Brendan Nyhan

Yesterday I complained about how few people would learn about the repeated debunking of Thomas Frank’s empirical claims. I want to make a more constructive point and recommend the two most important political science books of the year for general readers, both of which debunk claims promoted by pundits like Frank.

1. Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do. Andrew Gelman, David Park, Boris Shor, Joseph Bafumi, and Jeronimo Cortina use graphics and clever statistical analysis (hidden from the reader) to resolve the paradox of rich states voting Democratic and rich people voting Republican. You’ll never read David Brooks or Michael Barone the same way again.

2. Unequal Democracy: The Political Economy of the New Gilded Age. Larry Bartels has compiled a thorough analysis of various aspects of the politics of inequality. In particular, he undercuts the widely held perception that Republicans are better managers of the economy, showing that economic performance has been better under Democratic presidents since World War II.

Many people have complained that political science has forgotten about politics. These authors have not — their work deserves your attention.

Here’s yet another reminder of why you can’t trust the Wall Street Journal editorial page.

Earlier this month, the Journal wrote the following in an editorial blaming deficits on spending increases:

The real runaway train is what CBO calls a “substantial increase in spending” that is “on an unsustainable path.” That’s for sure.

In fact, however, the first quoted phrase comes on page ix of the report (PDF) and the second quoted phrase comes from page xii.

Here’s the first quote, which pertains to the year-to-year change from 2007 to 2008 and also blames “a halt in the growth of tax revenues”:

CBO expects the deficit to rise from 1.2 percent of GDP in 2007 to 2.9 percent this year (see Summary Table 1). The significant expansion in the deficit is the result of a substantial increase in spending and a halt in the growth of tax revenues.

And here’s the second quote three pages later, which pertains to long-term trends in entitlement and health care spending:

Over the long term, the budget remains on an unsustainable path. Unless changes are made to current policies, growing demand for resources caused by rising health care costs and the nation’s expanding elderly population will put increasing pressure on the budget…

The CBO then goes on to discuss expected growth in these costs over the next century and advocate adjustments in spending and taxes to reduce projected deficits.

The two quotes are substantially different in context and meaning. Yet the Journal jams them together without regard for accuracy. You just can’t trust them.

[By the way, the Journal’s argument in the editorial is misleading — the Center on Budget and Policy Priorities showed that 82% of the deterioration in the federal budget since 2001 that can be attributed to legislation comes from tax cuts (42%) and defense and homeland security spending (40%).]

I was amused that the “Swing State Project” blog included me in a strange diatribe against campaign expert Stuart Rothenberg:

Rothenberg goes even further:

But if the DCCC is going to go out of its way to promote certain races, it ought to be responsible for those selections.

Responsible how, exactly? Should the DCCC be tried before a jury of Stuart Rothenberg, David Broder and Brendan Nyhan on charges of slight exuberance in the service of politics? Should we be demanding that MAD Magazine devote an issue to mocking the D-Trip? Should we get out there and tar-and-feather Chris Van Hollen?

That would be quite a jury, although I’m not sure what I have to do with Rothenberg’s view of the DCCC’s targeting decisions.

Here’s another example of the bizarre relativism of “objective” news reporting — the New York Times describes John McCain’s support for private accounts in Social Security as “an approach that Democrats call privatization”:

Mr. McCain also stuck by his support for allowing workers to invest a portion of their Social Security payroll taxes in stocks and bonds, an approach that Democrats call privatization and that Mr. Obama has used to suggest Mr. McCain would subject retirees to excessive market risk.

Actually, John McCain called it “privatization” as recently as 2004 (via Think Progress):

Without privatization, I don’t see how you can possibly, over time, make sure that young Americans are able to receive Social Security benefits.

And of course, McCain wasn’t the only one — literally dozens of conservatives called it “privatization” before 2005. But once Republicans discovered that the word polls poorly, they ran away from it. The sad part is that the New York Times has acquiesced to their insistence that the term is some sort of Democratic invention.

Update 9/23 10:19 PM: Media Matters has more.

There’s been an outbreak of storytelling about the predictable decline of John McCain’s bounce and Sarah Palin’s favorable/unfavorable numbers.

For instance, Matthew Yglesias attributes Obama’s increasing numbers to the national focus on economic issues. Sam Wang calls it the “Palin Bounce.” Similarly, Ross Douthat blames the decline “primarily” on “negative press reports on her Alaska career… ongoing coverage of the still-simmering Troopergate scandal – and especially by her widely-watched, none-too-impressive interview with Charlie Gibson.”

From an epistemological perspective, we should be cautious about trusting any of these stories. The reason is that we have simpler explanations for both trends that have stronger empirical and theoretical support:

1. Presidential convention bounces tend to dissipate.
2. The poll numbers of national political figures become increasingly polarized when there is a two-way message flow about them (i.e. positive and negative information).

We can tell all sorts of (unprovable) stories about trends in polling numbers, which is what keeps pundits in business. At a minimum, it’s important to be precise about the claims that we make, keeping #1 and #2 above in mind. The relevant counterfactual for Yglesias is whether McCain’s bounce would have held up better had the national debate not shifted toward the economy. In the latter case, the question for Douthat is whether Palin’s numbers would have polarized as quickly if she had not been subject to as much negative coverage. Unfortunately, there’s no way to assess either of these scenarios.

More racial codewords are being directed at Barack Obama, who was called “uppity” by an anonymous Bush adviser:

A PRESCRIPTION FOR McCAIN, from one of the smartest Bushies: “I personally would make a lot of ‘accidental’ straight talk on the plane with reporters. Oh no. McCain was chatting with the press, slipped up and called the Paulson proposal totally lacking on detail at best and nothing more than bailout of Wall Street fat cats at worse. Oh no. McCain said it is time for a recusal by Paulson given that he is conflicted five ways from Sunday. Let staff say that they will see about clarification. Let McCain come back and say that he won’t apologize and in fact it is worse than what he said the first time. … And you also could guard against underwhelming debate performances, which McCain needs to be worried about given his ability in that regard and his history of performance. The tactics that got them to mid-September in a tie are not going to get them to 50 percent plus one in November. They need … an eye toward driving out the range of contrast that makes McCain different from Obama (action-oriented rhetoric v. grand prose; accessible v. uppity; humble servant of country v. arrogant).

Rep. Lynn Westmoreland (R-GA) previously called Obama “uppity,” as Think Progress points out. (By the way, what happened to Brad DeLong, who accuses Allen of making up the quote with no evidence?)

Similarly, Rush Limbaugh accused Barack Obama’s campaign of hacking Sarah Palin’s email account using racially coded language about Obama being “[a] community organizer, the street agitator, the Chicago thug”:

Obama, folks, is showing who he really is. A community organizer, the street agitator, the Chicago thug, clear the playing field, is finally being — it’s on display for everybody to see. Sarah Palin’s emails, personal emails, have been hacked, no doubt by Obama thugs. They dropped 30 people up there in Alaska trying to dig up dirt on her. Now they got some thugs that found her personal email address.

On the other side, Obama tried to exacerbate ethnic tensions in a Spanish-language ad that tries to link John McCain to inflammatory quotes from Rush Limbaugh that were taken out of context:

I’m expecting things to get much worse before this is over.

Update 9/23 10:17 PM: Brent Stapes has more on the disturbing racial discourse of the campaign in the New York Times.

Princeton’s Larry Bartels, who previously eviscerated Thomas Frank’s influential What the Matter with Kansas?, reports that a new conference paper by Steve Nicholson and Gary Segura provides yet more evidence that Frank’s thesis is overstated or wrong:

In What’s the Matter with Kansas, Thomas Frank argues that the Republican Party has
redrawn the landscape of contemporary American politics by displacing the Democratic
Party as the party of the working class. Although others have tested the electoral
implications of Frank’s arguments, no one has examined the phenomenon itself—the
belief that Republicans are the party of the working class. To test this claim, we created a
survey expressly for the purpose of examining the class-based images of the parties.
Following Frank’s argument, we examine whether the public believes Republicans are
the party of the working class and specifically whether the working class and Evangelical
Christians are most likely to hold this belief. Contrary to Frank’s argument, the
Democrats are the party most Americans associate with the working class. Furthermore,
we find that the working class and Evangelical Christians are no more likely to hold this
belief than respondents in other social classes or faiths.

Unfortunately, these critiques will reach .1% of Frank’s audience, which will never know how little empirical support his claims actually have.

Matthew Yglesias and Paul Krugman both suggest that taxpayers should get “a stake in the upside” (Krugman’s words) in return for bailing out the financial industry. But that’s unlikely to happen due to conservatives’ deep-seated hostility to government ownership of private industry. During the debate over Social Security privatization, many people pointed out that we could capture the upside of private equity investment (high returns) without the downside of individual accounts (high management costs, poor investing decisions) if the government simply invested Social Security funds in the market. But conservatives (including Alan Greenspan) killed this idea, warning of the dangers of government interference in private markets. Obviously the dire situation on Wall Street has changed the terms of the debate, making it possible for the government to buy almost 80% of AIG without protest, but it’s still not clear that the Bush administration would want the government taking equity stakes in all the companies it bails out.

For those who are interested, I’m doing another interview about the misperceptions research I’ve conducted with Jason Reifler. This one is with NPR’s Tell Me More with Michel Martin and will be taped live at 9 AM EST. Check your local listings for when it airs in your area. Alternatively, archived audio should be online tomorrow afternoon at the Tell Me More website.

Update 9/19 10:04 PM: Here’s the interview page which has a link to the audio clip.

I’ll be appearing on North Carolina Public Radio’s show The State of Things today at 12 PM EST to discuss my research with Jason Reifler on correcting misperceptions — you can listen live at WUNC’s website. Here’s the blurb:

Truth, Lies & Politics
Thursday, September 18 2008 by Daniel Zola and Frank Stasio

As the presidential election heats up, facts, figures and political fodder begin to fly. Reporters do their best to keep the candidates honest, but does fact-checking the claims really work? Duke Political Science PhD candidate Brendan Nyhan joins host Frank Stasio to discuss his research paper, “When Corrections Fail – The Persistence of Political Misconception” to uncover if the media are doing more harm than good.

Update 9/19 10:06 PM: Here’s the MP3 clip.