23 although these results suggest that the earnings

This preview shows page 18 - 20 out of 22 pages.

23Although these results suggest that the earnings of recent immigrants approach those of natives, they do not imply that the earnings of recent immigrants, will on aver-age, exceed those of natives. According to the IHIC model, the incentive for human capital investment decreases with age and as source-country human capital becomes more transferable; it suggests that the strength of the inverse relationship between initial earnings and earnings growth decreases with immigrant time in the United States. This theoretical expectation is supported in research following immigrants for 20 years. Duleep and Regets (2002) found that although the inverse relationship continues beyond the initial 10-year period (the earnings growth increase associ-ated with lower initial earnings continues beyond the initial 10-year period), it is about one-third of the 10-year effect. The decrease in the ratio of immigrant-to-native earn-ings growth rates is also apparent in the longitudinal data discussed in the section “Evidence from Longitudinal Data” below.24 This strong inverse relationship between relative entry earnings for an immigrant cohort and its subsequent rela-tive earnings growth rate has been explored theoretically and empirically in a number of recent papers (Duleep and Regets 1992, 1994a, 1994b, 1996a, 1996b, 1996c, 1997a, 1997b, 1999, 2002).
3OCIAL 3ECURITY "ULLETIN s 6OL. 78 s /O. 1 s 2008 5;25Their finding of a strong inverse relationship persists even when several methodological concerns are taken into account. In Duleep and Regets (1994a, 2002) a simple method to completely circumvent regression-to-the-mean bias in cohort analyses of entry earnings and earnings growth is introduced and used. In Duleep and Regets (1994a, 1994b, 2002), a method for testing the sensitivity of the estimated inverse relationship to the effects of emigra-tion is introduced and applied.26Refer to Duleep and Dowhan (1999a, 1999b, 2002) for earlier analyses using the Social Security matched longitu-dinal data focused on the trend in foreign- and native-born earnings growth and the diverse ways these data can be used to study immigrant economic assimilation.27The ratios are defined as [(Y10- Y1)/Y1] F/ [(Y10- Y1)/Y1]Nwhere Y1and Y10denote the beginning- and end-year earnings, and F and N denote foreign and native born.28 These foreign- and native-born differences in earnings growth rates are statistically significant at a .05 level. The 1984–1985 and 1986–1987 cohorts are exceptions to the pattern of increasing earnings growth, possibly reflecting the newly legalized Immigration Reform and Control Act (IRCA) immigrants, as well as relatively high unemploy-ment rates for these years.29These differences are statistically significant at a .05 level.30As discussed in note 23, the empirical fact of faster earnings growth for recent immigrants does not imply that recent immigrants will eventually surpass the wages of the native born. Theoretically, one would expect the relative earnings growth advantage of the foreign born to the native born to be highest in the initial years of earning in the United States and to decrease with time in the country. This

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture