The Effect of Age at Migration on Poverty Among Immigrants in Israel
Alisa Lewin, University of Haifa
Rebecca Raijman, University of Haifa
This study uses the "New Immigrant Survey 2010-2011" (n=3952), conducted by Israel''s Central Bureau of Statistics and matched with the 2008 census. This survey is a representative sample of immigrants aged 27-75 who arrived to Israel between 1990 and 2007.
Our findings reveal an effect of age at migration. Those arriving after age fifty have higher odds of being in poverty than those migrating at younger ages, even after controlling for country of origin, education and labor market participation, as well as gender and marital status. Language proficiency and social networks also affect the odds of being in poverty, but these effects do not differ by age at migration.
Immigrants differ not only in age at migration (age effect) and year of migration (period effect); they also arrive from different countries, that is, they evince an “origin effect.” Countries differ in levels of development, occupational structures, levels of human capital, and the extent that accumulated rights, savings, and pensions maybe transferred to another country. In addition, immigrants of different ethnic origin may be treated differently in the labor market: some groups may encounter discrimination while others may benefit from existing ethnic ties and networks. Immigrants also differ in their ability to transfer accumulated resources from their country of origin. For example, immigrants to Israel from the United States may import their savings and enjoy their social security benefits upon retirement, whereas immigrants from the Former Soviet Union (FSU) have, until very recently, forfeited their accumulated savings and pensions when they left their country of origin. Most vulnerable are those who migrate at an older age from countries in which they either did not accumulate retirement benefits or could not transfer them to the host country.
Data for the analysis are obtained from the "New Immigrant Survey 2010-2011" conducted by Israel''s Central Bureau of Statistics (CBS). The "New Immigrant Survey" is a nationally representative sample of 3,952 immigrants aged 27-75 who arrived in Israel between 1990 and 2007 and were also interviewed for the 2008 Israeli Census. The questionnaire used in the survey covers a broad range of topics related to immigrants'' socio-economic integration, as well as data regarding Hebrew proficiency and social networks. This survey was matched with Israeli 2008 census data from which we drew information on total household income, necessary for calculating poverty.
Our dependent variable is income poverty. Income poverty is a binary variable defined as 50% of the median equivalized household income. Equivalization was accomplished by dividing household income by the (standardized) number of household members. Standardization accounts for economies of scale by giving less weight to each household member in large households. This measurement procedure is similar to the way the official poverty line is calculated in Israel and in other OECD countries. Age at immigration is our main independent variable and we distinguish three age groups: arriving between age 6-19; 20-49 (reference category in regressions); 50 and older. Country of origin is another important variable in our study, and we distinguish four groups: FSU (the largest immigrant group in Israel today and the omitted category in regressions); Europe-America (and South Africa); Africa-Asia; Ethiopia. Language proficiency was measured as the mean of scores in fluency in reading, writing and speaking Hebrew. ''Social networks'' is a binary variable that indicates having Israeli friends. We control for education (in years) and labor force participation. We also control for gender and marital status.
Our findings show that age at migration indeed affects the odds of being in poverty, and those arriving after age fifty have higher odds of being in poverty than those migrating at younger ages, even after controlling for country of origin, education and labor market participation, as well as gender and marital status. Interaction terms between age at arrival and country of origin suggest that immigrants from the FSU arriving after age 50 have higher odds of experiencing poverty than other immigrants. Language proficiency and social networks also affect the odds of being in poverty. Hebrew fluency and having Israeli friends reduce the odds of poverty, net of all the variables in the equation, but these effects do not differ by age at migration.