Daniel L. Chen
Institute for Advanced Study
Toulouse School of Economics
Université Toulouse Faculty of Law
Centre National de la Recherche Scientifique
Center for Data Science, NYU Courant Institute of Mathematics Visitor, National Bureau of Economic Research (Winter, Summer)
oTree Open Source Research Association
Law and economics is divided between the consequentialist view that optimal policy should be based on calculations of costs and benefits and a non-consequentialist view that policy should be determined deontologically: from duties we derive what is the correct law–what is right and just.
Are there deontological motivations, and if there are, how might we formally model these motivations? What are the implications of things like deontological motivations for economics and policy, and what puzzles can we explain with deontological motivations that we cannot with standard models? What is the impact of law & economics on justice?
To answer these questions, his research has
- curated 12 terabytes of archival and administrative data on judges and courts where normative ideas incubate; the data bridge machine learning, causal inference, and normative theories of justice regarding equal treatment before the law and equality based on recognition of difference
- developed a programming language to study normative commitments in experiments, now used in 23 countries, 10 academic disciplines, private and public sectors, and local high schools
Some current themes on consequences, formation, and measurement of normative commitments (and applications in law) include:
- Law and Development tracing the incentives that led to what are now viewed as human rights violations
- Markets and Morality how market forces interact with normative commitments
- Behavioral Judging social and psychological, economic and political influences on legal ideas and production of justice
- Law and Legitimacy role of legitimacy in legal compliance
- Demography of Ideas economics of interpretation as a source of normative commitments
His research has been accepted in leading economics journals (American Economic Review, Econometrica, Quarterly Journal of Economics, and Journal of Political Economy), computer science journals (Journal of Machine Learning Research), double-blind peer-review law outlets (Stanford-Yale Junior Faculty Forum and Law and STEM Junior Faculty Forum), and press outlets (Washington Post).
The research has anchored successful applications with € 3 200 000 in grant budget awarded for “Origins and Effects of Normative Commitments”, “Positive Foundations of Normative Commitments”, and “Digital Humanities: Legal Analysis in a Big Data World” and received support from the European Research Council Consolidator Grant, Swiss National Science Foundation, and Agence Nationale de la Recherche. His work has also been supported by the MacArthur Foundation, Ewing Marion Kauffman Foundation, Social Science Research Council, Templeton Foundation, Earhart Foundation, Institute for Humane Studies, National Institute of Child Health and Human Development, National Institutes of Health, and National Science Foundation.
Papers
Articles Under Review
The Deterrent Effect of the Death Penalty? Evidence from British Commutations During World War I
American Economic Review, invited to resubmit; TSE Working Paper No. 16-706
Covering: Mutable Characteristics and Perceptions of Voice in the U.S. Supreme Court
Review of Economic Studies, invited to resubmit; TSE Working Paper No. 16-680; Y. Halberstam, A. Yu
Mapping the Geometry of Law using Document Embeddings
Science Advances, invited to resubmit; E. Ash
Government Expropriation Increases Economic Growth and Racial Inequality: Evidence from Eminent Domain
Economic Journal, invited to resubmit; TSE Working Paper No. 16-693; S. Yeh
Mandatory Disclosure: Theory and Evidence from Industry-Physician Relationships
Journal of Legal Studies, revise and resubmit; TSE Working Paper No. 16-716; V. Levonyan, E. Reinhart, G. Taksler
Judicial Analytics and the Great Transformation of American Law
Journal of Artificial Intelligence and the Law, revise and resubmit.
Policies Affect Preferences: Evidence from Random Variation in Abortion Jurisprudence
Journal of Political Economy, under review, 12/15; TSE Working Paper No. 16-723; V. Levonyan, S. Yeh
Mood and the Malleability of Moral Reasoning
Journal of Political Economy, under review; TSE Working Paper No. 16-707
Clash of Norms: Judicial Leniency on Defendant Birthdays
Quarterly Journal of Economics, under review; A. Philippe
Events Unrelated to Crime Predict Criminal Sentence Length
Science, under review; M. Loecher, N. Barry, L. Buchanan, E. Bakhturina
Non-Confrontational Extremists
Journal of the European Economic Association, under review; TSE Working Paper No. 16-694; M. Michaeli, D. Spiro
Priming Ideology: Why Presidential Elections Affect U.S. Judges
Journal of Legal Studies, under review; TSE Working Paper No. 16-681
Learning Policy Levers: Toward Automated Policy Analysis Using Judicial Corpora
Journal of Law, Economics, and Organization, under review; E. Ash, R. Delgado, E. Fierro, S. Lin
Economics Publications
Electoral Cycles Among U.S. Courts of Appeals Judges
Journal of Law and Economics, 60(3), 479-496, 2017; TSE Working Paper No. 16-704; C. Berdejo
What Matters: Agreement Among U.S. Courts of Appeals Judges NIPS16
Journal of Machine Learning Research, forthcoming; TSE Working Paper No. 16-747; X. Cui, L. Shang, J. Zheng
Decision-Making Under the Gambler’s Fallacy: Evidence From Asylum Courts, Loan Officers, and Baseball Umpires
Quarterly Journal of Economics, 131(3): 1181-1241, 2016; NBER Working Paper No. 22026; TSE Working Paper No. 16-674; T. Moskowitz, K. Shue
oTree: An Open Source Platform for Online, Lab, and Field Experiments
Journal of Behavioral and Experimental Finance, 9(1), 88-97, 2016, M. Schonger, C. Wickens
Are Online Labor Markets Spot Markets for Tasks? A Field Experiment on the Behavioral Response to Wages Cuts
Information Systems Research, 27(2), 403-423; TSE Working Paper No. 16-675; J. Horton
Can Markets Stimulate Rights? On the Alienability of Legal Claims
RAND Journal of Economics, 46(1), 23-65, 2015
The Construction of Morals
Journal of Economic Behavior and Organization, 104, 84-105, 2014; S. Yeh
Economics, Religion, and Culture: A Brief Introduction
Journal of Economic Behavior and Organization, 104, 1-3, 2014; D. Hungerman
Sparse Models and Methods for Optimal Instruments with an Application to Eminent Domain
Econometrica, 80(6), 2369-2429, 2012; A. Belloni, V. Chernozhukov, C. Hansen
Does Disclosure Matter?
Journal of Institutional and Theoretical Economics, 168(1), 120-123, 2012
Designing Incentives for Inexpert Human Raters
Proceedings of the ACM Conference on Computer Supported Cooperative Work, 2011; J. Horton, A. Shaw
Club Goods and Group Identity: Evidence from Islamic Resurgence During the Indonesian Financial Crisis
Journal of Political Economy, 118(2), 300-354, 2010
Islamic Resurgence and Social Violence During the Indonesian Financial Crisis
Institutions and Norms in Economic Development, MIT Press, ed. M. Gradstein and K. Konrad, 179-200, 2007
Income Distribution Dynamics with Endogenous Fertility
Journal of Economic Growth, 7(3), 227-258, 2002; NBER Working Paper No. w7530; M. Kremer
Income-Distribution Dynamics with Endogenous Fertility
American Economic Review, 89(2), 155-160, May 1999; NBER Working Paper No. w7530; M. Kremer
An Empirical Study Comparing the Controlled Random Search Procedure and the General Simulated Annealing Method for Function Optimization
Proceedings of the Fourth Annual District of Columbia Computer Conference, 1995
Law Publications (peer review)
Analysis of Vocal Implicit Bias in SCOTUS Decisions Through Predictive Modelling
Proceedings of Experimental Linguistics, 2018; E. Ash, R. Vunikili, H. Ochani, D. Jaiswal, R. Deshmukh
Machine Learning and Rule of Law
Computational Analysis of Law, forthcoming
Due Process and Voice in the U.S. Supreme Court
Computational Analysis of Law, forthcoming; M. Kumar
The Shareholder Wealth Effects of Delaware Litigation
American Law and Economics Review, 19(2), 287-326, 2017; TSE Working Paper No. 16-683; A. Badawi
Law and Literature: Theory and Evidence on Empathy and Guile
Review of Law and Economics, forthcoming; TSE Working Paper No. 16-684
Intermediated Social Preferences: Altruism in an Algorithmic Era
Advances in Economics of Religion, Vol. 158, 2018.
Tastes for Desert and Placation: A Reference Point-Dependent Model of Social Preferences
Research in Experimental Economics, forthcoming; TSE Working Paper No. 16-725
The Genealogy of Ideology: Identifying Persuasive Memes and Predicting Agreement in the U.S. Courts of Appeals
Proceedings of the ACM Conference on AI and the Law, 2017; TSE Working Paper, No. 17-783;
Early Predictability of Asylum Court Decisions
Proceedings of the ACM Conference on AI and the Law, 2017; TSE Working Paper No. 17-781; M. Dunn, L. Sagun, H. Sirin
Can Machine Learning Help Predict the Outcome of Asylum Adjudications?
Proceedings of the ACM Conference on AI and the Law, 2017; TSE Working Paper No. 17-782; J. Eagel
Perceived Masculinity Predicts U.S. Supreme Court Outcomes
PLoS-ONE, 11(10), e0164324; TSE Working Paper No. 16-682; Y. Halberstam, A. Yu
Investigating Variation in English Vowel-to-Vowel Coarticulation in a Longitudinal Phonetic Corpus
Proceedings of the 18th International Congress of Phonetic Sciences, 2015; C. Abrego-Collier, J. Phillips, B. Pillion, A. Yu
Does Appellate Precedent Matter? Stock Price Responses to Appellate Court Decisions of FCC Actions
Empirical Legal Analysis: Assessing the Performance of Legal Institutions, 2013; A. Araiza, S. Yeh
Can Countries Reverse Fertility Decline? Evidence from France's Marriage and Baby Bonuses, 1929-1981
International Tax and Public Finance, 118(3), 252-271, 2011
Religion, Welfare Politics, and Church-State Separation
Journal of Ecumenical Studies, 42(1), 42-52, 2007; J. Lind
Law Publications in non-peer review journals
What Kind of Judge is Brett Kavanaugh? A Quantitative Analysis
Cardozo Law Review, forthcoming; E. Ash
Distinguishing Between Custom and Law: Empirical Examples of Endogeneity from Property and First Amendment Precedents
William & Mary Bill of Rights Journal, 21(1081), 2013; S. Yeh
A Market for Justice: A First Empirical Look at Third-Party Litigation Funding
Journal of Business Law, 15(3), 2013; D. Abrams
'Not that Smart': Sonia Sotomayor and the Construction of Merit
Emory Law Journal, 61(4), 2012; G. Charles, M. Gulati
Trading Off Reproductive Technology and Adoption: A Response to Appleton and Pollak
Minnesota Law Review, 95(6), 2011; I. G. Cohen
Trading Off Reproductive Technology and Adoption: Do IVF Subsidies Decrease Adoption Rates and Should It Matter? Stanford-Yale Junior Faculty Forum
Minnesota Law Review, 95(2), 2010; I. G. Cohen
Writing for Public Audiences
Kavanaugh is radically conservative. Here's the data to prove it
Washington Post, Jul 10, 2018; E. Ash
Completed Drafts (submitted to journals)
Algorithms as Prosecutors: Lowering Rearrest Rates Without Disparate Impacts and Identifying Defendant Characteristics ‘Noisy’ to Human Decision-Makers Law and STEM Junior Faculty Forum NIPS17
D. Amaranto, E. Ash, L. Ren, C. Roper
Implicit Egoism in Sentencing Decisions: First Letter Name Effects with Randomly Assigned Defendants
TSE Working Paper No. 16-726
Insiders, Outsiders, and Involuntary Unemployment: Sexual Harassment Exacerbates Gender Inequality
TSE Working Paper No. 16-687; J. Sethi
The Political Economy of Beliefs: Why Fiscal and Social Conservatives/Liberals (Sometimes) Come Hand-in-Hand
TSE Working Paper No. 16-722; J. Lind
How Do Rights Revolutions Occur? Free Speech and the First Amendment
TSE Working Paper No. 16-705; S. Yeh
Judicial Compliance in District Courts
TSE Working Paper No. 16-715; J. Frankenreiter, S. Yeh
Markets, Morality, and Economic Growth: Competition Affects Moral Judgment
TSE Working Paper No. 16-692
Social Preferences or Sacred Values? Theory and Evidence of Deontological Motivations
TSE Working Paper No. 16-714; M. Schonger
A Theory of Experiments: Invariance of Equilibrium to the Strategy Method of Elicitation
TSE Working Paper No. 16-724; M. Schonger
Testing Axiomatizations of Ambiguity Aversion
TSE Working Paper No. 16-703; M. Schonger
Is Ambiguity Aversion a Preference?
TSE Working Paper No. 16-703; M. Schonger
Judicial Exits from the U.S. Courts of Appeals are Politically Motivated
TSE Working Paper No. 16-721
Is Justice Really Blind? And Is It Also Deaf?
M. Kumar
Using Machine Learning to Detect Human Rights Abuses
How Does Science Progress? A Statistical Approach to Postmodern Theories of Knowledge
Presented (Drafts available/submitted to conferences)
Constitutional Law
Ideas Have Consequences: The Impact of Law and Economics on American Justice
E. Ash, S. Naidu
Motivated Reasoning in the Field: Polarization of Precedent, Prose, Vote, and Retirement in U.S. Circuit Courts, 1800-2013
E. Ash, W. Lu
Implicit Bias in the Judiciary: Evidence from Judicial Language Associations
E. Ash, A. Ornaghi
Judicial Sentiments and Social Attitudes: Evidence from U.S. Circuit Courts
E. Ash, S. Galletta
Who Cares? Measuring Attitude Strength in a Polarized Environment
C. Cavaille, K. Van der Straeten
Is Ideology Infectious? Evidence from Repeated Random Exposure in the U.S. Courts of Appeals
Mitosis of Ideology: Polarization and Dissimilation in the U.S. Courts of Appeals
Precedent vs. Politics? Case Similarity Predicts Supreme Court Decisions Better Than Ideology
E. Ash
Affirm or Reverse? Using Machine Learning To Help Judges Write Opinions
E. Ash
Religious Freedoms, Church-State Separation, and Religiosity: Evidence from Randomly Assigned Judges
E. Ash
Law and Norms: A Machine Learning Approach to Predicting Attitudes Towards Abortion
K. Kwan, M. Maass, L. Ortiz
Criminal Law
Predicting Punitiveness and Sentencing Disparities from Judicial Corpora
E. Ash
Deep IV in Law: Automated Impact Analysis of Court Precedent and Application to Criminal Sentencing
E. Ash, X. Zhang, Z. Huang, R. Wang
How Prosecutors Exacerbate Racial Disparities: Screening Gaps, Race Effects, and Courtrooms Interactions
White Privilege in Law? The Reproduction of Racial Hierarchy in Sentencing Disparities
Procedure
Automated Fact-Value Distinction in Court Opinions
Y. Cao, E. Ash
Automated Classification of Modes of Moral Reasoning in Judicial Decisions
N. Mainali, L. Meier, E. Ash
Machine Prediction of Appeal Success in U.S. Asylum Courts
E. Ash
Tone of Voice Predicts Political Attitudes: Evidence from U.S. Supreme Court Oral Arguments
Y. Kadiri, Z. Pajor-Gyulai, T. Leble, E. Ash
Mimicry: Phonetic Accommodation Predicts U.S. Supreme Court Votes
A. Yu
Supreme Court Vacancies and Discretionary Opinion Writing in Federal Circuit Courts
E. Ash, W. Lu
Strategic Display of Emotions
A. Hopfensitz, J. Van Der Ven, B. Van Leeuwen
Contracts
Predicting Malpractice Risk Using Microdata on Treatment Choices
E. Ash, J. Van Prys, H. Wang, H. Zhao
Tax
Confusing Average and Marginal Tax Rates: Experimental Evidence
Employment Law
The Economics of Crowdsourcing: A Theory of Disaggregated Labor Markets
International Law
Gender Violence and the Price of Virginity: Theory and Evidence of Incomplete Marriage Contracts
Do Markets Overcome Repugnance? Muslim Trade Response to Anti-Muhammad Cartoons
Economic Distress Stimulates Religious Fundamentalism
Teaching
Contracts and Commitments
Experimental Economics: Sources of Normativity
Toulouse School of Economics (G1)
Contract Law
Duke Law (1L)
Positive Foundations of Normative Commitments
Institutional and Organizational Economics Academy (PhD), Hebrew Law (3L)
Hermemetrics Lab: The Economics of Interpretation
Harvard Economics (So)
Theorizing Cultural Differences: The Economics of Fundamentalism
University of Chicago Economics (Sr)
Law and Economics Reading List
Law and Economics (theory)
Toulouse School of Economics (G), Duke Law, cross-listed Economics (3L/G1)
Heremans Lectures in Law & Economics: Legitimacy, Law, and Recognition-Respect
KU Leuven (JD)
Machine Learning, Causal Inference, and Judicial Analytics
Toulouse School of Economics (G1) and (G2)
Natural Language Understanding and Computational Semantics Syllabus
NYU Courant Institute of Mathematics Center for Data Science (G2) (project advisor)
Machine Learning and Computational Statistics Syllabus
NYU Courant Institute of Mathematics Center for Data Science (G1) (project advisor)
Empirical Methods and the Law
Summer School: Behavioral Economics and Experimental Research (PhD), ETH Zurich (G1), Duke Law (3L)
Decision Theory
Harvard Engineering Sciences (G1) (teaching fellow)
Data Science Justice Collaboratory
We are always looking for curious, dedicated people interested in getting involved with research at the postdoctoral, doctoral, or pre-doctoral level. See research statement for background on current research. Brief summary of ongoing projects analyzing 12 terabytes of curated archival and administrative data on judges and courts: Slides_1 Slides_2
Data
1. Link administrative Medicare data to industry-physician relationships cleaned from litigation settlements (a comprehensive dataset is available through the Affordable Care Act) to examine the impact of disclosure laws and the impact of pharmaceutical company payments to doctors on prescribing, patient outcomes, and patient adherence.
2. Automate advances in high-dimensional econometrics for causal effects of court precedent where judges are randomly assigned. Apply method in legal areas where we have already hand-coded data (sexual harassment, eminent domain, free speech, abortion, church-state separation, affirmative action, gay rights, disability rights, campaign finance, capital punishment, criminal appeals, desegregation, sex discrimination, punitive damages, federalism, National Labor Review Board, environmental protection, National Environmental Policy Act, Federal Communications Commission, Title VII, First Amendment, Eleventh Amendment, standing, contracts, and corporate veil piercing are among the 25 polarized legal areas) to study the channels through which legal regulations have their effects. Use high-frequency data to buttress assumptions of exogeneity and precedence. Examine whether law shapes values and conceptions of rights.
3. Develop oTree for running real-time experiments in lab, online, or via smart device in the field. Show, e.g., how presence of deontological commitments can problematize widely used experimental methods involving random lottery incentive and strategy method; theoretically and empirically probe foundations of polarization; test assumptions in private law and judicial decision-making. oTree is open-source and an innovation on zTree.
4. Digitize universe of US Circuit Court cases from 1880 to 2013 (roughly 380,000 cases), the identities of randomly assigned judges sitting on the panels and authoring the opinions, the dissents and concurrences, the judges' biographies, the hand-labeled legal topic, the citation network among the cases, and 2 billion N-grams of up to length eight. Link publicly available Supreme Court datasets, US District docket datasets, geocoded judge seats, biographies of judicial clerks, 5% random sample hand-labeled for hundreds of features including vote ideology, oral arguments, and administrative data from Administrative Office of the US Courts (date of key milestones, e.g., oral arguments, when was the last brief filed, etc.) for measuring priming of identity, peer effects, perfectionism, partisan ways of persuasion, judicial innovation, and the genealogy of ideologies and schools of thought in temporally and spatially disaggregated text. Use algorithmic mechanism design, LASSO, and automated text analysis for court precedent project and citation data for econometrics of networks.
5. Digitize World War I British archival datasets, including universe of deserters reported in military diaries, police gazettes, and handwritten military trials, commuted and executed capital sentences, geocoded casualties, maps, officer lists, and order of battle to examine the role of legitimacy in legal compliance, effect of the death penalty on British vs. Irish soldiers, and potential long-run impact of demographic violence.
6. Curate universe of administrative data on 1 million refugee asylum and 15 million hearing sessions and their time of day across 50 courthouses and 20 years (with randomly assigned judges), 1 million criminal sentencing decisions in US District Courts from 1992-2009 (with randomly assigned judges), and hand-collected biographical data to study gambler's fallacy, implicit egoism, habit formation, racial contrast, mood, extraneous factors, and time of day effects on judges' normative commitments.
7. Link universe of individuals in a district attorney's office over a decade with many stages of random assignment, linked to administrative data on wages, education, credit, among other life outcomes, and past and current addresses for survey follow-up to measure, e.g., name letter effects, perceived legitimacy of law using a tool like oTree, and the long-run effects of forced migration.
8. Digitize speech patterns in US Supreme Court oral arguments since 1955 - longitudinal data on speech intonation (linguistic turns) are rare. Link to oral advocates' biographies, faces, clipped identical introductory sentences, and ratings of their traits. Test labor market treatment of mutable characteristics and persuasion, and mimicry between lawyers and Justices and among Justices over time using high-dimensional econometrics.
9. Over 1000 legal databases tagged and linked including all federal (supreme, appellate, district, bankruptcy, tax, patent, trade, customs, claims, unpublished) and state (supreme, appellate, district, tax, chancery, family, labor, unpublished) court cases to the earliest available date (some as early as 1778). Types of databases include code, statutes, bills, regulations, bulletins and notices, commission decisions, Attorney General opinions, rulings, statements, opinion letters, bill tracking, workers' compensation decisions, municipal codes, physician discipline decisions, market conduct examinations, issuances, directives, public health reports, FTC, IRS, EEOC, Department of Labor, Department of Defense, EPA, SEC, Federal Reserve, contract appeals decisions, legislative service, manuals, etc.
10. Case records collected from 24 High Courts and 3000 subordinate courts in India with details on over 8.7 million case records and 67 million hearings. Study (1) impact of court functioning on economic growth and inequality, (2) impact of economics, political, or psychological factors on court outcomes, (3) impact of court decisions or precedents on individuals' outcomes, and (4) artificial intelligence applications. Interested graduate and postdoctoral applicants should email a current CV, sample publication or manuscript, short description of research interests (2 pages or less), and names of 3 references. Undergraduate applicants should have a strong mathematical or computational background, broad knowledge in statistical methods, and experience with large data sets. For example, see 2016, 2017, and 2018 machine learning projects. oTree applicants should have Django, Python, and web development experience. Applicants should also have leadership and organizational skills as the position may require playing a leadership role as well as supervising research assistants.