The Bank of Canada has released a working paper by Koralai Kirabaeva titled Adverse Selection, Liquidity, and Market Breakdown:
This paper studies the interaction between adverse selection, liquidity risk and beliefs about systemic risk in determining market liquidity, asset prices and welfare. Even a small amount of adverse selection in the asset market can lead to fire-sale pricing and possibly to a market breakdown if it is accompanied by a flight-to-liquidity, a misassessment of systemic risk, or uncertainty about asset values. The ability to trade based on private information improves welfare if adverse selection does not lead to a market breakdown. Informed trading allows financial institutions to reduce idiosyncratic risks, but it exacerbates their exposure to systemic risk. Further, I show that in a market equilibrium, financial institutions overinvest into risky illiquid assets (relative to the constrained efficient allocation), which creates systemic externalities. Also, I explore possible policy responses and discuss their effectiveness.
He makes the point (tangentially) that the Efficient Market Hypothesis is dependent, in part, on an assumption of infinite liquidity:
Market liquidity is characterized by the cost (in terms of the foregone payo¤) of selling a long-term asset before its maturity.1 Two factors contribute to illiquidity in the market:a shortage of safe assets and adverse selection (characterized by the fraction of low quality assets in the market). On one hand, market liquidity depends on the amount of the safe asset held by investors that is available to buy risky assets from liquidity traders. Following the Allen and Gale ([9], [11]) “cash-in-the-market” framework, the market price is determined by the lesser of the following two amounts: expected payo¤ and the amount of the safe asset available from buyers per unit of assets sold. Therefore, this “cash-in-the-market” pricing may lead to market prices below fundamentals if there is not enough cash (safe assets) to absorb asset trades. On the other hand, market liquidity depends on the quality of assets traded in the market. In particular, adverse selection can cause market illiquidity if assets sold in the market are likely to be of low quality (as in Eisfeldt [25]).
He also explicitly considers liquidity risk as part of his model:
The long-term investment is risky not only because of its uncertain quality but also because of the cost associated with its premature liquidation or sale. Therefore, investors are exposed to the market liquidity risk through their holding of long-term assets. Holdings of the safe asset provide partial insurance against the possibility of a liquidity shock as well as against low asset quality realizations. In addition to the value as means of storage, the safe asset has value as means for reallocating risky assets from investors who have experienced a liquidity shock to those who have not. This is similar to the concept of liquidity value for ability to transfer resources in Kiyotaki and Moore [35].
In the course of determining the implications of his model the author examines the relative roles of private information and liquidity:
As a benchmark, I examine portfolio choice when investors have private information about their investment quality but the identity of investors hit by a liquidity shock is public information. Then I analyze the situation when the investors type (both liquidity needs and asset quality) is private information. In the latter case investors can take advantage of their private information by selling the low-payo¤ investments and keeping the high quality ones. This generates the lemons problem: buyers do not know whether an asset is sold because of its low quality or because the seller experienced a sudden need for liquidity.
His playing with the model leads to policy recommendations:
There are policy implications for government interventions during a crisis as well as for preemptive policy regulations. The e¤ectiveness of policy responses during crises depends on which ampli cation e¤ect contributes to a market breakdown. If it is due to an increase in liquidity preferences or to a small probability of the crisis then liquidity provision can restore the trading. However, if the no-trade outcome is caused by a large fraction of lemons or by the Knightian uncertainty about it, then it is more e¤ective to remove these low quality assets from the market. The preemptive policy response is an ex-ante requirement of larger liquidity holdings, which prevents market breakdowns during crises, especially if the economy is in the multiple equilibria range.
This last point is presumably part of the intellectual underpinnings of the Global Liquidity Standard in Basel III: A global regulatory framework for more resilient banks and banking systems. Other elements of this standard were discussed in the post Basel III.
He also provides intellectual underpinnings for a tax (deposit insurance premia?) on risky assets:
It should be noted that there is a moral hazard problem associated with government interventions during crises. If market participants anticipate government interventions then the optimal holdings of risky assets are larger. Therefore, a larger intervention is required. The moral hazard problem can be corrected if the liquidity provision at date t = 1 is nanced by a tax τ per unit of investment, which is imposed at date t = 0. The tax τx should be equal to the amount of liquidity λ that is required to restore market price to the level of p2,
[formula]
Imposing such tax increases liquidity holdings at t = 0 and prevents market breakdowns at t = 1, leading to a higher expected utility
I find it very disappointing that the author only examines a broad tax on risky holdings at time t=0. It would be more in line with the traditional role of a central bank to determine – given plausible assumptions – the required penalty rate for liquidity provision that would optimize welfare. Additionally, the welfare cost of a higher amount of liquid holdings is not addressed. And finally, investors – and the government – are assumed to know with perfect foresight which holdings are “risky” and which holdings are “liquid”. Holders of long-term Greek government bonds might be forgiven for questioning this assumption!
This last point is acknowledged by the author in his discussion of the Panic of 2007:
Financial institutions were exposed to systemic risk through securities holdings which had skewed payo¤s: they produced high returns in normal times but incurred substantial losses during the crisis. Before the crisis, many of these created securities were rated AAA, which implied a minimal risk of default. In particular, these assets were considered very liquid: if needed, these securities could be sold at a fair market price. During the crisis, the value of securities became more sensitive to private information.