View more on these topics

Chris Gilchrist: The collapse of financial theories

Most of us have some painful memories of the 2007-08 crunch. It is not surprising that most people in the business are keen to forget and to resume business as usual. This will not happen and if you believe it is happening now, you are heading for trouble.

One of the reasons we had the crunch was that the whole of economic and finance theory is wrong. By wrong, I mean it is based on assump-tions that are false, so the models it uses do not represent reality except in very limited ways.

Rational investors, maximising utility, symmetric risk preferences, normal distribution of returns…any so-called theory this full of holes would have been terminated long ago in any of the hard sciences. Only because of political factors has neoclassical economic and finance theory survived so long – in essence, it has served the interests of the rich and the bankers, who in the US have bought the regulators and politicians.

These may seem extreme assertions, but I refer you to Steve Keen’s Debunking Economics and Justin Fox’s The Myth of the Rational Market for the evidence.

There are lots of very big implications of the collapse of current theory, but for now I will stick to a few specific to the investment business.

Academics have known for years that investment returns are not “normally distributed”, so standard deviation is not a valid way of assessing them.

The reason they chose to stick with the bell curve or Gaussian distribution is that the maths is manageable, whereas the chaos maths pioneered by Benoit Mandelbrot now used in physics, weather forecasting and biology is very difficult.

The patches applied to normal distribution models to cope with “fat tails” are ugly ad hoc attempts to make the models look more like reality. But this does not work, as exemplified in the total failure of value at risk models (however adapted) to measure the real risk incurred by banks.

The conventional defence to the inaccuracy of standard distribution models is that they work most of the time or work except in exceptional circumstances.

But since there is no way of predicting when they will and will not work, using such models as a basis for decisions is simply unscientific. It can only rest on the belief that the user can judge whether the model will work at this time.

A user of modern portfolio theory methods would have to say to their clients: “Our model works most of the time but it will not cope with extreme events like crashes and panics and it has no means of predicting when such events will occur or how extreme they will be”.

If that is what they were told, how many clients would sign up to the use of such models?

The market is not efficient, probabilistic risk measures do not provide reliable grounds for decision-making and the coming revolution in economic and finance theory will, like those in meteorology and biology, emphasise “butterfly wing” effects that are simply not predictable.

As JM Keynes said, the real problem investors face is not risk, which is measurable, but uncertainty, which is not.

In the absence of genuine science, which Keynes was sceptical could ever happen with investment, it is entirely rational for practitioners to use judgement and heuristics – old fashioned rules of thumb – in making decisions.

Chris Gilchrist is the joint author of The Process of Financial Planning and editor of the IRS Report

Recommended

1

MP spells out auto-enrol role for IFAs

Former pensions minister and work and pensions committee member Oliver Heald says IFAs have an important role in helping employers through auto-enrolment. At a committee hearing in January, pensions minister Steve Webb said the availability of “fairly standard” products for auto-enrolment would make expensive tailored advice for small firms and lower-earners unnecessary. Last week, the […]

15

Nic Cicutti: The struggle to find FSCS funding solution

When your head has been smashed hard against a brick wall for ages, the fact that the force of collision has lessened slightly can sometimes feel like an enormous relief. Yet somehow, I very much doubt that the vast majority of IFAs will feel that way about the announcement the Financial Services Compensation Scheme interim […]

boulding

Lifting restrictions on Nest could break EU law, says Boulding

Legal & General pensions strategy director Adrian Boulding says removing restrictions on Nest, as suggested by the work and pensions select committee, could break European Union law. Nest was set up with a loan from the Government and will start operating in October, with an annual cap on contributions and a ban on transfers in […]

Clougherty leaves Aviva amid business shake-up and job losses

Aviva Investors chief executive of UK funds John Clougherty has left the firm amid a shakeup of the business and widespread job cuts. This week, Moneymarketing.co.uk revealed that Clougherty, who joined Aviva Investors in June 2005, had gone on gardening leave. He has beenreplaced on an interim basis by chief operating officer of the UK […]

Newsletter

News and expert analysis straight to your inbox

Sign up

Comments

There are 9 comments at the moment, we would love to hear your opinion too.

  1. excellent copy for use with suitability reports.

  2. Adrian Pickersgill 26th March 2012 at 3:37 pm

    It’s the truth, but as is so often the case the truth is not always simple and it certianly is tricky to explain in a way that the ‘powers that be’ would be happy with!

  3. Stochastic or any other model has limitations, in that the additional challenge is that most advisers using them are doing so to justify one route or another. Fantastic article

  4. Rather makes a mockery of the FSA’s attempts to quantify, codify and bureaucratise risk.
    I’m sure they will come out with a few more box’s we must tick!

  5. As far as pure investment is concerned we have probably passed the point at which investors gain on balance by taking advice.

    I say “on balance” simply because although advice almost certainly does add value I doubt if the benefit of any advice can offset the additional costs that have resulted from all the over-regulation of recent years.

    Investors who diversify, keep costs down and take advantage of tax breaks may well ask exactly what they are getting for the extra 1% pa of funds “under management” that their adviser is charging. What exactly do the “professionals” know that the investor does not ?

    Pretending to be able to pick funds which outperform or rebalancing and other forms of activity for its own sake will just not do.

  6. I was delighted to see mention of Mandelbrot, who demolished the current mathematical bases used in finance in his book in 2004. This came out virtually side by side with Nassim Taleb’s book that also took pleasure in putting the boot in. But they were preceded by 30 years by Burton Malkiel. Indeed there is a whole library of books, including the two mentioned in the article, that seriously question the fundamental assumptions of economics, let alone the mathematics used. And a lot of those books seriously pre-date the economic crisis. Yet those mathematics are still held in high esteem and, I suspect, will continue to be for years to come. The serious question is why is the finance world so ready to believe in witchcraft.A couple of weeks back the FT Magazine devoted a whole article to financial Quants and the adulation of the practitioners. The Black-Scholes theory has been thoroughly discredited on the back of subsequent research, but this has not apparently diminished its usage. There appears to be a delusion that because numbers can be manipulated, the answer is meaningful. The more esoteric the manipulation the more meaningful the result. The fact that neither history nor academia agree with this judgement does not deter practitioners in any way. What did come through from the FT article, possibly by accident, is that the quant techniques are more about trading than investment. They see discontinuities in short term patterns, thus allowing traders to make profits. It could be that there is little wrong in this. But there is a nagging question, found in Chaos Theory, of which Mandelbrot was such an influence, that such trading may be a fundamental disturbance in the normal patterns of the market.
    Gaussian statistics, upon which most of the techniques are based, assumes a discrete and self correcting environment. The economy is far from discrete in that it is influenced by everything from money to fraud, from rational to irrational, from politics to the irrational (sorry, that’s a tautology). There are rarely two people with the same interpretation of information (read Kahneman & Tversky’s original paper of 1974 for a good understanding of why), so why should it be self correcting.
    Most systems operate on the basis of self correction, in the main, otherwise we would be in a continuous state of chaos. It is not always clear why this societal self correction works, but it does. Until it goes on one of its self-reinforcing episodes when things end up in tears. I believe that the recent financial crisis was just such an episode.
    I would also conjecture that quants played an integral part in the process, because, by trading on discontinuities the traders ensured that the financial system was not able to self correct. Add to that the psychology of people who succeed for a while, whether by luck or good management, making large amounts of money, thus making them immune to controls, and the God Delusion sets in. Further self-reinforcement.
    Excessive risks are taken because the risk takers believe they are in control of the system. The mathematics tell them so – and this is re-enforced by their own sense of grandeur. The bubble builds.
    It should also be noted that Keynes expertise was more in probability theory than economics. His understanding of risk was therefore influenced by defined systems in which it probably is possible to measure risk. I seriously doubt that measurement of risk in the financial world is possible. It may be possible to estimate, with sizeable caveats, but not to measure. We do not even have a good working definition of risk, just a concept that we bandy about to impress people and with which the Regulator can beat people around the head, without themselves ever declaring any “measurable” guidelines. We need explicit risk guidelines. We know they will not be correct, but, for the moment they will be consistent, and will be open to discussion and research. Practitioners and consumers will learn to speak the same language.
    And that is the main danger of practitioner heuristics. Because they will not be consistent across the board they cannot be explained to the Regulator. To exacerbate this problem the Regulator is constantly talking in terms of certainties, of knowing risk levels and knowing attitude to risk, neither of which can be ascertained with anything approaching reasonable accuracy, nor with consistency over a period of time. Yet the public is led to believe that they can. And there lies a massive problem in current day finance. The Regulator is totally out of its depth with modern day finance because it pretends it knows more than it is possible to know, and thereby cons the public into expecting a level of certainty that is just not possible.
    For advisers to operate on the basis of their own “rules of thumb” is therefore tantamount to suicide because they do not hold the strings of power. The “rules of thumb” must come from the Regulator if the Regulator is determined to remake the industry in its own image.
    What is required is that the FSA/FCA set down specific parameters within which practitioners can make judgements. If practitioners operate within those specified parameters they should be immune; if they step outside they should be required to demonstrate that the exercise of that discretion was based on a justifiable interpretation of the information available – or they get hung, drawn and quartered. The parameters should be open to development, based on research and experience. That is the only way in which people learn.
    At the present time FSA Guidelines are so woolly they are putting sheep out of a job.
    And, more importantly, the general public should be told that perfect advice is an impossibility, now and forever. Caveat emptor.

  7. Jonathan Cooper 27th March 2012 at 10:18 am

    “What exactly do the “professionals” know that the investor does not ?Pretending to be able to pick funds which outperform or rebalancing and other forms of activity for its own sake will just not do.”

    First off very good article. With regard to this anonymous comment posted above which did make me smile.

    As IFA 25yrs + like to point out. If we could all use the magic of tracker and asset allocation computer portfolio`s why is it that the banks managed funds have very poor performance?

    There are good Accountants out there and very bad ones, there are good Solicitors and very bad ones; the difference in the fees they charge is enormous. You could use legal aid or your own DIY will good luck the revenue love these!

    Or you can pay to have your affairs professionally managed and yes there are very good funds and managers out there………………………………………………………

  8. My brain hurts.

    Roll on retirement!

  9. Great subject and a good article with some good comments.

    As Warren Buffet once said: “Read Ben Graham and Phil Fisher, read annual reports, but don’t do equations with Greek letters in them.”

Leave a comment