In search of a goldilocks exit strategy

Jean Pisani-Ferry calls for a credible exit strategy to help calm down bond markets and keep inflation expectations low. He believes that there is no contradiction between providing budgetary and monetary support in times of recession while committing to strict targets for the future- the two are in fact rather complementary.

When policymakers meet nowadays, at the G8 summit in L’Aquila or elsewhere, the discussions naturally revolve around exit strategies, in other words: when and how will economic policy return to normality?
It is only natural that policymakers wonder. Whether or not they are deemed appropriate and sufficient, actions taken by governments and central banks to support banks and stimulate the economy are genuinely extraordinary. Budget deficits in several countries have entered unknown territory (at least in peace time) and central bank balance sheets bear no resemblance to what they were two years ago. Impatience is thus perfectly understandable. But this does not make the task easy.
The first question to be asked is when retrenchment should start. The answer is neither too early, as a hasty withdrawal of support could precipitate a double-dip recession (as in Japan 1997), nor too late as debt build-up and the taking on by central banks of a de facto fiscal role should be
ended as soon as feasible without severe damages to growth. But the best analysis cannot guarantee a Goldilocks exit and against the background of a still very weak economy it is advisable to err on the side maintaining support for a little too long.
But an exit should be prepared. In fact the announcement of a credible exit strategy would help calm down bond markets and keep inflation expectations low. There is no contradiction between providing as much budgetary and monetary support as needed today and committing for tomorrow to strict targets, rather the two are complementary. The US arguably lacks medium-term commitments and this threatens the effectiveness of current actions. Europe’s policies are better anchored in the medium term in spite of the shortcomings of the Stability Pact. So a good approach for European policymakers at a time when the public wonders about the next step would be to preannounce their exit strategy, while making the start of actual normalisation contingent on a sustained improvement of economic conditions.
This however leaves two problems unresolved. The first is that as soon as conditions improve, most governments are likely to embark on consolidation simultaneously. This has often been done successfully in the aftermath of financial crises, but the difference here is that all countries are in recession. Against the background of wealth losses and unfinished deleveraging, private demand may be too weak to offset a major simultaneous cut to public demand. So there is a collective action problem that needs to be recognised and addressed. It will possibly be more difficult than the problem Europe faced at end-2008 when designing its (loosely coordinated) stimulus.
The second problem does not arise from coordination among countries but across policy instruments. Current exceptional measures are of three different types: support to the banking sector and individual banks through both government guarantees and central bank provision of liquidity against collateral (there are also public recapitalisations but the time horizon for state withdrawal may be longer here than on other fronts); overall monetary stimulus, involving in varying degrees non-standard easing techniques; and budgetary stimulus. Separations between the three categories are not watertight. Support to banks is in some respect hard to distinguish from overall monetary stimulus and central banks have embarked on quasi-fiscal initiatives. But it is conceptually useful to distinguish them. The question now is: who should normalise first? An obvious candidate for early exit is fiscal
policy since budgetary stimulus involves costs. So even though the lags involved in monetary policy are arguably longer, there are compelling reasons to remove budgetary support first. But the same applies to the withdrawal of bank support: it is only after banks have been adequately recapitalised and restructured that implicit subsidisation through the provision of cheap liquidity against collateral of uncertain quality can be ended. So the natural sequencing is to start with the cleaning up of the banking sector (keeping macro support broadly in place while this is being done), then to withdraw budgetary support, and to end with monetary policy normalisation.
This Goldilocks sequencing however raises difficulties. First, for the cleaning up of the banking sector to start, accurate information on the individual banks’ true health needs to be available, which in turns implies to stop agonising over stress tests. Second, the interrelated character of current policy actions implies that exit requires an unusual degree of coordination between governments and central banks. This only needs to be temporary until monetary and fiscal policy can again be separated, but in the meantime the situation requires sharing information, thoughts and plans – not an easy task within a single country and an even more daunting one within the euro area. Third, the very fact that monetary policy should normalise last creates a potential tension with other policy planks. Central banks are unlikely to accept their own normalisation being hostage to potential government procrastination.
These problems are far from insurmountable, but they need to be recognised, addressed and solved. This is why it is not too early to think about exit strategies, even though it is still too early to act.

Jean Pisani Ferry is Director of the Bruegel think tank in Brussels and Professor at the Université Paris-Dauphine.

Published in Börsen, El Pais, Financial Times Deutschland, La Repubblica and Neue Zuercher Zeitung.