWh)
1965 142 * * 1966 * * * 1967 * * *
1968 * * * 1969 * * * 1970 279 * *
1971 * * * 1972 396 * * 1973 411 * *
1974 470 * * 1975 603 * * 1976 730 * *
1977 795 155 5,129
1978 870 * * 1979 * * *
* Not available
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-13
1965
1966
1967
1968
1969
.1970
1971
1972
1973
1974
1975
1976
1977
1978
a
*
GREATER FAIRBANKS AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
a Total Fairbanks Area System
Energy Delivered ·Year-End
To Final Customers Customers Co~sumption/Customer
(HWh) (k\fu)
39,314 8,183 4,804
46 '669' 8,170 5, 712
* * *
61,378 9,344 6,569
76,896 10,023 7,672
90,542 10,756 8,418
106,4.16 11,184 9,515
120,950 11,487 10,529
132,834 11,825 11,233
153,782 . 13,261 11,600
190,380 13,877 13,719
193,671 15,419 12,561
197 '772 17,197 11,500
177,913 17,524 10,153
Net of Tok
Not available
H-14
GLENNALLEN-VALDEZ AREA
RESIDENTIAL ELECTRICITY CONSUMPTION
Copper Valley Electric Association (CVEA)a
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(~fu) (kWh)
1965 1,445 432 3,345
1966 * * *
1967 * * *
1968 * * *
1969 * * *
1970 2,133 561 3,802
1971 2,611** 676** 3,862
1972 1,528 324 4, 716
1973 2,887 680 4,246
1974 3,751 935 4,012
1975 7,656 1,48.7 5,149
1976 10,234 1,758 5,821
1977 10,895 1,601 6,805
1978 9,545 1,539 6,202
1979p 9,354 1,588. 5,890
a Glennallen and Valdez stations.
PPreliminary, based on data from various sources.
* ** Not available Valdez only
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-15
TABLE H.3. HISTORICAL CONMERCIAL-INDUSTRIAL-
GOVERNMENT CONSUMPTION DATA
H-16
GREATER ANCHORAGE AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Chugach Electric Association (CEA)
Energy Delivered
To Final Customersa
Year-End b
Customers Consumption/Customer
(m.Jh) (kWh)
1965 52,920 964 51,605
1966 60,601 1,047 53,626
1967 71,561 1,135 57,532
1968 84,513 1,381 55,652
1969 94,565 1,678 51,544
1970 108,298 2,040 53,087
1971 128,675 2,126 60,525
1972 163,566 2,449 66,789
1973 194,973 2,579 75,600
1974 208,855 2,835 73,670
1975 231,377 3,036 76,211
1976 264,731 3,494 75,767
1977 289,394 4,208 68' 772
1978 303,263 4,331 70,021
1979p 320,365 4,414 72,579
aCommercial and other (public authorities).
b Other only since 1970.
cOther only since 1970.
PPreliminary, based on data from various sources.
c
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-17
_GREATER ANCHORAGE AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Anchorage Municipal Light and Power (AMLP)
· Energy Delivered Year,-End
To Final Customersa Customers Consumption/Customer
(MWh) (k'tfu)
1965 92,889 2,071 44,852
1966 104,663 2,058 50,857
1967 116,157 2,060 56,387
1968 121,490 2,107 57,660
1969 135,306 2,115 63,974
1970 159,538 2,159 73,894
1971 181,374 2,226 81,480
1972 205,288 2,315 88,677
1973 233,312 2,350 99,282
1974 250,409 2,417 103,603
1975 289,296 2,464 117,409
1976 339,550 2,675 126,935
1977 365,510 2,800 130,539
1978 372,511 2,885 129,120
1979p 396,811 2,933 135,292
aCommercial and industrial.
PPreliminary, based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-18
GREATER ANCHORAGE AREA
CO}lliERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Matanuska Electric Association (MEA)a
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(MHh) (kWh)
1965 16,441 412 39,905
1966 17,187 425 40,440
1967 18,172 490 37,086
1968 17,471 500 34,942
1969 18,148 511 35,515
1970 19,311 594 32,510
1971 22,239 599 37,127
1972 26,264 675 . 38,910
1973 28,252 739 38,230
1974 30,630 800 38,288
1975 38,756 980 39,547
1976 48,296 1,128 42,816
1977 57,263 1,265 45,267
1978 66,699 1,307 51,032
1979p 71,255 1,315 54,186
aPalmer and Talkeetna stations.
PPre1iminary,.based on data from various sources.
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-19
GREATER ANCHORAGE AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
.ELECTRICITY CONSUMPTION
Homer Electric Association (HEA)a
Year-End Energy Delivered b
Tc Final Customers Customers c Consumption/Customer
(~.fu)
1965 23,419 416
1966 28,707 493
1967 30,705 561
1968 49,421 687
1969 63,155 705
1970 69,887 813
1971 75,955 861
1972 70,382 797
1973 74,194 831
1974 78,517 981
1975 88,714 1,066
1976 105,239 1,232
1977 122,512 1,355
1978 129,493 1,422.*
1979p 137,727 **
a d K · · Homer an ena~ stat~ons.
bCommercial, industrial, and public buildings.
cPublic buildings only since 1970.
dPublic buildings only since 1970.
PPreliminary, based on data from various sources.
* ** Yearly average Not available
(kWh)
44,647
48,032
.52' 989
63,086
72,852
85,962
88,217
88,308
89,283.
80,038
83,221
85,421
90,415
91,064
**
d
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-20
-.·"·{
GREATER ANCHORAGE AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSU}~TION
Seward Electric System (SES)
Energy Delivered
To Final Customersa
Year-End b
Customers Consumption/Customer
(MHh)
1965 2,989 131
1966 4,109 124
1967 4,267 117
1968 4,367 129
1969 4,814 116
1970 5,695 178
1971 6,513 194
1972 . 7,680 184
1973 8,436 194
1974 8,640 199
1975 11,174 204
1976 12,080 260
1977 10,842 232
1978 12,409 274
1979 * *
aCommercial, industrial, and public authorities.
bPublic authorities only since 1970.
cPublic authorities only since 1970.
* Not available
(k\Vh)
15,349
17,702
14,197
14,969
18,552
31,994
33,572
41.739
43,485
43,417
54,775
46,462
46,733
45,288
*
c
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-21
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
GREATER ANCHORAGE AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Total Anchorage Area System
Year-End Energy Delivered
To Final Customers Customers a Consumption/Customera
(Mt.fu) (kWh)
188,658 3,994 47,235
215,267 4,147 51,909
240,862 4,363 55,206
277,262 4,804 57' 715
315,988 5,125 61,656
362,729 5,784 62,.713
414;756 6,006 69,057
473,180 6,420 73,704
539,167 6,693 80"557
577,051 7,232 79~791
659,317 7,750 85,073
769,896 8,789 87,598
845,521 9,860 85,753
884,375 10,219 86,542
aNumber of customers are slightly underreported and
average consumption calculations are slightly overestimated
prior to 1970 due to incomplete information prior to 1970 on
the following customer categories: Chugach "other," Homer
public buildings, and Seward public authorities.
H-22
.';~~
J
'l
.
'
GREATER FAIRBANKS AREA
COMMERCIAL-INDUSTRIAL-GOVERN}ffiNT
ELECTRICITY CONSUMPTION ·
Golden Valley Electric Association (GVEA)
Energy Delivered Year-End
To Final Customers Customers Consumption/Customer
(MWh) (kWh)
1965 25,850 523 49,426
1966 28,982 591 49,039
1967 30,830 576 53,524
1968 41,585 634 65,915
1969 49,284 703 70,105
1970 69,064 844 81,829
1971 84,295 914 92,226
1972 92,758 916 101,264
1973 98,744 973 101,484
1974 102,342 1,132 90,408
1975 133,972 1,209 110,812
1976 138,735 1,365 101,637
1977 155,426 1,649 94,255
1978 157,202 1,675 93,852
1979p 155,436 * *
PPreliminary, based on data from various sources.
* Not available
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-23
GREATER FAIRBANKS AREA
C0~1MERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Fairbanks Municipal Utility System (FMUS)
Energy Delivered Year-End b
To Final Customersa Customers
(MWh)
1965 29,348 795
1966 30,.394 876
1967 * *
1968 36,321 835
1969 41,928 876
1970 49,496 1,044
** ** 1971 48,761** 1,015**
1972 43,115 1,086
1973 52,079 1,081
1974 59,273 1,110
1975 76,787 1,133
1976 80,440 1,165
1977 85,037 1,185
1978 85,466 1,179
1979 * *
aCommercial and other (public) categories.
b Other only since 1970.
c Other only since 1970.
Consumption/Customer
(kWh)
27,810
25,862
*
33,101
36,397
47,410
48,040
39,701
48,177
53,399
67 '773
.69,047
71,761
72,490
*
* ** Not available. Based on author's estimate of street
lighting requirements due to aberration in reporting method for
these years.
c
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-24
GREATER FAIRBANKS AREA
CO}lliERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Alaska Power and Telephone, Tok
Energy Delivered Year-End
To Final Customersa Customers Consumption/Customer
(MWh)
1965 1,343 * 1966 * * 1967 * *
1968 * * 1969 * * 1970 1,537 *
1971 * * 1972 1,949 * 1973 2,663 *
1974 2,526 * 1975 2,652 * 1976 2,730 *
1977 4,089 75
1978 4,514 * 1979 * *
aCommercial, industrial, and other categories.
* Not available
(kWh)
* * *
* * *
* * *
* * *
54,520
* *
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-25
GREATER FAIRBANKS AREA
COMMERCIAL-INDUSTRIAL-GOVERID-lENT
ELECTRICITY CONSUMPTION
Total Fairbanks Area Systema
Energy Delivered Year-End b
Consumption/Customerb To Final Customers Customers
(MWh) (kWh)
1965· 55~198 1,318 41,880
1966 59,376 1,467 40,474
1967 * * *
1968 77 ~906 1,469 53,033
1969 91,212 1~579 57,766
. 1970 118,560 1,888 62~797
1971 133,056 1,929 68,977
1972 135~873 2,002 67,869
1973 150~823 2,054 73,429
1974 161,615 2,242 72,085
1975 210,759 2,342 89~991
1976 219~175 2,530 86,630
1977 240,463 2,834 84,849
1978 242,668 2~854 85,027
a Net of Tok and University of Alaska.
b Number of customers are slightly underreported, and average
consumption calculations are slightly overestimated prior to 1970
due to incomplete information prior to 1970 on the following cus-
tomer categories--FMUS other (public).
* Not available
H-26
*
GREATER FAIRBANKS AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
University of Alaska
Energy Delivered To
Final Consumers
(MWh)
1965 * 1966 * 1967 *
1968 * 1969 * 1970 21,768
1971 * 1972 29,567
1973 31,913
1974 27,646
1975 28,259
1976 27,195
1977 25,644
1978 * 1979 *
Not available
SOURCE: Federal Energy Regulatory Commission, Power System
Statement.
H-27
GLENNALLEN-VALDEZ AREA
COMMERCIAL-INDUSTRIAL-GOVERNMENT
ELECTRICITY CONSUMPTION
Copper Valley Electric Association (CVEA)a
Energy Delivered b Year-End
To Final Customers Customers c Consumption/Customer
(MWh) (k\fu)
1965 4,188 141 29,702
1966 * * *
1967 * * *
1968 * * *
1969 * * *
1970 7,235 240 30,146
1971 7,657** 249** 30,751
1972. 3,842 122 31,492
1973 8,130 371 21,914
1974 10,193 354 28,794
1975 16,062 426 37,704
1976 22,465 455 49,374
1977 31,307 491 63,762
1978 28,604 495 57,786
1979p 26,917 492 54,709
aGlennallen and Valdez stations.
bCommercial, industrial, and other (public buildings).
cPublic building customers only since 1970.
PPreliminary, based on data from various sources.
* ** Not available Valdez only
c
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-28
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
TABLE H.4.
MISCELLANEOUS ELECTRICITY CONSUMPTION
(Primarily Street Lighting)
(MWh)
GREATER GREATER
ANCHORAGE AREA FAIRBANKS
6,907 2,429
5,439 2,585
6,322 1,560a
8,875 2,207
10,257 2,216
11,845 2,289
13,682 11,197
14,086 2,984
15,940 3,361
16,609 3,354
18,619 8,945
18,.542 7,193
17,707 3,467
21,362 5,864
aGolden Valley Electric Association only.
* Not available.
AREA
GLENNALLEN-
VALDEZ AREA
0
* *
* * 115
136
134
151
97
130
152
171
182
SOURCE: Federal Energy Regulatory Commission, Power System Statement.
H-29
TABLE H.5. HISTORICAL RAILBELT EMPLOYMENT
H-30
f,
{;
GREATER ANCHORAGE AREA
'(. NONAGRICULTURAL WAGE AND SALARY EMPLOYMENTa
):.
(thousand) ~',
J:
t.
I Census Division
,.
~'· ~· "" Matanuska-~:.
~: Year Anchorage Kenai-Cook Inlet Susitna Seward Total ~ ... ,,
::·: 1965 30,704 1,756 1,085 621 34,166
1966 31,519 2,465 1,139 645 35,.768
1967 32,958 3,678 1,07.5 638 38,349
1968 34,021 4,470 988 602 40,081
1969 37,789 4,144 1,002 626 43,561
1970 39,667 3,546 1,142 689 45,044
1971 44,616 3,454 1,415 774 50,259
1972 48,252 3,818 1,447 809 54,326
1973 50,630 4,049 1,610 862 57,151
1974 58,716 4 ,.487 1,786 934 65,923
1975 69,561 5,595 2;149 1,152 78,457
1976 73,019 6,473 2,398 1,137 83,027
1977 76,997 7,340 2,653 1,155 88,145
1918 76,942 6,565 3,083 1,226 87,816
aDoes not include active-duty military or self-employed.
SOURCES: 1975 through 1978: Alaska Department of Commerce and Economic
Development, Division of Economic Enterprise, "Numbers: Basic
Economic Statistics of Alaska Census Divisions," November 1979.
1965 through 1974: Alapka Department of Labor.
H-31
Year
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
GREATER FAIRBANKS AREA
NONAGRICULTURAL HAGE Al.'iD SALARY EMPLOYMENTa
(thousand)
Census Division
Fairbanks Southeast Fairbanks
28,740 2,044
28,118 2,615
24,868 819
21,662 874
Total
11,511
11,767
11,929
12,383
13,901
14,377
14,648
15,583
15,480
19,749
30,784
30,733
25,687
. 22,536
~oes not include active-duty military or self-employed.
SOURCES: 1975 through 1978: Alaska Department of Commerce and
Economic Development, Division of Economic Enterprise,
"Numbers: Basic Economic Statistics of Alaska Census
Divisions," November 1979.
1965 through 1974: Alaska Department of Labor.
H-32
a
GLENNALLEN-VALDEZ AREA
NONAGRICULTURAL WAGE AND SALARY EMPLOYMENTa
(thousand)
Year Valdez-Chitina-Whittier
1965 757
1966 826
1967 773
1968 667
1969 786
1970 859
1971 1,087
1972 906
1973 988
1974 1,529
1975 4,763
1976 8,049
1977 4,199
1978 2,043
Does not include active-duty military or self-employed.
SOURCES: 1975 through 1978: Alaska Department of Commerce and
Economic Development, Division of Economic Enterprise,
"Numbers: Basic Economic Statistics of Alaska Census
Divisions," November 1979.
1965 through 1974: Alaska Department of Labor.
H-33
TABLE H.6. HISTORICAL RAILBELT HEATING DEGREE DAYS
H-34
Season
1960-61
1961-62
1962-63
1963-64
1964-65
1965-66
1966-67
1967-68
1968-69
1969-70
1970-71
1971-72
1972-73
1973-74
1974-75
1975:-76
1976-77
1977-78
1978-79
Normal
GREATER ANCHORAGE AREA
HEATING DEGREE DAYS
(Annual)
Anchorage Homer
10,261 10,026
11,524 10,696
10,406 9,966
10,781 10,108
11,064 10,577
11,174 10,840
11,384 10,278.
9,997 9,793
10,779 10,410
9,351 9,446
11,670 11,340
12,077 11,471
11,555
11,223 10,807
10,868 10,389
11,115 10,694
9,050 8,901
9,701 9,668
9,395
10,911 10,364
Talkeetna
11,160
12,621
11,304
11,845
11,991
12,499
11,947
11,364
11,851
10,631
13,280
13,406
12,203
12,235
12,081
12,340
10,268
10,984
11,708
SOURCE: "Local Climatological Data," National Oceanic and Atmospheric
Administration, Environmental Data and Information Service,
National Climatic Center, Asheville, N.C.
H-35
GREATER FAIRBANKS AREA
HEATING DEGREE DAYS
(Annual)
Fairbanks
Season
1960-61 14,009
1961...;.62 14,786
1962-63 13,692
1963-64 14,912
1964-65 15,009
1965-66 15,688
1966-67 14,544
1967-68 13,671
1968-69 14,664
1969-70 12,939
1970-71 15,215
1971-72 15,141
1972-73 13,547
1973-74 14,391
1974-75 13,808.
1975-76 14,683
1976-77 12,674
"1977-78 13,635
1978-79 13,802
Normal 14,344
SOURCE: "Local Climatological Data," National Oceanic and Atmospheric
Administration, Environmental Data and Information Service,
National Climatic Center, Asheville, N .~c.
H-36
f."
~· ~.·. ~t 1~-•.... . .
..
~·-
1,.· .
~
I •••
I l~ ..
',
'
~-
GLENNALLEN-VALDEZ AREA
HEATING DEGREE DAYS
(Annual)
Gulkana
Season
1960-61 13~707
1961-62 14~551
1962-63 13~227
1963-64 14,183
1964-65 14,062
1965-66 15,054
1966-67 14,527
1967-68 13,290
1968-69 14~594
1969-70 12,975
1970-71 15,002
1971-72 15,810
1972-73 14,266
1973-74
1974-75 13,768
1975-76 14,061
1976-77
1977-78 13,966
Normal 13,937
Valdez
10,682
9' 911.
10,255
9,130
9,422
10,545
SOURCE: "Local Climatological Data," National Oceanic and Atmospheric
Administration, Environmental Data and Information Service,
National Climatic Center, Asheville, N.C.
H-37
TABLE H.7.
SEASONAL VARIATION IN ELECTRICAL ENERGY GENERATION
Summer Winter
July-August Sales December-January Sales
Percent Percent of Percent Percent of
of Annual Monthly Average of Annual Monthly Average
SOUTH CENTRAL
1964 13.1 78.6 21.1 126.6
1965 13.6 81.6 20.9 125.4
1966 14.2 85.2 21.0 126.0
1967 13.5 81.0 21.1 126.6
1968 13.1 78.6 22.0 132.0
1969 13.5 81.0 21.0 126.0
1970 13.4 80.4 22.1 132.6
1971 12.8 76.8 21.3 127.8
1972 12.5 75.0 21.9 131.4
1973 12.9 77.4 20.8 124.8
1974 13.2 79.2 20.8 124.8
1975 13.0 78.0 21.7 130.2
1976 13.0 78.0 20.2 121.2
1977 12.8 76.8 21.7 130.2
1978 13.6 81.6
INTERIOR
1964 14.4 86.4 21.2 127.2
1965 12.5 75.0 21.4 128.4
1966 13.0 78.0 22.3" 133.8
1967 14~9 89.4 21.4 -128.4
1968 . 14.6 87.6 22.5 135.0
1969 12.0 72.0 22.9 137.4
1970 11.7 70.2 25.1 150.6
1971 11.1 66.6 22.9 137.4
1972 11.4 68.4 23.6 141.6
1973 11.5 69.0 22.2 133.2
1974 11.2 67.2 23.7 142.2
1975 10.8 64.8 24.0 144.0
1976 11.9 71.4 21.8 130.8
1977 12.6 75.6 23.3 139.8
1978 12.9 77.4
SOURCE: Compiled from data supplied by the Alaska Power Administration.
H-38
APPENDIX I. A REVIEW AND COMPARISON OF
RAILBELT ELECTRIC PmVER REQUIREHENT PROJECTIONS
I.l. Susitna Riv.er Basin: A Report on the Potential Development of
Water Resources in the Susitna River Basin of Alaska by U.S.
Department of Interior, sponsored and prepared by the Bureau
of Reclamation, Alaska District Office, 1952.
This report did not present a detailed analysis of the electric load
beyond the accompanying Figure I.l which depicts a 30-year forecast.
Residential space heating was assumed to be 50 percent electric by 1970,
resulting in an average residential use of 10,000 KWh annually and total
consumption of 635,000 }fifu. The market area (equivalent to the present
study minus the Glennallen-Valdez area) was projected to have a popula-
tion of 260,000 in 1970. Farm requirements were projected at 66,000 Ml~
based on 3,800 to 4,000 farms by 1970. Commercial and municipal require-
ments were projected at 256,000 Ml~ for the same year. Projected large
industrial electricity users included an electrified railroad, minerals,
petroleum, synthetic fuels from coal, and local industries together con-
! suming 1,732,000 MWh in 1970.
~
;l
1.2. Devils Canyon Project: Alaska, Feasibility Report by U.S.
Department of Interior, Bureau of Reclam~tion, Alaska District,
1960.
Gross generation requirements in this study are projected for 22 years
for utilities, large industry, and military as shown in Figure I.2 and
Table I.l. Simple growth rates ranging from 10 to 12 percent annually were
FIGURE 1.1.
(/)
n::
~
0
H I I
N
1-
1-
<!
3:
0
_j -X':
z
0 -
....J
....J -
::'!
-POWER REQUIRMENT
L_
-/
1---BY CLASS OF USE /
t--4 800 ACTUAL a ESTIMATED FUTURE /
L1 t--_/·
lv / v
1---,~
"' / ./ 1---'><v~ / v
-4000 TOTAL POTENTIAL /
-ENERGY REQUIREMENT <v'?' / v
f---' I ~ /~"'v
f /<v~CJ/
I--I /~~/ t--3200
t--/ Vc.,<y
/ / 'V
1---ACTUAL ESTIMATED FUTURE I / / .... ~-t--
-<-_,_ "'~ I--
t--2400 I / 00
I v v ~'>
t--'
t--J 1/. v 0<v
-I I "?-~
I /I 'V
-1600 j I v -
-I 'I
r--./ V; ,......., V, v -f---8 00 v v --"'
1--/ L ~~LII AR'f ,...../ 1---_,..,..v l--c--v 1---~ I I 1---·~ ~ ~ f.---:--OTHER -r--~ 1--~ /_ ..-=
194 0 ·44 48 52. 56 60 64 68 72 76 80
GALAN D E.. Y. E R..S
0
Q) 3,400
3,200
10
"-
3,000
2,800
2,600
2,400
2,200 If)
0:::
::::>
0
I
2,000
t-
t-
), c::x:
1,800 3:
0
..J
~
1,600 LL..
0
If)
z
1,400 0
...J
...J
1,200 ~
1,000
800
co
<t
600
<t 400
<t
200
0
<t
()) -
FIGURE 1.2 .
rz3 . Utility
[Sj ·Lorge Industrial
~ 40 Of.:, of Military
Recorded Projected
Firm Energy Generation
With Interim Capacity
Needed-Assumed To
Be Steam
60 65
CALENDAR
I-3
// M i scellaneous Hydro
/////////////////
70 75
YEARS
i
!
·I
I
I
'
TABLE I.l.
Projection of Total Energy Load
Devil Canyon Power Market Area
Unit: Million kilo,.mtt-hours
.:..:. ita:;:y 'l'o·:~a:f.
Calendar Ncnr~.:l.litary lot·.n.s 1o<=,d 1oat't
·-----Sup:Pl:.ed Supplied supplied supplied
Year Largo by by by by
Utility inO.ustria.l Tot.a.1 .t. ,.1 '+ \!• .• 1.. J. .,y p:c'l),')':!~t prc,i~ct pro.~;~t
1960 287 22 309 309
H 1961 317 22 339 339
I 1962 351 19 370 370 ~
1963 388 16 404 4ot~
1964 429 14 443 443
1965 474 19 . 493 1 ~93
1966 524 21 545 545 ' ...
1967 580 22 602 602
1968 6lJ.1 25 666 666
1969 711 42 773 270 503 158 661
1970 789 51 • .8110 2'!0 5?0 158 728
1971 876 6o 936 270 . 665 158 82!+
1972 972 75 1,01~7 270 777 158 935
1973 1,077 99 1 ;7"' 270 oo6 158 1~06'-~ , ... 0 "' 1974 1,191 137 1,328 270 1,058 15G 1 o~~" 1::..J.O
1975 1,315 184 1,499 270 1,229 158 1,387
1976 1,4!.~9 2h2 1)691 270 1, l~?.l 158 1,579
1977 1 '::;\CI:+ 305 1 ~··::9 270 1,()29 158 1,~(87 s -.. 1 _,
1978 1,75J 384 2,13'7 270 1,867 158 2,025
1979 1,92~~ 463 2,39). 270 2,121 1~:3 2,~279
1980 2 ,:'...21 51~.:1 ,..., 1;-63 2j'O 2J 3~J3 158 2,5,1 '-1-.J 1.981 2,333 . 6l6 2,949 270 2,6'(9 158 2,837 1.982. 2,566 695 3,261 519 2 .~7~2 153 2,900
------·-----
·0
)(")
v\
' ... . 0J
I •• ru
r-1
\,;)
CIJ
" ('I')
"' ,(l
Ll'\
" C'J
TABLE I. 2.
Projection of Alaska's Electric Power Requirements, Electric Utility Systems, and Nonutility lnstallaHons
Region and type of load Load center
number
(fig. 4)
Northwest. ................................... .
Utility ..•................ 1,2,3 ......... .
Do.................. (') ·
Nonutility. . . . . . . . . . . . . . . . (')
Southwest. .. · ......•.................•........
Utility ................... 4, 5, 5 ......... .
Do.................. (')
. Nonutility (1) ••••• 0 ••••••••• 0
<:>outncenrral ..................................
Anchorage-Kenai .......... 9 to 13 .........
Utility ...............................
Nonutility (military) .............•.....
Other areas ............... 7. 8. 14. !.), .....
Utility .............................. .
Nonutility ........................... .
Utility.. . . . . . . .. . . .. . (1)
Nonutility (1) 0 0 ••••••••••
Interior ......................................
Fairbanks ................. 15 ............•
Utility ...............................
Nonutility (military) ...................
Utility •...•.......... (1) •••••••••••••
Nonutility ............ (') .............
Southeast .....................................
Utility ..•...........•.... I7 to 24· ........
Nonutility (industrial) ...... 20 and 23 ......
Utility ................... (1) .............
Non utility .........•...... (1) .•...........
Total utility requirement. ................
Total nonutility require-
ment .................................
Total Alaska ..................... · ....... ·····
I Scattered uonload center loads.
1965
Energy
mwh.
52,927
8, 219
468
44,2·10
154,293
7, 038
I, 255
146 000 ' 643,473
563, 749
406,604
157, I45
56. 030
22, 917
33, 113
7,494
16 200 ' 368,860
239,669
I05, 857
132,802
2, 191
I27, 000
419,942
155,023
246, 621
2, 298
16,000
720,371
919, 121
1, 639,495
Peak de-
mand
(mw.)
12.09
1. 86
• I8
10.05
35. I1
I. 55
. 25
33 30
144.07
126. 51
92.65
33.85
11. 75
5.05
6. 70
2. 10
3 70
81. 89
52. 23
25. 16
27.07
• 55
29. II
59.84
33. 75
31. 60
. 78
3. 70
163.92
179.08
343 .00
1975 1985
Energy
mwh.
Peak de-Energy
mand mwh·.
Peak de-
mand
(mw.)
72~ 690
I8, 100
700
53,890
I89, 800 .
I2,800
2,000
175 000 ' I, 184,240
1, 364, 720
I, 137, 840
226, 880
88 620
50,660
37,960
ll, 600
19 300 ' 654, 130
500, 110
275,850
2U,260
4,020
I 50,000
609,050
323,370
263,000
3,680
19,000
I, 840,620
I, 169,290
3,009, 9!0
(mw.)
I6. 52
4.I2
.30
12. 10
43.I5
2. 85
. 40
39 90
324.48
297.79
249. 79
48.00
19.29
Il. 09
8.20
3.00
4 40
H4. 71
109. 25
54.26
45.00
. 95
34.50
111. 04
70.79
35.00
• 85
4.40
408.40
231.50
639. 90
IlO, 680
44,790
I, 100
64, 790
237,990
24·, 790
3,200
210 000
3, 647, 890
3,412,090
3, 201, 190
240,900
165 000
118,690
46,310
I8,900
2I 900 ' 1, 145,680
967,980
72I, 350
246,G30
6, 700
I71,000
959, 730
668,630
263,000
6,100
22,000
4, 815,440
I , 286,530
6, 101, 970
25. I9
10.24
.35
14.60
51.05
5.5I
.55
45 00
784.94
739.49
689.49
50.00
35.85
25.95
9.90
4.60
5 00
256.29
215.89
I 54.69
51. 20
I.40
39.00
183. 24
141. 94
35.00
1. 30
5.00
1, 046.02
254. 70
1, 300. 72
NOTE.-1965 utility actual, non utility partly estimated; 1975 and 1985 estimated.
I-5
assumed for the utility load. Large industrial requirements were
generally unspecified but were projected to total 695,000 llivh by 1982.
Military requirement s were assumed to remain constant at a level of
360,000 MWh annually.
1.3. Alaska Power Survey, Federal Power Commission, 1969.
Utility and non-utility power requirements were projected to 1985
for all regions of the state. As shown in Table 1.2, the sum of the
Southcentral and Interior requirements (excluding Glennallen-Valdez and
large industrial but including military) were projected to be 4,410,000
MWh in 1985. This projection of growth was predicted upon a population
estimate of 410,000 in 1975 and 550,000 by 1985.
I.4. 1974 Alaska Power Survey, U.S. Department of Interior,
Alaska Power Administration, 1974; 1976 Alaska Power Survey,
Federal Power Commission, 1976; and Devils Canyon Status
Report, U.S. Department of Interior, Alaska Power Adminis-
tration.
The load estimates of these reports are identical and thus dis-
cussed simultaneously. Tables I.3 and I.4 present the utility and total
load projections by reg{on through 2000, and Figure I.3 graphically
depicts the statewide load growth projected. The Southcentral and Yuko n
regions generally conform to the railbelt as presently defined except
that the Alaska Power Administration includes Kodiak as part of the
Southcentral region. As shown in the tables and figure, this was the
I-6
Region
Southcentral
Yukon (Interior)
H
I
-...I
Total
South central
Yukon (Interior)
Total
South central
Yukon (Interior)
Total
TABLE 1.3 •
... •. Regional Utility Load Estimates, 1972-2000
Actual Requireme nts
1972
Peak
Demand
1000 KW
224
69
293
Annual
Energy
Million KWH
1,037
307
1,344
Peak
Demand
1000 KW
680
200
880
610
180
790
530
160
690
1980
Estimated Future Requirements
1990 2000
Annual
Energy
Million KWH
2,990
870
3,860 .
2,670
780
3,450
. 2,340
680
3,020
Peak Annual Peak Annual
Demand Energy Demand Energy
1000 KW Million KWH . 1000 KW Million KWH --------~--~----~
Higher Rate of Growth
1,640 7,190 3,590
460 2,020 970 --
2,100 9.210 '4, 560
Likely Mid Range of Growth
1,220
340
1' 560 '
Lower
980
270
1,250
Rate
5,350
1,500 --
6,850
of Growth
4,290
1, 200
5,490
.·
2,220
600
2,820
. 1,470
390
1,860
15,740
4,230
19,970
9, 710
2,610
12,320
6,430
1, 730
. 8,160
Note: Estimated future peak demand based on 50 percent annual load factor.
Source: Alaska Power Survey, Technical Advisory Committee on Economic Analysis and Load
Projection.
~--------__________ .. _ ----------· ..
H
I
00
Region
Southcentral
Yukon (Interior)
Total
Southcentral
Yukon (Interior
Total
Southcentral
Yukon (Inter;or)
Total
TABLE 1.4.
Regional Total Load Estimate, 1972-2000
Actual Requirements Estimated Future Requirements
1972 1980 1990
Peak Annual Peak Annual Peak ·Annual Peak
Demand Energy Demand Energy Demand Energy Demand
1000 KW Million KWH 1000 KW Million KWH 1000 KW Million KWH 1000 KW
Higher Rate of Growth
317 11 ·465 990 51020 5,020 30,760 7,190
115 542 ' 330 1,610 760 3,980 1,390 --
432 2,007 1,320 6,630 , 517 80 34,740 8,580
Likely Mid Range of Growth
790 3,790 1,530 7,400 3,040
280 1,310 470 2,270 910
1,070 51100 2,000 91670 3,950
Lower Rate of Growth
650 3,.040 ·1,160 51430 . 1,790
250 ·1,140 370 ·1,760 530
900 4,180 1,530 .71190 2,320
2000
Annual
Energy
Million KWH
40,810
7,000
47,810
15,300
4,610
19,910
8,510
2,540
11,050
Note: Assume 80 percent annual load factor for industrial requirements; 50 percent for . utility requirements.
Higher estimate includes nuclear enrichment facility in 1980's with requirements of 2. 5 million kilowatts.
Sou-rce : A.la.s 'ka. Pov..re-r Su-rvey • Technical Advisory Committee on Economic Analysis and Load Pro'ection.
~ ... ...
11
i3 ... ...
::l
J'
11
>. .. ... ... ...
)
d
1 ..
1
5
~
~
!J ...
)
::l .. ..
I)
u
j
j ..
) ... .. ..
!) ...
~
j
I)
I) ..
I) ... ..
-·-·-
0
"l
~
0
~
Q)
a
0
0
()
~
~
(/)
)o
Cl
~
0
!!< ~
rl
;1.
rl
ACT UAL
<
1965 1970 ..
()
0
)o
. ::1
0
(/)
FIG URE 1.3.
ALASKA TOTAL ENERGY REQUIREMENTS
(1965-2000)
Larg er energy r eq uireme nt s du e t o l arge in dus tr i a l l oad , suc h cs
a nuc l ear fuel enr ich.men t pla nt or ot her very large .en ergy int ensive
fac i li t y.
PROJEC TED
19 8 0 19 90 1995
YEAR
-··----
I-9
60
50
4 0
30 (/)
0:::
::::>
0
:X:
2 0 1-
1-<! :::
<!
0::: w
1-
10
9
8
7 (/)
6 1-z w
5 ::E w
4 0:::
::::>
0 w
3 0:::
>-~
0::: w 2 z w
I
2000
----
first study which attempted to delineate a range of forecasts. The
1980 "likely midrange rate of gro\vth" forecast is somewhat high.
The projection for utility load utilized the 1980 load projections
of the actual utilities. These are sho\vn, for railbelt utilities, in
Table I.S. The high and low estimates for 1980 were 20 percent above
and below, respectively, those aggregated estimates. For the subsequent
decades, growth rates were assumed in each case based upon unidentified
population projections. The growth rates declined over time, reflecting
the assumtion of increasing appliance saturation, efficiency, and utili-
zation of conservation measures. .In the "likely midrange" case, the
growth rates were assumed to be 7 and 6 percent, respectively, for the
1980s and 1990s.
Military requirem~ts were assumed to grow at 1.7 percent annually.
The industrial load growth estimate was based upon a State of
Alaska Department of Commerce and Economic Development study of possible
.industrial projects. This estimate is shown in Table 1.6. The assump-
tions behind these estimates are:
1. a high probability of major new petroleum, natural gas,
coal, and other new mineral production and processing;
2. significant further developments in timber processing;
and
3. good possibilities that Alaska energy and other re-
sources \vill attract energy intensive industries.
The estimates vary basically in the speed with which developments
occur except that the high case includes a nuclear enrichment facility
I-10
H
I
I-'
I-'
.. .:; ..
TABLE 1.5.
Utility Load Estimate Extended to 1990
Projected Requirements Generation
1972 1980 1990
Location and Utility Symbol
Southcentral
City of Anchorage (AML&P} ~
Chugach (CEA) ]/
Include:
Homer (HEA) ]}
Kenai (HEA) 3/
Seldovia-Port Graham (HEA) ]/
Sewat·d (SL&P) ]/ ·
Tyonek Y.
Palmer-Talkeetna (MEA) 1/
[S"Ubtota ;·~-A~~h~rage-Cook In 1 et
Yukon (Interior}
Fairbanks Area .
Peak Annual
· Demand Generation
1000 KW Million KWH
71.9 31 o. 1
. 123. 5 600.0
(16. 5) (85.1)
(4.7) (22. 1)
(0.8) (3.4)
Not Available
Not Available
15.9 . 73.9
211.3 984.0
..... ··-···-
Peak
Demand
1000 KW
123.0 360.0
(36.5)
(7.9)
~1.6~ . 5.2
(2. 1 )'
54.0
537.0
City of Fairbanks (FMU) 3J 21.0 90.0 . 51.8
Golden Valley (GVEA) 1( 45.9 211.0 131.0
IS u btota 1 , Fa i rba _n k_s_A_r_e_a ____ ....:..6..::...:..6. 9 __ __.=.c:30 1 • 0 ___ 1-=-=82. 8
. ···-·----~-----~-·--·----.. ---~---·--··-·· .. --·---··-·----··
Annual Peak Annual
Generation De mand Generation
Million KWH 1000 KW Million KWH
658.0 314.0 1 ,680.0 1,800.0 1,065.0 5,370.0
(190.0 (90~0) (490.0)
{37.6) ( 14 .. 9) (71.1)
(6.8~ ~3.5~ (16.1l (27.9 8.0 (45.0
(8. 7) (3.0) ( 12. 3
225.0 181 .0 850.0
2,683.0 1 ,560. 0 7,900.0
........ -·. :-
221.0 161.1 687.3
610.0 379.0 1,790.0
8~31=·=0 ======~40_:_1 ___ ?~~~!._:.: _____ _
Region
H
I ....
N
Southeast
Southcentral
Yukon (Interior)
&eutheast
Southcentral
Yukon __ q_~!~~i9_r)
~~
Southcentral
Yu):con (Interior)
TABLE I.6.
.. Assumed Regional Industrial Power Requirements, 1972-2000
A~tual Requirements
Peak
Demand
1000 KW
4
58
··--·-····-
1972
Annual
Energy
Million KWH
-··
Peak
Demand
1000 KW
~
260
70
Estimated Future Requirements
1980 1990
Annual Peak Annual Peak
Energy Demand Energy Demand
Million KWH 1000 KW Million KWH 1000 KW
High Rate of Development Assumed
-i,-4-1-0 -4-3-e
1,820 3,330
5-1G-
3,540
350
2000
Annual
Energy
Million KWH
490 240 ----~~--~----------
~-{}
23,340
1,680
3-;-9-9-e
24,810
2,450
-<!&
130
40
-5-e
70
30
6-3-0
910
280
-35{)
490
210
Mid Range of Development Assumed
-t.l:t) l-;-4-1<) -431) -3fol{)
260 1,820 . 760 5,330
-··,---·· 70 .. _____ 4_99 ______ ?.~_0 ____ 1 _~_680 --
Low Development Assumed ·
-9& ~ -ti-E} 1-;-4-!7-0
130 910 260 1,820
40 280 70 490 ....... ·--·------·-·-· -------------------·-·----·----------
'
in Southcentral with a peak load of 2.5 million kilowatts. All estimates
assume that mlnerals other than petroleum and natural gas could account
for the majority of industrial requirements by 2000. The relevant
facilites within the railbelt were assumed to be coal mining, the nu-
clear fuel enrichment plant, metal mining, metal processing plants, and
. metal dredging operations.
1.5. Reassessment Report on Upper Susitna River Hydroelectric
Development for the State of Ala ska by Henry J. Kaiser
Company, 1974.
This study projects net energy for the Anchorage and Fairbanks areas
(but not Glennallen-Valdez) through 1990. The forecast is sho'vn in
Table 1.7. The forecast used a model which related number of customers
in the residential and commercial-industrial sectors to population and
employment projections. 1980 energy sales per customer were forecast
at 13,000 and 14,000 in Anchorage and Fairbanks in the residential
sector and 148,000 and 118,000 in those respective areas for the com-
mercial-industrial sector.
1.6. Southcentral Railbelt Area, Alaska: Interim Feasibility
Report: Hydroelectric Power and Related Purposes for the
Upper Susitna River Basin, Alaska District Corps of
Engineers, Department of the Army, 1975.
The previous Alaska Power Administration report was updated for
this study with no significant modifications in methodology but some
changes in assumptions which result in a reduction of the projections.
I-13
TABLE 1.7.
~-·o··.;··: ~< :·•: -···
Anchorage-Cook Inlet Fairbanks Total
Net Peak Net Peak Net Pea k
Energy Load Energy Load Energy Lo a d
Year million thousand million thousand million thou s and --kwh kw kwh kw kwh k w
Actual
1968 559 . 125 161 43 720 168
1973 1,090 213 318 73 1,408 286 -
Forecast
1975 1,400 271 410 87 l, 810 3 58
1980 2,740 513 803 164 3,543 6 77
1985 5,080 928 1,354 266 6,434 1 , 194
1990 9,420 1, 721 2,281 434 11, 701 2 ,1 55
I-14
td
In this study, the actual areas that would be served by a railbelt
electric faicility were the basis for projections so some small com-
ponents of the previously defined Southcentral and Yukon regions are not
included. Projections are shown in Table 1.8 and Figure 1.4.
The utility estimates are some~11hat lower than earlier projections
reflecting this as ~11ell as the analysis of t>-.70 additional years of his-
torical data. Beyond 1980, the growth rates are identical to the pre-
vious studies. These are as follm11s:
high range
likely midrange
lower range
1980-1990
9%
7%
6%
1990-2000
8%
6%
4%
Military load growth is now projected at 1 percent annually
rather than 1.7 percent.
The most significant change in this study from the previous work
is a reduction in the projected industrial load which accounts for the
majority of the difference between the total load projected in this and
the previous study. The 1980 midrange projection has declined from
1,190,000 ~fiVh to 350,000 MWh. The specific industrial developments
assumed in the projections are presented and shown here as Table 1.9.
1-15
TABLE 1.8.
Actual Reguirements Estimated Future Reguirements
Type of Load 1974 1980 1990 2000
Peak •Annual Peak Annual Peak Annual Peak Annual
Demand Energy Demand Energy Demand Energy Demand Energy
Area 1000 kw Million/kwh 1000 kw Million/kwh 1000 kw Million/kwh 1000 kw Million/kwh .
Utilities High Rate of Growth
H Anchorage 284 1,305 650 2,850 1,570 6,880 3,430 15,020 I ......
0'\ Fairbanks 83 330 160 700 380 1,660 800 3,500 --Total 367 1, 635 810 3,550 1,950 8,5 40 4,230 18,520
Likely Mid-Range Growth
Anchorage 590 2,580 1,190 5, 210 2,150 9,420
Fairbanks 150 660 290 1, 270 510 2,230 ---Total 7 40 3,240 1,480 6,480 2,660 11,650
.
Lower Rate of Growth
Anchorage 550 2,410 1,010 4,420 1,500 • 6. 570
Fairbanks 140 610 240 1,050 350 1,530
Total 690 3,020 :1.,250 5,470 1,850 8,100
TABLE 1.8. (cont.)
Actual Reguil-ements Estimated Future Reguirements
Type of Load 1974 1980 1990 2000
Peak Annual Peak Annual Peak Annual Peak Annual
Demand Energy Demand Energy Demand Energy Demand Energy
Area 1000 kw Million/kwh 1000 kw Million/kwh 1ooo kw Million/kwh 1000 kw Million/kwh
National Defense
H Anchorage 33 155 35 170 40 190 45 220
I Fairbanks 41 197 45 220 50 240 55 260 I-' --"--1 Total 74 352 80 390 90 430 100 480
bdustrial High Rate of Development Assum e d
Anchorage 10 45 100 710 2,910 20,390 2,920 20,460
Fairbanks l/
Mid-Range Development Assumed
Anchorage 50 350 100 710 410 2,870
Fairbanks 1/ --
Low Development Assumed
Anchorage" 20 140 50 350 100 710
Fairbanks 1/
··--------··--···-... ·----· -·
TABLE 1.8. (concluded)
Actual Reguirements Estimated Future Reguirements
Type of Load 1974 1980 1990 2000
Peak Annual Peak Annual Peak Annual Peak Annual
Demand Energy Demand Energy Demand Energy Demand Energy
Area 1000 kw Million/kwh 1000 kw Million/kwh 1000 kw Million/kwh 1000 kw Million/kwh
Combined Utility, National Defense, and Industrial Power Reguirements
H .
I
f--1 Higher Growth Rate 00
Anchorage 327 1,505 785 3,730 4,520 27,460 6,395 . 35,700
Fairbanks 124 527 205 920 430 1,900 855 3,760 ·----Total 451 2,302 990 4,650 4,950 29,360 7,250 39,460
Likely Mid-Range Growth Rate
Anchorage 675 3,100 1,330 6,110 2,605 12,510
Fairbanlcs 195 880 340 1,510 565 2,490
Total 870 3,980 1,670 7,620 3,170 151000
Lower Growth Rate
Anchorage 605 2, 720 1, 100 4,960 . 1,645 7,500
Fairb ank s 185 830 290 1,290 405 1, 790
Total 790 3 ,55 0 1 ,390 6,250 2,050 9,290 ·-· ...
·------·---
-----. ---------· &
-----------:----.-. -------·-···. . --..
.Lota.L
H
I
t--'
\.0
f'JU
----·------.. -------.
:FIGURE 1.4.
ESTIMATED FUTURE POWER ·REQUIREMENTS
1974-2000
2. ANNUAL ENERGY REQUIREMENTS
40.ooor----r---rr----r----r----r====F:::::::=l I i I . . . I ·-·--· ·---···· .. -.. ·---~-.. ---·-·. ·---··:··---,-. ··-·-···-···-·····-··-·: ···--~---·-·· -~·-·--· ----~~-·-···. ·-··· .. -··-···--·-I -··--... -j· . . .. ···-···---·----
--------··· .. ·····+-·--·--·------~---+-------·----· --1----···----· ---·· --·· ....... ·-;.f:o-'?J ----~---·· ··--·----·! ..
i : i I ~c} I
l I I <?J~
I ! , ~~ : ::r: 10000 -· .. ... .. .... ·····. --, --.. ,. _________ i_, ___ .. __ ...................... .,... __ ,. _______ , .. _ --------·-·----1··---.. -·
~ . ! · : I I I t\00qe ! z . I i . I ~\0' I
0 5000 ---~----·-· -·-t------·--· .. --~-----'--------~--------I ----' . o\e ----1 ---------------L ·· :J 4000 -------------.. --1--------~-.. ------:--·-!-.. ___ .. _____ --, -----\..!)"lei <t.s\1~---------~--j-------------.. --r----·--· --------
~ 3000 ------------r ------------~,-~-~:~:-=-----,--------------·-r ------------------1-------... -!-----------
>-, . I ! ?i 2000 ---· ----.............. !··--. . ............
1
.. -· .. -----------
1
-----.... ----.... T -. .. .. i
W I Note: Includes estimated annual energy requirements Q ! 1 ! for utility, notional defense, and industrial power
· ------r--T
1
-----systems in the Anchorage-Cook Inlet and Fairbanks---
_J I I • <r . 1 · Tanana Volley areas. ! !
=> I I ! ·. I . I i :
~ · :~~ _:=:~:-: ~-:--F:·: ____ ~--!T-~= --=---r=-=-~:=_-.:·r=~-~---~-==~r~~~ ~ ~:~--~ =-: --
---------------1 ---------:--~---------~~ ------· -----+----· -----1"----· r -····· ·-
··:·.·:-··-................ : ---.---------.... -~ ---, ....... _. _____ ·--· .... T-.. ·-----··----.. ·-·--~-...... -· --.. --· -·:---.. -.... .. .i.. .... -.-
. I ! I j I
l 'I I l ,
1
! 00 A.?. A.-September 1975
I L-----~----~~----~--------~-----~-----~~----~~
1965 70 75 80 85 90 95 2000
YEAR
1.),2 90
TABLE I. 9.
Industrial Capacity in ~·iW
Rate of
Development Lov Range 1·1id Rane;e High Range
·Year 1980 1990 2000 1980 1990 2000 1980 1990 ~
Anchore5e .Area:
Kenai Peninsula:
Chemical. Plant Y ll 11. 11 12 14 16 1.3 16 20
LNG PLant l./ .4 .4 .4 .4 .5 .6 .5 .6 ·1
New Plent 1.0 10 10 1.0 10 10 10 10
1/.
Refinery -2.2 2.2 2.2 2.2 3 4 3 4 5
Timber Y 2 3 5 3 5 5 5 5 5
other Vicinities:
Coal Gasification 10 10 250 10 250 250
Mining and Mineral
Processing 5 25 -5 25 50 25 50 50
Nuclear Fuel
Enrichment 2500 2 500
Timber 5 7 5 7 7 7 7 1
1-l'e'W' Cit;[ 17 30 10 30 70 30 70 70
-TOTAL (rounded) 20 50 100 50 100 410 100 2910 2920
Fairbank3 Area ~/
Source: 1974 Alaska Power Survey Technical Advisory Committee Report on
Economic Analysis and Load Productions, pages 81-89.
"
!I Existing Installations
2/ Timber _processing and oil refinery loads totaled l e ss than 10 MW.
I-20
A population growth rate of 3 percent annually is assumed in the
study, resulting in estimates of 410,000 in 1980 and rising to 740,000
in 2000.
L.7. Electric Power in Alaska, 1976-1995 and ''Alaska Electric
Power Requirements," Alaska Review of Business and Economic
Conditions, Institute of Social and Economic Research,
University of Alaska, 1976.
This study estimated growth in electricity requirements for the
entire state based upon a model of the Alaskan economy and detailed
assumptions concerning customer growth and average consumption rates.
Projections were made based upon two sets of economic assumptions and
four sets of electricity use assumptions. This resulted in a signifi-
cant range of projections, a reflection of the uncertainty surrounding
both the future of economic growth of the state, and electricity use
patterns. Table 1.10 shows the range of estimates of sales for utili-
ties for. the Anchorage, Southcentral, and Fairbanks regions (a larger
region than the railbelt).
The economic projections assumed growth rates in population for
the state of between 3.8 and 4.8 percent with the growth concentrated
in Southcentral Alaska.
I-21
Anchorage,
Southcentra1,
& Fairbanks
LOWEST
HIGHEST
TABLE 1.10. 1976 ISER ELECTRIC POHER
REQUIREMENTS PROJECTIONS
Average Annual
Total Energ;y: Sales Grmvth Rates
1975- 1975-1975-
1974 1985 1995 1980 1985 1995 .
1,468 3,697 8,092 9.1 8 .8 8.5
1,468 7,787 20,984 17.6 16.4 13.5
I-22
Military electricity requirements are assumed to remain constant
over time.
Self-generated industrial electricity requirements are not spe-·
cifically modelled.
1.8. Alaskan Electric Power: An Analysis of Future Requirements
and Supply Alternatives for the Railbelt Region, Battelle
Pacific Northwest Laboratories for Alaska Division of
Energy and Power Development and the Alaska Pmver Authority,
1978.
This report did not do a load growth analysis but summarized and
interpreted several earlier studies, including those of the Institute
of Social and Economic Research and the Alaska Power Administration (APA).
Individual load growth studies of several railbelt utilities were also
utilized in selecting a narrow projection band from the broader band
represented by the earlier analyses. The industrial scenarios developed
by the APA were somewhat modified as well. The resulting proj ec.tions
for the railbelt are shown in Table 1.11 and Figure 1.5.
I. 9. Upper Susitna River Project Pmver Harket Analyses, U.S.
Department of Energy, Alaska Po\ver Administration, 1979;
Southcentral Railbelt Area, Alaska, Upper Susitna River
Basin, Supplemental Feasibility Report by the Corps of
Engineers, 1979; and Phase I Technical Memorandum: Electric
Power Needs Assessment, Southcentral Alaska Water Resources
Committee, 1979.
This report is a continuation of the work done by the Alaska Power
Administration and again provides load requirements projections for
I-23
Year
1974
1980
1990
2000
TABLE I. 11. -
I
Range of Railbelt Annual ·consumption (Includes use by utili ty :
and industrial customers likely to be part of an intertied \
. system. Excludes national defense and non-intertied users.) ,
Annual Consumption -Compound Annual Gro0th Rate
I
1.6 B kHh !
I
2.6 to 3.4 B kl-!h 8.4 to 13. 4 ( 197 4-1980) I
i
15.3 (1980-1990) I
8.5 to 10.8 B HJh 9.6 to I
!
16.0 to 22.5 8 kHh 4.0 to 10.2 (1990-2000) '
I-24
y
\
Vl
:r: :s::
~
Ll..
0
Vl z
0
-l
-l
FIGURE I.5.
10, 000
. -···-·--------------------------,-
1976 ACTUAL
~
/
, """f.1975 ACTUAL
1974 ACTUAL
lOOG ----·-··----------· --------------------1
-----------------------..,.------1
lQQL-----L-----~-----------L~--------~----~
1975 1980 1990 2000
Most Likely Range of Utility and Industrial Annual Consumption;
Intertied Railbelt Area
I-25
utility, military , and industry electricity users. The overall projec-
tions are summarized in Table 1.12 and Figure 1.6.
Utility requirements are projected on the basis of e x plicit popu-
lation estimates (provided by the ISER econometric and demographic model)
multiplied b y average annual consumption per capita estimates. The
estimate of per capita growth in average annual electricity consumption
is based upon growth in this indicator for Alaska utilities during the
interval of 1973 to 1977. This is presumed to capture the effect of
conservation over the period, and, furthermore, the growth rate is
assumed to decline over time to reflect additional conservation measures
and saturation of appliances, etc. For the three "consumption per
capita" scenarios, the following annual growth rates were assumed:
Time Period Scenario
High Medium Lmv
1980-1985 4.5% 3.5% 2.5%
1985-1990 3.5 3.0 2 .0
1990-1995 3.0 2.5 1.5
1995-2000 2.5 2.0 1.0
2000-2025 2.0 1.0 0.0
These growth rates were applied beginning in 1980. In the interim ,
growth rates of net generation were assumed to be 12 percent annually
for Anchorage and 10.3 percent for Fairbanks. The high and low estimates
l-26
TABLE 1.12.
Rail belt Area Energy Forecast
(Gl.ffi)
1977 .1980 1990 2000 2025 ··--(Historic)
Utility:
High 3,410 8,200 16,920 38,020
Mid 2,273 3,155 6,110 10,940 17,770
Low 2,920 4,550 7,070 8,110
National Defense:
High 348 384 425 544
Mid 338 338 338 338 338
Low 330 299 270 210
Self-Supplied Industry:
High 170 2,100 3,590 8,490
Xid 70 170 630 1,460 3,470
Low 141 370 550 1,310
Total:
High 3, 928 10,684 20,935 47,054
Mid 2,681 3,663 7,078 12,738 21,578
Low 3,391 5,219 7,890 '9, 630
Trend @ 1973-77 annual/growth: (3,215) (10,270)-(33,000) (601,000)
l-27
,(f) ··o:
·~ ·o ~ ~I
N
00
. FIGURE I. 6 •
. I 0 0 1000 ,__--------------------~---:-~~ .. ~ .... ::-:: ..• -:-: .... --:_:-::-.: ... -:-. :-:-: .. 7.:.::7., :: .... :-:":" .. ':":'" ..• :::-. ~--:::::--:-:-1
· )O,'ooo .I
.·ao,·aao ·TOTAL :RAILBELT AREA .
. :7-910~0 ! ~-.· . , . •. ·. : ' .
. _:s_o,ooo ENERGY :FORECA.ST
,-5 0,000
,, ,__tl ...... ·:
.Up p er Su s itna Project ~ower .Market Analysi~
'. 6 .: -• • • : : • ' • • • ' • • • • • I · ' ~ . ' • . ::
I 4'L~.·.RRPi -
I .,,'30,,090·
... I ~. j • • ,. • • !
I
.2o,ooa :-
.• ..• :,.•• I;', '
.LOW
• • ••• ,~ •• • ..... ."~> •.; ~ •• ,. : ·' •
.... ··•· .. ··· .. ,I
·.-. ,j. ··:;·.·.:·.:·
for 1980 appear to be based on estimates of growth between 1977 and 1980
which are 29 percent higher and 27 percent lower than the mid case re-
spectively.
Two population estimates are utilized. These are presented in
Table I.l3.
TABLE 1.13. POPULATION ESTIMATES
(thousands)
Year Anchorage-Cook Inlet Fairbanks Statewide
Low High Low High Lmv High
1980 239 270 60 62 500 514
1985 261 320 68 77 563 641
1990 299 407 75 95 618 790
1995 353 499 82 114 680 947
2000 424 651 90 140 743 1,158
2025 491 904 99 179 820 1,485
The result of t>.;o population projections and three per capita con-
sumption projections is six scenarios. The high/high and lmv/lmv see-
narios became the high and low projections, respectively, while the
average of the high/low and low/high became the midrange final forecast.
Military requirements were projected to remain constant in the mid-
range estimate, +1 percent in the high case, and -1 percent in the low
case.
I-29
The self-supplied industrial load projection is based upon the
earlier Battelle report which, in turn, is based upon the earlier APA
work. The high range forecast is summarized in Table I.l4. The only
difference between the cases is that the midrange does not include
the aluminum smelter and the low range contains neither the smelter
nor the ne'" capital city.
TABLE I.l,4. SELF-SUPPLIED INDUSTRIES FORECAST
HIGH RANGE
Existing refinery (2.4 MH)
Existing LNG plant (. 4 to . 6 H\v)
Coal gasification (0 to 250 MH)
New city (0 to 30 HH)
New refinery (0 to 15.5 MH)
New LNG plant (0 to 17 MIV)
Mining and mineral plants (5 to 50 M\v)
Timber (2 to 12 M\v)
Existing chemical plant (22 to 26 M\V)
Aluminum smelter or other energy
intensive industry (0 to 280 M\v)
A comparison of the estimates developed for this report with
earlier studies by-APA was done and is presented as Table I.l5. It
indicates a slight downward shift in the projections.
Finally, a compilation and extrapolation of forecasts done by the
utilities themselves was compared to the study results. As Table I .l6
shows, the aggregated utility forecasts are considerably above those
of the study.
I-30
TABLE 1.15.
CONPARISON OF UTILITY ENERGY ESTUIATES
1976 MARKETABILITY REPORT, UPDATE OF 1976, AND 1978 ANALYSIS
Upper Susitna Project Power Market Analysis
Anchorage-Cook Inlet Fairbanks-Tanana Valley Total Railbelt
•I Forecast· 1976 Update 1978 1976 Update 1978 1976 Update 197 8
I ~ Range .··r Re~ort of 1976 Forecast ~eeort of 1976 For e cast Report of 1976 Forecast . Year
1974 Historic 1,305 1./ 1,189.7 ll 330 353.8 1,635 1,543.5
1975 High 1, '•89 377 1,866
·· Mid 1,467 371 1,838
Low 1,450 367 1, 816
Historic 1,413.0 450.8 1,863.8
1976 High 1,699 430 2,129
H Hid 1,649 lfl7 2,066 I
UJ Low 1, 611 407 2,018 f-'
Historic 1,615.3 \ 468.5 2,083.8 \.;
' 1977 High 1,939 t.9o 2,429
Mid '· 1,853 469 2,322
Lm-1 1,790 453 2,242
Hi s toric 1,790.1 1,790.1 482.9 482.9 2,273.0 2,273.0
1980 High 2,850 2,660 2, 720 700 720 690 3,550 3,380 3,410
Mid 2,580 2,540 2,500 660 690 655 3,240 3,230 3, 155
Low 2,410 2,460 2,300 610 660 620 3 ,020 3,1 2 0 ·2, no
1990 Hi gh 6,880 6,300 6,630 1,660 1,700 1,570 8,540 8,000 8,200
Mid 5,210 5,000 4,880 ; 1, 270 1,360 1,230 6,480 6,360 6,110
Lo\-1 4,'•20 4,410 3,590 1,050 1,180 960 5,470 5,590 4,550
. -)· .. -~ooo·-..-nigh 15,020 13,600 13,920 3,500 3,670 3,000 18,520 17,270 16,920
Mid 9,LI20 8,950 8,960 2,230 2,440 1,980 11,650 11,390 10,940
LoH 6,570 6,530 5, 770 1,530 1,750 1,300 8,100 8,280 7,070
lJ 1974 historic data revised between 1975 and 1978. APA. 11/78
-----------·-·------------···------------~--------------------------------------------
TABLE 1.16.
Utility Forecasts 1978 Susitna Forecasts
Energy (G'i·TH) High ~fid Low
1980 3,344 3,410 3,155 2, 920
1985 6, 277 5,460 4,455 3,630
1990 10,965 8,200 6,110 4,550
1995 17,748 11,600 8,140 5,690
2000 26,550 16,920 10,940 7,070
Peak (H'i-1)
1980 725 778 720 667
1985 1,377 1,244 1,021 830
1990 2,986 1,873 1,396 1,039
1995 3,835 2,645 1,858 1,293
2000 5,641 3,865 2,497 1,617
I-32
Glennallen-Valdez projections were made using the 1976 Copper Valley
Electric Association Power Requirements Study as a base. Total energy
was projected to be 74,000 HWh in 1980, 134,000 11'\ffi in 1990, and
240,000 Mlffi in 2000.
I.lO. Differences Between Present and Previous Studies
I.lO .A. ECONOHIC PROJECTIONS
Differences exist in the economic projections among the various
studies because of both different predictions about what large scale
projects will be undertaken within the state and when they will occur
(the gas pipeline construction is -an example) and different assumptions
apout how the support sector of the economy and the population responds
to economic development.
In both these areas, the present study has some benefit of hind-
sight which earlier studies have not. The 1979 APA study, for example,
utilized a 1980 statewide population estimate ranging between 500,000
and 514,000. Present projections place the number closer to 420,000.
Part of the error was the result of overly optimistic projections of
large project activity which have not yet materialized but which to a
large extent the present study expects to occur in the early 1980s.
A second component of error ~vas the fact that the entire cyclical
pat.tern of economic behavior in response to the construction of the
oil pipeline was not captured in the economic projection technique.
I-33
Only the data from the expansion portion of the cycle was internalized
into the model but not the contraction portion of the cycle. As a
result, the economic model itself contained some upward bias .
I.lO.B. ELECTRICITY CONSUMPTION PROJECTIONS
The recent APA study as well as the earlier ISER study developed
projections based upon the product of population and consumption per
capita. The present study has generally lower growth rate assumptions
for per capita consumption based upon explicit estimates of saturation,
use patterns, and conservation measures. Another distinguishing
characteristic of the present study is that conservation measures re-
duce consumption growth rates until they have been completely implemented.
Subsequently, growth may actually accelerate. The 1979 APA study assumed
that the reduction in the rate of growth of electricity consumption after
1973 was the result of the implementation of conservation efforts grow-
ing out of the 1973 oil embargo and higher energy prices. This study
conclu~es that the reduction in the growth rate was not conservation-
induced because, particularly in Anchorage, there was no, and even
today is not, price incentive for the conservation of electricity.
I.lO. C. SELF-SUPPLIED INDUSTRIAL PROJECTIONS
Earlier studies done by APA and Battelle all trace their projections
of self-supplied industrial electricity consumption to a list of pro-
jects compiled in the early 1970s in a State of Alaska Department of
Commerce and Economic Development, Division of Economic Enterprise pub-
lication entitled "Power Demand Estimates, Summary and Assumptions for
I-34
the Alaska Situation." Rather than a projection, the report is simply
a list of potential projects with their related energy requirements.
Such a "list" is not felt to be appropriate as the basis for sound
electric pm.;er requirements planning.
I.lO.D. DEFINITIONS
The present study projects total sales of energy to final consumers.
This is a smaller quantity than net energy for system \vhich is the con-
cept used in some earlier studies. The difference is transmission and
distribution losses and energy unaccounted for.
I-35
APPENDIX J
BIBLIOGRAPHY
Alaska Department of Commerce and Economic Development, Division of
Economic Enterprise. The Alaska Economic Information and Reporting
System. Quarterly Report. January 1980; April 1980.
"The Alaska Economy, Year-End Performance Report 1978." Vol. 7.
"Numbers: Basic Economic Statistics of Alaska Census Division."
Novemb2r 1979.
Alaska Department of Commerce and Economic Development, Division of Energy
and Power Development. "1979 Community Energy Survey."
Alaska Department of Community and Regional Affairs. "Selected 1970 Census
Data for Alaska Communities." Part V Southcentral Alaska. March 1974.
Alaska Department of Labor. "Alaska's Economic Outlook to 1985." Juneau,
Alaska. July 1978.
Statistical Quarterly. Second Quarter 1979.
Alaska Department of Labor, Employment Security Division. Alaska Labor
Force Estimates by_ Industry and Area 1974. Juneau, Alaska. October
1975.
Alaska 1970 Census Atlas: Population by Election Districts.
Juneau~ Alaska. n.d.
Alaska Department of Revenue. "Fuel, Conservation, and Individual Credits
Relative to the Individual Income Tax." Juneau, Alaska. January 1980.
"Revenue Sources FY-1979-81." Quarterly Update. March 1980.
Alaska Department of Revenue, Petroleum Revenue Division. Petroleum
Production Revenue Forecast. Quarterly Report. March 1980.
Alaska Division of Planning and Research. Alaska Statewide Housing Study
1971. Vol. II. December 1971.
"Alaska Power Survey." 1969 and 1976 by Federal Power Commission. 1974
by Alaska Power Administration.
Alaska Public Interest Research Group. Funding Proposals for Conservation/
Renewable Energy Research, Development, Demonstration, Education and
Outreach. Draft. March 12, 1980.
Alaska Public Utilities Commission. Annual Reports to the Legislature.
Alaska State Legislature, Budget and Audit Committee. Energy Related
Budget Items. March 12, 1980.
American Gas Association, Department of Statistics. 1972 Gas Facts.
Arlington, Virginia. 1973.
"Anchorage Real Estate Research Report." Volume III. Anchorage Real
Estate Research Committee. Anchorage. Fall 1979.
Andreassen, Arthur. "Changing Patterns of Demand: BLS Projections to
1990;11 Monthly Labor Revietv. December 1978.
Applied Economics Associates, Inc.; R & M Consultants, Inc.; and Science
Applications, Inc. "The Role of Electric Potver in the Southeast
Alaska Energy Economy." Prepared for the Alaska Pmver Administration.
Juneau, Alaska. March 1979.
Arlon R. Tussing and Associates, Inc. "Background Information on the
Electric Power Industry in Alaska and Its Regulation with a History
of the Proposed Susitna River Hydroelectric Project." For the
Alaska State Legislature, House of Representatives, Power Alternatives
Study Committee. For Committee Use Only. December 31, 1979.
"Susitna Hydropmver: A Review of the Issues." Prepared for the
Alaska State Legislature. April 15, 1980.
Arthur D. Little, Inc. Alaska's Mineral Resources as a Base for Industrial
Development. Report to the State of Alaska, 1962.
Potential for Use of Alaska's Energy Resources. Report to State
of Alaska, 1962.
Baring-Gould, Michael, Marsha EnTin Bennett, Peg Hargis, and Joyce Taylor.
"Valdez City Census." University of Alaska. Anchorage. August 1978.
Baughman, Martin, and Paul Joskow. "The Effects of Fuel Prices on Residen-
tial Appliance Choice in the United States." Land Economics.
February 1975.
"Energy Consumption and Fuel Choice by Residential and
Commercial Consumers in the United States." Energy Systems and Policy,
Vol. 1 No. 4. 1976.
Baumgartner, Mark, and Charles Backus. "Energy End Use: Alaska's Northern
Railbelt ." For the House Pmver Alterhatives Study. Alaska Federation
for Community Self Reliance, Inc. February 1980.
Baumgartner, Mark, and Samuel Skaggs. "Energy Conservation Potential for
Alaska's Railbelt." Draft prepared for Alaska Center for Policy
Studies by the Alaska Federation for Community S~lf-Reliance, Inc.
Harch 1980.
J-2
Bowling, S.A., and C.S. Benson. "Study of the Subarctic Heat Island at
Fairbanks, Alaska." Geophy sical Institute. Fairbanks, Alaska.
June 1978.
Bracken, E.O. "Power Demand Estimators, Summary and Assumptions for the
Alaska Situation." Alaska Department of Economic Development.
April 1973.
Breeding, L.E. State of Alaska Energy Conservation Program: Phase III
Evaluation. Anchorage, Alaska. December 15, 1976.
Brown, R.V., A.S. Kahr, and C. Peterson. Decision Analysis for the
Manager. Ne\v York: Holt, Rinehart and Winston. 1974.
CH2M-Hill. "First Phase of Feasibility Assessment for Hydroelectric
Development at Grant and Crescent Lakes." Anchorage, Alaska.
October 1979.
California Energy Commission, Assessments Division. "California Energy
Demand 1978-2000, A Preliminary Assessment." (Staff Draft)
Sacramento, CA. August 1979.
"Appendix A: Analysis of Residential Energy Uses," 300-76-034,
1976.
Carroll, T; Owen, Robert Nathans, Philip F. Palmedo, and Robert Stern.
"The Planner's Energy Workbook." Brookhaven National Laboratory
and State University of New York Land Use and Energy Utilization
Project. Sponsored by the Office of Conservation and Environment
of the Federal Energy Administration.· June 1977.
Casper, David A. "A Less-Electric Future?" Energy Policy, pp. 191-211.
September 1976.
Charles River Associates, Inc. Analysis of Household Appliance Choice.
June 1979. Prepared for EPRI-EA-110.
"Development of a Commercial Sector Energy Use Model for New
York State." Prepared for New York State Energy Office, Albany,
New York. August 1979.
Christian, Jennifer H. "Alaska's Oil and Gas Industry, 1959-71."
Alaska Review of Business and Economic Conditions; Vol. IX, No. 2.
April 1972.
Code of Federal Regulations. Title 10, Part 430.
Cohn, Steve, Eric Hirst, and Jerry Jackson. "Econometric Analyses of
Household Fuel Demands." Oak Ridge National Laboratory. Oak
Ridge, Tennessee. March 1977.
Conopask, J.V.
for Small
Economic
Service.
"Data Pooling Approach to Estimate Employment Multipliers
Regional Economies." U.S. Department of Agriculture,
Development Division, Economics, Statistics, and Cooperative
1978.
J-3
Conway, Christopher J. "The Impact of Conservation and Interfuel Sub-
stitution on Electrical Load Growth 1974-1993." Presented to the
Select Committee on Hydro Affairs. Toronto, Ontario. February
1979.
Cro-v1, Robert E., James H. Hars, and Christopher J. Conway. "A Pre-
liminary Evaluation of the I.S.E.R. Electricity Demand Forecast."
-Working Paper #1. Toronto, Canada.
deLeemv, Frank.
Institute.
The Distribution of Housing Services.
Washington, D.C. 1974 .
The Urban
Demand and Conservation Panel of the Committee on Nuclear and Alterna-
tive Energy Systems. "U.S. Energy Demand: Some Low Energy
Futures." Science, pp. 142-152. April 1978.
Dolezal, Patricia L., and Richard L. Ender.
Municipality of Anchorage." Anchorage
Anchorage, Alaska. September 1976.
"1976 Population Profile:
Urban Observatory Program.
Edison Electric Institute. Economic Growth in the Future. New York:
McGr mv-Hill.
Electric Power Research Institute. "How Electric Utilities Forecast:
EPRI Symposium Proceedings," R. T. Crm.; (ed.) EPRI EA-1035-SR.
March 1978.
Ender, Richard L. "The Opinions of the Anchorage Citizen on Local Public
Policy Issues." Anchorage Urban Observatory. December 1977.
England, Gerald M. "Cogeneration at the University of Alaska Heating
Plant." The Northern Engineer. Summer 1978.
Erickson, Gregg. "The Natural Gas Industry in Alaska." Alaska Review
of Business and Economic Conditions. February 1967.
Fairbanks North Star Borough, Community Information Center. "1978
Fairbanks Energy Inventory." Special Report No. 4. Fairbanks,
Alaska. July 1979.
Community Information Quarterly. Fairbanks, Alaska. Winter
1979.
Community Information Quarterly. Fairbanks, Alaska. Spring
1980.
Fairbanks North Star Borough, Impact Information Center Final Report.
Chapter 7: ENERGY. Fairbanks, Alaska. 1977.
Federal Energy Administration. "Energy Consumption in Corrnnercial
Industries by Census Division--1974." National Energy Information
Center. Harch 1977.
J-4
"Residential and Commercial Energy Use Patterns, 1970-1990."
Project Independence Blueprint. Final Task Force Report. Vol. 1,
November 1974.
Federal Energy Administration, Energy Research and Development Adminis-
tration. "Tips for Energy Savers." August 1977.
Federal Energy Administration, National Energy Information Center.
Energy Consumption in Commercial Industries by Census Division--
1974 .. Washington, D.C. March 1977.
Federal Po\,er Commission. Form 12 annual reports of utilities, Annual.
Federal Register. Vol. 42, p. 36648, July 15, 1977. "Energy Conservation
Program for Appliance."
Flaim, Paul 0., and Howard N. Fullerton. "Labor Force Projections to
1990: Three Possible Paths." Honthly Labor Review. December 1978.
Fryer : Pressley : Elliott, Inc. "Alternative Energy Resources of
Alaska's Railbelt." Prepared for Alaska Center for Policy Studies.
April 1980.
Goldsmith, Scott. "Alaska Electric Power Requirements: A Revie\.;r and
Projection." Alaska Review of Business and Economic Conditions,
Vol. XIV, No. 2. June 1977.
"Alaska's Revenue Forecasts and Expenditure Options." Alaska
Review of Social and Economic Conditions, Vol. XV, No. 2. June 1978.
Man-In-The-Arctic-Program Model Documentation. Institute of
Social and Economic Research. 1979.
Goldsmith, Scott, and Lee Huskey.
An Economic Impact Analysis."
lature by Institute of Social
Alaska. April 1978.
"The Alpetco Petrochemical Proposal:
A report to the Alaska State Legis-
and Economi :: Research. Anchorage,
"Structural Change in the Alaskan Economy: the Alyeska
Experience." Paper presented at the 29th Alaska Science Con-
ference. August 15, 1978.
Goldsmith, Scott, and Tom Lane. Oil and Gas Consumption in Alaska
1976-2000. For the Alaska Royalty Oil and Gas Development Advisory
Board and the 1978 Alaska State Legislature by the Institute of
Social and Economic Research. Anchorage, Alaska. January 1978.
J-5
Goldsmith, Scott, and Kristina O'Connor. "Historic and Projected Oil
and Gas Consumption." For the Alaska Royalty Oil and Gas Develop-
ment Advisory Board. January 1980.
Halvorsen, Robert. Econometric Models of U.S. Energy Demand. Lexington,
Massachusetts: Lexington Books. 1978.
Hausman, Jerry A. "Individual Discount Rates and the Purchase and
Utilization of Energy-Using Durables." Bell Journal of Economics.
Spring 1979.
Heberlein, Thomas
Electricity
and Policy.
A. "Conservation Information: The Energy Crisis and
Consumption in an Apartment Complex." Energy Systems
1975, Vol. 1, No. 2.
Henry J. Kaiser Company. Upper Susitna River Hydroelectric Development.
Report to the State of Alaska, 1974.
Hirst, Eric, and Janet Carney. "The ORNL Residential Energy-Use Model:
Structure and Results." Land Economics. August 1979.
"Residential Energy Conservation." Energy Policy. September
1977.
Hirst, Eric, and Bruce Hannon. "Effects of Energy Conservation in
Residential and Commercial Buildings." Science, pp. 656-661.
August 17, 1979.
Hirst, Eric, William Linn, and Jane Cope. "A Residential Energy Use
Model Sensitive to Demographic, Economic, and Technological-Factors."
Quarterly Review of Economics and Business. Summer 1977.
Haag, Rush. The Promise of Power. Potential Economic Development in
Southeast Alaska. Alaska State Yukon-TAIYA Commission and Alaska
Department of Economic Development. Juneau, Alaska. January 1972.
"Housing Units Authorized by Building Permits." Compiled by Department
of Housing and Urban Development.
Hitchens, Diddy, Richard L. Ender, and G. Hayden Green. Anchorage
Municipal Housing Study. Anchorage Urban Observatory Program.
Anchorage, Alaska. 1976.
Hitchens, Diddy R., Richard L. Ender, G. Hayden Green, and Marsha
Bennett. "A Profile of Five Kenai Peninsula Tmvns." Bureau
Hanagement and Urban Affairs and Anchorage Urban Observatory.
of
1977.
Ide, Ed~vard A. "Estimating Land and Floor Area Implicit in Employment
Projections.'' Prepared for U.S. Department of Transportation.
July 1970.
J-6
Jackson, Jerry R. "An Economic-Engineering Analysis of Federal Energy
Conservation Programs in the Commercial Sector." Oak Ridge
National Laboratory, Oak Ridge, Tennessee . January 1979.
Jackson, Jerry R., Steve Cohn, Jane Cope, and WilliamS. Johnson.
"The Commercial Demand for Energy: A Disaggregated Approach."
Oak Ridge National Laboratory. Oak Ridge, Tennessee. April 1978.
Jackson, Jerry R., and WilliamS. Johnson. "Commercial Energy Use: A
Disaggregation by Fuel, Building Type, and End Use." Oak Ridge
National Laboratory. Oak Ridge, Tennessee. February 1978.
Kirkwood, Craig and Sioshansi, R. Perry. "Review of ISER Draft Report."
Woodward-Clyde Consultants. April 1980.
Kreinheder, Jack. Memorandum to Representative Bill Miles regarding
Pacific LNG Project, Research Request No. 30. House Research
Agency, Juneau, Alaska. February 29, 1980.
Kruse, Jack. "Fairbanks Community Survey."
Economic Research. Fairbanks, Alaska.
Institute of Social and
1977.
"Fairbanks Petrochemical Study." Community Information
Center. Fairbanks, Alaska. March 31, 1978.
"Research Notes: Fairbanks Community Survey." Institute
of Social and Economic Research. Fairbanks, Alaska. December 1976.
Lane, Ted. Lower Cook Inlet Petroleum Development Scenario: Economic
and Demographic Impact Analysis. Alaska OCS Studies Program. 1980.
Lin, William, Eric Hirst, and Steve Cohn. "Fuel Choices in the House-
hold Sector." Oak Ridge National Laboratory. Oak Ridge, Tennessee.
October 1976.
Loll, Leo M. A Study of Technical and Economic Problems--State of
Alaska. For Alaska Department of Economic Development by the
Institute of Social, Economic and Government Research. Fairbanks,
Alaska. August 1967.
Love, James, and Eric Meyers. "Draft Legislation prepared for the Alaska
Center for Policy Studies." January 20, 1980.
"Matanuska-Susitna Borough 1976 Census." Matanuska-Susitna Borough.
Palmer, Alaska. December 1976.
Midwest Research Institute. "Patterns of Energy Use by Electrical Appli-
ances," Final Report. Prepared for EPRI, EA-682. January 1979.
Miller, Kent, and Scott Goldsmith. Energy Consumption in Alaska:
Estimate and Forecast. Prepared for Alaska Department of Commerce
and Economic Development, Division of Energy and Power by the
Institute of Social and Economic Research. Anchorage, Alaska.
January 1977.
J-7
New England Power Pool (NEPOOL) and Battle Columbus Laboratories. "Mod-
els for Long Range Forecasting of Electric Energy and Demand."
June 30, 1977. (Available from Ne,.;r England Power Planning, 175 Brush
Hill Avenue, West Springfield, MA 01089; revised and updated version
forthcoming.)
Norgard, Jorgen. "Improved Efficiency in Domestic Electricity Use."
Energy Policy, pp. 43-56. March 1979.
Olsen, Marvin, et al. Beluga Coal Field Development: Social Effects and
Management Alternatives. Prepared for Alaska Division of Energy and
Power Development and U.S. Department of Energy. May 1979.
Parishi, Anthony J; "Electricity Use No Longer Soaring: Nation Expected
to Benefit in SO's." New York Times, 6 April 1980.
Personick, Valerie A. "Industry Output and Employment: BLS Projections
to 1990." Monthly Labor Review. April 1979.
Porter, Edward. Alaska OCS Socioeconomic Studies Program. Bering-
.Norton Statewide and Regional Economic and Demographic Systems
Impact Analysis. Institute of Social and Economic Research.
Anchorage, Alaska. January 1980.
"The Economic Impact of Federal Energy Development on the State
of Alaska." Harvard University. March 1, 1977.
Reaume, David M. Speech drafted for Chuck \.Jebber, Commission · of Alaska
Department of Commerce and Economic Development. February 1980.
Robert W. Retherford Associates. "Management.Audit: Electrical System
Performance-Engineering Aspects Evaluation." Matanuska Electric
Association, Inc. Palmer, Alaska. September 1979.
"Power Cost Study 1979-1993." Copper Valley Electric
Association, Inc. Anchorage, Alaska. Harch 1979.
Rogers, George W. "Analysis of the Alaska Economy and Its Future Outlook."
Financing Alaska's Employment Security Program, Vol. 2. Alaska
Employment Security Commission. Juneau, Alaska. October 1958.
The Future of Alaska: Economic Consequences of Statehood.
Baltimore: The John Hopkins Press. 1962.
Rural Electrification Association. "Power Requirements
2 Hatanuska, Matanuska Electric Association, Inc.
Hay .1978.
Study." Alaska
Palmer, Alaska.
"Power Requirements Study."
Association, Inc. Homer, Alaska.
Alaska 5 Kenai, Homer Electric
Hay 1978.
J-8
"Power Requirements Study." Alaska 8 Chugach, Chugach
Electric Association, Inc. Anchorage, Alaska. May 1976.
"Power Requirements Study."
Valley Electric Association, Inc.
Alaska 18 Copper Valley, Copper
Glennallen, Alaska. March 1979.
"Pmver Requirements Study 1980-1985." Alaska 6 Golden Valley,
Golden Valley Electric Association, Inc. Fairbanks, Alaska.
Saunders, Norman C. "The U.S. Economy to 1990: Two Projections for
Growth." Monthly Labor Revietv. December 19 7 8.
Schlaifer, R. Analysis of Decisions Under Uncertainty. Netv York:
McGraw-Hill. 1969.
Scott, Michael J. "Prospects for Bottomfish Industry in Alaska."
Alaska Review of Social and Economic Conditions. April 1980.
Southcentral Alaska's Economy and Population, 1965-2025: A
Base Study and Projection. Report of the Economics Task Force,
Southcentral Alaska Water Resources Study (Level B) to the Alaska
Water Study Committee by the Institute of Social and Economic
Research. Anchorage, Alaska. January 31, 1979.
"Southcentral Alaska Water Resources Study (Level B)." Phase I.
Technical Memorandum. Electric Power. Electric Power 1-lork Plan
Committee and Alaska Hater Study Committee. March 27, 1979.
Spetzler, CarlS., and Carl-Axel S. Stael Von Holstein. "Probability
Encoding in Decision Analysis." Management Science. November
1975.
Spurr, Stephen, et al. Rampart Dam and the Economic Development of
Alaska. School of Natural Resources, University of Michigan, Ann
Arbor. 1966.
"Status Report, House Power Alternatives Study Committee." Alaska State
Legislature. January 30, 1980.
Sweeney, Mark P., Joseph E. Feline, Katherine A. Miller, and Robert E.
Fullen. "Natural Gas Demand and Supply to the Year 2000 in the
Cook Inlet Basin of South-Central Alaska." Prepared for Pacific
Alaska LNG Company, Los Angeles, California. November 1977.
Swift, W.H. et al. Alaskan Electric Power: An Analysis of Future Re-
quirements and Supply Alternatives for the Railbelt Region. For
Alaska Department of Commerce and Economic Development, Division
of Energy and Power Development, and the Alaska Power Authority
by Battelle Pacific Northwest Laboratories, Richland, Washington.
Harch 1978.
J-9
Alaskan North Slope Ro y alty Natural Gas: An Analysis of Needs
and Opportunities for In-State Use. For Alaska Department of
Commerce and Economic Development, Division of Energy and Power
Development by Battelle-North~vest, Richland, Washington. September
1977.
Taylor, L.D. "The Demand for Electricity: a Survey." Bell Journal of
Economics. Spring 1975.
Tuck, Bradford. "A Review of Electric Power Demand Forecasts and
Suggestions for Improving Future Forecasts.'' Draft Sections.
Prepared for The House Power Alternatives Study Committee.
February 1980.
U.S. Department of Agriculture, Forest Service. _E_I_S ____ v~s~·-=B~o~r~a=x~M=i=n=i~n~g~
Access Road for Quartz Hill Proposal. 1977.
U.S. Department of the Army, Corps of Engineers, Alaska District. South-
central Railbelt Area, Alaska. Interim Feasibility Report. Hydro-
electric Power and Related Purposes for the Upper Susitna River
Basin. December 12, 1975.
U.S. Department of Commerce, Bureau of Census. Annual Housing Survey:
1977. "General Housing Characteristics, Part A." Current Housing
Reports, Series H-150-77. 1979.
Annual Survey of Manufactures 1976. Fuels and Electric Energy
Consumed: Industry Group and Industries; States by Industry Group;
Standard Metropolitan Statistical Areas by Major Industry Group.
Census of Housing: 1970 Detailed Housing Characteristics.
Final Re p ort HC(l)-B3 Alaska. Washington, D.C.: Government
Printing Office. 1971.
Census of Housing: 1970.
Final Report HC(l)-A3 Alaska.
Printing Office. 1971.
General Housing Characteristics.
\.J'ashington, D.C.: Government
Census of Population: 1970 Detailed Characteristics. Final
Report PC(l)-D3 Alaska. Washington, D.C.: Government Printing
Office. 1972.
Census of the Population: 1970 General Social and Economic
Characteristics. Final Report PC(l)-C3 Alaska. Washington, D.C.:
Government Printing Office. 1971.
"Demographic, Social, and Economic Profile of States: Spring
1976.'' Current Population Reports, Series P-20, No. 334.
Washington, D.C.: U.S. Government Printing Office. 1979.
Housing Vacancies: Fourth Quarter 1979. Current Housing
Reports. February 1980.
J-10
Projections of the Number of Households and Families: 1979
to 1995. May 1979.
Statistical Abstract of the United States. Washington, D.C.
Various years.
U.S. Department of Commerce, National Oceanic and Atmospheric Adminis-
tration, Environmental Data Service. "Local Climatological Data."
Asheville, North Carolina. 1978. Various issues.
U.S. Department of Energy. Information Kit: The National Energy Act.
Office of Public Affairs, Washington, D.C. November 1979.
"A New Start: The National Energy Act." Washington, D. C.:
Government Printing Office. 1979.
U.S. Department of Energy, Alaska Pmver Administration. "Potential
for Residential Heating Energy Conservation and Rene\vable Resource
Utilization In Southeast Alaska." Mimeograph. January 1980.
Upper Susitna River Project Power Market Analysis. March 1979.
U.S. Department of Energy, Energy Information Administration. End Use
Energy Consumption Data Base: Series 1 Tables. Washington, D.C.
June 1978.
U.S. Department of Interior. Alaska Natural Resources and the Rampart
Project. Washington, D.C.: Government Printing Office. 1967.
U.S. Department of Interior, Alaska Power Administration. "Devil Canyon
Project, Alaska." Status Report. May 1974.
U.S. Department of Interior, Bureau of Land Management. Proposed Five-
year OCS Oil and Gas Lease Sale Schedule, March 1980-February 1985.
Final Environmental Statement.
U.S. Department of Interior, Bureau of Reclamation, Alaska District.
"Devil Canyon Project-Alaska." Feasibility Report. May 1960.
"Susitna River Basin. A Report on the Potential Development
of Water Resources in the Susitna River Basin of Alaska." August
1952.
U.S. Department of Interior, Office of Minerals Policy and Research
Analysis. "Final Report of the lOS(b) Economic and Policy Analysis."
December 15, 1979.
U.S. Environmental Protection Agency. Alaska Petrochemical Company,
Refining and Petrochemical Facility Valdez, Alaska. Appendix,
Vol. II. 1979.
J-11
Wellen, Paula H. "Interior Alaska Energy Requirements." Fairbanks
North Star Borough, Community Information Center. Fairbanks,
Alaska. February 1980.
Wertz, Russ. "Housing Survey." Draft. Fairbanks North Star Borough.
Fairbanks, Alaska. 1980.
J-12