We examine the impact of government funding for R&D—and defense-related R&D in particular—on privately conducted R&D and its ultimate effect on productivity growth. We estimate longitudinal models that relate privately funded R&D to lagged government-funded R&D using industry-country level data from OECD countries and firm level data from France. To deal with the potentially endogenous allocation of government R&D funds, we use changes in predicted defense R&D as an instrumental variable. In many OECD countries, expenditures for defense-related R&D represent by far the most important form of public subsidies for innovation. In both datasets, we uncover evidence of “crowding in” rather than “crowding out,” as increases in government-funded R&D for an industry or a firm result in significant increases in private sector R&D in that industry or firm. On average, a 10% increase in government-financed R&D generates a 5% to 6% additional increase in privately funded R&D. We also find evidence of international spillovers, as increases in government-funded R&D in a particular industry and country raise private R&D in the same industry in other countries. Finally, we find that increases in private R&D induced by increases in defense R&D result in productivity gains.

We first estimate the direct effects on local workers’ earnings and housing costs from increases in local labor demand caused by gains in city-level manufacturing productivity. We find that local workers benefit from productivity growth, even after subtracting increases in housing costs. These gains are larger for local less educated workers, such that productivity growth reduces local inequality. We then propose and implement a new transparent method of estimating indirect effects of local productivity growth on earnings and housing costs of workers in other cities. We find that these general equilibrium effects are economically important and disproportionately benefit college-educated workers.

Empirical researchers interested in the causal effect of the endogenous regressor often use instrumental variables. When few valid instruments are available, they typically estimate restricted specifications that impose uniform per unit treatment effects, even when these effects are likely to vary. We show that in these cases, ordinary least squares and instrumental variables estimators identify different weighted averages of all per unit effects, so the traditional Hausman test is uninformative about endogeneity. We develop a new exogeneity test that works even when the true model cannot be estimated using IV methods as long as a single valid instrument is available. We revisit three recent empirical examples to demonstrate the practical value of our test.

We estimate the effect of extreme weather on life expectancy in the United States. Using high-frequency data, we find that both extreme heat and cold result in immediate increases in mortality. The increase in mortality following extreme heat appears mostly driven by near-term displacement, while the increase in mortality following extreme cold is long lasting. We estimate that the number of annual deaths attributable to cold temperature is 0.8% of average annual deaths in our sample. The longevity gains associated with mobility from the Northeast to the Southwest account for 4% to 7% of the total gains in life expectancy experienced by the U.S. population over the past thirty years.

Critics argue that electronic voting is vulnerable to fraud. We test whether voting technology affected electoral outcomes in the 2000 and 2004 presidential elections. We find a positive correlation between use of electronic voting and George Bush vote share. The effect could have been large enough to influence the final results in some swing states. While this pattern would appear to be consistent with allegations of voting irregularities, a closer examination suggests this interpretation is unlikely. We find no evidence that electronic voting had a larger effect in swing states, or in states with a Republican secretary of state. We also find that electronic voting has a negative effect on turnout rates of Hispanics (who tend to favor Democrats). Electronic voting was more likely to be used in counties with a higher fraction of Hispanics; especially in swing states.

Most Americans are now in some form of managed care plan that restricts access to services in order to reduce costs. It is difficult to determine whether these restrictions affect health because individuals and firms self-select into managed care. We investigate the effect of managed care using a California law that required some pregnant women on Medicaid to enter managed care. We use a unique longitudinal database of California births in which we observe changes in the regime faced by individual mothers between births. We find that Medicaid managed care reduced the quality of prenatal care and increased low birth weight, prematurity, and neonatal death.