We propose an approach to extend the standard framework of dynamic games to settings with multi-product firms. Our approach applies to industries with a large number of products offered by a small number of firms.

Temporary price reductions (sales) are common for many goods and naturally result in large increases in the quantity sold. Demand estimation based on temporary price reductions may mismeasure the long-run responsiveness to prices. In this paper we quantify the extent of the problem and assess its economic implications. We structurally estimate a dynamic model of consumer choice using two years of scanner data on the purchasing behavior of a panel of households. The results suggest that static demand estimates, which neglect dynamics, (i) overestimate own-price elasticities by 30 percent, (ii) underestimate cross-price elasticities by up to a factor of 5, and (iii) overestimate the substitution to the no-purchase or outside option by over 200 percent. This suggests that policy analysis based on static elasticity estimates will underestimate price–cost margins and underpredict the effects of mergers.

Temporary price reductions (sales) are common for many goods and naturally result in a large increase in the quantity sold. We explore whether the data support the hypothesis that these increases are, at least partly, due to demand anticipation: at low prices, consumers store for future consumption. This effect, if present, has broad economic implications. We test the predictions of an inventory model using scanner data with two years of household purchases. The results are consistent with an inventory model and suggest that static demand estimates may overestimate price sensitivity.

This article provides information on research regarding the effect of changing library acquisition strategies on academic journal pricing. A basic model of the demand of libraries for economic journals was estimated. Using a basic model structure, the acquisition of one publisher's journals by another may in theory lead to either higher or lower prices. The research extends the analysis in several ways. It estimates demand functions that allow for more library heterogeneity. It also considers the likely unilateral price effects of various acquisitions and plans to stimulate equilibrium price effects of various bundling strategies.

Storable products allow consumers to time their purchases to exploit price fluctuations. It has been documented that during promotions consumers buy more. The additional purchases are potentially intended not only for current use, but to be stockpiled for future consumption. This paper discusses the predictions of a consumer inventory model and reviews the available evidence. We then discuss the implications for demand estimation and present estimates of the economic magnitude of the dynamic effect of storability.

This paper examines the construction of a price index based on an estimated-demand system. In principle the method examined can produce a price index that takes account of the introduction of new products and quality changes in existing products. However, I isolate two key assumptions that have to be made in order to interpret the demand estimates into welfare measures. Using estimates of a brand-level demand system for ready-to-eat cereal, I demonstrate the empirical importance of the assumptions. For the data I use, depending on the interpretation of the demand estimates, a price index can range between a 35% increase over the five years examined to a 2.4% decrease.

One of the puzzles of store-level scanner data is the lack of a dip in quantity sold in the weeks following a promotion. Such a dip is predicted by a consumer inventory model. During a promotion consumers buy more, not only for current consumption, but stockpile for future consumption. The predictions of such a model have been confirmed by household-level data yet seem harder to find in aggregate brand- or category-level data. We re-examine this puzzle and reach two conclusions. First, the effects at the household-level are present, but are much smaller than previously found. Our estimates are different because we control for household heterogeneity in a more general way than most previous work. This suggests that since the effects are small they might be harder to spot in aggregate data. Second, we show that the dip is present in the aggregate data, once we control for additional promotional activity, like feature and display. The latter has an opposing dynamic effect that masks the existence of the post-promotion dip.

This article analyzes generalized method of moments estimation when the sample is not a random draw from the population of interest. Auxiliary information, in the form of moments from the population of interest, is exploited to compute weights that are proportional to the inverse probability of selection. The essential idea is to construct weights for each observation in the primary data such that the moments of the weighted data are set equal to the additional moments. The estimator is applied to the Dutch Transportation Panel, in which refreshment draws were taken from the population of interest to deal with heavy attrition of the original panel. It is shown how these additional samples can be used to adjust for sample selection.

Information-theoretic alternatives to general method of moments (GMM) use over-identifying moments to estimate the data-generating distribution jointly with the parameters of interest. This paper demonstrates how these estimates can be interpreted when the sample is not a random draw from the population of interest. I make explicit the selection probability implied by the empirical likelihood and exponential tilting estimators, two commonly used estimators in this class. In addition, I propose an alternative estimator that corresponds to a logisitic selection model. The small sample properties of the estimators are demonstrated with a Monte Carlo experiment.

We explore the relationship between shelf prices and manufacturers' coupons for 25 ready-to-eat breakfast cereals. We find that shelf prices are lower during periods when coupons are available. This result is inconsistent with static monopoly price discrimination under a broad range of assumptions. We present evidence that is inconsistent with both dynamic theories of price discrimination and explanations of couponing based on the vertical relationship between manufacturers and retailers. We find support for models of price discrimination in oligopoly settings as well as suggestions that firmwide incentives may induce managers to use coupons and price cuts simultaneously. Finally, lagged coupons have a positive effect on current sales, suggesting that coupons are used to induce repurchase.

The ready-to-eat cereal industry is characterized by high concentration, high price-cost margins, large advertising-to-sales ratios, and numerous introductions of new products. Previous researchers have concluded that the ready-to-eat cereal industry is a classic example of an industry with nearly collusive pricing behavior and intense nonprice competition. This paper empirically examines this conclusion. In particular, I estimate price-cost margins, but more importantly I am able empirically to separate these margins into three sources: (i) that which is due to product differentiation; (ii) that which is due to multi-product firm pricing; and (iii) that due to potential price collusion. The results suggest that given the demand for different brands of cereal, the first two effects explain most of the observed price-cost margins. I conclude that prices in the industry are consistent with noncollusive pricing behavior, despite the high price-cost margins. Leading firms are able to maintain a portfolio of differentiated products and influence the perceived product quality. It is these two factors that lead to high price-cost margins.

Estimation of demand is at the heart of many recent studies that examine questions of market power, mergers, innovation, and valuation of new brands in differentiated-products markets. This paper focuses on one of the main methods for estimating demand for differentiated products: random-coefficients logit models. The paper carefully discusses the latest innovations in these methods with the hope of increasing the understanding, and therefore the trust among researchers who have never used them, and reducing the difficulty of their use, thereby aiding in realizing their full potential.

Traditional merger analysis is difficult to implement when evaluating mergers in industries with differentiated products. I discuss an alternative, which consists of demand estimation and the use of a model of postmerger conduct to simulate the competitive effects of a merger. I estimate a brand-level demand system for ready-to-eat cereal using supermarket scanner data and use the estimates to (1) recover marginal costs, (2) simulate postmerger price equilibria, and (3) compute welfare effects, under a variety of assumptions. The methodology is applied to five mergers, two of which occurred and for which I compare predicted to actual outcomes.

A central part of the “new empirical industrial organization” has been the study of market power in homogenous-product industries. This paper discusses extension of these methods to differentiated-products industries. The requirements for identification of conjectural variation parameters are shown to be hard to satisfy in practice. An alternative menu-approach is discussed.

We compare outcomes obtained by sellers who listed their home on a newly developed For-Sale-By-Owner (FSBO) web site versus those who used an agent and the Multiple Listing Service (MLS). We do not fi…nd support for the hypothesis that listing on the MLS helps sellers obtain a signi…cantly higher sale price. Listing on the MLS shortens the time it takes to sell a house. The diffusion of the new FSBO platform was quick, with the market share stabilizing after 2 years, suggesting it managed to gain a critical mass necessary to compete with the MLS. However, the lower effectiveness of FSBO (in terms of time to sell and probability of a sale) suggests that the increasing returns to network size are not fully exploited at its current size. We discuss the welfare implications of our fi…ndings.

For many products the average price paid by consumers falls during periods of high demand. We use information from a large supermarket chain to decompose the decrease in the average price into a substitution effect, due to an increase in the share of cheaper products, and a price reduction effect. We find that for almost all the products we study the substitution effect explains a large part of the decrease. We estimate demand for these products and show the price declines are consistent with a change in demand elasticity and the relative demand for different brands. Our findings suggest, that for the data we examine, loss-leader models of retail competition are not the main explanation for price declines.