Over the past decade, India’s BJP-led national government has undertaken various reforms to welfare programs, aiming to reduce the scope for local officials to manipulate beneficiary selection or divert funds. Nevertheless state and local governments retained considerable discretion over beneficiary selection and the local delivery of central government funded welfare benefits, and have maintained exclusive control over state-funded programs. We use longitudinal household survey data from the state of West Bengal to evaluate the effects of these reforms on pro-poor targeting of welfare programs, and political competition between the BJP and the TMC, the incumbent political party in the state. We find that welfare programs were better targeted toward poorer and socially disadvantaged households in the period 2014–2018 than in 2010–2013. However the welfare reforms do not explain why the BJP’s vote share increased, as its electoral gains from the new nationwide programs were outweighed by their losses, stemming from the increased support for the TMC from recipients of new state and pre-existing benefits.
Multinational firms can use internal debt to shift taxable income across jurisdictions. To limit such profit shifting and protect the corporate tax base, many countries use thin capitalization rules that restrict interest deductibility. These rules often include escape clauses that exempt firms below specified thresholds, but such thresholds may create incentives for avoidance within corporate groups. This paper evaluates Norway’s 2014 thin capitalization rule using difference-in-differences, bunching, and regression discontinuity designs. We find that exposed firms sharply reduce leverage after the reform, driven by lower intra-group borrowing and interest costs, with no detectable effects on firm performance. At the corporate-group level, leverage remains unchanged, indicating that firm-level responses primarily reflect intra-group debt reallocation. We document substantial bunching below the rule’s exemption thresholds, especially among newly created subsidiaries, and show that internal debt is disproportionately shifted to these bunching firms. Although firms eliminate most of their initial exposure to the rule itself, this happens largely through reduced internal lending, and we find suggestive evidence of higher taxable income in the medium run. Because we observe only the Norwegian parts of multinational groups, we cannot provide conclusive evidence on cross-border profit shifting. Overall, the findings show that firm-level escape clauses generate strategic avoidance responses within corporate groups, while also suggesting that the reform may have protected the Norwegian tax base.
This paper provides direct evidence that shared incarceration can facilitate the formation of post-release criminal partnerships. Using Norwegian administrative data linking prison spells to police case identifiers, we measure network ties through subsequent co-offending. We exploit time-based variation in exposure within facilities by comparing inmate pairs in the same facility who differ only in whether their spells overlapped. Overlap is associated with a 37% increase in five-year co-offending relative to comparable non-overlapping pairs, robust to extensive controls and timing-based restrictions. Effects are stronger in environments and for pairs where interaction is plausibly more intense, highlighting network formation as a plausible mechanism.
Can digital technologies improve the efficiency of social safety net programs? This paper examines the impact of digitizing payments in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) on participant outcomes. We hand-collect the rollout schedule of payment digitization and link it to WIC participation and birth outcomes across multiple datasets. We find that WIC participation increases and birth outcomes improve among mothers exposed to digitization. Suggestive evidence indicates that payment digitization raises WIC participation by increasing awareness of the program and reducing stigma at the point of purchase, and that higher participation may, in turn, improve birth outcomes by increasing health insurance enrollment. Our back-of-the-envelope calculations suggest that payment digitization is fiscally appealing, with a marginal value of public funds of 0.94. These findings can inform evaluations of further digitization in safety net programs, such as online redemption.
To address rising income inequality, governments typically rely on two levers: predistribution policies that implement inequality reduction before incomes are earned—e.g., minimum-wage laws—and redistribution policies that implement inequality reduction after incomes are earned—e.g., taxes and transfers. We explore whether support for reducing inequality depends on the choice of policy lever. Using incentivized experiments with the general population of the US and Sweden (N=2510), we study inequality reduction under predistribution and redistribution, focusing on whether inequality reduction is implemented before or after income is earned. We find that timing matters: subjects in both countries reduce inequality more under predistribution than redistribution. Furthermore, subjects in the predistribution treatment more often invoke equal effort to justify reducing inequality, while those in the redistribution treatment more often invoke adherence to the initial agreement to justify preserving it. Overall, our results imply that the choice between predistribution and redistribution can influence public support for reducing inequality.
I study redistribution between inactive, unemployed, and employed agents who are heterogeneous in their preferences and their productive skills in the labor market and at home. The government considers that redistribution across skills is desirable, but redistribution across preferences is not: if everyone has the same preferences, equality is optimal, but if everyone has the same skills, laissez-faire is optimal. In turn, this leads to a simple sufficient statistic to assess any reform: any extra dollar of transfer should go to the agent whose consumption is the furthest below what she would have consumed if everyone had average skills. In high-income countries, the gap between the lowest consumption and the average production is empirically much larger in the labor market than among inactives. Hence, introducing an inactivity benefit, or basic income, would neither be welfare-improving nor optimal for that social welfare function, regardless of labor supply elasticities.
Departmentalized instruction, in which teachers specialize in one or more core subjects and instruct multiple groups of students in a day, has become increasingly prominent in elementary schools. Departmentalization involves a potential trade-off: specialized assignments exploit teacher comparative advantage across subjects but introduce coordination costs across teams of teachers instructing common students. Using 8 years of data from Massachusetts and a difference-in-differences design, we estimate the effects of departmentalization on student achievement. Overall, we find that departmentalization has positive effects in English language arts (ELA) and science and either null or positive effects in math. Consistent with prior findings on teacher specialization, teachers are less effective when specializing in math and no more effective in ELA than when teaching self-contained classrooms. Rather, the improvements are driven by departmentalized schools assigning teachers to their stronger subjects.
In this paper, we study the effects of the Women’s Labor Subsidy (WLS), introduced in 2012, on women’s labor outcomes. We use detailed administrative records for more than 2.5 million women that allow us to analyze employment and income trajectories before, during, and after the implementation of the subsidy. Using a regression discontinuity design that exploits the fact that eligibility is contingent upon falling below a specified vulnerability score cutoff, we analyze how eligibility and being a program beneficiary affect labor outcomes. Our results indicate no significant effects of WLS eligibility on women’s employment, months worked, or income. We also find null effects on labor outcomes for program beneficiaries. The null effect is relatively consistent across different subgroups of eligible women.
Federal judges—who enjoy lifetime tenure and constitutionally protected salaries—represent an especially hard test case for incentive-based bureaucratic reform. I study the “six-month list,” a reform requiring U.S. federal courts to publicly identify judges with overdue matters. Using a regression discontinuity design and other methods, I find that matters most exposed are resolved approximately 14% faster than those least exposed, with larger effects among younger, non-white, and female judges. The speed gains come with trade-offs: upfront time savings are partially offset by downstream delays, and more-exposed matters are more likely to be reversed on appeal. A bunching analysis estimates aggregate time savings of approximately 4%, demonstrating that non-monetary levers can shift behavior even among highly insulated elite professionals.
Using Measure 114’s narrow passage in Oregon as a natural experiment, we study how new gun regulations affect firearm demand. Background checks, a proxy for demand, rose 13.9% in anticipation of the referendum and surged 157% immediately following the election. After judicial intervention halted the law’s enactment, demand returned near pre-election levels. Temporal displacement/harvesting does not explain the demand spike: after eighteen months, we still observe a substantial cumulative increase of 63,000 excess firearm-related background checks. Administrative data reveal significant within-state heterogeneity. This evidence underscores the paradoxical effect of gun-control policies, offering a cautionary lesson to policymakers.
We assess whether layering an ultra-poor graduation program on Ethiopia’s Productive Safety Net Programme (PSNP) protects women and their households from localized weather shocks. Using panel data from a large cluster-randomized trial matched with detailed weather records, we identify community-level exposure to dryness shocks. Women who are exposed to dryness shocks experience deteriorating nutritional status (in conjunction with declining household food security and assets) and a significant increase in the risk of intimate partner violence. However, access to the graduation program fully buffers women from any increase in intimate partner violence, and partially buffers women and their households against other adverse effects. Further analysis suggests that improved household savings likely drives the program’s protective impact.
People with mental illness are frequently discriminated against and stigmatized. We introduce a new measure of perceived depression stigma and examine the causal effect of perceived stigma on help-seeking in a sample of 1844 Americans suffering from depression. We document that a large majority of our participants overestimate the extent of stigma associated with depression. An informational intervention aimed at reducing perceived social stigma decreases the demand for psychotherapy. Merely prompting individuals to think about stigma has a similar effect. A mechanism experiment reveals that our interventions increase both participants’ optimism about future mental health and their readiness to rely on friends and family, thereby reducing the perceived need for therapy.
We use population-wide data from Finland to provide evidence on the impact of earnings disregard policies on part-time work during unemployment spells. The share of part-time workers among benefit recipients increased sharply from about 10% to nearly 18% over a few years after the implementation of earnings disregards in unem ployment benefits and housing allowances, which allowed individuals to earn up to 300 euros per month without reductions in their benefits. Using variation in the impact of the reforms on incentives between individuals eligible for different types of benefits, we estimate that the share of individuals working part-time during unemployment increased by 4.1-4.5 percentage points (23-32%) due to the implementation of earnings disregards, correspond ing to a part-time work participation elasticity with respect to the participation tax rate of approximately 1. We find no evidence that earnings disregards crowd out full-time employment, but instead find moderate positive longer-run employment effects.
International trade depends on the expectation that courts will protect commercial interests when disputes arise. With the growing number of international conflicts, it is crucial to know whether this expectation is met in practice. We examine the effect of the 2014 annexation of Crimea on the enforcement rate of foreign decisions by Russian and Ukrainian judges. We assemble novel data on the entire universe of Russian and Ukrainian court decisions concerning the enforcement of foreign decisions resolving cross-border commercial disputes. Using a difference-in-differences design, we find that the enforcement rate of Ukrainian (Russian) decisions in Russia (Ukraine) fell by 26% (33%) after the annexation. We show that courts increasingly justified refusals through discretionary legal grounds and undue-notification claims. Our findings provide the first causal evidence that international conflict affects both judicial decision-making and compliance with commercial treaties.
The current opioid epidemic is the worst drug epidemic the U.S. has ever faced. We examine the role of social connections in its spatial spread. Having many friendship links to counties with high opioid exposure positively correlates with overdose death rates. To address identification concerns, we exploit two quasi-natural experiments that led users to substitute illicit opioids for prescription drugs. Having more friends exposed to counties affected by these shocks leads to more overdose deaths. We also provide suggestive evidence that spillovers affect different stages of addiction, are influenced by both supply- and demand-side conditions, and are attenuated in counties with more cohesive local friendship networks. Our results show how social connections contribute to the spread of the opioid epidemic.
We study the impact of Information and Communication Technologies (ICT) adoption on female employment outcomes in Vietnam using rich firm- and individual-level data. Our identification exploits exogenous variation in ICT access driven by geographic connectivity conditions and a nationwide network upgrade in 2007. We find that ICT adoption increases the female share of employment within firms, primarily through within-firm reallocation in the formal sector. We further provide evidence consistent with a task-reallocation channel: ICT is associated with a shift in labor demand away from routine and manual tasks toward social, customer-based, and service-oriented activities, where women are more likely to have a comparative advantage.
