The impact of international fragmentation and the role of the US dollar

Below are Mr Tombini’s remarks in policy panel 2.7 (“Rising impact of financial fragmentation on the international monetary and financial system”).

Increasing geopolitical tensions pose the danger of geoeconomic fragmentation. A notable development on the trade side has been the decline in the share of US imports sourced from China. On the financial side, there is much talk of reducing the reliance on the US dollar, with a number of initiatives grabbing the headlines.

Let me structure my remarks as follows: I’ll first say a few words on how I see the international role of the US dollar. I will then explore the consequences of a dollar-centred global monetary system before drawing some lessons for the international safety net.

The dollar remains the number one currency in the global financial system by several measures. The dollar’s dominance is particularly stark in FX markets, where it is on one side of 88% of all transactions. This is quite noteworthy because just over 20 years ago many expected the dollar to lose market share to the euro. This has not happened. BIS Triennial Surveys show that the dollar’s share in global FX trading has remained stable while the euro’s share fell from 38% in 2001 to 31% in 2022. By comparison, the renminbi’s share is still at 7%, despite its rapid ascent in recent years.

The US dollar is also the most important currency in international banking. The size of foreign currency assets and liabilities of the banking system has grown immensely over the last two decades, but the share of the dollar in those assets and liabilities has remained remarkably constant at around 60%. The dollar also dominates debt security issuance and trade invoicing, although in the case of the latter with significant differences across regions.1 

So why does the dollar remain so important?

The size of the US economy is clearly a factor. But it is certainly not the only one. The US accounts for just over 20% of global GDP and just over 10% of global trade. If size were the main determinant, the euro and the renminbi would rank on par or even ahead of the dollar.

I believe that the dollar’s importance owes to the depth and accessibility of the US market, in addition and on top of the importance of the US economy. There is also a great deal of path dependency.

As Governor of the Central Bank of Brazil, I was a reserve manager, among other things. Similarly, as a member of the BIS’s senior management I share responsibility for our banking operations. In both cases, I found that there really is no alternative to the dollar in the short and medium term. International reserve managers tend to seek safety, liquidity and return, in roughly that order. US Treasuries are hard if not impossible to beat on the first two criteria, at least. US Treasuries remain the closest thing to a global safe asset despite budget stand-offs, averted and non-averted shutdowns and rating downgrades.

They are also the most liquid fixed income assets in the world. Liquidity in the Treasury market has declined in recent years, but so has that in most other markets. It is therefore not surprising that around 60% of international reserves continue to be denominated in US dollars.

Going a bit deeper, the depth of the dollar market owes to a confluence of factors that is hard to replicate elsewhere.

First is the size of issuance. While total public debt in the euro area is also very high, liquidity is fragmented, which limits the amount that can be traded at any given point in time.

Second, the dollar is freely convertible.

The third point is the diverse investor base.

Fourth is the rich ecosystem of ancillary players, ranging from rating agencies to custodians and market infrastructures.

Last but not least, US authorities, in particular the Federal Reserve, have shown that they care deeply about the smooth functioning of the Treasury market and other financial markets. They have put in place a number of backstops to keep these markets open. Let me just mention the Foreign and International Monetary Authorities, or FIMA, facility at the Federal Reserve Bank of New York, where central banks can exchange their Treasury securities for cash.

All this does not mean that the role of dominant reserve currency cannot flip. Indeed, we have seen several changes in the past, from the pound sterling to the dollar in the early 1920s, back to sterling after the US left the gold standard some 10 years later, and back to the dollar some years later. We have also seen periods in which a dominant currency competed with several others, for instance the French franc in the 1920s and the euro today (Eichengreen and Flandreau (2009)). We could well see a move towards more diversified reserve portfolios or regional vehicle currencies.

Of course, a global financial system based on one lead currency does have drawbacks. The most important relates to international spillovers. Every economic policymaker knows that changes in US interest rates will have major implications for financial conditions in their economy. There is also a large academic literature that substantiates this point (see Caballero and Upper (2023)). Importantly, in addition to the trade channel that we learned about at university, there is a financial channel, based on currency mismatches on either the borrower’s or the lender’s balance sheet. This links key macroeconomic variables to the dollar exchange rate. Importantly, the financial channel works in the opposite direction to the trade channel: a weakening of the domestic currency against the dollar may result in a contraction rather than the expansion predicted by the trade channel.

Spillovers will not disappear in a global financial system with multiple lead currencies, even if they might be dampened by diversification.

Crucially, there are also ways of mitigating spillovers in a world of dollar dominance.

The first line of defence is clearly putting one’s house in order. In addition to strong fiscal and external positions, this includes having strong macroeconomic policy frameworks in place that give policymakers the tools to respond adequately to a wide range of shocks.

Unfortunately, strong fundamentals are not sufficient to prevent countries from running into difficulties. This provides a rationale for an international, ideally global, safety net, that could help countries smooth the adjustment to shocks and bridge potential liquidity gaps. Bilateral arrangements such as central bank swap lines or regional agreements, such as the Chiang Mai Initiative Mutualisation in Asia and the European Stability Mechanism in Europe, have an important place in the global safety net. But they also have their limitations, namely that they need to fit into the political and economic context of the granting country. Only multilateral institutions such as the IMF can provide universal and rule-based access for all.

But the global safety net only works if the institutions at its core have adequate processes and sufficient firepower to fulfil their functions. Regarding the latter, it is important to replenish the firepower of the IMF. In 1980, the IMF’s total resources amounted to roughly 2.3% of global external liabilities. Today, the quota resources of the IMF stand at less than 0.25% of global external liabilities. Borrowed funds can lift the IMF’s lending capacity to just above 0.5% of global external liabilities. This is not very much and could undermine the IMF’s ability to deal with liquidity problems in a large economy or several medium-sized ones at once. We urgently need to prepare for such a possibility.

Let me conclude by summarising my two main points:

  1. While we may see a shift towards a global financial system with multiple lead currencies at some point in the future, any changes are likely to remain gradual. The dollar is likely to remain the dominant currency for some time.
  2. There are ways to deal with the international spillovers that the current arrangements imply. In particular, we need to: (i) further strengthen our macro-financial policy frameworks; and (ii) ensure there is an effective global safety net.


Caballero, J and C Upper (2023): “What happens to EMEs when US yields go up?“, BIS Working Papers, no 1081.

Eichengreen, B and M Flandreau (2009): “The rise and fall of the dollar, or when did the dollar replace sterling as the leading reserve currency?” European Review of Economic History, vol 13, no 3, pp 377–411.

1      Ninety-six per cent of trade in the Americas is invoiced in dollars, compared with 74% in Asia-Pacific and only 23% in Europe.

Leave a Reply

Your email address will not be published. Required fields are marked *