January 12, 2009
International Imbalances: Measurement and Implications
Paul Kedrosky has observed that a statistical analysis (word cloud) of the American Economic Association session titles, or even of the papers, leads to the impression that the economics profession has been relatively uninterested in the ongoing financial and economic crisis. Unfortunately, this observation misses ignores the fact that session proposals are submitted a full eleven months ahead of the ASSA meetings. Think back to January 2008, and the terms ascribed to those who warned of a severe slowdown ("alarmist", etc.), and the whole discussion is cast in a different light.
Furthermore, the correlation between scheduled paper titles and actual is, shall we say, fairly loose. In any case, the main messages of the papers are usually discussed in the context of current events.
I'll illustrate this last point by discussing the only complete session I made it to (most of my time was spend interviewing job candidates): "The Capital Flows behind Financial Globalization" (organized by Kristin Forbes). I had to make it since I was presiding. As it turns out, it was a very lively session, and -- despite the title -- quite relevant to ongoing events. (And if you don't think this was timely, take a look at Brad Setser's post today.)
Figure 1 from Kristin Forbes (2008).
First on the agenda was my former colleague Joshua Aizenman. He actually presented results from two papers (presentation here). In "Globalization and the Sustainability of Large Current Account Imbalances: Size Matters," [pdf] Aizenman and Yi Sun conjecture how China's current account will evolve over time, in a context where China is a large player. From the abstract:
This paper evaluates the sustainability of large current account imbalances in the era when the Chinese GDP growth rate and current account/GDP exceed 10%. We investigate the size distribution and the durability of current account deficits during 1966-2005, and report the results of a simulation that relies on the adding-up property of global current account balances. Excluding the US, we find that size does matter: the length of current account deficit spells is negatively related to the relative size of the countries' GDP. We conclude that the continuation of the fast growth rate of China, while maintaining its large current account/GPD surpluses, would be constrained by the limited sustainability of the larger current account deficits/GDP of courtiers that grow at a much slower rate. Consequently, short of the emergence of a new "demander of last resort," the Chinese growth path would be challenged by its own success.
In a second paper, Joshua presented results indicating that the United States served as the "demander of last resort", wherein US current account deficits appeared to be a key determinant of LDC current account surpluses, even after accounting for other factors (demographics, etc.). This paper is available on Aizenman's website here. In that paper, Aizenman and Yothin Jinjarek write:
We identify an asymmetric effect of the US as the “demander of last resort:” a 1% increase in the lagged US imports/GDP is associated with 0.3% increase of current account surpluses of countries running surpluses, but with insignificant changes of current account deficits of countries running deficits.
The discussant, Jeffrey Frankel, lauded the paper for its eschewing complicated intertemporally optimizing models of the current account balance, in favor of an approach that could more useful be empircally implemeted. He also observed that perhaps the current crisis dealt a serious blow to the proposition that the reason why capital had flowed to the US was our "deep and sophisticated capital markets" or the credibility of our financial institutions, but something more simple such as profligacy. (Obviously, this is a view I am very sympathetic to , , .) On a more substantive note, he observes that capital account openness should enter into the authors' specification interactively rather than as a levels effect -- that is capital account openness should determine the easy of financial flows, rather than their sign. In the end, he agreed that these papers were useful in thinking about how the imbalances arose, and how they may now be in the process of abrupt shutoff.
In "Why Do Foreigners Invest in the United States?" [pdf], Kristin Forbes writes:
Why are foreigners willing to invest almost $2 trillion per year in the United States? The answer affects if the existing pattern of global imbalances can persist and if the United States can continue to finance its current account deficit without a major change in asset prices and returns. This paper tests various hypotheses and finds that standard portfolio allocation models and diversification motives are poor predictors of foreign holdings of U.S. liabilities. Instead, foreigners hold greater shares of their investment portfolios in the United States if they have less developed financial markets. The magnitude of this effect decreases with income per capita. Countries with fewer capital controls and greater trade with the United States also invest more in U.S. equity and bond markets, and there is no evidence that foreigners invest in the United States based on diversification motives. The empirical results showing a primary role of financial market development in driving foreign purchases of U.S. portfolio liabilities supports recent theoretical work on global imbalances.
I'll observe that this paper represents an impressive compilation of disparate datasets, so people interested in a thoughtful analysis of the data (as opposed to hyperbole and anecdotes) should consult this paper.
The discussant, Linda Tesar, observed that while the paper did not provide any big surprises, it was interesting to see how the trends evolved over time. In particular, the Gourinchas and Rey (2007) finding of higher returns on US holdings of foreign assets than foreignors earn on their holdings of US assets held true even after the adjustments of Curcuru, Dvorak and Warnock (2008).
More importantly, Tesar took issue with the view that if foreign returns were high (even after risk adjustment) and US returns low, it was a puzzle why foreigners invested in the US. But portfolio theory suggest that just because returns are low, the optimal level of investment in the US is not zero, as long as markets are not perfectly correlated. The question is how much investment, not whether to invest in the US.
The last two papers were closely related in terms of subject matter: measurement.
Pierre Olivier Gourinchas and Helene Rey produced a new version of their tabulation of the US external accounts, in order to generate new estimates of foreigners' returns on US assets versus Americans' returns on foreign assets. Their methodology
was basically a more elaborate version of related to the approach laid out in this paper, discussed briefly in this post "From World Banker to World Venture Capitalist:
US External Adjustment and The Exorbitant Privilege". Gourinchas describes it as an examination of "various assumptions (ours, [Warnock et al.'s, Milesi-Ferretti and Lane's]) and describe the implications of these different assumptions for the overall consistency of the balance of payments accounts. In particular, we explore precisely the implications of stuffing all the mismeasured stuff in the flows (as Curcuru et al do) or in initial positions (as Lane and Milesi-Ferretti suggest) and find that it yields unrealistic estimates of either the trade flows throughout the period, or net investment positions early in the sample, or both. This is our metric for judging whether some assumptions are plausible or not. Because, at the end of the day, things still need to add up." (edits added 1/14/09)
The discussant, Frank Warnock, wondered whether it made sense for academics to be constructing their own data. In particular, Gourinchas and Rey combine flow and stock data to impute stocks, even though the data were never constructed with this objective in mind (i.e., the data are not compatible). Instead, he argued that maybe it made more sense to utilize the estimates that statisticians in the Federal government were developing, especially to cover holes in the current statistics-gathering system. Some of Warnock's results in this context were discussed in this post, and can be found in his papers found here.
The deterioration in the U.S. net external position in recent years has been much smaller than the extensive net borrowing associated with large current account deficits would have suggested. This paper examines the sources of discrepancies between net borrowing and accumulation of net liabilities for the U.S. economy over the past 25 years. In particular, it highlights and quantifies the role played by net capital gains on the U.S. external portfolio and 'residual adjustments' in explaining this discrepancy. It discusses whether these 'residual adjustments' are likely to be originating from measurement errors in external assets and liabilities, financial flows, or capital gains, and explores the implications of these conjectures for the U.S. financial account and external position
While this paper seems unrelated to the ongoing crisis, during the rejoinder session, Milesi-Ferretti noted that -- using the framework outlined in the paper, and incorporating the fact that foreign equity markets had performed even worse than the US -- it was likely that the US net foreign asset position deteriorated substantially in 2008.
Kenneth Rogoff lauded the attempts by the authors to try to measure the US, and other country, net foreign assets. But he spent most of his time talking about -- how net foreign assets related to the ongoing crisis, relying much on his presentation to the joint AEA/AFA luncheon.
So, returning to the departure point for this post, don't be misled -- discussion of the ongoing crisis (and how global imbalances might have contributed) was everywhere at the AEA's, even if not explicitly in the session and paper titles.
Technorati Tags: American Economic Association, capital+flows, financial globalization, net foreign assets, current account imbalances, net external position, foreign direct investment, and foreign returns.
Posted by Menzie Chinn at January 12, 2009 03:10 PMdigg this | reddit
Regarding Figure 1, can you say "mercantile trade policy"?
Evidently Kristin Forbes can't! It is the most obvious explanation.
But that's just hyperbole and anecdote.
Posted by: Buzzcut at January 13, 2009 07:28 AM
You've missed the point of Kedrosky's (correct) criticism of economists: They have not looked at themselves in the mirror and tried to figure out why they did not anticipate our current global financial/crisis.
Just noting post-event that things are different--and worse--now than we previously anticipated is not the same thing by a long shot. Worse yet, it suggests economists will make the same mistakes in the future.
And, finally, on the proposal submission date a year ago, virtually none of these papers was submitted before December 2008 & they had plenty of time to be self-critical. Yet, one session's title is narrowly focused on the "subprime" mortgage problem. How out of touch with reality can one be?
Posted by: Terry at January 13, 2009 07:47 AM
Here, here Terry! I don't know who this Kedrosky guy is but he expressed something I've thought for quite a while (see comments on another thread re economists and mental illness). One of the ex-Mrs. Footwedges was in theater and had a great quip about theater critics; they are like "eunuchs at a gang bang." I wonder if that couldn't apply to a lot of experts . . .
Posted by: Footwedge at January 13, 2009 11:17 AM
Joshua Aizenman merely states the obvious. But the U.S. deficits are partly caused by currency mercantilism and this will stop (my guess) not because lenders fear for their assets, but because U.S. labor objects.
Kirsten's question may be backward. Rather than ask why foreigners are willing to lend to the U.S., ask why U.S. labor allows foreign currency mercantilism.
A researcher using data from Figure 1 may get gibberish for results. Much of the U.K debt holdings is really owned by other countries. And much of the Irish, Belgian and Cayman Island debt is owned by other countries or by U.S. muiltinationals.
Posted by: don at January 13, 2009 12:05 PM
The comment of Milesi-Ferretti:
"While this paper seems unrelated to the ongoing crisis, during the rejoinder session, Milesi-Ferretti noted that -- using the framework outlined in the paper, and incorporating the fact that foreign equity markets had performed even worse than the US -- it was likely that the US net foreign asset position deteriorated substantially in 2008"
Not so. The net foreign asset position of US improved dramatically on 2008... to the tune of 2 trillion dollars probably. All related to the defaults of CDOs and related debt. The US has been staging or going through peridic defaults of foreign debt (and stock values) since 1986. This last one, and the biggest by far, being the fourth one on the list. That is why the accumulated foreign debt does nor increase as much as current accounts deficits should inply.
Posted by: Talin at January 14, 2009 07:45 AM
With all due respect, eleven months is a short enough time horizon to gather a group of relevant-sounding papers.
You might laugh and argue, "no one could have predicted...". But that would be wrong. Maybe no one could have predicted the answers, but surely they could have predicted the questions. Academics have a wide choice of hypotheses to explore. It appears, as a group, they did not choose the right set of hypotheses.
Finally, if economists had insufficient idea of what questions to ask eleven months ago, then they probably are not asking the right questions today.
Posted by: David Peason at January 14, 2009 09:36 AM
Posted by: Terry at January 14, 2009 02:15 PM
"Think back to January 2008, and the terms ascribed to those who warned of a severe slowdown ("alarmist", etc.)"
This is a clue about one of the reasons that the college level economics profession has become a lagging indicator about the economy. Only one academician, Roubini, got it right. What an indictment of academic economics.
A small number (maybe OOM 10,000) of investment professionals got it right. But even though that is a small number, its much larger than the results from many professors at many colleges. That says something about the effectiveness of a profit incentive.
If academic economics wants to be taken seriously as a forward looking group (versus the current explain the past, theorize, and publish orientation) some very large changes are needed. These changes can only come from the Dept. Chair at many colleges. Otherwise, academic economics will continue to look through a rear view mirror.
Posted by: Mike Laird at January 14, 2009 02:43 PM
To me, Menzie makes the point that, although research do study current events, there is a lag between real world events and research that is properly peer reviewed and published (the lag for the journal of finance is at least a year).
This is why it I think it is critical that politicians seek the advice of experts and academics who are abreast of the latest research.
Posted by: MikeR at January 14, 2009 04:08 PM
I have to say this is the oddest line of comments I have ever seen. Mike Laird chastises academic economists for being complacent. But only a minority of academic economists work in the macro/monetary area, and even they do not typically comment in public venues on current events -- how would one know whether they had a complacent view of the world? If anything, I would say economists in the financial sector and government had a much more sanguine view of economic prospects than academic ones. (Personal perspective: I wrote about the threat from government overborrowing and twin deficits in mid/late 2005 -- and by far the most critical comments were from financial industry economists...so I must be talking to a different set of economists than Mike Laird talks to)
Terry argues that because most papers were submitted after the beginning of December, then they should have reflected the new economic situation. But, for instance, my paper on the ASSA program [pdf] was on the use of deviations from purchasing power parity to evaluate exchange rate misalignment. The paper's topic did not merit changing the conclusions in light of the financial crisis. That probably applied to a great number of papers.
Posted by: Menzie Chinn at January 14, 2009 04:11 PM
Its not true that only "macro" economists had a role in posing questions. Why didn't housing economists look at the prevalence of option arm's in California and their impact on house prices there? Why didn't public finance economists ask what the impact would be of a housing crash on state and municipal budgets, especially when it would likely be coincident with steep pension losses? Why didn't financial economists examine the impact of leverage on the correlation between asset classes? Why didn't labor economists look at the likely impact of low savings rates on unemployment benefits?
I could go an on. There are "micro" questions everywhere one looks.
As far as academic economists being "more critical" than those in the private sector, surely this presents a low bar. The question is not were you more critical, the question is, as a group, did you have any impact on either the stock of knowledge or the path of policy in such a way that could have helped to predict or prevent this crisis. The answer is a resounding "no".
BTW, I predicted this crisis by looking at the likely impact of option-arm loan penetration in California. When penetration reached the flatter part of the "S" curve, I theorized, prices would stabilize and the "true" level of delinquencies/defaults would become evident, which in turn would lead to steep declines in mortgage securities, and in turn...If I could do this working from my office at home in San Diego, why isn't there a single economist that produced a similar analysis?
Posted by: David Pearson at January 14, 2009 04:29 PM
How much access do you have to "economists" in the financial sector? Some give their research out for free, others you have to pay for (or trade with them) and the majority keep their mouths shut.
It does bother me when a "talking head" such as Diane Swonk calls herself an economist.
Posted by: MikeR at January 14, 2009 04:54 PM
MikeR: Well, between putting on conferences directed at the business community, serving on advisory councils, attending nonacademic conferences (all listed on my website), and acquaintances who left government/int'l financial institutions/academia for the financial world, I think I have plenty of access to economists in the financial sector.
Posted by: Menzie Chinn at January 14, 2009 07:48 PM
Menzie, I am not implying that you don't have frequent contact with industry. In fact, it looks like you have more connections than most professors. My point was that in the non-academic world, the smartest people are not necessarily the ones we here on TV.
Michael Lewis has a related antecdote about shorting investment banks:
Posted by: MikeR at January 15, 2009 08:13 AM
MikeR: Point well taken. I agree -- the correlation between intelligence/insight and visibility on TV is likely low. Thanks for the link; I'd only seen the concluding quotation before.
Posted by: Menzie Chinn at January 15, 2009 02:24 PM
David Pearson's comments above are correct. This economic crisis has origins in many fields (housing, monetary policy, int'l trade, tax, to name but a few), and the opportunity existed for many economists to comment on its emergence from many perspectives. Roubini's focus is international macro, but he found ways to comment on US mortgages. There is always a way, if there is an incentive. Roubini also found ways to comment outside the 1 year peer-reviewed publication cycle.
You are correct that most financial services economists were sanguine, but most of these folks publish with a marketing orientation. Its their job. Penetrating independent analysis is available only if you pay for it, or have a large portfolio at the few investment firms with independent analysis teams that have done well over the past year. These are the folks I referred to.
I'm simple minded enough to believe that important analysis doesn't have to be peer reviewed to be useful. The New Yorker is a good place to publish, IMO. Where were the articles by academic economists forecasting quantitatively the systemic results of the Fed's "free money" policies in 2003 to 2005? Where were the articles forecasting the eventual end of Mortgage Equity Withdrawals and the macro impact of that end? The examples could go on. We now know that these kinds of topics can be made into interesting, riveting reading, if approached well.
I don't blame individual economists. I believe most are caught in a system that rewards peer reviewed articles (ie. late for much impact) on very focused topics (too little integration or big picture). Its a system that does not serve society well when big systemic problems occur, and I believe we are entering an era with more big systemic problems than we have faced in the past 50 year era of the Cold War and its unwinding. A reward system problem can only be addressed by multiple Dept. Heads at multiple universities deciding on different ways to manage, reward, and appoint.
Posted by: Mike Laird at January 15, 2009 05:02 PM