Last month was the 60th anniversary of the conference at Bretton Woods, New Hampshire, which inaugurated the post-second world war international economic order. The flood of analysis that this occasion brought forth has concentrated on that meeting’s institutional progeny: the International Monetary Fund and the World Bank. But a bigger question needs to be addressed. It is whether floating exchange rates have proved to be the ideal replacement for the unsustainable adjustable exchange-rate pegs of the Bretton Woods monetary regime. The answer is: no. Consider the following features of the world economy today: the world’s richest country is, far and away, its biggest capital importer and net debtor; the two periods of sizeable net lending to emerging market economies over the past three decades ended in financial crises and sharp reversals in lending (see chart); and, since the most recent set of global crises, emerging market economies have, in aggregate, accumulated enormous quantities of foreign currency reserves. Participation in two recent conferences, both in Italy, has persuaded me that a single factor helps explain these phenomena: global currency instability. The first of these conferences - organised by the Aspen Institute Italia - discussed “how to live with imbalances”. The second - the Santa Columba conference, organised by Robert Mundell, the Nobel-laureate - considered “plans for a world currency”. Taken together, these meetings encouraged me to question a long-standing belief in the desirability of floating exchange rates. Look back to the last period of economic globalisation - the era before 1914. Under the gold standard, all countries used what was in effect one currency. While currency risk existed, it was limited. Today, however, it is huge. Since the breakdown of the Bretton Woods system in 1971, currencies have fluctuated widely. The concept introduced principally by Ricardo Hausmann, now at Harvard, to explain the consequences of currency instability for emerging market economies is “original sin”. This catchy phrase refers to “the inability of a country to borrow abroad in its own currency”.* Ninety-seven per cent of all debt placed in international markets between 1999 and 2001 was denominated in just five currencies: the US dollar, the euro, the yen, the pound sterling and the Swiss franc. Even well-run emerging market economies, such as Chile, cannot borrow in their own currencies. Whatever the explanation for this difficulty, the limited currency composition of global lending has powerful consequences for capital flows. By definition, net borrowing then creates a potentially lethal currency mismatch. When a currency falls sharply, net borrowers will experience large balance sheet losses. Many financial institutions will be wiped out as a result of the insolvency of any debtor that is burdened by large net foreign currency liabilities. Having either experienced this pain or watched other countries do so, competently run emerging market economies have tried to limit their net foreign currency liabilities. This has been particularly true of the Asian emerging market economies. These countries have attempted to preserve robust current account positions. They have also recycled inflows of foreign direct investment into official holdings of foreign currency assets, principally US treasury securities. Fortunately for them, it is vastly easier to keep currencies down than to prevent them from collapsing. Indeed, given the very low interest rates of the savings-surplus countries of east Asia, sterilisation of the monetary consequences of exchange-rate intervention aimed at keeping the rate down is close to costless. Yet this does not end the dilemmas confronting such emerging economies. As Ronald McKinnon of Stanford University has argued, the more successful countries are in strengthening their balance of payments and limiting net foreign currency liabilities, the more they will suffer from “conflicted virtue”.** They will, in other words, find themselves under growing political pressure to allow their currencies to appreciate. Such appreciations risk pushing low-inflation countries into deflation and, perhaps, even into a liquidity trap of the kind suffered by Japan in the 1990s. That has been a concern to China. Letting the currency appreciate while offsetting the contractionary impact through expansionary macroeconomic policies would risk creating a current account deficit. Should equity inflows halt, this would have to be financed by potentially destabilising net foreign borrowing. Now consider the consequences of a world in which emerging market economies, in aggregate, wish to limit net foreign currency borrowing, sustain robust positions in their underlying balance of payments and resist pressures for currency appreciation. Presume, in addition, that the world also contains rich, ageing societies with structural excesses of savings over investment and so enduring current account surpluses: Japan is the most important, but far from only, example.We would then expect the net private capital flow to emerging market economies to take the predominant form of equity investment, as we observe. We would also expect macroeconomic balance in the world economy as a whole to require at least some of the countries that can borrow easily in their own currencies to run large current account deficits. The US is, of course, the paramount example. A world in which borrowing abroad is hugely dangerous for most relatively poor countries is undesirable. A world that compels the anchor currency country to run huge current account deficits looks unstable. We should seek to lift these constraints. The simplest way to do so would be to add a global currency to a global economy. For emerging market economies, at least, this would be a huge boon.* Barry Eichengreen, Ricardo Hausmann and Ugo Panizza, “Currency Mismatches, Debt Intolerance and Original Sin”, NBER Working Paper 10036; ** Ronald McKinnon and Gunther Schnabl, “The return to soft dollar pegging in east Asia”, International Finance, forthcoming xref martin.wolf@ft.comI am well aware of the economic and political objections to this idea. But if the global market economy is to thrive over the decades ahead, a global currency seems the logical concomitant. In its absence, the world of free capital flows will never work as well as it might. This is a world I am unlikely ever to see. But maybe my children or grandchildren will do so.