The case for regular SDR issues: Fixing inconsistency in balance-of-payments targets

John Williamson

02 October 2009



The Special Drawing Right (SDR) was created by the IMF in the late 1960s as its very own gold-substitute and first allocated to IMF members in 1970. Subsequent issues in the following two years were agreed at the same time, and then a new 3-year period of modest allocations occurred in 1979-81. Despite the laughably ambitious aims for the new asset that were voiced by the Committee of Twenty in 1974, it subsequently went into a deep hibernation from which it was only woken recently as a consequence of the decision of the London G20 meeting to recommend a $250 billion allocation (Reisen 2009). Together with the call by the Governor of the People’s Bank of China to consider creating enough SDRs to allow countries like China to diversify some of their dollar holdings (Zhou 2009), this has rekindled interest in assessing whether the SDR could play a useful role in reforming the international monetary system to fit it for today’s world (Rodrik 2009).

Lots of difficult and ambitious reforms of the SDR, such as making it an asset that the private sector would be allowed and anxious to hold, are periodically proposed (Aiyar 2009). This column, based on a longer piece, does not presume any of these reforms. It merely argues that the world would be a better place if a substantial portion of the demand for holding additional international liquidity were to be met by creation of more SDRs instead of exclusively by expanding reserve currency holdings, as in the recent past. (By a “substantial portion” I mean perhaps half, which – to judge by the additions to reserves in the period 2002-06 – would imply SDR allocations of over $400 million per year; see Williamson 2009 for details.)

There are four differences in satisfying the increased demand to hold reserves by allocating SDRs rather than permitting countries to hold increased quantities of reserve currencies, which means principally dollars.

  • Reserve holders need to convert SDRs into dollars in order to use them, whereas dollars are immediately usable in intervention (Humpage 2009). It is in order to remedy this minor inconvenience that it is claimed to be necessary to extend holdings of the SDR to the private sector. Under present arrangements SDRs are converted by a central bank with dollars to spare when it is requested to do so by the Fund. No country is known to have encountered difficulties in making desired conversions.
  • If arbitrage serves to equalise the expected returns from holding any of the currencies that compose the SDR (as is implicitly assumed), then the expected return from holding SDRs will also be the same. However, there is a systematic difference in the second moment; for the typical country (i.e. excluding countries whose trade is dominated by the US), the variance of the SDR in terms of its imports is less than the variance of the dollar (by construction of the SDR as an average of the principal trading currencies).
  • Whether there is a net national advantage in one’s currency being used as a reserve currency is a much disputed subject, but no one has denied that there is a gross advantage. This gain is referred to as seigniorage (the profit from the privilege of issuing money), and under a reserve-currency system primarily takes the form of borrowing more cheaply. Under the SDR seigniorage is distributed around the world (in proportion to IMF quotas), which seems a fairer basis than to a single (or a handful of) reserve-currency country(ies).
  • Dollar reserves are supplied to the rest of the world insofar as the US runs an overall deficit (on current plus capital account). While in principle the US could run a current account surplus and a larger capital account deficit that supplied the world with the increased reserves it wishes to hold, the reluctance of most other countries to run large current account deficits is likely to make this impractical. An inconsistency in balance-of-payments targets of this type can be eased, or possibly eliminated, by SDR allocations.


Most of us can think of other reforms to the international monetary system that we would like to see. But it is quite difficult to think of other reforms that promise equally profound benefits to the world and which demand so little in the way of change. SDR allocation involves merely making use of what already exists, not making big and difficult reforms. Yet grasping this opportunity offers the surest way of reducing the inconsistency in payments objectives that currently looks to be the biggest obstacle to a strong recovery in the global economy in the post-Lehman world.


Aiyar, Anklesaria (2009). “An International Monetary Fund Currency to Rival the Dollar?” Washington: Cato Institute.

Humpage, Owen F. (2009), “Will special drawing rights supplant the dollar?”, 8 May.

Reisen, Helmut (2009), “Shifting wealth: Is the US dollar Empire falling?", 20 June.

Rodrik, Dani (2009), “Why Don't We Hear a lot More About SDRs?” Debate: Macroeconomics, a Global Crisis Debate,, 4 February.

Williamson, John (2009), “Why SDRs Could Rival the Dollar”, Peterson Institute Policy Brief, Number 09-20.

Zhou Xiaochuan (2009), “Reform of the International Monetary System”. Beijing: People’s Bank of China.



Topics:  Global economy Global governance

Tags:  reserve currency, IMF, Special drawing rights

Senior Fellow, Peterson Institute for International Economics