Referee Finance

Home of the Financial Intelligence Lab

  • Increase font size
  • Default font size
  • Decrease font size

What Happened to the Quants in August 2007?

E-mail Print PDF

What Happened to the Quants in August 2007?

Amir Khandani and Andrew Lo

in this paper, the authors revisit the recent events that have hit the small world of quantitative hedge funds, especially the ones involved in equity long short strategies and equity market neutral strategies (EMNs). Large unexpected losses occurred between August 6 and August 9, 2007 despite relative small market movements. The strategies rebounded on August 10, 2007 but after several hedge funds had to deleverage.

To illustrate these events, the authors use a simple quantitative mean reversion strategy introduced in Lo and MacKinlay (1990): every day, they buy US stocks that have outperformed and sell stocks that have underperformed in the previous day. The strategy performs because it provides liquidity to the market (for instance, when everybody wants to sell a stock, the price comes down and the strategy buys the stock, thus providing liquidity). The authors also assume no transaction costs, inflating the expected returns of the strategy. The time period is Jan 3, 1995 to August 31, 2007.
Over time, the strategy delivers a lower return and lower information ratio. To maintain a high return, the strategy needs to be leveraged (leverage is defined as the total gross value of investments divided by the equity capital). With a leverage of 8 to 1, the losses in August 7th-8th-9th, 2007 were significant at around  -28%. The strategy regained 24% in August 10, 2007.
The authors compare this episode with August 1998 during which hedge funds (LTCM) were also involved. Using data from the TASS database (that includes live hedge funds but also hedge funds no longer active) they do not find the same pattern of losses among EMNs hedge funds. They argue that the difference is due to fewer multi-strategy hedge funds in 1998 relative to 2007, the enormous growth of assets in the long/short equity space, the increased use leverage and the reduced liquidity caused by too many funds investing in the same strategies.
They then explore the unwind hypothesis. The pattern of losses on August 7th and 8th seems to be explained by the unwinding of some large EMN portfolios since stock indexes did not under perform during this period. The portfolios could have belonged to some multi-strategy hedge funds or proprietary trading desks that had to reduce risk to cover losses in credit or mortgages or face margin calls.
Funds with the same factor exposures as the unwinding funds would also be negatively impacted and have to reduce their exposure and deleverage. The authors think that the factors, such as value, earnings quality, or derived from financial ratios are today widely shared among quantitative and even traditional funds. The rebound of August 10th is furthermore taken as an indication that the strategies were not to be blamed but rather that it was an incidental liquidity event caused by the unwinding similar to a fire sale.
 They also examine the liquidity of the strategies by performing some autocorrelation analysis of the returns. High autocorrelation would be a sign of low liquidity. They find that indeed autocorrelation has increased and therefore liquidity has decreased over time. They also look at another pattern of correlation: the correlation between different hedge fund strategies between 1994 and 2000 compared to between 2001 and 2007. Correlation has clearly increased, especially with the multi-strategy, indicating that systematic risk among hedge funds is now greater than ever.
The authors conclude on some recommendations for regulation of hedge funds. Given the potential impact of hedge funds on systematic risk, they propose a Capital Market Safety Board that would study and closely monitor hedge fund activities.
 
Khandani, Amir E and Lo, Andrew W., "What Happened to the Quants in August 2007?" (September 20, 2007). Available at SSRN: http://ssrn.com/abstract=1015987

Lo, A. and C. MacKinlay, 1990, “When Are Contrarian Profits Due to Stock Market Overreaction?”, Review of Financial Studies, 3, 175-206.

 

Newsflash

You can follow us on Twitter  @fintlab and facebook