Skip to main content

Research

The Tobin Center supports policy-relevant research across Yale and beyond through the Pre-Doctoral Fellows Program, seed funding, and various forms of in-kind support. Tobin-supported research spans all of our main initiatives, from Health Policy to Climate, and also includes exploratory economics research projects with potential policy applications.

Discussion Paper
Abstract

There is widespread consensus that US infrastructure quality has been on the decline. In response, politicians across the ideological spectrum have called for increased infrastructure spending. Although the cost of infrastructure determines how much physical output each dollar of spending yields, we know surprisingly little about these costs across time and place. We help to fill this gap by using data we digitized on the Interstate highway system—one of the nation’s most valuable infrastructure assets—to document spending per mile over the history of its construction.

We make two main contributions. First, we find that real spending per mile on Interstate construction increased more than three-fold from the 1960s to the 1980s. The increase does not appear to come from states building “easy” miles first, since the increase is roughly unchanged conditional on pre-existing observable geographic cost determinants. Second, we provide suggestive evidence of the determinants of the increase in spending per mile. Increases in per -unit labor or materials prices are inconsistent with the pattern of the increase. But increases in income and housing prices explain about half of the increase in spending per mile. We find suggestive evidence that the rise of “citizen voice” in government decision-making caused increased expenditure per mile.

Proceedings of the National Academy of Sciences
Abstract

Randomized controlled trials (RCTs) enroll hundreds of millions of subjects and involve many human lives. To improve subjects’ welfare, I propose a design of RCTs that I call Experiment-as-Market (EXAM). EXAM produces a welfare-maximizing allocation of treatment-assignment probabilities, is almost incentive-compatible for preference elicitation, and unbiasedly estimates any causal effect estimable with standard RCTs. I quantify these properties by applying EXAM to a water-cleaning experiment in Kenya. In this empirical setting, compared to standard RCTs, EXAM improves subjects’ predicted well-being while reaching similar treatment-effect estimates with similar precision.