ALASKA ECONOMIC FORECAST FOR 1977

Daniel A. Seiver -Ïnstitute of Social and Economie Research

University of Alaska 707 11A11 Street~ Suite 206 Anchorage, Alaska 99501

Narch 2~ 1977

ALASKA ECONOMIC FORECAST FOR 1977

1. Sunmary and Hirrhlights

Alaska economie activity will experience a significant èecline

in 1977 as the trans-Alaska oil pipeline is completed. The unem-

ployment rate tvill be substantially higher than in recent years,

the rate of inflation (as measured by the Anchorage Consumer Priee

Index) vlill be the lmvest sin ce 1973. Almost all of Alaska 1 s support

sectors vlill be adversely affected, viith substantial employment de-

clines in services and trade, two of the largest groups of employers

in the state. Overall employment in the state Hill fall by 14 per-

cent belot-I 1976. This decline will occur in spite of assumed rapid

increases in employ~ent in the mining, manufacturing, and agriculture,

forestry, and fisheries sectors. Detailed forecasts:

U:NE?1PLOYHENT RATE: 12.1% for 1977 (highest ever recorded) 13.4% for first quarter of 1977 (highest in 3 years)

PRICES: 6.3% higher in 1977 (Anchorage C.P.I.) 5.6% for fourth quarter of 1977

CIVILIAN .EHPLOYHENT: Down 149& compared to 1976 last quarter of 1977 loi-;est since mid-1974

:&JI.

1

Trade:: llo~ 13.6% compared to 1.S7ô

Construction: Dawn. 4-6.4% c:ampared. ·to 1976

Transportaticm., CommU:nic:àtion~ and Pub.lic: Utilities Dmm 17.4-% compared ta 1976

Finance, Ins:u:ranc:e,. and Real Bst:ate: Dawn 18:.3% compared to 1976

State and Local GoverŒient: Up 2.6% con:rpared to 1976

2. Assumptions Hade for the Forecast*

United States

U. S . Consumer Priees: Up 7% over 1976

U.S. Wages:: Up 8% over 1976

Alaska

Hining Employment:

Manufacturing Employment:

Agriculture) Forestry~ Fisheries Employment:'

.;":'!:

Fede:ral Government Employment:

Up 15% over 1976

Up 15% over 1976

Ùp 15% over 1976

Unchanged in 1977

2

Pipeline Employment: 5,000 in 1977 (by quarter: 7,500; 8,000; 4-~000; 2~000)

Note: These assumptions are no.t derived):'rom I.S.E.R. ts econometrie madel. Rather, they are necë"SSary inputs to the madel (See Section 4). The accuracy of the forecast may dep:enFk.~"in part on the accuracy of these assumptions- · •'·'fù'""·: ·

*section 4 of this report explain~ the role of these assumptions in making forecasts of econocic activity.

~·o-:Food, lumber, and paper

3

3. Forecast Discussion

Figures 1-9 on the following pages dep.ict the extent of the

decline forecast for 1977. Pre.Frninary ernployrnent data for the

first two quarters of 1976 are also shown on the forecast graphs,

which can be cornpared with the 1976 results generated by the econo­

metrie model. ~·:

Figure 1 shoHs a rap.id increase for the state's unemployment

rate beginning in the fourth quarter of 1976. (Very prelirninary

data for the last half of 1976 suggests this rise bas in fact taken

place.)· The peak of 13.4 percent in the first quarter of 1977 is

higher than any quarter in the last three years, but still not as

high as the rate for 1974:1. In 1977, hOivever, unlike previous

years, the unemployrnent rate will not fall dramatically in the

second and third quarters, and again exceeds 13 percent in the

fourth.quarter. Thus, the annual average of 12.1 percent will be

the highest since record-keeping began in 19~~7.

Figure 2 presents our forecast for civilian Hage and salary

employrnent. The rnid-year surge of 1975 and 1976, as the pipeline

workforce rose to its peak, \vill be shai'ply attenuated in 1977.

By the fourth quarter of 1977, Hith the pipeline construction work-

force down to 2, 000, employrnent ~-rill fall to the lotvest level sin ce

...

.. The creation and testing of the model is discussed in Section 4.

..

1974:2, the commencement of the pipeline boom. The 1977 average

shOifS a 14 percent decline from 1976, the sharpest decline since

record-keeping began in 1947.

4

Priee inflation will continue the moderating trend of 1976;

with priees rising only 6.3 percent for the year, and increasing

at an annual rate of only 5.6 percent by the end of 1977. This

trend is graphed in Figure 3. The rate of inflation in Alaska is

measured by the change in the Anchorage Consumer Priee Index.

Priees in other areas of the state, notably Fairbanks, may increase

even more slowly. The 1977 rate of inflation is, however, still

rather rapid compared to the pre-1974 years of mild inflation~

The trade sector is one o~ the largest support sectors in the

Alaska economy, and employmen\ falls by 13.6 percent in 1977, or

approximately 3,500 employees. This decline (graphed in Figure 4)

accounts for a sizable portion of the overall decline in employment.

A major portion of this decline Hill occur in Anchorage, vThich ac­

counts fpr three-quar·ters of the trade employment in the state.

A slightly more severe decline occurs in the service sector

of the state economy, graphed in Figure 5. The annual percentage

decline of about 18 percent again will be concentrated in Anchorage,

the service center of the state. The fourth quarter of the year

will be the lovt point for 1977, vlÎth service employment falling to

J,

-_#'

'-~ .. --../

1 .

. ,

---·-----~-~··-·

5

its lowest level since the beginning of 1974. The annual percentage

decline in 1977 is equivalent to approximately 4,000 employees.

The construction sector, which exhibited phenomenal grmvth in

the 1974-76 period as a direct result of the pipeline~ will show

a dramatic decline in 1977 of close to 50 percent, as pipeline em­

ployment falls from a peak of over 20,000"in mid-1976 to 2,000 by

the end of 1977 •· (Figure 6) Non-pipeline construction falls also,

but by a rouch smaller amount (23 percent or 3,000 jobs). The

dramatic variations in pipeline employment have been~ and will be,

a prime determinant of the fluctuations in economie activity in

the si:ate.

The downward trend in employment in transportation, communica­

tions, and public utilities is graphed in Figure 7. Employment in

the transportation sector, by far the largest employer of the three,

falls the fast est ( 20 percent) ·in 1977, follm.ring a probable slight

decline in 1976 from 1975. Communications emplojtment .falls only

7.5 percent and public utilities, which comprises less than 10 per­

cent of this sector, falls 17 percent. The transport sector also

was directly influenced by the construction of the oil pipeline and

has been the first sector to experience post-pipeline weakness.

The forecasted employment decline in the finance, insurance,

and real estate sector is graphed in Figure 8. The annual percentage

d~cline of 18 percent may be a little stecper than will actually occur,

6

'based on very preliminary data for the last two quarters of 1976.

Again, the major portion of this decline will take place in Anchor­

age, t-~hich is the financial center of the state.

Last, but certainly not least of the support sectors of the

economy, is sta·te and local government. Employment in this sector

is forecasted ta rise in 1977, reflecting higher spending levels

by the state in fiscal year 1977 and especially fiscal year 1978,

beginning July 1, 1977. This increase in employ~ent will have an

ameliorating effect on the decline in economie activity, but cannat

prevent it. Figure 9 shows ·the forecasted trend in state and local

government employment. Employment in this sector showed little

growth in 1976, in contrast to the rapid groHth in most ether

sectors.

The I.S.E.R. short-run nodel cannet make regional forecasts.

Yet it is obvious that rouch of the forecasted decline in employment

Hill take place in the Anchorage area, vrhich accoun.ts for more than

half of the· state' s support sec tor em~:>loyment. The Fairbanks area

1-:ill no doubt suffer a greater relative decline in employment, and

probably higher unemployment also, given the·dramatic boom in the

area in 1974-1976.

~

..

Percent of Labor Force

16 -

14 -

12 -

10 -

8 -

6 -

l.j. -

2 -

\ \

\

·\ \ \

l

\ 1 \ 1 \ 1

...... ' 1 ........

........ 1 'J

1

1 1

1

1 1

1 1

" 1 \ 1 \

1

o- r 1 2 .3 l.j. 1

1976

\ \

\

Figure 1

UNEt1PLOYHENT RA TE

\ /

'----.../

2

1977

l 3

/

/ /

/

/

'I.J.

7

···- W·'ililif"":'

, ...... •

thousands

200 -

175 -

150 -

125 -

100 -

75 -

50 -

25 -

8

Figure 2

CIVILIAN HAGE AND SALARY EMPLQYHENT

-'\ -- \

\ \

/

f'" ',

\ /"--

'-, / ---, g ' / '

' ' / ' '-/

' ' '

0 -1

J ' 1 J • • 1 2 3 . 4 1 2 3 4

1976 1977

annual percent increase

16 -

14 -

12 -

l 10 -1 ~ ~ /;

8 -1 ··~ f; ,'! 1

1./ ' 1 1 '

1/

6 -1 ,tj/ ' 1 ' 1

;Î; ;; 4 -1 1~

2 -1 r / 1

0 -1

1

~ / '

! /' r

/ 1 /

/1 / /

[ ' / //, li lj 1/

,' 1 .' 1

/; 1/ /

// 1

f j

2 3

1976

(actual)

4 1 2·

Figure 3

INFLATION RATE (Anchorage CPI)

3 4

1977

(forecast)

9

thousands

40 -

35 -

30 -

25 -

20 -

15 -

10 -

5 -

0

-- "---... .......

/ ;- ',

1 2 3,

1976

....... , .......

4

..... .............. ....._ _

1

Figure 4

TRADE ENPLOYHENT

_ ..-..._

2

' ' ' '

1977

....... ....... -..... ___

3 4

10

thousands

40 -

35 -

30 -

25 -

20 -

15 -

10 -

5 -

/- ---, ' ' / '

/ '-..

Figure 5

SERVI CES EMPLOYt·lENT

..... - ........ -----~ ., ' .......

"-.. ........

....... --

. 0 - 1 1 1 1· 1 1 1 1 2 3 4 1 2 ( 3

1 4

1976 1977

i

11

. ~ ....

thousands

40 -

35 -

30

25

20 -;

15

10 -:

5

~

,.."\ // \

(/ \ If \

0 -· 1 1 1 2 3 .

1976

\

\ \ \ \ \ \

'"

\ \

~~

\ \

\ \

. ' ~ ..

12

Figure 6

CONSTRUCTION Et1_PLOYNENT

\ /"\ \ / \

\ / \ v/ \

\ \

1 2

\

3

' \ ' \

1977

\

'

4

·~·· ... ~,.

thousands

20 -

15 -i

1

10 -

5 -

---, r- ' / \

1 ' '\

13,/:c~'

Figure 7

TRANSPORTATION, COHMUNICATION, AND UTILITIES Eî1PLOYr·1EHT

/ ' '\ 1 \ Y ' ,A.. / ', / ',

' / ' ,....,.. ', ' " ' '

-,

' 1 0 - 1 3 l.j,

1 2 1

1 1

2 1 3

;--4

197€) 1977

"..:

"""t~

~~·-

thousands

a -

7 -

5 -

5 ,...

4 -

3 -

2 -

1 -

/ ....... ___ ,

' ' ' ' ....... ......... ......

114-

Figure 8.

FINANCE, INSUR:\NCE, AND REAL ESTA'tE ENPLOYMENT

...... _- --, ' ' '\.

' "'- - - -- ...

0 -~----~-------,...r------~-------r-----,-----~------T-----~----1 2

1976

3 4 1 2 3 4

1977

· ... )·"" ~

~

thousands

40 -

35 -

30 -

25 -

20 -

15 -

10 -

5 -

15

Figure 9

STATE AND LOCAL GOVERNt1ENT EHPLOYt1D~T

~__,--_______ __,

...... / ..... / ~ -...... ~

/

/

/

/----

0 - 1 1 1

2 3 4 1 1 2

1 3

l 4

1976 1977

··~·~.t'!<

•

4. Descriptipn and Background of I. S.E. R. 's Quarte,t':iYt:~.Econometric HodeF:

·-·, .... :. .; ~ -;- . ··~ ·· .. · ....... :.)

Beginning in 1978, the state of Alaska will receive almost.

one billion dollars per year in royalties and taxes from Prudhoe

Bay oil production •. The long-term consequences of this massive

flo:.; of "pe·tpodollars" have been studied by ·the Institute of

Social and Economie Researeh at the University of Alaska for the

past three years. SevePal long-run policy-oriented P-conometrie

models of the state have been constructed. These models are

d~iven by resoul"'ce development scenarios which project the employ-

~ent and revenue impacts of furthel"' development of Alaska's petre-

leu--n resources, and by state e:-:penditures, _a key policy variable

w .. l:ich, given the massive 11petrodollarTl flo:·is, can be manipulated

by policymakers Hith major effects on the economy. ~·:.;:

Long-term models of Alaska ts economy tend to shm·7 relatively

smooth and steady growth. Yet quite recent experience is demon-

strating once again the "boom-bust" nature of the Alaska eeonomy.

In order to captm'e the short-run eharaeteristics of the Alaska

economy, a quarterly econometrie nodel has been eonstructed and

ùA :-:;ore detailed and technical discussion of model construction, estir.1ation, and testing is contained in raniel A. Seivêr, 11A Quartcrly Hodel of the Alaska Eednomy," February, 1977. This r::odel was crcated using the TROLLsystemof the National Bureau of Economie Research, Înc •

....•.

16

.... These moclels are described in detail in David T. Kresge, "Alaska' s GroHth to 1990, 11 Alaska Revievr of Business and Economie Conditions, vol. 13, no. 1, January;'1976:"- --

; .. , ··:·'"' ..

. '

·v

·j

·~

.\J_.

,•

has heen tested against historical data, and used for the forecast

discussed above.

The structure of the model is diagrammed in Figure 10. The

symholic notation can he decoded hy referring to the symbol dic­

tionary on page 1 ~ • All non suffixes have heen deleted from

Figure 10 to simplify the diagram.

Employment ,in Alaska's support sectors is a function of local

17

-demand, vlhich is measured by real dispos able personal income (DPIRQS).

Almost all the demand-employment relations contain seasonal dummies,

reflecting the highly seasonal nature of the Alaska economy. Those-

variables ·Hhich do not exhibit seasonal val'Îation are marked ;-lith J'.

an asterisk. •· Several support sector employment equations have

sJ.ightly differing specifications. During the historical period,

much of the communications network '"as turned over to private in-

dustry hy the Federal government, and a special dummy has been

added to the communications employment equation to account for this

transfer. The transport sector, particularly trucking and air trans-

port, has been exogenously affected by oil development and pipeline

construction: a special dummy has heen added to the transport em-

ployment equation to capture this effect. During the pipeline con-

struction boom, construction employment has a large exogenous corn-

ponent (ECONXQ) l·1hich is forecasted separately based on periodic

xAppendix A lists all the equations in the model-and contair1s regres­. sion rcsults for· ali stochastic equations.

..

k···· .... •

. . Figure 10. ALASKA QUAR'rERLY Z.10DE:b.

c~.~~-=>­c~­c~)-

c "'!:US ):-

cl/EUS )--

-;>e----~->r~~'J~~-----1

c:~~;-)- "":.--,~ .. 0-.~--~ ...

r-~----->8 ~

( WEUS )- -->~

G

( \/EUS )- -- -->8 ~

( \l'tUS )-

Œ~-

'1

• t

-·

1 '

•

-,

1

1

t'

,. l 1

'; .

18'"

TRAD&

s u St:RVlCJ:S'

p

r TAANSPORTAtiOlf., •

0

a C0!-1:-:U!tiCATIONS

;

PlJBLIC. Ul'ILI.TIIS

s

E F.I.R.&-..

c

T P.MlUFACWIUNG-cOTdER

0

R

s

COHSTRUCTIOU •

STATE l~f!l LOC.O.t. GOVER!W.ENT

".&. x FLD::RAL GOVER!tHE!If· p 0 R T" AGP.ICULTUPJ::. FORE:STR'l.

FISHERlES

s E MAllJJFACTüRING-c T {Exc:ept Other) 0 R

MI!U:iG s

IŒ'l - - - - ~ stochastic relationship

( c~Î> )-

>;.-

~ ·- .. '-1\JI'

.;>J;;~ ";/v· , ,'

('

1 ,_

~>

- ... - -- .;.. _ÂJ.

'V' . 1

$;Y ~-· ~V/ -r::y--­

~-1:! li\ ~- -· l

.. 4J -- _.1

. ..

----------)>~ dcfinitional relationship

( • ) exogètJOu~ variable·

0 À

-k

"

cndo&enous varlable

rate: of chanr.e:

vnriablc lar,&ed k qu.ll'ters .. ~ "'""

tlO s1r,nHicant se..tsoraali~ var.iatlon

)

19

SYMBOL DICTIONARY

Sector Symbol Sector Symbol

Ag., For., Fish. A9 Hanufacturing (except other) !18 Communications Cl1 Hining pg

Construction CN Public Utilities PU Federal Government GF Services S9 Federal Hilitary Gr1 State and Local Governnent SL F.I.R.E. FI Trade Nanufacturing (Other) HO Transport

OTHER VARIABLES

Symbol

DNCSQ

DPIQ DPIRQ

DPIRQS E99SFQ

ECONXQ

EH9CQ LFCQ

Na me

Federal Pa}~ent for Alaska Native Claims Settlement Act

Disposable Personal Income Real Disposable Personal

In come DPIRQ Seasonally Adjusted State Government Non-Capital

Expenditures Exogenous Construction

Employment CiviliaL Employment Civilian Labor Force

Symbol

PC RPIQ S(i)

SFAC UQ URATE \VEUSQ HS99Q

VARIABLE PREFIXES

Coefficient - C Employment - EM

Wage Rate - t/R \vages and Salaries - l'IS

Example:

/EMr~ employment trade qm:œterly

D9 T9

Name

U.S.C.P.I. Alaska Relative Priee Index Seasonal Dummy for ith

Quarter {i-1,2)3) Seasonal Adjustment Factors Unemployment Unemp.lo:;went Rate U. S. Avel~age Heekly Earnings Total vlages and Salaries

Rate of Change - DEL {l)

__ _,.....--'

-~------

·~-.-

20

employment and employment projection reports obtained from the

pip~~ine construction consortium. The endogenous component of

construction is estimated first, using the·standard support sector

specification, and then exogenous construction employment is added

to de·terrnine total construction employment.

Hage rates (average earnings per quarter) in the suppor>t

sectors are functions of U.So quarterly average weekly earnings

(HEUSQ) o !:1ost of the h:.i.storical variation in support sector ï·Tage

rates can be explained with this variable and seasonal dummy vari-·

ables. Hm.;rever, the pipeline construction boom has had a major

impact on all 1-;age rates in Alaska.· To capture this effect, a

dummy variable, ECONXQ/Et-1CN'Q, has been added to most wage rate

equations.

The "exportu sectors of the madel have exogenously determined

employment levels, but endogenously determined ïvage rates o Federal

government ernployment has changed relatively little over the his- . . ~;,·;,-, ~:~

torical period. Host of the rr.anufacturing employment in.Alaska

consiste of food (fish) processing (!~5 percen·t of 1975 manufacturing

employment) and lwnber and paper manufactur.ing (35 percent of 1975

manufacturing employment). Host lumber and paper output is exported

to Japan, and food processing employment depends crucially on the

sizes of the relevant harvests of fish. It is thus not feasible

to tie Hhat mani1facturing employment Alaska does have to the

-21

national economy. The remainder of manufacturing employment is

responsive to local demand and is treated as a support sector.

Agriculture, forestry, and fisheries is a heterogenous sector

for v-Thich employment is also deterrnined outside the model. Alaska

has almost no agriculture. Forestry employment is heavily depend-

ent on supply management considerations, and fisheries employment

depends on supply considerations also. Total employment in agric~l-

ture, forestry, and fisheries comprises less than one percent of

total civilian employment.

Mining employr.1ent in Alaska is essentially petroleum employ-

ment, which is not sensitive to either local or national aggregate

demand conditions. Independent projections of mining employment

are contained in the resource development scenarios of I.S.E.R. ts

long-run models.

The state and local government sector is the policy sector of

this model. Host of the variation in state and in local government

wages and salaries can be explained by fiscal year state non-capitàl

expenditures. Thus, altering the level of the state budget has a

direct impact on v-lages and salaries in the government sec tor, and

thus has an indirect impact on support sector employment. Employ-

ment in the state and local government sector is defined as the

ratio of \.,rages and salaries to the \'lage rate, Hhich is determined 1

....

::'i1_. .... .., .. ,.. x ••·•·<~'•• ,,,.. .. ~•"•~"""'"'''''4'-....._• ,~.;:,,.,, ~• ,~~·--.'""'..,:.,j.~,.,_._..., .._; -:: ,f'"'-',;.;.~:-.<c""{','o-~,; .. ,.;.',.,;,_·-,.::·"' · ...... ·_ ;:·.;,~ ·-·.:· .. ~ ....... ,~ .. ~--··· .... ~._::~---~"--~-.

..

22

~by U.S. earnings. I.S.È.R. has estimated state non-capital expen-

ditures for fiscal year 1977 at $7LW million and fiscal 1978, $900

million, which reflects the first reeeipts of "petrodollars. 11

An estimate of Federal mili tary employment is sub·tracted from

the sum of sector employmt:m-t totals to determine total civilian

Hage and salary employment. Total Hages and salaries is similarly

the sum of all seetm's 1 wages and salaries. Disposable personal

ineome is determined by the relationship between Hages and salat'ies

and disposable personal incarne estimated on an annual basis. Dis-

posable personal incarne is then deflated by an Alaska relative pt>iee

index (RPIQ) to determine t'eal disposable personal incarne, and this

quantity is then seasonally adjusted~ based on histot'ieal seasonal

adjustment factors. The use of seasonally ~djusted disposable per-

sonal incarne in the support sector employment equations provides

an unambiguous interpretation of the seasonal dummy variables in

those equations.

The Alaska relative priee index is a regionally weighted func-

tian of the Anchorage, Alaska, consumer priee index, with an adjust-

ment made to reflect the difference in the level of priees betHeen

Alaska and the rest of the United States. Sinee Alaska produees

almost no consumer or produeer goods, mueh of the variation in

Alaska priees .ean be explained by U.S. priees. Sinee 1961, Alaska

priee;s have risen mo1~e slm-lly than u.-s. priees, reflecting reductions

·,:.

23

in transport costs and scale economies. The recent boom in Alaska

has shaHn, however, that local demand conditions can affect priees.

Thus, the priee level equation con·tains a proxy for the rate of

gro•·zth of local demand. In addition, in a quarterly madel, it is

appropriate ta specify a lagged response of Alaska priees to U.S.

priees. Th13 assumed geome·tric lag structure gives the familiar

lagged dependent variable on the right-hand side; in addition, the

regression v:as improved by lagging the U. S. C •. P. I. by one quarter.

The modeling of the state's labor market has been a difficult

task. No estimates of quarterly state population are available,

and interpolations are rnisleading since there is an important

seasonal component in net interstate migration ta Alaska. Popula­

tion is, therefore, not an input in the determination of the labor

force. Labor force is defined as the sum of total employment and

unemployment. Total employment is esse:ritially civilian vrage and

salary employment adjusted for self-employed and multiple job holders

and is thus a function of Et19CQ. Unemployment is a function of em­

ploy,,ent, seasonal d~mmies, the rate of grmvth of employment, a

special dumrny to account for a change in data collection methodology,

and the lev el of unemployment in the previoUs quar·ter.

The quartcl'ly madel has been estimated by ordinary least

squares regression. The historical period begins with 1965:1 and

ends \·:ith 1975:4.

.. .

24

Historical simulations have been run for the 1966:1 to 1975:4.

period, years ofrapid but uneven growth in the Alaska economy,

culminating with a sharp spurt induced by pipeline construction.

Figures 11 to 15 graph the simulated results for the historical

period again:st the actual data, for five key variables in the madel.

Table 1 lists the measures of "goodness of fit 11 for each variable

determined within the madel. These measm•es can shaH hmv \·wll the

madel 11 tracks" the historical period.

The true test of a forecast model, however, is its ability to

forecast beyond the historical data. The preliminary data for 1976:1

and 1976:2 (see Figures 1-9) indicate that the model can pass ·this

most rigorous test.

; J.

_1 ~

i

'

thousands

180 -

150 -

120 -

90 -

60 -

30 -

Figure 11

CIVILIAN WAGE AND SALARY EMPLOYMENT \

\ •

0 -~------------~~-----------------------~--------------------------------------~ 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

1\) <.n

..

Millions of dolléJ.rs

450 -

375 -

300 -

225 -

150 -

75 -

Figure 12

REAL DISPOSABLE PERSONAL INCOME

(Seasonally Adjusted)

0 -' ~~~--1966 1967 1968 1969 1970 1971

(,,-

1972 1973 1974 1975 1\) m

1

J 1

(1967=142.5)

240 -

Figure 13

ALASKA RELATIVE PRICE INDEX 200 -

160 -

::.-~ ·--- ·--~-.;.::.::.::::-

120 -

80 -

40 -

0 -~-------------------------------------------------------------------------------------------------~ 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

1 4

"' -..3

"

Percent of Labor Force

15 -

12.5 -

7.5-\

5 -

2.5 -

~\

Figure 1L!-

UNEMPLOYMENT RATE ·

1

1 1

1 1 1 1

1 1

1 1 '/

0 ------------------------------------------------------------------------------------------1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 "" co

., ~

' ! ..

l'

,.

"

thousands

30 -

Figure 15

TRADE EMPLOYMENT

25 -

20 -

15 -

10-

5 -

0 -~--------------------------------~-------------------------------------------1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

1\) 1.0

Table :t

HISTORICAL SIMULATION STATISTICS

Va:::--iable R · S ? a • ~l. • .... E. M./\.P.E. b Variu.ble ~ n.,a R • fv.~ • S . 1 • .!:, •

. b r1.A.P.E • ----

E1.:S9Q 1.;..68 3.89 HSSLQ 7.02 5.73 E~·~CNQ 1\\ 6.48 ,, 5.13 HRSSQ 3.23 2'. 63

~ 7.83 6. 54 EI·!F!Q t 'tlRCNQ 6.10 4.66

Er-!?GQ 7.96 6.96 1-iRFIQ 2 ~'26 1. 77 r-,. , •.• ~"C"Q ....,.... .. 5.39 4.31 HRPUQ 3:24 2.69

EHXOQ . 5. 92 4.'83 HRCt,lQ 5.68 4 •. 63 .E~1TSQ 4.39 3.24 "tiRA9Q :1.8.47 14.32 Er-m9Q 3 ··~ o'+v 3.01 WRP9Q 5.38 4.30 D?IRQS 3.03 2.54 HRGFQ 4.1.;.6 3.74 R?IQ 1.51 1.32 tVRT9Q 4.06 3.02

\.;'S99Q 3.05 2.43 '\·lRD9Q 2.18 1. 7S LFCQ 2.55 :1.. 97 i·iR!v!OQ 4. 53 3.65 Ei•19CQ 2.83 2.20 WRMSQ 4.99 . 4.11 URATE 9.85 ·a .14 HRSLQ 3.85 3.01

a = 1 r(Ai _Pi) r r b ~ [/Ai- PJl

R.:Ol •. S.P.E. i: . . • 100 M.A.P.E. = • 100

i Ai ~ Ai J N

N

A. = actual value in quarter i (A)

J. 0

Pi = predicted value in quarter i

N = number of observations