by Verne O. Sedlacek, President and CEO, Commonfund
On April 22nd the Deepwater Horizon drill platform sank in 5000 feet of water setting off the tragic series of events that has led to the daily reminder of the limits of technology. As dependence on technology increases, it becomes increasingly clear that there are no fail safe measures to guarantee that disasters won’t occur at some point. We have seen this time and time again. Whether it is the BP oil spill, hurricane Katrina or the space shuttle, increasing levels of complexity cannot be fully mitigated by increasing levels of technology.
Only 14 days after the disaster in the Gulf, the financial markets had its own blow out, the now infamous “flash crash”. As we all know by now, the market dropped 999 points in a matter of minutes and then just as quickly bounced back. During those 20 minutes $862 billion of value was erased. We were all left scratching our heads at the breathtaking events of that day after we got over the fear that the world was ending.
The immediate rumor was that a “fat fingered” trader put in a 1,000,000 instead of a thousand. That was quickly followed by theories about options or futures or ETFs or cyber terrorists. At this point almost two months later, there has been no official explanation from the exchanges or the regulators.
The seeds of this financial system “blow out” were, in fact, sown years ago. As we wrote to investors in the fall of 2003 (“A River Runs Through It”), the Grasso and specialist scandals at the New York Stock Exchange were likely to bring dramatic change to the operations of 11 Wall Street, and not all to the benefit of investors. The push to reduce the cost and time required for execution brought on by the emerging NYSE changes were the proverbial chaos theory butterfly in Philadelphia which led to the weather change in Boston. The demutualization of the NYSE to a shareholder company along with the significant growth in market share going to electronic systems of over 50 different venues (NASDAQ, ARCA, Bats, etc.) resulted in faster execution time and cheaper execution costs. It did however, as predicted, lead to the demise of the specialist system which had at least some responsibility to maintain orderly markets through the book building and the application of capital.
In 2003 I wrote, “While I believe that the electronic system works in stable times, it will tend to lead to increased volatility and bid/ask spreads when markets are under stress.” And there is little doubt that in the last several years the traditions of the NYSE have changed dramatically.
- The demise of the specialist system which had over ninety specialist firms in 1995 is now down to five;
- The percentage of volume flowing through the floor of the NYSE has dropped dramatically in the last decade from more than 50 percent to about 17 percent today;
- Floor staff has dropped in half in just the last five years from 2400 to 1200.
Where has this volume gone? It has gone to the electronic systems which depend not on humans to intervene but on computer systems.
So what happened on May 6th 2010 was almost inevitable. Interestingly it really started at the close on May 5th when there were least 280 public market securities that couldn’t be priced at the end of the day because of what appeared to be order imbalances. Those securities were finally priced around 8 pm. The market opened under significant stress with the crisis in Greece; however, it seemed to settle in the early afternoon and then as they say “all hell broke loose”. The collapse appeared to be triggered when the NYSE went to “slow mode” (a technical term that slows down execution from seconds to half a minute to a minute). This relief valve is intended to allow the specialist to build books to pick up bids at lower prices – ideally putting a floor at some price by generating interest or putting in specialist capital to provide for an orderly market.
When the NYSE went into “slow mode” the orders for execution were automatically diverted away from the NYSE (By design, the execution systems used by brokers require that the systems find the best bid for a buy (lowest) or the best offer (highest) for a sell.) When orders were routed away from the NYSE they hit electronic systems and automatically executed. The problem of course is that the deepest “book” was on the floor of the NYSE. This issue is compounded by what is now known as high frequency traders; these funds place bids and offers in the market with the expectation that they will execute buys and sells and make small profits on each trade. (In the old days funds couldn’t do this because the execution time was too long and the costs were too high.) Estimates are that these organizations account for two-thirds of the current volume on the exchanges. The high frequency traders, which provide capital to the market during normal times, pulled all of their bids almost instantaneously when the market started to crumble. So with no human intervention, the doomsday machine automatically executed trades at the best bid which may have been as low as a penny. Thus you had some positions that went to a cent in that 20 minute span.
In the aftermath, 20,800 trades were unwound that were at least 60 percent away from the market price when the plunge began. 59 percent of the cancelled trades took place on NASDAQ, 24 percent by Arca (the electronic exchanges owned by the NYSE), with the remainder spread across the other electronic platforms. There were no cancelled trades on the floor of the NYSE.
While the regulators and the exchanges have not provided a detailed explanation of the events of May 6th, there have been some proposals to rectify the situation. The first is to put circuit breakers around individual equities in the S&P 500 that are similar to market circuit breakers as a whole. For example should a stock drop a certain percentage (say 10 percent) over a five minute period, trading in the stock would be halted for five minutes to allow for price determination. The exchanges have also proposed to write the rules for breaking of trades that are the result of transactions that take place beyond certain boundaries.
So far it appears that they have not fixed the underlying problem of having different rules for the fifty odd different exchanges.
What does all this mean for institutional investors? I think as uncomfortable as it sounds we just have to be prepared for the trade-off that the industry bargained for: higher volatility for lower transaction costs and close to instantaneous execution. This means as long-term investors we have to be better prepared for more gut wrenching rides. This includes intraday, between days, weeks and months. Logically, as long-term investors short-term volatility should not bother us; yet we must overcome our very human characteristic of being acutely sensitive to short-term events.
Second rebalancing becomes a more important value-added aspect of long-term management. Higher levels of volatility should provide for a higher positive impact of rebalancing. I am not advocating that institutions rebalance every day but monthly rebalancing between liquid assets should improve long-term returns. We must of course overcome one of the other demons of human nature: buying something that is down and selling something that is up.
On a more technical level, ensure that your broker is ready to pull orders at a moment’s notice if markets enter a blow out state. If you use outside managers ensure that they have the “fail safes” to avoid selling into a terrible sink hole. In addition, stop loss orders should be a thing of the past. They set up the seller to get an execution far below what might otherwise be expected.
The world is now so dependent on technology that I am not sure we are capable of operating without it in all aspects of our professional and personal lives. This week we lost internet service at home and my family felt like they had entered the Stone Age. Yet, this technology dependence does have a dark side for which we must be prepared. We saw it in the blow outs in the Gulf and in the markets. Progress will not stand still and that is a great thing; however, from time-to-time we are going to experience the unexpected events.
Chaos theory postulates that long term predictions are impossible, due to the fact that widely divergent outcomes can result from very small differences in initial conditions. If you ever need a reminder look at the five day weather forecast. Thankfully, as long term investors our degrees of freedom in trying to forecast the future are much greater than those of the weatherman. Yet it is a constant struggle with our human tendencies to manage through unpredictable and volatile markets. So we must be guided by sound policy and governance, so when the butterfly flaps its wings we can withstand the storm.
Statements concerning Commonfund Group’s views of possible future outcomes in any investment asset class or market, or of possible future economic developments, are not intended, and should not be construed, as forecasts or predictions of the future investment performance of any Commonfund Group fund. Such statements are also not intended as recommendations by any Commonfund Group entity or employee to the recipient of the presentation. It is Commonfund Group’s policy that investment recommendations to investors must be based on the investment objectives and risk tolerances of each individual investor. All market outlook and similar statements are based upon information reasonably available as of the date of this presentation (unless an earlier date is stated with regard to particular information), and reasonably believed to be accurate by Commonfund Group. Commonfund Group disclaims any responsibility to provide the recipient of this presentation with updated or corrected information.