Can the Party in Power Systematically Win a Majority in Close Legislative Elections? Evidence from U.S. State Assemblies
Dahyeon Jeong & Ajay Shenoy
Journal of Politics, forthcoming
We study whether ruling parties can systematically win a slender majority of seats in close legislative elections, a phenomenon called "precise control." We test for discontinuities in two outcomes that, in the absence of precise control, should be smooth at the 50% cutoff: the probability density of the share of seats won, and the identity of the party that previously held a majority. We find robust evidence of precise control, but only in high-stakes state elections that determine which party controls Congressional redistricting. Its absence in other elections suggests precise control is a strategic option used at the ruling party's discretion. It shifts its strategy in high-stakes elections from seat maximization to majority-seeking, winning fewer seats but raising the chance it retains its majority. These tactics are disproportionately effective for the party defending a majority. It is 4 times more likely to win than to lose a close election.
Partisan Alignment Increases Voter Turnout: Evidence from Redistricting
Bernard Fraga, Daniel Moskowitz & Benjamin Schneer
Political Behavior, forthcoming
Partisan gerrymandering and polarization have created an electoral landscape where Americans increasingly reside in congressional districts dominated by one party. Are individuals more likely to vote when their partisanship aligns with the partisan composition of the district? Leveraging nationwide voter file data and the redistricting process, we present causal evidence on this question via a longitudinal analysis of individual-level political participation. Tracking turnout before and after a redistricting cycle, where the boundaries of congressional districts change, we observe what happens when registrants experience a shock to the partisan composition of their district. We find turnout increases for individuals assigned to districts aligned with their partisanship as compared to individuals in misaligned districts, consistent with voters deriving expressive benefits from voting for the winning party. By demonstrating how districting influences political participation, our findings suggest a new implication of partisan gerrymandering that may clash with other democratic goals.
Why Donald Trump Should be a Fervent Advocate of Using Rank-Choice Voting in 2024
Jonathan Cervas & Bernard Grofman
Carnegie Mellon University Working Paper, April 2021
This article builds off work by Devine and Kopko (2021) and Lacy and Burden (1999) who estimate a probit model of candidate choice from the CCES to determine the second choice of third-party voters. Using this model on 2020 election data, we show that the Libertarian candidate Jo Jorgenson probably cost Donald Trump victory in at least two states - AZ and GA. Additionally, the popular vote margin enjoyed by Joe Biden could have been between 260,000 and 525,000 votes less, using conservative estimates. The motivation of this paper is to give contrary evidence for two main misconceptions. First, that third-party candidates are "spoiling" elections for the Democrats. Our evidence clearly shows that third-parties have the potential to hurt either of the two main parties, but in 2020, it was Donald Trump who was hurt most; though not consequentially. Second, some reformers believe that Rank Choice Voting Benefits the Democrats; again, we show that, all else being equal, it is the Republicans that would have benefited by the change in rules, since the majority of third-party votes are going to the Libertarian candidate, and whose voters prefer Republicans over Democrats 60% to 32%.
Weather to Protest: The Effect of Black Lives Matter Protests on the 2020 Presidential Election
Bouke Klein Teeselink & Georgios Melios
Yale Working Paper, March 2021
Do mass mobilizations bring about social change? Prior research provides mixed findings on whether large-scale collective action helps protesters further their cause. This paper adds new evidence to this debate by investigating the causal impact of racial injustice protests on the 2020 presidential election. Following the death of G. P. Floyd Jr. on 25 May 2020, a series of Black Lives Matter protests erupted across the US. Using cross-county variation in rainfall as an exogenous source of variation in protests, we document a marked shift in support for the Democratic candidate in counties that experienced more protesting activity. As a consequence, BLM protests might have tilted the election in favor of the Democratic Party. We additionally document that BLM protests did not affect the overall turnout rate, which suggests that the increase in Democratic support primarily resulted from a progressive shift among undecided voters.
Trump Paradox: How Immigration and Trade Affected White Voting and Attitudes
Raul Hinojosa Ojeda & Edward Telles
Socius: Sociological Research for a Dynamic World, April 2021
Donald Trump presented immigration and trade as the cause of the diminished prospects of white working-class voters, the core of his political base. The authors' research -- the first that examines actual immigration and trade exposure with attitudes and Trump voting -- demonstrates that white voting for Trump was unrelated to immigration levels and, paradoxically, strongest in counties with low levels of trade. Anti-immigrant and antitrade attitudes more consistently and strongly explain voting for Trump and Republicans in 2016 and 2018 than actual immigration and trade. The authors also find descriptive support that over four years, Trump's false narrative unraveled as his support declined in those counties most exposed to immigration and trade. Although Trump elaborated a white nationalist narrative on the basis of anti-immigrant and antitrade politics that was widely accepted as truth, the authors show that virtually no aspects of Trump's simple narrative have any factual basis in actual reality.
Erroneous Beliefs and Political Approval: Evidence from the Coronavirus Pandemic
Matthew Lilley & Brian Wheaton
Harvard Working Paper, April 2021
Are politicians rewarded for good performance as implied by models of retrospective voting? This requires that public perceptions of performance are accurate. We examine the case of the coronavirus pandemic - during which surveys on governor approval abounded and objective measures of state outcomes were readily available. We conduct an incentivized survey in which respondents estimate how pairs of states have performed relative to one another in terms of deaths per-capita. We find that perceptions are only modestly more accurate than random guessing, and we find little to no evidence of partisan in-group bias. We find evidence, using data from SafeGraph and the Understanding America Survey, that these erroneous beliefs about state performance distort social-distancing behavior, suggesting both that our measure of beliefs is accurate and that erroneous beliefs come at a cost to society. Furthermore, running regressions with governor approval ratings as the outcome, we find that it is not the actual death rate - but rather beliefs about the death rate - that drive governor approval. This remains true after controlling for perceptions of how well states should have performed, setting aside factors of leadership/political competence. We replicate these findings in both an identical follow-up survey later in the pandemic and in a survey experiment.
Resemblance and Discrimination in Elections
Raluca Pahontu & Stavros Poupakis
London School of Economics Working Paper, March 2021
Discrimination affects hiring, mating and voting decisions. Whilst discrimination in elections mainly relates to gender or race, we introduce a novel source of discrimination: candidate resemblance. When candidates' partisanship is not known, voters select those that resemble most elected co-partisans. Using a machine learning algorithm for face comparison, we find a stronger resemblance effect for Republicans compared to Democrats in the US. This happens because Republicans have a higher within-party facial resemblance than Democrats, even when accounting for gender and race. We find a similar pattern in the UK, where Conservative MPs are more similar looking to each other than Labour. Using a survey experiment, we find that Tory voters reward resemblance, while there is no similar effect for Labour. We estimate that facial dissimilarity decreases the candidate's re-election probability by 5-14 percentage points. The results are consistent with an interpretation of this behaviour as a form of statistical discrimination.
Models, Markets, and the Forecasting of Elections
Rajiv Sethi et al.
Columbia University Working Paper, March 2021
We examine probabilistic forecasts for battleground states in the 2020 US presidential election, using daily data from two sources over seven months: a model published by the Economist, and prices from the PredictIt exchange. We find systematic differences in accuracy over time, with markets performing better several months before the election, and the model performing better as the election approached. A simple average of the two forecasts performs better than either one of them overall, even though no average can outperform both component forecasts for any given state-date pair. This effect arises because the model and the market make different kinds of errors in different states: the model was confidently wrong in some cases, while the market was excessively uncertain in others. We conclude that there is value in using hybrid forecasting methods, and propose a market design that incorporates model forecasts via a trading bot to generate synthetic predictions. We also propose and conduct a profitability test that can be used as a novel criterion for the evaluation of forecasting performance.
More Women Candidates: The Effects of Increased Women's Presence on Political Ambition, Efficacy, and Vote Choice
Mia Costa & Isabel Wallace
American Politics Research, forthcoming
The effect of women in politics is vitally important for the study of representation, yet evidence is mixed on the extent to which women's presence influences individuals' symbolic attitudes and behaviors. We use a priming survey experiment to examine how information about increased women candidates in the U.S. affects political ambition, efficacy, and future support for women candidates. We present several different patterns across gender and partisanship. Republicans report higher political ambition after hearing about more women candidates, even when those women are running for the opposite party. Men had higher political efficacy in response to more same-party women running, but not opposite-party women. Importantly, our evidence does not support the widespread notion that women's presence positively influences women's political efficacy or likelihood to vote for female candidates. The findings highlight the importance of considering the effects of women's presence not only for the group that is assumed to benefit.
The Impact of the Economy on Presidential Elections Throughout US History
Eric Guntermann, Gabriel Lenz & Jeffrey Myers
Political Behavior, forthcoming
As numerous studies in the US and elsewhere document, voters often hold incumbents accountable for recent economic circumstances. However, our knowledge of the conditions that allow voters to do so remains incomplete. In particular, most findings about economic voting come from studies of modern economies (post World War II). Modern economies have a host of characteristics that seem to lend themselves to economic voting. Their governments play a large role in the economy and have the Keynesian toolset necessary to influence the economy. Their voters are educated and have access to detailed economic data from ubiquitous media. Are these and other modern conditions necessary for economic voting? Would voters still hold politicians accountable even under adverse conditions? Using economic measures now available back to the 1790s, we study economic voting from the earliest days of the US Republic when none of these conditions were met. Voters, we find, appear to judge incumbent presidents on the economy all the way back to George Washington. Consistent with this pattern, we also find that the economy appears to shape presidents' decisions to run again throughout US history. These findings support recent comparative evidence that economic voting is pervasive across a variety of contexts.
Exposure to televised political campaign advertisements aired in the United States 2015-2016 election cycle and psychological distress
Jeff Niederdeppe et al.
Social Science & Medicine, May 2021
Methods: A secondary analysis of U.S. data on televised campaign ad airings from January 2015 to November 2016 (n = 4,659,038 airings) and five waves of a mail survey on television viewing patterns and self-reported medical conditions from November 2015 to March 2017 (n = 28,199 respondents from n = 16,204 unique households in the U.S.).
Findings: A 1 percent increase in the estimated volume of campaign advertising exposure was associated with a 0.06 [95% CI 0.03-0.09] percentage point increase in the odds of a respondent being told by a doctor that they have anxiety in the past 12 months. We observed this association regardless of the political party of the ad sponsor, the political party of the respondent, or their statistical interaction. We also observed this association for both Presidential campaign ads and non-Presidential (including local, state, and U.S. congressional election) campaign ads, providing evidence that these relationships were not driven by the unique divisiveness of the race between Donald Trump and Hilary Clinton. Some topic-specific models offered additional evidence of association between estimated volume of campaign advertising exposure and the odds of being told by a doctor that they have depression or insomnia, but these patterns were less consistent across models that utilized different categories of campaign exposure. Campaign ad exposure was not associated with cancer, which served as a negative control comparison.
Rock the Registration: Same Day Registration Increases Turnout of Young Voters
Jacob Grumbach & Charlotte Hill
Journal of Politics, forthcoming
Studies find that same day registration (SDR) laws increase turnout, but less is known about which kinds of voters are most affected. Young people are disproportionately burdened by traditional registration laws because they frequently change addresses and infrequently interact with government agencies providing registration services. SDR laws, which lower the cost of registration, should increase turnout most among young people. Laws that lower the cost of voting but not the cost of registration should be less effective at increasing youth turnout. Difference-in-difference estimates suggest SDR disproportionately increases turnout among individuals aged 18-24 (an effect between 3.1 and 7.3 percentage points). The effect of SDR on young voters is especially pronounced in presidential elections. By contrast, the effects of early voting and other reforms are smaller and do not consistently vary by age. The results suggest expanded SDR may produce a younger electorate.
Asian American Candidate Preferences: Evidence from California
Political Behavior, forthcoming
The diversity of the Asian American population presents challenges for theories of bloc voting, partisan voting, and descriptive representation. What cues (if any) do Asian American voters rely on? How informative are racial and partisan cues to Asian American voters. This article looks at the candidate preferences of Asian American voters in the 2018 election. I look at elections where an Asian American candidate was on the ballot and compare outcomes within district to the gubernatorial race (a race with no minorities on the ballot). I use surname-coded voter registration records and precinct-level vote returns to estimate Asian American candidate preferences as a racial group and by national-origin. I find strong evidence of national-origin preferences (i.e. Vietnamese for a Vietnamese candidate) among Asian American voters. In instances where the national-origin of the candidate and the national-origin of the voter did not align, voters seem to rely on partisan cues. National-origin preferences are sufficient enough that in one instance voters switched parties within the same election to vote for a candidate of the same national-origin. These findings have implications for theories of minority vote choice and challenges the existing literature on the strength of partisan cues.
From the Brady Bunch to Gilmore Girls: The Effect of Household Size on Economic Voting
Manuel Lago & Ignacio Lago
American Politics Research, forthcoming
This article examines whether household size affects economic voting. We argue when individuals are asked about national economic conditions and their personal financial situation that moderate or mid-range responses are more likely in multi-person households than in one-person households. The aggregation of personal economic evaluations within households reduces the variation in economic opinions across household members. As a result, it is harder for an individual to say that the national economic conditions and her personal financial situation are good or bad as the number of household members increases. Using individual-level data from the American National Election Studies from 1966 to 2016, the authors find that both evaluations of the national economy and personal economic conditions are endogenous to household size. The aggregate, state-level evidence from five presidential elections in the U.S. shows that the impact of the economy on the incumbent support increases the larger the number of one-person households.