Gödel’s Theorem, Cristal Champagne, and the Hedge Fund Hangover

The $3 trillion hedge fund industry has disclosed average annual returns of just 2% over the last three years, well below the average returns posted by major world index trackers with the S&P 500, MSCI ACWI, and Vanguard Global Bond Index Fund all gaining more than 15%, 8% and 4% respectively over the same period.

The hedge fund industry’s poor relative performance and investor complaints about high fees have culminated in more than $50 billion being withdrawn from hedge funds in the first 9 months of the year, the most since 2008 according to data compiled by Hedge Fund Research (HFR). While these withdrawals are insignificant in comparison to the total AUM of the industry, it could well be the start of a reversal of fortunes for many funds in the industry.

To some extent we are already seeing it. More than 500 funds were liquidated in the first half of 2016, on pace for the worst year for hedge fund closures since the financial crisis with some industry insiders predicting more than half of the 8,400 hedge funds globally may follow suit.

Many of the industry’s leading fund managers are blaming walls of passive money, algorithmic traders, central bank near-zero-rate policies, and political uncertainty for distorting markets. And to be fair in some instances they do have a point. Negative interest rate policies in Switzerland, Sweden, Denmark, Japan and the Eurozone with its 19 members are inflating asset prices globally and with investors increasingly searching for better yields in riskier assets, this is causing a disconnect between the underlying fundamentals and market performance for most assets.

However, this misses the main issue facing the hedge funds, which is increased competition and it is having a transformative impact on the industry. In 2000, there were just over 3,100 hedge funds managing approximately $456 billion on behalf of mostly high net worth individuals according to HFR. In 2016, there were over 8,400 funds managing nearly $3 trillion of assets, mostly for institutional investors.

With so many funds and so much AUM it’s becoming increasingly difficult for the industry to generate the alpha that is expected of them. Alpha itself is a finite resource which is easily competed away when too many industry participants with similar strategies are chasing the same market opportunities or looking to exploit the same inefficiencies. This itself is exasperated by industry hiring practices, where there is not enough talent to go around so the same people, with similar backgrounds and investment styles are poached from competitors, with dire consequences for returns.

In short, the problem the industry is facing appears to be one of homogeneity – the same talent, using the same inputs, the same process, and chasing the same alphas. Ultimately it is a problem of homogenous assumptions. That hiring a highly rated analyst will help boost P&L, that a tried and tested investment process doesn’t need to change or that fundamental indicators that have worked before will continue to work just fine. Assumptions of course are an important and necessary part of any investment process but analysts would be wise to ensure they always challenge and validate every and all statements on which their thesis relies.

Austrian mathematician Kurt Gödel’s Incompletion Theorem published in 1931 offers an interesting and relevant parallel to the investment industry of today. At a time when most of his peers were looking for a universal “Theory of Everything”, Gödel’s theory proved that there are always more things that are true than you can prove. Any system of logic or numbers will always rest on at least a few unprovable assumptions. This is true in maths but is equally true in finance. Anything you can draw a circle around cannot explain itself without referring to something outside the circle – something you have to assume but cannot prove. No matter how many inputs you use or how detailed your analysis gets, there will always be underlying assumptions which you cannot know which underpin your thesis.

To illustrate the point further, imagine a quant researcher with an unlimited budget, knowledge and capacity to build a super alpha computer which could identify, predict and monetize alpha almost instantaneously. It would of course need to rely on a certain set of economic rules and assumptions, many of which we know to be ‘true’ but cannot always be proved let alone predicted! Two interesting examples of this are illustrated below:

  • The law of supply and demand for example states that falling prices will increase demand. In 2006, US distributors of Louis Roederer’s Cristal champagne collectively lowered their prices in order to shift unmoved stock after a very public spat regarding the brand’s negative association with Hip Hop artists dampened demand for their products. But contrary to what most would have predicted, sales subsequent to the discounting decreased even further and Cristal fell rapidly in popularity from the 8th best-selling brand in 2005 to 63rd in 2006. Of course economists may point to the fact that Cristal is clearly a Veblen good; however, the point remains that classic assumptions that we take as a given cannot be so readily relied upon.
  • The law of one price states that identical assets should trade at similar prices. An example of this is Royal Dutch/Shell, which has both Royal Dutch shares (traded in Amsterdam) and Shell (traded in London). There is only one firm, the Royal Dutch/Shell Group, but based on the 1907 merger agreement, all cash flows are split so that Royal Dutch shares receive 60 percent and Shell shares receive 40 percent. Given this setup, the ratio of the market value of the Royal Dutch to the market value of Shell should be 1.5. However, this ratio has varied considerably from its theoretical value, from 30 percent too low in 1981 to more than 15 percent too high in 1996. Clearly on paper this appears to violate the law of one price but in practice the rationale probably had something to do the S&Ps decision to drop all foreign securities from its index in 2012, meaning US investors following an indexing strategy no longer had to buy the more expensive Royal Dutch line as they once did.

These perspicuous violations of logical economic principles which financial professionals hold to be true are more than just anomalies but highlight a more salient point relating to knowledge and investing. Analysts use inductive reasoning to extrapolate findings from a small sample to predict larger trends and this inherently means relying on assumptions which cannot be proved. Nearly all financial, economic and even scientific theories rest on the assumption that the universe is logical and based on fixed discoverable laws. This unfortunately is not the case, and understanding the limits of our knowledge and the assumptions on which our thinking relies is crucial to becoming a better investor. Woozle ethos is grounded on these principles. We aim to provide our clients with factual, on-the-ground primary research based solely on our interactions with industry insiders who know best and first how they are performing. In this way, Woozle can help clients validate and invalidate the underlying assumptions which underpin their investment thesis.

“We can invent as many theories as we like, any one of them can be made to fit the facts. But that theory is always preferred which makes the fewest number of assumptions” Albert Einstein

By | 2018-02-01T19:53:11+00:00 November 7th, 2016|Article|0 Comments