Knowledge that Transforms

It is shown how a comparison of price levels across a group of countries can be made by chaining bilateral price indexes across a spanning tree. It is argued that we should use the spanning tree whose resulting multilateral price indexes are least sensitive to the choice of bilateral formula. This minimum-spanning tree can be easily computed using Kruskal's algorithm. Results obtained by chaining Fisher indexes across a minimum-spanning tree are compared with the Penn World Table.

A Monte Carlo analysis of the coverage accuracy and average length of alternative bootstrap confidence intervals for impulse-response estimators shows that the accuracy of equal-tailed and symmetric percentile-t intervals can be poor and erratic in small samples (both in models with large roots and in models without roots near the unit circle). In contrast, some percentile bootstrap intervals may be both shorter and more accurate. The accuracy of percentile-t intervals improves with sample size, but the sample size required for reliable inference can be very large. Moreover, for such large sample sizes, virtually all bootstrap intervals tend to have excellent coverage accuracy.

In this paper we analyze the estimation of coefficients in regression models under moment restrictions in which the moment restrictions are derived from auxiliary data. The moment restrictions yield weights for each observation that can subsequently be used in weighted regression analysis. We discuss the interpretation of these weights under two assumptions: that the target population (from which the moments are constructed) and the sampled population (from which the sample is drawn) are the same, and that these populations differ. We present an application based on omitted ability bias in estimation of wage regressions. The National Longitudinal Survey Young Men's Cohort (NLS)—in addition to containing information for each observation on wages, education, and experience—records data on two test scores that may be considered proxies for ability. The NLS is a small dataset, however, with a high attrition rate. We investigate how to mitigate these problems in the NLS by forming moments from the joint distribution of education, experience, and log wages in the 1% sample of the 1980 U.S. Census and using these moments to construct weights for weighted regression analysis of the NLS. We analyze the impacts of our weighted regression techniques on the estimated coefficients and standard errors of returns to education and experience in the NLS controlling for ability, with and without the assumption that the NLS and the Census samples are random samples from the same population.

Random utility models (RUMs) are used in the literature to model consumer choices from among a discrete set of alternatives, and they typically impose a constant marginal utility of income on individual preferences. This assumption is driven partially by the difficulty of constructing welfare estimates in models with nonlinear income effects. Recently, McFadden (1995) developed an algorithm for computing these welfare impacts using a Monte Carlo Markov chain simulator for generalized extreme-value variates. This paper investigates the empirical consequences of nonlinear RUMs in the case of sportfishing modal choice, while refining and contrasting the available methods for welfare estimation.

Using data that permit a distinction between flows of workers (directly measured) and job creation and destruction (again, directly measured), we develop employment and job flow statistics for a representative sample of French establishments from 1987 to 1990. Annual job creation can be characterized as hiring three persons and separating two for each job created in a given year. Annual job destruction can be characterized as hiring one person and separating two for each job destroyed in a given year. When an establishment is changing employment, the adjustment is made primarily by reducing entry and not by changing the separation rates. There is considerable simultaneous hiring and separation, even controlling for skill group. Two-thirds of all hiring is on short-term contracts, and more than half of all separations are due to the end of these short-term contracts.

A simple construction that will be referred to as an error-duration model is shown to generate fractional integration and long memory. An error-duration representation also exists for many familiar ARMA models, making error duration an alternative to autoregression for explaining dynamic persistence in economic variables. The results lead to a straightforward procedure for simulating fractional integration and establish a connection between fractional integration and common notions of structural change. Two examples show how the error-duration model could account for fractional integration in aggregate employment and in asset price volatility.

We propose a procedure for measuring the effect of systematic discrimination in the income tax. Different socioeconomic groups are assumed to face different tax schedules. We show that a welfare loss is caused by the group specificity of schedules, the dollar value of which is our measure of discrimination. Defining vertical equity as the dollar value of the tax system's welfare superiority over an equal yield flat tax, discrimination equates to a loss in vertical equity. The Australian income tax is found to discriminate against wage and salary earners, causing a roughly 1% loss of social welfare in 1984.

We test the hypothesis associated with a standard assumption in the theoretical literature on learning: that economic agents interpret information identically. We use a data set based on a survey of Israeli business executives forecasting future inflation. One of the main advantages of using this data is that a major change in the inflation regime in 1985 can be treated as a natural experiment in new beliefs formation. We develop a methodology for testing this hypothesis and find evidence that is inconsistent with the identical-interpretation hypothesis, but is consistent with the proposed alternative.

In an oligopolistic product market, shareholders strategically use information on rival firms'performances when designing management-incentive contracts. When shareholders use industry performance information through relative performances evaluation (RPE), they evaluate their manager's effort more easily, but hinder collusive behavior in the product market. However, when compensation is positively linked to the industry performance through strategic group performance evaluation (SGPE), the credibility of a manager's commitment to product market collusion increases, and the sustainability of a collusive outcome increases. I test how industry performance affects management-incentive compensation using the data from 796 Japanese firms during the period 1968 to 1992. The results show that management compensation is positively linked to industry profit, suggesting the use of SGPE in management-incentive compensation. Cross-sectional analysis shows that the positive effect of industry profit on management compensation is higher in competitive industries than in concentrated industries. The positive effect is greater in slow-growing industries than in fast-growing industries. Empirical tests incorporating the risk component method show the same results. These results are consistent with the argument that, in a growing market or in a concentrated market, the value of SGPE diminishes as the value of commitment to collusion diminishes.

This paper contributes to the empirical literature that tests the implications of risk sharing across regions or countries. Rather than test the null of full insurance, the estimation method produces a parameter that measures the fraction of the annuity value of individual income that individuals of different regions (or nations) pool. Comparing the provinces of Canada, the states of the United States, and the G-7 countries, I find similar degrees of risk sharing within regions of Canada and the U.S. that exceed the risk sharing that occurs across countries. Specifically, more than two-thirds of the fitted annual variation in regional consumption is common to all regions compared to less than one-third in the case of G-7 countries.