Oliver Wendell Holmes’ 1858 poem “The Deacon’s Masterpiece” describes a perfected one-horse “shay,” a highly engineered carriage designed so that the failure of a single part could not cause an untimely breakdown. By eliminating the weakest links, the carriage performs flawlessly, at first. But the shay does not have a happy ending. It suddenly disintegrates with all the parts failing at once, leaving its rider dazed atop a pile of rubble. Holmes—the father of the eminent U.S. Supreme Court Justice—mocked the pseudo-scientific efforts of the overeducated Deacons of his day to engineer impractical structures.
In our domain, the Deacons are quants (financial engineers) and their Masterpiece is an overly complex quantitative investment strategy. The second week in August1 marks the four-year anniversary of the quant meltdown of 2007. While the events of 2008, including nationalization of Fannie Mae and Freddie Mac, the failure of Lehman, the bailout of AIG, creation of TARP, etc., have rightly received more scrutiny, August 2007 foreshadowed the global financial crisis and deserves more attention by today’s investors. Analyzing the underlying causes of the quant meltdown helps reveal the perils of complex quantitative strategies and highlights the difference between transparent and rules-based alternative beta strategies such as the Fundamental Index™ methodology and newer optimized approaches.
The Quant Meltdown
During the week of August 6, 2007, many large and previously successful hedge funds were forced to de-lever their portfolios and liquidate commonly held securities, resulting in simultaneous drawdowns of 30%, 50%, or worse. To make matters worse, these investments had been sold as risk-controlled and uncorrelated to the market. Khandani and Lo concluded that a “… deadly feedback loop of coordinated forced liquidations leading to deterioration of collateral value took hold during the second week of August 2007, ultimately resulting in the collapse of a number of quantitative equity market-neutral managers, and double-digit losses for many others.”2 Quantitatively managed enhanced index funds experienced similar simultaneous traumas, though the magnitude of losses was lower due to the lack of leverage.
None could have forecast the precise timing of the sudden liquidation of a large trading desk that catalyzed the quant meltdown of August 2007.3 But should we have been surprised that those funds failed catastrophically? After all, the quant funds of 2007 shared the same structural flaws as the highly engineered financial trading strategies that caused the stock market crash in 1987 and the implosion of Long-Term Capital Management in 1998.4
Inside the Black Box
To help avoid future meltdowns in our portfolios, investors need to look inside the black box of quant strategies. Simply put, quants use advanced statistical methods and high frequency data to create complex financial models. With experience, skill, and some luck, a few of these models successfully forecast future security price changes. In the short term, these strategies provide consistent trading profits and gather assets into associated funds.
Consistent profits can hide inherent risks, however. Most complex quant strategies have proven to be unstable. Markets evolve in response to the creation and adoption of these strategies. At first, the identified predictability in security price movements is reinforced as funds using the quant model, along with similar funds using similar models, begin buying and selling the same securities. Early success and clever marketing attracts large flows into the funds, which, in turn, drives the prices of securities held by these funds to unsustainable extremes. The result is a brittle price structure awaiting the inevitable crisis.
Leverage creates an even more toxic brew. In the years leading up to the quant meltdown in August 2007, the same models used to manage enhanced index funds (with relatively low tracking errors and high information ratios) were increasingly employed to create levered absolute return-oriented long/short funds. To facilitate the use of leverage, risk models were used to minimize country, sector, and other common factor risks. With all the risk seemingly wrung out of the strategy, ever more capital and leverage were applied.
Paradoxically, quantitative risk management was part of the problem. While risk models are useful tools for measuring risk, using models to tightly control risk is misguided and dangerous. Because no model is, or ever can be, a complete description of the complex dynamic system that is a market, all risk models fail to capture some risk. By eliminating all of the risks measured by their models, the quants transferred the risk in their funds into the areas their models could not measure and they did not understand.
Quant strategies produce remarkable profits in the early stages. But inevitably, the process becomes unstable and often ends with violent illiquidity events, such as the stock market crash of 1987, the Long-Term Capital Management-induced crisis in September 1998, and the quant meltdown in August 2007. The largest losses in those episodes were suffered by the most recent investors who were attracted by dazzling early performance records. Instead of consistent profits, the large later investors were stuck with shocking losses realized during fund liquidation as investors fled from the imploding strategies.
As Harry Markowitz stated in the middle of the crisis, “…the layers of financially engineered products… combined with the high levels of leverage, proved to be too much of a good thing.”5
Fundamental not Quant
Only four years after the last quant meltdown, over-engineered quantitative investment strategies are back. The latest incarnation is complexly optimized alternative betas. Such strategies attempt to engineer indices with the lowest possible volatility, the highest possible Sharpe ratio, or the maximum possible diversification. The more complex the engineering, the better the model performs in the backtest. As investors begin to adopt such narrow indices, early performance may be rewarding. Fund inflows will create buying and selling pressure on the same narrow set of securities. This pattern will create a brittle price structure resembling the Deacon’s Masterpiece and set the stage for the next wreck.
Recognizing the trouble with quants, should we eschew quantitative study of security price movements and abandon risk models? Of course not! Advanced statistical methods are invaluable tools to help us understand securities markets. Likewise, risk models help us measure, monitor, and decompose the risks in our portfolios. For example, with regard to the Fundamental Index methodology, we use quantitative methods to demonstrate how and why companies with low market prices relative to fundamental measures of company size provide higher returns than companies with high market prices relative to fundamentals. We use risk models to examine whether and how value priced companies have different risk characteristics than other companies.
The Fundamental Index methodology is far less complex and therefore less risky than a highly engineered quant model. Fundamental weights are simple, logical, and stable. Fundamental Index portfolios are transparently constructed and broadly diversified. The Fundamental Index strategy uses the time-tested technique of systematic rebalancing to capture the long-term return premium offered by the market’s excess volatility.
The following passage from Holmes’ poem descries the end of the one-horse shay. But it could easily be a fitting narrative to the quant strategies during that fateful week in August 2007.
“…it went to pieces all at once, —
All at once, and nothing first, —
Just as bubbles do when they burst.
End of the wonderful one-hoss shay.
Logic is logic. That's all I say.”
The performance of Fundamental Index strategies may break down occasionally over the long winding road to investment success, just as traditional index funds can create some nasty surprises. However, these setbacks are just that and eventually the Fundamental Index strategy’s simple and stable rebalancing process puts the portfolio back on track. That’s our logic. What do you say?