Network effect – Accounting for Externalities

Social organizations are not unique to humans. Ants, gathered in colonies, create buildings, the anthills. These constructions are made possible thanks to a separation of tasks in the organization.

Thus the young ants start by feeding the larvae and queens, then with a little more experience, they start to build the building, to finally become harvesting and defensive workers. This process is not fixed, but adapts itself according to the situation, which can for example in case of dearth, lead to a greater number of ants to harvest than to feed. In some species, an even more precise separation of tasks can be done according to the size of the ant and the shape of its mandibles.

This organization of work (animal Taylorism) allows the ants to survive: the individual benefit of this collaboration is thus positive. Economically, we can say that an individual receives a positive externality from the colony.

In economics, the internal organization of societies is also based on an organization of work, a socialization: it allows to guarantee yields, to lower the break-even point, to produce more. But this remains social factors (the organization) internal to the company.
Other similar factors also occur between companies. We talk about clusters, economic ecosystems, which allow to improve and facilitate the productive activity.

This gain in utility occurs thanks to the network effect, a phenomenon by which the real utility of a set depends on the number of its users. It should be noted that the network effect very often results from a positive externality but can also create nuisances.

This term can include interactions of different types. We will therefore try to qualify them, and then give an evaluation rule according to their origin.

Before evaluating the associated externalities, let’s try to qualify the networks in order to see them more clearly. We identify four types of networks:

Geographical networks,
Technological networks,
Commercial networks,
Social networks,

Technological networks appear thanks to the diffusion of an innovation. The use of this one is conditioned by a wide adoption. We can cite the internet, cell phones, minitel, fax, …
They are the result of technological innovations whose usefulness depends on their diffusion.

Geographic networks appear in clusters or basins that are favorable to economic development. Examples include Silicon Valley, the Ruhr region, and the Paris area. They are the result of a geographical concentration of resources, facilitating interactions and the development of the economy.

Social networks come from subsets of individuals sharing identical norms. Thus social networks are family, friendship, cultural, … They are the consequence of shared norms and common values, bringing together individuals. The term, overused to designate technologies that facilitate exchanges, must be limited, at least in the context of this demonstration, to individuals (the actors of a social network are the individuals, the company plays no role in it, except to propose an innovative technology).

Finally, commercial networks are characterized by common (economic) interests. They are born of existing or potential relationships between customers and suppliers.

This first division makes it possible to specify that three networks are the result of companies and/or States: technological networks, geographical networks, and commercial networks. The fourth, social networks, are the work of individuals (with possibly the State when we are dealing with administrative entities, driven by the public service).
Note that companies like Facebook first used a technological network (innovation), diffused it, exploited the need for socialization of their users and then transformed this network into a commercial network, when it was necessary to monetize the data in the company’s possession.

In the context of the balance of externalities, we will therefore not deal with social networks, which are a matter of individual choice.

Geographical networks are born by:

the will of states, wishing to develop economic clusters or poles of competitiveness,
the will of companies, strong enough to develop an ecosystem in a given region,

Concerning geographical networks, we generally speak of clusters or poles. In this context, we have defined the network as a means of increasing the usefulness of an asset through a geographic concentration of agents. London has thus become an important financial center, thanks to more flexible legislation for banks and the development of dedicated infrastructures (airport, the City, etc.).

These ecosystems (geographical concentration) are also born from the will of companies. Toulouse, a bastion of Airbus, has seen the development of a cluster of companies in the aeronautics sector. The State has also contributed to this by setting up dedicated schools and developing appropriate infrastructures. The network in this case is primarily geographical, but also becomes commercial thanks to the economic transactions between the various companies. Note that the geographical network in Toulouse is smaller than the commercial network of Airbus. The two sets of nodes constituting these two networks intersect but are not equal.

What are the economic benefits / drawbacks of this type of network?

For physical resources, this allows for gains in exchanges (logistical gains), and simplifies and optimizes exchanges. The counterpart of these gains is access to a non-renewable resource, land, which can dry up, with an effect on prices. For the moment, we will assume that there is an equilibrium between the logistical gains and the cost of access to resources, insofar as the only limiting resource, land, depends on a market that absorbs these externalities.
It should be noted that the gain from lower logistics costs is reflected in the price, and indirectly in the balance of externalities, with less resource consumption on the part of the firm that bears this cost.

A business network arises from the existence of an existing or potential link between a customer and a supplier.
If the company supplies a good, thanks to a commercial network, the benefit of the network is reflected in the turnover.

Commercial networks are based both on social networks (individual choices) and on innovations (choice of a supplier for the innovations it offers), proximity and costs.

In what cases can the social utility of a business network differ from the private utility? In both cases, this implies a crime of favoritism or corruption. In no case can a commercial network be a source of positive externalities. For negative externalities, the externality is calculated by assessing the impact of corruption and favoritism on society.
Only one case can be considered, and that is the power of intermediaries. However, this intermediary power must necessarily be based on an added value that this intermediary would have (volume, innovation, etc.). This brings us back to other considerations, which we deal with in other chapters.

Technological networks are born from innovations. Metcalfe’s law predicts that the usefulness of a network increases with the number of users. All innovations allowing to transport an object (material, information, …) follow this law. Thus the same rule (with the exception of one factor) applies to cars (the more cars there are, the more infrastructure there is, and the more infrastructure there is, the more cars there is), to public transport (the more passengers there are, the more buses there are, and the more buses there are, the more passengers there are), to the Internet (the more flows there are, the more infrastructure there is, and the more infrastructure there is, the more flows there are), and to social networks such as facebook, twitter, … (the more information there is on the networks, the more users there are, and the more users there are, the more information there is).

From this law, we can predict two sources of gains:

The user, with constant usage, will see the utility of the network increase (increased performance), thanks to other users using it,
The company will see the marginal cost decrease (volume effect, compared to fixed costs and investments)

From a company point of view, the company that proposes an innovation can claim to be a source of positive externality, if its technology becomes more widespread. This follows a life cycle (bell-shaped), until a new technology replaces the previous one. This externality is complex to evaluate:

How much responsibility does the company have for an individual’s innovation and talent?
What role does society play in technological innovation and its diffusion?
What is the share of the network effect of the internet in the development and therefore the network externalities of Facebook?

The evaluation of social gains for technological networks should not be limited to the evaluation of a simple social gain. Let’s take the example of a Google search.
If we first consider the social gain of a Google search without considering the value chain. This corresponds to the time spent searching in a library, on a given theme.
A small and arbitrary calculation: let’s fix the time to 1 hour of research with travel in a library, research of the cuts, …, versus an immediate result for Google. Then we can estimate the social gain (gross) at 6€ (the hour taken as a world average). Google claims in 2020, 2550 billion searches per year… that is a social gain of 15 000 billion dollars per year, about 17% of the Gross World Product.

We have a problem. Indeed, the gain from search comes at the end of the chain, and benefits from a network effect based on the posting of content online, on an Internet network, access from mobiles, …
Moreover, attributing 100% of the social gain to google is a quick reasoning: what would google be without the millions of contributors, free or paying, what would google be without the use of search tags, …

If we integrate a share (by taking the investments of google) versus the whole of the social investments that have allowed to contribute to the internet, then we fall into the opposite excess: we end up with a ridiculous share with regard to the whole of contributions, not only action to put on the internet, but the whole of the productions from Babylon to our days. Indeed, the transcription of the Pythagorean theorem on a wikipedia page, referenced on Google, must integrate the contribution of the scribe (the Wikipedia editor), but also that of Pythagoras, and of all the intermediaries who allowed the transmission of this knowledge.

We have explored both extremes. So, how to integrate the social contribution of Google. The only possibility is to integrate the research and development costs that allowed the search model to be developed, and thus to create an exception for innovations based on a network effect. To consider that the cost of this research is the effort required to obtain a search engine in the current environment.

This is not completely incoherent because without search engines, the internet would be a vast mess, in which users would quickly get tired, and would not have contributed as much. Google has created a monopoly for itself (which is also a source of negative externality in the economic field), but this has allowed a network effect, and a social benefit. We must simplify, at the risk of not coming up with anything.
The network effect as a source of externality is limited to a few actors. It must be codified, and a specific standard must be established to extract the social cost.

Transcribing the social benefit of a network effect is complex and depends on the economic model, the mode of diffusion of the innovation, the technological gap, …
It is thus necessary to rationalize the investments made (provided that the benefits are freely available (not a source of income), and to confront them with a gain (counterparty). In the case of Google, the search algorithm, made available to everyone, has made it possible to capture information and create a commercial value (monetization of traffic). We can thus estimate that the amounts invested at the beginning were invested with the hope of a monetization. In these complex cases, we must ignore the externalities involved. This alters the reading of the balance of externalities:

if we consider that a dazzling success is necessarily born thanks to the capture of externalities (monopoly, information not produced by Google, …) then the societal balance is in favor of Google (less societal debt),
if we consider that progress compensates for negative externalities (travel to the library,…), then the societal balance is in google’s disfavor (higher societal debt).

“He who only wants to act and speak rightly ends up doing nothing at all” (Nietzsche).
Thus, in this context, the integration of research and development costs in a balance of externalities for a technology that is free and available to all, seems to be the best possible compromise between integrating the notion of progress in the balance of externalities, and doing nothing: a necessary simplification with regard to other issues.

Leave a Reply Cancel Reply