Capitalists in the Twenty-First Century

Main article

Rising capital income has raised the possibility that the financial-capital rich may now dominate the ranks of America’s highest earners instead of the human-capital rich, but testing this has been challenging due to data issues around the classification of top-end business income. This paper analyzes deidentified administrative tax data, and estimates that for the typical top business owner about three-quarters of pass-through profits are returns to owner human capital rather than financial capital. The typical top earner, it seems, is still human-capital rich. The paper also shows the importance of careful attention to tax policy when interpreting economic measurement; tax incentives and rules determine how income is reported, which has important implications for key national accounts metrics.

Suppose you happen to meet a very high earner. How does she typically make her money? Is she likely a human capitalist, in that she makes most of her income from her labor or other human capital? Or is she likely a financial capitalist, in that she makes most of her income from her portfolio of publicly traded stocks and bonds and other passive financial capital? And what does the answer tell us about tax avoidance at the top of the income distribution?

The pioneering work of Piketty and Saez (2003) and Piketty, Saez, and Zucman (2018) showed that top earners take home a substantially larger share of national income today than they did in the 1980s. Rising income in the late 20^th century took the form of W-2 wages, including bonuses and stock options. But in the 21^st century, top incomes have risen in the form of business income and other non-wage income like interest and rent. Hence, there is reason to suspect that the financial-capital rich—those who derive most of their income from nonhuman capital—now dominate the ranks of America’s highest earners instead of the human-capital rich—the entrepreneurs and wage earners who derive most of their income from their human capital.

However, identifying the nature of top income is stymied by a key data question: what is the nature of income that accrues to the owner-managers of private businesses?

Most income at the very top of the income distribution is business income, not wage income or other capital income, and most of that top business income comes from private businesses. Standard data do not indicate whether private business income derives from financial capital like public stock dividends, or whether private business income instead derives from the owner’s human capital—defined as all inalienable factors embodied in business owners, including labor supply, networks, reputation, and rent-extraction ability.

Our paper makes progress on these questions by using deidentified administrative tax data to characterize top incomes in the twenty-first century.

The Tax Treatment of Private Businesses

To understand the data and the measurement problems, we first describe how most private businesses are taxed.

The first piece of the tax puzzle is that most top-business income comes from private “pass-through” businesses that are not taxed at the entity level. Instead, pass-through income passes through the firm to the owners who pay personal income taxes on their share of the firm’s income.

For example, compare the cases of Home Depot (America’s largest home improvement retailer) and Menard’s (America’s third largest). Home Depot is a public company and is therefore a “C” corporation (named after its subchapter of the Internal Revenue Code), which must face the annual corporate income tax. The corporate tax is paid before any dividend (or capital gains) taxes are paid by the firm’s owners on shareholder distributions from the after-corporate-tax income. In contrast, Menard’s is not publicly traded and is an “S” corporation, which does not face the corporate income tax. Instead, the annual income of Menard’s passes through pro-rata to its owners’ personal income tax returns, and its owners owe personal income taxes on that income in the year that it is earned. Dividends from Menard’s are never taxed. Partnerships are taxed similarly to S-corporations. A majority of U.S. taxable business income is now earned by pass-throughs, the vast majority of which is S-corporation or partnership income.

The second piece of the tax puzzle is that tax incentives and rules can lead pass-through owners to report their human capital income as business income (i.e., profit) instead of as wages or bonuses. When S-corporation owners declare their income as business income instead of wages or bonuses, they typically enjoy a 3.8 percentage point tax break. Tax law does not give the same tax break to partnership owners, but requires them to typically report their income as business income.

When putting these two pieces of the puzzle together, it is easy to see how private business income may predominantly represent human capital income rather than financial capital income. However, the extent to which private business income actually represents human capital income is an empirical question.

Basic Facts from Tax Data

We analyze deidentified administrative tax data that link pass-throughs to their owners for 11 million firms between 2001 and 2014.

The data reveal a vast world of business owners who prevail at the top of the income distribution. Most top earners own a pass-through business. In both their absolute number and the dollars they earn, top pass-through owners far surpass top public company executives.

What are these firms? A typical firm owned by someone in the top 0.1% of the income distribution is a regional business with $20M in sales and 100 employees, such as a large law firm, a beverage distributor, or an auto dealer.

Is it more likely that these owners passively earn income from their businesses as uninvolved investors, or instead actively earn income through involvement in firm operations? Simple statistics suggest the latter. The typical top pass-through owner owns a single pass-through that has one to three owners and whose income constitutes a large share of the owner’s total income. Hence, pass-through business income is not diversified. Pass-through industries tend to be relatively skill intensive. Perhaps most tellingly, pass-through owners are working age. Figure 1 shows that the age distribution of pass-through owners mirrors the age distribution of top wage earners, rather than the age distribution of people who earn most of their income in more clearly passive ways such as C-corporation dividends or other capital income.

Figure 1: Millionaire Pass-Through Owners are Working Age

.

Natural Experiments

To verify and quantify the approximate share of top pass-through income that represents human capital income, we would ideally measure the impact on firm profits of randomly forcing some pass-through owners to withdraw their human capital from their firms. Fortunately, there are two natural experiments that approximate the experimental ideal: the profit impact of owner death and the profit impact of owner retirement.

Figure 2a plots the time series of the impact on profits at firms that experienced a pre-age-65 owner death, relative to otherwise similar firms that did not experience an owner death. In the year of the owner death, profits fall sharply and permanently, even after we adjust for firm reorganizations after an owner’s death. Figure 2b shows a similar pattern after an owner retirement. Properly scaled and averaged, we estimate that about three-quarters of top pass-through profits are returns to owner human capital for the typical top business owner.

Figure 2a: Death of a Million-Dollar-Earning Owner

Figure 2b: Retirement of a Million-Dollar-Earning Owner

.

Punchline Numbers

We use our three-quarters estimate to gauge the share of top earners that earn a majority of their income from their human capital income—that is, the share of top earners that are human-capital rich. When classifying pass-through income as financial-capital income, a minority of top earners are human-capital rich. But when defining human capital income as wages plus three-quarters of pass-through income, we estimate that most top earners are human-capital rich. For example, among million-dollar earners, 67% derive most of their income from human capital income, while only 35% derive most of their income from wages alone.

Hence, if you run into a million-dollar-earner on the street, she likely earns most of her income from her human capital rather than from her financial capital.

As a complement to our cross-sectional analysis of top incomes, we also investigate their evolution since 2001. We find no role for a larger workforce in explaining higher top entrepreneurial income (i.e., business income plus wages earned from any private business they own). Instead, top entrepreneurial income growth can be accounted for by both rising labor productivity and a rising owner share of the value added. Thus, an explanation of rising top entrepreneurial income requires explaining both a growing pie and a growing owner-manager slice.

Conclusion

Our findings have three broad implications.

First, we find that the typical top earner is human-capital rich. This finding adds context to recent developments in understanding top incomes. Piketty and Saez (2003) find that U.S. top income inequality had returned to early-twentieth century levels, but the rentier income of the past had been replaced by the income of the human-capital rich. But Piketty (2014) and Piketty, Saez, and Zucman (2018) point to rentiers as possibly displacing the working rich in the twenty-first century. Our findings show that, despite recent work showing rising top financial capital income, the typical top earner is still human-capital rich. In particular, we highlight a class of entrepreneurs and other private business owners who are hidden from public view but whose human capital income is key for understanding top incomes in America. Although human capital is central, we stress that the returns to top owner-manager human capital need not be socially optimal and can include returns to unproductive rent seeking.

Second, the result that productivity growth explains a substantial part of entrepreneurial income growth aligns with recent work finding a key role for efficiency improvements in rising firm profitability (Autor et al. 2020; De Loecker, Eeckhout, and Unger, 2020). The increasing share of value added that accrues to owners also plays an important role. Since the owners in our context also appear to be managers and key workers, agency conflicts and zero-sum bargaining between executives and corporate boards are unlikely to explain the rising owner share of value added in our data.

Third, our findings show the importance of careful attention to tax policy when interpreting economic measurement. Tax incentives and rules determine how income is reported. In our context, “business income” reported to tax authorities is often in fact closer to wage income than to capital income. This finding has implications for national accounts. Pass-throughs rose steadily and dramatically in prevalence and quantitative importance after the Tax Reform Act of 1986 (Cooper et al. 2016). In ongoing work, we are finding that the rise of pass-throughs and the concomitant relabeling of wage income as business income h).

This article summarizes: Smith, M., Yagan, D., Zidar, O., & Zwick, E. (2019). Capitalists in the Twenty-first Century. The Quarterly Journal of Economics, 134(4), 1675-1745

Matthew Smith is at the U.S. Treasury Department, Danny Yagan is at UC Berkeley, Owen Zidar is at Princeton University, and Eric Zwick is at the University of Chicago.