Illiquidity Contagion and Liquidity Crashes

Abstract

Liquidity providers in a security often use prices of other securities as a source of information to set their quotes. As a result, liquidity is higher when prices are more informative. In turn, prices are more informative when liquidity is higher. We show that this self-reinforcing relationship between price informativeness and liquidity is a source of contagion and fragility: a small drop in the liquidity of one security propagates to other securities and can, through a feedback loop, result in a large drop in market liquidity. Furthermore, this relationship also generates multiple equilibria characterized either by high illiquidity and low price informativess or low illiquidity and high price informativeness. A switch from the latter to the former type of equilibrium generates a liquidity crash. We use the model to interpret the Flash Crash of May 6, 2010.

Capital Commitment and Illiquidity in Corporate Bonds

“…The literature also demonstrates that liquidity is important because it affects the valuation of corporate bonds. For instance, Chen, Lesmond, and Wei (2007) and Lin, Wang, and Wu (2011) document that illiquidity affects bonds' yield spreads, while Bao and Pan (2013) show that illiquidity contributes to the observed volatility of bond prices and Cespa and Foucault (2014) show that a lack of liquidity can hinder efficient price discovery.…”

Modelling illiquidity spillovers with Hawkes processes: an application to the sovereign bond market

“…Illiquidity spillover typically occurs within a few seconds (much faster than the sampling interval of 5 min in Bongaerts et al (2015)) and is faster in bonds that are related by a similar maturity. It is notable that trades are very sparse on MTS and our results can be seen as directly due to quoting activity, supporting a channel of 'informativeness' as in Cespa and Foucault (2014). Therefore, the main contributions of this paper lie in (i) proposing an event detection method for limit order book data that does not use a discrete time grid and therefore especially suited for modelling with Hawkes processes, and (ii) documenting extreme illiquidity contagion, i.e.…”

“…Indeed market liquidity has become an important channel in the theoretical literature on financial contagion (see e.g. Kyle and Xiong 2001, Garleanu and Pedersen 2007, Brunnermeier and Pedersen 2009, Cespa and Foucault 2014and the discussion in Bongaerts et al (2015). They relate to a large literature on commonality in liquidity pioneered by Chordia et al (2000) and documenting the correlation between †I.e.…”

You Need Three Butterflies to Cause a Hurricane

“…Many papers analyse in depth the impact that volatility and liquidity have on sharp market movements and in some of them also the role of stop-loss orders is mentioned, although only on a qualitative basis. In particular, Cespa and Foucault (2014) present a theoretical model in which liquidity providers learn information about an asset from the price of another asset and therefore explain how shocks specific to liquidity supply in one asset class propagate to other asset classes. Then, signals returned from the latter influence the former, creating a feedback loop.…”

“…Thus, adding more informed traders may add more contrasting signals, leading to learning complementarities, reducing price informativeness and eventually generating price jumps. The model proposed below makes use of, and links to, the models described in Cespa and Foucault (2014) and Goldstein et al (2013), and also includes stop-loss orders.…”