How the Tea Party movement is causing problems even for conservative Republicans
There's an interesting story in today's New York Times about Senator Robert Bennett's (R-Utah) troubles in winning his own party's nomination for re-election this year. Ordinarily, an incumbent like Bennett, with normally strong conservative credentials, no scandal, and lots of seniority, would have no problem, especially in receiving his own party's (re)nomination. But this is not an ordinary election cycle:
The dissatisfaction with Washington sweeping through politics is not only threatening the Democratic majority in Congress, it is also roiling Republican primaries. The Tea Party movement and advocacy groups on the right are demanding that candidates hew strictly to their ideological standards, and are moving aggressively to cast out those they deem to have strayed, even if only by participating in the compromises of legislating.
While I have my doubts that the Tea Party insurgency will lead to a significant number of Republican Senators losing their seats, one has to wonder at the strategic effects of all this - one of the problems of the 'Class of '94' was that they campaigned on a program of opposition, and once they actually held the reins of power they were either unable to implement policies or simply emulated the practices that they were elected to eliminate. While it seems unlikely (at this point) that the GOP will wrest control of either chamber as early as this election cycle, it's easy to see a similar fate befalling a Republican majority elected on a tide of Tea Party enthusiasm.
The Economist and ObamaCare counterfactuals
At lunch yesterday a liberal friend from Boston offered the comment that it was Scott Brown's victory in the Massachusetts Senate race that snapped the Democrats to attention and enabled yesterday's vote.
More Americans support the health care reform than oppose it
I'm a day late on this, but Gallup is reporting that 49% of Americans support the health care reform as passed by the House, with 40% opposing. Margin of error was ± 4%:
The health care vote will not help the GOP
From David Frum, via Joshua Tucker:
Conservatives and Republicans today suffered their most crushing legislative defeat since the 1960s.
It’s hard to exaggerate the magnitude of the disaster. Conservatives may cheer themselves that they’ll compensate for today’s expected vote with a big win in the November 2010 elections. But:
(1) It’s a good bet that conservatives are over-optimistic about November – by then the economy will have improved and the immediate goodies in the healthcare bill will be reaching key voting blocs.
(2) So what? Legislative majorities come and go. This healthcare bill is forever. A win in November is very poor compensation for this debacle now.
So far, I think a lot of conservatives will agree with me. Now comes the hard lesson:
A huge part of the blame for today’s disaster attaches to conservatives and Republicans ourselves.
At the beginning of this process we made a strategic decision: unlike, say, Democrats in 2001 when President Bush proposed his first tax cut, we would make no deal with the administration. No negotiations, no compromise, nothing. We were going for all the marbles. This would be Obama’s Waterloo – just as healthcare was Clinton’s in 1994.
Only, the hardliners overlooked a few key facts: Obama was elected with 53% of the vote, not Clinton’s 42%. The liberal block within the Democratic congressional caucus is bigger and stronger than it was in 1993-94. And of course the Democrats also remember their history, and also remember the consequences of their 1994 failure.
This time, when we went for all the marbles, we ended with none.
Frum 'gets it', in that he understands two key points:
- It's the economy that drives voting behavior, not specific pieces of legislation. Conservatives might be mad at the Democrats for 'ramming reform down the peoples' throats', but conservatives were hardly going to vote Democratic in the fall elections anyway.
- Once enacted, social welfare legislation is difficult, almost impossible to repeal. Social welfare programs create their own constituencies that are independent (no pun intended) from partisan identification. This is why Republicans have been so unsuccessful at dismantling New Deal or Great Society social welfare legislation. It's also why European countries with right-wing cabinets still have most of their welfare states largely intact.
The Democrats might lose a lot of seats in November, but if they do it's the economy that will be the cause.
Partisan differences and the 'public option'
I thought it would be interesting to look at historical patterns in partisan attitudes to government-sponsored health insurance. I looked at responses to the American National Election Survey question on self-placement on a seven-point scale, where a 'one' indicates the respondent favors an entirely government-run health insurance system, while a seven indicates the respondent supports only private health insurance.
What I found is that Americans, regardless of party identification, generally favor a mixture of both private and government health insurance. To be sure, Republicans are more favorable of private plans and Democrats more in favor of a government plan (with Independents somewhere in the middle), but there is surprisingly not a great deal of difference between the parties.
The first of the two graphs shows the mean value for members of each party, along with 'Independents' (N.B., 'leaners' were coded as partisans) on the seven-point ANES scale. What we see is a strong degree of consistency in attitude, especially among Democrats and Independents. Starting in the early 1990s Republicans became less favorable to the so-called 'public option', although this trend seems to have stopped to a large degree.
What I found is that Americans, regardless of party identification, generally favor a mixture of both private and government health insurance. To be sure, Republicans are more favorable of private plans and Democrats more in favor of a government plan (with Independents somewhere in the middle), but there is surprisingly not a great deal of difference between the parties.
The first of the two graphs shows the mean value for members of each party, along with 'Independents' (N.B., 'leaners' were coded as partisans) on the seven-point ANES scale. What we see is a strong degree of consistency in attitude, especially among Democrats and Independents. Starting in the early 1990s Republicans became less favorable to the so-called 'public option', although this trend seems to have stopped to a large degree.
The second graph shows the differences in mean position for each of the three groups. For example, the red line shows the mean difference in opinion between Republicans and Democrats. Two things are immediately notable:
- The differences in the average position between Democrats and Republicans has grown larger over time and continues to do so. The average difference between Democrats and Republicans during the 1970s was roughly one point on the ANES scale. In the 30 years between 1978 and 2008 it doubled. Clearly the divide between partisans, not party elites, has increased and in today's partisan environment, will probably continue to increase. This divide might help explain the seemingly great animosity between sided on the health care reform issue.
- The difference between Democrats and Independents was much greater in 2008 than in 1992. Both Bill Clinton and Barack Obama campaigned on health care reform, and on some sort of government-provided health care ('socialized medicine', if you will) as important parts of the platform. For each it was the dominant focus of their freshman year legislative agenda. Yet Obama was successful while Clinton was not. Why? One answer might be that the Democrats have larger majorities now than they had in 1992. Another reason might be that Democrats are now much more ideologically similar than they were in 1992, largely caused by the 1994 Congressional election all but decimating the Democratic Southern Caucus.
Ohio Issue 3, 2009
I'm trying to make some sense of the votes for and against Issue 3, a constitutional amendment allowing for a single corporation to hold a monopoly on casinos in the state, and granting them the right to construct up to three total in four designated counties around the state. So far I'm not having a lot of luck. I don't have individual-level data to look at, yet, if ever, so I'm stuck with county-level data. This constrains interpretation to a great degree. Some very general things are notable:
- Central Ohio voters were more likely than not opposed to the issue. Voters in Delaware and Franklin counties voted against the measure by roughly a 3-2 margin. Those in surrounding counties such as Pickaway, Madison, and Licking also opposed the amendment, albeit by a narrower margin.
- Voters in and near cities in the Northeast and Southwest, i.e. Cleveland, Cincinnati, and environs, supported the measure.
Harkin says he has the votes
Harkin says he has the votes to pass public option bill in the Senate - The Hill's Blog Briefing Room
Shared via AddThis
Shared via AddThis
How low can Obama's approval rating get and he still win re-election?
As long as he's at about 45% or above he should be OK.
100 day job grades
In a poll conducted from April 23-26, 2009 by the Opinion Research Corporation (on behalf of CNN), respondents were asked a series of questions, many of which focused on grading the performance of President Obama and the 111th Congress, 100 days in. Unsurprisingly, there was a strong relationship between how a respondent graded Obama and how they graded the Democratic congressional delegation. Oddly, a not very strong relationship existed between how Obama and the Republican delegation were graded:
(for details on how this plot was created, see here.)
Gender and non-partisanship
Women are far more likely to be partisans than men. This surprises me, and I can't come up with a good, theoretically sound reason for it. It is well established that women lean disproportionally towards the Democratic Party and men towards the Republican Party, but it also seems as if women lean towards partisanship in general, at least when compared to men. Think of it this way - women are just as likely to be Republicans as men are, but they are far more likely to be Democrats than men. And this pattern is consistent regardless of age or income.
Figure 1. Probability of non-partisanship among men and women, by age group, 2006.
Figure 2. Probability of non-partisanship among men and women, by age group, 2006.
Gallup partisanship trends, 2004-present
Gallup regularly asks poll respondents their party identification. The graph above shows the trend in party id since, roughly, the beginning of 2004. There is a general trend away from the Republican party, but not necessarily to the Democrats. It's also interesting to note that the anti-GOP trend has stopped, and possibly reversed, since the November 2008 election.
The pale dots reflect specific poll results, the wiggly solid lines represent LOWESS-smoothed trends, and the dotted lines represent linear trend estimates based on bivariate OLS regression.
The pale dots reflect specific poll results, the wiggly solid lines represent LOWESS-smoothed trends, and the dotted lines represent linear trend estimates based on bivariate OLS regression.
Who votes for green parties?
The educated middle-class left, it appears, at least in the European context. In looking at predilection towards voting for green parties rather than social democratic parties as part of a larger research project, I ran some quick and dirty logistic regressions, using data from the first (2002) wave of the European Social Survey. I'm looking at Europe (or at least the countries represented in the survey) as a whole, and also some individual countries. For the present time I am restricting my analysis to those respondents I have coded as "soft left", i.e., those who responded with a "3" or "4" on an 11-point (0-10) left-right scale. It is these voters whom I consider to be the most fertile "hunting ground", in a Downsian party space, for both social democratic and green parties. Table 1, below, shows that the 'typical' green or social democratic voter is just slightly left of a four on the 0-10 self-placement scale.
Table 1. Summary data for placement on a left-right scale of green and social democratic voters.
Next I will look at some proxies for middle-class status and their relationship with a predilection for voting either green or social democratic. Since my hypothesis is not only that persons who are currently middle class are, ideology being held constant, more likely to vote green than social democratic, but also (and this is actually the important past), those who have middle-class backgrounds, i.e., their parents were middle class as well, form the base of ecological party support. Therefore, two proxies for continued middle-class status are introduced, father’s level of education and father’s employment status – professional, middle class, working class.
1. Level of education
First, looking at the respondent's (voter's) level of education, there is in several countries in the survey a marked relationship in the likelihood to support either a green or social democratic party, based on level of education. This relationship is particularly evident in the case of Austria:
Figure 1. The relationship between education and the probability of voting either green (Grüne) or social democratic (SPÖ) in Austria.
The above graph shows Austria to be an almost perfect illustration of the effect of education on voting behavior among "soft left" respondents. the probability lines are almost symmetrical in their negative relationship - the less educated you are, the more likely you will vote for the SPÖ; the better educated, the greater likelihood you will vote Grüne, with the point at which the likelihood is the same being at roughly 15 years of education.
Figure 2. The relationship between education and the probability of voting either green (Vihreä liitto) or social democratic (Suomen Sosialidemokraattinen Puolue) in Finland.
Here we see a situation similar to the one in Austria, where "soft-left" voters likelihood to vote green can be seen as a function of education more than anything else, especially as we see education level as a proxy for social class (probably something we can get away with more readily in Europe than in the United States.)
Figure 3. The relationship between education and the probability of voting either green (Grüne) or social democratic (SPÖ) in Sweden.
Table 1. Summary data for placement on a left-right scale of green and social democratic voters.
Next I will look at some proxies for middle-class status and their relationship with a predilection for voting either green or social democratic. Since my hypothesis is not only that persons who are currently middle class are, ideology being held constant, more likely to vote green than social democratic, but also (and this is actually the important past), those who have middle-class backgrounds, i.e., their parents were middle class as well, form the base of ecological party support. Therefore, two proxies for continued middle-class status are introduced, father’s level of education and father’s employment status – professional, middle class, working class.
1. Level of education
First, looking at the respondent's (voter's) level of education, there is in several countries in the survey a marked relationship in the likelihood to support either a green or social democratic party, based on level of education. This relationship is particularly evident in the case of Austria:
Figure 1. The relationship between education and the probability of voting either green (Grüne) or social democratic (SPÖ) in Austria.
The above graph shows Austria to be an almost perfect illustration of the effect of education on voting behavior among "soft left" respondents. the probability lines are almost symmetrical in their negative relationship - the less educated you are, the more likely you will vote for the SPÖ; the better educated, the greater likelihood you will vote Grüne, with the point at which the likelihood is the same being at roughly 15 years of education.
Figure 2. The relationship between education and the probability of voting either green (Vihreä liitto) or social democratic (Suomen Sosialidemokraattinen Puolue) in Finland.
Here we see a situation similar to the one in Austria, where "soft-left" voters likelihood to vote green can be seen as a function of education more than anything else, especially as we see education level as a proxy for social class (probably something we can get away with more readily in Europe than in the United States.)
Figure 3. The relationship between education and the probability of voting either green (Grüne) or social democratic (SPÖ) in Sweden.
