The network of interlinked securities trading exchanges in the US brought about by Regulation NMS (National Market System), passed in 2005 by the Securities and Exchange Commission, must be seen as a colossal design failure. This is the inescapable conclusion of Thursday’s mini-crash on Wall Street, when many shares saw their prices drop by 90%, even 99%, in less than a minute.
According to Ron Rowland, quoted on Seeking Alpha, 167 ETFs and ETNs in the US market (17% of all exchange-traded products traded that day) saw intraday price declines of more than 90%. As we covered on IndexUniverse.com on Friday, some funds were effectively wiped out completely: the Rydex S&P equal weight ETF (NYSEArca: RSP) went from US$39 to US$0.0001 in the space of 32 seconds. That’s a 99.9998% decline, by the way.
For the time being, NYSE and Nasdaq appear to have weathered the storm by cancelling all trades involving price moves of more than 60% from the consolidated last price at 14:40 EST on Thursday. The recovery in equity markets from Thursday’s intraday lows has probably also helped to keep investor confidence up.
If you were the bidder at US$0.0001 for RSP, your trade would therefore have been annulled under US exchanges’ rules, though your predicament probably wouldn’t elicit much sympathy.
More seriously, if you had owned RSP and had placed a stop loss order at, say, US$38, just below the market price of earlier that day, you might have found yourself selling at, for example, US$16, if that was the next market price at which a trade was transacted. Your stop-loss sale would have stood, since a US$16 trade wouldn’t have exceeded the 60% threshold: too bad for you, and a horrible reminder of the perils of stop-loss orders and “slippage” when markets “gap”. Your sense of injustice would then have increased when you saw RSP trading back up to US$41.25 at the close.
And, even if you were simply an owner of RSP in your pension plan, the afternoon’s events might well have been enough for you to decide that you’d had enough of equity investing. Who could blame you if you cashed in your ETF and reinvested the proceeds in gold coins, or cash under the bed?
When I look at the details of what happened on Thursday, I’m reminded above all of an interview with James Rickards, published in welling@weeden in March this year.
Rickards argues that it is a catastrophic mistake to increase the scale of complex systems without examining the possible implications for failure.
“When we integrate exchanges, erase borders, come up with global players and global positions and global strategies on a massive scale, we have increased the interconnectedness in a way that’s going to lead to greater and greater catastrophes, which is exactly what we’re seeing. People are very good at estimating the first order efficiencies, but they seem to be blind to the second order costs,” Rickards is quoted as saying. He produces several real-life examples, from circuit breakers in power grids to watertight bulkheads in ship design, to illustrate the importance of limiting scale in complex systems.
In other words, the much-vaunted benefits of increased speed and lower costs in US securities markets that have come about as the result of expanding the interconnectedness of trading systems may not have been worth it. First order benefits – cheaper, faster execution – may be of little value if the overall result has been to increase the likelihood of a blow-up in markets like we saw last Thursday. Put another way, does the promise of cheaper trading outweigh the risk to you, as an investor, of getting hit with the kind of stop-loss trade I mentioned above?
Worse, if the risk of such market failures has been permanently increased by changes to system design, the next collapse may be much more damaging, even impossible to recover from.
In Europe, largely due to accidents of history, markets, exchanges and settlement systems are still fragmented. Operating on a smaller scale doesn’t mean that trading problems can’t occur – several market makers told me that both Borsa Italiana and NYSE Euronext were struggling to manage order flows during yesterday’s sharp rally. But the fact that Europe’s trading systems are still largely independent of each other means that, first of all, problems in market A aren’t necessarily transmitted to market B and, second, market makers and traders probably still have somewhere else to turn if they need to lay off risks and the first market isn’t operating properly.
Maybe Europe’s much-criticised market fragmentation isn’t such a bad thing after all?