From the August 01, 2007 issue of Futures Magazine • Subscribe!

Trading London and European hours

Globalization has created four distinct blocks: Asia, Europe (dominated by London), the United States and the Middle East. With the Internet allowing effectively instantaneous communication between these blocks, the retail market has become relevant; a new and vibrant independent market force.

Everything from the European Central Bank’s news conferences broadcast live via your computer’s media player to the words of governors of the U.S. Federal Reserve Bank posted at the precise time of a speech means that the difference between retail and institutional is less informational and more logistical.

In other words, we all now have access to the same information at the same time. It’s just the processing and execution of ideas based on that information that sets the professionals apart.

Beyond the number-crunching capabilities, institutional accounts are defined by their capability to link markets and products so that a single click can launch multiple orders across different exchanges and continents, resulting in a deluge of orders followed by a lull. Many times, it’s just prior to that lull when retail traders identify the right time to put on a trade — but only to watch the market suddenly become inactive.

This situation is one example of what is known as the liquidity mirage.


Since October 2006, three distinct liquidity mirage situations have played out in the markets. The first occurred on Globex in the E-mini S&P 500 December futures contract and spilled over into the big S&P 500 as well as the Nasdaq 100 and Russell 2000 contracts.

The trade occurred at 6:26 p.m. (EST) on the Sunday evening of Oct. 22, 2006. This was just as the new week began trading in the Asian hours. The bulk of the move began as the S&P 500 December futures went from 1375.70 to 1380.80, with a major buy order lifting all the offers.

Once the big S&P 500 contract traded 1380.80, it represented a trade 0.6 above the previous recent high of 1380.20. This triggered orders from large buy-stops resting above this key level. This created a domino effect and a surge of nearly 25 handles from the original open, taking the contract to 1399 (see “To the moon,” below).

Due to orders being queued on a first-come, first-served basis, the resting buy-stop orders would have been filled on the high of the move as the original large market order had precedence over the buy stops. This liquidity mirage scenario demonstrated how the stops failed to protect those trying to limit losses. Instead, they were likely filled with massive slippage on the high of the move after the initial buy order was satisfied.

By 6:31 p.m., the world had woken up to the trade, and the market was back at 1377.40. The market traded contracts with a nominal value of approximately $350 million, with a possible transfer of actual equity of about $25 million from losing accounts to the winners — in most cases those were also the accounts in tune to the earlier moves in the E-mini.

Relative to the E-mini, $350 million represents a fraction of the un-leveraged hedge fund community and does not affect the aggregate bottom line in a meaningful way. But consider the effects of the move on the individual retail level. The $1,250 per contract multiplied by 5,000 contracts (the original order was a size equivalent to 5,000 E-mini contracts) equals $6.25 million, or $12.5 million when the entire move, up then down, is considered. This represents for many retail accounts a 225% ratio of the minimum margin required by many brokers, or 25% of many retail account balances.

This move could have potentially wiped out 50% of many accounts.

The $12.5 million, therefore, may have been small in the grand scheme of things, but the effect on undercapitalized or overleveraged accounts was significant.

The second liquidity mirage event occurred in the Chicago Board of Trade’s (CBOT) e-CBOT market in the December 2006 gold contract.

On Nov. 12, 2006, the contract traded from $626 per ounce down to $608.80, then straight back to $626 within two minutes, just prior to the open of the U.S. session.

In the S&P 500 example, the cause was a market buy order for 5,000

E-mini equivalent lots. In the mini gold futures, it was a sell stop for 100 lots. Either way, both orders, when launched, took precedent over all other potential resting orders because they automatically became market orders. One difference is the gold order appeared confined to that contract while the S&P 500 order impacted the Nasdaq 100 and Russell 2000 futures.

The third incident occurred on Feb. 12, 2007, on Euronext Liffe in the Cac 40 March contract. It opened at 7 a.m. (GMT) or 2 a.m. (EST) and within one minute had plunged almost 100 points from the previous settlement, representing €1,000 ($1,341) per contract with a little more than 200 contracts changing hands on a market order (see “French slip,” below).

Page 1 of 3 >>
comments powered by Disqus
Check out Futures Magazine - Polls on LockerDome on LockerDome