Brad DeLong's Semi-Daily Journal: Global Imbalances in Historical Perspective

>[Brad DeLong's Semi-Daily Journal: Global Imbalances in Historical Perspective](http://delong.typepad.com/sdj/2006/10/global_imbalanc.html): http://papers.nber.org/papers/w12580.pdf LOSING OUR MARBLES IN THE NEW CENTURY? THE GREAT REBALANCING IN HISTORICAL PERSPECTIVE Christopher M. Meissner Alan M. Taylor Working Paper 12580 http://www.nber.org/papers/w12580 Abstract: Great attention is now being paid to global imbalances, the growing U.S. current account deficit financed by growing surpluses in the rest of the world. How can the issue be understood in a more historical perspective? We seek a meaningful comparison between the two eras of globalization: "then" (the period 1870 to 1913) and "now" (the period since the 1970s). We look at the two hegemons in each era: Britain then, and the United States now. And adducing historical data to match what we know from the contemporary record, we proceed in the tradition of New Comparative Economic History to see what lessons the past might have for the present. We consider two of the most controversial and pressing questions in the current debate. First, are current imbalances being sustained, at least in part, by return differentials? And if so, is this reassuring? Second, how will adjustment take place? Will it be a hard or soft landing? Pessimistically, we find no historical evidence that return differentials last forever, even for hegemons. Optimistically, we find that adjustments to imbalances in the past have generally been smooth, even under a regime as hard as the gold standard. 2 The unending feedback of the dollars and pounds received by the European countries to the overseas countries from which they had come reduced the international monetary system to a mere child’s game in which one party had agreed to return the loser’s stake after each game of marbles—Jacques Rueff, 19611 A remarkable amount of attention is now being paid to global imbalances, the growing U.S. current account deficit financed by growing surpluses in the rest of the world, most notably in the Asian “dollar bloc” and among the oil exporters. The talk is no longer confined to obscure academic and policy debates. With insufficient space in his weekly columns to devote to the issue, the Financial Times’ Martin Wolf launched a web site stating that “what is happening is extraordinary”; David Warsh considers the almost obsessive focus on the issue justified, since global imbalances constitute “the most exciting economic story of our times.”2 Exciting and extraordinary it may be, but a relentless focus on trends from the recent past, on the current announcements of each quarter’s balance of payments data, or on naive extrapolations into the future has left one important perspective rather neglected: how can the issue of global imbalances be understood in a more historical perspective? To address this question, we seek a meaningful comparison between past and present experience. We focus on the two eras of globalization: “then” (the period 1870 to 1913) and “now” (the period since the 1970s). We look at the special position in the global macroeconomy of the hegemons in each era: Britain then, and the United States now. And adducing historical data to match what we know from the contemporary record, we proceed in the tradition of New Comparative Economic History to see what lessons the past might have for the present. Although such an exercise in quantitative economic history could range far and wide, in this paper space limitations permit us only to look at what we consider two of the most controversial and pressing questions in the current debate. First, are current imbalances being sustained, at least in part, by return differentials? And if so, is this reassuring? If the U.S. can always earn some kind of “privilege” of this sort, then the degree of required adjustment will be reduced. Or, putting it another way, for any given trajectory of trade imbalances, we know that the current account and debt implications will look much more favorable or sustainable if such privileges persist. If not, any difficulties will be that much more pronounced. Second, how will adjustment take place? Will it be a hard or soft landing? It is possible, again, that adjustments will happen smoothly. Depending on the extent to which expenditure shifts rather than switches, countries might avoid dramatic real exchange rate movement. If up and down shifts are coordinated across countries, or if switching is unhindered by trade policies or other frictions, then global demand might hold up, and recession avoided. The fear is that adjustments might be much more abrupt, demand large changes in real exchange rates, lead to politically awkward realignments of trade, and cause recession for one or more players in the game. If such a hard landing is likely, then policymakers face the challenge of devising suitable countermeasures. 1 On the source of this quotation, see the comment by John Helliwell. 2 For Warsh on Wolf see http://www.economicprincipals.com/issues/05.04.10.html. For the Wolf forum see http://www.ftblogs.typepad.com/martin_wolf/. For up to the minute discussions see http://www.rgemonitor.com/blog/setser/. For a recent overview, see Eichengreen (2006). 3 Confronting these two questions, what insights can we take from the past? Summary To summarize our findings, on the persistence of privilege we find: o Among G7 countries today, the United States is not unique in being able to enjoy a “privilege” in the form of higher yields earned on external assets relative to yields paid on external liabilities. For the U.S. this has been worth about 0.5% of GDP to the U.S. in the years 1981 to 2003. Similar privileges are detectable for Japan and the U.K. France and Germany appear to have no privilege. Canada and Italy have negative privilege, or penalty. o In the years 1870–1913, the previous financial hegemon, Britain, enjoyed a similar yield privilege, also amounting to about 0.5% of GDP. o Measured as a differential in rates of yield, the U.S. privilege has been steadily declining since the 1960s, when it stood at around 3% per annum on all capital. It is now close to 1% per annum. Indirect measures may differ, and even the direct measures are subject to error. But if this trend continues, the U.S. will lose its privilege. o Direct and indirect evidence on rates of yield for Britain in the past also suggests small and declining rates of yield privilege from the 1870s to 1910s, a similar pattern. o For both the U.S. now and Britain then, declining rate of yield privilege meant that for a given leverage and a given composition of assets and liabilities, the income due to privilege (as a fraction of GDP) would have to shrink. In part this was offset either by expanding leverage (in the U.S case today) or by shifting composition to riskier assets with higher returns (in both cases). These shifts may not be able to proceed without limit. o It is often suggested that the U.S. might lose privilege if the net debt position grows too large. We find that rate of yield privilege has been correlated with the deterioration of the net external asset position in the postwar era. o In the historical British case, leverage and indebtedness were not an issue. British net external assets roughly equaled gross external assets, and Britain became a very large net creditor. But a net credit position did not preclude a loss of privilege, suggesting that even if the U.S. could reverse its net debt position, this would not protect its privilege automatically. o Rather, British experience suggests that over time, financial hegemons operating in a globalizing world face other pressures that squeeze privilege. Emerging markets mature and offer less outlandish risk-reward combinations, so the benefit of being a “loan shark” diminishes; the world becomes less risky as a whole; at the same time other rival financial centers emerge which can compete for lucrative business with the financial pioneer. o Most of these perspectives bode ill for the persistence of privilege. But if we add capital gains to yields we can estimate a total return privilege for the U.S. According to indirect estimates, total return privilege has risen since the 1960s. It also appears to have been steady in the 1980s and 1990s. Growing valuation effects have offset falling yield differentials, keeping up a total return privilege. It is unclear what mechanisms are driving these opposing trends. o Looking at indirect evidence on total returns on the U.K. domestic and foreign portfolio 1870–1913, we also find a total return differential, but one that is very volatile over successive decades, and with very little systematic privilege overall. o The large capital gains earned by the U.S. in the last 10 to 15 years are due to neither sustained price effects nor sustained exchange rate effects, both of which are close to zero on average; the effect is largely due to “other” capital gains. These remain a mystery, and until we understand them better, simple extrapolation of these trends may be ill advised. On adjustment we examine the behavior of current accounts and the processes associated with current account reversals for a broad sample of countries between 1880 and 1913. We attempt to verify whether there are any differences between the capital exporters like Britain, France, Germany, and the Netherlands, other core countries that import capital, areas that had recently been settled, also known as British offshoots (i.e., Australia, Canada, New Zealand and the United States), and less developed peripheral nations. Throughout we compare our findings to those from Edwards (2004) from the thirty years between 1970 and 2001. In particular we look at: summary statistics regarding the size of current accounts and incidence of reversals; the ability to sustain current account deficits or surpluses; connections between current account reversals, exchange rate movements and financial crises; patterns of movement of macroeconomic aggregates in the wake of large current account reversals including the growth effects of reversals. o We find that more developed countries and the offshoots were able to run higher current account deficits more persistently, and that these countries had very different patterns of adjustment. o In particular their current account reversals were generally associated with smaller real exchange rate fluctuations and less adjustment in the government surplus. o Overall, we do not find overwhelming evidence that current account reversals had negative consequences for the aggregate growth of income per capita in the core or periphery. (Although many reversals involved serious crises that surely did have major distributional impacts.) o Moreover, we are able to test some modern hypotheses with the historical data in ways that have not previously been done. We assess whether openness to international trade, financial and institutional development and currency mismatches played a role in adjustment. o We find little evidence that currency mismatches, openness to international trade or the level of institutional and financial sophistication (proxied very roughly by higher income per capita) altered the severity of output losses associated with reversals in the nineteenth century. o Nevertheless we do find some evidence that core Western European countries and the offshoots had lower growth losses in the adjustment process. Some countries even managed to see income rising in the face of reversals because previous investment was so productive. This offsets the negative growth experiences of other countries in the periphery leading to the finding that current account reversals were not systematically associated with output losses in this period.