Tipping point for high frequency trading

Although a trading error may have been to blame for the domino effect that knocked the Dow Jones Industrial Average down by almost 1,000 points on Thursday, May 6 it highlighted the damage that high frequency trading can inflict in the blink of an eye. It also proved the fragility of the post-Reg NMS market framework, and proves the need for government mandated pre-trade risk management, market surveillance and monitoring.
'Greek Thursday' - as I have dubbed it - could be the tipping point for HFT. Regulators are poised to decide upon new controls for high speed markets, and Greek Thursday might just be the clarification they needed. Here are some of the lessons that could be learned from the experience.
1. Pre-trade risk management is a necessity. Fat fingered errors are absolutely avoidable. Using pre-trade risk management tools would prevent fat fingered errors and/or breaching trade limits. That trading firms do not use them is unbelievable.
2. Smart order routing can also be stupid order routing. Many algorithms are designed to 'find and nail' liquidity, no matter where it rests or what the price. You must monitor and manage your algorithms in real-time.
3. All exchanges and ECNs should have to take a break when markets are volatile.  If an exchange such as NYSE institutes a trading pause due to volatility, your order routing can go to a venue where liquidity is less than desirable, and prices are downright miserable.
4. Orders should be tagged. Whether an order comes through from an algorithm or a sales trader or broker, it should have an electronic tag so that when regulators/exchanges/ECNs see an error they know where it came from and can respond accordingly.
5. Real-time market monitoring is a necessity. Electronic trading means that crashes such as that on Greek Thursday can happen in an instant. If a trading anomaly is spotted in real-time, preventive measures could be taken.
Days like Greek Thursday could repel retail or institutional investment - the very money the industry has been trying to lure back into the stock markets.  When prices started to collapse on Thursday, it became all too clear that everyone was on the same side - bearish. Algorithmic players had their pants taken down and their positions exposed to the world yesterday. This hardly builds confidence.