Knowledge that Transforms

This paper attempts to distinguish empirically real versus nominal sources of fluctuations in real and nominal exchange rates. The distinction is obtained by imposing the following restriction on the bivariate vector autoregression of real and nominal exchange rates over the current flexible rate period: nominal shocks are required to have no permanent effect on the level of the real exchange rate. Given this identification scheme, the author analyzes the dynamic effects and relative importance of real and nominal shocks with regards to exchange rates. The findings indicate that real shocks dominate nominal shocks for both exchange rate series over short and long frequencies. Copyright 1992 by MIT Press.

The composite cost function the authors propose combines a quadratic output structure with a log-quadratic input price structure and is well suited for examining economies of scope, subadditivity, and other important properties of multiproduct firms. To compare the composite model with an appropriate set of alternative functional forms, they develop a parsimonious--but general--specification that nests the standard translog cost function, the generalized translog cost function, a separable quadratic cost function, and the composite cost function. An application to economies of scope in banking confirms the advantages of the composite model. Copyright 1992 by MIT Press.

The impact of the stock of public capital on costs of production in the private sector is examined using annual data over the period 1958-89. A translog cost function is estimated and public capital is found to be a significant input. The estimates indicate that public capital has positive marginal product and that private and public capital are complements in production, rather than substitutes. Copyright 1992 by MIT Press.

This paper investigates if young internal migrants in the United States experience economic assimilation as they adapt to their new residential location. Using data from the National Longitudinal Survey of Youth, the authors examine how the hourly earnings of interstate migrants are affected by the number of years they have spent in their destination state. Their study indicates that internal migrants initially earn less than natives but that this wage differential disappears within a few years. Moreover, the initial wage disadvantage of internal migrants depends upon the distance moved and economic conditions in the destination labor market. Copyright 1992 by MIT Press.

This paper shows a simple method of approximating the exact distribution of the Durbin-Watson Test Statistic for first-order autocorrelation in a nonlinear model.The proposed Approximate Nonlinear Durbin-Watson (A.N.D.) test has good size and power when compared to alternatives.

Exports from Britain to the United States from 1900 to 1940 are examined to ascertain the effect of exchange-rate volatility on the volume of trade. In addition to using a rolling standard deviation measure of exchange-rate uncertainty, the conditional variance of the exchange-rate series modeled.as a generalized autoregressive conditional heteroskedastic process is used to generate an alternative measure of exchange-rate uncertainty. The results of estimation using the two measures of exchange-rate volatility suggest that increases in the volatility of the real exchange rate reduce the volume of trade. Copyright 1992 by MIT Press.

This paper analyzes the relationship between maternal labor supply and children's cognitive development, using a sample of three- and four-year-old children of female respondents from the 1986 National Longitudinal Surveys Youth Cohort (NLSY). Respondents in the NLSY were aged 21 to 29 in 1986; thus our sample consists of children of relatively young mothers. We show that for this group the impact of maternal labor supply depends upon when it occurs. Maternal employment is found to have a negative impact when it occurs during the first year of the child's life and a potentially offsetting positive effect when it occurs during the second and subsequent years. We find some evidence that boys are more sensitive to maternal labor supply than girls though the gender difference is not significant. The negative first-year effect is not mitigated to any great extent by the increased maternal income that accompanies it, though the increase in maternal income does appears to play an important role in producing the positive effect in the second and later years.

The comovement between exports and productivity observed in many countries suggests a direct link between these two variables. This paper tries to establish whether such a causal link exists for four developed market economies, using co-integration and Granger-causality techniques. These techniques offer a means to overcome serious problems encountered in previous attempts to examine this relationship, while recent trade theory suggests that the relationship between trade and productivity is fundamentally ambiguous. Both reasons call for more empirical evidence. The findings of the econometric analysis suggest that exports, productivity and the terms of trade move together in the long run in all countries except the United Kingdom.

What accounts for the apparent preference of firms to finance investment with internal funds? Recent theories stress information problems in capital markets, while older theories emphas ize the transactions costs of external finance. To test these competing hypotheses, the authors estimate the sensitivity of investment spend ing to internal funds across firms likely to face varying degrees of information problems and transactions costs. Several attributes are used to differentiate these firms. The results provide some support for information asymmetries as well as a source of the financing hierarc hy but indicate no significant role for transactions costs. Copyright 1992 by MIT Press.

This paper seeks to understand the recent history of U.S. external imbalances by identifying the "long-run tendency" of the U.S. current account balance and investigating its behavior. The procedure that is adopted is to estimate cointegrating regressions between U.S. exports and imports of goods and services. Estimates from cointegrating regressions between several measures of U.S. exports and imports show that up to about the end of 1983 the U.S. current account tended toward zero. Since that time, there has been an apparent structural shift resulting in a long-run tendency for a deficit in excess of $100 billion per year. Copyright 1992 by MIT Press.