Lost in correlation fatigue
Posted by Izabella Kaminska on Jul 28 09:40.
Alternate title: The end of valuation.
And this is why.
The following comes from Petromatrix’s Wednesday energy markets report:
Daily Trading Volume for WTI on Monday was at the lowest level of the year and Open Interest is not moving at all. As per our comments of the last 10 days, we do believe that crude oil is suffering from fatigue of trading correlations and it needs to regain some life of its own for traders to regain confidence that the oil markets are something else than an high-speed arbitrage generated by computers.
What has that got to do with valuation?
Well, if a fundamentals-driven market like oil can be mispriced because of high-frequency (HFT) and algorithmic trading — imagine the level of mischief potentially possible in equities?
On that subject, Joe Saluzzi of Themis Trading directed us to a very interesting second quarter report from Toronto-based money manager Friedberg Mercantile (as highlighted on Zero Hedge before us).
It’s worth a rehash though, since what it says is truly mind-blowing.
The group says it has experienced significant problems of late with its US equity hedge programme based on a market neutral strategy.
While the strategy had for many years earned above-average returns that were totally uncorrelated to S&P 500 returns, in the last five quarters it repeatedly disappointed.
And here’s Friedberg’s explanation as to why:
This persistence of unfavourable outcomes is a totally unique event in the 19-year history of the program. To boot, the cumulative loss for this five-quarter period, at 18.45%, has materially affected the fund’s performance. A rough estimate, taking into account the average allocation given to this program, puts this effect at approximately 500 basis points. What is noteworthy is that this has occurred despite the fact that we have not changed any of our procedures, neither the selection process nor the hedging formulas.
For some time we have speculated that the results have been affected by the lack of, and perhaps growing lack of, dispersion among stocks.
A recent article in The Wall Street Journal, entitled “The Herd Instinct Takes Over” (July 12), fully confirms this suspicion. The sub-title of the article is: “Component stocks’ correlation to S&P 500 at highest level since ’87 crash,” and the article goes on to say that the correlation has recently hit 83%, a remarkable number when one considers that it has averaged 44% since 1980.
Now we’d just like to emphasise the next point, since it relates to one of our favourite subjects — ETFs:
The growing popularity of exchange-traded funds, among other things, has leveled the returns of individual stocks and sectors. By maintaining (or freezing) over- and undervaluations, this phenomenon has become a nightmare for stock and sector pickers.
And, while it is true that this phenomenon raises hopes that, at some point in the future, large profits will be able to be made by exploiting these inefficiencies, such may not be the case for quite some time. In fact, structural changes such as the proliferation of exchange-traded funds and super-rapid computer-based bloc trading, activities that are totally unconcerned with valuation metrics and/or long-term trends, are still taking place and there is little or no prospect of this development coming to an early end.
And that is the key thing here. ETFs and high-frequency traders simply don’t care about valuations.
Joe Saluzzi — a prominent critic of HFT — explained the systemic implications of this as follows to FT Alphaville:
It means you can’t trust any valuation. I could have valued a subprime CDO better on May 6 than any equity. And it’s almost the same thing all day long. Valuations and prices have been divorced for a while. Just look at the volatility. It’s not like traditional trading. No wonder there are such increased correlations.
It’s a point also seized upon by Russia Today’s Max Keiser.
(And yes we know he’s slightly unconventional, but he makes a hugely valid point).
What these dynamics essentially mean, he says, is that Adam Smith’s invisible hand has for the time being, been handcuffed. Effective price discovery may have been lost.
Hence real deflation may also be actively disguised in the market.
Or as he puts it:
Price Discovery — the result of buyers and sellers simultaneously transacting in the market with the result being Adam Smith’s ‘Invisible Hand’ — means goods and services move around in the economy at mutually advantageous prices for all. It also means that everyone holding similar items have a benchmark or ‘price signal’ that tells them what these items are worth.
It is my thesis that the inflation, deflation debate is flawed because we no longer have reliable price signals. The overwhelming domination of program trading on various exchanges has fundamentally changed the way prices are created and represented in the economy. All ‘efficient market’ theories are dead.
Happy trading!
Related links:
Stock picks tough as asset paths correlate - FT
Did Socgen use ETFs to liquidate Kerviel positions? – FT Alphaville
Global correlation makes it harder than ever… – Business Insider
Is someone liquidating GLD? – FT Alphaville
This entry was posted by Izabella Kaminska on Wednesday, July 28th, 2010 at 9:40 and is filed under Capital markets, Commodities. Tagged with Adam Smith, Algorithmic trading, correlations, Deflation, Equities, etfs, Hft, High frequency trading, index trading, invisible hand, price discovery, valuation.
Market Data 
Markets 
AI
