Knowledge that Transforms

This paper estimates income and price elasticities for bilateral world trade. In addition to testing the properties of the error terms, the dynamic specification, and the assumption of parameter constancy, the analysis presents the first application of the Bank Spectrum estimator to bilateral trade flows for all countries. The paper finds that bilateral trade elasticities exhibit enough of a dispersion to suggest that the direction of trade is sensitive to changes in income and prices. Using the bilateral elasticities as raw data, the analysis obtains the associated multilateral estimates and finds that they are both consistent with the literature and suitable to addressing questions involving multilateral trade. But the evidence also reveals that sole reliance on multilateral elasticities conceals valuable information for both policy applications and empirical analyses of international trade.

Fairly recent data for about one hundred countries indicate that as the average level of schooling increases, educational inequality first increases and, after reaching a peak, starts declining in later phases of educational expansion. The turning point occurs when average schooling is about seven years. The observed empirical generalization, which seems quite robust, appears to have important implications for educational and distributional policies and for research on the linkage between education and income inequality.

This paper develops a two-country micro-theoretic model consistent with the Ricardian equivalence hypothesis. Specifically, tax increases used to retire government debt will not affect private spending or the current account balance. However, increases in government spending, regardless of the means of finance, can be expected to induce a current account deficit. An unconstrained vector autoregression shows some patterns in the recent U.S. data that appear to be inconsistent with the Ricardian equivalence hypothesis. Rigorous testing of the model, however, does not allow the authors to reject the independence of the record federal government budget and current account deficits.

Asymptotic Distributions of Impulse Response Functions and Forecast Error Variance Decompositions of Vector Autoregressive Models, The Review of Economics and Statistics, Vol. 72, No. 1 (Feb., 1990), pp. 116-125

Two implications of received theory are (1) mineral net prices rise at the riskless interest rate, and (2) in-ground value is equal to the current net price. Both propositions are false. A correct theory has been joined to mistaken premises. Mineral resources are inexhaustible. The economic problem is not the intertemporal allocation of a stock but coping with the cost of a flow of reserve accretions. Mineral scarcity and price are the uncertain fluctuating result of a tug-of-war between diminishing returns versus increasing knowledge. Hence minerals are risky assets. Development cost, finding cost, and user cost (the penalty for development/production today instead of tomorrow) are all substitutes. Hence change in any one is a proxy for change in any other. Development cost is observable, and has been stable in many countries for pro- longed periods. User cost was also stable in the USA. There is no sign of any pattern of gradual depletion and rising cost. A simple model of an individual reservoir explains observed relations of value and price. The rate of interest has both a positive and negative effect upon the rate of reservoir depletion. The net effect of a change is therefore weak. Expropriation of low-cost oil fields, had they been operated independently to maximize value, would have led to drastic increases in depletion rates. The fact of decrease proves collusive restriction of output to maintain prices.

A nonparametric frontier approach is used to calculate the overall, technical, pure technical, allocative, and scale efficiencies for a sample of 322 independent banks. The sample was drawn from the Federal Deposit Insurance Corporation tapes on the Reports of Conditions and Reports of Income (Call Reports) for the year 1986. The results indicate a low level of overall efficiency. The main source of inefficiency is technical in nature, rather than allocative. Separate efficiency frontiers are constructed to test the effect of branching. However, the distributions of efficiency measures for branching and nonbranching banks are not found to be different. Coauthors are Richard Grabowski, Carl Pasurka, and Nanda Rangan.

Many economic researchers have attempted to measure the effect of aggregate market or public policy variables on micro units by merging aggregate data with micro observations by industry, occupation, or geographical location, then using multiple regression or similar statistical models to measure the effect of the aggregate variable on the micro units. The methods are usually based upon the assumption of independent disturbances, which is typically not appropriate for data from populations with grouped structure. Incorrectly using ordinary least squares can lead to standard errors that are seriously biased downward. This note illustrates the danger of spurious regression from this kind of misspecification, using as an example a wage regression estimated on data for individual workers that includes in the specification aggregate regressors for characteristics of geographical states.

A multivariate time series model with time varying conditional variances and covariances, but constant conditional correlations is proposed. In a multivariate regression framework, the model is readily interpreted as an extension of the Seemingly Unrelated Regression (SUR) model allowing for heteroskedasticity. Parameterizing each of the conditional variances as a univariate Generalized Autoregressive Conditional Heteroskedastic (GARCH) process, the descriptive validity of the model is illustrated for a set of five nominal European U.S. dollar exchange rates following the inception of the European Monetary System (EMS). When compared to the pre- EMS free float period, the comovements between the currenciess are found to be significantly higher over the later period.