Knowledge that Transforms

This paper advances the literature on multicategory demand models by simultaneously handling more than one purchase decision of the household. We propose a two-stage bivariate logit model of incidence and quantity outcomes in multiple categories. Our results show that cross-category promotional spillovers are asymmetric between the two product categories of bacon and eggs. The total retail profit responds more to bacon price than to egg price. Promoting bacon is found to have a bigger impact on egg profit than the impact of egg promotion on bacon profit. We decompose (1) the total retail profits, as well as (2) the cross-category profit impact of a price promotion, into its two components, and find that (1) 23% (67%) of the total retail profit impact of a promotion on bacon (eggs) arises on account of quantity effects, and (2) 40% (33%) of the increase in egg (bacon) profit from promoting bacon (eggs) is on account of quantity effects.

We study two main questions in this paper: (1) How do spillovers of knowledge created by manufacturers' investments in process innovation affect channel structure and effort investment incentives? (2) What are the interactions between organizational incentives to form joint ventures and strategic alliances with competitors, and coordinate decisions vertically with downstream channel members? We focus on situations where spillovers are involuntary, firms' innovative activities are nonoverlapping, and firms benefit directly from the results of competitors' innovations. Under these conditions, we find that spillovers in process knowledge increase the likelihood of observing decentralized channel structures. Surprisingly, decentralized manufacturers invest more in process innovation than perfectly coordinated manufacturers do when spillovers are large. Moreover, in industries where large spillovers exist, horizontal cooperation among manufacturers induces higher levels of process innovation investments than channel coordination does. From a public policy perspective, however, the desirability of such cooperative arrangements among competitors depends on channel structure: joint ventures among decentralized manufacturers are more likely to meet the regulators' criteria of raising effort investments than cooperation among integrated manufacturers would be. Investment incentives are best provided when firms share their process knowledge and are buffered from subsequent price competition by independent retailers.

We propose a structural approach to measuring brand and subbrand value using observational data. Brand value is defined as the difference in equilibrium profit between the brand in question and its counterfactual unbranded equivalent on search attributes. Our model allows us to make this computation rigorously, taking into account competitors' and retailers' reactions in the real and counterfactual situations. We illustrate our method using quarterly city-level data on ready-to-eat breakfast cereals, and compare our brand value estimates with those obtained from previously used reduced-form methods. A key advantage of our methodology is that it provides estimates of the value of brands to firms—manufacturers and retailers—taking into account the brand's value to consumers as well as its impact on firm decisions.

New information technologies increasingly make it possible for service providers to adaptively personalize their service, fine-tuning the service over time for each individual customer, based on observation of that customer's behavior. We propose an “Adaptive Personalization System” and illustrate its implementation for digital audio players, a product category with rapidly expanding sales. The proposed system automatically downloads personalized playlists of MP3 songs into a consumer's mobile digital audio device and requires little proactive user effort (i.e., no explicit indication of preferences or ratings for songs). The system works in real time and is scalable to the massive data typically encountered in personalization applications. A simulation study shows the Adaptive Personalization System to outperform benchmark approaches. We implemented the Adaptive Personalization System on Palm PDAs and tested its performance with digital audio users. For actual users, the Adaptive Personalization System provides substantial improvements over benchmark approaches both in terms of the number of songs listened to and listening duration.

We analyze a large-scale field test conducted with a mail-order catalog firm to investigate how customers react to premium prices for larger sizes of women's apparel. We find that customers who demand large sizes react unfavorably to paying a higher price than customers for small sizes. Further investigation suggests that these consumers perceive that the price premium is unfair. Overall, premium pricing led to a 6% to 8% decrease in gross profits.

The Bass model has been a standard for analyzing and predicting the market penetration of new products. We demonstrate the insights to be gained and predictive performance of functional data analysis (FDA), a new class of nonparametric techniques that has shown impressive results within the statistics community, on the market penetration of 760 categories drawn from 21 products and 70 countries. We propose a new model called Functional Regression and compare its performance to several models, including the Classic Bass model, Estimated Means, Last Observation Projection, a Meta-Bass model, and an Augmented Meta-Bass model for predicting eight aspects of market penetration. Results (a) validate the logic of FDA in integrating information across categories, (b) show that Augmented Functional Regression is superior to the above models, and (c) product-specific effects are more important than country-specific effects when predicting penetration of an evolving new product.

In certain product categories, large discount retailers are known to offer shallower assortments than traditional retailers. In this paper, we investigate the competitive incentives for such assortment decisions and the implications for manufacturers' distribution strategies. Our results show that if one retailer has the channel power to determine its assortment first, then it can strategically reduce its assortment by carrying only the popular variety while simultaneously inducing the rival retailer to carry both the specialty and popular varieties. The rival retailer then bears higher assortment costs, which leads to relaxed price competition for the commonly carried popular variety. We also show that when the manufacturer has relative channel power, it chooses alternatively to distribute both product varieties through both retailers. Our analysis suggests, therefore, that when a retailer becomes dominant in the distribution channel, it facilitates retail segmentation into discount shops, carrying limited product lines, and specialty shops carrying wider assortments. We also illustrate how retailer power leading to strategic assortment reduction can lead to lower consumer surplus.

Many brands today introduce limited edition (LE) products as part of their product line. However, little is known about the conditions under which a brand should introduce an LE product or the competitive implications of doing so. We investigate this issue using a game theoretic model of a market where two brands compete for consumers who desire exclusivity. Our analysis shows that adding an LE product has a positive direct effect on brand profits through the increased willingness of consumers to pay for such a product, but also has a negative strategic effect by increasing price competition between brands. These effects result in different conclusions depending on the nature of brand differentiation. When brands differ in quality, we show that only the high-quality brand may gain in comparison to a scenario where there are no LE products. Although a low-quality brand may offer an LE product as a defensive strategy, its profits are lower than would be in a world without LE products because of the negative strategic effect. When we consider brands that are differentiated on a horizontal attribute such as taste, we find that the negative strategic effects cause lower equilibrium profits if both brands introduce LE products. Yet brands cannot avoid introducing LE products because they face a prisoners' dilemma.

We use a unique data set to estimate the price sensitivities of households in online and offline shopping channels when the same households shop across channels. We observe households that shop interchangeably at the online and the offline stores in the same grocery chain and investigate their purchase behavior in specific product categories. Although nearly 90% of households in our sample shop both at online and offline stores, we find that, across 12 vastly different product categories, these households exhibit lower price sensitivities when they shop online than when they shop offline. Our analysis accounts for observed and unobserved household heterogeneity as well as price endogeneity. The results hold for large basket-share categories and small basket-share categories, for consumer packaged goods and nonpackaged goods, for categories that are more likely to be purchased online because of their bulkiness or heaviness, and for categories that are more likely to be purchased offline because of their “sensory” nature. Households' price sensitivities are also closely related to demographics and inversely related to how far the households are located from the offline stores. Reasons for the lower price sensitivities in the online medium are discussed.

Reference groups influence product and brand evaluations, especially when the product is a publicly consumed luxury good. Marketers of such luxury goods need to carefully balance two important social forces: (1) the desire of leaders to distinguish themselves from followers and (2) the countervailing desire of followers to assimilate with leaders. In this paper, we examine the theoretical implications of these social forces for firm prices, product design, and target consumer selection. We show that the presence of reference group effects can motivate firms to add costly features, which provide limited or no functional benefit to consumers. Furthermore, reference group effects can induce product proliferation on one hand and motivate firms to offer limited editions on the other hand. We find that offering a limited edition can increase sales and profits. In some cases, reference group effects can even lead to a buying frenzy.