The authors show that expected returns across a range of diversified equity portfolios and asset classes can be explained by fluctuations in the capital share growth of national income. Positive exposure to risk associated with the changes in capital share produces a significant positive premium. The authors assume that most of the capital share is in the hands of the few wealthy households—that is, investors or owners. The fluctuation in wealth redistribution between the owners and workers—the people financing consumption with their wages and salaries—constitutes a source of systematic risk that can be priced.
What Is the Investment Issue?
The authors find that fluctuations in the capital share growth of aggregate national income are a source of systematic risk. These fluctuations have substantial explanatory power against the expected returns of a range of equity characteristic portfolios and asset classes. The portfolios in question can be formed on size and book-to-market value, size and investment, size and operating profitability, and long-run reversal and non-equity asset classes. According to the authors, positive exposure to capital share risk produces a considerable risk premium with estimated risk price estimates of the same sign and similar magnitude across portfolio groups.
The authors’ results are based on the recent development of inequality-based asset pricing models. When risk is shared imperfectly and wealth is concentrated among a small number of investors, capital share is a priced risk factor. These models attribute redistributive shocks to income shifts between wealthy households, which finance consumption from asset ownership, and workers. Wealthy households, with their income strongly related to capital share, constitute a marginal investor in many asset markets. Therefore, the proxy of consumption growth of the top 10% of the stock wealth distribution using household-level and income data exhibits substantial explanatory power for equity-characteristic portfolios.
How Did the Authors Conduct This Research?
The authors analyze quarterly data for a range of equity-characteristic portfolios from the period 1963–2013, sourced from Kenneth French’s website. They use portfolio data for other asset classes from previous literature, which is precisely referenced in the article. These portfolios include corporate bonds, sovereign bonds, and S&P 500 Index option portfolios sorted on moneyness and maturity.
The capital share is defined as 1 – LS, where LS stands for the labor share of the non-farm business sector as compiled by the US Bureau of Labor Statistics. The authors note several difficulties with this measure—in particular, the challenge of separating sole proprietors’ income into the labor and capital inputs.
They also explore two household-level data sources to obtain information on wealth and income inequality, the first being survey data from the Survey of Consumer Finances. According to the authors, it is the best source of micro-level data on households’ incomes in the United States. The other source is the capital income reported by households on their tax forms to the Internal Revenue Service.
The authors model a stylized, limited-participation endowment economy in which wealth is concentrated among a small number of investors. The remaining part of this society consists of workers who finance consumption with their wages and salaries. Workers have no risky asset shares and consume earnings from their labor. The authors develop an econometric model explaining the capital share as a stochastic discount factor, with familiar no-arbitrage Euler equations. The proxy for the wealthiest stockholders’ consumption is the product of aggregate consumption multiplied by the top group’s fitted share of income.
What Are the Findings and Implications for Investors and Investment Professionals?
The authors contribute to the existing literature that focuses on the search for an empirically relevant stochastic discount factor. This factor should have an econometrically valid relationship with the marginal utility of investors. Moreover, the research fits into the growing body of literature on the role of redistributive shocks where resources are transferred between shareholders and workers. Such shocks constitute a source of priced risk. The authors’ key observation is that the fluctuations in the capital share growth of national income form a source of systematic risk and, as such, have great explanatory power against the expected returns across a range of equity characteristic portfolios and asset classes.
They also suggest that capital share exhibits a strong positive correlation with top percentiles of stock market wealth distribution, versus a negative correlation with all the deciles between and including the first to the ninth. The product of the two composites—aggregate consumption—however, exhibits a negative correlation with expected results. Consequently, an asset’s exposure in the short- to medium-term fluctuation (four to eight quarters) in the capital share growth for the top decile shows strong explanatory power for expected returns of equity portfolios. The produced model, according to the authors, has greater explanatory power for size against book-to-market sort portfolios than, for example, the Fama–French three-factor model.
The methodology presented by the authors does not explain the cross-sections of expected returns for every portfolio type. In particular, it indicates no ability to explain expected returns of industry portfolios or foreign exchange and commodities portfolios. Finally, the momentum portfolios seem not to fit into the methodology presented.
The authors’ results should be valuable to investors who want to maximize their portfolio returns against an objective, macroeconomic factor. The work will also interest academics who research portfolio return and risk diversification, with a particular focus on redistributive shocks and their impact on portfolio results at the macro level.