Spot asset trading is studied where the only external source of value is an independent draw from a common information dividend distribution at the end of each of fifteen trading periods. Fourteen of twenty-two experiments exhibit price bubbles. This tendency to bubble decreases with trader experience. The regression of changes in mean price on lagged excess bids (bids minus offers in the previous period) supports the hypothesis that the intercept is minus the one-period expected dividend value, and the slope is positive, where excess bids measures excess demand attributable to homegrown capital gains expectations.

G. H. Hillier, T. W. Kinal, V. K. Srivastava, On the Moments of Ordinary Least Squares and Instrumental Variables Estimators in a General Structural Equation, Econometrica, Vol. 52, No. 1 (Jan., 1984), pp. 185-202

We are interested in estimating the average effect of a binary treatment on a scalar outcome. If assignment to the treatment is exogenous or unconfounded, that is, independent of the potential outcomes given covariates, biases associated with simple treatment-control average comparisons can be removed by adjusting for differences in the covariates. Rosenbaum and Rubin (1983) show that adjusting solely for differences between treated and control units in the propensity score removes all biases associated with differences in covariates. Although adjusting for differences in the propensity score removes all the bias, this can come at the expense of efficiency, as shown by Hahn (1998), Heckman, Ichimura, and Todd (1998), and Robins, Mark, and Newey (1992). We show that weighting by the inverse of a nonparametric estimate of the propensity score, rather than the true propensity score, leads to an efficient estimate of the average treatment effect. We provide intuition for this result by showing that this estimator can be interpreted as an empirical likelihood estimator that efficiently incorporates the information about the propensity score.

We study the implications of optimal insurance provision for long‐run welfare and inequality in economies with persistent private information. A principal insures an agent whose private type follows an ergodic, finite‐state Markov chain. The optimal contract always induces immiseration : the agent's consumption and utility decrease without bound. Under positive serial correlation, it also backloads high‐powered incentives : the sensitivity of the agent's utility with respect to his reports increases without bound. These results extend—and help elucidate the limits of—the hallmark immiseration results for economies with i.i.d. private information. Numerically, we find that persistence yields faster immiseration, higher inequality, and novel short‐run distortions. Our analysis uses recursive methods for contracting with persistent types and allows for binding global incentive constraints.

Does individual behavior in a laboratory setting provide a reliable indicator of behavior in a naturally occurring setting? We consider this general methodological question in the context of eliciting risk attitudes. The controls that are typically employed in laboratory settings, such as the use of abstract lotteries, could lead subjects to employ behavioral rules that differ from the ones they employ in the field. Because it is field behavior that we are interested in understanding, those controls might be a confound in themselves if they result in differences in behavior. We find that the use of artificial monetary prizes provides a reliable measure of risk attitudes when the natural counterpart outcome has minimal uncertainty, but that it can provide an unreliable measure when the natural counterpart outcome has background risk. Behavior tended to be moderately risk averse when artificial monetary prizes were used or when there was minimal uncertainty in the natural nonmonetary outcome, but subjects drawn from the same population were much more risk averse when their attitudes were elicited using the natural nonmonetary outcome that had some background risk. These results are consistent with conventional expected utility theory for the effects of background risk on attitudes to risk.

This paper applies revealed preference theory to the nonparametric statistical analysis of consumer demand. Knowledge of expansion paths is shown to improve the power of nonparametric tests of revealed preference. The tightest bounds on indifference surfaces and welfare measures are derived using an algorithm for which revealed preference conditions are shown to guarantee convergence. Nonparametric Engel curves are used to estimate expansion paths and provide a stochastic structure within which to examine the consistency of household level data and revealed preference theory. An application is made to a long time series of repeated cross-sections from the Family Expenditure Survey for Britain. The consistency of these data with revealed preference theory is examined. For periods of consistency with revealed preference, tight bounds are placed on true cost of living indices.

Consider a researcher estimating the parameters of a regression function based on data for all 50 states in the United States or on data for all visits to a website. What is the interpretation of the estimated parameters and the standard errors? In practice, researchers typically assume that the sample is randomly drawn from a large population of interest and report standard errors that are designed to capture sampling variation. This is common even in applications where it is difficult to articulate what that population of interest is, and how it differs from the sample. In this article, we explore an alternative approach to inference, which is partly design‐based. In a design‐based setting, the values of some of the regressors can be manipulated, perhaps through a policy intervention. Design‐based uncertainty emanates from lack of knowledge about the values that the regression outcome would have taken under alternative interventions. We derive standard errors that account for design‐based uncertainty instead of, or in addition to, sampling‐based uncertainty. We show that our standard errors in general are smaller than the usual infinite‐population sampling‐based standard errors and provide conditions under which they coincide.