Search results

Showing 1 – 50 of 5,912 results.
Self-published

Adopt or Innovate: Understanding technological responses to cap-and-trade (ICPSR 111981)

Released/updated on: 2020-07-23
Time period: 2000-01-01--2012-01-01
One important motivation for creating cap-and-trade programs for carbon emissions is the expectation that they will stimulate much-needed low-carbon innovation. I construct a new panel of British firms to investigate this hypothesis, finding that the European carbon market has encouraged greater low-carbon patenting and R&D spending among regulated firms, without necessarily driving short-term reductions in carbon intensity of output. This stands in contrast to past cap-and-trade programs, which have primarily spurred adoption of existing pollution control technologies, with little effect on innovation. I discuss how to reconcile these contrasting findings and implications for the future of carbon markets.

Self-published

Analysis Code for "The Black-White Gap in Non-Cognitive Skills among Elementary School Children" (ICPSR 117301)

Released/updated on: 2020-12-16
Time period: 1998-09-01--2014-05-01
Using two nationally representative datasets, we find large differences between black and white children in teacher-reported measures of non-cognitive skills. We show that teacher reports understate true black-white skill gaps because of reference bias: teachers appear to rate children relative to others in the same school, and black students have lower-skilled classmates, on average, than do white students. We pursue three approaches to addressing these reference biases. Each approach nearly doubles the estimated black-white gaps in non-cognitive skills, to roughly 0.9 standard deviations in third grade.
Self-published

Behavioral Risk Factor Surveillance System (BRFSS) Annual Survey Data, 1999-2019 (ICPSR 146342)

Released/updated on: 2021-09-03
Time period: 1999-01-01--2019-01-01
In 1984, the Centers for Disease Control and Prevention (CDC) initiated the state-based Behavioral Risk Factor Surveillance System (BRFSS)--a cross-sectional telephone survey that state health departments conduct monthly over landline telephones and cellular telephones with a standardized questionnaire and technical and methodologic assistance from CDC. BRFSS is used to collect prevalence data among adult U.S. residents regarding their risk behaviors and preventive health practices that can affect their health status. Respondent data are forwarded to CDC to be aggregated for each state, returned with standard tabulations, and published at year's end by each state. In 2011, more than 500,000 interviews were conducted in the states, the District of Columbia, and participating U.S. territories and other geographic areas.

The files in this deposit were downloaded from the CDC website by Julia Dennett, Yale University, and Toby Chaiken, J-PAL North America, and archived by Travis Donahoe, Harvard University and Michael Darisse, Cornell University. Additional information edited by Michael Darisse and Lars Vilhuber, Cornell University and American Economic Association.


Self-published

Cell Phone Access and Election Fraud: Evidence from a Spatial Regression Discontinuity Design in Afghanistan (ICPSR 118467)

Released/updated on: 2021-03-16
Time period: 2009-01-01--2009-01-01
This paper examines the impact of cell phone access on election fraud in the context of the 2009 Afghan presidential election. I combine cell phone coverage maps with unique data on the location of polling centers to accurately pinpoint which centers were exposed to coverage during the election. Results from a spatial regression discontinuity design along the two-dimensional coverage boundary provide considerable evidence that access to cell phones deters corrupt behavior. Polling centers just inside coverage areas report a drop in the share of fraudulent votes of about 4 percentage points while the likelihood of a fraudulent station goes down by about 8 percentage points. Analyses of the effect of coverage on citizen participation in election monitoring, election-related insurgent violence, and the tribal composition of villages suggest that the observed declines in fraud are likely attributed to cell phone access strengthening social monitoring capacity. From a policy perspective, these results illustrate how a widespread technology, namely cell phones, can exert a positive externality on institutional development via corruption deterrence.
Self-published

Children and the US Social Safety Net: Balancing Disincentives for Adults and Benefits for Children (ICPSR 163181)

Released/updated on: 2022-05-02
Economic research on the safety net has evolved significantly over time, moving away from a near exclusive focus on the negative incentive effects of means-tested assistance on employment, earnings, marriage and fertility to include examination of the potential positive benefits of such programs to children. Initially, this research on benefits to children focused on short run impacts, but as we accumulated knowledge about skill production and better data became available, the research evolved further to include important long run economic outcomes such as employment, earnings and mortality. Once the positive long-run benefits to children are considered, many safety net programs are cost-effective. However, the current government practice of limiting the time horizon for cost-benefit calculations of major policy initiatives reduces the influence of the most current economic research on the long run benefits. We conclude with a discussion of why the rate of child poverty in the US is still higher than most OECD countries and how research on children and the safety net can better inform policy-making going forward. 
Self-published

CMS Doctors and Clinicians Data, 2014-2021 (ICPSR 149961)

Released/updated on: 2021-09-28
Time period: 2014-01-01--2021-01-01
This deposit contains the Centers for Medicare & Medicaid Services’ (CMS) official data on doctors, clinicians, and groups listed on Medicare Care Compare.

The files in this deposit were downloaded from the CMS website and archived by Michael Darisse, Cornell University. Additional information edited by Michael Darisse and Lars Vilhuber, Cornell University and American Economic Association.



Self-published

Code and Data: Artifical Intelligence, Algorithm Design and Pricing (ICPSR 159401)

Released/updated on: 2022-04-07
We calculate the time path of prices generated by algorithmic pricing games that differ in their learning protocols. Asynchronous learning occurs when the algorithm only learns about the return from the action it actually took. Synchronous learning occurs when the AI conducts counterfactuals to learn about the returns it would have earned had it taken an alternative action. In a simple market setting we show that synchronous updating can lead to competitive pricing, while asynchronous can lead to pricing close to monopoly levels. However, building simple economic reasoning into the asynchronous algorithms significantly modifies the prices it generates.
Self-published

Code and Data for: Bargaining under the Shadow of Uncertainty (ICPSR 192422)

Released/updated on: 2024-09-30
In bargaining environments with stochastic future surplus, failing to delay agreement can be inefficient when the expected future surplus is sufficiently high.
Theoretically, such inefficiencies never arise under unanimity rule but can under majority rule.
Using a laboratory experiment, we find support for these predictions, both when the unanimity rule is predicted to be more efficient and when there should be no difference between the two rules.
We also find large point prediction deviations under the majority rule. 
We show these deviations can be explained by higher-than-predicted egalitarian sharing and a lower risk of being excluded from future agreements.
Self-published

Code and Data for: Beyond Competitive Devaluations (ICPSR 111342)

Released/updated on: 2022-03-22
Time period: 1980-01-01--2007-01-01
We propose a new perspective on how monetary and exchange rate policies contribute to international competitiveness, by focusing on implications for a country’s comparative advantage. We develop a two-country New-Keynesian model allowing for sectoral differences in the production of tradables: while in one sector firms are perfectly competitive, in another sector firms produce differentiated goods under monopolistic competition and subject to nominal rigidities, hence their performance is more sensitive to macroeconomic uncertainty. We show that, by stabilizing inflation and the output gap, monetary policy can foster the competitiveness of these firms, encouraging investment and entry in the differentiated goods sector. 
Self-published

Code and Data for: "Danish Flexicurity: Rights and Duties" (ICPSR 174561)

Released/updated on: 2022-11-02
Time period: 2019-01-01--2019-12-12
Denmark is one of the richest countries in the world and achieves this in combination with low inequality, low unemployment, and high income security. This performance is often attributed to the Danish labor market model characterized by what has become known as flexicurity. This essay describes and evaluates Danish flexicurity. The Danish experience shows that flexicurity in itself, i.e., flexible hiring and firing rules for firms combined with high income security for workers, is insufficient for successful outcomes. The flexicurity policy also needs to include comprehensive active labor market programs (ALMPs) with compulsory participation for recipients of unemployment compensation. Denmark spends more on active labor market programs than any other OECD country. We review theory showing how ALMPs can mitigate adverse selection and moral hazard problems associated with high income security and review empirical evidence on the effectiveness of ALMPs from the ongoing Danish policy evaluation, which includes a systematic use of randomized experiments. We also discuss the aptness of flexicurity to meet challenges from globalization, automation, and immigration and the trade-offs that the United States (or other countries) would face in adopting a flexicurity policy.
Self-published

Code and Data for "Detecting Potential Overbilling in Medicare Reimbursements via Hours Worked: Comment" (ICPSR 119829)

Released/updated on: 2020-11-23
Time period: 2013-01-01--2013-12-31
Code and data for "Detecting Potential Overbilling in Medicare Reimbursement via Hours Worked: Comment"

Abstract: Fang and Gong (2017) develop a procedure to detect potential overbilling of Medicare by physicians. In their empirical analysis, they use aggregated claims data that can overstate the number of services performed due to features Medicare billing. In this comment, I show how auditors can use detailed claims level data to better target improper overbilling.
Self-published

Code and Data for: Distributional Effects of the Canada U.S. Free Trade Agreement (ICPSR 191601)

Released/updated on: 2023-10-09
Time period: 1984-01-01--2004-12-31
This study extends the analysis of Kovak and Morrow (2022), who study the labor market effects of the FTA by comparing career trajectories for otherwise similar workers whose initial industries subsequently faced different tariff cuts under the FTA. Here, we focus on distributional impacts by examining how the effects of tariff cuts on employment and earnings differed for workers with different initial income levels. Our findings suggest that the effects of the FTA on earnings inequality were small, and the point estimates imply a slight reduction in earnings inequality among workers employed in manufacturing prior to the FTA’s enactment.
Self-published

Code and Data for: Does Cash Bail Deter Misconduct? (ICPSR 159582)

Released/updated on: 2023-10-16
Time period: 2017-01-01--2018-01-01
Courts routinely use low cash bail as a financial incentive to ensure that released defendants appear in court and abstain from crime. This can create burdens for defendants with little empirical evidence on its efficacy. We exploit a prosecutor-driven reform that led to a sharp reduction in low cash bail and pretrial supervision, with no effect on pretrial detention, to test whether such incentive mechanisms succeed at their intended purpose. We find no evidence that financial collateral has a deterrent effect on failure-to-appear or pretrial crime. This paper also contributes to the literature on legal actor discretion, showing that non-binding reforms may have limited impact on jail populations.

Self-published

Code and Data for Employed in a SNAP? The Impact of Work Requirements on Program Participation and Labor Supply (ICPSR 148041)

Released/updated on: 2023-01-26
Time period: 2007-01-01--2015-01-01
Work requirements are common in U.S. safety net programs. Evidence remains limited, however, on the extent to which work requirements increase economic self-sufficiency or screen out vulnerable individuals. Using linked administrative data on food stamps (SNAP) and earnings with a regression discontinuity design, we find robust evidence that work requirements increase program exits by 23 percentage points (64 percent) among incumbent participants. Overall program participation among adults who are subject to work requirements is reduced by 53 percent. Homeless adults are disproportionately screened out. We find no effects on employment, and suggestive evidence of increased earnings in some specifications. 
Self-published

Code and Data for: "Is New Platform Work Different from Other Freelancing?" (ICPSR 120686)

Released/updated on: 2020-11-16
The rise of freelance work in the online platform economy (OPE) has received considerable media and policy attention in recent years, but freelance work is by no means a new phenomenon. In this paper, we draw on I.R.S. tax records to identify instances when workers begin doing online platform work versus other freelance/independent contractor “gig” work for firms.  We find gig work occurs around major reductions in outside income, and document usage over the lifecycle. Our results provide suggestive evidence on motivations for entering into each type of work.  
Self-published

Code and Data for: "Is There Too Little Antitrust Enforcement in the US Hospital Sector?" (ICPSR 197602)

Released/updated on: 2024-10-23
Time period: 2002-01-01--2020-01-01
This is a repository for the code to replicate "Is There Too Little Antitrust Enforcement in the US Hospital Sector?" as published in American Economic Review: Insights.

The abstract of the paper is:

From 2002 to 2020, there were over 1,000 mergers of US hospitals. During this period, the Federal Trade Commission (FTC) took enforcement actions against 13 transactions. However, we find that 20% of these mergers could have been predicted, using the FTC's standard screening tools, to meaningfully lessen competition. We then show that, from 2010 to 2015, predictably anticompetitive mergers resulted in price increases over 5%. We estimate that approximately half of predictably anticompetitive mergers had to be reported to the FTC per the Hart-Scott-Rodino Act. We conclude that there appears to be underenforcement of antitrust laws in the hospital sector.
Self-published

Code and Data for: Learning by Offending: How do Criminals Learn About Criminal Law? (ICPSR 192283)

Released/updated on: 2024-07-10
This project investigates how criminals learn about criminal law. It uses the case of a natural experiment in which sentences were drastically increased for a specific type of recidivism in France. In the short run, media coverage of the reform did not seem to affect criminal behavior. However, people who had first-hand experience of the law later committed significantly fewer targeted crimes but the same number of nontargeted crimes. This pattern is consistent with a behavioral reaction based on a better understanding of the reform. This learning effect does not spread easily. While codefendants also learned, other criminal peers and defendants in the courtroom at the same time, but for a different case, did not.
This deposit contains code, nomenclature and read me file to replicate the findings
Self-published

Code and Data for: Market Entry, Fighting Brands and Tacit Collusion (ICPSR 138921)

Released/updated on: 2021-10-19
Time period: 2011-01-01--2014-12-13
We study a major new entry in the French mobile telecommunications market, followed by the introduction of fighting brands by the three incumbents. Using an empirical oligopoly model, we find that the incumbents' fighting brand strategies are difficult to rationalize as unilateral best responses. Instead, their strategies are consistent with a breakdown of tacit semi-collusion: before entry, the incumbents could successfully coordinate on restricting product variety to avoid cannibalization; after entry, this outcome became harder to sustain because of increased business stealing incentives. Consumers gained considerably from the added variety and, to a lesser extent, from the incumbents' price responses.
Self-published

Code and Data for: Measuring the Impact of COVID-19 on Businesses and People: Lessons from the Census Bureau’s Experience (ICPSR 138041)

Released/updated on: 2021-05-10
We provide an overview of Census Bureau activities to enhance the consistency, timeliness, and relevance of our data products in response to the COVID-19 pandemic. We highlight new data products designed to provide timely and granular information on the pandemic’s impact: the Small Business Pulse Survey, weekly Business Formation Statistics, the Household Pulse Survey, and Community Resilience Estimates. We describe pandemic-related content introduced to existing surveys such as the Annual Business Survey and the Current Population Survey. We discuss adaptations to ensure the continuity and consistency of existing data products such as principal economic indicators and the American Community Survey. 



Self-published

Code and Data for: Measuring Upward Mobility (ICPSR 187483)

Released/updated on: 2023-12-06
We conceptualize and measure upward mobility over income or wealth. 
At the core of our exercise is the Growth Progressivity Axiom: transfers of instantaneous growth rates from relatively rich to poor individuals increases upward mobility. This axiom, along with mild auxiliary restrictions, identifies an ``upward mobility kernel" with a single free parameter, in which mobility is linear in individual growth rates, with geometrically declining weights on baseline incomes. We extend this kernel to trajectories over intervals. 
The analysis delivers an upward mobility index that does not rely on panel data. That significantly expands our analytical scope to data-poor settings.

Self-published

Code and Data for: Reducing Frictions in Healthcare Access: The ActionHealthNYC Experiment for Undocumented Immigrants (ICPSR 177861)

Released/updated on: 2023-08-17
Time period: 2014-01-01--2017-01-01
In 2016, New York City designed and implemented an intervention to reduce frictions in accessing safety-net care: randomly making initial primary care appointments for 2,428 undocumented immigrants. We leverage a novel survey-administrative data linkage to show that the program increased self-reported access to primary care, leading to 23% fewer emergency department (ED) visits. High-risk individuals' ED visits fell by 32% on average, driving the aggregate effect. Preventive care also increased among individuals visiting sponsored clinics.
Self-published

Code and Data for: Reducing Medical Spending of the Publicly Insured: The Case for a Cash-Out Option (ICPSR 115603)

Released/updated on: 2019-12-06
Individuals' medical spending has both necessary and discretionary components, which are not, however, separately observable. This paper studies ways to improve upon existing public health insurance policies by using a framework where both the discretionary and necessary components of medical spending are explicitly modeled. First, using a simple theoretical framework, the paper shows that the key to reducing discretionary medical spending is to introduce a trade-off between nonmedical and medical consumption. Next, using a rich quantitative life-cycle model, the paper shows that this trade-off can be successfully implemented by introducing an option to substitute public health insurance with cash transfers.
Self-published

Code and Data for: Report for 2023 by the AEA Data Editor (ICPSR 198444)

Released/updated on: 2024-06-25
Time period: 2022-12-01--2023-11-30
This repository contains the code, data, and manuscript files for the 2023 report by the AEA Data Editor. If you are reading this on openICPSR, then only code and data are present.

Self-published

Code and Data for: Report for 2024 by the AEA Data Editor (ICPSR 212781)

Released/updated on: 2025-12-03
This repository contains the code, data, and manuscript files for the 2024 report by the AEA Data Editor. If you are reading this on openICPSR, then only code and data are present.
Self-published

Code and Data for: Self-fulfilling Prophecies in the Transition to Clean Technology (ICPSR 238528)

Released/updated on: 2026-04-02
Time period: 2019-01-01--2060-01-01
Technological lock-in has been a standard explanation for the slow take-off of clean innovation, but is hard to reconcile with forward-looking investors who anticipate the eventual switch to clean technologies. We provide an alternative explanation: strategic investment complementarities shape innovation and self-fulfilling prophecies can lead to delayed low-carbon transition. We analyze a standard directed technical change model with clean and dirty inputs. We find that when the two are good substitutes, two stable steady states can co-exist, each allowing multiple transitional paths. Optimal low-carbon transition requires a Pigouvian tax rule combined with a coordination device; commitment to a Pigouvian tax trajectory cannot solve a coordination failure.
Self-published

Code and Data for: Speculative Fever: Investor Contagion in the Housing Bubble (ICPSR 120446)

Released/updated on: 2021-01-28
Time period: 2000-01-01--2007-01-01
Historical anecdotes of new investors being drawn into a booming asset market, only to suffer when the market turns, abound. While the role of investor contagion in asset bubbles has been explored extensively in the theoretical literature, causal empirical evidence on the topic is much rarer. This paper studies the recent boom and bust in the U.S. housing market and establishes that many novice investors entered the market as a direct result of observing investing activity of multiple forms in their own neighborhoods and that “infected” investors performed poorly relative to other investors along several dimensions.
Self-published

Code and Data for "The Long-Run Effects of Government Spending" by J. Antolin-Diaz and P. Surico (ICPSR 198830)

Released/updated on: 2025-06-06
Time period: 1890-01-01--2015-12-31
Military spending has large and persistent effects on aggregate output because it shifts the composition of public spending towards R&D. This boosts innovation and private investment in the medium-term, and increases productivity, GDP and consumption at longer horizons. Public R&D expenditure stimulates economic activities beyond the business-cycle even when it is not associated with war spending. In contrast, the effects of public investment are shorter-lived while public consumption has a modest impact at most horizons. We reach these conclusions using Bayesian Vector Auto Regressions (BVAR) with long lags and 125 years of U.S. quarterly data, including newly reconstructed time series of government spending broken down by its main categories since 1890.
Self-published

Code and Data for: The Value of Software (ICPSR 226085)

Released/updated on: 2025-08-27
Software is one of the most important assets that needs to be priced in the digital economy.  It has emerged as a disruptive technology,  with companies primarily valued for their software offerings growing from 2% to 13% of market share between 1996 and 2023. We document persistent anomalies in growth forecasts and stock returns for software companies, indicating significant deviations from rational expectations over multiple decades. Our findings are consistent with Bayesian investors gradually learning about software's growing importance, highlighting how markets can be very slow to discern fundamental shifts from transient shocks in noisy data.

Self-published

Code and Data for: Washington Consensus Reforms and Lessons for Economic Performance in Sub-Saharan Africa (ICPSR 140821)

Released/updated on: 2021-07-26
Over three decades after market-oriented structural reforms, termed "Washington consensus" policies, were first implemented, we revisit the evidence on policy adoption and the effects of these policies on socio-economic performance in sub-Saharan African countries. We focus on three key ubiquitous reform policies around privatization, fiscal discipline, and trade openness and document significant improvements in economic performance for reformers over the past two decades. Following initial declines in per capita economic growth over the 1980s and 1990s, reform adopters experienced notable increases in per capita real GDP growth in the post 2000 period. We complement aggregate analysis with four country case studies that highlight important lessons for effective reform. Notably, the ability to implement pro-poor policies alongside market oriented reforms played a central role in successful policy performance.
Self-published

Code and Data for: Weighted Linear Discrete Choice (ICPSR 214961)

Released/updated on: 2025-04-08
Data and Mathematica code for "Weighted Linear Discrete Choice"

Abstract: We introduce a new model of stochastic choice that assigns each choice option a utility, along with a salience parameter reflecting economic frictions. We characterize our model behaviorally and investigate its comparative statics properties. We show that the model generates intuitive closed-form solutions in equilibrium settings where firms can choose price, quality, and advertising. In addition, we show that the model allows for flexible substitution patterns and changes in market shares across choice sets. We demonstrate that our model can be easily identified and can outperform alternatives in demand prediction.
Self-published

Code and Data for: What Difference Does a Health Plan Make? Evidence from Random Plan Assignment in Medicaid (ICPSR 175881)

Released/updated on: 2024-02-06
Time period: 2008-01-01--2012-01-01
Exploiting the random assignment of Medicaid beneficiaries to managed care plans, we find substantial plan-specific spending effects despite plans having identical cost sharing. Enrollment in the lowest-spending plan generates about 30% lower spending—driven by differences in quantity— relative to enrollment in the highest-spending plan. Rather than reducing “wasteful” spending, lower-spending plans broadly reduce medical service provision—including the provision of low- cost, high-value care—and worsen beneficiary satisfaction and health. Consumer demand follows spending: a 10 percent increase in plan-specific spending is associated with a 40 percent increase in market share. These facts have implications for the government’s contracting problem and program cost growth.
Self-published

Code and Public Data for "Aggregate Nominal Wage Adjustments: New Evidence from Administrative Payroll Data" (ICPSR 120647)

Released/updated on: 2021-01-28
Time period: 2008-05-01--2016-12-01
Using administrative payroll data from the largest U.S. payroll processing company, we measure the extent of nominal wage rigidity in the United States.   The data allow us to define a worker's per-period base contract wage separately from other forms of compensation such as overtime premiums and bonuses.  We provide evidence that firms use base wages to cyclically adjust the marginal cost of their workers.  Nominal base wage declines are much rarer than previously thought with only 2% of job-stayers receiving a nominal base wage cut during a given year. Approximately 35% of workers receive no base wage change year over year. We document strong evidence of both time and state dependence in nominal base wage adjustments.   In addition, we provide evidence that the flexibility of new hire base wages is similar to that of existing workers. Collectively, our results can be used to discipline models of nominal wage rigidity.
Self-published

Code and Public Data for: Increasing Degree Attainment Among Low-Income Students: The Role of Intensive Advising and College Quality (ICPSR 227282)

Released/updated on: 2026-02-04
Time period: 2014-01-01--2021-01-01
A college degree offers a pathway to economic mobility for low-income students. Using a multi-site randomized controlled trial combined with administrative and survey data, we demonstrate that intensive advising during high school and college significantly increases bachelor's degree attainment among lower-income students. We leverage unique data on pre-advising college preferences and causal forest methods to show that these gains are primarily driven by improvements in initial enrollment quality. Our results suggest that strategies targeting college choice may be a more effective and efficient means of increasing degree attainment than those focused solely on affordability.
Self-published

Code and Public Data for "Targeting In-Kind Transfers through Market Design: A Revealed Preference Analysis of Public Housing Allocation" (ICPSR 131162)

Released/updated on: 2021-07-15
Time period: 2010-01-01--2014-12-31
Code and instructions to reproduce analysis in the article "Targeting In-Kind Transfers through Market Design: A Revealed Preference Analysis of Public Housing Allocation."

Abstract: "Public housing benefits are rationed through waitlists. Theoretical work on public housing allocation has debated how much choice applicants should have over units, identifying a possible trade-off between efficiency and redistribution. This paper empirically establishes the existence and economic importance of this trade-off using waitlist data from Cambridge, MA. I estimate a model of public housing preferences in a setting where heterogeneous apartments are rationed through waiting time. Eliminating choice would improve targeting but reduce tenant welfare by more than 30 percent. Such a change is only justified on targeting grounds by a strong social preference for redistribution."
Self-published

Code and Replication Information for: "Harnessing Deductions to Increase Tax Compliance and Formalization" (ICPSR 235602)

Released/updated on: 2026-03-24
Time period: 2006-01-01--2015-12-31
Code, Replication Information and Data for "Harnessing Deductions to Increase Tax Compliance and Formalization". 
Abstract: We evaluate a tax reform in Ecuador that introduced generous deductions from personal income taxes (PIT), encouraging consumers to request receipts. The reform addresses tax evasion by targeting small self-employed businesses that mainly sell goods or services not subject to VAT but often evade income taxes. Exploiting plausibly exogenous variation in receipt demand due to the distribution of taxpayers across regions and professions, we find significant increases in reported profits among self-employed businesses exposed to the reform. We document spillover effects on the VAT system. Our net-revenue impact analysis suggests that the additional tax payments outweigh the foregone tax revenue.


Self-published

Code and Results for: Breastfeeding and Child Development (ICPSR 135401)

Released/updated on: 2023-09-28
Time period: 2000-01-01--2008-01-01
We show that children who are born at or just before the weekend are less likely to be breastfed, owing to poorer breastfeeding support services in hospitals at weekends. We use this variation to estimate the effect of breastfeeding on children’s development in the first seven years of life, for a sample of births of low educated mothers. We find large effects of breastfeeding on children’s cognitive development, with much weaker evidence of effects on non-cognitive development, and no effects on health during the period of childhood we consider. Regarding mechanisms, we study how breastfeeding affects parental investments in the child and the quality of the mother-child relationship.
Self-published

Code and Supplementary Data for: Prediction Errors, Incarceration, & Violent Crime (ICPSR 230362)

Released/updated on: 2026-03-24
Time period: 1995-01-01--2019-12-01
These files contain the code and public data for "Prediction Errors, Incarceration, & Violent Crime," American Economic Journal: Economic Policy. The complete data includes confidential data that cannot be included in the public repository under our IRB.  
Self-published

Code and Unrestricted Data for "Diffusion of Reproductive Health Behavior through International Migration: Effects on Origin-Country Fertility" (ICPSR 213741)

Released/updated on: 2026-02-12
Time period: 1992-01-01--2013-01-01
International migrants may facilitate the transmission of ideas across countries. We examine the impact of migrant exposure to reproductive health policies on origin-country fertility in the Philippines. We exploit temporal variation in destination-country reproductive health policies combined with spatial variation across Philippine provinces in their migration intensity and historical composition of migrant destinations. Migrant exposure to more liberalized reproductive health policies reduces origin-community fertility. This reduction is driven by increased adoption of modern contraceptives. Visible policy changes, such as commercial advertising of contraceptives, lead to this change in behavior. Firmly established family planning values moderate the fertility response. This is the code accompanying the article, as well as the unrestricted data. 
Self-published

Code for Advisor Value-Added and Student Outcomes: Evidence from Randomly Assigned College Advisors (ICPSR 142321)

Released/updated on: 2022-10-20
Time period: 2003-01-01--2016-01-01
This paper provides the first causal evidence on the impact of college advisor quality on student outcomes. To do so, we exploit a unique setting where students are randomly assigned to faculty advisors during their first year of college. We estimate advisor valued-added (VA) based on students’ first-year course grades. We find that having a higher grade VA advisor reduces time to complete freshman year and increases four-year graduation rates by 2.5 percentage points. It also raises  high-ability students' likelihood of enrolling and graduating with a STEM degree by 4 percentage points. The magnitudes of our estimated effects are comparable to those from successful financial aid programs and proactive coaching interventions. We also show that non-grade measures of advisor VA predict student success. In particular, advisors who are effective at improving students’ persistence and major choice also boost other college outcomes. Our results indicate that allocating resources towards improving the quality of academic advising may play a key role in promoting college success.
Self-published

Code for: AI-Powered (Finance) Scholarship (ICPSR 240109)

Released/updated on: 2026-02-12
This paper describes a process for generating academic papers using large language models (LLMs) and demonstrates this process’ efficacy by producing hundreds of complete  papers on stock return predictability, a topic well-suited for our illustration. After mining over 30,000 potential return predictors from accounting data, we generate “template reports” for 95 signals passing rigorous criteria from the Novy-Marx and Velikov (2024) “Assaying Anomalies” protocol. These templates detail signal performance predicting returns using a wide array of tests and benchmark performance against more than 200 documented anomalies. Finally, for each template we use state-of-the-art LLMs to generate multiple complete versions of academic papers with distinct theoretical justifications for the observed return predictability, incorporating citations to  literature supporting their respective claims. This experiment illustrates AI’s potential for enhancing financial research efficiency, but also serves as a cautionary tale,  illustrating how it can be abused to industrialize HARKing (Hypothesizing After Results are Known).