market leadership and relative market share

In looking for companies to invest in, market leaders are very often a good place to start.  They have the advantage of their large size, high visibility to existing and potential customers, and likely scale economies.

Of course, there are some common sense caveats to this.  We have to ask ourselves, for instance, whether the market a company claims to be addressing actually exists or is worth the effort to dominate.  If, say, a restaurant chain says it rules the market for left-handed diners in Cleveland, we might scratch our heads on either count.

Once we get past silliness, though, there is an important market attribute to consider–market concentration.  A useful way to approach this issue is through relative market share..  Relative share can be calculated in a number of ways, the simplest (and therefore my favorite) is to calculate the market share of a given competitor–usually either the firm you’re interested in or the number two in the market–as a percentage of the market leader’s.

In a market where the leader controls 60% of the market, #2 has 30% and #3, 10%, the relative market shares are 1, .5 and .17.  This is usually a very favorable situation for the leader.  Unless the overall market is growing at warp speed–in which case every firm is staking out territory as fast as it can and not worrying about what competitors are doing, #3 had better had a good niche strategy or else it’s toast.  #2 may be in a better position than #3 but may also be vulnerable to market share loss from #1.

A market where #1 has 35%, #2 has 33% and #3 has the rest is far different.  The relative market shares are 1, .94 and .91.  This is most likely an incredibly competitive market, where #1’s “leadership” may only be a function of its greater willingness to cut prices.  While the dynamics of such a market may be theoretically interesting, for investors–except, again, in the unusual situation of hyper-growth, it’s usually one to watch from the sidelines.




Leave a Reply

%d bloggers like this: