Abstract. Using novel micro data from Class I freight railroads, I find that capital maintenance demand is positive and has a price elasticity of two. Because maintenance is a tax-deductible input into capital production, it has two important effects on the transmission of capital tax policy. First, positive maintenance demand dampens the tax elasticity of capital because it is a tax shield in user cost. Second, elastic maintenance demand induces firms to substitute away from maintenance and toward investment when capital taxes decline. Together, these effects attenuate the traditional capital deepening channel emphasized by neoclassical investment theory. I show that the maintenance channel is quantitatively large in the context of the 2017 Tax Cuts and Jobs Act. A standard neoclassical model with maintenance forecasts a 0.6% increase in output per capita in the long run, which is half the increase predicted by an otherwise identical model without maintenance.

Abstract. Endogenous depreciation via capital maintenance is a simple but practically important channel for the transmission of capital tax policy. In practice, tax authorities tax the return on capital net of maintenance. With endogenous maintenance allocations, this implies that capital is relatively inelastic to tax policy because depreciation moves inversely with tax changes. The strength of this channel is mediated by the maintenance elasticity of depreciation, which naturally varies by capital type. Using a simple neoclassical model, I show that abstracting from the maintenance channel is normatively non-trivial. Although an unconstrained planner would still set the tax on capital to zero in a neoclassical model with homothetic preferences, the quantitative gain is roughly 30% as large as the benchmark Lucas (1990) case. In a practically relevant case, I examine optimal differential taxation when the Ramsey planner is constrained to raise revenue from capital. I prove that the planner would prefer to tax capital more when it has a higher maintenance elasticity of depreciation and a higher return net of maintenance compared to other capital types. That implies the current system---which taxes structures at thrice the rate on equipment---would be pushed toward uniform taxation.

Abstract. In this note, I extend smooth local projections to panel data and use Monte Carlo methods to explore the bias and variance properties of smooth local panel projections (SLPP). SLPP allows researchers to penalize the impulse respond toward a polynomial, while standard local panel projections (PLP) are nonparametric but result in theoretically unappealing IRFs because they are too lumpy. Relative to PLP, SLPP has nice properties in smaller samples. 

Draft | R Package (Dropbox) | R Vignette

Abstract. I study the welfare cost of deviating from the Friedman rule along the income distribution. As a first step, I provide a structural framework for studying heterogeneous agent money demand in the context of a shopping time model with idiosyncratic risk. With that framework, I estimate money demand curves for low-, middle-, and high-income households using microdata from the Survey of Consumer Finance over the period 1989-2019. Low-income money demand is perfectly interest-inelastic, while middle-income demand is roughly as elastic as aggregate money demand and half as elastic as high-income money demand. Using these demand curves, I quantify the welfare cost of deviating from the Friedman rule. Numerically, a 5% nominal interest rate costs 0.54%, 0.66%, and 0.76% of consumption for low-, middle-, and high-income households, respectively.

Abstract. Reflecting a widespread consensus that public infrastructure investment has become increasingly inefficient, the relative price of public investment has grown at 0.4% per year for the past sixty years. I quantify the negative effect of this trend on aggregate output and growth. If public investment efficiency had remained constant, output would be about 1.5% higher, which translates to more than half of total infrastructure spending. I embed infrastructure innovation in an endogenous growth model and estimate the necessary level of R&D expenditures necessary to offset the growing inefficiency wedge. The model is disciplined by evidence from quasi-experimental variation on the response of infrastructure investment productivity to government R&D expenditures on highways.

Abstract. Modern Money Theory (MMT) has risen to prominence in popular policy debates within macroeconomics. MMT economists argue for creating a job guarantee program, which they argue would generate price stability. Using a benchmark model of time consistency supplemented with a job guarantee, we conclude that once policymakers’ incentives are considered, the job guarantee does nothing to help stabilize prices. We compare this program to a competing proposal to maintain price stability and full employment, NGDP targeting.