Platykurtosity

There are times when political journalism comes off as a random trite-solution generator, a well-rehearsed circus act displaying the shocking underbelly of American politics that is entirely new, entirely horrifying, and suspiciously similar to last year. Let’s rip the beard off this sideshow and expose ten common tropes of political journalism that are utterly bunk:

Political independents buck partisanship and vote common sense

Every election year, journalists take it upon themselves to inform the two major political parties of the existence of a sacred “middle ground,” a wide swath of voters who carefully evaluate candidates and change their minds through the course of a campaign. Depending on who you ask, those who identify as “independent” are about 30% of registered voters, and those who say their minds aren’t made up yet are about 25% of the electorate at any given time. According to this trope, candidates must reign in their ideological rancor, pulling back the blinders of party affiliation to appease skeptical masses of undecided voters.

The reality is simpler: Independents are closet partisans. When people respond to a telephone poll, they often aren’t asked just to identify as “Democrat,” “independent” or “Republican.” Respondents who identify as “independent” are asked a follow-up, usually phrased something like this:

As of now, would you say you are leaning toward either of the major candidates? If you have to choose, would you say you are strongly leaning toward one of the major candidates, leaning slightly, or not leaning either way?

And if you track the self-identified “leaners,” they usually behave something like this:

And to political scientists, this is old news.

Honestly, this “closet partisan” hypothesis isn’t as well-established in the literature as some political scientists would like you to believe, but what is now virtually incontrovertible is that independents can also be classified as “low-information voters.” Not only are independents not neutral, they’re not particularly well-informed.

There is such thing as an elusive, unheeded “common sense” in politics

That is the hypothesis of Thomas Friedman, one of the chief architects of online third-party forum Americans Elect. In his words:

The Democrats are ready for more stimulus but have refused to signal any serious willingness to cut entitlements, like Medicare, that we know are unsustainable in their present form. The Republicans are all for spending cuts but refuse to accept any tax increases that we need to pay for the past and invest in the future. So what we’re basically saying as a country is that unless the market or Mother Nature make us pay, we are going to hand this whole bill over to our children.
Maybe it is just my friends, but I find more and more people completely disgusted by this situation and looking for a serious Third Party candidate who could run in 2012 and deliver the shock therapy to the corrupt, encrusted, two-party duopoly now running the show in America.

These beliefs imply politics should be easier than politicians are making it – that there are “commonsense” solutions to America’s problems that, absent wealthy vested interests and stubborn politicians, would be effortlessly and usefully applied. This, in turn, implies a few things about how political parties come to represent their own ideals:

• Parties exist irrespective of majority opinion, and their views are developed without input from the public.
• Politicians have rigid views on every issue, and will not change them even if the “problem” in question has an obvious solution.

There are too many obvious counterexamples for the first point to survive more than a moment’s thought. Every year, political parties spend millions of dollars on focus groups, public opinion surveys, and other means of discerning precisely what people view as “common sense.” Indeed, similar efforts are made to try to figure out what words to use in framing an issue (“healthcare reform,” anyone? “Socialized medicine,” perhaps).

One of the grandfathers of political forecasting, Stephen Rosenthal, found that the ability of a political party to “emphasize” its majority-approved positions and “downplay” its less popular views is integral to its success. He argues, for example, that de-emphasizing the Democratic Party’s recently reversed stance on civil rights was vital to John F. Kennedy’s electoral victory, and that the ability of the Republicans to identify Jim McGovern as the anti-Vietnam candidate helped him cruise to victory. Without adherence to “common sense” principles, these attempts to position ones party as having precisely those principles wouldn’t occur.

And on that note, does it sound like Friedman’s ideas are not nearly so ‘radical’ as he claims?

Such a Third Party would have a simple agenda: 1) Inject a short-term stimulus. 2) Enact Simpson-Bowles. 3) Shrink our presence in Afghanistan. 4) Raise automobile mileage standards. 5) Impose a gasoline tax to pay for a massive increase in government-supported scientific research and a carbon tax to pay for new infrastructure and stimulate clean-power innovation.

As of now, the President is about two and a half of four. “Radical,” indeed.

Third parties ruin elections

After Al Gore lost the 2000 election, the case was made that the Green Party “cost the Democrats” the election, and otherwise they would have won. The math here is simple: George Bush won 48.85% Al Gore won 48.84% of the vote, and Ralph Nader won 1.64% of the vote, so had Nader not been in the election, Gore would have taken his 1.64% and, with it, a slim but solid majority.

I call the assumption underlying this line of thought the “Transitive Property of Voters fallacy.” It may be generally true that informed voters vote for a “next best thing,” a politician that is closest if not exactly in line with their views, but it can’t be correctly inferred that an election between two candidates and an election of three candidates involves precisely the same voters. Doubtlessly, the Green Party in 2000 engaged in proportionally extensive voter registration and voter contact efforts as the Democrats and Republicans. If it possibly switched voters from other parties, the Green party campaign created new voters, as well.

This is the year for third parties

Every Presidential election year, voters are fed up. With the President. With Congress (but not their own Congressman).With everyone. But third parties face tremendous disadvantages: they aren’t legally institutionalized the same way the Democrats and Republicans are. They are by virtue of fitting under the umbrella term “third parties” smaller, and poorer. And, given what we now know about the finicky middle, their ideas aren’t so popular as one might infer from, say, the volume of web traffic occupied by self-identified libertarians versus their track record at the polls.

When Ross Perot took 8.4% of the electorate in 1996, he made Bill Clinton one of the lowest-vote share Presidential winners in the twentieth century. In 2000, Perot’s Reform Party’s share could be confused with a rounding error. The Reform Party operated from celebrity candidate to celebrity candidate until it ran out, and it died. Lack of party infrastructure and the host of daunting challenges facing third parties caused terminal flame-out after 1998.

Historians and political scientists consider the Democratic Party the most successful third party in America’s history. Well, not all of it, just about half. The southern flank, to be specific. In terms of policymaking, Southern Democrats acted as a successful pivot for Democrats elected outside the South and for the Republican party for decades.

Steering the institutions of one of the big parties toward one’s own desired agenda is much easier, and more effective, than attempting to establish a solid third party. Did anyone notice what happened when the “Tea Party” stopped producing candidates, and instead ran “Tea Party Republicans” in Republican primaries? They generally failed, but they had a better shot at such than, say, the “real” Tea Party, which, as you probably never heard, ran candidates in 2010.

Lengthy primaries hurt parties

The commentariat is concerned about how the length of the primary election effects the winner’s prospects in the general election. There are some intuitively reasonable arguments why long primaries may hurt the eventual nominee: A longer time spent trying to court party-registered voters means less time to court anyone else – and more time having to be spent saying things to appeal to party loyalists rather than, say, centrists. On January 12, 2012 – the night I wrote this section of this post – the big news is that Ron Paul and Rick Perry are vowing to stay in the fight to the bitter end, regardless the consequences of making Romney spend more time appealing to the median Republican rather than the median American.

For those of you stumbling upon this post late – post-November late, perhaps – you know where I’m going. There is little evidence that lengthy primaries hurt. In 2008, for example, the bloody Democratic primary didn’t hurt Obama, especially among Democratic voters who lived in a battleground state – i.e., those who had to bear the most blood to begin with. Elsewhere, political scientists wonder whether short, painless primaries can actually hurt candidates. Imagine the case of buyer’s remorse if you thought you were getting someone more conservative than William Scranton and wound up with Barry Goldwater, instead.

As then-Senator Obama learned in 2008, the act of conducting a primary is quite useful to the act of conducting a general. In either case, candidates scramble to register and turn out new voters, secure the donations of the important and insider’s approval from the powerful. Dramatic primaries have the added benefit of the added media attention that comes with a dramatic election. Fun fact: the Democratic primaries in Iowa and New Hampshire both took place on the same nights as the Republicans.’

The “base” must be “mobilized” and tended with care

Arguing against this idea may sound odd coming straight off arguing for the value of primaries in “activating” a “base,” but in a general election, the notion of a “de-motivated base” is hard to come by empirically. Regardless what you hear, “the base” is a relatively stable phenomenon, carried by – you guessed it – white middle-aged and older people.

Source: American Community Survey, or as I like to call it, “The Census, Jr.”

Smoothed for off-year election effects. Quick, spot the most “energizing” candidates. Bonus points for recognizing that Reagan II and John McCain ran in years with equal turnouts among three of the four age groups. The reasons voters give for voting are varied and complex, but few of them have to do with some “energizing” speech or event. There seems to simply be a type of person who is motivated to be involved in politics. Robert Erikson sums it up nicely: People vote because they are registered.

Losing independents means you’re losing an election

Actually, this one’s true – just vapid. This is because, given independents are simply closet partisans (see above!), losing independents is just one way to tell you’re losing everyone. Consider, for example, the loss among virtually all demographic groups the Democrats suffered in 2010:

John Side sums up:

Democrats lost votes among men, women, blacks, whites, Catholics, Protestants, young voters, old voters, etc. Sure, the shifts within these groups were not all identical, but 2010 is much more about the similar direction of the shift, not the differences among groups.

So odds are, if you’re losing independents, they are probably tied with many other groups among the most of your troubles.

Major policy victories increase likelihood of electoral victory

Enormous policy victories offer a temporary bump, then disappear. George H.W. Bush won a decisive military victory, his approval rating vanishing all the while. The Clinton Administration made historic headway on the deficit, and in 2000 George Bush enjoyed a 9-point polling lead on the issue of “the economy.” The Obama Administration oversaw the death of Osama bin Laden – it is yet to be seen how Mitt Romney will start on the issue of national security, but President Obama’s advantage is unlikely to last. The Monkey Cage has an excellent graph on just how long the Republican Party was able to “own” the education issue in the past quarter-century:

“Mavericks” are the real stars of politics

The political sobriquets of our time tell their own tales. Ron Paul is “Dr. No.” John McCain calls himself “the original maverick.” Dennis Kucinich is a “liberal firebrand.”

Of our examples, John McCain was found in a Congressional Quarterly study to have voted along party lines on 95% of votes taken in 2007-2008 (that same study shows Senator Obama voted along party lines almost 100% of the time).

Calling Ron Paul a “maverick” is an overstatement, and calling Dennis Kucinich a “maverick” is downright inaccurate. Consider each of their Project Vote Smart “Key Votes.” These are roughly 1,000 important votes taken over the past 14 years, and their positions on them. Let’s consider classical political “defection:” Your party introduces legislation with the intent of passing it, and you vote against it. From 1997-2005, Ron Paul’s defection rate was a steady, high 65% or so, as in he voted “nay” on about 65% of bills proposed by his fellow Republicans. Dennis Kucinich only had four years of majority control to show his “radical” colors, but even then his defection rate appears to have been much lower – around 30% over the course of four years.

Notice in the most recent year for which data is available Ron Paul’s defection rate drops significantly. At first I thought this was because of the size of the Republican majority – a stronger majority means the ability to more radically steer policy toward your preferred position, so your proposed bills are less likely to turn off radicals – but I suspect there’s another reason. In 2011, Ron Paul spent much of his time running for President, as he did in 2007. I went back and re-coded “defection,” to see when he was voting for or against the minority, for against the majority, or not voting. Turns out when Ron Paul votes “No,” regardless of party in power, it always puts him on the losing side – in other words, Ron Paul is a professional protest voter. His defections are virtually never ‘yea’ with the losing minority.

Kucinich is more predictable. From 2002-2009 he strictly opposed the Republican agenda; before that he was strongly bipartisan, and he was a protest vote while the Democrats controlled Congress.

Ron Paul is more of a maverick when his votes don’t matter, and he’s often too busy to hit the floor for his ideals. Dennis Kucinich meets few criteria to warrant the use of the term.

Polls don’t matter

The mother of all political myths concerns the realm of campaigning in which I got my start: polling. It needs no introduction: Don’t trust the polls. Numbers are unreliable. There are three kinds of lies: Lies, damn lies, and cliché sayings falsely attributed to American writers. Our final political white whale is the inaccuracy of polls. Polls are spectacularly predictive, and are getting more so all the time.

Nate Silver found that even merely acceptable pollsters were only off from the final results of given Federal elections in 2010 by about 1.8 points. Stochastic Democracy pointed out the flaws in Silver’s claims, ran a different method and arrived at much the same conclusion.

Much of this has to do with the competitive nature of polling, wherein the pressure to adopt new and adaptive technologies and methodologies drives progress inexorably forward to a possible maximum, but so far cell phones and Skype-only callers have done nothing to reverse this trend.