Numbers newly reported from the IMF’s COFER data base show that in the most recent quarter, the spring of 2009, the share of central banks’ foreign exchange reserve holdings that they allocate to dollars resumed its downward trend.   The dollar share has been gradually sliding since the beginning of the decade – perhaps because of the birth of a possible rival, the euro, in 1999, or perhaps because of the long-term path of tremendous fiscal and monetary expansion on which the United States embarked in 2001.   

          During the four quarters preceding the most recent one, the share of the aggregate portfolio that the world’s central banks allocated to dollars had temporarily reversed its downward direction.  Arithmetically, the main source of this increase in the dollar’s share was its appreciation against other currencies.   But another source was the action of central banks in industrialized countries, acquiring dollars more rapidly than other currencies.   The movement of the raw quantity shares can be seen in the first graph below, and the movement in the shares properly valued at current exchange rates in the second graph.   (I am grateful to Ted Truman and Dan Xie, both of the Petersen Institute for International Economics, for these graphs.)   

          Whether the temporary reversal from Q2 of 2007 to Q1 of 2008 is measured in quantity terms or in valuation terms, the phenomenon was presumably a (surprisingly strong) safe-haven reaction to the global financial crisis.  Apparently the recent easing of risk and liquidity concerns has now mitigated the flight into dollars.  The central banks that had shifted into dollars have begun to shift back a bit, into euros in particular.

          The gradual downward trend of the dollar’s share during the past decade is a continuation of the trend that began after the end of the Bretton Woods system: from the late 1970s until 1991.  The dollar’s share recovered from 1992 to 2000.  That temporary halt in the longer run trend may have been in part a result of the deficit reduction path that began with George H.W. Bush’s unpopular fiscal reversal and continued through the time of Bill Clinton achievement of fiscal surpluses, until George W. Bush took office and reinstated the chronic deficits.

          The usual response to worries that US macroeconomic profligacy will eventually end the dollar’s privileged position as lead international currency has always been that no asset constitutes a credible alternative for central banks to hold in their portfolios.   I have argued that, since 1999, the euro has constituted a credible alternative.   Based on econometric estimates of the determinants of central banks’ reserve holdings in research with Menzie Chinn, we have even gone so far as to report simulations that show the euro overtaking the dollar by 2022.  Many, like Truman, consider such speculation exaggerated.  They may be right.

           But the euro is not the only alternative to the dollar.  The yen, pound and Swiss franc remain viable alternatives for national authorities to put some of their reserves.  Furthermore, 2009 has seen the resurrection of two international reserve assets that had previously been written off as dead:  the SDR and gold.  My forecast is that we are gradually moving from the dollar standard to a global monetary system that features multiple reserve assets.

Share of central banks foreign exchange reserves allocated to dollars, 1999 QI – 2009 QII       (among industrial countries, among developing countries, and overall)

[Readers wishing to post comments are referred to the SeekingAlpha version.]

My friend Barry Eichengreen, together with Marc Flandreau, has written a column in today’s Financial Times, that appears under the headline “Why the euro is unlikely to eclipse the dollar.” The body of the article is a claim that network externalities and tipping points are not important, or perhaps that they once were but no longer are.

The first two steps of their argument are:
(1) a multiple-currency system is the historical norm. The dollar-denominated system that we have experienced for more than 60 years is an aberration, so network externalities (aren’t) important.
(2) The dollar surpassed the pound in the 1924-25, not in 1948, so lags and tipping phenomena are not important.

Regarding (1), I have always thought that Barry has a good point that a multiple-currency system has one advantage, that it gives the lead currency some competition, which discourages it from abusing its position by excessive deficits, money creation, inflation and depreciation. But I disagree that network externalities are not also important: there will always be an advantage to having a lead currency internationally, just as there is an advantage to having a single money within each country.

Regarding (2), I have no problem dating the pound’s loss of supremacy from the 1920s, if that’s what the eminent economic historians say. But I don’t see how this affects any of the arguments. For one thing, the US surpassed the UK in economic size in 1872, and in exports in 1915. So there is still a lag of between 10 and 53 years.

More importantly, these propositions have no bearing on the central claim that Menzie Chinn and I have made: based on our statistically estimated effects of economic fundamentals, such as the size of euroland — which has recently passed the US economy — the euro now has the potential to rival or even displace the dollar as lead currency. We think we have also found statistical evidence of inertia and non-linearity, which imply a tipping point. But this doesn’t really matter. The scenario that we most emphasize leaves the dollar with an estimated share of central bank reserves only a little less than that of the euro even in the long run. Regardless what one believes about how fast it will occur or how complete the de-throning will be, our claim that the euro will challenge the dollar stands.

My post last week suggested that the euro may overtake the US dollar as premier international currency. One might ask why this would matter. Some of the reasons it matters are economic: we would lose the “exorbitant privilege” of being able to finance our international deficits easily. But there are also possible geopolitical implications.

In the past, US deficits have been manageable because our allies have been willing to pay a financial price to support American global leadership;  they correctly have seen it to be in their interests.   In the 1960s, Germany was willing to offset the expenses of stationing U.S. troops on bases there so as to save us from a balance of payments deficit.   The U.S. military has long been charged less to station troops in high-rent Japan than if they had been based at home. Repeatedly the Bank of Japan, among other central banks, has been willing to buy dollars to prevent the U.S. currency from depreciating (late 1960s, early 1970s, late 1980s).   In 1991, Saudi Arabia, Kuwait, and a number of other countries were willing to pay for the financial cost of the war against Iraq, thus briefly wiping out the U.S. current account deficit.

Unfortunately, since 2001, during the same period that the US twin deficits have re-emerged, we have also lost popular sympathy and political support in much of the rest of the world. Now the hegemon has lost its claim to legitimacy in the eyes of many.  In sharp contrast to international attitudes at the dawn of the century, opinion surveys report that the U.S. is now viewed unfavorably in most countries.    Next time the US asks other central banks to bail out the dollar, will they be as willing to do so as Europe was in the 1960s, or as Japan was in the late 1980s after the Louvre Agreement?  I fear not. 

The decline in the status of the pound during the course of the first half of the 20^th century was part of a larger pattern whereby the United Kingdom lost its economic pre-eminence, colonies, military power, and other trappings of international hegemony.   As some wonder whether the United States might now have embarked on a path of “imperial over-reach,” following the British Empire down a road of widening budget deficits and overly ambitious military adventures in the Muslim world, the fate of the pound is perhaps a useful caution.   The Suez crisis of 1956 is frequently recalled as the occasion on which Britain was forced under US pressure to abandon its remaining imperial designs.  But the important role played by a simultaneous run on the pound, and President Eisenhower’s decision not to help the beleaguered currency through IMF support unless the British withdrew its troops from Egypt, should also be remembered.  

Question from The International Economy Survey of Experts: 
Ten years from now, which will likely be the next great global currency?

My answer: 
Contrary to fevered popular speculation in the 1990s, the yen and the mark never had the potential to challenge the dollar as premier international currency:  their home economies were smaller than the US and their financial markets less well developed and liquid than New York.   The euro, however, is a credible challenger:  Euroland is roughly as big as the United States.  Indeed, evaluated at the most recent exchange rates, the euro economy has just now surpassed the US economy in size.   Also the euro has shown itself a better store of value than the dollar.   These are two of the most important determinants of international reserve currency status.

To be sure, rankings of international currencies change only very slowly.    Although the US surpassed the UK in economic size in 1872, in exports in 1915, and as a net creditor in 1917, the dollar did not surpass the pound as number one international currency until 1945.   In 2005, when Menzie Chinn and I used historical data on central bank holdings of foreign exchange reserves to estimate the determinants, even our pessimistic scenarios did not have the euro overtaking the dollar until 2022.   Thus we could not have asserted that the dollar would be dethroned “ten years from now.”  But the dollar has continued to lose ground.    We have now updated our calculations, particularly to recognize that London is usurping Frankfurt’s role as the financial capital of the euro, notwithstanding that the UK remains outside of EMU.   It is also relevant that — at the most recent exchange rates — the GDP of euroland has surpassed that of the US.   Now we find that the tipping point could come within the ten-year horizon: the euro could overtake the dollar even as early as 2015.

Simulation of central banks’ reserve holdings: € passes $ around 2015 .
Scenario: Only accession countries join EMU in 2010 (UK stays out),
but 20% of London turnover counts toward Euro area financial depth,
and currencies depreciate at the 20-year rates experienced up to 2007.
Source: Chinn and Frankel (2008, Figure 7).