STRUCTURAL CHANGE IN THE ALASKAN ECONOMY: THE ALYESKA EXPERIENCE. and. Lee Huskey Assistant Professor of Economics

Transcription

1 STRUCTURAL CHANGE IN THE ALASKAN ECONOMY: THE ALYESKA EXPERIENCE by 0, Scott Goldsmith Assistant Professor of Economics and Lee Huskey Assistant Professor of Economics Institute of Social and Economic Research University of Alaska Anchorage* Fairbanks* Juneau paper presented at 29th Alaska Science Conference August 15, 1978

2 Inst. of Social & Economic Res University of Alaska 707 A Street, Suite 206 Anchorage, AK Structural Change in the Alaskan Economy: The Alyeska Experience by O. Scott Goldsmith and Lee Huskey In the last decade, Alaska has experienced very rapid growth, Stimulated by petroleum development and the largest privately financed construction project in history, the population grew from 285 thousand in 1968 to 413 thousand in an increase of 45 percent, Over the same period, personal income grew from $1.13 billion to $3.98 billion--an increase of 252 percent. This growth has been accompanied by structural change in the Alaskan economy, Not all sectors have responded to the petroleum stimulus to the same degree, so that the economy in 1978 is not only much larger than in 1968 but also somewhat more "mature," This structural change has not yet been analyzed. In framing future state economic policy, it is necessary to have not only an understanding of the present structure of the economy, but also the process of change by which the economy has evolved, This paper investigates the change which the Alaskan economy has undergone in the previous ten-year period, using the Alyeska pipeline project as the focus for the analysis, The change is examined in light of both the theories of economic development and of regional economics, The implications of this structural change for the future growth of the economy are discussed,

3 I. THE THEORY OF STRUCTURAL CHANGE The economic relationships which determine the flow of income, goods, and services are determined by the structure of the economy. Although the structure of consumption is important, it is the structure of production which is of primary interest in regional analysis. 1 The structure of the economy's productive sector can be defined by the distribution of employment, gross product, and income among industries. The structure of an economy influences the economy's overall level of activity, the level of prices, and seasonal and cyclical stability. The structure of an economy is also importantly related to its growth; the structure both affects and is affected by growth. A better understanding of the relationship between growth and the structure of the economy is needed to properly analyze the long-run growth of a region. I.a, Regional Growth Economic base theory is widely used in regional analysis, The theory states that a region grows primarily as a result of increased export activity to other regions. Supply factors such as transportation costs and labor supply are not considered to influence growth. The determinants of the demand for a region's exports are not assumed to be influenced by activity within the region. The expansion of the export sector leads to an expansion of the local support sector, which serves the export sector

4 1-2 and the population associated with it. The relationship between these sectors is usually described by a multiplier similar to the Keynesian income or foreign trade multiplier. One of the strongest statements in support of this theory was made by North. He argued that the growth of exports was the most important reason for growth in a region; he presented economic base theory as a long-run theory of economic growth. 2 In response, Tiebout argued that the theory was only valid in the short run and was not a theory of economic development. Tiebout pointed out that nonexport sectors such as government and local investment may generate growth even in the short run, Tiebout argued that the importance of exports as a determinant of regional income is inversely related to the size of the region. Tiebout pointed out that while an individual family may be totally dependent on exports, the United States exports very little and the world, of course, exports nothing, 3 The growth of a region may also result from improved technology and an increase in trade within the local economy, Increased intraregional trade results from the expansion of the goods and services provided by the local support sector, In spite of these reservations, the relation between the sector where the growth is determined by external demand (the exogenous or basic sector) and the sector where growth is determined by local demand (the endogenous or support sector) is still the primary concern in regional economic analysis. A major problem in examining the relation between the basic and support sectors is to determine which industries in a region are basic

5 I-3 industries, Since the size of the multiplier will be determined by this disaggregation, the distinction is crucial to regional analysis. The problem results because every industry has both basic and nonbasic components; the greater the level of aggregation, the more important this problem becomes. An Alaskan example is the construction industry which includes a basic component such as pipeline and federal governmentsponsored construction, a nonbasic component such as the construction of housing units, and an investment component which in the short run may be exogenous, while it is endogenous in the long run. An important support sector such as services has a relatively large basic component in hotel and motel service employment serving the tourist industry. The definition of support sector does not allow exact classification of industries; for example, are the direct suppliers to an export industry basic or nonbasic? Leven suggested that most traditional methods for separating these sectors misstate the importance of the basic sector. 4 The definition importantly affects the result of all regional analysis. I.b. The Effect of Growth on the Structure of the Economy The growth of the economy leads to changes in its structure; these changes result from both supply and demand factors. Structural change can result from a change in the structure of demand as changes in incomes and prices affect the structure of consumption, The effect of the growth of incomes on consumption is described by Engel's law; those goods which are income elastic will be an increased proportion of consumption as incomes grow, Kuznets suggested this demand effect was

6 I-4 partially responsible for the changing industrial structure he observed.. 5 as per capita incomes grew. In open economies, changes in demand may only change the distribution of imports unless supply conditions lead to the production of the goods locally. The cost of producing a good within a region relative to the cost of importing it (production cost plus transport cost) determines whether the good will be produced in the region. Regional growth does alter production cost, If economies of scale are obtained in production, the average cost of production falls with increases in the rate of production, Firms which can achieve economies of scale can lower their unit cost as the demand for their product increases. Growth in demand through both increases in income and population will allow firms to reduce their costs through internal economies of scale. Growing economies also provide external scale economies, These allow firms to lower their unit production costs at all levels of production; external economies include such things as better distribution and warehousing systems, improved labor markets, and local suppliers. Growth by increasing the market size allows local firms to produce at lower costs and to substitute local production for imports; this leads to a change in the structure of the economy. Central Place Theorists refer to the size of place at which an industry enters the economy as the industry's threshold size. This is the market size where the industry can obtain the scale economies necessary to compete with imports. Industries whose goods are most expensive

7 I-5 to transport and which have no significant scale economies have the smallest thresholds; services such as barber shops and medical clinics are examples of industries with small threshold sizes. The substitution of local production for the import of a good or service as regions grow past their threshold size leads to structural change. Chenery observed that in the growth of national economies, import substitution is one of the major factors responsible for the changing patterns of industrial pro. 6 d uct1on. Local and state governments may also experience certain threshold effects which lead to a greater-than-proportional expansion of government. For a particular service, governments may be subject to scale economies which reduce the average cost of the service. But as regions pass certain threshold sizes, governments may provide additional services not previously provided locally. The addition of these services could lead to greater-than-proportional growth in government employment. Import substitution is not the only production factor causing structural change which results. Growth will also affect the producers' optimal input mix which will affect the economy's structure. In determining the level of production, producers also determine their optimal mix of inputs to produce that level; most important is the capital-labor mix. As production increases, producers may change the relative amounts of labor and capital used in production. Whether the amount of labor per unit of output increases or decreases depends on the technology used in production and the relative prices of the inputs. The change of this ratio will influence the relative industrial distribution of employment and the economy's structure.

8 I-6 I. c. The Effect of the Economy's Structure on Its Growth The structure of an economy is not only affected by growth, but it also affects growth. Chinitz suggested that the structure of the export sector influences important determinants of growth such as bank lending patterns an d entrepreneurs h 1p.. 7 An export sector made up of large selfcontained corporations does not provide as great a catalyst for growth as an export sector dependent on other local industries. The structure of the export sector may also influence growth through backward and forward linkages. These linkages describe the process where the demand or supply generated by the export sector leads to growth of local suppliers or processors. The recently proposed Alpetco project is an example of a forward linkage from the Alaskan petroleum sector. Export industries with a greater propensity for backward and forward linkages will have a greater effect on regional growth. The structure of the economy will influence the economy's response to a major exogenous change. This relation is most often described in terms of income or employment multipliers. Multipliers describe the relation between exogenous change and the economy; they are most often used to project the effect on the economy of a given change. The region's industrial structure will determine how much of the incomes generated by a new or increased export activity will be spent locally. Because this is the relation described by the multiplier, the structure will affect the multiplier.

9 1-7 The multiplier has been observed to vary over time in a given region as well as across regions of varying size. These observations point directly to the importance of the structure of the economy on the multiplier and regional growth The size of the region influences the size of the multiplier, Ullman and Dacey observed that the prof tota 1 emp 1 oyment int portion o. h e support sector grows as regions. grow. 8 Regions which produce more of the goods and services they consume have a greater response to exogenous change, since they have fewer leakages from the economy, The change in the economic structure with increased size leads to a change in the multiplier relation, Time, as well as size, has been found to influence the size of the multiplier, McNulty has shown that the explanatory power of a simple economic base model is improved when the time period between the change in the basic sector and the nonbasic response is extended, 9 Others have posited a long-run adjustment process to exogenous change, so that the multiplier increases over time. 10 Such an adjustment process is a process of structural change, In the short run, economies respond to external change by simply expanding existing operations, while changes requiring investment and planning need longer lead times, Because of this, changes in the economy's structure as a response to some exogenous change may change the multiplier,

10 I-8 I.d. Structure and Growth The relationship between the structure of an economy and its growth is important when attempting to assess the long-run effect of a particular exogenous change on an economy. This relation is particularly important in small areas where the change is large relative to the size of the local economy. Large areas can absorb larger exogenous changes without a change in their structures, so the traditional multiplier may be appropriate. In small regions, there is likely to be a greater possibility of structural change in response to the exogenous change. The response may be much larger than that measured by the traditional multiplier which does not account for structural change. Structural change resulting from a given exogenous change is more probable in a small region because of the relative size of the change to the regional economy. Such an economy is small, measured by its existing stock and additions to either capital or labor. Exogenous changes in either of these will constitute major changes to the economy. Traditional economic analysis such as economic base theory describes the economic reactions to marginal changes. Major changes will affect the structure of the economy. Small, immature economies traditionally have underdeveloped support sectors; that is why small economies import more. There is more opportunity for reaching scale economies in the support sector and achieving import substitution. Once the range of support sector activities has been achieved in a mature economy, growth will merely expand these sectors, not add additional ones. The size and

11 1-9 maturity of an economy influences the effect on economic relationships of exogenous changes to the economy, Growth is accompanied by investment, In small areas experiencing relatively large changes, this investment is of two kinds. First, there is investment which allows the economy to expand with the exogenous change; this allows a proportional growth described by the multiplier. The second type of investment allows a broadening and deepening of the economy which leads to changes in the economy's structure, Since invest~ ment must accompany this growth, the timing of growth may be affected. The short-run response of the economy to growth will be less than the long-run response, since time to discover the opportunities for investment as well as to actually make the investment will delay the investment response, These delays may also lead to a lumpiness of growth, When the export industry reaches a given size, suppliers or producers enter at once, leading to big jumps, The same may be true of the support sector if thresholds of many industries are reached, Another cause of lumpy development is the existing capacity of the economy. Capacity constraints occur in both labor and capital. The more rapid the exogenous growth, the more important will be supply constraints, Economies can expand smoothly as long as there is excess capacity. Once a capacity constraint is reached, it will slow growth until either labor or investment increases the capacity,

12 I-10 The concern about the links between structural change and growth will be emphasized by examining the recent history of the Alaskan economy. The question is one faced in any small area attempting to assess its long-range growth, particularly areas in the western U.S. facing large-scale energy development. The concerns brought by this link are not new; the debate between balanced and unbalanced growth in development literature has addressed many of these issues, The focus on the growth of small, thinly developed regions such as Alaska makes this link important, Traditional multiplier analysis is appropriate when changes are marginal; expansion of the analysis to consider the links to the structure of the economy is necessary when the exogenous change is major,

13 II. THE STRUCTURE OF THE ALASKAN ECONOMY: GENERAL PERSPECTIVE Table 1 describes the changing pattern of employment in Alaska over the period since Although total employment has fluctuated in response to stimulus from the basic sectors of the economy, which includes government, commodity producing industries, and parts of the construction industry, a direct relation between the basic and nonbasic sectors is not obvious. The basic sectors have not always grown in the same direction as total employment. Military employment has been generally declining since the early 1950s, while civilian federal government has grown slowly. The commodity producing industries declined in employment during the 1950s; however, the growth of the petroleum industry has reversed that trend. In 1951 federal government, commodity production, and construction accounted for 73.8 thousand jobs out of a total of 99.7 thousand. Twenty years later, in 1971, they accounted for only 65.9 thousand jobs out of a total of thousand. Jobs in the basic sectors fell by 7.9 thousand, while total employment grew by 28.7 thousand. These changing relationships between the basic and support sectors are revealed in the ratios of total employment to certain components of total employment calculated in Table 1. Defining basic employment to include military and civilian federal, construction, and commodity producing, the ratio of total to basic employment has ranged from a low of 1,28 in 1953 to a high of 2.23 in The ratio has been increasing throughout the period with the most rapid growth of the ratio in

14 Table 1. HISTORICAL PATTERN OF COMPONENTS OF ALASKAN EMPLOYMENT ( ) (thousands) Government Other 1 Total/Basic Total/Basic Year Total Military Federal State/Local Construction Commodity Production Distributive 2 1/( ) 1/( ) H H I N other commodity producing includes agriculture-forestry-fisheries, mining, and manufacturing 2 Distributive includes transportation, communications, public utilities, trade, finance, insurance, real estate, services. Source: Before 1967: George Rogers, Alaska Regional Population and Employment, ISEGR, University of Alaska, December 1967, p. 63. After 1966: State of Alaska Department of Labor, Alaska Labor Force Estimates, annual.

15 II-3 the late 1950s and early 1970s. Essentially, the table shows the relatively rapid growth of the support sector of the economy. Whereas in 1950, there were 39 employees in the nonbasic or support sector for every 100 basic employees, by 1976 there were 121 support sector employees for every 100 basic employees. Over the 27-year period, for every change in employment in basic sectors of 100, there has been a 267-employee change in the support sectors. The fastest growing component of the support sector has clearly been state and local government which increased from 2.1 thousand to 29,3 thousand employees. If one assumes that they constitute basic employment, the ratio of total to basic employment increases from 1.34 in 1950 to 1.66 in The relative increase in the size of the support sector is independent of state and local government classification. The relation between the support and basic sectors has not only varied over time but also over space. In Table 2, estimates of basic employment, defined to include state and local government, mining, manufacturing, construction, federal government, and agricultureforestry-fisheries, have been used to rank ratios of total to basic employment for Alaskan labor market areas. These ratios for 1976 ranged from a low of 1.07 in the Prince of Wales labor market area to a high of 2.43 in Anchorage. This provides some indication of the spatial distribution of the support sectors of the economy relative to the basic sectors and indicates that the two are by no means coincidental.

16 Il-4 Table RATIO OF TOTAL TO "BASIC" EMPLOYMENT IN ALASKA LABOR MARKET AREAS Basica Basicb Total Total/ Total/ L3bor Areas Employment-'A Employment-B Employment Basic A Basic B Prin ::e of Wales* Yukon Koyukuk"< 3,654 3,271 4, Aleutians>'< 2,290 1,960 2,619 l.1l Haines:>', Barrow''' 5,905 5,253 6, Southeast Fairbanks,', 2,211 1,954 2, Outer Ketchikan* i".ngoon,'< Valdez:>', 6,225 5,492 7, Kuskokwim": Upper Yukon~'< Kodiak,l: 2,888 2,272 4, Sitka* 2,193 1,650 3, Cordova* , Br.istol Bay Borough* HrangeLl-Petersburg* 1, , Ju:i.eau 6,150 1,675 9, Seward:'< , Kobuk~~ , Bristol Bay* Kenai 4,037 2,944 6, Wade Hampton* Betheli, 1, , Skagway-Yakutat* , :Matanuska-Susitna 1, , State 92,900 63, , Ketchikan 2,724 1,680 5, Nome7< , Fairbanks 13,292 8,654 27, A:.::chorage 30,023 20,575 73, ~ote: Basic Sectors include Mining, Manufacturing, Construction, Federal Government, and Miscellaneous, alncluding state and local government. 0 Lxcluding state and local government. -~Division of Employment between basic and nonbasic sectors estimated by authors oeccuse of incomplete reporting resulting from disclosure regulations.,ource: Alaska Department of Labor, Alaska Labor Force estimates by Area, 1976.

17 II-5 It is more appropriate to include state and local government employment in the support sector in a statewide analysis. At the local level, there is more possibility that such activity is basic or export oriented. If state and local government were alternatively included as a support sector activity, the ratios would range from 1.26 in the Yukon-Koyukuk to 5.96 in the Matanuska-Susitna area. This illustrates the sensitivity of analyses to the determination of basic and nonbasic industries. The uneven distribution of this employment across the regions of the state accentuates the differences in regional structure as measured by employment. It is also evident that it is difficult to generalize about the size of the state and local government component of total employment on the basis of type of labor market area. Wide variation exists in both urban and rural parts of the state. To explore this somewhat further, Table 3 shows the ratio between state and local government employment and population in thousands for all labor areas of the state. This ratio in 1976 ranged from a low of 37 in Wade Hampton to a high of 238 in Juneau. There do not appear to be any obvious economies of scale in the provision of government services, at least with respect to population. Variation in state and local government employment among labor areas is also related to income. Table 4 indicates that the ratio of state and local government employment to personal income (in million$) varies from a low of 1.02 in Prince of Wales to a high of 9.52 in Ketchikan. Again, it appears difficult to generalize about this ratio varying in any systematic fashion among labor market areas of the state.

18 II-6 Table STATE AND LOCAL GOVERNHENT EMPLOYMENT PER THOUSAND POPULATION State and Locala Labor Area Government Employment Population Wade Hampton 159 4,287 Southeast Fairbanks 257 6,811 Aleutian Islands 330 8,282 Yukon-Koyukuk 383 8,238 Anchorage 9, ,179 Valdez-Chitina-Whittier ,000 Matanuska-Susitna ,010 Kenai-Cook Inlet 1,093 16,753 Kodiak 616 9,366 Kuskokwim 181 2,738 Barrow 652 9,609 Kobuk 332 4,788 Bethel 587 8,389 State 29, ,289 Nome 474 6,644 Upper Yukon 148 2,012 Wrangell-Petersburg 389 5,218 Haines 142 1,850 Sitka 543 6,883 Skagway-Yakutat 229 2,812 Seward 281 3,395 Prince of Wales 176 2,000 Bristol Bay Borough 112 1,252 Fairbanks 4,638 51,511 Ketchikan 1,044 11,394 Cordova-McCarthy 246 2,353 Bristol Bay 395 3,500 Angoon Outer Ketchikan 324 1,761 Juneau 4,475 18,760 b Employees per Thousand Population a.alaska Department of Labor, Alaska Labor Force Estimates by Area, balaska Department of Labor, July 1 Estimates of Resident Population.

19 II-7 Table STATE AND LOCAL GOVERNMENT EHPLOYHENT RELATED TO PERSONAL INCOME Personal Incomea Per Capita Labor Area (million $) Personal Income Prince of Wales ,650 Nome ,939 Seward ,629 Haines ,270 Skagway-Yakutat ,543 Kuskokwim ,967 Cordova-McCarthy ,522 Kobuk ,720 Bethel ,696 Angoon 3.7 7,490 Barrow ,913 Juneau ,485 Bristol Bay ,543 Outer Ketchikan ,928 Southeast Fairbanks ,614 Aleutian Islands ,792 Valdez-Chitina-Whittier ,538 Anchorage ,715 Fairbanks ,857 Kenai-Cook Inlet ,312 State ,616 Sitka ,693 Matanuska-Susitna ,773 Upper Yukon ,294 Wrangell-Petersburg ,931 Kodiak ,783 Wade Hampton ,362 Bristol Bay Borough ,304 Yukon-Koyukuk ,062 Ketchikan ,628 Employees per $1,000, , a U.S. Department of Commerce, Bureau of Economic Analysis, computer printouts (residence adjusted).

20 II-8 The structure of the economy influences indicators of economic activity other than employment levels. Two such indicators important in Alaska are seasonality and unemployment, The change in the structure of employment over time has had a marked impact on the overall seasonality of employment. Table 5 shows the level of employment in the highest, lowest, and average employment month for several industries over a 25-year period, Also included are ratios of high and low employment to average, This provides a rough measure of seasonality in that industry (uncorrected for variation in secular growth rates,) A value of l+g for the ratio of peak month to average month employment, where g is the secular growth rate of employment in the industry, indicates a lack of seasonality in the industry. Looking at total nonagriculture wage and salary employment, the table shows that the seasonality factor has declined significantly as the economy has expanded, At least two factors have been operating to reduce seasonality, First, the relatively rapid growth in the nonseasonal support sectors and state and local government have tended to reduce overall seasonality. Second, seasonality in some of those sectors traditionally associated with seasonal employment has declined, This is most obvious in the manufacturing sector as the result of growth of its nonfood processing components, It has also occurred in the mining and construction sectors, where technological change has contributed to an extension of the work season, Seasonality has not been eliminated by the changed pattern of employment, but in relative terms it has declined, 11

21 ~=H=~:~ ¾l(,i., $~1fu~i144i.WAA.w~~k :. ~,.. ~uu.w1~~,'m~ Ta l:q..!l..i:.. SEASQNALUY IN_THE_~ASKAN 1----;;;-----,--- ECONOMY 1950 I '60,... ~,...:..(,," r 1975 ratio to I ratio to ratio to ratlo to ratio to ratio to employment aver.:.ge employment avera:;:e em;:-ioyment all'.<"!:.ig.::, '.::m?lojmer.t average I employment average I employment average TOTAL High 55, , , , I 103, I 182, Low 30, , , , , , Average 41,800 48,900 56,~00 70,500 92, ,300 MINING High 2, , , I 1, I 3, lr, Low 1, , , Average 1,900 1,300 1,100 1,100 3,000 3,800 H H CONSTRUCTION I High 9, , , , , , " Low 2, , , , , , Average 6,300 6,400 5,900 6,400 6,900 25,900 MANUFACTURING High 13, , , , , , Low 1, , , , , , Average 5,700 5,000 5,800 6,300 7,800 9,600 GOVERNMENT High 16, , , , , , Low 15, , , , , , Average 15,700 19,400 22,700 29,700 35,600 47,200 SUPPORT High 14, , , l 29, l 41, I 80, Low 10, , , , , , Average 12,000 16, ,300 26,900 38,900 73,800 Source: Alaska Department of Labor, Nonagricultural Wage and Salary Employment by Place of Work.

22 II-10 In contrast, the change in the structure of employment has not had a significant impact on the rate of unemployment. Table 6 shows that, if anything, the rate of unemployment in Alaska has been edging upwards in the long run,

23 II-11 Table 6. EMPLOYMENT Al\JD UNEMPLOYMENT IN ALASKA (thousand) Unemployment Year Employment Unemployment Labor Forcea Rate l !f b ll b aprior to 1970, this is a work force concept (place of work rather than place of residence determined). b New methodology. Source: Alaska Department of Labor, Alaska Labor Force Estimates, annual.

24 III, THE ALYESKA EXPERIENCE III.a, The Years Immediately Preceding Construction Figure 1 charts the changes in employment by sector since In the peak impact year of 1976, total nonagricultural wage and salary employment was about 79 thousand above the 1970 level. Half of the growth, 39 thousand, occurred in the support sectors (defined to include transportation-communication-public utilities, trade, services, and finance-insurance-real estate) which in 1970 accounted for 40 percent of total employment. By 1977, the support sectors accounted for 48 percent of total employment, a much larger share of the total. The structure of the economy, in terms of employment percentages, had changed significantly over that period. Most of the growth occurred after 1973 when construction work on the pipeline actually commenced. This is particularly the case for the construction sector which had almost no growth between 1970 and In contrast, state and local government employment growth was evenly distributed between the periods 1970 to 1973 and 1973 to Growth in mining, manufacturing, and miscellaneous employment was strong, primarily in the post-1973 period. Total federal employment declined significantly between 1970 and 1973 and has remained fairly constant since then, (This is a reflection of a decline in military employment which is. not in the total figure.) Is it possible, from this general pattern, to separate out that component of total growth which can be accounted for by the construction

25 III-2 Figure 1. CHANGES IN ALASKAN EMPLOYMENT BY SECTOR 1970 TO 1977 T o--,\. 'ri.l (.0. so.i -"l '-t\r./,, /-----: SlfPf'oRT 30 I,, ST:4.TE"lf LOCAL I,GME(sflJ/1:cflT io - o/ C ()f'jstlw cnc1j '75 MlrJ;,,;c,,, N\1\Nl'. FA.Cf(.( ~/!JG;; M1SC. F!Jc.R4,Lc ({/j'lf 4 Iii l;l..i) /(I I l l1l/ 1'i?..Y Source: Alaska Department of Commerce and Economic Development, The Alaskan Economy: Year End Performance Report 1977, Juneau 1977, and The Alaska Economic Information and Reporting System Quarterly Report July 1978.

26 111-3 of the Alyeska project? As discussed in section I, there are a variety of reasons why a region's economy will grow, only one of which is an increase in activity in an export or basic sector activity. This is clear from looking at the period from 1970 to 1973 before construction activity began. Between 1970 and 1973, the combined employment of the mining, manufacturing, miscellaneous, and federal government (including military) sectors declined by 4.6 thousand, or 8 percent. Counteracting this decline was a growth in state and local government employment of 5.9 thousand, or 32 percent. If for purposes of illustration these sectors were considered the basic sectors, then because total employment increased by 11.4 thousand (including military), the ratio of total to basic employment change between 1970 and 1973 was 8.77 (11.4/( ))! Clearly, the increase in basic sector employment of 1.3 thousand alone could not account for this change in total employment. At least two other factors were also operating. First, income was increasing in the state and this stimulated growth in employment without basic sector increases in employment. In spite of a decline in military employment, military wages and salaries increased because of large wage rate increases in the military. And total federal outlays in Alaska increased over this period by 28 percent largely as a result of increases in the budgets of the Departments of Defense and Health, Education, and Welfare. Perhaps in response to

27 III-4 this, medical services was the fastest growing component of the support sector, increasing by 50 percent from 2.2 thousand to 3.3 thousand. Second, some economic activity occurred in anticipation of the construction of the pipeline. This undoubtedly accounts for a significant portion of the increase in activity in the support sector. The most rapid growth in services during this period with the exception of medical services was in the category of "other services" which grew by 40 percent from 5 to 7 thousand. This consists of repair services, amusement services, legal and educational services, nonprofit organizations, and "not elsewhere classified." Thus in assessing the growth during the pipeline construction years, it should be kept in mind that generally rising incomes was a contributing factor but also that some of the impacts, in terms of anticipatory spending, had already occurred. III.b. The Construction Years Figure 2 shows the changes in employment in the basic sector, in Alyeska construction employment, and in the support sector (defined now to include nonpipeline construction and state and local government) since In all years, support sector growth has exceeded that of the basic sector, By 1977, total employment had increased over 1973 by 51.4 thousand, and of this total, 8.9 thousand was basic. (This yields a ratio of 5.77 for the period.)

28 III-5 Figure 2. EHPLOYMENT CHANGES DURING ALYESKA CONSTRUCTION cco 1 E.r,.Pu)Y!Af.l>,. I " ,. 3UPPORT 30 j io j lo dc ~------~--'-~ S,6 Source: See Figure 1.

29 III-6 The lower diagram in Figure 2 shows the change in employment from year-to-year in both the basic and the support sectors. A lag in support sector response to changes in basic activity can be seen to be operating. In 1976, basic employment grew by 1 thousand, while support sector employment grew by 8,6 thousand, partially in response to the 10.8 thousand growth in basic sector employment in the previous year. It is difficult to determine the actual Alyeska employment levels and the industry or occupational mix of employment from the available data. Table 7 shows an estimate of total Alyeska employment by quarter and an estimate of Alyeska construction employment after adjusting for job turnover and nonconstruction employment reported by Alyeska. In the peak construction year of 1976, annual average construction employment was estimated at 15 thousand employees, which is considerably higher than originally had been expected. (See Appendix.) A considerable percentage of employment reported by Alyeska (26 percent in 1976) was estimated to be allocated properly to nonconstruction employment categories, primarily business services, transportation and warehousing, communications, and trade. The requirements of Alyeska const~uction thus required the direct employment of over 5 thousand in support activities. In a large and mature economy, these services would have been available without Alyeska having to supply them directly. As such, they would have been represented by increases in economic activity of the existing firms providing those services. The state employment statistics allocate Alyeska employment into the relevant categories so that a significant portion of Alyeska employment is not reflected in construction.

30 III-7 Table 7. ALYESKA ENPLOYMENT ESTINATES: QUARTERLY AVERAGE ( thousands) Adjusted Adjusted Business Construction Service Adjusted Othera Quarter Alyeska Reported Employment Emplo~en_J::_ Employment 1974: : : : : : :average : : : : :average : l : : a Includes primarily transportation and warehousing, communication, retail trade, wholesale trade. Source: Department of Labor unpublished worksheets.

31 III-8 Other basic sectors were expanding between 1973 and 1977 and account for the gap between Alyeska employment and total basic employment. In particular, oil and gas industry employment virtually doubled, driving mining employment from 2 to 4.4 thousand. The manufacturing sector grew by 1.4 thousand, but undoubtedly a significant percentage of that increase was the growth in nonbasic components of manufacturing. Large employment increases occurred in printing, petroleum refining, metal fabrication, and machinery, all of which primarily serve the local market in Alaska. Tourism and the Native Corporations are two additional sources of increase in the basic sector of the economy over this period. It is difficult to allocate employment increases in the service sector to tourist-related and nontourist-related activities, but it has been estimated that the number of visitors has grown from 180,000 in 1972 to 300,000 in If each of these 120,000 additional visitors contributed $500 to Alaskan personal income, the impact would be $60 million, or 1.5 percent, of 1976 personal income. This seems unlikely to have generated a significant part of the rapid long-run growth during this period, although it does reflect a recurring increase in activity rather than a one-time event such as pipeline construction and for this reason may be more significant than the figures indicate. Between 1973 and September of 1977, approximately $349 million had been distributed to the 13 regional Native Corporations. 13 This is equivalent to about $70 million annually over this period, although

32 III-9 the initial payment in 1973 was the largest. A portion of this money has been spent on goods and services in Alaska which have a multiplier effect, while some has been spent on the transfer of assets which do not involve an increase in economic activity, Some of the activities which would appear to be having a multiplier effect may actually be "backing out" other investment rather than increasing the total. Thus, it is impossible to assess the amount of basic activity which the Native corporations have generated. It appears to be roughly on the order of magnitude of the tourism industry. Of the change in basic employment since 1973, the majority has been pipeline construction related, It does not seem that the "wedge" in Figure 2 between Alyeska and total basic employment could be the cause of the maintenance of a high level of support sector activity after To suggest that such is the case implies that the growth in basic employment attributable to Alyeska in 1974 and 1975 did not cause the growth in support sector employment in 1974 and This does not appear reasonable. Alternatively, it has been suggested that the strength of the economy in 1977 was due to the fact that expansion in the nonbasic sectors from 1974 through 1977 "was primarily a delayed response to earlier growth in 'oil fired' state government spending and to federal spending. 1114

33 III-10 Between 1973 and 1976 fiscal years, the real value of federal outlays in Alaska remained virtually unchanged, except for a 1975 increase in the Department of Transportation budget. 15 In contrast, state and local government continued the pattern of expansion begun in the period 1970 to State and local government employment grew by 7 thousand with virtually all of the increase, 6.8 thousand, occurring at the local level. (The transfer of the state-operated schools to local control took place during this period.) A large increase in grants and shared revenues from state to local government allowed the rapid expansion at the local level which, in turn, was possible first because of and later in spite of Prudhoe Bay lease bonus revenues. The state and local government budgets grew in response to the availability of these revenues but did not contract in 1976 when those revenues were essentially depleted, Rather, the reserves tax was temporarily instituted. This seems to argue for a demand-related growth in government services between 1973 and 1977 rather than one driven by the supply of available funds. In fact, between 1973 and 1976, real state operating expenditures as a percentage of income declined slightly. 16 (Thus, it seems difficult to put state government behavior in this period into the category of basic activity, for it seems to have been demand driven.) Between 1970 and 1977, state and local government employment did grow by about 13 thousand employees and between 1970 and 1978, federal expenditures by 28 percent. It seems unlikely, however, that the economy which grew by 62 thousand employees in three years and then declined by

34 III thousand,.leaving it about 52 thousand higher in 1977 than 1973, was being sustained by this amount of growth in government. III.c. ~Indications of St~ The structure of the economy changed during the period of pipeline construction. The support sector has grown rapidly relative to the basic sector. It is not possible to establish causality from pipeline construction to support sector growth definitely because other factors have contributed to the support sector growth such as the normal increase in real incomes of individuals and the expansion of other basic sectors, but the pipeline was certainly an important contribution. Another contributing factor not yet mentioned was the synergistic effect of related developments or development possibilities (further oil and gas prospects, spending of state revenues, gas pipeline, capital move) on expectations of economic agents. Clearly, the lag effect mentioned earlier is also operating to keep the support sector strong. It will decline in 1978 but not return to its pre-pipeline relationship with the basic sector. It will not return to its old relationship, not because the tremendous exogenous shock to the Alaskan economy provided by the pipeline did not cause a large response in the economy, but because the exogenous shock caused the economy to grow, to mature; and this structural change resulted in a new relationship between the basic and support sectors. One example of this process of growth which has occurred is the Prinz Brau brewery which was established in Anchorage in Previously,

35 III-12 all beer consumed in the state was imported from the lower 48. Without a change in the structure of the economy, an increase in the size of the Alaskan market for beer would have resulted in the importation of more beer. The establishment of the brewery changed the structure of the economy by changing the labor and capital requirements in Alaska associated with the market for beer. The individual components of the support sector experienced different growth patterns during the pipeline years as shown in Figure 3 where percent differences in employment from the base year of 1973 are plotted. State and local government, trade, and finance-insurance-real estate have grown smoothly over all years. They have generally grown more slowly than the other sectors, except for finance-insurance-real estate which grew 80 percent between 1973 and These sectors appear to be more flexible in the short run to changing economic conditions and to grow in response to perceived patterns of change of a somewhat longerterm nature than the other support sectors. Transportation-communicationpublic utilities, services, and construction were more responsive to immediate changes in the basic sectors, providing much of the direct support in terms of infrastructure and business services to the pipeline project. Because of this, it is in these sectors that employment has fallen off with pipeline completion. It is interesting to examine in somewhat more detail the growth in some of the specific sectors. Space does not allow a complete analysis, but finance-insurance-real estate and services are two sectors where one

36 III-13 Figure 3. PERCENTAGE CHANGES IN EHPLOYHENT SINCE 1973.,,,,._ - / 1'(1-"f t,.,,. 7S YEAR Source: See Figure 1.

37 III-14 might expect import substitution to be occurring. Table 8 shows in an aggregate way what occurred in those sectors between 1973 and 1977 third quarter. Reclassification of firms because of revision of the Standard Industrial Classification (SIC) manual prevents one from drawing many conclusions from the data. It is possible to see some patterns, however. Holding and investment companies, miscellaneous business services, social services, and insurance carriers all grew by considerably more than 100 percent. Much of the growth in holding and investment companies could be attributable to the Native corporations, while business service growth is closely related to Alyeska. Social services was not a separate category in 1973, but the growth in child care services partially accounts for the large employment level in that category. Growth in insurance carrier employment is related to population growth and increased business activity. The declines of 46 percent in the education services and 13 percent in nonprofit membership organizations as well as the slow growth of 4 percent in personal services is attributable to the SIC revision which created the social service sector. If these sectors were combined, growth would be calculated at 39 percent. Other slow growing sectors include securities and commodities brokers at 5 percent and other amusements and recreational services at -6 percent. A reasonable hypothesis explaining slow growth in these sectors would be the existence of scale economies in the provision of

38 III-15 Table 8. EMPLOYMENT CHANGE IN BANKING AND SERVICES DURING ALYESKA CONSTRUCTION YEARS Sector Employment Percent Change 1973 to 1977 SIC 60 Banking 61 Other Credit Agencies 1, , Securities & Commodity Brokers 63 Insurance Carriers Insurance Agents 65 Real Estate , Combined Real Estate, Insurance, Loans, Law 67 Holding & Investment Co , Hotels & Other Lodging 72 Personal Services 2, , Misc. Business Services 7 5 Au to Repair 2, , Lt Misc. Repair 78 Motion Pictures Other Amusements and Recreational Services 80 Medical Services 327 3, , Legal Services 82 Educational Services 83 Social Services 86 Nonprofit Membership Organizations ,288 1, ,374 1, Misc. Services 1,729 2, Note: Some categories not comparable because of revisions of SIC codes. Source: Alaska Department of Labor, Statistical Quarterly, 1973:3 and 1977:3.