09 November 2016

Strained relations

US foreign exchange operations and monetary policy in the twentieth century

Michael D Bordo, Owen F Humpage and Anna J Schwartz, National Bureau of Economic
2015, University of Chicago Press, 496 pages,
ISBN 9780226051482

Reviewer: William A Allen, Economic & Financial Consultant

I begin this review by declaring a personal interest. I was the head of the Bank of England foreign exchange division from 1990-94, and in that capacity responsible for conducting extensive foreign exchange operations. So I know something about the subject, but I may have unconscious prejudices.

The book under review was begun as long ago as 1990, and one of the distinguished authors, Anna Schwartz, died in 2012. Its main conclusion is that foreign exchange market intervention adds nothing of value to the armoury of monetary policy instruments, and may detract from the effectiveness of monetary policy. It is hard to disagree that much foreign exchange intervention in the most actively traded currencies from, say, the end of Bretton Woods up to the financial crisis, and probably beyond, did more harm than good, and it is hard to think of many important things that it achieved. Intervention in currencies with floating exchange rates generally lacked a clear and necessary purpose, while intervention to defend fixed-but-adjustable rates all too often failed, at great expense. The evidence that the authors supply, from the period 1973-95, generally supports their conclusion. What troubles me about the book is that it often appears to assert its conclusions, which are ostensibly about the foreign exchange operations of the United States, as universal truths, rather than conclusions whose truth depends on the surrounding circumstances.

For example, chapter 1, reviewing the theoretical underpinnings of intervention, says that, “Most empirical studies find the relevant elasticities to be either statistically insignificant or quantitatively negligible,” (p 9), leading to the conclusion that the portfolio balance effect of intervention is insignificant. Over what period? Were the elasticities negligible in the 1930s, when the U.S. began foreign exchange intervention? And in what countries? The paper by Edison, written in 1993, which the authors cite in support of their statement does not look back as far as the 1930s, nor does it look beyond currencies with liquid markets. And, as early as pages 9-10, the authors dismiss the market microstructure approach to exchange rate determination in a single paragraph on no better grounds than that, “It is at odds with the macroliterature.” Nor is there any discussion of ‘excess volatility’ – that is, the observation that post-Bretton Woods floating exchange rates have been more volatile than earlier advocates of floating expected, and the theory that they have been more volatile than was justified by economic fundamentals. Perhaps the authors do not believe in excess volatility, but it would have been nice to know what they think about it, and about whether it could provide a justification for official intervention, particularly as one of the most egregious cases, the vertiginous ascent of the dollar in the early 1980s and its subsequent decline, falls well within the period to which they devote most attention.

In an epilogue, the authors, for the first time, qualify their conclusions about the ineffectiveness of intervention as a weapon of policy separate from domestic monetary

policy. The epilogue contains a four-page section on foreign exchange operations in developing and emerging market economies, which admits that the portfolio balance, expectations and market microstructure effects may make intervention effective in some countries (p. 358). The section is however full of sweeping generalisations about emerging markets, and indeed fails to describe the term ‘emerging market’. It points out, correctly, that intervention has the power to undermine commercial market-makers, but provides no evidence on the point.  The epilogue also contains four pages on Switzerland, in which it is not made as clear as it might be that there is now really no distinction between domestic monetary policy and foreign exchange policy in Switzerland: nearly all of the assets of the central bank are gold and foreign exchange.

The authors betray their own doubts about the universality of their conclusions in the last paragraph of the whole book, which reads as follows:

We have based these conclusions on an analysis of US exchange market policies during the twentieth century and, as noted, primarily from the perspective of the Federal Reserve System. While we offer them as lessons on intervention in general, we understand that the US experience may be unique in some undetermined manner. In that regard, historical studies of other countries’ experiences with foreign-exchange intervention should be a profitable venue for further research. (p. 343).

My interpretation of the last two sentences is that the authors are not convinced that their conclusions apply beyond those countries with broad and liquid foreign exchange markets. If that interpretation is correct, I share their doubt. I also doubt whether their conclusions apply to the United States, the United Kingdom and France in the 1930s, for the reasons given above. The book would have been more persuasive had it been cast as an examination of the conditions in which foreign exchange intervention can be useful. As it is, the book begins by setting out a rather narrow set of possible justifications for intervention, which are then largely dismissed a priori, followed right at the end of the book by a rather compressed list of exceptions and qualifications.

Indeed, for the most part, the authors appear to think and write within a framework constrained by a particular strand of the macroliterature. Perhaps all historians are the captives of some theory or other, but these ones seem unusually tightly confined. Thus, in chapter 2, which is about foreign exchange market policy in the United States before 1934, when the Exchange Stabilisation Fund was set up, they write that ‘Under the classical gold standard, disturbances to the balance of payments were automatically equilibrated by the Humean price specie flow mechanism’ (p 28). In one sense, this statement is innocuous: balance of payments disturbances gave rise to gold flows, other things being equal, and the gold flows must have promoted equilibriating adjustments.  However, it is a statement about theory, not about history: histories of the operation of the gold standard in the 19th century invariably emphasise the importance of capital flows.

Chapter 2 contains some curious and surprising factual errors and questionable judgments. The British Exchange Equalisation Account was not disbanded in 1979 (p 37); it lives on to this day. Bills of exchange were not first developed in England in the seventeenth century (p 38): they go back at least as far as the Middle Ages. The statement on p 50 that, after stabilizing the franc in 1926, France ‘converted foreign exchange (both sterling and dollars) into gold’ is not wholly true: the Bank of France sustained losses when Britain suspended the gold standard in 1931. The authors characterization of the speculative attack on sterling in 1931 as ‘ultimately reflecting a seriously deteriorating fiscal problem’ (p 52) is highly tendentious: the lack of liquidity of the Bank of England relative to the U.K.’s external short-term liabilities, and the overvaluation of the pound, are far more promising candidates. The authors go on to say, apparently seriously, that ‘Ultimately the gold exchange standard collapsed but debate still continues as to whether it would have survived absent the shock of the Great Depression’ (p. 52).

Chapter 3 is on the activities of the Exchange Stabilisation Fund between 1934 (when it was set up) and 1961, but mainly between 1934 and 1939.  On the evidence presented by the authors, and by Howson (1980), foreign exchange intervention in the 1930s after the collapse of the gold standard was quite successful. In particular, the Fed, like its British counterpart, sold pounds for dollars when they were expensive in 1935-36 and bought them back when they were cheaper in 1938-39.

The authors’ comment on the Tripartite Agreement of 1936 that ‘it is not obvious to us that the Americans and British would have been worse off had the French devalued the franc unilaterally, and there had been no Tripartite Agreement’ (p.  74). Well, maybe they wouldn’t have been any worse off, but the comment misses the point. A unilateral devaluation of the French franc was not on the menu, because after its pre-election promises, the Popular Front government felt unable to devalue the franc without an international agreement of some kind on exchange rates. The alternative to the Tripartite Agreement might, as Clarke (1977) shows and as the authors acknowledge, have been the imposition of German-style exchange controls by France; the avoidance of those was surely an important achievement. Again there are factual errors: the Belga was invented in 1926, not 1935 (p 86); it was a unit of five Belgian francs and its introduction was thus a kind of rebranding. The Belga was intended to be the unit of Belgian currency traded in foreign exchange markets; it was not a new currency.

The authors are on more secure factual ground after 1961, mainly because they can rely on the excellent records of the Federal Reserve. Perhaps in some respects, they rely too much on history according to the Fed. The problem of the outflow of gold from the United States, which preoccupied the United States and other countries from the early 1960s onwards, had a technical solution, foreseen at Bretton Woods, namely an increase in the official price of gold. The U.S. authorities did not wish to contemplate this solution, and went to great lengths, ultimately including the destruction of the Bretton Woods system, to avoid it. One reason for their aversion to it was that it would threaten the reserve-currency status of the dollar; another was that it would reward the two largest gold producers, South Africa and the Soviet Union, both of which were pariah countries at the time. However, fundamentally it was U.S. monetary policy that created the losses for holders of dollars and threatened the dollar’s reserve-currency status, and put upward pressure on the gold price which helped the gold producers.  The United States opposed a rise in the official price of gold for reasons of national interest and international politics; the argument for it deserves much more than the cursory treatment it gets on p 129.

And when the authors discuss the generosity of the Fed’s swap line assistance to Britain in 1963 (pp.169 – 170), they neglect to mention that the Bank of England delayed the increase in Bank rate which eventually took place in February 1964 for two months at the urgent request of the U.S. Administration, including President Johnson, in the interests of heading off pressure on the dollar. Co-operation worked both ways.

There is also a surprising but important analytical error in chapter 4 (p.121). The authors say that the U.S. official purchases of forward dollars raised the cost of speculation against the dollar. In fact, they did the opposite: by increasing the price of forward dollars compared to what it would otherwise have been, they enabled dollar bears to sell dollars at a higher price. The effectiveness, if any, of official purchases lay in taking some of the dollar bears out of the market and preventing them, at least temporarily, from depleting the U.S. gold stock.

The book contains some interesting facts, particularly about the 1930s, and provides a useful summary of U.S foreign exchange operations after 1961, but its analysis is disappointing, and it cannot be regarded as a detached and authoritative account of the subject.


Capie, F.H. (2010), The Bank of England, 1950s to 1979, Cambridge University Press

Clarke, S.V.O. (1977), Exchange rate stabilization in the 1930s: negotiating the tripartite agreement, Princeton Studies in International Finance no 41.

Edison, H. (1993), The effectiveness of central bank intervention: a survey of the literature after 1982, International Finance Section, Department of Economics, Princeton University, Special Papers in International Economics no 18.

Howson, S. (1980), Sterling’s managed float: the operations of the Exchange Equalisation Account 1932-39, Princeton Studies in International Finance no 46.

James, H. (2001), The end of globalization, Harvard University Press.

Usher, A.P. (1914), ‘The origin of the bill of exchange’, Journal of Political Economy, vol. 22 no. 6, pp 566-576.