ETFs crashed by flash

Still, 68% of the trades that were reversed that day concerned ETFs. “At the end of that time span, a lot of people, especially individual investors were perhaps saying that ETFs are not for me,” says Seddik Meziani, professor of finance and economics at Montclair State University.

Speaking at the Exchange Traded Forum organized by Radius Financial Education last week, Meziani characterized the U.S. Securities and Exchange and Commodity Futures Trading Commission investigation into the flash crash as “descriptive.” It plots how sells by one mutual fund firm in the E-mini market —E-minis are S&P 500 futures priced at $50 per contract on the Chicago Mercantile Exchange—unleashed a river of algorithmic trading that burst the dam, as the market was first flooded with buys and sells and then, just as suddenly, there was no liquidity at all.

In that respect, despite that fact that trading in ETFs now accounts for 25% of the volume on U.S. exchanges, ETFs “were more like a casualty, like many other things, than they [were] the reason for the flash crash,” Meziani says.

The liquidity lesson
And quite a casualty they were, at least for those itching to sell or buy. “ETFs are supposed to be tracking faithfully, at least the most liquid ones, they are supposed to be tracking very faithfully the underlying portfolio,” Meziani notes. That day, it didn’t happen.

It wasn’t because of a general market sell off. It wasn’t because, with debt problems looming in Europe, “someone spotted smoke and screamed fire.”

The E-mini sale, 75,000 contracts leveraging $50 times the value of the S&P 500, was a normal trade, except that it occurred within the space of 20 minutes instead of five or six hours.

High-frequency traders —professional day traders who are in and out of the market in milliseconds—pumped up the volume on their own black-box signals. And then they exited.
These types traders are often saluted for providing liquidity to the markets, but this time, all they supplied was volume before retreating behind their black boxes.

As a result, the bid/ask of some blue chip shares gapped out from a few pennies on the spread to $5. “That tells you there’s a problem with liquidity right there,” Meziani says.

ETFs didn’t escape the slaughter, creating problems because ETF providers “were saying, especially for the most important ETFs, you can buy and sell it at the same time of the day. Prices are refreshed in a matter of seconds.”

The prices that were refreshed were disastrous—or once-in-a-lifetime opportunities, depending which side of the trade the investor was on.

Stock market “tsunami”
But it’s more problematic than that, Meziani argues. “ETFs suffered more than other financial products compared with the stocks and bonds comprising the underlying portfolio.” There’s an asymmetry here. While ETFs comprise 25% of market volume, they accounted for 68% of the reversed trades.

The arbitrage mechanism, whereby designated brokers profit from narrowing the difference between an ETF’s price and its underlying components to almost nil, broke down. ETFs did not track, during those 20 minutes of crisis, the underlying net asset value of the portfolio. Meziani likens it to a tsunami. Except it’s the opposite. A tsunami washes over shores with huge waves of water, leaving destruction in its wake. A stock market tsunami comes like a wave—tremendous trading volume —but dries up the waters of liquidity, leaving destruction in its wake.

That, he says, “should not happen for the most liquid ETFs.”

Kathleen Morarity, a partner at Katten Muchin Rosenman LLP in Chicago, who was involved in the launch of the first ETF —SPY —suggests that part of the problem in the U.S. is that market rules have changed. Specialists are no longer in place. Formerly, they were required to step in to maintain an orderly market, risking their own capital along the way to balance sellers and buyers. Instead, market-makers are under no particular obligation, and may simply step aside when the weather turns.

That is what Meziani thinks happened on May 6: market-makers withdrew.

First, the bids dried up. Then, the asks dried up. Where normally there would be a bid/ask spread of 3.5 basis points, it quickly became 700 basis points—3.5 cents versus seven bucks, Meziani explains.

Those who didn’t trade that day were unaffected. Those who did, well, they tested the liquidity of the market. “Those who did not panic, they basically left the market unscathed,” says Meziani.

There’s a lesson here for ETF investors. Liquidity isn’t daily volume. It’s the bid/ask spread.

Scot Blythe is a Toronto-based freelance financial journalist.

(11/02/10)