“Income can’t be used to predict political opinion”

by Andrew Gelman on September 30, 2011 · 17 comments

in Media,Political Economy

What really irritates me about this column (by John Steele Gordon) is not how stupid it is (an article about “millionaires” that switches within the very same paragraph between “a nest egg of $1 million” and “a $1 million annual income” without acknowledging the difference between these concepts) or the ignorance it displays (no, it’s not true that “McCain carried the middle class” in 2008—-unless by “middle class” you mean “middle class whites”).

No, what really ticks me off is that, when the Red State Blue State book was coming out, we pitched a “5 myths” article for the Washington Post, and they turned us down! Perhaps the rule is: if it’s in the Opinions section of the paper, it can’t contain any facts? Or, to be more precise, any facts it contains must be counterbalanced by an equal number of inanities?

Grrrrr . . . I haven’t been so annoyed since reading that New York Times article that argued that electoral politics is just like high school. Who needs political science or economics when you can resolve all confusion with an appropriate Simpsons reference?

P.S. I’m not opposed to someone arguing for upper-income tax breaks. But couldn’t the Post have found a conservative who could make the case in an intelligent way?

P.P.S. A commenter caught this cute bit that I hadn’t noticed:

Gordon writes, “Today, $1 million in the bank generates only about $50,000 per year in interest.”

Where is this U.S. bank that’s paying 5% on deposits?

Perhaps that will be Gordon’s next column: “Five myths about banking.” Myth #1 will be that banks are paying 5% interest on deposits.


dr September 30, 2011 at 10:09 am

Oh, that article is terrible! I had to read most of the paragraphs twice to figure out which side of the debate the author was on. Take the line your commenter brought up about the money paid in interest on a million dollars. What’s even more remarkable than the rate of return is the way that claim is used. The argument being made is that because a million dollars only earns $50k per year in interest, and the median US household income is about $50K, therefore millionaires aren’t rich. WHAT?

Thomas September 30, 2011 at 10:59 am

According to the NY Times exit polling, McCain carried the $50,000 to $74,999 demographic. Isn’t that the middle class?

Andrew Gelman September 30, 2011 at 11:12 am


In the “middle class” category, the exit polls give the following:
$30-50K Obama 56 – McCain 43
$50-75K McCain 49 – Obama 48
$75-100K Obama 51 – McCain 48
$100-150K: Obama 48 – McCain 50

I’d call that Obama and McCain splitting the middle-class vote, not McCain carrying it.

matt w September 30, 2011 at 11:36 am

And surely the exit polls aren’t so precise that we can say with much confidence at all that McCain actually won more votes in the $50-75K bracket anyway. (Not to mention that, if the median income is $50K, the $30-50K bracket has as much right to be called the middle class.)

Thomas September 30, 2011 at 1:15 pm

I have a larger family than most, so under the federal poverty guidelines, my family would be in poverty if we had an income at the bottom of the range you show. Its hard for me to understand that as obviously “middle class”.

Matt W–Median family income is north of $60,000, and the exit polling refers to family income.

Now we’re arguing about the definition of middle class. I thought it would be obvious that Gordon had this wrong.

Andrew Gelman September 30, 2011 at 1:28 pm


Look at the numbers above. Obama wins in $30-50K and $75-100K. McCain wins in $50-75K and $100-150K (actually, I think I meant to type $100-200K there). All the margins above $50K are close. I can’t see how anyone can take those numbers and say that McCain won the middle class vote. “They split the middle class vote” is more like it. So, yes, I think Gordon had it obviously wrong on this one.

Thomas September 30, 2011 at 5:11 pm

Andrew, that’s just a renewed expression of your view that the middle class extends from a family income of $30,000 to $150,000 (or $200,000). The $50,000-$75,000 income band is the closest to the middle quintile distribution of any of the bands, and for that reason alone a choice to call that the middle class would seem to be to be definitionally appropriate. There’ s certainly nothing obviously wrong about it. Your choice, in which roughly 75% of families are deemed middle class, isn’t obviously wrong either, but I would say that defining the term that capaciously may make it useless (except perhaps as a political statement).

matt w September 30, 2011 at 2:09 pm

According to the U.S. Census, “Real median household income fell between the 2008 and 2009 ACS–decreasing by 2.9 percent from $51,726 to $50,221.” Amounts seem to be inflation adjusted to 2009 dollars (according to footnote 2).

Tyson October 4, 2011 at 4:54 pm

Gordon himself defines what he means by middle class:
In 2008, for example, Obama won the votes of 60 percent of those with a family income under $50,000 and 52 percent of those earning more than than $200,000. McCain carried the middle class.

He means between $50,000 and $200,000. Since the median income, according to Gordon, is $50,000, I guess he thinks half of the country is lower class, or else he made $50k the cut-off so he could say McCain “carried” the middle class, but he didn’t carry it by much, based on Gelman’s numbers above.

Someone at the Heritage Foundation says middle class is the middle 60%, which is $25,000-$100,000. By that measure (which seems more reasonable to me, but what do I know), it’d be more accurate to say that Obama “carried” the middle class than the reverse. Split the middle class is probably the least debatable.

By the way, the Post put up a correction that $1 million in the bank would actually earn only $5,000, so really those millionaires are in the lower class, since they would need to earn $45,000 more year to be middle class. Oh wait, that sounds silly. So what the author really meant to say was a combination of investment returns and interest earns 5%.

Tyson October 4, 2011 at 4:55 pm
GoRox September 30, 2011 at 2:03 pm

Median family income is north of $60,000???!! When did everyone get raises?

Thomas September 30, 2011 at 5:18 pm

Median family income is not the same thing as median household income. It isn’t clear to me which measure the poll is meant to relate to.

matt w September 30, 2011 at 8:45 pm

Median family income is the median income of all households with more than one person in them. Unless you think that the exit polls discarded people who live by themselves, it has to be household income, doesn’t it?

Dennis September 30, 2011 at 2:14 pm

Couldn’t agree more, especially since his first point seems to be that inflation exists.

Tobin Grant September 30, 2011 at 3:21 pm

Amen. Amen. Amen.

Preach it.

andrew September 30, 2011 at 5:26 pm

Who needs political science or economics when you can resolve all confusion with an appropriate Simpsons reference?

People can come up with Simpsons references for everything. Forty percent of all people know that.

Tyson October 4, 2011 at 5:30 pm

Here’s another funny one – Myth #5 is that a millionaire’s tax won’t hurt investment. Maybe he is right that a millionaire’s tax will hurt investment a little, I don’t know. But for his proof, he cites a 2001 Congressional study http://www.house.gov/jec/tax/taxrates/taxrates.pdf
This Congressional Study was making the case that cutting the top tax rate from 36% to 33% would increase tax revenues from that group (dynamic revenue change). Did it? I looked at a few tables on tax revenues by income level http://www.cbo.gov/publications/collections/taxdistribution.cfm
and it looks to me like the Bush tax cuts for the rich made tax revenues from that group drop, by a lot, not increase.

So evidence to dispel the Myth that taxing millionaires won’t hurt investment is a (Republican-written) study that is guilty of the Myth that cutting taxes on millionaires will increase tax revenues.

Comments on this entry are closed.

Previous post:

Next post: