Most of us are familiar either with Metcalfe's law or the general principle of it: networks are more valuable the more people are on it. The common example used to illustrate is a telephone. One person on a telephone network is useless. Two people is an improvement. With millions, you are far more likely to be able to call the person you want to.

Recognizing this principle flips common arguments about connecting rural areas on their head. Ensuring that people in rural areas are connected benefits everyone -- it is not charity. Connecting people in rural areas increases the value of the connections in urban areas, creating more value for everyone.

But the flip side may be too rarely considered. As these networks have grown in size (and therefore value), the cost of those who are excluded from them also increases significantly. This means that while the costs of not connecting rural areas are high today, those costs will be even greater in coming years. The argument is rather intuitive, but for those who want to learn more, Rahul Tongia and Ernest J Wilson III published an academic paper this year in the International Journal of Communications.

The abstract for "The Flip Side of Metcalfe's Law: Multiple and Growing Costs of Network Exclusion" is here:

The study of networks has grown recently, but most existing models fail to capture the costs or loss of value of exclusion from the network. Intuitively, as a network grows in size and value, those outside the network face growing disparities. We present a new framework for modeling network exclusion, showing that costs of exclusion can be absolute, and might, at the extreme, eventually grow exponentially, regardless of underlying network structure. We find that costs of exclusion can also be spread to the “included” through several mechanisms such as parallel networks, and we also highlight how future research needs to capture the interaction of alternate or parallel networks to the network at hand. Backed by empirical evidence, this will have wide-reaching policy and design implications, particularly for the role of subsidies or direct intervention for...