Knowledge that Transforms

This paper reconsiders the empirical evidence connecting inflation to its higher-order moments. In particular, we examine the statistical properties of the observed positive correlation between the sample mean and the sample cross-sectional skewness of price changes. This correlation has attracted substantial attention over the years and has recently been the focal point of a debate among macroeconomists. We show that the sample mean-skewness correlation suffers from a small-sample bias that accounts for the entirety of the observed correlation. In other words, we establish that one of the stylized facts in the literature on aggregate price behavior need not be a fact at all.

Nonlinear modeling has become popular in applied macroeconomics. Successful attempts include Beaudry and Koop's CDR (current depth of the recession) model of real GNP, and various STAR (smooth transition autoregression) models of industrial production. However, these models have not been directly compared. We compare CDR and STAR models of U.S. real GNP and industrial production. We find (i) within sample, the CDR model fits slightly better than the STAR model; (ii) out of sample, the CDR model forecasts better than the STAR model; and (iii) the CDR model generates very different dynamics than the STAR model.

This paper reexamines the evidence relating core inflation to cross-sectional inflation asymmetry using statistical measures that are robust to the criticism of Bryan and Cecchetti. The results here suggest that there does exist significant positive correlation between core inflation and cross-sectional inflation asymmetry, but only at the monthly frequency. Furthermore, a sampling problem is highlighted which underscores the importance of careful Monte Carlo analysis when exact small-sample distributions are unknown.

This study investigates the dispersion of price levels within highly disaggregated markets by examining plant-level product records from the U.S. Census of Manufactures. The paper estimates the effects of inflation on price dispersion through cross-sectional variation in the drift rate of average input costs within a market, arguing that, in several models that relate inflation to price dispersion, the effects of cost increases on dispersion is similar to the effects of general inflation. We also disentangle the effects of aggregate and idiosyncratic shocks on price dispersion. In general, we find that the higher the drift rate of input costs of a given commodity, the larger the amount of price dispersion. The standard deviation of idiosyncratic shocks also is positively correlated with the degree of price dispersion.

Traditional analyses of the income tax greatly underestimate deadweight losses by ignoring its effect on forms of compensation and patterns of consumption. The full deadweight loss is easily calculated using the compensated elasticity of taxable income to changes in tax rates because leisure, excludable income, and deductible consumption are a Hicksian composite good. Microeconomic estimates imply a deadweight loss of as much as 30% of revenue or more than ten times Harberger's classic 1964 estimate. The relative deadweight loss caused by increasing existing tax rates is substantially greater and may exceed $2 per $1 of revenue.

We use household-level data covering a ten-year period (1984 to 1993) to investigate the likely determinants of the Italian recession of 1993, the first year after WWII when private consumption fell. Consumption fell most for working-age households and for the self-employed. Our evidence is consistent with the response to permanent negative shocks due to the major pension reform of 1992 and the introduction of stricter tax-compliance measures for the self-employed. This is still true when we control for the role played by job losses and the collapse of the retail sector that characterized the early 1990s.

© 1999 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology

Previous property tax capitalization studies assume that families itemize, that they save in taxable assets, and that real interest income is taxed. However, many families do not itemize, many families invest in tax-deferred assets, and nominal interest income is taxed. As a consequence, prior studies likely misspecify the property tax capitalization equation for roughly ninety percent of their samples. Taking federal tax provisions into account increases the precision of our estimated capitalization rate. In addition, our results suggest that biases in prior studies likely contribute to the variety of capitalization estimates in the literature.

The ability of yield spreads to predict changes in long-term interest rates implied by the expectations hypothesis is usually rejected. It is suggested that this rejection is often caused by high persistence in the spread when standard inference is employed. Instead, the asymptotically valid method of Cavanagh et al. (1995) is applied to monthly U.S. data from 1952:1-1991:2. The persistence of the spreads seems to have varied over time, and in subsample analysis, the expectations hypothesis cannot be rejected at the long end of the maturity spectrum.

This paper examines the sources of productivity growth for the, U.S. telecommunications industry from 1935 to 1987. These years encompass both the pre-and post-AT&T divestiture periods. We formulate a structural model that accounts for both changes in the cost and the demand side of the industry. We measure the contributions of aggregate demand, information intensity of the economy, price-cost margins, relative factor prices, direct and indirect effects of technological progress, and R&D investment on total-factor productivity (TFP) growth rate. We show that TFP growth rate as conventionally measured is a seriously biased measure of rate of technical change in this industry.