Market definition refers to the process of determining the set of products
those are substitutable or interchangeable to the reference good for the
purpose of analyzing the antitrust issues at hand. However there are lots of
competition law cases in which the reference good consists of complementary
goods, and/or relevant goods are not substitutable but complementary each
others. Aftermarkets, two-sided markets, and cluster markets are the typical
examples. Since goods are not substitutable in those examples, the mechanical
application of SSNIP test or critical loss analysis to define relevant good market
could make serious mistakes. The most possible mistake will be to define
relevant market narrower than actual market since SSNIP test does not
consider the complementarity among goods. Therefore this article try to verify
the most important factors which should be considers when aftermarkets,
two-sided markets, and cluster markets are defined. In case of aftermarket,
the extent of switching cost, and information asymmetry between consumers
and firms, trade practice should be considered to decide whether equipment
and aftermarket service are defined as the same market or not. In case of
two-sided market, the most important factor that makes difference from
single-sided market is cross-network externality. Two distinct groups of users
in the two-sided platform become complementary goods to each other through
cross-network externality. Thus the feedback effect of the action which take
place in one side should be considered during market definition procedure. In
case of cluster market, transactional complementarity, economies of scope and
discount rate of cluster price, unbundling cost, extent of technological
integration, demand and trade practice should be considered to define the scop
of clustering.