Only The Democrats

Can turn what should be a sure thing into an angst ridden few days. The only thing more fulfilling than rooting for Celtic in European competition, or the Mariners prior to 1995, is consistently setting myself up for disappointment by supporting the Democrats.

I have several disorganized sets of semi-organized comments on this. First, the validity and reliability of polling in this election, second, the structural and contextual issues facing the Democrats, and finally, what the worst case scenario could mean.

538 has been doing their usual comprehensive, rigorous, and analytical job of covering the polling side of this race, but even they are flailing about at night, in a deep fog, on a Pacific Ocean beach, after drinking way too much vintage port, unable to work out east from west, north from south, blindfolded. With the tide coming in. (Yes, this is personal anecdote, save for the blindfold. But then it wasn't really necessary). Polling is going to be particularly problematic in this electoral setting. Polling houses rely on likely voter models that vary based on a series of assumptions. These assumptions are usually, but not always, defensible in a normal electoral setting such as a Presidential or even mid-term year, where we have heaps of past empirical data to model, from which we can estimate probabilities of who is more likely to turn out and who less likely.

A mid January election in the state of Massachusetts held in the current political climate does not afford such a wealth of data, needless to say. I would be extremely reluctant to use polls conducted in this electoral context as the empirical base for a projective model formulated during a "normal" year (2008) for Senate elections to estimate a probability of either a Brown or a Coakley victory tomorrow. Indeed, Franklin reports that the polling in this race is highly varied; where in general elections we find various poll estimations within +/- 5% 95% of the time, the results here are all over the map. This is not at all surprising, since nobody has a clear idea what the electorate will look like tomorrow.

However, there are a few certainties that we can cling to, and none warm my heart. First, turnout will be lower on Tuesday than in November 2008. As structural variance in turnout affects different socio-economic categories at different, and largely predictable, rates, it's safe to say that the drop off in Republican turnout will be less extreme than Democratic turnout. In a needlessly close rate hinging on independent voters, this matters, even in a state where Democrats heavily outnumber Republicans.

Second, while we can point to NY-23 for evidence of how the tea-bagger label or support can be toxic in certain settings, a year in to the administration, the present race can be assuming more of a protest vote dynamic. Obama and the Democrats are becoming associated with the economic malaise. While Obama is right to suggest that it takes more than one year to reverse what eight years caused, this evidential argument doesn't necessarily play. Democrats can't run against Bush indefinitely. There comes a point where the party has to present a positive message about its policies, and while this has always been a stretch for the Democratic Party in my lifetime, the effectiveness of "running against the past" will attenuate with time. As this is essentially a by-election where the fate of the executive branch is not at risk, and most people don't really understand how the U.S. Senate operates (i.e. that 60th vote does matter, like it or not), those content in their support for the current administration and the Democratic Party have less incentive to turn out than the tea bagging set, those swayed by seductive populist appeals, those independents who have reasoned concerns with the approach of the administration, or even those, independent and Democrat, who just want to send a message of protest or frustration with the current climate.

The Democrats, both in Congress and in the administration, are the incumbent party of power now, and will start to bear the burden of blame for current conditions.

Finally, as fellow Massachusetts pol Tip O'Neil famously said, all politics is local. Coakley has, at best, run a complacent campaign. At one extreme, the campaign can be portrayed as a campaign of entitlement -- she expected to coast to victory. I can understand the national Democrats' overlooking the Mass. Senate race, but the actual candidate still has to win, and convincing her base to support her in a special election in January is critical. She hasn't. Massachusetts has elected Republicans state-wide, and until 2007 had a lock on governor for 16 years. They're not scared of electing Republicans, and in an election that superficially appears to not matter, very well may do so.

The problem is that, as we know, this election does matter in both tactical and strategic senses. Tactically, it could (and probably would) torpedo health care reform. Of course, the Democrats can then blame the Republicans for this failure, but a) that's the only good that would come from a Brown victory, and b) this argument, too, assumes that average citizens understand that 59% of the vote in the US Senate is not enough. They won't get that, but rather see a party with 59% of the votes in the Senate failing miserably to pass the Administration's signature issue.

Strategically, the implications are clear. The Democrats can, and will, be portrayed both as free-spending socialists who can't be trusted with our tax dollars, yet simultaneously be portrayed as unable to simply and effectively govern with huge majorities in both Houses of Congress and the Presidency. We can't trust the Dems because they're pointlessly spending all our money, yet we can't trust the Dems because they can't seem to spend all our money.

This is before the symbolic value of it being Massachusetts, in Teddy Kennedy's seat, is exploited even superficially. The tea baggers and Republicans writ large will go ape shit, and the Democrats will embrace chicken little. Both will be over-reactions, but the potential damage caused by not passing health care reform should not be underestimated.

If Brown should win, I don't think the Democrats will suddenly lose the House in 2010. The conditions are different than 1994, which is another post entirely. They'll lose seats in both the House and the Senate, but anybody who knows anything about House elections already know that the Dems will lose seats in the House.

Will Brown win? Silver suggests a tend back towards Coakley, and these numbers don't include the weekend blitz by the Democrats (contra Silver, as he points out, Franklin reaches a different estimation). I wouldn't be surprised either way, yet I have faith that Coakley will pull it out. But it's nothing more than faith, and it pisses me off that it should come to nothing more than faith in Massachusetts of all places. Rationally, structural and contextual variables are aligned against the Democrats in this special election: significantly lower turnout favors the Republicans, the Democrats are now the incumbent party of government lending a protest-vote dynamic that typify special elections, and Coakley has run a magnificently shitty campaign.

However, in a state as blue as Massachusetts, the Democrats should have overcome these hurdles due to the built in partisan advantage. They haven't.

Read more...

Wingnuts hand seat to Democrat, Republican leaders apologize for trying to stop them

In today's "Newt Gingrich Letter", a publication that occasionally appears uninvited in my inbox:

In retrospect it is clear Dede Scozzafava should never have been nominated because she was far too liberal to be acceptable.

Republican leaders in New York must recognize that Mike Long and the Conservative Party in that state have to be consulted before decisions are made. The national conservative movement is a force that has to be recognized and respected.

Read more...

It's The Turnout, Stupid

As I feared, breathless, fatally flawed analysis follows the election results in NJ, VA, and to a lesser extent, NY-23.

Most of this article is remarkably sound, but the first paragraph sets the wrong tone:

The Republican victories in the races for New Jersey and Virginia governors put the party in a stronger position to turn back the political wave President Obama unleashed last year, setting the stage for Republicans to raise money, recruit candidates and ride the excitement of an energized base as the party heads into next year’s midterm elections.

It's difficult to argue that the Republicans are not in a stronger position in general, but as I argued yesterday, these elections had precious little to do with either Obama or national politics. The Democrats lost those two gubernatorial elections because they were either unpopular, ran lame campaigns, or both. It wasn't a newly energized base that swung the races; rather it was a combination of independents breaking R and a good chunk of the "Obama coalition" staying home.

Which we knew they would all along. Minorities, the young, the less wealthy, new voters do tend to stay home in odd years (and while I anticipate an uptick in turnout amongst these groups in 2010, it won't come near the level of 2008). These were the demographic categories that largely put Obama (way) over the top in 2008.

This, of course, is simply bollocks:

CNN senior political analyst Gloria Borger says that the projected GOP win in the Virginia gubernatorial race - and exit polls showing independents voting overwhelmingly for Republican Bob McDonnell - could be a problem for the president.

"This is a signal to this White House they have some problems right now – particularly on the economy and on the deficit," she said.

The White House did have a problem in Virginia: they knew that the Democrats had a mediocre candidate, which is specifically why they didn't invest too much time there. While this will be played up as a "signal" to the White House, it simply isn't. As several sources are reporting, the electorate that turned out in Virginia last night went 51%-43% for John McCain; in 2008 Virginia went 53%-46% for Obama.

The same was largely true in New Jersey. While the exit poll data do not appear to have asked vote choice in 2008, Obama's approval rating amongst those who voted in 2009 in NJ is 57%: yet 27% of those approving of Obama's job performance voted against Corzine.

In some ways, I think that there is more good than bad in these results for the Democrats. In New Jersey, for example, the 18-29 age cohort voted for Corzine by a fairly decent majority of 57% -- the only cohort to give him a majority (granted that cohort only represented 9% of the electorate, which is not terribly surprising given what I've been suggesting the past two days). While this did not hold in Virginia, of those that did turn out, the younger cohorts were more supportive of Deeds than the older cohorts. It would appear that a greater percentage of the "Obama coalition" did turn out in New Jersey whereas in Virginia they were more likely to stay home, which is a further nail in the coffin of those who argue that Obama has no pulling power. He campaigned heavily (for a sitting President) in New Jersey, and basically ignored Virginia.

And then there's NY-23. While it has been argued that NY-23 voted for Obama, therefore is not as Republican (or at least as Conservative) as conventional wisdom suggests, I suspect that the same patterns of turnout applied there as well as in NJ and VA. This one is the only one of the three that perhaps touched on national issues (though I'll stick to my original analysis that it was going to swing on local politics all along) as it ended up being a confrontation between a moderate Democrat and a Palinesque Republican. That the wingnut lost in a traditionally Republican district in an election where turnout patterns would strongly favor the Republicans if they ever do suggests that the 'return to core conservative values' model of reform in the Republican Party is not a winner, at least in non wingnut constituencies, and that perhaps 2009 was more about the Republicans than the Democrats / Obama all along.

UPDATE: and of course, leave it to the Brits to get it hilariously wrong:

Major Republican victories in two states last night left the fate of President Obama's signature health reforms in doubt and Democrats licking their wounds a year before the 2010 mid-term elections.

. . .

A year ago, Mr Obama became the first Democrat in 44 years to carry Virginia in a presidential race. This time voters expressed concern about major Obama initiatives on energy and stimulus spending as well as healthcare. Exit polls showed that independents broke heavily for Mr McDonnell.

. . .

"This reflects a sea change in the electorate," said Michael Steele, chairman of the Republican National Committee.

Sigh.

Read more...

Election Over-reaction

Several races today have received an inordinate amount of attention as tests of public opinion regarding the policies and effectiveness of the (still) nascent Obama administration. The Democrats will lose two of the three races being hyped, and very possibly all three.

Many will argue that this will be a setback for Obama, especially considering the political capital he has expended in NY-23, and especially in the New Jersey Gubernatorial election.

This is, largely, bunk. While special elections (NY-23) or even Gubernatorial elections can be suggestive of public opinion towards national politics (the 1991 special election for the Senator of PA is a good example here), this is relatively rare in American politics. Gubernatorial elections are about state politics, not national politics. The incumbent in New Jersey, Jon Corzine, is in a statistical draw at the moment because he's Jon Corzine. As Silver reports at 538, Corzine has not polled higher than 44%, and 53% of likely voters have an unfavorable opinion of the incumbent. This is a race about neither the Obama administration nor the relative popularity of Republicans in a blue state; rather it's a race about the unpopular incumbent governor facing off against a lackluster challenger and a third-party spoiler.

In Virginia, an argument that this is a referendum on national rather than state politics is more credible, but where Silver assigns a 3-1 split between national and local, I'd go no more than 50-50, and only then if I were sniffing glue. Contextual factors specific to the state matter. Deeds, the Democratic nominee, participated in a three-way primary that, while he handily won, could not have helped position him against Bob McDonnell, the Republican nominee. Second, while these very two faced off for Virginia Attorney General in 2005, and McDonnell only won by something like 300-odd votes, McDonnell has spent the last four years as a state-wide elected official, while Deeds did not. I'm not suggesting that this gives McDonnell an insurmountable edge in the rematch, but it does give him a marginal (perhaps very marginal) advantage.

Finally, and most critically (for all three races), the composition of the electorate will be significantly different in November 2009 than it was in November 2008. When turnout decreases, as it does for off-year elections and especially for odd-year elections, the underlying composition of the electorate is altered at differing rates. Those with lesser levels of education, lower levels of income, lower age, and less attachment to place drop off at a much higher rate than the wealthy / educated / home owners / etc. It's not terribly difficult to make the leap (actually, a small hop) from this to speculating (correctly) which party will benefit from the changed demographic composition of lower turnout.

As for NY-23, it's a Republican district. Yes, there was significant ticket splitting in 2008, but it's a strong Republican district: a Democrat wasn't even on the ballot in 2002. Perhaps it is not a wingnut Republican district, but it looks likely that it will be represented by a credible wingnut following this election.

Indeed, as Silver points out, the Republicans in the two Gubernatorial races aren't exactly hyping up their Republican street-cred. Granted neither are the Democrats especially so, but the Republicans are running from their label. If these were truly referenda on national politics, one would expect to see this distinction made more plain.

None of this is to suggest that Democrats still have it easy. The Republicans and right wingers are, as usual, far more adept at framing the narrative and mobilizing their support. Furthermore, the Democrats will lose seats in the House in 2010. But then aside from 1998 and 2002, the incumbent party in the White House always loses seats in the off-year Congressional elections. But, the results that I will wake up to tomorrow will not have me terribly concerned about the fate of the Obama administration, progressive (or even centrist / moderate) politics, or the fate of the free world. Furthermore, seeing how the New York Yankees can't possibly have won the World Series by (my) tomorrow morning, it's all good.

And all the better if Sarah Palin takes credit for any or all of these election results.

Me, I'll be watching R-71 in Washington State and Prop 1 in Maine . . .

Read more...

Plebiscitarianism run amok

In the process of voting, I find myself pondering "King County Charter Amendment #1: Repeal of Section 350.20.30 and Portions of Article 9 - Transitory Provisions" Here's the text in full:

Shall those no longer relevant portions of King County Charter Article 9 relating to the county's prior transition to a home rule charter and King County Charter Section 350.20.30, relating to the county's transition to a metropolitan form of government, be repealed, as provided in Ordinance # 16484?"

If you're anything like me, you tried to read this several times, but kept falling asleep. Apparently, we're being asked to vote on whether outdated language that refers to another no longer present section of the charter should be removed. I was tempted to vote no as an irrational protest vote against the madness that is the King County Charter. There's a reasonable debate to be had about governing by plebiscite, but I can't really figure out what a defense of copyediting by plebiscite might look like.

In other news, King County people please vote! I think we'll win the big three (Yes to human rights and decency (71), No to disastrous state budgeting practices and Tim Eyman (1033), No to Susan "Sarah Palin II" Hutchinson) but these off year elections are somewhat unpredictable and losing any of these would drive me to drink.

Read more...