Lessons for Risk Management - Wilmott and Rowe

Great event organised by PRMIA and IAFE last night at Goldman's London offices with a long title:

 "A Little Thought Goes A Long Way and Lessons for Risk Management from the Current Crisis".

The event was moderated by Giovanni Bellossi of FGS Capital, and featured speaking slots by Paul Wilmott and David Rowe of Sungard. Here are my notes on the evening, please forgive any innaccuracies, and please persevere through some of the techy quant stuff, as their general points are well worth understanding.

• Giovanni quoted from Nassim Taleb about how VAR is invalid and that mainstream financial mathematics should be banned (or words to that effect, see earlier post on Taleb)

• He added that whilst what Taleb says cannot be ignored, he said that despite the current crisis and its causes that we should not "throw the baby out with the bathwater" and added that Taleb "...is not only able to recognise a cow but also knows how to milk one."

• Giovanni said that financial mathematics has much to offer and that whilst VAR is simply a number, one of its great benefits has to make one measure of risk simple and compelling enough to get traders and risk managers talking.

Paul Wilmott then took the floor and put forward his thoughts:

On Taleb and the Black-Scholes Model

• Paul mentioned that he and Taleb were great friends, and whilst he agreed with much of what Taleb says he has areas of disagreement, particularly over the use of the Gaussian distribution in finance and its implications for "fat tail" events
• Paul Googled "Taleb" and found more entries for Taleb than for Stephen Hawkin which shows how much attention had come his way due to the "Black Swan" debate
• He thinks that he and Taleb are the "Marmite of finance" (for those of you not in the UK who do not know Marmite, it is a sandwich spread that you either love or hate, never anything inbetween)
• He suggested that every quant needs a much more fundamental and practically grounded understanding of financial mathematics.
• Paul refered to some work (mentioned by Giovanni) that Peter Carr of Bloomberg had done on discrete daily hedging that showed that this option replication technique could remove up to 85% of the risk and that all quants should know about this 15% error term when trying to calculate an option price to the Nth decimal place.
• He described how in the past he had set up a volatility arbitrage hedge fund, wanting to improve upon the flawed assumption of the Black-Scholes (B-S) model that volatility is constant and to build the world's best volatility model for option pricing.
• Paul said that he did build the world's best volatility model (?!), but soon found it took too long to calculate, so he reverted back to B-S and has become an unfashionable fan of the model and its assumptions.
• He added that many of the variants on B-S to overcome its limitations have made the model worse and harder to calibrate.
• In some part due to Taleb's opinions on fat tails of distributions, B-S and other models are now very unpopular but Paul claims that not many people have actually bothered to robustly test the B-S model or take a practical, evidence based approach such as that adopted by Peter Carr.

• Paul then showed some example charts and said that with a limited number of opportunities for regular time-period hedging it was not valid to use risk-neutral pricing whereas if the same number of hedges could be used optimally (implying at irregular time periods) then risk-neutral was valid and hedging could be more effective. He emphasised that this was the kind of practical stuff that a quant should know and that quants show know less about esoteric complex financial mathematics.

Correlation

• Paul said that of all of the issues that need addressing in mathematical finance, the one that he has very few answers on is correlation.
• He showed that even basic questions about correlation are poorly understood, even by quants - a question he asks some quants was that if two asset prices both start out at 100, and they have a correlation (of returns) of 1 (perfect correlation) what is the price of the second asset after a year if the first moves to 200. The answer is not 200, and he showed how assets could diverge in overall direction but still have a correlation of 1 or rise together with a perfect negative correlation of -1.
• Paul illustrated how correlation was a very blunt measure that is mis-used by people to summarise the highly complex and historically unstable relationships between assets driven for example by industry sector success (leading to +ve correlation) or competitive success (leading to -ve correlation)
• As a result, he said that financial products whose value depends on correlation should not be transacted in any great size and moved on to the example of CDOs, where a CDO with 1,000 underlying mortgages has been modelled with 1/2 million correlations all assumed to be 0.6. Why this assumption should be made was his main point.

Sensitivity to Parameters

• His main point here was that a constant should not be varied, otherwise it is not a "constant", in particular focussing on volatility used in the B-S model and the calculation of Vega as prices are moving.
• Paul added that sensitivity measures may apply locally and is such may look comparible from one situation to another, but quants need to understand how outputs respond over a wider range of inputs, and not to be inhibited by accepted practices and beliefs.

Complexity

• Models need to be robust and transparent, and that quants should aim for the mathematical sweet spot.
• Paul put forward the following analogy that at least when driving an old car over a long distance, you knew that the car was likely to break down at least once, but you also knew that it was likely that you could fix it. Contrast this with driving a modern sports supercar and finding that it has (unexpectedly?) broken down - you don't know how to fix it, you do not complete your journey and it costs you an ordinate amount of money to put things right...

Self-Referential Feedback

• Paul described here how the hedging of derivatives contracts in the underlying markets can cause price movements in underlying markets that cause derivatives contracts to re-price that cause more hedging in the underlying markets...
• He was critical of credit derivative pricing as being too complex and too "mathsy" (...but had to admit that he had also endorsed some of this work at the time)

Calibration

• Paul said that model parameter calibration is the devil's work...
• He refered us to inverse problems in mathematics as a background to this issue in mathematical finance.
• He emphasised how markets and price behaviour is fickle and driven by human opinions and behaviours
• He said that on-going and regular re-calibration of a model is very, very likely to mean that the model is wrong (he had a particular example of calibrating a particular model he hates where vol is a function of underlying price and time.

David Rowe, Sungard's specialist spokesman on risk management, then took over from Paul and set out his five topics for discussion:

• Statistical Entropy - fundamentally that information can only be extracted from data, with the emphasis on extraction of information (from that already in the data) rather than creation of new information.
• Structural Imagination - that we need to be aware of how the market assumptions we make are themselves a model and that we need to spend more time on thinking about what could happen outside our current understanding or market experience.
• Self-Referential Feedback - the feedback loops in pricing, risk management and economics
• Complexity and Dark Risk - when you add (untested) complexity of a model to limited data sets you get a recipe for disaster.
• Alternate Means of Valuation - when the primary means of valuing a security is not available (illiquid markets anyone?) then what is the secondary means of calculation value.

Some further notes from David's talk:

• AAA rating should imply a failing once every 10,000 years, with some super senior CDO tranches being rated as better than AAA - David pointed out that even as recently as the early 1990s there were problems in the US housing market that indicated that AAA did not mean what it was taken to mean.
• On structural imagination, David said that quants and risk managers must look for unrepresented variables in a model and track them early to monitor their effects
• On feedback he cited an example where increased returns drove product innovation which drove up (CDO) volumes, which caused underwriting standards to fall, that allowed further complexity, that then led to unreliable risk estimation which then led to more product innovation... and so on.
• He suggested that quants adopt the "second means of valuation" mantra in a similar way to credit specialists always having the mantra when assessing credit of "what is the second means of repayment" (e.g. a lien on a house) when the primary means (mortgage payments) goes away.
• David showed a nice classification from an IASB paper on classifying financial instruments:

Level 1: fair values measured using quoted prices in active markets for the same instrument.

Level 2: fair values measured using quoted prices in active markets for similar instruments or using other valuation techniques for which all significant inputs are based on observable market data

Level 3: fair values measured using valuation techniques for which any significant input is not based on observable market data

David additional proposed the interesting level of "Level ?" for some products, and said that obviously more attention needs to spent on Level 2 and 3 instruments under conditions of reduced (non-existant?) market liquidity.

Summary Session:

Paul and David then answered some questions from the audience:

• Paul said that some risk managers lacked the imagination necessary for good risk management, being confined in standard procedures, beliefs and ways of doing things. He wants risk managers who are good at thinking laterally.
• Paul said that risk management was often an afterthought, not part of the trading process.
• David said that VAR has proven useful despite its weaknesses, in his opinion preventing failures from non-extreme events regardless of the recent extremes
• David said that in answer to Taleb's criticism of using history in modelling, it quite frankly is all we have to go on. He quoted Mark Twain in that:

"History does not repeat itself but it does rhyme"

The talks were interesting, and even on points that have been discussed elsewhere both speakers had some interesting slants and good analogies. But maybe I am biassed, as the wine afterwards wasn't bad either!...

Best execution 2009 - July 1st 2009

A few summary points I took from the Best Execution Europe 2009 event courtesy of Incisive Media that I attended yesterday morning.

The event started with a presentation by Michael Fridrich, Legal and Policy Affairs Officer of the European Commission:

• From what Michael was saying then in my view, it seems that the EU is using the G20 declaration on financial stability in April as a remit to regulate in many areas (not all of which related to the current crisis, see last paragraph in this post)
• He said that the EU is currently working on removing national options/discretions with respect to financial markets in order to create a single EU rule book and combining this with stronger powers for supervisors including much harsher sanctions against offending institutions
• They are also reviewing the necessary information provided to investors in OTCs, even if the investors qualify as "professional investors" under Mifid.
• The EU is currently reviewing Mifid and the Market Abuse Directive (called "MAD" which is at least humorous...)
• EU is also unsurprisingly looking at the regulation of Credit Ratings Agencies (CRAs) given their involvement in rating CDOs and other structured products

So in summary it was a civil servant PR exercise with few surprises, other than we are going to regulate anything that moves. On to a panel debate on "build vs. buy" for execution management software. I will try and put my obvious vendor bias to one side in summarising this one:

• The panel summarised that this decision was about the usual issues of time to market and what is an institutions core IP
• A senior IT manager from JPMorgan said they both build and buy - but given the size of their organisation and the need to innovate they do build a lot
• The COO of Majedie Asset Management said that "build" was "20th Century" and the IT should focus now on "assembly"
• He added that if IT lead a procurement process he finds this tends to lead to more proprietary solutions than if business is managing it.
• He summarised that business people should have the mandate to define inputs/outputs to a requirement and that IT were not qualified to do this.
• Putting it more controvertially he suggested that IT people should work for IT companies
• The JPMorgan guy responded that "assembly" of external components can lead to excessive staffing in managing all the plumbing, and that build in house could build a more generic and targetted platform that would need less management
• The moderator summarised the build vs. buy decision as one of balancing time to market and how bespoke a solution is alongside of looking at the risks for buying of 1) integration risk 2) vendor risk and for building of 1) delivery risk 2) key man risk

The debate on this was pretty standard, but the guy from Majedie was at least controvertial in what he was saying, (including at one point that "investment management does not scale"). I assume he is trading simple products and as such is able to outsource more than the JPMorgan manager. My own slant is that more vendor products need to be designed to integrate easily with the IPR of a financial institution i.e. less black box.

Tom Middleton of Citi then did a presentation on (equity) market liquidity and market fragmentation:

• He started by saying the Smart Order Routing (SOR) was like "Putting Humpty-Dumpty back together again" from all the sources of liquidity now available under Mifid.
• Being no expert in SOR, I was excited (?) to learn a new term which was "finding Icebergs" - apparently an "Iceberg" is a large non-public ("dark")  order being posted with a much smaller public trade order.
• He said that market fragmentation will increase further but there will be less trading venues as the market consolidates.
• New algorithms will be developed more specifically for trading on dark pools of liquidity
• Clearing and settlement costs are still high across Europe which limits the usage of small size orders in trading but trading volumes will continue to grow
• The drive to ever-lower latency will also continue
• Usage of SOR will grow

Tom's presentation was then followed by a panel debate on Smart Order Routing:

• A manager from Baader said that the German area market of Europe was not very sophisticated yet, with most German clients specifying exactly where the trade should be executed hence nullifying the need for SOR.
• Deutsche Bank (DB) mentioned that having both US and EU operations had helped them get SOR in place for the EU quicker given their US experience.
• UBS and Baader both said that Algo trading and SOR are increasingly integrated and will merge with the Algo define what and how to trade and the SOR component determining where
• DB said that a "tipping point" towards usage of SOR in the EU will occur when more than 20% of trading occurs away from the primary exchanges.
• DB said that 60% of US liquidity was due to algorithmic trading and that there were now no EU barriers to this happening in European markets and bringing with it increased liquidity, although issues such as not having a consolidated market tape for trading made things more difficult
• Neonet said that clearing and settlement costs were still a barrier to widescale SOR adoption.
• IGNIS Asset Management said that SOR was a "high touch" service for them, requiring SOR vendors to be very responsive and client focussed. In selecting SOR vendors they were concerned with data privacy and also with having a real-time reporting facility to see how orders were being filled.

And finally (at least before I had to leave) there was a presentation by Richard Semark of UBS on Transaction Cost Analysis (TCA):

• He was surprised to find that there were not many presentations around on TCA
• TCA vendors are behind the times and are not up to date with current developments
• Historically TCA was about what had happened (about 3-4 months ago!)
• Mifid has driven fund managers and traders to talk more and TCA is a key part of this conversation
• It is hard to look bad against traditional TCA measures such as VWAP if a stock is always rising or always falling, and this can hide a lack of performance and "value add"
• Using "Dark" for non-displayed liquidity has been a publicity disaster for the electronic trading industry
• Much Smart Order Routing (SOR) is still based on static tables of trading venues that are updated on a monthly or quarterly basis
• Market share by volume of a venue is not necessarily correlated with obtaining the best prices in the market
• TCA should be based upon a dynamic benchmark that responds to the market and trades done not against a static one
• Trade performance is not linear with trade size which is an incorrect assumption in much of TCA
• Trade risk (variability in outcomes) deserves more focus
• Portfolio TCA is much more complicated where the trading of a single stock cannot be looked at in isolation of its effects on the whole portfolio
• Real-Time TCA is becoming ever more important to clients since it allows them to understand more of what is going wrong/right with filling an order
• TCA providers are not doing a good job for clients, not using the right data or answering the right questions for clients

Not sure who the TCA providers he refers to are, but maybe I should find out to see what they offer...

 

 

 

 

Risk as Sales Support?

Article in FTFM yesterday saying that the risk function is being ignored by asset managers when formulating new financial products.This seems consistent with some recent comments from one risk manager who said that their role was a lot to do with sales support i.e. to convince potential investors that the asset manager has good risk management capability. Given all the discussions on the sell side about the role of risk managers and the risk function, sounds like the debate should open out more onto the buy-side too.

OTC Valuation by SGSS

Given all the recent attention that OTC derivatives have received (see Geithner letter), then a topical update on the work we have done with Societe Generale Security Services (SGSS) on OTC and structured product valuation services has been written up on Securities Industry News. The work involved extensive integration with Mysis Summit, where our TimeScape data and analytics management system is used to provide "Golden Copy" of market, reference and derived data for the derivative products being valued. The section on TimeScape says:

"The Summit FT solutions are integrated with SGSS' market data software tool TimeScape, licensed from London's Xenomorph in November 2007. This produces a "golden copy" of end-of-day prices from 15 different information suppliers. The unit also processes information related to 70 different currencies and 5,000 volatility surfaces, which give three-dimensional views of how much and fast a security can move up or down. With Summit's product, each surface can include between 200 and 500 data points."

From talking to some of the SGSS team at our recent user group, the thing they most seem to value about TimeScape is its ease of use in describing and managing any kind of product, allowing product and market data specialists to use and customise the system without the need for specialist technology knowledge. This echos some of the things that were said about TimeScape after a demo to Lab49 last year. 

Less risk on the buy-side?

Interesting but counter-intuitive survey results discussed on the Advanced Trading blog, suggesting that risk function has lost status at buy-side institutions.

Capital requirements for Asset Managers

Article in the FT today saying that the Financial Services Authority (FSA) has criticised asset managers for poor risk management, and that these failures might force it to impose higher capital requirements on some institutions.

The Investment Management Association (IMA) countered by saying that the FSA guidelines on capital requirements for asset managers were unclear, but also added that as asset managers did not hold client-owned assets on their balance sheets they did not need to hold capital against these assets unlike the banks.

I understand this last point by the IMA, but surely given an institutions fees (aka revenues) derive mainly from fees for managing these assets, surely the IMA is not doing itself any favours by effectively suggesting that the (currently volatile) value of these assets are not relevant from a institutional risk point of view? Poor investment performance leads to redemptions, leads to reduced fees, leads to concerns over institutional stability, leads to more redemptions etc, etc.

Anyway, interesting that this is receiving some regulatory attention and maybe buy-side risk management will soon be moving beyond helping to market and sell the latest investment product...

RiskMinds - Robert Merton

Robert Merton gave the opening talk this morning on the subject of sovereign wealth funds...and immediately digressed into talking about the current credit crisis. As with Shiller yesterday, he is advocating more and better financial theory that has learnt from recent problems rather than saying the mathematics is invalid. He was heavily critical of the pricing models used for CDOs and the like (more of which in a later post).

An interesting point was that he reminded the audience that vanilla loans and mortgages contained embedded put options on the assets of the issuer, and that as a result the recent dramatic decrease in value of this kind of instrument is not purely due to ten sigma movements in markets, but rather lower movements in market variables but combined with greatly increased sensitivity (delta) to these inputs as markets decline and become more volatile. Put another way, he does not hold with the fashionable premise of the Black Swan of extreme events explaining all that we have been experiencing.

On sovereign wealth funds, he suggested that they should concentrate on their unique advantages as investors/counterparties in the market, such as credit worthiness and access to liquidity, and focus much less on stock picking and timing to allocate investment (he cited recent investments in US investment banks by CIC as an example). He proposed that sovereign wealth funds should sell that which costs them nothing (e.g. liquidity) and that others needs. He ended his talk by suggesting the sovereign wealth funds may (only may) step in to fill the gap left by the dramatic downturn in hedge fund activity in the market, as he classified both types of institution as "lightly regulated" and able to get around the "institutional rigidities" faced by the banks. So maybe the sovereign wealth funds are not the international bogey-men that the press have been making out recently?...

MiFID Market Data Deterioration

Members of the Investment Management Association (IMA) are not happy about the quality of market data following from the first year of MiFID being with us. According to an article in the FT this morning, a survery of IMA members has voiced concern over the fragmentation of trading venues leading to a deterioration in the quality of market data available.

Most institutions seem positive about the benefits of competition that multiple trading venues such as Chi-X, Turquoise and BATS bring, but IMA members would like to see a centralised venue where prices are consolidated and made available to the market (for free?), similar to the "consolidated tape" available for US markets.

Sounds like the institutional need for centralised data management across different departments and systems is also becoming apparent at a market and exchange level following MiFID. Multiple trading venues and decentralisation is necessary to bring the benefits of competition, but these benefits unsurprisingly do not come without costs or issues (see earlier post).