Working Papers

Analysis of projects generating 80% of total offset credits issued by ARB under its U.S. Forest projects offset protocol shows that 82% of the credits generated by these projects likely do not represent true emissions reductions, due to the protocol’s leakage accounting methods. The total quantity of over-crediting across these 36 projects equals approximately 80 million tons of CO2. For context, the U.S. Forest Protocol has generated 80% of the offset credits in California’s cap-and- trade program; the estimated over-crediting is equal to one third of the total expected effect of California’s cap-and-trade program on emissions during 2021-2030.

I investigate medium- and long-term impacts of the Great Recession on post-recession college graduates. Most so-called “scarring” models emphasize effects of initial conditions that attenuate over the first decade of a worker’s career. But early career recessions may also have permanent effects. I decompose the recent cohorts’ experience into transitory time effects, medium-term scarring, and permanent cohort effects. Cohort effects are strongly cyclical. Medium-term scarring explains only half of this cyclicality. The long-run cumulative effect of the recession on graduates’ employment is more than twice as large as the immediate effect.

Accumulating evidence indicates that environmental temperature substantially affects economic outcomes and violence, but the reasons for this linkage are not well understood. We systematically evaluate the effect of thermal stress on multiple dimensions of economic decisionmaking, judgment, and destructive behavior with 2,000 participants in Kenya and the US who were randomly assigned to different temperatures in a laboratory. We find that most dimensions of decision-making are unaffected by temperature. However, heat causes individuals to voluntarily destroy other participants’ assets, with more pronounced effects during a period of heightened political conflict in Kenya.

The genius of the Framers lay in identifying and systematically planning for the known pathologies of democratic government. That said, most of their evidence was limited to Greek and Roman history. This gave little warning of the disastrous polarization that would destabilize European mass democracies over the next two centuries. This paper asks how the Framers might have extended their design had they understood these dangers.

We start by noting that the well-known “median voter theorem,” which holds that successful American political parties must position themselves near the center, depends on very special assumptions about how public opinion is actually distributed. This implies that American politics can and probably will behave very differently as polarization increases. This paper presents a typology of possible polarizations, and argues from both theory and history that each is associated with its own unique political style. Significantly, only some of these styles favor consensus politics. Others are confrontational, with extremists deliberately sabotaging government to coerce opponents. Recent government shutdowns are an extreme expression of these tactics.

One peculiarity of coercive politics is that it depends at least as much on political passion (“intensity”) as raw vote totals. Asking whether such politics are democratically legitimate necessarily forces us beyond the familiar language of one-man-one-vote (“OMOV”) theories that count all votes equally. This philosophical question also has a practical side. After all, no real government can go on passing laws that increase public anger forever. The paper develops a simple baseline model of intensity-weighted voting and asks how familiar American rules like supermajorities, presidential vetoes, and filibusters have modified OMOV to avoid oppressive outcomes in the past. In doing so, we rely heavily on European historical precedents and ask how these might change in American circumstances.

We argue that coercive politics, while sometimes pathological, is an essential tool for measuring and accommodating voter intensity. It follows that reform should aim less to suppress coercive methods than to make them less costly. We argue that suitably reformed versions of government shutdowns, supermajorities, sunset legislation, regular order, and stiffened rule of law incentives offer the fastest path to restoring cooperative politics.

This paper shows that greater global spatial correlation of productivities can increase crosscountry welfare dispersion by increasing the correlation between a country’s productivity and its gains from trade. We causally validate this general-equilibrium prediction using a global climatic phenomenon as a natural experiment. We find that gains from trade in cereals over the last halfcentury were larger for more productive countries and smaller for less productive countries when cereal productivity was more spatially correlated. Incorporating this general-equilibrium effect into a projection of climate-change impacts raises projected international inequality, with higher welfare losses across most of Africa.

Intergenerational income transmission varies across commuting zones (CZs). I investigate whether children’s educational outcomes help to explain this variation. Differences among CZs in the relationship between parental income and children’s human capital explain only one-ninth of the variation in income transmission. A similar share is explained by differences in the return to human capital. One-third reflects earnings differences not mediated by human capital, and 40% reflects differences in marriage patterns. Intergenerational mobility appears to reflect job networks and the structure of local labor and marriage markets more than it does the education system.

In this project we evaluate the application of the case management and treatment delivery practices developed under the HOPE probation strategy to pretrial individuals who are conditionally released from jail subject to criminal justice supervision.  In the jurisdiction we study (Honolulu, Hawaii), defendants on supervised release are typically monitored by pretrial officers located at the county jail.  The revocation of supervised release occurs once a defendant has failed to comply several times with a set of pre-specified conditions, including but not necessarily limited to refraining from drug use and additional criminal activity, maintaining contact with the assigned pretrial officers, and making all scheduled court dates.  The intervention we evaluate applies random drug testing in conjunction with swift, certain, consistent, and proportionate sanctions to pretrial misconduct.  That is to say, misconduct is met with quickly administered arrest and re-incarceration, yet subsequent jail spells are proportionate to the seriousness of the violation.  The intervention also includes drug treatment interventions for those who repeatedly fail drug tests (or who request treatment services) and direct interaction following each violation with the presiding judge of a court devoted to HOPE probation as well as HOPE pretrial defendants. 

Between September 2014 and August 2016, felony defendants who failed to make bail and who were granted supervised release were randomly assigned to either status-quo pretrial services or to the HOPE pretrial treatment group.  We use administrative data on drug tests, revocations, supervised release case dispositions, and criminal history records to assess whether applying HOPE to individuals on pretrial supervised release impacts various measures of pretrial misconduct, criminal case disposition, and post-disposition arrests.  Our findings are the following:

(1) HOPE treatment group members experience more pretrial supervised release revocations most of which are better characterized as modifications but fewer permanent revocations ending the supervised release term relative to control group members. 

(2) Treatment under HOPE pretrial reduced the proportion of drug tests resulting in failure.  The drug test failure rate for treatment group members was roughly 21 to 30 percent lower than the comparable failure rate observed for the control group with the difference statistically significant.

(3) HOPE treatment did not impact total jail days served between the supervised release date and the disposition date for the criminal case.  However, treatment group members serve jail days earlier in their supervised release term while control group members serve more jail days later.

(4) Average total pretrial arrests occurring after supervised release does not differ significantly between the treatment and control group.  However, treatment group members are significantly and substantially less likely to be arrested with a new criminal charge.

(5) Treatment group members are less likely to be convicted and less likely to be convicted for a felony.

(6) We do not find statistically significant effects of treatment on post-disposition arrest outcomes.

The synthetic control method (SCM) is a popular approach for estimating the impact of a treatment on a single unit in panel data settings. The “synthetic control” is a weighted average of control units that balances the treated unit’s pre-treatment outcomes as closely as possible. The curse of dimensionality, however, means that SCM does not generally achieve exact balance, which can bias the SCM estimate. We propose an extension, Augmented SCM, which uses an outcome model to estimate the bias due to covariate imbalance and then de-biases the original SCM estimate, analogous to bias correction for inexact matching. We motivate this approach by showing that SCM is a (regularized) inverse propensity score weighting estimator, with pre-treatment outcomes as covariates and a ridge penalty on the propensity score coefficients. We give theoretical guarantees for specific cases and propose a new inference procedure. We demonstrate gains from Augmented SCM with extensive simulation studies and apply this framework to canonical SCM examples. We implement the proposed method in the new augsynth R package.