Saturday, August 15, 2009
Saez: New Income Concentration Data
Emmanuel Saez (UC-Berkeley, Department of Economics) has updated his income concentration data to reflect the latest statistics from the IRS's Statistics of Income Division on August 3, 2009: Striking it Richer: The Evolution of Top Incomes in the United States (Update with 2007 Estimate):
From 2006 to 2007, average real income per family grew by a solid 3.7%. Average real income for the top percentile grew faster (6.8% growth), further increasing the top percentile income share from 22.8 to 23.5% (Figure 2). Year 2007 is therefore the second highest year on record since 1913 almost equalling 1928, the record year when the top percentile share reached 23.9% (Figure 2). Even within the top percentile, the gains from 2006 to 2007 are extremely concentrated. The top .01% (top 14,988 US families, making at least $11.5m in 2007) share increased from 5.46% in 2006 to 6.04% in 2007 leaving well behind the 1928 peak of 5.04% (Figure 3). This shows that 2007 was an incredibly good year for the super rich. Year 2007 was actually also quite good for the bottom 99 percent of US families as their average income grew by 2.8%. This is the best annual increase since 1998. Real income growth for the bottom 99% had been very meagre during the Bush expansion starting in 2002. Even including 2007—a good year for ordinary US families-the top percentile captured 65% of total real income growth per family from 2002 to 2007 (Table 1).
[Click on figures and tables to enlarge.]
Hat tip to Ted Seto (Loyola-L.A.), who notes that Saez's methodology "significantly understates disparities in economic income":
Saez' data is pre-tax. Here's a brief description of his methodology:
We define income as the sum of all income components reported on tax returns (wages and salaries, pensions received, profits from businesses, capital income such as dividends, interest, or rents, and realized capital gains) before individual income taxes. We exclude government transfers such as Social Security retirement benefits or unemployment compensation benefits from our income definition. Therefore, our income measure is defined as market income before individual income taxes
Note, therefore, that the data series exclude tax-exempt interest, unrecognized gain, and other tax-advantaged economic income, most of which tends to accrue predominantly to the benefit of upper-income taxpayers. As a result, the report probably significantly understates disparities in economic income.
Update: Bloomberg, Richest in U.S. Had Record Share of Income in 2007, Studies Say.
https://taxprof.typepad.com/taxprof_blog/2009/08/saez-new-income.html
Comments
The article also apparently excludes the ballooning real growth in a multitude of pre-tax employee benefits such as health insurance, IRAs, 401(k)'s, and flexible-spending plans?
Posted by: John | Aug 16, 2009 5:58:34 PM
... but we'll never know unless someone in this field has the integrity to conduct an impartial analysis of the question
And how would one get tenure doing that?
If you want to get along in academia, you have to play to their prejudices.
Posted by: Borris | Aug 16, 2009 3:20:18 AM
In the discussion of economic inequity, I recommend the book "Century of Difference" by Fischer and Hout. And for much more depth on this particular topic, go to their academic web site at http://ucdata.berkeley.edu/rsfcensus/ and check out the working papers, particularly "Differences Among Americans in Living Standards Across the Twentieth Century."
Posted by: J. Wiedwald | Aug 15, 2009 11:32:50 PM
I don't know why you can't see it. Lay an inflation adjusted chart of any broad-based stock index over the top 1% graph: OMG, correlation!
I don't understand: what is it that they put in the water at graduate school that makes you all think you can distribute income more efficiently than the market can? (Especially you lawyers!)
Posted by: Morgan | Aug 15, 2009 8:57:59 PM
"the data series exclude tax-exempt interest, unrecognized gain, and other tax-advantaged economic income, most of which tends to accrue predominantly to the benefit of upper-income taxpayers. As a result, the report probably significantly understates disparities in economic income."
Unreported income, from the underground economy or tax evasion, is concentrated on the low end.
But here's the biggest distortion:
"We exclude government transfers such as Social Security retirement benefits or unemployment compensation benefits from our income definition. Therefore, our income measure is defined as market income before individual income taxes"
Therefore? What kind of non sequitur is that? The preceding sentence in no way explains why pre-tax data should be considered more relevant than after-tax data. If you don't include government benefits, excluding tax payment compounds the error, it does not mitigate it. (Teen to parent: "I stole $100 from your wallet; therefore I wrecked the family car.) Two wrongs don't make a right in any case.
Excluding income taxes would reduce the reported disparities dramatically. I believe that researchers in this area avoid showing statistics on after-tax incomes because they have a political agenda. I give much less credence to researchers with a political agenda, because I suspect them of hiding inconvenient truths. When I catch an author concealing crucial information or making obviously incorrect assumptions in order to slant a result, I never take him seriously again.
Saez is brilliant, as anyone can see by attempting to read one of his papers. He could be so much more valuable to all of us if he'd just play it straight. I have a feeling that the facts are on his side, but we'll never know unless someone in this field has the integrity to conduct an impartial analysis of the question.
Perhaps the scrupulously balanced Tax Policy Center could take it on, if they can find anyone willing to put his name to numbers that show a less alarming story. The inequality numbers are bound to take a major turn downward anyway starting with the 2008 market crash.
Posted by: AMTbuff | Aug 15, 2009 2:27:59 PM
I looked at some of the papers at http://ucdata.berkeley.edu/rsfcensus/ and came away with yet another question:
Why do researchers appear to believe that redistribution of income would have any effect on inequality of happiness? What if, as seems much more likely to me, inequality of income is a result of differing ability to function well in society in general? Poor people don't just lack money; they lack ability to function well in multiple areas. Giving them money won't solve their non-money problems, and it might even make those problems worse. All at the cost of de-motivating people who area able to function well.
Researchers need to question the implicit assumption that redistribution will make the recipients equal in happiness to those who earned the same income through their own efforts. I don't believe it. We need to distribute living skills and productive habits much more than we need to distribute income. Teach a man to fish, and not only will he eat for life but he'll be happier than if you hand him a fish every day.
Posted by: AMTbuff | Aug 16, 2009 6:47:58 PM