Finding the Right Spot

Wal-Mart is the king of the retail discount industry, or at least that's what everyone says. Without access to detailed data, however, that claim may be hard to prove. New research looks into the industry players' location decisions and reveals just how formidable Wal-Mart really is.

A store's location is one of the most important decisions an owner has to make to set himself apart from his competitors. This is especially true since a store's location cannot be changed as easily as, for instance, the price of a product.

In choosing a location, a store owner must consider a number of factors such as the size of the market, the cost of labor, and the logistics of moving products from suppliers to the site. But equally important is taking into account other retailers that may already be operating nearby or could potentially move in. This presents a crucial tradeoff. A retailer may find a market's large population and low cost of labor attractive, but so would its competitors.

With this in mind, University of Chicago Booth School of Business professor Ting Zhu and Vishal Singh of New York University looked at what's behind the location choices of three dominant players in the retail discount industry: Wal-Mart, Kmart, and Target. In their recent study "Spatial Competition with Endogenous Location Choices: An Application to Discount Retailing," Zhu and Singh analyzed whether these retailers are attracted to different types of locations, whether it pays to compete in the same market, and whether the threat of a nearby competitor varies across firms, distances, and store formats.

Zhu and Singh also are targeting a larger goal. They want to understand how retailers compete with each other despite limited information. "We want to know why Wal-Mart is so successful, how many customers go to Wal-Mart, and what kind of advantage the retailer has," explains Zhu. "But it's very hard for researchers to get detailed data at the consumer level."

Information about store locations, however, is publicly available. By observing the retailers' location choices, the authors have found an indirect but effective way to overcome this challenge and get a better picture of the competitive interactions among the biggest players in this industry.

A Location Game

If a retailer has chosen a certain location and the store has been open for an extended period of time, then the assumption is that the store is profitable. If it is the only retailer around, then that's likely because it is viewed by competitors as too powerful to compete against it. Zhu and Singh used this as a starting point to create a model where Wal-Mart's, Kmart's, and Target's store location choices are an outcome of a game between these players.

A retailer's payoff from choosing a location depends not only on market conditions in an area, but also on its expectation of how competitors will respond if it moves into that location. Wal-Mart's decision to open a store, for instance, would depend on whether it thinks Kmart and Target are going to enter that market and how far away they would be from each other. This reflects the fundamental tension faced by these firms: the desire to be in an attractive location while wishing to insulate themselves from competitors by choosing a site that's farther away.

One important feature of Zhu and Singh's analysis is that firms are allowed to be different, both in terms of their preference over market conditions and their impact on other retailers. Previous studies showed a firm's profits depended partly on the total number of firms competing in the market rather than the actual identities of their competitors. In reality, however, firms care about who they are up against. Kmart's profits, for instance, may differ depending on whether it faces Target or Wal-Mart, and whether that Wal-Mart is a discount store or a larger supercenter.

The Power of Wal-Mart

People who live closer to a store are expected to have a stronger impact on its profits than residents living farther away. But Zhu and Singh also found that those who live in more distant areas matter too–but only if you're Wal-Mart. This is especially true for Wal-Mart's supercenters. It suggests that Wal-Mart has a greater ability than its rivals to attract customers who live farther away.

Wal-Mart and Target both like markets where a larger proportion of families have children, but differ when it comes to their preference over income and education. Wal-Mart tends to enter markets with lower incomes while Target seems to prefer higher income areas as well as locations where a larger percentage of people have a college degree. Zhu and Singh's results are consistent with common perceptions of how Target and Wal-Mart position themselves. Target has traditionally marketed itself as an upscale discounter. Wal-Mart, on the other hand, has established itself as the low price leader in the industry.

Zhu and Singh also found that the presence of rivals reduces a retailer's profits, but the effect diminishes with distance. Thus, by choosing a location that is sufficiently far from its competitors, firms can successfully shield their profits. But not all firms have to try so hard. Their analysis reveals that the impact of retailers' location decisions on their competitors' profits varies significantly across firms. While Kmart can affect Wal-Mart's profits when it sets up shop nearby, its influence is much weaker than the other way around. Similarly, when Kmart is located more than 10 miles away from Wal-Mart, Kmart's impact on Wal-Mart's profits is only about one-fourth of Wal-Mart's impact on Kmart at the same distance.

Wal-Mart's strength becomes even more apparent if one looks at its supercenters–stores that include a full-service supermarket unlike a regular discount store. At the time that Zhu and Singh wrote their study, about half of Wal-Mart's stores were supercenters and only a much smaller fraction of Target and Kmart stores were supercenters. They found that the impact of a Wal-Mart supercenter dies out more slowly with distance compared with the impact of its competitors' regular stores. This means that a Wal-Mart supercenter has considerable power over a larger area, and even competitors that are relatively far away may be vulnerable.

Another way to portray these dynamics is to calculate how much demand would be affected if another store moved in its neighborhood. For instance, Zhu and Singh found that placing a new Wal-Mart store beyond 10 miles would be equivalent to reducing a Kmart store's population by 20,900, compared with only 2,000 for a Target. Thus, a Target store is able to hold out well against a distant Wal-Mart store. However, if a Wal-Mart supercenter moves in, the reduction in a Target store's population jumps to 18,900.

In a related paper, Zhu, Singh, and Mark Manuszak of the Federal Reserve Board ask what the minimum population for a regular Wal-Mart discount store and its supercenter should be in order to be profitable in that market. One would think that a supercenter's threshold ought to be higher because it is a bigger store. But Zhu and her coauthors found that the required thresholds are actually quite similar, which means that a supercenter does not need a significantly larger population to support its size. One reason is that people seem to be willing to travel the extra miles to shop at a supercenter. Also, people may go there more often because it carries more products, which would make the customers' average "purchase basket" much larger.

Monopoly, Duopoly, Oligopoly

Zhu and Singh's analysis can also be used to predict whether one or all players will enter a market. And as their results show, that depends on market conditions and the stores' effect on each other. "A discount retailer does not always pick the most attractive spot for its stores," says Zhu. "If a retailer thinks that Wal-Mart is an intimidating player, then it may be better off locating a little farther away even if market conditions there are not as good."

Suppose that there are two possible locations in a certain market–one with a higher level of population and income than the other. A Wal-Mart supercenter may be willing to build a store in the location with the lower population because it knows it can attract customers from distant areas. The site with the higher population and income, on the other hand, may be attractive to Target and Kmart. But Zhu and Singh found that these two retailers will only enter the market under very favorable conditions. Otherwise, they will probably stay away because of Wal-Mart's considerable reach.

Thus, while it is possible that two or all three players could move in the same market, the odds that a Wal-Mart ends up by itself are much higher than a Target or a Kmart enjoying a monopoly. Target does manage to get a small monopoly region, but only in areas with very high income levels. Kmart is unable to operate as a monopolist in this simulation exercise that only looks at changes in income and population. Zhu and Singh note, however, that it may be able to do so in areas close enough to its headquarters.

Zhu says their study may be useful to those who would like to understand how certain changes in market conditions or policies can affect the structure of a market. "Some counties and states give subsidies to firms," explains Zhu. "As a policymaker I have to ask myself, 'If I give a benefit to this firm, will it enter my market and how will it affect the other players?'" Looking at the entire picture can surely help form better policies.

"Spatial Competition with Endogenous Location Choices: An Application to Discount Retailing." Ting Zhu and Vishal Singh.