The Nonissue of Income Inequality

[I’ve written essays in semesters past that I’ve been meaning to post here. This is the longest of them. For ease of reading, I’ve removed the in-text citations and inserted the references as hyperlinks. In addition, I’ve made some (mostly minor) revisions in language in a few places. Otherwise, what you’re about to read is what my professor graded me on.]

Among the most politically fraught issues of recent memory, economic inequality has been referred to as “the nation’s biggest economic challenge in the coming decade” and “the defining challenge of our time.” Certainly, we hear enough about it in the news: notions that decades-long wage stagnation and economic immobility are the inevitable lot of the ninety-nine percent, while the wealthiest Americans see increasingly obscene gains (ostensibly at the expense of everyone else, and the result of a rigged system), are practically received wisdom in 2019.

However, this popular wisdom, as is so often the case, is misleading. Different studies produce contrasting results, and, depending on the methodology and assumptions a researcher chooses, the stories of American prosperity can vary tremendously. Viewed through the proper lens, Americans today are a great deal more prosperous than we were in the past. Furthermore, the system isn’t rigged to favor the wealthy over “the rest of us”—in fact, there is little reason to believe that money “wins” elections, nor is it clear that money “buys” sitting politicians. This isn’t to say that everything is peachy, of course. There are ways in which the lives of everyday Americans can be improved substantially through government actions. By examining each of these areas—methodologies and assumptions, campaign finance concerns, and policy implications—it quickly becomes evident that economic inequality is among the biggest nonissues of our time.

And so we go.


We frequently hear contradictory statistics thrown about regarding the nature of economic inequality in the United States, but there isn’t anything sinister or mysterious about this. Economist Stephen J. Rose of the Urban Institute reviewed half a dozen studies of economic inequality that took place between 2003 and 2018 and found that differences in methodologies led to a range of significantly varied results. At one end of this spectrum, a study conducted by economists Thomas Piketty and Emmanuel Saez (“narrative study”) finds that all the gains from 1979 to 2014 accrued to the wealthiest ten percent of Americans, while the median income in the United States simultaneously fell by eight percent. This is in keeping with the standard economic inequality narrative. At the other extreme, a study conducted by the Congressional Budget Office (“counter-narrative study”) finds that while the wealthiest ten percent of Americans did accrue a substantial portion of aggregated income gains (about forty-six percent), the median income nevertheless rose by fifty-one percent. Why the stark difference?

The figures upon which the narrative study is based are pre-tax/pre-transfer incomes of individual tax filers, not including employer benefits. To account for inflation and other changes in prices, the study uses the CPI-U-RS (consumer-price-index research series using current methods) as their price deflator. By contrast, the counter-narrative study uses the post-tax/post-transfer incomes of the size-adjusted American household (because the median household has shrunk over time) and elects to include employer benefits. Its price deflator is the PCE (personal consumption expenditure)—which is arguably superior to the CPI-U-RS on several dimensions.

The main difference between these two methodologies concerns total compensation versus money income alone. Because individuals only directly benefit from income that is leftover after taxes, and because transfer programs such as Social Security and Medicaid provide substantial income to individuals at the bottom end of the income distribution, excluding these considerations from any analysis of economic inequality produces a distorted view of the present financial well-being of individuals, particularly at the bottom of the distribution. Likewise, employer-provided benefits comprise an increasing proportion of total compensation over time, and ignoring them gives us the false impression of stagnation amongst the ninety-nine percent. The upshot of these methodological adjustments is that inequality in total compensation grew very little between 1992 and 2010, and the reason may come as a surprise: the poor in America are seeing their incomes increase faster than the rich.

To be fair, it’s true that the middle class is disappearing, as we so often hear. Between 1979 and 2014, the middle class declined from 38.8% to 32.0% of the total population. Usually, the analysis stops there, and the politicking begins. But this is a mistake because it leaves out a crucial question: did they fall down the income ladder? Happily, no. In fact, the bottom two quintiles (that is, fifths of the income distribution) also decreased in size, falling from 24.3% to 19.8% and from 23.9% to 17.1%, respectively. Meanwhile, the upper-middle class skyrocketed in size from 12.9% to 29.4% between 1979 and 2014, and the rich increased in proportion from 0.1% to 1.8%. Contrary to the implications of the popular wisdom, Americans are ascending the income ladder over time.

This raises a familiar objection: “Studies show that economic mobility is dead in America. If you’re born wealthy, you’ll stay wealthy, and if you’re born poor, you’ll stay poor.” This, too, is misleading. Addressing the concept of economic mobility requires that we acknowledge the fact that “the rich” and “the poor” are just analytical classifications, not persistent groups comprised of the same people throughout time. In discussions on economic inequality, most don’t take the time to consider that distinction, but it’s an important one. To get a clearer perspective on economic mobility, we need to follow the same people through time.

Researchers at the University of Michigan have done just that, following thousands of individuals between the ages of 25 and 60 since 1968 and compiling a widely used dataset called the Panel Study of Income database (“PSID”). Drawing on the PSID, a recent study found that nearly 70% of the population will spend at least one year in the top quintile of the income distribution, usually during the period of their greatest lifetime earnings (ages 45 through 54). They also found that more than half of Americans will spend at least one year in the top 10%, and more than 1-in-3 will spend at least one year in the top 5%. Finally, they found that more than 1-in-10 will spend at least one year in the top 1%.

Furthermore, a study published by the Federal Reserve Bank of Dallas that followed individuals for a decade-and-a-half found that just over five percent of those who started out in the lowest quintile at the beginning of the study remained there 15 years later. By contrast, the study found that 14.6% who began at the bottom had moved up one quintile, 21% had moved up two quintiles, 30.3% had moved up three quintiles, and fully 29% had risen from the bottom quintile all the way to the top quintile by the end of the 15-year observation period.

Purveyors of the popular wisdom often balk at such results, pointing to studies that compare economic mobility in the United States with other countries, which find that the U.S. fairs poorly in relative terms. However, this seems to be because individuals in other countries are more likely to fall down the income distribution than are Americans, a fact that’s obscured because economic mobility is usually measured in such studies without differentiating upward from downward movement.

None of this is to suggest that economic mobility is rising. In fact, most research finds that economic mobility has been relatively stable for the last two decades, though some research indicates a decline relative to the above findings. This is not of any special moment, however. As economist Steven Horwitz notes in one such study,

As the total amount of income grows, the width of each quintile expands as well, requiring progressively more income each year to move up from one quintile to another. If what concerns us about mobility is the ability of households in, say, the bottom two quintiles to move up the income ladder, the spreading out of the quintiles will mask actual improvement within each quintile [emphasis in original].

Bearing this in mind, even the more pessimistic findings paint a far rosier picture than the one painted by the popular wisdom.

Though it is certainly encouraging to many that total compensation is growing and that there is reasonable economic mobility in the United States, to some it is not clear that these are the relevant metrics by which to measure economic inequality. Consumption is the ultimate end of income, one might contend, and inequality of consumption between rich and poor has risen. This is, in fact, true. However, comparisons in the consumption of goods are better than they first appear, a fact which is borne out by the data. Between 1963 and 2014, consumption inequality between the richest 10% and the poorest 10% rose by only 7% for the median household, and it fell by 20% for single mothers and single individuals, groups for whom economic inequality simultaneously grew most during that period.

What does that look like for the poor in America, in practical terms? While the nominal dollar cost of goods rose between 1959 and 2013, inflation-adjusted prices declined precipitously when considered in terms of “hours of work.” For example, in order to purchase a bundle of eleven appliances—washing machine, gas clothes dryer, dishwasher, refrigerator, freezer, gas stove, coffee pot, blender, toaster, vacuum cleaner, and color television—the average poor household had to work 885.6 hours in 1959. By 2013, the average poor household would only have to work 170.4 hours.

And it gets better. Dartmouth economist Bruce Sacerdote finds that between 1970 and 2015, there was “a 164% increase in consumption for… households [below the median income]” when accounting for the quality, reliability, and feature improvements that are ignored by the CPI. Even granting the dismissal of the facts that “rich” and “poor” do not necessarily describe the same people over time, and that total compensation has grown faster for individuals in the lower income quintiles than for those in the upper quintiles, it must nevertheless be admitted that, in practical terms, being poor in the United States in this decade is far better than being poor in any decade past.


There is concern, however, that rising economic inequality threatens democratic institutions. The conventional wisdom concerning money and politics makes sense, at least at first blush: the more money one can spend, the more influence one has on the outcome of an election and the more influence one can exert on the legislator thereafter. But this is an empirical question, and there are more than one hundred years’ worth of data with which to test this assertion.

In the first place, as it turns out, real campaign spending as a fraction of national income did not grow at all in the 120-year period of 1884-2003, a period in which the United States federal government grew tremendously in size, scope, and spending capacity. If the standard explanation were correct, we would expect to see spending on political campaigns—best understood as bidding in auctions—rising relative to GDP as the capacity of the government to grant economic favors increases.

Now, campaign spending in all 2016 federal elections totaled $6.5 billion, a sum which is cast as an enormous amount of money by those alarmed by economic inequality. But compared to the federal government’s annual expenditures, which amounted to $4 trillion in 2017, it is a minuscule sum. With a prize worth well over 600 times the total campaign expenditures of every congressional and presidential candidate, the level of campaign spending is actually paradoxically low when viewed through the lens of the popular wisdom.

To illustrate, allow me to adapt an example given by economist Steven Landsburg: imagine yourself at an auction bidding for a six-thousand-dollar platinum coin. What is the maximum amount you would be willing to bid for that coin? Unless you’re completely irrational, the very most you should be willing to bid is six thousand dollars. If you’re the only person in the room, you could win the coin with the minimum bid of one dollar, thereby netting yourself a tremendous gain of $5,999. But if there is another bidder, your measly minimum bid would quickly be beaten, and as bidding continued the six thousand dollar-mark would soon arrive. In the presence of a second bidder, you won’t get away with buying the coin for anything less than its face value because the other bidder stands to gain as much as you do from winning the auction at any bid less than six thousand dollars. So, even if you must incur debt to obtain that six-thousand-dollar coin, it would be a perfectly rational thing to do, even if you were to win the coin at its face value.

Now stipulate that, win or lose, both of you must pay the amount of the highest bid. Because of the nature of this rule, both of you face that much more incentive to win the coin because neither of you wants to lose both the bid and your money. However, now there’s no good time to stop bidding, which removes the erstwhile incentive for either of you to stop bidding at face value. Instead, when the six-thousand-dollar mark arrives, bidding suddenly becomes a matter of who loses the least money, and it continues far, far beyond where it rationally should have stopped.

Return now to the view of political-spending-as-auction and consider four years’ worth of government largesse, worth $4 trillion per year. Why is so little spent on influencing election outcomes? One might respond that “only” $1.2 trillion annually is actually discretionary spending, but that doesn’t change the point in the slightest—if we are to accept the idea that political spending wins elections, we must ask how politicians consistently go home with a prize worth several thousand times more than what they are individually willing to bid, especially given the large number of competing bidders. As in the second scenario, win or lose, all parties in an election lose their expenditures. So, if they expect that campaign spending will win the election for them, all parties should be willing to spend whatever is necessary to win, even if it means incurring billions of dollars in debt, for there is no good time to stop “bidding” until no more debt can be incurred, and losing means being unable to defray their losses with government largesse. Again, however, this is not what we see. Political-spending-as-auction is an interpretive framework that fails to explain observable events reasonably.

Instead, campaign spending might be compared to the research and development undertaken by private firms each year. In 2017, the five highest corporate R&D budgets were Amazon at $22.6B, Alphabet at $16.6B, Intel at $13.1B, Microsoft at $12.3B, and Apple at $11.6B. The market value of these companies that year were $423B, $579B, $171B, $509B, and $754B, respectively, meaning that these companies spent between 2.3% and 7.7% of their total respective market capitalizations on R&D. By comparison, the United States Federal Government, an entity worth $4 trillion, saw election spending of $6.5 billion, amounting to 0.16% of its total “market cap,” or between 2.1% and 7.1% of what private firms were willing to spend, relative to their respective market caps. If money were deterministic in federal elections, it seems unlikely that interested parties would be willing to spend such a proportionately minute amount of money in “political R&D.”

The best explanation of campaign spending is that it is a form of political participation on the part of voters. In this case, we would expect to see individual contributions get higher as we go up the income distribution. However, we would also expect to see factors other than money influencing the outcomes of elections, and to a greater degree. Finally, we would expect to see that contributions have a minimal impact on the voting patterns of legislators, especially compared to other factors. In fact, the data bear these expectations out: while individual contributions do rise with income, it is ultimately the preferences of voters, party affiliation, and the policy preferences of the larger party that bear greatest sway over election outcomes and legislator voting. The influence of campaign contributions is marginal at best, not deterministic in any meaningful sense.


Does this mean there are no concerns to be had over the potential relationships between the super-rich and the government? Not at all. There is a reason lobbying is a billion-dollar industry. In their 2016 book, Equal is Unfair: America’s Misguided Fight Against Income Inequality, journalist Don Watkins and finance Ph.D. Yaron Brook write,

There are genuine barriers to opportunity, and the deck is becoming stacked against us—but not because ‘the rich’ are too rich and the government is doing too little to fight economic inequality…

When a bank or auto company that made irrational decisions gets bailed out at public expense, that is an outrage. But the root of the problem isn’t their executives’ ability to influence Washington—it’s Washington’s power to dispense bailouts.

On their view, the problem is that the United States Federal Government has too much power to tax, spend, and regulate, and that, contrary to popular wisdom, stripping it of much of its existing power in these realms will lead to greater prosperity among the masses and less income concentration at the top.

In fact, there’s good data to support their position. The Fraser Institute’s Economic Freedom of North America Index (“EFNA”) evaluates each state in the United States on twenty-eight economic variables that fall under the three general headings “measures of government spending,” “measures of taxation,” and “measures of labor market freedom,” then combines them into a single index on an ascending scale of relative freedom from 1 to 10. The higher on the scale, the freer the state. Drawing on data from Fraser’s EFNA Index from 1981 to 2009, economists Antony Davies and Megan Teague and political scientist James R. Harrigan find that the states with the most economic freedom consistently have higher rates of household income, income per capita, population growth, in-migration, and economic growth. Likewise, the most economically free states see lower rates of unemployment, poverty, income inequality, uninsured residents, taxes, and debt.

Internationally, a study of 58 countries of varying politico-economical freedom and development over the period 1980-2010 found that, in accordance with the conventional wisdom, increasing economic freedom from a relatively low starting point resulted in increased economic inequality. However, the study found that beyond a certain degree of economic freedom there is an inflection point where the relationship between economic freedom and economic inequality reverses and incomes between rich and poor begin to converge—that is, a point at which the poor begin to see a disproportionate share of the benefit of economic freedom. Crucial in these findings is that transitioning to the degree of economic freedom necessary for incomes to begin converging requires restricting the ability of governments to intrude into the activities of their respective economies.

Indeed, heavy regulation exacerbates income inequality. Economists Patrick McLaughlin and Laura Stanley find that adding 3.5 regulatory steps to the process of opening a small business leads to a 5.6% increase in the share of the income held by a country’s wealthiest 10%. Not only does this imply that deregulation is the friend of would-be small business owners, but it also implies that regulation is the friend of large businesses, who can afford the legal teams necessary to comply with the legislative barriers erected by government to keep out the small firms who can’t cope with the sea of bureaucratic red tape.

On the surface, this might seem to run counter to this author’s overall thesis. After all, wasn’t the claim made previously that campaign contributions, per se—and, by extension, economic inequality—pose no threat to democracy? How does one square the finding that seventy-five percent of the literature indicates that donations do not buy the votes of legislators or have a deterministic influence upon their patterns of voting—a finding that even holds in a post-Citizens United world—with the fact that special interests do, in fact, lobby governments for favors? If the endeavor were simply a waste of money, why would interest groups engage in it to begin with?

Summarizing the findings of previous researchers, political scientist Amy McKay suggests a plausible explanation: politicians elicit money from lobbyists, but any contract that is created by these contributions, implicit or otherwise, is unenforceable. Furthermore, “contributions on both sides of an issue tend to balance out,” usually resulting in no change in policy. Taken together, along with the fact that only ten percent of businesses engage in lobbying efforts, these facts suggest that lobbying, in the main, is not an irresistible force wielded by the super-rich to fool naïve legislators into giving them preferential treatment. Rather, it is the rational response of private parties to public servants empowered to offer up favors in the first place, and its outcomes are far from guaranteed.

Therefore, the best policy that an inequality-distressed government can pursue is a policy of economic non-intervention. Inasmuch as the government has no power to hand out favors to the wealthy, the wealthy will have no incentive to seek favors from the government. In the United States, we have a culture of separation between church and state—we should extend that culture of separation to state and economy, for the data clearly support it.


The data show that life in America—economic life, anyway—is continually improving, year over year. The poor and lower-middle classes are graduating into higher income brackets over time because economic mobility is alive in the United States, and democratic institutions are reasonably safe from corruption by the monetary influence of the wealthy. No, things are not perfect—the fact of extensive government influence in the economy creates perverse incentives for interest groups and legislators to fraternize—but it’s no ailment which more freedom from state intervention couldn’t ameliorate. The popular wisdom, which marks economic inequality as a crisis in need of rectification, is mistaken. The reality is that economic inequality is really no threat at all, making it one of the biggest nonissues of our time.