Clarifying America’s great inequality debate

BOSTON — Debates about inequality trends in the United States have leapt from the pages of academic journals to leading media outlets. While conservatives have long questioned whether US inequality has really increased, The Economist recently weighed in, concluding that “the idea that inequality is rising is very far from a self-evident truth”. Unfortunately, this debate has muddled several issues in an unhelpful manner.

There are different notions of inequality, each of which is relevant to a different question and complicated by its own unique measurement challenges. The most straightforward metric is labor-income inequality, which refers to what high earners receive relative to low earners. When we talk about how workers with college degrees are faring compared to those with just a high-school diploma, we are also talking about labour-income inequality.

Of course, measuring labor income is not a simple matter, because some earnings go unreported, and some very highly paid individuals deploy strategies to make their labor income look like capital income (which is taxed at a lower rate). Moreover, when it comes to determining if real (inflation-adjusted) wages have increased, there is a vigorous debate about whether the consumer price index is overstating true inflation. But even after accounting for these issues, there is no doubt that labour-income inequality has surged at least since 1980, and that the trend has continued since the post-2008 Great Recession.

This trend stands in stark contrast to the post-war era, when labour-income inequality was stable or declining. From the 1950s to the early 1970s, workers with a high-school diploma or less enjoyed real wage growth at the same rate as those with a college degree or more. But this pattern of shared prosperity ended sometime in the late 1970s and early 1980s. While the real earnings of workers with college degrees has continued to rise steadily, workers without one now earn less today than they did in 1980.

Contrary to what The Economist suggests, this general pattern is not in question. While a recent paper by David Autor, Arin Dube, and Annie McGrew shows that wages at the bottom of the distribution did finally start increasing around 2015, leading to a notable compression between the top and the bottom of after 2020, those at the bottom still earn much less, relative to the top, than they did in 1980.

A second definition of inequality rests on overall (pre-tax and transfer) income, which includes not just labour income but also income from dividends, capital gains, and business earnings reported on tax returns. The problem with this metric is that business income is not always reported, and other forms of capital income appear in tax records only when capital gains are realised (such as when someone sells stock for more than they paid for it).

Still, there is broad agreement on what has happened to “observed total income inequality” or “fiscal income inequality”, which simply captures total incomes on tax returns. Here, the share of the top 1 per cent has increased from about 8 per cent just before 1980 to almost 18 per cent by 2019; when capital gains are included, it rises to over 21 per cent.

Much of the current debate stems from seminal work by Thomas Piketty and Emmanuel Saez, and a complementary methodology they developed together with Gabriel Zucman. This trio allocates unreported capital income in a way that closely follows the distribution of reported capital income, thus finding a broadly similar increase in the overall income share of the top 1 per cent compared to its share of observed fiscal income. But now, recently published work by economists Gerald Auten and David Splinter challenges the trio’s famous findings.

Part of the disagreement concerns untaxed capital income, whether because of tax evasion or various exemptions, such as those that apply to corporate retained earnings and income in various retirement accounts and trusts. Since this untaxed component is now estimated to account for nearly 90 per cent of all capital income, the question is how it is distributed. Auten and Splinter assume that untaxed capital income is much more equally distributed than Piketty, Saez and Zucman do, and they estimate the top 1 per cent’s claim to this income to be much less than its share of observed capital income (15 per cent versus about 50 per cent for taxed capital income today).

There are good reasons why the top 1 per cent may have a lower share of untaxed capital income compared to taxed capital income (for example, many small businesses do not report their incomes and many middle-class Americans have retirement accounts). Given the many options available to the very wealthy for tax avoidance and evasion, however, it seems unreasonable to assume that they truly command such a small share of untaxed capital income. Moreover, even with the Auten and Splinter adjustments, the top 1 per cent’s share of overall income increased from 1980 to today, albeit by significantly less than what Piketty, Saez, and Zucman find.

When Auten and Splinter, and some in the media, argue that there has been no increase in inequality, they are referring to yet another important indicator: the level of inequality that exists after taxes and transfers. This one is particularly tricky to measure, because there is quite a bit of redistribution embedded in the US tax code, and the country’s broader tax-transfer system is extraordinarily complicated. For example, determining who receives employer-paid benefits and retirement income is far from straightforward.

Here, Auten and Splinter make further adjustments and arrive at their headline finding that the top 1 per cent’s after-tax-and-transfer share has remained roughly constant, at around 8 per cent, since the 1960s. But, because Auten and Splinter’s estimates of the top 1 per cent’s share of overall income is likely to be understated (owing to their treatment of untaxed capital income), their estimates of the top 1 per cent’s after-tax-and-transfer share is also probably lower than it should be. The debate on this point will surely continue.

But such debates should not obscure what matters most in the story of the US economy since World War II. After three and a half decades in which all demographic groups largely benefitted from economic growth, the pattern of shared prosperity unraveled. Though some of the resulting social and economic costs have been neutralised by taxes and transfers, that doesn’t change the fact that the market economy, together with the technological trends it has engendered and the globalisation it has encouraged, has malfunctioned and generated a huge amount of inequality.

Daron Acemoglu, Institute professor of Economics at MIT, is a co-author (with Simon Johnson) of “Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity” (PublicAffairs, 2023). Copyright: Project Syndicate, 2024.

www.project-syndicate.org