Optimal Dynamic Taxation with Distinctive Forms of Social Status Attainment

AuthorJuin‐Jen Chang,Tsung‐Sheng Tsai,Hsueh‐Fang Tsai
Date01 April 2019
DOIhttp://doi.org/10.1111/sjoe.12283
Published date01 April 2019
Scand. J. of Economics 121(2), 808–842, 2019
DOI: 10.1111/sjoe.12283
Optimal Dynamic Taxation with Distinctive
Forms of Social Status Attainment*
Juin-Jen Chang
Academia Sinica, Taipei 115, Taiwan
jjchang@econ.sinica.edu.tw
Hsueh-Fang Tsai
Fu-Jen Catholic University, New Taipei City 242, Taiwan
013640@mail.fju.edu.tw
Tsung-Sheng Tsai
National TaiwanUniversity, Taipei 106, Taiwan
tstsai@ntu.edu.tw
Abstract
We examine the role of both consumption- and wealth-induced social comparisons in setting
dynamic optimal income taxation. Under complete information, state-invariant labor income
taxes are used to remedy the externality caused by consumption-induced social comparisons,
while state-contingent capital income taxes are used to remedy the externalities caused by both
consumption- and wealth-induced social comparisons. Under incomplete information, distinct
types of agents are subject to an identical marginal capital income tax, which removes social
comparisons. To solve the information problem, low-productivity agents could be subject to a
lower marginal labor tax than high-productivity agents, which contradicts the traditional result
in the Mirrlees–Stiglitz models.
Keywords: Dynamic taxation; state-contingent policy; status motives in consumption and wealth
JEL classification:E62; H21; H23; H30
I. Introduction
It has been well documented that agents’ quests for social status (or
concern for their relative position in society) has major consequences for an
economy.1Status-motivated preferences have been supported by the rapid
*Wethank Been-Lon Chen, Jang-Ting Guo, Ching-Chong Lai, Ping Wang, C. C. Yang, and two
anonymous referees for their helpful suggestions and insightful comments on an earlier versionof
this paper.Their inputs have led to a much improved paper. Financial support from the Ministry
of Science and Technology, Taiwan is gratefully acknowledged. Any remaining errors are, of
course, our own responsibility.
1Social-status seeking has been widelystudied in other contexts, such as in analyzing asset pricing
(e.g., Gal´ı, 1994; Bakshi and Chen, 1996; Campbell and Cochrane, 1999), wealth distribution
C
The editors of The Scandinavian Journal of Economics 2017.
J-J. Chang, H-F. Tsai, andT-S. Tsai 809
development of the empirics of social status and happiness studies.2Based
on such social-status attainment, the normative literature indicates that a
“rat race” with conspicuous consumption and wealth gives rise to negative
externalities for others, which call for government intervention in that the
competitive equilibrium does not yield a Pareto-optimal resource allocation.
Therefore, a positive income tax is called for to correct the overconsumption
resulting from social aspirations in consumption and the overaccumulation
of capital resulting from social comparisons in wealth.3
In this study, we address the role of social-status seeking in setting
the dynamic optimal taxation (on both capital and labor incomes) in a
heterogeneous agent model. There are two salient features of the model.
First, we consider motives of status seeking, based both on consumption and
on wealth (capital), in a unified framework. With rare exceptions, previous
research in the literature has investigated the case with either consumption-
or wealth-based status-seeking motives in isolation (see below for a more
detailed discussion).
The so-called Veblen effect (Veblen, 1899) arises from the desire to
achieve social status by signaling wealth through conspicuous consumption.
An agent’s peers infer the agent’s wealth after observing a subset of the
agent’s purchases, and this causes agents to distor t their consumption
toward the purchase of visible positional goods, such as luxury clothing,
cars, and housing. People care about relative wealth because conspicuous
consumption is related not just to wealth, but additionally to relative
wealth (Cole et al., 1995). Both relative consumption and relative income
(wealth) have been empirically proved to be an important status-seeking
motive. Although Veblen’s theory implies that there exists a positive
relationship between relative wealth and conspicuous consumption (Bagwell
and Bernheim, 1996), these two motives might give rise to rather
different effects on labor supply and economic growth (Tournemaine and
Tsoukis, 2008; Chang et al., 2018). Because of a joint status concern
for consumption and wealth, in this paper, we offer distinct implications
for optimal taxation, as a consumption-induced motive can interact in
the opposite way to a wealth-induced motive to affect agents’ saving
behaviors.
(e.g., Futagami and Shibata, 1998; Corneo and Jeanne, 2001; Tsoukis, 2007), the patterns of
growth (Corneo and Jeanne, 1997; Rauscher, 1997; Carroll et al., 2000; Liu and Turnovsky,
2005), and inefficient allocations (Fisher and Hof, 2000; Dupor and Liu, 2003; Alonso-Carrera
et al., 2004).
2See Truyts (2010) for a survey of the relevantempirical studies.
3See Ljungqvist and Uhlig (2000) and Dupor and Liu (2003) for the optimal taxation with social
aspirations in consumption, and see Kurz (1968) and Zou (1995) for the optimal taxation with
social aspirations in wealth.
C
The editors of The Scandinavian Journal of Economics 2017.
810 Optimal dynamic taxation and social status
Second, this is a two-period Mirrlees–Stiglitz (Mirrlees, 1971; Stiglitz,
1982) model with incomplete information and capital accumulation in
which the government cannot observe agents’ productivity and the agents’
capital accumulation and endowment are various. To remove the distortion
caused by social comparisons, income taxes, such as Pigouvian taxes,
play a corrective role in achieving the social optimum. To solve the
information problem arising from unobserved productivity, income taxes
also play a redistributive role in screening agents’ tr ue productivity. Thus,
optimal taxation responds to status-seeking behavior quite differently under
a complete information and an incomplete information environment, as
the corrective and redistributive roles of income taxation interact under
incomplete information.
Our findings modify the conventional results of optimal taxation and
contribute new insights into the literature on both social-status seeking and
optimal dynamic taxation (or the new dynamic public finance). We show
that not only do distinctive types of social-status seeking play quite different
roles in the design of optimal labor and capital taxation, but also optimal
taxation offers different policy implications for the cases of complete
information and incomplete information. Under complete information, the
optimal marginal labor income taxes are positive, responding only to the
externality caused by consumption-based social comparisons. However, in
contrast to the conventional notion mentioned above, the optimal marginal
capital income taxes could be positive or negative, depending on the relative
magnitude of externalities in consumption-based and wealth-based social
comparisons.
Moreover, the optimal labor income tax is state-invariant, but the optimal
capital income tax is state-contingent, owing to the intertemporal preference
effect. In an influential study, Ljungqvist and Uhlig (2000) show that in
the presence of status seeking in only consumption, a constant optimal
income tax can correct the overconsumption period by period without
any intertemporal considerations. By contrast, we find that when agents
exhibit their status to others not only through consumption but also through
wealth (capital), a state-varying income tax is called for in order to remove
the distortion caused by the intertemporal preference effects of various
social-status attainments. In particular, the optimal income taxation can be
countercyclical in relation to business cycles, which stands in sharp contrast
to the Keynesian demand-management policy (i.e., the optimal tax policy
affects the economy countercyclically via procyclical taxes).
Under incomplete information, the optimal capital income taxation does
not discriminate against the agent’s type, while the labor income taxation is
used to screen the agent’s productivity type. The social planner simply levies
an identical marginal capital income tax rate on distinct types of agents,
because capital income taxation only plays a corrective role, removing the
C
The editors of The Scandinavian Journal of Economics 2017.

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