Do Equity Markets Care About Income Inequality? Evidence from Pay Ratio Disclosure

Stephan Siegel is Professor of Finance and Business Economics at the University of Washington. This post is based on a recent paper, forthcoming in the Journal of Finance, by Professor Siegel; Yihui Pan, Associate Professor of Finance at the University of Utah; Elena S. Pikulina, Assistant Professor of Finance at the University of British Columbia Sauder School of Business; and Tracy Wang, John Spooner Professor of Finance at the University of Minnesota Carlson School of Management.

Stagnant middle-class wages but rapidly-increasing incomes by high earners have led to a growing debate about income inequality in the U.S. However, it is largely unknown how U.S. financial markets and shareholders assess the dispersion in pay between a firm’s top executives and rank-and-file employees. Understanding equity markets’ assessment of income inequality is important because equity markets allocate capital and send valuation signals to firms, informing and possibly shaping corporate policies that contribute to or mitigate income inequality. In this paper, we address this question by exploiting a new rule that required U.S. publicly traded companies to report the ratio between CEO pay and median worker pay for the first time in 2018.

Experimental and survey evidence suggests that many individuals are averse to pronounced income inequality. Aversion to inequality could be self-centered or reflect concerns about future economic growth or stability in broader society. For some individuals, high income inequality could violate their views about a fair allocation of resources. Whether such attitudes are important among financial market participants is an open question, especially as wealthy Americans, who are more likely to be equity investors, have been found to be more accepting of inequality than the rest of the population (Cohn et al. (2019)).

The average pay ratio across the approximately 2,300 U.S. firms that reported their pay ratios for the first time in 2018 is 145, while the median is 65. We find that firms reporting a higher pay ratio experience a significantly lower market reactions than firms reporting a lower pay ratio. Specifically, a one-standard-deviation increase in pay ratio decreases a firm’s seven-day cumulative abnormal return by about 42 basis points (bps). The negative market reaction persists at least several months after the initial pay ratio disclosure. It is also robust to controlling for contemporaneously disclosed CEO or worker pay, suggesting that financial markets react to within-firm pay disparity independently of pay levels. The U.S. market reaction to high pay ratios therefore seems to be at odds with earlier findings from the UK that higher pay inequality is primarily a reflection of better managerial talent (Mueller, Ouimet, and Simintzi (2017)).

Next, we examine whether the negative market reaction to high pay ratios represents a cash flow effect due to expected negative reactions by customers, employees, governments, or rather a discount rate effect due to inequality-averse shareholders’ reduced demand for firms with large pay disparities. If the cash flow channel is important, we expect the market response to be stronger for firms with more inequality-averse stakeholders. Alternatively, if reduced demand due to shareholders’ inequality aversion is driving the observed relation, we expect to see lower announcement returns for firms with more inequality-averse shareholders.

While shareholders’ or other stakeholders’ inequality aversion is not directly observable, we proxy for a firm’s exposure to inequality-averse employees, customers, or local governments using the political leaning and the redistributive policies of the states in which the firm operates. Similarly, we capture the inequality aversion of a firm’s investors using the political leaning and redistributive policies of the home states of a firm’s shareholders. For institutional investors, we also construct an alternative measure of investors’ revealed social preferences as the average MSCI KLD Social Index score of their 2017 portfolio stock holdings.

We find that firms with more inequality-averse shareholders experience a significantly more negative market response to high pay dispersion. This result holds using both the location- and the holdings-based preferences measures. However, firms’ exposure to cash flow risk because of other inequality-averse stakeholders does not seem to play a significant role in explaining the cross-sectional variation in market reactions.

Finally, we examine whether in 2018 the portfolio rebalancing of institutional investors with different degrees of inequality aversion varies with portfolio firms’ reported pay ratios. Controlling for investor and firm fixed effects, we indeed find that in 2018 institutional investors with stronger inequality aversion reduce their allocation to high-pay-ratio stocks more so than other institutional investors. Interestingly, it is institutional investors’ revealed social preferences, not their environmental or governance preferences, that explain cross-sectional differences in both the initial market reaction and the subsequent portfolio rebalancing response to the newly available pay ratios.

Taken together, our results suggest that equity markets are concerned about income inequality and assess high within-firm pay dispersion negatively. While it is difficult to distinguish between investors’ preferences and their subjective beliefs, our findings suggest that investors’ inequality aversion, rather than concerns about future cash flows, is the dominant channel through which within-firm pay disparity affects firms’ equity valuations. Finally, our results allow for the possibility that investors, through their portfolio decisions and impact on firms’ valuations, may affect corporate culture and policies (Heinkel, Kraus, and Zechner (2001), Hart and Zingales (2017)). Our results thus hint at an alternative and possibly complementary mechanism to the political process to restrain inequality (Kuziemko et al. (2015), Pastor and Veronesi (2021)).

The complete paper is available for download here.

Both comments and trackbacks are currently closed.