Knowledge that Transforms

This paper studies the effect of foreign competition on the extent of forecasting bias. I focus on two biases often described in the behavioral economics literature: overoptimism and excessive belief in trends. Using data from firm-level surveys in five African countries, I show that firms that do not face foreign competition generate forecasts of sales growth that have greater trend and optimism biases than firms that have foreign competitors. I further provide evidence that these erroneous forecasts have real effects on firms' inventory management. © 2006 The President and Fellows of Harvard College and the Massachusetts Institute of Technology.

This paper distinguishes between target-earnings and life cycle motivations for return migration by examining how Philippine migrants' return decisions respond to major, unexpected exchange rate changes in their overseas locations (due to the Asian financial crisis). Overall, the evidence favors the life cycle explanation: more favorable exchange rate shocks lead to fewer migrant returns.A10% improvement in the exchange rate reduces the 12-month return rate by 1.4 percentage points. However, some migrants appear motivated by target-earnings considerations: in households with intermediate foreign earnings, favorable exchange rate shocks have the least effect on return migration, but lead to increases in household investment.

Using a unique set of household level panel data, we estimate the effect of capital gains on saving by asset type, controlling for observable and unobservable household specific fixed effects. The results suggest that the decline in the personal saving rate since 1984 is largely due to the significant capital gains in corporate equities experienced over this period. Over five-year periods, the effect of capital gains in corporate equities on saving is substantially larger than the effect of capital gains in housing or other assets. Failure to differentiate wealth affects across asset types results in a significant understatement or overstatement of the size of their impact, depending on the asset.

In an effort to alleviate the perceived growth of a digital divide, the U.S. government enacted a major subsidy for Internet and communications investment in schools starting in 1998. In this paper, we evaluate the effect of the subsidy—known as the E-Rate—on Internet investment in California public schools. The program subsidized spending by 20%–90%, depending on school characteristics. Using new data on school technology usage in every school in California from 1996 to 2000 as well as application data from the E-Rate program, the results indicate that the subsidy did succeed in significantly increasing Internet investment. The implied first-dollar price elasticity of demand for Internet investment is between −0.4 and −1.1 and the greatest sensitivity is seen among urban schools and schools with large black and Hispanic student populations. Rural and predominantly white and Asian schools show much less sensitivity. Overall, by the final year of the sample, there were approximately 68% more Internet-connected classrooms per teacher than there would have been without the subsidy. Using a variety of test score results, however, we do not find significant effects of the E-Rate program, at least so far, on student performance.

This paper provides new evidence on the predictive power of dividend yields for U.S. aggregate stock returns.Following Miller and Modigliani, we construct a measure of the dividend yield that includes all cash flows to shareholders.We show that this alternative cash-flow yield has strong and stable predictive power for returns, and appears robust to a battery of tests that have been proposed in recent critiques of the predictability literature.

From 1940 to 1990, a 10% increase in a metropolitan area's concentration of college-educated residents was associated with a 0.8% increase in subsequent employment growth. Instrumental variables estimates support a causal relationship between college graduates and employment growth, but show no evidence of an effect of high school graduates. Using data on growth in wages, rents, and house values, I calibrate a neoclassical city growth model and find that roughly 60% of the employment growth effect of college graduates is due to enhanced productivity growth, the rest being caused by growth in the quality of life. This finding contrasts with the common argument that human capital generates employment growth in urban areas solely through changes in productivity.

This paper presents a monopolistic competition model of trade and multinational production that incorporates asymmetric trade barriers and international differences in production costs. The model predicts the functional form for the dependence of U.S. exports and multinational production on tariffs, distance, and production costs. To deal with simultaneity, we estimate the nonlinear equations of U.S. exports and multinational production simultaneously. In the estimation, we also include country fixed effects and allow for endogenous location choice by firms. The estimation yields reasonable estimates of the structural parameters, including the elasticity of substitution. Based on the estimates, we then simulate the effects of trade liberalization. We find that the elimination of tariffs worldwide would increase U.S. exports by 3.0% and U.S. multinational production by 21.7%. This large expansion of U.S. overseas production mainly results from an expected increase in the number of U.S. foreign affiliates in response to tariff reductions.

This paper compares technologies across space and time on the basis of factual and counterfactual substitution elasticities and argues that differences in estimated substitution elasticities should be decomposed into two counterfactual components. While the first component is designed to indicate how the ease of substitution is altered by varied economic circumstances, the second addresses the question of how technologies would compare under genuinely comparable situations. This argument is illustrated by the example of energy-price elasticities of capital before and after the oil crisis of the early 1970s.

Previous experimental research has shown that people's decisions can be influenced by options they have encountered in the past. This paper uses PSID data to study this phenomenon in the field, by observing how long people commute after moving between cities. It is found, as predicted, that (i) people choose longer commutes in a city they have just moved to, the longer the average commute was in the city they came from and (ii) when they move again within the new city, they revise their commute length, countering the impact their origin city had on their initial decision.