Please activate JavaScript!
Please install Adobe Flash Player, click here for download

Oec. Magazin 2

mation, but its heterogeneity. «In the European Union, there are about 30 big quoted banks. That’s not big data. Add other significant players elsewhere, and it starts to get more challenging, but it’s still not big data. But once you inclu- de various models of potential distress scenarios and add some scope for error in the numbers (to create what physicists call an ensemble of scenar- ios), then it’s big data.» Digging for data Volume and heterogeneity are not his only hur- dles. Much of the data Battiston craves is secret. Banks’ trading positions, for example, are on public proprietary information. Even central banks and regulators, while enjoying greater access, don’t see the full picture. Regulators are now taking greater steps to access more data: the EU, for example, is com- piling a partial register of relevant banks’ acti- vities. But a big part of the struggle is creating reliable «synthetic» data for modelling. «Esti- mating systemic risks is difficult, but not im- possible. You need the data, or at least reliable synthetic data. But it is possible to construct models», he says. The research, funded by Swiss government and EU money, is not sold commercially, but disseminated through academic papers and computer code. «This is all open source, public service work.» But it is clearly of considerable value, considering institutions like the European Central Bank and the Bank of Italy are using it. Battiston stresses the research is not intended to predict the next crisis or tell regulators what to do. «We are providing tools. The regulators and central bankers are making the decision, because decisions require political considera- tions. We are not politicians, but scientists. Our goal is to build a framework that can enrich the toolbox used by national banks and the ECB in stress testing techniques, by incorporating net- work effects that can be very substantial.» But he is gratified by the recognition. «The ECB has been using our methodology on the data as an example of real time monitoring of systemic risks. The Bank of Italy too.» «The University of Zurich is already a very well perceived institution in economics and fi- nance in Europe. So it’s natural we should be dealing with an area, like systemic risk in finan- cial networks, that’s gaining so much traction and interest. It’s very stimulating intellectual- ly. But even more important, it’s addressing an urgent and compelling issue for society. What we’re doing here is at the frontier of research in networks in finance. I’m honoured we’re being recognised for this.» Oec. Dezember 2014 19 Stefano Battiston is SNF Professor at the Department of Banking and Finance. His work applies the complex networks approach both to the empirical analysis of large economic networks and the modelling of their dynamics. Oec. Dezember 201419