Working Papers

Time-Limited Subsidies: Optimal Taxation with Implications for Renewable Energy Subsidies (with Owen Kay)

Revise and Resubmit at the Journal of Political Economy

Pigouvian subsidies are efficient, but real-world subsidies are not Pigouvian when they have a limited duration. We show that such “time-limited subsidies” must be paired with investment subsidies to be efficient and that the change in production when a subsidy ends is the sufficient statistic for the optimal subsidy duration. We examine the US Renewable Energy Production Tax Credit for wind energy and find that firms reduce output by 5-10% when the ten-year subsidy ends, resulting in a loss of 6,000 GWh of energy annually—a figure that will increase as more turbines age out of subsidization.

Download Draft (Draft: Aug 2022)

2020: The Outstanding Third Year Paper Award

Capital Utilization in Production Function Estimation (with Michael Gmeiner)

A method to simultaneously estimate capital utilization rates and production functions is developed that exploits the fuel input choices of producers. Using Chilean data from 1979-1996 it is shown that the elasticity of output with respect to labor is systematically lower when accounting for capital utilization, and the elasticity of output with respect to capital is systematically higher. Total factor productivity is shown to be procyclical even when estimated independently of capital utilization. Output elasticities and capital utilization exhibit correlations with input quantities and investment which justify the validity of results.

Original SSRN Draft (Significant Revisions including Census RDC Proposal in Process)

Work in Progress

How Does Where You Live Change What You Think? Evidence from Volunteer Missionary Service for the Church of Jesus Christ of Latter-day Saints (with Tanner Eastmond)

$10,000 Pilot Funded and in the Field

As society grapples with questions about racism, an important question is how to change racial attitudes. This paper examines to what extent exposure to individuals of other races affects racial attitudes later in life. We do this in the context of missionaries for the Church of Jesus Christ of Latter-day Saints who spend 18-24 months immersed in the culture of their (geographically limited) missions. We survey former (Mormon) missionaries. Leveraging the plausible exogeneity of mission assignment within the United States to racial attitudes, results from the first wave of the pilot suggest that whereas volunteers assigned to white areas have similar racial attitudes of to national averages, those assigned to minority-rich areas show much less racial resentment later in life and manifest much more willingness to learn about race and racism in the United States.

Public Good Perceptions and Polarization: Evidence from Higher Education Appropriations (with Reuben Hurst and Andrew Simon)

We conduct two field experiments to understand drivers of polarization in the demand for government expenditure, using US state spending on public higher education as a concrete example. We first provide information to taxpayers about their own state’s graduation rate at public colleges, and its rank relative to other states. Information increases average preferred spending by 5 percent and closes the partisan gap in preferred policies by 32%. The reduction in polarization is due to differences in partisan reasoning, where Republicans are more likely to interpret a high graduation rate as a signal that their state is more productive. Information also affects how taxpayers communicate their preferred policies with elected officials. The treatments increase the likelihood that taxpayers, especially Republicans, write a (positive) message to a state representative of the opposing party.

(New Draft Coming Soon)

From Value Added to Welfare Added: A Social Planner Approach to Education Policy (with Tanner Eastmond, Nathan Mather, and Julian Betts)

Though ubiquitous in research and practice mean-based “value-added” measures may not fully inform policy or welfare considerations when policies have heterogeneous effects, affect multiple outcomes, or advance distributional objectives. In this paper we formalize the importance of heterogeneity for calculating social welfare and quantify the importance of heterogeneity in an enormous public service provision problem: the allocation of teachers to elementary school classes. Using data from the San Diego Unified School District we estimate heterogeneity in teacher value added over the student achievement distribution. Because a majority of teachers have significant comparative advantage across student types, allocations that use heterogeneous estimate of value added can raises scores by 34-97% relative to using standard value added estimates, with even larger gains if the social planner has heterogeneous preferences over groups. Because reallocations benefit students on average at the expense of teachers' revealed preferences, we also consider a simple teacher compensation policy, finding that the marginal value of public funds would be infinite for bonuses of up to 16% of baseline pay. These results suggesting that using information about effect heterogeneity might improve a broad range of public programs—both on grounds of average impacts and distributional goals.

Why Choose Career Technical Education? Disentangling Student Preferences from Program Availability (with Brian Jacob)

Recent research finds that participation in high school Career and Technical Education (CTE) can have substantial economic benefits, but the returns to CTE vary dramatically by field and across students. Some programs for some students are quite beneficial, but many are not. Unfortunately, there is no research on the determinants of CTE course-taking, and thus no guidance for policymakers seeking to improve student outcomes. In this paper, we utilize student-level longitudinal data on 11 cohorts of Michigan high school entrants to examine how students make decisions regarding CTE participation. We first document stark differences in access to and participation in CTE programs across gender, race and income groups. Next, we develop and estimate a discrete choice model to understand how both supply (accessibility) and demand (student preferences) factors contribute to these gaps. Our analysis yields three main findings: (i) the sex gap is driven entirely by differences in student preferences; (ii) the income gap is driven entirely by differential access to CTE programs; (iii) the race gap is driven by factors operating at the school level - that is, all students (regardless of race) at predominantly Black schools participate less in CTE, due to what we conjecture is a combination of supply and demand factors. Finally, our results suggest that the ability to take CTE classes without traveling to a different school building is a critical determinant of student participation, a fact that highlights important trade-offs in the design of CTE delivery models.

The Effects of Career Technical Education (with Brian Jacob)

Growing income inequality has brought a resurgence of interest in high school career and technical education (CTE) programs. This paper explores how CTE programs offered in traditional high schools affect educational attainment using data from the universe of students in public high schools in Michigan. To allow for effects heterogeneity by program type as well as by observed and unobserved student characteristics we leverage within-school variation in CTE availability over time to estimate a discrete-choice demand model that allows us to selection correct graduation rates in each CTE program. Participating in any CTE programs increases student persistence through high school, especially for low achieving students and especially in years with lower compulsory education requirements. Whereas Professional programs (such as business and health sciences) increase graduation rates but have no effect on college enrollment, vocational programs (such as welding, cooking, or autoshop) induce students to enroll in two-year colleges more and in four-year colleges less. Preliminary evidence shows that students select into vocational programs on graduation gains, but not into professional programs.

Dynamic Inefficiencies of Production Subsidies: Technological Progress, Investment, and the Race for Wind Resource

Firms often face simultaneous trade-offs when making entry and location decisions. For example, in the US wind industry, location decisions face a static tradeoff between wind resource and proximity to transmission, and entry decisions face a dynamic trade-off between early entry in preferred locations and waiting for technological progress. I show that market failures like inter-firm externalities from transmission investment can break Pigouvian subsidies calibrated only to pollution externalities, generating dynamic distortions by inducing entry at inefficient times and in inefficient places. The key parameters for this distortion are the effect of the market imperfection on entry timing and the degree of complementary between productivity (in this case wind resource) and technology (feasible turbine rotor diameter). I estimate the responsiveness to transmission spillovers using an event study design and find that entry increases by 10% after new infrastructure is built near a feasible site.