The second edition of this acclaimed graduate text provides a unified treatment of two methods used in contemporary econometric research, cross section and data panel methods. By focusing on assumptions that can be given behavioral content, the book maintains an appropriate level of rigor while emphasizing intuitive thinking. The analysis covers both linear and nonlinear models, including models with dynamics and/or individual heterogeneity. In addition to general estimation frameworks (particular methods of moments and maximum likelihood), specific linear and nonlinear methods are covered in detail, including probit and logit models and their multivariate, Tobit models, models for count data, censored and missing data schemes, causal (or treatment) effects, and duration analysis. Econometric Analysis of Cross Section and Panel Data was the first graduate econometrics text to focus on microeconomic data structures, allowing assumptions to be separated into population and sampling assumptions. This second edition has been substantially updated and revised. Improvements include a broader class of models for missing data problems; more detailed treatment of cluster problems, an important topic for empirical researchers; expanded discussion of "generalized instrumental variables" (GIV) estimation; new coverage (based on the author's own recent research) of inverse probability weighting; a more complete framework for estimating treatment effects with panel data, and a firmly established link between econometric approaches to nonlinear panel data and the "generalized estimating equation" literature popular in statistics and other fields. New attention is given to explaining when particular econometric methods can be applied; the goal is not only to tell readers what does work, but why certain "obvious" procedures do not. The numerous included exercises, both theoretical and computer-based, allow the reader to extend methods covered in the text and discover new insights.

/pdf/econometric-analysis-of-cross-section-and-panel-data-3jzr8emmtf.pdf

Econometric Analysis of Cross Section and Panel Data

Most papers that employ Differences-in-Differences estimation (DD) use many years of data and focus on serially correlated outcomes but ignore that the resulting standard errors are inconsistent. To illustrate the severity of this issue, we randomly generate placebo laws in state-level data on female wages from the Current Population Survey. For each law, we use OLS to compute the DD estimate of its “effect” as well as the standard error of this estimate. These conventional DD standard errors severely understate the standard deviation of the estimators: we find an “effect” significant at the 5 percent level for up to 45 percent of the placebo interventions. We use Monte Carlo simulations to investigate how well existing methods help solve this problem. Econometric corrections that place a specific parametric form on the time-series process do not perform well. Bootstrap (taking into account the autocorrelation of the data) works well when the number of states is large enough. Two corrections based on asymptotic approximation of the variance-covariance matrix work well for moderate numbers of states and one correction that collapses the time series information into a “pre”- and “post”-period and explicitly takes into account the effective sample size works well even for small numbers of states.

/pdf/how-much-should-we-trust-differences-in-differences-407p70a9b4.pdf

How Much Should We Trust Differences-In-Differences Estimates?

https://www.oxfordmartin.ox.ac.uk/downloads/academic/The_Future_of_Employment.pdf

The future of employment: How susceptible are jobs to computerisation?

This paper examines the reflection problem that arises when a researcher observing the distribution of behaviour in a population tries to infer whether the average behaviour in some group influences the behaviour of the individuals that comprise the group. It is found that inference is not possible unless the researcher has prior information specifying the compisition of reference groups. If this information is available, the prospects for inference depend critically on the population relationship between the variables defining reference groups and those directly affecting outcomes. Inference is difficult to implossible if these variables are functionally dependent or are statistically independent. The prospects are better if the variables defining reference groups and those directly affecting outcomes are moderately related in the population.

/pdf/identification-of-endogenous-social-effects-the-reflection-rurdh6yhqn.pdf

Identification of Endogenous Social Effects: The Reflection Problem

▪ Abstract This paper assesses and synthesizes the cumulative results of a new “neighborhood-effects” literature that examines social processes related to problem behaviors and health-related outcomes. Our review identified over 40 relevant studies published in peer-reviewed journals from the mid-1990s to 2001, the take-off point for an increasing level of interest in neighborhood effects. Moving beyond traditional characteristics such as concentrated poverty, we evaluate the salience of social-interactional and institutional mechanisms hypothesized to account for neighborhood-level variations in a variety of phenomena (e.g., delinquency, violence, depression, high-risk behavior), especially among adolescents. We highlight neighborhood ties, social control, mutual trust, institutional resources, disorder, and routine activity patterns. We also discuss a set of thorny methodological problems that plague the study of neighborhood effects, with special attention to selection bias. We conclude with promising ...

ASSESSING "NEIGHBORHOOD EFFECTS": Social Processes and New Directions in Research

In recent years, the publicly controlled sector has accounted for about 67 percent of all students in four-year institutions of higher education, whereas in 1897 the figure was only 22 percent.' The transition was early and swift, however, for on the eve of the United States' entry into World War II almost 50 percent of all students were in public-sector institutions. Thus, by 1940 more than 60 percent of the century-long expansion in public-sector enrollments (four-year) had already taken place. The relative rise of public higher education occurred decades before the rapid postwar acceleration of college enrollments and is apparent even without including community colleges. Differences across states in public-sector enrollments and in public support for higher education were considerable in the past but narrowed markedly after 1950. In 1929, enrollments in publicly controlled institutions averaged 6 per 1,000 residents in the Mountain and Pacific states, but only 0.8 per 1,000 residents in New England. State and local spending on higher education was $2,057 per 1,000 inhabitants in the Mountain states, but just $458 per 1,000 inhabitants in New England. In 1900 the cross-state coefficient of variation of public-sector enrollments per capita was 1.14, but this fell to 0.60 by 1929 and stands today at about 0.21. The convergence for the private-sector figure was far less (0.88 in 1900 and 0.70 in 1994). Changes in the cross-state variation in spending per capita and in publicsector tuition are similar. Despite the potent forces of market integration (Caroline Hoxby, 1997), many higher-education indicators exhibit substantial long-term persistence. A strong relationship exists at the state level between privatesector enrollments per capita today and those in 1900 (p = 0.64). A strong correlation (p = 0.46) also exists between public highereducation spending per capita today and the year of statehood (see also John M. Quigley and Daniel L. Rubinfeld, 1993). States admitted earlier have stronger private sectors and less well-supported public sectors, and the relationship is more complex than a simple difference between the Northeast and the Far West. Both the public and private highereducation sectors evolved in the 1890-1940 period. Fundamental changes in the creation and diffusion of knowledge, such as the specialization of disciplines, the professionalization of many occupations, the secularization of higher education, the increased role of research, and the ascent of the "university" form, altered the industrial structure of higher education. These changes increased the scale and scope of institutions and gave a productivity advantage to certain public-sector institutions. t Discussants: Robert P. Inman, University of Pennsylvania; Christopher S. Jencks, Harvard University. * Department of Economics, Harvard University, Cambridge, MA 02138. We acknowledge generous support from the National Science Foundation for our research and from the Russell Sage Foundation for our 1997-1998 academic leave. We thank Cheryl Seleski and Kerry Woodward for their superb assistance in the coding of the higher-education data set. Conversations with Caroline Minter Hoxby, James Poterba, and Marcus Stanley have helped clarify some of the ideas in this paper. We thank them all. ' In the interest of brevity, all citations for recent data are U.S. Department of Education (1996) and "today" means 1994. For all other data sources, see Goldin and Katz ( 1998). The share of total enrollments accounted for by publicly controlled institutions in 1994 was 78 percent when two-year institutions are also included. The term "public sector" in this paper means "publicly controlled sector."

The Origins of State-Level Differences in the Public Provision of Higher Education: 1890-1940

In this paper we provide theoretical and empirical analyses of an asymmetric-information model of layoffs in which the current employer is better informed about its workers' abilities than prospective employers are. The key feature of the model is that when firms have discretion with respect to whom to lay off, the market infers that laid-off workers are of low ability. Since no such negative inference should be attached o workers displaced in a plant closing, our model predicts that the postdisplacement wages of otherwise observationally equivalent workers will be higher for those displaced by plant closings than for those displaced by layoffs. An extension of our model predicts that the average postdisplacement unemployment spell of otherwise observationally equivalent workers will be shorter for those displaced by plant closings than for those displaced by layoffs. In our empirical work, we use data from the Displaced Workers Supplements in the January 1984 and 1986 Current Population Surveys. We find that the evidence (with respect to both re-employment wages and postdisplacement unemployment duration) is consistent with the idea that laid off workers are viewed less favorably by the market than are those losing jobs in plant closings. Our findings are much stronger for workers laid off from jobs where employers have discretion over whom to lay off.

Layoffs and Lemons

This article discusses trends in alternative work arrangements in the United States using data from the Contingent Worker Survey (CWS) supplements to the Current Population Survey (CPS) for 1995 to 2017, the 2015 RAND-Princeton Contingent Work Survey, and administrative tax data from the Internal Revenue Service for 2000 to 2016. Based on cyclically adjusted comparisons of the CPS CWS, measures using self-respondents in the CPS CWS, and measures of self-employment and 1099 workers from administrative tax data, we conclude that there has likely been a modest upward trend in the share of the U.S. workforce in alternative work arrangements during the 2000s. We also present evidence from Amazon Mechanical Turk suggesting that the basic monthly CPS question on multiple job holding misses many instances of multiple job holding.

/pdf/understanding-trends-in-alternative-work-arrangements-in-the-ocz7tmsy2p.pdf

Understanding Trends in Alternative Work Arrangements in the United States

To understand the impact of high-poverty neighborhoods on families, we collected data from participants at the Boston site of HUDs Moving To Opportunity (MTO) demonstration. MTO randomly assigned housing vouchers to applicants living in high-poverty public housing projects. The vouchers allowed families to move to private apartments, typically in lowerpoverty neighborhoods. This paper reports the results of our qualitative fieldwork which included observation of the operation of MTO in Boston and in-depth interviews with participants. This qualitative work had a profound impact on our MTO research. First, it caused us to refocus our quantitative data collection on a substantially different set of outcomes, primarily in the domains of safety and health. In our subsequent quantitative work, we found the largest program effects in the domains suggested by the qualitative interviews. Second, our qualitative work led us to develop an overall conceptual framework for thinking about the impacts of high-poverty neighborhoods on families and the ways in which moves to lower poverty neighborhoods might affect these families. We observed that fear of random violence appears to cause parents in ghetto families to focus a substantial portion of their daily routine on keeping their children safe. In later quantitative research, we confirmed that parental monitoring intensity was reduced among families offered housing vouchers. We further hypothesized that the need to live life on the watch may have broad implications for the future prospects of these families  including potential impacts on childrens development and on the mothers ability to engage in activities that would lead them to become economically self-sufficient, although sufficient data to assess this hypothesis are not yet available. Third, our fieldwork gave us a deeper understanding of the institutional details of the MTO program. This understanding has helped us to make judgements concerning the external validity of our MTO findings, and has prevented us from making some significant errors in interpreting our quantitative results. Fourth, by listening to MTO families talk about their lives, we learned a series of lessons that have important implications for housing policy. For many of the things we learned, it is hard to imagine any other data collection strategy that would have led us to these insights.

Bullets Don't Got No Name: Consequences of Fear in the Ghetto

This paper reviews the evidence on the efficacy of neighborhood and school interventions in improving the long-run outcomes of children growing up in poor families. We focus on studies exploiting exogenous sources of variation in neighborhoods and schools and which examine at least medium-term outcomes. Higher-quality neighborhoods improve family safety, adult subjective well-being and health, and girls' mental health. But they have no detectable impact on youth human capital, labor market outcomes, or risky behaviors. In contrast, higher-quality schools can improve children's academic achievement and can have longer-term positive impacts of increasing educational attainment and earnings and reducing incarceration and teen pregnancy.

/pdf/achieving-escape-velocity-neighborhood-and-school-32ey4t525c.pdf

Lawrence F. Katz

Papers

The Origins of State-Level Differences in the Public Provision of Higher Education: 1890-1940

Layoffs and Lemons

Understanding Trends in Alternative Work Arrangements in the United States

Bullets Don't Got No Name: Consequences of Fear in the Ghetto

Achieving Escape Velocity: Neighborhood and School Interventions to Reduce Persistent Inequality