Showing posts with label Systemic Risk. Show all posts
Showing posts with label Systemic Risk. Show all posts

Dec 28, 2009

Control Leverage

Key question is whether 'adding more control' will stabilize financial institutions like banks, insurance companies or pension funds.....

With all the - apparently failing - new legislation of the last decade already in place and new control measures like Solvency II and the strengthening of the Basel II Framework ahead, one might - at least - question whether we're on the right track with this intensified 'control approach'.

Will adding more control
empower or paralyze financial institutions?



In other words: Is the Control Leverage Effect positive or negative?

Insurance
In an FT-Adviser article called 'Solvency II costs are unsustainable', Joy Dunbar reports that the ABI (Association of British Insurers) has warned that the costs of implementing Solvency II regulations could destabilize the industry across Europe.

To gain more control (financial stability), European Insurers are obliged to implement Solvency II measures by the end of 2012, starting already in 2010.

Impact Solvency II
The increasing control costs and capital demands of Solvency II will have an enormous impact om the insurance market:
  • Recapitalization: Insurers need to acquire fresh equity capital (billions of Euros) in the market
  • Over-Capitalization: More 'dead' capital is created in financial institutions, resulting in declining investment returns in insurance.
  • Market shake out: Companies will exit the market
  • Pricing effects: prices (premiums) will rise, cover will be reduced

Banks

Whereas the European Insurers are on a more or less 'blind track' with regard to the implementation of Solvency II, the banks - according to chairman of the Basel Committee Mr Wellink - stressed that "decisions on the final proposals and their calibration will be made only after a thorough analysis of the impact assessment and the comments received on the consultative documents. The Committee will ensure that implementation of the new standards is consistent with financial market stability and sustainable economic growth".

The real problem
One doesn't have to be an actuary or financial expert to conclude that we're at the end of the road where adding more of the same type of control measures will substantially stabilize our system.

Without diving deep into real life quantitative analyses, let's get a helicopter-view and take a look at an average 'Control-Return Matrix' to do some 'rule of thumb' exercises...

Rule of thumb Control-Return analyses

Phase I
A few decades ago, starting in the good old sixties of the twentieth century, there where only limited control measures in place (control=0). The average Return on Equity (ROE) of a company was (e.g.) 6% and although Value at Risk (VaR) didn't yet exist as such, the 6% ROE could easily swap between (e.g.) +15% and -50%.

Financial markets where not that developed as today (no derivatives, , CDS, etc). Systemic risk was almost non-existent and accounting principles where based on the simple and relatively stable method of 'historical cost'.

The need for 'more control' was clear to everybody. More control implied lower costs, 'more opportunity insight' and 'more risk control'.
More control turned out to be a good investment and would lead to realizing a better return (ROE) in combination with a lower risk (Var) and a higher 'upward potential'. Every stakeholder was happy.

Phase II
Getting into the eighties and nineties of the twentieth century, 'control' had done its major job and still did, as it was able to manage the few relatively small recessions in those years.

With the help of the oncoming heavy computers, the first baby steps regarding new risk management techniques and ALM (Asset Liability management) were taken.

This way major risks (VaR) could further reduced, sometimes at the cost (expense) of a small reduction of the ROE. But this small effect was largely compensated by the 'fallacy high returns' in the high trust market.

Phase III
At the beginning of the Twenty First Century a new recession made clear the financial environment had substantially changed:
  • New techniques, models and the use of modern computer software led to new markets and new products like derivatives
  • Markets became global, (on face) transparent, in open competition
  • A lack of insight with regard to systemic risks
  • Differences in local supervision, legislation, administration and accounting rules, led to a complex, non-transparent global market.
  • In order to be able to compare companies, they had to be valued at 'market value', implicating the birth of more volatile (stock) markets....
  • Step by step, the public and media became more conscious. Investors and consumers understood that even if a 0.5% VaR level would be further reduced, it wouldn't make any sense because it would be always overshadowed by the non-trackable, nor manageable, risk of let's say 1 à 2%. And moreover, who would trust his money to a bank that would go bankrupt once every 50 or 100 years....

Investors, Boards, Managers, everyone lost their handrail....

In the recent decade (2000-2010) things got worse :
  • Existing control and accounting systems would locally differ and failed to meet the complex demands of the new markets
  • Supervisors en regulators, normally ahead of the market, were suddenly one step behind and unable to catch up given the actual system of supervision
  • It had become clear that new financial products ( e.g. CDOs, CDSs, subprime mortgages, swaps, swaptions) had been introduced without a good understanding of their financial construction or risk
  • Turbulence in the markets. Relatively stable stocks of big international firms, suddenly appeared remarkably unstable, due to new volatile markets/products and 'fair value accounting'.
  • The once so well controlled VaR risk exploded, due to these new types of risk in the market, the fair value accounting principles and the spooky systemic risk.

Way out

Like everyone else - totally flabbergasted - supervisors and regulators immediately grabbed the traditional emergency brake of 'more control'.

Unfortunately, more 'traditional' control in phase III will not have the same effect as in phase I or II. The effects of more traditional control in phase III will be:
  • Substantial but unsure decrease of ROE and 'upward potential'.
    The effects are not known upfront and can't be estimated well.
    Sure is that the costs of extra control and 'dead money' will have a negative impact on the ROE.

  • Unknown and questionable reduction of VaR risks, as one thing is sure: the new type(s) of (VaR) risks can not be estimated by our retrospective based models. Probably, all efforts in vain, the remaining actu(ari)al risk level will not be substantially reduced.

  • Trying to 'catch' more 'safe' risk levels (lower α , VaR) will lead to over-capitalization and 'dead' money in the balance sheet and an unbalanced growth of derivatives.

  • The market of derivatives continuous to grow.

    The notional value of derivatives held by U.S. commercial banks increased $804 billion in the third quarter to $204.3 trillion.

    This, despite the statements of Fed Chairman Bernanke who says he wants to avoid the possible risk of a future speculative bubble.

    And despite of Treasury Secretary Geithner who says he wants to reform financial regulation to avoid a future debt disaster.

  • Because the real issues of the financial crisis where not solved, but only covered up with government help (money), new uncontrollable 'bubbles' will keep showing up.

Solutions
probably the best solution is not 'more control', but

Other Control

Examples of 'other control' are:

  • Obligatory report and central registration of all derivatives under one worldwide supervisory. This way systemic risk analyses won't be 'guess statistics' anymore and can be managed. System risk is one of the weirdest risks to tackle, as is illustrated by the next article:

    Why Your Friends Have More Friends Than You Do

    Although the Exchange Commission has taken some serious steps in 2009 to regulate and strengthen the over-the-counter ("OTC") derivatives, this process will probably not be rigorous and fast enough to prevent a possible new bubble or collapse.
    All OTC market products should be asap standardized on a centrally administered basis.

  • Limit and control the derivatives market. Maximize the derivative market in respect to the 'normal' market. Limit each companies derivatives in line with his equity. New regulation should also be developed with regard to participating in non defensive (strange) derivatives (e.g. define max. exposure multipliers).
    If not the next bubble is a fact!

  • New derivatives should be subject to approval ('no objection') by the regulator before market launch.

So it all comes down to the 'right control' leverage.
It's either positive leverage with 'new other control' or negative leverage with 'more of the same traditional control' and waiting for the next bubble. What do you prefer as an actuary?

Sources:
- Contagion in Financial Networks
- Testimony Concerning OTCs (Over-the-Counter Derivatives )
- OCC’s Q3 2009 Report on Bank Trading and Derivatives Activities
- The bigger and riskier monster....
- Tarp facts: The Troubled Asset Relief Program
- The Investment Fallacy

Dec 11, 2009

Systemic Risk

In an excellent paper called 'Defining and Measuring Systemic Risk', professor Sylvester Eijffinger of the Tilburg University discusses actual developments around one of the most interesting risk topics of this moment: systemic risk (not to be confused with systematic risk).

Just a short warming up to actually download and read this excellent article:

ESRB
Main target of the 2010 launch of the European Systematic Risk Board (ESRB) is trying to identify and avoid future financial crises before they start. This implies that ESRB's main issue is 'how to detect systemic risks '. All this -of course - under the lead of the European Central Bank (ECB).

First of all the ECB does not have a clear concept of systemic risk, nor in the academia there exists a generally accepted definition. However, the G10 definition provides a good starting point:

Systemic risk
Systemic risk is the risk that an event will trigger a loss of economic value or confidence in, and attendant increases in uncertainty about, a substantial portion of the financial system that is serious enough to quite probably have significant adverse effects on the real economy


This still sounds pretty complex, and it is.
To get the right feeling, take a look at the next diagram illustrating a network of Credit Default Swaps (CDS) contracts:

In his blog 'complexity is our enemy' Steve Hsu, Professor of physics at the University of Oregon, explains in short and in simple words the principles and problems of the Credit Default Swap Market.

Hsu perfectly illustrates why some financial institutions are 'too connected to fail', as opposed to 'too BIG to fail'. Systemic risk is all about complexity.

New early warning models

There are several new models that can predict a financial crisis. Key challenge is to find a model with an indicator that predicts a potential crisis (just in time) with high probability, while at the same time minimizing errors of type I errors (missing crises) and type II false alarm).

One indicator can be qualified as the best current performing indicator: 'The global private credit gap', by Alessi and Detken (2009). This method predicts 82% of the crises correctly and has a 32% share of false alarms. 95% of the crises (price boom/bust cycles) are signaled in at least one of the 6 preceding quarters and the difference in the conditional and unconditional probability of a boom following a signal is 16%

Individual Institutions’ Contribution to Systemic Risk
For measuring risks of individual banks, a measure called CoVaR was developed by Adrian and Brunnermeier. The CoVar model measures the marginal expected shortfall (MES) as used in Value at Risk (VaR) as well as the systemic expected shortfall (SES).
Eijffinger's Conclusion
Finding new early warning instruments that are effective, easy to use, and independent of the interest-rate instrument seems to be an impossible task. And yet there is a solution according to Sylvester Eijffinger: "Central banks should give the growth of (broad) money supply more prominence in their monetary policy strategies."

The ECB with its often criticized monetary pillar may have a head start. Important central banks, such as the Bank of England and the United States Federal Reserve, kept their key interest rates too low for too long leading to a long period of double-digit growth in money supply.

The ECB was more cautious. To be sure, the fall of he risk premium on financial markets, the development of all kinds of exotic derivatives, and these derivatives’ subsequent misuse sowed the seeds for this crisis, but those factors could not have caused the crisis without the plentiful rainfall that allowed those seeds to grow.

Finally
What can pension funds and insurers learn from this?
The answer is simple:
  • Make Risk Management top priority nr. 1
  • Develop and implement in advance - cross financial institutions - early warning models.
  • Insist upon regulators to create a world wide central registration data base that registers and reports all possible derivate transactions in the financial market. Every financial institution has to report every transaction in a preformatted form.
  • New financial products are subject to approval ('no objection') by the regulator before market launch.
This way regulators will have a complete transparent view cross financial institutions. Systemic problem solved.

Sources
- Eijffinger:Defining and Measuring Systemic Risk
- The global private credit gap
- CoVaR modelM
- Steve Hsu: complexity is our enemy