Roger Vicquéry and Kevin Hjortshøj O’Rourke
While the collapse of the Bretton Woods system in 1973 has traditionally been seen as heralding a major shift towards floating exchange rates, the extent of this transition away from fixed arrangements has been called into question by a ‘New Consensus’ view. We provide a new index to measure exchange rate fixity at the global level, which restores the conventional account of international monetary history over the last 70 years: according to our measurement global exchange rate fixity is now only about a third of its Bretton Woods level. We highlight how this transition to floating arrangements was largely driven by anchor currencies ceasing to be pegged to one another.
The standard narrative of the contemporary evolution of the international monetary system, often framed within the context of the international macroeconomic trilemma (Obstfeld et al (2005)), suggests that the world transitioned decisively towards floating exchange rates after the collapse of the Bretton Woods system in 1973. This shift was seen as more compatible with open international capital markets compared to the fixed exchange rate regime that characterised the Bretton Woods era.
However, this conventional wisdom has been challenged by a ‘New Consensus’ view (Ilzetzki et al (2022)). Reinhart and Rogoff (2004) first introduced classifications of exchange rate regimes based on actual exchange rate behaviour rather than official declarations. They argued that when focusing on de facto rather than de jure exchange rate arrangements, post-1973 exchange rates appear much more fixed than previously thought. Ilzetzki, Reinhart and Rogoff (2019), henceforth IRR, extended and updated the original country-level classification. When aggregating up country-level classifications at the global level, by computing the share of countries with fixed-exchange rate regimes (with or without GDP weighting), IRR posit a strong continuity in exchange rate arrangements from the Bretton Woods era to the present along two key dimensions. First, they argue that the prevalence of flexible exchange rate arrangements is only marginally higher today than it was before 1973. Second, they contend that the US dollar’s role as a monetary anchor is as prevalent and, by some metrics, more significant today than it was during the Bretton Woods era. We revisit both conclusions in a recent paper (O’Rourke and Vicquéry (2025)).
One mechanical reason why IRR find higher shares of countries with fixed-exchange rate arrangements, both in raw terms and weighted by GDP, is the classification of eurozone members as having fixed-exchange rates. Their approach is consistent with the macroeconomic trilemma, ie individual countries within the eurozone have given up independent monetary policy, which allows them to achieve a fixed-exchange rate alongside free capital flows. However, it contrasts with the International Monetary Fund’s classification of these countries as floaters. While it is difficult to argue that countries like Ireland or Portugal have floating currencies, so is to consider that Germany moved from a flexible to a fixed-exchange rate with the creation of the euro in January 1999, and that the euro area as a whole is not a floating entity. There is then a degree of arbitrariness in indices of global exchange rate fixity that rely on such judgment calls. For example, if the eurozone countries were to pass a threshold of political integration for them to be considered a single, floating entity, the measures of global exchange rate fixity underpinning the ‘New Consensus’ would shift discontinuously downward.
To address this issue, we introduce (O’Rourke and Vicquéry (2025)) a new method to aggregate country-level exchange rate regime classifications: an index that reflects the probability that two units of GDP, randomly selected anywhere in the world, will come from countries whose currencies are pegged against each other. Such an index boils down to computing the total number of GDP-unit matches involving either fixed-exchange rates or a common currency (thus including within-country matches), relying on the classification of IRR, and dividing this by the total number of possible GDP-unit matches worldwide. The index therefore varies from 0 – a situation where each unit of global GDP has its own currency, all of which float against each other – to 1, in which there is a single world currency or a fixed-exchange rate regime encompassing all currencies.
Our index is therefore invariant to reclassifying the eurozone from being a collection of 20 separate countries, all pegged to each other, to a single entity. In our measurement, such a reclassification would simply shift some GDP-unit fixed-exchange rate matches from being between countries to being within one. Another useful feature of our index is that it allows us to consider indirect fixed-exchange rate relationships, eg the fact that, during the Bretton Woods era, peggers to the British pound were also indirectly pegged to the US dollar, given the former anchoring to the latter. Finally, a key distinctive feature of our index is that it considers the fact that countries that are in a pegged relationship vis-à-vis some partners might well be floating against other partners.
Chart 1 compares our index to measures of the global share of countries with fixed-exchange rate regimes, with or without GDP-weighting, which underly the conclusions of IRR.
Chart 1: Prevalence of fixed-exchange rate arrangements now and then
Note: The chart compares the baseline index presented in O’Rourke and Vicquéry (2025), encompassing all possible ranks of indirect pegs, to a version of the same index relying on direct pegs only, and to aggregate measures of global exchange rate fixity analogous to the ones presented in IRR (the share of countries with fixed-exchange rate regimes, with and without GDP weighting). A higher value indicates a greater prevalence of fixed-exchange rates.
Our baseline index shows a larger shift from floating to fixing, compared to IRR-type measures, in the aftermath of WWII. This is driven by indirect pegs, as anchor currencies became pegged to one another. Indirect pegs are crucial in explaining the high level of fixity during the Bretton Woods period. Both types of indices capture a decline in global exchange rate fixity after President Nixon suspended the convertibility of the US dollar into gold in August 1971 – bringing to an end a key aspect of the Bretton Woods system – and an increase in fixity starting in the 1990s. However, our index aligns with the view that flexible exchange rate regimes have become more prevalent since the 1971 Nixon Shock. Currently, only about 25% of GDP matches are pegged, versus around 75% during the Bretton Woods’s heyday, indicating that global exchange rate fixity is now one third of what it used to be before 1971. This contrasts with IRR-style measures, which show that around 70% of global exchange rate regimes (close to 50% on a GDP weighted basis) have been consistently fixed since the 2000s. Our index also shows the relevance of accounting for indirect pegs when assessing the evolution of exchange rate arrangements in recent history. Comparing versions of our index computed with or without indirect peg links shows that most of the post-Nixon Shock discontinuity can be accounted for by the fact that major anchor currencies stopped being pegged to one another via US dollar anchoring.
Our index can also be tweaked to look at a separate question: the prevalence of anchoring arrangements to a certain currency, regardless of whether exchange rates are fixed (for example as part of a managed float). Here, we focus on anchoring to the US dollar. The nature of the matches is in this case different as anchoring is asymmetric: while the United Kingdom might anchor to the US dollar, the opposite is not true, although anchoring might then result in a symmetric pegging relationship between the US and the UK.
Chart 2: Prevalence of US dollar anchoring now and then
Note: The chart compares the baseline index of US dollar anchoring presented in O’Rourke and Vicquéry (2025), encompassing all possible ranks of indirect anchoring, to a version of the same index relying on direct anchoring relationships only, and to aggregate measures of US dollar anchoring analogous to the ones presented in IRR (the share of countries anchored to the US dollar with or without GDP weighting). A higher value indicates a greater prevalence of US dollar anchoring.
Chart 2 again contrasts our US dollar anchoring index, with or without indirect linkages, to the share of countries anchored to the US dollar, with or without GDP weighting. Consistent with the claim by IRR that dollar anchoring is by some metrics now higher than it was during the Bretton Woods era, the share of countries anchored to the dollar has increased from around 40% prior to the Nixon Shock to more than 50% today. The GDP-weighted measure shows current levels of dollar anchoring slightly lower (roughly 70%) than during the Bretton Woods peaks (roughly 80%). Our index, however, tells a different story. Considering only direct anchoring, dollar anchoring declined from a peak of roughly 40% of GDP-unit matches to a stable level of 20%–25% post-Bretton Woods. Including indirect anchoring shows a halving of global US dollar anchoring since Bretton Woods, from nearly 100% of GDP-unit matches to around 50% today. Interestingly, the rise of dollar anchoring in the ‘fear of floating’ 1990s is almost entirely driven by indirect linkages, ie emerging markets finding themselves indirectly anchored to the same currency.
Our new measurement of global exchange rate fixity potentially sheds new lights on other important secular trends in the international monetary system, including the dominant currency paradigm (Gopinath et al (2020)) and the global financial cycle (Miranda-Agrippino and Rey (2022)). For example, the global rise of dominant currency pricing (Boz et al (2022)) as well as the decline of FX volatility among major currencies (Iltzetzki et al (2020)) since the end of Bretton Woods can be thought of as a partial substitute for declining exchange rate fixity.
Roger Vicquéry works in the Bank’s Global Analysis Division and Kevin Hjortshøj O’Rourke is a Professor of Economics at Sciences Po and Directeur de Recherche at the CNRS.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
Share the post “How fixed are global exchange rates?”
