Positive network externalities exist when consumer benefit increases with the number of adopters of a network.  This effect can be direct or indirect.  Direct externalities arise when the attractiveness of a network increases directly with the number of consumers; for example, a telephone network.  These externalities are indirect when an increase in consumers results in an increase of complementary products provided for the network.  For example, more software is made available for a video game system as more people adopt that system.

This paper analyzes competition between multiple standards when indirect network effects are present.  Implications are obtained for the evolution of market leadership over the product lifecycle in an industry characterized by this hardware/software relationship.

There is a tendency for a dominant standard to emerge in industries characterized by indirect network effects.  This is demonstrated by Viscusi et al. (2001):  “In a market characterized by network effects, one product often becomes the standard (or the market ‘tips’ to one product).  This outcome is especially likely if, as in the software industry, there are large economies of scale (or increasing returns).”  Of particular interest is failure to adopt a technologically superior hardware system because of large network effects associated with an inferior system (see the seminal works of David (1985) and Liebowitz & Margolis (1990) for opposing views).

Results show this outcome is rare.  Analytical and computation evidence suggest a higher quality hardware system is rarely dominated by a lower quality system with a larger installed base of consumers.  The indirect network effect is seldom large enough to drive a higher quality entrant from the market.  The elasticity of market share with respect to quality is significantly larger than the elasticity of market share with respect to network size.

Data from the home video game industry provides empirical support.  Econometric evidence suggests markets share is roughly eight times more sensitive to hardware quality than network size.  Finally, estimates suggest an increase in an entrant’s relative quality has a larger absolute effect on the probability of entrant success than a similar increase in the size of the installed base of an established competitor.

References:

David, P. A.  1985.  “Clio and the Economics of QWERTY,” American Economic Review, 72(2), 332 – 337.

Liebowitz, S. J., and Margolis, S. E.  1990.  “The Fable of the Keys,” Journal of Law and Economics, 33(1), 1 – 25.

Viscusi, W. K., Vernon, J. M., and Harington, Jr., J. E.  2001.  Economics of Regulation and Antitrust, 3^rd ed.  MIT Press, Cambridge, Massachusetts.