The recent depreciation of the US dollar sparked debate about its possible dethroning in the near future. Such discussion is not new. It has occurred repeatedly each time the US dollar faces adverse circumstances. However, the recent combination of financial turmoil, a significant current account deficit, and bear speculation provides a favourable background for renewed concerns. The attention these phenomena draw themselves is understandable. But do they warrant worry about the dollar?
One view, which has been mapped onto contemporary empirical evidence in a noted paper by Menzie Chinn and Jeff Frankel, is that rankings of international currencies change only very slowly, but when they change they do so with a bang. As the story goes, although the United States surpassed the United Kingdom in economic size very early (between 1860 and 1875 depending on estimates), the dollar did not surpass the pound as the number one international currency until 1945. When it did, however, the pound sterling was permanently dethroned.
Now, Chinn and Frankel note, the US dollar has steadily lost ground in the recent past. London is usurping Frankfurt’s role as the financial capital of the euro, notwithstanding that the United Kingdom remains outside the EMU. They say that a tipping point could come within the ten-year horizon: the euro could overtake the dollar even as early as 2015.
The reason usually put forward to account for the existence of tipping points is network externalities, or if you prefer, agglomeration economies. Currencies are a bit like malls: you go there because you expect to find a variety of services. Other customers do the same and sellers understand it. In the end, the mall is amply furnished with all amenities. And thus malls or more generally markets may survive somewhat longer than the reasons that led to their emergence in the first place. Yet a point comes where the underlying geography does change so much that previously active malls become deserted. Customers just start going somewhere else. We observe the collapse of a formerly booming market place. Such are the economics of tipping points.
But what do we really know of the reasons for the long delay in the ascent of the US dollar during the second half of the nineteenth century? In recent work I co-authored with Clemens Jobst, we took a fresh look at the issue. Gathering information on the international reach of every single currency in the world, we articulated a model that enabled us to test for alternative determinants of international currency status. We found strong evidence of size effects: the currencies of large trading powers tended to circulate widely. We also found evidence of financial persistence. The currencies that were actively sought by a large number of countries tended to be more liquid and, as a result, were more attractive, which boosted their use and liquidity. Figure 1 summarizes the evidence. It compares the outcome from a history-free scenario where there is no persistence with the actual performance of alternative currencies. As seen, the US dollar was penalised by history, while some European currencies (in particular the French franc) were rewarded with a larger actual share than the history-free counterfactual suggests.
Figure 1 The history-dependent circulation of currencies, 1900
Note: The graph presents the number of foreign markets in which currencies circulated.
Sources: Flandreau and Jobst (2008).
Next, we examined whether the model we estimated generated strict persistence – whether currencies that had once reigned supreme would always reign supreme (this is essentially a property of the elasticity of monetary demand functions to externality effects). Results unambiguously rejected this stronger form of persistence. One prediction of our model was that the US dollar ought to have managed to rival the pound sterling in the interwar period, rather than post-1945 as conventionally believed.
Figure 2 simulates the incidence of an increase, other things being equal, of US trade with the rest of the world. As can be seen, the simulated equilibrium is slightly above the observed one. History matters, but not that much. Or if you prefer this one: we cannot lay all the observed persistence at the door of network externalities.
Figure 2 US trade and the US dollar, 1900
Note: Popularity is measured as the percentage of markets using the US dollar.
Sources: Flandreau and Jobst (2008)
Compelling evidence tells us history matters, but its importance ought not to be overstated.
The importance economists assign to agglomeration economics stems from the mismatch in the first half of the 20th century between the dollar’s role in global commerce and the US’s share of the global economy. Our evidence suggests that fundamentals can explain this mismatch.
First, the US, a protectionist economy in the 19th century, was not a leading trade power in Europe until fairly late, so its share of global GDP mis-measures its importance in trade. Second, the US and New York were not near the routes of European commerce, so the US was not conveniently located to capture a significant share of trade financing; distance matters since the trade consignments themselves are natural collateral for such financing.
In plain words, a large fraction of the “delayed” ascent of the US dollar in the 20th century is fully in line with fundamentals – once they are properly measured. We do not require references to the forces of history and agglomeration economics.
Occasionally, it takes an economic historian to remind his economist colleagues that history may not matter as much as one would want to believe.
Marc Flandreau and Clemens Jobst (2008) “The Empirics of International Currencies: Network Externalities, History and Persistence” (forthcoming, Economic Journal).
Will the Euro Eventually Surpass the Dollar as Leading International Reserve Currency? (2007). Menzie Chinn and Jeffrey A. Frankel, in G7 Current Account Imbalances: Sustainability and Adjustment, R. Clarida (ed), University of Chicago Press.