Macroeconomics, Public Finance, Political Economy, Heterogeneous-Agent Modeling
Working Papers: (click title for link)
Abstract: Fiscal policy analysis in heterogeneous-agent models typically involves the use of smooth tax functions to approximate complex present tax law and proposed reforms. In this paper, we explore the extent to which the tax detail omitted under this conventional approach has macroeconomic implications relevant for policy analysis. To do this, we develop an alternative approach by embedding an internal tax calculator into a large-scale overlapping generations model that, while conditioning on idiosyncratic household characteristics, explicitly models key provisions in the Internal Revenue Code applied to labor income. We find that for a comparative-static steady state analysis of a given tax policy change, both approaches generate similar policy-induced patterns of macroeconomic activity despite variation in the underlying patterns of household tax-preferred consumption and labor supply behavior. However, this variation in underlying behavior is associated with significant quantitative and qualitative differences in macroeconomic aggregates along the transition path immediately following a policy change. Consequentially, although the use of unconditional smooth tax functions may be a reasonable modeling simplification for steady state analysis of tax policy, caution should be taken for their use in transition path analysis within heterogeneous-agent models.
Abstract: Macroeconomic models used for tax policy analysis often simultaneously abstract from two features of the US tax code: special tax treatment for preferential capital income, and the joint tax treatment of ordinary capital and labor income. In this paper, we explore the extent to which explicitly accounting for these tax details has macroeconomic implications within a heterogeneous-agent model. We do this by expanding the Moore and Pecoraro (2018) overlapping generations model to include distinct corporate and non-corporate firms so that the business income distributed to households can be separated into ordinary and preferred capital income. Household income tax treatment is then determined by an internal tax calculator that fully accounts for interaction among income bases while conditioning on idiosyncratic household characteristics. Relative to a conventional approach where household income taxation is determined by independent labor and capital income tax functions that do not distinguish between ordinary and preferred capital income, we find that our innovations have implications for household behavior and economic aggregates - especially the tax consequences of changes to the returns to labor and capital - when analyzing a subset of tax provisions from the recently enacted “Tax Cuts and Jobs Act”. Our findings imply that the abstracting from tax detail may come at the expense of correctly accounting for incentives and estimating macroeconomic responses to tax policy changes.
Abstract: Analysis of fiscal policy changes using general equilibrium models with forward-looking agents typically requires the modeler to assume a counterfactual adjustment to some fiscal instrument in order to achieve the debt sustainability implied by the government's intertemporal budget constraint. Since the fiscal instrument chosen to close the model can induce economic behavior unrelated to the policy change in models where Ricardian Equivalence does not hold, noise may be introduced into the analysis. In this paper we use such an overlapping generations framework to examine the impact of alternative fiscal closing assumptions on projected changes to economic aggregates over the ten-year `budget window' following a change in tax policy, assessing the extent to which the noise associated with a particular fiscal instrument can be mitigated. We find that while quantitative differences in projected macroeconomic activity can be observed across alternative fiscal instruments, these differences tend to shrink as the date that fiscal instruments begin to adjust is delayed into the future. Since the particular fiscal instrument chosen to achieve debt sustainability can then become relatively unimportant, the reliability of policy analysis obtained using this class of models may be improved.
Publications: (click title for link)
1. “Macroeconomic Modeling of Tax Policy: A Comparison of Current Methodologies." National Tax Journal, Volume 70, No. 4, December 2017, pp. 819-836. (with Alan Auerbach, Itai Grinberg, Thomas Barthold, Nicholas Bull, Pamela Moomau, Rachel Moore, et al.)
Abstract: The macroeconomic effects of tax reform are a subject of significant discussion and controversy. In 2015, the House of Representatives adopted a new “dynamic scoring” rule requiring a point estimate within the budget window of the de cit effect due to the macroeconomic response to certain proposed tax legislation. The revenue estimates provided by the staff of the Joint Committee on Taxation (JCT) for major tax bills often play a critical role in Congressional deliberations and public discussion of those bills. The JCT has long had macroeconomic analytic capability, and in recent years, responding to Congress’ interest in macrodynamic estimates for purposes of scoring legislation, outside think tank groups — notably the Tax Policy Center and the Tax Foundation — have also developed macrodynamic estimation models. The May 2017 National Tax Association (NTA) Spring Symposium brought together the JCT with the Tax Foundation and the Tax Policy Center for a panel discussion regarding their respective macrodynamic estimating approaches. This paper reports on that discussion. Below each organization provides a general description of their macrodynamic modeling methodology and answers five questions posed by the convening authors.
Abstract: The classic democratic theory of redistribution claims that an increase in market income inequality causes an increase in the size of government through majority voter support for an offsetting expansion of redistribution. I argue that the predicted inequality-redistribution relationship can break down when voters face uninsurable idiosyncratic risk with respect to future labor income and a timing differential between tax collections and government outlays. This is formalized using an incomplete market heterogeneous-agent DSGE model with majority voting and `time-to-build' policy, which suggests the collective demand for redistribution will not necessarily increase with growing income or wealth inequality. This result implies that even with equal political power among voters, democracies do not have a systematic mechanism to offset rising inequality as contrary to popular belief.
Abstract: This article utilizes a unique data set to examine the relationship between a group of potential explanatory variables and educational corruption in Ukraine. Our corruption controls include bribing on exams, on term papers, for credit, and for university admission. We use a robust nonparametric approach in order to estimate the probability of bribing across the four different categories. This approach is shown to be robust to a variety of different types of endogeneity often encountered under commonly assumed parametric specifications. Our main findings indicate that corruption perceptions, past bribing behavior, and the perceived criminality of bribery are significant factors for all four categories of bribery. From a policy perspective, we argue that when bribe control enforcement is difficult, anti-corruption education programs targeting social perceptions of corruption could be appropriate.
Abstract: The classic democratic theory of redistribution claims that an increase in the mean-to-median (MM) income ratio causes a majority coalition in the electorate to collectively demand more redistribution. The functional dependence of redistribution on the MM income ratio is tested in parametric and nonparametric regression frameworks using an OECD panel dataset. While theparametric regression model is found to be misspecified rendering subsequent inference invalid, the robust nonparametric regression model fails to uncover evidence that the MM income ratio is relevant for predicting redistribution.