|
Article Excerpt
A consumer's demand for a durable good is governed not only by his individual preferences but also by preferences of other market participants. This interdependence of preferences arises from the inevitable resale of durable goods. If most people prefer goods with certain features, original buyers conform and choose goods with these features even if they do not like them. Using a matching model, we show there is always conformity in equilibrium. The incentive to conform is strongest for long-lived durables and for people who trade frequently. If average preferences are sufficiently strong, there is always too little conformity in equilibrium.
"Conformity, in terms of size, condition & features, tends to support your home's market value more than anything else." (1)
1. Introduction
* Durable goods survive for long periods of time and are often possessed by many different people over the life of the good. Because many durables change hands from time to time, the efficient provision of these goods should to some extent reflect the preferences of many potential owners. Indeed, if durables are exchanged very frequently, they should have features that cater to the average preference in the market rather than the preferences of a single individual. As a result, there should be less product variety in durable goods markets--the distribution of varieties available for purchase should be compressed relative to the underlying distribution of preferences. Put differently, there should be conformity in markets for durable goods.
Market forces provide incentives for conformity in the provision of durable goods. These incentives typically arise through resale concerns. Because durables are often traded in secondary markets, resale concerns influence the original purchase decisions and thus encourage conformity. In contrast, nondurable goods are consumed by a single person and thus there is no incentive to conform to the average preference.
For example, new houses often have features and styles that differ only superficially from one house to the next. New housing developments are often derided because they consist merely of "cookie-cutter" houses or "McMansions." These houses are virtually the same--most have cathedral ceilings, walk-in closets, built-in jacuzzis, mud rooms, and so forth. Of course, many of these features are desirable but it seems unlikely that preferences are so aligned as to justify such a homogeneous mix of products. The apparent homogeneity among new houses may instead be an efficient market reaction. Rather than catering to individual tastes, builders conform to the average taste anticipating the eventual resale of the house.
Conformity also likely arises for durables other than housing. For example, because used cars are frequently traded, there may be pressure to conform in automobile purchases. A new car buyer might purchase a car with an automatic transmission even if he prefers manual transmission because of resale concerns. The used car market, because it is localized and fragmented, creates an incentive to conform to the average preference. Conformity may also arise in markets for business capital. In this case, firms face a tradeoff between having capital equipment that fits their specific production purposes and having capital that has a high resale value. Because capital goods that are valued by many other firms can be easily resold, firms have some incentive to conform by employing capital that reflects the needs of other firms. As it does in our model, the incentive to conform increases with the durability of the capital goods.
We analyze these issues with a matching model in which agents buy and sell a long-lived durable good that must be resold from time to time. Although the specific function of the durable is not important for the analysis, we refer to the durable as a house. There are two types of houses and agents differ according to their preferences over the two types. Frictions in the resale market imply that agents are not perfectly matched with others who have the same preferences. Thus, if someone buys an unusual house, he runs the risk that he will not be able to sell it if he needs to move. Resale concerns can be so strong that the individual chooses to purchase a good that he dislikes relative to other available goods. When this occurs, we say that the individual is conforming to the market.
In equilibrium, there is a tendency to conform to the average preference. Rather than being a knife-edge phenomenon, conformity is the typical outcome in markets for durable goods. Because resale concerns increase with durability and the incidence of trade, there is greater conformity in markets for long-lived durables and for people who trade frequently. Although agents do not care directly about the preferences or actions of other agents, in equilibrium they act as though they do. Because they buy and sell a common set of goods, durability and trade endogenously align the preferences of the agents.
The equilibrium level of conformity is often not socially optimal. Surprisingly, the model suggests that, if preferences are sufficiently strong, there is too little conformity in durable goods markets. There are two reasons for this inefficiency. First, by conforming, agents reduce search costs. If they have a house that few others want, they will have difficulty selling it if they need to move. Of course, the original builder has an incentive to conform to reduce the severity of these search costs. However, the search costs affect both buyers and sellers. Because the original owner only internalizes his own search costs, he has too little incentive to conform. Second, even if the house is sold, it is possible that the house will not be an ideal match for the new owner. The seller typically does not fully internalize the social costs incurred when the buyer settles for a house that is not ideal for his needs. Because he will likely be matched with someone who has typical preferences, by conforming, the original owner would reduce the number of mismatches. Yet, unless the seller captures all of the surplus from trade, he does not fully internalize this cost and again, there is too little incentive to conform.
The rest of the article is organized as follows: Section 2 presents the model, describes optimal behavior, and defines and characterizes the equilibrium. Section 3 presents the main results of the article. We show that conformity increases with durability and the incidence of trade. Section 3 also considers the welfare implications of conformity. Section 4 discusses the results and considers several possible extensions. Section 5 discusses related literature. Section 6 concludes.
2. Model
* We consider a continuous-time matching model in which agents can own one of two types of a durable good. Although as we discussed above our analysis holds for many durable goods markets, we take the good in our model to be a house. We denote the two types of houses as type a and b. The houses could differ along many dimensions. For example, type a houses could be "traditional" houses while type b could be "modern" houses. Alternatively, type a could be a two-story house with a small yard while type b might be a one-story house with a large yard. Every agent must have a house in every period.
We normalize the utility functions so that all consumers get a flow utility of I from living in the type a house. Consumers have different tastes for the type b house. Specifically, each consumer has an individual taste parameter z which quantifies their preference for type b houses. For a consumer with taste parameter z [member of] R, the flow utility from living in a b house is 1 + z. Thus, the flow utility for an agent with a given z and a given house x [member of] {a, b} is
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]
We assume that z is distributed over the population according to a distribution function F. The support of F is a subset of R.
From time to time agents switch houses. Agents may switch their house for one of two reasons. First, the house may "die." We think of this as capturing normal depreciation but it may also include extreme idiosyncratic events such as fires, and so forth. When this occurs, the agent must build a new house. We refer to this event as the "build shock." We assume that the build shock obeys a Poisson process with an exogenous arrival rate [delta]. An agent who gets the build shock decides which type of house to build and incurs a building cost c. Because we want to focus on heterogeneity in tastes, we assume the building cost is the same for all agents and for either type of house.
Second, the agent may be required to trade his house, which we refer to as the "trade shock." Agents who get the trade shock must move out of their house and into a new house. We allow agents to differ in the likelihood of receiving the trade shock. Thus, some agents move often whereas others do not. Heterogeneity in the likelihood of trade allows us to distinguish between individual trade hazards and aggregate trade hazards. We assume that for each agent, the trade shock obeys a Poisson process with an exogenous arrival rate y. The arrival rate is distributed over the agents according to the distribution G. The support of G is restricted to the interval [0, [infinity]]. Here y = corresponds to an individual who never needs to trade and y = [infinity] implies that the individual...
|
