Knowledge that Transforms

We consider expectations of the form E[logy|x] = Σj=1d αj log xj as a good starting point for a more general analysis. We show why this naturally leads to the following flexible functional form: E[y|x] = f(Σj=1dhj(xj)), where f(ċ) and the hj(ċ)'s are estimated by cubic splines. The main objective of this paper is to provide a straightforward method to estimate E[y|x]. We demonstrate the usefulness of this approach by estimating gasoline demand from the 1994 RTECS data set, and in doing so, uncover interesting relationships of income and age to expected gasoline use.

This paper summarizes the dramatic changes in relative educational attainment by men and women over the past three decades. Stock measures of education among the entire adult population show rising attainment levels for both men and women, with men enjoying an advantage in schooling levels throughout this interval. Cohort-specific analysis reveals that these stock measures mask two interesting patterns: (a) gender difference at the cohort level had vanished by the early 1950 birth cohort and has been reversed in sign ever since; (b) for several cohorts, attainment rates were flat for women and flat and falling for men. This last is puzzling in the face of the large college premia that these cohorts observed when making their schooling choices. We present a simple human capital model showing how the anticipated dispersion of future wages should affect educational investment, and find that a model which includes measures of future earnings dispersion fits the data for relative schooling patterns quite well.

This paper calculates elasticities of demand for race groups in South Africa, government-revenue-maximizing tax rates, and excess burdens associated with taxes. A change in tax policy can be the political engine of income redistribution with appropriate taxes and subsidies on different commodities. This paper compares both semiparametric and parametric estimators with the censored least absolute deviation and censored maximum likelihood in calculating demand equations and elasticities. It is found that cigarettes and milk are the two commodities that generate the most government revenues from whites per unit of government revenues from blacks.

This paper examines data from U.S. federal tax returns to shed light on whether the timing of death is responsive to its tax consequences. We investigate the temporal pattern of deaths around the time of changes in the estate-tax system periods when living longer, or dying sooner, could significantly affect estate-tax liability. We find some evidence that there is a small death elasticity, although we cannot rule out that what we have uncovered is ex post doctoring of the reported date of death.

Wolf demonstrates that trade within the United States appears substantially impeded by state borders. We revisit this finding with improved data. We show that much intranational home bias can be explained by wholesaling activity. Shipments by wholesalers are much more localized within states than shipments from manufacturing establishments. Controlling for relative prices and the use of actual, rather than imputed, shipment distances also reduces home bias estimates.

Inspired by the recent literature on aggregation theory, this paper introduces HITS, a semiparametric model of consumer demand that allows for diversity in tastes. The strong variation of budget shares observed across income groups has two possible origins: the individual income effect, and taste differences between poor and rich households. Consumer surveys reporting repeated cross sections do not permit the direct measurement of these two effects. In HITS, linear heterogeneity allows the GMM estimation of structural coefficients on an aggregate series. The joint density of spending and tastes is then recovered from cross sections by a nonparametric procedure involving a deconvolution. We estimate the model on British data (1968–1998) and report that taste heterogeneity explains a large fraction of the variation of budget shares with income.

Empirical research over the last decade has uncovered predictive relationships between the slope of the yield curve and subsequent real activity and inflation. Some of these relationships are highly significant, but their theoretical motivations suggest that they may not be stable over time. We use recent econometric techniques for break testing to examine whether the empirical relationships are in fact stable. We consider continuous models, which predict either economic growth or inflation, and binary models, which predict either recessions or inflationary pressure. In each case, we draw on evidence from Germany and the United States. Models that predict real activity are somewhat more stable than those that predict inflation, and binary models are more stable than continuous models. The model that predicts recessions is stable over our full sample period in both Germany and the United States.

We investigate whether fluctuations in U.S. inflation rates are better described by infrequently occurring large shocks or by frequently occurring small shocks. We estimate a model that encompasses the two hypotheses within the framework of non-Gaussian state-space models. Our results indicate support for infrequently occurring large shocks, but this weakens somewhat once we allow for outliers and conditional heteroskedasticity. It appears that, for the purpose of forecasting monthly U.S. inflation rates, recognizing the distinction between frequent small shocks and infrequent large shocks does not matter much once outliers and conditional heteroskedasticity are allowed for.

We extend the local-to-zero analysis of models with weak instruments to models with estimated instruments and regressors and with higher-order dependence between instruments and disturbances. This framework is applicable to linear models with expectation variables that are estimated nonparametrically, such as the risk-return tradeoff in finance and the effect of inflation uncertainty on real economic activity. Our simulation evidence suggests that Lagrange multiplier confidence intervals have better coverage in these models. We apply these methods to excess returns on the S&P 500 index, yen-dollar spot returns, and excess holding yields between 6-month and 3-month Treasury bills.

No significant convergence between Canadian and American income per capita occurred during the first ninety years of the twentieth century. This lack of convergence does not appear to have been due to technological dependence, input price distortions, or diseconomies of scale within the Canadian manufacturing sector. The evidence presented in this paper is based on total factor productivity measurement, statistical testing, and counterfactual experimentation using data from national statistical agencies and firm-level sources.