On Tuesday, nearly two weeks after the unprecedented stock market sell-off and rebound, the US Commodity Futures Trading Commission (CFTC) and the Securities & Exchange Commission (SEC) published their Preliminary findings regarding the market events of May 6, 2010. At 151 pages you can see why it has taken so long; the SEC says it has “sourced and analyzed price, time, and volume data on over 19 billion shares executed on May 6, and quote data representing the best bid and best offer for over 7,800 securities, for each exchange, for each millisecond during the trading day.”
However, the preliminary findings are not very illuminating: the six working hypotheses and findings of the commissions could be summarised as the market hit a liquidity air-pocket and then stop-losses kicked in. But the market disruption “was exacerbated by disparate trading rules and conventions across the exchanges,” according to SEC chairman Mary Shapiro.
The hypothesis of the commissions – which seems more a statement of obvious fact and less of a hypothesis – is that the situation was worsened by the use of liquidity replenishment points (LRPs) by the NYSE (and only by the NYSE.) “When an LRP is triggered, trading on the NYSE will go slow and pause for a time to allow additional liquidity to enter the market,” says the report.
LRPs are designed to dampen price volatility. They might serve that purpose when the market resides on a single exchange, however, the US stock market has fragmented over the last five years – the combined market share of the NYSE and Nasdaq has declined from 75% to around 50%. Rather than contain the sell-off, LRPs probably accelerated it as orders intended for the NYSE were re-routed elsewhere. The proposed remedy is to co-ordinate the use of LRPs or other circuit breakers by all US stock venues when a stock or index falls more that 10% in a five-minute period.