We need a multilateral consultation on how to avoid global deflation

Olivier Jeanne

03 March 2009



With the scope for monetary policy apparently exhausted – policy interest rates have been reduced to very low levels – policymakers seem to have pinned their hopes on fiscal stimulus. The only hope left for monetary policy is that we will avoid deflation. However, it is not clear how central banks would respond if deflationary pressures were to become stronger. The uncertainty resulting from central banks’ failure to clarify their anti-deflation policies is damaging confidence and – as I will argue below – carries risks for the international economic order.

To defuse those risks, I would propose the organisation of a round of "multilateral consultation", under the auspices of the IMF, on how to avoid deflation. The concept of multilateral consultation was tried in 2006-07, when China, the Eurozone, Japan, Saudi Arabia, and the US discussed the problems posed by global financial imbalances.1 My proposal is to have a second round of consultations focused on the risks of global deflation. This is a modest step, but it would be one in the right direction.

The eventual return to monetary policy

Global deflation, it is true, is still a risk, not a reality. However, it does not seem premature to start more open and frank discussions of what monetary policy can and cannot do to alleviate the global financial crisis. The fiscal and financial policies currently being implemented to deal with the crisis are likely to disappoint. Those policies do not address the main structural reason behind the slump in demand – a large debt overhang in the private sector inherited from the years of credit boom. A countercyclical fiscal stimulus may help smooth out of a cyclical downturn but – as logic, as well as the US Great Depression and the Japanese “lost decade”, suggests – piling up public debt on top of private debt does little to cure a structural debt overhang.

As for the policies currently being implemented to address problems with banks and credit markets, their shortcoming is that they are primarily focused on the supply side of the credit crisis – getting credit flowing again from the financial sector. This is certainly a very important part of the equation. However, it does very little about the demand side – the depressed level of solvable demand resulting from the debt overhang in the real sector. If one wants to avoid a protracted and painful period of deleveraging through high saving and low demand, other policy actions will be needed.

As the failure of fiscal and financial policies to deliver a vigorous recovery becomes apparent, attention will naturally turn to the possible roles of monetary policy.

Monetary policy in a credit crisis

Monetary policy matters in different ways. First, and obviously, monetary policy-makers should not let the economy become entrenched in a Fisherian debt-deflation spiral. What is not obvious, however, is the appropriate way of avoiding such a spiral. The Japanese experience of the 1990s does not yield clear-cut lessons. The theoretical literature on the Japanese liquidity trap suggests that the only way central banks can avoid deflation is by a credible commitment to generate some inflation. However, inflation expectations do not mechanically follow from quantitative easing or announcing an inflation target (as the US Fed did recently) (Jeanne and Svensson, 2007).2 More generally, the monetary orthodoxy that has prevailed in the last three decades, and the institutional reforms that it inspired, aims at preventing excessively high, not excessively low, inflation.

Another question is the appropriate objective for the inflation rate in a severe credit crunch. Assuming that central banks can avoid deflation, (i.e., if they can do 2% rather than -2%), then there is no reason that they should not be able to raise the inflation rate to 4 or 6%. Indeed, my understanding of “flexible inflation targeting” is that this would be the right thing to do in a credit crunch. Inflating away the excessive amounts of debt may not be the worst way of deleveraging, compared to the alternatives.3 If the inflation rate is equal to the target on average, then it stands to reason that it should be higher than the target in a credit crunch, when the marginal return on inflation in terms of output is higher. That is, an inflation-targeting central bank should try to converge toward the inflation target from above, not from below. This can be viewed as a way of implementing, through monetary policy, the kind of GDP contingencies in debt contracts that Robert Shiller has been advocating in his work on macro-insurance (see Jeanne 2008a and 2008b).

Obviously, this poses problems for the credibility of the monetary policy framework. Will expectations lose their nominal anchor and will long-term interest rates increase to levels that hurt the very borrowers that we want to help? This is certainly a risk, but overemphasising it would betray a lack of confidence in the very concept of inflation targeting, in which credibility is achieved by explaining to the public how the actions of the central bank are consistent with its objectives. An inflation-targeting central banker should be able to explain that in a credit crunch flexible inflation targeting means overshooting the target for a limited time – and that inflation will return to the target as the credit crunch is alleviated. This makes the pedagogy of monetary policy more challenging in a credit crunch than in normal times, but who has an easy job these days?

Be prepared: Start the debate now

I do not expect wide agreement on those views. But my point is that we need to start an open and rational debate on these issues with central banks. This debate should involve various constituencies and should not be resisted by central banks as an infringement on their independence. The best foundation for this independence, after all, is the central banks’ ability to explain their policies to the public. Furthermore, their independence might more seriously compromised, in the long run, if the debates – which will take place anyway – took the form of a sterile opposition between a conservative orthodoxy and a rising clamour of populist critiques. A round of multilateral consultations involving the Treasuries and central banks of significant countries could help a lot by mapping out the options that can be considered for monetary policy. Getting the main central banks to agree on a basic set of principles would reduce the fog of Knightian uncertainty that can only prolong the crisis.

There is another reason to make this debate multilateral and involve the IMF. The current lack of clarity on what monetary policy can and cannot do is dangerous for international economic cooperation. The future conflicts over monetary policy may arise not only between central banks and domestic constituencies, but also between countries. Down the road, that could lead to trade protectionism. For example, imagine the protectionist pressure in Europe if the US adopted a strategy of higher inflation, which would depreciate the dollar (even though the primary purpose of US monetary policy would be to alleviate the real debt burden of US households, not increase export competitiveness).

One may draw a parallel between the current situation and the monetary problems of the interwar period. The old view that the interwar monetary instability was due to beggar-thy-neighbour competitive devaluations was challenged by a new view that emphasises the deflationary effects of the Gold Standard system (Eichengreen 1992). According to this view, the depreciations reflected the fact that countries unshackled themselves from the unsustainable constraints of the Gold Standard, and rightly so. Trade protectionism resulted from the fact that there was no shared understanding of the underlying problem and of the required policy actions. In retrospect, there should have been a coordinated global monetary stimulus. Although the Bretton Woods system was built on different premises, the new view might be more relevant for our times.

Editors' Note: This column was initially posted with an erroneous IMF affiliation for the author. This was a typo. This column in no way represents the views of the IMF, its management or its board.


Eichengreen, Barry, 1992, Golden Fetters: The Gold Standard and the Great Depression, 1919-1939, Oxford University Press
Jeanne and Svensson, 2007 “Credible Commitment to Optimal Escape from a Liquidity Trap”, American Economic Review 97, 474-490
Jeanne, Olivier, 2008a “What Inflation Targeting Means in a Credit Crunch
Jeanne, Olivier, 2008b, “Inflation Targeting and Debt Crises in the Open Economy: A Note” forthcoming in Festschrift volume in the honour of Alexander Swoboda

1 In the words of the IMF, the multilateral consultation approach “allows a frank and focused discussion of the issues among the key players, while ensuring that–given the IMF’s universal membership–the rest of the international community is also involved.”
2 Lars Svensson and I showed that the commitment to inflation could be achieved by a small open economy through an exchange rate depreciation target., However, this mechanism does not work if the rest of the world also tries to inflate.
3 This of course assumes that debt is in domestic currency and of a reasonably long maturity, which is not the case in many emerging market countries.



Topics:  Global crisis

Tags:  IMF, global crisis debate, debt-deflation, global stimulus

Professor of Economics, Johns Hopkins University; visiting Senior Fellow, Peterson Institute for International Economics and CEPR Research Fellow