The opening sentence of Robert Mundell’s 1963 paper “Capital mobility and stabilization policy under fixed and flexible exchange rates” — one of the two most influential in a series of pathbreaking papers he published in the late 1950s and early 1960s — is curious: “The world is still a closed economy, but its regions and countries are becoming increasingly open.”
“Still”? Was Mundell thinking of a future with interplanetary trade, so that eventually the world as a whole wouldn’t be a closed economy? OK, he probably wasn’t, but if he was, it would have been in character. Mundell, who passed away on 4 April, was an economist ahead of his time.
Specifically, those seminal papers were written in an era when many of the restrictions imposed on international transactions during the Depression and WWII were still in place. Britain’s foreign exchange controls persisted until Margaret Thatcher came to power; France didn’t abolish its controls until 1989. Yet in those papers Mundell envisaged a world with high mobility of capital and perhaps other factors of production; indeed, his stabilisation paper made the strategic assumption of perfect capital mobility, with money flowing instantly to equalise rates of return across countries.
And over the decades that followed, as capital flows surged and fixed exchange rates gave way to floating rates, Mundell’s work provided an essential guide.
In what follows, I’ll try to explain Mundell’s contribution to economic thought and policy.
Let me admit from the outset that the trajectory of Mundell’s ideas makes this a tricky project. Most of his influence within the economics profession comes from a handful of brilliant papers written when he was very young; most of his public prominence came from arguments he made later in his career, which often seemed to conflict with his earlier work.
Now, great economists often change their views over time, as they should when new information arrives. Mundell, however, changed his whole intellectual style; if you were to read his Nobel lecture without knowing who wrote it, you might never have guessed that it was the same man who devised those crisp little models several decades earlier.
But let me begin with those models, which remain the foundation of modern international macroeconomics.
When Mundell was awarded the Nobel Prize, I was among a number of economists who noted that his most influential work seemed inspired by Canadian experience. In retrospect I may have understated the case: the Canadian model arguably underlay all three of Mundell’s key contributions to international macroeconomics.
As I already pointed out, in the late 1950s and early 1960s capital movements were in general circumscribed by extensive controls. Yet Mundell found it useful to posit a world of perfect capital mobility, partly for analytical clarity, but also because it was “a stereotype towards which international financial relations seem to be heading.” And Canada, “whose financial markets are dominated to a great degree by the vast New York market,” was, he suggested, already pretty much there.
It seems plausible, then, to guess that Canadian experience contributed to Mundell’s consistent early focus on the role of capital mobility, and factor mobility in general, in the international economy. This focus was already apparent in 1957 when he published “International trade and factor mobility”, a still widely read paper arguing that trade could substitute for factor movements and vice versa.
Canada’s openness to capital flows wasn’t its only distinctive feature. In a world of fixed though adjustable exchange rates, it stood out for having spent an extended period allowing the loonie — the Canadian dollar — to float freely, something it had to do if it was to have any monetary independence. Surely this distinctive Canadian experience helps explain why Mundell focused so early on macroeconomic policy under a floating-rate regime, which was academic speculation for most countries at the time but lived reality for his home nation.
And Canada’s decision to let the loonie float also offered a concrete example of the Impossible Trinity implied by his 1963 paper. A country can’t have free movement of capital, a fixed exchange rate, and effective monetary policy – it must choose two out of the three.
There was also one more thing about Mundell’s home nation that was special in the 1960s and that remains special today — the country’s unusual economic geography. Although Canada’s land area is huge, its climate ensures that the vast bulk of its population lives in a fairly narrow but very long strip just north of the US border; Vancouver and Toronto are 2,000 miles apart. Canada is in effect closer to the US than it is to itself.
Canadian geography clearly influenced Mundell’s vision in the 1961 paper that rivals his stabilisation paper in influence, “A theory of optimum currency areas”. He worried that a flexible exchange rate wouldn’t do much for Canada, both because the economic bases of the country’s east and west were so different and because, he argued, they didn’t constitute a single labour market. This led naturally to the idea that factor mobility is a key determinant of whether or not nations should have their own currencies and/or allow their currencies to float.
So, Mundell in effect used Canadian experience to motivate questions about how open-economy macroeconomics would work in a world where markets were being freed up. What did we learn from his answers?
The 1963 paper on stabilisation policy laid out Mundell’s contribution to what is generally referred to as the Mundell-Fleming model. Boughton (2002) has argued that it really should be called the Fleming-Mundell model — Fleming’s similar 1962 paper came first — and that much of the analysis can be found in earlier work by James Meade. I don’t want to dismiss that debate as meaningless. However, Mundell’s version ended up being the one most people cited, because it offered a dramatic policy ‘hook’: it said that the effects of both monetary and fiscal policy depend crucially on the exchange rate regime.
Specifically, Mundell’s analysis said that monetary policy can’t be used at all under fixed rates, while it becomes super-effective under floating rates, operating not through traditional channels but via the exchange rate. Fiscal policy, by contrast, is effective under fixed rates, but under floating rates fiscal expansion crowds out net exports, offsetting any stimulative effect.
How have these results held up in practice? The combination of high capital mobility and fixed rates clearly does eliminate any role for domestic monetary policy. Under Europe’s Exchange Rate Mechanism, everyone simply took it for granted that the Bundesbank was setting monetary policy for the entire European Monetary System.
The supercharging of monetary policy under floating rates is also a more or less established fact. For example, the tight money imposed by Thatcher to control British inflation was transmitted to the UK economy largely via a huge surge in the value of the pound, which slammed manufacturing competitiveness. However, monetary policy still seems to affect interest rates even under floating rates.
The fiscal side is a lot less clear-cut, in part because the underlying assumption of the Mundell-Fleming model — that the money supply is held constant under floating rates — doesn’t describe actual policy and never really has. Instead, central banks target interest rates, which makes the effect of fiscal policy on rates more about policy response functions than a mechanical consequence of changes in money demand.
But how is it that central banks can choose the interest rate? In Mundell’s original model, perfect capital mobility was supposed to ensure equality of interest rates across currencies, even under floating rates. This clearly isn’t true; just since the 2008 financial crisis we’ve seen the ECB raise rates in 2011 when the Fed didn’t, and then the reverse story from 2015 to 2019, when the Fed raised rates while the ECB stayed on the floor.
The answer is that even with perfect capital mobility, interest rates can differ if the exchange rate is expected to change — a point driven home by Mundell’s student Rudiger Dornbusch in his celebrated 1976 paper, “Expectations and exchange rate dynamics”, which was built on a Mundell-Fleming foundation. In general, allowing for expectations, especially if investors expect exchange rates to return toward some normal level, softens the Mundell-Fleming results: monetary policy works through domestic channels as well as the exchange rate, fiscal policy only produces partial crowding out through net exports. But it was the hard-edged nature of Mundell’s initial analysis that pushed international economists to consider the implications of exchange rate expectations.
So, Mundell wasn’t the last word on open-economy macro, but he pushed the field toward much sharper, more realistic assessments of how macro works under floating rates.
Mundell and Dornbusch, I’d argue, also had a sort of meta effect on open-economy macroeconomics, because they showed that an IS-LM-type model with slow price adjustment could lead to interesting, even exciting conclusions. This had the effect of keeping international economics relatively Keynesian even during the high tide of equilibrium macro — or, as I’d put it, keeping the field relatively sane.
There is, of course, some irony here, given Mundell’s eventual adoption as the patron saint of supply-side economics. But I’ll get there.
Like his work on stabilisation policies and the exchange rate regime, Mundell’s paper on optimum currency areas presented a dramatic, hard-edged analysis that stimulated a great deal of further analysis by others. This further analysis introduced considerations beyond Mundell’s original premise, and also somewhat softened his original point. But Mundell provided the essential starting point.
One way to think about Mundell’s original formulation was that he in effect asked: what makes a country a meaningful economic unit? For only if a country really is such a unit can it benefit from exchange rate flexibility.
Mundell’s answer was actually prefigured in earlier work; much of what he said can be found, once you know what you’re looking for, in Milton Friedman’s 1953 essay making the case for flexible exchange rates. But Mundell introduced most economists to the notion that high factor mobility – in particular, enough labour mobility that a nation could be regarded as having a single, integrated labour market – was the key. He suggested that Canada didn’t meet this criterion.
Subsequent work pointed to other criteria. In 1963 Ron McKinnon argued that an optimum currency area needed to be big enough that it wasn’t too open, that is, had a high share of its economy devoted to production of nontraded goods and services. A few years later Peter Kenen argued for the importance of fiscal integration, which could offset shocks to particular regions within a currency union (Kenen 1969).
The response to the euro crisis added to the list, highlighting the importance of banking union and, in an instant classic by Paul De Grauwe, the willingness of the central bank to act as lender of last resort (De Grauwe 2011).
As the list of conditions for currency union expanded, the discussion also moved from black and white to shades of grey. Early contributions seemed to suggest fairly simple yes-or-no criteria: you should form a currency union if and only if you can use this one weird trick. Like most arguments in economics, this eventually became an analysis of trade-offs, the benefits of a single currency against its costs, with the various criteria for an optimum currency area understood as factors that mitigated the cost of asymmetric shocks to regions within the area.
How well did this approach hold up over time? As I argued in 2012, the march to the euro offered a kind of test of optimum currency area theory. Europe falls well short by most criteria for a currency area: labour mobility is limited, fiscal integration trivial, banking union still an unfulfilled dream. This stands in stark contrast to the US, which has managed a continent-wide currency area but both has high labour mobility and provides a de facto safety net for regional economies suffering idiosyncratic slumps.
Sure enough, the euro area suffered severe difficulties from 2010 to around 2015. Euro advocates will insist that the advantages of the single currency outweigh these problems — and the mechanics of euro exit are so difficult that even nations under severe stress remained in the union. Still, the usefulness of the optimum currency area framework seemed to have been confirmed.
But a funny thing happened on the way to the euro: even as many economists invoked optimum currency area theory as a reason to be eurosceptical, Mundell himself was an enthusiastic advocate, so much so that he was called (with dubious justification) the ‘father of the euro’. Furthermore, supply-sider anti-Keynesians declared the economist whose early work helped keep international macroeconomics fairly Keynesian their intellectual founder.
How can we understand this part of Mundell’s intellectual journey?
It’s a lot harder to track Mundell’s later intellectual evolution than his earlier work, because he stopped leaving as much of a paper trail. Much of what the world believes about Mundell’s later thinking comes not from his own writing but from others’ accounts of his views, especially the journalist Jude Wanniski’s 1975 Public Interest article “The Mundell-Laffer hypothesis — a new view of the world economy”.
Still, Mundell didn’t dispute these accounts, and some of what did write — in particular, his Princeton International Finance Section essay “The dollar and the policy mix: 1971” — both confirms the general thrust of these articles and seems to indicate where they were coming from. And I think we can more or less reconstruct how the man who brought Keynesian analysis to the open economy and highlighted the difficult trade-offs in creating a currency area became the father of supply-side economics and the euro.
Here’s how I believe it went. At some point in the 1960s, Mundell came to believe that currency depreciation was ineffective and useless, because the law of one price always prevails. In Wanniski’s account, he believed that “devaluation has no real effects, but results only in price inflation in the devaluing country”. Given this belief, the optimum currency area is the whole world, and at any rate as much of it as you can get to adopt a single currency; hence the euro.
In the interim, however, Mundell was concerned about the international role of the dollar, which he considered endangered by US payments deficits. In a 1962 paper, “The appropriate use of monetary and fiscal policy for internal and external stability”, he had argued that a country facing external deficits and unemployment, which was the situation of the US at the beginning of the 1970s, should combine tight monetary policy with expansionary fiscal policy. So, he prescribed that mix for America.
As far as that analysis went, it suggested that expansionary fiscal policy could take the form of either higher spending or tax cuts. But over the course of the 1960s, Mundell appears to have grown increasingly convinced that taxes — especially rising marginal rates as inflation pushed people into higher tax brackets — were retarding economic growth. So, he became an advocate, more specifically, of a mix of tight money and tax cuts. Hence his putative role as father of supply-side economics.
If this depiction of Mundell’s views seems to lack the pristine clarity of his seminal work, that’s because, as I noted at the beginning, his intellectual style had changed. It’s not easy to summarise the argument of his 1971 paper on the dollar, which seems to roam among speculations about US global leadership, a restatement of Milton Friedman’s natural rate hypothesis, and a half-dozen other topics. If you like it, you could call it protean; if you don’t, woolly-minded.
In any case, it wasn’t the sort of thing that would made its way into graduate economics syllabi. As I noted earlier, while Mundell turned his back on sticky-price, more or less Keynesian macroeconomics, his student Rudi Dornbusch helped keep sticky prices at the heart of open-economy macro with his overshooting paper.
And Mundell’s assertion — or rather Wanniski’s assertion that he asserted — that depreciation has no real effects was implicitly denied by Dornbusch and explicitly rejected by another of his students, Michael Mussa, in his 1986 paper “Nominal exchange rate regimes and the behavior of real exchange rates: Evidence and implications”. Mussa showed that nominal exchange rate changes, far from being offset by price movements, seemed to cause one-for-one movements in real exchange rates, and also that the volatility of real exchange rates moves dramatically with the exchange rate regime, which seems to rule out reverse causation. As Mussa argued, the behaviour of real exchange rates offers prima facie evidence for sluggish price adjustment.
The stylistic break between early and later Mundell, and the apparent empirical failure of his later assertions, accounts for his peculiar role in economic discourse. Mundell’s early work remains a key building block for research and policy modelling, and is routinely cited after half a century. His later views drew acclaim from politicians and policy entrepreneurs, but have basically been ignored by professional economists. Economics, it turns out, follows the model, not the man.
But those models really were both pathbreaking and breathtaking. International macroeconomics is largely a house that Mundell built, even if he himself eventually chose to move out and seek lodging elsewhere.
Boughton, J (2002), “On the origins of the Fleming-Mundell model,” IMF Working Paper 02/107.
De Grauwe, P (2011), “Managing a Fragile Eurozone” VoxEU.org, May.
Dornbusch, R (1976), "Expectations and Exchange Rate Dynamics," Journal of Political Economy 84: 1161-76.
Fleming, M (1962), "Domestic financial policies under fixed and floating exchange rates", IMF Staff Papers 9: 369–379
Friedman, M (1953), The Case for Flexible Exchange Rates", in Essays in Positive Economics, University of Chicago Press.
Kenen, P (1969) “The Theory of Optimum Currency Areas: An Eclectic View”, in R Mundell and A Swoboda (eds), Monetary Problems of the International Economy, University of Chicago Press.
Krugman, P (1999), “O Canada,” Slate.
Krugman, P (2012), “Revenge of the optimum currency area”, NBER Macroeconomics Annual Vol. 27.
McKinnon, R I (1963), “Optimum currency areas”, The American Economic Review 53(4): 717-725.
Mundell, R (1957), “International trade and factor mobility,” American Economic Review 47(3): 321-335.
Mundell, R (1961), “A Theory of Optimum Currency Areas”, American Economic Review 51(4): 657–65.
Mundell, R (1962), “The appropriate use of monetary and fiscal policy for internal and external stability,” IMF Staff Papers 9: 70-79.
Mundell, R (1963), "Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange Rates," Canadian Journal of Economics 29: 475-485.
Mundell, R (1971), “The dollar and the policy mix,” Essays in International Finance, no.85.
Mundell, R (2000), “A reconsideration of the Twentieth Century,” American Economic Review 90(3) 327-340.
Mussa, M (1986), "Nominal exchange rate regimes and the behavior of real exchange rates: Evidence and implications," Carnegie Rochester Series on Public Policy 25: 117-214.
Wanniski, J (1975), “The Mundell-Laffer hypothesis, a new view of the world economy,” The Public Interest No. 47.