by Mariacristina De Nardi, Giulio Fella and Fang Yang,Voxeu.org
Appeared originally at Voxeu.org 22 December 2015
Thomas Piketty’s “Capital in the Twenty-First Century” quantified the evolution of wealth inequality and concentration over time and across a number of countries. This column examines existing macroeconomic models of wealth inequality through the lenses of the facts and ideas in Piketty’s book. It further examines the importance of the mechanism that Piketty champions – post-tax rate of return on capital. Gaps in existing knowledge and directions for future research are identified.
Piketty’s book Capital in the twenty-first century is, in the author’s own words,
“… a book about the history of the distribution of income and wealth.”
Among other interesting and important facts, the book quantifies the evolution of wealth inequality and wealth concentration over time and across a number of countries. Wealth is highly concentrated and its distribution is skewed with a long right tail (see also Atkinson 1983, Wolff 1995, 1998, and Rodriguez 1998) – a small number of very rich individuals hold a large share of total wealth in the economy. The book documents that the share of aggregate wealth in the hands of the richest individuals displays a U shape over time, trending downward for most of the 20th century and then increasing from the 1980s onward (see Figure 1). In other words, wealth has become more concentrated over the past 35 years.
Figure 1. Wealth inequality: Europe and the US, 1810-2010
Though the book discusses a number of mechanisms affecting wealth inequality – the role of tax progressivity, top income shares, and heterogeneity in saving rates and inheritances – it singles out a ‘fundamental force for divergence’ in the size of the difference between the post-tax rate of return on capital and the rate of output growth (see Figure 2). According to this mechanism, a higher post-tax rate of return increases the rate at which past accumulated wealth compounds, thus magnifying wealth inequality. Conversely, a higher rate of output growth reduces wealth concentration by increasing labour earnings and, therefore, saving by individuals whose main source of income is labour earnings. In Piketty’s view, the effect of these two forces is big, and changes in the rate of capital taxation and output growth can explain the dramatic evolution of wealth concentration over the past century. It should be noted that according to Piketty it is the difference between the net rate of return on wealth and the output growth rate that affects wealth inequality. Furthermore, he does not distinguish between changes in the rate of output growth due to changes in total factor productivity (TFP) versus those in the population growth rate.
Figure 2. After tax rate of return vs. growth rate at the world level, from Antiquity until 2100
In a recent paper on Piketty’s book and macroeconomic models of wealth inequality, we provide two contributions. First, we take stock of the existing literature on models of wealth inequality through the lenses of the facts and ideas in Piketty’s book and highlight both what we have learned so far and what we still need to learn in order to reach more definitive conclusions on the mechanisms shaping wealth concentration. Second, we explore the quantitative importance of the mechanism proposed by Piketty by evaluating the effect of changes in the rate of return on capital and the rate of output growth on wealth inequality in a quantitative model capable of generating realistic wealth inequality.
In the first part – that takes stock of the existing models of wealth inequality – we initially discuss the (mostly analytical) literature aiming to account for the observation that the right tail of the wealth distribution is well approximated by a Pareto distribution (See, e.g., Benhabib et al 2011, and Aoki and Nirei 2015, for two recent contributions). This strand of the literature provides the main theoretical underpinning for the mechanism, emphasized in Piketty’s book, according to which wealth concentration increases with the difference between the average net rate of return on wealth r and the trend rate of growth of aggregate output g. Multiplicative idiosyncratic random shocks to the rate of return on wealth are the main mechanism that generates wealth concentration in this class of models. While Piketty sees the rate of output growth as unambiguously reducing wealth concentration, according to some of these models output growth due to TFP increases can either reduce or increase wealth concentration depending on the environment (e.g., Aoki and Nirei 2015). For tractability, this literature abstracts from key aspects of reality, such as the determinants of the heterogeneity in rates of return (for instance due to entrepreneurship and portfolio choice), the life cycle, and the observation that bequests are luxury goods (consistent with the evidence that saving rates are strongly increasing in wealth, as documented by Dynan et al 1993, and, more recently, Saez and Zucman 2014). In addition, Gabaix et al (2015) show that these models imply a transitional dynamics of wealth concentration that is much too slow to account for its empirical evolution over the last 35 years.
Endogenous heterogeneity in saving rates in response to earnings and expenditure shocks (including, possibly, medical and nursing home expenditures during retirement), and in some cases endogenous heterogeneity in rates of returns, are instead at the centre of the quantitative models that we discuss next. The comparative advantage of this literature is its emphasis on understanding the forces that shape differences in saving behaviour and rates of return and on quantifying the importance of such heterogeneity in accounting for wealth inequality in rich quantitative models. Previous work has convincingly emphasized that entrepreneurial activity (Cagetti and de Nardi 2006), a luxury-good bequest motive (De Nardi 2004), heterogeneity in patience across families (Krusell and Smith 1998, Hendricks 2007), and high earnings risk for top earners (Castaneda et al 2003) help explain the high degree of wealth concentration (De Nardi et al 2015 investigate the latter point using tax data on earnings). Among these, entrepreneurial activity and luxury bequests appear to be the most promising. However, it is not clear to what extent each of these forces interacts with the others and jointly contributes to wealth inequality because, at least so far (with the exception of De Nardi and Yang 2015), most of these forces have been studied in isolation. There is also work to do in determining to what extent these quantitative frameworks can match the observed large differences in wealth inequality, both across countries and over time. Promising work by Kaymark and Poschke (2015) shows that these models, appropriately matched to data, succeed in explaining the evolution of wealth inequality and top wealth shares in the US over the last 50 years. More specifically, they show that increasing wage dispersion during this period is a major driver of trends in inequality and that changes in taxes and transfers to seniors account for nearly half the observed increase in the wealth concentration.
In our second part, we assess both the strength of the (r-g) force and the extent to which, as conjectured by Piketty, it is the difference between the rate of return on capital and output growth that drives wealth concentration. We carry out our analysis within a rich quantitative model that can account for the observed inequality in both wealth and earnings. We show that changes in the rate of return on capital and TFP have only small effects on wealth inequality, while changes in output growth due to population growth have very large effects. The key intuition is that the rate of return and the TFP growth rate affect all households in a similar way. In contrast, a change in the population growth rate affects different categories of individuals through two main channels. First, changes in the population growth rate affect the number of people among whom a bequest is divided and therefore the average bequest size. If, for instance, the population growth rate decreases, inherited estates tend to be larger and since richer people leave larger bequests, a reduction in population growth generates more wealth inequality. Second, a fall in the population growth rate changes the demographic structure and in particular it increases the fraction of older people, who tend to be richer. Thus, the rate of return on capital and of output growth are not perfect substitutes in their effect on wealth concentration when output growth is due to population growth. In fact, an increase in the rate of return on capital raises wealth concentration substantially less than a fall in the rate of population and output growth by the same amount.
- Aoki, S and M Nirei (2015) “Pareto distribution in Bewley models with capital income risk”, Mimeo, Hitotsubashi University.
- Atkinson, A B (1983) “The economics of inequality”, Clarendon Press, Oxford.
- Benhabib, J, A Bisin, and S Zhu (2011) “The distribution of wealth and fiscal policy in economies with finitely lived agents”, Econometrica 79: 123–157.
- Cagetti, M and M De Nardi (2006) “Entrepreneurship, frictions and wealth”, Journal of Political Economy, 114(5): 835-870.
- Castaneda, A, J Diaz-Gimenez, and J V Rios-Rull (2003) “Accounting for US earnings and wealth inequality”, Journal of Political Economy, 111(4): 818-857.
- De Nardi, M (2004) “Wealth inequality and intergenerational links”, Review of Economic Studies, 71(3): 743-768.
- De Nardi, M, G Fella, and G Paz Pardo (2015) “Fat tails in life-cycle earnings and their implications in structural models”, Mimeo, UCL.
- De Nardi, M, G Fella and F Yang (2015) “Piketty’s book and macro models of wealth inequality”, NBER working paper no 21730.
- De Nardi, M and F Yang (2015) “Wealth inequality, family background, and estate taxation”, NBER working paper no 21047, Forthcoming, Journal of Monetary Economics, 2016.
- Gabaix, X, J M Lasry, P L Lions, and B Moll (2015) “The dynamics of inequality”, NBER working paper no 21363.
- Hendricks, L (2007) “How important is preference heterogeneity for wealth inequality?”, Journal of Economics Dynamics & Control, 31: 3042-3068.
- Kaymak, B and M Poschke (2015) “The evolution of wealth inequality over half a century: The role of taxes, transfers and technology”, Mimeo, McGill University. Forthcoming, Journal of Monetary Economics, 2016.
- Krusell, P and A Smith, Jr. (1998) “Income and wealth heterogeneity in the macroeconomy”, Journal of Political Economy, 106(5): 867-896.
- Piketty, T (2014) Capital in the twenty-first century, Harvard University Press.
- Rodriguez, S B, J Diaz-Gimenez, V Quadrini, and V Rios-Rull (2002) “Updated facts on the US distributions of earnings, income, and wealth”, Federal Reserve Bank of Minneapolis Quarterly Review, 26(3): 2-35.
- Wolff, E N (1992) “Changing inequality of wealth”, American Economic Review, 82(2): 552-558.
- Wolff, E N (1998) “Recent trends in the size distribution of household wealth”, Journal of Economic Perspectives, 12(3): 131-150.
About The Authors
Mariacristina De Nardi is professor at University College London, a senior economist in the research department at the Federal Reserve Bank of Chicago, a research fellow at IFS, and a faculty research fellow of the National Bureau of Economic Research. Her research focuses on savings, wealth inequality, Social Security, health, medical expenses, entrepreneurship, and taxation.
De Nardi’s research has been published in the Review of Economic Studies, the Review of Economic Dynamics, the Oxford Review of Economic Policy, the Journal of Political Economy, the Journal of Monetary Economics, and the American Economic Review.
Before joining the Chicago Fed in July 2005, De Nardi was an assistant professor at the University of Minnesota.
De Nardi received a B.A. from the University of Venice in Italy and a Ph.D in economics from the University of Chicago.
Giulio Fella:School of Economics and Finance, Queen Mary University of London
Fang Yang is an assistant professor at Louisiana State University. Her research focuses on housing, consumption, wealth inequality, social security, and higher education.
Yang’s research has been published in Review of Economic Dynamics, Journal of Economic Dynamics and Control, International Economic Review, European Economic Review, Journal of Labor Economics, Journal of Monetary Economics, and other publications.
Before joining Louisiana State University in August 2013, Yang studied at Peking University in China (B.A. and M.A) and the University of Minnesota (Ph.D.) and worked at the University at Albany – State University of New York as an assistant professor.