Black Monday 25 years later: Are algorithms any less of a threat?

Mon Oct 15, 2012

Email a friend
  • To include more than one recipient, please seperate each email address with a semi-colon ';', to a maximum of 5 email addresses

Today's post Dodd-Frank environment contains a number of supposedly more stringent controls on the markets.


By Jason Scharfman

This month marks the 25th anniversary of Black Monday, a tragic event when fortunes were lost for most, and made for some. When an incident such as the Dow plunging on October 19, 1987occurs, people often demand swift action. The ensuing change can be quite public and broad. Despite the media and regulatory fanfare that occurred after reports of algorithmic trading gone wild on Black Monday, one can still quite reasonably pose the question: Are markets more protected today than we were in 1987? Such questions certainly hold more weight when framed in the context of the 2008 global financial crisis and even more recent algorithmic trading disasters, such as the Knight Capital fiasco.

The post-mortem on Black Monday has a number of lessons about operational controls, risk management and algorithmic trading, which still ring true today. For starters, much in the same way it may be easy to blame hedge funds for the most recent financial crisis in the headlines, it is easy to blame computer-based trading for Black Monday. We now know that there were many factors that contributed, including poor monetary policy and the decline of the global bond markets. In reflecting on the 25th anniversary of this event, investors and regulators should not look to algorithmic trading as a scapegoat while ignoring these other causes.

That being said, the risks surrounding algorithmic trading, or "program trading" as it was known in 1987, are still alive and well. Even after Knight Capital, algorithmic traders, be they humans or robots, continue to toy with markets. Recent evidence of this includes reports of algorithms testing markets by requesting hundreds of quotes in milliseconds and accounting for over 4% of all quote traffic in the U.S. stock market during the week of September 30, 2012, alone, according to market data firm Nanex.

From a regulatory perspective, as compared to 1987, today's post Dodd-Frank environment contains a number of supposedly more stringent controls on the markets. Additionally, the global banking system is purportedly more insulated from total financial calamity via enhanced minimum capital requirements and cash buffers in case of emergency. But do these measures make investors and markets safer?

What investors and fund managers have learned over time is that regulation is not in and of itself a panacea. Regardless of how much oversight is in place, the risk of a crash caused by an algo can't be completely removed. Just like the markets, these types of risks are fluid and constantly changing. When one type of risk is regulated, a new one that rule makers never thought of will crop up. Strapped for resources and under-informed, the regulators cannot effectively keep up with the new speed of markets. Fund managers and investors must also be key partners in attempting to prevent such crashes. Even if the next Black Monday is around the corner, is there anything investors can do about it? The short answer is No individual investors cannot generally move markets but, investors can take measures to limit their exposure to such black swan events in several ways.

Non-investment related risks, commonly referred to as operational risks, also played a key role in causing Black Monday. While regulators and institutions have implemented some measures to protect against these risks, ultimately it is up investors to vote with their dollars and not allocate capital to fund managers that do not have solid algo trading control frameworks.

These controls should be in place in a number of different areas. Before an algorithm can go rogue it has to be created, tested and put into production. Good fund managers that employ high-frequency algorithmic trading strategies understand that this path to live production is fraught with opportunities for a wide variety of issues that could cause an algo to go haywire. Additionally, investors should take measures to understand what controls a fund manager has in place to shut down an algorithm if it goes rogue. For example, are fund personnel automatically alerted if certain limits are breached? Do the right people have authority to shut down an algorithm if it enters a nosedive?

Even if Black Monday Part 2: The Return of the Rogue Algo comes around, investors can protect their money with fund managers via detailed operational due diligence. These types of reviews provide insight into how seriously a fund monitors the controls in rolling out and monitoring algorithms. Without such due diligence, investors run the very real risk of being washed away in the next robot-driven downward market.

Jason Scharfman is managing partner of hedge fund operational due diligence company Corgentum Consulting.

ISSN: 2151-1845 / CDC10004H

Popular Searches on HFI