Progressive Tax and Inequality in a Unionized Economy*

• Date: 01 January 2020
• DOI: http://doi.org/10.1111/sjoe.12328
• Published date: 01 January 2020

Scand. J. of Economics 122(1), 38–80, 2020

DOI: 10.1111/sjoe.12328

Progressive Tax and Inequality in a

Unionized Economy*

Chun-Chieh Huang

Fu Jen Catholic University, New Taipei City 24205, Taiwan

cchuang@mail.fju.edu.tw

Juin-Jen Chang

Academia Sinica, Taipei 11529, Taiwan

jjchang@econ.sinica.edu.tw

Hsiao-Wen Hung

Tamkang University, New Taipei City 25137, Taiwan

hwhung@mail.tku.edu.tw

Abstract

In this paper, we develop a heterogeneous-agent, endogenous growth model of a unionized

economywith distinct progressive tax schedules on labor and capital income. Withtime preference

heterogeneity,the effective labor force, balanced growth,and income inequality are endogenously

determined, and these interact with each other.A reduction in the degree of progressive labor tax

yields a “double-dividend”in terms of reducing income inequality and boosting economic g rowth,

while capital income progressivity displays the usual growth–inequality trade-off. Particularly,

the double-dividend effect becomes more pronounced when unionization is declined or trade

unions become more wage-oriented, leading to the so-called “Cheshire cat” phenomenon.

Keywords: Progressive taxation; trade-off between inequality and growth; unionization

JEL classification:D31; J52; O15; O40

I. Introduction

The potential trade-off between economic growth (efficiency) and income

distribution (equality) has been a central issue ever since modern economic

growth theory began. As for policy-makers, more aggressive redistribution,

typically through more progressive income taxation, can lead to lower

economic growth. The beneficial effect of a more progressive tax system

on income distribution is then traded off against the efficiency loss arising

from distorting labor supply and capital accumulation.

*Wethank Ping Wang, Been-Lon Chen, Ching-Chong Lai, Jang-Ting Guo, Hsieh-Yu Lin,Wei-

Neng Wang, the anonymous referee, and participants at the 13th Annual Conference of the

Association for Public EconomicTheor y andAEI-Five Joint Workshop2014 for their comments.

Weare also g rateful for financial support from the Ministry of Science and Technology,Taiwan.

C

The editors of The Scandinavian Journal of Economics 2018.

C-C. Huang, J-J. Chang, and H-W. Hung 39

Can progressive income taxation really promote income equality? The

decline in tax progressivity associated with the 1986 Tax Reform Act has

increased income inequality in the United States (Altig and Carlstrom,

1999; Li and Sarte, 2004). However, in a cross-country comparison, the

OECD (2012) finds that more progressive taxation does not necessarily

result in a more equal income distribution. For a large panel of countries,

Duncan and Sabirianova Peter (2016) also show that tax progressivity is not

effective at reducing income inequality: relative to the poor, the rich are able

to more effectively respond to a more progressive tax schedule by reducing

their working hours or concealing their hard-to-verify income (e.g., capital

income). To compare the effect of the tax share versus income share over

the past four decades, Hartman (2002) and Bastagli et al. (2012) conclude

that progressive taxation has failed to reduce the disparity of real incomes,

given that income inequality has increased in most advanced economies

and many developing economies over the past few decades.

Does more progressive income taxation entail a trade-off between

economic growth and income distribution? Despite a vast body of literature

on the link between growth and inequality, no general consensus has

emerged. The empirical evidence is rather inconclusive on this trade-off

in the presence of more progressive income taxation. Alesina and Rodrik

(1994), Persson and Tabellini (1994), and Perotti (1996) find evidence

of a negative relationship between inequality and growth, while Partridge

(1997), Forbes (2000), and Frank (2009) find a positive relationship.1By

identifying structural breaks in economic growth for 140 countries, Berg

et al. (2012) have calculated “growth spells”, and they have found that the

growth duration (sustained growth) is positively related to the degree of

equality in the income distribution. Indeed, as shown by the Kuznets cur ve

(Kuznets, 1955; Barro, 2000), the trade-off between income inequality and

economic growth is not clear-cut, and it depends on different development

stages. As is evident, the policy relevance of progressive taxation is still a

controversial issue that needs a further thorough investigation.

In this paper, we address these important issues by incorporating three

novel features in a heterogeneous-agent, endogenous growth model: (i)

separate progressive tax codes for labor and capital income; (ii) a unionized

labor market; and (iii) an endogeneous effective labor force, which accounts

for both the employment rate and labor hours per worker. In the model,

heterogeneity arises from two sources (i.e., time preferences and capital

endowments), reflecting the empirical finding whereby average patience

explains a considerable fraction of the variation in wealth accumulation

1See Røed and Strøm (2002) or Wellerand Rao (2008) for a comprehensive survey.

C

The editors of The Scandinavian Journal of Economics 2018.

40 Progressive tax and inequality in a unionized economy

and income (see Dohmen et al., 2016).2It is crucial to distinguish between

labor and capital income taxes, because both tax schedules exhibit different

trends and vary widely across countries.3More importantly, the two

distinct taxes, as we show later, have very different policy implications

for inequality.

It is also important to introduce trade unionism into the model. Over

the past few decades, two influential countries – the United States and

the United Kingdom – with the largest declines in unionization have also

experienced the biggest increases in income inequality. The decline in

union density in the United States can account for about 20 percent of

the rise in wage inequality during the 1980s (Card, 1996; Dinardo et al.,

1996; Freeman, 1996; Fortin and Lemieux, 1997). Likewise, the trend of

de-unionization has also increased pay inequality in the United Kingdom

(Machin, 1997; Card et al., 2004; Visser and Checchi, 2011), Canada (Card

et al., 2004), New Zealand (Wallerstein, 1999), and other OECD countries

(Kahn, 2000). Although union densities have fallen sizably, trade unionism

still plays a crucial role in the determination of wages, as collective

bargaining coverage is still high.4More importantly, trade unionism is

interconnected with progressive taxation. In a unionized economy, the

impact of more progressive taxation crucially depends on the distribution

of the labor supply for different income groups, and on the employment

rates they face. The importance of the inequality in hours worked has been

stressed by the OECD (2012) and Krueger et al. (2010). The employment

rate is equally important, as inequality in labor earnings goes hand-in-hand

with low employment rates in several European Union countries, including

Belgium, Finland, France, and Italy (see Dreger et al., 2015). In the model,

both the employment rate and labor hours are endogenously determined.

Any tax progressivity change will affect the bargained employment rate

and wage rates, which in turn will influence the individual labor supply

decision at the hours-of-work margin. The intensive or extensive margin of

the effective labor force will alter growth and inequality. In the benchmark

model, we characterize a physical capital-driven balanced-growth-path

2The heterogeneities in time preferences and wealth endowment are important factors in

explaining the observed income inequality in the United States (Krusell and Smith, 1998;

Hendricks, 2004) and other industrialized countries (see Garc´ıa-Pe˜nalosa and Turnovsky, 2007).

3In OECD countries, the progressivity of labor taxes increases because of a cut in social security

contributions and an introduction of in-work tax benefits, targeted at lowerincomes, but capital

taxes become less progressive because of a decline in top marginal individual income tax rates.

Nevertheless, the degree of progressivity of the labor income tax is commonly higher than that

of the capital income tax.

4For example, the union density in France and Spain is roughly 10 percent, yet the collective

bargaining coverage in these countries is 92 and 68 percent, respectively.

C

The editors of The Scandinavian Journal of Economics 2018.