The Truth Behind Discount Brokerage

Since I read Flash Boys, those ads for discount brokerage now appear differently to me. I used to think that when, say, TD Ameritrade offers commissions of $9.99 a trade it’s because their platform is so efficient, so geared to enjoy economies of scale, that this low rate was sufficient to generate revenue in excess of its costs. But after reading Michael Lewis’s latest book it’s now dawned on me that an important element in the business model of these firms is to sell their order flow on High Frequency Trading (FHT) outfits or “dark pools”. The payment for order flow concept is a market-based recognition that many investors represent a reliable source of profit for the counterparties to their trades over and above the commissions they pay. So the $9.99 commission doesn’t represent the full return to (in this case) TD Ameritrade from their participation in your business.

It’s all naturally legal and therefore held to be beyond reproach. And perhaps the clients on such terms are naive for assuming that their visible cost of execution (i.e. the commission) was the only cost. But wouldn’t most people like to know if their orders were in effect the subject of a bidding war among the HFT crowd? Wouldn’t you think that the broker is getting you best execution not in the literal sense as defined in the regulations, but actually setting about to do that? If a discount broker can sell your orders on to a profit-seeking algorithm, they may still be providing you with a service but are not obviously working in your best interests.

It just seems as if there’s been a colossal error of judgment. The client might well be staggered to comprehend the economics of the online broker. The HFT apologists are no doubt equally shocked that anybody else is shocked. Hasn’t all this been disclosed? Well, technically I guess it has, but you can’t blame the average retail investor for wondering who they can trust. The brokerage model is full of the potential for principal-agent conflict. Dark pools and HFT algorithms are the latest manifestation. At a minimum, this is a PR disaster. The burden of proof is on those who equate volume with liquidity, who find nothing offensive in computer software being implemented to front-run orders. When it’s worth $300 million to build a perfectly straight fiber-optic line from Chicago to NY so as to transmit orders in a fraction of the time it takes us to blink, the casual observer may be forgiven for assessing that something is very wrong.