An interesting paper here by Jean Tirole at the Toulouse School of Economics (HT: Nicholas Gruen). It seems to echo the "liquidity as a public good" concept that Joshua Gans and I independently developed in our work on the Australian securitisation market. We argued that it was this idea that provided the rationale for government intervention in markets that had collapsed due to an external shock. In early 2008 we commented:
"The central tenet of this proposal is that a basic level of liquidity in key economic markets is a ‘public good’. The policy imperative here is reinforced by the fact that severe market dislocations, such as the credit crunch that we are presently observing, are becoming increasingly common and more quickly transmitted in today’s highly networked world. The presence and apparent regularity of these extreme events is consistent with recent academic innovations in the so-called ‘behavioural finance’ and ‘extreme value theory’ literatures.
In standard finance theory, academics, and the commercial practitioners that follow their prescriptions, have all too often made the erroneous assumption (for analytical purposes) that asset returns are ‘normally distributed’ (ie, virtually never subject to events like the 1987 stock market crash or the 2001 tech wreck) and that financial markets are ‘frictionless’—ie, investors always benefit from perfect liquidity and price-discovery. These are, by way of example, some of the essential assumptions underpinning the ‘Capital Asset Pricing Model’ (CAPM), which is widely used around the world by investors and their advisors. Up until recently, the assumption of perfect liquidity and return normality were condition precedents in almost all financial models used by financial market participants.
In the real world, however, investors are finding that they are increasingly faced with periods of profound illiquidity, extremely poor price discovery, and, in certain cases, complete market failure. In the financial market history of the last two decades, there are numerous examples of this illiquidity problem and governments acting to remedy it. In 1998 the massive hedge fund LTCM confronted severe illiquidity when the Russian government defaulted on its debt obligations, losing some US$4.6 billion in less than four months (LTCM was also hit by a sudden convergence in the ‘correlations’ of all of the assets it held, which it had previously assumed to be uncorrelated and hence well-diversified). Of course, at that time the US Fed acted to facilitate a bail-out of LTCM by a consortium of investment banks.
In the past eight months, major institutions around the world have been subject to the specter of extreme illiquidity in the market for many of their debt securities, which has in turn made price discovery near impossible (ie, how do you value assets for which there are virtually no prices, and when prices do exist almost all participants—including the regulators and government—agree that they represent dramatic deviations from any understanding of fair market value).
One of the primary problems here is that academics, practitioners, and regulators are discovering that financial markets are not always ‘efficient’ in the sense that was popularized by University of Chicago financial economists such as EugeneFama33 (1965). This assumption of market efficiency has dramatically changed the financial market landscape and led, for instance, to the prolific use of ‘index’ funds provided by State Street, Vanguard and others. The market efficiency paradigm in turn hinged on the belief that investors are in aggregate highly rational ‘agents’ that are not subject to systematic behavioural biases. This assumption can in turn be traced back to the work of the US economist John Muth who developed the so-called ‘rational expectations’ theory under which individuals and institutions make forecasts about the future without any fundamental error or bias. That is, investors’ expectations are, on average, accurate. This rational expectations hypothesis has underpinned much macroeconomic analysis of the last half century.
More recently, though, pioneering academics such as Kahneman and Tversky34 —the former of whom received the Nobel Prize in 2002— and Richard Thaler have applied principles from psychology, sociology and anthropology to document that in practice people behave in a manner that can deviate strikingly from the equilibrium predictions of the efficient markets hypothesis (and rational expectations in particular).
This makes intuitive sense if we cast our minds back through history and consider the speculative fads and crashes of the Dutch tulip mania, the emergence of junk bonds in the early 1980s, the related 1987 stock market crash, the late 1990s tech craze and the inexorable tech wreck of 2001. Over the last 20 years a large body of evidence has built up illustrating that humans are fallible and subject to a wide range of biases, including irrational loss-aversion, framing, use of heuristic rules of thumb, hindsight biases, and cognitive dissonance (ie, avoiding information that conflicts with our assumptions).
Many authors, such as Barberis, Shleifer, and Vishny35 and Daniel, Hirshleifer, and Subrahmanyam36 have demonstrated that there can be major mispricings, non-rational decision making, and return anomalies in financial markets due to these behavioural biases. In particular, the tendency of humans to identify fictitious ‘patterns’ in otherwise random return sequences, and for us to be consistently ‘over-confident’ in our assessment of our own forecasting abilities, can result in significant market over- and under-reactions in asset price returns (eg, consider the tech boom and subsequent crash). Behavioural economists have also found evidence of the anecdotally well-known market phenomenon of ‘herding’ and ‘groupthink’ whereby strongly anomalous market-wide effects can materialise where there is collective fear and greed (again consider the wild and seemingly irrational—at least judging by the actions of central banks—swings in the risk appetites of global debt investors before and after the US sub-prime crisis).
It is now accepted by many economists that these behavioural biases that plague human decision-making under uncertainty can cause extreme asset price bubbles and subsequent crashes. In parallel with these innovations in the field of behavioural finance, academics have also started to accept that capital market returns are not ‘normally distributed’, but rather characterized by ‘fat-tails.’37 The presence of these fait-tails or so-called ‘black swans’ in asset returns, which suggests that extreme events (such as the 1987 crash or the current credit crunch) can occur with far greater regularity than the predictions of a ‘normal’ distribution, is also consistent with the tendency of investors to irrationally herd in one positive or negative direction, which can perpetuate clusterings of extremely positive or negative outcomes, such as that which we are observing today…
In the presence of highly uncertain prices, institutions are reluctant to lend to one another as they do not have sufficient visibility on the value of the collateral that they will use as security. This propagates potentially enormous problems for the financial system at large as transactions that were previously considered to be nearly risk-free are subject to perceptions of ‘counterparty risk’. This is precisely what happened with Bear Stearns, which on 10 March 2008 reportedly still had US$17 billion in cash. A few days later, the leading US investment bank Goldman Sachs announced to the world that it would no longer serve as a counterparty in Bear Stearns’ transactions. Goldman’s actions shattered confidence in Bear Stearns’ ability to service its obligations and meant that it could no longer raise any short-term debt funding to underwrite its working capital requirements. Once again, the Fed was forced to step in and inject liquidity into a market that had failed: in particular, the Fed took Bear Stearns’ otherwise illiquid and unpriceable assets as security and lent JP Morgan the US$30 billion that it needed to buy Bear Stearns.
It should be clear that market failures and the absence of price discovery suggest that the provision of a minimum level of liquidity can be construed as a ‘public good’. While in practice it is hard for any good to unconditionally satisfy the two key conditions of a public good—namely ‘non-rivalness’ and ‘non-excludability’—many come close to approximating them (eg, the light from a lighthouse, clean air, and market infrastructures). It is well known that markets can fail to produce sufficient quantities of such goods, which is referred to as the ‘public good problem’. As a technical aside, there may be an argument that market liquidity is ‘rival’ but ‘non-excludable’, in which case it may be more appropriately classified as a ‘common pool resource’. In any event, you have similar problems to those found with public goods, albeit that in this case they are known as the ‘tragedy of the commons’."
In his abstract, Tirole similarly observes:
"The recent crisis was characterized by massive illiquidity. This paper reviews what we know and don't know about illiquidity and all its friends: market freezes, fire sales, contagion, and ultimately insolvencies and bailouts. It first explains why liquidity cannot easily be apprehended through a single statistics, and asks whether liquidity should be regulated given that a capital adequacy requirement is already in place. The paper then analyzes market breakdowns due to either adverse selection or shortages of financial muscle, and explains why such breakdowns are endogenous to balance sheet choices and to information acquisition. It then looks at what economics can contribute to the debate on systemic risk and its containment. Finally, the paper takes a macroeconomic perspective, discusses shortages of aggregate liquidity and analyses how market value accounting and capital adequacy should react to asset prices. It concludes with a topical form of liquidity provision, monetary bailouts and recapitalizations, and analyses optimal combinations thereof; it stresses the need for macro-prudential policies."
