The forthcoming article “Informational lobbying and commercial diplomacy” by Calvin Thrall is summarized by the author below.

Foreign affairs are primarily managed at the bilateral, state-to-state level. Further, while leaders themselves get involved in important relationships, bilateral diplomacy is typically delegated to the foreign policy bureaucracy: diplomats serving in the state’s foreign embassies. Diplomats are responsible for managing bilateral relationships across a wide range of issue areas—from commercial policy, to immigration, to scientific cooperation, among others—and they are given substantial discretion over their allocation of time and effort across issue areas. Yet we know little about how diplomats decide which issues to prioritize. What determines the content of diplomacy, across bilateral relationships and time?

I argue that the policy issues that diplomats prioritize are affected by their sources of information about their host state. In order to gauge the most pressing issues in a bilateral relationship, diplomats must first collect intelligence on what is happening in the host state. Due to resource and personnel constraints, diplomats typically rely on various organizations in the host state—business associations, human rights NGOs, expatriate groups, etc—to supply much of their intelligence. I argue that these host state organizations are akin to interest groups engaging in informational lobbying: by providing diplomats with intelligence about an issue of relevance to them, these groups subsidize diplomatic effort in that issue area and increase the likelihood that the embassy dedicates more time to their preferred issue. When diplomats receive more information about a specific issue area, they are more likely to allocate greater attention to that issue.

I evaluate this argument in the context of the expansion of one type of influential interest group: American Chambers of Commerce (AmChams), which are composed of U.S. firms operating abroad in a particular country (e.g. AmCham China, AmCham France). They advocate for pro-business policies in areas such as taxation, investment, and trade. AmChams proliferated widely in the post-WWII period, and provided substantial information to U.S. diplomats about commercial issues in the states where they opened. If my theory is correct, diplomats should shift their attention towards commercial issues once an AmCham begins operating in their host state.

How to measure diplomatic attention at the country-year level, since diplomatic activity is classified? I use modern natural language processing techniques to extract information about diplomatic activity from nearly 1,500 oral history interviews with former U.S. diplomats (conducted by the Association for Diplomatic Studies and Training), allowing me to determine how much different embassies focused on commercial issues over time. I show that, when a new AmCham opens, diplomats in that host state go on to pay greater attention to commercial issues relative to states without AmChams. Further, leveraging the fact that U.S. diplomats rotate to new embassies every three years, I show that individual diplomats spend more time on commercial issues when they are rotated to a state with an active AmCham.

These results highlight an underappreciated channel of foreign policy influence: lobbying by interest groups that occurs “beyond the water’s edge.” Future research should study the effects of this type of lobbying—in particular, that of “bilateral lobbies” such as AmChams—on broader foreign policy phenomena.

About the Author: Calvin Thrall is an Assistant Professor of Political Science at Columbia University. Their research “Informational lobbying and commercial diplomacy” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “The politics of rejection: Explaining Chinese import refusals” by Sung Eun Kim, Rebecca L. Perlman, and Grace Zeng is summarized by the author(s) below.

Governments have long been known to use health and safety standards as an underhand way of impeding imports while feigning an open trading posture. This paper examines a less explored explanation for such regulatory barriers. We show that in the same way that product standards can be used as a subtle form of protectionism, so too can they be used as a means of punishing or coercing one’s trading partners. Indeed, we theorize that some of the same characteristics that make health and safety standards attractive to governments as a form of protectionism – particularly the ambiguity surrounding their intent – also make these measures attractive as a way to retaliate against foreign nations. 
 
In order to evaluate our theory, we collect and translate original data from 2011-2019 on China’s use of import refusals. An import refusal is the rejection of an import at the border, supposedly because the product in question fails to meet domestic health, safety, or environmental rules. Combining our data on import refusals with data on political tensions involving military actors, we demonstrate that during the period under study, China has systematically increased its bilateral use of import refusals in the wake of foreign policy squabbles. This has allowed China to exert political pressure on trading partners with which it is at odds, while maintaining plausible deniability about its actions. 
 
Our research contributes to the broader literature on the source of regulatory barriers to trade and the extent to which political tensions harm economic relations. More narrowly, our findings offer insight into China’s use of economic coercion, a topic that has been of growing interest both to political science scholars and to policymakers. While there have been numerous anecdotes suggesting that China has, at times, used import refusals as a form of political retribution, to date there has been a lack of evidence demonstrating a systematic relationship between instances of import refusals and those of heightened conflict. Our findings provide the first evidence of a consistent and measurable connection between China’s rejection of imports and the existence of political tensions.  

About the Author(s): Sung Eun Kim is an Associate Professor at the Department of Political Science and International Relations at Korea University, Rebecca L. Perlman is an Assistant Professor of Political Science at the University of California, Berkeley, and Grace Zeng is a Ph.D. candidate in the Department of Politics at Princeton University. Their research “The politics of rejection: Explaining Chinese import refusals” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Encouraging loyalty and defection: The impact of campaigns on tactical voting in Britain” by Lucas Núñez is summarized by the author below.

In multi-party elections using winner-take-all electoral systems, voters often face strong incentives to cast a tactical (or strategic) vote: a vote for a less-preferred party but with higher chances of winning. Tactical voting has received widespread attention in the voting behavior literature, predominantly focusing on measuring its extent. In recent United Kingdom General Elections, about a third of voters face such incentives, and between a third and half of them cast a tactical vote. While some correlates of tactical behavior are known, they are typically non-actionable; that is, not factors over which electoral participants have agency during the course of an electoral campaign. As such, our understanding of tactical/strategic voting behavior remains limited by a mostly passive view of electoral participants.  

In this article I focus on one such actionable factor: parties’ direct outreach to voters. Using panel data for three separate UK General Elections, I estimate how contact by the different parties influences voters’ decision to cast a tactical vote, an effect separate and distinct from any changes in preferences that may simultaneously occur. That is, rather than measuring traditional conversion/persuasion effects, I measure the impact of parties’ outreach voters’ decision to remain loyal to or defect from their most preferred party.  

I show that contact by a voter’s most preferred party encourages loyalty to the most preferred party, thus reducing tactical voting by about 4 percentage points. On the other hand, contact by a voter’s best alternative party (a less-liked party with higher chances of winning) encourages defection from the most preferred party, thus increasing tactical voting by about 5 percentage points. Additionally, in partial counterfactual estimates, I show that had parties not conducted any direct voter outreach, tactical voting overall would have decreased by about 2 percentage points.  

These effects of campaigns on tactical voting stand in contrast to those in the persuasion literature in the United States, which typically finds null or minimal effects. However, there are important differences in context: the US is a two-party system with few cross-pressured voters; the UK, on the other hand, is a multi-party system with a significant number of cross-pressured voters and voters facing strong tactical incentives. My findings show that campaigns matter in the latter circumstances.  

On the practical side, my findings suggest that candidates’ communications highlighting tactical/strategic situations in their constituencies, as parties in the UK often do, are a valuable strategy. On the normative side, the increase in tactical behavior could enhance voters’ political efficacy and help them elect a “less bad” government in the short term. In the long term, however, the overall tactical behavior encouraged by parties’ campaigns could prevent minority views from maturing into viable alternatives due to tactical defections.

About the Author: Lucas Núñez is an Assistant Professor of Political Science at the Schar School of Policy and Government at George Mason University. Their research “Encouraging loyalty and defection: The impact of campaigns on tactical voting in Britain” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Trains, trade, and transformation: A spatial Rogowski theory of America’s 19th-century protectionism” by Kenneth Scheve and Theo Serlin is summarized by the author(s) below.

Do the groups that benefit from economic change become more powerful and get to implement their preferred policies? Many theories in political economy assume so, but the empirical record isn’t so straightforward. In the US in the late 19th Century, falling transportation costs integrated agriculture into the global economy, but the exporters were politically marginalized; the country elected Republican politicians who implemented high tariffs. This phenomenon occurred at the local as well as the national level. Using voting records and a measure of protectionist sentiment estimated from newspapers, we show that places in the agricultural hinterland that gained transportation connections became more protectionist. 

We argue that geographic mobility conditions political responses to economic change. Building on recent developments in urban and international economics, we combine a majoritarian model of policy with a spatial trade model. Internal transportation improvements reduce the cost of accessing the global economy, and cause the exporting sector—agriculture in this period—to expand. Economic development induces both internal and international migration, which alters the composition of the electorate. In the period we study, the migration of laborers into agricultural regions diluted the power of landowners. In our model, the high rate of migration between regions, and the relative sizes and factor intensities of the manufacturing and agricultural sectors, generates the prediction that laborers, regardless of location, should have aligned with the manufacturing sector which supported tariffs. The net result of these changes was an increase in the basis of support for protectionism. 

We take the model to county-level data on economic and political outcomes over the period 1880-1900. Our empirical strategy examines how counties responded to distant changes in the transportation network that reduced the cost of accessing the global economy. We find these transportation cost reductions did benefit the agricultural sector—agricultural output and the value of farms both increased—but led to large population increases that reduced the share of farmers in the population. Politically, expanding trade access was associated with a shift away from the Democratic Party—which advocated for low tariffs—and towards protectionism. 

Including economic geography in a model of politics generates predictions about what policies agents want and how economic change alters policy outcomes that differ from traditional political economy models. Our theory assumes a high rate of geographic mobility, which was a feature of 19th century America. Modeling and estimating barriers to geographic mobility should provide insights into the relationship between economic and political change across different contexts.

About the Author(s): Kenneth Scheve is the Dean Acheson Professor of Political Science and Global Affairs at Yale University and serves as Dean of Social Science for the Faculty of Arts and Sciences and Theo Serlin is a Ph.D. candidate in Political Science at Stanford University. Their research “Trains, trade, and transformation: A spatial Rogowski theory of America’s 19th-century protectionism“ is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Austerity, economic vulnerability, and populism” by Leonardo Baccini and Thomas Sattler is summarized by the author(s) below.

Fiscal austerity – spending cuts and tax increases to reduce the public deficit –hurt economically vulnerable voters, e.g. low-skilled workers. We show that these vulnerable voters turn towards populist parties when their government implements austerity.  

Our analysis covers the period from 1991 onwards and examines election results in electoral districts across Western European countries. Vulnerable voters are workers with a low level of education, those in the manufacturing sector, and those with routine jobs. These voters are exposed to the greatest risk to lose their jobs because of globalization or automation. A standard austerity package increases vote shares of populist parties between 3 and 5 percentage points in districts with many low education / manufacturing / routine job workers. The same package has no effect on populist vote shares in districts with fewer vulnerable voters.  

The results have several important implications. First, governments matter. Policies implemented by governments have a variety of ways to moderate the adverse effects of economic transformations, e.g. through redistribution. However, if policies fail to do so or if they enhance the unequal distributional consequences of globalization and automation even further, populist parties can exploit the growing anti-globalization sentiment among dissatisfied voters. Second, austerity matters. Governments may be able to build a pro-austerity coalition and a majority of voters may not object against or even approve of austerity. Still, the political consequences of austerity measures are unavoidable: Many vulnerable voters turn away from traditional parties transforming politics for a long time to come. Third, vulnerability matters. Voters hold very diverse attitudes about fiscal policy that seem most strongly affected by their ideological views. But when it comes to cast a ballot, voters’ material interests, i.e. the effect of austerity on their welfare, is a defining factor to understand their political preferences and behavior. 

About the Author(s): Leonardo Baccini is an Associate Professor in the Department of Political Science and an associate member in the Department of Economics at McGill University and Thomas Sattler is a Professor of International Relations in the Department of Political Science and International Relations at the University of Geneva, Switzerland. Their research “Austerity, economic vulnerability, and populism” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Playing politics with traffic fines: Sheriff elections and political cycles in traffic fines revenue” by Min Su and Christian Buerger is summarized by the author(s) below.

Politicians often make strategic decisions around elections to influence voters’ choices. For instance, they frequently adopt popular budgetary measures, like increasing spending or cutting taxes, just before an election to appeal to their constituents. While these tactics are common among national and state politicians, our research delves into the local arena by examining the behaviors of county sheriffs during election years. 

County sheriffs play a pivotal role in the United States. They oversee a wide range of public safety issues beyond the jurisdiction of city and state police departments. Since sheriffs are directly elected by county constituents, they are one of the most visible and influential actors in local politics. This unique position provides an opportunity to analyze how elections can influence local policy decisions. 

We selected a policy area – traffic fines – sheriffs’ pre-electoral policy manipulation. Traffic stops represent the most common interaction between the public and law enforcement. Additionally, sheriffs wield significant discretion in establishing and enforcing traffic fine policies. We anticipate that, during an election year, sheriffs may reduce traffic fine issuance to prevent voter discontent and negative perceptions of law enforcement performance. 

To test our expectation, we utilized a panel dataset encompassing 57 California counties from 2003 to 2020, spanning four election cycles. Employing fixed-effects models, we investigated the presence, duration, and magnitude of a political cycle in a county’s traffic fines revenue. Our findings were revealing: in sheriff election years, traffic fines revenue decreased by 30 percent. Notably, this decline was short-lived, observed only in the election year. Furthermore, the drop was even more pronounced when an incumbent sheriff sought reelection or when the election was highly contested. 

Our study bridges three pivotal domains of research: political science, law enforcement, and public administration. By illustrating the influence of local elections on traffic enforcement outcomes, our study introduces a new dimension to our understanding of the intricate relationship between politics and policies at the local level. Considering the prevalence of traffic stops, our findings can aid policymakers and reform advocates in gaining a deeper understanding of the electoral motivations that could shape policy initiatives and the implementation of policing practices.

About the Author(s): Min Su is an Associate Professor in the Department of Public Administration at Louisiana State University and Christian Buerger is an Assistant Professor at the O’Neill School of Public and Environmental Affairs at IUPUI. Their research “Playing politics with traffic fines: Sheriff elections and political cycles in traffic fines revenue” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Evaluating (in)experience in congressional elections” by Rachel Porter and Sarah A. Treul is summarized by the author(s) below.

Candidates for U.S. Congress with a history of holding elective office (as city councilors, state legislators, mayors, etc.) are thought to have a sizeable electoral advantage over candidates lacking an elected background. Having run and won office before, experienced candidates have an established campaign infrastructure and network of donors they can draw on from their campaign’s outset. Experienced candidates also benefit from their public service reputation because voters attribute desirable traits like honesty, integrity, and competence to candidates with an elected background. Because experienced candidates systematically beat amateurs in elections, prior elected experience has become the standard indicator for challenger “quality” in congressional research.   

Our paper asks whether experienced candidates still possess an electoral “edge” over amateurs. To do this, we compiled data on the political backgrounds, personal characteristics, and campaign fundraising for all candidates who appeared on the ballot for the U.S. House of Representatives between 1980 and 2020. We uncover a surprising finding: candidates without prior elected experience are entering the U.S. House today at rates not seen since the widespread adoption of primaries. From the early 1980s to mid-2010s, three-quarters of newly elected House members possessed elected experience; conversely, just half of freshmen members elected from 2016 to 2020 held prior office. 

We attribute this decline in the proportion of experienced candidates entering Congress to electoral losses. Experienced candidates today are more often failing to gain their party’s nomination in safely partisan open seats and, instead, are losing to amateurs who go on to win uncompetitive general elections. We investigate why experienced candidates are losing more often and show that, in modern elections, amateurs increasingly possess the kinds of electoral benefits that have traditionally advantaged experienced candidates. In particular, we find that amateurs compose the majority of top campaign fundraisers today. We also find that amateurs now possess the kinds of valence characteristics that give candidates an advantage in congressional elections. Among Republicans, we find that candidates are advantaged if they are political “outsiders” with no experience in politics—elected or otherwise. Among Democrats, we find that candidates are advantaged if they are female or non-white. Notably, the vast majority of female and non-white Democrats running and winning today lack elected experience. 

Our findings offer a reconsideration of long-held theories regarding candidate success in congressional elections. Our work aligns with trends documented within American and comparative politics toward growing public dissatisfaction with politicians and anti-establishment sentiment among voters. Lastly, this article raises important questions about the implications of electing political amateurs to Congress. 

About the Author(s): Rachel Porter is an Assistant Professor of Political Science at the University of Notre Dame and Sarah A. Treul is an Associate Professor of Political Science at the University of North Carolina at Chapel Hill. Their research “Evaluating (in)experience in congressional elections” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Citizens as a democratic safeguard? The sequence of sanctioning elite attacks on democracy” by Marc S. Jacob is summarized by the author below.

In various democracies worldwide, including Türkiye, Hungary, and Poland, democratically elected incumbents have eroded crucial democratic institutions during their terms. Many of these politicians have managed to maintain electoral support despite widespread public endorsement of democratic governance. This raises important questions: Why do voters often fail to oust these politicians in elections? And are there specific segments of the electorate, such as politically informed, liberal, anti-majoritarian, or moderate voters, who might be more inclined to withdraw their support from such politicians? 

A crucial aspect of how citizens can remove undemocratic politicians from office revolves around the sequence in which political elites come to power, undermine democratic institutions, and how voters respond in elections. In electoral democracies, elites are initially elected into office and, over time, can either bolster or undermine democratic institutions, such as the integrity of electoral bodies. Voters can only directly respond when elites violate democratic principles in the subsequent election, meaning that voters can only react retrospectively to undemocratic elite behavior. Another implication of this sequence is that voters must disapprove of this elite conduct, and a sufficiently large number of them must translate this disapproval into revised vote choices. 

To investigate how voters assess and ultimately react in their voting choices to undemocratic elite behavior, I conducted a survey experiment in Poland. This experiment simulated the real-world election sequence, where participants were placed in a hypothetical runoff election between candidates from the two principal political alliances in Poland (the incumbent national conservative Law and Justice party and the liberal Civic Coalition). After selecting a candidate, participants were informed whether their chosen candidate won or lost the election. In the event of a victory, they were also informed if their candidate had undermined electoral integrity by refusing to appoint members from another party to the Electoral Commission. In case of defeat, they were informed whether their candidate conceded to the election’s winner. Participants then evaluated this behavior in terms of approval and perceived adherence to democratic principles and indicated whether they would support that politician in the subsequent election. 

The results indicated that Polish voters generally disapproved of their preferred candidate’s undemocratic conduct. However, this disapproval had minimal impact, if any, on reducing support for politicians who had violated electoral principles. Even though more informed and liberal-democratic-oriented voters were more critical of undemocratic conduct, the results suggested that segments traditionally seen as more supportive of democracy would not necessarily remove undemocratic politicians from office. 

What follows from this evidence? Citizens tend to dislike undemocratic conduct even when committed by politicians they support. However, this disapproval mostly does not manifest in their political behavior. Effective containment of the decay of democratic institutions requires changes in political behavior, but this study suggests that most citizens are unwilling to reconsider their choices even when they oppose undemocratic elite conduct. 

About the Author: Marc S. Jacob is a postdoctoral fellow with the Polarization Research Lab at Stanford University and recently completed his Ph.D. at ETH Zurich, Switzerland. Their research “Citizens as a democratic safeguard? The sequence of sanctioning elite attacks on democracy” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Great expectations: The effect of unmet labor market expectations after higher education on ideology” by Loreto Cox is summarized by the author below.

Higher education has massively expanded around the world—graduation rates in the OECD countries increased from 31% in 1995 to 49% in 2017 (OECD 2017). On average, higher education graduates earn 56% more than high school graduates (OECD 2017). Yet, there is great variation in outcomes and many graduates benefit less than they expected (or not at all) from their education. Teichler (2007) found that 19% of European college graduates find their work situation four years after graduation worse than expected. Data collected for the current study shows that in Chile, one year after graduation, 65% of graduates earn less than they expected when they finished their studies. 

What are the political consequences of these unmet expectations? After years of investing effort and money, how does realizing that the labor market does not value their studies as expected affect graduates’ political stances? The effects of unmet expectations on graduates’ political behavior have received little scholarly attention. Expectations play a fundamental role in individual satisfaction and are likely to matter in higher education. 

This paper focuses on the effect of the gap between expected and actual outcomes from higher education on ideology regarding government vs. individual responsibility. I study the case of Chile, which has experienced one of the world’s greatest expansions of higher education enrollment since 1990. My research strategy relies on two different empirical methods: a survey experiment and a panel analysis, which produce consistent results. I use a large sample and novel two-wave panel with an embedded experiment that I designed and implemented with the collaboration of 49 higher education institutions, which together enroll 72% of the country’s undergraduates. The panel follows students as they enter the labor market, surveying 14,233 higher education students or recent graduates around the time of graduation in 2016 and administering a follow-up survey to 3,948 of them one year later. 

In the experiment, subjects were asked how much they expect to earn after graduation and then half of them randomly received information on the average labor outcomes of past graduates of their institution who were awarded the same degree, based on administrative data. Larger gaps between respondents’ expectations and past graduates’ outcomes resulted in more pro-government ideology. The analysis of the respondents’ actual gaps after one year yields similar results. Both sets of results show stronger effects for unmet expectations than for exceeded expectations, in line with loss aversion. The consistent experimental and panel results appear to come from a reduction in perceptions of social mobility: unmet expectations erode perceived social mobility, which in turn implies a pro-government, leftward shift.  

Theoretically, this finding adds a new, previously understudied dimension to the long-standing association between material conditions and ideology: the role of expectations. It may also help explain the politicization of higher education and the emergence of protests among the young and educated, who often promote left-wing ideologies in Chile and elsewhere.

About the Author: Loreto Cox is an Assistant Professor at the Escuela de Gobierno, Pontificia Universidad Católica de Chile. Their research “Great expectations: The effect of unmet labor market expectations after higher education on ideology” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.

The forthcoming article “Can Close Election Regression Discontinuity Designs Identify Effects of Winning Politician Characteristics?” by John Marshall is summarized by the author below.

More than one hundred published articles now use what I call politician characteristic regression discontinuity (PCRD) designs. These designs compare outcomes of interest across districts that narrowly elected politicians who differ in an observable characteristic X, such as gender, incumbency, party affiliation, or partisan alignment with other levels of government, from the candidate they defeated. PCRD designs are appealing and important because of their potential to investigate whether the characteristics of politicians matter for government responsiveness, electoral success, or citizen participation and welfare.

This is a non-standard application of the regression discontinuity (RD) design. RD designs typically compare units above and below a threshold; indeed, close elections can identify an average individual-level effect of getting elected by comparing politicians who narrowly win to those who narrowly lose. In contrast, PCRD designs compare districts narrowly won by one type of politician to districts narrowly won by a different type. By connecting the standard RD to this non-standard application, this article seeks to clarify what effects PCRD designs can and cannot identify.

A key conceptual decision for researchers using PCRD designs is to define what is and is not part of their treatment of interest, and thus their target estimand. I consider two leading estimands, which are both local to districts with close elections between candidates with and without the characteristic of interest. Akin to conjoint experiments, the first is the average effect of electing a politician with (binary or binarized) characteristic X, holding all their other politician characteristics Z constant (usually averaging over the distribution of Z in close elections between candidates with and without characteristic X). Rather than isolate the effect of X, the second estimand is the average effect of the bundle of characteristics that come with possessing X relative to not possessing X. Applied researchers consider both estimands, although many papers should state their estimand more explicitly.

I focus mostly on the challenges of identifying the effect of a particular X, using university education as another example in this blog. While separating the effect of a leader’s education from other characteristics may be theoretically appealing, PCRD designs yield biased estimates under the relatively weak standard RD assumptions for two reasons. The obvious reason is that politicians’ characteristics are usually correlated. Consequently, university educated politicians naturally differ from less educated politicians in other ways, like their policy preferences. This article highlights a subtler reason: PCRD designs introduce a form of post-treatment bias, even if university education were randomly assigned, by conditioning on a close race. For example, if electorates prefer more educated politicians, then the average university educated politician who narrowly wins must be less desirable to voters in other ways – what I call compensating differentials, such as advocating unpopular policies. The direction of bias will depend on the application, although PCRD designs may understate the effect of X when the candidate characteristics that appeal to voters affect district outcomes similarly.

My analysis has several implications for using PCRDs to isolate the effect of a single politician characteristic. First, the bias due to compensating differentials only disappears where the potentially large set of such compensating differentials do not influence a candidate’s vote share or do not affect the outcome of interest. Such assumptions, which are far stronger than the usual RD continuity assumption, are rarely empirically plausible. Averting the natural correlations between characteristics requires further assumptions, which may rely on manipulations that are not possible in isolation.

Second, if neither assumption holds, the need for compensating differentials implies that we should expect to find imbalances across other politician characteristics. While close elections still guarantee continuity in district-level covariates, balance tests for politician-level characteristics now serve to illuminate the nature of the compound treatment. Instead of implying an PCRD design is valid, failing to detect differences between observable politician characteristics suggests that there are differences in unobserved characteristics or between combinations of characteristics.

The second estimand, which is carefully articulated and exemplified by Hall (2015), embraces a bundled treatment. Rather than attempt to distinguish the effects of X from Z, this approach obviates the problem of politician-level confounding by subsuming all (observed and unobserved) politician characteristics into a broad conception of treatment. My identification result demonstrates that, under the standard RD assumptions, PCRD designs capture the effect of an average bundle of characteristics for politicians defined by X (weighted by the correlation between X and each Z) relative to an analogous average bundle for politicians who do not possess X, among the set of politicians in close races.

Beyond the appeal of requiring weaker identifying assumptions, the value of an estimand encompassing many characteristics depends on the application. The broad treatment is likely to be most useful when the analyst cares about bundles of characteristics that approximate available policy choices, such as whether political parties should recruit university educated candidates. Because some characteristics in the bundle are unobserved and each characteristic’s contribution to the overall effect is not identified, the bundled approach is less useful for testing theories which seek to estimate effects of a particular politician characteristic holding other characteristics equal, such as whether rising education has improved governance.

Ultimately, PCRD designs work differently from standard RD designs. To identify effects of specific well-defined treatment in the way that standard RD designs usually do, PCRD must impose strong additional assumptions. To maintain the relatively weak standard RD assumptions, researchers must embrace a bundled treatment that may be policy-relevant but less useful for accumulating evidence about theoretical mechanisms and informing high-dimensional policy decisions.

About the Author: John Marshall is an Associate Professor of Political Science at Columbia University. Their research “Can Close Election Regression Discontinuity Designs Identify Effects of Winning Politician Characteristics?” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.