HOW do you define a market? If it is America's market for office products, that depends on whom you ask. According to Staples, an office-supplies superstore chain that has since last September pursued a $4 billion takeover of Office Depot, its bigger rival, office products are sold in a fragmented $185 billion market in which the merged firm would have a meagre 5-6% share. According to the Federal Trade Commission (FTC), which on March 10th voted to block the deal, the merged group would dominate a much narrower “office-supply superstores” market, pushing up prices for consumers. After Staples and Office Depot agreed to sell some stores to a rival two days later, the FTC seemed set to back down. But does its tough “niche-market” approach to trustbusting make sense anyway?

The FTC's view is that Staples is one of only three key players--the third is OfficeMax--in a market characterised by “a unique combination of low prices and one-stop convenience”. The agency found that in cities where Staples and Office Depot compete, prices are lower than where they do not; a survey by Prudential Securities found that prices were lower still where all three firms did battle. So a divestment-free merger, believes the FTC, would have “substantially” reduced competition in 40 metropolitan markets, and allowed the merged firm--which will have annual sales of about $10 billion and, on the FTC's definition, a two-thirds share of the superstore market--to exert undue influence on suppliers and shut out potential rivals.

Nonsense, says Staples, which opened America's first office superstore in 1986. The firm reckons its paperclips-to-PCs superstores compete with everything from family businesses, mail-order companies and warehouse clubs to giants such as Wal-Mart, CompUSA and Circuit City. Nonetheless, to appease the FTC, Staples and Office Depot agreed on March 12th to sell 63 of their 1,149 superstores to OfficeMax for $109m. Staples has also said it might be willing to sign a consent decree promising to pass on cost savings from the merger to consumers.

Whether the FTC's analysis of the merged firm's market power would ever have stood up in court is debatable. To assess a market's competitiveness, trustbusters use a measure of concentration known as the Herfindahl-Hirschman Index (HHI). This is calculated by squaring the percentage share of each firm in a market, then adding them all up. A market cornered by a single monopolist would therefore have an HHI of 10,000; one shared equally among 50 firms would have an HHI of 200. Typically, the FTC rates an individual market as highly concentrated if its HHI exceeds about 1,800.

On this scale, the office-products market as defined by Staples is highly competitive, with an HHI in the low hundreds, whereas that defined by the FTC, with only three big players, is already highly concentrated, with an HHI of well over 3,000--a rating that will soar above 5,000 following the merger. So if a court were to accept the FTC's definition of the office-products market, it would also be likely to buy the argument that an already concentrated market will become stiflingly uncompetitive after the merger takes place.

The main snag with the FTC's theory is that prices in the overall office-products business (ie, the market as Staples defines it) have fallen dramatically in the decade since the advent of the office superstores, suggesting that their competitive impact extends far beyond the FTC's narrowly defined superstore market. Competitors and consumers appear to treat superstores as only one of many substitutable types of retailer in the office-products market--and not as a unique “sub-market”. If that view is correct, the true HHIs for the market in most cities may be low. Supporting documents examined by the FTC during its inquiry suggest that in San Francisco the HHI may be as low as 57; in Detroit, 75; and in Los Angeles, 108. Such numbers indicate that there exists intense competition.

Had the case gone to court, it is just conceivable that the FTC might have won the day. But, on the evidence available, the market behaviour of consumers and non-superstore retailers of office products, supported by a decade of falling prices and profit margins in the business, suggests that the FTC's attempt to redefine and narrow the market is deeply flawed. Now that the paperclips are settling, the agency might do well to reassess a method that seems to run counter to common sense.