The Consensus Among Economists on Multisided Platforms and the Implications for Excluding Evidence That Ignores It

David Evans, Jun 13, 2013

Multisided platforms create value by bringing two or more different types of economic agents together and facilitating interactions between them that make all agents better off. These platforms play critical roles in many economically important industries. There is considerable material on this subject; Richard Schmalensee and I have provided a survey of the economics literature on multisided platform literature, as of 2012, with particular emphasis on antitrust applications.

This note draws out some of the implications of that literature for the reliability of economic models, tools, and theorems that have been commonly relied on in antitrust analysis. It makes seven key points, each of which is elaborated on below.

1.     There is a well-developed, non-controversial, peer-reviewed economics literature on multisided platforms.

2.     The fundamental defining characteristic of multisided platform businesses is that they supply multiple customer groups and that the demands of these customer groups are interdependent.

3.     It is now well accepted among economists that a number of economically significant industries are based on multisided platforms.

4.     Economic models that account for interdependent demand among customer groups yield profit-maximization conditions that differ, in significant and important ways, from economic models that do not account for such interdependent demand.

5.     There is no reason to assume that models that do not consider interdependent demand apply to multisided platform businesses where such interdependent demand is important.

6.     Many economic theorems, tools, and models that are routinely used in antitrust analysis are not reliable, at least not without explicit modification, to account for interdependent demand.

7.     Economic analysis of antitrust issues that fails to account for interdependent demand for multisided platforms-either by explicitly considering this demand in the models relied upon or by accounting for possible biases from not doing so-is not reliable and should not be given weight by courts or competition authorities.

The last point has obvious implications for considering Daubert motions concerning economic experts in cases involving multisided platforms.

Links to Full Content