Systemic Risk at Baruch College
February 11, 2012
Baruch College hosted the Capco-sponsored "Institute Paper Series in Applied Finace" on Thursday. I assume this is a further follow-up event to the one they did at NYU Poly last year (see some notes here). I have put some notes together below, my apologies in advance to the speakers for any innaccuracies or ommissions in putting my thoughts together:
Systemic Risk Presentation
First part of the day started with a presentation by Viral V. Acharya of Stern on systemic risk. I have always found systemic risk an interesting topic, given the puzzle of how do you dis-incentivise an organisation from increasing risks in the wider financial system when the organisation itself will not directly (or wholey) face the consequences of this "external" risk increase.
Viral started his presentation with some great jokey graphics, one of a the HQ of a bank going up in flames with fireman hosing the flames with banknotes not water. He mentioned the definition of systemic risk given by Daniel Tarullo, Governor of the Federal Reserve (I couldn't find the definition, but primer paper here). He asked how Lehman was allowed to fail when the likes of Fannie Mae, Freddie Mac, AIG, Merrills, CitiGroup, Morgan Stanley, Goldman Sachs, Washington Mutual and Wachovia were not and offered assistance in one way or another. He said there was not enough capital in the system to stop Lehmans failure but that he saw Lehmans as the catalyst for the recapitalisation of the American banking system, not the cause. He later implied that Europe had so far lacked such a catalyst for action in the European banking system.
Viral said that he wanted to put forward an ex-ante regulation that would force a bank to retain additional capital to account for the systemic risk it produced. He said that the banking system was obviously much safer than it had been a few years back, but suggested that whilst the system could now withstand say the failure of a large organisation such as Citigroup, in his opinion it would struggle to survive the failure of Citigroup and a Euro default happening at the same time. Viral said that the current Dodd-Frank regulation on systemic risk was not a healthy one in that if a large institution fails, banks of capitalisation of over $50B are jointly taxed to assist in the consequences of the failure. Viral viewed this as a big dis-incentive against a healthy bank (say a JPM) from stepping in to purchase the failing institution before the failure, as JPM would know that it would be taxed anyway on the bailout.
In Viral's model, he defined a crisis as a 40% market correction, and assumed that non-equity liabilities repayed at face value in such a crisis. Given there is not much real data around for a 40% correction, he used data obtained from 2% correction events observed, then extrapolated from the 2% to the 40% level. He said that the question that needed to be asked was whether in such a crisis scenario that a bank like JPM would retain 8% capital. He emphasis that the level of capital chosen was somewhat arbitary but rather more importantly were the assumptions in the model of crisis, since the capital models used in regulation today are based on average losses not crisis-level losses. Using this and related models, Viral showed that the banks exhibiting the most systemic risk were Bank of America, JPM and the Citigroup (for more background and a complete list see Stern's V-Lab ).
Viral said the restructuring of Dexia (exposed heavily to peripheral sovereign debt) was the "Bear Stearns of Europe" (exposed heavily to peripheral MBS), but that is restructuring was not large enough to cause a more widespread re-capitalisation of the European banking system. Dexia was ranked as one of the safest banks in the Europe-wide stress tests of 2011, given that the Basel risk weightings did not apply any haircut to European sovereign debt. This was another critiscism that Viral levelled at Basel in that the risk weightings are static and do not reflect changes in market conditions.
Viral then joined a panel debate on systemic risk chaird by Linda Allen of Baruch, joined by Jan Cave of FDIC, Sean Culbert of Capco, Gary Gluck of Credit Suisse and Craig Lewis of the SEC.I have tried to bring out some of the main themes/points of the discussions below:
– The Balance Between Risk to the System and Risk to the Economy
There was a lot of debate on the secondary effects of regulating systemic risk and increasing capital charges on banks, and its wider effect on the general economy. Craig put forward the argument that too high capital requirements would stifle lending and in turn stifle the wider economy (arguably the "bigger" systemic risk maybe?). He argued for a balance to be found and that the aim should not be to eliminate risk in the system completely. I guess Craig was taking the banker's view, but the rest of the panel seemed to agree that the point was a valid one.
– Basel III
All agreed that Basel III was an improvement but there was still much more to be done. Gary was critical of Basel III calculation remaining too static, but Jason described how Basel III had removed many debt-like assets from the capital calculation which was good however. Jason also described how Basel I had been a simple framework (and good for that) but was tinkered with with VAR encouraging assets to be moved to trading book to reduce capital charges. Basel II then introduced the Internal Model approach and over ten years capital requirements were continued to be lowered, with CDO's attracting a 56bp capital charge during this time down from 8%. Enforcement of Basel III on both liquidity risk and capital was considered as key for coming years.
– Liquidity Risk
There was general consensus that pre-2007 liquidity risk was not talked about enough and there were no standard ways of calculating its level. Jason said that pre-2007 the regulators had not modelled what happens when the counterparties start running. Gary said that he questioned whether some of the current calibrations of liquidity risk were correct.
Sean raised the point that Volcker was likely to impact market-makers and hence impact liquidity (see earlier post on this).
Sean also mentioned that Rehypothecation of Assets has not been debated enough and had only received scant attention in Dodd-Frank (maybe see recent article on Thomson-Reuters on MF Global)
– Europe (and more Basel)
General consensus that Basel III capital requirements will constrain GDP growth in Europe. Viral seemed to have the strongest views here, saying the Europe needed a bank recapitalisation program just as the US had gone through, and that such a program would be a big boost to economic confidence. Viral remains deeply sceptical on the success of Basel III – for example all of the 2007 failures were supposedly from well capitalised insitutions under Basel I and II. Viral says that the problem is not the level of capital (8% or 12% etc) but the method of modelling the shock. A good point from Gary I thought was his premise that politics in relation to sovereign debt was playing its part in undermining the calculations and approach of Basel III.
– Too Big to Fail?
One audience question was "is too big to fail simply too big?" and should the largest organisations be broken up into more manageable parts. Viral answered that he was not in favour of a size constraint and cited that some large institutions, notably JPM, Rabobank and HSBC had been relatively robust successful during the recent crisis. He did however qualify this response by saying that he was in favour of a size constraint if the large size reached was due to implicit banking guarantees from the government, and that he would like failing large banks to be broken up into smaller pieces.