REDISTRIBUTION AND FISCAL UNCERTAINTY SHOCKS

Date01 August 2020
DOIhttp://doi.org/10.1111/iere.12449
Published date01 August 2020
INTERNATIONAL ECONOMIC REVIEW
Vol. 61, No. 3, August 2020 DOI: 10.1111/iere.12449
REDISTRIBUTION AND FISCAL UNCERTAINTY SHOCKS
BYHIKARU SAIJO1
University of California, Santa Cruz, U.S.A.
This article studies the impact of fiscal uncertainty shocks. In micro data, noncapital holders reduce consump-
tion persistently in response to an increase in fiscal uncertainty whereas capital holders do not. Motivated by
this evidence, I introduce limited capital market participation and show that it magnifies the fall in economic
activity due to a fiscal uncertainty shock and induces macroeconomic comovement. This is because the limited
participation model captures individual uncertainty about redistribution. When agents are ambiguity averse, this
uncertainty about redistribution has first-order effects. As a result, the model successfully matches the empirical
responses of macro and household variables.
1. INTRODUCTION
In this article, I study a business cycle model with limited capital market participation and
Knightian uncertainty that can successfully match the empirical responses of macro aggre-
gates and household-level variables to fiscal uncertainty shocks. With limited capital market
participation, the model captures individual uncertainty about redistribution that is absent in
representative agent models. When agents are ambiguity averse, this uncertainty about redistri-
bution has first-order effects because it shows up as heterogeneous worst-case scenarios. Thus,
household heterogeneity significantly magnifies the fall in economic activity due to a capital
income tax uncertainty shock and induces comovement of macro variables.
To study the empirical effects of fiscal uncertainty shocks, I first estimate a structural vec-
tor autoregression (VAR) using measured fiscal volatilities. I include in the VAR not only
standard macro variables, but also household-level consumption measured from the Consumer
Expenditure Survey (CEX). According to the VAR, output, consumption, investment, and
hours all drop in response to an increase in the volatility of innovations to capital income
tax. In addition, real wages decline and the economy experiences a mild deflation. The capi-
tal income tax volatility shock is quantitatively relevant: for example, at eight-quarter-ahead
horizon, it explains 21% of forecast error variance of real per capita GDP. At the household
level, the impulse response is heterogeneous. In particular, while noncapital holders reduce
their consumption persistently, capital holders do not. This suggests that the macroeconomic
contraction after the capital income tax uncertainty shock could be at least partially driven by
the consumption cut by noncapital holders.
Motivated by this finding, I construct a New Keynesian business cycle model featuring lim-
ited capital market participation. In addition, I assume that agents have recursive multiple
prior preferences and perceive uncertainty not only as risk but also as ambiguity (Knightian
Manuscript received August 2018; revised April 2019.
1This article was prepared in part while the author was a visiting scholar at the Institute for Monetary and Economic
Studies, Bank of Japan. I thank their hospitality during my stay. The Editor (Iourii Manovskii) and four anonymous
referees provided excellent comments. I also thank the seminar participants at Bank of Japan, Keio University,
Hitotsubashi University, Kyoto University, San Francisco Fed, Stanford Institute of Theoretical Economics, USC
Marshall, and UC Santa Cruz for helpful feedback. Please address correspondence to: Hikaru Saijo, Department of
Economics, University of California Santa Cruz, 401 Engineering 2 Bldg - 1156 High Street, Santa Cruz, CA 96064.
Phone: +831-359-8168. E-mail: hsaijo@ucsc.edu.
1073
C
(2020) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social
and Economic Research Association
1074 SAIJO
uncertainty).2Under this preference representation, agents lack confidence in assigning proba-
bilities to relevant events and act as if they evaluate plans according to the worst-case scenario
drawn from a set of multiple beliefs. As in Bianchi et al. (2017) and Ilut and Saijo (2020), the
belief sets are tied to the measured volatilities of these fiscal instruments and are parameterized
by intervals of conditional means. In turn, as in Fern´
andez-Villaverde et al. (2015) and Born
and Pfeifer (2014), each fiscal instrument is allowed to display time-varying volatility in its
innovations. Hence, a fiscal volatility shock means an increase in the width of the intervals.
The limited capital market participation model matches the data well. As in the data, the
model reproduces sizable and simultaneous falls in output, consumption, investment, and hours
while also successfully matching the dynamics of prices such as real wages and inflation. In
contrast, the representative agent counterpart of the model in which all agents participate
in the capital market misses key features of the data. In particular, macro quantities such
as output and hours fall too little, consumption does not fall initially, and real wages stay
constant.
To understand the mechanism, first consider the representative agent model. An increase in
uncertainty about the capital tax rate increases the width of the set of the conditional mean
for the one-period-ahead capital tax and thus the representative household acts as if future
capital tax rates are higher. Due to a lower after-tax return on capital, households reduce
their investment. In contrast, consumption does not fall because it becomes cheaper relative to
investment.3Because of the intertemporal substitution of labor supply, households work less: a
lower perceived after-tax return on investment reduces the return on working. However, since
consumption does not fall, the reduction in overall aggregate demand is mitigated and thus
equilibrium employment and output decline only mildly.
Next, consider a model with limited capital market participation. The key feature of the
limited capital market participation model is that the worst-case beliefs are heterogeneous.
Capital holders fear a higher future capital tax rate. Noncapital holders, in contrast, act as if the
future capital tax is lower. Lower capital tax makes noncapital holders worse off since, assuming
transfers are equally distributed across households, lump-sum transfers are lower (lump-sum
taxes are higher) because of lower government revenue. As a result of this perceived negative
income effect, noncapital holders reduce their consumption. Because the consumption cut
by noncapital holders outweighs the mild consumption increase by capital holders, aggregate
consumption declines. Thus, aggregate demand declines and markups increase due to sticky
prices. The increase in markups leads to lower wages and labor and as a result output falls
significantly. To sum up, in the heterogeneous household model, noncapital holders perceive a
negative income effect because the model captures individual uncertainty about redistribution
that is absent in the representative agent model. Because this individual uncertainty manifests
itself as heterogeneous worst-case scenarios, household heterogeneity has first-order effects:
relative to the representative agent model, a capital income tax uncertainty shock generates a
sizable fall in output, investment, consumption, and hours.
The rest of the article is organized as follows: After reviewing the relevant literature, in
Section 2, I use a structural VAR to investigate the empirical effects of fiscal uncertainty shocks
on macro aggregates and household-level consumption. In Section 3, I introduce the model.
Section 4 describes the solution procedure and the Bayesian impulse-response matching method
to estimate the model. Section 5 presents the results and Section 6 concludes. Additional details
and results are collected in the Online Appendix.
2The multiple prior utility was axiomatized by Gilboa and Schmeidler (1989). Epstein and Schneider (2003) introduce
a recursive version that is consistent with dynamic optimization.
3It is important to point out that this is not a generic feature of the representative agent model. For example,
in Fern´
andez-Villaverde et al. (2015) and Born and Pfeifer (2014), fiscal uncertainty shocks generate a reduction in
consumption. However, as I show below, using a Fern´
andez-Villaverde et al. (2015) parameterization and assuming
expected utility, the model understates the decline in real activity compared to the VAR and counterfactually predicts
an increase in inflation and the nominal interest rate.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT