Journal Articles
Payment Shocks to Medicare Advantage Plans and Their Impact on Plan Generosity
with Niru Ghoshal-Datta and Michael Chernew
Forthcoming, Health Affairs
Beginning in 2024, CMS implemented version 28 (V28) of the Medicare Advantage (MA) risk adjustment algorithm, which altered payments to MA plans. This paper examines how these risk-adjustment changes, which were expected to reduce payments to most plans, affected plan premiums, cost sharing, and benefits. Our findings suggest that V28 did not significantly affect plan premiums, but was associated with modest increases in cost-sharing and reductions in some supplemental benefits, thus partially contributing to broader declines in supplemental benefit generosity in MA during this period. Overall, our results suggest that modest payment cuts are likely to lead to small reductions in non-premium benefits. However, the value of these foregone benefits to beneficiaries and the potential impact of larger cuts remain uncertain. Policy makers must weigh whether the fiscal savings resulting from reduced payments justify potential declines in benefit generosity, which remains higher than historical norms.
The Optimal Geographic Distribution of Managed Competition Subsidies
with Amil Petrin, Robert Town, and Michael Chernew
Forthcoming, RAND Journal of Economics
Paper (September 2026) · Monte Carlo code (September 2022)
Governments often subsidize consumers’ purchases from competing firms rather than provide goods directly. We study managed competition, where firms compete for consumers and subsidies that follow them. We develop a framework for choosing the optimal market-level subsidy schedule with heterogeneous consumers and endogenous firm prices and product characteristics. We apply it to Medicare Advantage, which offers private insurance as an alternative to Traditional Medicare. We estimate product-characteristic policy functions and solve for Nash equilibria in revenues to compute counterfactual equilibria under alternative subsidies. The consumer-welfare-maximizing budget-neutral schedule raises aggregate annual consumer welfare overall by $7 billion relative to current policy.
Shifting Tax Incidence: Evidence from the Washington State Cannabis Market
with Benjamin Hansen, Kendall Houghton, and Caroline Weber
Journal of Policy Analysis and Management, 45(3): e70041, 2026
Published version · NBER Working Paper
We study how prices respond when a 25% gross receipts tax remitted by cannabis manufacturers was eliminated in Washington state and the retail excise tax was simultaneously increased from 25% to 37%. Standard theory suggests that this change should have increased welfare for manufacturers, retailers, and consumers; instead, our analysis shows that the reform unexpectedly shifted benefits toward manufacturers at the expense of retailers and consumers, who faced higher tax-inclusive prices post-reform. We hypothesize that this outcome was driven by behavioral factors such as anchoring and loss aversion. Our findings add to a growing body of evidence that tax reforms can affect market outcomes in ways not predicted by standard economic models, offering a cautionary lesson for policymakers considering similar reforms.
Early Evidence on the Effects of Open Mobile In-App Payment Systems from South Korea
with Boyoon Chang
Journal of Competition Law & Economics, nhaf034, 2025
Published version · Paper (June 2024 draft)
Mobile app platforms are highly concentrated—Apple and Google each distribute over 75% of the total number of apps installed on the relevant devices. These platforms generally require developers to use a built-in system for processing payments with a commission rate of 30% for both the initial purchase of an app and any subsequent in-app purchases. In September 2021, South Korea became the first country to ban this lock-in; purchases made in the country may be conducted through any billing system developers wish. We analyze the short-term impact of this policy change (and Apple’s subsequent adaptation) on demand for apps and revenue using difference-in-differences techniques and data on apps offered through Apple’s App Store from a leading app analytics firm from January 2021 to December 2022. We find no evidence that the policy change generated substantive changes in South Korea’s app marketplace.
Capturing and Harvesting in Medicare Advantage Plan Design
Health Affairs Scholar, 2(6): qxae077, 2024
Published version
Consumers in health insurance markets have inertia stemming from the desire to maintain relationships with providers and other frictions involved in switching plans. In other markets that feature inertia, suppliers respond with pricing strategies that vary by market share: lowering markups to capture consumers when market shares are low and raising markups to harvest profits once market share has been established. I tested for this behavior in the Medicare Advantage (MA) market by examining how MA plan sponsors changed the financial terms of their plans in response to changes in market share from 2007 to 2021 using a first-difference model with fixed effects. I found evidence that plans increase premiums, copays, and out-of-pocket limits when market shares increase. The results imply that for every 1% increase in market share, plan sponsors subsequently increase out-of-pocket costs by 1% in the following year.
Changes in Rail Rates for U.S. Commodity Grain Shipments Over Time
with Wesley W. Wilson
Research in Transportation Economics, 102: 101359, 2023
Published version
The deregulation of the U.S. railroad industry has largely been considered a success, as costs and rates have fallen due to changes in the mix of traffic and industry consolidation. However, rates did not fall as quickly as costs, and since 2000 rates have been rising while many measures of cost have remained relatively stable. We investigate these changes in rail rates using a sample of agricultural shipments from 2000–2016. We provide evidence that even after controlling for changes in cost drivers such as fuel, the relationships between prices and determinants have changed over time, suggesting that railroad pricing rules have driven increases in rates faced by shippers.
Reducing Medicare Advantage Benchmarks Will Decrease Plan Generosity, But Those Effects Will Likely Be Modest
with Michael Chernew, Amil Petrin, and Robert Town
Health Affairs, 42(4): 479–487, 2023
Published version
Concerns that Medicare Advantage (MA) plans are overpaid have motivated calls to reduce MA benchmarks, the dollar amounts set by the Centers for Medicare and Medicaid Services (CMS) against which MA plans bid to set premiums and fund extra benefits. However, cutting benchmarks may lead to higher MA enrollee premiums and decreased plan generosity. We assessed the relationships between MA benchmarks and plan generosity and benefits. We estimated that a $1,000 per year decrease in benchmarks would lead to small increases in annual premiums of about $60 and increases in annual deductibles of about $27. Copays would also increase modestly, and the propensity to offer benefits would generally decline by less than 5 percentage points, with the greatest impact being on the availability of dental, hearing, and vision benefits. These results suggest that although cuts to MA benchmarks would adversely affect plan generosity, those effects would be modest.
Watching the Grass Grow: Does Recreational Cannabis Legalization Affect Retail and Agricultural Wages?
with Sichao Jiang
Journal of Cannabis Research, 4(1): 42, 2022
Published version · Paper (January 2022 draft)
Background. Over the past several years, cannabis has become legal for recreational use in many US states and jurisdictions around the world. The opening of these markets has led to the establishment of hundreds of cannabis production and retail firms with accompanying demand for labor, leading to concerns about spillover effects on wages from incumbents.
Methods. We study the markets for agricultural and retail labor in Washington and Colorado from 2000 to 2019 using differences-in-differences with synthetic controls. We employ employment data from the Quarterly Census of Employment and Wages, state-level demographic data from the US Census Bureau, and agricultural data from the National Agricultural Statistics Service. We use the least absolute shrinkage and selection operator (LASSO) for variable selection and classification and regression trees (CART) for chained imputation of missing values.
Results. We find little-to-no evidence of a significant difference in weekly wages per worker generated by cannabis legalization: the log of the weekly wage per worker decreases by 0.013 in Washington’s agricultural sector (p value 0.091) and increases by 0.059 in Washington’s retail sector (p value 0.606). Results in Colorado are qualitatively similar. These results are limited in part by the short post-legalization period of the data.
Conclusions. Cannabis legalization is unlikely to negatively impact incumbent agriculture or retail firms through the labor market channel.
Vertical Integration and Production Inefficiency in the Presence of a Gross Receipts Tax
with Benjamin Hansen and Caroline Weber
Journal of Public Economics, 212: 104693, 2022
Published version · NBER Working Paper
We quantify the effects of a gross receipts tax (GRT) on vertical integration for the first time. We use data from the Washington state recreational cannabis industry, which has numerous advantages including a clean natural experiment: a 25% GRT imposed on cannabis firms was subsequently replaced by an excise tax at retail. We find the short-run elasticity of vertical integration with respect to the intermediate good net-of-tax rate is −0.15 and the long-run elasticity is about twice as large. We find these incentives lead to large output losses – production increases by 23 percent when the GRT is eliminated.
The Effect of Cannabis Legalization on Substance Demand and Tax Revenues
with Boyoung Seo
National Tax Journal, 74(1): 107–145, 2021
Published version · Paper (January 2019 draft) · News coverage in the Wall Street Journal
Cannabis advocates argue that legalization will increase tax revenues. However, if legal substances are substitutes, cannabis revenues may cannibalize other taxes. We document substitution between legal cannabis products and alcohol and tobacco products using detailed administrative and scanner data from Washington State. We estimate a flexible demand system for legal substances and find legalizing cannabis leads to a 15 percent decrease in alcohol, mainly by liquor and wine, and 5 percent decrease in cigarette demand. Approximately 40 percent of Washington’s 2015 cannabis revenue was cannibalized from preexisting sources. We find that Washington’s current substance taxes, even after accounting for substitution, are on the upward-sloping side of the Laffer curve.
Federalism, Partial Prohibition, and Cross Border Sales: Evidence from Recreational Marijuana
with Benjamin Hansen and Caroline Weber
Journal of Public Economics, 187: 104159, 2020
Published version
Marijuana is partially prohibited: though banned federally, it is available to 1 in 4 U.S. adults under state statutes. We measure the size of the interstate trade generated by state-level differences in legal status with a natural experiment: Oregon allowed stores to sell marijuana for recreational use on October 1, 2015, next to Washington where stores had been selling recreational marijuana since July 2014. Using administrative data covering the universe of Washington’s sales and a differences-in-discontinuities approach, we find retailers along the Oregon border experienced a 36% decline in sales immediately after Oregon’s market opened. We investigate the home location of recent online reviewers of marijuana retailers and find similar cross-border patterns. By the end of Washington’s 2018 fiscal year, our results imply that Washington had earned between $44 million and $75 million in tax revenue from cross-border shoppers. These cross-border incentives may create a “race to legalize.”
Sharing the Sacrifice, Minimizing the Pain: Optimal Wage Reductions
Economics Letters, 196: 109503, 2020
Published version · Paper (August 2020 draft)
Faced with temporary revenue shocks, employers may implement wage reductions — a common response to COVID-19 for U.S. universities. I provide a framework for optimal reductions when labor is inelastic wherein a planner balances “minimizing the pain” and “sharing the sacrifice.” I show that for a broad class of utility functions the optimal schedule includes (1) a cutoff wage below which the reduction is zero and (2) weakly progressive reduction rates above the cutoff wage. I illustrate the results using data from a large U.S. university.
Taxing the Potency of Sin Goods: Evidence from Recreational Cannabis and Liquor Markets
with Benjamin Hansen, Boyoung Seo, and Caroline Weber
National Tax Journal, 73(2): 511–544, 2020
Published version · Preprint
Cannabis is legal to purchase for over 28 percent of U.S. citizens. A central argument used in public campaigns for cannabis legalization has focused on the tax revenue that legal cannabis markets could generate. Recently, some policy makers and politicians have debated switching from traditional ad valorem taxes to taxes on potency, aiming to reduce the potential externalities associated with highly potent products. In this paper, we construct a theoretical model to predict the implications of a potency-based tax in an environment with market power. We then estimate the demand for cannabis potency based on administrative records of sales and potency from Washington state. We finish by conducting counterfactual analyses comparing revenue and potency outcomes from potency-based taxes versus the traditional price-based taxes.
Environmentalism, Stimulus, and Inequality Reduction Through Industrial Policy: Did Cash for Clunkers Achieve the Trifecta?
with Wesley W. Wilson and Nick Wood
Economic Inquiry, 58(3): 1109–1128, 2020
Published version
The 2009 American Cash for Clunkers program, which subsidized consumers who scrapped old vehicles and purchased new vehicles, was promoted by appealing to multiple constituencies. We evaluate the policy and alternatives according to its stated goals: emissions reductions, economic stimulus, and reducing inequality. We calibrate a dynamic partial equilibrium portfolio model to match consumer expenditure data from 1998 to 2011 focusing on heterogeneity across cars and trucks. We find the program generated $0.17 in environmental benefits, $0.28 in consumer surplus, and $0.31 in net discounted additional spending per subsidy dollar. Since subsidies largely went to middle-income infra-marginal consumers, the program exacerbated consumption inequality. We evaluate alternative policy designs and find no policy which simultaneously improves all outcomes.
Early Evidence on Recreational Marijuana Legalization and Traffic Fatalities
with Benjamin Hansen and Caroline Weber
Economic Inquiry, 58(2): 547–568, 2020
Published version
Over the last few years, marijuana has become legally available for recreational use to roughly a quarter of Americans. Policy makers have long expressed concerns about the substantial external costs of alcohol, and similar costs could come with the liberalization of marijuana policy. Indeed, the fraction of fatal accidents in which at least one driver tested positive for tetrahydrocannabinol has increased nationwide by an average of 10% from 2013 to 2016. For Colorado and Washington, both of which legalized marijuana in 2014, these increases were 92% and 28%, respectively. However, identifying a causal effect is difficult due to the presence of significant confounding factors. We test for a causal effect of marijuana legalization on traffic fatalities in Colorado and Washington with a synthetic control approach using records on fatal traffic accidents from 2000 to 2016. We find the synthetic control groups saw similar changes in marijuana-related, alcohol-related, and overall traffic fatality rates despite not legalizing recreational marijuana. ( JEL K42, I12, I18)
Governance Structure and Exit: Evidence from California Hospitals
with Wesley W. Wilson
Review of Industrial Organization, 53(1): 31–55, 2018
Published version
Most inpatient and emergency health care services in the U.S. are delivered by non-profit organizations. To understand the impact of policies that are designed to affect competitive outcomes in hospital markets, it’s important to understand whether the “non-profit” structure changes the behavior and competitive conduct of firms. Given the complexity of the product space within which hospitals operate, we focus on more easily interpreted decisions within the hospital market: entry and exit. Using comprehensive administrative data for the universe of California hospitals from 1980 to 2013, we document the observed entry and exit behavior. We estimate flexible exit policy functions and demonstrate a difference in behavior between for-profit and non-profit firms that exists after accounting for several observable characteristics of hospitals. We find differences in observed behavior: this is a finding that strongly suggests that there are differences in the underlying objective function of the various firms.