Why a new study gives a misleading view of inequality in America

Elise Gould & Josh Bivens

In recent years, researchers have debated the simple question of whether inequality has risen a lot or a little in the United States over the past half-century. Lots of arguments in this debate surround highly technical issues like, “Should the income of owners of ‘pass-through businesses’ be reported as wages or business profits?” or “Is income that is not reported on tax returns mostly earned by rich or middle-class households, and how do you know?”
But we’ve identified available data that sidesteps nearly all these complexities and demonstrates that inequality has indeed risen enormously: what individual Americans earn in the labor market. Take one measure of labor market earnings – the pay (including benefits) of the 80% of workers who are not managers or supervisors at work. For decades before 1980, these workers’ hourly pay tracked economy-wide productivity growth tightly. Then productivity growth slowed significantly, but hourly pay growth collapsed even faster, leading to a growing gap between these typical workers’ pay and overall growth. That wedge of missing pay for typical workers went either to workers at the top or to business owners.
The most important driver of this wedge between typical workers’ pay and economy-wide productivity is the growing concentration of labor income at the very top of the wage distribution. Due to excess concern about confidentiality, most public data available to researchers inconveniently suppresses information about the wages of the highest earners by assigning a uniform (and too-low) “top-code.” But one enormously valuable data source does not – the Social Security Administration (SSA) data on annual wage earnings. This data sorts individual workers by what they earn each year and allows us to see if earnings growth at the top outpaced growth for the vast majority. This is straightforward; there is no unreported data or allocation decisions to be made. It is simply measuring if individual earnings are growing more unequal over time – and they absolutely are. The latest SSA data demonstrates how vastly unequal earnings growth has been between 1979 and 2022. Over that period, inflation-adjusted annual earnings for the top 1% and top 0.1% skyrocketed by 171.7% and 344.4%, respectively, while earnings for the bottom 90% grew just 32.9%. This unequal growth has seen a rising share of total earnings in the US economy accumulate at the top of the wage ladder.
The share of earnings for the bottom 90% fell 9.7 percentage points between 1979 and 2022, while the share of earnings for the top 5% grew 8.8 percentage points. The very top – the top 0.1% of wage earners – nearly tripled its share of total earnings from 1.6% in 1979 to 4.6% in 2022. These gains at the very top are mirrored by data on CEO pay in the United States. Compustat, a financial database, tracks executive pay at all publicly owned US firms. In this data, inflation-adjusted pay for CEOs of the 350 largest publicly owned US firms grew 1,209% (not a typo) between 1978 and 2022. Further, the gap between CEOs and typical workers has grown dramatically over the last few decades. In 1965, CEOs were paid 21 times as much as the typical worker. By 2022, CEOs were paid 344 times as much as the typical worker.
All of this straightforward data on earnings clearly shows a huge rise in inequality of pay in the US economy. Moreover, labor income is far more equally distributed among US households than income derived from wealth (for example, the top 10% of US households holds about 85% of all corporate stock – including indirect ownership through 401(k)s and other retirement vehicles). Hence, if labor income is becoming so much more unequal, are we really to believe that ownership of corporate equities or other financial assets has become radically more equal to make up for this? No data suggests this is true. Arguing about whether inequality has significantly increased seems like a real step backwards in recent years. The real debate should be about why and what to do about it. The fact that so much inequality comes from the labor market is a real clue that much of the answer is that policy changes intentionally disempowered typical workers in their bargaining over wages with employers. Reversing it will require efforts to re-empower workers, through expanded unionization, keeping unemployment rates consistently low, buttressing key labor standards that have atrophied, blocking employer efforts at wages suppression and restraining wage growth at the very top.
CNN