Setting the Record Straight on Income Inequality

In The Myth of American Inequality: How Government Biases Policy Debate by Gramm, Ekelund, and Early, their methodological message is simple: “The Census Bureau is accurately measuring what it has chosen to measure, but it is not measuring the right things” (p. 3). Their empirical message is also simple but a lot more controversial:

• There are certainly people who are physically or mentally unable to care for themselves and have fallen through the cracks in the system that delivers transfer payments, but, for all practical purposes, poverty due to a lack of public or private support has been virtually eliminated in America. (p. 4)

The post-tax, post-transfer poverty rate, they argue, is not the 12.3% the Census Bureau reported in 2017, but about 3%. The enormous difference is because the Census Bureau is not measuring the right things.

As Gramm, Ekelund, and Early make their argument, they kick the legs out from under the conventional wisdom about inequality. Taxes in the United States are extremely progressive, and the rich pay a higher share of taxes than they earn in income. Income mobility is alive and well. “Tax cuts for the rich” made “the rich” reorganize their affairs so that, as they quote JFK, they spent less time and money on “avoidance of taxes” and more on “production of goods” (pp. 115-116). “Tax cuts for the rich,” they argue, in 1962 and 1986 actually led to the top 1 percent and the top 10 percent paying higher percentages of their incomes in taxes.

One of their most striking findings compares the real income distribution today to the real income distribution of the 1960s. After accounting for taxes and transfers, Gramm, Ekelund, and Early argue that the top three quintiles in the 2017 income distribution and two-thirds of the households in the second quintile from the bottom have incomes that would put them in the top quintile of the 1967 income distribution. They recount a series of facts that we take for granted about the diffusion of indoor plumbing, air conditioning, cars, and televisions. Half of households in the bottom two quintiles own homes (p. 28). Restaurant meals used to be the privilege of the rich, but in 2017, “the average bottom-quintile household spent 34 percent of its food budget away from home,” much more than the 27% of top-quintile households in 1967 (p. 146). For the modern rich, cooking is a hobby, not a necessity.

To get an idea of what this means, watch an episode from the first season of The Wonder Years, which ran from 1988-1993 and was set in 1968-1973. Would we call the solidly middle-class Arnold family of the late 1960s and early 1970s “poor”? Would we say they hadn’t achieved the American Dream? Their standard of living would have put them in the bottom 40% and maybe even the bottom 20% of the 2017 income distribution. To the extent that the American Dream is harder to achieve today, it is because we have redefined it to mean two (or three) cars rather than one, central heating and air conditioning, a microwave, a programmable coffee pot, multiple streaming subscriptions feeding to multiple giant flat-screen TVs, computers, mobile devices for everyone in the family, an Xbox, and multiple bathrooms (with toilets that would flush and showerheads that would work if they weren’t regulated to reduce water use).

In his book The Vision of the Anointed: Self-Congratulation as a Basis for Social Policy, Thomas Sowell referred to what he called “A-ha!” statistics that support left-wing narratives, at least superficially. A bit of further examination, however, causes things to come apart. Talking about what has happened to “the poor,” “the rich,” or “the middle class” across long periods is dangerous because the people at the bottom of the income distribution in 2023 are not the same people who were in the bottom of the income distribution in 1983.

First, there is the fact that income distribution statistics are comparing people at different life stages. I was at the bottom of the income distribution as a graduate student in 2002, but it would have been silly to say I was “poor.” I moved up after my wife and I married and created a new household that merged her income with mine. I was in another place in the income distribution in 2012 as an Assistant Professor with three young children, and I occupied yet another place in the income distribution as the Margaret Gage Bush Distinguished Professor of Economics in 2022. I will likely fall down into a lower income quintile should I ever decide to retire (not likely).

Second, there is the fact that the population is always changing. Suppose you have five people. One earns $25,000; one earns $48,000; one earns $50,000; one earns $90,000; one earns $100,000. Then two things happen simultaneously: everyone’s income rises by $1000 and four people enter the labor market. One earns $24,000, two earn $30,000, and another earns $200,000. This means we now have one person earning $24,000, one person earning $26,000, two earning $30,000, one earning $49,000, one earning $51,000, one earning $91,000, one earning $101,000, and one earning $200,000. You can expect a flood of headlines about how median earnings have fallen to $49,000 even as the average income in the t​​op quintile has risen by about 50%. The ratio of top to median income has risen from 2:1 to about 4:1, the ratio of top to bottom income has risen from 4:1 to more than 8:1, and people with high incomes are “capturing” a larger share of the total income. The headlines paint a picture of an economy that is working for only the few at the top while everyone else suffers stagnation or decline even though everyone is better off. The real story—that people are earning higher incomes—disappears under the non-story that the median income has fallen not because any actual person’s income has fallen but because the population has changed.

The Myth of American Inequality answers the “what’s the point?” question people have about the arcane, minute details of academic research. Measurement is a lot harder than it appears at first. There are well-known problems of technological improvement—a car in 2023 and a car in 1993 or 1973 are very different things—and then there are questions about what to count as “income.” There are wages and salaries, but what about benefits? Should transfers like Medicaid and food stamps be counted as “income”? One of the book’s critics has pointed out that the payments Medicaid makes go to doctors rather than patients and therefore complicate the story, but the patients receive medical services that are, we hope, at least as valuable as what the doctors are being paid for them. Food stamps are not cash, but they spend like cash—at least on approved items (the political economy of how something gets and stays on that list is a dissertation waiting to be written). Do official measures overstate inflation? The answers change how we make policy, write op-eds, argue at the dinner table, and write textbooks.

Since the Census Bureau (and economists) are not measuring the right things, it’s hardly clear we are going to make the right policies. So what should we do in light of this new story about postwar inequality? First, Gramm, Ekelund, and Early argue, we must get the theory straight and do the math right. It is one thing to look at pre-tax, pre-transfer inequality and be horrified. It is something else to think you can get less pre-tax, pre-transfer inequality with taxes and transfers: you cannot, by definition. If we want to know the real story about actual, on-the-ground inequality for living, breathing people, we need to look at measures that account for those taxes and transfers. Once we do so, we get a very different story from what we read in the New York Times.

All this makes me wonder why the left does not embrace this book and take a victory lap on behalf of Lyndon Johnson’s Great Society. It seems like, “even Phil Gramm admits we won the war on poverty” would be cause for celebration. All I have seen from the left, however, is breathless indignation about how the authors are heartless because they say poverty has fallen and extol Ebenezer Scrooge’s thrift, and wrong because their approach jars with the claims of economists like Thomas Piketty that ignited the “Occupy” movement in the Great Recession’s aftermath.

Second, Gramm, Ekelund, and Early propose changes to how we do welfare with special attention to the work disincentives very high implicit marginal tax rates create. If the 70% marginal tax rates Alexandria Ocasio-Cortez proposed a few years ago would be disincentives to produce at the top of the income distribution, then surely similar implicit rates are disincentives to produce at the bottom of the income distribution. If slashing top marginal tax rates in the 1960s and 1980s unleashed forces of production, then slashing implicit marginal tax rates in 2023 would have a similar effect.

Third, the authors go after occupational licensing. When I teach about licensing in my classes, I talk about the medieval guilds and South African apartheid. The logic is the same in all these cases: an interest group enlists the state to protect itself and its members’ incomes from potential competitors. Lest one think these rules are about public safety, they write,

• New York State recently added a new requirement that entry-level shampoo assistants in beauty parlors and barber shops must complete a five-hundred-hour training course at an average cost of $13,240 before they can practice this complex art that most of us perform daily without mishap. Of course, three of the four regulators issuing this requirement have economic interests in companies that sell the required training. (pp. 181-182)

Fortunately, occupational licensing has gotten at least some bipartisan attention—and perhaps more people will come to understand that the poor and downtrodden wouldn’t need so much of our “help” if we weren’t spending so much on trying to “protect” them.

I would add another, though I admit it is a pipe dream: we should stop caring so much about inequality per se. It’s reasonable to care about poverty. Having enough food, clothing, and shelter to lead a meaningful life matters, and I am pleased to see people’s possibilities expanding year after year. It’s unreasonable to care about inequality per se. In a commercial society, whether your neighbor has more does not mean you don’t have enough. The qualifier “in a commercial society” is very important. Capitalism is the greatest positive-sum game in history. Competition for admission to elite universities is fierce, but have you been robbed if the kids whose parents read to them every night and hired math tutors get into Stanford, Harvard, and Penn while your kids have to settle for Samford, Haverford, and Penn State? If the extra effort these families put into getting their children into an elite school means they have more skills and higher earnings capacity, then everyone wins. These kids have only been able to get into an elite school by equipping themselves with the tools they need to serve people better. It only becomes a problem when people use those tools to rule rather than serve—by, say, establishing licensing boards and commissions to micromanage the lives of the unshampooed masses.

“As a nation,” Gramm, Ekelund, and Early write, “we need to get our facts straight.” I agree, and I hope this book moves the policy needle. Resistance will be fierce, of course, given that many people’s careers, incomes, and self-images are inseparable from the “rising poverty and inequality” story and the welfare state we have built to deal with it. The Myth of American Inequality is an important contribution, however, that shows how the story is built on a foundation of theoretical and empirical sand.

Footnotes

[1] Phil Gramm, Robert Ekelund, and John Early, The Myth of American Inequality: How Government Biases Policy Debate. Rowman and Littlefield, 2022.