Knowledge that Transforms

Angus Deaton discusses the ambiguity that arises from using different definitions and data sources for individual income or consumption levels in world poverty measurement. Should one rely on the direct information on individual consumption or income provided by national representative household surveys, or should consumption and income figures be scaled up or down so that means coincide with National Accounts (NA) data? It is generally the case that consumption expenditure per capita estimated in the NA is higher than the mean expenditure per capita obtained in surveys: thus scaling up leads to lower poverty estimates than when surveys are used. It is also the case that the difference between the two estimates tends to widen over time, so that trends are not more reliable than poverty estimates at one point of time. Deaton analyzes in detail the reasons for this divergence and concludes that NA-scaled survey data are in some sense faulty, whereas a pure consistency argument pleads in favor of using survey data at their face value.

Scholars working on the border of economics and psychology have documented many contexts in which individual decision-making is unreliable and might be improved by paternalistic interventions. Against this mounting body of negative evidence, economists' default belief in consumer sovereignty has been motivated primarily by theory rather than evidence. The goal of the present study is to see whether there is direct evidence supporting economists' faith in consumer sovereignty in a simple context. We address this question by presenting direct evidence that consumers' own purchases generate between 10% and 18% more value, per dollar spent, than items received as gifts.

We use a data set describing ownership of productive assets in the carpentry trade to evaluate several factors influencing the allocation of asset ownership between an employer and his employees. The findings suggest that the allocation involves a tradeoff between two incentive effects influencing how the employee uses the asset and what the employer decides it should be used for. In particular, the allocation of ownership hinges on whether an asset is easily lost or stolen, which favors employee ownership, and whether the employer's task assignment affects the asset's depreciation, which favors employer ownership. There is also evidence that more expensive assets and assets that are shared by more than one employee are more likely to be owned by the employer. The results suggest that a general theory of asset ownership should be able to take account of at least these effects.

This paper provides data on the output and factor payments of new goods for every four-digit industry in the U.S. manufacturing sector in the late 1970s and 1980s. For the entire manufacturing sector, the new goods' average skilled-labor intensity exceeds the old goods' by over 40%, and new goods can account for approximately 30% of the increase in the relative demand for skilled labor. Because new goods provide a direct measure of technology, this paper offers new evidence that technology has shifted demand in favor of skilled labor, consistent with the technology skill-complementarity hypothesis.

Why have some countries done so much better than others over the recent past? This paper sheds light on this issue by providing a decomposition of the change in the distribution of output per worker across countries over the period 1960–1998. We find that most of the change in shape of the world distribution of income can be accounted for by a very substantial increase in the social returns to capital accumulation. In contrast, we do not find significant effects coming through changes in the effect of initial conditions or through increases in the importance of education.

Relative prices are nonstationary and standard root-T inference is invalid for demand systems. But demand systems are nonlinear functions of relative prices, and standard methods for dealing with nonstationarity in linear models cannot be used. Demand system residuals are also frequently found to be highly persistent, further complicating estimation and inference. We propose a variant of the translog demand system, the NTLOG, and an associated estimator that can be applied in the presence of nonstationary prices with possibly nonstationary errors. The errors in the NTLOG can be interpreted as random utility parameters. The estimates have classical root-T limiting distributions. We also propose an explanation for the observed nonstationarity of aggregate demand errors, based on aggregation of consumers with heterogeneous preferences in a slowly changing population. Estimates using U.S. data are provided.

This note examines the implications of mean-reverting mea-surement error for two influential literatures based on longitudinal survey data: (1) the literature on real wage variation over the business cycle and (2) the literature on intertemporal substitution in labor supply. Accounting for mean-reverting measurement error suggests that real wages may be even more procyclical than indicated by recent longitudinal studies. We also find that the instrumental variables estimator commonly used in intertemporal substitution studies is inconsistent if changes in earnings and hours of work are measured with different degrees of mean reversion, but the magnitude of the resulting inconsistency appears to be small.

Recent research documents the importance of uncertainty in determining macroeconomic outcomes, but little is known about the transmission of uncertainty across such outcomes. This paper examines the response of uncertainty about inflation and output growth to shocks documenting statistically significant size and sign bias and spillover effects. Uncertainty about inflation is a determinant of output uncertainty, whereas higher growth volatility tends to raise inflation volatility. Both inflation and growth volatility respond asymmetrically to positive and negative shocks. Negative growth and inflation shocks lead to higher and more persistent uncertainty than shocks of equal magnitude but opposite sign.

Knowing the responsiveness of state spending to federal subsidies along different dimensions allows for the optimal design of joint federal-state programs. Welfare is an important case in point: states have the ability to choose both the extent of welfare eligibility and the intensity of benefits provided through the program. This paper estimates the sensitivity of state spending to separate federal subsidies for increasing benefits and for increasing recipients. Because the federal match rate schedule changed several times during the early years that I study, I am able to estimate elasticities in a way that is not biased by the endogenous relationship between income, spending, and federal contributions. I find that state behavior is quite sensitive to these federal subsidies (and much more sensitive than a simple OLS regression would imply). A 10% increase in the cost of benefits causes a 3.8% decrease in benefit amounts, whereas a 10% increase in the cost of recipients causes a 2.8% decrease in the number of recipients. Cross price elasticities are positive, implying a substitutability of extensive for intensive generosity and making an analysis of total spending without such a decomposition misleading. States appear sensitive to their neighbors' benefit levels, and may also use nonincome recipiency requirements to adjust to changes in prices. These results suggest that the federal government has untapped policy instruments at its disposal to affect the nature of welfare spending.

The paper outlines a methodology for analyzing daily stock returns that relinquishes the assumption of global stationarity. Giving up this common working hypothesis reflects our belief that fundamental features of the financial markets are continuously and significantly changing. Our approach approximates the nonstationary data locally by stationary models. The methodology is applied to the S&P 500 series of returns covering a period of over seventy years of market activity. We find most of the dynamics of this time series to be concentrated in shifts of the unconditional variance. The forecasts based on our nonstationary unconditional modeling were found to be superior to those obtained in a stationary long-memory framework and to those based on a stationary Garch(1, 1) data-generating process.