|
Description
I present an ordered-search model that, in contrast with random-search models, yields an intuitively appealing equilibrium in which there is price dispersion, prices and profits decline in the order of search and consumers with lower search costs search longer and obtain better deals. These features of the search equilibrium hold regardless of whether consumers are informed of prices prior to searching.
1. Introduction
* Imagine a consumer who enters an Oriental bazaar and walks by a row of sellers, observing their prices along the way. The consumer first has to pass the sellers located closest to the entrance in order to obtain offers from those located further away. Intuitively, the consumer is unlikely to find the best deal at the entrance to the bazaar. The reason is that sellers located at the entrance recognize that they have some market power over those consumers who will not shop around due to high search costs. These sellers will charge higher prices than those located down the line. Building on this intuition, I develop a simple model of a search market in which consumers sample firms in a predetermined order.
In the ordered-search model, there exists a search equilibrium (characterized by price dispersion and nontrivial consumer search) even though (i) the product is homogeneous, (ii) firms are identical except for the order in which their offers are evaluated, and (iii) consumers differ only in their search costs. In the equilibrium, prices and profits decrease in the order of search, consumers engage in comparison shopping, and consumers with higher search costs search less and end up paying more. I show that these features of the equilibrium hold regardless of whether consumers are informed of prices prior to searching.
Examples of markets in which firms know the order of consumer search are not confined to spatially structured markets, such as the Oriental bazaar. In recent years, it has become technologically feasible to track consumer searches online. Through partnerships with advertising companies and price-comparison sites, sellers can acquire information on prior consumer visits to other sellers and then offer a price quote accordingly. (1) Similarly, in labor markets, the value of an offer extended to a job candidate may depend on the past and future options available to that candidate.
Another recent phenomenon is that of online pay-for-placement search engines, such as Overture.com, that take payments from merchants and display at the top of the list in search results those sellers who pay the greatest amount. (2) For some popular keywords, advertising firms are willing to pay over $7 per click-through to be listed at the top of the displayed search results. This supports the idea that sellers value top positions given by search engines and that a consumer search is not uniformly random. (3) Recent empirical studies of online marketplaces document a strong relationship between the ordinal ranking of results and consumer choices. Smith and Brynjolfsson (2001) show that consumers buying books at a shopping bot, EvenBetter.com, are much more likely to select an offer displayed first in search results as well as offers displayed on the first page. Ellison and Ellison (2004) find that ordinal ranking of sellers at a price-comparison site, Pricewatch.com, better predicts sales than do listed prices. (4)
The predictions of the ordered-search model are in agreement with the empirical and experimental evidence. Price dispersion is a well-documented phenomenon in both traditional and online markets. From the seminal article by Stigler (1961) to recent studies of online markets by Baye, Morgan, and Scholten (2004) and Baye and Morgan (2004), researchers have documented significant and persistent differences in prices of seemingly identical products. Experimental evidence is consistent with the idea that, when firms know the history of consumer search, they price discriminate based on this information. In experiments by Deck and Wilson (2006), sellers charge more when they are positioned higher on the consumer search list. I show that, for a predetermined search order, equilibrium pricing indeed has such a pattern.
In contrast, the literature on (uniformly) random-search markets reveals that, when consumers engage in a costly sequential search, monopoly pricing is the unique equilibrium strategy for the firms (Diamond, 1971). The result holds for any positive search costs and any number of firms in the market. The reasoning is that a firm with the lowest price has an incentive to increase its price slightly if the price is below the monopoly level because a marginal price increase does not reduce the firm's sales. Another feature of the Diamond paradox is that it arises in a model of sequential search, but because there is no price dispersion in the equilibrium, consumers search at most one firm. Therefore, the equilibrium that arises in random-search markets does not describe the way search markets operate in the real world. A number of resolutions to the Diamond paradox are offered by subsequent literature. (5) Yet, the assumption of uniformly random search has been maintained by most researchers.
The rest of the article is organized as follows. In Section 2, I present a benchmark (full information) model of ordered search in which firms simultaneously choose their prices and consumers make purchase decisions that are conditional... |
More articles from RAND Journal of Economics
Seasonality in the U.S. motion picture industry., March 22, 2007
Corporate strategy and information disclosure., March 22, 2007
Collusion and dynamic (under-) investment in quality., March 22, 2007
Consumer preferences and product and process R&D., March 22, 2007
Outsourcing, information leakage, and consulting firms., March 22, 2007
Looking for additional articles?
Click here
to search our database of over 3 million articles.
|
