Sveriges Riksbank Prize in Economic Sciences · 2021
Joshua D. Angrist
He showed how lotteries and accidents of policy can reveal cause and effect, giving schools and governments evidence they can trust.
The Nobel citation: “for their methodological contributions to the analysis of causal relationships”
- Born
- September 18, 1960, Columbus, OH, USA
- Shared with
- David Card, Guido W. Imbens
- Affiliation at the time
- Massachusetts Institute of Technology (MIT), USA
Economics prize
2021
Shared with 2 other laureates.
Age that year
61years
Born in 1960.
Sources cited
19
Fact-checked September 24, 2026.
- He left high school after 11th grade and got into Oberlin College off the waitlist, despite poor grades. At first he needed tutoring in math.
- His PhD used the Vietnam draft lottery. In the early 1980s, white veterans were earning about 15% less than comparable men who had not served.
- He became an Israeli citizen in 1983 and spent two years in the Israel Defense Forces as a paratrooper, with deployments in Lebanon.
- One of his best-known studies used a 12th-century rule from Maimonides, which caps classes at 40 pupils, to measure how class size affects learning in Israel.
- In 2020 he volunteered for the Phase 3 trial of Moderna's COVID-19 vaccine. His father had caught polio a few years before the Salk vaccine arrived.
The breakthrough
The LATE theorem: what natural experiments can and cannot tell us
Does another year of school raise your pay? Or are people who stay in school just different from people who leave? Economists usually cannot run experiments to find out, so Angrist looked for accidents of chance that work like experiments. For his PhD he used the Vietnam draft lottery, which called men up by birthday, to measure how military service affected their later earnings. With Alan Krueger he used the season people were born in. Because of school-start dates and leaving-age laws, children born early in the year could legally quit school sooner, and on average they got a little less schooling. But a lottery or a birth date only nudges some people. Some men with safe lottery numbers volunteered anyway, and some who were at risk of the draft never served. So what does a study like this actually measure? In 1994 Angrist and Guido Imbens gave a precise answer. If a few clearly stated assumptions hold, the standard method, called instrumental variables, measures the average effect for one group: the people whose behavior the chance event changed. They called this the local average treatment effect, or LATE. Think of a discount coupon. It tells you about the shoppers who bought only because they had the coupon, not about the shoppers who would have bought anyway or never would.[2],[4],[5],[6],[7]
“I’m fortunate to be the son and grandson of enterprising and resilient people who aspired to build strong families and meaningful lives even when uprooted.”
What it meant for humanity
Angrist's methods did not cure a disease or build a machine. They changed how societies find out whether their policies work. Much older research compared people who had made different choices, and that kind of comparison often confuses cause with effect. The Nobel committee credits Angrist, Imbens and David Card with transforming empirical research. It says the LATE framework has been widely adopted and has made studies more transparent, because researchers must now spell out their assumptions. Angrist has put the approach to direct use in schools. Massachusetts charter schools with more applicants than seats must admit students by lottery, so each admissions season works like a randomized trial. Angrist and colleagues used these lotteries to show that Boston-area charter schools, many following a strict "No Excuses" model, raised test scores a great deal. Students who started furthest behind, including many Black students, gained the most. The same research found that charters in suburbs and rural areas did no better than local public schools. It also found that winning a seat at an elite exam school in Boston, New York or Chicago did not raise achievement by itself, a result his team called the "elite illusion". His draft-lottery work measured the long-run cost of Vietnam-era service to veterans' earnings. Through Blueprint Labs at MIT, which he co-founded in 2011, and the company Avela, launched in 2020, his group helps school districts design admissions systems and learn from their data. His textbooks with Jörn-Steffen Pischke and a free online course teach these methods around the world.
- In lotteries for KIPP Lynn, a charter middle school in Massachusetts, being offered a seat raised math scores by about 0.36 standard deviations. Angrist calls this a large and statistically precise effect.[5]
- Lottery studies by Angrist's team found that a year in a Boston charter middle school raised math scores by about a quarter of a standard deviation, with the biggest gains for disadvantaged students and English learners.[15]
- Using the draft lottery and Social Security records, he found that in the early 1980s white Vietnam-era veterans were earning about 15% less than comparable men who had not served.[7]
- In January 2010 Massachusetts raised the charter cap in its lowest-performing districts from 9% to 18% of a district's budget, just as lottery studies were showing large gains. A 2016 study counted 13 new charters opened in Boston since then.[17]
- His studies of exam schools in Boston, New York and Chicago found that winning a seat did not raise achievement. In Chicago, applicants who only just qualified did worse on the ACT.[5]
- His textbooks with Jörn-Steffen Pischke, Mostly Harmless Econometrics and Mastering 'Metrics, and a free online econometrics course have taught these methods to students around the world.[11],[12]
Impact in numbers
Angrist's contribution is a way of learning, not a product. The LATE framework tells researchers exactly what a lottery, a birthday cutoff or a policy rule can prove, and what it cannot. That discipline now runs through labor economics, education research and field experiments, and his textbooks have trained a generation of applied researchers. The findings it produced have shaped real debates: the earnings cost of the Vietnam draft, whether class size matters, whether urban charter schools help disadvantaged children, and whether elite exam schools add value. We make no quantified claim. His studies measured effects of policies that others designed and ran, so crediting him with the students those policies reached would double count their work. The influence of better evidence on policy is real, but it cannot be honestly counted.
EducationEconomyFundamental science
No number is given here on purpose. Some contributions cannot be counted honestly, and we would rather describe them than invent a figure.
The double edge
No physical harm is linked to Angrist's work, but its limits and its uses are debated. The economists Angus Deaton and James Heckman argue that LATE describes only a group of people who cannot be named, the "compliers", and so it may not answer the questions policymakers actually face. Angrist and Jörn-Steffen Pischke reply that better research design has brought a "credibility revolution" to economics. In 1995 other researchers showed that some of his early quarter-of-birth estimates were open to "weak instrument" bias. In 1993 he and Alan Krueger had withdrawn an already-accepted paper after realizing its estimates were badly biased. Later methods, some of which he helped develop, showed that the main 1991 findings still held. His charter school findings became part of a bitter political fight. His co-authors cited them in 2016 to argue for lifting Massachusetts' charter cap, but voters rejected the expansion amid fears that charters drain money from district schools. When he revisited the Maimonides class-size study with Israeli data from 2002 to 2011, the class-size effects came out close to zero.
- Minor
Critics question what LATE really measures
Angus Deaton wrote that he found it hard to make sense of LATE, because it averages effects over a group that cannot be identified in advance. He cited work by James Heckman and Sergio Urzúa showing cases where LATE can mislead about the underlying economics. Angrist and Pischke respond that clear research designs have made empirical economics far more credible.[10],[13],[14]
- Minor
Weak-instrument bias in early estimates
In 1995 John Bound, David Jaeger and Regina Baker showed that some estimates in the 1991 Angrist-Krueger schooling paper could be biased toward simpler estimates. In 1993 Angrist and Krueger had withdrawn an accepted draft-lottery schooling paper for the same reason. Newer methods later showed the main 1991 findings still held.[1]
- Moderate
Charter school evidence in a polarized fight
Angrist's lottery studies were central to the case for charter expansion in Massachusetts. In 2016 voters rejected a ballot measure to add up to 12 charters a year, amid worries that charters drain money from district schools. A 2018 study found that charter expansion actually raised per-pupil spending in traditional schools.[15],[16]
- Minor
A celebrated class-size result did not hold up in later data
The 1999 Maimonides Rule study found that smaller classes raised 4th- and 5th-grade test scores in Israel. When Angrist and colleagues repeated the design with Israeli data from 2002 to 2011, the estimated class-size effects were close to zero.[5]
Against the odds
Angrist grew up in a secure, middle-class American Jewish family. No source reports that antisemitism blocked his education or career. His family's history, though, was shaped by persecution. His father's mother's family fled pogroms in Odessa after 1905 and landed in Galveston, Texas. His mother's parents came to Montreal from small towns in Lithuania in the mid-1920s. The relatives who stayed behind, including his grandmother's parents, were murdered by the Nazis in 1943. German killing units and local Lithuanian collaborators killed about nine in ten of Lithuania's Jews. In Montreal, a grade-school teacher pushed his mother to change her name from Sarah to Shirley, implying that Sarah sounded too Jewish. His wife's parents were Holocaust survivors who met in a displaced persons camp in Italy. As a student in London in 1980-81, he found British students hostile to Israel; their national union had "no-platformed" pro-Israel speech. That hostility sparked his interest in Zionism. He chose to move to Israel and serve in a combat unit, with deployments in Lebanon. In October 2018 a gunman killed 11 worshippers at Pittsburgh's Tree of Life synagogue complex, the deadliest attack on Jews in US history. One of the dead belonged to his parents' congregation, Dor Hadash.
1905
Persecution
His paternal grandmother's family fled pogroms in Odessa after 1905 and emigrated to Galveston, Texas.[1]
1943
Family killed
His maternal grandmother's parents and other relatives who stayed in Lithuania were murdered by the Nazis in 1943. German killing units and local collaborators killed about nine in ten of Lithuania's Jews.[1],[18]
—
Discrimination
In Montreal, a grade-school teacher pushed his mother to change her name from Sarah to Shirley, implying that Sarah sounded too Jewish.[1]
1984
War
As an Israeli paratrooper he was deployed to Lebanon, mostly in the Bekaa Valley, near the end of a war he later called ill-considered. He says some friends were less lucky than he was.[1]
2018
Antisemitic attack
The October 2018 Tree of Life synagogue shooting in Pittsburgh killed 11 people, including a member of his parents' congregation, Dor Hadash.[1],[19]
Jewish background
Both his parents are Jewish. His mother grew up in a Yiddish-speaking home in Montreal, the daughter of immigrants from Lithuania. His father's family came from Ukraine and Poland. His parents met at a Hillel-run Jewish summer camp. He had his bar mitzvah at Pittsburgh's Congregation Dor Hadash. As a student he was drawn to Zionism. He moved to Israel in 1982, became a citizen in 1983, served in the army and married an Israeli. He taught at the Hebrew University from 1991 to 1996, and today he holds both US and Israeli citizenship.[1],[8],[9],[12]
Key dates
September 18, 1960
Born in Columbus, Ohio, where his parents were graduate students at Ohio State University; the family moved to Pittsburgh in 1962.[1],[2]
1977
Leaves Taylor Allderdice High School after 11th grade and works full time in facilities for people with disabilities.[1]
1982
Graduates in economics from Oberlin College and moves to Israel to study at the Hebrew University.[1],[11]
March 1983
Becomes an Israeli citizen. That summer he is drafted for two years in the IDF and serves as a paratrooper, including in Lebanon.[1]
August 1985
Marries Miriam Poznanski on a kibbutz and starts a PhD at Princeton.[1]
1989
Earns his Princeton PhD with a thesis on the Vietnam draft lottery. Joins Harvard as an assistant professor.[1],[3],[11]
1990
Publishes his draft-lottery study of veterans' lifetime earnings in the American Economic Review.[7]
1991
With Alan Krueger, publishes the quarter-of-birth study of schooling and earnings. Returns to Israel to teach at the Hebrew University.[1],[4],[8]
1994
With Guido Imbens, publishes the local average treatment effect (LATE) framework.[2],[6]
1996
Moves to MIT, having been offered a tenured post there.[1]
1999
With Victor Lavy, publishes the Maimonides Rule class-size study, one of the first regression discontinuity studies in economics.[1],[5]
2011
Co-founds MIT's School Effectiveness and Inequality Initiative, later renamed Blueprint Labs.[1],[5]
October 11, 2021
Shares the economics prize: half to David Card, the other half jointly to Angrist and Guido Imbens, for methods of analyzing cause and effect.[3]
December 8, 2021
His recorded Nobel lecture, "Empirical Strategies in Economics: Illuminating the Path from Cause to Effect", is released.[5]
Sources
- 1.Joshua D. Angrist - Biographical · NobelPrize.org (Nobel Foundation), 2022
- 2.Joshua D. Angrist - Facts - 2021 · NobelPrize.org, 2021
- 3.Press release: The Prize in Economic Sciences 2021 · NobelPrize.org (Royal Swedish Academy of Sciences), 2021
- 4.Popular information: Natural experiments help answer important questions · NobelPrize.org (Royal Swedish Academy of Sciences), 2021
- 5.Empirical Strategies in Economics: Illuminating the Path from Cause to Effect (Nobel Prize lecture) · NobelPrize.org (Nobel Foundation), 2021
- 6.Scientific Background: Answering causal questions using observational data · Committee for the Prize in Economic Sciences, Royal Swedish Academy of Sciences, 2021
- 7.Lifetime Earnings and the Vietnam Era Draft Lottery: Evidence from Social Security Administrative Records · American Economic Review 80(3) (author copy, MIT Economics), 1990
- 8.Israeli-American Joshua Angrist shares Nobel Prize in economics · Jewish Telegraphic Agency, 2021
- 9.Former Pittsburgher awarded Nobel prize in economics · Pittsburgh Jewish Chronicle, 2021
- 10.The Credibility Revolution in Empirical Economics: How Better Research Design Is Taking the Con out of Econometrics · Journal of Economic Perspectives 24(2), American Economic Association (Angrist and Pischke), 2010
- 11.Joshua Angrist · Wikipedia
- 12.Avela Founder Josh Angrist Wins Nobel Prize · Avela Education (press release via PR Newswire), 2021
- 13.Instruments, Randomization, and Learning about Development · Journal of Economic Literature 48(2) (Angus Deaton; author version, Princeton University), 2010
- 14.Comparing IV With Structural Models: What Simple IV Can and Cannot Identify (NBER Working Paper 14706) · National Bureau of Economic Research (James J. Heckman and Sergio Urzúa), 2009
- 15.Massachusetts Charter Cap Holds Back Disadvantaged Students · Education Next (Sarah R. Cohodes and Susan Dynarski), 2016
- 16.Fiscal and Education Spillovers from Charter School Expansion (CEP Discussion Paper 1577) · Centre for Economic Performance, London School of Economics (Matthew Ridley and Camille Terrier), 2018
- 17.Can Successful Schools Replicate? Evidence from Boston Charter Schools (SREE conference abstract) · ERIC, US Department of Education (Cohodes, Setren and Walters), 2016
- 18.Lithuania · United States Holocaust Memorial Museum, Holocaust Encyclopedia
- 19.Pittsburgh synagogue mass shooter sentenced to death after victims face him in court · NBC News, 2023
Fact-checked on September 24, 2026 by a separate AI fact-checking pass that re-opened the sources, with 5 corrections made. How we check
Suggest a correction