Home Finance Fiscal Policy and Income Inequality

Fiscal Policy and Income Inequality

by internationalbanker

By Rodrigo Caputo, Associate Professor of the Faculty of Economics, Universidad de Santiago de Chile (USACH)




The COVID-19 pandemic has led to a worldwide fiscal response estimated at nearly $12 trillion, or about 12 percent of the global gross domestic product (GDP). Despite this, the evidence suggests that those in low-income deciles and minority groups were disproportionately hurt. In fact, while nearly all governments have provided financial support to households, advanced economies account for the bulk of it since they have the fiscal space to finance larger deficits, and their central banks have been able to help through purchases of government or corporate securities. The fiscal response in low-income and developing countries, which are also highly unequal societies, has been restricted by tighter financing constraints, according to Furceri, Loungani, Ostry and Pizzuto.1 As a result, in low-income and unequal societies, the pandemic has had far-reaching impacts beyond health, inducing adverse social, economic and political outcomes. It has also contributed to political polarization in these economies.

Even before the pandemic, income inequality was a growing concern for economists, politicians and policymakers. In the years leading up to the pandemic, the gap between the rich and the poor widened in many places, and the pandemic exacerbated these trends.

Income inequality, as pointed out by José Gabriel Palma,2 is “a particularly complex phenomenon, which is often blurred by layers of distorting veils which sometimes make it resemble a hall of mirrors”. Palma identifies some stylized facts about the distribution of disposable income after fiscal policy is considered: Inequality is highly unequal across countries and is particularly disparate among middle-income countries. In turn, some stylized facts of market inequality are: i) a significant deterioration since the 1980s confined to the OECD (Organisation for Economic Co-operation and Development), Eastern Europe and Russia, and China and India; ii) most OECD countries have attained low levels of disposable-income inequality after government interventions, which can be highly inefficient processes; and iii) emerging markets have suffered extreme inequalities before and after fiscal policies have been implemented.

The traditional models used to understand the effects of inflation-taming measures, such as interest-rate hikes, are limited in their ability to capture the complex dynamics of the modern economy and, more precisely, the role of income inequality. As emphasized by Kaplan, Moll and Violante3, these models often assume that all households and firms have the same characteristics and respond in the same way to policy changes, which is not a realistic assumption. In short, traditional Representative Agent New Keynesian (RANK) models ignore income and wealth inequalities and assume that what’s good for the typical consumer, as defined by the models, must be good for the broader economy. This is at odds with the relevance and prevalence of income and wealth inequalities across the globe.

The good news is that the profession has developed a class of macroeconomic models—used to understand both monetary and fiscal policies—that explicitly incorporates wealth and income distributions across households and firms. These models, known as Heterogeneous-Agent New Keynesian (HANK) models, combine heterogeneous-agent models with New Keynesian models (the basic framework for studying monetary policy and movements in aggregate demand). They are particularly useful because they can capture the distributional effects of policy changes on different groups in society (see Kaplan et al.4). This is important for policymakers who want to design policies that are both effective and equitable.

Heterogeneous-agent models are well suited to explore the trade-offs in designing tax systems that balance the needs for revenue with concerns about income inequality. (One challenge in using heterogeneous-agent models is that they can be more computationally intensive than traditional models. This is because they require more data and more complex algorithms to simulate the interactions between different agents in the economy. However, advances in computing power and data availability have made it easier to use these models in practice.) In an important contribution, Heathcote, Storesletten and Violante5 concluded that the optimal level of fiscal progressivity depends on a range of factors. If individuals face high degrees of income risk, for example, because they work in a volatile industry or have unpredictable earnings, then progressive taxation can help smooth consumption and reduce the likelihood of poverty. In other words, a progressive tax system provides “social insurance” against adverse and unpredictable income shocks (such as the recent pandemic). In this case, a progressive tax system provides insurance that is absent from the market and increases social welfare; ex-post households are less likely to reduce consumption even when adverse scenarios materialize. This is precisely how fiscal policy responded during the pandemic in countries with progressive tax systems. The lack of social insurance in emerging and developing economies during the pandemic exacerbated pre-existing social inequalities.

Now, a progressive tax system will not be exempt from social costs. Progressive taxation is often seen as a way to reduce income inequality, but it can also create disincentives to work and invest. This may generate an output contraction, which is more severe if the more productive households and firms, subject to higher taxes, reduce output, labor supply and, eventually, investment.

When Heathcote, Storesletten, and Violante6 parameterized the theoretical model with the United States and quantified the net impact of the various forces for and against progressivity, they concluded that welfare gains could be obtained by making the US tax and transfer system less progressive. More precisely, reducing the average (income-weighted) marginal tax rate from 34 percent to 26 percent would be socially optimal. This result is importantly determined by two forces. The first is that progressivity discourages skill investment, thereby reducing pretax wages and labor supply. A second force limiting the optimal degree of progressivity is the need to provide public goods. The more net revenue that must be collected to finance public consumption, the less progressive the tax system should be. (The intuition behind this result is that individual households tend to underinvest in skills and work too little. This is because they do not internalize the value of the public goods that additional tax payments can finance. Lower marginal tax rates, associated with lower progressivity, reduce the gap between private and social returns.)

The result in Heathcote, Storesletten, and Violante7 contrasts most of the existing literature. For instance, Emmanuel. Saez8 derives optimal income-tax formulas using compensated and uncompensated earnings elasticities concerning tax rates. This method suggests that marginal rates for labor income should not be lower than 50 percent and may be as high as 80 percent. In a recent contribution, Fabian Kindermann and Dirk Krueger9 share similar results: The optimal tax system for top earners is characterized by high marginal tax rates and a progressive tax schedule. High marginal tax rates are necessary to finance social-insurance programs that provide income support to households in the event of idiosyncratic adverse income shocks. The progressive tax schedule ensures that the tax burden is distributed more heavily to top earners, who have a greater ability to pay, and less to low- and middle-income earners, who are more likely to experience income shocks. Hence, it has been found that high marginal labor income-tax rates on top earners are an effective tool for social insurance, even when households have high labor-supply elasticity. More specifically, marginal tax rates for the top 1 percent of earners of 79 percent are optimal as long as the model earnings and wealth distributions display a degree of concentration, as observed in US data.

The evidence so far supports a progressive tax system in the US, although there is no consensus on the degree of progressivity the system should have. In a recent blog, the World Bank stated the importance of maintaining or increasing the progressivity of the global tax system.10 Three reasons explain this prescription. First, the general population broadly supports progressive tax and transfer systems. Second, progressive taxes help reduce inequality. Third, people are more willing to pay taxes when taxes are progressive and far less willing when taxes are not progressive. The challenge for policymakers and politicians is to find a way to implement a progressive tax system without inducing excessive efficiency costs to society.



1 Industrial and Corporate Change: The rise in inequality after pandemics: can fiscal support play a mitigating role?”, Davide Furceri, Prakash Loungani, Jonathan D. Ostry and Pietro Pizzuto, July 1, 2021, Oxford University Press, Volume 30, Issue 2, Pages 445-457.

2 Development and Change:Behind the Seven Veils of Inequality. What if it’s all about the Struggle within just One Half of the Population over just One Half of the National Income?”, José Gabriel Palma, July 1, 2019, Volume 50, Issue 5, Pages1133–1213.

3 Finance and Development:The Very Model of Modern Monetary Policy,” Greg Kaplan, Benjamin Moll and Giovanni L. Violante, March 2023, International Monetary Fund.

4 Ibid.

5 The Quarterly Journal of Economics: “Optimal Tax Progressivity: An Analytical Framework,” Jonathan Heathcote, Kjetil Storesletten and Giovanni L. Violante, November 2017, Volume 132, Issue 4, Pages 1693–1754.

6 Ibid.

7 Ibid.

8 The Review of Economic Studies:Using Elasticities to Derive Optimal Income Tax Rates,” Emmanuel Saez, January 2001, Volume 68, Issue 1, Pages 205–229.

9 American Economic Journal: Macroeconomics: “High Marginal Tax Rates on the Top 1 Percent? Lessons from a Life-Cycle Model with Idiosyncratic Income Risk,” Fabian Kindermann and Dirk Krueger, April 2022, Volume 14, No. 2, Pages 319-366.

10 World Bank Blogs: “Why does the progressivity of taxes matter?”, Christopher Hoy and Chiara Bronchi, November 9, 2022.



Rodrigo Caputo is an Associate Professor in the Faculty of Economics at the Universidad de Santiago Chile. He was a Research Associate at the Centre for Experimental Social Sciences, Nuffield College, University of Oxford. He holds a Ph.D. in Economics from the University of Cambridge.


Related Articles

Leave a Comment

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.