We offer an econometric framework that models consumer's consideration set formation as an outcome of her costly information search behavior. Because frequently purchased products are characterized by frequent price promotions of varying depths of discounts, a consumer faces significant uncertainty about the prices of the brands. The consumers engage in a fixed-sample search strategy that results in their discovering the posted prices of a subset of the available brands. This subset is referred to as the consumer's “consideration set.” The proposed model is estimated using the scanner data set for liquid detergents. Our key empirical results are: (i) consumers zincur significant search costs to discover the posted prices of the brands; (ii) whereas in-store displays and feature ads do not influence consumers' quality perceptions of the brands, they significantly reduce search costs for observing the prices of the brands; (iii) per capita income of consumer's household significantly increases her search costs; and (iv) the consumers' price sensitivity is seriously underestimated if we were to assume that consumers get to know all the posted prices at zero cost. The proposed model is also estimated for the ketchup category to enable us to do cross-category comparisons of consumers' price search behaviorprice uncertainty, consumer search, consideration set, quality uncertainty, consumer learning, bayesian updating, structural model, econometric estimation

Reward programs, a promotional tool to develop customer loyalty, offer incentives to consumers on the basis of cumulative purchases of a given product or service from a firm. Reward programs have become increasingly common in many industries. The best-known examples include frequent-flier programs offered by airlines, frequent-guest programs offered by hotels, and frequent-shopper programs offered by supermarkets. Despite the widespread business practice of reward programs, research efforts on reward programs, particularly in marketing, have been scarce. Our paper takes an important step towards understanding the design of reward programs and its implications on pricing strategies. We study a market that consists of two segments: heavy- and light-user segments. The key distinction between the two segments is that the heavy-user segment purchases in each period and thus is a candidate for the reward programs. In contrast, the light-user segment exits the market after one purchase and is not in a position to exploit reward programs. An important feature of our model is that we allow for different price sensitivity between heavy-user and light-user segments. Our model closely examines the type of rewards. A reward worth a dollar to the consumer might have different cost implications for the offering firm, depending on the type of reward. For example, cash rewards have higher unit reward cost () for the firm than a free product of the firm, such as an airline ticket or long-distance minutes (). Specifically, we examine an interesting puzzle observed in the marketplace. Several firms offer a cash reward or a product made by the firm, such as jackets, electronic items, etc. These firms could offer their own product as rewards and significantly lower their cost. We examine whether there is any reason for such a seemingly suboptimal practice. Our analysis shows that reward programs weaken price competition. By offering the incentives for repeat purchases, reward programs increase a firm's cost to attract competing firms' current customers. Because firms gain less from undercutting their prices, equilibrium prices go up. Moreover, as consumers become unwilling to switch because of potential rewards, the firm with a larger market share in the heavy-user segment charges higher prices. Therefore, a low price in the first period, which leads to a larger market share in the heavy-user segment, will always be followed by a high price in the second period. In our model, consumers are rational and can correctly anticipate firms' incentive to offer lower prices initially to enroll them into the reward programs. Our paper offers an explanation as to why the type and amount of reward may vary across the programs. We identify two determining factors for the selection of rewards: size and relative price sensitivity of the heavy-user segment. We find that in a market with a small heavy-user segment that is also much more price sensitive than the light-user segment, it is optimal for firms to offer the rewards. The intuition is based on...

The authors investigate whether and how pricing and promotional activities influence prescription choice behavior using a comprehensive panel of physicians and data on competitive price and promotional activities. The authors find that physicians are characterized by fairly limited price sensitivity, detailing and samples have a mostly informative effect on physicians, and physicians with a relatively large number of Medicare or health maintenance organization patients are less influenced by promotion than other physicians are.

We study the signalling strategy of a principal who is privately informed about its high demand potential to an uninformed risk-neutral agent. We analyze the model in the context of a contract between a franchisor and a franchisee. We examine the distortions of a two-part pricing scheme necessary to credibly inform the franchisee (agent). We also study whether the inability of the franchisor (principal) to observe the agent's effort moderates or exaggerates the distortions from the first-best two-part pricing scheme. A surprising outcome is that even though the principal incurs greater signalling cost, the magnitude of distortion in the two-part scheme is smaller when service is unobservable than when it is not. Thus, a signalling strategy employing the fixed and variable fees is harder to detect under moral hazard. Empirical studies failing to control for moral hazard may incorrectly conclude that signalling strategy does not occur. We later consider a three-part scheme to verify whether or not the scheme reduces signalling cost. Interestingly, we find that there exists a unique three-part scheme that results in the first-best profit even in the presence of two-sided information asymmetries. While the two-part scheme can never achieve the first-best profit, the three-part scheme always achieves the first-best profit. The costless three-part separating scheme relies on variable income to induce the agent to undertake first-best service. The reliance on variable income to alleviate moral hazard contrasts with the two-part scheme's focus on reducing the variable income to overcome the inability to monitor. The finding provides additional empirical implications.franchising, channels, marketing, signalling, game theory, pricing, asymmetric information models

Our work represents one of the first attempts to assess the impact of IT (information technology) on both process output and quality. We examine the optical character recognition and barcode sorting technologies in the mail sorting process at the United States Postal Service. Our analysis is at the application level, and thus does not involve the aggregation of IT impact over multiple processes. We use data from 46 mail processing centers over 3 years to study the IT impact. We also use a set of factors in our model to account for differences in input characteristics. Our results show that mail sorting output significantly increases with higher use of IT. In addition, IT improves quality which in turn enhances output. We also find that input characteristics exert considerable influence in determining the output and quality of the mail sorting operation. For example, while absenteeism tends to decrease output and quality due to its disruptive consequences, a higher fraction of barcoded mail seems to enhance both performance measures.information technology, business value, productivity, quality, business process evaluation