Comment

Previous articleNext article FreeCommentMichael P. DooleyMichael P. DooleyUniversity of California, Santa Cruz, and NBER Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreThe paper by Curcuru, Thomas, and Dvorak is an important contribution to the debate about the sustainability of recent U.S. current account deficits. If estimates of the U.S. net international investment position (NIIP) published by the Bureau of Economic Analysis are correct, U.S. net international debt at the end of 2007 was about $2.4 trillion and was growing at a decreasing rate (light solid line in Curcuru et al.’s fig. 2). NIIP estimated by cumulating the same agency’s estimates of the current account deficit stood at about $6 trillion at the end of 2007 and was growing at an increasing rate (broken line in fig. 2).The conclusion that the U.S. current account is unsustainable is usually based on projections for U.S. current account deficits. If recent history is repeated, however, the level of NIIP will continue to grow more slowly. If the factors that slowed NIIP growth are reversed, the outlook would be much more alarming.If lower levels and growth rates for reported NIIP have been generated by large persistent yield differentials in favor of U.S. investors, U.S. NIIP will continue to grow relatively slowly. Curcuru, Dvorak, and Warnock (2008) presented evidence that estimates for much higher returns for U.S. investors within asset classes are based on an inappropriate merging of data sets. In particular, they argue convincingly that revisions in reported stocks of U.S. international assets and liabilities are not entirely due to realized returns. Nevertheless, in that paper and the current paper there remains a small yield differential in favor of the United States, and “valuation effects” calculated by the authors account for about a $2 trillion lower estimate of the NIIP in 2007.A problem associated with Curcuru et al.’s conclusion is that it leaves about $1.7 trillion of reported balance of payments data without a home. The authors’ rejection of the hypothesis that the $1.7 trillion belongs in differential returns would be strengthened by finding plausible alternative homes. In general, their task is to reduce estimates of 2007 NIIP from reported positions data or to increase estimates of the NIIP calculated from revised balance of payments data.The results of their search are nicely summarized in table 6. The original reconciliation between balance of payments and position data leaves $1.7 trillion homeless. Scenario A incorporates revisions to both data sets and leaves only $0.36 trillion homeless.The largest revision is a $1.4 trillion reduction in balance of payments net financial capital inflows to the United States from 1989 to 2007. The largest component is in bonds, where their analysis suggests that inflows were overreported and outflows underreported in the balance of payments data. Note that if this were the only adjustment, the implied cumulated statistical discrepancy in a revised balance of payments would increase from about zero to $1.4 trillion. This would be unfortunate because it moves the homeless balance of payments data into another shelter, the statistical discrepancy, without solving the problem. A skeptic might move the statistical discrepancy back into net capital inflows or, as Milton Freidman suggested, into investment income.But while less capital came to the United States in scenario A, less is needed because the authors estimate that exports were underreported over the interval by about $0.5 trillion. Evidence for this comes from trade data from Canada and Mexico, where their reported imports from the United States exceed U.S. recorded exports. In general, it is plausible that imports that are subject to tariffs are better recorded than exports. Other missing net credits include net sales of residential real estate and net sales of financial derivatives. Combining these revisions, Curcuru et al. estimate that financial inflows were about $1 trillion smaller than the original balance of payments accounts suggest but that higher merchandise exports limit the increase in the statistical discrepancy to about $0.5 trillion.Finally, a number of adjustments to the reported NIIP in 2007 suggest that the level of U.S. debt is about $0.5 trillion larger than reported. The bulk of this is residential real estate owned by nonresidents but not included in the reported NIIP.One way to summarize the authors’ results is that the “correct” number for the level of NIIP at the end of 2007 was $3 trillion, not $2.4 trillion, and it is growing at about a constant rate. The implications for sustainability, however, are as vague as the concept of sustainability. It still looks to me that the reconciliation between the current account and the NIIP provides tremendous latitude for slowing the rate of growth of U.S. debt relative to a naïve compounding of current account deficits. But it is also plausible to increase the rate of growth of debt relative to cumulated current accounts as valuation effects are reversed, for example, as the dollar appreciates.This is not the first time that measurement problems have proved crucial in judgments concerning the sustainability of international debts. In Dooley et al. (1986), we were interested in the sustainability of emerging market countries’ debt and faced similar inconsistencies between data on stocks and flows of international financial transactions. In that case, data for stocks of gross debt compiled by the World Bank were clearly more reliable and much larger than the implied level of debt derived from cumulating debtor countries’ balance of payments data. Our conjecture was that the residents of debtor countries were probably buying foreign financial assets and that the implied statistical discrepancy was motivated by capital flight from debtor countries. Our conclusion was that political risk was unlikely to abate and that private capital flight would continue to contribute to sovereign governments’ debts that were not sustainable.In the U.S. case, Curcuru et al. have effectively questioned the optimistic view that U.S. debt will continue to grow slowly because of an exorbitant privilege in rates of return. They show that alternative holes in the data can account for the surprisingly slow growth in published data for the U.S. NIIP. The implication for growth rates for NIIP looking forward, however, is not clear. If exports continue to be underreported and if the revised statistical discrepancy results from underreporting investment earnings, NIIP will continue to grow slowly relative to the recorded current account deficit. If the discrepancy results from underreporting increases in net financial liabilities, NIIP will grow much more rapidly. Henry Wallich used to say that if we do not know how many dogs and cats live on the dark side of the moon, we should assume it is 50‐50. This prior would generate a slight deceleration in the growth of NIIP as U.S. trade adjusts to recent changes in dollar exchange rates.ReferencesCurcuru, Stephanie E., Tomas Dvorak, and Francis E. Warnock. 2008. “Cross‐Border Returns Differentials.” Quarterly Journal of Economics 123 (November): 1495–1530.First citation in articleGoogle ScholarDooley Michael, William Helkie, Ralph Tryon, and John Underwood. 1986. “An Analysis of External Debt Positions of Eight Developing Countries through 1990.” Journal of Development Economics 21, no. 2:283–318.First citation in articleGoogle Scholar Previous articleNext article DetailsFiguresReferencesCited by Volume 5, Number 12009 Article DOIhttps://doi.org/10.1086/595999 Views: 98 © 2009 by the National Bureau of Economic Research. All rights reserved.PDF download Crossref reports no articles citing this article.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.