The paper uses 1980 to 2000 Panel Study of Income Dynamics (PSID) data to study the short-run effect of a stock market appreciation on U.S. household income inequality. Fixed-effects regressions suggest that a stock market appreciation raises the incomes of stockholder households more than non-stockholder households. The Gini coefficients derived from the regressions reveal a perceptible but rather volatile increase that can be attributed to the stock market appreciation, especially for the latter parts of the 1980s and 1990s. When averaged by decade, the stock market appreciation raises the Gini coefficient by about 2% for the 1980s and by 3% for the 1990s.
Since the 1970s the U.S. has experienced a trend toward increasing income inequality. The increase in the skill premium resulting from a greater demand for skilled workers along with skill-biased technological change is considered to be a key cause of this trend (Acemoglu, 2002; Berman, Bound, & Griliches, 1994; Johnson, 1997, Krueger, 1993 and Krusell et al., 1997; Murphy & Welch, 2001). This factor alone may explain about two thirds of the increase in income inequality.
Apart from the skill premium there are a number of other factors that are thought to be influencing the trend in income inequality.1 The boom in the U.S. stock market is one of them (Clark, 2002 and Piketty, 2003; Piketty & Emmanuel, 2001; Smith, 2001 and Wolff, 2000). In the past 20 years, the S&P500 stock market index has increased by more than five times. This has raised stockholders’ wealth significantly compared to non-stockholders’ wealth (Smith, 2001). As wealth is a potential source of income, the rising wealth inequality may also bring about more income inequality. This may be of potential interest for tax policy, for example, the taxation of capital gains. But although there are several studies about the stock market's effect on wealth inequality, there are few on the stock market's effect on income inequality. Yet, from the point of view of public policy and taxation, the income distribution and not the wealth distribution is at the center of attention. The focus on income manifests itself in the fact that the governments of all industrialized countries are primarily redistributing income rather than wealth.
Clark (2002) argues that the recent stock market boom has raised income inequality by causing a real income decline of the poor. Piketty and Emmanuel (2001) and Piketty (2003) find that the secular increase in the income share of the top tenth percentile of households is for the most part a capital income phenomenon. Das and Mohapatra (2003) conclude that the income share of the top quintile of the income distribution grows at the expense of the “middle class” when emerging stock markets are liberalized, while the share of the lowest income quintile remains effectively unchanged.
No research appears to have focused on the effect of the recent U.S. stock market appreciation on income inequality by comparing the income response of households that participate in the stock market with that of households that are not. Also, studies that examine the relationship between stock market growth and income inequality do not always condition on GDP growth. Conditioning on GDP growth is of potential importance in this context because there is substantial evidence that GDP growth decreases income inequality (Beach, 1977 and Blank, 1989; Blank & Blinder, 1986; Hirsch, 1980; Thorton, Agnello, & Link, 1978).
The purpose of this paper is to analyze empirically the short-term impact of the U.S. stock appreciations of the 1980s and 1990s on income inequality while accounting for any effect of GDP growth. The paper uses the Panel Study of Income Dynamics (PSID) data. The sample is split into two groups: stockholders and non-stockholders. If the stock market has any role to play in the observed increase in income inequality, then an examination of its differential impact on the incomes of stockholders and non-stockholders is a logical starting point of the investigation.
The remainder of the paper is organized as follows. Section 2 discusses the potential short-run links between income inequality and a stock market appreciation. It sets the stage for the empirical testing strategy. Section 3 introduces the data, the sample selection method, and some evidence on the incomes of stockholders and non-stockholders. Section 4 provides panel data estimates of the impact of stock market growth on the household incomes of stockholders and non-stockholders. It also reports estimates on how much the Gini coefficient of income inequality has been raised by the stock market appreciations of the 1980s and 1990s. Section 5 briefly summarizes the paper and draws some conclusions.
The paper has employed household income and demographic data from the University of Michigan Panel Study of Income Dynamics to study the short-run effect of the stock market appreciation of the 1980s and 1990s on U.S. household income inequality. The panel data regressions run on the panel data reveal that the incomes of households which hold stock are significantly more sensitive to changes in the S&P500 stock index than households which do not hold stock. This result is robust to alternative definitions of what constitutes a stockholding and what a non-stockholding household.
For stockholder households, the elasticity of household income with respect to the growth rate of the S&P500 index is about 0.1. By contrast, the growth rate of GDP does not influence incomes above and beyond what is implicitly controlled for by such household-specific panel variables as hours worked. The regression results suggest that the stock price increases of the 1980s and 1990s have also raised the incomes of non-stockholder households. This implies that the stock market boom has had an indirect trickle-down effect on the incomes of non-stockholders. However, as the incomes of non-stockholders react significantly less to stock price changes than the incomes of stockholders, the regression results provide some indirect evidence for the idea that the stock market boom of the 1980s and 1990s has contributed to the rising income inequality in the U.S.
To study the impact of the stock market boom on income inequality more directly, the predicted income series from the panel data regressions are employed to construct Gini coefficients for each year of the sample. One series of Gini coefficients takes the stock market boom as given. A second counterfactual series of Gini coefficients is constructed on the assumption that the stock market boom did not impact household incomes. By comparing the two series of Gini coefficients the stock market appreciation is calculated to have raised the Gini coefficient between 1.5 and 2.1% on average for the 1980s and between 2.5 and 3.2% on average for the 1990s. Behind these averages, however, hides a significant amount of volatility in the impact of the stock market boom on income inequality and relatively little persistence. The volatility in the sensitivity with which the Gini coefficients react to the stock market boom is likely related to the volatile nature of how households realize accrued capital gains. Based on the experience of the Congressional Budget Office, capital gains realizations are not easily tied to either stock price movements or changes in the tax laws related to capital gains.
The results of the paper suggest that a stock market appreciation does raise income inequality, but that this increase in inequality is temporary and also rather unpredictable in terms of timing. Longer runs of significant stock market gains, such as experienced in the 1980s and 1990s, may give the false impression of a permanent shift in income inequality due to the stock market. They do not provide a solid reason to change tax policies, such as those related to capital gains, with the intent to affect income inequality.
