Ladies and Gentlemen ,
I would like to thank the Peterson Institute for the invitation to speak in this panel.
In my remarks today, I wish to consider the current debate on competitive devaluations and the peril of a currency war.
Large policy spillovers in an increasingly integrated world economy naturally raise the possibility of inefficiencies. National policy-makers, who pursue a domestic objective and ignore the externalities they impose on other countries, may find themselves in equilibria that entail collective welfare losses. Policies that are efficient for the global economy are not necessarily optimal from the national vantage point, and there is a temptation to shift the burden and costs of adjustment abroad, and maximise the net gains accruing to national residents.
A classic example of the application of this framework to monetary policy is Ragnar Nurkse’s well-known analysis of the devaluations that took place in the interwar period. Since then, concerns about spirals of competitive devaluations have motivated the design of institutions and rules – from the Bretton Woods arrangements of 1944 to the Group of Twenty – to prevent countries from adopting “beggar thy neighbour” policies . The lessons from the crises of the interwar period continue to have an impact on policy discussions today, as the world economy is adjusting to the Great Recession.
Some central banks in major industrialised countries are currently pursuing, or have announced policy measures to stem entrenched deflationary pressures and bring down unemployment levels. Because policy interest rates are at or close to their lower bound in these economies, monetary easing has taken the form of non-standard measures, such as forward guidance concerning the path of future policy rates, and/or quantitative easing. These measures have had domestic effects (e.g. raising inflation expectations); but as it is a standard outcome of monetary policy easing, have also exerted some depreciation pressures on the currency of the respective country. The question I want to address is the following: should policy-makers, especially those in large countries whose currency is appreciating, be concerned by these developments? Do the latter pose a challenge to the G20 international cooperation policy agenda?
Taking mainly the perspective of a large currency area, I will first outline the conditions under which the answer to the above question is “No”. In the process, I will also argue that there is a misleading “Panglossian” view that is emerging, namely that competitive devaluations are an efficient way to coordinate national policies on the global task of taking the world economy out of a deflationary trap. I will then contend that competitive devaluations, if aggressively pursued, would be quite harmful, due to the constraints faced by policy-makers in most countries in today's environment.
Typically, sustained exchange rate movements affect policy objectives (e.g. inflation and output) and hence demand offsetting policy actions. Other things being equal, these policy actions, by resisting the exchange rate movement to a degree, will re-establish the original target outcomes over the relevant horizon. So, to the extent that such offsetting policy actions are unconstrained and also pose no trade-offs, and the exchange rate is not per se a target (namely under a floating exchange rate regime), there is little or no scope for effective currency manipulations.
This is by and large the environment in which the ECB operates. The exchange rate is not a policy target for the ECB; however, it matters for price stability and growth and, as such, it is part of the overall assessment of the appropriateness of the policy stance. If contrary to the objective of medium term price stability, it may thus trigger offsetting policy actions.
An important aspect of the consequences of an exchange rate movement resulting from a monetary policy shock abroad is that it brings about few trade-offs for domestic monetary policy-makers. Currency appreciation would tend to have effects of the same sign on inflation and growth. However, exchange rate movements reflecting other forces, particularly those resulting in severe misalignments, would not have such benign consequences. Currency swings disconnected from fundamentals, or brought about by misguided policies, have the potential to become disruptive.
The empirical evidence on the cross-border effects of standard monetary policy shocks in major economies suggests the following. A monetary easing, say in the US, while depreciating the dollar, is associated with a fall in net exports and a global output expansion; interest rates also decrease in other countries.  Therefore, if foreign policy measures, e.g. in the form of forward guidance, can be readily matched with domestic policy actions, we should not worry about the risk of a loss of competitiveness. In addition, a foreign policy expansion might not require a domestic response that fully neutralises the exchange rate movement if there are sources of overall positive spillovers, in particular through increased foreign demand. Objectives in terms of inflation and output gaps can then be achieved despite some currency appreciation.
The previous argument should not be confused with an emerging “Panglossian” perspective  which posits that competitive devaluations triggering an exchange rate “arms race” would be welcome in the current conjuncture. The argument put forward is that, contrary to a tariff war that leaves everybody worse off, such a race would generate a needed global, coordinated monetary expansion. This view rests on the assumption that monetary policy-makers should do more than they are currently doing. Chasing an exchange rate depreciation would make up for a perceived lack of action and deliver the appropriate symmetric policy response. However, there is no such lack of action in advanced economies, where monetary policy has been and is likely to remain very accommodative. Moreover, the “Panglossian” approach to coordination may create difficulties of its own, as I shall illustrate later on.
It would be simplistic to think that countries should always put up with any exchange rate movement generated by policy actions abroad, when the latter aim to address domestic problems and are not carried out through overt currency manipulations. Let me list here three reasons why there are limits to non-cooperative policy responses to exchange rate movements following foreign monetary policy shocks.
First, it is true that starting in the mid-1980s the economic literature by and large finds only modest gains from international policy cooperation.  This consensus, although not undisputed,  has been reinforced by the paradigm shift in international macroeconomic theory from the Mundell-Fleming-Dornbusch model to the New Open Economy Macroeconomics.  However, the workhorse open-economy model underlying those results describes a benign, frictionless environment, with only moderate macroeconomic interdependence. These theoretical economies are in an ideal state of “divine coincidence”: one policy instrument suffices to achieve all targets, quite independently of policy decisions abroad.  In the real world, a number of market frictions – in particular of a financial nature – have been identified as channels of large cross-country spillovers and, hence, as potential sources of non-trivial gains from cooperation – particularly concerning unconventional policies.  The ever stronger global integration of trade in goods and services and, in particular, in financial markets makes the issue of central bank cooperation ever more relevant today.
Second, a group of large, systemically relevant economies has not been affected symmetrically by the crisis. As noted by Barry Eichengreen , this is an important difference with the interwar period. Insofar as they have decoupled from advanced economies, the main policy concern in emerging market economies has been inflationary pressures, and in some cases, growth rates and asset prices that are too strong rather than too weak. Rather than resorting to tighter macroeconomic policies, these countries may instead be tempted to fight back against currency appreciation. To deal with the ensuing pro-cyclical bias, they might opt for second-best measures designed to limit capital inflows and/or moderate their domestic impact; the effectiveness of which remains unascertained. 
Likewise, small open economies which find it desirable to peg their exchange rates also have little room to react, unless the easing is undertaken in their anchor currency. Even then, their ability to deliver domestic monetary easing may be hampered by the lack of domestic assets available for quantitative easing.
Third, central banks in the advanced countries are close to reaching the limits of their conventional policies, or are already undertaking unconventional ones, such as quantitative easing. The operational difficulties that would surround additional non-standard measures and doubts over their traction on domestic aggregate demand suggest that there is at least an implicit cost in responding to further negative developments, as an exchange rate appreciation would be for many economies. This is particularly likely to be the case when the transmission of monetary policy is hampered by persistent structural rigidities, as it is the case in large parts of the euro area today. In situations in which it becomes difficult to generate monetary policy stimuli for domestic economies, the “coordinated” expansion unwittingly resulting from a currency war would not provide more room for manoeuvre for policy-makers. To the contrary, the “Panglossian” approach may only result in a global inflationary bias.
In the light of these constraints, it would be a matter of concern if countries were to directly pursue overt competitive devaluations, particularly by resorting to large purchases of foreign assets.  Moving to expenditure-switching policies would in effect create a zero sum game and expose global markets to serious risks of an escalation of trade and financial protectionism.
As much as exchange rate movements are the natural result of policies aimed at achieving sound and legitimate domestic targets, such as price stability, and as long as other central banks are not constrained in their ability to take offsetting actions, there is no reason for concern. Keeping one’s own house in order, however, is quite different from sticking one’s head in the sand. Especially in the current conjuncture, the room for policy actions might be reduced, or countervailing interventions might have become more costly. Then, large and sustained currency swings might hinder the achievement of domestic targets.
I am inclined to believe that global economic circumstances are currently far from the ideal case of “divine coincidence” where one policy instrument would suffice to achieve all targets. Although they are on a clear path to recovery, financial markets in advanced economies are still fragile and the real economy is at best giving timid signs of improvement. There are some concerns that non-standard policies would not gain enough traction on domestic aggregate demand because of deleveraging by domestic agents, including banks. This would leave the exchange rate as the residual macro-policy instrument, with forward guidance but especially quantitative easing mainly operating via external channels. Under these circumstances, the temptation to divert global demand and foreign capital towards the domestic economy at the expense of other ailing countries could be dangerously alluring. It is during times like these that the international community has to display an ability to reach mutually beneficial agreements, while recognising the legitimate ambition of each country to solve its domestic problems.
The economic literature highlights the dangers of individual countries pursuing “beggar they neighbour” or “burden-shifting” policies. I believe this message is well understood by central bankers . We strive to conduct policies in a cooperative spirit as much as possible. There have been several initiatives at global level to foster monetary cooperation. The commitment by G20 leaders to build a more stable and resilient international monetary system, as reaffirmed at the Cannes summit, is echoed for instance in the CGFS report on global liquidity , which develops a framework for the analysis of global liquidity cycles, as well as in the IMF’s work on the multilateral aspects of policies affecting global capital flows . The recent G7 statement on exchange rates also reflects this cooperative spirit, as well as the G20 communiqué subsequently adopted.
This kind of cooperation seems different from the way it is treated in theoretical models, where cooperation takes the form of policy-makers agreeing on a configuration of policies to maximise global welfare.  I do not think this difference matters. For me, the main value added coming from the economic literature consists in identifying and quantifying the implications for welfare of spillovers across economies and policies. Cooperation between the ECB and other major central banks is and will remain extensive and effective, building on a common ground of shared experience and mutual trust.
Thank you for your attention.
I wish to thank Luca Dedola and Giovanni Lombardo for their contributions to this speech and Georges Pineau for his remarks. I remain solely responsible for the opinions contained herein.
See B. Bernanke (2013), "Monetary Policy and the Global Economy", Remarks at the Department of Economics and Suntory and Toyota International Centres for Economics and Related Disciplines, Public Discussion in Association with the Bank of England, London School of Economics, London, 25 March; and B. Eichengreen (2013), "Currency wars or international policy coordination?", University of California, Berkeley.
See the key contributions by S. Kim and N. Roubini (2000) “Exchange rate anomalies in the industrial countries: A solution with a structural VAR approach”, Journal of Monetary Economics, Vol. 45 pp. 561-586; and Kim (2001) “International transmission of U.S. monetary policy shocks: Evidence from VAR's,” Journal of Monetary Economics, Vol. 48, pp. 339-372.
“In truth, the so-called “currency war” is not all bad: if everyone debases their currency, the world will get closer to the monetary stimulus necessary to outweigh excessive fiscal zeal”, from “G20 in search of recovery”, Financial Times 14 February 2013, http://www.ft.com/intl/cms/s/0/cec13222-76a2-11e2-8569-00144feabdc0.html#axzz2O5GVvOZi
See the classic contribution by G. Oudiz, and J. Sachs (1984), “Macroeconomic Policy Coordination among the Industrial Economies,” Brookings Paper on Economic Activity 15(1), pp. 1-76. For a survey of this literature, see the book by M. Canzoneri and D. Henderson (1991), “Monetary Policy in Interdependent Economies”, MIT Press, Cambridge MA.
W. McKibbin (1997). “Empirical evidence on international economic policy coordination”, in M. Fratianni, D. Salvatore and J. Von Hagen (eds.), Handbook of Comparative Economic Policies, Greenwood Press.
The seminal paper studying international monetary cooperation in this new paradigm is the now classic paper by M. Obstfeld and K. Rogoff (2002), “Global Implications of Self-Oriented National Monetary Rules”, Quarterly Journal of Economics, pp. 503-535. This paper found that the welfare gains from monetary cooperation are very small.
This term was used by Blanchard, O., & Gali, J. (2007) (“Real Wage Rigidities and the New Keynesian model”; Journal of Money, Credit and Banking, Vol. 39, pp. 35-65) to refer to closed economy models where all gaps can be easily closed by policy-makers. More generally, it has been used to describe all cases, also in open economy models, where trade-offs are absent or negligible.
See, for example, Dedola, L., P. Karadi & G. Lombardo (2013), “Global implications of self-oriented unconventional policies”, Journal of Monetary Economics, forthcoming.
See B. Eichengreen ( op. cit.).
It has been argued however that capital flows to emerging market economies have been driven as much by "pull" factors at home as by "push" factors originating abroad, see M. Fratzscher, M. Lo Duca and R. Straub (2012), “A Global Monetary Tsunami? On the Spillovers of US Quantitative Easing,” CEPR Discussion Paper, No. 9195. Also, as recalled by B. Bernanke (op. cit.), empirical simulations suggest that the drag on the competitiveness of emerging market economies and the output spillovers arising from accommodative monetary policies in the advanced economies are roughly offsetting.
See J. Gagnon (2013), “The Elephant Hiding in the Room: Currency Intervention and Trade Imbalances”. Peterson Institute Working Paper 13-2.
 IMF (2011): The Multilateral Aspects of Policies Affecting Capital Flows, ( http://www.imf.org/external/np/pp/eng/2011/101311.pdf))
International exchanges of information do not necessarily produce better outcomes, as shown e.g. by J. Frankel and K. Rockett (1988), “International macroeconomic policy coordination when policy-makers disagree on the model”, American Economic Review 78. For a survey of this literature, see also A. Ghosh and P. Masson (1994), “Economic cooperation in an uncertain world”, Blackwell, Oxford.
Reproduction is permitted provided that the source is acknowledged.