Introducing a new eReport: Reforming the international monetary system

Emmanuel Farhi, Pierre-Olivier Gourinchas, Hélène Rey, 19 September 2011



On the face of things, the ad hoc international monetary system seems to be working well enough. The US dollar dominates international financial transactions, and one asset class (US Treasuries) serves as a global reserve or safe asset. This system, however, has its weaknesses.

  • When the financial storm erupted in autumn 2008, investors worldwide steered their portfolios towards the safe havens – US dollars and US government debt.
  • This surge would have had severe adverse consequences for currency prices, interest rates, and asset markets worldwide, had it not been accompanied by an unprecedented level of cooperation and coordination among major central banks.

This collaboration allowed the large-scale provision of US dollar liquidity to the broader markets via various swap lines the Fed set up with selected central banks (Goldberg et al 2011). The situation was also improved by the broadening of collateral criteria, and a substantial increase in the resources at the disposal of the IMF. This temporary provision of liquidity in a time of severe financial stress restored confidence in markets and helped stop a potentially damaging deleveraging process.

Deep fault lines

This success, however, revealed deep fault lines in the system.

  • What drives the demand for safe assets in more normal times and in stress times?
  • Who supplies global reserve assets, and why?
  • Could the international monetary system also be, in some indirect way, partly responsible for the financial fragility of the world economy?

In a recent CEPR report (Farhi et al 2011), we argue that the optimal provision of global liquidity is a central feature of any well-functioning international monetary system.

Because few countries can offer truly safe assets, and the availability of liquidity in times of crisis is not guaranteed, we argue that there are severe distortions in the demand for safe assets even in normal times, distortions that can accentuate financial fragilities and exacerbate economic volatility.

In short, the main structuring idea that guides our report is that the world economy is characterised by a chronic and severe shortage of reserve assets –or, with some abuse of language, ’safe assets’.

The principal characteristics that determine the reserve potential of a financial asset are its safety and its liquidity; investors must be assured that the asset will not lose its value and that this value can be quickly realised. This is a rare characteristic, since the conditions that lead a country to liquidate part of its reserves are often associated with periods of economic stress and of low liquidity in world markets.

The unmistakable sign of this shortage of safe assets is a persistently low level of world real interest rates, and of the yields of the different classes of safe assets, be they sovereign debt of fiscally responsible countries, highly rated corporate bonds, or highly rated tranches of securitised products.

  • Our diagnosis is that the current functioning of the international monetary system is not only inefficient, but that it also has a number of perverse effects that undermine the stability of the world economy.
  • We propose a number of reforms. The common thread of these proposals is to increase the conditional supply of liquidity and reduce its unconditional demand.

Two key levers are 1) encouraging the transition to a more multi-polar international monetary system, and 2) improving the global management of liquidity.


The global shortage of safe assets arises from factors both on the demand side and on the supply side. A significant fraction of the global demand for reserve assets is due to precautionary strategies. This is a form of self-insurance against macroeconomic shocks through the accumulation of a buffer or risk-free assets.

However, this is an imperfect form of insurance which amounts to accumulating non-contingent assets. If financial markets were complete, a portfolio of financial assets would improve over self-insurance.

Of course, in practice, financial markets are not complete.1 Without insurance contracts or complete markets, self-insurance through precautionary savings is the only alternative.

This desire to self-insure makes perfect sense at the level of a country, but it can lead to excessive aggregate saving and excessively low interest rates.

This inefficiency is the result of a market imperfection – a pecuniary externality –and it can be the source of macroeconomic inefficiencies. Low interest rates encourage leverage, which often leads to fragility and instability, especially in the financial sector. Excessively low interest rates can spark off perverse risk-taking phenomena often referred to as the 'search for yield'. Environments characterised by low interest rates are also prone to speculative bubbles – for instance, in housing or commodity markets. Such bubbles are, by their very nature, fragile. Their emergence and disappearance create excess macroeconomic volatility, which reinforces precautionary saving behaviours and thus creates a vicious cycle.

There are also market imperfections associated with the supply of safe assets.

The supply of reserve assets

A strong demand for reserves generates an incentive to create safe assets. Thus, the scarcity of safe assets puts the financial sector under pressure. The recent expansion of the securitisation industry can be viewed as a collective attempt to create safe assets via the pooling and tranching of risk. Similarly, some governments responded to this pressure by relaxing fiscal discipline. This process leads to the creation of quasi-safe assets. Yet the sudden realisation that such assets are not actually safe induces violent market adjustments, which increase macroeconomic volatility.

The incentive to create safe assets affects more than the issuance and structuring of financial assets; it also has an impact on the issuance and structuring of the liabilities of financial institutions. Indeed, it enhances the attractiveness of short-term, risk-free debt because the demand for such a safe asset is strong. The problem with short-term debt is that it weakens balance sheets and increases the risk of financial distress and gives rise to fire-sale externalities.

Finally, the decline in real interest rates increases the probability of a liquidity trap, with depressive effects on the global economy.

To reduce – or eliminate – the shortage of reserve assets, it is possible to influence both the supply and the demand for these assets.

Solutions: Multipolar world

The dollar and US treasuries still play a central role in the international monetary system – they are the reserve assets par excellence. Our diagnosis of shortage of safe assets indicates the emergence of a modern version of the celebrated Triffin dilemma.2 There is a growing asymmetry today between the fiscal capacity of the US (the ’backing’ of US Treasury bills) and the stock of reserve assets held abroad – in other words, the US external debt.

Therefore, it can only be a matter of time before the world becomes multipolar. The emergence of this multipolar world is in itself a source of stabilisation for the world economy. By increasing the supply of reserve assets, a multipolar world naturally solves the Triffin dilemma.

It is also important to recognise that the development, liquidity, and openness of emerging countries’ financial markets will naturally lead to an increase in the global supply of reserve assets. This evolution towards a multipolar world is inevitable and must be encouraged. To this end, it is also crucial for the currencies of emerging countries to become freely convertible (eventually), so that assets denominated in these currencies may be considered truly liquid.

Improving the global management of liquidity

Self-insuring reserve accumulation is inefficient. It would be more efficient to establish a form of insurance contract between countries at the global level. Doing so would help alleviate the world’s chronic shortage of safe assets and preclude the associated negative consequences.

One may wonder why such insurance arrangements do not already exist. The explanation likely involves agency problems and the costs associated with market infrastructures. For this reason, an international agency such as the IMF would have a clear comparative advantage in the management of agency problems. It would also be able to catalyse the coordination required to create a large-scale insurance infrastructure, and would have the financial strength and credibility to perform the insurance functions, collect insurance premiums, and to discipline the resulting moral hazard.

There are several possible ways to achieve partial or full realisation of this objective.

  • Proposal 1. Systematise and sustain the provision of international liquidity in the form of swap agreements between central banks.

A possibility would be to ‘multilateralise’ swap lines by centralising the organisation of swap lines at the IMF. Doing so would replace a complex network of bilateral swaps with a star-shaped structure. The IMF would be at the centre of this system and would enter into swap agreements with the central banks of participating countries. The IMF could then redistribute the liquidity to countries in need during crises by simultaneously entering into a swap agreements with a liquidity-issuing country and with a country in need of liquidity.

  • Proposal 2. Strengthen and expand the facilities of the IMF – including the Flexible Credit Lines, Precautionary Credit Lines, and Global Stabilisation Mechanism – and extend IMF financing mechanisms, such as the New Arrangements to Borrow, so that the IMF can borrow directly from the markets.
  • Proposal 3. Develop deposit facilities in foreign exchange reserves with the IMF (reserve pooling arrangements) that will serve participating countries better than self-insurance.

Let us close by saying a few words on special drawing rights (SDRs).

The SDR debate

According to our analysis, the global provision of liquidity need not involve the issuance of SDRs; nor does it require ’anchoring’ the system through coordination of foreign exchange policies. Special drawing rights are complex and poorly adapted to the liquidity needs of the global economy. Their use – which can be justified under certain limited conditions – would not, in itself, cure the structural inefficiencies of the international monetary system. And a monetary anchor assumes that the priorities of monetary policy (economic and financial stability, including stable prices) can be changed in favour of external objectives. However, such an evolution is neither feasible nor clearly desirable. Nevertheless, SDRs could be an indirect way of moving certain currencies towards convertibility.

  • Proposal 4. Include the yuan in the SDR basket to facilitate emergence of a private market for SDRs, and allow the IMF to issue SDR-denominated debt.


Farhi, Emmanuel, Pierre-Olivier Gourinchas, and Hélène Rey (2011) Reforming the International Monetary System, Centre for Economic Policy Research, September.

Linda S. Goldberg, Craig Kennedy, and Jason Miu (2011), “Central Bank Dollar Swap Lines and Overseas Dollar Funding Costs”, FRBNY, Economic Policy Review, May.


1 There are multiple reasons. Some of them involve agency problems. Others arise from the costs of creating market infrastructures for financial instruments.

2 As Obstfeld (2011) puts it: “A Triffin dilemma arises any time increasing demand for a reserve asset strains the ability of the issuer to supply sufficient amounts while still credibly guaranteeing or stabilising the asset’s value in terms of an acceptable numeraire” (see “the International Monetary System : Living with Asymmetry”, mimeo, August).


Topics: Exchange rates, Global crisis, International finance, Monetary policy
Tags: exchange-rate policy, International Monetary System

Professor of Economics at Harvard University and CEPR Research Affiliate

Professor of Economics at the University of California, Berkeley and CEPR Research Fellow

Professor of Economics, London Business School and CEPR Research Fellow