Working Papers

Contaminated sediments in rivers, lakes, and harbors around the world result in diminished ecological health, degradation of environmental resources, economic losses, and, in rare cases, impacts to human health. Despite the ongoing interest in the cleanup of contaminated sediments in rivers and harbors, little progress has been made in reducing the number of contaminated sites worldwide. Much of the difficulty in advancing this cause can be attributed to the high cost of sediment cleanups and the difficulty in assigning financial responsibility for the cost of the cleanup. Simple schemes dependent on identifying polluters are fraught with underlying complexity. More elaborate approaches tied in with waterfront redevelopment show some promise but are yet to be applied routinely. New advances in the understanding of how sediments may, or may not, factor into circularity pose new challenges and opportunities, with the potential to complement new funding paradigms. The most promising possibilities for achieving circularity in sediment management lie in a kind of “punctuated circularity,” which requires idiosyncratic, project-based beneficial use opportunities. However, these ideal situations are likely to remain rare for the foreseeable future, without advancements in technology and regulatory approaches, as well as development of market demand for the products made from contaminated sediments.

For most of US history, the federal government let states conduct congressional elections with only minimal interference. This changed with Congress’s passage of the Voting Rights Act (1964) and the Supreme Court’s one-man-one-vote (“OMOV”) decision in Carr (1965). Yet even then the Supreme Court stopped short of overruling gerrymandered congressional districts. Fifteen years ago, Justice Kennedy acknowledged that workable approaches still did not exist, but challenged litigants to do better. The Supreme Court’s devastating opinion in Rucho v. Common Cause (2019) marks the end of this experiment.The question remains whether some fresh departure can satisfy the Court’s objections.

This paper investigates the leading candidates. The most conservative possibility is to tighten traditional visual criteria like “contiguity” and “compactness” to make them binding. We present detailed numerical arguments showing that such rules would have to be so stringent as to deprive legislators of practically all discretion in drawing lines. This, however, would split communities at random. We argue that this (a) violates OMOV just as a deliberate gerrymander would, and (b) disrupts grassroots networks that voters rely on to educate themselves, making votes less valuable to those who cast them. It follows that OMOV is always improved by replacing random districting with determinate rules that track communities.

Our second candidate implements this strategy. Remarkably, recent social science research shows (a) that education and social pressure across voters play an important role in shaping community opinion, and (b) that the probability of such interactions between voters can be reliably estimated from an inverse square law. We describe a new open source software program that uses these insights to track community and test it against county-level population data from Texas. We show that the resulting maps are visually similar to those produced by legislatures; are fully determinate; efficiently balance OMOV constraints against respect for community; and are robust against manipulation. The chief downside is that some districts are discontinuous, although this is rare and could be mitigated even further. In the meantime, our algorithm provides a valuable safe harbor for States hoping to avoid future court challenges, an attractive model for reform legislation, and a transparent benchmark for exposing gerrymanders to public scrutiny.

Research Summary

In late 2014, California voters passed Proposition 47 that redefined a set of less serious felony drug and property offenses as misdemeanors. We examine how racial disparities in criminal court dispositions in San Francisco change in the years before (2010-2014) and after (2015-2016) the passage of Proposition 47. We decompose racial disparities in court dispositions into components due to racial differences in offense characteristics, involvement in the criminal justice system at the time of arrest, pretrial detention, criminal history, and the residual unexplained component. Before and after Proposition 47 case characteristics explain nearly all of the observable race disparities in court dispositions. However, after the passage of Proposition 47 there is a narrowing of racial disparities in convictions and incarceration sentences that is driven by lesser weight placed on criminal history, active criminal justice status, and pretrial detention in effecting court dispositions.

Policy Implications

The findings from this study suggest that policy reforms that scale back the severity of punishment for criminal history and active criminal justice status for less serious felony offenses may help narrow racial inequalities in criminal court dispositions. Efforts to reduce the impact of racial inequalities in mass incarceration in other states should consider reforms that reduce the weight that criminal history, pretrial detention, and active probation status has on criminal defendants’ eligibility for prison for less serious drug and property offenses.

We evaluate whether California’s state proposition 47 impacted state violent and property crime rates. Passed by the voters in November 2014, the proposition redefined many less serious property and drug offenses that in the past could be charged as either a felony or misdemeanor to straight misdemeanors. The proposition caused a sudden and sizable decline in county jail populations, a moderate decline in the state prison population, a decrease in arrests for property and drug offenses, and a wave of legal petitions filed for retroactive resentencing and reclassification of prior convictions. We make use of multiple strategies to estimate the effect of the proposition, including state-level synthetic cohort analysis, within-state event study estimates based on state-level monthly time series, and a cross-county analysis of changes in county-level crime rates that exploit heterogeneity in the effects of the proposition on local criminal justice practices. We find little evidence of an impact on violent crime rates in the state. Once changes in offense definitions and reporting practices in key agencies are accounted for, violent crime in California is roughly at pre-proposition levels and generally lower than the levels that existed in 2010 prior to a wave major reforms to the state’s criminal justice system. While our analysis of violent crime rates yields a few significant point estimates (a decrease in murder for one method and an increase in robbery for another), these findings are highly sensitivity to the method used to generate a counterfactual comparison path. We find more consistent evidence of an impact on property crime, operating primarily through an effect on larceny theft. The estimates are sensitive to the method used to generate the counterfactual, with more than half of the relative increase in property crime (and for some estimates considerably more) driven by a decline in the counterfactual crime rate rather than increases for California for several of the estimators that we employ. Despite this sensitivity, there is evidence from all methods tried that property crime increased with, a ballpark summary of five to seven percent roughly consistent with the totality of our analysis. Similar to violent crime, California property crime rates remain at historically low levels.

We develop empirically-grounded estimates of willingness-to-pay to avoid excess mortality risks caused by climate change. Using 40 countries' subnational data, we estimate a mortality-temperature relationship that enables global extrapolation to countries without data and projection of its future evolution, accounting for adaptation benefits. Further, we develop a revealed preference approach to recover unobserved adaptation costs. We combine these components with 33 high-resolution climate simulations, which produces a right-skewed distribution of global WTP with a mean of $38.1 per tCO2 under a high emissions scenario. Projections generally indicate increased mortality in today's poor locations and higher adaptation expenditures in rich ones.

This study investigates the rise of legislative restrictions on foreign funding to NGOs, a global trend occurring not only in aid-dependent, non-democratic regimes but also more recently in developed economies and democracies. I create a dataset of 65 countries with legislative restrictions and differentiate between developed and developing economies, as well as consolidated and transitioning democracies. I use panel data, polynomial interpolation, and logistic regression to examine predictors of foreign funding restrictions, in particular, foreign aid, natural resource extraction, conflict risk, and regime type. By collapsing the wide variety of cases into regime type, and within regime type, differentiating between resource-rich countries as well as countries prone to global risk in terms of national and subnational power, my framework accounts for the rise of legislative restrictions across diverse political and economic contexts, and pays close attention to the paradox of the rise of foreign funding restrictions in developing country democracies. Overall, the presence of the extractive industry is highly significant across regime types and even in democracies in particular as the majority of countries with restrictions are resource-rich.

We develop a simple cross-sectional research design to identify causal effects that is robust to unobservable heterogeneity. When many observational units are dense in physical space, it may be sufficient to regress the “spatial first differences” (SFD) of the outcome on the treatment and omit all covariates. This approach is conceptually similar to first differencing approaches in timeseries or panel models, except the index for time is replaced with an index for locations in space. The SFD design identifies plausibly causal effects, even when no instruments are available, so long as local changes in the treatment and unobservable confounders are not systematically correlated between immediately adjacent neighbors. We demonstrate the SFD approach by recovering new cross-sectional estimates for the effects of time-invariant geographic factors, soil and climate, on long-run average crop productivities across US counties — relationships that are notoriously confounded by unobservables but crucial for guiding economic decisions, such as land management and climate policy.