JONATHAN CHAIT JULY 6, 2010
-
Read Later
READ LATERAvailable only to subscribers. SUBSCRIBE TODAY
-
Listen
ARTICLE AUDIO
- Font Size
Jonathan Cohn, a.k.a Citizen Cohn, a.k.a. Mahmoud Abdul-Rauf, has a reply to our old American Prospect editor Bob Kuttner's Huffington Post column taking the Obama administration to ask for its moderate ways. I'd like to make a reply to one particular part of Kuttner's argument, where he asserts that Democratic electoral losses in November will be due to Obama's lack of ideological resolve rather than to structural factors:
Come November, as Republicans break out champagne, the usual commentators will offer the usual alibis and silver linings.
The party of the newly elected president always loses Congressional seats. Not always: viz. Roosevelt, 1934, or Bush II, 2002. The two men shared nothing, except resolve in a crisis. That should tell you something. Where's Obama's resolve?
This is not persuasive. Here's a chart showing the relationship between economic conditions and midterm election results:
First, note that almost every president loses seats in midterm elections. The state of the economy influences the magnitude of the losses, though other conditions can exert significant influence. But the general trend is that midterm elections are bad for the president's party, and slow income growth is even worse. Ronald Reagan had a lot of "resolve," but he still lost a lot of seats in 1982.
Kuttner cites two notable exceptions to the pattern of the president's party losing midterm election seats. The first is 2002. I think it's pretty clear that the 9/11 attacks had an unusually powerful role here. The second is 1934. That's not on the above chart, which begins in 1950. Is that another exception? Actually, no. Personal income grew an astronomical 12.7% in 1934.
So we're down to one exception to the rule: 2002. Locating a single exception to a well-established trend is not a good reason to ignore the trend.
9 comments
Ah, the Wehner fallacy makes an appearance on the left, in the person of Robert Kuttner.
- liberal reformer
July 6, 2010 at 4:57pm
"A well-established trend" from a plot that's appears to be very nearly random if one ignores the line drawn across it? Changes in average income certainly affect elections, but the data shows that the effect is not very strong, and that other factors are much more important. Consider the difference between 2002 and 2006 - from a gain to a 30-some seat loss for the Republicans. The difference sure wasn't the economy. Likewise 1954 - 1958, and 1962- 1966, similar economic conditions, very different results. While I agree that "resolution" may not explain much, you can't predict at all accurately based on changes in income.
- K_Wilson
July 6, 2010 at 5:03pm
Here's the standard political science explanation for why the president's party usually loses seats in midterm elections. Presidential races bring out many more voters than midterm elections. The voters who show up for the presidential elections but decline to vote in midterms are the sort of voters who tend to have weak attachment to a political party and are relatively ill-informed. They are more strongly influenced by topical issues that favor one presidential candidate and one party over another in any given race. So they are inclined to vote a straight ticket. This gives a kind of artificial boost to candidates of the president's party, leading that party to win a slew of narrow victories in districts that usually vote for the other party. But these voters are unlikely to show up in the midterm elections two years later and thus the district is likely to return to its "natural" partisan alignment. The bigger the coattail effect, the larger the number of vulnerable seats in the midterms. Obama had pretty big coattails, thus the Democrats will have a huge number of vulnerable seats to defend. Combine that with the worst economy in 80 years, and it looks like lights out for the Dems. Note that the biggest outliers in Jonathan's chart are Clinton 98 and Bush 02 -- both followed presidential elections in which the candidates had very weak coattails. Actually, Bush had no coattails at all in 2000 since he lost the popular vote.
- gorton
July 6, 2010 at 7:53pm
Doesn't anyone on the TNR staff read what is published here? JC, please take a gander here and them make your comments: http://www.tnr.com/article/politics/75965/hey-senator
- tnmats
July 6, 2010 at 9:34pm
This is an intellectually lazy and basically dishonest post by Chait and by the first three commenters. Did any of the above read the Kuttner post? One sentence out of context from the Kuttner post is worthy of Cheney-- not Chait. The point of the Kuttner article is an extensive description of so many missed opportunities by Obama and the Dems. The post by Chait doesn't refute the post by Kuttner.
- drofnats1
July 6, 2010 at 10:26pm
The point about 02 ought to apply as well to the least squares trend-line that's been plotted. It's an outlier of epic proportions in direct relation to the epic impact on the citizenry of that attack and should be removed, or at least down-weighted. Bearing in mind, Bush's popularity ratings went from abysmal for so early in a Presidency to well over 90% after the attacks. So, probably pays to remove it quantitatively as well as qualitatively- eyeballing it here, my guess is the slope change will be radical (and illustrate, e.g., Regan's outperformance vis a vis baseline, which intuitively makes more sense to me given how he swept into office, and success at blaming problems on the Carter administration ad nauseum). There is plenty of scope for criticizing Obama drofnats, and I don't think this post is meant as a response to those criticisms. It is meant instead to refute the single evidentiary claim that Democrat's and Obama's current disfavor with the electorate is a function of those failings. At least primarily, the answer is clearly no (unless one makes the IMO meritless argument that the state of the economy is a function of those failings).
- I Majorajam
July 7, 2010 at 9:42am
- Brilliant! Find the one characteristic FDR and Bush 43 shared and it will explain why they were the only presidents in the modern era who didn't lose seats in the mid-term of their first term. "The two men shared nothing, except resolve in a crisis." Hey, that was easy. The premise is sound, it must be a character trait rather than the diverse dynamics in any particular era. And we don't need to consider what 1934 and 2002 did not share. I did not consider FDR and Bush 43 were character clones regarding that defining feature, the key to their success and our destiny was that rare flash of resolve. Like Halley's Comet but probably with Rove pulling the strings. Resolve must be damn powerful. It was as if we had the same persons in power during the same year because their connective trait was their resolve. Imagine, but for the Twenty-second Amendment Bush was probably on his way to four terms.
- michaelg
July 7, 2010 at 10:28am
drofnats1 has missed the point. The data Mr. Chait has used to support the idea that the economy is a major, if not the major influence on the outcome of midterm elections does not show that at all. The scatter is so great that the change in average disposable income is essentially useless for predicting the election result. Now, it may very well be true that the state of the economy is a major factor; I expect it's rather more important than "resolve", but this particular evidence doesn't support that conclusion. Beware magical thinking about the power of presidential will and the Mandate of Heaven.
- K_Wilson
July 7, 2010 at 10:31am
- Hey, take it easy on the Mandate of Heaven. That's the name of my new party if it appears Sarah has a chance. I have a box of oracle bones, ready to roll.
- michaelg
July 7, 2010 at 11:48am