Childhood Neighborhood Conditions and the Persistence of Adult Income

38 Pages Posted: 7 Sep 2012 Last revised: 25 May 2015

Date Written: October 15, 2012


Economists have long been interested in the role of childhood neighborhood and family characteristics in explaining adult income. In the past, the importance of family traits has been emphasized over neighborhood characteristics in explaining income, and persistence - if included at all - has been treated as homogenous across individuals. The importance of persistence should not be overlooked because it is vital to the understanding of the long term effects of variables on income, the likelihood of a person to be trapped in a low income trap, and the time taken to recover from a negative shock. In this paper, I build upon the pre-existing literature by adding an individual level heterogeneity in income persistence and show that it is affected by childhood neighborhood variables. I apply a two-step correlated random effects GMM on the Panel Study of Income Dynamics Dataset and find that average persistence is about 0.38, with the poor having a higher persistence of income than the rich. This implies that the poor are affected more when the macroeconomy changes. I also find that the poor take a longer time to recover from a negative shock. My simulation exercise using this model shows that reducing idleness rate in childhood neighborhood of those earning in the bottom quartile of the income distribution can help to increase overall social welfare the most in the long run, when compared to other variables.

Keywords: income persistence, correlated random coefficient, GMM

JEL Classification: C1, D6, J

Suggested Citation

Islam, T. M. Tonmoy, Childhood Neighborhood Conditions and the Persistence of Adult Income (October 15, 2012). Regional Science and Urban Economics, Vol. 43, No. 4, 2013, Available at SSRN: or

T. M. Tonmoy Islam (Contact Author)

Elon University ( email )

Elon, NC 27244
United States


Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics