This paper demonstrates how corner solutions raise difficulties for the specification, estimation, and use of incomplete demand systems for welfare measurement with disaggregate consumption data, as is common in the outdoor recreation literature. A simple analytical model of consumer behavior is used to elucidate the potential biases for welfare measurement arising from modeling the demand for M goods as a function of M + N prices (N > 1) and income when individuals do not consume all goods in strictly positive quantities. Results from a Monte Carlo experiment suggest that these biases can be substantial for large-scale policy shocks when prices are highly correlated.

This paper analyses the world demand for fibers using the system-wide approach with three dimensions—product × space × time. We investigate to what extent differences in international consumption patterns of fibers can be explained by differences in incomes and prices faced by different consumers. A novel approach to cross-country consumption comparisons is employed to avoid the troublesome problem of what exchange rates to use when converting data into a common currency unit. We use data from the ten largest consuming countries to estimate demand systems and then examine how they perform in predicting consumption patterns in a large number of out-of-sample countries.

Using IRI Infoscan data pertaining to six types of spaghetti sauces and employing an extension of the demand systems framework developed by Duffy, estimates are obtained of own-price, cross-price, and total expenditure elasticities as well as own- and crossproduct advertising elasticities. We augment the Duffy model through the use of a polynomial inverse lag mechanism to deal with the carryover effects of advertising. We also account for the impacts of features in newspaper fliers, in-store displays, and coupons. Advertising efforts by industry leaders in spaghetti sauce produce positive own-advertising elasticities (ranging from .000058 to .0168) and negative cross-advertising elasticities (ranging from –.000003 to –.0094). Own-price elasticities are in the elastic range, and nearly all compensated cross-price effects are positive, indicative of Hicksian substitutes.

A previous study showed that imposing economic restrictions improves the forecasting ability of food demand systems, thus warranting their use even when they are rejected in-sample. This article evaluates whether this result is due to economic restrictions enhancing degrees of freedom or containing nonsample information. Results indicate that restrictions improve forecasting ability even when they are not derived from economic theory, but theoretical restrictions forecast best.