A tale of three trilemmas, by Dominique Strauss-Kahn (Cambridge Union Society (UK) – March 9, 2012)
Policy trilemmas have always been a potent way of describing economic policy trade offs and some of these trilemmas are especially powerful to understand our current predicament and future challenges. I would like to present three of them, which I regard as particularly useful to analyse the world economy and to provide hints for solutions. They are in principle completely independent from each other but I see them as deeply inter-connected so it is interesting to think of them concurrently as a system.
1.There is the canonical one known since the beginning of the 1960s as the Mundell-Flemming one.
It has always been an important reference for all international economists. It is the so-called impossible trinity that Mundell and Flemming in the early 1960s derived from the classic IS-LM model in a small open economy.
1.1. It posits that one such economy cannot simultaneously seek to have an independent monetary policy, an open capital account and a fixed exchange rate regime. This idea has been a key justification for the liberalization of capital movements and has also led a number of countries to accept the flexibility of their exchange rate in order to regain sovereignty over their monetary policy when the gold standard ended. It also had important consequences for the international monetary system as a whole which evolved gradually after the demise of Bretton Woods into a “non-system” loosely governed by an implicit anchor (the dollar) and the supposed disciplining and stabilizing effects of flexible exchanges rates and free movement of capital.
1.2. We found out subsequently (and painfully from the end of the 1990s until now) that the stabilizing features of this non-system were much weaker than initially thought. There are many reasons for that.
1.2.1. At the micro level, there was a lack of understanding of the imperfect nature of financial markets, our underestimation of the consequences of financial frictions, our deliberate dismissal of the impact of asset bubbles and in one barbaric word, our disregard for behavioural macroeconomics.
1.2.2. The other reason, at the macro level, is simpler. On paper, this monetary arrangement was meant to have stabilizing features in a general equilibrium. That is, if applied internationally and systematically, inflation targeting, free capital movement and flexible exchange rates applied globally ought to produce both a stable and efficient international monetary system. This was one of the reasons why the IMF and the US Treasury (to name a few) embarked in a far-reaching promotion of capital account liberalization and flexible exchange rate regimes while Central Bankers slowly adopted inflation targeting as a framework that was meant to allow delivering both internal and external stability.
Yet the reality, which we have failed to acknowledge, was that a large part of the world — for good or bad reasons— refused to adopt this model. Half of the world, and what turned out to be the most populous and dynamic economies in the world, rejected vigorously this duet on which the international monetary system came to rest. This contradiction between the theory of the international monetary system and its actual reality became an important source of instability and today justifies its profound reform.
1.2.3. This has been the subject of a large amount of research over the years but for the most part this effort has remained largely normative and has failed to take properly into account national preferences and that of emerging economies in particular. The underlying assumption seemed to be that the world would eventually end up converging towards an accepted intellectual consensus about international macroeconomics —the one proposed by advanced economies. This consensus never happened. But surprisingly enough we behaved as if there was a consensus. For decades, China and India (but not only them) have been keeping a fixed exchange rate and everybody acted as if the reality was not different from the model. Even after the Asian crisis when it became obvious that most of the Asian countries were not following this track, the mantra remained simplistic: globalization will go on and in a decade or so the renminbi will be a flexible currency and the Chinese capital account will be open. The debate about global imbalances would probably not have even started if the system had worked as planned. It never did but we wanted to look at the world with our own glasses.
1.3. Today, this consensus is still lacking and it seems unlikely that the crisis will get us any closer to it. There however seems to be a growing consensus on at least two points that need to be addressed:
i. World reliance on any one currency and on a currency whose underlying economy is likely to become relatively less prominent is not an optimal solution. This calls for either:
- a new hegemonic currency as Kindleberger would have probably thought (the renminbi?),
- a new truly multilateral currency system which involves a strong euro area and a new Asian currency of gravity,
- or finally a real global currency (something akin to the bancor of Keynes) probably based on a basket of existing ones which would allow to internalize the Triffin dilemma but would at the same time raise a sensitive question over its issuance —a topic central bankers usually get very ticklish about. I sometime read that the SDRs issued by the IMF could play this role. There is certainly a bigger role to play for the SDRs but as it stands, the SDR is not a real deliverable currency; hence it cannot be the answer.
ii. The crisis has also showed the inadequacy of our international financial safety nets.
These have evolved considerably through the crisis. The IMF has expanded its toolbox in a major way but a salient feature of today’s global safety nets architecture is that they also rest on stronger regional pillars, something the European crisis has recently confirmed. The cooperation between regional and international safety nets is still an experiment. All the system would be at risk if the IMF worked only as a backstop for regional structures because it would probably act too late and in a way that weakens conditionality considerably. What we need is a real partnership between the IMF and regional mechanisms including the European Stability Mechanism or the Chiang Mai Initiative in Asia for instance. But there is no clear framework for cooperation at this stage. And, even worse, the new European safety net has been designed without a clearly articulated coordination process with the IMF, as if the two other members of the troika would always agree with the Fund.
The recent past has shown that it is not the case –far from it. From this point of view, a post-mortem on the Latvian program (where the IMF and the European Commission differed on the peg) will be very helpful not to talk about Greece (where the IMF and the Europeans differed on the need and size of debt restructuring).
iii. In addition, FX swap lines between central banks, which have proven to be an important liquidity backstop, remain largely governed by central banks’ discretion and bilateral negotiations. This is potentially dangerous as it reduces predictability and effectiveness and increases the number of potential cracks in the international financial safety nets.
But this debate about the reform of the IMS (that I have addressed in a recent lecture in Beijing) raises an important global governance question. Reserve currency issuers enjoy global benefits, but also have global responsibilities, and there are no structures to make them accountable for it. Managing those two global public goods that are financial stability and global money requires a governance framework that is still deficient. This is something that the second trilemma studies in more details.
2. The second trilemma I would like to present is that of Dani Rodrik (2011)
It departs from pure economic theory and connects economics to politics. It argues that one cannot simultaneously have national self-determination (i.e. sovereign Nation States), democratic governments and deep global economic integration. Rodrik has always been more doubtful about the positive forces of globalisation than most of his peers but his arguments have always been more subtle and powerful than hard globalisation sceptics. His arguments lead to important questions about the degree of globalisation we should seek and the shape and form of international governance we should pursue.
2.1. According to Rodrik, the choices are the following.
(i) We can choose to give way to hyperglobalisation, restrict democracy in an attempt to reduce transaction costs and improve global economic efficiency.
This would bring the world as a whole closer to its economic production frontier but probably at the cost of equality, social cohesion, human health and the environment. This is the choice that some economies have made already.
(ii) We can decide to limit globalisation to sustain a degree of democratic governance. This involves such things as putting sand the wheels of hyperglobalisation and there is indeed a growing literature that revisits the economic benefits of financial globalisation with a much more sceptic outlook. Even the IMF started thinking in some cases of ways and means to curb undesired and unnecessary international financial flows under my directorship. Many have been surprised to see the IMF arguing that in some cases and under some conditions capital controls could be an effective –even if temporary– way to deal with large amount of cross-border capital flows that may be very volatile and create asset bubbles. But it is clear to me that the healthy functioning of the IMS depends crucially on orderly cross-border capital flows and hence “rules of the road” are necessary.
(iii) Or finally we can choose to globalise democracy and therefore limit the importance of Nation States in favour of a globalised governance mechanisms or institutions.
2.2. This last idea sounds utopian for sure but there are therefore somewhat less ambitious versions of it: international regulatory bodies fall under that category, multilateral surveillance as conducted by the IMF is an another example and international economic coordination is a slightly more ambitious one.
At this point I would like to say that when I reflect on the crisis, I consider the unprecedented international policy cooperation its most important legacy. Facing a common threat of impending economic disaster, countries came together to do what was needed.
Today countries face different challenges and so the commitment to multilateralism is on the wane. But even though we know that there can be no domestic solutions to global economic problems, momentum for international policy coordination is quickly fading.
For all their merits, all of these supranational policy action devices in fact lack democratic legitimacy and accountability and as a result, their scope for action and enforcement ought to be limited. The only such ambitious experiment of extension of democracy over and above national frontiers is the European Union, which is effectively an attempt to create a space for democracy and self-determination in a grouping that can actually face and shape globalisation in a way that no Nation-State alone in a globalised economy can. Yet, the management of the current crisis in the Eurozone does not portray the system as very effective, to say the least.
3. This leads me to the third trilemma, which regards Europe and its governance structure.
In a recent paper, Jean Pisani-Ferry (2011) from Bruegel argued that there is a fundamental incompatibility between the monetary union, national banking systems and the absence of joint responsibility over national public debt.
3.1. This is indeed at the core of the current governance crisis in Europe and relates to the previous trilemma in the sense that Europe is yet an unresolved compromise between national and supranational governance. This is particularly visible both in the areas of financial and budgetary policy.
3.1.1. Europeans have made the choice of a monetary union, which its founding fathers saw as a way to precipitate further political integration. European citizens were initially sceptical but voted in favour of it by a small margin in 1992 by ratifying the Maastricht Treaty. They understood then that although this would probably weaken their national protection inside Europe, it would strengthen Europe’s ability to resist the levelling forces of hyperglobalisation. In Rodrik’s trilemma, they accepted to partially depart from national citizenry and pooled their democratic prerogatives above the Nation State. Yet this choice was a compromise that required a number of accommodations and as a result, Europe’s current governance structure is still an unstable equilibrium. In particular, Europeans accepted to create a Community that would have substantial power over Nation States but not all powers.
3.1.2. Fiscal policy remained a national prerogative with, on one hand, the now infamous no-coresponsibility clause (article 125 of the Lisbon Treaty), and on the other, the refusal to grant the European Union the autonomous right to raise taxes, spend and borrow. Together, this essentially barred the road towards closer fiscal integration. Budgetary matters remained an inherently national prerogative loosely governed by rules and possible sanctions (the Growth an Stability Pact)
3.1.3. Finally, as national banking systems expanded across the continent rapidly with the removal of restrictions to capital movements and the advent of the single currency, national banks maintained a large degree of home bias in their holdings of sovereign debts and they continued to be perceived and used as national champions or captive arms of their governments that could be swayed into the indirect conduct of national industrial policies.
More importantly, the financial supervision, regulation and resolution frameworks remained essentially national. Banks could have operations and take risks across the continent but their supervision and resolution in the case of failure remained both a national prerogative and a national risk to the sovereign balance sheet. During the crisis, the lack of cross-border resolution scheme has been one of the biggest problem we had too face. Even if this crisis has brought some modest progress towards supranational regulation and supervision (the De Larosiere report and the creation of the European Systemic Risk Board for instance) the question of home versus host country supervision, regulation and resolution is still largely unsolved. The reality is that banks, in life and in death, are still governed by national authorities marred with principal-agents conflicts. This has far reaching consequences for fiscal policy but also for monetary policy as it leads the European Central Bank to behave accordingly by maintaining a degree of national responsibility over a number of risks (Emergency Liquidity Assistance, Expansion of the collateral pool in the context of the LTRO to name a few).
3.2. All in all, this trilemma can be resolved in three ways.
i. First, a partial or a complete disbanding of the monetary union. I do not believe that this is going to happen. First for political reasons, it will be very costly for any Head of State or Government (especially in Germany and France) to be seen as the liquidator of more than half a century of European dream. Second, because there is no process in the treaty to do that. Of course, countries are sovereign and they may unilaterally decide to leave. But the simple idea that some country (say Greece) could be easily kicked out if it doesn’t want to seems very unlikely to me. As a result, it is probably more desirable to keep the monetary union intact and choose to either move towards a form of banking federalism or fiscal federalism.
ii. The former would essentially centralize both the supervision and the contingent liability embedded in national banking systems through a pan-European resolution framework guaranteed jointly and severally by member states and a supranational guarantee of deposits. With this mechanism in place, fiscal transfers would occur indirectly through sovereign defaults and bank failures that would be resolved by this authority. This system may work, but it will be very disruptive and have consequences for financial stability and Europe’s potential output.
iii. Alternatively (or simultaneously), one could choose fiscal federalism and end the work that was started with the creation of the monetary union by recognizing that there hasn’t been any conclusive experience of a monetary union outside of a fiscal union. But fiscal federalism is a vague notion and it could take several forms. One of these options includes a relatively modest degree of centralisation and would probably appear more politically appealing. Others require greater coordination, deeper mutual surveillance and stronger controls. But all of these options limit transnational transfers to a small and possibly insufficient level. They are also institutionally and democratically rather complex to set up, and create an important risk of technocratic capture and risks of social disenchantment. All in all, they probably altogether offer a mediocre economic return on political investment.
The other option is closer to that of standard fiscal unions. It would require the expansion of a supranational budget. It would involve granting the EU the power to tax and spend and it would inherit new prerogatives.
Institutionally and politically, this would require some changes to the current framework. The European Parliament should be made to embody the European polity more fully through the creation of transnational lists and it would need to be more representative of European demography and its dynamic. The parliament would then elect the president of the Commission who would truly head the executive branch as well as the European Council and the Eurogroup (this is all possible within the confines of the Lisbon Treaty by the way). This would not only provide the euro area with a real budget and fiscal policy but it would also provide the continent with the executive power that it has been missing in the worse moments of this crisis.
3.3. Possible but is it likely?
The political move toward a stronger fiscal union seems very difficult to make. Moreover in many European countries, euroscepticism is on the rise not on the wane.
The inability to solve the Greek case by accepting the unavoidable losses and sharing it among the Eurozone partners rather than relying on the idea (now proved wrong) that the Greeks alone will be able to repay their debts is one of the main reason for the current lack of confidence. The recent conclusion of PSI go in the right direction, but obviously it is not enough, official sector involvement is needed too to return Greece to solvency and regardless of how politically unpleasant it might be, this discussion should start soon.
So far the increasing snowball of debt has been pushed forward coupled with austerity policies that make the problem worse not better. None of the underlying problems of the Eurozone have been addressed: no central budget, no institutional centre, no lender of last resort, limited expansion of monetary policy (which would help solve the competitiveness problem if inflation were comparatively lower in the debtor and problem countries) even if the LTRO have been very useful, not to talk about the lack of labour force mobility.
European leaders have shown during this crisis that they can only move when at the edge of the cliff. May be is it time to think more strategically, and lead boldly. If not the risk is a long period of low growth with its accompaniment of social unrest and the risk of a growing and more profound rejection of Europe as a project.
4. Let me try to wrap up by underlining two points.
4.1. It is interesting to see how these three different trilemmas are in fact deeply interrelated.
Let’s put the three trilemmas in a table. For each line, only two items can be kept, one as to be rejected to meet the conditions of each trilemma. Table A shows a possible state of the world where the items in the first diagonal (in red) have been rejected.
Table A (see page 7 of the paper : 120310 DSK Cambridge last )
(i) To comply with the Mundell-Flemming trilemma, no fixed exchange rate (i.e. flexible exchange rates) but independence of the monetary policy and open capital accounts.
(ii) Following Rodrik, one choose the preserve democracy and deep economic integration at the expense of less national sovereignty
(iii) And for the Eurozone, according to Pisani-Ferry, we keep the monetary union and in the absence of fiscal solidarity, we have to reject national banking policies.
This state of the world looks very much like the so-called Washington consensus combined with the current rules of the game in the Eurozone. The problem is that a part of this world does not exist and the other part doesn’t work. Not all currencies are flexible, peoples are less and less inclined to abandon their national sovereignty, and the underlying European banking crisis has showed the limits of the current European banking regulation, supervision and resolution model. This table A illustrates the policy responses we have built and highlights their weaknesses.
Another table is possible (see table B) which figures an international monetary system closer to the reality and one we can actually manage. In the real world, some countries have decided to tinker with the impossible trinity and they keep a managed currency and a fairly closed capital account (China and India…). Curbing national sovereignty or relying on international coordination devices can also fail and we should therefore think of taming globalisation and especially financial globalisation to resolve the Rodrik trilemma. Finally, the solution of the European crisis is found in some form of fiscal solidarity.
Table B (see page 7 of the paper : 120310 DSK Cambridge last )
But until now, policymakers have refused to look at the world as it is as opposed to how they would like it to be. In Europe, in particular, the refusal to move rapidly towards fiscal federalism raises the risk of Europe’s failure.
4.2. What are the consequences if Europe fails?
If Europe fails, that is if Europe fails to successfully create a polity that supports a form of pooling of sovereignty above nation states, then hopes for international cooperation in general will vanish as well. In the event of Europe’s failure, either hyperglobalisation would then crush democracy or we would experience a substantial phase of deglobalisation.
Indeed, what this tale tells us is that unless we abandon our commitment to democracy or we learn how to convince citizens to go beyond their national sovereignty we have no other choice but taming globalisation, and particularly financial globalisation, at least as long as the globalisation of democracy lags behind.
This tale teaches us also that it is illusory to dream of a world were all international adjustments will be achieved by market forces alone. It is useless to argue against some countries’ fixed exchange rates when those countries (sometime for good reasons) do not want to open their capital account completely and too rapidly. Hence, to make the IMS work we need to accept the reality check and enhance the role of institutions like the IMF that facilitate external adjustments in stormy weather and reduce their likelihood in fair weather but the IMF’s own governance has to improve for it to remain legitimate and relevant.
Finally, Europe has a special place in the debate about globalisation not so much because of its economic size but because of the importance of its original governance experience for the world at large. In practice, the technical problems are simple; the only non-disruptive way to deal with a monetary union is to accept some fiscal integration, period. The political problems are more challenging. While Europe remains the most advanced attempt at dealing with the forces of globalisation in a democratic fashion and is therefore crucial to ever solve Rodrik’s trilemma, it has to ensure both the internal credibility of its project to convince its own citizens of its relevance and its external credibility to continue to weigh on the world.
What we all need to keep in mind is that to design an efficient IMS, to manage globalisation, to solve the debt crisis and to build Europe, the challenge is not a technical one, it is a democratic one.