THE EFFECTS OF STATE EXPENDITURES ON RURAL POPULATION SETTLEMENT AND INTRASTATE POPULATION MIGRATION. Volume I. Theodore Lane Will Nebesky Teresa Hull

Transcription

1 THE EFFECTS OF STATE EXPENDITURES ON RURAL POPULATION SETTLEMENT AND INTRASTATE POPULATION MIGRATION Volume I by Theodore Lane Will Nebesky Teresa Hull Institute of Social and Economic Research University of Alaska December 1982

2 Volume I Table of Contents List of Tables V List of Figures VJ. l. Preface.. ix Chapter 1: Background I-1 Chapter 2: Patterns of Population Settlement Chapter 3: Expenditure Patterns Chapter 4: The Relationship of Expenditure Patterns and Other Factors to Population Settlement I-5 I-35 I-81 Ref er enc es I-91 iii

3 LIST OF TABLES Page Number I.l. Size Distribution of Communities 1-5 I,2, Population Change: (Non-Urban Communities),, I-9 I.3. Places with Population Declines and Net Out-Migration: I.4. Characteristics of In- and Out-Migration Places (Non-Urban Communities) I-15 I.5. Net Migration by 1980 Place Size (Non-Urban Places: ) I-16 I.6. I. 7. r.8. I.9. Relative Distribution of Non-Urban Places: 1970 and 1980 by Net Migration. Net Migration by Census Division Total 1981 Rural Expenditures Average Per Capita Expenditures by Size of Place, I-16 I-18 I-35 I-36 I. 1 Per Capita 1981 State and Federal Expenditures by Percent Native in 1980, I-38 I.11. I.12. I I.15. Total 1981 Per Capita Expenditures by Geographic Area, 1981 Total Operating Expenditures by Program Percent of Total Individual Program Operating Funds Obtained by Communities of 1,000-5,000 Population 1981 Total Capital Expenditures by Program 1981 Total and Per Capita Operating Expenditures by Percent Native and by Program Category I-40 I-42 I-44, I-4 7 I-50 V

4 List of Tables (continued) Page Number I. 16. I. 17. Percentage Distributions of 1981 Operating Expenditures by Percent Native, Place, and Program I Total and Per Capita Expenditures by Percent Native and by Program Category 1-53 I.18. I.19. Percentage Distributions of 1981 Capital Expenditures by Percent Native, Place Size, and Program Study Regions and Corresponding Census Divisions Percentage Distribution of 1981 Operating Expenditures by Geographic Region 1-71 Percentage Distribution of 1981 Capital Expenditures by Geographic Region 1-76

5 LIST OF FIGURES Page Number I.l. Cumulative Distributions of Population and Places I I.3. I.4. I.5, I.6. Alaska Regions Distribution of Places by Percent Native Changes in Population and Percent Native Net Migration: Regional Center Net Migration. I-7 I-8 I-10 I-29 I-31 I.7, I I,13, I,14. Distribution of Total Income-Transfer Expenditure by Size of Place.,,, I-45 Education Expenditures as a Proportion of Total Operating Expenditures by Place Size and by Native Cohort,,,,,,,,, I-59 Housing Expenditures as a Proportion of Total Operating Expenditures by Place Size and by Native Cohort,,,,, I-60 Income Transfers as a Proportion of Total Operating Expenditures by Place Size and by Native Cohort,,,,,,,,,, I-61 Per Capita Capital Expenditures by Native Population Cohort,,,,,,,, I-63 TCU Expenditures as a Proportion of Total Capital Expenditures by Place Size and by Native Cohort,,,,, I-65 Education Expenditures as a Proportion of Total Capital Expenditures by Place Size and by Native Cohort,, I-67 Housing Expenditures as a Proportion of Total Capital Expenditures by Place Size and Native Cohort,,,,,, I-68 vii

6 PREFACE During 1981, the Institute of Social and Economic Research (ISER) was requested by the Office of the Governor to study the relationship between state expenditures and rural population settlement patterns. We immediately realized that it was impossible to keep federal expenditures out of the study. First, some federal and state programs may substitute for one another. If a village were unable to obtain state community development funds to construct a bulk fuel storage tank, it might have sought assistance from a federal agency such as EDA. Second, the current national administration is cutting back many programs and moving others into block grants which will be administered by the state. As a result, historical distinctions between federal and state expenditures are becoming increasingly blurred. Finally, and most important, the distinction between federal and state expenditures may be important for identifying the source of an event, but it makes no difference in the impact of the event. The location of a high school, a hospital, or an airport in a particular village may influence its pattern of migration. Whether the same school, hospital, or airport is funded from federal or state expenditures would not be expected to make any difference. Much of the effort that went into the study consequently was involved in the construction of an expenditure data base and a village net migration data base. The patterns between these two data bases were then compared to determine if they were related. The village net migration data base was developed from data available from the U.S. Census of Population, 1970 and 1980, and the Vital Statistics Section of the Alaska Department of Health and Social Services. These data were combined to identify over 200 places in Alaska for which we could calculate net migration over the decade.l The data base of federal and state expenditures was built separately for operating expenditures and capital expenditures. It soon was obvious, however, that building such a data base back to 1970 extended far beyond the resources available to the study. We consequently decided to concentrate on documenting recent expenditures since these data were more readily available. The most complete data we were able to obtain were for Less complete data (particularly on the federal side) were available for FY Additionally, we were able to obtain a limited set of historical data on federal capital expenditures at the village level. 2 lsee Volume II, Sources and Definitions of the Data. ix

7 Throughout this study, we received invaluable assistance from many persons in state government, federal government, and with Native nonprofit organizations too numerous to be named. Our special thanks, however, go to Ann DeVries, House Research Agency, who allowed us to use the original worksheets from her study of state expenditures by election district in Alaska. Perry Shipman, Vital Statistics Section, Alaska Department of Health and Social Services, provided us with two essential computer runs with virtually no turnaround delay. Through the efforts of Frank Mcchesney, Economic Development Administration, U. S, Department of Commerce, and Al Robinson of the U.S. Department of Housing and Urban Development, we were able to obtain our only historical data. Finally, our heartfelt thanks go to Mr. Ted Volin, computer consultant extraordinary, who managed to organize our array of data into an organized data base capable of being analyzed. Chapter One presents a summary of the research on Alaska's population settlement patterns conducted in Alaska during the past decade. Chapter Two is a discussion of population settlement patterns in Alaska. The discussion covers nonurban, nonmilitary places for which we were able to estimate population growth and net migration over the past decade. These estimates are discussed with respect to their ethnic, place size, employment, and geographic location characteristics. Chapter Three discusses federal and state expenditure patterns for the same nonurban places. Capital and operating expenditures are discussed separately to see if different trends emerge. These expenditure data were only available for FY 81, and they are related to the 1980 place characteristics data. Finally, Chapter Four presents some conclusions and speculations about various factors which appear to be influencing the pattern of population settlement in nonurban Alaska. X

8 CHAPTER ONE BACKGROUND Introduction Many researchers have commented on the long-term trend of Alaska's Native population to migrate into the state's urban areas. George Rogers stated that this settlement pattern changed Natives from being established village residents into a "floating population" which lives in Anchorage and Fairbanks. 1 In a similar manner, William Alonso and Edgar Rust commented, "In the 1960s, it became clear that a major Native migration to the cities was underway. Its significance was confirmed, but probably understated, 11 by the 1970 Census. 2 Most recently, the Alaska Department of Labor stated, "Migration in Alaska tends to draw Natives from the rural areas to the urban areas. Due to this fact, the urban Native population expanded at a rate nearly four times greater than the 113 corresponding rate for the rural Native population. While there is general agreement about the trend, there is no consensus about either its causes or its implications for population settlement in Alaska. Arthur E. Hippler studied twelve villages in the Kobuk-Seward Peninsula area and concluded that the data 11 fragmentary as it is at present, still strongly suggest that whether regional service centers such as Bethel, Kotzebue, Barrow, and Nome continue to grow in size and attract migrants from smaller villages, and whether Natives continue to move into Fairbanks and Anchorage, many smaller villages will not disappear or even decline 11 dramatically in the foreseeable future. 4 The basis for this conclusion was that the absolute number of persons in the population below the age of 15 was so great (particularly given the high Native birth rate) that even with substantial net out-migration, there would still be enough adults to maintain the size and viability of most villages. lgeorge W, Rogers, Alaska Native Population Trends and Vital Statistics, (Institute of Social, Economic, and Government Research, University of Alaska, Fairbanks, Alaska), ISEGR Research Note, November william Alonso and Edgar Rust, The Evolving Pattern of Village Alaska (Federal-State Land Use Planning Commission for Alaska, Anchorage, Alaska), Commission Study No. 17, March 1976, 3 Alaska Native Report (Research and Analysis Section, Alaska Department of Labor, Juneau, Alaska), January Arthur E. Hippler, Some Observations on the Persistence of Alaskan Native Village Populations (Institute of Social, Economic, and Government Research, University of Alaska, Fairbanks, Alaska), Research Note No, Al, September 1969,

9 Dorothy M. Jones studied a series of villages in the Aleutian Chain, however, and came up with a somewhat different conclusion.5 She found that most inter-village migration is to nearby villages with which the migrants had familiarity and either family or friendship ties. She found the prevailing pattern to be one where "members of villages lacking local means of earning an income move to the nearby village that offers the best job 11 opportunities and community services. 6 Her findings indicate that the mobility of village residents is directly related to their familiarity with other villages and the availability of transportation systems. Although she found that the policies of government and industry favor the location of community services, facilities, and job opportunities in centralized places, she cone luded that the "isolation" of the village was a key determinant of its long-term survival. The pattern of Alaska Natives to move, at least initially, to places relatively close to their village home was also found by Hippler and Wood.7, In a study of several Eskimo villages, Norman A. Chance found that the integration of local employment o portunities to the village culture was a key factor in migration. 8 In one place, the development of wage employment opportunities involved no loss of respect or prestige for older village residents and supported traditional leadership patterns. This village had little out-migration, and its population doubled in over six years. Another village studied by Chance had few opportunities for local work, and those which existed acted to underline the traditional village lifestyle. This village declined by almost 100 residents in ten years and few of the out-migrants, or their children, ever returned to the village. Another case study found that attitudinal 5Dorothy M. Jones, Aleuts in Transition: A Comparison of Two Villages (University of Washington Press, Seattle, Washington) 1976; Changes in Population Structure in the Aleutian Islands (Institute of Social, Economic, and Government Research, University of Alaska, Fairbanks, Alaska), ISEGR Research Note A2, December 1970; Patterns of Village Growth and Decline in the Aleutians (Institute of Social, Economic, and Government Research, University of Alaska, Fairbanks, Alaska), ISEGR Occasional Paper No, 11, October 1973, 6Patterns of Village Growth, op, cit,, p, 34 7 Arthur E. Hippler, From Village to Town: An Intermediate Step in the Acculturation of Alaska Eskimos (Center for Urban and Regional Affairs, University of Minnesota, Minneapolis, Minnesota) October 1970; and Arthur E, Hippler and John L. Wood, "Alaska Eskimo and Indian Village Populations--the Social Meaning of Demographic Change," Etudes Inuit Studies (Volume 1, No. 1), Norman A, Chance, The Eskimo of North Alaska (Holt, Rinehart, and Winston, San Francisco). I-2

10 shifts of village residents resulting from increased dependence on wage employment and a desire for modern conveniences (and services) were representative of the out-migrants. This study, however, found that almost half of the people who left the village either returned to it or maintained close contact with friends or parents who were still living there.9 Summarizing many of the trends that had been observed by prior researchers, Alonso and Rust concluded that smaller places had an overall downward trend in their populations while larger places showed steadier, larger rates of population growth. They concluded that federal and state transfer payments, stockholder income for members of regional and village corporations, the growth of jobs, and the increased availability of public and semi-public services (facilities) were all acting to support the continuity of Native villages in Alaska, although larger places were rece1.v1.ng greater support than smaller ones. On the other hand, high Native birth rates, increased accessibility to urban areas, increased availability of radio, television, and telephones, and the absence of adequate wage and salary jobs were acting to move the Native population from the villages and into the urban areas. On balance, they cone luded that the "net migration of Natives to the larger cities will accelerate for another decade, but not at the rate which would cause more than a minor decline in the population of the village system [and] the larger places will continue to grow more than the smaller ones, and many of the smallest-size classes (under 100 persons, for example) will dee line In reviewing the literature on rural populatin settlement in Alaska, several things stand out. First, most researchers have limited their investigations to Native villages. Yet, a significant number of small places (villages) in Alaska are not dominantly Native. Of the 209 places for which we were able to collect consistent data between 1970 and 1980, 147 were places of 250 or fewer residents, Of these places, 38 (about 26 percent) had populations with 50 percent or less Natives. Among the smallest places, those between one and one hundred persons, 25 out of 65 places (approximately 37 percent) had 50 percent or fewer Natives in their population. Furthermore, among all persons residing in places with populations of one to one hundred persons, slightly over 40 percent were non-native. Most of these places with a majority of the population non-native appear to have identifiable locational patterns. They are clustered in the areas surrounding the Anchorage and Fairbanks Census Divisions or else they are located along highway 9Holly Reckard, "A Case Study of Copper Center, Alaska" (Alaska OCS Socioeconomic Studies Program), Technical Report No. 7. loalonso and Rust, The Evolving Pattern of Village Alaska, op, cit., P 4. I-3

11 or railroad transportation corridors. By contrast, many of the places with a majority of their population Native are isolated and can only be reached by air or, in the summer, by water. Nonetheless, in talking about rural population settlement patterns in Alaska, non-native places represent a significant percentage of the total and cannot be excluded from any comprehensive analysis. A second characteristic of most past research on rural population settlement is that reliance was placed primarily upon case study methodology. As a result, these studies normally focused on particular ethnic groups and regions of the state. Some studied the Aleuts; some the Athabascans; others the Eskimos. With the exception of the Alonso and Rust study (which used secondary data), all of the studies were so specific to a part of the state or an ethnic group within the state's total population that the conclusions drawn from them lacked generality. Most studies have researched the migration behavior of small, specific populations. None was capable of providing generalized conclusions about statewide rural settlement patterns. A third, and final, observation is that all of the studies on rural population settlement and migration implied the importance of public policy, but none of them investigated public policy explicitly. In one or another of the studies, references are made to the importance of cash employment (or economic development in general), the availability of educational (primarily secondary education) institutions, the accessibility to health care, the importance of transportation, and the importance of media accessibility. In Alaska, all of these involve public policy. Yet, none of the studies explicitly investigated the causality between different types of expenditures and village population patterns. Are schools more important than hospitals, and how does either one compare with the importance of job availability? The answer to such policy questions has usually been assumed or dealt with in an anecdotal manner. Because of this, a substantial effort is made in this study to develop a statewide data base from which a description of rural population settlement patterns could be derived and compared to patterns of federal and state expenditures. 1 1 The urban places are defined as the census divisions of Anchorage, Fairbanks, Juneau, and Ketchikan. I-4

12 CHAPTER TWO PATTERNS OF POPULATION SETTLEMENT Urban Concentration Although Alaska is the largest state in the union, with 586,400 square miles (approximately one-fifth of the total area of the other 49 states combined), it has a small population which is highly concentrated in a few urban places, While over 200 nonmilitary places are identifiable from the 1980 Census of Population, only four of them (Anchorage, Fairbanks, Juneau, and Ketchikan) were urban places with 5,000 or more residents, Over 125 places had populations of 250 or less, and almost a quarter of all places in Alaska had fewer than 100 residents, The extreme concentration of Alaska's population is shown in Table I. l and illustrated in Figure I. 1. The dashed diagonal line in the figure represents the polar case where all places in Alaska have equal population sizes. The other polar case is represented by the dashed line at the top of the figure, extending outward from 100 percent on the vertical axis. It represents a situation where all of the state's population is concentrated in a single urban community with the rest of Alaska devoid of people, The closeness with which the solid line (which represents Alaska's actual cumulative distributions of population and places) approaches the single community polar case graphically illustrates the state's extreme population concentration. On a scale of zero to one (with zero being equal population in all places and one being all population concentrated in a single place), Alaska would have a score between 0,8 and 9. TABLE I. 1, SIZE DISTRIBUTION OF COMMUNITIES* Places 1980 Population Number Cumulative Percent Distribution Number (OOOs) Cumulative Percent Distribution 100 or Fewer Persons Persons Persons ,000 Persons 4 1,001-2,500 Persons 13 2,501-5,000 Persons 3 Over 5,000 Persons , , *Communities stratified according to their 1970 population levels, I-5

13 100% 90 % 80% 70% Figure 1.1 Cumulative Distributi ons of Populations and Places --~7 60% '" (.) a: -"' 50% 0 '" > ~ 40% / ::i / E :, 30% u / / 20% / / 10% / / 'if/. / 10% 20% 30 % 40% 50% 60% 70% / / / / / / / / / 80% 90% 100% / Cumu lativ e% of Popul ation Et hnic Conce n trat io n Alaska ' s population in 1980 was predominantly White (77 percent) and Native (16 percent).1 The largest concentrations of Alaska ' s Native population are in the Northwest and Southwest regions, where they average ove r 80 percent of the total regional population (Figu r e I.2).2 Togethe r, these two regions contain about 42 percent of all Natives living in Alaska: The greatest relative concentration of Whites in Alaska occurs in the Southcentral region, excluding the Anchorage urban area. The Matanuska - Susitna and Kenai - Cook Inlet areas, which comprise the nonu r ban parts of the Southcentral region, have populations which were over 90 percent White in 198 lalaska Natives consist of Eskimos, Aleuts, and American Indians. The remaining 7 percent of the population was Black (3.4 percent), Asian and Pacific Islanders ( 2.0 percent ), and others (1.6 percent). 2 The third largest number of Natives in Alaska (8,953) live in Anchorage, but th e y only represent about 5.1 percent of this urban area's total population. I - 6

14 Figure 2. '. Alaska Regions Northeast Southwest //// //// Urban Areas //// //// I - 7

15 When looked at in te r ms of the individual places for which we had data, almost 58 percent of all nonmilitary places in Alaska had populations which wer e 80 percent or more Native. An additional 24 percent had populations that were 80 percent or more White~ while only about 18 percent of the places had populations where no ethnic group accounted for mor e than 80 percent of the total. This is illust r ated in Figure I ~ 3 ~ While a smaller number of places were either 95 percent Native or White in 1980 than was true in 1970, mos t pl a ces in Alaska continue to be either predominantly Native or White communities. The Native communities tend to be smaller and clustered in the Northwest and Southwest r egions, while t he White connnunities tend to be larger and clustered in the Southcentral and Southeast regions~ Figure Distribution of Places by Percent Native , "' 70 ~ a: ~.0 E :J z % 20% 30% 40% 50% 60% 70% 80% 90% 100% Percent Native I- 8

16 Population Changes Fifty of the 210 nonurban places (24 percent) for which we have data experienced population declines between 1970 and 198 All but seven of these places had populations of 250 or less in 197 The total number of these places declined from 147 in 1970 to 129 in 1980; while the average size of places with 100 or less residents actually declined from 72 to 67 persons between 1970 and 1980 (Table 1.2). Overall, the percentage of total places in Alaska with populations of 250 or less declined between 1970 and 1980 from 70 percent to 61 percent. TABLE I. 2. POPULATION CHANGE: (NONURBAN COMMUNITIES)* Number of Average Average Average Average% Place Size Places Poeulation % Native % Female Yrs. By No. of Persons ,321 1, ,559 3, *Excludes Sitka The percent of the Native population declined, on average, for most places in Alaska, with the exception of the smallest ones (those with populatins of 100 or less). In general, the larger the percentage growth in a place I s population, the greater the dee line in its percent of population Native. At the same time, the larger the percentage growth in a place's population, the larger the growth in its number of Native inhabitants (Figure I.4). Taken together, these patterns appear to describe a trend for Natives to move from smaller, ethnically homogenous communities into larger communities which are somewhat more ethnically diverse. I-9

17 Figure 1.4 Changes in Populat ion and Percent Native (by place size category) 50 AO - 30, , ~----' Percentage Change in Proportion of Population,Native Percentage Change in Population size.. I-10

18 There was little change 1.n the percentage of females between 1970 and 1980 in the places for which we had data, although there appeared to be a very slight tendency for the percentage of females to dee line among the places which experienced overall population declines. On average, the percentage of persons 15 through 35 years of age increased dramatically among all the places in Alaska, going from 16,5 percent in 1970 to 38,9 percent in 198 This appears to reflect the aging of the village baby boom which occurred in the late 1960s and early 1970s. 3 In 1970, there was no apparent pattern to the percentage of the population 15 through 35 years of age, However, in 1980 this percentage varied systematically with place size, with larger places having a higher percent of their populations in the age grouping than smaller places. The correlation coefficient between the percent change in the 15 through 35 population and the percent change 1.n population among the 210 places between 1970 and 1980 was.67. Among places with population growth between 1970 and 1980, the proportion of the population 15 through 35 years grew almost 20 percent faster than was true among places with population declines. Taken together, these trends imply that the age cohort is shifting out of Alaska's smaller communities and into larger one s. Employment Changes The nonurban places in Alaska experienced a combined growth of over 17,500 persons employed between 1970 and 1980, with the average place having an increase of about 84 employed persons over its 1970 level of 26 The greatest increase in employed persons occurred in the largest commun1.t1.es, and the correlation coefficient between changes in persons employed and 1970 population was.85. This, of course, is reasonable. The larger communities had more people and more employed people in 1970, and their absolute growth of employed people between 1970 and 1980 would be expected to be the largest. This interpretation of the data is supported by the fact that when communities' rates of empoyment growth or decline between 1970 and 1980 are compared with their 1970 populations, the relationship disappears, having a correlation coefficient of only.09. The percentage change in employment communities also was unrelated to their (The correlation coefficient was.07.) among the state's nonurban rates of population growth. This surprising lack of a 3 see Arthur E. Hippler, "The Demographic Youth Bulge: One Reason for Acculturation Difficulties Among Alaska Natives, 11 Polar Record, Vol, 10, No, 114, p I-ll

19 relationship between the rate of growth (or dee line) in employment and population is probably caused by several factors. First, over 60 percent of the state's nonurban communities in 1980 had populations of 250 persons or less, and these smaller places have a large component of subsistence activity in their economies.4 Consequently, the relationship between changes in cash employment and changes in population would be expected to be weak. Second, the closure (or opening) of a small fish processing plant, lumber mill, or airport fixed-base operation will cause large percentage changes in a small community's total employment. Yet these changes will have little effect on a small community's population since the elasticity of labor supply for cash employment appears to be quite high in smaller communities, particularly if the employment is seasonal or part time.5 Finally, it appears that state spending on capital projects acts in two offsetting ways. For a larger community with an expanding economic base and population growth induced (in part, at least) by economic opportunity, the state provides assistance in responding to population pressure by funding a variety of social overhead capital projects. For small communities whose economies are not expanding and which remain concentrated in subsistence activities, the state appears to try to provide a minimum level of amenities (such as housing, education, and infrastructure) by funding a variety of capital improvement projects. These projects provide some level of employment for small communities even though their economies (and populations) are either constant or declining. While this two-fold interpretation of state capital spending impacts requires further research to establish with certainty, it is supported by the fact that (a) the correlation between the percentage in employment and 1981 per capita capital expenditures was positive and 7 5 among Alaska's nonurban communities; (b) no other economic or demographic variable investigated was correlated to the percentage change in employment with a coefficient greater than.17; and (c) capital expenditures per capita varied directly with the remoteness of small communities in the northern and western regions of Alaska and inversely with their location in nonurbanized parts of the Southcentral region. (Although these correlations were quite weak, about.21 and -.17, respectively, they were the highest obtained among the ten demographic and economic variables investigated in addition to per capita capital expenditures.) 4see Lee Huskey, "The Determinants of Native Labor Force 11 Participation in Rural Alaska, paper delivered to the 22nd Annual Meetings, Western Regional Science Association, 1983 (forthcoming), pages 1 and 16. 5Huskey, op. cit., page 9. I-12

20 Net Migration A place's net migration between 1970 and 1980 equals the difference between its actual 1980 population and the population it would have had if (a) no one (including the new born) had moved away and (b) no one had moved in between 1970 and 198 Formally, this is defined as Net Migration= 1980 Population - (1970 Population+ Births - Deaths) It follows from this definition that a community can be growing even though it is experiencing net out-migration. For example, a community with a 1970 population of 100 persons that had 25 births and 5 deaths between 1970 and 1980 would have had a 1980 natural population of 120 persons. If its reported 1980 census population was 115, it would have grown by 15 persons (15 percent) over the decade but still have had a net out-migration of 5 persons (or -5 percent of the 1970 population). The net migration of -5 persons in our example does not mean that only five persons moved away from our hypothetical place. Thirty persons may have moved away while 25 other persons moved 1n. Net migration represents the difference between the much larger gross in-migration and gross out-migration flows. 6 Table I. 3 shows the number and percentage of Alaska places that had population declines and net out-migration between 1970 and 198 While less than a quarter of the places (24 percent) had population declines between 1970 and 1980, over half (52 percent) had net out-migration. Particularly striking is that only two of the 24 places in Alaska with populations of 500 or more persons showed declines between 1970 and 1980, while 12 (or half) of them had net out-migration. Population losses were highly concentrated in small places (particularly those of 250 or fewer persons), but net out-migration was experienced widely among places of all sizes. 6 Although not included 1n this study, many researchers have commented on the importance of these gross migration flows. See William Alonso and Edgar Rust, The Evolving Pattern of Village Alaska (Federal-State Land Use Planning Commission, 1976); pg. 27ff. Particularly in smaller, predominantly Native places, residents may leave annually to fish or obtain seasonal jobs in construction or fish processing plants. Others leave to be close to their children who attend boarding schools. Some families may move to the city for several years and then return again. Little data exist on these gross migration flows, and much of the information about them is anecdotal. While this study does not address these seasonal and annual gross migration patterns, they should not be forgotten for they may be an important component in understanding the pattern of life in smaller places (or villages) in Alaska. 1-13

21 TABLE I.3. PLACES WITH POPULATION DECLINES AND NET OUT-MIGRATION: Place Size by No. of Persons Over 5000 Total Places with Places with Net Population Declines Out-Migration Total No. of Places Number Percent Number Percent o.o o.o Among the 211 nonurban places in Alaska reported in Table I. 3, 107 had net in-migration and 104 had net out-migration between 1970 and 1980 (Table I.4). The average place grew at a rate of almost 3. 5 percent annually during the decade. In-migration places, however, grew at an annual rate of about 6.5 percent while the rate for out-migration places was less than 1.0 percent. The nonurban in-migration place's percentage of Native residents became slightly smaller even though their absolute numbers of Natives increased during the decade. There was little change in their proportion of females, but their proportion of persons between 15 and 35 years of age more than doubled. Turning to the net out-migration places, they averaged a larger population in 1970 than did the in-migration places; but they grew much more slowly, losing about 70 percent of the population growth they would have had if no net out-migration had occurred. They had fewer births per hundred persons in the 1970 populations than did the in-migration places, but that is to be expected since most migrants are in prime child-bearing ages. On the average, the out-migration places had a higher percentage of their populations Native than did the in-migration places; and in contrast to the in-migration places, this percentage went up between 1970 and 198 I-14

22 TABLE 1,4, CHARACTERISTICS OF IN- AND OUT-MIGRATION PLACES (Nonurban Communities) All Places Net In Migration Net Out Migration Number of Places Avg, 1970 Population per place Avg, 1980 Population per place Percent Change Avg. Net Migration per Place Avg. Net Migration as Percent of 1970 Population , 9 Births as% of 1970 Population Deaths as% of 1970 Population ,0 12, Percent Native 1970 Percent Native , Percent Female 1970 Percent Female , Percent Years 1970 Percent Years , , A stratification of net in- and out-migration places by their 1980 place size is presented in Table I. 5, While the percent of total places with net in- and out-migration was virtually the same, they were distributed differently with respect to place size, Almost one-third (31. 7 percent) of the total out-migration places had 100 or less population in 1980 while only about 18,7 percent of the net in-migration did. Among larger places, the opposite pattern held, with almost a quarter (23,4 percent) of the in-migration places having populations of over 500 in 1980, compared to only percent of the net out-migration places, When the relative size distribution of nonurban places is compared for 1970 and 1980 (Table 1,6), it is evident that almost all of the change occurred among the smallest and largest communities, Places with 100 and less people fell from 32 percent to 25 percent of the total between 1970 and 1980 while those with populations of over 1,000 grew from 1 percent to 9 percent, Even among communities experiencing net I-15

23 out-migration, the larger places continued to show absolute growth of population. Overall, the trend appears to be one of migration from smaller to medium-sized communities and from medium-sized to larger ones, with the smallest communities experiencing both absolute and relative population declines. TABLE I.5. NET MIGRATION BY 1980 PLACE SIZE (Nonurban Places: ) 1980 Place Size by No. of Persons Net In-Migration Net Out~Migration Total Places No. Places Percent No~ Places Percent Total TABLE I.6. RELATIVE DISTRIBUTION OF NONURBAN PLACES: 1970 AND 1980 BY NET MIGRATION In-Migration Out-Migration Total Places Number of Places Places Persons '¼ Total

24 Finally, among places in the state with less than half their populations Native, 26 percent experienced net out-migration when classified by their 1970 Native distributions; when classified by their 1980 dis- tributions, 29 percent did. As many Native majority places gained from migration as lost. At the same time, one out of every two Native majority places experienced net out-migration, while only about one out of four non-native majority places did. Net Migration by Geographic Location The 211 nonurban places for which we had data were organized into census divisions (CDs) and are reported in Table I. 7. By way of reference, the four urban places all had absolute population growth between 1970 and 1980; one of them (Ketchikan) grew at a slow rate and experienced net out-migration during the decade. Of the four urban places, Ketchikan and Juneau had declines in their percent of Natives in the population, while Ketchikan was the only urban place which did not have an absolute increase in its Native population. Overall, the number of Natives residing in the state's four urban areas grew by almost 9,000 persons between 1970 and 198 This increased the percent of Alaska's Native population living in the urban places from about 23 percent in 1970 to about 32 percent in 198 SOUTH CENTRAL All but two of the places in the Matanuska-Susitna and Seward Census Divisions (CDs) had net in-migration between 1970 and 198 Most of these places were small at the beginning of the decade. Five of the seven places in the Matanuska-Susitna CD and two of the three places in the Seward CD had less than 100 residents in 197 By 1980, these seven communities averaged 147 residents, and two had increased in size to over 350 persons. The most rapidly growing places were in the Matanuska-Susitna CD, where the average place increased in size by more than two-and-a-half times between 1970 and 198 Both Matanuska-Susitna and Seward's places averaged about 95 percent non-natives, and in none of their places did the Native population exceed ten percent. The Kenai-Cook Inlet CD presents a somewhat different picture. Five of the nine net in-migration places were like the majority of ones found in the Matanuska-Susitna and Seward CDs--small, over 90 percent non-native, and growing rapidly. Two (English Bay and Port Graham) were over 80 percent Native and growing moderately. I-17

25 TABLE I. 7. NET MIGRATION BY CENSUS DIVISION Net Migration Population % Native Number Avg. % of 1970 Average Average of Places Number Population SOUTHCENTRAL Mat-Su In-migration Out-migration Total Kenai-Cook Inlet In-migration Out-migration , Total Seward In-migration Out-migration Total PRINCE WILLIAM SOUND Cordova-McCarthy In-migration ,164 1,879 13, Out-migration 0 Total ,164 1, Valdez-Chitina-Whittier In-migration , Out-migration Total SOUTHEAST Skagway-Yakutat In-migration , ,9 Out-migration Total Sitka In-migration ,057 2, Outmigration Total ,057 2, I-18

26 TABLE I.7 (Continued) Net Migration Population % Native Number Avg. % of 1970 Average Average of Places Number Population Haines In-migration Out-migration Total Angoon In-migration Out-migration Total Wrangell-Petersburg In-migration ,245 1, Out-migration ,029 2, Total ,506 1, Prince of Wales In-migration Out-migration Total Outer Ketchikan In-migration o.o 1.3 Out-migration Total NORTHEAST Upper Yukon In-migration Out-migration Total Southeast Fairbanks In-migration Out-migration Total Yukon-Koyukuk In-migration Out-migration Total I-19

27 TABLE I,7 (Continued) Net Migration Population % Native Number Avg. % of 1970 Average Average of Places Number Population NORTHWEST Barrow In-migration Out-migration ,234 1, Total , Kobuk In-migration Out-migration Total Nome In-migration Out-migration Total SOUTHWEST Wade-Hampton In-migration Out-migration Total Bethel In-migration Out-migration , Total Kuskokwim In-migration Out-migration Total I-20

28 TABLE I.7 (Continued) Net Migration Population % Native Number Avg. % of 1970 Average Average of Places Number Population SOUTH Bristol Bay In-migration Out-migration o 76.0 Total , Bristol Bay Borough In-migration 0 Out-migration , Total Kodiak In-migration Out-migration Total Aleutian Islands In-migration Out-migration Total I-21

29 The final two in-migration places (Homer and Soldotna) both had over 1,000 residents in 1970, and fewer than 5 percent of either of their populations were Native. Both places approximately doubled in size between 1970 and 1980, with net in-migration accounting for more than half (54 percent) of Homer's total population growth and under a quarter (23 percent) of Soldotna's. PRINCE WILLIAM SOUND The Cordova-McCarthy CD had only one place about which we had complete data, the city of Cordova. It grew moderately between 1970 and 1980 (about 6 percent annually). Its small Native population (about 14 percent in 1970) more than doubled in absolute size over the decade and represented about 18 percent of the population in 198 The Valdez-Chitna-Whittier CD had five places with net inmigration between 1970 and 1980; however, over 85 percent of their combined population growth and 90 percent of their combined net in-migration occurred in the city of Valdez, home of the TransAlaska Pipeline's southern terminal and oil port. Over 80 percent of Valdez's growth came from net in-migration, and its percent of the Native population declined from 15 percent in 1970 to 8 percent in 198 Five places in the Valdez-Chitna-Whittier CD had net outmigration between 1970 and 198 Two of the places had populations in which non-natives were a majority (Gakona and Edgerton), and these places had an average population gain of about O. 2 percent annually during the decade. Two of the out-migration places (Mentasta and Tatitlek) were over 95 percent Native and had an average population loss of about -3 percent annually. On average, the five net out-migration places had absolute population losses. SOUTHEAST Four of the six places in the Skagway-Yakutat CD had fewer than 250 residents in 197 Among these four small places, one (El fen Cove) had net out-migration and an absolute population loss during the decade, while the remaining three (Gustavus, Pelican, and Yakutat) were both growing absolutely and having net in-migration. One of the four smaller communities (Yakutat) had a predominantly (63 percent) Native population while the other three were predominantly non-native. I-22

30 The two larger places were Hoonah and Skagway. Hoonah's population declined absolutely between 1970 and 1980, and it had substantial net out-migration. Its Native residents increased by about 50 persons over the decade; and by 1980, Hoonah's population was 84 percent Native--up from 71 percent in 197 The other larger place was Skagway. It had net out-migration during the 1970s even though its population increased slightly (about 1 percent annually). Skagway's population was about 95 percent non-native. The Sitka CD is dominated by the city of Sitka, which grew from about 6,100 persons in 1970 to about 7,800 persons in 198 The percent of Sitka's population which was Native remained constant at about 23 percent between 1970 and 1980, implying that its actual Native population increased by about 1,000 persons (from a 1970 base of about 760) over the decade. Two of the places in the Haines CD grew absolutely and had net in-migration between 1970 and 198 They were the small community of Klukwan, which was over 85 percent Native and grew from about 100 to 135 persons (with a net in- migration of about 20), and the city of Haines, which was about 80 percent non-native and grew from about 460 to about 990 persons (net in-migration, 435). The Angoon CD contains one larger-size place, Angoon, which had almost no net migration over the decade, and one smaller-size place, Tenakee Springs, which grew from about 85 to 140 persons. Tanakee Springs, which is over 90 percent non- Native, was one of the rare places which reported more deaths than births during the decade. As a result, its net in-migration was almost 85 percent of its 1970 population. The city of Angoon, over 90 percent Native, grew slowly during the decade, with the Native population growing in absolute terms, but at a slower rate than the total. The Wrangell-Petersburg CD had one out-migration place--the city of Wrangell. It grew by less than 1 percent annually between 1970 and 1980 and had a net out-migration of about 220 persons. It was about 80 percent non-native. Two places in the CD had net out-migration. One, Kake, was predominantly Native (over 85 percent). Its population grew slowly (about 2 percent annually), but its number of deaths per 100 in the population was about twice as high as the average of all places in the state. Its natural growth (i.e., the excess of births or deaths) was conse- quently low and resulted in a small amount of net in-migration. The other place with net in-migration, Petersburg, was about 88 percent non-native in both 1970 and 198 Its population increased moderately (about 4 percent annually), and over half (57 percent) of its absolute growth was net in-migration. The Prince of Wales' CD had three in-migration and three outmigration places. Most of the in-migration occurred in one place, Craig, which in 1980 had about a one-third (33 percent) Native I-23

31 population. The two smaller in-migration places were over 75 percent Native in 1980, and they grew at slightly slower rates. The three places in the Prince of Wales CD which had net out-migration also had absolute declines in their populations. Two had births per 100 residents in 1970 of only about half the state average for all places. All three had less than 25 percent of their populations Native in 1970, and all three had this percentage rise over the dacade. The Outer Ketchikan CD had one small, predominantly non-native place (Hyder) which had few deaths (3), fewer births (2), and net in-migration. The other two places in the CD had net outmigration. One (Metlakatla) was over 85 percent Native, and the other (Annette) was about 90 percent non-native in 1970 but became more Native over the decade. While Annette's absolute population declined over the decade, its number of Native residents almost tripled (from 19 to 54 persons), and their percent of the population rose from about 10 to 40 percent. NORTHEAST The Upper Yukon and Southeast Fairbanks CDs each had twice as many places with in-migration as they did with out-migration. Net in-migration in the Southeast Fairbanks CD occurred at four small places (less than 250 residents in 1970). One of these (Tok) had an absolute population increase of close to three- fold between 1970 and 198 It had more net in-migration during the decade than its total population level in 197 Southeast Fairbanks' out-migration places, on average, also had absolute increases in population. Most (85 percent) of the net out-migration in the Southeast Fairbanks CD occurred in one place (Delta Junction), where the number of births per hundred residents in 1970 was twice the average of all places in the state. This resulted in net out-migration even while Delta Junction's absolute population increased by about 3 percent annually between 1970 and 198 Its population was over 95 percent non-native. The Upper Yukon CD had five smaller places and one larger place with net in-migration between 1970 and 198 The five smaller places accounted for about half (48 percent) of the total population growth and about two-thirds of the total in-migration. Five of the six places had Natives as a majority of their populations. The three net out-migration places in the Upper Yukon CD all had 1970 populations of under 250 persons; all were 90 percent or more Native, and two of the three had absolute population declines during the decade. I-24