“A Very Uneven Playing Field: Economic Mobility in the United States”

Sep 14, 2022

CID Faculty member Pablo Mitnik published “A Very Uneven Playing Field: Economic Mobility in the United States” in CID’s discussion paper series earlier this year. The paper was published by the American Journal of Sociology in January 2024.

He sat down with CID to discuss his inspiration and process.

CID Faculty Pablo Mitnik listens to a session at the RC28 Summer Conference


CID: What is your paper about?

Pablo: “In this paper, my co-authors and I present results from a new U.S. data set that has been assembled from tax and other administrative sources to provide high-quality evidence on intergenerational income mobility and the transmission of economic advantages from parents to children. This data set allows us to take on the methodological problems that have complicated previous efforts to estimate intergenerational earnings and income elasticities (IGEs), which are crucial measures of mobility in the field (the larger an IGE, the less mobility there is). Unlike previous research with similar administrative data, we find quite low mobility or, equivalently, that a large share of economic inequality is transmitted across generations. For instance, we find that approximately two-thirds of the inequality between poor and well-off families is passed on to their children.”

“Comparing our results with fully comparable results for Denmark, we find that there is about twice as much income mobility in Denmark as in the U.S. With the help of a formal model, and relying on empirical evidence produced by sociologists, economists and political scientists, we argue that the much lower mobility of the U.S. cannot be attributed to any single factor. Instead, it is the joint result of virtually all factors governing the mobility process, including the returns to human capital, the amount of public investment in the human capital of low-income children, the amount of socioeconomic residential segregation, and the progressiveness of the tax-and-transfer system. For each of these four factors, the U.S. has opted for policies that are mobility-reducing, with the implication that any substantial increase in economic mobility will likely require a wide-ranging package of reforms that cut across many institutions. These institutions are, to a large extent, the same institutions responsible for the country’s very high level of income inequality, for instance the laws and regulations governing the capacity of workers to organize and exercise ‘countervailing power’ in the labor market and the country’s quite meager welfare state.”

CID: Why was it important to you to study this?

Pablo: “The transmission of economic advantages from parents to children and intergenerational income mobility in the U.S. have long been studied by social scientists. Moreover, the field has seen a surge of research in recent years, with mobility scholars relying more and more on massive datasets based on high-quality tax data and related administrative information. However, partially due to this unprecedented access to new data, there was in our view marked uncertainty regarding the true level of income mobility in the country, as measured by IGEs. Indeed, while twenty years of survey-data IGE estimates had led to the tentative conclusion that U.S. mobility is very low, the new wave of administrative-data studies suggested that there was either a lot of mobility or at least as much mobility as that found in other late-industrial countries. We thought that this was the result of downward biases generated by various methodological shortcomings in these studies, and that it was imperative to produce new administrative-data estimates as unaffected as possible by those biases.”

CID: How did you get started with this project? 

Pablo: “This paper has been very long in the making. In fact, it is the second version of a paper whose first version was released in 2018 as a Stanford Center on Poverty and Inequality Working Paper.  It started as a simple paper where we would present and discuss the main results of a joint project between researchers of Stanford CPI and the Statistics of Income Division of the Internal Revenue Service. However, it quickly evolved into a much more ambitious paper that also makes a number of methodological and theoretical contributions and articulates and supports its arguments with the help of various formal models (all included in appendices).”

CID: What remaining questions should be addressed by future research on this topic?

Pablo:“There are so many! Let me mention just a few that, for various reasons, are particularly salient in my mind these days. On the conceptual front, we need to specify very carefully the relationships among the various measures of intergenerational income mobility employed in empirical research. People often use various mobility measures—for instance the IGE, which focuses on monetary income, and the rank-rank slope, which focuses on income rank or relative position in the income distribution—as if they were interchangeable, that is, measures of the same thing rather than measures of importantly different although related things.  We do a little bit of this work of conceptual clarification in one of the sections of our paper, but substantially more needs to be done. Closely related, I believe it would be beneficial for the mobility field that we develop a much better understanding of how our measures of income mobility relate to the sophisticated notions of inequality of opportunity we find in the theories of distributive justice advanced by philosophers and a few economists. I have done a tiny bit of this in a 2019 Sociological Science paper (with two co-authors), but this wasn’t a central goal of that paper and so we barely uncovered the tip of the iceberg, so to speak.”

“On the empirical front, there are two areas of work that I think need sustained attention. The first is the macro-level relationship between cross-sectional inequality and intergenerational mobility, across both periods and places (countries or subnational jurisdictions like states). Previous empirical research in this area has produced mixed findings, but I suspect that this is in part because the mechanisms involved interact in very complicated ways. I believe that combining empirical studies with formal models that carefully specify these mechanisms may be one way of making progress. Second, I think that we social scientists need to pay much more attention to the impact that wealth inequality has in various domains, including the role it plays in generating inequality of opportunity. We need to understand better how family wealth interacts with other family resources, like education and income, and descriptive characteristics, like gender and race, to shape people’s life chances.”

Read the full paper.

About Pablo Mitnik

Pablo A. Mitnik joined the Stone Center for Inequality Dynamics (CID) as Assistant Research Scientist in September 2021. His research focuses on intergenerational mobility, economic inequality, labor markets, and statistical methods. A large part of his recent work has aimed at improving the methodological approaches, statistical models, and data used to measure mobility and inequality of opportunity.

Some of these methodological contributions can be found in three articles published in Sociological Methodology and Sociological Methods and Research. His empirical research has relied widely on the Statistics of Income Mobility Panel, a dataset based on tax and other administrative data whose construction he has coordinated. Key results from this research have been reported in a Sociological Science article on family-income mobility in the U.S. and a NBER book chapter that compares inequality of opportunity for income in Denmark and the US. Results from other research on trends in class mobility in the U.S. he has conducted can be found in The Annals of the American Academy of Political and Social Science.

Pablo’s current projects focus on wealth inequality and the transmission of wealth across generations, the role of gender and marriage in family-income mobility, the measurement of economic mobility in the U.S. with new censored regression and copula-based selection models, the formal modeling of the policy and institutional determinants of intergenerational mobility, and the methodological and conceptual foundations of research in mobility and inequality of opportunity. He obtained his Ph.D. in Sociology at the University of Wisconsin-Madison, was a postdoctoral scholar at Stanford University, and held several research positions at Stanford’s Center on Poverty and Inequality before joining CID.



More Like This