Flash crash reforms may not be enough
19 October 2012 London
Image: Shutterstock
Researchers have claimed that US ‘flash crash’ reforms will fail to avert another market meltdown unless they are coupled with new liquidity-based circuit breakers.
Academics Dr Giovanni Cespa of Cass Business School, City University London, and Thierry Foucault of HEC School of Management in Paris, have studied the May 2010 crash, which saw stock prices plummet and recover within minutes.
In their new paper, ‘Illiquidity Contagion and Liquidity Crashes’, they argued that high frequency trading has made markets become progressively intertwined, meaning that liquidity suppliers in one asset class increasingly rely on the prices of other asset classes to set their quotes.
This creates a self-reinforcing positive relationship between price informativeness and liquidity that can result in different market ‘regimes’ displaying inverted correlations between the two, according to the authors of the paper.
“Markets can therefore hover in different liquidity states for the same set of underlying fundamentals. This means that for a given underlying, the market can be in a high or a low illiquidity regime,” said Cespa.
“In the former state [high illiquidity / low price informativeness], aggregate order realisations due to temporary price pressures trigger huge price adjustments. In the latter [high price informativeness / low illiquidity], the same realisations command milder price movements. Thus, the price impact of orders of a certain size is not univocally determined. Within this framework, a switch from the low to the high illiquidity equilibrium is what causes a flash crash.”
The limit-up / limit-down circuit breaker system coming into force in the US in 2013 may not prevent future flash crashes.
In the paper, authors argued that illiquidity-based circuit breakers should be introduced alongside price-based circuit breakers to stop trading when market-wide depth falls below a specified threshold.
“This liquidity evaporation may materialise in one market first, triggering a spiral that drags all assets into the illiquid regime,” said Cespa. “Price based circuit breakers do not necessarily offer a good protection against such illiquidity spirals because the latter may happen without trades and therefore without changes in prices.”
“[Illiquidity-based circuit breakers] could be an effective way to block an illiquidity spiral at its inception and thereby help traders to re-coordinate on a regime with higher liquidity.”
The paper also looks to explain the performance of exchange-traded funds (ETFs) during the flash crash. The authors found that exchanges successively broke all of the transactions that occurred with a price drop in excess of 60 percent during the crash. ETFs accounted for nearly 70 percent of the securities that were involved in those transactions.
Cespa explained: “The fact that the market can hover into two liquidity states has implications for the ability of cross-market arbitrageurs to provide liquidity.”
“Indeed, cross-market arbitrageurs diversify the risk associated with their position in one asset by taking an offsetting position in another (correlated) asset. As specialised dealers are the counterparties to this offsetting trade, liquidity provision by cross-market arbitrageurs rests on liquidity provision by specialised dealers.”
“As a liquidity crash is characterised by a drastic reduction in specialised dealers’ liquidity supply, it leads cross-market arbitrageurs to curtail their liquidity provision as well. This can explain the severe price disruption in the ETF’s market, as well as why the dispersion of price differentials between assets can increase considerably during a liquidity crash, leading to the perception of major price dislocations.”
Academics Dr Giovanni Cespa of Cass Business School, City University London, and Thierry Foucault of HEC School of Management in Paris, have studied the May 2010 crash, which saw stock prices plummet and recover within minutes.
In their new paper, ‘Illiquidity Contagion and Liquidity Crashes’, they argued that high frequency trading has made markets become progressively intertwined, meaning that liquidity suppliers in one asset class increasingly rely on the prices of other asset classes to set their quotes.
This creates a self-reinforcing positive relationship between price informativeness and liquidity that can result in different market ‘regimes’ displaying inverted correlations between the two, according to the authors of the paper.
“Markets can therefore hover in different liquidity states for the same set of underlying fundamentals. This means that for a given underlying, the market can be in a high or a low illiquidity regime,” said Cespa.
“In the former state [high illiquidity / low price informativeness], aggregate order realisations due to temporary price pressures trigger huge price adjustments. In the latter [high price informativeness / low illiquidity], the same realisations command milder price movements. Thus, the price impact of orders of a certain size is not univocally determined. Within this framework, a switch from the low to the high illiquidity equilibrium is what causes a flash crash.”
The limit-up / limit-down circuit breaker system coming into force in the US in 2013 may not prevent future flash crashes.
In the paper, authors argued that illiquidity-based circuit breakers should be introduced alongside price-based circuit breakers to stop trading when market-wide depth falls below a specified threshold.
“This liquidity evaporation may materialise in one market first, triggering a spiral that drags all assets into the illiquid regime,” said Cespa. “Price based circuit breakers do not necessarily offer a good protection against such illiquidity spirals because the latter may happen without trades and therefore without changes in prices.”
“[Illiquidity-based circuit breakers] could be an effective way to block an illiquidity spiral at its inception and thereby help traders to re-coordinate on a regime with higher liquidity.”
The paper also looks to explain the performance of exchange-traded funds (ETFs) during the flash crash. The authors found that exchanges successively broke all of the transactions that occurred with a price drop in excess of 60 percent during the crash. ETFs accounted for nearly 70 percent of the securities that were involved in those transactions.
Cespa explained: “The fact that the market can hover into two liquidity states has implications for the ability of cross-market arbitrageurs to provide liquidity.”
“Indeed, cross-market arbitrageurs diversify the risk associated with their position in one asset by taking an offsetting position in another (correlated) asset. As specialised dealers are the counterparties to this offsetting trade, liquidity provision by cross-market arbitrageurs rests on liquidity provision by specialised dealers.”
“As a liquidity crash is characterised by a drastic reduction in specialised dealers’ liquidity supply, it leads cross-market arbitrageurs to curtail their liquidity provision as well. This can explain the severe price disruption in the ETF’s market, as well as why the dispersion of price differentials between assets can increase considerably during a liquidity crash, leading to the perception of major price dislocations.”
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
