February 23, 2006
A Critique of Pure Dark Matter
Net income on the balance of payments might be essentially zero in 2005q4
There's been a tremendous flurry of interest in the Hausman-Sturzenegger dark matter explanation for why the net income account in the balance of payments has remained positive despite the measured net indebtedness of the United States. The hypothesis, I believe, has been ably critiqued by Matt Higgins et al., Brad Setser, and my colleague Don Nichols, among others (not to mention the investigation by CBO of differential rates of return on FDI).
I'm going to tackle this issue from another vantage point. And that is to ask oneself if the net income series that the Hausman and Sturzenegger hypothesis hinges upon will remain positive. My guess is that that outcome remains unlikely.
The figure below shows the net income series for the US, according to the December 2005 balance of payments release. Three observations: First, net income has been negative as recently as 2005q2. Second, even when the series has been positive in recent quarters, the value has been essentially zero, well within rounding error.
Source: BEA, Balance of Payments, December 2005 release; and author's calculations.
Third, 2005q4 net income is as likely to come in negative as positive. I estimated over a 1995q1-2005q3 period a relationship between US income receipts and German and UK short term interest rates, and the US gross asset position (as estimated by Lane and Milesi-Ferretti, interpolated using capital flow data from the IMF, and extrapolated for the 2005q1-q4 period using an exponential trend estimated over the 2002q1-2004q4 period). All coefficients are statistically significant, and the adjusted R-squared is 0.93, SER of $4 billion where mean value is $74 billion. A dynamic forecast is then generated for 2005q4. A similar regression is estimated for US income payments, where the right hand side variables are the short term US interest rate and the US gross liabilities position. The adjusted R-squared is 0.92, SER of $5 billion where mean value is $69 billion.
The resulting forecasted net income for 2005q4 is shown in the figure (as the red line) as -$502 million, on a seasonally adjusted, quarterly basis (or -$2 billion on an annualized basis). Basically, this result occurs because the US short term interest rate is rising relative to the foreign interest rate.
Source: IMF, IFS, averages of daily data.
To the extent that US short rates have risen -- and are likely to continue to rise in the coming months -- US interest payments are likely to grow (most of the foreign holdings of US Treasury securities are at the short end of the spectrum).
The point of this exercise is not really to make a prediction regarding the figure to be released in mid-March. Rather, it is merely to illustrate how close the recorded net income series is to zero, and that it could easily turn negative within the next few quarters. Using the Hausman-Sturzenegger approach, dark matter would evaporate if these trends continue.
Late Addition: I changed the title from "Will Dark Matter Go the Way of Cold Fusion?" after realizing that Willem Buiter's paper on this subject also invoked cold fusion.
Posted by Menzie Chinn at February 23, 2006 02:57 PMdigg this | reddit
"In 2004, according to Hausmann and Sturzenegger, the US got $30 billion more on its overseas investments than it paid on its external debts, which implies $600 billion of assets (with a 5% return). Yet formally the US had $2.5 trillion in (net) external debt, which implies (with a 5% return) net payments of $125 billion. Voila. $3.1 trillion in dark matter."
Hmm, I read the above as saying there would still be 2.5 trillion of dark matter, even if the income balance was zero.
Posted by: Heiko Gerhauser at February 24, 2006 06:23 AM
Excellent observation on the possibility that March or future quarters could go negative. We believe that the "multiplier effect" running in reverse will be observed through "dark matter".
The Nattering One thinks that the "dark matter" is there. But, the old "three card monty" through accounting chicanery to minimize tax liabilities causes distortion in the numbers, which makes its mass and effect difficult to quantify.
It's like trying to quantify the government or church's black market profits.... you really don't think these 3rd world countries pay their loans with profits from clay pot sales, do you??
We leave you with this to ponder: The Chinese who invented creative book keeping, keep four sets of books, one for external auditors, one for the tax man, one for internal purposes and one for their wives. Let the debate rage on.
Posted by: The Nattering Naybob at February 24, 2006 06:26 AM
Very good post ... I think the discussion about DM is very interesting because the questions it raises are fundamental;
1. Do we have global imbalances or not and how much do they matter?
2. The indestructable US economy; i.e. dark matter makes a huge CA deficit wither away?
3. The ability of official statistics to track the flow of factors in a knowledge based economy.
Ohh and btw, I have made it my personal obligation to map out the flurry of the DM discussion over at my own blog mainly because I find it very interesting and as I said fundamental for the practice of macro-economics; you have been blogged as well :).
Posted by: claus vistesen at February 24, 2006 07:33 AM
I should have cited this paper by Matt Higgins, et al. directly critiquing dark matter.
I also omitted a citation of this post by Brad DeLong.
Finally, this paper by John Kitchen on predicting the components of the current account has recently been brought to my attention; it presents a much more detailed approach than incorporated in many other expositions circulating.
Posted by: menzie chinn at February 24, 2006 01:54 PM
Fair point, and a fair summary of my argument. if the net income position turns slightly negative as US external debt rises -- i.e. adding say $600 b to the US net debt in 2005 erases the $30b income surplus, the stock of dark matter doesn't fall. For the stock of dark matter to actually fall, (net) income payments have to rise faster than the stock of US debt.
That said, Hausmann and Sturzenegger do not just argue that the US has a constant stock of dark matter, but they argue that the stock of dark matter the US has will continue to rise, offsetting ongoing deficits. Recently they have conceded that it may not rise at quite the same pace as it did from 2001-2004. But based on historical experience, they still expect the stock of dark matter to increase by between 2-3% of US GDP, of over $250b, every year.
Menzie -- I too need to read the Kitchen paper.
Posted by: Brad Setser at February 24, 2006 03:37 PM
"Will Dark Matter Go the Way of Cold Fusion?" would have been an absolutely terrible title.
It appears that Brian Naranjo, Jim Gimzewski, and Seth Putterman at University of Los Angeles have actually gotten it to work.
This experiment has been repeated successfully and other scientists have reviewed the results: it looks like the real thing this time.
I don't think that is the point you wanted to make.
Posted by: Mark at February 24, 2006 04:33 PM
Thanks for the additional links; they will go swiftly into an update.
Especially Matt Higgins' paper is concise and to the point which may also be its weakness, but I like it simple.
Kitchen's paper might also be worth a look as it argues that USA will have no difficulties servicing its debt in the future.
"An important result in the
base projection, however, is that the cost of servicing the U.S. net international debt remains very
low � at much less than one cent per dollar of GDP."
Posted by: claus vistesen at February 25, 2006 04:32 PM
Very interesting discussion.
It occurs to me that the "light matter" of the equation, the ultimate source of the high rate of return on FDI, may be related to the structural pricing power of producing goods at low cost in the 3rd world and selling at a high margin in the the developed world markets. The margin for domestic corporations is at a record level. How long can this persist?
At some point, the value of the dollar has to decline so that domestically produced goods are competetive in the world market or domestic production costs have to deflate. Assuming that the higher rates of return are as a result of lower wage and operating costs in the third world, the disparity between dollar based domestic and third world incomes have to equilibrate. If the dollar does not decline, then domestic incomes will decline with an complementary decline in domestic personal consumption expenditures (PCE). Killing the golden goose.
I think we have already seen the beginnings of the process. The dollar is artificially inflated in value through currency manipulation on the world markets. Domestic PCE has been supported through mortgage equity withdrawl (MEW), even in the face of flat to declining real income growth, allowing the US consumer to party hardy while defraying the equilibration of the monetary markets.
I have a sense that the current state is best defined as a Nash "stable disequilibria".
The current account deficit equation also ignores (correct me if I'm wrong), off sheet "borrowing" of social security receipts by the government.
Hopefully, one of you brilliant gentlefolk will discuss the relationship between "asset inflation" and general inflation.
Posted by: STARK at February 25, 2006 05:03 PM
don't think that is the point you wanted to make.
Actually the dark matter - cold fusion analogy might be very good now that 'cold fusion' was shown to 'be posible'...
(1) Both have little practical value - changes nothing in the real world.
(2) Both give false hope of an easy magical 'solution' to our difficult problems.
(3) Both are mostly of 'academic' interest. It won't play in Peoria very well.
There maybe some 'dark matter' out there but it doesn't make our balance of payments problem any less of a problem. It is a red herring in that respect.
Posted by: dryfly at February 26, 2006 10:59 AM
In the end most of the explanations amount to variations on the theme of that the US has high paying (if potentially risky) long term FDI abroad, while much of the lending to the US takes the form of shorter term bonds with low yields but presumably lower risk in the US (although I understand that major multinational banks are now including a calculation of the probability of a default on its securities by the US government into their regular copula value at risk calculations).
It is pretty straightforward. With the net indebtedness continuing to increase at a rate between a half and a whole trillion US $ per year, there would have to be an enormous increase in dark matter to keep the income part of the BOP positive. I see no major change in the composition of US FDI abroad, but we do see rising interest rates in the US, especially on all those shorter term securities. There is simply no way the income part of the BOP account can remain positive for much longer. The only question is when it will go negative, and once it does, it will almost certainly stay there and get worse.
The real fun and games in all this will be when that income account gets to the neighborhood of negative $100 billion or more (less actually) per year. Then we may see some truly unpleasant stuff start to happen.
Posted by: Barkley Rosser at February 27, 2006 01:18 PM
What few have noticed is how the income surplus that the current account statistics reports is largely a accounting fiction. Little Ireland with just 4 million people for example had a income deficit of nearly $30 billion. This however is not a case of Ireland having a $600 billion net external debt , but it is simply a result of the practice by multinational (particularly American)corporations to have internal pricing policies that transfers profits to Ireland because of its low (12.5%) corporate income tax. This artifically boosts Ireland's reported trade surplus while increasing its factor income deficit.
This for America means that its trade deficit is overestimated by the official numbers, while the factor income post falsely reports a surplus.
Posted by: Stefan Karlsson at February 27, 2006 01:19 PM
I'm glad to have sparked this interesting discussion. I thought the "dark matter" issue might have played itself out by the time of this post.
I do want to stress that my main point was that net income is likely to go negative, and I've heard nothing to dispell this view.
I might mention one aspect of the revaluation effect on U.S. assets abroad, denominated in foreign currencies: it relies upon continued depreciation without an exchange risk premium arising. Even if no risk premium develops, it is important to remember secular dollar depreciation is not costless. This occurence would mean the terms of trade would worsen over time, and this in turn means that U.S. GDP measured in terms of foreign goods and services decreases over time (this is the terms of trade adjusted GDP series reported, for instance, by the Penn World Tables) relative to the real GDP expressed in units of home goods and services.
Posted by: menzie chinn at February 28, 2006 11:59 AM