Wednesday, December 8, 2010

The Crash of 2:45 - An Autopsy

This article appeared in Corporate page of The Edge Malaysia, Oct 30 – Nov 6,2010

By Jasvin Josen

High Frequency Traders. Flash Trades. Algorithmic Trading. – these uncommon phrases have suddenly become familiar in the news since the May 6 “flash crash” that occurred at 2:45 pm and ended twenty minutes later.

It later turned out that the crash was actually sparked by a $4.1bn sale of stock index futures by a single institutional investor. On Oct 1, the U.S. Securities and Exchange Commission (SEC) and the U.S. Commodity Futures Trading Commission (CFTC) provided an in-depth analysis of what actually happened that day, in a 104 page report (referred as “the Report”)

This article will describe the chronology of events which brought on the biggest one-day point decline (998.5 points) on an intraday basis, in Dow Jones history.

9:00 A.M.
On the morning of May 6, 2010, New Yorkers woke up to adverse news on the European debt crisis that had been looming all afternoon across Europe. The market was demanding higher premiums for bearing additional risk. CDS premiums rose on some European sovereign debt while the Euro was suffering a downward pressure.

1:00 P.M.
It is no surprise that the despair spread swiftly to the U.S. markets. By 2.30 pm, the S&P 500 Volatility Index (VIX) increased 22.5%, while investors shifted towards “quality” investments such as gold and Treasuries. Selling pressure had pushed the Dow down about 2.5%.
The risk began to manifest itself in individual equities on the New York Stock Exchange (NYSE) and Liquidity Replenishment Points (LRPs) began to trigger far above average levels.
LRPs are mini circuit breakers that the NYSE applies to individual equities during unusually volatile markets. These are not the standard circuit breakers that were put in by the SEC after the 1987 Black Monday crash - that would have required a 10% downward move in the Dow Jones.
When a share price falls rapidly (1% to 5%; the limit varies for different stocks and daily trading volume), the LRP kicks in to temporarily convert the trading of a stock from an automated market to a manual one. Trading on the NYSE for that stock will “go slow” and automatic executions will cease for a time period ranging from a fraction of a second to a minute or two, to allow the manual market to contribute additional liquidity before returning to an automated market.
According to the Report, that day, between 2:30 p.m. and 3:00 p.m., more than 1000 securities triggered LRP events lasting more than 1 second, compared to a “normal” day average of only 20-30 such events.
At about the same time, the E-Mini S&P 500 futures (E-mini), the most active stock index instrument plunged by 55% from early-morning levels. Another very active stock index instrument, the S&P 500 SPDR exchange traded fund (“SPY”), also declined by 20%. Chart 1 and Chart 2 show the volume and price movements of the E-mini and SPY ETF that day.

The E-Mini is often accepted as the primary price discovery product for the S&P 500 Index. In other words, investors perceive these future contracts to transmit information into prices for the S&P 500 index.

Chart 1 :  E-Mini Volume and Prices on May 6, 2010
 Source:  Findings Regarding the Market Events of May 6, 2010, by SEC and CFTC, Sep 30, 2010

Chart 2: SPY ETF Volume and Prices on May 6, 2010
 Source:  Findings Regarding the Market Events of May 6, 2010, by SEC and CFTC, Sep 30, 2010

2.32 P.M.
Against the setting of unusual high volatility and thinning liquidity, at 2:32 p.m., a large “Seller”, a mutual fund complex initiated a program to sell 75,000 E-Mini contracts (valued at approximately $4.1 billion) as a hedge against the risk of a market downturn.

Generally, a customer makes a choice of how much human judgement is involved in executing a large trade. He could engage an intermediary to manage the position. He could also choose to manually enter orders into the market. Or, he could execute a trade via an automated execution program. The algorithm in the program will have in-built conditions to execute trades according to price, time and/or volume.
The Seller chose the third option. Its automated execution algorithm (Sell Algorithm) was programmed to feed orders into the June 2010 E-Mini market to target an execution rate of 9% of the previous minute trading volume, but without regard to price or time.  The Sell Algorithm executed the sale in just 20 minutes. Without doubt, it appears that the Seller wanted the order to be completed as quickly as possible, at any price. Or could it be an erroneous condition in the algorithm? The Report falls short in probing into this aspect.

The Report identifies High-Frequency Traders (HFTs) and intermediaries (i.e. market makers and brokers) to be the likely buyers of the initial batch of orders submitted by the Sell Algorithm. As a result, these buyers built up temporary long positions.

HFTs are trading firms that use high speed systems to monitor market data and submit large numbers of orders to the markets. They utilize quantitative and algorithmic methodologies to trade with the exchange. We will explore this subject further in the next article.

2.41 P.M.
Before the 75,000 contracts were fully taken up, the HFTs then aggressively began to sell E-Mini contracts in order to reduce their temporary long positions. This is consistent with the HFTs’ typical practice of trading a very large number of contracts, but not accumulating a significant inventory in either direction.
As they quickly proceeded to sell their contracts to the market, they only found other HFTs, who in turn bought and quickly resold the contracts, creating a “hot potato effect” that created huge volume.
A very useful lesson pointed out by the Report - Especially in times of significant volatility, high trading volume is not necessarily a reliable indicator of market liquidity.
The Sell Algorithm, which had not completed the sale of the 75,000 E-mini contracts, responded to the increased volume by increasing the rate at which it was feeding the orders into the market, even though the initial orders were possibly not yet fully absorbed.
In the five minutes before 2:45 p.m., prices of the E-Mini fell by more than 5% and prices of SPY declined over 6%. Buy-side depth was already a serious issue.

2.45 P.M.
Trading on the E-Mini was paused for five seconds when the Chicago Mercantile Exchange (“CME”) Stop Logic Functionality was triggered. During that split-second, sell-side pressure in the E-Mini somewhat eased and buy-side interest returned. When trading resumed at 2:45:33 p.m., prices stabilized and shortly thereafter, the E-Mini began to recover, followed by the SPY.
The Sell Algorithm continued to execute the sell program until about 2:51 p.m. as the prices were rapidly rising in both the E-Mini and SPY.
The second liquidity crisis occurred in the equities markets. The Report confirmed that many market liquidity providers temporarily paused trading in reaction to the sudden price declines observed in the E-mini and SPY markets. Based on their respective individual risk assessments, some market makers widened their quote spreads, others reduced offered liquidity, and a significant number withdrew completely from the markets.

The severe dislocations observed in many securities were short-lived. As market participants had time to react and verify the integrity of their data and systems, buy-side and sell-side interest returned and an orderly price discovery process began to function.

3.00 P.M.
By this time, most securities had reverted back to trading at prices reflecting true consensus values.

4.00 P.M.
After the market closed, the exchanges jointly agreed to cancel (or break) trades that were more than 60% away from their 2:40 p.m. prices.
As liquidity completely evaporated in a number of individual securities, some participants with instructions to sell (or buy) found no immediately available buy interest (or sell interest). This resulted in trades being executed at irrational prices as low as one penny or as high as $100,000, as a result of so-called “stub quotes”.

When a market maker’s liquidity has been exhausted, it may (on some markets) submit a stub quote – an offer to buy or sell a given stock at a price so far away from the prevailing market that it is not intended to be executed, such as a penny (buy order) or $100,000 (sell order). This is to comply with its obligation to maintain a continuous two-sided quotation.

The next day, Tim Geithner, Treasury secretary, spoke to Mary Schapiro, chairman of the SEC, and other regulators to discuss the issue. “We cannot allow technological problems, regulatory loopholes or human blunders to spook the markets and cause panic.”
In the next article, we explore the lessons learnt from this crash and indeed if it is possible to avoid spooking the markets again.

No comments:

Post a Comment