Market Power and Inequality

Joshua Gans is Jeffrey S. Skoll Chair of Technical Innovation and Entrepreneurship and Professor of Strategic Management at University of Toronto Rotman School of Management; Andrew Leigh is a Member of the Parliament of Australia; Martin C. Schmalz is Associate Professor of Finance at University of Oxford Saïd Business School; and Adam Triggs is Director of Research at the Asian Bureau of Economic Research at the Australian National University Crawford School of Public Policy. This post is based on their recent article, forthcoming in the Oxford Review of Economic Policy.

For over a century, the idea of the United States as a “nation of shareholders” has been a powerful one. This notion has its roots in attempts by the New York Stock Exchange to broaden its political base by ensuring that more Americans owned at least a handful of stocks, and Cold War comparisons of the United States as a shareholding democracy with the central planning of the Soviet Union. The myth of a market dominated by “mom and pop” investors has been used to argue that policies which boost corporate earnings are good for all Americans—because all citizens own a stake in America’s corporations.

But stock owners are not a representative cross-section of society. Most stocks are held by the richest. Inequality in stock holding is far more pronounced than inequality in consumption or income. Moreover, while consumption inequality has stayed stable, inequality in corporate equity holdings has grown considerably over the past generation.

We explore the implications of this for one area of the economy: excess market power. Economic theory suggests that monopoly pricing benefits shareholders but hurts consumers. If share ownership and consumption were equally distributed across the population, then we might be less worried about monopolies behaving badly. Excess market power would hurt households in their capacity as consumers, but if everyone equally owned shares in the monopolists, then those same households would enjoy higher stock market returns. Monopoly profits would be returned to stock owning households as partial compensation for the harms they suffered as consumers.

In Inequality and Market Concentration, When Shareholding Is More Skewed Than Consumption (forthcoming later this year in an issue of the Oxford Review of Economic Policy on inequality), we use data from the Survey of Consumer Finances and the Consumer Expenditure Survey to calculate the distribution of consumption, income and corporate equity across US households over a period spanning nearly three decades, from 1989 to 2016.

We begin with consumption. In 1989, we find that the 20 percent of families with the lowest incomes accounted for 9 percent of all expenditure, while the 20 percent of families with the highest incomes comprised 38 percent of all expenditure. In 2016, the lowest-income group still accounted for 9 percent of consumption, while the highest-income group’s share of consumption had risen by only 1 percentage point, to 39 percent. Across this 27-year period, expenditure shares remained remarkably stable.

With income, the story is a familiar one of rising inequality. In 1989 and 2016, the poorest fifth had 3 percent of pre-tax family income. But the top fifth of families saw their share of income rise from 57 percent in 1989 to 64 percent in 2016. Put another way, the bottom group’s share remained miniscule, the top group’s share rose by 9 percentage points (or one-sixth), and middle America saw its share diminish.

For corporate equity, we find that the lowest-income fifth of families had 1.1 percent of corporate equity in 1989, and 2.0 percent in 2016 (over the same timespan, the second-bottom quintile share went from 3.5 percent to 1.6 percent, so the total share of corporate equity of the bottom 40 percent fell). By contrast, the highest-income quintile had 77 percent of corporate equity in 1989, and 89 percent of corporate equity in 2016. Hence, corporate equity is considerably more skewed than expenditure or income, and has become considerably more skewed over the past three decades.

Even if the shares had remained unchanged at their 1989 levels, excess market power would have exacerbated inequality, because stock holdings were considerably more skewed than consumption. But because consumption inequality remained little changed, while inequality in stock holdings worsened, the effect of market power on inequality was even more substantial in 2016 than a generation earlier.

To calculate the quantitative impact of market power on the distribution of income, we draw upon a model developed by Sean Ennis, Pedro Gonzaga and Chris Pike at the OECD. In a paper that is forthcoming in the same issue of the Oxford Review of Economic Policy, they show that impact of market power on inequality can be estimated from the average mark-up, the labour income share, average savings rates, the marginal propensity to save, and observed income and wealth shares. When we implement this approach using our data, we calculate that in 2016, removing market power would cause the bottom 60 percent income share to rise from 19 percent to 21 percent, and would cause the top 20 percent income share to fall from 64 percent to 61 percent. In other words, market power contributes to income inequality, but is far from the only driver of inequality.

Our study aims to help draw together two strands of literature. As the World Inequality Report recently showed, most advanced nations have seen an increase in inequality over the past generation. Meanwhile, a growing body of evidence points to an increase in market power, both in terms of rising market concentration and increasing markups. A burgeoning literature suggests that superstar firms are capturing increasingly high market shares, allowing them to use their market position to earn excess profits. Given the inequality in stock ownership, market power may continue to increase inequality in the future.

The complete article is available for download here.

Both comments and trackbacks are currently closed.

One Comment

  1. Lucas Wozny
    Posted Friday, January 25, 2019 at 6:39 pm | Permalink

    Professors Gans, Leigh and Schmalz,

    I think it’s fascinating that the second-bottom quintile share of corporate equity is lower than the lowest quintile’s share. Why would this be?