« Turning Up the Heat | Main | An Unexpected Addition to the Order of the Shrill? »

December 24, 2004

General Glut Foresees Balancing Down

When the U.S. current account deficit returns to something more normal, will it be because a falling dollar has boosted net exports and encouraged or been accompanied by policy changes that boost saving, or will it be because a fall in investment has reduced U.S. incomes, caused a recession, and so diminished imports? General Glut argues for the second:

General Glut's Globblog: Brad DeLong today gives a tepid endorsement of a "soft-landing" scenario for the US current account deficit.

To restate: (1) the dollar falls, (2) as a result net exports rise, (3) export and importing-competing industries hire workers, (4) unemployment falls, (5) wages start rising and bring rising inflation with them, (6) the Federal Reserve raises interest rates to stop any inflationary spiral, (7) the economy cools off as higher interest rates reduce construction and investment spending and raise unemployment back to its natural rate.

It could happen--if exports react rapidly and substantially to the falling dollar, and if the rising long-term interest rates that diminish employment in construction and investment-goods production are somewhat delayed...

But well before we get to step [7], it would be nice to see some actual evidence that we can even get to step [2]. Thus far we have had nearly three years of [1] ("the dollar falls") and not a bit of [2] ("net exports rise"). From February 2002 to December 2004 the dollar has fallen 16% in real terms (per the broad dollar index) while the trade deficit has grown in real terms about 60%. I keep hearing the "delayed effects" mantra, but it's beginning to wear really thin for me.

What if there is no necessary connection at all between [1] and [2]? What if the US is the global price-setting market, and thus rather than importing inflation via a falling dollar which would cut into import consumption, the US exports deflation instead and maintains import consumption at current or even rising levels?

The key mechanism to correct the CA deficit is not a falling dollar, but -- as I discussed early last week -- rising savings.

International economists are divided into those who look at prices like exchange and interest rates and how they affect imports, exports, and capital flows; and those who look at the circular flow of national income and how the savings-investment balance determines net exports. We have airtight theoretical arguments that these two analytical roads lead to the same place. So why is it that those who focus on exchange rates ("falling desire to hold dollar-denominated assets reduces the value of the dollar") are more likely to see a happy, balancing-up resolution while those who focus on the circular flow ("falling desire to hold dollar-denominated assets leads to a sharp fall in the financing available for investment and a spike in interest rates") are more likely to see an unhappy, balancing-down resolution?

Posted by DeLong at December 24, 2004 11:45 AM

Comments

All this talk about rising exports with a falling dollar. My question is which exports where? What American products has the world been on the edge of its seat waiting to buy, but couldn't or wouldn't because of a strong dollar? Our illustrious gas guzzlers? Our dazzling plasma TVs? Oh wait, we don't make TV sets anymore.

People want American pop culture (even if they say they don't), and they're already buying it. People want certain American tech products (although those will likely face stiff competition from Asia in coming years and decades), and they're already buying that stuff.

And wouldn't we have to have a sea change in attitudes about spending generally in Asia before some great new wave of export-led growth in America could happen? Japan you'll note has been stuck in the deflationary gutter for years and Japanese consumers still haven't stepped up to the plate.

Also, I for one believe that those grim projections about economic growth that the Bush administration has been using to sell social security privatization may have a grain of truth, even if they don't in any intelligible way support the privatization thesis (poor economic growth of course bodes well neither for the social security system or the markets, so why bother privatizing?) The point is that we may be facing a serious demographic slump as the spendthrift, nearly 80 million strong boomer generation enters retirement (and presumably spends less), and the frugal, less than 40 million strong generation x enters middle age.

Posted by: Robin the Hood at December 24, 2004 12:11 PM


PS I agree completely with this:

"What if the US is the global price-setting market, and thus rather than importing inflation via a falling dollar which would cut into import consumption, the US exports deflation instead and maintains import consumption at current or even rising levels?"

And what happens with a falling dollar *and* falling consumption in America, without rising consumption elsewhere? Seems to me a recipe for long-term deflation.

Posted by: Robin the Hood at December 24, 2004 12:16 PM


Surely it is because there are multiple solutions to the mathematical problem, and because the two specialities focus on different parts of the solution space? I think it is the understanding of the behavior of markets in time that makes the difference. Seems to me the exchange rate economists tend to believe that the response of increased production happens in same time scale as the stimulus of increased demand, whereas the circular flow economists will point out that any response will have a time lag during which there will be a period of deflation--that it takes time to, for instance, build factories, hire and train employees, and so forth.

Posted by: Randolph Fritz at December 24, 2004 12:16 PM


Brad asks why those who focus on rates have such a different view from those who focus on circular flow.

Why is it that so many economists appear unable to remember from one moment to the next that Say's Law is a kind of fallacy? The classical model, where movements in general price level, "naturally" lead toward a full-employment equilibrium, in this, the best of all possible worlds, is a load of crap. Even in neo-classical finery, (put on to reconcile it to the Keynesian General Theory), it is seductive to a certain sort of "optimistic" mind.

Bottom line, though, is that prices are, as they used to say when I was in college, "sticky," and a great many markets, for various structural reasons, do not clear. Short run, long run, many markets do not clear, ever, and couldn't clear if they wanted to. An analytical fetish for asserting that a fall in the value of the dollar will result in increased exports/reduced imports, because of Price Mechanism changes, and that that Price Mechanism effect will dominate all other factors, is just a fetish.

Anyone, with even a passing acquaintaince with the history of the past fifty years of currency valuation change is likely to realize that the circular flow analysis works a good deal better in explaining what happens. Duh!

Posted by: Bruce Wilder at December 24, 2004 12:19 PM


I think the previous commenters are getting at the same issue but surely the answer to Brad's question is that the circular flow people are people who at some level see the usefulness of the simplest Keynesian multiplier model with no role for prices at all, despite all the debunking of that model over the last 40 years.

Posted by: P O'Neill at December 24, 2004 12:48 PM


Reporting to duty, General Glut!

Hi, Brad! We discuss you all the time. You are so reasonable, in the old fashioned way, always assuming things are normal. Well, we are falling off a cliff, sir, badly. And if one looks only at one thing or another, it isn't so hideous but all things in concert? A disaster.

Like us, you hate Bush and his buddies and what they are doing, but you don't seem nearly alarmed enough! The Titanic America is not only badly listing, the cold water is already flooding the lower decks and the upper decks? The chairs are sliding and the drinks are spilling....

Posted by: Elaine Supkis at December 24, 2004 02:13 PM


o/~ I'm in the dance band on the Titanic
Sing "Nearer my god to me!"
The iceberg's off the starbord bow
Won't you dance with me? ~/o

I think, really, we've a long way before we're sunk, though.

Posted by: Randolph Fritz at December 24, 2004 02:42 PM


Brad, as a physicist it seems to me that economists are so in love with equilibrium that they seem uninterested in looking at kinetics seriously.

All the viewpoints in the argument seem to be saying "well, in the long run entropy is maximized", a perfectly true, and frequently useless statement, while no-one seems to have a QUANTITATIVE theory of just how the approach to maximum entropy occurs, the sort of thing that in physics we might call transport theory or in chemistry chemical kinetics.

I understand that the problem is a whole lot harder, but dammit, people have been modelling this stuff for fifty years. Why are we stuck with this sort of hand-waving "Joe claims a will take 3 yrs to happen, followed by b; while Fred claims a will take 6 months to happen meaning that it'll be followed by c"?

Posted by: Maynard Handley at December 24, 2004 03:33 PM


Hypotheses based on a layman's understanding Mandelbrot on markets (I think they're even relevant, though they really belong in an open thread):

1. The social order of economies is hierarchical, hence the distribution of property and the size of market transactions tend to follow power-law distributions.

2. Different economically-significant activities take different amounts of time, and time to start and stop, hence markets exhibit memory.

3. The larger the shift in an economy, the more transactions must be undertaken to account for the shift and the more quickly those transactions must be settled, hence market time varies.

Turning to the subject at hand, I think the key difference between the two groups is in their understanding of the interaction between market time and economic "memory." The exchange-and-interest rate economists assume that memory is captured entirely in interest rates and that major non-linearities will not occur, whereas the circular flow economists assume that memory is "deeper" and the system less responsive. I don't think we really know enough to predict this, myself.

Further speculations: it is a commonplace of the "new economy" that it requires much less physical capital and concentration than the old; there is much less need for large-scale industrial facilities and concentrations of labor, though there are perhaps still reasons to build them. If that is so, I suppose the views of the exchange-and-rate economists will turn out to be valid.

That and a nickel...

I don't think I want to pursue a PhD in this topic, really not.

Posted by: Randolph Fritz at December 24, 2004 03:34 PM


"Brad, as a physicist it seems to me that economists are so in love with equilibrium that they seem uninterested in looking at kinetics seriously."

"Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is long past the ocean is flat again."--JM Keynes.

I think that insight, though, has yet to be fully incorporated in economics.

Posted by: Randolph Fritz at December 24, 2004 05:45 PM


General Glut wrote:
"... Thus far we have had nearly three years of [1] ("the dollar falls") and not a bit of [2] ("net exports rise"). From February 2002 to December 2004 the dollar has fallen 16% in real terms (per the broad dollar index) while the trade deficit has grown in real terms about 60%. ..."

1. But most of the fall in the dollar has been against the Euro. The change in the dollar/yen rate has been retarded by massive Japanese market interventions, the dollar/yuan rate has not changed at all, and Japanese exporters have been adapting by rapidly shifting production to China. So the effective change in rates must have been much less than 16%.

2. There is inevitably a lag between a change in exchange rates and a change in trade patterns, even for commodity goods for which it is easy to change suppliers. For non-commodity goods with substantial technical content, there is a hysteresis factor in addition to the lag; the cost of qualifying a new supplier is significant, it will take time for a new supplier to tool up, and if the machinery of the US factory that once produced the product has been packed up and shipped to China, and the work force fired, there will need to be a substantial change in exchange rates before anyone will try to revive a US-based source. (Note that this is one of the most important reasons why exchange rates should never be allowed to get as far out of balance as they have -- these hysteresis effects can nullify true comparative advantage.)

3. Why does General Glut not consider the possibility that had it not been for the shift in exchange rates, the trade deficit might have grown in real terms much more than 60%?

Posted by: jm at December 24, 2004 06:40 PM


Circular flow is an illusion. Things are pushed out and reabsorbed in different places.

Posted by: cloquet at December 24, 2004 07:24 PM


Leaving model building for a moment, you must have an idea of what industries you expect could soon start exporting again as a result of a dollar drop that would allow a smooth landing? At what dollar/rembi exchange rate could we say expect to take back 50% of lost Walmart business or some similar benchmark?.
As with all other professions, Economics must also face moral issues. Here, I would ask what are the advantages and disadvantages of a hard versus another soft landing. Following-up with what Walensa said - we are not much a superpower.
If that is so then is it advantageous to try and remain so? Or would a hard landing (like a nervous breakdown) force us to reconsider who we are and who we cannot be.
Best holiday wishes to you and your family.

Posted by: aeolus at December 25, 2004 12:06 AM


Brad asks:
"So why is it that those who focus on exchange rates ("falling desire to hold dollar-denominated assets reduces the value of the dollar") are more likely to see a happy, balancing-up resolution while those who focus on the circular flow ("falling desire to hold dollar-denominated assets leads to a sharp fall in the financing available for investment and a spike in interest rates") are more likely to see an unhappy, balancing-down resolution?"

Because neither theoretical school is including the rest of the various factors into the equation.

The dollar is falling not because the current accounts balance is so out of whack, but because the US government is simply printing money to cover its operation---and appears to intend to do so indefinitely at an ever increasing rate for as long as it can get away with it.

The dollar is being driven down primarily because there is a lack of long term confidence in the dollar. The "solution" to the eventual US debt crisis will be high inflation---and people don't want to be holding a whole lot of treasury bonds paying 5% interest when inflation is running at 25%.

Politics plays a role as well----people were fine with dollar hegemony when the US foreign policy was not being run by crazy people. But there is also a sense that US power and ambition must be curbed, and reducing the power of the dollar as a medium of exchange is one way to achieve that. (This is why the European central banks are doing nothing to stop the decline of the dollar relative to the Euro.)

The exchange raters are under the impression that the US can become significantly more competitive in the international marketplace---but the only way that will happen is if there is a significant decline in the US standard of living. The decline of the dollar may mean that the third world workers who produce our apparel are now making the equivalent of $1.50 an hour instead of $1.20 an hour, but that obviously doesn't make the US work force significantly more competitive. A US company who is going to invest in additional productive capacity still has incentives to do so overseas (especially with all the "free trade" agreements), significantly reducing the theoretical effect of the dollar decline on exports.

The circularists are under the impression that rising interest rates will spur additional savings in the US. And although this may eventually result, the real impact of higher interest rates will recessionary. US consumer debt is astronomical, and a very large chunk of that debt is "variable rate" credit card debt. Raising interest rates will mean that people have to devote more of their income to their monthly credit card debt---they will be able to buy less, and the incentive will be to pay off that debt, and not invest their money because the interest rates being charged by credit card companies will always be higher than that being offered in interest bearing investments. (This of course is likely to result in credit card companies having more "savings"---but the economy's real engine is consumer purchases, not business investment.)

In sum, the reason that the two sides are at such odds is that both appear to be working from false premises involving not only the reason for the decline of the dollar, but the expected impact of their proposals.


Posted by: paul_lukasiak at December 25, 2004 05:27 AM


>This is why the European central banks are doing nothing to stop the decline of the dollar relative to the Euro

I wish I believed that, but I'm pretty much convinced at this point that the ECB is full of self-hating Europeans.

The rest of it rings true, though.

Posted by: a different chris at December 25, 2004 05:56 AM


"The exchange raters are under the impression that the US can become significantly more competitive in the international marketplace---but the only way that will happen is if there is a significant decline in the US standard of living."

It would help if we made things people actually wanted too.

Posted by: The Chesire Cat at December 25, 2004 12:10 PM


Aren't you economists trying to linearize complicated nonlinear problems? Correct me if I'm wrong, but most of what you talk about sounds like exploration of (or arguments about) the detailed shape of whatever local minimum your system happens to be sitting in at a given moment. Eventually there are major devaluations, or market crashes, and the overall pattern of the landscape changes. At that point we get a metastable situation where the system doesn't know where to go - for examples, traders no longer know when to buy or sell in order to move around the local contours. How about Monte Carlo studies to determine where the various local minima lie under various scenarios? Does anyone do that? Surely there are aspects of the current situation which could be modeled better using different assumptions such as "stickiness" and so forth?

Posted by: Ralph at December 25, 2004 12:16 PM


I would also like to say that Paul Lukasiak's comment, above, sounds spot-on to me. This administration is driving us all down into a deep hole. How (or if) we ever get the vehicle back up to ground level again is apparently none of their concern.

Posted by: Ralph at December 25, 2004 12:24 PM


Another dynamic to consider is that there is a growing movement abroad for boycotting American goods. So, whereas a weak dollar would theoretically lead to an increase in exports, this may be tempered by those not wishing to buy our goods purely for political reasons. This may be where our poor foreign policy comes home to roost alongside poor fiscal policy.

Posted by: [MAC] at December 25, 2004 12:31 PM


I should know more economic history than I do. Are there significant examples of large national economies that experienced sustained long-term growth that did not (a) protect their domestic markets to at least some extent and (b) produce and export a surplus using whatever means was necessary? Historically, it would seem that the Netherlands, Great Britain, and the United States all followed that route on their way to their peaks. Since WWII, Japan, Taiwan and South Korea all seem to have followed that route. Today, China and India seem to be started down the same path.

Given examples, did any give up on (a) and (b) and still maintain robust growth over a long term?

Posted by: Michael Cain at December 25, 2004 03:04 PM


Michael Cain

Take a look at Australia which seems forever to have run a trade deficit.

Posted by: anne at December 25, 2004 03:44 PM


Theoretical economics trades in models--perhaps necessarily. And most useful models employ ceteris paribus assumptions that, in real life, are falsified. Explanations in economics are instrumental: a "true" story is one that enables prediction and control.
Whether the story matches with or corresponds to reality is another matter, and might be irrelevant to prediction and control--
modern economic theory as Ptolemaic science in drag.
YET many economists and politicians employ realist rhetoric--speaking as if obviously their models corresponded with reality. That is simply childish--even if irresistible.

Posted by: TB at December 25, 2004 10:29 PM