A ‘‘building code’ ’ for preference measurement is needed in a world in which many expressions of preference are constructed when people are asked a valuation question. Construction of preferences means that preference measurement is best viewed as architecture Ž building a set of values. rather than as archaeology Ž uncovering existing values.. We describe potential faults in the process of preference construction, offer guidelines for measuring constructed preferences Ž a ‘‘building code’ ’. to mitigate these faults, and discuss how the code must be sensitive to the purpose of the valuation Ždesign vs. prediction.. Key words: constructive preferences, value measurement, decision aiding JEL Classification: D80

Community is a core construct in social thought. Its intellectual history is lengthy and abundant. Community was a prominent concern of the great social theorists, scientists, and philosophers of the nineteenth and early twentieth centuries (e.g., Dewey 1927; Durkheim [1893] 1933;

Innovative industries are often characterized by rapid product turnover. Product longevity may be driven by both a product’s position within a market as well as its position within a firm’s larger product portfolio. However, we have little understanding of the relative importance of these factors in determining product turnover and how they interact as an industry evolves. Although researchers have invested substantial effort in analyzing firm survival and turnover, there are far fewer studies of the determinants of product survival and turnover. We use hazard rate models and count regression models to describe the behavior of firms and their products with a new and detailed database on the laser printer industry. We show, first, that competition and market structure variables have a large impact on both speeding product exit and delaying product entry. Second, there is some evidence that firms that have maintained a high market share for a number of years keep their products on the market longer than those with lower market share. Finally, firms with high innovative capacity tend to enter markets frequently, but withdraw their products at average rates. Firms with strong brands tend to introduce few products and withdraw their products slowly. With these findings, the paper links product entry and exit decisions to the broader literature on firm strategic and product

ABSTRACT. Firms face competing needs to expand product variety and reduce production costs. Trade policy affects firm investments in product variety and production processes differently. Access to larger markets enables innovation to reduce costs. Although firm scale increases, foreign competition reduces markups. Firms react by narrowing their product varieties to recapture these lost markups. I provide a theory detailing this conflicting impact of trade policy and address welfare gains from trade. Accounting for firm heterogeneity, I show support for the theoretical predictions with firm-level innovation data from Thailand’s manufacturing sector which experienced unilateral home tariff changes during 2003-2006. Acknowledgment. I am grateful to Bob Staiger for continued guidance and to Charles Engel and Steven Durlauf

We present a theory of individual choice in which the decisionmaker focuses more on, and hence weights more heavily, attributes on which the options in her consideration set are more different. Consistent with evidence on salience in monetary choices, our model predicts that the decisionmaker is biased toward options whose advantages are concentrated in fewer attributes. In intertemporal choice, because a single period’s choice can lead to a different concentration of consequences than a lifetime perspective that integrates many choices, the model often predicts time inconsistency in behavior. The decisionmaker exhibits present bias in “lifestyle ” decisions whose consequences are distributed over many future dates, but also overcommits to an increasing number of future goals with a single large benefit each. In response to the bias toward concentration, profit-maximizing firms design products with one core attribute, and split prices into as many pieces as they can. A strong firm designs products which are strong on its competitor’s weak attribute, while a weak firm copies the strong firm’s strength. We also propose a theory of consideration-set determination in which the agent considers the set of options that maximizes a combination of utility and differences between attributes.

The authors are listed in reverse alphabetical order. They gratefully acknowledge the support of IRI/Europanel, who provided the data on which this study is based. They thank the participants at the MSI Conference on Competitive Reaction for constructive comments on an earlier draft of this work. Financial support by the Flemish Science Foundation (F.W.O) under grant G.0145.97 and the Research Council of the Catholic University of Leuven under grant OT.96.4 is greatly How do competitors react to each other’s price-promotion and advertising actions? How do these reactions influence the net sales impact we observe? We answer these questions by performing a large-scale empirical study of the short-run and long-run reactions to promotion and advertising shocks in over 400 consumer product categories, over a four-year time span. Competitive reaction can be passive, accommodating or retaliatory. We first develop a series of expectations on the type and intensity of reaction behavior, and on the moderators of this behavior. These expectations are assessed in two ways. First, vector-autoregressive models quantify the short-run and long-run effect of a promotion or advertising action on competitive sales and on competitive reactions. By cataloging the numerical results, we are able to formulate

(Please do not cite or circulate without permission)

this paper. The predominant form is called a covariance structure model (CSM), or a structural equation model (SEM)

r marketing academic researchers and practitioners. In the 1990s, a whopping 81% of new products introduced were brand extensions (Keller 1998). While many new products generally fail, every year the most successful ones tend to be brand extensions (Aaker 1991). Starbuck's ice cream, Nokia cellular phones, and Nike sportswear are just a few recent successful examples. In light of their importance, brand extensions have received much attention in academic research during the past decade. One of the most robust findings in past research is the role of the semantic relationship between the parent brand and extension category in determining brand extension evaluations (Aaker and Kel- *Shi Zhang is assistant professor, and Sanjay Sood is assistant professor, both at the Anderson School, University of California, Los Angeles, Los Angeles, CA 90095; e-mail: shi.zhang@anderson.ucla.edu; sanjay.sood@ anderson.ucla.edu. The authors contributed equally to the article. They acknowledge the financi

As the use of the Internet continues to grow among consumers, firms need to assess the requirements of individual branding-strategies in the context of electronic commerce. Although branding has been an integral part of the marketing mix for a long time, the issue of brand orientation and consumer loyalty in the context of the Internet has not yet seen the same attention.