Explaining the association between rates of in-migration and out-migration

A positive association between rates of in-migration and out-migration across locations has long been observed. This apparent paradox has usually been attributed to the impact of in-migration on population composition. We argue here that the intrinsic structure of population dynamics may also induce this effect, and we develop a formal model of this process. Empirical analyses using migration data for states in the U.S. over 1955–1960, 1965–1970, and 1975–1980 show that the intrinsic mechanism may swamp the compositional effect.     

Volatility in the geographical structure of short-run US interstate migration

"Migration is often thought to be a long-run phenomenon. However, it is shown here by means of labor-force migration estimates derived from the Continuous Work History Sample over the period 1958-1975, that US interstate migration is quite volatile over the short run with respect both to its adjustment to macroeconomic fluctuations and to its geographic structure. A hierarchical clustering routine is utilized to analyse yearly estimates of interstate gross in-migration and out-migration so as to identify the temporal stability and characteristics of the geographical patterns of migration. States enter the national migration linkage tree at different levels for different years, depending upon the level of short-run economic activity." It is also shown that "when national economic conditions worsen, labor-supply adjustments are more localized and the degree of interstate interdependence is less than in times of economic boom."     

Migration in the urban-rural hierarchy of China: insights from the microdata of the 1987 National Survey

Microdata from the 1987 National Population Survey are used to analyze internal migration in China in the 1980s. The focus is on the impact of migration policies on rural-urban migration. "There are two main findings. First, although the migration policy resulted in a very low migration level and systematic distortions in migration schedules, its encouragement of downward migrations was very ineffective, whereas its control on rural-to-urban migrations was partially weakened by the strong upward aspirations of rural families awakened by recent economic reform. Consequently, net in-migration contributed substantially to the growth both of city and of town populations. Second, although the level of education had a strong positive effect on the migration propensities both of males and of females in general, it had a strong negative effect on the migration propensities of females at the time of marriage, a finding which suggests that the families at subsistence income level tended to marry their daughters to grooms in other communities in order to reduce the risk of familial income shortfalls."     

An economic analysis of migration in Mexico

This paper analyzes internal migration in Mexico over the 1960–1970 period. A model of the determinants of migration is specified and estimated for aggregated interstate migration flows. The results show that distance serves as a significant deterrent to migration, that higher destination earlings levels are attractive to migrants, and that regions with high unemployment rates experience lower rates of in-migration. An unanticipated finding is that regions with higher earnings levels have greater rates of out-migration.The data are disaggregated to examine separate migration relationships for each state. The results are that distance is a lesser deterrent for those migrants with more accessible alternatives, that higher earnings levels reduce the deterring effects of distance, and that regions with higher earnings levels have lower associated elasticities of migration.It is concluded that economic factors have played a crucial role in internal migration and thus in the changing occupational and geographic structure of the Mexican labor force.This research was supported by Grant Number 1-RO-1-HD08567-01 from the Population and Reproduction Grants Branch, Center for Population Research, National Institute of Child Health and Human Development, United States Department of Health, Education, and Welfare. Part of this study was completed at the Food Research Institute, Stanford University, where J. R. Ladman was a Visiting Scholar. We are grateful to Barry Edmonston and Dudley Kirk for their helpful comments on an earlier draft of the paper. Responsibility for remaining shortcomings remains ours.     

Migration and the quasi-labor market in Russia

"This paper explores the twin concepts of labor demand and labor mobility during the Soviet and post-Soviet periods. The study uses a detailed data set on labor stock, industrial labor demand, and labor flows for the 1980s in the Yaroslavl' Oblast, and data on migration and regional labor markets for all Russian regions in the 1990s. Contextual features, such as the social contract, full employment, methods of labor allocation, and a generally low rate of geographic mobility, distinguish the centrally planned quasi-labor market from the labor market in capitalist democracies. The findings suggest that net in-migration induces employment change in the current period rather than in a future period. The job creation effects appear concurrent with migration during the Soviet period. In the post-Soviet period, migration and employment relationships are not predictable based on the same relationships during the Soviet period."     

Age articulation of U.S. inter-metropolitan migration flows

A fundamental divide in migration research has existed between aggregate studies of movement among geographic regions and micro studies of individual migrant behavior. Micro-scale studies have highlighted the importance of stage of life-cycle in predicting movement propensities, whereas many aggregate studies have focused on age-aggregated data summed over all origins for in-migration and over all destinations for out-migration. In this paper we show that if data for functional metropolitan-centered regions are employed, and if origin-destination specific streams of movement are analyzed, the age-specific patterns of inter-metropolitan migration within the United States cluster into distinctive patterns of flow representative of key stages of the life-course. In order to expose and portray a rich, age-articulated geography of U.S. migration we aggregate county-to-county migration flow data from the 1990 census for extended metropolitan regions: Bureau of Economic Analysis (BEA) Economic Areas. We calculate destination-specific out-migration rates broken down into 17 age groups for each origin-destination-specific migration stream between pairs of Economic Areas and present the results of a factor analysis of these flow-specific age profiles. We use the factor scores to cluster the very large number of origin-destination-specific age profiles and find that seven characteristic types emerge reflecting key mobility stages of the life course. We analyze the distinctive characteristics of the migration flows in each cluster and based on the prevalence of flows of each type within streams of gross in- and gross out-migration we present a typology of the 172 BEA Economic Areas. Our conclusion is that better understanding the age articulation of origin-destination-specific flow patterns would help advance regional science migration research. Received: January 2001/Accepted: June 2002 Earlier versions of this paper were presented at the 47^th North American Meetings of the Regional Science Association International, Chicago, Illinois, USA, November 9–12, 2000 and at the 12^th International Symposium Hosted by the Executive Secretariat of the Pacific Regional Science Conference Organization, Tokyo, Japan, October 1–3, 2000. The authors thank Joseph Persky, Roger Bolton, Hiroyuki Shibusawa, and Roger Stough for helpful suggestions made at those two presentations. The authors gratefully acknowledge the expert computer programming assistance of Lucy M. Carruthers of the Center for Computing and Information Technology of the University of Arizona. We also thank Chris Henrie, Ph.D. student, University of Arizona, for constructing our base map of BEA Economic Areas. Frank Heins would like to thank the National Research Council of Italy (Short-term Mobility Program 1999) for financially supporting a stay at the Department of Geography and Regional Development, University of Arizona. During his stay the groundwork for this research was laid. David Plane would like to acknowledge the support of the Population Division, U.S. Census Bureau for his 2001–2002 academic year visiting research position during which the final analyses and revisions were made.     

Population migration and economic restructuring in the United States

"This article demonstrates that sectoral employment shifts associated with the migration pattern changes of the 1970s are very different than those for the period 1955-60 to 1965-70. Changing competitiveness for jobs in manufacturing and other traditional basic sectors of the economy cannot account for the greatly accelerated levels of core-periphery net outflow that have been the dominant characteristic of interstate movement during the 1970s and 1980s. Instead, an interconnected set of activities that includes government, services, trade, and construction is associated with the broadscale shifts in the geographic pattern of the United States' population. The causal linkage from migration to employment change assumed heightened importance during the 1970s."     

Short-term forecasting of internal migration

A new methodological approach to the forecasting of short-term trends in internal migration in the United States is introduced. "Panel-data (or longitudinal-data) models are used to represent the relationship between destination-specific out-migration and several explanatory variables. The introduction of this methodology into the migration literature is possible because of some new and improved databases developed by the U.S. Bureau of the Census.... Data from the Bureau of Economic Analysis are used to investigate the incorporation of exogenous factors as variables in the model." The exogenous factors considered include employment and unemployment, income, population size of state, and distance between states. The author concludes that "when one...includes additional parameters that are estimable in longitudinal-data models, it turns out that there is little additional information in the exogenous factors that is useful for forecasting."     

Models for the analysis of interregional migration

"Asymmetric square tables, such as those arising from interregional migration, can be analysed by separating the skew-symmetric and symmetric components. A least-squares analysis of the skew-symmetric part can indicate the degree of complexity of model that is consistent with data and this can be combined with some suitable model for the symmetric part. The joint model may then be fitted by maximum likelihood based on suitable distributional assumptions. This approach is used for an analysis of Australian interstate migration for l960-l966 and indicates a model with independent in-migration and out-migration rates proportional to a symmetric function of population sizes and interstate distance."     

On the contribution of out-migration to changes in net migration: A time series confirmation of Beale's cross-sectional results

Recent cross-sectional studies of population migration patterns have consistently failed to find any relationship between the economic conditions prevalent in an area and the propensity of its inhabitants to emigrate. Changes in the rate of out-migration apparently occur only slowly as the age composition and mobility history of the population change. Thus, in the short run, changes in out-migration are small and contribute little to changes in net migration. This article tests this model of the migration process by making a time series analysis of gross and net migration between Japanese prefectures. We find that, among rural areas, changes in out-migration are a consistently significant component of changes in net migration. If changes in net migration reflect changes in economic conditions in these areas, then one can only conclude that the latter do have an effect on out-migration rates, an effect which is missed by cross-sectional studies. Beale's graphical analysis, which is also cross-sectional, is a notable exception and would have predicted our results.This study was supported in part by a grant from the National Science Foundation.     

Analysis of urban-rural population dynamics of China: a multiregional life table approach

Multiregional life tables at various ages developed by Rogers and Willekins are used to examine the urban and rural dynamics of population (fertility and out-migration expectations) in China. Period-cohort survival rates are used in the calculation of survival probability. A unit radix is assumed in the calculation of the life table for each age. The life expectation of the female population a birth is 73.59 years in urban areas and 72.46 in rural areas. Male life expectancy is 69.73 years in urban areas and 58.99 in rural areas. Life expectancies are higher for rural males aged over 20 years and females aged over 60 years because of urban and rural differences in out-migration and mortality rates. Life expectancy for the rural males over age 60 is 16.55 and for urban males over age 60 is 16.00. Rural females over age 70 have a life expectancy of 12.25 years and urban females over age 70 have a life expectancy of 12.04 years. In the calculation of fertility expectations, the occurrence-exposure fertility rates of period cohorts are estimated from the forward fertility rates of period cohorts by means of forward mortality rates. The occurrence-exposure fertility rate of life table cohort y is calculated from the occurrence-exposure fertility rates of period cohorts r (y) and r(y + 1). A female born in an urban area is expected to have 1.82 children in an urban area and .06 children in a rural area. A female born in a rural area is expected to have 2.10 children in a rural area and .35 children in an urban area. Rural population has a much higher net reproduction rate than urban population. In the calculation of the net migraproduction (spatial out-migration expectation), forward out-migration rates of period cohorts are used for estimation of occurrence-exposure out-migration rates of period cohorts. Mortality rates are included in the calculation of population at risk. Out-migration rates are modified to account for nonsurvival migrants using mortality rates of the destination region. The occurrence-exposure out-migration rate of the life table cohort y is estimated similarly as the fertility rates of life table cohorts. A female born in an urban area is likely to make .0451 migrations out of the region; a female born in a rural region is likely to make .3491 out-migrations. A male born in an urban area is likely to make .0475 out-migrations; a male born in a rural area is likely to make .2903 out-migrations.     

新时期的“小城镇、大战略”——试论人口高输出地区的小城镇发展机制

在全面认识我国城镇化发展趋势的基础上指出,要使超过2亿农民工平稳完成城镇化转型的进程,除了要关注人口导入地的城市群资源集聚效应以外,还必须重视人口输出地的农村经济中心——即小城镇在城镇化进程中的作用文献分析和案例研究表明,小城镇发展的动力机制不仅仅在于工业化,更重要的是农村人口对城镇服务的内在需求,包括住房、教育和人居环境等多年来小城镇发展滞后于城市,其背后有着一系列的社会、经济和制度等原因为此提出,新时期小城镇发展仍然是城镇化进程中的一项"大战略",但是发展思路需转变总之,人口高输出地区的小城镇是我国未来健康城镇化发展的重要组成部分,关系着我国健康城镇化是否能够顺利推进和实现也可以说,新时期的"小城镇、大战略"要有新的内涵。     

Interstate migration of college freshmen

We examine the economic determinants of interstate migration of college-bound freshmen, using state-level data. Our analysis provides a robust explanation of the striking differences among the U.S. states in out-migration of college-bound freshmen. States that provide more educational choices and higher quality education services, charge lower tuition, have broad-based merit scholarship programs and have lower income levels tend to retain a higher percentage of their college-bound freshmen at home. We thank Andrew Mason, Sang-Hyop Lee and Rhonda Sharpe and two anonymous referees for helpful comments and suggestions.     

Immigration and the City: Some Demographic Facts and Issues

Unquestionably, migration has become a predominant factor in the growth of all mainland capital cities of Australia, partly because of the relatively low fertility extant, partly because of the level of immigration to Australia and, also, because of high levels of interstate internal migration. In fact, the migration components oif growth have varied for decades between Australian cities – with strong overseas migration and internal migration together resulting in annual rates of growth of over 3 per cent per annum in Perth, and strong internal migration and lesser but significant immigration sustaining almost comparable growth rates in metropolitan Brisbane. Much of the internal migration net gain to these two cities comprises migration from Sydney and Melbourne, particularly the former.     

Population change and migration in Britain since 1981: evidence for continuing deconcentration

Recent trends in spatial distribution and internal migration in Britain are reviewed using small-area statistics from the 1981 and 1991 censuses. "The results indicate that the differentials in the population growth rate between metropolitan and nonmetropolitan Britain narrowed somewhat between the 1970s and the 1980s, but the negative relationship between urban status and population change remained very clear. Moreover, contrary to the experience of the U.S.A. and a number of European countries, in the mid-1980s Britain saw a resurgence of nonmetropolitan growth which had widespread impact across the country."     

Recent trends in migration between core and peripheral regions in developed and advanced developing countries

"The main finding of this article is that net internal migration to the core regions in the countries of the developed world, which subsided in the 1970s, increased in the 1980s, although not to the level of the 1960s. In some countries of northwest Europe there is a balance now in net flows between core and periphery. In the countries of the periphery of Europe and Japan net internal migration to the core regions increased slightly in the 1980s. Net migration flows to the periphery have completely reversed in Canada, and net flows out of the core regions of the United States have been significantly reduced. In eastern Europe, however, there is still moderate net migration to the core regions without any interruption as seen in western Europe, North America, and Japan. In South Korea and Taiwan rates of net migration to the core regions have been reduced from their high levels of the 1970s, but they are still quite high and show no clear sign of a break from the past."     

Key Relationships Between Urban Development and Neighborhood Change

Changing neighborhood conditions cannot be understood apart from the overall urban development process. U.S. urban development has been a form of technological progress beneficial to most households, but harmful to those left in concentrated poverty areas, and to many forced to move by neighborhood change. Heavy past in-migration to metropolitan areas was accommodated by a process of households upgrading through movement to “better” neighborhoods—thus causing neighborhood change. Where such in-migration has slowed, upgrading in place may now be socially preferable. Cities faced with net declines in housing demand can accommodate themselves through several different strategies best determined at the local level.     

Are small communities at risk of population loss?

Small communities almost universally worry about out-migration and the negative effects of out-migration on community viability. Using Oregon community-level data and applying the threshold estimation method of Hansen (Econometrica 68(3):575–603, 2000), we are able to identify population thresholds that distinguish small communities from their larger counterparts based on significant structural differences in factors affecting net migration. Our results suggest that smaller communities are more at risk of population decline than larger ones. After controlling for spatial spillovers from neighboring communities, the average net migration rate is 3 % in the larger communities (roughly above 5,000 population), 2 % in the mid-sized communities (roughly between 1,250 and 5,000) and $-3$  % in the smallest communities (roughly less than 1,250). Other things equal, geographic isolation from large cities and low wage rates provide some protection from net out-migration for the smallest communities, but even for the smallest places, a larger population base lowers the risk of net out-migration.     

The effects of in- and out-migration on urban growth in Turkey (1965–85) and a comparison with the developed countries

Lowry's (1966) study about in- and out-migration rates started a series of discussions on pull and push effects, net migration, and urban growth. This paper presents the results of a similar study, done in Turkey, that tests whether Lowry's hypothesis is valid for developing countries. The results of this study indicate that, for the Turkish case, the effects of migration on urban growth had been contrary to those found in the developed countries, but that, as Turkish development increased over the time period 1965–85, the effects of migration increasingly came to resemble those of developed countries.     

Multiregional estimation of gross internal migration flows

"A multiregional model of gross internal migration flows is presented in this article. The interdependence of economic factors across all regions is recognized by imposing a non-stochastic adding-up constraint that requires total inmigration to equal total outmigration in each time period. An iterated system estimation technique is used to obtain asymptotically consistent and efficient parameter estimates. The model is estimated for gross migration flows among the Canadian provinces over the period 1962-86 and then is used to examine the likelihood of a wash-out effect in net migration models. The results indicate that previous approaches that use net migration equations may not always be empirically justified."