Knight Capital Group had a dual life.
It touted itself as the largest independent wholesale market maker in the US, catering to large retail brokers such as Etrade, Fidelity, Scottrade, TD Ameritrade, and Vanguard.
These firms use Knight for executing their stock orders from individual investors, rather than sending those orders directly to a stock exchange, like the New York Stock Exchange. That’s because Knight will typically pay these brokers, lowering their costs compared to sending that same order to the NYSE. But until August 1, no one ever thought that a firm like Knight could lose $440m in 30 minutes.
Yet Knight was also a huge proponent of high-frequency trading. In fact Tom Joyce, its chief executive, testified to the US House of Representatives only in June about how there “has never been a better time to be an investor” and that “high speed computers, dark pools, etc are not the problem”.
Clearly, high-speed computers were exactly the problem as the algorithms moved so quickly that the brightest minds within Knight could not comprehend the gravity of the situation or react quickly enough to stop it.
Nor is this an isolated incident. Months earlier, the Nasdaq stock market debuted the highly anticipated Facebook initial public offering, which was one of the most highly publicised flotations in history and was expected to “bring investors back” to the stock market. That hype subsequently turned out to be one of the market’s biggest embarrassments as Nasdaq suffered technical glitches, sending a domino effect of losses through the industry and costing trading firms and investors more than $400m.
Add in the now-infamous “flash crash” of May 2010 which sent the Dow Jones Industrial Average plummeting 600 points in just 20 minutes, and the US markets have seen an ever increasing incidence of trading anomalies which are destroying investor confidence. Today, the individual investor in the US feels that the market is rigged as a giant casino, against which they have no chance.
Today’s stock exchanges boast execution speeds of less than 500 nanoseconds (one billionth of a second). By contrast the human mind by most estimates takes 300-700 milliseconds (one thousandth of a second) to make a simple decision. That means our exchanges operate more than a million times faster than the human mind.
Post “flash crash” there were a number of brilliantly designed proposals to help fix the problem so a repeat would not be possible. In all, the regulators implemented well over a dozen rules designed to slow things down and allow the human to react to the reality of the situation. But none of those fixes seem to be curing the next successive issue. Why?
One reason is the pace of regulation, which has increased fivefold over the past five years and has created successive layers of rules that have made the US markets ever more complicated to navigate. There used to be roughly five exchange filings to the Securities and Exchange website a month. Now there is usually no fewer than five per day. On Wednesday the exchanges introduced 16 rules.
Another is the pace at which automated trading algorithms have evolved in the markets. In less than a decade they have claimed so much of the equity trading landscape. Tabb Group estimates that on most days HFT can account for the majority of trading on US equity markets. Humans have been replaced by machines as they trade much more efficiently than a human ever could. Many HFT firms claim their strategies add robust liquidity to the markets and much more than a human. While in some cases it may be true, many times these machines are also feeding opportunistically off investors and siphon off precious returns from their portfolios.
Restoring investor confidence must be the number one goal of our regulators. In order to get our markets back on track we must take a more holistic approach to rule making by slowing down the pace of regulation. We must look at peeling back layer upon layer and build a new foundation. We should not seek to dismantle the machines, but we should create processes to control those machines so they cannot control us, such as certification and ample testing before deploying the machine and algorithm in the market.
We must also arm our regulators with the tools to see how these machines are interacting in the market for the benefit of investors, so that as the markets evolve the regulators are better equipped to enact policy in a proactive way, rather than the current reactive state we’ve been relegated to. In order to restore the crises of confidence in our markets, above all we must move with all due speed.
SOURCE:https://www.ft.com/cms/s/0/0b32795a-df9f-11e1-9bb7-00144feab49a.html#axzz25no51pum
GLTAL!