Alinsky-Style Statistical Tricks

[Marshall]

Barack and Joe: Social Justice

By Michael Barry


In 'splaining things to Joe the plumber, then-candidate Obama leaned hard on the unfairness of our current political-economic system:

We haven't given a break to folks who make less [the $250,000] and, as a consequence, the average wage and income for just ordinary folks, the vast majority of Americans, has actually gone down over the last eight years.

Boy -- is that a commonplace.
Before we get to the data, I have to say this proposition, that things have gotten worse (this is in 2008, remember) for the "folks," to me is intuitively false. Is that bias? I don't think so; I may be wrong, but my conclusion is based on simple, admittedly totally anecdotal, experience. I look around. I see some people who get lucky -- they get a bunch of stock options and their company takes off. I see some people who exploit an opportunity to loot public funds -- I think of Franklin Raines. Guys on Wall Street make fortunes. Is any of this new? Most people I see, rich or just upper-middle-class, went to school, worked hard. Steve Jobs, Warren Buffet, Lloyd Blankfein -- yes, America, there are some really rich people out there. But these guys all started in the middle and worked for what they got. And they had talent and, probably, some luck. But I think vision and a willingness to pursue it had a lot more to do with it than luck.
In all "advanced" countries, income (and wealth) is a bell curve, with some very rich people at one end and poor people at the other and a lot of people in the middle.
More importantly, there are very obvious correlations between income on the one hand and education (and, particularly, certain kinds of education) and age (AKA experience). Ours is not, generally, an unfair or dishonest economy -- it could not produce wealth if it were. Hard work + utility = wealth, in America.
To the data
Let's begin by defining rich. As I understand it, this is routinely taken to be (more or less) the top 1 percent in income. In 2005 the top 1% was $348,000. Let's begin by noting that this definition (1) is biased against people who earn their money rather than inherit it. If you just happen to own a lot of stuff, say because your grandfather gave it to you, you may not even show up in the top 1%. This is an index of wealth production, not wealth consumption. And the definition (2) makes no distinctions within this top 1% group. I hate to tell all the other 99%, but for some, say living in San Francisco or New York with a couple of kids, $348,000 is nice but it isn't rich.
Here's the New York Times (from 2007) on income inequality:

Income inequality grew significantly in 2005, with the top 1 percent of Americans -- those with incomes that year of more than $348,000 -- receiving their largest share of national income since 1928, analysis of newly released tax data shows." And, check this out: "The new data also shows that the top 300,000 Americans collectively enjoyed almost as much income as the bottom 150 million Americans.

What a juicy factlet.
The NY Times article is based on research being done by two (economics) professors -- Emmanuel Saez of the University of California, Berkeley and Thomas Piketty of the Paris School of Economics. And it does indeed show, based on income tax return data, a marked trend towards income inequality, producing the statistics the Times quoted. Interpretively, this "trend" is widely understood to be a bad thing: according to Prof. Saez, "[i]f the economy is growing but only a few are enjoying the benefits, it goes to our sense of fairness. It can have important political consequences."
Sounding a theme that we hear a lot, Robert Greenstein, executive director of the (left-wing) think-tank the Center on Budget and Policy Priorities, says: "The nation faces some very tough choices in coming years. That such a large share of the income gains are going to the very top, at a minimum, raises serious questions about continuing to provide tax cuts averaging over $150,000 a year to people making more than a million dollars a year, while saying we do not have enough money" to provide health insurance to 47 million Americans and cutting education benefits.
Not so fast
At the risk of being accused of confirmation bias, I read the article by Alan Reynolds of Cato reviewing Saez's and Piketty's work. Reynolds notes the following:
(1) Use of tax returns to gauge income has been criticized, because tax policy induces behavior, particularly with respect to the reporting of income, that distorts the data. As a general matter, any reduction in top rates will increase reported income, as the penalty (taxation) for reporting income goes down. There are of course perfectly legal ways to avoid reporting income. Just ask GE.
Reynolds notes, for instance, that Piketty and Saez themselves explain (in a 2001 paper) that "a significant part of the gain [in top income shares] is concentrated in two years, 1987 and 1988, just after the Tax Reform Act of 1986." The '86 Act, which dramatically reduced taxes on income (in return for eliminating deductions), obviously encouraged the reporting of income that was previously buried in tax shelters.
Reynolds provides a long list of tax code changes that increase reported income. But the most significant one was the move by smaller businesses away from "Sub-Chapter C" income, which is not reported on personal returns, to "Sub-Chapter S" income, which is. When you account for this change, a significant portion of the inequality trend (according to Reynolds, over half) vanishes.
There is also a problem that Thomas Sowell has commented upon. The Saez-Piketty numbers are based on not individuals, but tax reporting units. It's really difficult to unpack this to make a reasonably transparent oranges-to-oranges comparison. Consider: a single individual making $348K = one rich unit; so does a couple filing jointly. Are these units equivalently rich?
Sowell has been an emphatic critic of data based on "household" income because family size has gone down over time. So while family income may have gone down or not increased much, much of the "trend" can be accounted for demographically rather than as a matter of social justice.
(2) The other real problem with all this data is that it ignores transfer payments -- all the non-taxable income that, in our Great Society, the "rich" must pay to the "poor" (and sometimes the not-so-poor). When you add in the effect of transfer payments, the so-called trend vanishes: from 1988 to the present, the percentage of national income of the top 1%, adjusted for shifted business income and for transfer payments, has held steady in the high 8's/low 9's.
Personally, I find Reynolds arguments persuasive. But the counterfactual position, that the rich are getting rich, informs most of our public discourse. Including Obama's exchange with Joe. And it provides the pretext for...taking Joe's money and giving it to...the waitress Obama was just talking to. I wonder whom that waitress is going to vote for?
"Fairness"
Any parent knows what baggy notions fair and unfair are. Can we not discount a huge piece of these social justice complaints as simply narcissism: an heroic ability to see things from one's own point of view? To repeat: in this country, work hard at something useful, learn, try to understand what other people want and are willing to pay for, strive -- do all those things and you are going to do pretty well. You may not be in the top 1%. Most of us aren't (duh). But you'll be fine.