The simple analytics of systemic liquidity risk regulation

Enrico Perotti, Javier Suarez, 16 March 2011



The recent crisis has provided a clear rationale for the regulation of refinancing risk – an aspect that was overlooked in the pre-Crisis agreement on international banking coordination known as “Basel II" (see Acharya and Merrouche 2009).

The liquidity proposals in the current agreement, Basel III, rely on quantity-based requirements to contain liquidity risk as a risk externality. They take the form of larger liquidity buffers and fixed limits to the size of unstable funding. But tight fixed ratios are seen by industry as very expensive, and they are likely to be significantly watered down. In the absence of other tools, the task of containing systemic liquidity risk would remain unfulfilled.

  • Short-term funding has advantages, as it can support rapid expansion of credit over GDP (for instance, because insured deposit supply expands slowly, or because short term lenders do not need to become informed about bank credit risk.

However, given its limited commitment, short-term funding creates refinancing risk.

Sudden withdrawals may:

  • Lead to disruptive liquidity runs (Diamond and Dybvig 1983), and
  • Cause fire sales which spread losses, or
  • Trigger counterparty risk externalities for exposed intermediaries (Brunnermeier 2009; Allen et al. 2010).

Each bank's funding decision has an impact on the vulnerability of asset prices to liquidity runs. So even if an individual bank takes into account its own exposure, it will not internalise its contribution to system-wide cost – a classic externality (Perotti and Suarez 2009). The wedge between the net private value and the social cost of short-term funding ensures that banks rely excessively on short-term funding.

A recent example is the massive build up in wholesale funding ahead of the crisis. Overnight (repo) secured credit feeding the final stage of the securitisation wave grew explosively during 2002-2007 to a volume over $10 trillion (Gorton 2009). Rapid withdrawals forced an unprecedented liquidity support by central banks, undermining their control over the money supply. The need to contain future accumulation of liquidity is thus a core challenge for macroprudential policy.

New insights on regulating systemic risk

In the tradition of externality regulation led by Weitzman (1974), in a new CEPR discussion paper we assess the performance of Pigouvian taxes (aimed at equating private and social liquidity costs) and quantity-based regulations in containing this systemic externality (Perotti and Suarez 2011). As in Weitzman, the optimal regulatory tool depends on the response elasticity of banks, recognising that the regulator is information-constrained in targeting individual bank characteristics.1

Our results show how the industry response to regulation depends on the composition of bank characteristics, as banks differ in their credit ability and their incentives to take risk. Banks earn decreasing returns to expand credit to their (monitored) borrowers, so better banks naturally wish to lend more. Shareholders of less capitalised banks gain from investing in poor gambles, since they retain the upside and shift downside risk to the public safety net (or alternatively because they are run by self-interested and overconfident managers, which view excessive risks as profitable). We find that depending on the dominant source of heterogeneity, the socially efficient solution may be attained with Pigouvian taxes, quantity regulations, or a combination of both.

Bank variation in credit quality

When banks differ only in capacity to lend profitably (reflecting credit assessment capability or access to credit opportunities), a simple flat-rate Pigouvian tax on short-term funding (possibly scaled up by the systemic importance of each bank, e.g. to capture contribution to counterparty risk) implements the efficient allocation. The intuition is that liquidity risk levies allow better banks to lend more, without requiring the regulators to be able to identify them. In this context, a quantity approach such as a net stable funding ratio or a liquidity coverage ratio (such as those proposed by the Basel Committee on Banking Supervision in December 2009) would be highly distortionary, reallocating credit away from banks with good lending opportunities.

Net stable funding ratios which impose an upper limit on short-term debt do reduce overall liquidity risk, but redistribute liquidity risk inefficiently across banks. Banks with better credit opportunities will be constrained, while the reduced systemic risk actually encourages banks with low credit ability (for whom the requirement is not binding) to expand.

Buffers are either ineffective or procyclical

Liquidity coverage ratios that require banks to hold fractional reserves of liquid assets against short-term funding work as a de facto tax, but with poor incentive effects. In boom times, the yield on liquid assets equals the cost of short-term liabilities, as it was the case prior to the crisis. In this stage, buffers impose no cost on carrying liquidity. So banks will simply increase their gross short-term funding, so that their net short-term funding are as high as in the unregulated equilibrium. The only effect is an artificial bank demand for liquid assets – traditionally kept in money market mutual funds rather than banks.

When the spread between liquid asset yields and bank borrowing costs is positive, a liquidity requirement operates as a tax on short-term funding, but the effective tax rate will be market determined – the tax rate will equal the product of the buffer requirement per unit of short-term funding times the interest spread. In the recent experience, interbank spreads over safe assets has been minimal just as aggregate liquidity risk was building up, while it escalated once the crisis started. Unless buffers are adjusted frequently, they would contribute to procyclical effects, relaxing the opportunity cost of unstable funding in good times and increasing it after a shock.

Bank variation in solvency incentives

Banks do not vary only on credit quality, but also in solvency incentives. These reflect capitalisation and charter value, or behavioural determinant of risk-taking, such as overconfidence). This case qualifies the results radically. Low charter value (or more risk-loving) banks have incentives to gamble to shift risk to deposit insurance and other forms of support (Keeley 1990).

We show that when bank decisions are overwhelmingly driven by gambling incentives, banks are not properly deterred by levies, while quantity constrains are more effective. Both short-term funding limits (e.g. a net stable funding ratio) and capital requirements can contain risk shifting by limiting the scale of lending. Levies will not be very effective because the most gambling-inclined banks will also be the most inclined to pay the tax and expand their risky lending. In this case, quantity instruments such as net funding or capital ratios are best to contain excess credit expansion.

Both quantity limits and liquidity charges

Our analysis on the relative merits of price versus quantity instruments suggests that combining them may be adequate for the simultaneous control of gambling incentives and systemic risk externalities. However, this presumes that the regulator controls can only act on regulatory tools related to liquidity risk. If strengthening capital requirements is an effective strategy for the control of gambling incentives (e.g. Hellmann et al. 2000), the case for regulating liquidity risk with levies on short-term funding is considerably reinforced.

The approach identifies recommendations for the use of ratio and charges. For instance, levies may be less costly to adjust than ratios. They might be easier to change, as in the likely case that regulatory ratios are embedded in an international agreement while levies are under control of a macroprudential authority. More importantly, they would impose lower adjustment costs than changing funding volumes on short notice.

Similarly, changes in levies are less likely to induce procyclicality than buffers, since the effect of a Pigouvian charge is more directly controlled by the regulator rather than being set by the interaction of a (controlled) quantitative requirement and the (freely fluctuating) market price of the rationed quantity.

For preventive purpose, controlling time-varying liquidity risk may be best achieved by a combination of some (variable) levies aimed to control the generation of risk externality by normal banks, while ratios may be aimed at containing the risk-appetite of gambling banks.


Acharya, V and Ouarda Merrouche (2009), “Precautionary Hoarding of Liquidity and Inter-Bank Markets: Evidence from the Sub-prime Crisis”, CEPR and NBER.

Allen, Franklin, Anna Babus, and Elena Carletti (2010), "Financial Connections and Systemic Risk", NBER Working Paper 16177, July.

Basel Committee on Banking Supervision (2009), International Framework for Liquidity Risk Measurement, Standards, and Monitoring, Consultative Document, December.

Brunnermeier, Markus (2009), "Deciphering the Liquidity and Credit Crunch 2007-08", Journal of Economic Perspectives, 23:77-100.

Diamond, Douglas, and Philip Dybvig (1983), "Bank Runs, Deposit Insurance, and Liquidity", Journal of Political Economy, 91:401-419.

Gorton, Gary (2009), "Slapped in the Face by the Invisible Hand: Banking and the Panic of 2007", paper prepared for the Federal Reserve Bank of Atlanta's 2009 Financial Markets Conference, May.

Hellmann, Thomas, Kevin Murdock, Joseph Stiglitz (2000), "Liberalisation, Moral Hazard in Banking, and Prudential Regulation: Are Capital Requirements Enough?", American Economic Review, 90:147-165.

Perotti, Enrico, and Javier Suarez (2009), "Liquidity Insurance for Systemic Crises", CEPR Policy Insight  31, February.

Perotti, Enrico, and Javier Suarez (2011), "A Pigovian Approach to Liquidity Regulation", CEPR Discussion Paper 8271, March.

Poole, William (1970), "Optimal Choice of Monetary Policy Instruments in a Simple Stochastic Macro Model", Quarterly Journal of Economics, 84:197-216.

Weitzman, Martin L. (1974), "Prices vs. Quantities", Review of Economic Studies, 41:477-491.


1 The approach is also related to the classical discussion by Poole (1970) on the optimality of price or quantity monetary policy instruments when the system to regulate is affected by several types of shocks.

Topics: International finance, Microeconomic regulation
Tags: financial regulation, Pigouvian tax, systemic risk

Professor of International Finance, University of Amsterdam and CEPR Research Fellow

Professor of Finance at CEMFI, Madrid and CEPR Research Fellow