Did financial risk as a discipline actually fail? I dont think so….
I have not worked in the Financial Risk industry and therefore a neutral observer. I have been reading on the build-up, causes (global and country-specific) and manifestations and want to add to the literary debate that has been overshadowed by media attention, political blame-shifting, regulator-bashing with blanket statements such as ‘they did not understand the risks they were taking’. In this post, I argue that ‘Risk Management in the Financial Industry has been no more /less effective than in other industries’. This is a contemporary theme and is being developed by economists and international experts, but I am tempted to add my few cents:
Opinion 1 – We do not need exotic financial products.
When ‘government bonds’ were first introduced and sold in Dutch secondary markets, they were perceived no differently than the Credit Default Swaps (CDS). Cautiousness and apprehension to innovation (with the ‘old guard’ being most cautious and change resistant) is understandable, correct and providentially imbibed into our genetic code.
Within decades of bond market inception, equity and commodity markets on the lines of bond markets had been functional without any studies into contagion risks between the different markets (atleast I have found no evidence of such studies!!). So, the notion that securitisation market was toxic, too risky and not fully tested is no more different than accepting that we did not test our financial innovations since 1600’s!!
To stop innovation is therefore impossible as it will just be circumvented through regulatory arbitrage or worse, create ‘too big to fail’ financial houses with significant systemic risk as seen in the cases of AIG and Lehman Brothers (both managing off the shelf /proprietary derivatives) owing to the absence of formally regulated markets. Some of you may argue that regulatory supervision on a global scale will overcome this possibility, but again this would be akin to staying with the Gold Standard which have constrained global economic development through resources, which are sparingly located geographically and more unbalanced.
My conclusion is that this opinion will always be held in the same nostalgia of a 50 year old who remarks that ‘things were better in my time’.
Opinion 2 – Financial experts created exotic instruments that they did not fully understand.
OK, here the financial risk experts are partly at fault. One of my financial risk expert friends told me that their VaR models did not model correlation between default risk of commercial property and residential property, resulting in comparatively lower capital ratios. But again, here is the point – Organisations will always seek to maximise their profits (if you argue that they should be lending, wake up!! Banks are not public sector organisations. If society wanted them that way, they should not have let politicians to mutualise /list on stock exchanges). I am actually proud that Indian regulators are old-fashioned and insist on regulating credit-deposit ratios.
Securitisation as a concept was perhaps the best concept any Risk expert would have come up with. I would love to have been the guy who developed the concept. You see, securitisation has existed since 1990’s (see the Bank of International Settlements website or ‘The Economist’ from FT). Developing a security with 70% prime, 20% sub-prime and 10% junk asset classes is a superb way of maximising returns while mitigating default risk. Before you start thinking of residential mortgage-backed securities and say, ‘aha this is what caused the problem’; wait:
- Credit card, student-debt and many other assets have been securitised in this way, although the proportion of good/ bad and ugly may change based on the investor’s attitude to risk / willingness for premium.
- It was not the securitisation process that resulted in the problem; it is the age old problem of leverage!! The same leverage that brought down LTCM (remember options pricing models from the Nobel Prize winners?)
If financial industry had stuck to traditional deposit-loan ratios of 60-70% (as in the good old days which thankfully still can be found in India!!); we would never had a problem. The problem is that western financial industries (in the goal of profit maximisation and shareholder return) attenuated by agency problems (bonus structures), created highly leveraged institutions (the Bank of England reports will show leverage falling since 2007, but significantly beyond 100%. So, even a minor default in the smallest asset group within the security would be sufficient to kill the organisation.
So it is not the securitisation market (by inference the risk management), but organisational goal of profit maximisation through leverage (borrowing from money markets on a day-to-day basis to compensate for the low-saving Americans and British) that led to the problem. If you are saying why did they, again I request you to wash your eyes and read this loud – ‘Banks are not social enterprises, they need to make money and will continue to do so’. If you don’t want that, create a public bank and stay with 60-70% loan-deposit ratio!!
Opinion 3 – Quantitative models (such as Value at Risk) did not reflect the true nature of portfolio, exposure and contagion risks.
This is based on the data range used in the stochastic models and as with Options Pricing formula, the data range for calculating correlation between defaults of different asset sectors and the sudden extrapolation in interest rates could not have been modelled as base-case. For a start, interest rates increasing from 1% to 5% within a very short span of time would not have been identified as a risk. Even if it was identified as a risk and modelled, a case of interest rate doubling would have been a reasonable assumption to make, given the long term average.
Although a scenario model could have picked up these eventualities in the form of a ‘fan chart’, it would be highly unlikely for Management decisions to be based. Even if a Risk Manager did stick his head up and say that the correlation between two asset groups should be modelled, then s/he would have been pointed to the Basel rules and how peer organisations were modelling such eventualities.
Risk managers used linear interpolated data to model exposure risk but if the data was incorrect (such as granting mortgages to NINJA – no income, no jobs, allowances) then risk models can hardly pick that up. That is a question for corporate responsibility within the supply chain. Similarly, if Senior Management chose to ignore /avoid professional opinion (as shown in the sacking of HBOS Risk Manager for highlighting the accelerated rate of exposure growth to Senior management), then an inhibition towards honest discussion is created (see my other post on Risk Management and Organisation culture under the heading of Power culture).
Risk Models therefore were well developed and the best that industry as a whole had been using, although the input data could have been better captured and used. For the problem to manifest itself in extremes were reasons of ‘leverage’ and reliance on money market funds (also called overnight lending / LIBOR rates).
Opinion 4 – If regulatory bodies had supervised the financial industry better, we would not have had a significant systemic risk.
Even with the best regulation, I wonder how many of the following points could be reasonably monitored..
- Would regulators monitor mis-selling at the point of sale? Remember the size and breadth of a single American state. As of now, we as a society have struggled to manage loan sharks and therefore harder to regulate.
- Could regulators control the use of money market funds? Doing so would significantly reduce the ‘money velocity’ (remember the M x V = P x D, where M = Monetary supply, V = Velocity at which money moves within the system, P = price of goods, D = aggregate demand). It is exactly because of negative velocity that central banks have ‘eased the money supply’. i.e., to keep the left hand side of the equation constant so that D increases. See the Wikipedia article on ‘Quantity theory of money’.
- Could regulators monitor packaging of securities & quantification into securitisation models? In theory you could. But goes back to point 1, how do you know that the securities rated ‘prime’ were actually ‘prime’. In any way, the central banks across globe ultimately ended up lending money to banks at virtually zero rates for the same securities. How do you calculate the CDO pricing for something you don’t know? This is not even forgetting that in a flexible labour economy, a ‘prime’ borrower can become a ‘sub-prime’ and vice-versa in a very short span of time…
- How can regulators set/ implement realistic levels of deposit-loan ratios without causing market punishment on specific organisations or even industries (short selling, etc)?
- With a limited resource pool to draw from, how would regulators monitor other financial transactions (insider trading, loan sharks, etc)?
- How can regulators overcome ‘moral hazard’ and competence issues? Regulatory staff would need to be located at the client location and can be ‘poached, ‘influenced or will have an affinity’ over a longer horizon.
- Without sucking significant liquidity, lending (short and long term); how can you overcome the reinstatement of a barrier similar to Glass-Stegall.
Opinion 5 – This is an American problem
Financial service industries of Iceland, UK, Japan, Germany, France and Italy are all exposed and in many cases have high leverage ratios – in similar problems. Basel accord that set consistent capital ratios and took so long to implement has proven itself to be inflexible to reacting to local problems, although one can argue that the failure of fiscal policies in some of the above-mentioned countries has placed higher emphasis on failure of Basel accord. A ‘one size fits all’ approach sacrifices flexibility for simplicity and uniformity.
Global trade balances further accentuate the issue with approximately 70% Chinese foreign reserves (formal figures are not issued by the People’s Republic) held in Dollar asset classes, but are not the problem. America can just print itself out of the problem, being the world’s reserve currency – a luxury that other countries do not possess, courtesy Bretton Woods!!
The fundamental problem is that the Western world’s innovative culture that led to industrial revolution, navigation, etc has all but stopped, especially in those sectors with ability to provide long term mass employment. Implementation of existing technology in mass employment sectors – schools, infrastructure projects that are crumbling has also reduced. Credible interpretation of Keynesian Economics has been absent (except in China) with most additional borrowing helping unproductive assets (pension black holes, repayment of debt interest, etc).
It is the absence of investment categories in productive sectors that has resulted in allocation of capital in what has been described as socially-useless asset classes. Allocation of capital to productive, long term sectors with potential for long term employment will automatically reduce default risk, reduce bank leverage, reduce fiscal deficit and provide sustainable growth. Only with entrepreneurship can employment be created, deficit reduced and sustainable fiscal headroom created (as most Economics lecturers would have been saying when we were asleep in class!!)
Financial risk management cannot be blamed for the failure of financial system. Several macro and micro-variables are contributory (savings glut from Asia, lack of innovation in mass employment sectors, fiscal indiscipline, agency problems, low savings culture in Western World because of effective welfare state, etc). Nonetheless, there are some lessons for the financial risk industry:
- highlighting scenario models and stress tests results to Board level,
- studying correlation between related asset groups (residential and commercial mortgage backed securities, for example) and then disseminating the results to wider industry through Institutes, Technical papers, Seminars, etc.
- to attract non-financial risk professionals into the industry – challenge the ‘status quo’.
- model the effects of peer default to stress the models.
- feedback loop between actual defaults and those predicted, to improve input data.
I welcome your comments………….
Posted on February 20, 2010, in Economics and tagged correlation, credit crunch, Economics, financial risk, quantity theory of money, regulation, securitisation, stochastic, systemic risk. Bookmark the permalink. Leave a comment.