TaxProf Blog

Editor: Paul L. Caron
Pepperdine University School of Law

A Member of the Law Professor Blogs Network

Wednesday, February 20, 2013

More on the New Emmanuel Saez Income Inequality Data

Following up on Monday's post on the new income inequality data from Emmanuel Saez (UC-Berkeley): 

Chart 11

What should we make of these numbers? One approach is to stress the increase in wealth inequality, deploring the gains of the top 1% while lamenting the decline in the income of the remainder of the population. But this approach is only half right. We should be uneasy about any and all income declines, period. But, by the same token, we should collectively be pleased by increases in income at the top, so long as they were not caused by taking, whether through taxation or regulation, from individuals at the bottom.

This conclusion rests on the notion of a Pareto improvement, which favors any changes in overall utility or wealth that make at least one person better off without making anyone else worse off. By that measure, there would be an unambiguous social improvement if the income of the wealthy went up by 100% so long as the income of those at the bottom end did not, as a consequence, go down. That same measure would, of course, applaud gains in the income of the 99% so long as the income of the top 1% did not fall either.

This line of thought is quite alien to thinkers like Saez, who view the excessive concentration of income as a harm even if it results from a Pareto improvement. Any center for “equitable growth” has to pay as much attention to the first constraint as it does to the second. Under Saez’s view of equity, it is better to narrow the gap between the top and the bottom than to increase the overall wealth. ...

Rather than focus on “equitable growth,” the President should focus on flattening the income tax and deregulating labor markets. Today’s constant emphasis on progressive taxation and government intervention in labor markets will continue to lead the country, especially the middle class, on a downward path.

(Hat Tip: Bill Turnier.)

Tax | Permalink

TrackBack URL for this entry:

Listed below are links to weblogs that reference More on the New Emmanuel Saez Income Inequality Data:


" But, by the same token, we should collectively be pleased by increases in income at the top, so long as they were not caused by taking, whether through taxation or regulation, from individuals at the bottom."

Except that money buys political and economic power, and this inequality means that a small fraction of the population can buy and sell and abuse everyone else.

And of course, wages for most others have grown much more slowly (or not at all), so policy changes have redistributed income up relative to the situation 40 years ago.

This was the Reagan era--make taxes less progressive, gut labor unions, deregulate everything. So clearly making taxes more progressive (higher on the very rich, lower on everyone else) and re-regulating the labor market (protecting most people's most important source of income, their wages) would help the middle class.

Money buys power, buys more money, buys more power.

Of course, Richard Epstein is quite familiar with the ways in which those who advocate for the interests of the Rich and powerful are financially rewarded by them.

Posted by: Anon | Feb 20, 2013 4:11:41 PM

Does Anon have support for his/her comment that "re-regulating the labor market" would help the middle class? It will likely help some, but will it hurt more than help? For example, if you make terminating employment extremely difficult, then employers will hesitate before hiring. So the unemployed suffer. And wouldn't removing the fear of termination tend to undermine productivity? Also, the government's exercise of power has unintended consequences. Usually, the greater the power, the worse the unintended consequences. Of course, sometimes the consequences are obvious but the government doesn't care: note the EEOC recent attempt to eliminate discrimination against hiring felons.

Posted by: daniel | Feb 20, 2013 5:41:24 PM

Leaving aside the misleading truncation of the vertical axis, the line starts moving just after the enactment of the Subchapter S Revision Act of 1982, then it really takes off with the 1986 Tax Reform that reduced tax rates to the point where it became a no-brainer to report as much business income as possible on personal tax returns. The income of multimillion dollar businesses moved from Form 1120 (not included on the above chart) onto Form 1040 (included on the chart). Furthermore, the 1986 tax reform outlawed longstanding tax sheltering techniques, in effect legislating an increase in reported taxable income among former participants in tax shelters.

In other words, the pre-1982 and the post-1986 numbers are drawn from different tax worlds. They are not comparable without making the adjustment for the movement of business income onto personal returns.

The Saez data has a chart, number 4B, that breaks out business income. It shows exactly the jump I described. There is also a jump in realized capital gains which could be attributed to increased realization at lower tax rates as well as to heavier use of stock options after the 1986 tax reform.

Tax reform was designed to induce more realization and more reporting of taxable income at the high end. It accomplished this objective, bringing more money into the Treasury. Saez thinks that's bad. I don't. I call it a success.

If minimizing inequality in reported taxable income is paramount there is one sure-fire solution: Tax all income at 100%. Everyone will report zero. Problem "solved".

It's intellectually dishonest for people who definitely know better to silently ignore the elasticity of taxable income. Everyone knows that this elasticity is very high for high-income individuals. They hire tax planners to make sure they pay the least possible. The genius of the 1986 tax reform was that it made those schemes pointless or illegal or both. Thus the jump in "income".

Posted by: AMTbuff | Feb 20, 2013 6:21:51 PM


On the question of whether minimum wage laws cause unemployment, the debate among serious economists is now between the answers "no" and "probably by much less than they help boost total earnings at the bottom."

Read Card & Krueger.

Posted by: Anon | Feb 20, 2013 8:18:34 PM

And on the question of whether maximum hours laws and mandatory minimum annual vacation times undermine productivity per hour, the answer is once again almost certainly not.

Posted by: Anon | Feb 20, 2013 8:20:36 PM

Epstein is an iconoclast, but he has the courage to think in an original fashion, unlike Saez who basically says the same thing in a slightly different way each time

Posted by: michael livingston | Feb 21, 2013 1:57:38 AM

Michael Livingston,

Epstein is an ideologue, not an iconoclast. There's a difference.

What he will say is completely predictable. Just read Ayn Rand, or go to a Republican Party fundraiser on the upper east side of Manhattan, and you've pretty much got it down. Granted, Epstein is far more intelligent and personable than Ayn Rand, and as libertarian ideologues go, he's top shelf quality. He's also very direct and doesn't mince words. So if you need an ideologue, I highly recommend him.

But who needs ideologues?

Saez is an empirical economist. He measures what is going on in the world and reports his findings.

The reason he always says that inequality is getting more extreme is because that's what the data unambiguously shows.

Being "creative" with the data isn't something to be praised--it's called fraud.

AMTBuff is wrong. He's repeating a claim made by Reynolds, an academic failure who found refuge at the Cato institute, and whose claims have been thoroughly debunked by Saez.

The existing tax data likely understates income inequality because most of the unreported income (i.e., unrealized capital gains, deferred comp, etc.) is at the top of the income distribution, while most of the reported income (wages) is at the bottom. Very few "middle class" people are maxing out their 401(K) contributions or have saved anywhere near enough to comfortably retire.

It's true that the U.S. has both seen income inequality increase and taxes become less progressive, but the increase in income inequality isn't an artifact or misreporting created by changes in tax rates. There have been huge policy changes over the last 30 years in the U.S.--evisceration of labor unions, allowing minimum wages to fall in real terms, reducing regulation and worker protections, free trade agreements, reduced funding for anti-poverty programs and infrastructure, relaxing anti-trust laws and allowing mergers to monopoly--which have increased inequality.

Basically, tax cuts for the rich are part of a right wing policy toolkit that inevitably increases inequality.

Posted by: Anon | Feb 21, 2013 6:14:34 AM

Study after study shows labor's share of income falling and capital's share of income rising, and it's true across data sources.

The authorities on income inequality are Thomas Piketty and Emannuel Saez (a Berkeley PhD Economist). The OECD has also done excellent studies.

ATM Buff / the Cato institute seems to be saying, ignore the fact that the top 1% of earners own businesses and get their income from capital, and the inequality is then much less.

Yes, if you ignore the fact that the rich are very rich and make money from their money, then you've successfully ignored just how rich they are.

Inequality is not an artifact of tax reporting. It shows up in studies of life expectancy, in studies of political funding, in the growth of luxury consumption, in studies of occupational wages, in the share of income going to capital vs. labor--basically in every data source you can imagine, as long as it isn't too aggressively top coded and has fine grained data at the top.

It also shows up in the Survey of Consumer Finances by the Federal Reserve.

What is particularly galling is the way folks like the Cato Institute take money from the rich in order to deny that the rich have the money and power to influence policy.

Posted by: Anon | Feb 21, 2013 6:24:45 AM

Here's another Alan Reynolds (Cato Institute) classic in denial:

"No Housing Bubble Trouble", Washington Times, Jan. 2005

Yes, there wasn't a housing bubble in the mid 2000s. And yes, inequality hasn't grown since the 1970s.

And yes, I have a bridge I would like to sell you.

Posted by: Anon | Feb 21, 2013 6:42:13 AM

Anon, you missed my point. It's not that rich people have business income, it's that their business income pre-1986 IS NOT INCLUDED in the chart at the top of this post. Income reported on corporate returns is not included in the Saez data. A whole lot of that income moved onto personal returns post-1982 with Subchapter S revisions and in post-1986 due to Tax Reform.

The most salient feature of the graph, the 1980's jump, is therefore an artifact. Here's a chart that makes my point:

I read the Saez response. It evades this issue by digressing into a discussion of capital gains. The chart I linked speaks for itself and demolishes the Saez evasion.

Saez is very smart. He could probably make his point effectively without fudging his methodology. I can't understand why he doesn't just play it straight. But then I also can't understand why superb athletes like Lance Armstrong or Barry Bonds would decide to break the rules rather than accept the limits of their natural talents.

Posted by: AMTbuff | Feb 21, 2013 10:41:49 AM

Which would at most explain a one time jump in 1982 and 1986, but none of the subsequent increase in inequality.

Posted by: Anon | Feb 21, 2013 11:10:27 AM

But you are not accurately describing Saez's methodology or his response.

Saez explains that if business income wasn't showing up in individual returns pre-1982 / 1986, it should have showed up as capital gains, so the change in tax treatment wouldn't actually distort overall income, because income has to show up either as business income or as capital gains.

And Saez did another check where he excluded business income and capital gains, which confirmed his original results.

From Saez's response:

"Indeed, our series focusing exclusively on W2 wage and salary income (therefore excluding both business income and capital income), show that the top 1% wage income share has increased from 6.4% in 1980 to 11.6% in 2004, no doubt a very large increase as well. "

Emmanual Saez has won the 2009 John Bates Clark Medal--the pre-Nobel Prize in Economics--and is an endowed chair at Berkeley. His research has survived the most rigorous of peer review. He's been cited more than 8,000 times.

Alan Reynolds dropped out of night school at Cal State Sacramento after 3 years, apparently with no advanced degree, and works at the propaganda factory know as the Cato Institute.

As far as I can tell, Reynolds hasn't published anything in a real peer reviewed economics journal (and no, the Cato Institute Journal doesn't count!).

Who do you think is more likely to make a dumb mistake or put a thumb on the scale to try to distort the data?

The academic failure completely dependent on donations from high income individuals who want their tax rates to stay low?

Or the super star economist with complete financial independence and unimpeachable integrity?

Posted by: Anon | Feb 21, 2013 11:26:44 AM

"Indeed, our series focusing exclusively on W2 wage and salary income (therefore excluding both business income and capital income), show that the top 1% wage income share has increased from 6.4% in 1980 to 11.6% in 2004, no doubt a very large increase as well."

The chart 4B that I mentioned shows wage data as well as all the other categories. Among the dozen or so charts there is not one that focuses only on wage income. In effect, the wage data is buried so that casual readers such as reporters or bloggers will not notice it.

As Saez notes in your quote, the wage data is reasonably strong by itself. So why focus on suspect methodology (comparing business income across a tax policy discontinuity) when correct methodology would suffice? That's my Barry Bonds point.

Just once I'd like to see somebody present real data without any slant and let the reader reach his own conclusions. The problem is that everyone who studies "income inequality" has an agenda. I wish there were one investigator who did not, but if you didn't have an agenda why would study something so dry?

Posted by: AMTbuff | Feb 21, 2013 3:12:36 PM

Some people think income inequality is fascinating.

Economists and statisticians spend their lives studying distributions.

If you're going to study distributions, what could possibly be more interesting than the distribution of money?

I'd say Saez and Piketty come as close to a neutral, unbiased, scientific look at inequality as we will ever see. Them, and on a good day, the OECD.

What agenda could they possibly have?

Inequality is in their personal best interest. They happen to be quite rich.

According to public records, Saez made $370K in 2011 from U.C. Berkeley alone (i.e., ignoring investment income or side consulting), which puts him comfortably in the top 1%.

Posted by: Anon | Feb 22, 2013 7:04:00 AM

And the reason not to focus on wage income is because it underestimates inequality.

Capital gains and business income are less equally distributed than wages. The rich make money off their money; the poor depend on their labor.

So why try to hide just how unequal income is by focusing only on wages?

That is, unless you're a Cato Institute hack paid to hide the truth about inequality?

Posted by: Anon | Feb 22, 2013 7:09:30 AM

Wealth is where the real inequality is. Data on wealth is scant so people study income. Data on true economic income is scant so people study reported taxable income. Adjusting for tax policy discontinuities is hard so people don't do it.

It's like the old joke about looking for your lost wallet down the street from where you dropped it because the light's better there.

In my opinion there are two kinds of success: success earned by delivering value to customers and success earned through political influence. Gates, Buffett, and Jobs are examples of the former. The latter include thousands upon thousands of businessmen who make inside deals with politicians and the politicians who magically manage to get rich making those deals. Real estate deals at the local level are corrupt because nobody is minding the store. At the federal level we have Wall Street executives that make trillions of dollars of "heads I win, tails taxpayers lose" bets sanctioned by government.
The public idolizes rich people who earned their wealth. The public detests people who got rich through political influence.

Gaining wealth primarily through political influence is the hallmark of poorly run countries all over the world, from Argentina to Zimbabwe. The people of those countries don't like it either.

What proportion of America's 1% got there through political influence? I have no idea, but it's not small. That's an inequality study I'd love to see. A steady increase in that percentage would be a true indicator of trouble ahead.

Posted by: AMTbuff | Feb 23, 2013 8:09:53 AM

Furthermore, if the proposed solution to wealth inequality is more government intervention, how will that reduce the importance of political influence in gaining wealth? Isn't such a solution likely to make the problem of politically acquired wealth even worse?

Posted by: AMTbuff | Feb 23, 2013 9:48:01 AM

There are illegitimate ways to gain wealth that don't involve public corruption or government influence--like externalizing your costs through pollution (think the Koch Brothers, Exxon, the automotive industry, most manufacturers); defrauding your business partners and customers (much of Wall Street before the financial crisis); exploiting information asymmetries to charge for unnecessary services (mechanics; plumbers; doctors); bribing individuals in positions of power at other corporations (i.e., Wall Street working the ratings agencies); privatizing public assets at fire sale prices (what Wall Street is attempting to do now to the GSEs; what oil companies are trying to do with leases on public lands).

Most of the illegitimate ways to become rich depend on minimal regulation and minimal government regulation, or shrinking the government through privatization

Which is why the most unequal countries have the smallest public sectors, and why expanding the public sector and regulation is the solution to illigitimate wealth creation and corruption and inequality.

Posted by: Anon | Mar 3, 2013 2:46:30 PM

As for Warren Buffett and Bill Gates and Steve Jobs, they're more of a mixed bag than is usually presented in the media.

Warren Buffett is the son of a prominent stock broker and a member of the U.S. Congress. Buffett's early career, and his ability to initially attract investors, largely depended on his father's connections.

His subsequent career depended entirely on connecting investors with companies--he rarely changed the management team or changed corporate policies at the companies he bought.

Basically, all he does is Vet companies and reassure investors that they're going to get good value. He's in the business of connecting people.

Is having connections to investors really so different from having connections in government?

Bill Gates is a similar story. His dad was a law firm partner at a predecessor of K&L Gates, and a very wealthy and well connected guy. Gates was able to get investors early on because of his father's connections.

And his career was again built on connecting investors, with programmers, with customers (initially IBM, not consumers). Gates didn't actually program any of the software that made him rich. He's just a middleman, trading on his connections in different parts of the business world.

Is that so different from having connections in government?

Steve Jobs wasn't born rich, but he was born with a remarkable gift as a salesperson. When it came to early Apple Wozniak did all of the work (coding software; building hardware), and Jobs was just the slick frontman. Again, Jobs just connected programming talent with investors and customers.

He didn't actually do most of the work.

I'm not sure why connections to business people are a legitimate source of wealth, but connections to government officials is illegitimate.

It seems like this depends on some kind of visceral, anti-government religion, not any realistic assessment of the fairly subtle differences between corporate and public bureaucracies.

Posted by: Anon | Mar 3, 2013 3:08:03 PM