Home | Topics | quant

Acquisition of RiskMetrics by MSCI Barra could standardise VaR measurement: ISSUE

Printer-friendly versionPDF version


Acquisition of RiskMetrics by MSCI Barra could standardise VaR measurement and increase systemic risk, says EM Applications

London - 9 March 2010 Bob's Guide

MSCI Barra's (NYSE: MXB) acquisition of RiskMetrics (NYSE: RISK) unites two market leaders in systems used by the asset management industry to calculate their Value at Risk (VaR). Once the firms are merged, it is possible that the majority of the global asset management industry will be calculating VaR using models and systems produced by just one company. This will make it more likely that these firms will take similar positions, which will increase systemic risk in the financial sector.

Currently the methodologies used by MSCI Barra and RiskMetrics are radically different, as are their perspectives on risk to clients. The need for cost savings and product consistency must mean that, following the merger, all their clients will be seeing risk in the same way. And the brand power of two industry giants joined together will make it more likely that other approaches will not be taken into account as excessive faith will be put into the risk figures produced by this huge company.

This spells danger, both for investors relying on such figures and for the financial system as a whole. A rule of markets is that if everybody believes in the same thing it will be given too much weight, and the potential losses from an alternative outcome are considerable.

The scenario would be similar to a merger of the two largest credit rating agencies, Standard & Poors and Moody's. While the merged company would benefit from cost savings and would enjoy substantial resources for carrying out detailed and expert credit analysis, the public loss of a different perspective on credit risk would surely outweigh the private gain.

Once a specialist topic for quantitative experts, the measurement of Value at Risk entered the mainstream when its use contributed to the banks' underestimating of their capital needs during the credit crunch. Nicolas Taleb, in particular, has been a trenchant critic of the naive use of risk calculations saying, in relation to VaR, that: "It is the uniqueness, precision and misplaced concreteness of the measure that bother me."1

The Turner Review2 also criticised VaR calculations where they were based on short term observations, which led to the pro-cyclicality of risk sensitive behaviour. This meant excessive risk was being taken based on benign conditions at the time, not taking account of less favourable longer term trends.

The clear conclusion of the post credit crunch analysis is that risk cannot be boiled down to a single number and that a multiplicity of measurements and analyses are needed in order to gain a full picture of possible exposures.

Peter Ainsworth, Managing Director of EM Applications, said: "Although a specialist area, the credit crunch demonstrated what a key role in global financial stability is held by the methods used to calculate investment risk. While there is always a need for advances in the methods used, it is also important for a diversity of methods to be used as this is the best guarantee of overall stability. MSCI Barra with RiskMetrics will hold a dominant position in methods used for investment risk estimation, which has the potential to lead to concentrated positions and systematic risk."



Roy HodgsonThe issue as I see it is clearly one of methodology and an understanding of VaR's as statistical measures. As this article states in terms of The Turner Review, it is of limited value considering only short term time series data. It is for precisely this reason that we advised a Tier 1 client last year that relying entirely and only on a variance/covariance VaR for regional level risk reporting is entirely insufficient.

It is only by taking a balanced VaR view considering an extended multi year time data series, variance/covariance and then Monte Carlo projections and simulations that a holistic picture can be generated which is the basis of Siag market risk methodology. Then; via full portfolio back testing and stress testing to full revaluation considering all VaR analyses and metrics a high degree of probability and accuracy may be achieved within a given temporal horizon. The shorter the horizon - the more accurate the measure. The longer the time series considered - the more accurate the measure. The more scenarios considered - the more accurate the measure. The more recent the market volatility data etc - ad infinitum.....

It is presumably precisely for these reasons that the Basel Committee are recommending a Stressed VaR based upon a 12 month time series during periods of elevated volatility as a further check and balance to VaR analyses in order to include dramatic stress events, albeit that no system can predict Black Swans which are by their nature random and unpredictable events. I would equally argue that the current crisis was no Black Swan and was entirely self evident from the sub prime asset bubble and the radical over leveraging of complex structured securities and CDS coverages, high PD's, inaccurate ratings, inadequate reserve ratios, over exposure to high risk derivatives etc; however many chose to ignore the signs until the market imploded. + 500 bn in write downs in Europe alone in 2009 should emphasize my point here.

VaR consists of a range of different types of analyses and reports to produce a balanced view of risk exposure. There is no one size fits all VaR which covers all eventualities and this is why we deploy the balanced methodology we recommend.

Perhaps the real challenge here is that many do not understand VaR and how to use it correctly in a range of different scenarios and including a range of VaR variants considering past, present and probable / possible futures with weighting for stress scenarios and events. While we still have Tier 1's relying on just one type of VaR analysis only as we know to be the case, it is an incomplete view of risk exposure and accordingly an unreliable metric because of incorrect methodology application - not because of faults in the measures themselves. The correct revaluation of portfolio should therefore be a risk adjusted valuation, with the highest possible analyses reporting frequency to include actual market volatilities and for this a range of VaR analyses and reports are required; not just one type.

And to Mr Taleb: Var is not a measure - it is a range of measures. This only goes to prove my point as to the lack of understanding of VaR, its variants, and how to use these metrics correctly; seemingly even in high levels of Academia.



Short URL
Asymptotix on Twitter

Are the key legislative pillars such as Basel II & III, UCITS IV and Solvency II forcing you to re-examine how you identify, measure and manage risk and capital?

Asymptotix work closely with our partners to help clients develop a more proactive, systematic and integrated approach to governance and risk management to deliver proper value.

Asymptotix can offer the support you need to deliver on time. Read more...

Is the goal of your website to sell services or products, educate, or collect data?

A positive customer experience is vital to conversion, no matter what your conversion goals may be. Our designers and developers will create a positive experience to maximize your conversions and deliver the optimal return on your investment. We strive to find the perfect balance between the web site’s design and functionality.

Asymptotix implements interactive solutions for European companies. From corporate websites to social communities, our clients will tell you an investment in building a scalable online experience will deliver long-term tangible benefits.

Based in Luxembourg we can help you all over Europe. Our multi-lingual team can work with projects and speak your language! Read more...