Income Taxation and the Diversity of Consumer Goods: A Political Economy Approach*

• Date: 01 July 2019
• Author: Michael T. Dorsch,Paul Maarek,Renaud Bourlès
• DOI: http://doi.org/10.1111/sjoe.12307
• Published date: 01 July 2019

Scand. J. of Economics 121(3), 960–993, 2019

DOI: 10.1111/sjoe.12307

Income Taxation and the Diversity of

Consumer Goods: A Political Economy

Approach*

Renaud Bourl`es†

Aix-Marseille University,FR-13001 Marseille, France

renaud.bourles@centrale-marseille.fr

Michael T. Dorsch

School of Public Policy at Central European University, H-1051 Budapest, Hungary

dorschm@ceu.edu

Paul Maarek‡

Universit´ePanth´eonAssas, FR-75006 Paris, France

paul.maarek@u-paris2.fr

Abstract

After-tax income inequality has risen since the mid-1990s, as increases in market income

inequality have not been offset by greater fiscal redistribution. We argue that the substantial

increase in the diversity of consumer goods has mitigated mounting political pressures for

redistribution. Within a probabilistic voting framework, we demonstrate that if the share of

diversified goods in the consumption bundle increases sufficientlywith income, then an increase

in goods diversity can reduce the political equilibrium tax rate. Focusingon OECD countries, we

find empirical support for both the model’smicro-political foundations and the implied relation

between goods diversity and fiscal policy outcomes.

Keywords: Diversityof goods; probabilistic voting; redistribution; taxation

JEL classification:D72; D78; H24

I. Introduction

In the last decades, many countries have experienced a strong increase in

market income inequality that fiscal redistribution did not offset, leading to

*We are grateful for the thoughtful comments of participants at the annual meetings of the

European Public Choice Society (Groningen), at the ASSET conference (Grenada), and at the

LAGVconference (Marseille). In par ticular,we received comments that improved the paper from

Tok ´e Aidt, Nicolas Berman, Pierre Boyer,and C ´ecilia Garcia-Penalosa.Two anonymous referees

provided constructive criticisms that greatly improvedthe paper. Of course, all remaining errors

are our own. Some of this research was carried out while M. Dorsch and P. Maarek were visiting

the Aix-Marseille School of Economics and they are grateful for their hospitality and financial

support.

†CNRS, EHESS, Centrale Marseille, Aix-Marseille School of Economics (AMSE).

‡LEMMA.

C

The editors of The Scandinavian Journal of Economics 2018.

R. Bourl`es, M. T. Dorsch, and P. Maarek 961

a significant increase in net income inequality; see Figure 1 (Piketty, 2014).

This led the International Monetary Fund to state in its last report (Gupta

and Keen, 2014) that reductions in the generosity of welfare benefits and

less progressive taxation have decreased the redistributive impact of fiscal

policy; see Figure 1.1We seek to add a new perspective to this now classic

puzzle in political economy: why democracies do not redistribute more and

what are the political economic limits to redistribution. More specifically,

we investigate why democracies have not offset the strong increase in

income inequality over the last decades.

We relate this limit to redistribution to the considerable increase in the

variety of consumer goods available in developed economies; see Figure 1

and Broda and Weinstein (2004, 2006), Hummels and Klenow (2005),

and Arkolakis et al. (2008) for empirical evidence.2This increase seems

mostly due to the increases in trade3and in R&D spending, including the

effects of R&D spillovers from abroad (Coe et al., 2009).4The key idea

of our paper is that this increase in goods diversity might have moderated

the tendency for a society to compensate the increases in inequality by

redistributing more. Figure 1 shows that average income tax rates on

average incomes have decreased and that fiscal freedom (as measured by

the Heritage Foundation) has increased.5Central to our argument is the

intuition that the welfare impacts of greater goods diversity might have been

heterogeneous across individuals, depending on the share of differentiated

goods in individuals’ consumption bundles. In this paper, we analyze how

such heterogeneous welfare effects of growth in the diversity of goods might

have affected individual preferences for fiscal redistribution, given that

private consumption becomes more valuable as goods diversity increases.

We also show how such a change in policy preferences might have affected

the equilibrium income tax rate within a political economic model.

More specifically, we build a model of probabilistic voting in which

voters spend their net income (after taxes and transfers) in private markets

1Gupta and Keen (2014) show that between the mid-1980s and mid-1990s, fiscal policy offset

about two-thirds of the 3.1 percentage point increase in market income inequality. Over the

subsequent decade (mid-1990s to mid-2000s), fiscal policy only offset about one-fifth of the 2.2

percentage point increase in the market income inequality.

2Bils and Klenow (2001) estimate that growthin the number of varieties of goods has accelerated

since 1980, and Broda and Weinstein(2006) estimate that the number of varieties (at the ten-digit

level) availableto consumers from trade flows has risen from 71,420 in 1971 to 259,215 in 2001.

More importantly, the number of goods categories (within which there are several varieties of

one good) has increased from 7,731 in 1972 to 16,390 in 2001.

3For example, an increase in diversity is an essential feature of trade models with monopolistic

competition (Krugman, 1979).

4In the theoretical part of our paper, we consider this increase in goods diversity as exogenous.

5For details concerning these variables, werefer readers to the empirical section of the paper.

C

The editors of The Scandinavian Journal of Economics 2018.

962 Income taxation and the diversity of consumer goods

Fig. 1. Empirical evidence from OECD countries

Notes: This figure shows the evolution over the last decades of income inequality, the variety of goods, and fiscal

outcomes. Note that the Gini coefficient is increasing with the degree of inequality,and that the Heritage Foundation

index is increasing the degree of fiscal freedom. All variablesare expressed as the yearly average over the OECD.

for two types of goods: one homogeneous and one that is a composition

of varieties. The diversity of goods then corresponds to the number of

varieties in the composite good. The key mechanism of our model comes

from a different allocation of income between those two types of goods

for different income levels. The intuition for this mechanism goes back to

Engel’s Law (Engel, 1857), which states that the share of food in household

spending decreases with income, and which we suppose extends to other

basic goods (clothing, shelter, transport, energy, health and sanitation, etc).

The share of normal goods and services (i.e., those that are not necessities)

in the consumption bundle should increase with income as a result (for

empirical evidence, see Bils and Klenow, 2001; Henry, 2014, 2015).

We assert that basic goods are produced mainly in competitive domestic

markets that did not benefit from the massive productivity gains of the past

decades, or from the accompanying increases in trade volumes (for instance,

international prices for many food commodities have increased over the last

decades). In this case, the introduction of new goods and services and the

subsequent increase in diversity should mainly affect the quantity of normal

differentiated goods available for consumers, which are often produced in

C

The editors of The Scandinavian Journal of Economics 2018.