Research

Abstract: We show that increases in the share of workers reporting self-employment to the IRS are not associated with changes in firm-reported payments to "gig'' and other contract workers after 2005 but are driven primarily by self-reported earnings of individuals in the EITC phase-in range. We examine a regression discontinuity design that generates exogenous variation in tax rates at the end of the year after labor supply decisions are already sunk and find tax code incentives increase self-employment reporting conditional on actual labor supply. We show that reporting effects have grown over time as knowledge of the tax code spreads.

Abstract: The COVID-19 pandemic triggered a large and immediate drop in employment among US workers, along with major expansions of unemployment insurance and work from home. We use Current Population Survey and Social Security application data to study employment among older adults and their participation in disability and retirement insurance programs through the second year of the pandemic. We find ongoing improvements in employment outcomes among older workers in the labor force, along with sustained higher levels in the share no longer in the labor force during this period. Applications for Social Security disability benefits remain depressed, particularly for Supplemental Security Income. In models accounting for the expiration of expanded unemployment insurance, we find that the loss of these additional financial supports is associated with a drop in older adult unemployment rates and an increase in Social Security Disability Insurance claiming. Social Security retirement benefit claiming has rebounded to pre-pandemic levels, but has shifted from offline to online applications. 

Abstract: The COVID-19 pandemic and associated mitigation strategies exacted a large economic toll on large portions of the United States population. For older and disabled workers, the effects could be more persistent and fiscally costly than the impacts experienced by young, healthy workers due to the spillovers onto Social Security. We use Current Population Survey, Social Security administrative data on applications for retirement and disability benefits, and Google Trends data to assess the impact of COVID-19 on older adults age 50-70. We find that employment for this group dropped substantially more than would have been predicted prior to the pandemic: employment for 50-61 year olds was 5.7 pp (8.3 percent) lower, while employment for 62-70-year- olds was 3.9 pp (10.7 percent) lower. For people aged 50-61, unemployment and labor force exits due to reasons other than disability and retirement represented 63 and 30 percent of the employment decline, respectively. For those aged 62-70, the two largest components of the reduction were unemployment (50 percent) and retirement-driven labor force exits (30 percent). We find evidence of declines in reporting a labor force exit due to disability (4-5 percent), applications for disability insurance (15 percent), and Google search intensity for disability (7 percent). Retirement benefit claiming remains largely unchanged overall, though we find evidence that applicants substituted towards filing for benefits via the internet. We explore potential mechanisms and find evidence for both supply- and demand-side explanations.

Abstract:   This paper documents the extent and growth of gig work in the United States using Internal Revenue Service tax records. We discuss advantages of using tax data to learn about gig work and address several important methodological considerations. We find that around 12 percent of the 2018 workforce participated in contract-based gig work. Apart from the emergence of platform-based transportation work — which typically is a secondary earnings source for participants — the overall share of the workforce relying on gig work as a primary job has remained relatively constant over time. We document key differences in the work done by high- and low-income individuals.

Media coverage: LA Times National Review

Abstract: The Affordable Care Act (ACA) authorized the largest expansion of public health insurance in the U.S. since the mid-1960s. We exploit ACA-induced changes in the discontinuity in coverage at age 65 using a regression discontinuity based design to examine effects of the expansion on health insurance coverage, hospital use, and patient health. We then link these changes to effects on hospital finances. We show that a substantial share of the federally-funded Medicaid expansion substituted for existing locally-funded safety net programs. Despite this offset, the expansion produced a substantial increase in hospital revenue, reflected in an equivalent increase in operating surplus. We do not detect significant improvements in health outcomes as measured by patient mortality, although the expansion led to substantially greater hospital and emergency room use, and a reallocation of care from public to private and better-quality hospitals.

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

Media coverage: WSJ Vox Cato

Abstract: The Affordable Care Act (ACA) includes several provisions designed to expand health insurance coverage that also alter the tie between employment and health insurance. In this paper, we exploit variation across geographic areas in the potential impact of the ACA to estimate its effect on health insurance and labor market outcomes in its first four years. Our findings indicate that approximately 70 percent of the increase in health insurance coverage since 2013 is due to the ACA. We also find that these increases in health insurance coverage did not result in statistically significant changes in labor market outcomes.

Media coverage: Vox Stanford SIEPR Brief 

Abstract: Privatization has been shown to increase growth and profitability of public firms. However, effects on consumers are understudied. We study potential trade-offs in the US hospital sector where public control declined by 42% over 1983–2019. Private operators may improve hospitals’ financial performance, but a focus on profitability may adversely affect access to care for certain patients. Using national data across all hospitals and patients, we study 258 hospital privatizations over the 2000–2018 period. Private operators improve profitability so that hospitals generate a modest surplus, primarily by increasing mean revenue per patient. However, this is partly achieved by differentially reducing the intake of low-income Medicaid patients, who are typically less profitable than other groups due to lower reimbursement rates. While other patients appear to be absorbed by neighboring hospitals, Medicaid patients experience an aggregate decline in utilization at the market-level, which we interpret as a decline in access to care. Hospital privatization therefore partially offsets the benefits of providing publicly funded health insurance through Medicaid, and our estimates imply it is quantitatively important. The aggregate decline in Medicaid volume is detected only in more concentrated hospital markets, suggesting market power is a key driver. 

Abstract:  The rise of platform gig work in recent years has meant more workers have self-employment earnings, which are not subject to tax withholding but generally covered by information reporting on 1099 returns. In 2017, many gig workers suddenly and unexpectedly stopped receiving a 1099 for their work activity, providing an opportunity to study the impact of third-party reporting on tax compliance for gig workers. We first exploit exogenous variation in receiving a 1099 information return using a border-county design in two states where all workers continued to receive a 1099. We find no impact of receiving a 1099 on whether one files a tax return, but find that each dollar reported on a 1099-K increases self-reported profits by 17 cents. We then use the state-level filings to impute the size of the national online platform workforce in 2017 and 2018. We estimate that 770,000 gig workers did not receive information returns due to the change in third-party reporting practices by online platforms and that $560 million in profits went unreported on income tax filings as a result.  

Abstract: We study the extension of Unemployment Insurance (UI) benefits to groups of workers typically excluded from the UI system—new entrants, the self-employed, and gig workers—implemented as part of the United States policy response to the 2020 COVID-19 crisis. Although enacted at the federal level, the state-level implementation resulted in large differences in the roll-out of benefits across jurisdictions in practice. We exploit this cross-state variation to estimate the causal impacts of these UI expansions on the labor market earnings, education choices, and mortality of affected groups using a spatial regression discontinuity design. We find that every additional dollar in UI payments received by self-employed workers led to a 30 cent reduction in their earnings, and find bigger reductions among platform-based gig workers of 48 cents. More generally, we find larger earnings reductions among self-employed workers in industries less affected by the pandemic. We find that the reduction in work reduced mortality among older gig workers by 0.36 percentage points that can be plausibly linked to reduced exposure to COVID through work—but find no health benefits among the self-employed more broadly. 

Abstract: A growing number of American workers earn income through platforms in the gig economy which provide access to flexible work (e.g. Uber, Lyft, TaskRabbit). This major labor market innovation presents individuals with a new set of income smoothing opportunities when they lose their job. I use US administrative tax records to measure take up of gig employment following unemployment spells and to evaluate the effect of working in the gig economy on workers' overall labor supply, skill acquisition, and earnings trajectory. To do so, I utilize penetration of gig platforms across counties over time, along with variation in individual-level predicted propensities for gig work based on pre-unemployment characteristics. In the short run, I show an increase in gig work following an unemployment spell and that individuals are correspondingly better able to smooth the resulting drop in income. However, individuals stay in these positions and are less likely to return to traditional wage jobs. Thus, several years later, prime-age (25-54) workers' income lags significantly behind comparable individuals who did not have gig work available. Among older workers (55+), I find an increase in gig work corresponds to a postponement of Social Security retirement benefits and a reduction in receipt of Social Security Disability Insurance (SSDI).

Abstract: We document the dynamics of tax-based measures of work mediated by online platforms from 2012 through 2021. We present a measurement framework to account for high reporting thresholds on some information returns using returns from states with lower reporting thresholds to provide a more complete estimate of total platform work. Updating data through 2021 allows us to provide the most comprehensive estimates of the COVID-19 pandemic on tax filing behavior. We find that the number of workers receiving information returns not subject to the 1099-K gap increased dramatically during the pandemic, with least 5 million individuals receiving information returns from platform gig work by 2021, nearly all from transportation platforms. We present evidence that the availability of expanded unemployment insurance benefits resulted in many individuals who were platform workers in 2019 not reporting any self-employment income in 2020-2021. At the same time, other services done by platform gig workers increased dramatically by at least 3.1 million people between 2019 and 2021. Interestingly, the broader 1099-contract economy follows a different trend, declining during this period, suggesting the challenges for tax administration are largely concentrated among platform gig workers, at least through 2021.

Media coverage: NY Times

Abstract: We examine the universe of tax returns in order to reconcile seemingly contradictory facts about the rise of alternative work arrangements in the United States. Focusing on workers in the “1099 workforce,” we document the share of the workforce with income from alternative, non-employee work arrangements has grown by 1.9 percentage points of the workforce from 2000 to 2016. More than half of this increase occurred over 2013 to 2016 and can be attributed almost entirely to dramatic growth among gigs mediated through online labor platforms. We find that the rise in online platform work for labor is driven by earnings that are secondary and supplemental sources of income. Many of these jobs do not show up in self-employment tax records: approximately 44 percent of the overall growth in the 1099 economy comes from people who do not file self-employment taxes. Examining the relationship between 1099s and self-employment tax records more generally, we find that the previously documented increases in self-employment tax filings since 2007 are largely driven by workers without 1099s. We discuss implications of these findings for tax administration and measurement of alternative work using tax data.

Media coverage: CNBC HHS

Abstract: We use administrative tax data to provide a detailed examination of demographic and economic characteristics of self-employed individuals, how their share of the workforce has changed over time, and the implications of that change for benefits coverage. We find that essentially all of the increase in self-employment since 1999 is among individuals with little or no business-related deductions, who appear to be almost exclusively providing labor services. Furthermore, large differences in benefits coverage exist; the self-employed are less likely to be covered by health insurance or to participate in or make contributions to a retirement account than employees.

Abstract: Wikipedia has been successful despite its purely decentralized production. Given its size, it is important to know whether editors produce the articles in an efficient manner. I introduce a new measure of dynamic efficiency and use it to evaluate the ordering of Wikipedia's article creation. At any given point in time, editors have created roughly 85% of the maximum possible views. An article's probability of being created is significantly and substantially increasing in its relative number of connections (links from other Wikipedia pages). Frequent editors tend to produce highly-viewed articles while infrequent editors tend to produce highly-connected articles.