[Previous] [Main Index] [Next]

Tuesday, August 19, 2003


An economics professor at UC Berkeley who runs a good web site, Brad DeLong expresses some stimulating thoughts about the size and duration of the US trade deficit . . . presumably in manufacturing and agricultural goods plus fuels and raw materials, as well as in services. See the chart on the page linked to here. (The Trade Account refers to exports and imports in the goods just mentioned. Add in Trade in Services and Unilateral Transfers, in which the US consistently runs a surplus, and you get the Current Account . . . itself highly negative for a couple of decades, save for 1991, but not nearly so large as the trade in goods itself.)

DeLong's puzzled about the size of the trade deficit (presumably, the current account) and its persistence at high levels with some ups-and-downs for more than two decades now. Why, presumably, doesn't it adjust in textbook fashion towards more balance in exports and imports on the trade side? The buggy prof left a trio of lengthy rejoinders at DeLong's site; revised and amplified, they will be found here in a few moments. First, though, we need to summarize DeLong's thought-provoking reflections that prompted the buggy replies.

Note that by the time you've worked your way through DeLong's argument and the three sets of buggy replies, you should be able to understand much better, let's hope, what the balance of payments is, the causes of the US trade deficit, how that deficit relates to the rest of GDP, and whether it's a big problem or not. And, to top it off --- because of some bugged-out comments in reply to someone else at the DeLong site --- whether or not the US is a particularly violent country compared to other industrial countries. On the latter score, expect some surprises.


A Trio of Likely But Incomplete Causes

DeLong notes that there are three immediate and likely causes of the persistent and growing trade deficit since the 1970s . . . none of which in his view fully exhaust the size of that deficit or its long-term duration for two decades now. Specifically, large US trade deficits are:

1) Due at times to macroeconomic mismanagement in the US --- a point he doesn't clarify, and may be wrong as you'll see below because, as it happens, despite claims to the contrary, there's no clear correlation between US federal deficits or surpluses and the size of the US trade deficit.

2) Due also, in part, to macroeconomic mismanagement abroad . . . . by which DeLong presumably means overtight monetary and fiscal policies that produce slower growth in the EU and Japan and most of East Asia than is desirable, in which case faster US growth is bound to increase our imports from those countries and regions and slow down our export sales there. That's much more likely to be true.
3) And reflect the big statistical problems that US Commerce department --- specifically its Bureau of Economic Analysis that handles national income accounts --- has encountered in tracking billions of economic transactions undertaken annually by American citizens and residents here with the outside world. This problem, note, has grown especially acute as our economy has moved from a largely "closed one", with little trade in goods and services and even financial inflows and outflows in the early post WWII-era, to a much more "open economy" in which the imports and exports of goods alone has risen from around 4% of GDP in the 1950s to triple that these days. Even that stat is misleading. We actually export more than 25% of our mfg. goods and about 40% of our agricultural ouput, even as services have internationalized rapidly too. (Most services, of course, aren't tradeable across boundaries.) And all at a time, remember, when the economy has grown not just more integrated with the outside world but far more difficult to track as information and communications technologies and various kinds of new service industries --- think of Amazon, McDonald's, Walmart, Microsoft, IBM, Morgan-Stanley, and Dell --- have come to mark the American economy . . . even if most jobs are created and held in small business firms. The result? These various tracking problems show up as a large item in the US balance of payments known as "statistical discrepancy."

We'll return to this matter later.


DeLong Argues That Political Explanations Are Therefore Needed

What is particularly stimulating in DeLong's analysis is some outright political explanations to supplement these three economic and statistical causes and explain the large and stubborn size of the US trade deficit. To understand his argument, a little background needs to be sketched in here about the balance of payments and how it's organized and uses double-entry book-keeping. In particular, the US balance of payments divides into three parts:

1) The Trade Account: exports and imports in mfg. and agricultural goods and in fuels and raw materials:

2) Trade in Services and Unilateral Transfers: exports and imports (credit and debit items) here too:

--- for instance, travel by US citizens abroad and by foreigners here; or the use of US airline carriers by foreigners to reach us and vice versa; or brokerage and insurance and banking services used abroad by Americans and by foreigners here; or exports of American films and TV and imports from abroad; remittance of profits earned by US multinational abroad to the US, and vice versa for foreign multinationals operating here that send profits back to Japan or Europe; and --- unilateral transfers --- things like foreign aid or Mexican workers sending money home every month to Mexico.

Together, these two parts add up to the Current Accout. In the US, it's running around $300 billion annually on the debit side: hence the current account is running a deficit of about $450 - $500 billion projected for this year, roughly 4.0 4.5% of GDP. See the BEA chart. That leaves . . .

3) The Capital Account, both short-run inflows and outflows as well as long-term.

Short-term inflows and outflows of capital in the US involve the purchases of financial assets like bonds and corporate equities. Long-term inflows and outflows usually refer to non-liquid made investments by Americans to purchase a fixed financial asset abroad --- say, the purchase by Microsoft of a German mfg. firm for $300 millions that is then integrated into Microsoft's global multinational activities; and vice versa, the purchase by a German firm, say Mercedes of Chrysler, in this country. Microsoft's acquisition of the German computer firm is a debit item in the US balance of payments for the year when it's undertaken: it has to exchange dollars for euros in order to make the purchase --- which is an outflow of dollars abroad --- in the same way that you or your friends have to exchange dollars to buy a German-made Mercedes that is imported here (strictly speaking, of course, it's the US import-firm that has to exchange the dollars for euros, and there are further complications because of the ways in which double-entry bookkeeping changes the liabilities of US banks on this score). Oppositely, Mercedes' purchase of Chrysler for billions of dollars is an inflow of foreign money into the US for that year, and it's recorded as a credit item.

Owing to double-entry bookkeeping, any deficit on current account is automatically going to be offset by an equivalent surplus on capital account. This is easily shown by national income accounts, of which the balance of payment is a sub-item. We'll do the simplified math later by way of illustration: it's actually simple to follow. (See the addendum at the end).


Back to DeLong's argument.

The unusual and stimulating political analysis has three points, all of which he refers to as a "exorbitant privilege" for the US.

To clarify briefly, a few more background comments need to be sketched in for non-economists. For one thing, the US $ is by far the most important currency used in world trade (even Brazilians trading with China use it). For another, even more significant thing, it's by far the most important of the major currencies --- the euro and the Yen are the other two --- used by both private parties (individuals and firms) for investment purposes and by central banks for holding their foreign currency reserves. In 2001, if I remember correctly, the $ accounted for about 65-70% of various central banks reserves around the world; and though that has probably fallen somewhat since then as both private and central banks have adjusted their portfolios away somewhat from the dollar and into euros, it's still likely to be close to 60%. That's what DeLong means by the US being a key currency (along with the Yen and the euro).

And other background knowledge needed? Yes. Something critical.

In particular, if foreign investors --- whether individuals or firms or private banks or central banks --- desire to hold US financial assets (stocks and bonds; real estate; US businesses from Chrysler to MGM), they will have to exchange Yen and euros and Chinese Yuan and Mexican pesos to acquire dollars. In that case, the price of the dollar against these other currencies tends to rise for these financial-investment reasons. The consequence? As the dollar rises, US businesses are able to sell fewer goods abroad (US exports) that are now more expensive in Europe and Asia and Latin America and elsewhere, and oppositely, American households and businesses buy more and more cheaper imports. (Even with the dollar falling about 20% against the euro since the end of 2001, and much less against the Yen and the Brazilian currency and Canadian dollar, it has really fallen only 7% or so on a trade-weighted basis that takes into account our trade with the euro-countries and trade elsewhere in the world. Simultaneously, the dollar has actually risen against the Mexican Peso and remains the same against the Yuan and other Asian currencies.

Note, finally, to complete the background, that whether the term "exorbitant privilege" is sound or not --- in the buggy professor view, as you'll see, it's lopsided --- the three points are worth pondering and seem persuasive, at any rate in part.


"The first exorbitant privilege is that foreign central banks prefer to hold their reserves in dollar-denominated assets--and as the world economy expands, they want to hold more and more of such dollar-denominated assets. That finances a component of the trade deficit: after all, if they want to buy dollar-denominated assets, they first need to acquire dollars by convincing us to buy imports. (Moreover, this preference for dollar-denominated assets means that dollar real interest rates are a bit lower and the foreign exchange value of the dollar a bit higher than would be otherwise.)

"The second exorbitant privilege is that rich people in many foreign countries think that dollar-denominated assets--large sums of money in the Vanguard funds or somewhere in Citigroup--are an important part of their political risk insurance portfolio. Suppose something bad happens in domestic politics, and you need to make a run for it in the Learjet (or, worse, have to paddle away in a small rubber boat and hope that the smugglers you have hired to guard you aren't close readers of The Strategy of Conflict). Large amounts of $$$$ stabled in financial assets inside the jurisdiction of the U.S. government make post-exile life a lot more pleasant. Thus there is an extra part of the trade deficit which is really a service export: the U.S. selling political risk insurance to the rich in foreign lands

"The third exorbitant privilege is that even if the rich abroad are confident about the political stability and economic prospects of your native land, the United States is still a very, very nice place to live in many, many ways. Lots of people living elsewhere know this, and think that even if they don't want to live in America, the odds are very good that some at least of their children or grandchildren will. How do you give your descendants the option of living in America--of becoming Americans? Investing a good chunk of your family wealth in America--and spending time there--is a good way to raise the odds that whichever of your descendants wish to become Americans will be able to slip through whatever immigration barriers will be raised in ten, twenty-five, or fifty years.

"So there you have it. Some of America's trade deficit is not really there--the result of errors and omissions in the data, a "statistical discrepancy." Some of America's trade deficit is there, but is not "unsustainable": the portion of America's trade deficit that is the result of its three "exorbitant privileges" can continue until the age of the world changes: American can keep selling international reserve and liquidity services, political risk insurance services, and future immigration options to the central banks and rich of the rest of the world for a long time to come.

"Only that portion of the trade deficit that is neither (a) a statistical discrepancy, (b) the result of "exorbitant privilege", nor (c) clearly a short-run and transitory cyclical phenomenon is cause for concern. How large is that worrisome component of the trade deficit? I don't know. It bothers me that I don't know--because I am supposed to."

Posted by DeLong at August 16, 2003 10:49 AM | TrackBack




These are stimulating set of comments, Professor DeLong, especially the references to the political motivations of various foreign investors in the US economy. Some responses follow:

[1] Statistical Discrepancy. Your reference to it seems accurate, but you don't pin the causes down (admittedly speculative for any of us). a) Lots of exported services --- not least in information and communications industries --- are probably not reported to the Commerce Department. Think of a law firm in New York doing business by the Internet with a Canadian client. b) Lots of intra-firm trade across borders from one branch of a multinational with a US base to another branch that is never reported to the Commerce Dept. c) And lots of accounting manipulations, not least transfer-pricing that allows multinationals --- not just US but foreign --- to reduce their overall tax burdens.

[2] Exorbitant Privileges

The political motivation is nicely brought out by your categories, but the term "exorbitant privileges" seems excessive and tendentious and ignores just old-fashioned economic exchange. Consider

a) Central Banks rolling up dollar assets. Yes, but the counterpart is that they --- or at least their governments when they're private --- also want and seek to promote export-led growth, especially into the US economy. The new euro bank might have been an exception from 1999 on, concerned to make the new euro system work, but we do know that Berlin, Paris, and Rome among other government in the EU didn't mind at all a falling euro from $1.18 to 0.86 between 1999 and early 2002 . . . since which time, of course, the euro has risen to around $1.14.

b) Privilege Two -- rich people and businesses investing here --- seems overdone in the case of stable countries like Japan or the EU region or maybe even for lots of rich Indians. No doubt the political motive is greater for rich Arabs and Latin Americans and Asians not in Japan or India. But even they, surely, like the enormous offerings --- range, depth of bond and stock markets, dynamism of the US economy's long-term growth prospects, certainty of no capital controls etc --- that amount to a major economic motive too.

c) There is a huge imbalance in the global system, with the EU, Japan, and China waiting for an economic revival here. This is discussed in an article at my site on the Chinese economy: a short-term problem for the US (using an undervalued Yuan, back by huge dollar buyings daily for months now, to push such a surging export-drive in certain US industries --- a $100 billion plus deficit at an annual rate), but a long-term promise.

See http:www.thebuggyprofessor.org Article on China, August 2nd, 2003.

Posted by: michael gordon on August 16, 2003 04:58 PM



[1] The Loss of mfg. jobs in the US economy --- concerns about which were voiced by others who have commented on DeLong's argument at his site: This fear, observe carefully, has been around for decades now, voiced loudly in some quarters, as though it betokens catastrophe for the US economy. The reality? Far from that being the case, our economy has created about 50 million new jobs, on a net basis, since the post-WWII turning point. And especially in the 1990s, most of the jobs were created were in new service industries with average or higher-than-average wages. See the Clinton Council of Economic Advisers' White Paper on this, published in early 2000: "20 Million Jobs: January 1993 - November 1999.

[2] One fear with more substance to it that was voiced in the 1970s, 1980s, and early 1990s had to do with the growth of labor productivity: would a largely service-oriented economy be able to foster growth in productivity as fast as one based more on mfg? (In fact, even in the fast-growth period of productivity advances in the 1950s and 1960s, mfg. accounted for less than 40% of the labor force, and was steadily declining then too.) As we now know since 1995, this fear too has proved wrong. Productivity advances have doubled at an annual clip, and we now know with greater certainty they weren't just a blip on the long-term growth of productivity caused by excessive investment in information-and-communication technologies.

If anything, it's the EU economies on the Continent that -- with a much higher percentage of the labor force in mfg. than we have (in Germany, well over double our 12%) --- that now find themselves stuck with pokey growth in productivity and excessive wage costs in mfg. compared to other rapidly industrializing countries in Pacific Asia and Brazil and Mexico. Japan, too, is stuck with more mfg. jobs than its firms are competitive in, internationally viewed.

[3] A third fear connected with the declining mfg. sector in the US economy has to do with the problems of low-wage labor since the mid-1970s. Between 1975 and 1995, the bottom 20% were clearly losing out in terms of wages: overall, they suffered about a 15-20% decline in real wages --- a big problem. Some of that had, almost certainly, something to do with the loss of about 5 million mfg. jobs in the smoke-stack industries and autos in the Mid-West (which underwent, all the same, a remarkable decade-long transformation into a high-tech region), though how much isn't something that has been established.

What now appears to be the case, given the rapid increase in low-income wages between 1996 and 2000 --- they actually rose faster than the wages of the top 20% --- is that this fear, while not entirely misplaced in those days, was wrong. The key change after 1995 was the big reduction in unemployment, first below 5.0% (with most economists howling that this was lower than the natural rate of unemployment, inflation sure to surge soon in the wake), then below 4.5% and toward 4.0%.

Labor in low-skilled and mid-skilled jobs was so scarce that businesses in the Mid-West were going into prisons and securing the release of non-violent offenders whom they then trained to take jobs in those business firms. The result? A huge jump in low-income wages. In parts of California, starting wages were double the minimum wage in parts of the Bay area and Los Angeles.


[4] But wait. That doesn't mean some concerns being voiced about lost jobs in mfg and other industries more recently --- around 2 million or so since the start of 2001 --- aren't justified, given the strange performance of the job market since the technical end of the recession in the fall of 2001. In particular, the huge inflow of Chinese exports --- pushed by an undervalued Yuan that the Chinese government maintains by selling large quantities of Yuan daily for dollars and dollar-denominated financial assets (up to $300 million daily for months now) --- doesn't create a short-term problem, above all in the few industries like textiles and wood-products where those exports from China are concentrated.

Our trade deficit with China is running well over $100 billion annually, far higher than even our deficit with Japan's . . . which, in any case, despite the Japanese government's efforts to keep the Yen from appreciating, has fallen from about $60 billion in 2001 and 2002 toward a annualized level of about $40 billion this year. And the rise of the Yen, about 7-8% against the $, has eroded the profits of lots of big Japanese firms, many of them, in consequence --- especially in autos --- accelerating the relocation of production to this country. What's to be done with the Chinese-created problem without outright long-term protection? I've grappled with this in a lengthy analysis published on this site: http://wwww.thebuggyprofessor.org See the article there for August 2nd on China, "A Short-Term Problem, A Long-Term Promise for the US economy." (It's the first of two-parts. The long-term side --- generally favorable to the US economy's trade and investment relationship with China --- will be set out soon. As a political scientist, I add, I'm especially interested in the implications of growing Chinese GDP and wealth for the Pacific-regional and global distribution of power.


[5] On a trade-weighted basis, the dollar has fallen about 7-8%. As someone noted here, the fall has been against the euro mainly, the Canadian dollar, the Brazilian currency, and the Yen. It has risen against the Mexican peso and not budged against the rest of Asian currencies. Fortunately, the last published statistics about the US trade balance is that it fell dramatically in the early summer. Exports were up sharply, imports down even more. That was to be expected, what with the lengthy period in which a dollar depreciation against certain key countries takes before relative price changes show up.

Even then, the length has been spun out and the effects lower because the EU economies and Japan's are so stagnant. In such circumstances, as DeLong notes, a rapidly growing economy here --- probably at a 3.0% annualized rate in the spring, far higher more likely since the end of June --- is bound to suck in more imports than otherwise, while finding it harder to locate customers in those slow growing countries for US exports.

That creates a major problem of imbalance in the global economy, which is hard to justify in 2003 . . . decades after the EU and Japan became wealthy countries with huge internal markets. Even China --- with a population of 1.3 billion --- is far too dependent on export-oriented growth (produced mainly from coastal mfg. centers, with a colossal influx of multinational investment from abroad).

Again, this is discussed at the buggy site in two other very recent articles besides the one on China:

"The Job Market and Unemployment in the US and the EU"

"Foreign Competition and Industrial Competitiveness,"


[6] US, Violent Crime, Other Havens. (This too is in response to another comment-poster at the DeLong site).

Derider, the Australian visitor here, notes something about crime --- violent and otherwise --- that is, I fear, a misconception: the US is a particularly crime-ridden country, again violent or otherwise. He raises the matter, then asks why --- since Australia, his home country, has financial attractions and is also known for being low in crime compared to the US --- Australia doesn't get more of the politically motivated investment. Well, there are multiple reasons, but let's stick just with the crime-side.

In fact, as UN surveys of crime-victims undertaken every 4-5 years since 1989 by a Dutch university team show, the US is in the bottom third of industrial countries in both the incidence of crime, violent and non-violent, and trends. Australia, it will surprise Derider no doubt, tops the list of violent crimes per capita, and you are 6 times more likely to be mugged in London these days than in New York. The US population, moreover, shows the most confidence in our police, and simultaneously is found to show the least fear about going out into public spaces.

These stats --- which parallel those gathered by the US government in crime-victim surveys --- are also supported by a different data base that we and other governments use: crimes reported to the police and investigated. Interpol gathers these and publishes comparative studies too. If you're interested, you can find my comments on these --- plus links to other articles that deal with them, by me and others --- at another web-site where I posted on Friday:

"US vs. EU Crime"

It's a high-quality site and so is the commentary left by visitors. You'll find the comments I left using the Dutch survey --- its links too, and links to other articles of mine on this --- if you scroll down the lengthy comments to my name or the buggy professor.

Posted by: michael gordon on August 17, 2003 09:26 AM



1] My first set of comments above [1] has an omission that can lead to a wrong impression: " . . . our economy has created about 50 million new jobs, on a net basis, since the post-WWII turning point. . ."

Add after the post-WWII turning point, the year "1975". That's a turning point because the unusual rapid growth of GDP in West Europe, the US, and Japan between the end of the 1940s and then was brought to a jolting stop by the first oil-price hike and the resulting dislocations caused to industrial economies (and not just them alone).

Essentially, GDP growth in West Europe and Japan fell between 50 - 67%, and has never been improved on since except for Ireland and Britain and Spain, Portugal, and Greece. In the US, the slower grower between 1945 and 1975 --- fully to be expected in terms of convergence theory catch-up growth, with follower countries once launched on a path of sustained growth almost always growing faster than the lead-country --- GDP growth fell off far less: roughly by a third. What did fall by half or more was the growth in labor productivity.

[2] In the 1990s, the US economy recovered. In fact, it had been growing faster in GDP than the EU average in the 1980s, but it was only in the 1990s that our GDP advance and especially the renewed vigor in labor productivity growth made themselves felt. (Of the major industrial countries, only Britain has also improved noticeably its GDP and productivity growth since the 1975 turning point. Ireland, 4 million people (Britain has 57 million), has surged in economic performance and become the richest country in the EU!)

[3] As a result, by 2001, the EU average in per capita income compared to the US had fallen from around 85% at the start of the 1990s to around 68% according to EU stats, and the level of labor productivity had fallen back to 79%. (I posted on this recently at this site, with some comments too about a different study by Robert Gordon on this topic. For an expanded and updated version, see the buggy prof article, "Some Curious Problems with Cross-Country Comparisons of Labor Productivity: The EU and the US"

[4] Of the other industrializing countries, only Pacific Asia maintain high growth rates of GDP after 1975. Even then, by the early 1990s --- a few years before the 1997 financial meltdown --- growth was slowing in both North and SE Asia, and noticeably so. The exception, slightly, was China in the post-Mao era, though there have always been problems with making sense of Chinese official national income stats, and the use of deflators for inflation. Since 1997, the sustained high growth of the Chinese economy --- though officially a couple of percentage points lower in GDP growth --- seems doubly suspect . . . to the point that even the former Prime Minister, Zhu Rongji, publicly complained that the stats were bogus, a fabrication of local Chinese officials in the regions around the country, plus CP members, wanting to look good in Beijing. For a discussion of the Chinese stats and their dubious quality, based on an excellent study by Thomas Rawski of Pitt, see the buggy article for August 2nd:


Posted by: michael gordon on August 17, 2003 09:58 AM


Replies: 1 Comment

I agree with your refutations of the 3 "exorbitant" privileges. Furthermore, central banks are small players in the international currency markets.

I believe the reason the US has been able to sustain a trade deficit for so long without depreciation is that Japan and the EU have much higher savings rates and their financial institutions channel a large proportion of these to the US because (i) they want to geographically diversify, (ii) they believe(d) in better prospects for US companies than for their counterparts in Japan or Europe.

It was interesting to see how these capital inflows fell as soon as the US markets fell. I guess these inflows have picked up again somewhat lately, given the positive earnings forecasts from US companies, hence the $ has done better lately than in quite some time.

All it means is that more and more of the US is owned by Japanese and Europeans (although they tend to be investors when markets are high...).This is not much different from myself borrowing from my bank because I spend more than I earn. This is OK as long as my earnings increase fast enough. If Japan and Europe stopped investing in the US, I guess there is only one way for the $ : down, which should solve the trade balance somewhat, but not totally.

Does this make sense ?


Posted by Tom Dechaene @ 08/20/2003 07:50 AM PST