When working with panel data, many researchers wish to estimate the direct effects of time-varying factors on future outcomes. However, when a baseline treatment affects both the confounders of further stages of the treatment and the outcome, the estimation of controlled direct effects (CDEs) using traditional regression methods faces a bias trade-off between confounding bias and post-treatment control. Drawing on research from the field of epidemiology, in this article I present a marginal structural modeling (MSM) approach that allows scholars to generate unbiased estimates of CDEs. Further, I detail the characteristics and implementation of MSMs, compare the performance of this approach under different conditions, and discuss and assess practical challenges when conducting them. After presenting the method, I apply MSMs to estimate the effect of wealth in childhood on political participation, highlighting the improvement in terms of bias relative to traditional regression models. The analysis shows that MSMs improve our understanding of causal mechanisms especially when dealing with multi-categorical time-varying treatments and non-continuous outcomes.

In the context of recent economic and financial crisis in Europe, questions about the power of the core executive to control fiscal outcomes are more important than ever. Why are some governments more effective in controlling spending while others fall prey to excessive overspending by individual cabinet ministers? We approach this question by lifting the veil of collective cabinet responsibility and focusing on intra-cabinet decision-making around budgetary allocation. Using the contributions of individual cabinet members during budget debates in Ireland, we estimate their positions on a latent dimension that represents their relative levels of support or opposition to the cabinet leadership. We find some evidence that ministers who are close to the finance minister receive a larger budget share, but under worsening macro-economic conditions closeness to the prime minister is a better predictor for budget allocations. Our results highlight potential fragility of the fiscal authority delegation mechanism in adverse economic environment.

In this paper, we show that state stability exhibits a persistent and robust non-monotonic relationship with economic development. Based on observations in Europe spanning from 1 to 2000 AD, regions that have historically experienced either short- or long-duration state rule on average lag behind in their local wealth today, while those that have experienced medium-duration state rule fare better. These findings support the argument that both an absence as well as an excess of state stability are bad for economic development. State instability hinders investment for growth, while too much stability is likely indicative of elite capture and subsequent stagnation of innovation.

I propose a unified explanation for parties' joint policy and emphasis decisions which bridges saliency theory and spatial analyses of party campaigns. Party platforms are anchored by the policy preferences of activists, core supporters and target voters, leading parties to disproportionately emphasize issues where their policies are popular with all key constituencies. However, which voters a party targets relates to its historical electoral performance (“party size”). Traditionally successful (“major”) parties emphasize issues where the policies preferred by activists and core supporters are generally popular, but smaller (“minor”) parties emphasize issues where their preferred policies may be unpopular but are distinctive. Using recent European data and various empirical strategies, I show that this account has significant explanatory power beyond existing party typologies and theories of issue selection.

Do elections motivate incumbent politicians to serve their voters? In this paper, we use millions of service requests placed by residents in US cities to measure constituency responsiveness. We then test whether an unusual policy change in New York City, which enabled city councilors to run for three rather than two terms in office, improved constituency responsiveness in previously term-limited councilors’ districts. Using difference-in-differences, we find robust evidence for this. Taking advantage of differential timing of local election races in New York City and San Francisco, we also find late-term improvements to responsiveness in districts represented by reelection-seeking incumbents. Elections improve municipal services, but also create cycles in constituency responsiveness. These findings have implications for theories of representative democracy.

The striking under-representation of women in Japan has been partly attributed to gender stereotypes and prejudice toward female leadership among voters. We examine whether and to what extent candidates get rewarded or punished when they deviate from the behavioral expectations associated with their gender roles and images. Our conjoint experiment results in Japan demonstrate that not only are female candidates disadvantaged compared to their male counterparts, but also that they could lose support when they diverge from gender-based behavioral expectations. Our findings suggest that female candidates face a difficult dilemma in that they must weigh the cost of losing support for failing to conform to gender-based expectations, against the general loss of support they would incur for conforming to these expectations.

Behavioral economists and social psychologists have shown that extrinsic motivations can crowd out intrinsic motivations to act. This study examines this crowding out effect in the context of legislative behavior. By exploiting the federal nature of Swiss elections, we examine if response rates to requests of voters residing inside or outside a candidate's district vary based on the electoral competition candidate legislators face. We report two main findings. First, we find a high response rate among Swiss candidates (66 percent) which remains high for voters who reside outside a candidate's district (59 percent) suggesting that intrinsic motivations are a key driver of constituency effort. Second, the response to voters who reside inside a candidate's district is more pronounced for candidates confronted with a high degree of electoral competition. This suggests that extrinsic motivations are important for constituency work, but at the same time their presence might crowd out intrinsic motivations. This evidence suggests that the relationship between electoral competition and responsiveness might be less straightforward than assumed.

Globalization has been one of the biggest driving forces of the last half century. There has been substantial disagreement about the impact that increased international integration has on income inequality. Though most agree that globalization positively affects economic output, it is no surprise that it leads to relative winners and losers within nations. The question that remains is where in the income distribution are these relative gains and losses occurring? We offer a broader picture of globalization's effects on inequality by using a dynamic compositional approach to test the impact of globalization and relative factor endowments on the composition of income. Using data from four countries, we model the effects of globalization on quantiles of the income distribution. Our findings suggest that globalization has substantial (and divergent) effects across income strata, and that these effects differ across nations based on relative factor endowments.

Coincidence Analysis (CNA) is a configurational comparative method of causal data analysis that is related to Qualitative Comparative Analysis (QCA) but, contrary to the latter, is custom-built for analyzing causal structures with multiple outcomes. So far, however, CNA has only been capable of processing dichotomous variables, which greatly limited its scope of applicability. This paper generalizes CNA for multi-value variables as well as continuous variables whose values are interpreted as membership scores in fuzzy sets. This generalization comes with a major adaptation of CNA’s algorithmic protocol, which, in an extended series of benchmark tests, is shown to give CNA an edge over QCA not only with respect to multi-outcome structures but also with respect to the analysis of non-ideal data stemming from single-outcome structures. The inferential power of multi-value and fuzzy-set CNA is made available to end users in the newest version of the R package cna.

Spatial conditionally autoregressive (CAR) models in a hierarchical Bayesian framework can be informative for understanding state politics, or any similar population of border-defined observations. This article explains how a hierarchical CAR model is specified and estimated and then uses Monte Carlo analyses to show when the CAR model offers efficiency gains. We apply this model to data structures common to state politics: A cross-sectional example replicates Erikson, Wright and McIver’s (1993) Statehouse Democracy model and a multilevel panel model example replicates Margalit’s (2013) study of social welfare policy preferences. The CAR model fits better in each case and some inferences differ from models that ignore geographic correlation.

Researchers in different social science disciplines have successfully used Facebook to recruit subjects for their studies. However, such convenience samples are not generally representative of the population. We developed and validated a new quota sampling method to recruit respondents using Facebook advertisements. Additionally, we published an R package to semi-automate this quota sampling process using the Facebook Marketing API. To test the method, we used Facebook advertisements to quota sample 2432 US respondents for a survey on climate change public opinion. We conducted a contemporaneous nationally representative survey asking identical questions using a high-quality online survey panel whose respondents were recruited using probability sampling. Many results from the Facebook-sampled survey are similar to those from the online panel survey; furthermore, results from the Facebook-sampled survey approximate results from the American Community Survey (ACS) for a set of validation questions. These findings suggest that using Facebook to recruit respondents is a viable option for survey researchers wishing to approximate population-level public opinion.

We examine the causal effect of legislative activity on private benefits, which have been largely neglected by previous research in legislative studies. By relying on a natural experiment in New Zealand, where randomly selected Members of Parliament (MPs) are given the opportunity to propose legislation, we find evidence for a causal relation between proposing a (successful) bill and the private benefits MPs receive, in terms of gifts and payments for services. We conclude that the allocation of private benefits depends on legislative performance.

Using data from the World Values Survey, we analyze the extent to which value consensus exists within countries. To do this, we introduce a statistical model which allows us to generate country-level measures of cultural heterogeneity. Our statistical approach models each country as a mixture of subcultures that are shared across the world. Our results demonstrate that value consensus varies substantially across countries and regions.

Political scientists use expert surveys to assess the latent features of political actors. Experts, though, are unlikely to be equally informed and assess all actors equally well. The literature acknowledges variance in measurement quality but pays little attention to the implications of uncertainty for aggregating responses. We discuss the nature of the measurement problem in expert surveys. We then propose methods to assess the ability of experts to judge where actors stand and to aggregate expert responses. We examine the effects of aggregation for a prominent survey in the literature on party politics and EU integration. Using a Monte Carlo simulation, we demonstrate that it is better to aggregate expert responses using the median or modal response, rather than the mean.

The welfare state literature argues that Social Democratic party representation is of key importance for welfare state outcomes. However, few papers are able to separate the influence of parties from voter preferences, which implies that the partisan effects will be overstated. I study a natural experiment to identify a partisan effect. In 1995, the Labour Party (Ap) in the Norwegian municipality of Flå filed their candidate list too late and could not participate in the local election. Ap was the largest party in Flå in the entire post-World War period, but have not regained this position. I use the synthetic control method to study the effects on welfare spending priorities. I find small and insignificant partisan effects.