The case for a macro prudential framework

By: Casper de Vries, Erasmus University Rotterdam

The world is awash with cash. This has created a surge in asset prices around the globe. It has not been translated in inflation thanks to the production glut of the new Asian tigers. The global imbalances have created a benign economic environment of high growth. The globalization also made economies much more interdependent. At the first blinking of the Chinese stock markets this week, world wide asset markets all corrected in similar fashion. How can we understand such dramatic re-pricing in unison?

 

Consider the cross-plot of daily stock returns of the ING Bank against those of the ABN/AMRO bank in Figure 1. To interpret this cross-plot, Figure 2 provides a remake of this plot on basis of the bivariate normal distribution, with the same means, variances and correlation. The normal distribution is the standard fare in finance and risk management.

 

 

 

Two features of the first cross-plot jump into the eye. There are far more outliers in the actual returns than in the simulated normally distributed returns. Over the twelve years of this sample, daily losses and gains in the order of 8% or larger were not uncommon, while such returns are almost absent from the normal based plot. This is known in the market as the fat tail property of financial return distributions. The normal density declines exponentially fast, but in reality this goes at a power rate, yielding many more `outliers’. Economists often ignore outliers, but for risk management these realizations are essential. It explains the focus on the Value-at-Risk downside risk measure in the Basel I and II accords.

 

 

 

The other, perhaps more worrisome feature, is that the extreme realizations almost invariably occur together. Systemic risk appears to be much higher than in the fancy `normal’ world. The interdependence of the banks’ returns in the extremes is due to the network externalities of the banking sector with its numerous linkages. Banks’ portfolios are connected due to their exposures in the inter-bank deposit market, syndicated loans, and macro risk drivers such as the interest rate and hence monetary policy of the central bank. One shows that these linkages in combination with fat tailed risk drivers generate outliers which mainly occur jointly, i.e. along the diagonal in Figure1; whereas this cannot happen in a normal world. In case of the two Dutch banks, the interdependency is even stronger due to their cross holding of shares. The Dutch coziness is currently under fire of a hedge fund, threatening to break up inefficient ABNAMRO, much to the chagrin of the DNB. But perhaps supervisors should think twice before opposing such a break-up, as it might reduce the systemic risk by reducing the linkages in the sector. Not to be overlooked, it might also enhance competition.

 

Diversification in the banking sector is often considered a common good as it smears the risks across multiple parties. The Basle accords do stimulate this attitude as the focus is on the risk of the individual bank. Consider the hypothetical world in which all banks hold equal stakes in all exposures. This would safeguard individual banks from failures due to small losses on some exposures. If these loss making exposures were instead concentrated in a few banks, these might go under. On the flip side, a banking sector with proportionate stakes would be exposed to huge systemic risks: Any circumstance which would make one bank go under, would make all others fail. This thought experiment shows that the supervision of the banking sector should be much more focused on the sector as a whole, rather than proceeding on a per bank basis. The two figures, moreover, show that even without such an equal proportions banking sector, the current linkages are already producing much higher interdependency than in the normal world which underlies the Basel approach to supervision. A systemic approach would also be more in line with the original motive for overregulation in this sector, given that the banking sector due to its clearing and settlement functions is so important for the entire real economy.  For example, research has shown that the EU banking system as a whole is still more systemically stable than the US system, thanks to the fragmentation on a nation by nation basis.

 

Interconnectedness of bank returns manifests itself both through the micro-financial channels, but also through the macro policy channels. The lax monetary policy stances around the globe in combination with financial innovation, such as the gearing banks attain by their funding of hedge funds, has increased the leverage of the banking sector as a whole. This has made the banking sector increasingly dependent on prudential monetary macro policies to prevent systemic breakdowns. Sharp corrections such as seen in China this week can be understood in this way. New supervisory, regulatory and macro measures should therefore be introduced only gradually, to prevent sudden shocks. Moreover, the Chinese jitters reverberated through world stock markets, as these have become increasingly interlinked. In analogy with the Figure 1, extreme outcomes in different stock markets thus do occur together. Macro diversification is inherently difficult. World monetary policy has been too lax for too long. The liquidity overflow, though, should only be put back in the box through a gradual tightening of the money markets, so as not to add to current uncertainty. In summary, both from a micro-financial based perspective as from a macro point of view, it is important to shift the focus of current individual bank based supervision towards a macro-prudential framework.

 

I am grateful to Oleg Tschernizki in helping me with the first draft.

 

---