مهاجرت، امنیت اجتماعی، و رشد اقتصادی

This paper studies the effect of population aging and international migration on economic performance. Fertility is endogenized so that immigrants and natives can have different fertility rates, which provides a more realistic view of policy effects. Fertility is an important determinant to the tax burden of social security since it affects the quantity and quality of future tax payers. We find that introducing immigrants into the economy can reduce the tax burden of social security. If the survival probability of young agents to old age (or the replacement ratio) is high enough, the growth rate of GDP per worker for an economy with international migration will be higher than for a closed economy. Regarding migration policies, our numerical results indicate that economic growth rate of GDP per worker will first decrease then increase as the flow of immigrants increases. Attracting more skilled immigrants will enhance economic growth.

Over the past century longevity has steadily increased in countries which have experienced economic growth. The upward trend of life expectancy at age 65 for female and male in Canada, France, Japan, the United Kingdom and the U.S.A. during the period from 1980 to 2005 is presented in Fig. 1a and b. One implication of longevity growth is that more resources must be devoted to supporting the elderly, and governments have often raised social security expenditures. Fig. 1c shows that the ratio of pension expenditure to GDP increased along with life expectancy in Canada, France, Japan, United Kingdom and U.S.A. over the period from 1980 to2005.1However, for most industrialized countries a decline in mortality is accompanied by a reduction in fertility over the course of development. The graying of many countries' populations has led many to worry about the future tax burden of social security. In short, the concern is that expenditures on old-age entitlements will continue to grow even as there are fewer young people available to contribute to the tax base. There is a huge amount of literature devoted to the increasing social security burden caused by aging populations and possible solutions to this problem. These studies can be classified into two categories. The first line of research focuses on the sustainability of a pay-as-you-go (PAYG) system or on possible reforms to the social security system. The interaction of public investment in education, social security and growth is investigated by Kaganovich and Zilcha (1999) and Pecchenino and Pollard (2002). Based on a two-sector growth model, Zhang et al. (2001) compare the effects of mortality decline on long-run growth under funded and unfunded social security schemes. Recently, Groezen et al. (2003) and Fenge and Meier (2005) suggest that the burden of social security can be mitigated by using child allowances. With child allowances, the cost of raising children becomes lower and parents will have stronger incentives to have more children.2 The second category of previous research considers the problem of increasing social security burden in an economy with international migration. This research line demonstrates that introducing immigrants can alleviate the increasing burden of social security for developed countries. Razin and Sadka (1999) show that even introducing low-skilled immigrants who are often beneficiaries of the welfare state into an economy can be beneficial to all income and all age groups. With a dynamic set-up, adult immigrants share the burden of social security with natives upon their migration into the host country. Although these immigrants will create a new welfare burden in the future, the burden will be shared by newly introduced immigrants as long as the economy is ever-lasting. By using the data from Current Population Survey (CPS) and Public Use Microdata Sample (PUMS), Lee and Miller (2000) project the fiscal impact of immigrants and their descendents. They find that the net present value contributed by an immigrant is always positive to the Old-Age, Survivors, and Disability Insurance (OASDI). In this paper, we follow the second line of research to revisit the issue of an increasing social security burden in an economy with international migration. Besides examining the impact of introducing immigrants on the tax burden of a social security program, we also explore immigration's effects on key macroeconomic variables such as fertility, educational investment and economic growth. There are two major differences between this paper and existing research. First, while previous literature tends to treat fertility as an exogenous variable, fertility decisions are endogenized in this paper.3 It is now well-known that parents' decisions about fertility and the educational investments of their offspring are interdependent (Becker et al., 1990).4 So a migration model allowing parents to make decisions on fertility and educational investments in their children can give a more realistic view of policy effects. Second, we assume that the level of immigrants' human capital can be different from that of natives. Different levels of human capital cause immigrants and natives to make different choices about fertility and investment in their children's education which will in turn affect economic growth. Furthermore, fertility matters when considering an economy with heterogeneous agents, because it will also affect the future population structure and the quality of labor force. Both population structure and quality of labor force are important determinants of the burden of social security. We develop an overlapping generations model of social security within which adults make decisions about consumption, fertility and investment in their children's education. In order to compare the impact of longevity in an economy with international migration with that in a closed economy, we first consider a simple, closed economy with a PAYG social security program.5 We show that a balanced-growth-path (BGP) equilibrium exists in such an economy. An increase in the survival probability of young agents to old age will increase the tax rate necessary for social security and reduce after-tax income. Reduced after-tax income will cause adults to have fewer children (income effect). But the lower after-tax wage rate will also reduce the time cost of raising children and will motivate adults to have more children (substitution effect). We find that the income effect will dominate the substitution effect and an increase in the survival probability will reduce fertility and investment in children's education. This will slow economic growth. On the other hand, a higher survival probability to old age will induce more savings which increases economic growth. If the survival probability is high enough, the former effect will dominate the latter, and the increased survival probability will retard economic growth. We also find that an increase in social security payments will raise fertility and decrease educational expenditure if the degree of altruism is sufficiently low. Hence, under this condition an increase in the social security payment will lower economic growth. Next, we consider an economy with international immigrants. In order to study the impact of migration policy regarding the quality of immigrants, we assume that immigrants possess a different level of human capital from natives.6 The existence of heterogeneous agents makes the model more complicated, so we simulate the model to quantify the effects of migration policy. We find that introducing low-skilled immigrants into the economy can reduce the social security tax rate since immigrants have higher fertility. A migration policy which allows the amount of immigrants whose human capital level is 94% of natives to be 2% of the population of natives in every period can lower the social security tax rate by 1.83% in the long run. If the survival probability of young agents to old age (or the replacement ratio) is high enough, the growth rate of GDP per worker for an economy with migration will be higher than for a closed economy, provided that the gap of human capital between immigrants and natives is not large and the flow of immigrants is small. Comparing the numerical results under endogenous-fertility and exogenous-fertility models, we find that there will be biased estimation for the effects of the survival probability to old age on fertility, social security tax rate and economic growth if one does not consider the endogenous change in fertility. The benefits brought by the immigration policy would be exaggerated for an aging economy if fertility is not endogenized. We also study the impact of migration policies by changing the quantity and quality of immigrants. Our numerical results show that the economic growth rate will first decrease and then increase as more workers are allowed to migrate into the host country. On the other hand, attracting more skilled immigrants will enhance economic growth. The remainder of this paper is organized as follows. In the next section we develop an economy with international migration and a PAYG social security scheme. A closed economy is also analyzed in this section. In Section 4, we carry out numerical analysis to determine the effects of longevity, social security payments and migration policies on economic performance. The conclusions are presented in the final section.

This study presents an analysis of the impact of introducing immigration on economic performance in an overlapping generations model with a social security system. Our analysis shows that introducing immigrants into the economy can help reduce the social security burden. If the gap in human capital between immigrants and natives is not too large and the flow of immigrants is small, the growth rate in an economy with migration will be higher than that in a closed economy when the survival probability (or the replacement ratio) is sufficiently high. Concerning migration polices, we find that there is a U-shape relationship between the economic growth rate and the flow of immigrants. Although increasing the flow of immigrants can reduce the tax burden of social security, it is not necessarily good for economic growth. On the other hand, increasing the education level of immigrants is beneficial for economic growth, but it raises the tax burden of social security if immigrants are low-skilled workers. A few notes are worth discussing. First, throughout the paper, we focus on the effects of legal immigration. If we also consider illegal immigration into the analysis, then migration may help reduce the tax burden of social security even more. This is because illegal immigrants are not eligible for social security, but their children who are considered to be natives will participate in the social security program and need to share the burden of social security. Second, we use a simple three-period OLG model in this paper to study how immigration affects the burden of social security through the channel of fertility. It would be more interesting if we study this issue by extending the model to a life-cycle model with more periods of life (e.g., a period length of at most 5 years). A life-cycle model will allow us to study the effects of a particular immigration policy which allows certain ages and skills of workers to immigrate on the social security burden for the host country. We conclude by suggesting two possible directions in which our model can be easily extended and applied. First, we assume that all immigrants can stay in the host country permanently. It would be interesting to extend the model to study the effects of temporary immigrants. Secondly, our model ignores the social costs of introducing immigrants such as cultural conflict. Adding such costs may reduce economic growth and social welfare.