Economic network structure refers to the overall makeup or topology of an economic system’s network of relations. Many parameters can be identified as affecting this structure such as the overall degree of connectivity, the number of nodes, the average path length, the systems diameter etc. Primary among these parameters is network density. Network density describes the portion of the potential connections in a network that are actually connected.
The density of connections within a system is a key metric in understanding its behavior and overall make-up in that it tells us how connected the system is. Going from low connectivity to high connectivity involves a regime shift within the system. One way of understanding this process of change is through what is called unbundling. When we get a reduction in transaction costs, a homogenous monolithic system can be broken up into small parts that are distributed out and then re-coordinated through this network of connections. This is a very deep and fundamental transformation to the system’s entire architecture. But during the process, it goes from being a bounded system to becoming a more complex distributed network.
For example, up until a few hundred years ago, universities used to teach each individual almost all the technical knowledge that we had within just a few years. Students would then go out and apply this knowledge in some working environment. With the rise of modern science and engineering, this body of knowledge has become far too large and complex for any single person to learn. So today we do not go on trying to teach everything to everyone. At some stage, universities started to break all this knowledge up, that is to say, unbundle it into different departments. Students now learn within specific domains of expertise. They then go out into the workplace and join businesses that function as the mechanisms for re-integrating all this different knowledge and applying it to some set of tasks.
With the rise of information technology, the reduction in trade tariffs and proliferation of the free market ideology, this unbundling process is essentially what has happened to our economies over the past few decades. Factories, production processes, services, national infrastructure systems, and even whole national economies, have through privatization and globalization become unbundled. Components are no longer bound within centralized structures but are increasingly distributed out and coordinated through multinational corporations, global markets, and the Internet. This is only set to continue as the Internet becomes an ever more important platform for production and exchange. We can outsource tasks and bring together teams from the other side of the planet in a few clicks. Today small businesses can build and orchestrate global supply networks online. This unbundling process is an important mechanism for dealing with a high level of complexity that monolithic systems are not designed for. It is central to dealing with the kind of volatile, uncertain and complex environments that large organizations find themselves within in the 21st century. This unbundling process is part of the significance of networks within post-industrial economies.
Networks are a very informal system of organization, which is in contrast to our more traditional formal organizations like hierarchies whose structure has to be predefined, and then the components fit into it. For example, we have different job roles in our business and employees have to fit into them. In these formal organizations, we would typically define the whole structure to the business before it starts operating. And once defined, it is temporarily closed; meaning anyone can not just walk in off the street and take up a job in the business. We have to first decide that we need a new position, define how this position will relate to all the other positions, and then go through the process of finding and onboarding a new employee. This is the same for all of our Industrial Age economic institutions and systems of organization; from production processes to governments to banks. They are all well-defined formal closed systems with a top-down chain of command that are optimized for relatively static and stable environments.
Although this formal design may sometimes be the case with networks such as corporate IT networks, networks are typically not like this; they are typically informal. They are often user-generated, without a predefined formal structure. People, businesses, traders and other institutions just make connections whenever the marginal benefit is greater than the marginal cost. If we give people mobile phones, they will start talking to their friends. If we reduce trade barriers, companies will start trading across borders. If financial institutions can invest in high-growth emerging markets with good returns, then they will create these financial connections. This is the nature of networks. They just grow in an organic fashion wherever the transaction costs are low enough and payoffs are high enough. No one planned or designed or even really managed the networks that run our global economy on its many different levels.
The density of connections within a network is rarely evenly distributed out. Whether we are talking about a social network or transportation network, some parts of the network will face higher transaction costs and will have a lower density of connections compared to other subsystems that have a higher density of connections. For example, the transportation network of Nepal has a low density due to high costs of making transport connections within the Himalayas, while other parts of the global logistic network like the Netherlands will be densely interconnected. This will give us a heterogeneous topology to the network that will significantly influence how something will flow across it and how quickly or slowly resources in the global network will permeate into any specific subset.
The key to modeling all of this is the idea of network clustering, which is an important factor in the overall topology to a network. When an agent comes to make a connection the cost and benefits of connecting to different nodes will typically not be evenly distributed. For example, because of the regulatory framework, trading within Europe is easier between entities inside the EU than trading with a country outside of Europe. Due to this, the European economy forms a cluster within the global economic network, where the nodes in this sub- network interact more often and more intensely between each other than with other nodes outside of this sub-system and this is an example of cluster. This clustering will again give the network a heterogeneous topology. And it is this topology that influences the flow of resources across the network; which is really the big question that we are interested in because it is going to define who gets what, that is to say, the macro resource distribution pattern. The wealth is in the network, every component slice of the pie is in their access to the network.
This is a very different vision to resource allocation. In our traditional paradigm, resource allocation is all about equilibrium, the efficient market hypothesis. Markets reach an equilibrium and clear. From the network perspective, an equilibrium where all prices across the market equalize would be a perfectly homogeneous network, which might exist in very simple environments. But heterogeneity across the network is much more likely, particularly when we use multiplex network models to capture how markets can be affected by heterogeneity within other socio-cultural networks.
Going back to our previous example of the European economy, when the Union was economically-integrated, people thought that the higher wages in Germany would reach equilibrium with low wages in the southern European countries. This has not happened though. The network has remained heterogeneous most likely due to its interaction with other socio-political, cultural and spatial clusters within the multigraph. This is a good example of the kind of complex multi-dimensional analysis that needs to be done in order to get a real picture of what complex economic systems actually look like. Saying that things will equilibrate is nice in theory and may work in practice if we are dealing with very simple systems, but often the reality of these global systems is one of a much messier complex heterogeneous network.
Lastly, as somewhat of a side note, we will discuss informal economic systems. In economic analysis, we typically spend a lot of time focused on formal economic systems such as large corporations or national economics. What typically goes relatively unnoticed is the informal economy. For example in India, 84 percent of employment comes from the informal economy. There are 11.7 million undocumented immigrants in the USA all working in some capacity in the informal economy. AirBnB and car sharing services like Uber form ever larger informal economies. All of these are still very much outside of our traditional systems of formal economic organization. These informal network structures have developed in an organic fashion, wherever there was a need for an economic function to be performed that lay outside of traditional institutions, whether that be Mexico cartels supplying drugs to the US market or cheap counterfeit goods to the Chinese. Network theory is a very appropriate tool to help us reason about these informal systems, how they develop, grow and interact with formal economic systems. And as we get this unbundling process of formal organizations that we see with the rise of networked organizations, the boundary between these two will likely become somewhat blurred.
1. Introduction to Social Network Methods: Chapter 8: More Properties of Networks and Actors. (2017). Faculty.ucr.edu. Retrieved 10 June 2017, from http://www.faculty.ucr.edu/~hanneman/nettext/C8_Embedding.html 2. The great unbundling. (2007). The Economist. Retrieved 10 June 2017, from http://www.economist.com/node/8559758 3. (2017). Mitpress.mit.edu. Retrieved 10 June 2017, from https://mitpress.mit.edu/sites/default/files/titles/content/alife14/978-0-262-32621-6-ch119.pdf 4. (2017). Samoa.santafe.edu. Retrieved 10 June 2017, from http://samoa.santafe.edu/media/workingpapers/09-09-038.pdf 5. Identity Economics. (2017). Google Books. Retrieved 10 June 2017, from https://goo.gl/OhxEl6