Increases in Inequality Have Been Overestimated

Stephen Rose of the Urban Institute (not exactly a right-wing or libertarian think tank) compares recent studies measuring changes in inequality and finds that although inequality has increased the Piketty and Saez (2003) results, which generated a tremendous amount of discussion and research, are very likely over-stated.

The results from at least four studies were compared for three measures of income change: change in median incomes, share of growth captured by the top 10 percent, and the changing income share of  the top 1 percent. In all cases, Piketty and Saez (2003) were the outlier, showing the most increased inequality. And in all three measures of income change , Piketty, Saez, and Zucman (2018) found much less growth in income inequality than Piketty and Saez (2003).

This brief does a meta-analysis of different findings to estimate a “consensus” level of change…I find that instead of stagnating, real median incomes grew by just over 40 percent (1 percent a year) from  1979 to 2014;  the top 10 percent of the income ladder captured 45 percent of income growth from 1979 to  2014; and the share of the top 1 percent grew 3.5 percentage points.

All studies find that income inequality rose after 1979, but common perceptions that all income gain went to the top 10 percent and middle class incomes stagnated (or even declined) are wrong.

Russ Roberts also has several good videos showing how the numbers can be cut in various ways.


Still waiting for a funnel graph.

Unless you used a log scale, it would look more like a thumbtack than a funnel.

A funnel graph is a way to do an meta-analysis. In a funnel plot you plot many studies so that N is on the Y-axis and the effect on the X-axis.

Why would a funnel plot look like a thumbtack in this case?

I agree that it would not look like an inverted funnel. It would probably demonstrate publishing bias by being skewed in one particular direction. Your own comments below illustrate the zeitgeist.

Oh, I thought you meant a funnel plot of the income or wealth distribution, not of the studies in the literature. There aren't that many studies, so you wouldn't have much to plot yet.

I'm still going to vote for populists like Trump or Bernie. You can't stop me. Even with your fake news.

'and the share of the top 1 percent grew 3.5 percentage points'

Income? Sure, that is believable.

Because how many people in the 1% in America have been stupid enough to care about increasing income when the real money has been in all that non-income 'compensation,' starting with using stock as an incentive (leading to the now apparently completely accepted opportunities for self-aggrandizement for those able to award themselves stock and determine how a company treats its share price as corporate policy).

I think you misunderstand the 3.5% statistic. It’s not the increase in income, it’s the increase in the share of income. Big difference.

Good point, Ron. Tax info shows a much bigger income share change for the 1%, from 8% to 20%.

Qualitatively, inequality is even less than the numbers say. Historically, there were many things that only the rich could afford, but now there are reasonably good downmarket substitutes for almost everything. For example, today unlike in history, middle class people have the same legal rights as the very rich; eat a similar diet and have similar clothes; have entertainment options that are a close substitute for what the rich have (like movies v. plays); and use similar forms of transportation (the difference between a coach air seat or used Honda and private jet or Maserati is much smaller than historically when non-rich people couldn’t afford to travel at all). Thus, a middle class income in a developed country today gets you 90% of the way to a rich lifestyle, which wasn’t always the case. Maybe that’s why studies show that utility from more income flattens after about $75,000.

This is a great post, and everything you say is both true and widely known. But the fact is that discussions of "inequality" are nothing more than an excuse to raise taxes, which is why the Dems will never let them die.

There actually could be more "inequality" given the big individual differences in talent, drive and motivation. We've kind of smoothed all that over.

Maybe the one place inequality does show up is not daily lifestyle but in an emergency -- you get sick, your car dies, your job goes away -- you don't have that buffer that the very affluent can turn to.

Income inequality is a farce. Given how the US can simply print more money, the upper limit of income is effectively infinite. However, the lower limit of income is effectively zero. Therefore, over time (and generous QE from the Fed) the difference between rich and poor (or inequality) will grow

The big individual differences cannot possibly be accounted for by talent and effort alone. Jeff Bezos has gained ~$40 billion in net worth this year, while the average brain surgeon made $750K. Are you seriously going to contend that the individual talent and effort of Jeff Bezos is the equivalent of 53 brain surgeons . . . or that maybe, just maybe, we have an economic system with a notable amount of pure economic rent built into it?

Sorry, that's 53,333 brain surgeons!

Mr. Rose might suggest 53,333 overstates the issue. 50,000 perhaps would be more accurate. Problem solved. Inequality is not an issue.

Yeah Bezos created a hellava more value than the average brain surgeon. He should definitely be compensated this much. He changed my life and that of millions. No brain surgeon has had that much impact on so many lives.

Stop being so envious. Try to do what Bezos did. He helped the poor more than you would ever even if you could take the money of the rich and redistribute it the way you think is "fair".

(And thank God you can't.)

Yeah, but by 2018 Bezos had already had that impact. The billions he raked in this year are just a bonus.

Also the "impact on so many lives" that Bezos has had is a product of not just his talent and effort, but the talent and effort of tens of thousands of people, 99.99% of whom have never made a total of a billion dollars, let alone 40 in one year. Without those people, Bezos wouldn't have impacted jack shit.

Moreover, the fact that we are talking about this particular tech entrepreneur, and not the thousands of others who didn't manage to catch fire in a similar way is largely a matter of very marginal differences in positioning. Was Bezos thousands of times more talented and hard working than all of them? No, he was very slightly better than them in a market with network effects, which means winner-take-all. That is not a merit-based reward, which is the whole point. Pre-tax incomes are not meritocratic.

If the people working for Bezos are only marginally less talented then he is, then they could open up their own businesses or find other work with other firms (in competing or entirely different fields), but they don't, so maybe they aren't so talented after all (rather than dismissing it merely as a matter of positioning).

And even if it could be reduced to economic rents, then those other workers have an even weaker claim to collect any share of those rents than Bezos.

If there is any specificity in human capital (which there obviously is, skills are not interchangable), and there are network effects, then marginal differences will lead to massive gains, and the losers in that competition will have nowhere to go that could yield comparable gains. This is not a difference of talent, it's a structural problem. By insisting on boiling down inequality to individual differences, you are basically treating the rules of the economy as a natural given. Why do that?

As for rents, that Bezos has some sort of legal claim to them is true, but is an example of the same error -- of assuming that the way the rules of the economy are right now is the only* way they could be set up.

* Or the "best" or the most "natural" . . . It's unclear which flavor of knee-jerk capitalism defense you subscribe to.

The median ages for the top 10/1% are usually pretty steadily around 50-60, so we would expect nearly all of that cohort from 1979 to be dead today, thus the implication they've "captured" anything is somewhat misleading -- more accurate to say they were replaced in the cohort by new high earners (formerly younger and poorer) who earned a greater proportion of total income than those they replaced.

McArdle also pointed out a long time ago that changes in tax and related law, such as LLCs, made income a more attractive vehicle over time.

Consumption inequality is a more meaningful measure -- income inequality also tends to assume flatter distributions produce more consumption, but it's not at all clear what ratio would actually maximize living standards (particularly with respect to leisure).

Between nations and therefore world consumption inequality might also be slightly flatter compared to GDP - and (from Random Critical Analysis's latest excellent blogpost on relationships between consumption and healthcare spending - Not so sure about income though.

Many studies have determined that the very wealthy (the top .1%) hide much of their income and the assets that generate that income, as high as one-fourth of their income and assets. It's concerning, because the very wealthy are avoiding payment of taxes, thereby shifting the burden to those with lower incomes who don't have the ability to hide their income. What happens to that hidden income and those hidden assets? They aren't invested in productive capital in the U.S. that's for sure. As I've expressed many times, my concern about high levels of inequality is its affect on financial and economic instability, something that history seems to confirm. We rely on rising asset prices for prosperity, which is a recipe for instability. Even now the Fed is facing a dilemma: raise interest rates to tame asset prices but risk an economic downturn, or maintain interest rates at or below the current level but risk asset bubbles and financial crisis. It's the inequality, stupid!

You're actually just pissed that the real estate developer down the street from you has more stuff than you do. Quit pretending to be an altruist.

A tax on invisible assets will bring that money back from Never-Never Land!

Well, that's nice. Now can we please devote some attention to improving living conditions among the impecunious rather than to the kvetching of politicians and professors about the standard of living among business executives and successful proprietors? For starters, get the hoodlums off the street and into jails and prisons where they belong, get the incorrigibles out of the schools and into county detention centers where they belong, and replace a dozen different grant programs and coupon programs with a negative income tax or wage subsidy? Why is it the people yapping about 'equality' also seem to believe we must leave no social worker behind?

Exactly, we should concentrate on the petty criminals getting away Scott Free, rather than the powerful or connected. There is no inequality to see here. Fake news.

How did that link get there? Is it someone at Twitter hacking my messages just like they did my ex friend Jeff Sessions? There is nothing to see here.

They did it again! I'll have to ask my friend-for-now Geraldo Rivera weather this constitutes illegal harassment.

You mean the 'petty criminals' who have produced a homicide rate in my hometown 8x that of the surrounding suburbs?

Haven't you heard? The mass incarceration problem is a bunch of people getting locked up for weed. Any evidence to the contrary is fake news.

@Tabarrok to the Pillory: ..............>> "kvetching of politicians and professors"

Yes indeed. But this silly fixation upon economic "Inequality" reflects a much deeper problem of leftist ideological dominance in our academic/political/media establishments.

This "Inequality" meme is classic Marxist framing -- Tabarrok is casually feeding the beast here & granting credibility to that destructive meme.

I think this study misses Piketty's larger point: it's not the top 1% in INCOME that's distressing to some, it's the top 1% share of WEALTH (wealth is net worth: assets minus liabilities) that increases over time during peace (r > g in Piketty's thesis). That said, I think wealth and income are correlated, but I bet that wealth changes much more slowly than income (is lost more slowly, takes longer to gain than income, since wealth is the integral of income and accumulates). As Oprah Winfrey once said, the fist ten million is the hardest to accumulate, after that, the capital works for itself. In our family we've found that to be true. Even with 1% interest rates we make more money on interest than probably many Americans make in a regular job. But we've still not made the coveted "living on the interest of the interest" threshold, like the 0.1% do.

Bonus trivia: to be in the top 1% requires a minimum net worth of $10M. It's kind of nice, for one thing, you don't have to work, just let your capital work for you. That's what I did (quit my six figure job in my 40s, not missing it either, though you do have to take a hit in income and you end up begging for an allowance from your parents, but no different IMO than depending on fickle clients and customers). Also reading now the prophetic Kevin Phillips late 1980s book "Politics of Rich and Poor" which got a lot of Reaganomics right; he also predicted in the late 1960s the Republican revolution of the 1980s.

Ray is right: it's rising asset prices that maintain and increase the wealth (and income) of the top 1%. The problem is that with each generation there are more and more to support in the lifestyle to which they have become accustomed, thus increasing the need for ever increasing rising asset prices to increase the wealth and the income. It's no coincidence that rising inequality (from the later 1970s and early 1980s) occurred at the same time as the consolidation of the wealth of America's wealthiest families in investment partnerships intended to increase the rate of return through increases in asset prices, the need to increase the rate of return resulting form the proliferation of family members. Of course, increasing inequality and the consolidation of wealth makes it harder and harder to increase the rate of return to support the ever increasing number of family members.

What distinguishes "less unequal" and "more equal" from . . . "equal"?

When, that is, does "equality" become exactly the egalitarian notion it is supposed to be, with all the intrinsic equalities that the notion requires?

(What does "more equal" mean, perzacktly? [When do we begin to clear our heads of the odious comparative forms to which decades of advertising copy have habituated our superlatively cognitive and rational minds?])

When do we begin to jettison "equality" for the rank political idealism that it is? Why must American devotion to its tattered Constitution require deference to 18th century political idealism in an age of political fantasy?

Somehow fixing "inequality" requires that other equally vague and annoying term -- "social justice."

At inequality can be more or less measured. When I ask colleagues about “social justice” and how we might define it, and how we might measure justice, I get incredulous looks and comments to the effect that we are far, far away from social justice.

'Why must American devotion to its tattered Constitution require deference to 18th century political idealism'

Because the idea that all citizens are equal in the eyes of the law is one of the most revolutionary concepts that the American Revolution implemented.

Perhaps it is the case, then, that political idealism inevitably yields political fantasy . . . . discrepancies between innocuous idealist assertions and sordid political realities are being perceived these days: with a Constitution over two centuries old, we are due and overdue for a complete Constitutional overhaul that few of our "equal citizens" have the stomach for, in spite of all growing evidence of political indigestion and distress.

'Perhaps it is the case, then, that political idealism inevitably yields political fantasy'

Everything ends.

'we are due and overdue for a complete Constitutional overhaul'

Or not - the Constitution is an amazingly revolutionary document, and the temper of the times (just as it was at the time of the Contitution's creation) seems opposed to free speech or free exercise of religion, to name two more of the Constitution's triumphs.

We can go back and forth forever, of course.

The daftest statistical "discovery" I've seen reported lately - in a blog thread - was that the mode age of death in the UK (or maybe in England and Wales) changed only a little from 1840 to 1930.

The secret of this madness turned out to be that the original gatherers of the data had done the calculation neglecting all deaths before age 10. (As of course they had said.)

It did remind me that my first stats lecturer had urged us all to buy "How To Lie With Statistics".

I am totally against equality. I just think poor people in the United States should have more money on average than they do now and that rich people in the United States should pay more tax. But there should be sumptuary laws introduced that allow the rich to wear codpieces with their size based upon how much tax they pay which would allow everyone to see how extremely unequal they are.

Mr. Rose perhaps could advise whether plus/minus one foot overstates target accuracy on a North Korea nuke with a 100 square mile firestorm impact. Suppose five minutes overstates the time for a billion-dollar annual income to make a $15,000 minimum wage income. What might Mr. Rose suggest?

Price floors can be closely related to income inequality in the United States of America and around the world. Minimum wage creates unemployment around the workforce and work place. This minimum wage subject will cause little disturbance in the American economy as a whole. Airline and aeronautics programs are responsible for massive technological growth and therefore, creation of minimum wage jobs and higher paying private firm positions. Banking industries invest in numerous materials science importation projects that will further widen the income separation among the top ten percent and bottom ten percent of the world economy. Deregulation plays a major role in the re-involvement of this income inequality in 1979.

Income versus wealth seems a huge distinction, no? What % of 'reward' do you think the rich can 'hide' from the taxman? And the poor?

It does seem that the idea of no gain among poor may be overstated, but I think people are responding to unequal growth more than absolute. What's the great Menchen quote: being wealthy means having more money than my wife's sister's husband...

Although it's an important topic to study, an academic or researcher presenting their "findings" isn't going to convince people that there's no "inequality" if they feel that there is, and, G-d forbid, they might just have a point.

We need concerted efforts to massively increase inequality. That is the only way to defeat the SJWs and progressives.

Mr. Perlman, are you suggesting a return to the most profitable economic system in U.S. history? Owners in said system received lots of “value”. Concurrently, another economic system transitioned to an industrial revolution. Perhaps you could expand on the type of economic system you envision.

@Chuck, Here is a plagiarized response to your question -

"Time, resources, and effort spent on ‘redistribution’ are time, resources, and effort not spent on production."

By the way I am Alex Perlman's father.

The above response has been plagiarized by me from

To complete the answer to your question - I envision a society in which poor people are ruthlessly punched, kicked, and slapped until they begin to produce something and begin paying taxes.

A pleasure to meet you Mr. Perlman. Is their supporting data for either of your comments? Comments without data, while sometimes entertaining, perhaps are a missed opportunity.

Inequality is economically meaningless in a capitalist democracy and emblematic of a collectivist oligarchy.

Well stated, Willitts. Have any thoughts on how to get us out of the mess we’re in?

@Chuck, I am George's father. We are not at all in a mess. I want to see even more inequality to emerge and it to be the norm. I want a society in which the less equal are constantly humiliated and made to remember that they suck and that they are worthless pieces of sh**. Oh by the way - in such a society the rich actually celebrate while the poor are being slapped, and kicked in their butt all the time.

Does the society you describe have a name?

I am George's grandma. That society is called "normal".

Well stated. Have any thoughts on how to get us out of the mess we’re in?

The headline here is that increases in Economic Inequality(post 1979) have been overestimated based on the findings of multiple studies. These studies were conducted in response to the Piketty and Saez Study of 2003(P/S'03) that is referenced in this post. The P/S'03 claims that only the top 1-10% in our society are reaping any real benefits of Economic Growth. However, as mentioned above according to the Urban Institute and multiple studies, this is not the case. The studies looked at Median Income trends and % Share of Economic Growth. Based on the numbers shown it is easy to see that rising economic inequality is definitely happening. That being said P/S'03 paints this as an epidemic. I am not knowledgeable enough to dispute this, however based on my understanding of this post I find it does seam that the problem and effects are being overstated. Economic Inequality is a problem. That being said, if the quality of life of the bottom 90% are also improving at an acceptable rate then Economic Inequality isn't necessarily a bad thing. To say that their is NO benefit to Economic Growth for the bottom seams shortsighted. For example the Internet and Tech Boom along with increasing advancements in medicine have made a noticeable impact for the better on Americans. Technology is now much more accessible and affordable than in decades past. People are living longer and happier thanks to advances in Medicine & Mental Health Science. Those are things that have benefited Americans. Not just Rich Americans. It is common to assume that Economic Inequality is somehow bad. But as any Economist would tell you, it's not that simple. Look at the British Industrial Revolution for example. Economic Inequality was absolutely increasing but Economic Growth did not solely impact those at the top. The Promethean Growth that occurred at that time period impacted people of all classes. New inventions and increased productivity benefited society from top to bottom. Much like the Tech Boom that is occurring today. Yes, Economic Inequality is increasing. But, so is our Standard of Living. Both need to be talked about in order to paint a fair picture of a complex problem like Economic Inequality. Especially, in order to accurately assess whether Economic Growth is affecting the bottom 90% for the better. Is it disproportionate, yes. But is it bad? Well, that's a different story. But a resounding yes is not the answer.

Given yhe the extraordinary fact that US life expectancy is decreasing it's pretty clear that the quality of life of a large fraction of the population are not improving.

Don’t panic. Mr. Rose’s studies show everything’s under control.

In 1960, Americans lived five-years longer than citizens of Cuba. Now, thanks to the marvels of the economic freedom index (U.S. scores 75%, Cuba 32%), life expectancy in the U.S…oops…just fell behind Cuba.

It seems to be widely regarded that the skills premium hypothesis (that was already circulating 40+ years ago) as a driving force behind greater wage equality must be abandoned. The reasons normally cited are mainly the disparity at the upper tip of the distribution and the failure of much broader college graduation rates to mitigate inequality growth. But with data updates like this, I think the skills premium hypothesis is looking better and better. There is really no reason to think that a college degree will magically bestow all possessors with economically useful skills and it is also quite reasonable that the greatest benefit will flow to the most extremely skilled individuals. Notable income growth above average at both the 10% and 10% percentile level is perfectly compatible with a skills premium story for me. And what we know about disparity in income across firms related to firms' quality and performance seems to support this as well, providing lots of anecdotal evidence. It is not enough to just be a programmer - if you want to make really good money you need to be a Google-class developer.

Mpowell, nice comment. To address inequality, go up the income ladder another billion (literally). Inequality is not about who’s on the street and who’s on top of the curb, it’s about the guys up in the clouds.

From wikipedia:

"By 2012, about 50 million Americans were food insecure, approximately 1 in 6 of the population, with the proportion of children facing food insecurity even higher at about 1 in 4.[1]

Hunger has increasingly begun to sometimes affect even middle class Americans. According to a 2012 study by UCLA Center for Health Policy Research, even married couples who both work but have low incomes will sometimes now require emergency food assistance"

You can talk about inequality as a theoretical discussion. Or you can look at what it means day to day. With so many rich people in America, why are so many so poor and hungry? What kind of country are we really.....

Comments for this post are closed