printlogo
ETH Zuerich - Homepage
print   

Paper: ETH-RC-14-011

Title: How Damage Diversification Can Reduce Systemic Risk

Authors: Rebekka Burkholz*, Antonios Garas, Frank Schweitzer


Abstract:

We consider the problem of risk diversification in complex financial networks, where nodes are financial actors and weighted links represent financial obligations (credits/debts). Whereas in classical financial theory diversification generally is considered a rational strategy, from a complex networks perspective diversification implies exposure to possible failures of other actors and hence increased individual risk. This insight has sparked a multitude of recent works on the issues of risk and cascading failure, taking a complex networks perspective. In the past results about risk and failure in financial networks mainly centered around scenarios where investments are diversified (i), while the second strategy of diversification in credit networks has been neglected: the diversification of debts (ii). From a mathematical perspective this is mainly due to the fact that the analytical treatment of case (i) is based on a heterogeneous mean-field approximation, which does not take into account that different network neighbours of a node can have different impact on that node -- a feature that is critical for the treatment of case (ii). Here we present a framework that allows the analytical treatment of combined investment and debt diversification in credit networks. We present the models, the analytical solutions for failure risk as well as stylized facts about the impact of investment and debt diversification strategies on the system. Most importantly we find that diversification is a double-edged sword: whereas nodes that are highly diversified in their debts or investments are crucial for the stability of the system as a whole, they also have a high individual failure probability. Furthermore, we show how the developed analytical treatment can be generalized to weighted networks of (almost) arbitrary network topologies and link weight distributions.


Keywords: cascading processes, diversification of risk, configuration model

Manuscript status: Preprint

JEL codes:
PACS numbers:



Local copy of the paper: ETH-RC-14-011.pdf


Submission date: 13-12-2014


Wichtiger Hinweis:
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
folgender Seite.

Important Note:
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.
More information