Whither the Dollar? The outlook for the world’s reserve currency


For some time now players in the world’s currency markets have been debating how long the US dollar can continue to maintain its value in the face of the large and persistent US current account deficit.  The recent sharp fall in the dollar’s value, especially against the Euro, has prompted investors to ask whether this is the start of a major correction which sees the dollar move to a substantially lower level.  Even the Americans themselves – better known for a studied indifference to the external value of their currency[1] – have begun to talk about the foreign exchange market, with no less a person than Alan Greenspan opining publicly on the outlook for the dollar and the effect of the trade deficit.

The immediate cause of the dollar’s current decline – the position of the US current account, and the imbalance between saving and consumption in the US economy that is driving it – is not in dispute.    Of more interest to investors are the two questions “how low will the dollar go?” and (perhaps more importantly) “will it come back?”  And with central banks among the very largest of investors in the dollar, through the enormous holdings of US Treasuries in official reserves[2], these questions are being asked with no less interest among central bank reserves managers than they are by other market participants worldwide.

The long term prognosis for the dollar

The most important issue in answering these two questions is to distinguish between short term movements in the dollar – even if substantial – and the long term outlook for the currency, which concerns more fundamental issues such as the position of the dollar as the world’s reserve currency. On this hangs the second of our two questions, for if we are witnessing the start of a long term and more permanent move in the dollar to lower levels, the response from investors will be very different.

There is no doubt that one day, the dollar will no longer be the world’s reserve currency. No currency holds that position for ever, just as no country holds the position of hegemon or superpower for ever.  At the height of a country’s or currency’s power, it is hard to imagine that power ever ending – the British Empire in the last years of the 19th century looked set fair to go on for ever, and the role of sterling likewise.  But there is a natural cycle of such power, and it seems to be around 100 years long[3].   This cycle has broadly held good ever since the time of Portuguese supremacy from about 1450-1530, through the time of the Spanish, who were dominant from 1530 to about 1640, the Dutch from 1640 to about 1720, the French from 1720 to say 1815, the British from 1815 to 1920, and from then until the current day the United States.

This suggests that the time of US hegemony – while not about to end overnight – is in its last 20 to 30 years or so.   That would fit quite neatly with much touted view that China, or possibly India, will emerge as the dominant power of the 21st century.  And if the era of American dominance passes, then the era of the US dollar as the world’s reserve currency will also pass.  At the moment neither the Renminbi nor the Rupee qualify as a potential world reserve currency, but one day they will both be freely traded, and with the size of their respective economies behind them they will command respect and challenge the dollar’s role.  Which one takes over that role and becomes the reserve currency of choice, and when, one cannot at the moment say, but what we can say is that it is most unlikely that the dollar will preserve its unique position beyond say 2030.

And when the dollar does relinquish its current role, there is every reason to expect it to settle at a substantially lower level. Not only will investors and central banks have less reason to hold dollars, but also when their current holdings (especially treasury holdings) mature, they will increasingly seek to be repaid in the new “hard” currency.  In other words there will be less demand for dollars either for transaction balances or as a “store of value” currency[4], and increasing pressure on the US to acquire foreign currency to finance itself.

However, markets, and even central banks, seldom have the luxury of planning for events which may or may not happen in 25 years! And few investors, even those planning for the very long term, are able completely to ignore market movements in the interim.  So while we recognise that the era of the dollar will eventually come to a close, investors need to look more to the shorter term outlook when shaping their portfolios.

The immediate outlook for FX markets

Most market commentators agree on the main causes of the dollar’s current weakness, viz excessive US consumption or deficient US savings, both of course sides of the same coin.  These are the results of policy decisions, and so, like any policy decision, they are reversible – there is no reason why the US could not correct its imbalances and return to a position of greater equilibrium if it wished to, and if it did so, this would probably lead to a period of renewed dollar strength.  And although at the moment the US administration seems determined to stay with its current policies, as long as a rebalancing of the US economy cannot be ruled out, our view is that the dollar is not inevitably and irrevocably on its final slide yet, and it is at least possible that the dollar will regain its strength over the cycle – maybe in 2 years, maybe in 5.

Given this analysis, we suspect that most central banks, like many other investors, will hold onto a substantial part of their current dollar positions, not least because they have too much already invested in the dollar trade to give up now. There may be, and indeed there already has been, some central bank currency reallocations at the margin, but we do not see a major shift out of dollars from central bank reserves managers just yet. Like any investor with a major position, they face the challenge that selling significant amounts of their dollar holdings would merely risk intensifying the devaluation of the dollar assets they still retained, and the gain from any sales would be overwhelmed by their unrealised losses on their remaining portfolio.

Nor would these losses be insignificant. For many central banks, dollar assets form a major part of the asset side of their balance sheets, and a substantial dollar depreciation would be very damaging for their profits, as assets denominated in dollars fall in value, while liabilities, being in domestic currency in the main, do not[5].

As an example, consider a country where the reserves are all in dollars and represent 20% of GDP (several countries have holdings of this magnitude or even higher). A 25% decline in the dollar versus the domestic currency would create an unrealised loss of 5% of GDP, which when transferred to the balance sheet would require a similar sized transfer from the profit & loss account to cover it.  Since most central banks pay a significant proportion of their profits to their national exchequers, this results directly in a substantial loss of revenue for the government.  And in the case of really large revaluation losses, the effect may be to exhaust not just the profit & loss account but also the central bank’s equity, leading to a direct need for recapitalisation from the government, which is both costly to the national budget and very damaging to central bank independence.

For Asian central banks, who hold the major portion of the world’s reserves, there are other considerations as well. Major shifts out of dollars would inevitably lead to a general unpegging of the region’s currencies (most of which are at the moment more or less formally tied to the dollar), which would precipitate considerable uncertainty and jeopardise regional trade.

So, while the US appears largely indifferent to the state of the dollar, preferring to maintain domestic consumption, its fall could have serious ramifications for other countries beyond just terms of trade effects.  What can other countries do to make the US authorities change their attitude of studied neglect, especially as appeals to the US to reduce the twin deficits are clearly falling on deaf ears?

As President Clinton reputedly discovered to his chagrin in 1993-94, there is one way in which even the most determined administration can be made to listen to market moves, and that is through the link between bond yields and the stock market.   The stock market cannot for ever ignore the bond market, as sharp moves up in long term yields will be negative for the US economy.  And even the president cannot for ever ignore a falling stock market.  So the trigger for a reaction from the US will probably be when rising yields in the bond market spark a fall in the stock market, and as a result, if the falling dollar is seen to be driving up bond yields to painful levels, then the US may re-think its position on its currency.

What might bring about such a move in US bond yields?  As we have discussed, it is unlikely that there will be major sales of central bank holdings, and even if there were, the short duration of most official holdings would mean that the effect on longer Treasury yields would be muted.  Instead, any rise in longer yields would be mainly driven by perceptions of rising inflation or a fall in demand for new issuance.  Yields have already started rising as a result of the recent moves in the dollar, as investors fear that imported inflation will start to have an effect, and further dollar depreciation can only exacerbate these concerns.  And a bear market will make it ever more difficult for the US Treasury to sell its bonds at their regular auctions, especially if central banks prove reluctant to take part.

In summary, we do not see the current decline in the dollar as irreversible. On the one hand, the world’s creditor central banks will probably aim to resist too great a dollar slide:  not only would it damage the world economy, now heavily reliant on the United States as “the consumer of last resort”, but the direct costs for the central banks themselves would, as we have shown, be very high as well.  And on the other hand, although the US sometimes seems to act as though their current account is “America’s deficit but the World’s problem” (to paraphrase the famous quote that “the US dollar is our currency and your problem”), perhaps the financial markets will show the US that it is their problem as well, and even the current administration may eventually decide it should act to stem the dollar’s slide.

Looking to the medium term

It is one of the consequences of the dollar being the dominant currency that market effects to counter imbalances in the US economy are very much more muted and weaker than they are for any other country.  The position of issuer of the reserve currency allows the US to run greater imbalances for longer, and borrow more freely (and at a lower rate) than other countries can, and this allows the imbalances to persist and build until they are substantial[6].

In the present position, the US imbalances are indeed significant.  Exports are running at around half the level of imports only, a gap which it is very difficult to solve simply by increasing US exports.  Instead, any resolution of the deficit would have to include an element of import reduction as well.  This can be achieved in one of two ways only – import substitution (sales of home produced items in place of imported goods), or if that is not sufficient, a straightforward cut in overall consumption.

Neither of these can be achieved in a short timescale, even if the political will was there in Washington to do so.  Import substitution requires long lead times as American manufacturers regain domestic market share, and overall cuts in consumption require a different political agenda.  And the level which the dollar would have to fall to in order to bring either or both of these about is probably lower than the creditor nations of the world could accept.

So while we do not think the dollar has embarked on its final descent into being an “ordinary” currency (ie, not a reserve currency), and while we therefore think that the current move is part of a cycle not a structural change, so that the US currency will eventually strengthen again from its current levels, it is also possible that the current trend to a weaker dollar may be in place for some time. Indeed, one could argue that it has already been in place for 4 years, since the euro touch $0.83 in October 2000.

The main medium to longer term consequence may well be to make central banks and other investors more wary about building up such huge dollar balances next time, and we envisage that, while central banks will not be keen to sell large amounts of dollars now, at this level, they may be quicker to move out of the dollar into other currencies such as the euro (or possibly gold) on any sign of dollar strength. Which in turn might suggest that the next dollar peak will be lower than it might otherwise have been[7]. But we believe that the greenback has not lost its status as a reserve currency yet, and as long as it has that status, central banks will continue to hold, and need, major holdings of America’s much battered dollar.


[1]              Best illustrated by then Secretary of the Treasury Donald Regan’s comment “There is nothing wrong with the value of the dollar.  It is still worth 100 cents” (in a discussion with the author, when the dollar was approaching record levels against all currencies in late 1984).

[2]              Foreign central banks are estimated to hold well in excess of $2 trillion of US dollar-denominated securities, of which around $1.85 trillion are in Treasuries   (Source: US Treasury, September 2004).

[3]              This recalls the English saying “Rags to rags in 3 generations” – broadly, the first generation earns wealth and power, the second preserves it and the third, which has no knowledge of the hard work which generated the wealth, squanders it.

[4]              Evidence from when Sterling relinquished its role as the reserve currency suggests that the country losing that status suffers a prolonged period of pressure on its currency as the “post-dated cheques” of its borrowing come home and creditors demand payment, and a prolonged period of sub-trend growth as efforts are made to meet those creditors’ demands.  In this light, the main consequence of President Bush’s economic strategy, with its very free use of America’s current ability to borrow, may be to hasten slightly the demise of the dollar as the reserve currency, and more importantly to prolong and deepen significantly the cost which the US will eventually have to bear as all their borrowings fall due.

[5]              The current accounting trend towards more market-based valuations of reserves, plus a move towards reflecting even unrealised exchange losses in Profit & Loss accounts rather than in reserve accounts, will exacerbate this.  The greater transparency which these accounting changes are designed to engender will come at the cost of greater volatility in central bank accounts, and greater risk that losses will eclipse capital and make central banks technically insolvent.

[6]              One consequence of this is that trends in the reserve currency tend to last for longer, like the rise in the dollar to 1985, and its fall thereafter all the way to its low point in 1995 (recall that in March 1995, the $‑DM rate was around DM1.35, equivalent to $1.45 to the euro).

[7]              An interesting analogy is with the gold market, which suffered from “central bank overhang” for much of the 1990s as central banks looked to lighten their holdings on any sign of strength in the gold price.  The market only stabilised when central banks agreed at the end of the decade to limit their sales to avoid a “devil take the hindmost” rush for the exit.  Given the size of central bank dollar holdings, it may take similar co-ordination to avoid a similar instability when central banks decide to exit the dollar.

This essay was first published by State Street Global Advisors as part of their Insight programme