Districting that minimizes partisan bias

The shapes of electoral districts determine how votes translate into seats. When districts favor certain political parties, electoral results can be disproportionate and the public may lose faith in the political process. Disagreement about appropriate district shapes is subjective, rarely resolved, and often leads to lawsuits. Previously, many authors have called for objective districting criteria. We offer a novel synthesis of models that enables the proactive comparison of district maps, by relating a planar graph partition, the single-member plurality rule, the maximin decision rule, and any agreed measure of partisan bias with a territorial map and historical vote results. Historical vote totals avoid the complexity and uncertainty associated with counterfactual models of vote swing. Districting plans could be objectively compared on such criteria as party proportionality or compact shape to reject plans with worse bias. Objective tools to reduce partisan bias in district maps could boost collaborative participation, increase perceptions of fairness and justice, and reduce costs.

Policy, Ignorance, and the Will of the People

It is well established that the general population tend to lack in-depth knowledge about key political and policy matters. What are the implications for policymaking? This chapter considers this question in the context of immigration policy, reporting first on a focus group study which offers evidence that reported desires for a reduced number of immigrants might ultimately reflect a desire for immigrants of (perceived) high quality, not a reduction in overall quantity, where quality is defined in terms of fiscal impact. The chapter then argues that public preferences for such “good immigrants” are problematic, deploying a number of counterfactual models that suggest that such preferences are based on mistaken beliefs, and arguing that they thereby likely fail to reflect what the person truly desires. These findings extend beyond immigration policy and serve to highlight the often-overlooked problem that policies implemented with reference to popular sentiments might not capture “the will of the people.”

Beyond Behavior: Prevalences, Penalties, and Ethno-racial Inequalities in Poverty

Ethno-racial differences in poverty are substantial and persistent in the US. To explain these differences, scholars have relied largely on behavioral explanations, which argue that poverty is the result of high prevalences of problematic behaviors or “risks.” Given substantial differences in the prevalence of risks, scholars intuit that ethno-racial differences in poverty are explained by disproportionately high prevalences of risks in Black and Latino populations. However, these approaches rely heavily on untested assumptions regarding the relationship between risks and poverty rates. Using the 1993-2016 Current Population Survey and the Urban Institute’s TRIM3 model to derive high-quality estimates of income and poverty, I confirm persistent and substantial ethno-racial differences in poverty. Next, I employ a prevalences and penalties framework to compare risks in Black, Latino, and white-lead households. This framework is then leveraged to estimate counterfactual models to predict Black and Latino poverty rates given alternative prevalences of risks. The findings demonstrate that if the prevalence of risks for Black and Latino Americans was equal that of whites, poverty rates would remain over twice as high for Black and Latino individuals compared to whites. Furthermore, even when risks are eliminated for Black and Latino Americans, poverty remains substantially higher compared to whites. These findings undermine behavioral approaches to understanding poverty and point to the need for scholars to pursue alternatives, including structural and political explanations.

A Comparison of Different Data-driven Procedures to Determine the Bunching Window

In this article, we compare three data-driven procedures to determine the bunching window in a Monte Carlo simulation of taxable income. Following the standard approach in the empirical bunching literature, we fit a flexible polynomial model to a simulated income distribution, excluding data in a range around a prespecified kink. First, we propose to implement methods for the estimation of structural breaks to determine a bunching regime around the kink. A second procedure is based on Cook’s distances aiming to identify outlier observations. Finally, we apply the iterative counterfactual procedure proposed by Bosch, Dekker, and Strohmaier which evaluates polynomial counterfactual models for all possible bunching windows. While our simulation results show that all three procedures are fairly accurate, the iterative counterfactual procedure is the preferred method to detect the bunching window when no prior information about the true size of the bunching window is available.

The effects of the lower ignition propensity cigarettes standard in Estonia: time-series analysis

BackgroundIn 2011, the lower ignition propensity (LIP) standard for cigarettes was implemented in the European Union. Evidence about the impact of that safety measure is scarce.ObjectiveThe aim of this paper is to examine the effects of the LIP standard on fire safety in Estonia.MethodsThe absolute level of smoking-related fire incidents and related deaths was modelled using dynamic time-series regression analysis. The data about house fire incidents for the 2007–2013 period were obtained from the Estonian Rescue Board.ResultsImplementation of the LIP standard has reduced the monthly level of smoking-related fires by 6.2 (p<0.01, SE=1.95) incidents and by 26% (p<0.01, SE=9%) when estimated on the log scale. Slightly weaker evidence was found about the fatality reduction effects of the LIP regulation. All results were confirmed through counterfactual models for non-smoking-related fire incidents and deaths.ConclusionsThis paper indicates that implementation of the LIP cigarettes standard has improved fire safety in Estonia.

STRUCTURAL EQUATIONS AND BEYOND

Recent accounts of actual causation are stated in terms of extended causal models. These extended causal models contain two elements representing two seemingly distinct modalities. The first element are structural equations which represent the “(causal) laws” or mechanisms of the model, just as ordinary causal models do. The second element are ranking functions which represent normality or typicality. The aim of this paper is to show that these two modalities can be unified. I do so by formulating two constraints under which extended causal models with their two modalities can be subsumed under so called “counterfactual models” which contain just one modality. These two constraints will be formally precise versions of Lewis’ (1979) familiar “system of weights or priorities” governing overall similarity between possible worlds.

What Would Have Been Is Not What Would Be: Counterfactuals of the Past and Potential Outcomes of the Future

Epidemiology is often described as the basic science of public health. A mainstay of epidemiologic research is to uncover the causes of disease that can serve as the basis for successful public-health interventions (e.g., Institute of Medicine, 1988; Milbank Memorial Fund Commission, 1976). A major obstacle to attaining this goal is that causes can never be seen but only inferred. For this reason, the inferences drawn from our studies must always be interpreted with caution. Considerable progress has been made in the methods required for sound causal inference. Much of this progress is rooted in a full and rich articulation of the logic behind randomized controlled trials (Holland, 1986). From this work, epidemiologists have a much better understanding of barriers to causal inference in observational studies, such as confounding and selection bias, and their tools and concepts are much more refined. The models behind this progress are often referred to as ‘‘counterfactual’’ models. Although researchers may be unfamiliar with them, they are widely (although not universally) accepted in the field. Counterfactual models underlie the methodologies that we all use. Within epidemiology, when people talk about a counterfactual model, they usually mean a potential outcomes model—also known as ‘‘Rubin’s causal model.’’ As laid out by epidemiologists, the potential outcomes model is rooted in the experimental ideas of Cox and Fisher, for which Neyman provided the first mathematical expression. It was popularized by Rubin, who extended it to observational studies, and expanded by Robins to exposures that vary over time (Maldonado & Greenland, 2002; Hernan, 2004; VanderWeele & Hernan, 2006). This rich tradition is responsible for much of the progress we have just noted. Despite this progress in methods of causal inference, a common charge in the epidemiologic literature is that public-health interventions based on the causes we identify in our studies often fail.