A Better Culprit

by Larry Levin

When the final (and ludicrous) SEC report was released regarding the Flash-Crash and blame was laid at the feet of Waddell & Reed, I immediately debunked it. Its order of 75,000 was said to be the culprit, yet that size routinely trades throughout the day - on heavy volume days - yet never causes a Flash-Crash.

A reporter from Barron's has a better culprit.

The full story can be read here

The Real Flash-Crash Culprits

MEET THE FLASH-CRASH scapegoat. A report by regulators blamed May's spectacular market break on a single trade by a single "mutual fund complex" identified in the press as Waddell & Reed.

This was as ludicrous as blaming Mrs. O'Leary's cow rather than lax building codes for the Great Chicago Fire. The official explanation of the May 6 tumult, which saw the Dow plunge by nearly 1,000 points before largely recovering, does not hold up beyond a reasonable doubt.

In fact, the jargon-encrusted back pages of the report suggest that far greater damage was inflicted on investors that day by their brokerage firms. The brokers abandoned them to the wildfire. Call it progress: Never before have so many lambs been roasted so quickly...

HERE'S WHERE THE REGULATORS' story starts to fall apart. CME Group, owner of the exchange where the E-minis trade, said the sell order was consistent with market practices. Furthermore, only half the order had been entered as the market fell. And it had been broken up into small orders—nine out of every 100 coming into the market. In any event, this one trade couldn't have spooked investors because the market is anonymous. Traders didn't see a single, large seller. What they saw was continuous action.

The fact is, high-frequency traders and brokerage houses acting as market makers did more to drive down prices. They stopped buying and started selling.

The brokerage firms' behavior was particularly galling, though by no means illegal. They stopped automatic execution of customer orders, also known as internalization, which on most days accounts for nearly 100% of retail trades.

A brokerage firm will try to match one customer's order with that of another customer in-house. If the firm can't make the trade, it sends the order on to an executing broker. The big ones are Knight Capital, Citadel and UBS. The executing broker will generally take the opposite side of the customer order because retail customers tend to buy high and sell low, so it's easy to make money off them.

In the rare instances when an executing broker demurs, he sends the trade to a dark pool, usually one owned by his firm. (Dark pools are electronic-trading venues where institutional investors trade stocks away from the public stock exchanges.) If the dark pool can't execute the trade, it is sent to one of the stock exchanges. This largely automated process occurs in sub-seconds.

On May 6 when the market fell out of bed, the report says blandly, some of these players reduced executions of sell orders but continued to execute buy orders. In other words, they'd sell stock to a retail customer but wouldn't buy stock from a retail customer. They wanted to get rid of their own inventories, not accumulate more shares. So they sent the customer sell orders onto the swamped stock exchanges.

Here's one measure of the damage: Twenty thousand trades, totaling 5.5 million shares, were executed at a price 60% or more away from pre-Flash-Crash price levels, and thus later were deemed invalid. At least half those were retail orders. And, of course, that says nothing of the countless trades done at discounts of less than 60% but still large.

IT WAS A VICIOUS CYCLE. Retail stop-loss and market orders were converted to limit orders by internalizers prior to routing to the exchanges. A limit order requires the trade to be executed at a specific price, whereas a market order is the best price available. If the limit order wasn't filled because the stock's price had fallen, it was kicked back to the internalizer who, in turn, set a new, lower limit price and resubmitted it. Orders were kicked back multiple times because prices were collapsing so rapidly. They followed the prices down, "eventually reaching unrealistically low bids," as the report puts it.

Brokers meant well. Chris Nagy, managing director of order-routing strategy for TD Ameritrade, told me by e-mail, "In seeking best execution, a broker may use various methods to help ascertain a better price for the client. This tactic is somewhat common to protect unknowing investors from wild price swings, although May 6 was a whole different animal. It's important to note that when this type of strategy is used, it's generally sub-second, and many exchanges don't accept market orders."

SEC Chairman Mary Schapiro expected the report to boost investor confidence. Ba-a-a-h: The goat story is too hard to swallow.

Previous Day's Trading Room Results:
Trade Date: 10/11/10
E-Mini S&P Trades*
(before fees and commissions):

1) FT sell @ 10:22am at 1162.25 = -0.25 (1)

2) Algorithm positions (1)

3) “Reading the Tape” positions (1) combined Secret’s, Algo, & “Reading the Tape” total… +1.50

Sign up as an AvidTrader Member to receive "The Technician" Value Area's each day. The market then has an 80% chance of filling the Value Area. Many traders familiar with the Value Area and the techniques that go along with it use it to help them decide what trades to do each day. Join and see how this technique can help you trade more successfully!