James Glattfelder aims to create a data-driven understanding of the interactions that comprise our global economy. The study
The Network of Global Corporate Control, which he co-authored, highlighted vulnerabilities in the economy that are due to how power and control flow through our financial systems. He will be speaking at Wired Money on 1 July.
Wired.co.uk: You describe yourself as a complexity scientist. What does that mean?
James Glattfelder: I was working in a small quantitative hedge fund in Zurich, where we tried to build trading algorithms for the foreign exchange market. My background is physics and what I realised studying physics was that while there is so much we can explain in the world with equations and technology -- from making smartphones to putting rovers on Mars -- somehow there are things all around us we can't really grasp with equations. For example, we can't really understand swarms of birds, or ants, or markets.
That's why I got interested in complexity science, as it's loosely called. I was quite amazed that often when you look at complex behaviour from far away [it seems very complicated], but when you go close up, it's the result of a few simple rules of interaction. This means that there is a way to understand complexity -- you can start to map it in computers and simulate things. It's based on this very simple idea of interaction, in which you get agent-based modelling and the whole field around it.
So when did the switch come about from working at a hedge fund to co-authoring peer-reviewed papers?
I was loosely interested in this [complexity science] through my work at the hedge fund and thought I'd like to know more about this. Alongside my work I went up to the technical university here and started a dissertation and basically the study which came out of that [The Network of Global Corporate Control] was the result of one one of the papers I did there.
There was this amazing set of coincidences that made it happen.
First all, when I started, I was a PhD student and I was just generically interested in complex systems. I just wanted to know more about how they work, but didn't have any preference to what domain my study should be in. Whatever: biology, physics, computer science, economics. But the professor had just bought this huge global data set, so I arrived there ready to start to start on my PhD and they said 'OK, James you work on this economic data.' So when people ask why we studied these things, and how we knew people would be interested, we didn't. It's total coincidence.
The paper showed that around 0.01 percent of the shareholders studied controlled around 80 percent of global companies. We're all aware of inequality, so why did your study get so much attention?
When we did the first study which compared 48 countries, and this got picked up a bit by the press and people discussed it. So we realised that there was an interest in this, but never realised it would become such a big thing. Then we published the global study, so we went from a cross country analysis to a global analysis. In the exact week that Occupy Wall Street movement went from being a US thing to being a global thing, the New Scientist prominently placed our study on their front page saying "the math behind the protest".
The funny thing was that if the article had come out a week or two earlier or later, I'm sure this kind of viral effect wouldn't have happened. This study came out and New Scientist noticed and for three weeks the telephone was ringing and emails were going crazy, it was a real viral phenomenon.
Can you explain what the findings were?
What we looked at was this global database of ownership data, with something like 13 million entries in it. The reason we looked at ownership data is that it's actually very hard to get information in a financial context because a lot of it is undisclosed, but ownership is actually published. The idea is that it does give you a snapshot of an organisational thing going on in the economy.
Ownership gives the shareholder voting rights, so if you have more than 50 percent in a company, you have a lot to say as a shareholder.
So we tried to take this idea and formalise it and build a methodology out of it, tried to figure out who are the most important shareholders based on how many shares they have in what companies and how big those companies are -- which we looked at in terms of operating revenue. And then we looked at the network effect. So, what companies do these companies have shares in, and so on, all the way down stream. And this is something that had never been done on a global scale. Some people had done similar things in a national context, but never in a global context.
But regarding our results, we found that the distribution of power was an order of magnitude, more concentrated and more uneven than the distribution of income or wealth. We expected it to be skewed but at this level it was quite unprecedented. You know, 140 companies having the potential to control 40 percent of the network, in a network of 600,000 nodes. That's quite striking. The second thing that is important and that is coming up on the radar of regulators is this idea of systemic risk, where it's not about individual entities but about interconnection. The idea of 'too big to fail' is basically very naive, because these entities are embedded in a network of their dependencies.
Can you explain that further?
What we found was a small core of powerful shareholders, or power players, in the network. If you look who's in that core, it's all these top players and they're highly interconnected with each other, so they all own shares in each other. This is something that shouldn't be there if you look at standard or classical economic theory. Some people naively think that a high degree of interconnectivity is good, because you're diversifying and sharing the risk -- this is actually true but only up to certain thresholds of interconnectivity. Again, you can only find this out if you simulate the system, but at one point it peaks and then it goes into a state that if one of the nodes experiences distress, the distress spreads through the system like an epidemic. And this is something you don't see from the balance sheet -- you have to do network analysis.
The bottom line for us is that we wanted to raise these questions. As physicists we're not really trained with the whole implications and philosophy of economics. What is hoped is that we open the door for other people to pick this up and do new studies into whether this is really true and the implications. And just have people think more in a networked way in economics, in that everything depends of everything else.
Is the problem that the system we have in place was never designed?
That's one of the key issues in complexity. You have this phenomenon of emergence in that suddenly the system displays characteristics and organisational structures that are not visible from the interaction of agents. Suddenly it just lifts itself to this new intelligence level and has features that no one put in there, they just come out of the complexity of the system. I think this is one of the important perspectives which I don't hear about at all. We see extreme global inequality as problematic. But this distribution of inequality is something that pops up everywhere in nature and seems to be somehow important for the stability of complex systems. And it's only in the context of humans that we think 'OK, this isn't something we want because this is very undemocratic.' So the perspective that the economy is how it is due to conspiracy theories, or because people are greedy is wrong: this is the totally natural, normal distribution you would expect in any normal complex system.
So, if this is a universal law of nature and we agree that we need to change and combat it, then we have to do this from a totally new perspective. People don't say this very often, but basically we don't know how these systems work. We created them, but really they have a life of their own. I think that's something that should change in the philosophy of how we deal with these problems. More people should say 'we don't know the right thing to do, we've got a few clues let's try a trial and error idea', rather than coming from an ideology where each political party says 'look, we know exactly what we need to do'. These entrenched ideologies are just problematic.
James Glattfelder will be appearing at Wired Money, on 1 July, 2014. Tickets are on sale now: see wired.co.uk/money14 for a full speaker list and further information. Wired subscribers receive a 10 percent discount.
This article was originally published by WIRED UK
