We study the impact of foreign tax planning strategies on the foreign and domestic economic activity of US multinational corporations (MNCs). We use confidential IRS tax data to identify firms that adopt complex tax planning strategies that target mismatches with Irish, Dutch, and Luxembourg tax law. We document that these MNCs routed a large and growing share of foreign earnings through these tax planning structures. We estimate effects of these tax planning strategies using a difference-in-differences framework. These MNCs engage in behavior related to profit shifting, increasing their foreign holdings of intangible assets and their cash held abroad significantly more than MNCs without these structures. Adopting MNCs also experience stark reductions in foreign effective tax rates. Finally, we estimate the effects of these structures on economic activity and find that their adoption increases domestic R\&D and payroll, and foreign capital accumulation.

This paper uses micro data from country-by-country reporting of more than 3 600 large multinational companies operating in 238 jurisdictions to analyze global profit shifting. These companies report 7% of their global profits in jurisdictions with effective average tax rates below 5%, but only 0.4% of their employees and 3% of their tangible assets are located there. We find that globally, these companies reduce their tax burden by EUR 53 billion (15% of their overall tax payments) by shifting profits to low-tax countries. 60% of the profit shifting is carried out by the10% largest multinational companies. We also investigate profit shifting channels and provide evidence suggesting that the location of IP and equity in low-tax countries as well as the provision of loans to entities in high-tax countries play a key role for tax planning.

Multinational firms (MNEs) often pay no tax in high-tax countries because they

shift a large fraction of their taxable income to tax havens. We build a model of tax policy and investment that incorporates unobserved heterogeneity in MNEs’ profit-shifting capability and different costs of setting up a tax minimization network. The model matches the distribution of taxable profit and investment in detailed UK tax returns data. We use the model to quantify the policy tradeoff between raising tax revenue by combating tax avoidance (via, for example, a Global Minimum Tax) and attracting investment. The results solve a longstanding puzzle in the existing profit-shifting literature: our model reconciles the differences between previous micro- and macro-level estimates of profit-shifting elasticities by accounting for extensive margin decisions (to report positive or no taxable profit in a jurisdiction). We test the model’s predictions using a quasi-experiment from Italy.

Some developing countries exhibit remarkable intergenerational persistence at the top of the earnings distribution. This persistence cannot be uniquely credited to differences in human capital. Social networks also play a role in determining wages. Using a unique data set, we show the intergenerational transmission of employers across parents and children in Chile. Children whose parents were in the bottom part of the income are most likely to work at the same company. However, higher income premiums are observed among children whose parents were in the top part of the income distribution. The correlations are significant even after controlling by county, high school, college, and surnames fixed effects.

We study the optimal one-shot tax reform in the standard incomplete markets model where households differ in their wealth, earnings, permanent labor skill, and age. The government can provide transfers by raising tax revenue and has several tax instruments at its disposal: a flat capital income tax, a flat consumption tax, and a non-linear labor income tax. The optimal fiscal policy funds a transfer that is nearly 50 percent of GDP through a combination of very high taxes on consumption and capital income. The labor tax schedule has a high average rate but is also moderately progressive. We find an identical outcome when policy is instead determined by majority voting. Finally, we offer suggestive empirical evidence that households' preferences for tax and redistribution are more strongly associated with political identity than economic status.

This paper investigates the role of Social Security reforms in explaining the increase in labor supply of older men across cohorts and evaluates the impacts of additional Social Security policy reforms. I develop and estimate a rich dynamic life-cycle model of labor supply, savings, and Social Security application that captures the key structure of the Social Security system, while taking into account uncertainties in health, survival, wages, and medical expenditures. The model accurately matches the observed life-cycle profiles of employment, hours worked, and savings by health for men in the 1930s birth cohort. It shows that Social Security reforms account for over 73% of the observed rises in labor supply by the 1950s cohort, with reforms to the Social Security earnings test being the most important. Additional policy experiments suggest that fully eliminating the earnings test and reducing retirement benefits by 23% would further increase older-age participation by 3.4 and 5.1 percent.

With an average replacement rate of roughly 40 percent, Social Security (SS) benefits provide an important source of retirement income for older Americans. However, the total size of lifetime benefits a household receives is a complex function of factors including age of benefit claim and labor supply decisions. The intricacies of the rules related to this program are likely to lead to many misinformed households who make decisions that do not maximize the size of the total lifetime benefits received. In this work we ask: What are the welfare costs of this misinformation pertaining to Social Security rules? We build a life-cycle model of consumption, savings, labor supply, and Social Security application decisions as well as heterogeneity in education, marital status and SS program knowledge. Using this model, we run counterfactual experiments to better inform public outreach policies.

This paper estimates the wage and employment effects of wage subsidies using a large 2015 national-level reform in France that provides additional financial support to poor working households. While the aim of this policy is to promote work, it can incidentally reduce wages in response to an increase in the labor supply. Using administrative data and a shift-share IV design leveraging variation in the exposure to the reform based on the socio-economic composition of the local working-age population, I show that labor markets exposed to an increase in wage subsidies experience an increase in the growth rate of the number of hours worked and a decrease in the growth rate of the average hourly wage. I find no significant effect on pre-tax labor earnings growth at the local labor market level. These effects suggest a pass-through of wage subsidies to wages equal to 37% on average.

Globally, the last decade has seen rapid growth in the prevalence of individuals earning income from digital platforms that mediate work for the solo self-employed---gig work. Policymakers suffer from a lack of evidence that quantifies the benefits of gig work and how prospective policies affect worker welfare. I use administrative data, which spans the UK’s food delivery market, to estimate worker surplus in this typical gig labor market. Evidence that workers learn about their own value of gig work over time, which a new survey corroborates, allows for the identification of the joint distribution of gig work valuations and outside options. Estimates imply a median monthly surplus for a gig worker equal to one-third of the median employee's monthly income, or £673. In terms of policy, attaching fixed benefits to gig work is likely to be self-defeating if platforms pass on the associated costs to workers through, for example, lower hourly earnings.

We estimate the causal effects of a pandemic-era wage subsidy program in Canada on job losses and business closures. Our estimates use administrative microdata and a regression discontinuity strategy to estimate the effects of marginal changes in the wage subsidy rate. The estimated net wage elasticity of employment was 0.11, implying a small aggregate employment effect of the pro- gram and an estimated fiscal cost per job saved of nearly $200,000 per year. Subsidy payments caused a small but persistent reduction in business closure rates during subsequent waves of the pandemic, and increased earnings of existing employees. In all, our results suggest the subsidies did little to preserve job matches, but played a greater role in the overall social insurance response to the pandemic.

People frequently share stories on social media with little additional context. I explore how this type of coarse messaging affects receivers’ beliefs about the veracity of stories forwarded to them. Using a set of experiments in India with ~800 pairs of real-life friends, I find that receivers over-interpret sharing as a sign of a story's veracity and discount other reasons for sharing. As a result, receivers’ beliefs in shared stories increase irrespective of the sharers’ belief in them, with false stories accruing greater credibility. By comparing receivers’ updating behavior on shared stories to their updates from (i) learning the sharer’s beliefs directly and (ii) computer-generated signals, I find several mechanisms contribute to this aggregate bias: receivers overestimate sharers’ ability to discern a story’s veracity; mispredict how sharers’ beliefs map to sharing decisions; and exhibit base-rate neglect and under-inference while updating, which give the least believed stories the greatest advantage.

Power, defined as the asymmetric control of valued resources, affects most human interactions. Yet there is little observational evidence on how power affects real-world behavior and resource allocation. We examine this question in the setting of the doctor-patient encounter. We exploit the quasi-exogenous assignment of 1.5 million patients to physicians in US military emergency departments, using their difference in military ranks to measure their power differential. We find that, keeping patient rank constant, “high-power” patients (i.e., who outrank their physician) receive greater effort and resources from physicians and have better outcomes than “low-power” patients. We also find that: (i) low-power patients suffer on days their physician cares for a high-power patient; (ii) doctor-patient concordance on race and sex is important to such power dynamics; and (iii) within-physician effort is higher for patients recently promoted than those about to be promoted. Thus, power confers nontrivial advantage to its possessor.

Psychotherapy has been shown to be an effective tool to overcome mental illness, yet a large fraction of the population suffering from mental illness do not seek psychotherapy. In this paper, we provide evidence on the role of misperceptions in impeding the take-up of online psychotherapy. Using a sample of 3000 individuals with depression we study the causal role of perceived social stigma and perceived effectiveness in shaping the demand for therapy. Lowering perceived social stigma leads to a reduction in the demand for therapy. Increasing the perceived effectiveness of therapy increases demand. Our design allows us to disentangle the relative roles of belief changes and attention in driving observed treatment effects. We uncover a primacy of beliefs that has implications on how to design policy.

Developing countries often use two taxes to collect revenue from firms, a turnover tax for small firms and a profit tax for large firms. In this paper, we investigate the optimal design of this tax system and what aspects of the environment influence its optimal structure.

We study a tax system where firms with outputs above a particular level (threshold) must pay the profit tax. In this setting, a set of firms will reduce their outputs to the threshold to instead pay the turnover tax, i.e., firms bunch at the threshold.

Introducing firms' tax avoidance activities in the form of cost over-reporting and under-reporting of output enable firms to manipulate their reported outputs to meet the threshold, rather than lowering their actual outputs. Thus, tax avoidance activities create additional bunching of firms at the threshold.

This paper derives necessary and sufficient conditions for allocations to be incentive compatible in multidimensional screening problems that satisfy a generalized single crossing property. We then devise a numerical method based on these results to solve multidimensional screening problems. Importantly, our numerical method can be applied to multidimensional screening problems for which existing approaches cannot be applied. We apply this method to several numerical examples in the context of multidimensional optimal taxation. In addition to illustrating how to apply our theoretical results and implement our numerical method, our simulations highlight the importance of bunching in optimal multidimensional taxation. Finally, we prove that bunching must occur in multidimensional optimal taxation problems when the social planner has sufficiently redistributive preferences.

When should tax policy be used to influence political donation behavior? In a model of electoral politics where campaign spending is financed by citizen donations, inequality of political influence favoring the "donor class" can arise. Adopting the normative stance that such inequality is undesirable (for either welfarist or non-welfarist reasons), I characterize the optimal nonlinear tax/subsidy schedule on political donations. Sufficient statistics for optimal policy include not only donation demand elasticities, but also the marginal efficacy of campaign spending, and the effect of taxes on the sensitivity of donation behavior to candidate policy platforms. Using numerical simulations, I provide proof-of-concept results showing that this framework can rationalize real world policies such as the nonlinear subsidy schedules present in Canada which feature generous marginal rates of subsidy on the first dollar of political donations, with a marginal rate of subsidy that declines in donation amount.

What is the optimal income tax and Social Security policy? We set up a rich quanti- tative model in which a Ramsey planner jointly chooses both programs by maximizing the welfare of currently alive and future generations. We find that optimal pension system generosity depends strongly on the chosen social welfare function. There exist reforms that deliver Pareto improvement across (but not within) generations.

Success in life increasingly depends on key skills that allow people to thrive in education, the labor market, and their interactions with others. In this paper, we emphasize creativity as a key skill that is essential to open-ended problem solving and resistant to automation. We use rich longitudinal data to study the relationship between people's creativity measured in childhood and their individual attributes and life outcomes. We find that childhood creativity predicts labor market and educational success: more creative individuals earn more during the course of their careers, work in higher occupational categories, and reach higher levels of educational attainment. Our analysis of attributes further suggests that creative individuals have a package of practical skills that allows them to thrive in work environments where learning from experience is important. We combine insights from our findings with evidence from psychology to propose creativity-improving interventions that could lead to substantial economic benefits.

Information campaigns have the potential to influence decisions, but how should these campaigns be evaluated in the presence of heterogeneous responses? For example, information on the returns to education can lead to the continuation or discontinuation of education. We propose a method that uses the distribution of prior beliefs to identify heterogeneous responses to treatment. The method is used to analyze an information provision RCT in Ibadan, Nigeria implemented at a critical juncture when adolescents had to decide to stop or continue schooling. The average effect of the intervention is a 3.8 percentage point increase in the high school dropout rate one year, from a baseline of 9.4 percent. Twice as many students changed their decisions due to the intervention, with 2 percent deciding to continue schooling. Our results help reconcile mixed evidence on the effectiveness of information RCTs in education and uncover mechanisms.

I study the hypothesis that household sorting around school facilities explains a significant share of school and residential segregation in the United States. The U.S. pays for most school facilities with local funds, determined largely by the local electorate's willingness to approve the issuance of municipal debt ("school bonds"). I evaluate a 2001 California reform that decreased the approval threshold for school bonds from two thirds of votes cast to 55 percent. This resulted in new local funding in communities where bond proposals had previously failed. Existing research shows that bond passage increases home values. I look for evidence that this is associated with changes in the demographics of enrolled children via a Tiebout mechanism. I find that the reform increased facilities outlays by up to $4,500 per student over 12 years. Increased investment reduced income segregation by up to 1.5 standard deviations, but had no effect on racial segregation.

This paper develops a model to explore whether a higher detection probability for offshore tax evaders—e.g. because of improved exchange of information between countries and/or digitalization of tax administrations—renders it optimal for governments to introduce a voluntary disclosure program (VDP) and, if so, under what terms. We find that if the VDP is unanticipated, it is optimal for a revenue-maximizing government to introduce a VDP with a low or even negative penalty. When anticipated, however, the VDP is neither incentive compatible nor optimal as it induces otherwise compliant taxpayers to evade tax. A VDP can then only be beneficial if tax evasion induces an external social cost beyond the direct revenue foregone, e.g., due to adverse effects on overall tax morale. We also find that governments should relax enforcement if the VDP itself provides more powerful incentives to come clean.

We use a rich administrative dataset on individual tax returns from 2008 to 2015 to

analyse the behavioural and distributive effects of a flat tax (FT) reform introduced in

2011 for residential property income in Italy. Using a difference-in-difference identification strategy, we address five research questions: (i) does the FT increase the probability of declaring a positive rental income to the tax authorities? (ii) does the FT increase the declared tax base? (iii) is the reduced tax burden shared with the tenant? (iv) does the FT affect the overall tax revenue? (v) who are the gainers of the policy? The estimated intention-to-treat effects suggest that the decrease of tax evasion is limited whereas tax burden reduction is large, it is not shared with tenants and it mostly benefits top-income taxpayers. Overall, top 1% of property owners reap about 20% of the overall lost tax revenues.

This paper analyzes the effects of a tax policy reform on high-income individuals' decision to transition from employment to unincorporated self-employment. We use the introduction of a flat tax for unincorporated business owners in Poland. The marginal tax rate for the highest-income business owners decreased from 40% to a flat rate of 19%, while the highest-income employees remained subject to a progressive tax schedule with a top rate of 40%. We find a 23% increase in the probability of high-income earners switching to self-employment after the reform. Among those who made the transition, we observe a rise in the proportion of long-term solo-self-employed individuals. These findings suggest that the tax reform increased the attractiveness of self-employment as an alternative to employment but did not necessarily increase entrepreneurship among those who switched to self-employment.

We examine the relation between administrative income data and survey reports for the self-employed and wage-earners from 2000 - 2015. The self-employed report 40 percent more wage and self-employment income in survey than in tax administrative records; this nets out differences between these two sources shared by wage-earners.

We provide causal evidence that less third-party reporting for the self-employed increases the self-employed gap by exploiting a well-known administrative data artifact -- one and two dependent self-employed respondents exhibit substantial bunching at their respective first EITC kinks in administrative records. In contrast, these two groups have no observable survey income differences.

We apply state-of-the-art methods of welfare evaluation to a computational OLG model of real-world complexity. The model was developed by Van Rymenant et al. (2022) to study the effects of the dramatic decline in U.S. federal estate taxes since 1980. The model features firms, a fiscal government, and a rich household sector of overlapping generations. It captures several important heterogeneities between households, different tax instruments, and several key decision variables from the debate regarding the effects of estate taxation.

Keeping the estate taxes at their much higher levels of 1977-1979 (compensated by a lump sum transfer, or by a lower capital income tax rate or a lower labor income tax rate), leads to higher welfare levels for the bottom 55% of households, and higher equality of opportunity. Strong welfare losses are incurred only by households just below the top of the welfare distribution: the (grand)children of the very wealthy.

Inequalities are often regarded as more acceptable when they reflect differences in effort, rather than in luck. Effort and luck are, however, often intertwined - elements of luck may decide if an individual can even exert effort. We study redistributive behavior when luck fully determines whether an agent works. Using survey experiments in general population samples in the United States and Sweden, we document how spectators, tasked with redistributing income between agents, grant working agents not only more earnings, but more implied utility, than non-workers. One reason is that pre-redistribution earnings are treated as a stronger reference point when agents work, even when the ability to do so is randomly determined.

Transfer receipt is voluntary and costly, generating "self-targeting" through selective take-up among the eligible. How does self-targeting select on need, and what are its policy implications? We show self-targeting is advantageous in eight U.S.\transfers: On average, recipients have lower consumption and lifetime incomes than eligible nonrecipients with similar current incomes. Due to self-targeting, these transfers provide 50 to 75 percent more to the consumption-poorest and lifetime-poorest than would automatic transfers that are distributionally equivalent by income. Self-targeting makes automatic transfers undesirable: We estimate the social benefits of self-targeting are approximately six cents per transfer dollar, generally exceeding the social costs of ordeals.

In this paper I use a labor market equilibrium model and a social welfare analysis for comparing between an employee-based EITC and an employer-based system. Since the labor supply of the low-wage workers is relatively rigid, firms enjoy market power that might be used in detriment of the policy-maker's goals when implementing an EITC system. While the employee-based system avoids coordination among firms, and thus reduces their market power, an employer-based system allows for automatic take-up at the initial stage of the program's implementation, which constitutes an advantage. These pros and cons in both systems shall be weighted one against the other; for that purpose, I run simulations using empirically plausible parameters taken from data of countries that implemented an EITC. The simulations show that in terms of social welfare the employee-based EITC tends to dominate the employer-based system.

After international organizations such as The Organisation for Economic Co-operation and Development (OECD) advocated for national governments to adopt accrual accounting, the number of countries that switched from traditional cash-based accounting to accrual accounting or amended former cash-based accounting is increasing. This change in the accounting system is referred to as the New Public Management (NPM). One of the core elements of NPM is enhancing budget transparency, efficiency, and accountability of decision making using business-like management tools such as the double-entry bookkeeping method. This study examines the impact of the “local public account revolution” on the efficiency of Japanese local governments using a stochastic frontier approach and panel data. We present evidence that preparing business-like financial statements may increase the efficiency of local governments.

This paper adds to the literature by examining fiscal competition and deficit financing by local governments in a developing country, China. We examine a unique revenue source in China, land-use premiums (a type of property tax), in a panel dataset consolidated at the prefectural level from 2006 to 2016. Our results indicate that fiscal competition in land-use premiums exists and is stronger among wealthier than poorer local governments, a result that supports the view of Cai and Treisman (2005) that competition among asymmetrically endowed regions can lead to less discipline. Moreover, we find higher local deficits are associated with lower land-use premiums, a result that suggests that the local government does not fill any fiscal gap with own revenues.

The key question of this paper is whether the conditions of financial distress recently experienced by Italian municipalities can be at least partially imputed but also to the inadequacy of financial resources some municipalities suffer compared to the needs of their populations and territories.

Starting from a multidimensional definition of financial distress, we investigate this issue by exploiting the new mechanism of equalization transfers recently implemented in Italy at municipal level which provides an exogenous measure of the resource gap possibly suffered by each municipality compared to their needs assessed on a standardized basis.

The estimation results show that the Italian municipalities suffering from a level of resources below what is necessary to provide public services at standard levels are, ceteris paribus, more prone to run into financial difficulties. By the same token, large cuts in central government transfers have a statistically significant effect on financial vulnerability at municipal level.

This paper studies tax policy interaction for both mobile and immobile tax bases using German municipality-level data. We exploit exogenous changes in local tax setting associated with state debt reduction programs in Hesse and Northrhine-Westfalia to learn about strategic interaction among local governments when setting their taxes. Our results suggest strong and significant tax policy responses both in corporate as well as in property tax rates. Our estimates imply response function gradients in the range of 0.3 to 0.7, depending on the type of tax and state. Policy spillovers from property tax rates remain very local, which is consistent with yardstick competition behavior.

This paper uses newly-linked administrative data to quantify the long-run taxpayer cost of undergraduate education for a full cohort of English university students. For each student, we estimate initial government outlay, lifetime student loan repayments, and the effect of going to university on lifetime tax payments. We find that the average long-run taxpayer cost of undergraduate education is substantially negative. On average, higher lifetime tax payments alone outweigh the initial government outlay even at a high assumed discount rate. However, this average obscures large heterogeneity by gender and subject studied. For nearly half of all students, the long-run taxpayer cost is positive, as student loan repayments and the effect on lifetime tax payments do not make up for initial outlays. The effect on lifetime tax payments drives the long-run taxpayer cost of undergraduate education even in a system funded almost entirely by income-contingent student loans.

Section 304 of the Credit Card Accountability, Responsibility, and Disclosure Act (2009) limited the marketing and sale of credit cards to college students, nudging them away from these high-rate products. While it reduced card use, we document the nudge raised student loan balances by 8.4%, 15% among the less affluent. To assess the benefits of this substitution, we design a survey that reveals prevalent sub-optimal financial decision-making among students tied to higher card debt. Model-based evidence demonstrates how these results imply the policy raised welfare. A complementary analysis indicates higher grade-point averages and on-time graduation rates resulting from the policy.

Several studies have investigated determinants of children’s skill development, but the results are inconsistent due to data limitations. This study conducted an original questionnaire survey and examined the effects of parental time on children’s cognitive and non-cognitive skills. The time spent by mothers helping children to study encouraged children to study longer, but the effect on test scores was small. However, study time spent with fathers had a much larger effect, with test scores increasing by 0.09–0.11 points for an additional 100 minutes per week. Parental time also promoted children’s non-cognitive skills.

We investigate how the gender gap in confidence affects the views that evaluators (e.g., employers) hold about men and women. If evaluators fail to account for the confidence gap, it may cause overly pessimistic views about women. Alternatively, if evaluators expect and account for the confidence gap, such a detrimental impact may be avoided. We find robust evidence for the former: even when the confidence gap is expected, evaluators fail to account for it. This “contagious” nature of the gap persists across many interventions and types of evaluators. Only a targeted intervention that facilitates Bayesian updating proves (somewhat) effective.

This paper studies the role of social connections in women's career advancement. We investigate whether access to a larger share of female peers in business school affects the gender gap in senior managerial positions. Merging administrative data from a top-10 US business school with public LinkedIn profiles, we first document that female MBAs are 24 percent less likely than male MBAs to enter senior management within 15 years of graduation. Next, we use the exogenous assignment of students into sections to show that a larger proportion of female MBA section peers increases the likelihood of entering senior management for women but not for men. This effect is driven by female-friendly firms, such as those with more generous maternity leave policies and greater work schedule flexibility. Qualitative interviews reveal that some of the mechanisms behind these results include emotional support, job referrals, and information transmission.

Prior studies find a significant inequality in time spent by men and women in heterosexual households on child-related tasks even when both parents work full time. This inequality has been linked to gender inequalities in a wide range of labor market outcomes, human capital accumulation, and economic mobility. We investigate an important potential source of this inequality: external demands for parental involvement. We pair a novel theoretical model with a large-scale field experiment and find that mothers are 1.4 times more likely to be contacted than fathers by their child’s school. We decompose this inequality into discrimination stemming from differential beliefs about parents' availability versus other factors to better inform policies aimed at closing gender gaps. We experimentally test several solutions and present evidence that gender norms are in part responsible for the inequality that we document.

This paper analyzes how different screening practices affect gender equity in hiring. I transform tens of millions of high-dimensional, unstructured records from Brazil's public sector into selection processes with detailed information on candidates, evaluators, screening tools, and scores. Exploiting a federal provision that required the use of more impartial hiring practices, I find that increasing screening impartiality improves women's evaluation scores, application rates, and probability of being hired. To understand which design choices reduce gender disparities, I combine variation in how job processes complied with the reform requirements with a model of hiring in which evaluator bias, tool bias, and screening precision jointly determine relative hiring outcomes by gender. Screening changes that limit discretion in existing hiring practices or add new impartial screening tools reduce the gender hiring gap, while policies that eliminate subjective screening tools are ineffective because the loss of screening precision outweighs the reduction in evaluator bias.

Tax designation has been a popular attribute of the tax plans in a rising number of countries, yet evidence of its effects on tax compliance remains scarce. We conduct an online experiment with 830 Czech taxpayers who are self-employed or standard employees. Our approach mimics the actual tax designation mechanism: it allows subjects to express their preferences for how a part of their taxes is used by redirecting some of the money to a non-governmental organization (NGO). We exogenously vary the presence of the tax designation mechanism, the possibility to choose the recipient NGO from a list, the tax rate, and the use of tax revenues. We find no consistent significant effects of the tax designation mechanism on overall compliance, though for employees, we do find a small effect on the probability of them being fully compliant.

We study the causal effects of household tax credits on the willingness to demand legally provided services using two survey experiments with 1.974 German homeowners. We vary the type of the tax credit, the refund rate, and whether the tax reduction is made salient. All tax credits increase the willingness to pay for offers with invoice. The effectiveness is significantly higher when two attractive features are combined: consumer-friendly implementation plus high rate or the reduction is made salient. About two thirds of respondents who would use the tax credit would have demanded an offer without invoice also without the subsidy.

This paper investigates the effect of audits on tax compliance by exploiting a threshold-based Value Added Tax (VAT) desk audit program in China and administrative tax return data. We estimate the causal effect of audits on firm behavior using a regression discontinuity design. Our results show a significant compliance effect of the audit, with firms reporting more taxable output the following year. Additionally, firms report more input credit, with more reported input at higher credit level. Despite the fact that we find no significant effect on average VAT remittance, larger taxpayers tend to remit less, which may be attributed to higher consulting fees. Mechanism analysis indicates that the compliance effect is primarily driven by paper trail transactions. Importantly, we find evidence that firms tend to purchase more counterfeit invoices after audits, which highlights the need for enhanced tax enforcement measures.

This paper uses laboratory experiments to investigate the effects of the taxpayer's trust in the tax authority on tax compliance. First, an investment game is used to measure the trust between subjects in the experiment. Then, we examine the tax compliance and auditing decisions under different trust and information environments in a tax experiment involving both of the tax payer and the tax authority. The experiment results show that information about trust induces significant and heterogeneous impacts on tax compliance: the information of tax authority’s trustworthiness increases the tax compliance of low income taxpayers while has an opposite effect on high income ones, and lead to the highest overall welfare. The information of the taxpayer's trust in the tax authority reduces the auditing probability in the early rounds, especially with a low fine rate.

This paper examines the behavior of bureaucrats, implementing India's public employment program, as their expectations of being audited change. Exploiting random assignment to audit timing, I find: the rate of deterrence (in financial misappropriation) is increasing in bureaucrats’ expectations; bureaucrats evade detection by adjusting the timing and type of misappropriated expenditure. I interpret these findings with a model to analyze how information communicated about audit risk should be designed. I estimate a sufficient statistic from the model to solve for the optimal signal and analyze counterfactuals. Fully-informative signals would have persuaded bureaucrats to misappropriate USD35m less compared to uninformative signals.

We show that bureaucrats can exploit discretion in hiring decisions to engage in forms of favoritism that hinder organizational performance. We do this in the context of the Mexican federal judiciary. The arrival of a judge at a circuit results in the hiring of 0.05 relatives to key staff positions within the following year on average, a figure which is probably a lower bound of the overall effect. Moreover, we find that related appointees are negatively selected in terms of educational level and that their hiring leads to a reduction in the court's productivity, which indicates that such hires are motivated by rent-seeking rather than by efficiency purposes. Importantly for personnel policy, nepotistic hires are concentrated among judges who have been sanctioned for administrative offenses, those assigned to courts located in their state of birth, and those in higher-ranking positions.

Developing economies are characterized by limited compliance with government regulation, such as taxation. Resources for enforcement are scarce and audit cases are often selected in a discretionary manner. We study whether the increasing availability of digitized data help improve audit targeting. Leveraging a field experiment at scale in Senegal, we compare tax audits selected by inspectors to audits selected by a risk-scoring algorithm. We find that inspector-selected audits are more likely to be conducted, to uncover tax evasion and to detect larger amounts of evasion. We show, however, that the tax administration invests less manpower in algorithm-selected cases, and that algorithm-selected audits may generate less corruption, based on survey results. In ongoing work, we attempt to unpack the algorithm’s (dis)functioning and the relevance of human capital in the audit selection and implementation process.

To what extent does government advertising influence media content and how does media content influence citizens’ political perceptions? The institutional features of Austria’s media market involving highly-discretionary media funding practices allow us to test for two-way favors between the government and the media. Linking comprehensive data on ministerial payments to daily newspapers and large-scale data on media content, we document that payments and coverage about the minister are positively correlated. The correlations persist even when examining within newspaper-minister pair variation, indicating that the effect goes beyond the ideological proximity between newspapers and politicians. We show that government advertising crowds-in positive reporting and increases slant. Finally, we combine the analysis of how advertising influences media content with an approach that estimates the causal effect of media content on citizens’ voting intentions. The results indicate that media content has a significant impact on citizens’ voting intentions.

This paper empirically analyses how online budget transparency affects the electoral performance of mayors running for re-election. The main testable hypotheses are that: (1) voters reward higher budget transparency with higher vote shares for the incumbent mayors; and (2) the eventual positive effects of pre-electoral increases in expenditures or reductions in taxes on the incumbent's vote share decrease as transparency increases. These hypotheses are tested by estimating fixed-effects models on a panel comprising up to 519 Croatian cities and municipalities in the local elections of 2017 and 2021. The empirical results provide clear support for both hypotheses. First, higher online local budget transparency is associated with larger vote shares. Second, increases in expenditures in the year before elections only lead to greater vote shares for the incumbent when the level of online local budget transparency is low or very low.

This article tests the expansionist austerity hypothesis for Brazil between January 2000 and December 2020 through Nonlinear Autoregressive Distributed Lags (NARDL) models. Among the results, a long-term relationship was verified between GDP, gross and net public debt, total, direct and indirect taxes, in addition to total, mandatory and discretionary expenses. With regard to the short-term dynamics, measured by the Error Correction Models, the shocks that affect the Brazilian economy dissipate, however, the adjustment dynamics is too slow for all the fiscal variables tested. It was also verified that the short-term adjustment on the expenditure side is slower in relation to taxes and debt. Finally, the analysis of dynamic multipliers reports that austerity policies in Brazil, undertaken on the expenditure, tax or debt side, are expansionist.

For over fifty years, the IMF and other international organizations have been trying to compare the tax performance of different countries. Over this time, both data availability and sophistication of statistical methods have improved considerably. However, these new tools cannot eliminate the fundamental problem of imperfect information as the observed tax outcome is a joint product of unobservable tax effort and tax capacity. This paper aims to serve as a refresher on the purposes and assumptions underlying this kind of analytical models and to reassess the best technical solutions out of currently available modeling approaches given improvements in the availability of longitudinal data and advances in dynamic panel analysis made over these fifty years. Using the Pooled Mean Group (PMG) estimator in this paper, we attempt to disentangle the long-run relationship between tax capacity and economic development from short-run adjustments in response to economic cycles and other transient shocks.

This paper examines the impact of tax shocks on economic activity in Japan. Following the narrative approach pioneered by David and Christina Romer, we construct a comprehensive dataset of 486 tax changes enacted in Japan, categorizing them into endogenous tax changes and exogenous tax changes. Our dataset complements existing datasets for the United States and the United Kingdom. The baseline results indicated that an one percent of an exogenous tax changes would lead to 2.66 percent drop of output over three years. Exogenous tax increases exert a significant adverse influence on output, with consumption as the main transmission channels. On the other hand, deficit-driven tax changes may be less costly than other tax hike. While it depresses residential investment, deficit-driven tax changes may be less costly on consumption than other tax changes and corporate investment consistently experiences a prolonged expansion.

This paper examines age-specific individual preferences for the legal retirement age. Based on a theoretical model we develop the hypothesis that retirees prefer a higher legal retirement age than workers, whereby just retired individuals prefer the highest retirement age. We corroborate the hypothesis empirically with a fuzzy regression discontinuity design and show that just retired individuals are indeed most in favor of an increasing retirement age. We conclude that in aging societies the political feasibility of raising the legal retirement age increases. This result is of political relevance especially with regard to the expected retirement wave of the baby boomer generation.

How do households use retirement savings accounts in retirement? The answer to this question is important for tax policy pertaining to retirement savings. I shed light on this question by studying how households respond to Required Minimum Distribution (RMD) regulations, which mandate withdrawals from retirement accounts upon reaching a specified age. Using data from the Health and Retirement Study and a regression discontinuity design, I estimate the causal effects of aging into RMD regulations. First, I establish the direct effects of RMDs in my setting and show a sharp increase in withdrawals from Individual Retirement Accounts (IRAs). Next, I provide new evidence on the indirect effects of RMDs and show a concurrent, discontinuous increase in inter vivos transfers. The results indicate that some households ultimately use IRAs to facilitate intergenerational gifts, holding wealth in the tax-advantaged accounts until required to take distributions and then passing resources to children.

Old-age poverty is a problem in many countries around the world. This is particularly true for women, as the gender gap in retirement income is large. Knowledge of these issues is therefore important, as is financial literacy. I conduct a large representative online survey of 3,000 respondents in the United States to examine respondents’ beliefs about the gender pension gap and whether information about the gender pension gap changes their fairness views, perceived drivers of the gap, and policy views. I find, that treated respondents are more likely to agree that the gender pension gap has an important direct effect on their own lives and that it is unfair. In addition, the unequal distribution of care work and gender differences in wages are seen as more important drivers, and support for policies such as targeted financial education increases significantly.

We investigate the causal relationship between language and people's attitudes towards private and public debt. We provide different sets of empirical results: descriptive evidence from Switzerland's language border, a randomized controlled trial in Austria, Germany, Switzerland, the Netherlands, Sweden, and three countries where English is an official language, and a natural language processing analysis of parliamentary speeches. The experimental component of our research design consists of the randomization of wording in questions about private and public debt. We exploit the fact that the commonly used word for ``debt" also means ``guilt" in German, Dutch, and Swedish to explore how respondents' attitudes towards private and public debt are affected if those negatively connotated words are substituted with neutral ones.

The results show that neutral words trigger higher approval for private and public debt in German-speaking countries. The study also provides evidence that language is strategically used in the political arena.

To address the global challenge of small business finance, China has launched a fintech program that offers unsecured loans to small firms, using tax credit and tax payment information from the tax bureau and digitalization technology. We find that this program effectively improves small firms' access to loans, increasing both the probability of obtaining a loan and the loan amount. It also leads to better firm performance, including higher revenue, return on assets, and survival rate. The success of the program depends on its general inclusiveness among creditworthy taxpayers, which allocates loans based on tax information, prioritizes new borrowers, and reduces the impact of traditional financial metrics and geographic distance from banks. The positive loan effects on firm performance increase with tax credit and payment and are often stronger among the new borrowers. Therefore, it may be social optimal to spare credit rationing on expensive private information for small firms.

In the past three decades, government support of higher education has nearly tripled in Australia. Most of the growth observed in its support of an income-contingent loan scheme that permits domestic students to defer payment of tuition until the students have left university and have started earning an income above a minimum threshold by the government. When this program was adopted (1989), it was applauded for its effort to introduce tuition-fees without negatively affecting low-income student enrollments. Early research suggested that the income-contingent repayment scheme did not negatively affect participation by low-income students. Internationally, research has suggested that the burden of student debt negatively affects earnings, occupational choices, and family decisions. A touted solution for addressing increasing debt burdens is through income-contingent loans. This paper analyses a long-standing program using an extensive panel data set on taxpayers in Australia to explore the impact of income-contingent loans on post-schooling outcomes.

This paper studies the unequal incidence of corporate taxes among workers and its consequences for macroeconomic outcomes. Using French employer-employee data and multiple reforms over the period 2009-2019, I find a positive firm-level elasticity of labor income with respect to the corporate income net-of-tax rate, which varies substantially across firms with different technologies. Using variation in exposure to national reforms at the local labor market level, I show that the corresponding macro-elasticity is larger in magnitude, suggesting an amplification mechanism in general equilibrium. I develop a model featuring firm heterogeneity and imperfect competition in the labor market, which connects these two facts by deriving a macroeconomic amplification from micro-level heterogeneity in tax incidence. I calibrate the model to match the empirical results, and evaluate the short vs. long run, and micro vs. macro impact of corporate tax reforms on income inequality and the share of corporate tax burden borne by workers.

This paper brings two strands of the literature together and studies how corporate taxes causally affect land, or specifically commercial rents and commercial property purchasing prices, by combining the specific system of business taxation in Germany with unique data on real estate prices.

Our analysis combines administrative panel data on the LBT rates of German municipalities with housing data from F+B. Our identifying variation consists of over 17,000 tax hikes between 2008 and 2018. The large number of tax changes allows us to control for potentially heterogeneous time trends across regions or industries. We use non-parametric event study designs to estimate how real estate prices develop after a change in the LBT. We find that real estate prices react to corporate taxation.

Simplifying tax policy comes with costs and benefits. This paper explores simplification options for the taxation of MNEs, an area where administrative and compliance costs of the current rules are large. Simplified approaches seek to reduce these costs by relying on an approximation of the true tax base, potentially distorting resource allocation. We examine the efficiency cost of transfer pricing simplification theoretically and empirically. Using a sample of 300,000 firms located in 22 countries, we estimate that common transfer pricing practices reduce efficiency between 0.25 and 2.2 percent of total factor productivity across sectors. Focusing on the manufacturing sector, we then observe that simplification more than doubles sectoral inefficiency on average. However, large differences exist, with moderate efficiency costs in several sectors.

We examine how venture capital incentives affect financing for start-ups. We use the introduction of a tax decution for angel and VC investment in China as a quasi-experiment to identify the impact of the policy and find that the tax incentive leads to more VC investment into eligible start-ups. Larger and more experienced VCs appear to be more responsive to the tax incentives. In contrast to recent evidence from the US by Denes et al. (2020), we find that the early-stage investor tax incentives encourage firm entry. We also find that VC investment into eligible start-ups crowds out that into non-eligible ones.

We revisit the classical result that labor bears approximately zero of the incidence of corporate taxes in a rich general equilibrium framework with endogenous choice of occupation, organizational form, investment, and financing structure. Financial frictions and the traditional tax advantage of debt over equity generate misallocation. Further increases in the corporate tax reallocate production from C corporations to pass-through businesses on the intensive and extensive margin. This exacerbates misallocation, reducing labor productivity and wages. Overall, we find that both labor and capital bear a large share of the corporate tax incidence, while entrepreneurs are net beneficiaries of the tax change.

We highlight an opportunity for improved measurement of a key data item in corporate tax research, a firm’s tax residence. Some countries define tax residence based on a firm’s location of incorporation, some on a firm’s location of headquarters, and some consider both locations. Because no firm-level tax residence database exists, studies typically apply a uniform assignment of either the location of incorporation, headquarters, or center of business activity. We use a novel algorithm that embeds the residency laws of 150 countries to accurately assign tax residence. We reassign the tax residence of a considerable fraction of firms relative to standard proxies, and provide evidence that reassignment significantly affects inferences. For instance, for cross-border mergers and acquisitions with a US acquiror, 16% of the deal value involves an acquiror that is reassigned. Moreover, reassigned firms are systematically different from other firms along several dimensions, including effective tax rates.

We study how national implementation shapes the success of global tax agreements. In an unprecedented act of global coordination, countries have recently implemented a global multilateral network of automatic information exchange (AEI). Two parties, the sending, and the receiving countries need to be able to process information to make AEI work. We build a database on country characteristics for the sending and the receiving countries. On the sender level, we build a dataset relying on newly published OECD statistics on AEI enforcement. Using micro-level data, we find a significant increase in bank transfers from tax havens to Norway from deposits owned through several layers of secrecy after the local introduction of AEI. We provide evidence of variation in responses depending on enforcement intensity in the sending countries. On the receiver level, we operationalize a new dataset from the OECD with approx. 100 variables on the capacity of the tax administration.

This paper examines the impact of profit shifting on firm growth using balance sheet data. Using firm balance sheet data, we find that multi-establishment domestic firms tend to be larger than their multinational counterparts with subsidiaries in tax havens. We attribute this to the prioritization of tax savings, where firms may locate new establishments in low-tax countries at the expense of domestic expansion. We develop a novel firm-dynamic model with multi-establishment to explain this empirical observation, where firms can expand by either growing existing establishments or opening new ones domestically or in low-tax foreign locations. Our model shows that tax planning results in fewer and smaller establishments for multinationals, leading to lower level of output and employment. Our

findings suggest that when firms put minimizing their tax bill as their objective, they

may forgo a higher level of output and employment

Uganda has one of the lowest corporate income tax collection rates in sub-Saharan Africa while offering generous tax incentives and introducing new ones to attract investment and stimulate economic growth. The study uses administrative tax data and a new dataset on Ugandan tax incentives to evaluate the impact of these incentives on firm profits and employment. The initial findings suggest that firms benefiting from tax holidays experience increased profit before tax, but this should not be affected by the incentive, and there is no clear evidence of an employment effect in the country. The study highlights the importance of evaluating the impact of tax incentives and the role of administrative tax data in evidence-based policymaking in developing countries. The results provide guidance for Ugandan tax policymakers on the effectiveness of specific tax incentives.

This paper uses schooling decisions to estimate curvature in utility of consumption in developing countries. We show that the elasticity of marginal utility of consumption (EMUC) can be recovered from three sufficient statistics: (1) the price effect of schooling; (2) the income effect of schooling; and (3) the consumption cost of sending a child to school. We estimate the consumption cost of schooling using variation from the randomized roll-out of Progresa, a conditional cash transfer program in rural Mexico. Using estimates of income and price effects in the Progresa setting from Bergstrom and Dodds (2021), we estimate the EMUC at 1.6.

My job market paper shows that monopoly matters for investment in institutions. The next question is why. This paper presents evidence for an answer that monopoly is the de facto institution protecting private property rights absent formal ones. Unlike in capitalist economies, high-skilled workers in communist economies prefer the state sector even without wage premiums. The Chinese data from 1992 to 2006 shows that high-skilled workers work for the state sector much less for wage differentials (even not at all for high-skilled managers) than the low-skilled workers but more for rent differentials, measured by the asset per employee ratio in the state to the nonstate sector, which can be as high as 26.6 percent for high-skilled managers. Higher-skilled workers join the state sector for better positions with richer monopoly rents and higher capacity protecting them from being taken away.

While behavioral non-price interventions (“nudges”) have grown from academic curiosity to a bona fide policy tool, their relative economic efficiency remains under-researched. We develop a unified framework to estimate welfare effects of both nudges and taxes. We showcase our approach by creating a database of more than 300 carefully hand-coded point estimates of non-price and price interventions in the markets for cigarettes, influenza vaccinations, and household energy. While nudges are effective in changing behavior in all three markets, they are not necessarily the most efficient policy. We find that nudges are more efficient in the market for cigarettes, while taxes are more efficient in the energy market. For influenza vaccinations, optimal subsidies likely outperform nudges. Importantly, two key factors govern the difference in results across markets: i) an elasticity-weighted standard deviation of the behavioral bias, and ii) the magnitude of the average externality.

When faced with higher managerial taxes, mutual fund managers who personally invest in the funds they manage take on greater risk. By exploiting the enactment of the American Taxpayer Relief Act 2012 as an exogenous tax shock, we observe that co-investing fund managers increase risk-taking by 8%. Specifically, these managers adjust their portfolios by investing in stocks with higher beta. The observed effect appears to be driven by agency incentives, particularly for

funds with a more convex flow-performance relationship and for managers who have underperformed compared to their peers in the past two years. Such taxinduced behavior is associated with negative fund performance. We highlight the role of co-investment in transmitting managerial tax shocks to mutual funds.

Personal autonomy has been argued to be fundamental to a person's well-being and is often discussed as an important driver of economic and political behavior. Yet, the empirical foundations and relevance of preferences for autonomy are not well understood, in part because they are difficult to measure. We propose an incentivized elicitation method that allows to measure intrinsic preferences for autonomy. Our method solves the main identification challenge by controlling for any instrumental value that may be attached to autonomous choice. We establish the existence of intrinsic preferences for autonomy in two large online samples and show substantial heterogeneity in this preference. We further assess the precision and quality of our measure and study the antecedents of intrinsic preferences for autonomy by relating it to existing personality scales as well as socioeconomic characteristics. Finally, we test the association of our preference measure with other preferences, attitudes and beliefs.

This study investigates how labor market conditions at graduation affect individual’s labor market outcomes when facing employment shocks in later career, specifically due to plant closures. Our findings reveal that university graduates who entered the labor market during periods of high regional unemployment experience larger declines in employment probabilities and are more likely to work in smaller firms after facing a plant closure event. Furthermore, we find that vocational school graduates with higher regional unemployment rates at graduation have better income trajectories following a plant closure, despite being employed in firms with lower productivity and wages per worker. We also report how labor market entry conditions affect labor market outcomes in the early career. These results highlight the lasting impact of labor market entry conditions on employment, income, and employer quality, which may influence individuals’ response to future shocks in their careers.

Compared to other industrialized countries, Germany still has a relatively low female labor supply. The income taxation scheme is often seen as an obstacle to increasing labor supply of married women. Whereas a transition from the joint income taxation to an individual taxation scheme is difficult due to constitutional reasons, there is scope for reforms of the withholding tax on employment income. We quantify possible labor supply effects, using a structural discrete choice labor supply model. We investigate how different assumptions about couples’ decisionmaking influence estimated reform effects. The standard approach, assuming a joint utility function that is maximized with respect to household income, results in minimal labor supply reactions and likely underestimates potential effects. We test alternative assumptions about the role of individual incomes in the utility function and show that increases are limited to about three percent of hours worked even in the most optimistic scenario.

Unemployment insurance (UI) eligibility requires a claimant to have lost their job through no fault of their own. Approximately 10% of claims are deemed ineligible solely based on the job separation reason. Using the systematic variation in separation-based eligibility approval rates across UI claim processing offices and examiners in California from 2002 to 2019, we show that receiving any UI benefits causes approximately 2 additional weeks of nonemployment. By replicating existing research designs for other UI policy margins within the California data, we conclude the efficiency costs of UI benefit expansions through separation-based eligibility criteria are lower compared to those of expansions through monetary eligibility, weekly benefit amount, or potential benefit duration.

Meritocracy implies paying people according to their performance. It thereby comprises two principles: (i) paying individuals with higher performance more (fair inequality) and (ii) paying people with equal performance equally (fair equality). While the literature has mainly focused on the fair inequality dimension, we lack evidence on the importance that people attach to fair equality. We empirically investigate individuals’ fairness preferences regarding both equality dimensions. In this paper, we for the first time (i) document the existence of preferences for fair equality and fair inequality, (ii) assess the relative importance of both equality concerns and (iii) assess individuals’ willingness to incur a cost for establishing fair equality and fair inequality.

Generalised trust is essential for collective action, which is at the heart of many societal problems. Institutional quality has been proposed as a determinant of generalised trust, though while the correlation between the two is strong and robust, the evidence on the causal link is scant. We show that this relationship is causal. We first experimentally expose individuals to institutions of different quality, operationalised as their ability to prevent corrupt behaviour. We then measure generalised trust using a trust game. The results show that institutional quality drives generalised trust and that this effect is generated by the mere doubt that corrupt behaviours might succeed, even without knowledge of occurrence or success of such behaviours. Cross-country comparisons with novel data support our results. Our contributions are the first causal experimental evidence on the link between institutional quality and trust and a novel experimental design for modelling institutional quality in laboratory settings.

We conduct three field experiments to understand the drivers and consequences of polarization for US state spending on public higher education. The first two experiments study polarization in taxpayer demand. We provide different types of information to shape beliefs about the social return on investment. Our treatments decrease the political partisan gap in ideal policies by up to 32\%, with differences in partisan reasoning from information as a key mediator. Information also affects how taxpayers communicate their ideal policies with elected officials. Specifically, information leads taxpayers to be more likely to write a (positive) letter to an official of the other party--- a reduction in affective polarization. In the third experiment, we send the taxpayers’ letters to a randomized subset of elected officials to study the supply of policies. Officials who receive their constituents’ demands engage more with higher education issues in our correspondences.

What do markets for voluntary climate protection imply about people’s valuations of environmental protection? I study this question in a large-scale field experiment (N=255,000) with a delivery service, where customers are offered carbon offsets that compensate for emissions. To estimate demand for carbon mitigation, I randomize whether the delivery service subsidizes the price of the offset or matches the offset’s impact on carbon mitigation. I find that consumers are price-elastic but fully impact-inelastic. This would imply that consumers buy offsets but their willingness to pay (WTP) for the carbon it mitigates is zero. However, I show that consumers can be made sensitive to impact through a simple information treatment that increases the salience of subsidies and matches. Salient information increases average WTP for carbon mitigation from zero to 16 EUR/tCO2. Two complementary surveys reveal that consumers have a limited comprehension of the carbon-mitigating attribute of offsets.

Decarbonization is a significant transition risk for communities historically built around the fossil fuel industry. This paper uses a current example of a regional decarbonization, the decline in the Appalachian coal industry between 2011 and 2018, to understand how individuals are harmed by a decline in local fossil fuel extraction. We use individual-level credit data and exogenous variation in coal demand to identify how the coal mining industry’s decline affected household finance outcomes. We find the coal decline caused broad-based negative impacts, decreasing credit scores and increasing credit utilization, amounts in third party collections, bankruptcy rates, and the number of individuals with subprime status. The drop in credit score was most pronounced between the 20th and 40th percentiles. Our results show that individuals living in fossil fuel extraction regions are likely to experience personal financial costs from decarbonization even if they do not directly work in the affected industry.

One non-price mechanisms that can help encourage water conservation is a water audit. Yet, little is known about their efficacy, efficiency, and cost-effectiveness. This paper fills the research gap by implementing a natural field experiment in the UK. Our experiment involves randomly allocating 45,000 customers to a control group or to groups that are provided with different encouragements to take-up an online water audit. Our analysis yields three findings. First, providing participants with financial incentives to participate in the audit significantly increases audit take-up. Using the positive encouragements, we find that the water audit reduces water consumption by 17% for two months. Second, incentivizing uptake of the audit does not pass a benefit-cost test. We also implement a marginal value of public funds approach to considering benefit and costs, and reach a similar conclusion. Third, we find that targeting of high users doubles the effectiveness of the financial incentive interventions.

Many governments aim to give disadvantaged firms an equal opportunity to compete for government contracts and will use diversity in awards as a measure of success. I show theoretically that, when ex ante identical disadvantaged firms differ in an identifiable but irrelevant characteristic, diversity in awards may not translate into equity in opportunity—as buyers may discriminate within the disadvantaged group. Subcontracting data on Disadvantaged Business Enterprises in New Mexico show that inequities can arise in practice.

Despite being set-asides widely used in procurement, their effects on competition and efficiency are a priori ambiguous. We pool a decade of US federal procurement data to study this empirical question. To circumvent the lack of exogenous variation in our data, we first use a random forest to calculate the likelihood of a tender being set aside. We then estimate the effect of restricted tenders on pre- and post-award outcomes using an inverse probability weighting regression adjustment. We find that set-asides prompt more firms to bid: the increase in targeted bidders more than offsets the loss of untargeted. During the execution phase, set-aside contracts incur higher cost overruns and delays. The more restrictive the set-aside, the stronger these effects. In a subset of our data, we leverage an unexpected spike in set-aside spending to study implications for firm dynamics. We do not find evidence of increasing recipients' performance over time.

Veteran-owned businesses are given preferential treatment in the allocation of procurement contracts from the U.S. Department of Veterans Affairs, which awards over $30 billion in contracts annually. We exploit a 2016 Supreme Court ruling that significantly increased the scope of these set-asides, to study the impacts of preference programs on both the targeted businesses and public procurement. The policy change increased the share of contracts awarded to the target population, service-disabled veteran-owned small businesses. While awards increased most for existing vendors, the policy also led to significant entry of new vendors, including those who had previously failed to win contracts. We also show that the policy led to an increase in survival for targeted firms. We find no evidence of relevant spillovers to awards by other federal agencies, nor of a decline in contracting execution performance by vendors.

This paper examines the fiscal and distributional consequences of introducing a means test in the U.S. Social Security program using a calibrated heterogeneous-agent overlapping-generations model. I find that a means test, i.e. conditioning benefit payments on a household's assets and/or earnings, leads to a higher implicit tax on old-age resources, but has desirable distributional effects. Cutting benefits to households with more than 200% of median income by 50% leads to only a 1% percent decline in Social Security benefits per retiree, but an almost 5% reduction in the overall size of the program. This means test also brings about a leftward shift in the earnings and benefits distributions. The effect of this policy change on capital and labor are minimal, as wealthier households respond to the means test by working and saving more in early life.

This paper studies redistributive transfers between inactive, unemployed and employed agents. In a model with fully heterogeneous preferences and arbitrarily unequal skills, labor market inactivity arises from home production and disutility of participation. The social objective champions the ethics of equality of opportunity while upholding the Pareto principle. In the Mirrleesian second-best, welfare analysis is reduced to a sufficient statistic. Its empirical application suggests that an inactivity benefit would not be welfare-improving in most high-income countries. Overall, the equity gains of introducing a basic income with respect to equality of opportunity are tenuous, whatever its efficiency costs.

Employers in low-unemployment risk industries rarely lay off workers but regularly pay unemployment taxes, while employers in high-risk industries regularly lay off workers but only pay a portion of the cost of unemployment benefits in taxes. High-risk industries are thus subsidized by low-risk ones and face lower labor costs. This paper investigates whether unemployment insurance subsidies enable high- risk industries to expand their employment and evaluates the implications for welfare. Using data from two US states and a differences-in-differences approach, I estimate a larger elasticity of labor demand in high-risk industries compared to low-risk industries, which is consistent with higher unemployment taxes reducing the employment share of high-risk industries. Since a higher employment share in high-risk industries increases the benefit claims that governments need to finance through general taxation, the eleasticity of this share with respect to the unemployment tax represents a fiscal cost on the government budget.

Severe consumer opposition toward a broad tax base in Japan has made it difficult to increase value-added tax (VAT). This study empirically examines how Japanese VAT is shifted, using a database of purchasing records of customers. The empirical results show that a comprehensive VAT increase brings about a full shift. The introduction of the reduced rate has been highly effective in reducing the VAT burden and prices of foods and beverages. While the government and consumers have been successful in earning tax revenue and evading tax burdens, respectively, the firm sector has suffered serious losses in part. This study also examines the effect of another policy device that deregulates the price adjustment in advance. The results show that firms level off prices before and after the VAT increase by increasing the pre-tax price beforehand. These findings call for sophisticated policy tools facilitating the pass-through.

Revenue Authorities in the Global South have recently adopted new policy tools to improve domestic revenue mobilization through taxes. One such tool is a withholding system for value added tax. In this study, we investigate the impact of adopting a VAT withholding (WVAT) system on VAT collection in Zambia. While similar systems exist in many countries, empirical research into their impact is still limited and inconclusive. Using a difference-in-differences approach, we estimate how introducing such a system in Zambia affects the value added, sales, purchases, and output VAT of affected firms. We observe significant positive impacts after the reform on those indicators that the withholding agent reports on and remits to the tax authority. We find that the reform improves compliance by introducing a change in the tax remitting liability in the transaction chain.

In 2020, Germany reduced the VAT rate on menstrual products by twelve percentage points. We estimate the pass-through of the VAT rate cut to prices of products affected by the tax reduction and identify pricing spillovers on complementary products for which the VAT rate remained unchanged. For this purpose, we use a unique data set of daily prices of almost 5,000 hygiene products from the online shop of a large German supermarket chain. We find a full and immediate pass-through of the VAT cut to prices of menstrual products. Prices of complementary feminine hygiene products, which were not subject to the VAT change but are offered by producers of menstrual products, such as panty liners, increased after the reform (by 5 to 10 percentage points) with no sign of phasing out. The results are robust to various alternations of the control group, taking into account potential general equilibrium price effects.

We use a cluster randomized field experiment to study the effect of performance monitoring on the efficiency of public procurement in Chile. In collaboration with the Chilean Public Procurement Office, we provided monthly reports on the purchasing performance of individual procurement officers and services to a sample of 8,300 procurement officers in 184 public services. We randomly varied whether the individual performance was disclosed to managers (public) or not (private). After 5 months of treatment exposure, we find that the reports generated significant reductions in overspending in the public treatment group but not in the private treatment group, with the mechanism being the extrinsic motivation of procurement officers. We further find that the treatment effect is mediated by the extent of alignment that managers and officers have regarding the value (utility) of efficiency, suggesting that organizational culture plays a key role in explaining the impact of performance monitoring.

Governments play a crucial role in regulating the behavior of both citizens and firms to ensure the provision of public goods and services. Public procurement can foster economic activity, but it also presents moral hazard problems. To combat this, governments impose buying rules, but their effects still need to be well-determined. In our paper, we analyze the impact of changes in procurement regulation on buying manipulation and the mechanisms behind it in a middle-income country between 2002-2021. We document a relevant manipulation in the simpler procedure to the detriment of the stricter ones. We quantify an excess of 3,357 purchases and the relative excess of mass of 0.04 by the effect of having a threshold. As the thresholds change, we find a higher response when the government restricts the rules with a smooth counterfactual. However, if we construct the counterfactual with the previous threshold, we find symmetric results.

This paper investigates the effects of network externality on two countries’ policy competition in attracting a multinational enterprise (MNE). The two countries with a different number of consumers (“friends” of the MNE) endogenously set a tax/subsidy, and there is a local firm (an“enemy” of the MNE) in the country with a large market size. The larger country with the local firm becomes more attractive for the MNE because the resulting larger supply amplifies network externality. The MNE’s location in the larger country can benefit consumers in all countries, and it can also benefit the local firm despite the fiercer competition with the MNE. Fiscal competition increases the likelihood of the smaller country hosts the MNE when the network externality is small, but it increases the likelihood of the larger country hosts the MNE when the network externality is large.

We study how local governments’ debt financing decisions relate to the term structure of interest rates. Using a novel database of loans taken out by local governments in England, we find that the yield spread is the main determinant of the local governments’ adjustable-rate loan share. Our findings demonstrate that these public sector institutions minimize their short-term costs at the expense of long-term public welfare gains. Consistent with these observations, we find that the effect of the yield spread is stronger for financially-constrained local governments that experience negative cash flow shocks. We also show that the adoption of adjustable-rate loans allows local politicians to maintain lower taxes suggesting that myopia related to political economy considerations can be an important channel driving the borrowing decisions of local governments.

We establish a framework to quantify the welfare effects of decentralized policies. Local policies result in benefit-spillovers, mobility of households and firms, and interjurisdictional fiscal externalities that are not internalized by the government enacting the policy. Their magnitudes are measured by a new metric, the “marginal corrective transfer” (MCT), the share of funding the federal government should provide to induce a locality to internalize these interjurisdictional externalities. Formally, the MCT is estimated as the wedge between the marginal value of public funds (MVPF) of the locality enacting the policy and the MVPF of the entire federation. To calculate this wedge, we develop a rigorous framework for distinguishing between the benefits and costs that are internal and those that are external to the enacting locality. The MCT enables comparisons of local policies, allowing the federal government to prioritize policies based on the relative external benefits and costs.

We study tariffs in a general equilibrium dynamic model with and without search frictions between heterogeneous exporting producers and importing retailers. The model’s competitive equilibrium is inefficient because it has matching, participation, and aggregate productivity externalities. We find the global tariffs that internalize all externalities. We characterize optimal import tariffs, chosen by each country’s social planner, with and without strategic considerations. We show that optimal tariffs in the model with search frictions may be negative because of market thickness effects. Finally, we characterize how unilateral optimal tariffs with search vary with matching efficiency and relative country size.

Trade credit late payments are a severe problem in many economies. In Poland, in 2013 – 2015, debtors were forced to adjust income tax base - deductible costs resulting from unpaid liabilities to improve payment discipline. This natural experiment enables us to answer whether excluding late payments from deductible expenses shortens days payable outstanding. To achieve this goal a panel of 56,364 Polish firms from 2010 to 2019 is analysed. We apply a difference-in-differences approach and use fixed effects and Arellano-Bond dynamic panel data. Our results show that during 2013-2015 firms cut the days payable outstanding, which acknowledges that the examined instrument is an effectual tool for reducing payment congestion. Furthermore, higher effective income tax rates in 2013 – 2015 for firms with overdue liabilities confirm that the unreliable contractors applied the analysed provisions. Therefore, excluding late payments from deductible costs is a good way to reduce intentionally delayed payments.

Did increases in mask supply slow the spread of COVID-19? Rwanda licensed and incentivized textile manufacturers to produce high-quality masks at the start of the COVID-19 pandemic; we exploit spatial variation in exposure to mask manufacturing through textile trade networks within an event-study design using receipt-level tax data. Licensing domestic mask manufacturers conservatively reduced mask prices by 8.8% and reduced monthly growth in COVID-19 infections (proxied by demand for anti-fever medicine) by 12%. The dynamics of our results suggest that increased mask quality explains reduced infections, in a context where there was strict enforcement of mask mandates and informal markets for masks.

Can the complexity of income tax systems impact labor supply decisions? We study this question in the context of withholding taxes paid by married couples in Germany. In a first step, we document with a survey that less than 20% of the interviewed married individuals understand that withholding taxes do not impact the final income tax burden. In line with this finding, we then estimate that married women adjust their labor supply in response to reform-induced variation in the size of withholding taxes. Importantly, the German institutional setting allows couples to partly redistribute the withholding tax burden from one partner to the other, and the majority shifts parts of it from the husband to the wife. Our findings suggest that the increased withholding tax burden of married women in Germany contributes to their low labor supply. They also highlight that governments should be aware that overwithholding distorts labor supply incentives.

Married couples filing a joint return put the male name first 88.1% of the time in tax year 2020, down from 97.3% in 1996. The man’s name is more likely to go first the larger is the fraction of the couple’s allocable income that goes to him, and the older is the couple. Based on state averages, putting the man’s name first is strongly associated with conservative religious attitudes, religiosity, and a survey-based measure of sexist attitudes. Risk-taking and tax noncompliance are both associated with the man’s name going first, but the opposite is true for charitable giving

Economic theory suggests that the land market is a powerful device to achieve optimal resource allocation, but there is little credible evidence to back this up. This article addresses the topic by discussing fiscal incentives for allocating urban commercial and residential land in China. Using parcel-level data on state-owned land transfers in major Chinese cities from 2007 to 2019 and the spatial radius matching method, we find a significant price discount on commercial parcels compared to adjacent residential ones, caused by differences in the tax-generating capacity of land. We further find that tax incentives increase commercial land supply, but a rise in rental prices counteracts up to 73% of this effect. These empirical results reveal that a well-developed land market can mitigate distortions caused by discriminatory tax policies when local government revenue sources include land rent or land value-based taxes.

This paper examines the impact of land use policy on the local labor market and housing market in China. There are two primary land use purposes: residential land and commercial land. We found that the more commercial land that the Chinese local government grants, the lower the regional unemployment rates, the higher the regional wages, and the higher the housing prices. We then built a model to investigate the mechanism behind the aforementioned empirical facts. Higher commercial land use attracts more firms and creates more jobs but reduces the land available for building houses. The combination of rising housing demand and reducing housing supply brings up the housing prices in high commercial land use areas.

During the COVID-19 pandemic, the Australian Government allowed eligible individuals to withdraw up to A$20,000 (around half median annual wage income) from their retirement accounts, ordinarily inaccessible until retirement. These transfers were much larger than those considered in the literature and represented a liquidity shock, holding lifetime income constant. Using administrative and bank transactions data, we find a high marginal propensity to spend given the transfer size of at least 0.32–0.49 within eight weeks, and that persistently poor financial health strongly predicts withdrawal and spending. We conclude that ‘behavioral’ factors, rather than low liquidity itself, best explain these observations.

Can simply changing the public definition of a normal retirement age impact retirement, without any associated financial incentive changes? We study a reformulation of the retirement system in Finland that relabeled retirement ages with only modest and continuous changes in financial incentives. We find that relabeling matters: both graphical evidence and estimated hazard models reveal an enormous change in retirement when individuals face a newly defined ‘‘normal retirement” age. We also find evidence consistent with increased retirement ‘‘regret,” as the marginal workers induced to retire by relabeling are more likely to return to work

People often do not respond to changes in incentives in key economic decisions. The existence of passive saving behavior is well-established in retirement saving decisions and in mortgage refinancing decisions despite large changes in tax incentives and interest rates. Yet, we don't know whether it is the same people who respond passively to changes in incentives across decisions, and, consequently, how foregone financial gains are distributed. We analyze this question with Danish administrative data in a multiple quasi-experimental setting, where we compare individuals’ responses to multiple reforms of saving incentives for retirement saving and mortgage refinancing. Our preliminary results document a large share of passive savers in all decisions, but we find little correlation in passive behavior across decisions. Our findings thus suggest that it is not the same people who are passive in different decisions.

Empirical evidence suggests that individuals often evaluate options relative to a reference point, especially seeking to avoid losses. We conduct a welfare analysis of reference-dependent preferences. We decompose the welfare effects of variation in reference points and prices into direct and behavioral effects and describe how these effects depend on whether or not reference dependence reflects a bias. In a simple, empirically grounded model of loss aversion, lowering reference points robustly improves welfare; the welfare effect of a price change depends critically on normative judgments. We derive sufficient statistics characterizations of our main welfare effects. We illustrate the theoretical results with an empirical application to reference dependence exhibited in German workers' retirement decisions. Simulation and sufficient statistics results suggest positive welfare effects of increasing the Normal Retirement Age, but ambiguous effects of financial incentives to postpone retirement. We illustrate how adopting alternative models of reference dependence modifies key welfare effects.

Wage garnishment allows creditors to deduct money directly from workers' paychecks to repay defaulted debts. We document new facts about wage garnishment between 2014-2019 using data from a large payroll processor who distributes paychecks to approximately 20% of U.S. private-sector workers. As of 2019, over one in every 100 workers was being garnished for delinquent debt. The average garnished worker experiences garnishment for five months, during which approximately 11% of gross earnings is remitted to their creditor(s). The beginning of a new garnishment is associated with an increase in job turnover rates but no intensive margin change in hours worked.

Indebtedness is a common feature of American financial life and debt originating from health care is particularly widespread. In the United States, there is more medical debt in collections than all other sources of debt combined and roughly one in five Americans has a medical debt in collections (Kluender et al., 2021). Debt relief in other contexts has shown the potential for large beneficial effects, improving financial well-being, increasing earnings, and even reducing mortality (Dobbie and Song, 2015; Dobbie et al., 2017; Di Maggio et al., 2019). This paper analyzes the impacts of two randomized control trials (RCTs) which relieved $173 million of medical debt in collections for 84,210 patients. We present outcome measures drawn from proprietary surveys, debt collections data, and credit reports.

We evaluate the effects of the 2020 student debt moratorium that paused payments for student loan borrowers. We show that the payment pause led to a sharp drop in student loan payments and delinquencies for borrowers subject to the debt moratorium, as well as an increase in credit scores. We find a large stimulus effect, as borrowers substitute increased private debt for paused public debt, while borrowers used the new liquidity to increase borrowing on credit cards, mortgages, and auto loans rather than avoid delinquencies. The effects are concentrated among borrowers without prior delinquencies, who saw no change in credit scores, and we see little effects following student loan forgiveness announcements. The results highlight an important complementarity between liquidity and credit, as liquidity increases the demand for credit even as the supply of credit is fixed.

I study how debt relief impacts household saving and consumption compared to cash transfers by analyzing borrower responses to federal student loan forbearance in the 2020 CARES Act. Borrowers manage liquidity from the payment pause non-optimally, with many making large voluntary prepayments on 0%-interest student debt instead of repaying higher-interest obligations. These borrowers correctly prioritize repaying high-interest debt when receiving stimulus checks. This indicates a flypaper effect in which borrowers treat liquidity from debt relief as non-fungible with other windfalls. As such an effect predicts, borrowers display a marginal propensity to spend (MPX) out of forbearance liquidity less than half the size of their MPX out of stimulus checks. I field a student borrower survey to understand mechanisms, and formally incorporate results into a lifecycle model. Finally, I show that the flypaper effect has quantitatively large impacts on the effectiveness and cost of forbearance as a counter-cyclical fiscal policy tool.

In this paper, we exploit a quasi-random variation in public-sector employment from an episode of the relocation of public-sector entities and estimate local employment multipliers. The estimated multiplier is positive, statistically significantly different from zero, and persistent over time: an introduction of 1 public sector employment increases the private sector employment by 1 unit, almost completely driven by the services sector. In line with the literature, we document that the effect of public employment on private employment is highly localized. In addition to changes in private employment, we also find that the relocation led to a positive net-inflow of residents into the treated neighborhood; this effect is also localized. Lastly, we estimate heterogeneous local employment multipliers and provide evidence that this heterogeneity is shaped by the size of public sector shocks, different types of relocation, and the extent of migratory responses.

How does economic activity respond to tax shocks and how do these responses depend on the distributional targeting of the tax change? We shed light on the potentially stimulative effects of tax cuts using thirty years of US tax filing data. Focusing on state income taxes, we show that state-level measures of employment respond positively to tax cuts targeted at the bottom of the income distribution, but find no evidence that economic activity responds to tax shocks aimed at the highest earning households.

Expected inflation has few real effects in purely private economies, but this is not the case when the tax system is not neutral with respect to inflation. In practice, tax systems are not neutral—though some have attempted to be so in the past—and this paper provides a comprehensive overview of the most relevant non-neutralities drawing both on existing literature and showing new illustrations and evidence of the effects. The paper shows, for example, how taxing inflationary gains can have tremendous impact on effective tax rates—even at relatively low rates of inflation. It also shows how partial adjustment—for only some types of incomes—can create additional distortions. A new empirical analysis reveals how the erosion of the value of depreciation allowances through inflation affects investment. Finally the paper discusses policy options to address such non-neutralities.

Exploiting regional differences in local public spending during the Great Depression in early 1930s Japan, we examine the impact of expansionary fiscal policies on suicide prevention. Analysis of historical regional panel data from 1899 to 1938 suggests that increased local spending helped mitigate the rise in suicides during the Great Depression. Additionally, we investigate variations in the effects on subgroups defined by industry ratios, revealing stronger mitigation in comparatively urbanized regions. Furthermore, we explore the mechanisms connecting fiscal policy and suicide rates, concluding that economic and familial factors play a significant role.

I consider the case for the minimum wage alongside (optimal) income taxes when workers differ in both wages and working hours, such that a given level of income corresponds to multiple wage rates. This renders the minimum wage unambiguously desirable in a discrete-type model of the labor market. Desirability of the minimum wage is a priori ambiguous in a continuous-type model of the labor market. Compared to the minimum wage, income taxes are less effective in compressing the wage distribution but more effective in redistributing income. Desirability of the minimum wage depends on this trade-off between the "predistributional" advantages of the minimum wage and the "redistributional" advantages of the income tax. I derive a desirability condition for the minimum wage and write it in terms of empirical sufficient statistics. A numerical application to the U.S. suggests a strong case for a higher federal minimum wage.

This paper analyzes whether a minimum wage should be used for redistribution on top of taxes and transfers. I characterize optimal redistribution for a government with three instruments – labor income taxes and transfers, corporate income taxes, and a minimum wage – using an empirically grounded model of the labor market with positive firm profits. A minimum wage can increase social welfare when it increases the average post-tax wages of low-skill labor market participants and when corporate profit incidence is large. Also, the minimum wage can help the government redistribute efficiently by preventing firms from capturing income subsidies and from enjoying high profits that cannot be redistributed via corporate taxes. Event studies show that US state-level minimum wage reforms have increased average post-tax wages of low-skilled labor market participants and reduced corporate profits in services industries. A sufficient statistics analysis implies that US minimum wages typically remain below their optimum under the current tax system.

We revisit the seminal result by Chamley (1986) and Judd (1985) from the perspective of fiscal institutions. Using a model that incorporates institutional frictions and the finite policy planning horizons of policymakers, we illustrate another mechanism leading to positive optimal capital taxation. The numerical results based on a tractable model calibrated to the U.S. economy suggest that the current U.S. tax scheme is consistent with near-optimal fiscal policy when policymakers have finite planning horizons and imperfect altruism toward household welfare under subsequent governments.

This paper discusses tax implications of the rapidly evolving use of cryptocurrencies. It lays out the main tax design and implementation challenges across taxes on income (mining proceeds or capital gains), wealth (including inheritances), and transactions (VAT and sales taxes). The paper also elaborates on possible future measures in a rapidly changing environment. Opportunities for estimating potential tax evasion/avoidance are limited due to the lack of relevant data in light of the quasi anonymity of cryptoassets. Available evidence suggests, however, that ownership of cryptoassets is highly concentrated at the top raising profound issues of taxing the rich. The revenue at stake worldwide is plausibly in the tens of billions of dollars, but it is unclear how much is recoverable. While much attention has been given to the taxation of capital gains from crypto, back-of-envelope calculations suggest that VAT and sales taxes are likely significant issues.

We incorporate a model of frictions into the bunching-based elasticity estimator to ra- tionalize diffuse bunching around kinks and mass above notches in empirical distributions. Model agents draw a sparse set of opportunities from a Poisson process, approximating a broad class of frictions including search costs, inattention, and lumpy adjustment; the pre- dicted density depends on the standard structural elasticity and a money-metric “lumpi- ness parameter.” We estimate the model using administrative tax data on South African small-businesses, recovering moderate elasticities of taxable income between 0.2 and 0.3 at higher incomes, and larger elasticities at low incomes. Firms appear to treat the bot- tom kink as a notch, and firms with paid tax practitioners exhibit sharper bunching, driven primarily by lower frictions rather than a higher elasticity.

This paper characterises the Laffer curve of each individual taxpayer in a schedular multi-rate income tax with income shifting. Analytical expressions for the revenue-maximising tax rate and the revenue-maximising elasticity are provided for the individual taxpayer and the aggregate population, as well as new estimates of the Elasticity of Taxable Income (ETI). Applying these to the Spanish income tax in 2011 demonstrates that 49.46% (58.49%) of the taxpaying population in the non-savings tax base (savings tax base) was on the 'prohibitive' side ('normal' side) of the Laffer curve. On average, these taxpayers were 6.59 points (24.73 points) above (below) the maximum of the Laffer curve. The fraction of total tax revenue lost through behavioural responses amounts to 53.77%. However, this fraction varies by population subgroup and decreases when we account for income-shifting responses, suggesting the presence of fiscal externalities in the Spanish PIT.

This paper studies the interstate effects of decentralized taxation and spending when individuals can work from home (WFH), a possibility that decouples a state’s population and employment levels. Our main findings show that a shift from a non-WFH economy to WFH reduces employment and raises the wage in high-tax states, with larger effects under source taxation. Once WHF is established, an increase in a state’s tax rate either reduces employment further while raising the wage (source taxation) or leaves the labor market unaffected (residence taxation). We show that the residence-taxation equilibrium is efficient.

This study analyzes the impact of Hometown Tax Donation (HTD), a unique local fiscal system in Japan, on local government efficiency. It allows residents to make donations to local governments of their choice, receiving deductions on payment of local and national taxes, equivalent to the amount donated. Therefore, tax revenue will outflow from the donor residents’ municipalities to other regions, whereas it will inflow to recipient municipalities from other regions. This HTD may cause misperception of tax prices, thereby leading to inefficient provision of local public services. This study uses stochastic frontier analysis to quantitatively analyze the impact of HTD on the inefficiency of local governments. The findings reveal that municipalities whose revenues are more dependent on HTD tend to be more inefficient. Thus, providing implications for local public finance on the importance of decentralization.

This paper analyzes how binding term limits shape the allocation of intergovernmental grants. We employ recent difference-in-differences and synthetic control methodologies, accounting for heterogeneous treatment effects, to analyze the effects of mayoral term limits on the allocation of grants to Portuguese mainland municipalities over the period 1998-2021. We find that municipalities with term-limited mayors tend to receive smaller amounts of conditional (non-formula-determined) grants in (or close to) local election years than municipalities with reelection-eligible mayors. The strongest effects are found for capital grants, national and from the European Union. However, we also find that this result does not generalize to the entire four-year term. On average, there are no statistically significant differences in the grants received by treated and control municipalities. Our results are consistent with the view that the absence of electoral incentives induces term-limited mayors to exert lower effort than reelection-eligible mayors to obtain conditional grants when elections draw nearer.

This study attempts to explore whether local public services induce internal migration. On the basis of a simple model, I developed a theoretical model of residential choice following boundary reforms of jurisdictions. Theoretical predictions are tested empirically using Japanese municipal-level data. The following results are obtained. First, mergers have positive impact on net migration and in- and out-migration. Second, in line with theoretical predictions, the differences in population and per taxpayer income are positively correlated with net migration. Third, geographical distance has no impact on net migration, whereas economic condition and local fiscal capacity are significant determinants of choice of residence.

We document that both economic and political factors are important for financing decisions in Norwegian municipalities. Municipalities with higher disposable income have more debt per capita, and municipalities with more funds and reserves have less short-term debt. Political fragmentation is associated with more debt per capita, more short-term debt, and more use of money market debt. The larger the fraction of socialist seats on the municipal board, the more municipal debt.

Just as in the for-profit sector, enforcement can encourage compliance in the nonprofit sector too. This paper studies both donor and nonprofit responses to alleged or confirmed misconduct being investigated by the media or governmental agencies. Using advisory notices published by Charity Navigator, I find no evidence that donors respond to these events, yet charities take steps to improve, albeit temporarily. In particular, spending on program services, as measured by the program service expense ratio, increase by 3 percentage points following the publication of an advisory notice. To offset this increase in program spending, spending on fundraising and administrative expenses decline, although this decline appears only temporary, with spending returning to baseline after 3 years. These results help to quantify the value of a crucial accountability tool in the nonprofit sector.

Charitable organizations are vital in funding public goods, but they often face costly fundraising efforts that result in competition among charities. To explore this competition, our study presents a model where charities use informative advertising to attract individual donors. We find that this competition leads to inefficient fundraising as charities incur excessive advertising costs, and the inefficiency increases as available funds increase. Our research highlights the implications of increased deductibility rates for charitable donations, such as intensified competition and potential impact on the quality of services provided by charities. We used data from the IRS, Kantar Media, and Charity Navigator to validate our predictions and found that increasing the deductibility rate leads to considerable increases in competition between charities. Our findings underscore the need to consider trade-offs associated with tax incentives for charitable giving and find ways to promote a more sustainable and impactful charitable sector.

Utilizing the near universe of tax filings, we document both the extent of donations of art to non-profit organizations in the U.S. and their use as a method of tax avoidance. Non-profit organizations hold sizable assets, worth $12.4 trillion in 2019, with 2.65% identified as holding art, valued at $5.5 billion. While only 20% of organizations required to disclose the value of their collections and donations do so, these organizations consistently write down the value of artwork donations by 13%. This effect is amplified by donation valuation methods more likely to be influenced by the donor, but driven primarily by donations to private foundations, which write down by an average of 98.7%. Private foundations are associated with poor internal controls. Tax loss due to avoidance is great, as back of the envelope calculations estimate between $1.4 billion and $3.9 billion in income tax losses over our 9 year sample.

Using U.S. administrative corporate tax data, I provide novel estimates on profit shifted back to the United States after closure of the Double Irish. Multinationals funneled an estimated $1.2 to $1.4 trillion in profits to low-tax jurisdictions via the Double Irish from 1998 to 2018. I estimate that firms that used the Double Irish redirected $59 billion in royalty payments to the United States in 2020. A difference-in-differences analysis, preceded by a first-stage propensity score matching technique, suggests the average Double Irish user reports $609 million more in royalty payments after full closure. I use a log specification to demonstrate royalty payments shifted, although large, are driven by outliers. These redirected royalty payments comprise 31 to 38 percent of profit within the Double Irish in 2018. My estimates on the size of the arrangement suggest the current literature may underestimate the U.S. tax base shifted abroad.

Since 2015 more than 140 countries have cooperated in the OECD's BEPS project to fight multinational tax avoidance. Several actions, most importantly measures against tax-treaty shopping, require changes to double taxation treaties. The OECD designed a special instrument - the MLI - to allow for a swift implementation of BEPS-related treaty changes. In this paper, we show that MLI take-up is incomplete, we present (partly surprising) correlates of the take-up decisions and develop a theoretical model to rationalize the observed takeup pattern. A key insight is that conduit countries can benefit from anti-treaty shopping laws as firms have incentives to scale-up real activity in conduit nations to be exempted from the new anti-treaty shopping rules. Consistent with that, conduit nations' MLI take-up rate is high. An early empirical evaluation of the impact of the BEPS project on treaty shopping indicates a modest drop at best. Firms indeed enhance their real activity in conduit nations.

We exploit the new country-by-country reporting data of multinational corporations, with unparalleled country coverage, to reveal the distributional consequences of profit shifting. We estimate that multinational corporations worldwide shifted over $850 billion in profits in 2017, primarily to countries with effective tax rates below 10%. Countries with lower incomes lose a larger share of their total tax revenue due to profit shifting. We further show that a logarithmic function is better suited for capturing the non-linear relationship between profits and tax rates than linear or quadratic functions. Our findings highlight effective tax rates’ importance for profit shifting and tax reforms.

To combat multinational firms’ profit shifting activities, many countries implemented transfer pricing documentation rules. While there is first evidence on potential benefits of such provisions, the evidence on the effects on firms’ compliance burdens and tax authorities’ administrative costs is rather scarce. In this project, we exploit the introduction of French transfer price documentation requirements and track whether affected firms change the size of their tax departments in a difference-in-differences approach. Using administrative data, we find that the reform did not pose an economically significant cost burden on firms. After the reform, treated firms increase the number of tax employees by 4 % and the wages associated with these employees by 5 % compared to control firms. In addition, we document that firms do not seem to outsource these activities to external tax advisors.

Borrowers with revolving debt on multiple accounts fail to exploit price differences in their accounts while making their repayments and leave a significant amount of money on the table. Constructing a simple repayment environment in the laboratory, we test the role of a set of behavioral mechanisms that would directly inform the design of consumer protection policies. We find that providing salient interest rate disclosure has no effect while disclosing the interest rate in a fee format has modest effects. On the other hand, providing an opportunity to purchase automated financial advice reveals that subjects are predominantly aware of their choice inefficiencies and are relatively good at using financial advice. Our results suggest that promoting and subsidizing consumer financial technology applications that provide automated financial advice would be a substantially more effective way of protecting consumers from simple arbitrage failures than conventional disclosure policies.

We demonstrate how the decision to avoid planning and saving for retirement can arise as an optimal choice made by forward-looking, utility maximizing households. In our general equilibrium, overlapping generations, life-cycle model, households select the length of their planning horizon optimally each period, in addition to making a consumption-saving decision. In simulations of our model, households avoid planning and saving for retirement while young and, as a result, arrive at retirement with just over half of the wealth that they otherwise would have accumulated had they planned for their entire remaining lifetime each period, as in a standard life-cycle model. We add a government that operates a pay-as-you-go Social Security system to our model and explore its impact on planning, consumption, and wealth over the life-cycle. We find households respond to Social Security by choosing shorter planning horizons in middle age, which reduces the welfare gains of the program.

Minority homeowners pay more than their fair share of property taxes to fund local services, but are less likely to appeal their taxes to reduce the excess burden. We conduct a framed field experiment to understand the drivers of property tax appeal behavior and the racial appeal gap. We randomize homeowners to receive a combination of information about the financial benefits, a guide of how to appeal, and a financial incentive for appealing. Information frictions are an important determinant for why few homeowners appeal. Homeowners are 417 percent more likely to appeal when provided with a guide that reduces these frictions. We provide evidence that the racial gap is due to minority and non-minority homeowners facing different types of information frictions. Beliefs about the financial benefits from appealing do not explain the gap. We find no evidence that racial differences in appeal outcomes are driven by differences in appeal quality.

By exploring a class of non-standard consumer preferences for an OLG economy with Social Security, this paper not only provides the mathematical equivalence of the three well-known behavioral models for over-consumption, in terms of the key mechanism, but also contributes to the literature on intergenerational distribution related to the pension system. The proposed model in the paper is versatile enough to incorporate over-consumption, as well as under-consumption, in a unified framework among the decision makers who deviate from the standard consumption behavior due to non-exponential time discounting, temptation utility and self-control cost, or low loss aversion. Furthermore, by utilizing an integrated scheme for social security system which allows partial intergenerational transfers, this paper can determine the transition effect from a movement toward a more funded social security system when consumers are prone to over-consume.

We use individual-level tax administrative data to estimate income inequality for formal sector income earners in South Africa. The tax data offers better coverage of the top end of the distribution, allowing us to track changes in the income shares and sources of income for top income groups over time. We exploit the longitudinal nature of the tax administrative data to examine income mobility across the distribution and, in particular, the top income groups. Our results indicate that income inequality for formal sector earners in South Africa remained stable over the 2011 – 2018 period. We also find that mobility for formal sector earners is low, particularly at the top end.

In collaboration with the Uganda Revenue Authority, I conduct a randomized evaluation of a low-cost, easy-to-implement text-based tax encouragement scheme, testing the efficacy of sending messages and looking at variation across message content. I find a 6x the cost rate of return on average, which increases to 13x when considering an enforcement-focused message. This average effect masks substantial heterogeneity along existing tax capacity, measured via a novel, granular, nationwide dataset on governmental service provision and economic activity. I find that the treatment effect is stronger when tax capacity is lower, and that this relationship is strongest in areas with recent, high-visibility investments in services. These results are consistent with a role for fiscal exchange in tax compliance, where in a semi-voluntary context, the success of efforts to increase compliance may be driven by perceived provision of services and the visibility of the state.

This study investigates a Consumption Goods-Based Withholding Tax Regime in Brazil. We leverage the tax-administrative data of a representative Brazilian State (Rio Grande do Sul) with a reform in 2016 that excludes a set of goods from the withholding tax regime (Tax substitution – ST) to document the impact on tax collection, investments, and sales revenue reporting for both industries/manufacturers and retailers involved in those production chains. Using the dynamic differences-in-differences approach (Callaway and Santanna, 2021), we find an increase in tax collection of R$2.5 million due to the increase in tax collection on retailers (150%), without a significant loss in sales revenue reported of the manufacturers and retailers. We still find an increase in large frims' investments both retail an industry. The results are sensitive to the size of the companies, showing more precisely estimated positive effects for larger companies.

I test the claim that VAT’s in-built third-party information trail, togetherwith electronic filing, deters tax evasion automatically. Using a reform which utilizes information beyond VAT trail and authorizes a computerized risk analysis system to accept or reject tax credits in real time, my difference-in-differences estimates show that claims declined by fifty percent. Based on firm heterogeneity, the response ranges from thirty to ninety percent. Ten percent of treated firms were fake, created for missing trader fraud. Lower bound estimate of increase in net VAT collection at country level is ten percent. I find that traditional VAT enforcement mechanisms of cross matching, audit and recovery fail to deter evasion in developing countries but a risk based real time enforcement system is effective. Because this system eliminates the need for reverse charge, its tax policy implications extend to developed countries.

Welfare weights measure the value of providing an additional dollar of consumption to individuals in society. We identify the welfare weights assigned by the general population of the U.S. to individuals in society using a large-scale real stakes experiment. The welfare weights obtained from our experiment are sufficient statistics that can incorporate various ideals and can be directly used to evaluate reforms. We find that the welfare weights of the general population of the U.S. (i) exhibit, on average, a low to moderate degree of progressivity, (ii) predict their policy preferences, (iii) are slightly more progressive than the welfare weights implied by the existing policies in the U.S., and (iv) are much less progressive than the welfare weights commonly used in the literature.

This paper studies the impact of economic deprivation on radical voting. Using IV estimation, individual panel data and a unique dataset covering different indicators of economic deprivation as well as federal election outcomes at the county-level in Germany, we examine whether economic deprivation influences the support for radical parties. Our results suggest that an increase in economic deprivation at the county-level has a sizeable effect on vote shares of radical parties; notably on the electoral success of the populist right-wing party AfD in the federal election of 2017. The effect of regional economic deprivation on individual political preferences is constant across all income groups. Our findings show that a rise in relative economic deprivation decreases satisfaction with democracy and may undermine trust in the political system across all income groups. Economic deprivation thus might be a threat to democracy and political stability.

Within a general equilibrium model, this paper identifies a novel animal welfare externality that occurs if the private animal friendliness in a market economy falls short of the social animal friendliness used by the social planner when determining the efficient allocation. The animal welfare externality causes an inefficiently high quantity and an inefficiently low quality of animal food. Correction of this market failure is attained by taxing animal food output and subsidizing animal food quality. With consumer and producer heterogeneity, reg- ulation is the same but sector-specific, with a more intense regulation in the sector with the worse living conditions of animals.

CO2 emissions are disproportionately caused by more affluent consumers. In the political debate, this fact has triggered the demand for income redistribution and wealth taxes not only to reduce inequality but also to reduce CO2 emissions. This paper calculates the possible size of such a green dividend of redistribution in 26 countries and concludes that, for most EU countries, it is negative if the redistribution is efficient, in the sense that it keeps average incomes constant. If the redistribution introduces inefficiencies that lead to total income losses, the negative green dividend, otherwise associated with additional redistribution, may be avoided.

The economic debate underlines the reasons why discount rates of infrastructure projects should be similar, regardless the public or private source of financing, during the forecast period when flows are risky but predictable. In contrast, we show that the incompleteness of contracts between governments and private firms beyond the forecast period (i.e., when flows of net social benefits are state-contingent) entails expected terminal values that are systematically larger under government rather than private financing. This effect provides a new rationale for applying a lower discount rate in the assessment of projects under public financing as compared to private financing. Our results are particularly important to properly consider the impact of uncertainty that is determined by alternative climate scenarios on investments in infrastructures

This paper provides the first causal evidence that gender affects the information an individual receives about careers. We conduct a large-scale field experiment in which real college students seek career information from 10,000 working professionals. We randomize whether a professional receives a message from a male or a female student. When students ask broadly for information about a career, female students receive substantially more information on work/life balance relative to male students. This gender difference persists even when students specifically state an interest in learning about work/life balance. In contrast, professionals mention workplace culture to male and female students at similar rates. We develop a methodology to combine our experimental estimates with student preferences for professionals. Allowing students to choose which professionals they interact with does not reduce gender disparities in access to information, nor does it align the information students receive with the information they demand.

This paper examines gender differences in advice seeking behavior. We introduce a novel experimental framework in which subjects have the option to either solve a task on their own, or with the help of costly advice. Over two treatments, we vary the amount of information that advisees receive about advice quality. We use two types of questions, mathematical and verbal, to test the effect of stereotyped domains. Our findings suggest that women seek less advice than men. This result is driven by two behavioral patterns. First, women seek less advice when they are in the information treatment. Second, men seek more advice on verbal tasks. Advisor gender does not matter. These results suggest that gender may affect the decision to seek advice.

Many universities created temporary optional information policies—e.g., by allowing students to replace grades with “credits” and by making test scores optional—during the early part of the COVID pandemic. As universities consider whether to make some of these policies permanent, it is essential to understand their equity implications. To study the equity implications of information-optional policies, we leverage data from two large-scale natural experiments at two highly selective universities, that introduced optional “credit” (pass/fail) grading that students could choose after already learning their grades. We find that female students are significantly more likely to reveal damaging performance information—grades below their GPA—when given access to the optional grading policy. That is, our findings show that optional information policies can create unexpected disparities because women are more transparent than men about their performance. Additional results from student surveys as well as follow-up experiments speak to the potential drivers of this transparency gap.

We consider the effects of high stakes and performance feedback on the gender gap in winning competitions. Despite their intuitive appeal, we show that high stakes and feedback have theoretically ambiguous effects on this gender gap and that their effectiveness depends on gender differences in beliefs and costs of competing. We run a laboratory experiment with 292 undergraduate college students and find that neither intervention reduces the gender gap in winning. We explain that this may depend on high stakes being not high enough, on the sample not having more women with pessimistic beliefs, and on low serial correlation in rank performance. We discuss the policy implications of our findings.

The COVID-19 crisis reminded again that foreign investor participation in local-currency government bond markets in emerging market economies (EMEs) may create substantial instability in the markets. This paper sheds light on this issue by studying how, during financial stress periods, foreign investors of different types may behave differently in the markets. Specifically, assembling weekly data on net purchases of local-currency government bonds by several types of foreign investors over the 2008-21 period for Indonesia, we show that when global financial risk is high, net selling of the bonds by foreign mutual funds destabilizes the market by increasing the yield (i.e., decreasing the price) significantly. Our interpretation is that foreign mutual funds, often prone to redemption demands from the end investors during stress periods, destabilize the market by resorting to fire sales.

Specialization within couples implies financial risks for individuals who take on the majority of unpaid household and care production. They forgo human capital accumulation and individual labor income while financially relying on their working partner. In opposite-sex couples, these individuals are predominantly women. Combining administrative and survey data from Germany, we document significant specialization penalties in lifetime income for married women that extend to post-divorce periods. Importantly, we show that women also face specialization penalties in marriages with no children. While marriage as a legal contract can provide insurance against this risk of specialization by ensuring financial compensation upon divorce, the long-term risks of specialization are often neglected. To understand whether spouses incorporate the (limited) insurance in their specialization decision, we exploit a divorce law reform that reduces the insurance value of marriages by cutting alimony rights.

We estimate the subnational employment and GDP multiplier of Brazil's 2020 federal cash transfers to vulnerable households. Using two-stage least squares regressions we estimate a formal employment multiplier and then apply an analytical transformation to recover an implied GDP multiplier in the range of 0.5-1.5. The lower bound of this range lies below most estimates in the literature, which may result from the exceptional constraints imposed by the pandemic on supply chains and consumption. Nevertheless, even using the lower end of our range implies that federal cash transfers played an important role in supporting employment and GDP.

This article tests the hypothesis of fiscal dominance for the Brazilian economy between 2003/01 and 2022/10 through the models of Thresholds Autoregressive Approach (TAR) and Markov Switching Regime Regression. The results for both methods suggest that the relationship between public debt and short and long-term interest rates is non-linear. TAR models suggest that the country enters fiscal dominance from a public debt of 71% of GDP in the case of short-term and 73% in the case of long-term interest rates. The models of Transition of Markovian Regimes indicate that from these indebtedness points, the effects of debt on interest rates intensify.

I consider the private provision of a public good within a community, whose members are inclusive Kantian or Nashian optimizers. It is analyzed how a change in group size due to an additional Nashian joining the community affects the mixed Kant-Nash equilibrium. Kantians face a trade-off when admitting a Nashian into the community. Even though the new member is a contributor, there is a negative externality, as the Kantians’ hypothetical cost of public good provision, which is already perceived too low, is falling in group size. The burden of public good provision may partially be shifted from the Nashians to the Kantians. Starting from an entirely Kantian community, the first Nashian always raises utility in an interior solution. Yet, the number of Nashians tolerated by a fixed number of Kantians may be finite. As the number of Nashians increases, the entire Nashian subgroup may switch from contributors to complete free-riders.

We examine Tug of War contests with the Blotto specification. Players have fixed effort budgets and must allocate these budgets to a sequence of battles. The outcome of each battle is a function of the efforts allocated to that battle. The player who first wins L more battles than the opponent wins the contest. We derive a pure strategy, subgame perfect equilibrium for the case where the contest success function that governs each battle is a generalized Tullock function. In equilibrium, players invest the same percentage of their remaining resources into each battle. The value of this percentage depends on how close each player is to winning the contest. Escalation of efforts occurs when the player with the smaller budget is close to winning; at the same time, the probability that a player wins any individual battle remains constant along the entire equilibrium path.

This paper analyzes the impact of corporate social responsibility (CSR) on the total provision of public goods in a framework in which consumers who may make such voluntary contributions to public goods via CSR are also voters who decide on the level of taxes to finance publicly provided public goods. The main result indicates that, relative to an economy in which all public goods are

publicly financed, the introduction of CSR lowers the total amount of public goods, as voters rationally anticipate that higher CSR will partially offset the consequences of lower public funding. The results offer a cautionary tale about the promotion of CSR in an economy with heterogeneous preferences for the public good.

Using anonymised administrative data on the population of UK taxpayers, we show that – in line with high-profile anecdotes about the tax affairs of the rich – effective average tax rates (EATRs) decline at the top of the distribution of income and capital gains. We also document substantial variation in EATRs within remuneration level: a quarter of those in the top 1% pay headline rates, while another quarter pay at least 9pp less than the headline rate. Most of this effect is driven by the composition of remuneration, with investment income having lower tax rates and capital gains having lower rates still. If all individuals with income above £100,000 paid the headline rates, this would raise tax revenue on income and gains by £23bn on a static basis, an increase of 27% in the tax paid by this group.

Combining a variety of survey and administrative data, this paper measures the progressivity of taxes and transfers for each of the US states and contrasts it to progressivity at the federal level. Our findings are fourfold: (i) the tax and transfer system is progressive at the federal level; (ii) state and local tax and transfer systems are close to proportional, on average: the regressivity of state consumption and property taxes neutralizes the progressivity of state income taxes and transfers; (iii) there is substantial heterogeneity across states, and its key determinant is the choice of the tax base (sales and property vs income); (iv) Democrat-leaning states tend to have more progressive systems, but richer and more unequal states tend to me more regressive.

We investigate the behavioral response of workers to the labor income tax schedule taking into search costs in the formal labor market, the informal sector, and the effect of publicly provided goods. We extend the theoretical framework f Chetty et a (2011) to show that, unless search costs are negligible, informality fosters bunching at kinks of the tax schedule, which brings to larger estimates of the elasticity of the formal labor income to tax changes. Smaller search costs also inflate bunching. We empirically assess such theoretical predictions using Brazilian matched employer-employee administrative data reported by the firms, to test formal wage earners elasticity of earnings with respect to changes in marginal tax rates from the Brazilian federal income tax schedule. We show that individuals bunch at kink points of the tax schedule in 2008 and the estimated bunching mass and elasticity vary according to theoretical testable predictions

We investigate the effect of personal income tax (PIT) rates on the number of hours worked as a self-employed individual. Using the rotating panel data from the Annual Social and Economic Supplement (ASEC) of the Current Population Survey (CPS) from 2003 to 2019, we estimate IV regressions in first differences to exploit changes in the tax code for identification. We also distinguish between self-employed owners of incorporated and unincorporated businesses and examine their differential responses. The findings reveal that higher individual-specific marginal PIT rates increase the hours worked among the self-employed with incorporated businesses, which could be explained by the availability of tax avoidance strategies. In contrast, we do not find a significant response to PIT tax in hours worked among unincorporated entrepreneurs. We also find that higher corporate income tax rates discourage switching to incorporated self-employment and encourage switching to unincorporated self-employment.

In this paper, we study the extent to which ministers engage in regional favoritism. We are the first to provide a comprehensive analysis of a larger set of the governing elite, not just focusing on the primary leader. We hand-collect birthplaces of this governing elite globally. Combining this information with extended night-time luminosity and novel population data over the period from 1992 to 2016, we utilize a staggered difference-in-differences estimator and find that birthplaces of ministers globally emit on average roughly 7% more nightlight. We do not find evidence that this is driven by, or induces migration to their home regions. The size of our data set lets us investigate heterogeneities along a number of dimensions: political power, ministerial portfolio, and the institutional setting.

The theory of economic voting, which describes that voters judge governments by how well they manage the economy, has found mixed support in empirical studies. Previous studies have typically employed data for large geographical units such as municipalities. We have collected a novel fine-grained dataset on the precinct-level, including a total of 23,880 electoral precincts which allows us to capture the previously unobserved within-heterogeneity. We link the electoral precincts with unique address-level data on socio-demographic and economic characteristics. For causal identification, we use the exogenous economic shock arising from the Covid-19-pandemic as a Bartik-instrument. The Covid-shock exerted heterogeneous effects across industries and, hence, precincts depending on their industrial structure. Our paper provides important new insights into the longstanding debate on the role of economic performance for voters’ electoral choices. Confirming the economic voting theory, we show that voters support left-wing parties if they suffered income losses arising from the Covid-19-pandemic.

When governments cannot perfectly enforce taxation, they may seek to exchange services for voluntary citizen tax compliance. This fiscal contract requires that tax morale responds to public service provision. In this paper, we present experimental evidence of the impact of local public infrastructure on tax compliance, leveraging a large public investment experiment and individual property tax records from Mexico City.

I study an effort by the Ecuadorian government to mitigate offshore tax avoidance through the imposition of an outflows tax. First, I find that the outflows tax led to a sharp decrease in dividend payments to tax havens. I document a decrease by 66% in dividend payments sent to tax havens relative to non-havens following an increase in the relative cost of transacting with tax havens by 5%. Using administrative data on shareholder-company linkages to identify individuals highly connected to tax havens, I find that these individuals increased their domestic income reporting by 40% and paid 55% more in personal income taxes per year. This response was mainly driven by domestic capital income flows and independent labor income and not through repatriation behavior or wage labor income. These results suggest the substantial scope for countries to act unilaterally in mitigating tax haven use and increasing tax collections.

Over recent years, a growing number of countries have enacted rules that require multinational enterprises (MNEs) to document their intra-firm trade prices and show that they are set as in third-party trade. The rules aim to limit strategic trade mis-pricing and profit shifting to lower-tax affiliates. In this paper, we quantify the regulations’ fiscal and real effects. Testing ground is the introduction of the French transfer price (TP) documentation requirements in 2010. Drawing on rich firm-level data, we show that affected MNEs increased their tax reporting in France, while simultaneously lowering real investments. There are also cross-border effects: while foreign affiliates in high-tax countries remain unaffected by the reform, treated multinationals invest less at foreign low-tax locations. We show that the observed investment response roots in reform-induced increases in firms’ effective tax costs; there is no indication that MNEs respond to compliance burdens associated with the new TP documentation rule.

Leaks of confidential documents from companies that facilitate the creation of secretive corporate structures in offshore jurisdictions have become a major source of information about the world of financial secrecy, with far-reaching consequences for the individuals involved. In this paper I look at the effects that the leaks have on individuals not directly involved in the leaks, but using schemes and tax havens exposed by the leaks. I use the leaks in a fixed-effects model at the bilateral level to assess their impact on cross-border bank deposits and portfolio investment. I find that offshore leaks negatively influence the use of the implicated offshore jurisdictions: the more pronounced is the presence of a given offshore jurisdiction in an offshore leak, the higher is the effect on inward cross-border financial flows to that jurisdiction.

We show that an unintended consequence of property tax limitations, such as assessment limitations, can expose households to more systematic risk despite decreasing the variance of property tax payments. Using a state border discontinuity design for the universe of U.S. residential properties, we show that this increase in risk translates to homeowners facing relatively larger increases in property tax obligations during market downturns. We develop an innovative measure of tax risk using Arrow-Debreu securities. The price of tax risk varies from $0 to $2,000 a year across states due to various combinations of assessment, levy, revenue, and mill rate limitations. Mortgage distress substantially increases with the price of tax risk. Specifically, moving from a no-limit state (e.g., New Hampshire) to a state with multiple types of limitations (e.g., Michigan) increases the probability of mortgage distress by 1% as a result of increased tax risk.

This paper investigates whether homebuyers display a particular type of cognitive bias called left digit bias when it comes to future property tax payments. Using a regression discontinuity method, this paper shows that homes with property tax obligations that are just under a thousand-dollar threshold sell for 0.5% more than homes with property tax obligations that are just over a thousand-dollar threshold. For the median home in the sample, this type of cognitive error results in an over-payment of $2,370. The estimated left digit bias is larger in areas with low rental rates, suggesting that owner-occupiers are more susceptible to this cognitive error than professional landlords. Additionally, the size of this cognitive error grows over the sample time period, which coincides with the rise in popularity of online real estate websites such as Zillow that provide access to property tax data for prospective home buyers.

This paper studies the causal impact of property taxation on long-term regional development. To this aim, we exploit a unique natural experiment, the Taille, during the Ancien Régime France. The Taille was one of the main tax between the 14th century and the French revolution, whose tax base varied discontinuously around a specific tax border. The southern area, under the historical influence of the Roman Empire, levied a property tax, based on the cadastre, while the northern area levied an (archaic form of) income tax. Using original historical data at regular time intervals, and very fine spatial resolution data, we estimate a Spatial RDD model. We find that the area subject to property taxation experienced a greater regional development in terms of population, number of firms, and income.

Accurate valuation of the tax base provides an important foundation for evaluating the equity and efficiency of property tax. This paper tests whether assessment outcomes are impacted by how frequently local governments update their property tax base. We first examine the effect of revaluation lag on assessment uniformity with Virginia’s tax assessing jurisdictions, using IV and control-function approaches. Results show four to six percent deterioration in assessment uniformity per additional year of interval between revaluations. Then we analyze the impact of switching from irregular assessment to shorter and regular (specifically annual and triennial) cycle on assessment outcomes using data on New York's municipalities. We employ difference-in-differences and event study models with various strategies to address potential selection bias. We find improvement in assessment uniformity from annual but not triennial assessment. However, committing to both cycles are associated with higher administrative cost on assessment, while not imposing additional property tax burden.

How are optimal taxes affected by reputation building and imperfect information in labor markets? In this paper, I build a model of labor markets with incomplete and asymmetric information where job histories play a crucial role in transmitting information about workers' productivity, which allows us to better understand the efficiency and distributive consequences of imperfect monitoring and screening in labor markets, and the tradeoffs the government faces when setting taxes. Optimal taxes are described by generalized versions of standard redistributive and corrective taxation formulas, which depend on labor wedges: the marginal contribution to output relative to the increases in lifetime earnings that result from supplying one extra unit of labor at each period. Combining estimates from the literature and new estimates using data from the Health and Retirement Study, I find that the corrective component of taxes is likely to be large, especially at the top of the income distribution.

Although the existing U.S. code through its Generation-Skipping Transfer Tax levies a higher tax on estates passed directly to grandchildren rather than to children when children are still alive, previous work on estate taxes has largely ignored the effect of estate-tax rates that depend on the age of heirs. Using a two-period overlapping-generations model, we examine how steady-state welfare varies as we change the estate tax rate imposed on young vs old heirs. In our baseline calibration, as in reality, the estate tax does not generate a large amount of revenue relative to the labor tax, but we find that welfare can be improved, on account of a higher present value of lifetime income, if, in contrast to the existing code, we set the tax rate paid by young heirs to zero and raise the tax on old heirs to maintain total tax revenue.

This paper characterizes the optimal labour and capital income tax schedules when access to income shifting opportunities is heterogeneous. I analyse a two period model where individuals face a variable as well as a fixed cost when shifting income between the labour and capital income tax base. The optimal level of income shifting depends on the progressivity of the tax system, both locally and globally. The intensive shifting responsiveness will have the largest implications for optimal tax rates at those income levels where income shifting is mostly present. Calling for a lower (higher) tax rate on labour (capital) income. The extensive shifting responsiveness will lead to changes that have implication for the overall progressivity of the tax system. Increasing (decreasing) the global progressivity of the labour (capital) income tax system.

How ought risk to be shared between two groups of agents: one which must be provided incentives, workers, and another which need no longer be, retirees?

To answer this question, we must assess how incentives affect workers' ability to carry risk and how backloading incentives through entitlements optimally vary across states of nature.

Perfect risk sharing is optimal if utility from consumption is log or aggregate productivity growth is i.i.d. Departures from perfect risk sharing increase welfare if more risk is born by retirees (workers) when productivity growth is persistent (mean reverting).

Numerically, the departure from perfect risk-sharing seems to be small. We discuss different alternatives to implement the optimal allocation.

Using administrative plant-level data I study how capital stock taxation affects firms and workers. Before a reform in 1998, the German local business tax was levied on two bases: profit and the capital stock. I exploit this unique setting to identify the effects of capital stock taxation. Businesses that experienced a larger tax cut increased investment and wages. However, aggregate employment was not affected in the long run. In robot-adopting sectors, firms that experienced larger tax cuts reduced employment compared to less exposed firms. This provides support for an interaction between capital taxation and automation. Comparing the estimates to the literature on corporate income taxation, I find slightly larger investment effects and smaller wage and employment effects.

This paper presents a simple theoretical model that highlights the differences between capital income taxes and wealth taxes. The focus is on a distinction between two asset classes. One asset class provides financial returns whereas the other asset class provides flow utility to its owners. In this setting, both capital income taxes and wealth taxes discourage saving. Moreover, capital income taxes distort the portfolio composition of individuals. With preference heterogeneity, the optimal capital taxes are non-zero. Depending on the nature of preference heterogeneity, either the optimal capital income taxes may be non-zero, the optimal wealth taxes may be non-zero, or both.

This paper analyzes drivers of individual wealth accumulation using machine learning methods. Wealth accumulation is measured as the absolute difference in individual wealth recorded in two consecutive wealth surveys of the German Socio-Economic Panel (SOEP) in the years 2002, 2007, 2012 and 2017. Random Forests are employed to select and order the most important drivers from a long list of potential factors like socio-demographics, employment variables, and savings behavior. To gain a more fine-grained understanding of wealth dynamics, different types of wealth holders are defined and their asset-specific wealth growth is analyzed separately. The analyses reveal that while commonly identified drivers of wealth growth (such as income and inheritances) are important, lesser known mechanisms, such as the type and location of real estate or personality traits, are also found to be influential.

This paper documents a strong Hometown-Growth-Wealth nexus exploiting large spatial and intertemporal variation of economic growth across German regions 1920-2019. We use a new dataset that oversamples wealthy individuals and includes information on the county of birth, current residence, individual wealth, inheritances and gifts, and parental background. A one percentage point increase in economic growth in a person's birth place since birth increases individual net wealth in adulthood by roughly 400 Euro. This translates into a growth effect of 90,000 Euro for a person born in the declining industrial Ruhr area between 1940 and 1959, while one born in Munich has one of 252,000 Euro. Our mediation analysis reveals that about two thirds of the total growth effect can be attributed to our set of mediators, within which intergenerational transfers are the most important being responsible for 90% of the indirect effect.

The expansion of high-speed broadband Internet is of central importance to public policy and the digital transformation. We analyze the causal effect of fast wireline Internet access on real estate prices in Germany. Our identification strategy exploits the quasi-experiment of German states’ different preferences for broadband expansion in rural areas, which induced variation that was plausibly exogenous to house buyers and tenants. Using a spatial RDD and a novel micro-dataset, our main findings for 16 Mbit/s broadband reveal significantly positive capitalization effects of 4 to 7 percent. The results correspond to increases of average property prices by about 13,575 euros and of monthly rents by 28 euros. For higher Internet speeds, we document still significantly positive capitalization effects but find diminishing returns to Internet speed. We show that the capitalization effects are driven by households’ current demand for fast Internet, as higher broadband availability translates into higher uptake by households.

Commercial gambling legalization is a popular means of raising tax revenue in the United States, despite disagreement over the economic and social effects of widespread gambling access. This paper investigates the local effects of video gambling legalization in Illinois, where almost 7,000 establishments now operate up to six slot machines. Using variation in municipal legalization in an event-study framework, we find mixed evidence that legalization increased municipal revenue. While tax revenue increased on average, these gains were offset by the expiration of federal aid grants following the Great Recession. Despite these patterns, we estimate that legalization produced a significant, sustained increase in local public spending of 4.2%. This result is consistent with the salience of expected gambling revenue enabling local governments to justify budget increases. In ongoing work, we use our empirical framework to evaluate the effects of legalization on household credit usage to evaluate welfare effects.

This paper examines the role of large-scale renovations in driving gentrification, particularly in rental housing in Sweden. Using comprehensive Swedish register data, we find that renovations coincide with changes in the socioeconomic composition of properties. Our findings suggest that turnover increases in the years around a renovation and that average earnings increase by 13 percent across all properties and 73 percent for low-income properties. However, we find no conclusive evidence of spillover effects on neighboring properties, raising questions about the broader impact of renovations on gentrification.

Taxes can affect the efficient working of the market for corporate takeover. Tax implications from takeovers depend on joint characteristics of firms involved in reorganization but sorting among firms render the assessment of tax influence difficult. We take advantage of the development in econometrics of matching models and estimates a model that explicitly

account for sorting behaviors. We examine a sample of M&As transactions between publicly traded corporations in Japan from 1996 to 2018. Our estimates thus far suggest a limited influence of losses carried forward and debt finance. The analytical framework in our paper can be extended to cross-border M&As and other issues in public economics.

This study provides a game-theoretical model to investigate whether the availability of dispute resolution procedures induces multinational enterprises (MNEs) to set transfer prices in a tax-aggressive way. We find that this is typically true. However, in some cases, the availability of dispute resolution procedures may induce the taxpayer to report more profit in the high-tax country than he or she would do in the absence of such procedures. Simulations reveal that this result is more likely to occur if the tax rate differential is small. Moreover, our results suggest that in a world with dispute resolution mechanisms in place, a minimum taxation, higher administrative costs for mutual agreement procedures, or higher audit probabilities affect the transfer pricing decision, i.e., the profit shifting behaviour of MNEs.

This paper studies how corporate tax hikes transmit across countries through multinationals' internal networks of subsidiaries. We build a parsimonious multicountry model to underscore two opposing spillover effects: While tax competition between countries generates positive investment spillover, intra-firm production linkages predict negative spillover. Using subsidiary-level data and exogenous corporate tax hikes, we find that local business units cut investment by 0.4% for a 1% increase in foreign corporate tax. This result highlights the importance of production linkages in propagating foreign tax shocks, as the supply-chain-induced negative spillover dominates the positive spillover effect suggested by the conventional wisdom of tax competition.

The COVID-19 pandemic affected most companies' profits negatively, but some companies did exceptionally well, recording excess profits during the pandemic. In this paper, we estimate the scale of the excess profits and the tax revenue potential of an excess profits tax, an additional tax levied by governments on companies’ excess profits. To estimate excess profits, we develop a trend-adjusted average earnings methodology. We apply the methodology to firm-level consolidated Orbis data to estimate that large multinational corporations with subsidiaries in the EU generated excess profits of $447 billion in 2020 (42% of their total profits in 2020). Using country-by-country reporting data, we estimate the excess profits arising from each EU member state and find that EU member states could together raise $6 billion with an excess profits tax of 10%.

Many governments have engaged in policy experimentation in various forms to resolve uncertainty and facilitate learning. However, little is understood about the characteristics of policy experimentation, and how the structure of experimentation may affect policy learning and policy outcomes. We collect comprehensive data on policy experimentation conducted in China over the past four decades. We find that the experimentation exhibits two characteristics that complicate policy learning. First, about 90% of the experiments exhibit positive sample selection in terms of a locality’s economic development. Second, promotion-driven local politicians allocate more resources to ensure the experiments' success, and such effort is not replicable when policies roll out to the entire country. These results suggest that, while China’s bureaucratic and institutional conditions make policy experimentation possible at an unparalleled scale, the complex political environments can also limit the scope and bias the direction of policy learning.

A burgeoning literature recognizes that the efficacy of the state is crucial for economic growth and citizen welfare. However, much of that literature abstracts away from the institutional details underlying state capacity. We develop a theory that provides a working definition of state capacity and how it is provided and maintained. We conceive of the state as a knowledge hierarchy, or an information-processing institution that passes problems up a set of organizational layers until a layer with the required expertise solves it. Knowledge hierarchies are costly to establish and operate, and politicians differ in policy preferences and public goods valuations. We embed this structure in a simple political economy framework, where politicians may idle parts of the state depending on electoral prospects, thus reducing output. In conjunction with high partisanship, this gives the state designer incentives to distort the state away from efficient levels of capacity and specialization.

Policymakers routinely make high-stakes funding decisions. Assessing the value of a program is difficult and may be affected by bounded rationality. In two experiments with U.S. policymakers and the general public, we find that respondents’ valuations of programs are inelastic with respect to the program’s impact. We design and test two portable decision aids—one that compares programs side-by-side and another that aggregates multiple features of impact into a single metric. The two decision aids increase elasticity by 0.20 on a base of 0.33 among policymakers and by 0.21 on a base of 0.21 among the general public. We provide evidence that the cognitive difficulty of translating impact-relevant information into policy decisions helps explain our findings.

Measuring product differentiation and substitutability is a key concern in the analysis of consumer demand. We develop and analyze a general yet tractable model of random choice with product differentiation in a multi-attribute setting. We show the analyst can separately identify vertical and horizontal product differentiation from binary comparison data alone. We characterize the binary choice rules that arise from our model using four easily understood axioms. In multinomial choice, we show the intersection of our model with the classic random utility framework yields random coefficients with an elliptical distribution. We provide applications to consumer demand and choice under risk.

We study the generation, transmission, and effects of causal narratives - narratives which describe a (potentially incorrect) causal relationship between variables. In a controlled experiment, we show that exogenously generated causal narratives manipulate the beliefs and actions of subjects in ways predicted by behavioral theory. We then show how to `grow' these types of narratives organically by asking subjects who observe a dataset to advise future subjects about action choices. Homegrown causal narratives are more likely to be shared and more influential than other types. Finally, we show that factual, statistical information does not eliminate the power of causal narratives.

Individuals often change their decision-making in response to complexity, as has been discussed for decades in psychology and economics, but existing literature provides little evidence on the general characteristics of these processes. We introduce an experimental methodology to show that in the face of complexity, individuals resort to "procedural" decision-making, which we categorize as choice processes that are more describable. We elicit accuracy in replicating decision-makers' choices to experimentally measure and incentivize the choice process' describability. We show that procedural decision-making increases as we exogenously vary the complexity of the environment, defined by the choice set's cardinality. This allows for procedural reinterpretations of existing findings in decision-making under complexity, such as in the use of heuristics. Participants' meta-choices provide suggestive evidence of the welfare consequences of procedural choice.

How much do an individual’s choices impact health in the long-run? Effectively combating public health concerns like the ongoing obesity epidemic require a clear understanding of the cumulative effect of health behaviors on the evolution of health. Using a novel dataset which follows individuals over a period of 30 years and collects detailed data on their medical conditions and health behaviors, we document the long-run relationship between health, weight, and health behaviors. We develop a structural life-cycle model that incorporates key components of health behavior (exercise, caloric intake, and diet quality) and ex-ante fixed heterogeneity (e.g., genes, childhood conditions, etc.). Despite fixed heterogeneity playing an important role, we show that health behaviors in adulthood are central in explaining the maintenance of good health over the life cycle.

This paper studies the effects of the privatization of the pension system on workers’ reported earnings, employment and retirement behavior, and income in old age. We analyze a cohort-based reform to the pension system in Uruguay that transitioned from a pay-as-you-go system with defined benefits into a mixed system with retirement accounts. We find significant labor supply responses to the privatization. First, workers in the privatized system are more likely to be employed in their fifties. Second, workers in the privatized system report significantly higher earnings early-on, seemingly driven by lower tax evasion. Regarding income in old age, we find little differences on income and poverty rates across the two systems. However, decades later we find that a significant share of workers opted-in for a de-privatization option, especially among those who did not choose the most profitable savings option or whose career profiles favor defined benefits formulas.

Decisions about the level at which to govern are among the most important made by states, yet our understanding of decentralization remains limited. In this paper, I provide empirical evidence on the importance of administrative capacity in decentralization. I focus on Ugandan reforms in 2009-10, and use rich data on sub-district units to follow them in a difference-in-differences framework. By comparing across different outcomes in a theoretically motivated way, and by comparing “parent” to “new” districts, I find evidence of the importance of both administrative distance and administrative capacity. I also present direct evidence on post-split decreases in administrative capacity for newly created districts, and argue that this channel is the most important in explaining post-split worsening of infrastructure and economic growth for newly formed districts. These findings provide rigorous, empirical evidence on the importance of implementation in reform, and the consequences that may result if it is lacking.

Many developing countries revised their high import tariff regimes in last three decades and replaced it with VAT. I utilize kinks produced by minimum value addition thresholds to estimate evasion of VAT post-importation. I use administrative data set for the universe of VAT returns filed in Pakistan for tax years 2009 to 2016 to estimate evasion by firms exclusively engaged in imports. I estimate an average evasion rate of nearly 50%. Using changes in thresholds over years, I provide evidence that this minimum tax collection is the best case scenario for revenue efficiency. The firms show strong bunching at or below threshold with about 40-60% of the firms showing bunching behavior. These results show that importers have strong tendency to misreport their sales which implies that VAT’s impact on informal sector is only minimal absent these arbitrary thresholds.

Delegating tax collection to informal leaders could raise revenue but risks undermining their local accountability. We investigate this possibility by exploiting whether city chiefs in the DRC were randomly assigned to collect property taxes. To measure accountability, we study the other side of the social contract, i.e., chiefs' distribution of resources during a government-run cash transfer program. Consistent with citizens' preferences, chiefs who collected taxes allocated more program benefits to poorer households, making fewer inclusion and exclusion errors. We find no difference in the likelihood that they pocketed benefits or allocated them to family members. Overall, citizens updated positively about chiefs who collected taxes. We provide evidence that collector chiefs were more likely to target poor households because tax collection allowed them to learn about household needs. In contrast to concerns of `decentralized despotism, we find evidence of accountability benefits from delegating tax responsibilities to local leaders in low-capacity states.

This paper presents Distributional National Accounts (DINA) for Germany from 1992 to 2017. The DINA method established by Piketty et al. (2018) combines national accounts, individual income tax returns and SOEP household survey data to build series on the distribution of pre- and post-tax national income. We discuss income composition across the pretax income distribution and the age distribution. Further, we explore who benefited from income growth between 1992 and 2016 and compare pretax and disposable income growth patterns. Lastly, we highlight methodological issues that arise from linking microdata with national accounts, a crucial step in the DINA method, and their implications for inequality estimates.

This paper combines detailed administrative records on the universe of the Dutch population with national accounts aggregates to provide a thorough description of income inequality before and after taxation and government spending. Accounting for domestic and foreign retained earnings has a substantial impact on inequality, raising the top 10% share of pre-tax national income from 29% to 31%. Overall, the tax system is regressive due to high consumption taxes and a low tax burden on capital income. The entire reduction in inequality - the top 10% income share falls to 26% - comes from government spending that is targeted at the bottom of the distribution.

We introduce a new theory and new estimation method for optimizing frictions with a piecewise linear constraint. Allowing frictions to depend on observables, we estimate why agents do not behave as standard frictionless models predict. Our methods are not limited to public finance and apply to a general class of mixture models and any of the four possible piecewise linear constraints, 1) slope increase (convex kink), 2) slope decrease (concave kink), 3) intercept increase (convex notch), or 4) intercept decrease (concave notch). We demonstrate these methods in three of these four settings. Individual income tax returns with a, 1) convex kink and, 2) concave kink implied by the EITC. New Jersey real estate transfer taxes with a 3) convex notch. We document which covariates account for a substantial share of optimizing frictions and provide elasticity estimates that explicitly control for optimizing frictions.

Many recent attempts have been made to measure the elasticity of taxable income (ETI), spanning many countries and time periods. Existing methods have, however, generally not considered that the ETI is not a primitive parameter, but rather depends on disparate aspects of a tax system. This raises questions about the external validity of ETI estimates based on data in one country (and period) for the ETI in other countries with different tax systems. We propose a new approach to estimate country-specific and time-specific ETIs with cross-country panel data, based on the correlated random coefficients model, that explicitly express the ETI as a function of structural determinants of the tax system such as citizens’ views about the government’s role in redistribution and the acceptability of tax evasion. We also develop methods for extrapolating estimated behavioral elasticities to countries that do not levy a particular tax, using wealth taxation as an example.

The German subsidy for rooftop solar panels is a cornerstone in the global efforts to deploy renewable energy. This paper exploits kinks in the nonlinear subsidy to estimate adopters' responses at the intensive and participation margin. The estimates show that, compared to a linear scheme, the government's subsidy reduces costs by 0.14%; an optimal nonlinear scheme would more than triple this gain. Methodologically, the paper proposes a novel semi-nonparametric estimator. It enables the evaluation of nonlinear taxes, subsidies, or prices when existing kink and discontinuity methods are inapplicable because agents simultaneously respond at the intensive and participation margin.

This paper shows that ubiquitous bunching estimators commonly used in the analysis of tax policy can be viewed as approximating experimental quantile differences. It characterizes assumptions and bias associated with bunching reformulated as a simple ratio estimator. Finally, the paper considers the extent to which distributions of potential income can be used as direct inputs to general welfare and tax optimality analysis.

Unexpected crises, such as armed conflicts, natural disasters and pandemics require immediate government interventions to protect the population. In democracies, policy makers need to balance the effectiveness of their intervention with respect for civil liberties, such as the freedom of movement and the right to privacy. What determines citizens' political approval of restrictive state interventions in a crisis? To answer this question, we explore the early moments of the COVID-19 pandemic in Germany, a crisis that especially highlighted tension between civil liberties and public health objectives. We exploit variation in mobility restrictions across states to estimate a difference-in-differences (DID) model and a DID-event study analysis. Our results show that stay-at-home orders increased individuals' satisfaction with democracy by 18% of a standard deviation. Turning to the mechanism, we show that the effect strongly depends on the perceived health threat and past exposure to an authoritarian regime.

The Norwegian government developed a compensation program for enterprises to sustain its business activity during the corona crisis. The amount of the compensation was scaled by the size of the revenue drop and relied on firms’ self-reported revenue. We analyze the reporting behavior of businesses in response to the program. We find strong evidence of the presence of strategic misreporting, but to a very limited extent. In total, we find that the actual support paid out is 5-10 percent higher than what it should have been, due to strategic misreporting.

Short-time work (STW) programs held a central stage in the policy response to the pandemic in Europe, subsidizing temporary reduction in hours worked. While there is a growing literature on their employment effects, it has remained silent on moral hazard. Yet, its rapid expansion and large upscaling likely fueled moral hazard. This papers intends to quantify behavioral responses to STW in the context of France during the pandemic. Using exhaustive establishment-level and worker-level data on STW take-up and quasi-experimental variations in employer contribution to the program, I study two types of behavioral responses (i) misreporting and (ii) changes in real economic behavior. I estimate a bunching response to a discontinuity in cost along the wage distribution which stems from a reporting margin only. Using a variation in employer contribution across industries, I quantify a misreporting response as well as real adjustments of labor demand.

Evaluations of business support programs often face the problem of accurately predicting which firms leave the market. In this paper we first show that a machine learning model better predicts firm exits than conventional methods. This improvement in prediction quality stems from both the estimation method and the use of high-dimensional firm data. Second, we show that firms with a lower predicted exit probability were more likely to use COVID-19 related state-support. However, since the firms with the highest predicted exit probabilities benefited most from the support we conclude that market cleansing was distorted, although less than previously thought.

This paper characterizes efficient tax subsidies for charitable contributions, and considers the properties of potential reforms. Contributions are underprovided in the absence of subsidies, and are misdirected if subsidies fail to account for the net costs that donors incur. It is costly for prospective donors to identify high-quality giving opportunities and to act on the basis of this knowledge, so there will be too few high-quality contributions if all giving receives the same tax treatment. A more efficient alternative is to offer generous tax subsidies that are partially or entirely recouped if recipient organizations subsequently experience precipitous contribution declines.

The permanent income hypothesis states that agents perfectly smooth consumption given a large, anticipated shock to income. Testing these implications is difficult given the endogenous nature of income and payment timing. We leverage exogenous variation in military bonus size and timing matched with donations from a large workplace charitable drive where soldiers contribute via payroll deductions during a fixed open enrollment period. Our findings show that soldiers are 5 to 10 percent more likely to contribute if they receive their bonus during the open enrollment period. We also show that the effect diminishes with age and bonus experience.

This paper analyzes donation behaviors from the perspective of religious beliefs. Using a transaction-level dataset from an Asian economy, we show that higher income uncertainty predicts more donations, especially for religious donations, and after negative income uncertainty and health shock. This pattern is inconsistent with existing explanations of donation, but can be explained by a “spiritual insurance” channel, whereby donators believe that giving reduces the probability of the bad state. Indeed, we find that those who donate to non-local religious organizations reduce their insurance purchases, suggesting that “spiritual insurance” channel can be influential for donation and the insurance market.

Organizations facilitate commitments to long-run giving and large gifts. Despite making a commitment to give, individuals may not donate. This study asks, what affects follow through? We partner with a charity whose donors commit to give 1% of their income for life. Donors drop out of the donation pathway at two key points: (1) after pledging but prior to the first donation and (2) at donation milestones, e.g. 6, 12 and 24 months. Using a field experiment design, we randomly assign donors to receive a recommitment request or no request, at these two key points, to see how this affects donations. Those who intend to follow through on an extreme giving promise don’t need a prompt of that promise. For those who have not started donating, prompts increase the probability of donating and amount donated. Prompts are ineffective for those who already started their commitment.

There seems to be more to voting than policy. Nevertheless, models of voting tend to focus on aggregate policy preferences. I model voting for redistribution under uncertainty over future incomes with imperfect memory and anticipatory utility. Imperfect memory gives votes meanings as signals of future income prospects. Anticipatory utility motivates concern over the meanings of votes. Voting becomes self-signaling: Voting for low tax rate is consistent with the desirable belief of high future consumption. Voting and policy preferences diverge and voting does not aggregate policy preferences: fewer voters vote for redistribution than prefer redistribution. Higher income risk increases the value of self-signaling which decreases the demand for redistribution. If voters do not perceive themselves as pivotal a policy trap where a decisive coalition votes against its best interest may arise.

This paper argues that political and market concentration level explains why developing economies often under-invest in institutional infrastructure and legal capacity. Economic growth challenges this equilibrium and incentivizes rulers to invest in institutional infrastructure complementary to physical infrastructure. Rulers make joint investments to expand market entry and size if they can secure greater rents and preserve institutions favoring concentration. Instrumenting market concentration level with the share of coal mining industry in local industrial output, the difference-in-differences analysis of Chinese data from 1997 to 2006 shows that the fiscal expenditure ratio of physical to institutional infrastructure rose 42% faster in provinces with market concentration indexes in the top quartile in 2000 (the year before China acceded to the World Trade Organization). The paper also presents a theoretical model proposing that investment in physical infrastructure rises faster than institutional infrastructure when the market concentration level increases.

Low criterion objectivity increases the scope for political-strategic use of interregional grants by allowing grants to be tailored to specific political interests. Using Danish large-scale 2020 equalization and grants reform as a case, we show that several newly-introduced grants have low criterion objectivity. We find that these grants correlate strongly with constituencies where the Incumbent (Social Democratic Party) and its parliamentary supporters and the reform coalition are well represented before reform. We do not find a robust relationship between party support and changes in grant schemes based on objective criteria. Leveraging polling station data we finally find strong positive correlation between new non-objective grants and votes for the incumbent and its parliamentary supporters in subsequent 2022 parliamentary elections. The paper contributes a new argument to the literature, points out the importance of objectivity for strategic use of grants and tests the argument empirically with new data using mixed methods.

Even under an authoritarian rule, the stability of power rests, at least to some extent, on popular consent. We study a global game of regime-change, where an authoritarian government can influence popular support via policy making. Citizens can challenge the government via a riot, whose chances to succeed increase with popular discontent, which is unknown to all. We show that public, unbiased information about consensus can be ex ante beneficial for the regime. Furthermore, we provide a full characterization of the regime's optimal policy. Our results provide a consistent explanation for the heterogeneous forms of limited media freedom often observed in modern non-democratic regimes and widely documented by the empirical literature.

In this paper, I consider optimal minimum taxation and the optimal taxation of intellectual property (IP) income in a setting, in which research and development (R&D) activities of multinational enterprises (MNEs) are suboptimally low due to positive cross-country spillovers resulting from R&D activity by MNEs. The R&D incentives set by an MNE's host country are inefficiently low in equilibrium. Allowing MNEs to access tax havens may help to overcome this inefficiency because it shelters MNEs' profit from foreign taxation. This forces foreign countries to contribute to the R&D incentive for domestic MNEs. In such settings, minimum taxes serve as a tool to fine-tune tax haven access. Regardless of the welfare-effect of R&D, a strictly positive minimum tax is optimal for each country. Uncoordinated minimum taxes may be excessively high, if R&D investment has a strong impact on productivity. Under certain circumstances, IP boxes are a welfare-improving substitute for tax havens.

The Global Minimum Tax (GMT) is applied only to firms above a certain size threshold. We propose a simple model of tax competition and profit shifting by heterogeneous multinational firms to evaluate the effects of this partial coverage of the GMT. A non-haven and a haven country are bound by the GMT rate for large multinationals, but can set tax rates for multinationals below the threshold non-cooperatively. We show that the introduction of the GMT with a moderate tax rate increases tax revenues in both the non-haven and the haven. A gradual increase in the GMT rate, however, triggers a sudden change in equilibrium from a uniform to a split tax rate, at which tax revenues in the non-haven decline. A gradual increase in the GMT coverage weakly raises world tax revenues, but may benefit the non-haven and harm the haven. We also discuss the quantitative importance of these results.

Since 2016, Country-by-Country reporting has provided tax authorities with information about multinationals' worldwide activities. It has been hailed as a game-changer for corporate taxation, enabling tax authorities to target multinational firms with high profits in tax havens. We model Country-by-Country reporting as increasing both tax planning and audit costs for profit-shifting multinationals, where the latter depend on the share of profits held in tax havens. Then, Country-by-Country reporting makes shifting profits from a high-tax country to a tax haven relatively more attractive than shifting from a low-tax country. Under plausible conditions, profit shifting from high-tax affiliates may increase relative to no Country-by-Country reporting. We confirm these changes in profit-shifting patterns using a staggered difference-in-differences design. The opposing effects for low-tax and high-tax countries also explain the mixed findings of previous empirical evidence on Country-by-Country reporting.

Illicit financial flows are intrinsically difficult to detect as they cannot be measured directly. In this paper, we develop a general, indirect method of estimating the scale of illicit financial flows, the so-called abnormal flows method. The method utilizes a bilateral gravity model to estimate the expected volume of cross-border financial flows and quantifies the unexplained flows to offshore financial centers. We denote these flows as an indicator of illicit flows. Applying this method to data on foreign direct investment, portfolio investment, and bank deposits, we observe that Europe and America are the primary destinations for illicit financial flows. We estimate that in 2020, tax havens and secrecy jurisdictions resulted in US$53 billion and US$3.1 billion in forgone tax revenue, respectively. Our findings also indicate a 19 percent increase in illicit financial flows in tax havens despite the implementation of automatic information exchange regulations, highlighting ongoing gaps in the system.

Using administrative data on the globally connected super-rich in the UK, we study the effect of a large tax reform on migration behaviour. Prior to 2017, offshore investment returns for `non-doms' - individuals tax-resident in the UK but with connections to other countries - were untaxed. Non-doms are strongly concentrated at the top of the income distribution: 86% are in the UK top 1%. A 2017 reform brought long-stayers into the standard tax system, reducing their effective net-of-average-tax rate by 18%. We find that emigration responses were modest: our central estimate is that the emigration rate increases by 0.26 percentage points for a 1% decline in the net-of-tax rate, and we can rule out increases larger than 0.4 percentage points. Dispelling fears that the targeted taxpayers were able to circumvent the tax hike, we find large average increases in income reported and tax paid in the UK of more than 150%.

Using administrative data on all immigrant arrivals to Canada from 1982 until 2016, we investigate how personal taxation affects immigration decisions. Using variation in country-of-origin tax rates and a variety of identification approaches, we are not able to detect statistically significant effects of taxation on immigration decisions. Our results suggest that while taxes may affect location choices of the very top earners, they are unlikely to play a qualitatively important role for regular folks. �

Does the removal of property tax-related disincentives to relocate increase migration flows? Combining IRS county-to-county migration data from 1993 to 2020 and comparative case study designs, we estimate that the portability provision led to a robust increase in individual and household in-state county-to-county migration outflows in Florida relative to county-to-county migration outflows in all U.S. states but Alaska, California and Hawai'i by 201.3 to 612 households and 270 to 889.6 individuals and null effects on in-state migration net inflows. We interpret these findings as evidence that the portability provision has a non-negligible impact on migration flows with important policy implications. Revising property tax code through the portability provision can increase in-state mobility and alleviate the lock-in effect of assessment limits while retaining state population. Robustness checks imply a natural amenity mechanism rather than home-ownership rates driving migration flows in support of the benefit, not the capital, view of property taxes.

This paper presents a model that relates the efficiency of environmental regulatory competition to firms’ excess-entry. The two main results are as follows. First, if emissions from production remain localized, environmental regulations are too tight and firm entry is insufficient because tighter environmental regulation in one country will have negative external effects through exacerbating firm entry with increased emissions in other countries. Second, if emissions from one country spillover to a global scale, just as CO2 emissions lead to global warming, the effect of emissions on welfare is the same for all countries no matter which country emits, so the positive and negative external effects of tightening of regulation in one country just cancel each other out, resulting in efficient regulation levels and firm entry

Whether local governments effectively implement environmental policies is crucial to successfully address global environmental challenges. However, scholars are still unclear about when and why local implementation of environmental policies fails, epecially in developing countries where high priority is often given to economic development rather than environmental protection and local officials are held accountable upwards. We investigate how local governments implement environmental performance assessment policy (EPAP) in China. Based on a sample of prefecture-level cities from 2005 to 2015 and a difference-in-differences estimator, we find that the EPAP has effectively reduced pollution emissions. However, the pollution abatement effect only emerges in assessment years but not in non-assessment years, leading to strategic implementation of environmental policy at the city level. The strategic implementation behavior is more evident in cities where emission reduction task is unfeasibly high and where city mayors have no environmental work experience or professional expertise.

This paper studies the impact of subsidies for Plug-in hybrid vehicles (PHEV) on carbon emissions. I show that subsidizing innovations without considering consumer behavior can harm the environment. I provide descriptive evidence on charging instances of PHEV and combine it with a structural model of demand for new passenger vehicles to evaluate the market outcomes had subsidies for PHEV not been in place. I show that PHEV subsidies were used by consumers to purchase larger and heavier vehicles and that consumers of PHEV seldom charge their vehicle. Taking into account the observed consumer behavior, I find that the elimination of subsidies for PHEV would have led to a yearly reduction of 167,139 tons of carbon emissions which are equivalent to the yearly carbon emissions 52,916 households emit due to energy consumption.

We exploit tax holidays in response to the invasion of Ukraine to study how the pass-through of gasoline taxes varies with market power. We are able to observe both the number of competitors faced by a gas station and its markup over marginal cost. Our findings are consistent with a model in which multiple equilibria are sustained in markets with a low concentration of gas stations, leading to more heterogeneity in pass-through, while only a competitive equilibrium is sustained in markets with many competitors. We present evidence that those gas stations that are able to collude, and charge high markups, only passed about 60% of the tax cut on to consumer prices. Gas stations with low markups, instead, passed on virtually the entirety of the tax cut.

Pigouvian subsidies are efficient, but subsidies with limited durations are not Pigouvian. When using "time-limited" output subsidies, the optimal policy subsidizes output and investment, where investment subsidies separately correct for the limit. Because the change in production when the subsidy ends is a sufficient statistic for the optimal duration, we estimate this statistic using the US Renewable Energy Production Tax Credit for wind energy. Wind facilities reduce generation by 5-10% when the ten-year subsidy ends, demonstrating that time limits distort production even in inelastic industries, and suggesting that adapting to limits is key to improving industrial policy elsewhere.

We address the question of optimal insurance against idiosyncratic productivity and health risk over the life-cycle and how the design depends on the demographic structure. For the analysis, we develop a quantitative overlapping generations model. During the working period, households face a progressive income tax code, which encompasses contributions to social insurance, notably unemployment insurance and retirement insurance. In retirement, households receive a, potentially, progressive pension income, which is partially linked to past earnings. We parsimoniously parameterize tax, transfer and pension schemes and retirement age and optimize jointly over this entire system of life-cycle insurance and redistributive instruments. We apply the model to study optimal insurance in light of ongoing demographic change processes in Germany.

The COVID-19 pandemic increased public debt and changed the income distribution in many countries. We use a numerical simulation approach to derive optimal nonlinear marginal tax rates for the pre-crisis and crisis periods. Our work contributes to the literature in two main ways. First, our results are informative about the desirable extent of redistribution in lower-middle-income countries in normal times. The analysis considers relevant features of a developing country, such as informality. Second, we assess the need for changes in tax and benefit policies in the wake of shocks.

This paper develops a new approach to calculate country-industry-year-specific forward-looking effective tax rates (FLETRs) based on a panel of 19 industries, 221 countries, and the years 2001 to 2020. Beside statutory tax rate and tax base determinants, the FLETRs account for typical country-industry-specific financing structures as well as asset compositions. We show that effective tax rates suffer from significant measurement error when the latter information is neglected, owing primarily to inappropriately assigned asset weights to statutory depreciation allowances. Our empirical analysis exploits the substantial variation in FLETRs over time to provide estimates of the tax semi-elasticity of corporate investment in tangible fixed assets. Based on more than 24 million firm entity observations, our results suggest a statistically significant tax semi-elasticity of -0.41, which is at the lower end of previous findings. We further show that different sub-groups of firms respond very heterogeneously to tax incentives.

Governments often subsidize private R&D using both grants and tax incentives. This paper studies whether they are complements or substitutes. I take a difference-in-discontinuities approach to examine small firms in the United Kingdom and find that increasing tax credit generosity enhances the effect of grant funding on R&D, suggesting that they are complements. Financing constraints are likely at play. The effects are most substantial for firms that appear constrained, and by implementing another quasi-experimental research design, I find that the instruments are substitutes for larger firms, which are usually less constrained. Some alternative explanations can be ruled out.

We study how firms respond to differential tax rates based on size using administrative tax data in Lithuania. The existence of a notch in the tax schedule faced by corporations allows us to use bunching estimators to calculate the revenue (0.35), cost (-1.29), and profit elasticities (7.41). We then leverage on the panel structure of our data to provide insight on whether the the tax system has an effect on the growth path of firms. We find that firms located close to the notch in one year will report systematically lower revenue growth than other firms.

We leverage lottery and administrative data to study the effects of a large-sized housing program on educational and labor market outcomes. Brazil's My House My Life program heavily subsidizes homeownership, albeit in remote areas, for millions of low-income families. We document that children of lottery-winner households move to worse schools as measured by a wide range of indicators. Moreover, school absences increase significantly. Further analysis shows that children in lower-quality schools are those who present higher absence rates. Our findings have far-reaching implications highlighting the need to design and target housing policies to stimulate important drivers of social mobility.

This study examines the effectiveness of a place-based policy targeting rural areas in Japan to address regional economic disparities. The policy offers financial incentives to regions with significant declines in population and an aging population to increase capital investment and public services. Employing a difference-in-differences (DID) estimation, this study compares designated municipalities with those narrowly missing designation. The results show that the policy has had a statistically significant impact on social capital expenditure, as well as an overall increase in public expenditure. However, there was no significant impact on in-migrant population size or per capita income. Interestingly, municipalities with a higher aging population had smaller effects despite receiving more fiscal expenditures. This study contributes to the ongoing debate surrounding the effectiveness of place-based policies in rural areas.

We use a unique administrative dataset on housing transactions in Tehran to provide evidence on the incidence and distortionary effects of taxes on rental properties. We exploit a particular feature of the tax code in the Tehran rental market, where the tax-exemption threshold depends on the property’s size. Substantial bunching occurs below the tax cutoff, suggesting strong behavioral responses to the tax kink. We also find higher after-tax rents above the kink. Based on these variations, we develop a structural framework with property taxes and filing costs to estimate the price elasticities of housing size supply and demand. We estimate a mid-run (10-year) price elasticity of housing size supply of 1.36 and a price elasticity of housing size demand of -0.17. We find high, but incomplete pass-through of the rental tax, implying that renters bear most filing costs.

Previous research on school accountability has shown that the disclosure of school-level results of a national standardized student achievement test has a heterogeneous impact on student achievement across schools. This paper, highlighting a type of standardized test that has no stakes for students (called a national assessment), sheds further light on circumstances under which the information disclosure has a desirable impact on student learning. Specifically, utilizing an unanticipated disclosure of school-level results of Japan's national assessment occurred only in one of the prefectures in 2013 and treating schools in other prefectures as a control group, we show that the information disclosure has a significantly more positive impact on student achievement when the school has more schools in close proximity (i.e., neighboring schools). The result is robust to the consideration of other possible conditioning factors of the information effect such as school budget autonomy.