The size of the critical mass as a function of the strength of network externalities: a mobile telephone estimation

ICT Diffusion in Developing Countries

“…Thus, the conviction that the 'critical mass' is easily detectable might be misleading and confusing. Baraldi (2012) provides new insights into the estimation of the size of the 'critical mass' rather than concentrating exclusively on its determinants (recall the works of, e.g., Grajek and Kretschmer 2012). To quantify 'critical mass', Grajek and Kretschmer (2012) define it as a function of the installed base and price.…”

“…Different approaches to the identification of 'critical mass' are offered by Grajek (2003Grajek ( , 2010, Kretschmer (2011, 2012), Baraldi (2004Baraldi ( , 2012, Arroyo-Barrigüete et al (2010) and Villasis (2008). Baraldi (2012) argues that the size of the 'critical mass' is determined by the strength of network effects. Following, e.g., Cabral (1990Cabral ( , 2006, they presume that due to the installed base effect the diffusion of products should continue even if prices remain unchanged.…”

“…To quantify 'critical mass', Grajek and Kretschmer (2012) define it as a function of the installed base and price. 30 Similar to Baraldi (2012), Arroyo-Barrigüete et al (2010) offer a conceptualization of the 'critical mass'. Consequently, Grajek and Kretschmer (2012) develop a structural model of demand with installed base effects.…”

“…To estimate the threshold level of the 'critical mass', they suggest that diffusion is highly endogenous and the process as such can be identified as multi-equilibrating. 30 Baraldi (2012) specifies the network effects as: Strong network effects allow for a lower installed base and higher prices to assure the sustainability of further diffusion, and the opposite is true in the case of weak network effects.…”

“…To estimate the size of the critical mass, Baraldi (2012) follows Rohlfs (1974), Katz and Shapiro (1985) and Economides and Himmelberg (1995a) and formalizes the inverse demand function as X i,t is thus the installed base, p i,t is price, GDP population i, t is GDP per capita, and g X i, tÀ1 ð Þreveals network externalities in country i at time t. To control for concavity, g X i, tÀ1 ð Þincludes a squared term for the lagged installed base.…”