|
|
One of the hallmarks of financial globalisation has been a growth in cross-border linkages (exposures) among banks. On the positive side, these linkages have been associated with new funding and investment opportunities, contributing to rapid economic growth in many countries (especially in the early part of the 2000s). But the growing financial linkages also have a “dark side”: the increased cross-border interconnectedness made it easier for disruptions in one country to be transmitted to other countries and mutate into systemic problems with global implications.
The potential harmful consequences of cross-border interconnectedness for domestic banking sector stability have been illustrated rather dramatically during the recent global financial crisis, when shocks to one country’s financial system were rapidly transmitted to many others. One of the upshots of the crisis was that considerable efforts have been devoted to better measuring “systemic importance” of jurisdictions around the world. There is a growing consensus that interconnectedness, together with size, should be a key variable in assessing the systemic importance of a jurisdiction from the viewpoint of financial stability (IMF, BIS, and FSB, 2009; IMF, 2010).
This paper aims to answer the following key question: when a country’s banking system gets more linked to the global banking network, does it become more stable, or less stable? The answer to this question is obviously relevant for policymakers and regulators in individual countries. To the extent that interlinkages help banking stability, should policies and regulations be designed to promote such cross-border interconnectedness? And to the extent that interlinkages are bad for stability, should policies and regulations aim to stop or, at least limit, the growth of such interlinkages?
IMF examines the above key question in two ways: First, they analyse it conceptually, using simulations. Second, they examine it empirically, based on a range of econometric approaches—parametric as well as non-parametric—that combine data on banking crises around the world with a comprehensive data set on cross-border banking linkages.
To preview the main results, their short answer to the above question is: it depends on the degree of interconnectedness. The relationship between the likelihood of a banking crisis in a country and the degree of integration of that country’s banking sector into the global banking network is far from trivial. IMF finds that in a country whose banking sector has relatively few linkages to other banking sectors, increased cross-border linkages tend to improve that system’s stability, controlling for other factors. In other words, within a certain range, connections serve as a shock-absorber. The system acts as a mutual insurance device with disturbances dispersed and dissipated. Connectivity engenders robustness. Risk-sharing – diversification – prevails. But at some point—which IMF estimates to be at about the 95th percentile of the distribution of countries in terms of interconnectedness—increases in crossborder links begin to have detrimental effects on domestic banking sector stability. At a yet higher point, when a country’s network of interlinkages becomes almost complete, the probability of a crisis goes down again.
One of the novel insights of the IMF paper is that it is important to distinguish whether the crossborder interlinkages are stemming primarily from banks’ asset side or from their liabilities side. IMF introduces measures that distinguish those two types of interconnectedness (which they call “downstream” and “upstream” interconnectedness), and find that the impact of changes in interconnectedness on banking system fragility are more significant for liabilities-side (“upstream”) interconnectedness than for asset-side (“downstream”) interconnectedness.
The paper is structured as follows. Section II provides a discussion of the relationship between financial interconnectedness and instability, positioning the paper as a bridge between the network analysis literature and the banking crisis literature. Section III presents the data, in particular the measures of interconnectedness and the measures of financial instability. Section IV provides a simulation exercise. Section V describes the estimation procedures and presents the results of both parametric and non-parametric estimates. Section VI concludes.