Defining segments and antitrust
We've argued that the power to define categories is a critical one. It's especially critical in the case of antitrust, where the definition of market segments often make the difference in a decision to oppose a merger or to ignore it.
That power is misapplied, according to a recent article in Forbes magazine (" The Little Picture," 3/29/2004). The author rails against the "circular logic of identifying thin markets that are - surprise! - highly concentrated." The examples cited are:
- The high-function enterprise software market -- invoked in the ongoing Oracle-PeopleSoft case
- The refrigerated pickle market -- used to deny the merger of Claussen and Vlasic in 2002
- The retail superpremium ice cream market -- which affected the Nestle-Dreyers merger in 2003, requiring some asset discards
- PIN debit network market -- which caused First Data and Concord to drop assets before merging in 2003
- The spiral-wound composite can market (as used for frozen orange juice) -- which stopped the merger of industry leaders Sonoco and Pasco in 2003
Of course, the opposite is very broad definitions of market that would allow just about any merger to take place. Indeed this is the big battleground of antitrust, that of segment definition. Any measure of concentration only reflects the definitions you start with.
But it's not just the antitrust officials that define narrow segments. Companies constantly redefine market segments they compete in so that they can tell investors and customers that they are #1 or #2 in a market. In the copier market, for example, Canon, Ricoh, and Xerox al claim to be #1, and they all define that #1status in different ways. And all companies describe themselves as the "leader" in something or other.
