These are the occupations on the opposite extremes of the liberal-conservative spectrum based on their campaign contributions according to research by Adam Bonica. As Andy put it: the conventional wisdom isn’t always wrong, even after using sophisticated statistical methods and hard data. See here for more on what campaign contributions can tell us about the ideology of PACs and candidates.
John Judis says yes:
[A] president’s political acumen—his ability to put the best light on his and his party’s accomplishments—can mitigate the effects of rising unemployment. That’s what Ronald Reagan and the Republicans achieved in the 1982 midterm elections…
Using economic models, some political scientists predicted that Democrats would pick up as many as 50 House seats. The Democrats also hoped to win back the Senate, which they had lost in 1980. But when the votes were tallied, the Republicans lost 26 House seats and kept their 54 seats in the Senate.
And Brendan Nyhan plows this argument under. The problems are many. Judis picks one forecasting model but ignores others that got the seat loss almost correct. And, even troublesome for Judis’s thesis, Reagan’s approval actually declined during 1982 — precisely the time when he was “putting the best light on his and his party’s accomplishments.”
The lesson has nothing to do with Reagan per se. It’s simply that presidents are far less powerful as communicators than Judis and others seem to believe.
The anti-immigrant Front National did much better than pollsters expected in recent French regional elections. Arthur Goldhammer reports that the French newspaper Le Monde suggests that this may be due to a Bradley effect: the idea that voters are reluctant to admit that they are voting against a popular black candidate or (by extension) for a party widely deemed to be racist. The Monkey Cage featured a lot of discussion on this issue during the 2008 Presidential election. The general consensus appeared to be that there was no evidence for a Bradley effect during the 2008 election and that the effect in general had been waning in recent years. My colleague Dan Hopkins concluded (non-gated, gated) in his brilliant study into the Bradley (or Wilder) effect that it is “the product of racial attitudes in specific political contexts, not a more general response to under-represented groups.” It seems like France represents another potential political context to study this effect. Indeed if, and I think Dan is right, political context conditions attitudes on racial and ethnic issues, then this should be a very fruitful area for comparative research.
.
I know all of you have been eagerly awaiting updates about the Dutch cabinet crisis I blogged about earlier. Well, here are some interesting recent developments. First, the most recent polls show that the PvdA (the red line in the graph below) continues to gain from resisting troop renewals in Afghanistan and forcing the cabinet crisis. The party has won nearly 10 seats in the pollls since the crisis.
Second, and more puzzling, two of the politicians who stood to gain most from these developments have decided in the last two days that they need to spend more time with their families (see also here). PvdA leader Wouter Bos, who was much beleaguered before the crisis but came out of it the undisputable leader of his party, made the announcement today. This follows a similar announcement by Camiel Eurlings, who was seen as the crown prince for current prime minister Balkenende.
Third, the recent polls and local election results imply that the prospect of Geert Wilders’ PVV becoming the largest party is a realistic one. In the Netherlands, the largest party almost always delivers the prime-minister. Speculation on what this would mean will have to wait until another blog post. In the mean time, go read why Daniel Pipes is enthusiastic about it and even calls him “the most important European alive today” while Ian Buruma is a little less impressed with Wilders.
Thanks to the wonders of podcasting, I can now listen to NPR’s Marketplace while jogging practically anywhere in the world. Because of this, I had the opportunity over the weekend to hear Megan McCartle claim on Friday’s Marketplace that - and I’m paraphrasing here a bit - if the Democrats went ahead and passed healthcare without Republican support, it would cost them the House and the Senate this November. Apparently, McCartle is not the only one to hold this viewpoint, as today’s Washington Post features a quote from Rep. Cantor essentially making the same point; the Post quoted him as saing that “if Democrats push the bill through on party lines, they will ‘lose their majority in Congress in November’”.
Really? Let’s parse this logic for a second here. Implicit in these claims seems to be the following two beliefs. First, if the Democrats don’t push through healthcare reform on party lines, they will hold on to their majorities in the House and the Senate. Second, if they do push healthcare through and susequently lose their majorities, it will be because of the Healthcare bill.
John has already addressed the first of these points in his prior posts on how the midterm elections will most likely largely be a function of the state of the economy (see here and here), so I’m going to set aside the question of how the Democrats would do in the 2010 midterms if they had never even brought up healthcare (answer: if unemployment is still at 10%, lousy).
Instead, I want to focus on the second question: are the Democrats somehow better off politically if they fail to pass a healthcare bill? Here are the numbers according to the most recent Gallup Poll. If the two parties can not reach an agreement, then 42% of Americans favor passing the Democratic plan, and 49% oppose it. Note that that 49% could include those who want a public option, a single payer plan, etc. But either way, 42% support the plan, 49% oppose it. If the question wording is changed to include the use of reconciliation, then support drops to 39% and opposition goes up to 52%.
So if the bill is passed, it will not have the support of the majority of the country. At the same time, if healthcare fails, then a significant portion of Obama’s first year(s) in office will have been wasted on a failed major policy agenda and the Democrats will be portrayed as divided, incompetent, etc. Which would you rather take into the midterm elections? The President/Congress that succeeded where Clinton, Truman, etc. had failed in the past by passing healthcare reform - but without the support of a majority of the population - or divided, incompetent, failure? Now I’m not saying that the fate of this bill will significantly impact who controls the House or Senate after the 2010 elections one way or another, but it seems to me that if it did, then not passing your most meaningful domestic policy objective would ultimately be more damaging to your political prospects then passing it with the support of only 40% of the country.
During the 2008 presidential election we submitted letters to the editor at 100 major U.S. newspapers as part of a field experiment to test whether interest in the letter depended on which candidate the letter supported. We find, contrary to what charges of a liberal media bias would suggest, that newspapers expressed more interest in pro-McCain letters than pro-Obama letters. Further, we find that papers were most likely to be interested in letters supporting the candidate they did not endorse, a result that is consistent with the idea that editors seem to be using their gate keeping powers to allow dissenting opinions to be heard.
That is from a newly published paper by Daniel Butler and Emily Schofield. An ungated version is here.
The forecast is Alan Abramowitz’s, discussed here. A couple of tidbits:
Contrary to many other analyses, however, the results of the forecasting model indicate that the main factors contributing to likely Republican gains in November are structural and do not reflect an especially negative political environment for Democrats. The current political environment only appears unfavorable for Democrats compared with the extraordinarily favorable environment that the Party enjoyed in both 2006 and 2008…
…Even under what might be considered a best-case scenario for Democrats, if President Obama’s net approval rating were to improve from a +5 to a +20, and Democrats were to regain a 10 point lead on the generic ballot, Democrats would still be expected to lose about 20 seats in the House.
Salon is starting a new feature called The Numerologist which is going to feature political scientists writing about politics using analysis that involves at least some reference to numbers or figures; I believe most posts will involve a figure or a graph. They asked me to write the first one - I think John is up on Wednesday - which featured the graph above. It was accompanied by short a blog post reiterating the point made by the chart that – at least for now – there are actually more Republicans than Democrats currently vacating seats in both the House and the Senate, a situation that was definitively not the case in 1994. Now this doesn’t mean that the Democrats won’t get hammered anyway due to the state of the economy or that Republicans aren’t retiring from safer seats than Democrats, but it is worth keeping the basic numbers in mind, something that seems to get lost in the media’s response every time another Democrat announces a retirement. For more, see the post at Salon.
A few months ago, I wrote that, based on the so-called generic ballot (surveys that as, “If the elections for Congress were being held today, which party’s candidate would you vote for in your Congressional district?”) and some research by Bafumi, Erikson, and Wlezien, the Republican Party looked to be in good shape in 2010.
Recently, Harry Joe (who took a statistics course from Joe Bafumi at Dartmouth!) crunched the early poll numbers, does the adjustment based on the Bafumi/Erikson/Wlezien regression, and comes up with a forecast that the Republicans will win by 8 percentage points (that’s 54%-46%); that is, 1994 all over again, but this time without the benefit (yet) of any Newt Gingrich-like figure (although I’m sure there will be many proto-Gingriches around to claim the credit if this all happens). Here’s Harry Joe’s plot:
For those who haven’t tired of this subject, Mark Blumenthal of Pollster has a nice column at the National Journal. He was kind enough to cite my post on the myths about independent voters. See also this piece by Alan Abramowitz, which Mark notes and I (somewhat inexcusably) had not seen before.
Mark’s column updates my chart of Obama approval and also discusses the implications of “independent leaners” for the Tea Party.
As Forrest Maltzman beautifully noted in his eulogy, Lee didn’t like to leave things undone. One exception is the only paper he and I wrote back in 2004, which means quite rightly that the blame must fall on me.
Our paper analyzed the extent to which daily perceptions of campaign negativity (measured in the Annenberg survey) tracked actual negative attacks by the Bush and Gore campaigns in the 2000 elections. Our answer: not at all. Instead we found that perceptions stayed remarkably consistent throughout the campaign with one major exception: on October 11 (the day of the second debate) a sudden but lasting change occurred (see the graph below). Up until then, the Bush campaign was thought to be more negative than the Gore campaign by about ten percentage points. Yet, Bush more than neutralized that difference in one day.
What happened? It wasn’t an unusually negative attack by Gore nor was it Gore’s performance in the second debate. Almost all interviews were held before the debate actually took place and the result holds when we remove the other survey respondents from the analysis. Instead, we suggest that it was a media effect:
Tens of millions of Americans watched the first debate, but the lasting effects of that evening’s developments – especially Gore’s exasperated sighs and overbearing demeanor – were not registered until more than a week later, when, on the eve of the second debate, virtually every discussion focused disparagingly on his first-debate performance.
A nice twist in the paper is that we applied a Bayesian change point model to identify the precise day in which a shift in perceptions occurred. We submitted the paper as a research note and received an invitation to revise and resubmit from a respectable journal. Nevertheless, we thought it would make more sense to wait until the 2004 Annenberg survey came out to see if similar sudden shifts in perceptions of campaign negativity occurred. Unfortunately, we never got around to this (and I promised Lee’s spouse not to remind him of this). Although I know it won’t help my Sigelman number, I have now made the paper available on SSRN (the link to it from my web-site had been broken for years).
I recently attended a talk by Dan Kselman about the relationship between electoral systems and corruption; a draft of the paper is available here. In the paper, Dan develops a model that includes the assumption that legislators in PR systems will identify a particular region within their district to target particular attention. There was a bit of a spirited discussion about the merits of that particular assumption, so I thought it might be interesting to throw the question out to the readers of the Monkey Cage. For those of you who have closely studied party competition in PR systems, does this sound like something you have observed in practice? Conversely, does this sound like something that does not happen in the country(ies) you have observed? Either way, please let us know in the comments section, especially if there are published works documenting either type of campaign strategies.
I also asked Dan to provide a brief synopsis of the argument for the readers of the Monkey Cage, and this is what he sent along:
The posted paper draft (not quite complete…) develops a game theoretic model to study legislators’ incentives to provide constituents in their electoral districts with particularistic goods and services (pork, social services, ombudsman services, etc). In single-member district systems, identifying the ‘target’ of such legislative particularism is fairly straight-forward: incumbents target such projects and services to residents of their single-member district.
In multi-member district systems, specifying theoretically the set of voters whom incumbents may target to receive particularistic goods and services is less straight-forward. For example, studies of clientelism in countries as varied as Brazil, Argentina, and Turkey have emphasized that clientelistic targeting requires ‘deep’ constituency relationships, and that incumbents have no way of maintaining such relationships with all voters in larger multi-member districts. As such, incumbents tend to maintain personal relationships in regional or municipal strongholds within larger electoral districts (aka bailiwicks). Similarly, research on Japanese politics suggests that incumbents often maintain personalistic relationships with particular professional subgroups within larger electoral districts.
On the other hand, one could also think of legislators in larger multi-member districts building a bridge which pleases all voters in that district, and not simply those in a well-defined geographic or professional strongholds. So, the question is this: how do legislators in multi-member district systems choose the subset of district constituents to which they will target their particularistic efforts? In the attached paper, the assumption is that legislators can choose (or choose not…) to develop personalistic relationships with subsets of voters in multi-member districts. Does anyone have examples outside of those mentioned above where this type of ‘carving out’ occurs? Or, does anyone have examples of situations in which particularistic efforts are targeted to the entirety of voters in a multi-member district?
In a post last week, I raised the question of why it was that whenever we think about economic voting in legislative elections as meaning that bad economic conditions hurt the incumbent candidate, we always consider the incumbent to be the candidate representing the president’s party. In a closed-list proportional representation electoral system, this makes perfect sense: the only option the voter has to vote for or against particular parties, and the voter has no say over which party members actually get to sit in the legislature. But in single member district electoral systems, as we have in the United States, angry voters could choose to throw the current office holder in their district out of office, regardless of whether that office holder was of the president’s party or not.
In response, John cited research by Alan Abramowitz suggesting that empirically this just doesn’t occur very often (although note Abromowitz is talking about the even more proximate cause of discontent with Congress, as opposed to just poor economic conditions):
Discontent with Congress does not lead to a general tendency to kick out incumbents. Occasionally voters do get upset and give the boot to a large number of incumbents—but they almost always take out their dissatisfaction on the members of only one party—the president’s party.
Larry Bartels offered a reason why this would be the case. Looking for a punishment effect of the type I suggested:
requires one to ignore lots of good evidence that voters see little connection between “Republicans in Congress” and their own Republican member of Congress.
And as John noted, Abramowitz also wrote that:
As the noted congressional scholar Richard Fenno has observed, Americans generally love their own congressperson even though they dislike Congress. They tend to see their own Senators and Representatives as the rare good apples in an otherwise rotten barrel.
So with that discussion in mind, here are some interesting results from a new CNN poll (for details, see here). The figure displays the percentage of respondents who believe (a) that most members of Congress deserve re-election and (b) that their representative deserves reelection in 2000 and 2010. (The bars do not sum to 100% because I did not graph the “no opinion” respondents.)
Two facts are quickly apparent. First, Americans are clearly much more dissatisfied with Congress now than they were a decade ago. Second, both in 2000 and 2010, Americans liked their own representative more than they liked Congress as a whole.
The interesting remaining question, though, is what to make of the fact that the proportion of Americans who believe their own member does not deserve to be re-elected is now close to the proportion believing their representative does deserve to be reelected? It is of course possible this is simply picking up partisan preference (ie., Democrats who have a Republican representative and visa-versa), but then shouldn’t we have seen a roughly similar pattern in 2000? It is also possible that the Tea Party movement and the liberal blogosphere are combining to make many Democrats and Republicans prefer potential primary challengers to their representative (and thus answer the my representative question with a “no”), but that they will return to the fold once primary season is over. Alternatively, the data could be consistent with the claim that we are approaching an election where anti-incumbency sentiment could work against both Democratic and Republican incumbents. Potentially further evidence in this regard is the fact that poll also reports that while 54% of respondents believe most Democratic members of Congress do not deserve to be re-elected, 56% believe most Republican members of Congress do not deserve to be re-elected.
Of course, if Abramowitz is correct that the only thing that matters in Congressional elections is presidential approval rate, then I suppose we can safely ignore these figures. I guess I would feel more confident doing so if we could find elections where similar proportions of the population were dissatisfied with their own representative - thus undercutting the micro-level mechanism Bartels and Fenno suggest is at play here - and we still found only candidates from the president’s party being punished as “incumbents”. Any data out there?
From a grad student in our department who used to work at the FEC:
Anecdotally, of the 300 committees I reviewed at the FEC, only one corporate PAC made independent expenditures over $5000 in the 2008 election and it was the Human Rights Campaign, which is a non-profit corporate PAC. The rest were unions (AFSCME, SEIU, etc.). I think labor unions are going to take much greater advantage of the decision than corporations - especially when it comes to things like GOTV drives and voter contact communications. I think the types of corporations that will get involved with that type of activity are the non-profit interest groups (Human Rights Campaign, NRA, Chamber of Commerce, etc.) as opposed to for-profit corporations like Goldman Sachs and Bank of America.
Were Republicans more likely to be elected governor in states without restrictions on corporate campaigning than in states with restrictions?
Was control of governorship more likely to flip from Democratic in 2002-05 to Republican in 2006-09 where business was not restricted?
Were Republicans better able to hold on to the governorship in 2006-09 in states where business campaigning was not restricted?
Did Republicans have more control of state legislatures prior to 2008 where business campaigning was okay?
Did Republicans have more control of state legislatures after 2008 where business campaigning was okay?
Are states that lean Republican in presidential races (a quasi measure of the ideology/partisan culture of states) especially likely to have no restrictions on business campaigning?
Are state and local taxes less progressive where business campaigning is okay?
Are states where business can campaign likely to have good scores on business climate from the Tax Foundation in 2010?
Did states where business can campaign improve their Tax Foundation business climate score between 2006 and 2010 more than states where they couldn’t?
Are states where business can campaign likely to rank better on business climate from Forbes in 2009?
Are states where business can campaign likely to rank better on the regulatory environment from Forbes in 2009?
Is there a difference between the average Forbes business climate ranking for states with and without restrictions?
Is there a difference between the average Forbes regulatory environment ranking for states with and without restrictions?
Is there a difference between the average Forbes business costs (labor, energy, and taxes) ranking for states with and without restrictions?
Is there a difference between the average Tax Foundation business climate score for states with and without restrictions?
The provisional answer to all of these question is “no.” The analysis comes from political scientist John Coleman at the University of Wisconsin-Madison (pdf). His findings dovetail with those of Lee Drutman and Nick Seabrook. As with their analyses, Coleman’s comes with necessary caveats, which he states in the conclusion. Nevertheless, as he writes:
But these data, as they are, recommend some caution toward the most extreme statements about the effects of the Court’s decision [in Citizens United] on election and policy outcomes.
Arlen Specter shares a birthday with Abraham Lincoln but is much, much younger. Lincoln would be be turning 201 if he were still alive today, and Specter is only 80. Which means, if he succeeds in his reelection campaign, he will be a mere 86 at the end of his next term in the Senate. Fortunately, the job entails no heavy lifting.
This is apropos of Josh’s post below. The New York Times story he cites says:
The poll suggests that both parties face a toxic environment as they prepare for the elections in November. Public disapproval of Congress is at a historic high, and huge numbers of Americans think Congress is beholden to special interests. Fewer than 1 in 10 Americans say members of Congress deserve re-election.
Here is some relevant analysis from Alan Abramowitz. He makes several important points. First, approval of Congress is strongly correlated with presidential approval, even under divided government:
The data show that when the president is more popular, Congress tends to be more popular and when the president is less popular, Congress tends to be less popular. Moreover, this is true even when Congress and the presidency are controlled by different parties.
Why? He posits that opinions of both are related to underlying structural features, such as the economy:
This may indicate that evaluations of Congress are influenced by evaluations of the president or that both are influenced by feelings about the condition of the country and the overall performance of the federal government.
And of course there is this truism, which somewhat undercuts the notion that incumbents are threatened by low congressional approval:
As the noted congressional scholar Richard Fenno has observed, Americans generally love their own congressperson even though they dislike Congress. They tend to see their own Senators and Representatives as the rare good apples in an otherwise rotten barrel.
To speak to Josh’s question about bipartisan anti-incumbent voting, Abramowitz writes:
Discontent with Congress does not lead to a general tendency to kick out incumbents. Occasionally voters do get upset and give the boot to a large number of incumbents—but they almost always take out their dissatisfaction on the members of only one party—the president’s party.
Finally, and most importantly, and please please shout this from the rooftops:
This brings up the most important point about evaluations of Congress. They have very little influence on how Americans vote in congressional elections. When it comes to choosing candidates for Congress, it is opinions of the president’s performance that matter.
The NY Times has released a new poll today which basically says that while voters are not thrilled with President Obama’s job performance, they are even less thrilled with Republicans in Congress. Particularly interesting is that attempts to tar the Republicans as the party of no combined with Obama’s gestures at bipartisanship seem to be working: 62% of respondents see Obama as trying to work with Republicans in Congress, while the same percentage believes that Republicans in Congress are not trying to work with Obama to get things done. This is not good news for Republicans in Congress, as large majorities of respondents think both Obama and Cogressional Republicans should be compromising to get things done (see here and here).
All of this raises the interesting question: what happens in an election when voters are angry at everyone?
When I was first reading the classic political science works on economic voting in US congressional elections, I was struck by the fact that the “incumbent” in all of these studies - as in, incumbents should perform worse when economic conditions are worse - were almost uniformly defined as as congressional candidates from the president’s party. Thus a congressman who had been in office for twenty years would not be considered an incumbent if he or she was from a different party than the president, while his or her first time challenger - who could have never held national office - would be coded as the incumbent, as long as he or she was in the same party as the president. Off the top of my head, I’m not aware of any study of economic voting in the United States which tracks whether poor economic conditions adversely effect incumbents - defined now as the actual office holder - from both political parties. I ask this because I wonder whether it is possible that we could be heading into an election where sitting office holders from both political parties risk the ire of the voters.
The only work on economic voting of which I am aware where the actual incumbent office holders are included in the analysis as such concerns work done in Poland (ungated draft and gated) and Ukraine (gated) by Kazimierz M. Slomczynski, Goldie Shabad, and Jakub Zielinski, although even in these countries they find that the effect of the economy is conditioned on whether an individual legislator is a member of the opposition party or the governing party (although there are also interesting findings in the Poland piece about legislators who attempt to avoid the voters’ wrath by switching to a new party).
So I’m interested in opening this up to discussion. More generally, is it time to revisit what we mean by “incumbents”? As I have noted in my own research, figuring out how to code incumbents in multi-party system, especially unstable ones, is often not intuitively obvious. More specifically, is there any literature out there tests whether voters ever punish actual incumbent office holders in the US regardless of which party they are in? How about any evidence of this occurring outside of the United States? And if such research exists, can it give us any insight into the coming 2010 US mid-term elections? Received wisdom from existing research on economic voting in the US would suggest that the president’s party will be in trouble. But then what are we to make of the NY Times poll showing that voters are even more disappointed with Congressional Republicans than they are with the sitting President?
I’ve been remiss in following up on my promise to discuss political science research on campaign finance. Fortunately, Nick Seabrook, a Ph.D. candidate in political science at the University of Buffalo, sends a back-of-the-envelope analysis that speaks to the potential consequences of the Citizens United decision:
In following the coverage of Citizens United in the mainstream media and blogosphere, two major criticisms of the decision have become apparent: (1) that the massive influx of corporate money into the electoral process will effectively allow corporate sponsorship to dictate election outcomes (let’s call this the McCandidate hypothesis); and (2) that this surge in independent corporate expenditures will disproportionately benefit Republican office seekers (the Fat Cat hypothesis).
While testing these propositions on federal elections will require waiting at least until the outcome of the 2010 midterms, state legislative elections offer an ideal quasi-experiment for analyzing the implications of the decision (24 states ban independent corporate expenditures in state races, 26 do not).
For another project, I put together complete election data for 2002 upper and lower house state legislative districts in 9 states, 4 that ban corporate expenditures (AZ, KY, NY, SD) and 5 that allow them (CA, FL, MD, NV, OR). We can test these two hypotheses by running some simple regressions. Here, state legislative election results are modeled as a function of incumbency, district partisanship, and candidate expenditures.
We can indirectly test the McCandidate hypothesis by splitting the sample between states that ban corporate electioneering and those that do not. If independent corporate expenditures have a significant effect, an observable implication of this would be that these other factors (incumbency, party, and money) should have less of an impact. However, there is no evidence that this is the case – the models explain just under 90% of the variation in the two-party vote in each case, and the coefficients and significance levels are pretty much the same (a Chow test confirms this). Elections appear to work similarly whether a state allows independent corporate expenditures or not, and if corporate money is affecting election results in these states, it is doing so at the margins.
What about the Fat Cat hypothesis? Recombining the data into a single model and introducing a dummy variable for states that allow corporate expenditures, the results are surprising. All else being equal, allowing independent corporate money actually works to the detriment of Republican candidates to the tune of between 2% and 3% of the two-party vote (b=-2.3; s.e.=.6). If we exclude New York from the sample (NY places restrictions on corporate spending but does not ban it outright), the coefficient is statistically insignificant. Similarly, if we run a logit model where the dependent variable is which candidate wins rather than their vote shares (same independent variables), the corporate dummy again fails to reach statistical significance. From these data, there is simply no evidence that Republican candidates are benefiting from corporate campaign involvement at the state level.
Of course, some caveats apply: we can’t really generalize too much from a sample of just 9 states in a single election year, and it’s unclear whether the dynamics of corporate electioneering may be different at the federal level than they are in state elections. Nevertheless, these results suggest that we should be cautious about simply assuming that Citizens United will inevitably produce wholesale changes in the functioning of our electoral system
Which state’s politics are the bigger trainwreck, those in Illinois or New Jersey? I am always happy to give the nod to New Jersey, but Illinois is coming on strong.
Voting geeks and political scientists can sometimes engage in quite vigorous arguments over which system of vote-counting is best. But entertainment value is rarely one of their criteria. It should be - and from my personal experiences as a tallyman in Ireland when I was a teenager, it is one on which PR-STV (proportional representation with a single transferable vote) scores very highly. In PR-STV, the voter votes for the candidates, recording her order of preference (Farrell 1, Tucker 2, … Sides 6 and so on). The votes are then counted. If a candidate reaches the quota with first preference votes, then she is deemed elected, and her surplus votes are distributed to other candidates, according to the second preferences recorded on them. If no candidate reaches the quota with her first preference votes, then the weakest candidate is eliminated, and his votes are distributed to the other candidates according to the second preferences, and so on, until all the vacant seats have been filled.
What makes this system entertaining is that much depends on the order in which candidates are eliminated. If one candidate goes first rather than another with a nearly equal share of votes, this can have significant knock-on repercussions for who gets elected and who doesn’t. The candidates are usually present at the count, and observing them, whey-faced, trying to figure out whether they will lose their jobs or not, as wizened old mountainy men, (who know their end of the constituency from decades of tramping its back-country roads and boreens, and have a good idea of where the second, third and fourth preferences are going) offer predictions based on the way that this or that ballot box seems to be going, is enormously entertaining for heartless teenagers with a predilection for politics.
Which brings me to my proposal. Under the new Oscars system, the ten Best Pictures nominees are chosen under some PR voting system, which may not be PR-STV but is likely to be at worst a closely related cousin. So the Oscar ceremony, rather than cutting to the smug accountant who presents the results as a fait accompli at the end of the ceremony, should instead be cutting back and forth to a vote counting process which would be happening simultaneously, live. Alongside the main Oscar broadcast, one might even have CNN running an “Oscars Voting Special” in which underemployed political pundits could opinionate on who is winning and who is losing, who deserves to win and so on. This would make gripping television for a substantial subsection of the population who is uninterested in Oscar dresses and awards for best fly-grip camera and so on. I can see three possible objections. First - that CNN political pundits don’t actually know anything interesting or useful about the movie industry. This seems to me to be both true and uncompelling - they don’t (with a few exceptions) know anything useful or interesting about politics either. And they can always hire me for a moderately outrageous retainer fee. Second - that the US public would be unlikely to find processes involving complicated math at all entertaining. For rebuttal, I offer you the countless millions of baseball statistics bores to be found in this fine country. Third - that this will lead inevitably to family friction over whether we should watch the Oscars-proper ™ or the Oscars Special Vote Count edition ™ from the Finest Names in News (or whatever the slogan is this week). This seems to me to be the most plausible of the objections (but since I’m the only person in our household who really knows how to use the remote-controls, I’m not particularly worried about it meself).
Political scientists believe unanimously that economic conditions play an important role in shaping public opinion. I don’t think you could find one who would suggest that a president one year into his term facing double-digit unemployment could avoid a significant drop in popularity.
One complaint I have with the mainstream media is its habit of ignoring such structural factors and explaining changes in public opinion almost solely as the outcome of ideological positioning by candidates and elected officials. In defense of the MSM, they do this in a bipartisan way. Whenever a party suffers electoral misfortune, the media will attribute it to a failure to heed to wishes of the center, coupled with demands that the losing party purge itself of ideological sin and embrace the moderate center.
The rest of the piece is an argument with a former aide to Karl Rove. For the purposes of this blog, I’m just happy to see political science get its due. Chait also notes that he did not over-intepret the 2008 election as a sweeping mandate, for which he deserves credit.
[Hat tip to Brendan Nyhan.]
This one is making the rounds:
From the Orleans Parish coroner’s race.
[Hat tip to Eric Lawrence.]
In our continuing series of election reports, we are pleased to have Paul Staniland, a Pre-doctoral Research Fellow at the Program on Order, Conflict, and Violence at Yale University’s MacMillan Center for International and Area Studies present the following post-election analysis of the 2010 Sri Lankan presidential election:
Sri Lanka’s presidential election, the first since the defeat of the Liberation Tigers of Tamil Eelam (LTTE) in May 2009, further entrenched the incumbent government of Mahinda Rajapaksa. Rajapaksa’s major opponent, General (ret.) Sarath Fonseka, was the chief of the army in the victorious struggle against the Tigers but had turned against Rajapaksa when he was stripped of real power after the war. The campaign was ugly and sometimes violent, marked by charges of war crimes, coups d’etat, vote rigging, treason, and corruption. Approximately 800 violent incidents were reported during the course of the campaign. The aftermath of the election included government charges that Fonseka had planned a coup against his government, while Fonseka is reportedly now considering going into exile.
Rajapaksa and his Sri Lanka Freedom Party (SLFP)-led coalition ultimately took 57% of the vote while Fonseka, the common opposition candidate, won 40%. Despite expectations of a tight contest, Rajapaksa dominated in ethnic Sinhalese areas and thus overcame Fonseka’s higher levels of support in cities and ethnic Tamil and Muslim regions. This map shows the demographic breakdown: Rajapaksa won the Sinhalese-majority southern and western parts of the country, while Fonseka did better in less-populous Tamil and Muslim areas in the north and east. What can we draw from both the process and the outcome of the election?
First, the government and SLFP used state resources and patronage to get out the vote. Fonseka was a threat because he could undermine Rajapaksa’s credit for defeating the LTTE. In response, the Rajapaksa campaign fully exploited the advantages of incumbency, in sharp contrast to a shambolic Fonseka campaign. Organizationally, the SLFP’s patronage machine was highly effective at mobilizing rural voters and at pulling away defectors from the opposition. Crucially, Rajapaksa’s campaign skillfully exploited its position in power. Rajapaksa was portrayed in the state media as the prime reason for the destruction of the LTTE while Fonseka’s gaffes on the campaign trail were ideal fodder for criticism and often-dubious accusations. Rajapaksa offered a clear Sinhalese nationalist political line that was linked to economic development promises. The election commissioner publicly protested these irregularities, but there is no evidence of massive vote rigging.
Continue reading "2010 Sri Lankan Presidential Election: Post-Election Report" »
…we found that individuals’ political affiliations could be accurately discerned from their faces…perceivers were able to accurately distinguish whether U.S. Senate candidates were either Democrats or Republicans based on photos of their faces…these effects extended to Democrat and Republican college students, based on their senior yearbook photos…these judgments were related to differences in perceived traits among the Democrat and Republican faces. Republicans were perceived as more powerful than Democrats. Moreover, as individual targets were perceived to be more powerful, they were more likely to be perceived as Republicans by others. Similarly, as individual targets were perceived to be warmer, they were more likely to be perceived as Democrats.
The study, by Nicholas O. Rule and Nalini Ambady, is here (via). If I am reading their results correctly, correct perceptions of U.S. Senate candidates were better than chance, but not by much (57% correct). In the yearbook study, the percentage correct was higher, (62%). The third set of findings — about power and warmth — dovetails with Danny Hayes’ work on party ownership of traits (pdf), which the authors cite.
I think all of this is quite interesting, although we shouldn’t push the conclusions too far. It’s clear that party identification is somewhat evident from appearance, but appearance is far from a definitive clue.
Charles Stewart of MIT crunches some data and reports:
Statewide, the big Brown vote shift was concentrated in the Republican parts of the state. Within Boston, the shifts came in the Republican and white parts of the city.
See the above graph, which focuses on Boston, tabulates the Brown vote relative to the McCain vote in each precinct, and then plots that against the percent minority.
And then this on turnout:
Statewide, the big turnout shift was concentrated in the Republican and wealthy parts of the state. Within Boston, turnout held firmest in the Republican and white parts of the city.
His analysis throws cold water on the “working class revolt” explanation of the MA election:
The geographical pattern of the vote shifts, showing that the shifts occurred in the Boston suburbs and exurbs, argues against the notion that this was largely a matter of angry working-class Democrats abandoning their party. More likely, it was borderline independents punishing the incumbent party for poor economic performance, along with Republicans who had no interest in George W. Bush coming home to support a national Republican who could string sentences together to make paragraphs.
More is here (pdf).
Lee Drutman, last seen here guest-blogging on corporate lobbying, sends along this quick analysis:
Will the Citizens United decision help Republicans more, since they’ve traditionally raised more money from corporations, who can now spend unlimited amounts?
As the New York Times notes, 24 states ban or restrict corporate contributions, and 26 allow unlimited contributions. So do Republicans do better in those states with unlimited contributions?
One way to assess this is to run some very simple regressions. We would expect that the Democrats’ share of seats in state legislatures in 2009 would be strongly predicted by Obama’s vote share in 2008. It is.
But what if we added a dummy variable for whether a state bans or restricts campaign contributions into this regression? In state houses, having a ban or restriction improves the Democrat share of seats by three percentage points on average, but the result is not statistically significant (b=.03; se=.03). In state senates, it has no effect (b=.01; se=.04).
This provides a cautionary note to fears that unlimited contributions will inevitably create a partisan bias in election outcomes. See also this piece by David Kirkpatrick in Sunday’s NY Times, which nicely digests the challenges in proving the consequences of campaign contributions.
But Lee notes:
On the other hand, there may be an effect on the policy positions of different parties, particularly the Democrats. This will definitely be worth watching.
More to come on this decision — and political science research on campaign finance — this week.
In our continuing series of election reports, we are very pleased to have Lucan Way of the University of Toronto provide the following commentary on last weekend’s first round of the Ukrainian presidential election:
The first round of Ukraine’s Presidential election, which reunited all of the major participants of the Orange Revolution save Leonid Kuchma, revealed major changes but some important similarities in the political system that existed in 2004.
The most important and encouraging change has been the introduction of functioning democratic institutions. Relative to 2004, media was much more open and diverse (if highly partisan) and electoral fraud was largely absent. Perhaps even more remarkably, the first round suggests that the use of state or “administrative resources”by incumbents no longer yields the same benefits it once did. Most strikingly, access to state resources by President Viktor Yushchenko translated into just 5 percent of the vote – a result that may be the worst performance by an incumbent in modern democratic history. Prime Minister Yulia Tymoshenko, who had backed Yuschenko in 2004 but now ran against him, did much better with 25% — enough to get into the second round with Viktor Yanukovych. However, she does not seem to have benefitted significantly from state access. Tymoshenko arguably bet heavily on administrative resources when she decided to become Prime Minister in late 2007 at which point there were already signs of an impending economic downturn in Ukraine. Given, her overwhelming desire to become President, Tymoshenko might have chosen to stay in opposition and let others take the blame for Ukraine’s 14 percent economic decline in 2009. Tymoshenko’s decision makes sense in light of the rules of the game that had dominated under Kuchma. Until 2004, access to the government had been a sine qua non for political viability. A government position was about the only way candidates could gain the necessary patronage, organizational resources and media attention necessary to mount a serious campaign.
However, in what is a very good sign for Ukrainian democracy, Tymoshenko appears to have suffered far more as an incumbent who had to take the blame for Ukraine’s dismal economic performance than she benefitted as a government official able to use of state resources to bolster her support. A preliminary picture of the relative benefits of administrative resources emerges from a comparison of her performance last Sunday with her party’s performance in 2007 parliamentary elections and Yushchenko’s performance in the first round of the Presidential election in 2004 when he suffered significant disadvantages. Overall, she got 5 percent less than her party did in 2007 and about 15 percent less than Yushchenko did in 2004. Administrative resources may have contributed to slightly improved performance in several eastern provinces, where Tymoshenko did better than in previous years. But, these regions were still overwhelmingly dominated by Yanukovych as in both 2007 and 2004.
Continue reading "2010 Ukrainian Presidential Election: Round 1" »
Yesterday, I cautioned against interpretations of the MA election. Brendan Nyhan has a nice post on that subject today. Then someone sent me John Judis’ piece in the New Republic, “Does He Feel Your Pain”? Judis — previously complimented here for his effort to engage political science — provides a thoughtful interpretation, but one that doesn’t convince me.
Here, I think, are the key bits from Judis:
Where he has lost ground—and where the Democrats have lost ground—is primarily among white working and middle class voters and senior citizens…
…If you look at national polls, Obama has suffered the greatest loss of approval among exactly the same groups. In the Pew polls, Obama suffered a drastic drop in support in the $30,000-$75,000 income group, from 63 percent to 17 percent approval in February 2009, to 53 percent to 35 percent disapproval in the January 14 poll. Among respondents over sixty-five years old, he went from 60 percent to 17 percent approval to 54 percent to 31 percent disapproval. In its January 2010 poll, Pew has a breakdown by race that is even more disturbing. Whites with some or no college—a rough designation for working-class whites—disapprove of Obama’s presidency by 54 percent to 36 percent.
…These two groups of voters have not viewed Obama’s presidency in a fundamentally different way from many other voters, but they, and particularly working-class whites, have been the prime source of a populist anger against the Obama administration. They have perceived Obama as robbing Peter to pay Paul—or more concretely, taking benefits from and imposing higher taxes on them in order to provide greater income and benefits to others.
Here are my thoughts. First, trends in income groups or age groups don’t mean anything unless we can show that these trends don’t simply derive from the fact that the groups differ in their underlying partisanship. People think about candidates and leaders largely as partisans, and 90% of Americans identify with or lean toward a party (see here). I would always, always start with trends among partisans, and then show if there are differences within partisan group based on income, etc.
Second, we should not assume that trends in the “working class” mean that voters are thinking about politics in terms of class (i.e., populism). For the most part — and political science has 40 years of findings in this regard — people don’t think about politics in terms of their individual self-interest. So I tend to doubt this claim:
The middle class and senior citizens see it as a program that taxes and takes benefits away from them in order to help those without insurance—the out groups—and to enrich the insurance companies themselves.
People may agree with such sentiments when pollsters prompt them to do so, but that doesn’t mean they have arrived at these sentiments on their own. These may only be rationalizations for beliefs that derive from much simpler reasoning processes (i.e., “I oppose health care because I’m a Republican and I don’t like Obama”). See Lee’s old post on rationalization.
Third, there is an implication in John’s article that policy trumps performance — that is, that voters are reacting to the Obama administration’s policy proposals rather than to the weak economy. But performance tend to matter more. Judis’s graph of the unemployment rate against Obama’s approval shows the apparent effect of performance. But then he makes this claim:
What I found in Obama’s case is that at the beginning of last fall, when Washington began debating his health care plan in earnest, his level of disapproval began to exceed the rise in the unemployment rate.
Click the graph to see what he’s talking about. I’m far from persuaded that the extremely mild divergence between Obama’s approval (declining) and the unemployment rate (steady) tells us anything. Political science suggests that is difficult for voters to cast votes based on policy.
And then, of course, there is the problem of figuring out why Massachusetts voters in particular made the choices they did — the subject of my first post. The Fabrizio exit poll is no help here, because we cannot we trust the stated reasons they gave (and, of course, we have no details as to the poll’s methodology anyway).
Ultimately, we can content ourselves with back-of-the-envelope math like Nate Silver’s. Really, we can’t do any more. But fundamentally, we know very little about why Massachusetts voters made the choices they did.
Yes, I know political science is a buzzkill. And no one gets paid to say “We don’t and can’t know.” But that’s what we should be saying.
Just as John predicted, now that Scott Brown has won the MA special election, we will be subject to a relentless barrage of explanations as to how the US political wold has been turned upside down in a matter of days.
As a public service announcement, therefore, we at the Monkey Cage once again won’t try to interpret what the election means, but instead bring you the following graph that displays the actual effect of the MA Senate election on the balance of power between the two political parties in Washington and across the nation’s statehouses.
Yes, the Democrats’ have lost their filibuster proof majority in the Senate.* And yes, I’m aware of the fact that there could be a ripple effect whereby more moderate Dems move to the right out of fear of losing the next election. But as to the actual balance of power in Washington today, the graph tells the story.
The 1993 Canadian Elections this was not.
______________
Let me lightly revise and expand my post from just before the NJ and VA gubernatorial elections in 2009.
Be prepared. You will see the pundits hold forth on what the Massachusetts special election “means” for the Republicans, the Democrats, Obama, Curt Schilling, and Lord knows what else.
The correct answer will be: I don’t know. This election could turn on national politics. It could turn on local factors and issues. It could turn on both, although we won’t know how important each was. Who wins can’t tell us why they won. And there won’t be an exit poll, which would give us only the most tentative interpretation anyway.
This is why I’ve struggled to write something lucid about this campaign. Yes, there’s been a striking trend in Brown’s favor. But we have no conclusive evidence as to what is driving this trend, just as we will not have any evidence about why he or Coakley won. My rank speculation is no better than anyone else’s.
The problem with speculating about the meaning of elections is that such speculation often devolves into a discussion of “optics,” “narrative” and other woolly terms. But the “narrative” only exists when commentators construct it. It’s not some independent entity that is magically affected by election outcomes. To talk about how the MA election affects some “narrative” is like saying, “The MA election affects how I am talking about the effect of the MA election.” Nor is it clear if changes in narrative have tangible consequences for things that actually matter — like presidential approval, upcoming elections, etc.
There is only one truly significant potential consequence of the MA race, and that’s what having 59 as opposed to 60 Senators will do to the prospects for certain pieces of legislation, notably health care reform.
Boris Shor writes:
A special Senate election is being held next Tuesday in Massachusetts to finish Edward Kennedy’s term. The candidates are Martha Coakley (D), and State Senator Scott P. Brown ®. . . . recent polls show Brown matching or even exceeding Coakley’s electoral support, in one of the most liberal states in the entire country . . .Brown is attracting very positive national and state Republican and conservative attention. On the other hand, State Assemblywoman Dede Scozzafava attracted very negative attention from conservatives in her special election campaign for the 23rd Congressional District of New York.
Brown is actually a liberal Republican who is to be found to the left of Dede Scozzafava! So why, then, the enthusiasm gap in support for the two? This post documents this assertion, and then answers this puzzle.
How liberal or conservative is [Scott Brown]? We have evidence from multiple sources. The Boston Globe, in its editorial endorsing Coakley, called Brown “in the mode of the national GOP.” Liberal bloggers have tried to tie him to the Tea Party movement, making him out to be very conservative. Chuck Shumer called him “far-right.”
In 2002, he filled out a Votesmart survey on his policy positions in the context of running for the State Senate. Looking through the answers doesn’t reveal too much beyond that he is a pro-choice, anti-tax, pro-gun Republican. His interest group ratings are all over the map. Business and gun rights groups typically rate him very highly, labor and and environmental groups have rated him both middling and high over time. The teacher’s union rated him low in 2001, and high in 2005.
All in all, a very confusing assessment, and quite imprecise. So how do we compare Brown to other state legislators, or more generally to other politicians across the country?
Zabranjeni spot “Stop crvenoj Hrvatskoj” from Milan Bandić on Vimeo.
The ad is from the recently concluded Croatian Presidential Election, where Ivo Josipovic (an academic(!) and composer) defeated Milan Bandic. The ad was launched by Bandic against Josipovic in an attempt to mimic the mid 1990s playbook of Vaclav Klaus and Boris Yeltsin by claiming that a victory for his opponent would return the nation to Communism. According to Robert Mackey in the New York Times, this accusation was based on the fact that Josipovic had promised to turn the map of Croatia red because the press shades areas of the country won by Josipovic’s party - the Social Democratic Party - in red. The claim was made all the more ironic by the fact that Bandic himself had been a member of the Social Democratic Party up until November of last year when he was expelled for - wait for it - deciding to run for president.
[Hat tip to Andrew Therriault and The Lede - NY Times News Blog]
The short answer is: maybe not. In recent elections, during the period from 1992-2004, turnout among Evangelical Christians increased. Conventional explanations emphasize campaign strategy, especially those that investigated the Bush campaigns of 2000 and 2004. For example:
Mr. Rove’s relentless focus on turning out more Republican voters, many of them evangelical Christians, was the critical factor in Mr. Bush’s victory, Republicans said.
In fact, there is new evidence that the mobilization of Evangelicals may have had more to do with demographics than campaign strategy. This is the conclusion of a newly published paper by Ryan Claassen and Andrew Povtak. They write:
We find that, contrary to conventional wisdom, the increase in Evangelical turnout appears to have been driven by social and demographic changes among Evangelicals rather than by a political strategy. In fact, controlling for social and demographic changes, we find more impressive turnout gains among other groups, such as black Protestants and the nonreligious.
And this:
In light of changes among Evangelicals that have made them more prosperous and better educated as a group, one would expect increased participation even in the absence of moral issues on the national agenda, born again candidates, and voter guides.
Moreover, it appears that an equally, if not more important story, involves increases in turnout among black Protestants and the non-religious:
…increased turnout among secular voters and black Protestants has received far less attention recently (no doubt because the Democratic party did not emerge the victor in the 2000 and 2004 presidential elections), the raw increases in turnout over time are even more impressive than those of Evangelicals and they appear to be less a function of sociodemographic changes in the groups.
The main caveat is that this account focuses only on turnout, not on the changing partisan loyalties of Evangelicals. Nevertheless, I think this piece poses a significant challenge to conventional wisdom.
Benjamin Kay writes:
I wonder if you saw Bruce Reed’s article The Year of Running Dangerously — In a tough economy, incumbency is the one job nobody wants. about the recent flurry of retirement announcements in the Senate but also less well publicized ones in the House. My understanding is that there is a known effect on retirement from census driven redistricting. We also happen to be in a census year but I haven’t read any journalists discussing that as as a factor. Do you have any insight into the relative explanatory decomposition of partisan politics, redistricting related concerns, and simple economy driven unpopularity in these retirement decision?
My reply: Retirement rates definitely go up in redistricting years (see, for example, figure 3 here), but that would be 2012, not 2010, I believe. The census is this year, but I don’t think they’re planning to redraw the district lines in time for the 2010 elections.
In any case, I imagine that somebody has studied retirement rates, to see if they tend to go up in marginal seats in bad economies. Overall, retirement rates are about the same in marginal seats as any other (at least, that’s what Gary and I found when we looked at it in the late 1980s), but I could imagine that things vary by year. The data are out there, so I imagine somebody has studied this.
P.S. I’ve never understood why anybody would want to retire from a comfortable white-collar job. But now, spending a year on sabbatical and relaxing most of the time, I can really understand the appeal of retirement.
One thing I learned in econ class in 11th grade was that government policy should be counter-cyclical (spending more in recessions and cutting back in boom times), but that there’s a lot of pressure to be pro-cyclical, which will tend to exacerbate business cycles. (Except I suppose they didn’t say “exacerbate” in 11th grade.) At a personal level, too, it’s natural to spend more when we have more and cut back when we aren’t doing so well. Every now and then you hear about a “rainy day fund” but my general impression is that these are never big enough to counter the business cycle.
Political parties seem to apply a similar pro-cyclical behavior in their congressional election campaigns.
Consider 2008. As expected, it was a good year for the Democrats, and so it was a logical time for them, as a party, to make some investments in new, young candidates. 2008 was the time they should’ve encourage lots of their incumbents to retire, because in that year they could win a lot of these districts without needing the incumbency advantage (estimated to be about 10% of the vote, i.e., enough to take you from 50% to 60%). Conversely, 2008 was the time for the Republican Party to hold on to what it had, and to keep all their incumbents in, trying to hold out until 2010 when the pendulum might swing back in their favor. But we didn’t see that—actually, something like 30 Republican House members retired in 2008. Republicans retiring, Democrats sticking around—that was a recipe for big Democratic gains. But then in 2010, or 2014, or whatever year it is when the Democrats get wiped out—then a bunch of their incumbents will probably retire, and boy will the Democrats wished they had put in younger incumbents back in 2008 when they had a chance!
This election cycle we’ve been seeing this happen from the other direction:Democratic Senators Christopher Dodd of Connecticut and Byron Dorgan of North Dakota announced this week that they would retire from their long-held seats rather than face uphill battles in this year’s midterm elections. Several House Democrats from conservative districts have said they will step down. The Democratic governor of Colorado, Bill Ritter, has made the same decision.The flip side of this was in 2008, when 84-year-old Frank Lautenberg ran for reelection in New Jersey. That was a Democratic year when the Democrats might’ve done well with just about anybody. (Or maybe not; I don’t really follow New Jersey politics and am just extrapolating from national polls.) When 2014 rolls around, they’re going to need to find someone new, and at that point they might wish they had an incumbent already in the slot.
What makes sense for the individual officeholder—stay in when you think you’ll have an easy win, and wait to quit when the going is getting tough—isn’t so helpful for the national party.
Dan Goldstein points to a draft article by Andreas Graefe and J. Scott Armstrong:
Continue reading "Predicting elections using the "most important issue" question" »
We are again pleased to have Gregory Weeks provide post-election analysis of the first round of the 2009 Chilean presidential and legislative elections:
In Chile, there will be a second round on January 17. Sebastián Piñera received 44%, to Eduardo Frei’s 29.6%, with MEO at 20% and Jorge Arrate with 6.2%.
MEO announced that he would not endorse either candidate, though made it clear he preferred that Piñera did not win. Jorge Arrate indicated he would support Frei, but with conditions.
Neither coalition gained a majority in the Chamber of Deputies. The Concertación won 54 seats, in addition to the three seats of the Communist Party, with which it ran in alliance. The right now has 58 of the 120 seats. A similar dynamic holds in the Senate. No coalition has a majority: the right now has 17 seats and the Concertación holds 19. However, Alejandro Navarro from the MAS Party leans clearly toward the Concertación (though he campaigned with MEO).
A few quick thoughts:
First, the campaign will be nasty, brutish, and short. Piñera’s lead is significant, but there will be a fight for MEO’s supporters. Piñera and Frei almost immediately insulted each other, while MEO insulted both.
Second, the Concertación is down but not yet out. The coalition was sagging under its own weight after 19 years in the presidency, but the center-left split at the presidential level did not damage its legislative results (see Matthew Shugart for a pre-election discussion of this point). The binomial system still provides a strong incentive to remain in a coalition.
Third, change is in the air no matter who wins the presidency. In the Chamber of Deputies over one-third (45 of 120) of the winners were elected for the first time, with diverse platforms. Many of those new faces are from the right. The conservative Independent Democratic Union had the most at 13, followed by National Renovation with 8.
David points me to this news article by Dennis Cauchon, which begins:
Continue reading ""In major flip, House Dems now represent richest regions"" »
Over at The Hill, Markos Moulitsas of The Daily Kos writes of a Brutal Poll for Dems. He reports on a Research 2000 poll suggesting that Republican are much more likely to vote in the 2010 congressional elections. He notes that:
Among Republican respondents, 81 percent said they were definitely or probably going to vote, versus only 14 percent who were definitely or not likely to do so. Among independent voters, it was 65-23. Among Democrats? A woeful 56-40: Two out of every five Democrats are currently unlikely to vote.
A look at key Democratic constituencies shows how demoralized the party’s base currently is. Among African-Americans, just 34 percent are likely to vote, versus 54 percent unlikely to do so. Republican-leaning white voters clocked in at 66-29. Only 41 percent of 18- to 29-year-olds, a key constituency for Democrats in both 2006 and 2008, are likely to vote, compared to 49 percent likely to sit things out.
The result of these trends?
Democrats will suffer at the ballot box.
It is all well and good to report these numbers, but without a reference point to compare them to, I don’t think it can really tell us anything about what is likely to happen in 2010. To put this another way, if you had asked me off the top of my head whether 18-29 year olds were more likely to vote in any off-year congressional election than “Republican leaning white voters”, I’d have picked the Republican leaning white voters.
Or another way of thinking about this: according to the United States Election Project, turnout in the 2006 off-year congressional elections for the country as a whole was 40.3%. If the Democrats, who still outnumber Republicans in terms of party-identifiers, can actually get close to 60% of their supporters to turn out in a mid-term election, mightn’t that be a particularly good outcome? Again, I’m not sure, but that’s because I don’t know what kind of turnout we need to see among Democrats in order for Democrats to perform well in a mid-term election.
I don’t doubt the central premise of the article - that there is an enthusiasm gap between supporters of the Republicans and Democrats at this point in time - but we need some reference points to get a sense of how these likelihood to vote numbers are likely to translate into actual election results. Perhaps the most useful would be participation rates in 2006 - can any readers of The Monkey Cage help out by providing commensurate numbers from that race to the ones Moulitsas has cited in the article? If so, please add them in the comments section, and I’d be happy to update this post later.
In response to my note on the limited ideological constraints faced by legislators running for reelection, Alan Abramowitz writes:
I [Abramowitz] agree—although they probably have less leeway now than in the past due to growing pressure toward ideological conformity within parties, especially GOP. But one thing that struck me as very interesting in your graph is that it looks like the advantage of a moderate voting record is considerably smaller now than it used to be, down from over 4 percentage points in the 1980s to maybe 1.5 points on average now. It suggests to me that the electorate has become increasingly partisan and that fewer voters are going to defect to an incumbent from the opposing party regardless of voting record. This could reflect more concern among voters with party control of Congress itself. Along these lines, one thing I’ve found in the NES data is a growing correlation between presidential job evaluations and voting for both House and Senate candidates over time.
My reply: Yes, that makes sense. The trend is suggestive although (as you can see from the error bars) not statistically significant. Recently I have not had my thoughts organized enough to write any articles on this stuff, but it feels good to at least post these fragments for others to chew on.
My colleague Pat Egan sends along the following observation about parallels between the 1982 midterm elections and the coming 2010 midterm elections:
It’s becoming increasingly clear that the economy — or at least the parts of the economy that matter with regard to elections — will not turn around to any significant degree by Election Day 2010. Unemployment is forecast to remain in the double digits; personal income growth is expected to be anemic; and even real GDP growth (the one bright spot) is projected to be only decent, not dazzling. In other words, in some ways we’re looking at a potential replay of the 1982 midterms, which took place two years after a charismatic president (Ronald Reagan) swept dozens of new members of his party into office. Plagued by the bad economy of the early 1980s (although not as terrible as the current one), Republicans lost 27 seats in the House. The parallels continue: Gallup currently has Obama’s approval rating at 52 percent; Reagan’s November 1981 rating was also 52.
The big difference between then and now? Party identification. In the 1980s, a strong plurality of Americans continued to identify as Democrats despite the Reagan Revolution. Today, the out-party enjoys no such advantage: recent surveys find the share of Americans affiliating with the Republican Party to be at or near historic lows as the number of Independents surges. A lesson that Obama and the Democrats might take from all of this is that— despite the town hallers, the birthers, and the tea partiers—they would do well to continue to carve out a distinctive approach to domestic policy that can be contrasted in 2010 to the lack of solutions proposed by an unpopular G.O.P.
In response to the Chen and Rodden paper, Michael McDonald writes:
I [McDonald] cover the issue of the application of redistricting criteria in the Midwest Mapping Project report available here.
I [McDonald] would quibble with the claim that compact and contiguous districts will “always” generate substantial pro-Republican bias. There are a number of caveats that cannot be addressed by an analysis of Florida’s congressional districts alone.
- It is well-known that single-member district systems tend to provide a seat bonus to the majority party. This bonus may be offset by the inefficiencies of Democrats’ residential patterns, thus producing a plan that has actually little or no bias. This may be true in a heavily Democratic state like Illinois (for the congressional districts), but may
not be true for a battleground state like Florida.
- These biases are most prevalent where the size of the districts to be drawn are smaller than the urban areas where Democrats are concentrated. It is relatively easy to unpack the Twin Cities to draw a balanced congressional plan for Minnesota’s eight congressional districts, but is it difficult to do so for the 134 state legislative lower chamber districts.
- The residential patterns in the state matter, too. Some states like Wisconsin tend to have a relatively balanced mix of Democrats and Republicans (here, in much of the southern part of the state). Thus, congressional districts in Wisconsin following traditional redistricting principles tend to have relatively little partisan bias.
- There is an issue of partisan symmetry to consider, as well. While there may be pro-Republican bias, generally, more Republican districts tend to be competitive. So, Republicans gain some bias in exchange for increased uncertainty. In a large electoral swing such as 2006 or 2008 towards the Democrats, the result can be that Democrats unseat many Republicans in competitive districts, giving Democrats a super-majority in, for example, Minnesota’s Senate even though the plan appears to have a pro-Republican bias.
Generally, I agree that compactness and other traditional redistricting principles tend to work against Democrats, especially in a state like Florida where Democrats are concentrated in a corner of the state. It is for this reason that I have cautioned the proponents of Florida’s redistricting ballot initiative that they may not fully realize their vision of partisan fairness through the application of traditional redistricting principles.
However, these caveats suggest that the pro-Republican bias is not universally true. The truth is more complicated and depends on the jurisdiction, the number of districts to be drawn, and the political circumstances. I recommend that careful consideration be given to the criteria one wishes to adopt for a given jurisdiction. If partisan fairness
is a goal, then as I have published elsewhere, then that criteria should be explicitly stated in the criteria, rather than hoping it will be achieved through seemingly-apolitical means.
And here’s a reference: Michael P. McDonald. 2007. “Regulating Redistricting.” PS: Political Science and Politics 40(4): 675-9.
Jonathan Rodden and Jowei Chen sent me this article:
When one of the major parties in the United States wins a substantially larger share of the seats than its vote share would seem to warrant, the conventional explanation lies in manipulation of maps by the party that controls the redistricting process. Yet this paper uses a unique data set from Florida to demonstrate a common mechanism through which substantial partisan bias can emerge purely from residential patterns. When partisan preferences are spatially dependent and partisanship is highly correlated with population density, any districting scheme that generates relatively compact, contiguous districts will tend to produce bias against the urban party. In order to demonstrate this empirically, we apply automated districting algorithms driven solely by compactness and contiguity parameters, building winner-take-all districts out of the precinct-level results of the tied Florida presidential election of 2000. The simulation results demonstrate that with 50 percent of the votes statewide, the Republicans can expect to win around 59 percent of the seats without any “intentional” gerrymandering. This is because urban districts tend to be homogeneous and Democratic while suburban and rural districts tend to be moderately Republican. Thus in Florida and other states where Democrats are highly concentrated in cities, the seemingly apolitical practice of requiring compact, contiguous districts will produce systematic pro-Republican electoral bias.
My thoughts:
Matthew Yglesias remarks that, when staking out positions, congressmembers are not very strongly constrained by the ideologies of their constituents.
Wow, that was a lot of big words. What I meant to say was: Congressmembers and Senators can pretty much vote how they want on most issues, whatever their constituents happen to believe. Not always, of course, but a representative can take a much more liberal or conservative line than the voters in his or her district or state, and still do fine when election time comes.
Yglesias gives some examples from the U.S. Senate, and I just wanted to back him up by citing some research from the House of Representatives.
First, here’s a graph (based on research with Jonathan Katz) showing that, when running for reelection, it helps for a congressmember to be a moderate—but not by much:
Being a moderate is worth about 2% of the vote in a congressional election: it ain’t nuthin, but it certainly is not a paramount concern for most representatives.
Continue reading "Politicians have a lot of leeway in how they vote" »
Seth Masket crunches the numbers and finds that economic growth is correlated with seat loss, but unemployment is not. He has more graphs and some commentary.
I’ve expressed some skepticism about the role of candidate appearance before. Now comes this new research from Matthew Atkinson, Ryan Enos, and Seth Hill:
We estimate the effect of candidate appearance on vote choice in congressional elections using an original survey instrument. Based on estimates of the facial competence of 972 congressional candidates, we show that in more competitive races the out-party tends to run candidates with higher quality faces.We estimate the direct effect of face on vote choice when controlling for the competitiveness of the contest and for individual partisanship. Combining survey data with our facial quality scores and a measure of contest competitiveness, we find a face quality effect for Senate challengers of about 4 points for independent voters and 1–3 points for partisans. While we estimate face effects that could potentially matter in close elections, we find that the challenging candidate’s face is never the difference between a challenger and incumbent victory in all 99 Senate elections in our study.
This takes the study of candidate appearance out of the laboratory and into the field, where selection effects (better looking candidates in competitive races) and confounding factors (competitiveness, party identification) cannot be ignored. The results are very modest effects that are not sufficient to change outcomes. Candidate appearance is less important than the hype often suggests.
As a follow up to Tuesday’s election, Politico’s Arena is discussing the question of whether Tea Party Conservatism is a help or hazard for Republicans seeking a return to power. Here’s my answer; feel free to add your own in the comments section below:
It’s a hazard. In a two-party system, the party that can position itself closest to the median voter is more likely to win elections. And in the United States, the median voter is a moderate (see figure below). While their may be particular districts and states where the median voter is more to the right or more to the left, at the end of the day the reputation of the national party still matters. Due to the nature of the primary system in this country – lower turnout, more participation by more ideologically extreme party members – the trick to winning elections for Republicans (Democrats) is usually to move to the right (left) in the primaries, and then back to the center in the general election. If the effect of Tea Party conservatism is to make Republican candidates need to move even farther to the right in primaries, then it is – on the aggregate – going to make it harder for the Republicans to win general elections, with NY-23 as Exhibit #1. The way the Democrats came back from the wilderness in the 1990s was to move to the center; it’s hard to imagine Republicans can come back by moving further to the right.
Greg Marx had two nice posts at the Columbia Journalism Review, in which he shows just how hard it is for reporters to resist interpreting these off-year elections as harbingers of Something Significant.
To buttress that point, via Brendan Nyhan comes Alan Abramowitz’s analysis of the predictive power of the VA and NJ elections, “What Happens in VA and NJ, Stays in VA and NJ”:
However, the results of the previous year’s gubernatorial elections in Virginia and New Jersey did not predict the results of the midterm elections. Not only is the estimated coefficient for the Virginia/New Jersey election variable small and statistically insignificant, but it is in the wrong direction: the better Republicans did in Virginia and New Jersey, the worse they did in the subsequent midterm election.
I also asked David Karol to comment on the race in NY-23. Like Hans Noel, he is an author of The Party Decides and also a forthcoming book, Party Position Change in American Politics.
David writes:
Clearly, the Republican leaders in New York State bungled this race badly. They should have nominated someone with moderate leanings, but more conservative than Scozzafava, perhaps someone who was pro-choice, but not pro same-sex marriage, for example. In New York major party politics is traditionally more corrupt, pork-oriented and less ideological than in some other states, so I guess the county leaders were tone-deaf regarding conservative sentiment in this case.
Contrary to my co-author and friend Hans however, I think that by undermining the GOP nominee the right-wingers certainly were bolting and did so first. If factions in a party do not respect their own nomination process that is a big problem for it. Alan Ware has shown that part of the reason party leaders supported the creation of the Australian ballot and, some years later, the establishment of primaries was that they needed a nomination process that would appear legitimate to losers in order to minimize splits. Part of being a successful party is accepting that you fight your fight inside the organization and then respect its verdict. I can see why various conservatives thought it was in their interest to bolt, even if it meant losing this seat to a Democrat, but that’s clearly what they were doing.
I can also see how conservatives (and Hans) would think that Scozzafava’s endorsement of Owens vindicates their attack on her in retrospect, but for politicians politics is often quite personal. Scozzafava is understandably embittered and her endorsement does not show that she would not have been a real Republican, albeit a moderate, had she been elected to the House. Her decade of service in Albany would seem to be more probative than her lashing out in anger following the extraordinary treatment she received. More speculatively, it seems less likely that she would have endorsed the Democrat had she simply been defeated in a GOP primary.
Relatedly, the fact that Scozzafava was nominated, not in a primary, but by local party leaders, while in keeping with the New York law for special elections, probably undermined her legitimacy once she was scrutinized. This is only one case, but as far as it goes it tends to undermine claims that party leaders would choose candidates more wisely than ordinary primary voters do.
Finally, the fact that the party label cut so little ice with voters has two implications: it supports claims that the scope of “party revival” has been overestimated at least in terms of affect and allegiance and suggests that party identification, which in many formulations is distinct from ideology, is probably not all that it is cracked up to be in the literature.
Vacancies in the House of Representatives are filled using special elections. These elections occur off the usual American electoral cycle, and the results of these elections are routinely portrayed by the American mass media as indications of what to expect in the next general election. We examine the predictive power of the results of special elections on the general election outcomes for the U.S. House of Representatives from 1900-2008 and find that those special elections that result in a change in partisan control do have predictive power for the general election.
That is from a forthcoming paper by David Smith and Thomas Brunell. Their central finding is this: if you compare a year where all of the special elections that produced seat changes were won by Republicans to a year where all of these elections were won by Democrats, we would expect Democrats to pick up about 15 seats in the next general election, on average.
They conclude:
We are not arguing that this relationship established here is necessarily causal; rather the results of special elections are a barometer of sorts that provide some information about the national political mood, which manifests itself in the general election that follows. One could, however, imagine a causal connection – high quality candidates witness the results of these special elections and interpret the results as an indicator of swing toward their party, which motivates them to run for office. Future research might investigate whether the emergence of high quality candidates is linked to the results of special elections.
The paper is here.
From Josh Marshall over at TPM:
The reason negative campaigns often lose isn’t necessarily because people don’t like negativity. It’s because you usually get into running a negative campaign because that’s the last card you’ve got to play. McDonnell could run a feel good campaign because he was kicking Deeds butt pretty much the whole way. Conversely, Corzine didn’t lose because he ran a negative campaign. He ran a negative campaign because he was incredibly unpopular with New Jersey voters. And making Christie equally unpopular was really his only path to victory. This is elementary.
This is so elementary that even I agree. In Virginia, Deeds, the trailer, did what a trailer should do: try to move undecideds and McDonnell supporters toward himself. Lacking a charismatic personality or a set of compelling policy appeals, his only resort was to attack, attack, attack. He shot his big gun — McDonnell’s M.A. thesis — early and often, and then he had nothing left that could have caught him up. Meanwhile, McDonnell, as the clear front-runner, was able to do what a front-runner should do: wage a more positive campaign that emphasizes his appealing personal qualities. The McDonnell campaign’s response to Deeds’ relentless attacks was, I thought, masterfully done — especiallyt an ad that featured glowing testimonials from several women who had worked with McDonnell when he was attorney general of Virginia.
The elementary point restated: It’s not so much that attackers lose as that losers attack.
[Hat tip to Henry]
I asked Marc Hetherington for his thoughts on NY-23. He has done work on the origins of partisan polarization and is a co-author of a textbook on political parties. I’ve previously noted his new book with Jonathan Weiler on authoritarianism.
Marc writes:
The race in New York’s 23rd congressional district exemplifies the disconnect between polarized political elites and much of the American public, which like moderate alternatives when they are available. Movement conservatives like Sarah Palin and Rush Limbaugh attacked the moderation of Dede Scozzafava, the Republican nominee, ultimately driving her from the race and causing her to endorse Bill Owens, her Democratic counterpart. This opened the door for the first Democratic victory in the district in more than 100 years. It seems very likely that, others things being equal, a candidate like Scozzafava would have been the choice over Owens in a two person race. The district choked on the type of candidate that Palin and Limbaugh favored.
In the mid-1990s, so called movement conservatives were seen by some as the salvation of the Republican party. Now, they have become its bane, driving the party farther to the right than many mainstream Republicans will accept. Although they have damaged the Republican brand on issues like immigration, in particular, the party’s victories in New Jersey’s and Virginia’s gubernatorial races demonstrate that candidates who present a moderate image to the public can still prosper, even if they aren’t particularly moderate.
Movement conservatives often trumpet poll data suggesting the conservative label has become more popular during the first year of the Obama administration. Importantly much of this increase is due to the South’s increasing embrace of the label. That is not happening in upstate New York, or in much of the rest of the country for that matter. Douglas Hoffman, the Conservative candidate who Limbaugh and Palin embraced despite the fact that he didn’t even reside in the district, might have been broadly acceptable to Republicans in rural Mississippi or Tennessee but not in an increasingly competitive district like the New York 23rd where Barack Obama won a narrow victory over John McCain last year. By not realizing this, movement conservatives provided Democrats with one of their only bright spots on a night when perhaps they warranted none.
Here is Seth Masket:
Here’s what Owens’ victory doesn’t mean. It’s not a sign of major national affirmation of Obama’s agenda or foreshadowing of Democratic victories in 2010. Yes, it’s pretty fascinating that this area will be represented by a Democrat in the House for the first time since the 1800s, but that has a lot to do with the unusual circumstances surrounding the race, like, say, the fact that the Republican dropped out last week and urged support for the Democrat.
Yes. The same is true of the gubernatorial races in NJ and VA: absent more evidence, we simply don’t know if they were referenda on Obama, on Corzine and Kaine, or on none of the above. Interpretations of elections depend on the reasons for voters’ choices. You can’t simply ask voters why they chose a candidate, or whether a particular factor mattered. People do not accurately report on their own mental processes. You can’t simply look at the at overall levels of opinions — what percentage approves of Obama or is dissatisfied with Corzine, etc. This is why, contra NPR this morning, there is no “contradiction” in the fact that a majority of New Jersey voters approve of Obama but voted for Christie. The key is the relationship of these opinions to candidate preferences. And those relationships cannot be accurately estimated without controlling for other relevant factors, such as voters’ party identification. There is absolutely no evidence on this score — unless someone wants to send me the raw exit poll data — and any interpretations being provided by pundits and media outlets are pure speculation.
(Especially when these interpretations are contradictory mush like this from Dan Balz:
Off-year elections can be notoriously unreliable as predictors of the future, but as a window on how the political landscape may have changed in the year since President Obama won the White House, Tuesday’s Republican victories in Virginia and New Jersey delivered clear warnings for the Democrats.
Unreliable, yet clear!)
Seth goes on to say this of NY-23:
So what does it mean?…But in general, the coalition of opinion leaders who were initially resistant to moderate Republican nominee Dede Scozzofava — including Sarah Palin, Rush Limbaugh, Michelle Bachmann, and a number of Tea Party spokespeople — needed to do two things to prove that they were the future of the party. They needed to push the moderate out of the race (check) and still get their chosen candidate, Doug Hoffman, elected (whoops). It was a somewhat risky strategy in that it sent a message that moderate politicians had no place in the party, but if they could win elections and remain ideologically pure, that’s an ideal position for the conservative activists.
So now what they’ve done is proven the importance of ideological positioning — if you nominate too extreme a candidate, you lose the election. So the folks in the Republican Party who sabotaged Scozzofava and rallied around Hoffman now look silly and more than a tad disloyal. They cost their party a seat in Congress.
Jon Bernstein disagrees:
I’m very skeptical about that. To outsiders, sure, the conservatives “look silly and more than a tad disloyal.” But I really don’t think that’s how most Republicans are going to interpret NY-23. I think they are going to consider the formal party officials of NY-23 the villains of the piece. For them, this will be the story of a near-outrage that was foiled by a determined group of conservatives, and despite all disadvantages the conservative candidate nearly won. From that point of view, it was the party officials of NY-23 who were the disloyal ones, and it was their mistake in nominating an unacceptable candidate that cost Republicans this seat.
I think most Republican elites will either believe that version of events, or act as if they believe it.
I think that Republican elites’ interpretation of this will depend on their preexisting opinions. The moderates will see it as further evidence that conservatives want ideological purity even if it risks defeat. Conservatives will probably think something along the lines Jon proposes. People interpret new information in light of what they already think, and I don’t see any reason to suppose that Republican party leaders are different.
Marc Ambinder tells us that the infamous Corzine “throwing his weight around” spot aimed at Chris Christie was by far the most downloaded ad in the New Jersey race. This of course tells us more about media buzz than about efficacity - but has anyone done any research on whether candidates’ perceived girth relative to their opponent helps predict their share of the vote? The measurement issues are obvious, but political controversies (viz. the ‘how did Mike Huckabee slim down’ controversy from the presidential primaries) suggest that it is at least worth considering as an explanatory variable …
In his book Making Votes Count, Gary Cox makes the by now well known prediction that in single member districts (i.e., electoral races where only one person can ultimately win the seat/position up for election), voters should ultimately settle on one of the top two candidates in an effort not to waste their vote. Knowing this, elites should congregate around the top two candidates as well, ultimately leading to only two viable candidates within a given district. This behavior, on the part of both elites and voters, is loosely known in the political science literature as strategic voting. With this weekend’s political developments in mind, I thought it might be interesting to revisit Cox’s predictions regarding strategic voting.
By far the best realization of Cox’s predictions can be found in the NY 23 house race, where, as predicted, the candidate running third in the poll - Dede Scozzafava - has dropped out of the race and thrown her support behind one of the top two candidates. Score one for Cox.
In New Jersey, however, we still see a three way race for governor, with independent candidate Chris Dagget staying in the race up until the end. Dagget’s behavior is not particular consistent with Cox’s predictions, as the polls are giving him very little realistic chance of winning at this point, indeed probably even less than Scozzafava had.
Voters, however, do appear to be desserting Dagget a bit in recent days, which, according to Cox, is what they should be doing. So perhaps we can give Cox a push on this one.
Finally, we come to Afghanistan, which alone among the three elections under consideration has the right institutions for managing strategic voting, in the form of a two-round majoritarian electoral system. In these types of electoral institution - under which only the top two vote winners move on to a second round if no one gets a majority in the first round of the election - voters can safely cast their vote for their preferred candidate in the first round knowing that they will still have a chance to cast a ballot between the top two choices in the second round, and therefore not “waste their vote”. According to the logic of strategic voting, we should never see a candidate drop out in the second round of a two-round majoritarian election, which is, of course, exactly what just happened in Afghanistan, joining another distinguished recent election (the 2008 Zimbabwe Presidential Election) in this disturbing trend. To understand such behavior, therefore, we’ll likely need to look elsewhere, and in particular to the literature on election boycotts. But that’s another post for another day.
Be prepared. You will see the pundits hold forth on the outcomes in Virginia, New Jersey, and NY’s 23rd district “mean” for the Republicans, the Democrats, Obama, Sarah Palin, Glenn Beck, and Lord knows what else.
The correct answer will be: I don’t know. One, two, three, or none of the elections could reflect national politics. They could easily turn on local factors and issues. Who wins can’t tell us why they won. Exit polls, if there are any, will give us only the most tentative interpretations.
Marc Ambinder anticipates most of inevitable spin, but even he can’t resist the temptation:
…unless GOPers sweep the night, it’s going to be hard for any reasonable person to conclude that November 3 was a referendum on the entire Democratic project.
No. Even if GOPers sweep the night, it doesn’t mean that November 3 was a referendum on the entire “Democratic project,” whatever that is. I can’t see the value for either party in making inferences from such a small number of races.
On the problems in interpreting election outcomes, see this post on “mandates” from November 2008.
My colleague Boris Shor has performed some analysis (jointly with Nolan McCarty) on the ideological positions of state legislators. The estimates are based on state legislative voting, which might make you wonder how you could possibly compare legislators in one state with those in another. The trick is that some state representatives (for example, Barack Obama) also end up in Congress. There are enough of these overlap cases that you can put legislators from all 50 states on a common scale.
Boris and Nolan most recently applied their method to compare Deirdre Scozzofava, a state assemblywoman running on the Republican ticket in special election in New York’s 23rd congressoinal district. Boris writes:
Scozzafava has been assailed from the right for being far too liberal. For example, the libertarian Wall Street Journal this morning wrote, “Democrats want to portray this race as a familiar moderate-conservative GOP split, but the real issue is why Ms. Scozzafava is a Republican at all. She has voted for so many tax increases that the Democrat is attacking her as a tax raiser. She supported the Obama stimulus, and she favors “card check” to make union organizing easier, or at least she did until a recent flip-flop. . .” The conservative National Review writes: “In spite of its having gone for Obama in 2008, the district’s history suggests that it is basically conservative; Ms. Scozzafava is basically not. Boy, is she not. . . .
Actually, though, Boris and Nolan find Scozzafava to be pretty much in the exact center on a national scale:
Her ideological “common space” score is 0.02. These scores, similar but far superior to interest group ratings, put state legislators around the country on the same scale with each other, as well as with members of Congress.
Being in the center nationally puts Scozzafava to the right in New York:
Scozzafava’s score puts her in the 58th percentile of her party, which makes her slightly more conservative than the average Republican legislator in Albany, so she’s a conservative in her [state] party.
Here’s Boris’s graph showing the estimated positions of Democratic and Republican legislators in all 50 states in the past decade:
The Republican Party appears to be particularly liberal in Massacusetts, Connecticut, Hawaii, New Jersey, Rhode Island, Oregon, Illinois, and Delaware (although not, as has been much remarked, in California). (The gray lines on the graph show the average ideologies ofcongressional Democrats and Republicans in approximately the same time period.)
In case you’ve ever wondered what your district (and that includes your US House of Representatives district, upper or lower Statehouse districts, and even local districts) looks like, there’s a new website out there that lets you find out at just a click of a button: Redistricting the Nation.com. Moreover, according to the press release I was sent about it (does this mean The Monkey Cage has officially arrived?), the site also allows you to compare the “compactness” score of the district, as well as find out who is in charge of redrawing boundaries following the 2010 census. Looks like it could be a fun teaching tool.
For what it’s worth, my district (NY District 8) comes up 10th, 3rd, 15th, and 18th out of 428 districts in various measures of compactness, where, according to the site, the lower the score, the more likely there is gerrymandering present. I’m sure the Hudson River has something to do with this, but, then again, I don’t think we’ve elected any Republicans to represent Greenwich Village any time recently…
This one is making the rounds:
Young men who voted for Republican John McCain or Libertarian candidate Robert Barr in the 2008 presidential election suffered an immediate drop in testosterone when the election results were announced, according to a study by researchers at Duke University and the University of Michigan. In contrast, men who voted for the winner, Democrat Barack Obama, had stable testosterone levels immediately after the outcome.
Here is more. As is often the case with such studies, the further consequences of the physiological changes are less clear. Here are possibilities:
Stanton said the scientific consensus suggests the testosterone response to fighting and competition in males affects their future behavior in a beneficial way. The loser chills out a bit so he doesn’t continue to press his case and perhaps become injured. In contrast, the winner may be motivated to pursue further gains in social status.
It would be interesting to know if testosterone changes after election outcomes are accompanied by changes in actual behavior.
Higher-income people are more likely to vote Republican, and we’ve seen this in many different subgroups of the population. Among whites, among blacks, among religious attenders, etc., the poorer voters among these subgroups are more Democratic and the richer ones are more Republican.
This got me wondering: What are the subgroups of the population for which this isn’t true? Or, more generally, how do rich and poor differ in their voting patterns, in different subgroups of the population.
Here’s we found, courtesy of the 2000 and 2004 Annenberg surveys. For each group, we’re looking at Republican share of the two-party vote intention among people in the upper third of family income, minus Republican share … among people in the lower third of family income:
(Click on any of these graphs to see larger versions.)
A striking pattern. The differences between rich and poor are much larger among conservative, Republican groups than among liberal, Democratic groups. At the very bottom of the graph above, you see a few groups where richer people are more likely to vote Democratic. All of these are groups that are mostly liberal and Democratic.
Continue reading "Rich voter, poor voter, red voter, blue voter: Usually but not always!" »
Over at Politico this morning, there is a story entitled John Boehner downplays Dow 10,000. In it, the author writes:
Dow 10,000 just isn’t that big a deal, House Republican leader John Boehner said Wednesday morning. And anyone who places significance on the stock market hitting this symbolic number is “certainly not talking to the American people.”
In responding to this story in Politico’s Arena, I offered the following comment:
Boehner’s recent comments sounds like wishful thinking on his part, but the reality is that he is may be right in terms of the potential effect of a stock market swing on future voting behavior. Studies of economic voting in the United States show a strong relationship between economic conditions and presidential vote (or vote for the president’s party in congressional elections), but the variables that have tended to matter have been more focused on economic grows (e.g., change in real disposable income) than the stock market. In fact, a 2001 study by Michael Lewis Beck and Charles Tien (gated) concluded that “despite the current talk about using stock market performance as an economic indicator [in economic voting models], we found it had no vote effect after all.”. In the post-communist countries of Eastern Europe, interestingly, recent research by Andrew Roberts shows that unemployment turns out to be the most important economic indicator in voting for incumbents (gated), although he doesn’t include any stock market based variables in his models. Of course, to the extent that the stock market is doing what it is supposed to be doing (admittedly a big if!) and serving as a leading indicator of future economic growth, then Obama – and Democratic congressional candidates – may indeed stand to benefit in the future.
I wanted to throw this question out to the readers of The Monkey Cage to see if there was any new interesting research out there since the Lewis Beck and Tien (2001) PS piece on the relationship between stock prices and voting behavior, either in the US or elsewhere. Please feel free to promote your own research!
Times change — quickly, I guess. Compared to just a year ago, or even a few months ago, the air in America is now “crackling with optimism.” Or so proclaims this article in today’s New York Times. And as a consequence, the Times story continues, the nation’s largest companies, trying to catch the wave of optimism, are ratcheting up the happy talk in their commercials.
That may or may not be so. I’ve seen nothing more than anecdotes — the ones in the Times story — that indicates that it’s true. (I handle the remote at Casa Sigelmana, and I don’t stick around for the commercials. So I have no data to report, or even first-hand impressions.) But let’s assume that the Times, which is of course correrct about everything else, is right about this, too. Seems like politicos would have caught on by now, too, doesn’t it?
I don’t have any hard data on the tone of this year’s batch of political campaign commercials to contribute, either — one reason being that this is an odd-numbered year, so there are few major political campaigns going on. But if one ongoing race — the Virginia gubernatorial contest between Republican Bob McDonnell and Democrat Creigh Deeds — is any indication, going negative is hardly a thing of the past. Here are a couple of examples:
Let’s see what happens more generally in the 2010 round, assuming that what the Times takes to be the current mood of rosy optimism still holds by then. Anybody wanna bet that candidate ads will be any less negative next time around than they have been in recent years?
(For whatever it’s worth, my perception is that Deeds’ ads have been decidedly more attack-oriented than McDonnell’s — perhaps because McDonnell has been in the lead throughout the campaign, and perhaps because the Deeds campaign has concentrated so single-mindedly on hitting McDonnell over the head with his M.A. thesis.)
In an earlier post, I noted some initial findings from Democratic groups, particularly Catalist, who were doing voter contact, registration, and mobilization in 2008. Today, Marc Ambinder describes a new report from Catalist with the following conclusions (in his words):
According to the analysis, those registered voters contacted by Catalist member groups turned out at a rate of 74.6%; the voters who weren’t turned out in proportions roughly equivalent to the national average — about 60.4%. In four states, the number of new votes cast by liberals exceeded Obama’s victory margin: in Ohio, Florida, Indiana in North Carolina. If you assume that only 60% of these voters chose Obama, the margin was still greater than Obama’s in North Carolina and Indiana, both essential to his victory. With the caveat that correlation does not equal causation, the report provides convincing, if not absolute, evidence that the progressive/Democratic data-mining and targeting operation measurably helped elect Barack Obama.
In a subsequent post, he links to the actual report (.doc). I’ve only skimmed the report, but here are two preliminary conclusions, both noted by Ambinder and Catalist. First, these data are based on correlations between voter contact and political participation. Correlations are not causality, but that doesn’t mean that they are useless. Particularly in light of the next point.
Second, these correlations line up very nicely with some political science evidence. Ambinder and Catalist note, apropos of the graph above (Figure 14 in the report), that in-person contacts seem more effective than phone calls and direct mail in increasing turnout. This is in line with Don Green and Alan Gerber’s work, as well as that of some other scholars.
But Catalist’s findings are eerily similar to another finding from political science research on voter mobilization. Consider the shape of the curves in the graph above. These show that the effect of mobilization efforts was greatest among those with a moderate propensity to vote. Those with a very low or very high propensity were not much affected — the former presumably because they are hard to mobilize and the latter because they are already very likely to vote.
Compare that graph to one from a recently published paper by David Nickerson and Kevin Arceneaux who analyze a set of 11 voter mobilization field experiments conducted from 1998-2003 to determine who is mobilized.
The y-axis — labeled “attempt-to-treat effect” — captures the same thing as the Catalist graph: the effect of some mobilization effort on turnout. Note that the shape of their “high salience” curve looks exactly like Catalist’s curve. It peaks a bit below the midpoint of the vote propensity scale, with a longish right tail. This makes considerable sense: the 2008 presidential election was nothing if not highly salient. Note also that the maximum effect they document, about 0.14 (or 14%), is almost exactly the same as the maximum effect in Catalist’s data. These similarities are uncanny. Nickerson and Arceneaux’s theory looks very, very good.
I can’t help but puff out my chest and declare: Yes, political science research can tell us about the real world.
In our continuing series of election reports from political scientists, I am pleased to present the following guest post from Stratos Patrikios and Georgios Karyotis of the Department of Government, University of Strathclyde:
The Greek parliamentary election of 2009 took place on October 4, two years ahead of schedule. The Panhellenic Socialist Party (PASOK) returned to power after five-and-a-half years of conservative administration by New Democracy (ND). The early dissolution of parliament has become the norm in Greek politics since 1974, with elections called early, usually in order to serve the incumbent’s electoral prospects. In the September 2007 election, New Democracy had renewed its electoral mandate on the basis of a sound economic performance and trust in Prime Minister’s Costas Karamanlis leadership skills. On the other hand, defeat had left PASOK in turmoil. While its leader, George Papandreou, successfully bounced off challengers to his leadership in nationwide elections in November 2007, questions over PASOK’s unity allowed the left-wing party SYRIZA to reach historically high levels of public support. Despite its weak majority of 152 in a 300-strong Parliament, ND remained ahead in the polls until the autumn of 2008 (see graph, data from Public Issue, website in greek):
However, ND failed to deliver on its reform plans in education, social security and public sector downsizing, amidst economic underperformance, internal opposition, as well as violent street protests. Karamanlis’ promise to restore public faith in the political system was further hampered by a string of corruption scandals, which involved high profile cabinet members. The slow reflexes of the PM to assume responsibility and isolate those implicated during his annual address at the International Exhibition in Thessaloniki in September 2008 proved a crucial turning point. PASOK overtook ND in the polls, while SYRIZA’s support gradually deflated. Meanwhile, the downturn in the global economy had affected the key sectors of tourism and shipping, bringing Greece to the brink of recession.
As most watchers of European politics probably know already, Ireland has just passed a constitutional referendum allowing it to accede to the Lisbon Treaty by a whopping majority of 67%. This is surprising, given that a similar vote last year (some small concessions were made in the meantime by Ireland’s EU partners) saw the Lisbon Treaty being rejected by a 6% margin. So what changed in the meantime?
Unfortunately, we don’t have any very good evidence from opinion poll crosstabs, exit polls etc. It may be that someone has conducted a flash poll in the aftermath of the referendum (Eurobarometer did this last time), which may provide a little more data. In the meantime, my thoroughly non-scientific, open to challenge etc rough ranking of the plausible reasons why it was different this time around.
(1) Better campaigning by the Yes side. In the last referendum, while the main political parties advocated a Yes vote (with the exception of the Greens, who were more or less in favor), they didn’t exert themselves especially hard to persuade voters until the closing days of the campaign, when it was too late. This time around, anecdotal evidence suggests that the Yes campaign was much better organized. It also had a single - and very articulate - spokesperson in the form of Pat Cox, former president of the European Parliament.
This was important, because the flash survey last time around suggested that the most important reason why people voted No was because they did not know enough about the Treaty (11% of No voters actually thought that the referendum would, on balance, be good for Ireland). The No campaign did an excellent job of convincing voters that if they did not understand the Treaty, their default position should be to vote No. They pointed out that Ireland would still continue to be a member of the EU. Pro-Treaty politicians did little to explain the Treaty’s presumed benefits - largely because they did not themselves have an especially good understanding of the Treaty, instead relying on generic nostrums about how good Europe had been for Ireland in the past. This time, in contrast, the Yes campaign did a better job in explaining what the Treaty did, and, more importantly, what it did not do, aggressively targeting some of the bogus claims that No proponents had made about abortion, minimum wages etc.
Campaigning effects often wash out in electoral politics, because both sides have similar incentives to win, and to organize as best they can. Here, in contrast, we saw a referendum where one side was far more organized than the other in the initial Lisbon vote, but where these differences evaporated in the second vote (where, if anything, the Yes side was better organized than the No side).
(2) Increased turnout. There was a significant increase in turnout from 53.1% last year to 58% this time around. We don’t have any data on who turned out this time who didn’t the last time, or vice versa. My best guess would be that we saw substantially increased turnout among middle and upper class voters (who are far more likely to favor the Yes position), and static or decreased turnout among working class voters (who are more likely to favor the No position in relative terms). My guess is based on the seat-of-the-pants perception that the salience of the vote was higher among middle class voters than last time round, and that working class voters had other things to worry about given (3).
(3) A change in economic circumstances. Ireland’s GDP is predicted to shrink by anything up to 12% this year, in contrast to last year when the vote was taken just before the economic crisis began to bite. This may be one reason for changing patterns in turnout (working class voters are more likely to be hurt by the recession, and hence perhaps less likely to vote). But it also may have decreased voters’ willingness to ‘go it alone’ against the rest of the EU. Elite consensus seems to have hardened around the opinion that Ireland would have been far worse affected by the recession had it not been an EU member. You could argue this both ways on the empirical merits - while liquidity support from other member states surely helped, a stringent monetary policy probably hurt and will continue to hurt Ireland (Iceland, which was far worse affected initially, may end up recovering more quickly). But whether the belief was justified or not by facts, it plausibly had significant ideological consequences, perhaps shifting the default position for many voters from ‘No’ to ‘Yes.’ Perhaps voter irrationality - the quick succession of an extremely severe recession after the last vote, had psychological consequences too.
(4) Perhaps most interesting was the effect which did not bark in the event. As best as we can guess, voters who blame the (extremely unpopular) government for the current economic crisis did not vote No as a protest against the government in anything like the numbers one might have expected. The referendum carried by an extremely wide margin, even against the backdrop of recession and a deeply unpopular bailout of the banks. This is surely in part due to the fact that the major opposition parties supported the referendum too. But it is surprising to me that there was not more evidence of a generic anti-system vote under the circumstances.
Again, take this all this analysis with a grain of salt - in the absence of proper data it is somewhere between loose hypothesizing with big fat ecological problems, standard-issue punditizing and sheer guesswork. But until and unless actual data becomes available, educated guesswork may be the best we can do.
In our continuing series of election reports from political scientists, I am pleased to present the following guest post from Professor Marco Lisi of the Social Sciences Institute of the University of Lisbon on the 2009 Portuguese legislative elections:
The European elections held in June 2009 inaugurated a series of three elections in which Portuguese voters had the possibility to choose their representatives at the European, national and local level, respectively. After the first absolute majority achieved in its history by the Socialist party (PS) in 2005, the government led by the Prime Minister José Sócrates had to face huge challenges, especially in terms of economic performance and reform of social policies. These factors – as well as personal scandals associated with the figure of the Prime Minister - slowly undermined the popularity of the government by the second half of the legislative session. Thus, the socialist lead over the main opposition party – the center-right Social Democratic Party (PSD) – increasingly diminished, especially after the leadership change occurred in 2008 with the election of Manuela Ferreira Leite as the new opposition leader. In the 2009 European elections the PS suffered a huge defeat, receiving only 26.6 per cent of the vote, while the PSD won 31.7 percent. Although in second order elections small parties are expected to perform better than in national competition, opinion polls predicted that the substantial loss of supporters for the PS would also harm the two smaller left-wing parties, the Portuguese Communist party (PCP) and the Left Block (BE). In fact, both parties substantially increased their electoral performance compared with previous results, obtaining more than 10 per cent.
The European election results not only strengthened the PSD leadership, but also consolidated the growth of extreme left parties. The steadily loss of the PS government’s public support seemed to increase the competitiveness of the legislative elections. Thus, at the beginning of the campaign for the legislative elections, held September 27, it was clear that achieving an absolute majority was just a dream. After the huge electoral defeat, Sócrates attempted to polarize the contest and to reduce the debate to the choice between the two main parties, the PS and the PSD. At the same time, he appealed to strategic voters, those who voted for small parties in June to punish the incumbent party, by emphasizing the danger of government instability and setting forth more leftist policies, not only through big state investments but also in terms of “post-materialist” policies. While the socialist leader adopted a twofold strategy in order to recover electoral losses with regard to both right and radical left parties, the PSD leader Manuel Ferreira Leite employed a low-profile strategy, trying to differentiate her style from Sócrates’ and focusing mostly on a retrospective evaluation of the socialist government.
Dan Balz and Haynes Johnson’s The Battle for America 2008 (here) is — as of this moment — the most complete narrative of the 2008 race. It covers the major and minor events, most of which will be familiar to readers of this blog. More importantly, it uncovers some nice tidbits, based on their reporting during and after the campaign. The book is certainly superior to the book based on Newsweek’s reporting — and in fact contradicts that the book on at least one key point. (See this previous post on the Newsweek book.) It’s pretty lively in general, although sometimes they adopt an odd bird’s-eye-view tone that leads to tip-toeing passages like this:
They [Republicans] reeled from self-inflicted wounds. GOP legislators and lobbyists went to prison on corruption charges. A House Republican was forced to resign over a sex scandal. A Republican senator attracted wide press coverage after an embarrassing incident in a men’s airport bathroom.
Of course, that’s a small matter. I’ll have more to say about the more important ways in which I think the book falls short. But in the meantime, let’s get to the good stuff.
Continue reading "Book Review: The Battle for America 2008" »
As part of our continuing series of international election analysis from political scientists, I am pleased that we now have our first example of pre-election and post-election analysis of the same election. (As a reminder, please feel free to email me directly if you are interested in writing on any upcoming elections.) Here is the post-election analysis of this weekend’s 2009 German parliamentary elections from Dominik Duell and Alexander Herzog of NYU:
The 2009 German parliamentary election is over and with more and more certainty Germany faces the end of the grand coalition and a new center-right government of Christian Democrats (CDU and CSU) and Free Democrats (FDP). The Social Democrats (SPD) received their worst result in the history of democratic elections since 1949. Recently, we argued in this blog that this was due to SPD’s inability to create an ideological label, which distinguished the party from their grand coalition partner CDU. This caused many voters on the left to deviate to the Green party and especially to The Left. So far, however, we have not said anything about where voters actually see themselves located within the political space. The figure below shows where survey respondents locate themselves on the left-right dimension and where they perceived SPD and CDU to be positioned.
We assumed that voters go for the party which is closest to them and moving to the right should thus have been beneficial to the Left and causing a gain among left-leaning voters. Voters, however, were all assembled around the center with highest density; that actually means that in moving to the right the SPD should have gained more votes around the center than the party should have lost on the left side. So what did the SPD actually lose in this grand coalition if not the perceived closeness to voters? The final argument might read like this: The Social Democrats lost their ability to be seen as the leader of the left. This on the one hand enabled The Left to become more mainstream by being the leading voice on formerly SPD-core ideological themes like social justice. On the other hand, and because of resentments among many Western Germans against the Left, this ideological “weakness” also triggered many former SPD voters not to turn out at all.
Continue reading "2009 German Parliamentary Elections: Post-Election Analysis" »
In our continuing series of election analysis from political scientists, we are pleased to have pre-election analysis of this weekend’s German Parliamentary Election from Domink Duell of New York University:
Next Sunday Germany’s first female Chancellor, Angela Merkel, is up for re-election. Approaching election day, the polls indicated a tight race between two coalitions: first, a possible center-right coalition of Christian Democrats (CDU) and Free-Democrats (FDP) and, second, the left-leaning Social Democratic Party (SPD), Greens, and, if accepted by the SPD, The Left (see here for a collection of most recent polls in English). After four years of a grand coalition between the CDU and SPD, facing the worst economic crisis in the history of democratic Germany, and earning criticism from all sides of the political spectrum, election outcomes were said to be more uncertain than ever before.
For years now many political scientists and journalists in Germany have interpreted the decreasing vote share of the two biggest parties, the Christian Democrats and Social Democrats, and the increasing amount of swing voters and people splitting votes, as the establishment of a very fluid five party system (see Niedermeyer 2003) in which voters tend to travel within and between ideological camps and in which parties should follow this movement by experimenting with a larger set of possible coalitions across borders of ideological segregation (see for example here for an essayistic summary of this point in German). In 2009, those trends may continue. Recent results from state-legislature elections saw the Christian Democrats loosing their absolute majority in Thuringia and Saarland. But rather than favoring the SPD, the losses empowered The Left and the Free Democrats. Polls also show that 40 per cent of voters are still undecided, a number that has not been reached so near to election day since 1949.
The question to be answered next Sunday is if we will indeed see another push towards indistinguishable ideological camps making voters even more indifferent, and discouraging them from making a clear decision for a new government made up by one ideological camp and led by one strong party? Taken to an extreme, will it be the case that, at some future point, all relevant parties are fully similar and voters are simply tossing a coin? Is it really true that there are no identifiable ideological blocks up in this election?
Let us go back to the 2005 election and look at voters’ perception of each party’s policy position. The graph below shows party positions on a left-right dimensions, derived by aggregating answers to a public opinion poll.
In this particular year, we see that only the two (then incumbent) parties, the Social Democrats and the Green party, were almost indistinguishable, but that a majority of survey respondents saw a difference between the SPD and CDU, with some overlap but still most of the time different. Furthermore, The Left (PDS at that time) clearly was perceived as a choice at the absolute left and FDP as one closely connected to the center-right camp. The grand coalition might have added diffusion about differences between the two big players by moving SPD and CDU closer together.
But if we now look at a similar graph for the year 2009 (above), we see that voters still perceive a difference between the CDU and SPD even after four years of shared government responsibility. The Left, however, gained in large numbers whereas the SPD lost in numerous state or local elections. Directly comparing how voter’s perceive the position of SPD and the Left on the major ideological dimension in this graph demonstrates that the Left, in the voter’s eyes, has made it’s way slightly further to the right occupying more and more space on the left side of the political spectrum. If we believe spatial theory saying that voters go for the party ideologically close to them, the Left is a good candidate to gain more from the losses of the SPD on next Sunday.
What does this mean for the upcoming election? As indicated by recent polls, it is less about the CDU losing or winning the election. They still score highest with around 35 per cent of the vote. It will be more about how far the SPD was able to really distinguish itself from their current coalition partner in the final weeks of their campaign so as not to lose more voters to the more leftist camp. If the recently more aggressive campaign by SPD leader Frank-Walter Steinmeier turns out to be successful, Germany might end up in another round of a grand coalition. If not, if the lack of a clear and differentiable ideological label for the SPD strikes, we will see a CDU-FDP coalition since the Left and the Greens will take away social-democratic votes. Ideology is not dead, not every coalition is possible, not every political experiment rewarded. The Social Democrats were punished for being open to experiments, trying to cover a broad ideological range, and not making their label clear enough to voters. The SPD failed in establishing itself as the leader of the ideological left.
[Note: This is updated from the original post, which inadvertently omitted the last two paragraphs. Sorry!]
When the tempest over Robert McDonnell’s thesis began, I didn’t think it would exceed the proverbial teapot. Now I’m beginning to wonder. In my wonderment is a lesson about how to identify the effects of campaign events.
The first lesson: here we are three weeks after the story broke, and only now is it possible to see a potential effect. The lesson, in other words, is “wait.” Above I’ve taken Pollster’s graph, and made the smoothing more sensitive. You can see an inflection point around September 1, when story broke. Another way to look at this is to compare the margin between the two candidates in two sets of polls: those in August and the first week of September, when McDonnell’s average margin was about 10 points, and the polls taken after the first week of September, when McDonnell’s average margin was 4.5 points. The race does seem to be narrowing.
The narrowing isn’t necessarily due to McDonnell’s thesis, of course. However, some other evidence suggests that it might. This leads to a second lesson: identify which groups are changing their minds.
Who is most likely to be turned off by conservative social attitudes, particularly with regard to gender rights and responsibilities? Probably swing voters who do not share these attitudes. And in today’s Washington Post, Anita Kumar and WaPo pollster Jon Cohen provide some evidence for this hypothesis. Compared to an August poll, the most recent WaPo poll found particularly significant losses for McDonnell among Northern Virginians and independent women.
The third lesson is: pin down the corollary attitudes that should be affected by the event in question. The Post did just that. (See here). Northern Virginians and independent women are more likely than other groups to identify McDonnell as “too conservative,” to say that his thesis makes them less likely to vote for him, and to oppose traditional gender roles.
Taken together, this evidence is still circumstantial, but it’s far better than what often exists amidst cable-newsy speculations about campaign effects.
What does this gang of political punchlines have in common? They were all major-party nominees for Vice President. Presidents and presidential candidates, by comparison, don’t seem so wacky. There was Nixon, but he counts in the vice-presidential ledger too. And there have been failed presidencies (Jimmy Carter, George W. Bush), but these dudes aren’t political jokes along the lines of John Edwards.
What’s going on here? Some possible explanations:
1. I’m not doing a careful historical study. If you list all the pres and vp nominations over the past century, say, you’ll find that the vp nominees are as a sober and reasonable bunch as the guys at the top of the ticket.
2. Just the fact of being president, not to mention what they do in office, bestows an air of seriousness. For example, had Bill Clinton lost in 1992, perhaps he’d be considered a Gennifer Flowers-sized joke. But, since he won, he’s evaluated on the accomplishments of his presidency. Conversely, Geraldine Ferraro is now little more than a political joke, but had she become president, she’d be bestowed a retroactive seriousness. And, if Aaron Burr had been president, maybe he would’ve been too busy to go around shooting people.
3. Presidential nominees really are vetted more closely—by other politicians and by the voters—whereas vp nominees are chosen pretty arbitrarily. For example, it’s hard to take Joe Biden seriously as leader of the free world, but he was considered good enough for the placeholder role of Democratic politico.
Whatever is going on, the contrast is pretty striking to me.
Earlier this week I announced a new Monkey Cage initiative to provide timely commentary on international elections from political scientists who are actively following developments in the country in question. I am very pleased that Ola Listhaug of the Norwegian University of Science and Technology agreed to provide a guest post on this week’s Norwegian elections. He writes:
The Norwegian parliamentary election held on September 14 resulted in a narrow majority of seats in the Storting for the red and green coalition of parties. The parties in the coalition, Labor, Socialist Left, and the Centre party received 86 seats against 83 seats for the parties on the right. Within the winning coalition, Labor received 35.4% of the popular vote, an increase of 2.7% from the previous election. The two smaller coalition parties both received 6.2% of the vote which was about status quo for the Centre party and a loss of 2.6% for the Socialist Left party. The Conservative party had 17.2% - an increase of 3.1% from the very low support in the 2005 election. The other party on the right, the populist Progress party received a record share of the votes with 22.9%, slightly higher (+.9%) than the vote share in 2005. Norwegian parties are from historical reasons classified as “left” or “right” or “socialist “ or “bourgeois”. The two remaining parties that achieved representation in the parliament, the Liberal party and the Christian People’s party normally are part of the right or bourgeois bloc of parties, but can also be seen as centre parties. Both parties did badly. The Liberal party was able to elect only 2 MPs from a vote of 3.9% , -2% from 2005 and the Christians declined by 1.2 to a record low support of only 5.5.
Several observers have noticed that the losing parties in the Storting received about 50 000 more votes than the winners. There are several reasons for this. The Liberal party fell below a threshold of 4% where a party can get an improved proportionality. Receiving 3.9% the party just missed the threshold and was left to win seats in each of the 19 counties (that elected from 5-17 seats). Second, Norway has an electoral system where votes in rural areas count more than in the urban parts of the country. This gives an advantage to parties with a relative strength in rural areas. In this case it gave an advantage to the parties on the left. That geography is part of the election rules is no accident but signifies the important role of cleavages in Norwegian politics. Looking at the geography of Norway – a large territory with huge distances between south and north and with sparsely populated areas in the north as well as in parts of the south, one can see the reason for giving a vote advantage to the rural periphery. The center-periphery dimension is part of Stein Rokkans conceptual model of Norwegian politics to explain the origin of political cleavages in the period 1880-1935. The cleavages of this period created the party system that we still see remnants of in elections in the twentieth century, but the relationship to the original has now weakened for more than 50 years. Maybe the most striking difference from the original model, is the decline of the center parties. This is not center parties in the Downsian meaning since they can have distinct, even extreme, positions on other dimensions than left-right. The parties have partly attempted to establish their own coalition alternative, but mostly been part of a center-right political bloc. One of the three important center parties, the Centre party (from the beginning a farmer’s party) has chosen to side with the left, from 2005 in the coalition led by the Labor party. The Liberal party has had trouble of finding a consistent issue profile, and even more to align in a coalition with the two parties on the right. The Christian People’s has cooperated with the Conservative party, but has more trouble with the Progress party. In sum, this means that a center that is moderate in left-right terms but distinct in support of religion, distinct regional cultures, and economic interest of rural areas, is almost gone.
Chris Kutz and Aaron Edlin pointed me to this book by Derik Parfit that made essentially the same argument that Aaron, Noah Kaplan, and I did, but twenty years earlier. Thus scooping by three years John Quiggin’s publication of the same idea.
The key idea—that rationality is not the same thing as selfishness, and that, in particular, voting is not a sucker’s game or a prisoner’s dilemma—is an important one, and I hope that my incessant promotion of this point will enable future researchers to be aware of it. So far, Parfit (1984) is our earliest reference.
Jeff Lazarus emails me with a different perspective on the question of why women in Congress seem to do better at porkbarrel politics:
As it happens, Amy Steigerwalt and I are working on the same issue as Anzia and Berry. We’ve been working with earmarks data instead of the FAADS dataset, but we have come to the same conclusion: female members bring home more bacon than male members.
However, we disagree with Anzia and Berry as to why. We agree that the difference stems from the fact that female candidates for office have a harder time winning than male candidates. On the other hand, we believe that the increased challenges faced by female candidates for office produce a subtly different effect. Female members of Congress look forward to their upcoming reelection with an understanding that they will have a harder time winning reelection than their male counterparts.
This prompts them to spend more time and energy on position-taking and credit-claiming activity than male members. Thus, we argue that the difference in pork barrel spending is not due to any difference in quality between male and female legislators (as Anzia and Berry argue), but rather to a difference in what part of the job members of different gender are likely to focus their efforts on.
Fortunately, the two models make different predictions about other areas of the legislative process. The selection model predicts that female legislators will be “better” in other areas of legislative behavior, such as writing bills which successfully navigate the legislative process. However, our evidence indicates that female members’ bills are actually less successful than male members’ bills. On the other hand, the effort model predicts that female members spend more time engaging in credit-claiming and position-taking activity. One type of credit-claiming activity is pork barrel spending, which got this whole conversation started. One type of position-taking
activity is bill authorship. We find evidence that female members of congress do indeed write more bills than male members of Congress. So on the whole we believe that the evidence for the effort model is much stronger than the evidence for the selection model.
Wanna make some easy money? Call your favorite bookie and get your money down — wager the mortgage if need be — on the outcome of this year’s Virginia gubernatorial race. It looks like a sure thing.
Maybe you’re hesitant to place a bet on this race because you don’t follow Virginia politics. Don’t worry. Here’s a little help to get you started.
Virginia’s governors are term-limited.One four-year term and they’re out the door. So you needn’t worry about the incumbency factor.
Virginia holds its gubernatorial elections in odd-numbered years (2009, 2005, 2001, …), presumably minimizing the effect of national-level trends. So don’t worry about coattails.
Virginia’s last two governors have been Democrats (Tim Kaine and Mark Warner).
This year’s candidates are Creigh Deeds (Democrat) and Bob McDonnell (Republican). McDonnell currently leads in the polls, but Deeds has gained ground.
Now, armed with all that information, are you ready to pick a winner?
Still not ready? Well, then, here’s one final piece of information, as provided by the dean of Virginia politics watchers, Larry Sabato.
According to Sabato, in the last eight gubernatorial elections, without a single exception, Virginians have elected the candidate of the party that does NOT occupy the White House.
I think eight out of eight should be good enough even for a cautious type like you. Let’s see now: Obama’s a Democrat, so Bob McDonnell, the Republican, will be the next governor of Virginia. You can count on it.
[Hat tip to Jim Todd]
They’re better at bringing home the pork. Brayden King at OrgTheory points to a new paper by Sarah Anzia and Christopher Berry.
We argue that the process of selection into political office is different for women than it is for men, which results in important differences in the performance of male and female legislators once they are elected. If voters are biased against female candidates, only the most talented, hardest working female candidates will succeed in the electoral process. Furthermore, if women perceive there to be sex discrimination in the electoral process, or if they underestimate their qualifications for office relative to men, then only the most qualified, politically ambitious females will emerge as candidates. We argue that when either or both forms of sex-based selection are present, the women who are elected to office will perform better, on average, than their male counterparts. We test this central implication of the theory by using legislators’ success in delivering federal spending to their home districts as our primary measure of performance. We find that congresswomen secure roughly 9 percent morespending from federal discretionary programs than congressmen. This amounts to a premium of about $49 million per year for districts that send a woman to Capitol Hill. Finally, we find that women’s superiority in securing particularistic benefits does not hurt their performance in policymaking: women also sponsor more bills and obtain more cosponsorship support for their legislative initiatives than their male colleagues.
To be less flip about it, Anzia and Berry’s paper is not (despite this post’s flip title) an argument for electing women to Congress. Instead, it is a claim that those women who do manage to clear the hurdles in their way are likely to be exceptionally good politicians. In a world where women politicians did not have higher barriers to entry than men, the effect they observe would presumably disappear.
With football season officially upon us (much to the dismay of fans in Buffalo and Washington), two quick notes seemed in order. First, my cousin Stephen Horowitz - the creator of the only cartoon about Bankruptcy Law of which I am aware - has a post on his cartoon’s website addressing the similarities between structured finance and fantasy football. Not directly related to political science, but seemed like the kind of thing that might interest readers of the Monkey Cage.
Second, my favorite 2009 APSA paper featuring college football games as part of the empirical evidence (and, ok, the only 2009 APSA paper I’m aware of that features college football games) can be downloaded here. Here’s the abstract for the paper by Healy, Malhotra, and Mo, entitled Personal Emotions and Political Decision Making: Implications for Voter Competence:
According to what criteria do citizens make political decisions, and what do these criteria say about democratic competence? An impressive body of evidence suggests that voters competently evaluate diagnostic information such as macroeconomic trends and their personal financial circumstances to reward good performance while ridding themselves of leaders who are corrupt, incompetent, or ineffective. However, what if some voters’ personal emotional reactions to events completely unrelated to public affairs influence their voting decisions? The conflation of personal emotions with political cognition challenges traditional conceptions of citizen competence and democratic accountability. We explore whether emotional reactions unrelated to incumbent performance affect voting behavior by assessing the electoral impact of local college football games, events that government has nothing to do with and for which no government response would be expected. On average, a win before Election Day causes the incumbent to receive about one percentage point more of the vote, with the effect being larger for teams with stronger fan support. We corroborate these aggregate-level results with a survey conducted during the 2009 NCAA Men’s College Basketball Tournament, where we find that sports-induced emotional change affects approval of President Obama and assessments of the health of the country. Voters’ decisions and attitudes are thus shown to depend considerably on events that affect their personal level of happiness even when those events are entirely disconnected from government activity. Our results provide new evidence on the significant limitations of the electorate’s capacity to hold elected officials accountable for their actions.
One question I would raise about the paper concerns the magnitude of the effects. We’ve always known that traditional voting models focusing on things like performance evaluation, policy positions, partisan preference, etc., are not going to be able to capture all aspects of the voting decision. With this in mind, is the 1% of the vote identified by the authors of this paper as being a function of one’s personal emotional state a lot? Put another way, does it really provide a “significant limitation” on the ability of the electorate to hold politicians accountable, or is it simply helping to unpack some of what has always been hidden in the error term?
One of the goals of the Monkey Cage is to inject the opinions of political scientists into popular discourse about contemporary political events (see our mission statement). With this in mind, we are going to try to add a new element to The Monkey Cage starting this fall: quick reports on global elections from the perspective of political scientists.
Here’s the motivation. Most of what most of us know originally know about elections tends to come from journalists. Political scientists are occasionally (or even often) quoted in articles, but usually we don’t enter the discussion on our own terms until months (or even years!) later when articles that include analysis of the election in question begin to appear, with the occasional Op Ed as the exception.
However, for most elections out there, there is some political scientist paying close attention to what is going on and/or some graduate student out in the field doing research. Our thought is to use the Monkey Cage to give this person a chance to share their thoughts more widely in the immediate aftermath of a given election. Individually, we may only really have a sense of what is going on in a few countries each, but collectively we probably have most of the globe pretty well covered. If successful, we can broaden what all of us know about particular elections immediately after they have occurred; in an ideal world, the Monkey Cage would simultaneously become a place where others can turn to find out what political scientists think about current elections, thus giving you a chance to disseminate your opinions more widely.
This would of course have to be done largely through guest blogs. I’ve already solicited a couple of these on an ad hoc basis in the past (e.g., following the Moldovan and Japanese election), but would like to try to make this a more systematized process in the future.
With that in mind, after the jump I’ve posted a list of elections coming up this fall. If you are interested in writing a guest blog (2-3 paragraphs, although could be longer if you prefer) for any of these elections, please email me directly at joshua_dot_tucker_at_nyu_dot_edu. The key would be to try to get the guest blog up with 24-48 hours after the election is over. I’d also be willing to post guest blogs before and after a given election, and there’s no reason we couldn’t have multiple guest blogs on a single election if there is interest.
For those of you who follow elections in your scholarly work, I’d love for you to try to include The Monkey Cage in post-election discussion, either by contributing a guest blog or by joining in the discussion here after elections. We would especially encourage comments relating your own research to contemporary electoral developments. And we are also happy to take suggestions on how this particular aspect of the blog could evolve in the future.
In the immediate future, does anyone want to write on either the Norwegian or Swiss elections this week? The NY Times suggests that Norwegian elections tend not to exhibit typical economic voting patterns - does anyone have any research that could shed light on this question?
List of upcoming elections follows below…
With last week’s historic Japanese elections in mind, I asked Princeton University professor Christina Davis for her thoughts on the election results. She kindly provided the following:
Japan’s August 30th election has been hailed as revolutionary. After fifty-five years as the ruling party with only an 8 month period out of power in 1993-94, the Liberal Democratic Party (LDP) was thrown out of office. The opposition Democratic Party of Japan (DPJ) won 308 of 480 seats in the Lower House for a decisive victory that gives it majority control. Through a coalition with two smaller parties it will also hold a majority in the upper house. What does this mean for Japanese politics?
First, alternation in power will give a much needed jolt to Japanese democracy. The ongoing rule of the LDP despite almost two decades of economic malaise had seemed an anomaly from the most basic expectations of democratic accountability. How could the LDP ride to success on the back of high economic growth in the first thirty-five years of its rule but not be held responsible for the crash of the bubble economy and stagnant growth of the 1990s? Retrospective voting theories would suggest that the Japanese people, who observed rising unemployment, bankruptcy, and social dislocation with alarm, would direct their anger at the incumbent party. They did to the extent that the support rate of the LDP has fallen steadily while unaligned voters grew to form 40 percent of the electorate. But voters feared that the opposition would not be competent to rule, the LDP benefited from all the powers of incumbency arising from years of directing the pork to favored regions and groups, and the political saavy and calls for change of Junichiro Koizumi resurrected the popularity of the LDP during his term as prime minister from 2001 to 2005. But when the Koizumi reforms brought rising inequality along with a return to economic growth, it contributed to backlash against the LDP. The party proved unable to reinvent itself this time as it suffered from a string of weak prime ministers and the steady erosion of support from key constituencies, in part the very result of the economic reforms launched by Koizumi. When the global economic crisis pushed Japan back into recession it was clear that the economic growth supported by Koizumi reforms could not withstand a global downturn of demand and left some sectors of society more vulnerable. Voters were informed on the eve of the election that Japanese unemployment had reached 5.7 percent, which would be positive figures in Europe but represented a historic high in Japan. This election suggests that finally the voters decided that risking a new set of leaders was necessary to change the direction of the country. Thirty percent of LDP supporters and 53 percent of unaffiliated voters cast their ballot for DPJ candidates.
Continue reading "Implications of the 2009 Japanese Elections" »
Seattle has a mayoral candidate named Jan Drago. I am thinking of voting for her just based on her name or the chance that, when debating the city’s proposed plastic bag tax/fee (which she is against), she will say “If it dies, it dies.” Or should I read the voter pamphlet and make an informed decision?
From a reader email to Bill Simmons. (This is the reference.)
Simmons’ reply:
I received this e-mail before Jan Drago failed to make the final three for November’s election, which raises the question, how could you not vote for Mayor Drago??? You know how many of us would kill to have a Mayor Drago so we could make “I can’t get over the size of this mayoral candidate!’ and “What started out as a joke of an election has turned into a disaster” jokes? The only way I wouldn’t vote for a Mayor Drago is if her opponent filmed a commercial in which they climbed a 20,000-foot mountain in Russia wearing only a parka and running boots, then reached the top and screamed, “Draaaaa-gooooooooooooo! Draaaaaaaaaa-goooooooooooo!” Actually, I’d still vote for Mayor Drago. And you know why? Because I vote for me. I vote for me!!!!
Did the Clinton health care plan affect the outcome of the 1994 elections? A few weeks ago, I suggested that it didn’t, and that a much more important indicator was dissatisfaction with the economy, combined with some local factors.
Today, James Morone suggests that the defeat of health care reform did affect the 1994 elections by “demoralizing” Democrats:
Many Democrats are moving to whittle back health reform in order to win over moderate, fence-sitting, frightened independents. Big mistake.
Go back and look at the midterm tsunami that swept the Democrats out of office the last time. The turnout for that wave was just 36 percent. Moderate, fence sitting independents don’t vote in midterm elections with a 36 percent turn out.
What really happened back in 1994? The Republican base - jubilant, mobilized, and angry - turned out. The Democratic base - dispirited, disenchanted, and demobilized - stayed home. As Democrats ponder which way to go in this latest round they ought to read the political lessons more carefully: Short term electoral success rests with the base, the people who got excited about “change we can believe in.” Long term electoral success rests in designing and pushing through a program that then grows very popular.
Ezra Klein cites this approvingly:
The political danger is not just that a failure on health-care reform will anger the electorate. It will also change the composition of the electorate. Dispirited Democrats will stay home. Energized Republicans will press their advantage. Add in that the wave of young voters who were energized by Obama’s campaign probably aren’t going to turn out for the midterm election anyway, and you’re looking at a pretty unfriendly landscape.
That’s why the midterms are dangerous for Democrats. Losing on health care and collapsing into recriminations and internal divisions pretty much guarantees that Democratic voters of all sorts are turned off. You don’t just win elections by being popular. You win elections by making sure that the people who like you turn out to vote.
What does the data tell us about turnout in 1994? Here is a graph of self-reported turnout based on ANES data:
Unsurprisingly, Republicans have historically maintained an edge in turnout, with the exception of 1982-1990. In 1994, turnout among both parties was up, relative to 1990. It was up more among Republicans (a 14-point increase) than among Democrats (a 5-point increase).
To me, this is pretty mixed evidence that 1994 was due to a “demobilized” or “dispirited” set of Democrats. It does appear that Republicans were energized, but it looks like Democrats were too, just less so. Moreover, if you estimate a model of turnout that controls for ethnicity and education, there is no significant relationship between party and turnout in 1994. Any difference between the parties appears spurious.
What about other measures of “spirit”? The ANES asks “Some people don’t pay much attention to political campaigns. How about you, would you say that you have been/were very much interested, somewhat interested, or not much interested in the political campaigns (so far) this year?” In 1994, there is an apparent gap between the parties: 34% of Republicans said “very interested” compared to 25% of Democrats. But again, the difference between the parties is negligible once you control for education and income: a shift from “strong Democrat” to “strong Republican” is associated with only a 4% increase in political interest.
In short, there is little evidence that Democrats were “dispirited” in 1994, much less by the failure of health care reform. Of course it’s good for Obama if health care reform passes, but we shouldn’t assume that its passage (or failure) will actually affect the 2010 election.
I’ve been reading Dan Balz and Haynes Johnson’s The Battle for America 2008 (here), about which I’ll have more to say. Here’s one juicy tidbit that I didn’t previously know. In the various phone calls that led up to Ted Kennedy’s endorsement of Obama, there was this one:
In late December, a 2003 tape recording of Obama made while he was still in the Illinois Senate became public. It was a comment on Kennedy’s efforts to pass a prescription drug bill. Obama had described Kennedy as “getting old and getting tired” and said the backers of a strong prescription drug bill should go after him. Obama called Kennedy to make amends. “Well,” Kennedy said when he picked up the phone, “you start the conversation.” Obama began to grovel, but Kennedy stopped him. He would let Obama off the hook, he said gently, because he had once mangled Obama’s name in a speech at the National Press Club the month Obama was sworn in as senator, calling him “Osama bin Laden” before finally stammering out his right name.
The Boston Review solicited several replies to Andy’s and my piece on the 2008 election. Those are now online.
The replies are from Michael Dawson, Richard Johnston and Emily Thorson, Rick Perlstein, and Mark Schmitt. They all pushed us to think more about our account, and they all are well worth reading.
Here is our response to these replies. One tidbit, prompted by Rick Perlstein’s generous call for more political scientists on TV news:
Will these efforts [to enter the public sphere] get political scientists invited to Joe Scarborough’s kaffeeklatsch? Probably not. The media ecology fetishizes novelty in reporting and certainty in commentary. And yet the academic study of elections shows that what is certain is almost never new, and what is new is almost never certain. We might only bore Fox & Friends with our scholarly qualifications and caveats, or simply look foolish trying to present our research in soundbites…That said, we are willing to risk it if we can take airtime away from yet another discussion of the Bradley effect. Wolf Blitzer: have your people call our people.
Many have noticed, and been surprised by, the partnership between the SEIU, Pharma, the AMA, and a few others to support health care reform. But is this a surprise? Some new research (gated; ungated) by Matt Grossmann and Casey Dominguez suggests otherwise.
Patterns of interest group ties resemble two competing party coalitions in elections but not in legislative debate. Campaign endorsement and financial contribution ties among interest groups are consistently correlated but legislative ties do not follow directly from electoral alliances…interest groups have distinct incentives to join together in a party coalition in elections but also to build bipartisan grand coalitions to pursue legislative goals.
Grossmann and Dominguez argue that legislative process introduces multiple considerations that may push interest groups out of their electoral coalitions and into more diverse alliances.
In other words, because groups may support the same policy for different reasons, there is ample room for strange bedfellows.
And on the question of fractures:
The Democratic coalition is not fractured into many small constituencies. The Democratic campaign and legislative networks are denser than equivalent Republican networks, with a core of labor organizations occupying central positions.
Here’s the picture (click to enlarge):
Will Jennings sends along this article on issue ownership for governing and opposition parties in the UK.
Too bad we didn’t just have a national election where voters got to decide between health care reform or health care status quo. That would have really helped clear things up …
Of course, McCain didn’t endorse the literal status quo, but this is only a misdemeanor in comparison to the claims about a mandate. Let’s review why this is not the correct interpretation of the 2008 election. Here’s Andy and me:
But do we really know what people want just because they support one candidate over another? We cannot assume that people voted for Obama based on detailed knowledge of his policy positions…Instead people often choose a preferred candidate and then rationalize issue positions to fit this preference. So we cannot interpret an election outcome as a wholesale endorsement of the winner’s policy proposals (or as a wholesale rejection of the loser’s).
It’s also worth noting that anyone who did choose Obama because they wanted “health care reform” would be entitled to oppose the specifics that have emerged from Obama, the House, or the Senate.
Obama’s election does not signal the dawn of a postracial era in U.S. politics. Rather, it reflects the current structure of racial politics in the United States—a division between those who favor color-blind policies and seek to keep racial discussions out of politics, and those who favor race-conscious measures and whose policies are often political liabilities. The Obama campaign sought to win support from both camps. Only if pervasive material racial inequalities are reduced can such a strategy succeed in the long run.
That is Rogers Smith and Desmond King, writing in The Dubois Review. The article appears to be ungated. Here is one other passage of note, this on the “multicultural challenge”:
It is a challenge that goes to the heart of Obama’s core promise: to embrace the diversity of Americans and yet to find ways to “bridge our differences and unite in common effort—black, white, Latino, Asian, Native American; Democrat and Republican, young and old, rich and poor, gay and straight, disabled or not,” as he put it in his Ohio “closing statement” near the end of the campaign. All Americans are to come to feel and act politically as “one nation, and one people” who will together “once more choose our better history.”
One reason this promise is so challenging is Americans do not agree on what constitutes their “better history.” Some see the spread of religious diversity and considerable secularity, for example, as advances for freedom. Others see those developments as a retreat from the United States’ true calling to be a shining “Christian nation.” Some believe their country’s “best history” centers on the realization of ideals arising in historically Anglo-American cultural traditions. Others see those cultural traditions as historically responsible for the repression of communities and identities that they regard as most valuable and most their own. Put more broadly, the difficulty is that it may well be impossible to give any specific content to the putative shared, unifying values and purposes of Americans, without appearing to fail to recognize and accommodate adequately the diversity of values and purposes Americans in fact exhibit.
The second perspective comes from Susan Fiske and colleagues:
Images of Black Americans are becoming remarkably diverse, enabling Barack Obama to defy simple-minded stereotypes and succeed.
Fiske’s work emphasizes two dimensions of stereotypes. Warmth encompasses beliefs about a group’s intentions — e.g. is the group threatening, peaceful, violent, etc. Competence encompasses beliefs about a group’s abilities — e.g., is the group lazy, hardworking, intelligent, unintelligent, etc.
Obama’s credentials prevented him from being cast as incompetent, though the experience debate continued. His legendary calm and passionate charisma saved him on the warmth dimension. Social class subtypes for Black Americans differentiate dramatically between low-income Blacks and Black professionals, among both non-Black and Black samples. Obama clearly fit the moderately warm, highly competent Black-professional subtype.
Here’s a map of how people placed different groups on these two dimensions. Note the difference between the placement of poor blacks and black professionals.
See also their discussion of habituation:
Finally, non-Black Americans are habituating to Blacks in politics, and to Obama in particular. This habituation proved critical because of the automatic emotional reactions of many non-Blacks to Black people. Although many non-Black Americans have rapid, unconscious vigilance reactions to Black faces, these immediate reactions can habituate, with familiarity and empathy.
The Obama victory was historic, but it was not surprising; Obama shifted, but did not redraw, the electoral map; race and class mattered, but not in the way people assumed they would matter; and partisan loyalty was powerful even as partisan defections like Colin Powell’s garnered headlines. Furthermore, it is simply too soon to tell how and how much campaign tactics mattered, whether this election’s outcome constitutes a realignment in voting behavior, and whether Obama has emerged with a mandate.
Andy and I have a new piece on the 2008 election in the upcoming Boston Review, which is now available on-line. The paragraph above summarizes its basic conclusion.
The piece grew out of our collective blogging on the 2008 election. Essentially, we distilled our blog posts and related (numerous) graphs into a longish document, which was then made much more coherent by Joshua Cohen, Deborah Chasman, and Simon Waxman of the Boston Review. I learned two lessons in the process. First, if anyone seeks an outlet for scholarly research rendered in a “popular” form, I can highly recommend the Boston Review. It is quite a pleasure to work with real editors who care about the social science and also can make it readable, as Josh, Deborah, and Simon do and did.
Second, for me, writing this piece confirmed even more the value of blogging. Thanks to this blog and Andy’s own blog, we could not only publish our interpretations of the campaign and the election outcome in “real time,” but also produce a coherent summary of those interpretations more easily than if we had written this article entirely from scratch. In short, blogging creates certain efficiencies: you fire off a quick posts and then over time those posts can accumulate into something more meaningful.
In the next several weeks, the Boston Review will publish several responses to our piece from other scholars, as well as reply from Andy and me. I’ll link to those when they are available.
The previous Moldovan parliamentary elections - held in April of this year - attracted quite a bit of attention in the blogosphere (including on The Monkey Cage) because of the use of Twitter by protestors following the election. As it turns out, Moldovans returned to the polls this Wednesday for another parliamentary election. I asked Grigore Pop-Eleches of Princeton University for his thoughts on the election, and he provided the following guest post. (Those interested in the implications of the election for the political science literature are encouraged to continuing reading to the last paragraph…)
Continue reading "2009 Moldovan Parliamentary Electons: Take 2" »
Nate Silver links to a Congressional Quarterly list of ratings for 2010 congressional races and concludes that, although these listings give a sense of which races are more likely to be competitive, the CQ chart doesn’t really say much about the chance that there will be a “wave” election that would switch partisan control to the Republicans.
The same day, Matthew Yglesias links to a recent Congressional Quarterly report entitled, “2010 House Outlook: Democrats Look Secure” and concludes that, yes, the Democrats look secure to keep their House and Senate majorities.
What should we believe? For the purpose of campaign strategy, you need to look at the races in each district, but to get a sense of what’s going to happen overall, I think the best approach is to look at the national vote. There’s lots of variation, but, overall, swings occur nationally.
Here’s a graph I made after the election, showing the average Democratic share of the two-party vote for the House of Representatives and for president for the past sixty years:
From this picture, it looks possible but unlikely that there will be a 6% swing toward the Republicans (which is what it would take for them to bring their average district vote from 44% to 50%). Historically speaking, a 6% swing is a lot. The biggest shifts in the past few decades appear to be 1946-48, 1956-58, and 1972-74 (in favor of the Democrats) and 1964-66 and 1992-194 (for the Republicans). I don’t know if any of these would quite be enough to swing the House majority. A more likely outcome, if the Republicans indeed improve in next year’s election, is for them to make some gains but still be in the minority.
The other factor helping the Democrats is incumbency, which helps lock in a congressional majority (as it did for the Republicans after 1994) by bumping up the vote shares of the new congressmembers elected in swing districts. In 2008, John Kastellec, Jamie Chandler, and I estimated that the Republicans would need something like 51% of the average district vote to have an even shot of winning a majority of House seats.
Continue reading "What might happen in the 2010 House elections" »
John links to this quote from Barney Frank:
Not for the first time, as a — a — an elected official, I envy economists. Economists have available to them, in an analytical approach, the counterfactual. Economists can explain that a given decision was the best one that could be made, because they can show what would have happened in the counterfactual situation. They can contrast what happened to what would have happened. No one has ever gotten reelected where the bumper sticker said, “It would have been worse without me.” You probably can get tenure with that. But you can’t win office.
I have two thoughts on this. First, I think Frank is a bit too confident in economists’ ability to “show what would have happened in the counterfactual situation.” Maybe “estimate” or “guess” or “hypothesize” would be a bit stronger than “show.” Recall this notorious graph, which shows the unintentional counterfactual of some economic predictions:
Second, I don’t know how Frank can say that about “no one has ever gotten reelected . . .” In Frank’s district in Massachusetts, it would take a lot—a lot—for a Democrat to not get reelected.
Ed Kilgore writes:
…one talking point heard often in denunciations of Democratic foot-draggers on health care is that as “everyone knows,” the failure to enact health reform in Bill Clinton’s first two years caused the Democratic midterm debacle of 1994.
…I have to say that no, it’s not at all self-evident that the failure of a Democratic-controlled Congress to enact universal health coverage was the primary cause. For one thing, there was a lot going on in November of 1994—a vast number of Democratic retirements, the final stage of the ideological realignment of the South (exacerbated by racial gerrymandering in the House), and residual resentment of a Democratic majority in the House that had been in place since 1954.
I agree that the failure of health care reform wasn’t really important to the Republican takeover. We should remember the broader climate in the country in 1994. I’ll quote from Gary Jacobson:
In 1992, 79 percent of the voters in the national exit poll thought the economy was in bad shape, and 62 percent of them voted for a Democrat for the House. In 1994, 75 percent said they were no better off financially than they had been two years ago; 57 percent thought the economy was still in bad shape, and 62 percent of this group voted for the Republican.
In addition, he notes that 57% of exit poll respondents thought the country was on the wrong track, and they tended to vote Republican.
Of course, it wasn’t just national issues that mattered. Local dynamics were also important. Jacobson again:
In sum, although all politics was not local in 1994, the electoral effect of national issues varied across districts and regions, depending on incumbency, the quality of candidates, the level of campaign spending, the partisan makeup of the district, and the behavior of the incumbent. Wielding potent campaign themes drawn from national issues, Republicans needed only reasonably attractive and well financed candidates to take seats from Democrats in districts that leaned Republican in presidential elections. Democratic retirements from such districts created a host of open-seat opportunities that well funded Republican candidates exploited to the hilt. Against incumbent Democrats, Republicans did best where they fielded experienced, well financed challengers against Democrats whose votes tied them to the Clinton administration in districts where the president (and his party more generally) were relatively unpopular-where, in other words, the local campaign could give their national themes the most extensive publicity, and the local context gave these themes their greatest resonance. Poorly funded Republican challengers were largely unsuccessful, even in districts where the Democrat should have been vulnerable. The central issues in the 1994 House elections may have been national, but how they played out depended strongly on local circumstances.
To the extent that we can attribute the Republican takeover to national issues rather than local dynamics, it’s more about the economy and the accompanying generalized dissatisfaction than any specific policy failure of Clinton’s first 2 years. The same will be true for Obama and the Democrats in 2006.
Chris Good at the Atlantic Monthly expands on Conor Clarke’s case (discussed by John below) against opinion polls.
At the Atlantic Special Ideas Report, Conor Clarke makes a case against polling, for, among other reasons, polls’ ability to influence mass opinion by reflecting it, accurately or inaccurately, and to effect a herd mentality: … Polls illustrate, in other words, the power of perception in politics. Fatigue sets in when one’s favorite candidate is down. If he’s down ten points, why even take the trouble of voting? In that sense, polls corrupt the experiment of an election by suggesting results beforehand; for all the statistical science that goes into them, they’re fundamentally anti-scientific. And perhaps more significantly, Conor suggests, people can change their opinions to side with the front-runner.
I’m not at all sure that I buy ‘they’re fundamentally anti-scientific’ bit - elections, despite Good’s metaphor, are not experiments, and are not supposed to be. But Good’s underlying animus is, I suspect, a reasonably common one. Good believes that voters should vote on the basis of their actual preferences for candidates, rather than their perception of whether other voters support the candidate in question. Opinion polls - because they provide evidence of others’ voting intentions - make voters more likely to be swayed into voting for candidates who would not be their first choice in an ideal world.In the language of political science, voters should vote sincerely rather than strategically.
This is not a ridiculous position to hold - as the social choice literature makes clear, strategic voting can lead to all sorts of indeterminacies, possibilities for manipulation via agenda control etc. And that’s not even to get into the social psychological literature on tendencies to group conformism etc. But there is a normative case to be made for strategic voting too. Sunshine Hillygus has a nice article on strategic voting in the 2000 Presidential election. She finds strong evidence that supporters of minority parties (in this case, Ralph Nader) are more likely to vote strategically, changing their vote to support the candidate closest to their own views (for the most part, Gore) who has a decent shot at winning. High information voters (as best as the data indicates) were more likely to switch support from Nader to Gore, as were voters in swing states. But clearly, a number of people did not vote strategically (according to the usual political science definition of the term) in this election, since Nader had no realistic chance of winning, instead expressing their sincere preferences for a no-hoper candidate.
those supporting Nader for expressive reasons were undeterred by the wasted vote appeal, and were the most likely to remain loyal. These voters might not be voting so as to alter the outcome but instead to send a message or signal. For these Nader supporters, they might consider their vote to be much like cheering for their favourite team in the stadium. If a voter views the vote as an end in itself, then there is little reason to respond strategically to the electoral environment.
Nader’s vote, as is well known, collapsed in the 2004 presidential election. There are many possible explanations for this collapse. But casual empiricism would suggest that one plausible reason why left-leaning 2000 Nader voters were unlikely to vote for him again in 2004 was that they felt they were badly burned by the 2000 experience - what had seemed like a good opportunity to signal their sincere preferences turned out to have momentous consequences for the governance of the country. This is, of course, highly unlikely to have been the only reason for the precipitous drop in support for Nader, but it does suggest that sincere voting may have its downside in a winner-take-all electoral system. And opinion polls - by providing information on who is likely to win or to lose (especially in multiple candidate primaries and the like) can allow voters to cast their votes in more efficacious ways.
A few weeks ago, I posted about an interesting paper by Richard Johnston and Emily Thorson. They found that McCain’s poll numbers tracked Palin’s favorability in an uncanny fashion:
Judgment on her was incontestably important. The correspondence between dynamics in her ratings and dynamics in McCain vote intentions is astonishingly exact. Her marginal impact in vote-intention estimation models dwarfs that for any Vice-Presidential we are aware of, certainly for her predecessors in 2000 and 2004. And the range traversed by her favorability ratings is truly impressive. But why? We are unaware of any theory that opens the door to serious impact from the bottom half of the ticket.
See the post for the graphs. Palin’s resignation got these graphs some penetration in the blogosphere (Matt Yglesias, Ezra Klein), which prompted Seth Masket to offer a different interpretation. Here are his thoughts and my responses.
I’m going to somewhat disagree with this [Johnston and Thorson’s] interpretation of the graphs. For one thing, it shouldn’t be a shock that a presidential candidate’s approval rating moves in tandem with that of his vice presidential candidate. If you’re going to vote for one, you’re going to vote for the other — people know that. (Indeed, it’s more surprising that Obama’s and Biden’s ratings don’t jibe more.)
I think it is more surprising. Vice-presidential candidates don’t have a consistently strong impact on the evaluations of the presidential candidates. Palin’s impact is unusual — certainly in light of 2000 and 2004, as Johnston and Thorson note — but also in light of the past 50 or so years of presidential elections, according to some other research I recently reviewed. Simply put, the media paid an extraordinary amount of attention to her, which easily could have made the public’s feelings about her more important than has historically been the case.
Second, while we would expect the incumbent party’s candidate (in this case, McCain) to bear the blame for a sour economy, we wouldn’t necessarily expect voters to make that connection overnight. It was a confusing and very sudden collapse in the financial sector that caught the public’s attention in mid-September. It probably took a few weeks of campaign noise and media coverage for voters to apply the information to the political sphere.
I think Seth is basically right here, although I would say that the public was already paying a lot of attention to the economy even before the financial collapse. In an August 2008 NBC-WSJ poll, 49% said that economic issues such as job creation, gas prices, and the mortgage crisis should be the highest priority of government; an additional 14% cited health care, which also manifests economic concerns. Only 37% cited non-economic issues, such as Iraq, terrorism, and immigration.
If the financial collapse mattered, it didn’t matter by in a strict mechanical way, with each bank failure or drop in the Dow causing spikes in Obama’s poll numbers. Phil Klinkner pointed this out soon after the election. My graph above denotes some of these events, along with the announcement of the vice-presidential candidates, the onset of the conventions, and the four debates.
If the financial collapse mattered, it did so in the way that Seth suggests — by continuing to heighten the salience of the economy such that it became a more important factor in vote choice. This in turn would have pushed the candidates’ poll numbers toward the “equilibrium” predicted by the candidates’ positions — a la the accounts of Tom Holbrook as well as Andy and Gary King.
But there is still this question of the McCain plunge in October. Here is Seth:
Finally, note where the McCain plunge begins — right around October 8th…What happened that week? Well, October 7th was the night of the second presidential debate, the town-hall format one that included McCain’s famous “that one” reference to Obama. That debate turned mostly on economic matters, with the candidates charting out different positions on how to rescue the economy. Sixty-three million people watched the debate - a 21% increase over the first Obama-McCain debate and the biggest viewership for a presidential debate since 1992.
So here’s my interpretation: By the 2nd week of October, voters recognized the economy was tanking. The high public attention on the 2nd debate gave voters an opportunity to apply their knowledge of the economy to the political world, and they chose to blame the economy’s collapse on the incumbent party. Palin was along for the ride on this. Her star was tied to John McCain’s, and his to the interaction of party and economics.
If you look at my graph above, you can see the dip after the Oct. 8 presidential debate. (“Plunge” is perhaps too strong; it’s a shift of a only few percentage points.) So Seth is right: it is tough to say whether the dip in Palin’s evaluations documented by Johnston and Thorson is causing the McCain dip, or the debate is causing the drip (and possibly the dip for Palin as well).
However, I’m less certain that the debate mattered by focusing on the economy. For one, Obama’s numbers don’t really increase after the debate. Shouldn’t he benefit? And McCain’s numbers bounce back a little bit after the debate. Why? If the debate somehow helped voters blame the Republican Party for the bad economy, shouldn’t this suggest a final decision of sorts? If it was the debate that caused the dip, its apparently temporary effect suggests, to me at least, something tied more specifically to McCain’s performance in that debate (i.e., “that one,” etc.).
Of course, all this is speculative. We’ll be seeing more and better research soon. In the meantime, if anyone has any suggestions for the graph above, let me know.
In comments to my earlier post, Monkey Cage reader Jim notes that his analysis of the 2008 American National Election Studies finds a significant relationship between education and feelings toward Palin, controlling for party identification. He sent me his models via email, and they look interesting. Especially noteworthy is that the model of attitudes toward Palin looks much different than a similar model of attitudes toward McCain. I wanted to dive in, especially because Jim’s analysis is in tension with the Pew data I analyzed in the post and seems to support Ross Douthat’s contention that Palin’s appeal is class-based. Here is some more analysis.
Like Jim, I measure attitudes toward Palin and McCain with the “feeling thermometer,” a 0-100 scale where 0 signifies very negative attitudes, 100 signifies very positive attitudes, and 50 signifies a neutral attitude. Below are graphs that plot the mean thermometer rating for each candidate across educational categories, with separate lines for each partisan group. (Independents who lean toward a party are counted as partisans. For this group, I combine those with college and graduate degrees because there are few respondents in the latter category.) By plotting each partisan group separately, I am allowing for the effects of education and party identification to “interact.”
Above is the graph for Palin. Obviously, Republicans have much more positive feelings toward her than do Democrats. But the key question is, does education matter?
The short answer is: yes, but only for Democrats, and then only somewhat. Republicans of every educational background tend to like her. Independents are pretty neutral, no matter their level of education. Simply put, among this large fraction of the population, Palin’s appeal (or lack thereof) has nothing to do with class, at least as captured by formal education.
Democrats with no college degree tend to be unfavorable, with averages in the 40s. Democrats with at least a college degree are even less favorable, with averages in the high 20s. Overall, there is about 12-point gap between Democrats with and without a college degree. This direction of this relationship confirms Douthat’s account, but the magnitude of the relationship is modest at best. And this relationship shouldn’t conceal this basic fact: Democrats tend not to like Palin, no matter what. This is confirmed in the more recent Pew data, which found that only 24% of Democrats had a favorable opinion of her.
But, as Jim’s analysis also found, education has even less to do with evaluations of McCain. The graph above shows this pretty clearly.
So Palin’s “appeal” is different than McCain’s. But is it linked to class? I remain unconvinced.
(For more, see Mark Blumenthal and Jennifer Agiesta.)
In a recent Pew poll, 44 percent of Americans regarded Palin unfavorably. But slightly more had a favorable impression of her. That number included 46 percent of independents, and 48 percent of Americans without a college education.
That last statistic is a crucial one. Palin’s popularity has as much to do with class as it does with ideology. In this sense, she really is the perfect foil for Barack Obama. Our president represents the meritocratic ideal — that anyone, from any background, can grow up to attend Columbia and Harvard Law School and become a great American success story. But Sarah Palin represents the democratic ideal — that anyone can grow up to be a great success story without graduating from Columbia and Harvard.
So says Ross Douthat in the NY Times. I have no idea whether Sarah Palin is a “great success story.” I guess getting elected governor and being nominated for the vice-presidency counts as success, although some might say it’s how you do those jobs rather than whether you simply hold them. And I’ll leave it to political theorists to parse Douthat’s contrast between meritocracy and democracy, and whether or how Obama and Palin somehow embody those ideals. (A lot of Obama’s story seems to me a “democratic” story, but that’s a separate post.)
Instead, I’ll just do the boring old social science thing and note how badly Douthat misuses polling data. Douthat wants to claim that Palin’s appeal is class-based. To back that up, he cites how favorably she is viewed among Americans without a college education. His view of that figure (48%) is a little opaque. I read him as saying something like, “It’s noteworthy that slightly more Americans have a positive impression of Palin than have a negative impression, and those positive impressions seem even more prevalent among Americans without a college degree.”
So here’s my point. If you want to make the argument that Palin’s support is based on class, and that “lower class” people have a more favorable view of her, it really really helps if people of different class backgrounds have different opinions of her.
And they really really don’t. Douthat cites these Pew data, but fails to tell you how people with some college or a college degree view her. It turns out that there are extraordinarily small differences based on education background. Here’s a graph:
That 7-point gap between the views of the college educated and the high school-educated implies a tiny class cleavage at best. Douthat’s rendering of Palin is not reflected in how the public sees her.
Two decades have now passed since John Petrocik introduced the concept of “issue ownership” into the lexicon of students of political campaigns. The basic idea of issue ownership is that voters associate certain issues with certain parties. In the U.S., for example, tax cuts would generally be considered a Republican issue, and environmental protection a Democratic one. The strategic implication is that parties should emphasize the issues that they “own” and should avoid playing on their opponents’ turf.
But how stable or enduring is a party’s ownership of an issue? That question is at the heart of Stefaan Walgrave, Jonas Lefevere, and Michiel Nuytemans’s experimental study (abstract here) of the ability of parties to use media appearances to create new issue turf for themselves and to protect the turf they have already staked out.
Rather than describing the entire study, I’m going to concentrate on the portion that I consider most interesting. This is the idea that parties can gain most by focusing on issues that no other party owns, can gain less by focusing on other parties’ issues, and can gain least by focusing on issues they already own; that last part may seem counterintuitive, but the idea is simply that when people already agree with you, there’s little to be gained by trying to persuade them.
Walgrave and his colleagues embedded an experimental component within a very large (n=more than 11,000) but nonrandom panel survey of Belgian voters. The experimental stimulus was a phony news item inserted into a real excerpt of the most popular television news program in Belgium. Almost 5,000 respondents were exposed to the fake story. These were respondents who volunteered to watch the news program; they weren’t randomly assigned to the treatment condition. In the excerpt, the news anchor introduced a party leader whose statement started with “The point of view of [the party] on [some issue] is that…” The five party leaders who were shown were real; they had agreed to participate in the study, and the positions they described for their party were real, too. In all, each party leader recorded separate 30-second statements on six different issues.
Respondents rated parties on 0-10 scales according to their competence to determine policy on the issue about which the party leader had spoken. When the leaders discussed an issue that no party owned, their party registered a significant credibility gain. When they addressed an issue owned by another party, they also profited, but not by as much. And when they talked about issues their party owned, they made no progress, though the results contained some evidence that they may have limited the losses that their party would otherwise have suffered. In sum, the results, though containing considerable noise (welcome to social science), were basically consistent with the idea that the researchers set out to test.
Along with recent non-experimental research, these results and others reported by Walgrave et al. on which I haven’t focused have some striking strategic implications. Most importantly, perhaps, “trespassing” on other parties’ issues not only does happen, but can be successful – especially for parties that receive a good deal of play in the media; for parties that have difficulty commanding attention, though, the superior strategy may be to stick to their own issues lest another party make off with them. It’s also known that much issue trespassing occurs toward the end of a campaign; early in a campaign, parties tend to concentrate on solidifying their base, but later on they try to reach out to supporters of other parties. But the authors are skeptical of the merits of this strategy; it’s fairly easy, they argue, to move up in the rankings on an issue, but it’s very difficult to dethrone the issue owner.
All in all, a pretty cool study – not a perfect research design by any means, especially in its heavy reliance on nonrandom assignment to the various experimental conditions, to the experimental component of the study, and to the study itself. But cleverly executed (wouldn’t the rest of us love to be able to use real-life party leaders in our experimental research?) and centered on issues that are important theoretically and have some important real-world implications.
Tyler Cowen writes:
Median voter theorem. It’s my [Cowen’s] first-cut account of a lot of what is going on in the newspaper headlines. Yet somehow I rarely see it mentioned, even when I read very prominent social scientists commenting on current policy.
I can see how an economist would think of the median voter theorem as central, as it’s the most economics-like of political laws. That and campaign contributions are the two explanations that best align with economic models.
The median voter theorem has limitations, though, which are essentially quantitative rather than quantitative. I’ll give the story, but first the graph:
Here’s the positive statement of the median voter theorem. A politician is trying to get elected will probably get more votes (all else equal) if he or she is a centrist rather than far to the left or the right of the majority of the voters. Similarly, if you’re trying to push a bill through Congress, to first approximation you can think of the legislators as aligned on a left-right axis, in which case if you can get the median congressmember (and everyone to the left or right of him or her) on your side, you’re golden.
Certainly the median congressmember is important: by definition, it’s that marginal vote you need to get a majority. But where do the median congressmember’s positions come from? Not necessarily from the median voter in his or her district. My research with Jonathan Katz (see the graph above), suggests that being a moderate is worth about 2% of the vote in a congressional election: it ain’t nuthin, but it certainly is not a paramount concern for most representatives. (The graph appears in chapter 9 of Red State, Blue State. If you’re interested in the median voter theorem and U.S. politics, I recommend that whole chapter, actually.)
I am sympathetic to Cowen’s larger point (also made by Matthew Yglesias), however, which is that it might be a mistake to assume that politicians of your political party agree with you, deep down, on the issues, and that they’re only voting differently because of expedience, craven political calculation, or whatever. It’s worth considering the hypothesis that lots of Democratic politicians do not share the values and policy preferences of lots of Democratic voters, and similarly for the Republicans. Given the diversity of public opinion, this really has to be true on some issues, and it very well might be true all over the place.
This last point, of course, is completely consistent with the idea that the median voter theorem is a “weak force” with much less importance than might be assumed from a casual examination of the political system.
Physics envy should now, I hope, lead us to discover political forces of gravitation, electromagnetism, and the rest. Only the political equivalent of string theory can unify all this. I’m sure it will turn up on the Arxiv soon.
Mark this down as a prediction: when scholars start writing articles about the recent Iranian presidential elections, this will be the quote they use to begin their articles:
“Statistics provided by the candidates, who claim more than 100 percent of those eligible have cast their ballot in 80 to 170 cities are not accurate — the incident has happened in only 50 cities.” - Abbas Ali Kadkhodaei, spokesman for the Guardian Council, as quoted in the NY Times
With the NY Times now reporting that the number of votes affected by regions where the Guardian Council admits that greater than 100% of voters cast ballots could number as many as 3 million (out of a total of 40 million total votes) from 50 cities, it is clear that this was not an isolated incident. So it seems legitimate to begin asking how this could have possibly happened.
To me, the most likely explanation would seem to be an either literal or virtual form of “ballot stuffing”, e.g., Ahmadinejad’s votes being inflated without any of the other candidates having their votes deflated. Hence, if enough stuffing occurred, the totals could eventually begin to surpass the number of eligible voters. In a literal sense, it is easy to understand why one might “stuff” but not “destuff”. Adding extra ballots is easy: you just get hold of extra blank ballots, fill in your candidate, and put them in the ballot box. “Destuffing”, on the other hand, involves getting hold of ballots that were already cast, filtering them to remove only votes for the opposition, and then getting the ballots for your candidate back into the ballot box. Not impossible, but clearly a lot more work than stuffing, especially when time is short.
But if, as my colleagues Bernd Beber and Alex Scacco have suggested (here and here), the ballot stuffing was more “virtual”, with totals just being adjusted at higher levels of aggregation, then the question remains: why would totals ever be adjusted beyond the number of votes counted? Is there something that keeps people who are fixing election results from wanting to adjust opponents’ results downwards as opposed to just revising their own candidate upwards? One explanation could be laziness: you are ordered to get your candidate a certain number of votes, and the easiest way to do that is just change his/her vote total. Another explanation could be fear of a paper trail: if you decrease votes at a reporting stage and the ballots actually exist, then perhaps there is more of a fear of getting caught? I’m interested in others’ thoughts on this matter, especially those who have worked specifically in the area of election monitoring and fraud detection.
All of this, however, bring me to the issue of election boycotts, which have have been used as an important tool of protest in cases where the question of whether elections will be free and fair is in doubt. If, however, it turns out that it is kind of an iron rule of election fraud that it is easier to revise vote totals up then down, then it seems that not boycotting an election serves another role besides allowing your candidate to accumulate votes: it also narrows the window in which the regime can inflate vote totals for its own candidates without getting “caught” by going over the number of actual votes count.
Put another way, one reason why we might now have 50(!)+ cities in Iran where the number of votes cast exceeds the total number of voters is precisely because Iranian citizens participated in the election in such record numbers. Had turnout been a bit lower, there would have been more margin for error in this regard, and perhaps less evidence of vote totals exceeding voters.
Alex Scacco and Bernd Beber follow up on their analysis of the Iran election data:
After we wrote our op-ed using the province-level data, we’ve now also done some preliminary tests with the county-level data. In the latter dataset, the last digits don’t appear fraudulent. Why might we find suspicious last digits at the province level, while, at the same time, Walter Mebane and Boudewijn Roukema find evidence that first and second digits are fishy at the county level? We can only speculate about what happened behind closed doors, but here is a scenario of top-down fraud that is consistent with the patterns found in the quantitative analyses mentioned above:
Continue reading "Combining findings at the Province and County Level from Iran's Election" »
From my new colleagues Bernd Beber and Alex Scacco:
In the past week, analysts have scoured the results from Iran’s presidential election, looking for evidence of fraud. In an op-ed in today’s Washington Post online, we offer a different take on the problem. The key idea in the piece is that people are poor randomizers: When humans try to fake numbers, they leave traces of their activity in the data. For instance, psychologists have found that people choose some digits more often than we would expect in a sequence of random numbers.
What distinguishes our approach from Walter Mebane’s note (discussed in a previous Monkey Cage post) is that we look at patterns of last digits in province-level vote returns. In a fair election, each number (0, 1, 2, etc.) should appear as often as any other in the last digit. But that’s not the case in the numbers from Iran: Among other things, wefind too many 7s and too few 5s. The deviations we find suggest there’s little chance the Iranian election results weren’t manipulated.
You can find the full op-ed here, as well as supplementary materials (including the annotated version of the op-ed and the data and code used in the analysis) here.
Stephen Hawking once wrote a paper where the title of the first slide (I write from doubtless imperfect memory) was something like “On the Breakdown of Physics in the Region of Space-Time Singularities.” The title of the second was “On the Breakdown of Physicists in the Region of Space-Time Singularities.” Ethan Zuckerman has a smart post which suggests that the real Twitter in Iran story is about the physicists rather than the physics. The interesting social phenomenon is not the use of Twitter by Iranian dissidents, but the media’s collective convergence on framing the story of what is happening in Iran around Twitter.
One of the reasons MSM outlets are so focused on social media is that they’re not able to deploy reporters to cover these protests. In some cases, the majority of reporting from the ground is coming from social media. It’s worth asking what the biases might be in amplifying those social media reports. Ahmedinejad’s supporters tend to be poorer, more rural, less educated and more likely to speak Farsi than Mousavi’s supporters - a picture of the protests via social media runs the danger of overstating Mousavi support or minimizing Ahmedinejad support.
Iran is one of the countries American and British media pay closest attention to. The use of social media for protest - especially to promote a protest to international audiences - is far from unique. But because there’s such strong media focus on Iran, and such interest in the use of social media for protest, this is a perfect storm for interest in this topic. …
I’ve been asking some of the reporters I’ve spoken with where they were on other recent social media and protest stories. Citizen media has emerged as one of the key spaces for journalism in Fiji in the wake of a coup government that’s censoring mainstream media. It’s been a key source of information in Madagascar as that country’s suffered through a violent change of government. (One reporter who I mentioned this to remarked that Madagascar was “just a speck of an island somewhere”. That speck is twice the size of Great Britain and has the population of Australia…) …
I’ve written at some length about homophily, the tendency of birds of a feather to flock together. Turns out that reporters flock, too. It’s somewhat amazing to me the extent to which reporters from really good newspapers are all asking the same questions. I’m glad that people are taking a close look at the phenomenon of social media in the Iranian protests - it’s an important, fascinating and worthwhile topic. But there’s a lot of topics out there, and I wonder whether we benefit from a thousand well-researched stories on this phenomenon rather than a hundred, and nine hundred other stories.
Ethan’s own research is on global attention profiles - the differences in the relative amounts of attention that news sources pay to different parts of the world. This is a nice demonstration of how this may work in practice. Ethan also links to an overview of the debate by Gaurav Mishna which suggests that there are less than 10,000 Twitter users in Iran, and that only around 100 of them seem to be active (Mishna provides no source for the second figure - I presume that it is taken from his own research). Even if there turn out to be more Twitter users in Iran than Mishna’s figures indicate, I strongly suspect that his underlying claim is correct, and that Twitter has played little, and perhaps no role in actually organizing protests.
The model of electoral behavior that Anthony Downs popularized in An Economic Theory of Democracy produces a paradox. The chances that I will cast the deciding vote in any election involving more than a few voters are so small that it is irrational for me to devote any effort whatsoever to voting; I might just as well stay home and let others do the work, secure in the knowledge that the election will come out exactly as it would if I had voted. The paradox is that people do vote, opening a huge gap between what a “rational” person (at least according to a narrow definition of rationality) should do and what hundreds of millions of people sctually do.
One possible reconciliation is the “minimax-regret” principle that Ferejohn and Fiorina ( here, gated) dredged up long ago to try to rescue the rational model. The idea is basically that people are motivated above all else to avoid making a big mistake. Imagine, Ferejohn and Fiorina say, that “your” candidate has just lost by a single vote and you didn’t vote. You’d feel just terrible. To avoid any possibility of feeling terrible, then, you’re likely to vote, even though you understand that the event that would make you feel terrible is extremely unlikely.
And here’s where some recent news out of Arizona comes into play. Every now and then it happens that the Ferejohn-Fiorina scenario, which sounds pretty far-fetched, actually does happen, as it did recently in Cave Creek, Aruizona, when an election for city council produced 660 votes for the incumbent and 660 for the challenger. A bunch of Cave Creekers who didn’t make it to the polls presumably spent the next day beating up on themselves for not voting.
All was not lost for some of these benighted souls, however, for the tie vote still had to be decided one way or the other.If Cave Creek were in Minnesota rather than Arizona, they probably would have decided it by staging an endless procession of recounts until they got down to some arbitrary totals that the last court hearing the final appeal would sign off on in a fit of exhaustion. Well, Cave Creek isn’t in Minnesota, and out on the rugged frontier they have a simpler way of settling such disputes. No, it’s not six guns at 30 paces. It’s … cutting cards. Here’s the story, as reported in the New York Times.
All of which leads me to suggest a revision of the probability term at the heart of rational models of voter turnout. It shouldn’t be “the (very low) probability that I could cast the deciding vote for my preferred candidate.” Rather, it should be “the (very low) probability that I could cast the deciding vote for my preferred candidate plus the (.5) probability that my preferred candidate would win the card-cut given that I have cast the tying vote.” A small step for models of voter turnout, to be sure, but one I consider worth a lead article in the APSR (or on second thought, in the AJPS). This is major.
When you hire someone to do a job for you, that arrangement is likely to work best if the two of you have the same goals. Or if your goals differ but are compatible, things are likely to work out pretty well, too. But if your goals are incompatible, then you’ve got a problem – a principal-agent problem.
In a recent analysis (abstract only) of political campaign consulting as an industry, Matt Grossman identifies just such a situation.
Consider how political consultants are compensated. The most common arrangement is for their pay to be based on the amount of service they provide to the campaign (producing and purchasing television or radio ads, carrying out direct mail campaigns, conducting surveys, and so on), as indicated by their total expenditures. Another common arrangement is a flat fee: a firm agrees to perform certain services for an amount that’s agreed-upon in advance. Somewhat less common, but still widespread, is a victory bonus, which can be added on to either of the first two arrangements. Finally, consultants might agree to be paid only if their candidate wins – a contingent-fee structure like that often used by personal-injury attorneys; that’s a rare arrangement for, in Grossman’s words, “consultants are not willing to bet the firm on a win guarantee.”
It is heavy reliance on the first arrangement – tying consulting fees to expenditures – that produces the potential principal-agent problem. If consultants were compensated according to the same logic that holds for candidates (only winning produces the desired payoff), the goals of the candidate and the consultants would be identical or at least compatible. But contingency fee arrangements, as just noted, are rare. As Grossman puts it, “Candidates get a job only by winning the most votes whereas consultants generate more revenue primarily by having their client spend the most money on their campaign.”
Consultants appear to have two sets of incentives associated with their revenue streams. First, they have a direct economic interest in having candidates spend money, especially on television [which is extremely costly]. This may come at the expense of direct mail, Internet campaigning, and GOTV drives. Second, since expenditures are likely tied to contributions, consultants also have an incentive to see that more money is raised. Rather than save money early in a campaign, consultants may be better off spending it and hoping that their candidates can raise more later.
This point shouldn’t be overstated. Grossman notes that consultants have to be concerned about their reputation, and if their actions generate suspicion that they are motivated primarily by a desire to pile up huge fees for themselves, their future revenues and the viability of their firm may suffer. Still, the economic incentive structure does open a potential conflict of interest between principal and agents.
Consultants are often implicated in the worst public complaints about modern campaigns. Their answer is typically to suggest that they work in their candidate’s interest, implementing what works. If business incentives create a division in client and consulting firm interests, however, we need to subject that claim to scrutiny.
Political consultants are often depicted as puppet masters who deftly manipulate the strings that enable candidates to dance to a particular tune. Grossman’s research, by emphasizing the possibility that the tune may be one that suits the puppet master’s interests more than the puppet’s, amounts to a call for further research on the business side of political campaigns and the difference it makes.
Despite facing some significantly high hurdles, which I have written about here, supporters of opposition leader Mir Hossein Moussavi are continuing to take to the streets and apparently to exert real pressure on the Iranian regime. In a somewhat stunning development, the Iranian Supreme Leader, Ayatollah Ali Khamenei, has endorsed a partial recount of the vote, although Mousavi has apparently rejected the offer of a recount, instead demanding a revote. As developments proceed in the next few days, I would keep a close eye on the following:
1) Will the planned recount go ahead, and or will Mousavi succeed in forcing a revote? While the latter still seems very unlikely, it would be an extremely significant development, demonstrating that the position of Khamenei is much weaker than we thought only a few days ago.
2) Will the security forces in Iran move beyond what they are doing now – basically low levels of violence and apparently detaining opposition leaders – and unleash a more concentrated show of force (e.g., something in the spirit of what happened in Tiananmen Square or Andijan, Uzbekistan)?
3) Will the Iranian authorities take further steps to shut down technological means of communication being utilized by the opposition, such as text messaging and Twitter?
4) Can the opposition continue to deliver large numbers of people into the streets as the protests head towards a second week? One of the truly fascinating things about the Orange Revolution in Ukraine was the sheer number of days that people continued to protest, even in the dead of winter. It will be very informative to watch if the protests in Iran have similar staying power, although of course this likely to be in part a function of the previous two points.
All of these points are informed by my view of thinking about protests following electoral fraud in terms of the individual protester, trying to ask when protesters will believe it is in their interest to take to the streets and when they will not. I have written about this in great detail in Perspectives on Politics (ungated).
On a closely related topic, John Rood has responded to my query regarding Just What is Iran? by writing:
Thanks for your piece via The Monkey Cage on Iran as “competitive authoritarian” regime. I thought the analysis could go a step forward by subdividing into what I’ve called “strong” and “weak” competitive authoritarian regimes,. The distinction is that in a weak regime, the supreme authority is occasionally compelled to accept election results with which it disagrees in order to maintain their power. (A strong regime is sufficiently powerful that it can invalidate election results at its discretion.) You mentioned that Khatami’s election was one example; the situation on the ground now suggests that there is a similar inflection point re: reviewing last week’s results.
John has more on this topic in a blog post at The Public Philosopher.
Drawing together the themes of protest, Twitter, and Iran that have dominated the Monkey Cage in the past few days, I published an Op-Ed piece today over on The New Republic online on lessons from the post-communist colored revolutions that may have been learned by the authorities in Iran. Not sure exactly what the blogging etiquette is on this sort of thing, but thought I would share the first two paragraphs here with readers of the Monkey Cage and then direct those of you that are interested to TNR to finish reading it:
As I watched events in Iran unfold at the end of last week, I couldn’t help but note the similarities to the “Colored Revolutions” that swept through the post-communist region in the middle of this decade. Pre-election polls predicted a surprisingly competitive election in an erstwhile authoritarian country. Following the election, both sides claimed victory amid allegations of serious electoral fraud. Supporters of the opposition candidate took to the streets, and even had a color — green — lined up to give them the moniker of the “green revolution.”
However, over the past three days, it has become apparent that Tehran is not turning into Kiev. While there are numerous important differences between Iran and the post-communist colored-revolution countries (Serbia, Georgia, Ukraine, and possibly Kyrgyzstan)—with the most notable being that ultimate executive power in Iran lies with the Supreme Leader, currently Ayatollah Ali Khamenei, who is not popularly elected—it does seem to me that the Iranian authorities may have learned a number of specific lessons from their less fortunate post-communist counterparts.
Continue reading the rest of From Kiev to Tehran? at The New Republic.
As a follow up to my previous post, things are apparently starting to get very interesting in Iran, with both sides declaring victory amid extremely high turnout, even though results are just starting to come in. A few things to keep an eye on as this develops:
If the authorities declare Ahmadinejad the winner, will the opposition forces supporting Mousavi have enough hard evidence to claim that the results were falsified? The BBC is currently reporting that Mousavi has claimed that his election monitors were denied access to polling stations, and that he has complained of “voting irregularities”.
Wikipedia has a fascinating collection of pre-election polls posted here. I couldn’t figure out exactly how to copy the table into the post, but the numbers are all over the place. Polls released just in June have Mousavi getting anywhere from 28% to 64% of the vote, and Ahmadinejad receiving between 23% and 63% of the vote [Caveat: some of these polls are nationwide, some are from “major cities” only]. Slightly more polls have Mousavi ahead. Pre-election polls may come into play if massive fraud is alleged.
Remember that this is a two round election. So if neither candidate wins a majority of the election in the first round - and there are four candidates competing in the first round - then a second would be held next week.
If opposition supporters take to the street, a key question will be if the regime will try to use force quickly. This crucially did not happen in any of the successful Colored Revolutions, and I have argued elsewhere that this can play an important roll in determining the size of subsequent crowds of protesters. I’d be interested in the opinions of anyone who knows something about Iran whether they think we would see a quick show of force.
It will be interesting to see if Twitter plays any role in coordinating opposition action in Iran. I am already seeing a lot of posts on Twitter when you search for “Iran Election”, but am not sure what, if any, role Twitter might be playing within Iran. Anyone know anything about this? (For more on the potential role of Twitter in facilitation protest in the aftermath of electoral fraud, see here)
For those of you who became enthralled with the EU Parliamentary Elections due to The Monkey Cage’s coverage last week (here, here, and here), I wanted to bring to your attention that the results can now be found on the EU’s Website in 23 different languages. In addition, they have some very nifty graphical displays of the results here. CNN also has a nice, very concise, summary, of the results from about half of the countries. All in all, the NY Times reports:
Only 43 percent of Europeans voted — a record low turnout, despite the financial crisis and compulsory voting in some countries. Far-right parties, opposed to the European Union and to immigrants from poor member countries, recorded gains, as did the Greens. Those who did vote weighed in largely on national issues.
As an aside, I continue to wonder what it is that we are supposed to conclude from the fact that EU parliamentary elections had “record low turnout”. 43% of the population turning out for what is by all accounts a “2nd order” institution actually strikes me as a fairly high percentage of the population to vote in such an election. Ought we to expect EU parliamentary elections to attract as many voters as national elections? If the answer is no, then what is the correct benchmark? Off-year congressional elections in the US? State and local elections? Granted, there is obviously some interesting information in the fact that turnout has decreased in every year that EU parliamentary elections have been held, but at what point does lower turnout become a “problem”? Does lower turnout really signify a loss in support for the idea of Europe, or does it just reflect a growing understanding on the part of the European population that, at the end of the day, the EU parliament is not all that important a political institution? (And yes, this could still be the case despite the paradoxical fact that turnout has gone down as the actual powers of the EU parliament have increased.)
Taking up Andrew’s challenge of why anyone “over here” might care about the EU Parliamentary Elections, I’ll throw out those who are interested in the phenomenon of strategic voting. For those outside of the academy, political scientists use the phrase strategic voting to refer to instances when voters do not necessarily vote for their top choice to win an election (which is known, by contrast, as sincere voting) because they care about something other than simply if their candidate wins the election. The most commonly described form of strategic voting
