Which Economy is it, Stupid? Part III: Depends on whether you are an investor, home owner, or renter

by Joshua Tucker on May 2, 2012 · 0 comments

in Campaigns and elections,Political Economy

Following up on my recent posts (here and here) regarding which parts of the economy are more likely to affect voting behavior, I wanted to alert readers to a new paper by University of Minnesota political scientists Wendy Rahn and Philip Chen. Here’s the abstract from the paper:

Nearly three years after the official end of the Great Recession, the American economy is starting to show, at long last, some “moderate” improvement as the nation heads into a presidential election campaign in which the economy is likely to be a central issue. Yet some parts of the macroeconomy are more improved than others. Housing prices, for example, after having fallen more than 30% (or more, in certain hard-hit areas) off their 2006 peak, have yet to bottom out and it may be years before any significant price appreciation returns. The labor market, while evincing recently some signs of strength (though perhaps induced by better-than-typical weather conditions), is still anemic, and the employment-to-population ratio, which is considered by many to be a better barometer of labor market conditions, is still at historic lows. The U.S. stock market, on the other hand, recent wobbles notwithstanding, has been buoyed by rebounding profits in the corporate sector and unconventional monetary policy. However, it is unclear whether the Fed will engage in any more rounds of quantitative easing, and U.S. financial markets remain vulnerable to the European debt crisis.

American households are buffeted by these macroeconomic forces to different degrees, and when conditions in these various spheres diverge, as in the aftermath of the Great Recession, groups that are differentially affected may respond politically in ways that generate new lines of cleavage that add complexity to our traditional economic voting models. Using monthly survey data from the Michigan Survey of Consumers over the period 1992 to 2011, we examine the impact of unemployment, inflation, house and stock prices, and real income growth on people’s retrospective assessments of family financial well-being. Our innovation is to compare the effects of these variables for different groups of households defined by their asset-holding status: investors (directly or indirectly) in the stock market, homeowners without risky financial assets, and renters. We indeed find that people respond to aggregate economic conditions in heterogeneous ways. In particular, investors’ well-being is directly tied to movements in stock prices and is unresponsive to short-term fluctuations in the labor market. Homeowners and renters, on the other hand, are strongly affected by shocks to unemployment, with homeowners additionally showing sensitivity to trends in house prices. Interestingly, and contrary to several economic forecasting models, real income growth does not significantly influence any of these groups, a null result that we attribute to the fact that over the time period we study, the benefits of economic growth have accrued mostly to a very small segment of public. We conclude that economic voting models in use by political scientists need to be move beyond their traditional focus on growth, inflation, and unemployment to consider newer sources of economic vulnerability and their effects on political behavior and electoral choice.

Now, I want to be clear that what Rahn and Chen are studying here is not the direct relationship between economic conditions and vote choice. Recall that the original post by Nate Silver which motivated my response was looking at the macro-level relationship between economic conditions and election outcomes. Underlying this relationship has to be some sort of micro-level model, the simplest of which would simply be that worse actual economic conditions make respondents think their personal economic situation is worse, which in turn makes that individual more likely to vote against the president (or some other incumbent candidate). This is generally known as “pocketbook” economic voting; other scholars have argued that economic voting is more “sociotropic”, in so far as voters are concerned about the overall state of the economy, and not their own personal economic situation (although of course these can be and often are correlated.)

To the extent that one believes pocketbook economic voting is important or prevalent, then Rahn and Chen’s findings have important implications for how we think about the link between economic conditions and election results. (Moreover, even if one thinks voters are more sociotropic, I’d guess that the factors they identify also affect sociotropic evaluations as well.) Namely, the search for a “silver bullet” economic indicator (like Silver’s embrace of job growth in this post) is doomed to fail not simply because we lack the necessary data to test these hypotheses thoroughly, but because different parts of the economy may matter to different parts of the electorate. Rahn and Chen’s work suggests that this intuitively plausible hypothesis has legs; future research extending this intuition into models that explain actual voting choices would be an exciting next step.

Rahn and Chen’s full paper is available here.

0 comments

Is Obama More Popular Than He Should Be?

by John Sides on May 2, 2012 · 9 comments

in Campaigns and elections,Presidency,Public opinion

I tackle that question in a new post at 538.  The analysis involves constructing a model of presidential approval from 1948-2008 and forecasting values for Obama.  On average he is about 9 points more popular than the model would predict.  Out-of-sample predictions for Obama and past presidents are here (click to enlarge):

Among these presidents, only Reagan and George W. Bush have actual first-term approval ratings that exceed their expected approval ratings at a level comparable to or great than Obama.

If anyone wants to read the 538 post, which has more details, and leave me some reactions here, I’d be grateful.  This analysis will hopefully be part of an initial chapter of the 2012 book, which will focus on the broader political and economic landscape leading up to the election.

9 comments

Is “Even Jimmy Carter” Even a Good Insult?

by John Sides on May 1, 2012 · 11 comments

in Campaigns and elections

Apropos of Romney’s “even Jimmy Carter” remark, from a January 2011 CNN poll:

From what you have heard, read or  remember about some of our past presidents, please tell me if you  approve or disapprove of the way each of the following handled their  job as president: Jimmy Carter?”

Carter’s approval rating wasn’t all that low: 53% of respondents approved, 39% disapproved, and 7% didn’t know.

Here is the breakdown by party: Dems – 74%, independents – 56% (vs. 35% disapprove), and Republicans – 28%.

This doesn’t strike me, on its face, as evidence that Carter is some powerfully resonant negative symbol.  He’s viewed favorably even by the majority of independents.  If this “message”—I use quotes because Romney’s remark sounded off-the-cuff and not like some pre-planned talking point—is going to resonate with anyone, it will mostly be Republicans, 90% of whom already plan to vote for Romney anyway.

11 comments

Mass Appeal: Crowd-Funding Political Science Research

by John Sides on May 1, 2012 · 2 comments

in Academia

This is a guest post from Aleks Ksiazkiewicz, a Ph.D. candidate in political science at Rice.

*****

One of the most difficult things about being a junior scholar (and perhaps a senior scholar too) is finding funds for small research projects and pilot studies. Crowd-funding, which relies on receiving small donations from many individuals, presents a new way to fill this gap.

Most of us have already been involved in crowd-funding in one form or another: children sell Girl Scout cookies, public broadcasters run pledge drives, and political campaigns solicit partisans for donations. The internet has allowed individuals to take this model to a whole new level by soliciting contributions from around the world as seed money for innovative ideas.

The crowd-funding phenomenon was first popularized in the art world by sites like RocketHub and Kickstarter, but it has recently moved into scientific research. The SciFund Challenge is one initiative that is harnessing this resource by providing a periodic venue in which scientists from any discipline can petition the masses to fund their research. The first round took place in the fall of last year and raised over $75,000 with projects receiving on average $1,500 and some receiving over $10,000.

While I am the only political scientist participating in the second round of SciFund, there are several projects that political scientists may want to check out (and perhaps choose to support): mygenopolitics project on genetic effects in political attitude change, Cathy Day on the effectiveness of farming subsidies in the United States, Karyn Boenker on public opinion on energy policy in Hawaii, Jennifer Valpreda on deforestation in Argentina, and Eran Elhaik on applying population genetics to historical study.

As a participant in the second round of SciFund, I have found the process of preparing a project to be very rewarding (and of course I hope the fundraising will be too!). Since the goal is to raise funds primarily from non-scientists (or at least non-specialists), I have had to think about how to explain my dissertation research in accessible language and in a way that provides a compelling narrative to lay individuals.

Moreover, part of the incentive structure is to provide rewards at different levels of donations and it was genuine fun to come up with cost-effective but attractive offerings, like a video lecture via Skype or a hand-knitted double-helix (since I work on behavioral genetics). SciFund has also helped me to connect with scientists in other disciplines within my university and elsewhere as part of reviewing each other’s project pages and coordinating fundraising strategy.

I encourage all of you to consider joining SciFund for its next wave once applications become available (or try any one of several other science crowd-funding services like IndieGogo, Petridish, or FundaGeek).

(An earlier version of this post was included in the International Society for Political Psychology Junior Scholar Committee spring newsletter.)

2 comments

Making Voting Compulsory (for Young People)

by John Sides on May 1, 2012 · 6 comments

in Campaigns and elections

Guy Lodge and Sarah Birch:

Increasing electoral turnout is not just a nice idea, it is something we must actively strive for if elections are to serve the needs of all citizens….So how can we increase rates of electoral participation, particularly among ‘hard-to-reach’ groups such as the young and the poor?
…by far the most effective – albeit controversial – way of boosting participation is to make voting compulsory….Calls for compulsory voting are, however, commonly met with the objection that it is a citizen’s right to choose not to vote…
To allay such fears we propose a more realistic approach which is to make electoral participation compulsory for first-time voters only. Voters would be compelled only to turn out – and would be provided with a ‘none of the above’ option. The logic behind this proposal is that people who vote in the first election for which they are eligible are considerably more likely to vote throughout their lives. Introducing an obligation for new electors to turn out once would thus go a significant way toward breaking the habit of non-voting that often gets passed from generation to generation, and could have a substantial and lasting impact on turnout.

I am going to side-step the normative question of whether turnout should be higher or lower.  For the sake of argument, I’ll assume that this is a worthy goal and focus my initial thoughts on the likely empirical effect of this plan, with the proviso that this is developed for the UK and I am no expert in British politics.

First, turnout is indeed habit-forming, according to other research by Donald Green, Alan Gerber, and Ronald Shachar (see here and here for summaries).  To quote from the second study, which involved a randomized experiment in New Haven, CT, during the 1998 election:

…voting in 1998 raised the probability of voting in 1999 by 46.7 percentage points.  Other things being equal, registered voters who did not vote in 1998 had a 16.6% chance of voting in 1999, as compared to 63.3% among those who voted in 1998.  By any standard this is a very large effect.

It is.  In the short run, some of that effect will probably decay.  And we don’t really know whether the effect would be as large for young people as for older people.  But still, after years and years of compulsory turnout for young people, the effect is likely to accumulate, relative to a counterfactual world with no such requirement.  I think there is a good chance that the proposal could succeed.

Second, this surge in turnout among the young is going to discomfit many politicians.  Lodge and Birch argue that there won’t be any partisan effects, since politicians of both sides will start to appeal to young people and presumably each party will win an equal number of the new votes of young people.  I am less certain.  The partisanship of young people depends a lot on the prevailing climate when they come of age politically.  The Pew Center put out some data from the U.S. a while back and found “generational differences that reflect the political climate at the time when individuals were forming their political identity and loyalties.”  The appeals of parties and politicians may be less effective than the overall impact of the climate.  So whichever party isn’t doing well among young people when the plan was introduced could easily oppose it for strategic reasons.

Third, if every person were required to vote when they were first legally eligible, I suspect that parties would start targeting young people much earlier in an effort to build party loyalty during adolescence.  Some people might find political appeals to, say, 14-year-olds, problematic.  I am agnostic about that, personally, but I would note that whatever people tend to dislike about political campaigns—negativity, lying, money, etc.—will now be increasingly targeted at minors.

Here is a related study by Sarah Birch.  Previous Monkey Cage posts on compulsory voting are here, here, here, and here.

6 comments

Do the French Discriminate Against Muslims Because of Gender Norms?

by John Sides on May 1, 2012 · 6 comments

in Comparative Politics,Public opinion

This is a guest post by Claire Adida of UCSD, David Laitin of Stanford, and Marie-Anne Valfort of the Paris School of Economics & Paris 1 Panthéon Sorbonne University.  We have previously discussed their research here and here.


*****


Muslims express more traditional views toward gender-based inequality than do non-Muslims. Relying on the World Values Survey, Steven Fish shows that Muslims are significantly more likely to agree that “a university education is more important for a boy than for a girl”, to think that “when jobs are scarce, men should have more rights to a job than women” and to support the idea that “men make better political leaders than women do” (p 181-193). Concomitantly, many correspondence tests have shown that nationals originating from Muslim dominated cultures face systematic discrimination in European labor markets. In France, Duguet, Léandri, L’Horty and Petit (ungated) show that callback rates received by applicants with North African sounding names are much lower than those received by applicants with French sounding names. In Sweden, Magnus Carlsson identifies substantial discrimination against applicants with Middle Eastern sounding names. And Kass and Manger show that applicants with Turkish sounding names are discriminated against in the German labor market.

Could discrimination against nationals originating from Muslim areas be partly accounted for by recruiters’ reluctance to hire individuals whose gender norms might threaten the firm’s female employees and thus its esprit de corps?

Answering this question presents methodological challenges.  First, one must isolate the source of discrimination faced by applicants originating from Muslim areas. Confounding factors complicate any causal claim. For example, do employers discriminate against people of North African, Middle Eastern or Turkish background or do they discriminate against Muslims? Second, one must confirm that gender norms between Muslims and Christians do indeed differ, with Muslims being less favorable to women than Christians. In addition to the issue of confounds, current cross-country survey data are compromised since their measures rely on self-reported, not actual, behavior. Third, even if it is found that Muslim gender norms tolerate or encourage discrimination against women, it remains to be demonstrated that the behavior of employers in regard to Muslim job applications in the European labor market is related to Muslim gender norms

We addressed the first challenge by running a correspondence test in France in 2009, holding constant the country of origin of the two applicants—French nationals of Senegalese background. Religion was the only differentiating characteristic: one of the applicants was Christian while the other was Muslim, thereby allowing us to isolate the Muslim factor from country of origin and other possible confounds in French labor market discrimination. Our results confirm that anti-Muslim discrimination is significant: the Muslim candidate is 2.5 times less likely to receive a call back for a job application than is her Christian counterpart. (This result previously appeared on The Monkey Cage here; see our paper for further details.)

Are Muslims less favorable to women than are Christians?

A unique identification strategy combined with a set of lab experiments in France allow us to meet the second and third methodological challenges. Members of two Senegalese communities (Serer and Joola), who are each united by language, culture, and race, but divided by religion (Muslims and Catholics), immigrated to France over the same period in the 1970s. Studying this population (as we did for the correspondence test) thus enables us to isolate the effect of religion on gender norms.

To do so, we compared the donations of 18 Senegalese Muslim and 11 Senegalese Christian players to men and women in a dictator game we conducted in the 19th arrondissement of Paris in 2009. In this classic experiment, subjects view pictures of people whom they have never met and are given money either to keep for themselves or to share with the person (the “recipient”) whose picture they are viewing. There were no penalties for keeping it all, and no one could influence their donations. They are therefore effectively “dictators.” Fewer donations to a particular group in the experiment is an indicator of discrimination.

In our version of the dictator game, dictators were shown the same set of six recipients on a large screen revealing only their faces and their ascribed first names. The dictators were asked to allocate from zero up to the five euros allotted to them for each recipient. Among the six recipients, three appeared to be rooted French (i.e., French-born subjects of French-born parents and French-born grandparents) and presumably Christian, one appeared to be a Muslim from North Africa, and two appeared to be black Africans. We varied the first names of the recipients such that dictators would see the same faces, but with different signals of religious identities via ascribed names. For example, for half the sessions, subjects viewed one of the black African recipients with a Christian name and the other with a Muslim name; for the other half of the sessions, this was reversed. This protocol enabled us to investigate whether Senegalese Muslim “dictators” treat women differently from men relative to Senegalese Christian “dictators.”

 


Our results confirm that in-group gender norms differ between Senegalese Muslim and Senegalese Christian players (see this paper for details). First, Senegalese Christian male players favor rooted French women over rooted French men, while Senegalese Muslim male players do not. And, given that rooted French male dictators also favored rooted French women over rooted French men, our experimental data indicate that Senegalese Christian gender norms coincide with rooted French gender norms while Senegalese Muslim gender norms do not. Second, Senegalese Christian male players favor their co‐religious women over their co-religious men, while Senegalese Muslim male players discriminate against their co‐religious women over their co-religious men. Third, gender norms are not reserved for men: Senegalese Muslim female players also discriminate against their co‐religious women over their co-religious men, thereby taking an active part in supporting an in-group norm that favors co‐religious men over women.

Is there a causal link between Muslim gender norms and French labor market discrimination?

To address this question, we need to know whether rooted French anticipate anti-women gender norms among Muslims. We addressed this question with a set of experiments in 2010 that brought together 50 rooted French players and asked them to guess the donations of our 2009 Senegalese Muslim and Christian dictators to our set of 6 recipients the figure above. We find that rooted French do not associate our Senegalese Muslim players with different gender norms relative to our Senegalese Christian players, thereby casting strong doubt on the notion that Muslims are discriminated against in the French labor market because French employers respond to a perception of gender discrimination among Muslims.

Our research reveals that the gender norms supporting inequality towards women as reported in the survey literature accurately represent the behavior of Muslims in France. But the link between those norms and the egregious discrimination in the French labor market has only been assumed. Our research questions that link.  The sources of anti-Muslim discrimination in Europe remain obscure, but our research reduces confidence that the answer lies in Muslim gender norms.

6 comments

Senator Rubio’s Support among Latino Sub-Groups

by Dan Hopkins on May 1, 2012 · 1 comment

in Blogs

As a recent Time magazine cover and a host of other articles remind us, Hispanics/Latinos are a sizable voting block in several states that will be fiercely contested come November, including Arizona, Florida, Colorado, New Mexico, and Nevada.  What’s more, shifts among this group were a critical factor in President Obama’s 2008 victory.  As Governor Romney considers would-be running mates, his campaign is likely to ask whether certain choices would increase GOP support among Latinos.  Certainly, commentators such as George Will, Jamelle Bouie, and Harry Enten are already doing exactly that.

Florida Senator Marco Rubio, who is of Cuban descent, is one prominent name in such conversations.  Here, one interesting question is about the extent to which Latinos in contested western states—-many of whom are of Mexican descent—would be more inclined to support a ticket with a Vice Presidential nominee of Cuban descent.  Is the shared pan-ethnic identity likely to be influential?

For some relevant background, we might consider data from the Latino National Survey, which was conducted by political scientists Luis Fraga, John Garcia, Rodney Hero, Michael Jones-Correa, Valerie Martinez-Ebers, and Gary Segura in 2005 and 2006.  Among its many questions, respondents were asked which term best describes them, and they were able to choose between “Hispanic/Latino,” their family’s country of origin, and “American.”  37.0% of the survey’s 2,881 citizens of Mexican heritage chose the pan-ethnic “Hispanic” or “Latino,” while 27.5% opted for the national-origin group “Mexican.”  Another 28.3% chose “American,” with the remainder declining to answer or volunteering “none of the above.”  So there is significant identification with the pan-ethnic identity—and also with Mexico.

Is that pan-ethnic identity likely to influence voting?  For those seeking an in-depth answer, be sure to see Matt Barreto’s new book, Ethnic Cues: The Role of Shared Ethnicity in Latino Political Participation.  (I hope to post on that book soon, when I’ve had a copy for more than a few hours.)  For our purposes here, one preliminary way to address that question is by examining vote shifts between Florida’s 2008 Presidential election and its 2010 Senate election.  Exit poll data indicate that Obama won Florida Latinos 57%-42%, and that Rubio won the same group 55%-45% (where we group Latinos who voted for Democratic nominee Kendrick Meek alongside those who voted for independent candidate Charlie Christ).  This is a marked shift given that overall GOP support remained quite similar across the two elections, in large part due to 2010’s unusual three-way race.  Still, as Scott Clement over at Behind the Numbers points out, Rubio did far better among Americans of Cuban descent than among other Latino groups.  And from the Latino National Survey, we know that roughly 34% of Florida’s Latino citizens report being of Cuban descent, with 28% reporting Puerto Rican descent and another 8% indicating Mexican descent.

We turn, then, to precinct-level election data released as part of a broader project by Stephen Ansolabehere and Jonathan Rodden.  Specifically, we can observe each precinct’s shift in GOP support from 2008 to 2010, and compare that with guesses about its Cuban, Puerto Rican, and Mexican populations estimated from the most proximate Census tract in 2010.  This analysis has the usual concerns about ecological inference—-we don’t know who within each precinct is voting for Rubio—-and we also need to keep in mind that the Census measures ancestry for residents, not for citizens.  So the results need to be taken with more than the usual a grain of salt: it is quite plausible that groups other than those measured here are actually driving the changes.  Still, this approach gives us a sense of the extent to which Marco Rubio out-performed John McCain in different precincts across a highly diverse state.


The Figure above summarizes the relationships, and includes coefficients from linear models predicting the change in vote share with the tract-level percentage of residents from each national origin group.  Shifts toward the GOP were pronounced in precincts near concentrations of residents of Cuban descent, as the left panel indicates.  Intriguingly, they were even more pronounced in precincts near concentrations of residents of Puerto Rican descent—check out the upward-sloping pattern in the middle panel.  But such shifts were also essentially unrelated to local Mexican-descended populations, as the right-most panel indicates.  From the precinct-level returns alone, our best guess is that Marco Rubio’s appeal in 2010 was different depending on the Latino sub-group in question.  Notice, too, which groups appear to be influenced: Puerto Ricans, who are U.S. citizens from birth, and Cubans, whose eligibility for political asylum might lead them to approach the contemporary immigration debate from a different angle.  Certainly, that’s a possibility to examine with individual-level data.

1 comment

Potpourri

by John Sides on April 30, 2012 · 2 comments

in Potpourri

  • Campaign spending matters more down the ballot.  Ezra Klein is right on this.  For more, see my first Moneyball post.
  • It’s too early to pay attention to the Electoral College.  Say it over and over.  For more, see Jon Bernstein.
  • Political Pipeline’s challenges for political scientists.
  • Does putting network anchors in campaign ads make those ads more credible?  Elizabeth Wilner and Ken Goldstein discuss.

 

2 comments

A Balanced News Diet After All?

by John Sides on April 30, 2012 · 10 comments

in Media

This is a guest post from Michael LaCour, a Ph.D. candidate in political science at UCLA.  The paper on which this post is based is here.

*****

The reemergence of a prominent partisan press has led many scholars to investigate partisan self-selection of news outlets. Political scientists and journalists have concluded that individuals are motivated to select media sources that match their own political views and avoid media sources that challenge their political views. However, an analysis of individuals’ actual media exposure patterns lead to conclusions about selective exposure quite different from previous research based on self-reported media exposure.

I use data collected by Integrated Media Measurement Incorporated (IMMI) – to measure individuals’ actual media exposure. Nine hundred and twenty panelists from the New York and Chicago media markets were given smartphones equipped with audio recognition software that continuously and  automatically captured their exposure to all radio and television media – over the course of the 2006 midterm campaign.

Results indicate that a majority of viewers consume little or no news and the remainder consume very high levels of local news  as well as an ideologically diverse set of partisan news programs.   The figure below measures panelists’ news diet among partisan sources on television and radio.   This is a net measure of how “balanced” individuals partisan news diet is (weighted by volume).  The figure displays the distribution of panelists’ partisan news diet in the form of separate kernel density estimates for Democrats and Republicans.

On this scale a high positive score represents a panelist watching and listening to media that is conservative. Conversely, a negative score indicates a panelist watching or listening to media that is liberal – and all the more so if the exposure occurs frequently at high volume. A detailed description of how media outlets are coded is here.


The unimodal distributions centered around zero, indicate that Democrats and Republicans have similar news consumption  patterns. Only a small unrepresentative subset of the public is sorting themselves into ideologically like-minded enclaves.

To summarize, most individuals do not refuse to hear the other side. In fact, most people consume predominately non-partisan  local TV newscasts, while tuning out news from partisan  sources altogether. Of those who do turn to partisan sources, most Republicans and Democrats have virtually indistinguishable news diets. Contrary to recent claims, there is little evidence that the electorate is self-sorting into “ideologically like-minded information cocoons” at the level being described by scholars and political commentators.

10 comments

Farmscape for Mayor

by John Sides on April 28, 2012 · 5 comments

in Campaigns and elections

A la the Murray Hill campaign in 2010, another business is running for office: Farmscape.  Their site is here.  “Bio” is above.  Here is video from when they registered the company to vote.

 

5 comments

← Previous Entries