Informative advertising and the Unitd States personal computer market: A structural empirical examination

“he who has a product to sell and goes and whispers in a well is not so apt to get the dollars as one who climbs up a tree and hollers” —Author unknown

Traditional models of consumer choice assume consumers are aware of all products in the market. This assumption is questionable, especially when applied to markets characterized by a high degree of change. In this study, I present an empirical discrete-choice model of incomplete information on the part of consumers—consumers may not be aware of all the products for sale in the industry. I model the probability a consumer knows a product as a function of advertising, consumer attributes, and product attributes.

I apply the model to the US market for personal computers (PC). The PC industry is a rapidly changing industry. On average over 200 new products are introduced by the top 15 firms alone every year. The structural model encompasses both consumer and firm optimal decisions treating advertising as influencing the set of products from which the consumer chooses to purchase. The impact of advertising on a consumer's ‘choice set’ depends on the characteristics of the consumer and the medium in which the advertising appears. The estimation technique incorporates micro- and macro-level data. I use data on PC purchases, advertising expenditures, and media exposure gathered from three different sources.

The estimates indicate that product characteristics, such as whether the computer has a Pentium chip, its CPU speed, and whether it is a laptop, have a significant positive effect on utility. The product characteristic driving substitution patterns, for all but Apple computers, is the form-factor (eg laptop). The estimates indicate there are increasing returns to advertising and that PC firms use advertising to target high-income households. For some firms advertising one product has positive effects on other products sold by that firm (relative to rival products). In addition, word-of-mouth or experience plays a large role in informing consumers. The average markup over marginal costs is 34%. While the top 5 firms have higher than average markups, charge higher than average prices, and engage in higher than average advertising. Results from merger simulations indicate that, should firms be banned from changing their advertising levels, industry prices would be lower.