Absorbed into the networks behind the systems we see

Its been a while since I have last posted here. The reason for my absence is two-fold.

Firstly, I am busy with a course offered by Coursera and the University of Michigan about Social Network Analysis (SNA). My business partners and one of our associates in Mesopartner are participating in this course. The course is 9 weeks long and I must admit that it is taking much more of my time than I originally anticipated.

The second reason I am hardly online is that the industrial policy in South Africa is starting to have positive effects on local industry. As I work mainly with the manufacturing sector on topics like innovation systems, industrialization, identifying and addressing market failures, and the competitiveness of regions, it means that there is suddenly an upsurge in demand. The demand is lead by state owned companies that are suddenly obliged to procure manufactured content locally, and by local industries that realize that years of underinvestment and fighting to survive against cheap and sometimes lower quality goods have left many sub-sector uncompetitive.

But these two reasons are also having an effect on each other. I have been applying many of the principles and tools of Social Network Analysis in my diagnostic work for the last 2 years, and for the last year I have been using SNA as my main diagnosis instrument. This recent course have simply forced me to read up more and more on many of the theories and the concepts behind the instruments I have been using. I am still trying to figure out how to do this kind of diagnosis fast, and how to teach these instruments and theories to the practitioners that we (Mesopartner) are working with around the world. At this moment the diagnosis that I am doing in valve, pump, tooling, automotive and industrial equipment is still slow and it takes all my attention.

What is the benefit of taking a SNA approach to sub-sector development?

  1. Well, firstly, a network diagnostic very quickly reveals whether there is a cluster or even a value chain. We often assume that these constructs are real, but in the last few years we have learned that just because all the actors that should be in a chain are there doesn’t mean that a value chain exists. Same goes for a cluster, just because all the elements are there doesn’t mean there is a dense network of cooperation, knowledge exchange and systemic competitiveness.
  2. Secondly, a network view assists with understanding the deeper relationships, trust patterns and information flows in a small part of a real system. These relationships makes it possible to predict how information flows, who the thought leaders are and how influential institutions, leaders, officials and business people are. This is directly relevant for my work with innovation systems.
  3. Lastly, Social Network Analysis also highlights how complex even a single link in a value chain can be. When you look at the spider web of relations, ownership structures, communication channels and knowledge spillovers, then you see how traditional development interventions have completely missed the leverage points.

All I can do at this moment is to commit to blog more frequently once this course is done. I will share some of the results of the industrial diagnosis that I am currently busy with in a few weeks time. Below I will give a sneak preview of the network map of the valve manufacturing cluster in South Africa. You will immediately see that some manufacturers (in red) and some foundries (in blue) are more connected than others. The yellow dots are valve manufacturers that are not yet part of the formal valve cluster structure. Hardly any additional analysis is needed to show that the more connected firms are the ones we should work with.

Cluster drawing 4

However, the additional analysis that we can run on this cluster further narrows the choices of whom to work with to get both the highest impact (in terms of both ability to grow their business, increase employment and meet customers needs) and in terms of getting the highest demonstration and spill over effects. The latter is important, because when you want to upgrade an industry you should prioritize firms that are able to create positive spillovers and that others are willing to follow. To do this kind of analysis we need a combination of qualitative and quantitative information, and we use specialized software applications. But more about this in a future post!

%d bloggers like this: