Robert LaLonde: The Pursuit of Accuracy in an Imperfect World

March 26, 2018

Robin Mordfin

Professor Robert LaLonde leaves behind an enduring legacy as a researcher, professor, advisor, family man, and friend.

Thirty years ago, Robert LaLonde fundamentally changed the way economists evaluate training and social programs. He spent the next three decades applying the highest levels of precision to his own work, while at the same time teaching a large contingent of younger scholars new methods that helped make them some of the most influential thinkers in their fields.

LaLonde, who passed away on January 17, was memorialized recently at Rockefeller Chapel on the University of Chicago campus he loved, by his family, friends, colleagues, mentors, and former students, many of whom shared memories of the late professor’s life, legacy, and enduring impact.

“Bob personified the vision of scholarship that animated our school,” said Robert T. Michael, Eliakim Hastings Moore Distinguished Service Professor Emeritus and Dean Emeritus, University of Chicago Harris School of Public Policy. “The school was created to promote theory-guided, statistically-sophisticated empirical research on enduring public policy issues. In both his research and his teaching, he lived the core values of our school. I will always love Bob for that.”

The focus of LaLonde’s research evolved over his academic career, but consistently examined themes of social justice, seeking to better understand the factors involved in labor outcomes, especially for segments of society that had become disadvantaged. His interests included, but were not limited to, the costs of worker displacement, the impact of unions and collective bargaining, and the economic impact of incarceration. His research was lauded for its rigor, as well as its relevance and accessibility, which ensured that his work often informed public policy design and implementation. Of his many contributions to research, it was his seminal breakthroughs on policy evaluation, however, that many consider his most important.

“His thesis work published in the American Economic Review on the value of randomized controlled trial experiments vis à vis conventional research approaches shaped how we think about this fundamental problem in designing research and deciding how convincing the evidence is from a given study,” noted Bruce D. Meyer, the McCormick Foundation Professor at Harris Public Policy.

LaLonde’s groundbreaking dissertation was reworked and published as “Evaluating the Econometric Evaluations of Training Programs with Experimental Data” in 1986. The paper examined the methods used to evaluate how an employment training program might affect participants’ earnings. He found that the econometric procedures used at the time did not replicate the results that had been experimentally determined. Some methods showed that training was detrimental to participants, while others showed it to be wildly successful. LaLonde compared these results to the results from experimental methods and found that the training program brought about moderate gains.

“It is now quite common to compare experimental methods with non-experimental methods to determine if the non-experimental methods are getting the ‘right’ answer,” Meyer added. “Because many of these studies suggest that non-experimental studies are not getting the right answer, there has been a seismic shift towards the use of experimental and quasi-experimental methods in economics in the past thirty years.”

In fact, according to LaLonde’s PhD advisor, Orley Ashenfelter, the Joseph Douglas Green 1895 Professor of Economics at Princeton University, “It became common among economists to refer to ‘LaLonde-izing’ a study if a particular piece of research sought to determine if strong conventional empirical work would replicate in an experimental result.”

After completing his graduate work at Princeton in 1985, LaLonde made his dissertation data available to anyone who wanted it so that researchers could apply different approaches to it. It has become the canonical data set for trying out non-experimental methods. Soon thereafter, LaLonde joined the Graduate School of Business (now the Booth School of Business) at the University of Chicago as an assistant professor. He also continued his research on job training programs, their effectiveness and their uses.

In 1987, he wrote, “Labor Market Adjustments to Increased Immigration” with Robert H. Topel from the Graduate School of Business (now the Isidore Brown and Gladys J. Brown Distinguished Service Professor at Booth). Together they examined earnings and employment data for immigrants and native-born workers from the 1970 and 1980 censuses. (The issue of illegal immigration was not addressed, as census questionnaires do not refer to legal status, only to place of birth.) They found that an increase in new immigrants in an area causes a small decrease in earnings for that group, but that this dissipates with time. They also found that an increase in new immigrants lowers the earnings of earlier immigrants, and thus the two groups substitute for one another. However, a finding perhaps of greater interest was that immigrants had only a small effect on the incomes of native workers, and they concluded that immigrants are absorbed easily into the American economy.

In 1991, LaLonde also began expanding his look at employment by examining unions and collective bargaining. He began with an article in the Chicago Law Review that he wrote with the late Bernard Meltzer, the Edward Levi Distinguished Service Professor at the University of Chicago Law School. In “Hard Times for Unions: Another Look at the Significance of Employer Illegality,” they concluded that the assumption that the decline of labor unions in the United States was largely due to illegal activities by employers was incorrect. The authors noted that while some companies did negotiate in bad faith, act in a discriminatory fashion, or refuse to recognize unions, these activities have not increased significantly, and therefore are not the only reason for the decline in unions.

LaLonde continued his research into employment with one of his most cited papers, “Earnings Losses of Displaced Workers,” which he wrote with Louis Jacobson (now at George Washington University) and Daniel G. Sullivan of the Federal Reserve Bank, which was published in the American Economic Review in 1993. The trio used a data set derived from Pennsylvania unemployment insurance records to determine what long-term effects unemployment has on workers by comparing incomes before and after their lay-offs. They found that long-time employees for a company, which in this data set largely consisted of those employed in the steel industry, lost 25 percent of their income per year after dismissal. They also learned that workers began experiencing losses before they actually separated from their employers and that their age and gender had little effect on their earnings. Finally, they discovered that such losses were large even for those who later found new jobs in similar firms.

“The quarterly records they used showed a huge reduction in earnings over a ten-year period, and it wasn’t just job loss – these people lost pensions, and the value in their homes, and so much more,” clarified Dan Black, deputy dean at Harris. “The thing you have to understand about Bob is that the red string that runs through all his work was his interest in disadvantaged people. All his work was about people in positions that needed to be better understood so that we could find ways to help them.”

LaLonde returned to the University of Chicago as a full professor at Harris in 1998, after three years as an associate professor at Michigan State University. As he continued his research, LaLonde expanded his interest into secondary education, where he found that non-experimental approaches worked as a better way to evaluate its value. He also found a new category of disadvantaged individuals to study – incarcerated women and their job prospects. In 2005, he wrote “The Impact of Incarceration in State Prison on the Employment Prospects of Women” with Rosa Cho. Using a data set created from administrative records, they examined the relationship between time spent in prison and employment rates for former prisoners in Illinois. Interestingly, they found that while women generally see a decline in employment rates in the time leading up to their incarcerations, time spent in prison does not appear to harm their employment prospects later on. Unfortunately, they also found that these gains in employment do not last and these women’s chances of being employed eventually fall back to pre-incarceration levels. For some groups of women, including those with several children, those who served longer, or who were incarcerated for drug-related crimes, their employment rates are higher, often because time in prison leads to them breaking their addictions.

LaLonde also considered what happens to the children of women in prison. In “Incarcerated Women, Their Children, and the Nexus with Foster Care,” he worked with four other scholars to look at how often children of incarcerated women spend time in foster care and how prison time affects foster care outcomes. Using prison records for the state and county prison systems alongside foster care records from Chapin Hall, the researchers found that three-quarters of children of incarcerated women begin their time in foster care at the time the mother begins her prison stay, while the remaining 25 percent had been in the system beforehand. The scholars discovered that reunification of the family is less likely if the child or children have already spent considerable time in foster care before the mother’s imprisonment, and when prison sentences are long.

“This is not a topic that makes you popular,” Black pointed out. “But again, Bob was looking to find out what was really going on.” LaLonde so loved doing research that he was still working on his own papers at the time of his death. And he was still considering additional research focused on skills training and education and their effects on jobs and income, new unionism, and the foster care experiences of children of incarcerated mothers.

A Consummate, Inspiring, Nurturing Teacher

As an enthusiastic member of academia, Professor LaLonde not only (co-)authored more than 50 papers, and was working on more, he helped hundreds of students to complete their graduate degrees. As director of the doctoral program at Harris from 2008 to 2013, he served on 21 dissertation committees and assisted numerous others with their research. “Bob worked his students very hard, which they really appreciated,” Black said. “There was a lot of rewriting and reworking of data. He would tell them that their approach was not quite right and help them find a new way to look at the information. He really trained them not to just accept their first outcome.”

Henry Schultz Distinguished Service Professor James Heckman concurs. “He was a consummate teacher and colleague. We trained many students. We worked on many Ph.D. committees together, in which he was very insightful in guiding students.” Heckman, Black, and LaLonde also spent many years teaching a cross-listed Harris Public Policy/Department of Economics three-part course called Analysis of Micro Data that allowed students to see how theory was used by leading economists in their research. LaLonde’s ability to straddle the Graduate School of Business, Harris, and the Department of Economics was powerful, Heckman says. “People like Bob are the glue that hold the University together. He joins its disparate parts.”

“I was fortunate enough to both take and serve as a teaching assistant for the sequence taught by Professors Black, LaLonde, and Heckman,” said Janna Johnson (PhD’12), assistant professor at the Humphrey School of Public Affairs at the University of Minnesota. “This year-long advanced PhD-level econometrics sequence was absolutely the most valuable course sequence I took during graduate school. Professor LaLonde shared his mastery of incredibly difficult but important duration analysis, which I now teach in my own PhD-level advanced applied econometrics class at the University of Minnesota, drawing heavily from my notes from Professor LaLonde's course.”

According to Cynthia Cook-Conley, who is Ph.D. program specialist and executive assistant to the deputy deans at Harris, during LaLonde’s tenure as head of the doctoral program, he expanded the program to more than 100 students studying in a number of different disciplines.

Professor LaLonde also brought economic theory into places other than academia. Jacob Rosch, MPP’13, now with the Reinvestment Fund, recalls when the economist shared anecdotes about a doctor of his who had proposed a diagnosis of two rare diseases. “When others might have accepted the physician’s recommendation without question, he was compelled to ask if the two conditions, each extremely rare, were known to have any relationship, because, of course, considered as two independent variables, the probability of both occurring simultaneously would be exceedingly unlikely.”

Sebastian Gallegos PhD’16, a post-doctoral scholar at Princeton, was LaLonde’s teaching assistant and advisee and remembered a Ph.D. econometrics class in which they studied panel data with models of wine pricing. “To the last lecture, Bob brought in the finest bottles of wine for us to taste. And he told us that there is actually an Association of Wine Economists! Bob’s Ph.D. advisor Ashenfelter created it, and of course, they have an academic journal, the Journal of Wine Economics.”

A Family Man and Friend Who Took an Interest in Everyone He Met

At the memorial, LaLonde was remembered as much for being a kind, compassionate, and courageous family man as he was for being a celebrated economist. An active and involved parent, his family came above all else. Just as with his students and colleagues, he was a naturally nurturing father, husband, and sibling who cherished time with his wife, Laura, and their three children, Elena, Eve, and Julian. Whether chatting with his children about their day at school or writing additional math problems for them to practice, being supportive of his wife in her wishes for her career or their family, coaching Little League, checking in on his siblings, or just reveling in being a dad, he was caring, kind, and considerate.

LaLonde particularly treasured when he and his family could escape to their cabin in northern Wisconsin, when they attended rock concerts and other performances together, or when they just enjoyed one of their favorite movies at home with popcorn and a cup of tea. He enjoyed, among other things, the Chicago White Sox, watching baseball and other sports with Julian, listening to comedy routines with Elena, learning about Eve’s new favorite TV series, exploring less traveled parts of the world with Laura, an engaging political debate, ice skating, walks along Lake Michigan or in the Wisconsin woods, family reunions in Syracuse, catching up with his daughter’s adoption group families, taking his children to Garfield Park Conservatory, and tasting fine wines, and he described economics as his hobby. It really wasn’t important what LaLonde did with his family, just that they enjoyed time together.

An Enduring Legacy

“A pioneering researcher who relentlessly pursued timely, important policy questions, Bob also held an unwavering commitment to his students that made him a beloved, much sought-after teacher and mentor,“ said Daniel Diermeier, provost of the University of Chicago, the David Lee Shillinglaw Distinguished Service Professor at Harris, and former Harris dean. “Bob leaves behind a lasting legacy of impact in his field, as well as on the many lives he touched here at the University.”

His long-time friend and colleague, Daniel Sullivan, now executive vice president and director of research at the Federal Reserve Bank of Chicago and a Harris lecturer, remembered LaLonde this way: “The true measure of a person is how you treat the people who can’t help you get ahead. Bob never failed that test of generosity. He treated everyone well. We all ought to do our best to be like Bob in that regard.”

Economics was an enormous part of Robert LaLonde’s life, and he was an important part of his many students’ and colleagues’ lives. He did fascinating and transformative research, and he trained an entire new generation of economists. In fact, his extraordinary impact will be felt well into the future: as Cook-Conley pointed out, 95 percent of his doctoral students have gone on to become academics themselves and are teaching his methods around the world.

One of those former students, Margaret Triyana, who now teaches at Notre Dame, offered this testament to LaLonde, while flanked by more than 20 of his other students at the memorial service: “We were so blessed to have him as a teacher, advisor and mentor. He pushed us, supported us and took us under his wing. Even when we doubted ourselves and everyone else around us was saying ‘No,’ he would quietly encourage us to simply ‘Go for it.’”