The behaviour of banks during the crisis and why another bank crisis is unavoidable

Several banks were bailed out as a result of the recent crisis. But, how did banks behave during the crisis - did banks reduce risk as the financial crisis unfolded, or did they do the exact opposite? This is an interesting question - and the key to understanding why future bank failures will happen again. 
Do banks reduce or increase risk as a response to a financial crisis? This question has been analyzed by many academics and has been the topic of many books. But, most of the thinking behind this topic can be understood by studying this example: 

During the financial crisis, a lot of the big European banks were already heavily invested in the mother of all carry trades. The mother of all carry trades? Sounds funny, but is essentially when banks were buying Euro periphery bonds during the crisis and - at the same time - most likely sell (or short) German government bonds. The bet behind this trade is that the yield spread between Euro periphery bonds and German government bonds will fall over time - so, essentially betting on the survival of the Euro zone. 

So, what happened during the crisis? Well, banks increased their bet on this trade consistently through the crisis. And the most important thing to realize is that banks that were already in trouble - banks with low capital ratios - were the banks that increased this bet the most. 

The reason why banks in general are happy about this trade is the belief that politicians will not allow the Euro zone to break up. Ok, so far so good. But there is another (far more important) reason why this risky trade was so popular with banks. The reason is regulatory arbitrage. 

Bank regulators demand that banks use a risk weight on all their investments. In theory, a risky investment will have a high risk weight and a less risky investment means a lower risk weight. A bank that is known to invest in risky assets will have to issue more capital (equity) to cover their higher level of risk weighted assets. A bank that is more conservative and invest in less risky assets, have less risk weighted assets and do not have to issue as much capital (equity) as the more risky bank. 

Back to regulatory arbitrage. The thing is - an investment in German government bonds and an investment in Spanish, Greek and Italian government bonds all carry the same risk weight. That risk weight is zero! Seriously, the regulators will have us believe that an investment in Spanish and Greek government bonds carry the same risk as an investment in German government bonds. That is absolutely ridiculous.

This means that European banks are incentivized - by the regulators (the government) - to buy peripheral bonds and at the same time the banks don´t even have to issue equity to cover the risk profile of the investment. This is obviously a regulatory arbitrage opportunity and the reason why this carry trade has been so popular with banks during the crisis. 

A lot of European banks increased their bet on the carry trade during the crisis because of regulatory arbitrage. So, the bottom line is: Regulators (the government) actually incentivized banks to take on more risk at the wrong time. You can read more about this in a recent paper from Stern School of Business Professor Viral Acharya called The Greatest Carry Trade Ever? Understanding Eurozone Bank Risks.

It gets even more perverse than that - the regulatory requirements are designed such that banks are incentivized to keep the capital ratios at a ridiculously low level, overall. The new Basel III accord fails to address the basic problem and bank´s equity can be as low as 3% of their total assets. That means that some banks - essentially - can become insolvent when assets drop by no more than 3%. Only 3%!

The point is, we will have a future new bank crisis - guaranteed. And, the reason is that bank regulators incentivize banks in the wrong direction. 



Below you will find some comments on the yield spread between Italian and German government bonds. 

As you can see in the graph below, the spread between the 10 year Italian bond yield and the 10 year German government bond has been on a roller coaster during the crisis. It seems that the Italian yield is now converging very nicely towards the "pre crisis level".

Picture from Bloomberg. 

The graph in the top left shows the yield on both German and Italian 10 year government bonds all the way back from January 2005 until last week. The graph in the lower left shows the difference, or the spread. The spread was peaking during the government debt crisis of 2011 and 2012. 

The question back then was if the EU would survive the crisis or not. It was not until the summer of 2012 when ECB President Mario Draghi assured the market that he would do "anything to save the eurozone" that spreads began to drop.

So, at what level will the spread settle? It is not easy to tell, but it looks like there is a little more convergence left. To argue that the spread has to come all the way back to its pre crisis level of around 10 - 20 bp. is probably not realistic.  A break up scenario is not out of the question going down the road - it´s not immediate, but it´s certainly something that can happen in the future, which is why the spread level cannot go all the way back to the pre crisis level. In the pre crisis era, a break up of the EU was all but academic and something that traders, analysts and economists were not considering seriously.