We investigate the economic value of cash management. In the legal marijuana industry, where only half of businesses have access to cash management services from a financial institution, we examine dispensary profitability using administrative and survey data. Our results show that businesses with cash management charge higher retail prices (8.3%), pay lower wholesale prices (7.3%), and have higher sales volume (19%). Together these advantages create a 40% increase in profitability. These results support our model in which a clientele effect and administrative costs drive profitability regardless of whether national banks, credit unions, or Fintech provide the cash management functions.

Brigham Young University

How large are the gains from knowledge sharing among co-workers? What frictions prevent the spread of information, and what management practices overcome these frictions? We conducted a field experiment with three active treatments in a sales company to address these questions. (1) Encouraging workers to talk about their sales techniques with a randomly chosen partner during short meetings substantially lifted average sales revenue during and after the experiment. (2) Worker-pairs given incentives to increase joint output increased sales during the experiment, but not afterwards. (3) Worker-pairs given both treatments realized small incremental sales gains over the meetings treatment alone. Providing encouragement for workers to initiate conversations resulted in knowledge exchange; incentives were insufficient. Our results highlight the importance of managerial interventions in facilitating knowledge spillovers.

New York University, Western Economic Association, Society for Institutional and Organizational Economics

We review the recent developments in the bunching literature, both when bunching is presented in the outcome variable and when it is presented in the treatment variable. We discuss issues related to identification, estimation, practical considerations, and suggest directions for future work.

Urban Economics Association Meeting

Bunching-type estimators use mass points in an observed distribution to estimate elasticities, such as the elasticity of taxable income with respect to the net-of-tax rate. We clarify the assumptions necessary to identify these elasticities using a single budget constraint that may contain any number of kinks and notches. Next, we derive the partially identified set for the elasticity and relate bunching estimators to censoring models. The result is an updated method that combines the insights of the bunching-type approach and a depth of knowledge from a vast econometric literature. Our approach results in quantitatively different estimates of the compensated elasticity of taxable income with respect to net-of-tax rate than bunching estimators in both Monte Carlo experiments and the context of the earned income tax credit.

University of Michigan, Dartmouth College*, American Economic Association*, International Institute of Public Finance*, National Tax Association, UCSD Workshop on Bunching Estimators and their Applications*, European Economic Association European Meeting of the Economic Socieity*.

We provide a method to track active prevalence of COVID-19 in real time, correcting for time-varying sample selection in symptom-based testing data and incomplete tracking of recoveredcases and fatalities. Our method only requires publicly available data on positive testing rates incombination with one parameter, which we estimate based on a representative randomized sample ofnearly 10,000 individuals tested in Utah. The method correctly predicts prevalence in two state-wide,representative randomized testing studies. Applying our method to all 50 states we show that trueprevalence is 2–3 times higher than publicly reported.

We quantify the intended and unintended consequences of increasing information disclosure by firms to the IRS using the 2011 redesign of IRS Form 3800. Our empirical strategy leverages novel variation in exposure to the redesign created by exogenous differences in tax years across firms. We find some evidence that the redesign was successful at increasing compliance after 2011, the intended consequence. We find, however, an unintended consequence of this redesign is that firms reported 23% more carry-forward general business credits in 2011. In this way, the increased disclosure in 2011 cost between $1.4 and $1.8 billion in corporate receipts.

National Tax Association Meeting*

We develop an urban model that incorporates: (1) heterogeneous sites; (2) fiscal and urban wedges; and (3) an endogenous number of cities, i.e., the extensive margin of urban development. Within- and across-city decreasing returns to scale cause agents to perceive their city as being too large in the socially optimal allocation. As a consequence, in equilibrium the largest cities on the most amenable sites are undersized, whereas the smaller cities on less amenable sites are oversized. We propose a test for optimal city size with heterogeneous sites extending the Henry George Theorem.

NBER Summer Institute*, Urban Economics Association, American Real Estate and Urban Economics Assoication.

We evaluate the ability of the income tax to insure household consumption against income shocks. To do this, changes in insurance are decomposed into active insurance changes (due to changes in the tax code), passive insurance change (due to changes in the income distribution), and residual behavioral insurance changes. From 1967 to 2010, both active and passive insurance changes had large, partly offsetting effects, such that the overall effect on insurance was minimal. In particular, active insurance changes decreased insurance and passive insurance changes increased insurance suggesting that, tax policy and the nonlinearity of the tax code influence consumption inequality.

ZEW Public Finance Conference, International Institute of Public Finance, National Tax Association

Does market power encourage or discourage investment? This is an open question due to theoretical ambiguity and empirical difficulties. The answer to this question is particularly important in the hospital market where market power has increased dramatically since the 1990s and health care spending has continued to skyrocket. We exploit an investment tax shock and data on the universe of US hospitals to test whether investment increases more or less in concentrated markets. We find that firms in concentrated markets increased investment by 4.4 percentage points ($1.7 million on average) more in response to tax incentives than firms in competitive markets. Further, firms' investment responses monotonically increase with market concentration and in the most concentrated markets, investment increased by 5.5% ($2 million on average).

National Tax Association, University of Tubingen, FTC.

Investor-level taxation may distort mergers and acquisition decisions when capital gains are taxed at a preferable rate relative to dividends. The intuition is that if the value of a target firm's assets are discounted by the dividend tax rate, but the proceeds from their purchase are taxed at the preferred capital gains rate, then the wedge between the tax rates can create a discount on the acquisition, distorting the efficiency of the market. To test for the existence and effects of this tax discount on merger and acquisition behavior, we exploit quasi-experimental variation created by the Jobs Growth and Tax Relief Reconciliation Act of 2003, which equalized dividend and capital gains rates, eliminating the tax discount. We find that acquiring firms that do not use internal funds for other tax-advantaged purposes performed higher quality acquisitions after the discount was eliminated. We find supporting evidence that firms substituted away from inefficient mergers and acquisitions and towards shareholder payouts after the rates were equalized. These results confirm the existence of the tax discount prior to 2003 and suggest that re-implementing the same wedge between dividend and capital gains rates would destroy between $82 and $214 billion of the value of mergers and acquisitions in the U.S. annually.

Oxford Centre for Business Taxation, National Tax Association.

We use administrative data on the population of corporations to investigate the distortion of corporate taxes on private firms, which account for 99% of all corporations. In response to a 9% increase in corporate tax rates, firms decrease income by 8.9%---5.5% due to reporting differences (e.g., tax shields) and 3.4% due to real differences (e.g., investment). We find differences in agency costs, investment frictions, and tax aggressiveness between firms using cash and accrual accounting and among small and large firms. Our estimates suggest that lowering the corporate tax rate from 35% to 25% would increase firm value by 16%.

Oxford Centre for Business Taxation, National Tax Association, Federal Reserve Board Applied Microecnomics Seminar, University of Oregon.

Typical censoring models have mass-points at the upper, lower, or both tails of an otherwise continuous outcome distribution. In contrast, we consider a censoring model with a mass-point in the interior of the outcome distribution. We refer to this mass point as "bunching" and use it to estimate model parameters. For example, economic theory suggests that for increasing marginal income tax rates, many taxpayers will report income exactly at the threshold where the tax rate increases. This translates into a censoring model with bunching at the threshold. The size of this mass point of taxpayers can be used to estimate an elasticity parameter, which summarizes taxpayers responses to taxes. This article introduces the command bunching , which implements new non-parametric and semi-parametric identification methods for estimating elasticities developed by Bertanha, McCallum, and Seegert (2022). These methods rely on weaker assumptions than what are currently made in the literature and result in meaningfully different estimates of the elasticity.

Using a firm-based dataset from a sales organization, we study employee responses to a staggered compensation shock, which reduced average take-home pay by over 7\%. We find minimal changes in sales, suggesting that real world compensation contracts may include features that limit extreme effort reductions. The compensation cuts did, however, contribute to a significant increase in the turnover rate of the firm's most productive employees. Changes in employee sentiment cannot explain the observed turnover responses, suggesting instead that productive employees' superior outside options drove their attrition.

Society for Institutional and Organizational Economics*.

We combine structural estimation with ideas from Machine Learning to estimate a model with information-based voluntary social distancing and state lockdowns to analyze the factors driving the effect of social distancing in mitigating COVID-19. The model allows us to estimate how contagious social interactions are by state and enables us to control for several unobservable, time-varying confounders such as asymptomatic transmission, sample selection in testing and quarantining, and time-varying fatality rates. We find that information-based voluntary social distancing has saved three times as many lives as lockdowns. Second, information policy effects are asymmetric: `least informed' responses would have implied 240,000 more fatalities by June 2020 while `most informed' responses would have saved 25,000 more lives. Third, our estimates suggest that contagion externalities from social interactions are large enough that a lockdown response could have been 25\% less costly for the median state and still saved an equivalent number of lives.

This project's aim was to generate an unbiased estimate of the incidence of SARS-CoV-2 infection in four urban counties in Utah. A multi-stage sampling design was employed to randomly select community-representative participants 12 years and over. Between May 4 and June 30, 2020, surveys were completed and sera drawn from 8,108 individuals belonging to 5,125 households. A qualitative chemiluminescent microparticle immunoassay was used to detect the presence of IgG antibody to SARS-CoV-2. The overall prevalence of IgG antibody to SARS-CoV-2 was estimated at 0.8%. The estimated seroprevalence-to-case count ratio was 2.4, corresponding to a detection fraction of 42%. Only 0.2% of individuals who had a nasopharyngeal swab collected were reverse transcription polymerase chain reaction (RT-PCR) positive. The prevalence of antibodies to SARS-CoV-2 in Utah urban areas between May and June was low and the prevalence of positive RT-PCR even lower. The detection fraction for COVID-19 in Utah was comparatively high. Probability-based sampling provides an effective method for robust estimates of community-based SARS-CoV-2 seroprevalence and detection fraction among urban populations in Utah.

We compiled serological SARS-CoV-2 antibody test data and prior SARS-CoV-2 test reporting from members of 9,224 Utah households. We paired these data with a probabilistic model of household importation and transmission. We calculated a maximum likelihood estimate of the importation probability, mean and variability of household transmission probability, and sensitivity and specificity of test data. Given our household transmission estimates, we estimated the threshold of non-household transmission required for epidemic growth in the population.

This paper clarifies issues debated by A. C. Pigou and Frank Knight about correcting inefficient use of congestible resources, focusing for concreteness on their original example of road congestion. Instead of government‐imposed Pigouvian access fees, Knight favoured access fees set by private toll‐setters. We consider the case of n≥2 congestible roads and an uncongestible road of arbitrary speed. Knight argued that in the case of a single congestible road, a private toll‐setter would always choose the toll that Pigou recommended, hence the allocation would minimize aggregate commute time without government meddling. We find instead that two or more toll‐setters would never choose Pigouvian tolls except in the special case of a sufficiently fast uncongestible road. Moreover, for uncongestible roads of slower speed, the allocation of motorists under Knight's proposal is almost never efficient. Whenever it is inefficient, motorists are strictly worse off when they pay tolls set by private firms instead of paying government‐imposed tolls, and aggregate toll revenue is also lower. Nevertheless, if the private sector does set tolls, then the full cost to motorists can be limited if the government provides an uncongestible alternative, such as a train, to offer potential competition along the same route.

UC Santa Barbara*, Society for Institutional and Organizational Economics.

State tax revenues continue, since the Great Recession, to experience elevated volatility relative to previous decades. The framework developed in this paper shows that the interaction between economic uncertainty and the riskiness of the portfolio the government holds drives tax revenue volatility. From 1970 to 2007 the average State tax portfolio became 7 percent riskier, and that trend continues through 2013. This observed increase in volatility and riskiness of tax portfolios is not sufficient to determine whether States are accepting unnecessary levels of risk. Specifically, States are constrained to accept additional volatility in exchange for additional expected tax revenues. To summarize this tradeoff, I develop a volatility index that estimates the efficiency of a State's tax portfolio. This index can be interpreted as the cost to States, in terms of lost expected tax revenue, from holding a tax portfolio off of their minimum variance frontier. I find that almost all States held riskier portfolios in 2007 than in 1970, but that 26 States held portfolios in 2007 that were closer to their minimum variance frontier than the portfolio they held in 1970. This suggests that the observed increase in tax revenue volatility is a consequence of States increasing their expected tax revenues and not due to undertaking unnecessary risks.

Natiaonal Tax Assocation Spring Meeting

We investigate the tax implications for the new recreational marijuana industry in the United States, which reached a size of $9 billion in 2017. We exploit administrative data from Washington state to evaluate market conduct, and we estimate the elasticity of supply to be 1.46 . In addition, we conduct a survey of marijuana producers and retailers in Colorado, Oregon, and Washington, calculating the elasticity of demand to be -1.84. We use these estimates to determine how much of the tax burden is borne by consumers. The answer depends on market conduct. In perfectly competitive markets, producers pay slightly more of the tax than consumers, but in a monopoly market consumers would pay most of the tax. Additionally, we calculate that the change in deadweight loss due to the tax is $63 million per year, or 48% of total marijuana tax revenues in 2015. This calculation, however, depends critically on estimates of consumption externalities.

International Institute of Public Finance

Despite expectations that state sales tax revenues would drop 8 to 20 percent in 2020, many state policy makers were surprised to see revenue increases during the pandemic. We also show that this windfall is associated with a reallocation of sales tax revenues away from urban areas and toward rural areas and spending shifts towards goods as well as e-commerce. Using a combination of administrative state tax data and a purpose-built consumer survey, we document three stylized facts. These stylized facts suggest that states may want to rely more on sales tax moving forward because the sales tax may have become more efficient, more equitable, easier to enforce, and exposes states to less tax revenue risk. In addition, these facts have implications for devolution and the apportionment of sales tax revenue across local governments.

National Tax Association Meetings

In the U.S., means-tested cash, in-kind assistance, and social insurance are part of a patchwork safety net, often run with substantial involvement of state and local governments. Take-up–participation among eligible persons–in this system is incomplete. A large literature points to both neo-classical and behavioral science explanations for low take-up. In this paper, we explore the response of the safety net to COVID-19 using newly-collected survey data from one U.S. state–-Utah. The rich Utah data ask about income and demographics as well as use of three social safety net programs which collectively provided a large share of relief spending: the Unemployment Insurance program, a social insurance program providing workers who lose their jobs with payments; the Supplemental Nutrition Assistance Program, which provides benefit cards for purchasing unprepared food at retailers; and Economic Impact Payments, which provided relatively universal relief payments to individuals. The data do not suffice to determine eligibility for all of the programs, so we focus on participation per capita. These data also collect information on several measures of hardship and why individuals did not receive any of the 3 programs. We test for explanations that differentiate need, lack of information, transactions costs/administrative burden, stigma, and lack of eligibility. We use measures of hardship to assess targeting. We find that lack of knowledge as well as difficulty applying, and stigma in the UI program each play a role as reasons for not participating in the programs.

Local governments across the U.S. heavily rely on the property tax to finance public goods and services, particularly K-12 education. According to data from the Census Annual Survey of State and Local Government Finances, property taxes accounted for over 70% of total local government tax revenue in 2019. Economists tend to favor property taxes due to the prospect of smaller deadweight losses arising from the inelasticity of housing supply relative to the supply of other tax bases. By contrast, regular surveys of homeowners point to the widespread unpopularity and perceived unfairness of the property tax compared to income and sales taxes. The role of property taxes in state and local government financing has become increasingly prominent as housing unaffordability has peaked in recent years. In 2022, more than 42.1 million U.S. households were cost-burdened, defined as spending more than 30% of their annual income on housing.

National Tax Spring Symposium

Rushes are a fundamental characteristic of the growth of many industries and cities. To explain these rushes, and better understand the mechanisms of growth, this paper develops a model centered on a new tradeoff between fundamentals and opportunities. Early growth in industries and cities depend critically on the opportunities they provide; whether from entrepreneurship human capital accumulation or land in cities. This analysis provides a blueprint for evaluating policies aimed at encouraging growth. These policies are particularly important in developing countries that are experiencing rapid urban growth both in current cities and the number of new cities.

Urban Economics Association Meeting

We show that land taxes are associated with higher density, neighborhood diversity, business formation, and other indicators of economic performance. To demonstrate this, we first estimate idiosyncratic land taxes (or subsidies) for over 2,000 counties in the US. We use the fact that property taxes are based on tax-assessed values rather than market prices. County assessors that, for idiosyncratic reasons, overvalue land relative to the market impose an implicit land tax. Consistent with these land taxes being idiosyncratic, we find substantial dispersion across counties in the US and within metropolitan statistical areas. Finally, we develop a model of land taxes and endogenous population to rationalize our results.

University of California Irvine, Georgia State University, Stanford University

NBER Summer Institute Urban Economics, National Tax Association Meetings

We investigate the impact of the federal Low-Income Housing Tax Credit on the diversity of income in neighborhoods. We collect data on LIHTC applications in Utah from 2000 to 2018 and compare neighborhoods in which developers’ LIHTC applications were accepted to neighborhoods where developers' applications were declined. We find that income diversity declined in neighborhoods with LIHTC developments. This decline is due to a reduction in the prevalence of the lowest-income households, not the highest-income households. We also find that the program increased the number of households across the income distribution, possibly indicating an increase in neighborhood desirability.

NBER Summer Institute Personnel Economics*

There is substantial variation in whether workplace training and mentorship programs are voluntary or mandatory. When programs are voluntary, many workers do not participate. We conducted a natural field experiment on a mentorship program in a sales call center where in one treatment arm, labeled the Mandatory-Condition, all subjects were either randomly assigned a mentor or not. A second treatment arm, labeled the Voluntary-Condition, required subjects to opt into the program before randomization into receiving a mentor. In the Mandatory-Condition, the mentorship treatment raised workers' daily revenue by 17% in their first two months of tenure. In the Voluntary-Condition, those who opted out of the program were substantially less productive than those who opted in, and treatment gains conditional on program participation were negligible. Comparing the conditions indicates that treatment effects are largest for workers who are most likely to opt out of participating in the program. We conclude that workplace programs can raise the productivity of lower performing employees but these workers may require inducements or mandates to participate.

Why do overconfident entrepreneurs fail to learn from frequent market feedback? Using two field experiments across almost 1,000 firms in Utah we follow for over a year, we explore the role of hindsight bias and causal misattribution. Both biases can potentially help sustain overconfidence psychologically, as hindsight bias creates a false memory of past mistakes while misattribution allows entrepreneurs to shift blame to external factors. We use two treatments to address biased memory and misattribution. First, under our "Error Reminder" treatment, entrepreneurs are shown past forecast errors to remove hindsight bias. Second, under our "Scientific Learning" treatment, we encourage entrepreneurs to develop a causal hypothesis about their firm and test this causal hypothesis empirically, to mitigate misattribution. We find that the Error Reminder treatment does not reduce overconfidence, because misattribution replaces hindsight bias to sustain overconfidence. In contrast, stronger engagement with hypothesis testing within scientific learning successfully reduces overestimation.

This paper investigates optimal tax policy when economic conditions are uncertain, creating risk in tax revenues. Optimal taxation with uncertainty balances costs from volatility and deadweight loss. Tax policy affects volatility through two channels; spreading risk between public and private consumption and hedging idiosyncratic tax base risk. Risk in tax revenues can be decomposed into economic uncertainty and systematic and unnecessary tax policy risk. This paper develops a volatility index that measures the cost of unnecessary risk associated with a tax portfolio. Using panel data from all US states for the years 1964--2013, I find that almost all states increased their portfolio risk during this period and half of these states exposed their tax revenues to more unnecessary risk. This suggests states could reduce their tax revenue volatility by rebalancing their tax portfolios without decreasing expected tax revenues.

Natiaonal Tax Assocation Meetings

Revenue volatility affects the welfare of U.S. states, which typically do not smooth their expenditures over the cycle but instead spend revenues as received. Between 2000 and 2014 U.S. state tax revenue volatility increased to 10.8% of revenues, up from 2.9% in the previous three decades. This increase was due to a combination of a rise in the volatility of state GDP, changes in tax policy that increased the exposure of tax revenues to state GDP, and other factors in the tax base. An Oaxaca-Blinder decomposition indicates that tax policy changes explain 59% of the increase in tax revenue volatility, and that increased state GDP volatility and all other changes account for 22% and 19%, respectively. This evidence implies that tax policy is responsible for much of the recent increase in state government revenue instability, and the welfare consequences that follow from it.

Stanford, UC-Berkely, UC-Irvine, University of Illinois, University of Kentucky, University of Nebraska, US Treasury, Amherst, University of South Carolina, Clemson, Utah Winter Business Economics Conference, International Institute of Public Finance, National Tax Association Meeting

We investigate team diversity and productivity in the setting of open source software. We find that team diversity is strikingly low compared to the contributor pool, with teams most frequently at the lowest level of diversity i.e., monoculture. To identify the causal effects of team diversity, we use teams' exposure to country-by-year variation in Internet access. The evidence is consistent with homophily, the preference to associate with similar people, leading to teams being trapped in inefficiently low-diversity states. Teams that escape the ``homophily trap" go on to higher diversity and increased project success relative to teams that do not.

We highlight the importance of signaling effects in determining whether public policy should be implemented at a decentralized or centralized level. For example, although a public policy may have the same direct effect if enacted at a state or county level, people may perceive these policies differently, leading to different indirect effects. We explore this mechanism using the patchwork of mask mandate orders in the U.S. from April to September 2020. State-wide mask mandates stimulate economic activity while also reducing COVID-19 case growth. Surprisingly, county-level mask mandates generally have the opposite effect, depressing economic activity. We argue that different unintended signaling effects can explain these differences in policy effects: households infer from county mask mandates that infection risks have increased in their local area and, therefore, socially distance more and spend less. In contrast, state mask mandates do not lead to similar local inferences, and thus overall, they stimulate the economy.

We use administrative tax data from 5 different countries to calculate the within-country corporate elasticity of taxable income and investigate differences between these estimates. Our estimates exploit the differential tax treatment of business income for firms earning positive and negative taxable income in a bunching framework. We develop two new estimators to overcome several challenges that are unique to the business context. We find meaningful differences in the elasticity, with Greece having the largest (1.2) and China having the smallest (0.30). The differences we find, however, are much smaller than the range found in the literature (0 to 5). This suggests that some of the difference in the estimates in the literature may be due to differences in method rather than fundamental firm-specific characteristics, e.g., industry or tax system-specific characteristics, e.g., level of credits and enforcement.

Oxford Centre for Business Taxation, National Tax Association Spring Meetings*

Homeowners face risk due to variation in annual property tax liabilities, which may result in financial distress and eventual mortgage foreclosure. By reducing the pro-cyclicality of property tax liabilities, we show that property tax limitations can expose households to greater systematic risk despite reducing intertemporal variation in tax amounts overall. We develop an innovative measure of tax policy risk using Arrow-Debreu securities and obtain simulated measures of risk that capture all of the key characteristics of states' property tax regimes. Using a state border discontinuity design and parcel-level data for the universe of U.S. residential properties, we show that a one standard deviation increase in tax policy risk ($\approx$ \$350) increases the probability of mortgage distress by approximately 0.24 percentage points. The magnitude of this unintended effect is somewhat larger than the increase in probability of mortgage distress associated with owning a home in disrepair and is approximately one seventh as large as the effect of moving between the third and fourth quartiles of the loan-to-value distribution (near the threshold for being underwater).

Michigan Tax Invitational (MTAXI)*, Urban Economics and Public Finance Conference*, Villanova*, Urban Economics Association conference*, National Tax Association*, International Public Finance*

Personal data markets have become ubiquitous. At the same time, the non-rivalry of data suggests that the social returns to personal data sharing will often exceed its private returns. Using a unique sequence of exogenous program evaluations for randomized COVID-19 testing among over 10,000 individuals in Utah, we analyze different tools to stimulate personal data sharing. We contrast the effectiveness of incentives for data sharing with mechanisms suggested by behavioral economics, including framing, image motivation, and identity. Our results suggest that incentives by themselves can easily backfire and are highly complementary with framing effects. Furthermore, image motivation and identity are an order of magnitude more effective in influencing data sharing than monetary incentives.