Why is the equilibrium network size can be less than the socially optimal size in networks w/ positive externalities for single system?

Question

For Context, I am reading http://web.mit.edu/rpindyck/www/Courses/NetEx19.pdf. The chapter talks about network externalities in single system like fax machine, email, vaccinations for measles. Paraphrasing from the book "In each of these cases, the marginal social benefit of one more person joining the “network” is greater than the marginal private benefit. Hence, the equilibrium network size can be less than the socially optimal size, so that the competitive market will not be efficient".

Why is equilibrium network size less than socially optimal?

Answer 1

The equilibrium network is smaller than that chosen by a social planner because agents do not internalize network externalities.

Vaccination against a contagious disease is a standard such example. Someone who is not vaccinated exposes others to potential infection and prevents others from, say, using public facilities. This is a negative externality for which he pays no cost. A social planner would therefore impose vaccination when an individual agent might choose not to.