Article Detail

Illix – A New Index for Quantifying Illiquidity

Journal 34: Cass-Capco Institute Paper Series on Risk

Tim Friederich, Carolin Kraus, Rudi Zagst

Illiquidity is a major issue in today’s risk management, yet there exists no straight-forward quantification of liquidity or illiquidity. We present eight possible measures of liquidity which are partially based on micro-structural market data and examine their evolution over time in the context of the development of financial markets. These eight measures are used for creating the new illiquidity index ILLIX. We outline the calculation of this index and show that it can describe well the liquidity situation in the North American market over the period between 1998 and 2009.

The illiquidity of financial markets has always been a widely acknowledged source of risk among practitioners, and its importance and the attention to it has increased during the recent financial crisis. Liquidity describes the “ease and speed with which a financial asset can be converted into cash or used to settle a liability,” as defined by the ECB. Being cognizant of liquidity has become an indispensable aspect of risk management. However, when we want to monitor liquidity, suddenly the question arises as to how this factor can be measured.

In contrast to stock prices and most financial data, liquidity itself is a hidden factor. Numerous papers in the literature have proposed using observable market factors as measures of liquidity or illiquidity. Chacko and Stafford (2004) define illiquidity as the spread between the fundamental value of a security and the price at which the security is actually traded. If this gap is large, illiquidity is high and vice versa. Besides this rather general definition, a variety of possibilities for measuring liquidity more directly have been discussed in the literature: Amihud and Mendel-son (1986) propose the bid-ask spread as the most intuitive measure of liquidity. Kyle (1985) shows that trading volume and market capitalization are positively related to the liquidity of the market. Amihud (2000) sug-gests the ILLIQ, describing the price response to the turnover, as a measure of illiquidity. The open interest also comprises liquidity information via examining the volume of derivatives traded [DraKoln (2008)]. Other measures are not defined for single securities but for entire markets or asset classes, such as the credit spread which quantifies the risk premium of corporate bonds [Huang and Huang (2003)]. The assumption here is that the higher the spread, the lower the liquidity. Adrian and Shin (2010) propose the LIBOROIS spread as an indicator of illiquidity observable through the interbank market. Finally, Whaley (2008) claims that the VIX, which describes the implied volatility on the U.S. stock market, is a good measure of how investors evaluate the liquidity of a market. All of these measures express liquidity from a different point of view.

Some studies have already examined the commonalities among several liquidity measures and concluded that the dependency between these measures is high, which means that they are driven by the same un-observable factors, i.e., factors which cannot easily be publicly quoted as a single measure and thus are not readily observable. Chordia et al. (2000) extensively analyze the cohesion of several measures of liquidity in regression analyses on single stocks and at an aggregated market level. They regress market- and industry-wide liquidity (as an average over a set of stocks) on the liquidity factors of single stocks and find high dependencies. Korajczyk and Sadka (2008) include price impact components in their analysis (together with spreads, turnover, and the Amihud measure) and find in a principal component analysis that the first three components can explain up to 55 percent of the variation of one liquidity factor across assets. However, their principal components measure variability of one liquidity factor over different stocks. They further analyze the canonical correlations of the different liquidity measures and find high correlation among them. Both of these studies, as well as Hasbrouck and Seppi (2001), focus on analyzing the price impact of liquidity, stating that systematic liquidity is a factor comprised in asset prices. The aforementioned studies concentrate on the ex-post analysis of the similarities and extensively scrutinize their impacts on, for example, price movements. However, they, as well as most other studies, focus on the commonalities of one liquidity measure among different stocks rather than on the commonalities between the different liquidity measures themselves.

In contrast to these outlined studies, in this paper we construct one single index of (il)liquidity which may serve as a trend indicator of the liquidity of a market. We propose an algorithmic methodology for a continuous calculation method (as opposed to ex-post) which takes into ac-count the findings of the previously mentioned research and identifies the common hidden factors. Among widely used measures based on asset price spreads and trading information we also include aggregated market information (in the form of credit spreads, the LIBOROIS spread, and the VIX). We use micro-structural market data of the 500 companies of the S&P 500 for our analysis. We combine the information comprised in these measures to achieve one comprehensive indicator, the illiquidity index ILLIX, which nicely explains the evolution of illiquidity in the North American market over the past 12 years. To the best of our knowledge, no research so far has included data gathered from the time of the financial crisis and beyond.


Leave a comment

Comments are moderated and will be posted if they are on-topic and not abusive. For more information, please see our Comments FAQ
This question is for testing whether you are a human visitor and to prevent automated spam submissions.