Association for Public Policy Analysis & Management

Besufekad Alemu, University of Minnesota

Introduction

Recent work on intergenerational economic mobility focuses on decomposing between the different causal mechanisms (see Black and Devereux, 2011[i]). These factors include family effects, neighborhood effects, and innate ability. Additionally, while there are an increasing number of studies that look at the causal mechanisms of intergenerational mobility for the overall population, those that analyze the intergenerational mobility of immigrants are limited. Analyzing the intergenerational mobility of immigrants can deepen our understanding of the immigrant adjustment experience. Simultaneously, this paper will add to the literature of networks impacts on social mobility.

I explore the following questions: (1) how do networks upon settlement affect the economic outcomes of immigrants and, (2) whether the different experiences of the parents significantly affect their children’s future labor market outcomes.

Model

Borjas (1992)[ii] develops a model that shows how ethnic capital at the time of initial settlement affects immigrant children. Ethnic capital is defined as the average quality, in terms of human capital, of the parent’s ethnic group. I modify the Borjas (1992) model to incorporate income that is heterogeneous with area of settlement. The premise is that income opportunities are different based on the networks into which immigrants settle. For example, in the simplest case, where networks have a positive impact on income, settling into a denser ethnic network will translate towards greater opportunities for first generation immigrants. This will allow the first generation to invest in the their children’s human capital. Hence, those individuals who initially settle into areas with higher ethnic density improve their own outcomes as well as those of their future generation.

Methods & Data

A thorough analysis of the impact of social networks on intergenerational mobility will require the ability to trace migrants from their initial areas and networks of settlement, to the first generation’s outcome, and the outcomes of later generations. In order to accomplish this, I use the full-count U.S. censuses from 1920 and 1940. Using an algorithm outlined in Feigenbaum (2015)[iii], I link immigrants who were between 0 and 15 years old in 1920, to themselves when they were 20 to 35 years old in 1940. This linking will form a panel data of the same individuals at two points in time.

While taking into account the social density of the networks upon initial settlement, I use econometric methods to compare the occupational scores of the parents’ generation in 1920 to the occupational scores of their children in 1940. I hypothesize social networks will have positive externalities on first as well as second-generation immigrants. This result would suggest that policies which facilitate connections between immigrant groups of the same background lead to more optimal levels of social welfare.