Dark matter and Phlogiston

by John Q on December 19, 2005

Given the unwillingness of the Bush Administration to offer any policy response to the massive growth in the US trade and current account deficits, it is not surprising to observe a steady stream of theories explaining that such deficits can be sustained indefinitely.

The latest, put forward by Ricardo Hausmann and Fredrico Sturzenegger of the Kennedy School of Government at Harvard, uses the catchy idea of ‘dark matter’, a reference to the unseen matter cosmologists believe is necessary to explain the observed properties of the universe.

Such hypothetical forms of matter have a patchy history. In the eighteenth century, physicists posited the existence of a substance called phlogiston, with negative mass, suppose to be emitted in the process of combustion. In reality, substances gained mass during combustion because they combined with oxygen.

Another hypothetical substance was the luminiferous ether of the 19th century supposed to carry light waves. Einstein’s theory of relativity eliminated any need for the ether and the idea has been discarded.

Hausmann and Sturzenegger use the idea of dark matter in a couple of ways. The first (fairly uncontroversial) is to argue that while the official accounts show that the US is a large net debtor, the fact that the income account has shown a small surplus for most of the past two decades suggests that the value of US assets overseas is understated. The unmeasured excess is referred to as dark matter.

The second idea is to argue that, since the income balance has been roughly stable over the past five years, the massive trade deficits of that period must have been offset by steadily increasing exports of dark matter, totalling around $3 trillion. Hausmann and Sturzenegger suggest that this dark matter consists of such things as seignorage on US currency, the expertise embodied in operations like EuroDisney and the capacity of the US to provide implicit insurance services.

Something like this must be true if it is assumed that the willingness of foreign lenders to support the trade deficit is the rational outcome of efficient markets. In this sense, the ‘dark matter’ hypothesis is very like phlogiston and the ether. No-one would believe in its existence if it were not necessary to make the theory work.

In 1999, after all, the US economy was riding high. The dotcom mania was in full swing and a stream of books promised a golden future, with an unending boom and the Dow rising to 36 000. The US government had balanced its books and was projecting a surplus of $US5 trillion over the period to 2010.

Is it really plausible that the last five years, with the collapse of the dotcom boom, the Iraq war, the emergence of chronic government budget deficit and weak employment growth have increased international faith in the US to the tune of $3 trillion, as claimed by Hausmann and Sturzenegger?

There have been some important positive developments in the US economy in this period, such as strong growth in output per hour. But this ought to have led to an inflow of equity capital. In fact, foreign investors have generally withdrawn equity capital from the US in this period.

There are some technical difficulties with the ‘dark matter’ analysis. A crucial analytical assumption is that assets should be valued according to the average value of the returns they generate. In fact, however, the riskiness of returns is equally important in asset pricing theory. The visible transactions in the national accounts show that the US has built up a large stock of short-term debt, which offers relatively low risk in the absence of an exchange rate collapse. By contrast, the putative dark matter is illiquid and almost certainly highly risky.

The real problem though is that Hausmann and Sturzenegger overlook the obvious explanation for the strength of the US income balance. For most of the period since 2000, US interest rates have been exceptionally low. Given the importance of short-term debt, this obviously improved the balance temporarily. As interest rates have risen, however, the income balance has turned negative, and this trend will accelerate as debt is rolled over.

One final point. If the sustainability of the US current account deficit is explained by ‘dark matter’ what does this say for Australia and other English-speaking countries with large deficits? It’s hard to believe that we share the supposedly unique capacity of the US for seignorage, insurance and so on. The ‘dark matter’ hypothesis actually makes Australia’s chronic deficits look less sustainable.

More on this from Brad Setser



des von bladet 12.19.05 at 4:50 am

In this sense, the ‘dark matter’ hypothesis is very like phlogiston and the ether. No-one would believe in its existence if it were not necessary to make the theory work.

The reason to believe in the ether (as I sometimes still do) is that waves are vibrations, and vibrations have to be vibrations of something, and the ether is just the sort of something that light waves vibrate in. (The Michelson-Morley effect is just a boundary-layer phenomenon in a viscous ether, says me. Sometimes.)


a 12.19.05 at 6:49 am

“The visible transactions in the national accounts show that the US has built up a large stock of short-term debt, which offers relatively low risk in the absence of an exchange rate collapse.”

Since an exchange rate collapse is possible, I’m not sure this can be counted as low risk. Also foreign investors have increasingly been chasing higher returns (MBSs, etc.).

“By contrast, the putative dark matter is illiquid and almost certainly highly risky.”

I’d disagree with this “almost certainly”. The U.S. is making a lot of money by getting partial or total management control of foreign entities and then having them institutue U.S. style management techniques. Look at all the European companies which are cutting back on workforces, farming out production to the Far East, and so on. The U.S. is just a big hedge fund, getting financing from prime brokers who turn out to be Asian central banks, and then chasing higher returns in equities. While equities are generically riskier, they are less risky if one knows that higher returns can be achieved by instituting a different U.S.-style management style.

Arguably the same is true of other Anglo-Saxon countries, which have an equity culture and are prone to invest in equities. The British certainly are like this, but I don’t know about Australia.


John Quiggin 12.19.05 at 7:18 am

I agree that an exchange rate collapse is quite likely, but presumably Hausmann and Sturzenegger don’t think so.


abb1 12.19.05 at 7:18 am

Isn’t this ‘dark matter’ the stuff every ponzi scheme is made of?


Daniel 12.19.05 at 7:27 am

No, it’s the premium over book value reflecting the fact that, on average, US firms’ overseas operations earn more than their cost of capital. Doesn’t seem an intrinsically ridiculous concept to me. On the other hand; in order to finance trade deficits it needs to be monetised, which could prove difficult and I suspect that the majority of the US/Overseas premium is simply a reflection of the average vintage of capital (to take an example, they capitalise Eurodisney’s earnings at a PE ratio of 20(?), but I wonder if they would have been prepared to do that a few years ago when it was piling up losses), in which case ROTW will start catching up pretty soon.

This looks to me like a situation which is sustainable indefinitely, although I note that nobody ever sold their soul to the devil for the promise of indefinite youth.


agm 12.19.05 at 7:47 am

Brad Setzer? Every single time, I have to reprocess the name to correct the first reading: Brad, not Brian…


Matthew 12.19.05 at 7:56 am

They give a figure for the UK, which is very high, at $234bn annual average. This is more than 10% of UK Gdp.


John Emerson 12.19.05 at 8:04 am

It sounds as though “dark matter” is recently-formulated hypothesis. As such, it might explain why things haven’t gone bad so far, but it granted that the hypothesis has only just been stated, we have no idea how far into the future it will continue to have its effect. Thus, the reassurance it brings is slight.

As a paranoid, I naturally assume that the Saudi and Chinese state banks buy American for political reasons, because those governments like having leverage on the US. As I understand, we attacked Iraq rather than Saudi Arabia because we can’t afford to attack Saudi Arabia (where the hijackers , their leader, and their money came from). We also don’t seem to seriously disagree with China much.


Robert Waldmann 12.19.05 at 8:08 am

typo alert “the official accounts show that the US is a large net creditor” should be “the official accounts show that the US is a large net debtor”.

The ether is, as des von bladet notes, a very natural assumption made because the alternative counter intuitive. In fact, it is perfectly possible to imagine that there is an ether and special relativity prevents us from measuring our motion relative to it (as Henri Poincare’ did independently from Einstein). The view that something unmeasurable should not be posited is really philosophical. Einstein held that view strongly because he was influenced by Mach (the guy with the speed of sound when acting as a philosopher) not because of Michaelson and Morley (the guys with the speed of light).

No one ever imagined that Phlogiston had negative mass. The guess was that things lost weight when they burned (which if you have ever looked at the ashes in a fire place does make a lot of sense). In fact, the modern view of burning wood is that most of it combines with oxygen and then is released as C02 and water vapor. Thus, as is plainly visible the ashes weigh less than the wood and most of the material escapes into the atmospheree.

The modern view of oxidation is based on studying the oxidation of mercury, a relatively obscure topic. Once the experiments were performed by Lavoisier the new theory took over quickly This happened largely because he showed that air is 20% something needed for oxidation which he named oxygen and once that is used up won’t oxidize anything. This observation was a complete surprise (unlike the discovery of oxygen first called dephlogistonised air).

You can look and look in the history of the natural sciences, but you have not yet found an idiocy of the sort which is common in economic theory.

I’d suggest checking the original definition of “organic” in organic chemistry. Now that was a silly idea. Also in a non silly but highly relevant example, there was the hypothesized planet Iaptes closer to the Sun than Mercury. This really was needed to fiddle a model so that the the theory (Newtonian gravity) fit the data. It’s not there but people claimed to have seen it. The anomoly is no longer anomalous given Einstein’s theory of gravity and no one has seen Iaptes in decades.

Oh one other thing. I first heard of the dark matter theory in a debate between Paul Krugman and a famous economist who will remain nameless in 1988. Less dark matter was needed then. Krugman took your side of the debate. He cleaned the floor with the other guy using exactly the arguments you used.

Great minds think alike.


abb1 12.19.05 at 8:21 am

…US firms’ overseas operations earn more than their cost of capital

What does it mean, exactly? There’s the forex market, probably the most open, most dynamic market possible. The forex says that $1=E0.834028. Does it mean that there’s also some dark-matter dollar that is equal to E1.5 or something? Could you link to the dark-matter forex website, please?


Barry 12.19.05 at 9:57 am

Isn’t this just an echo of late 1990’s arguments? We learned then that the market can get irrational by trillions of dollars, and stay that way for a couple ofyear, even though many people pointed out the irrationality.

There was no shortage of “it’s different this time!” arguments, honest or dishonest. I guess that it’s really hard to be an economist, when major markets can exhibit huge, lingering anomalies.


Sean 12.19.05 at 11:23 am

There is a famous astronomical example in which dark matter was posited and later found: anomalies in the orbit of Uranus were explained by a new planet, which we now know and love as Neptune. Hopefully direct detection of non-baryonic dark matter in the universe is on the horizon.


Sebastian Holsclaw 12.19.05 at 11:37 am

“Given the unwillingness of the Bush Administration to offer any policy response to the massive growth in the US trade and current account deficits…”

Can someone point me to what a good policy response to the fact that lots of Americans buy overseas-created goods would look like? What do you do about that? Impose massive tariffs? Would you suggest something like Warren Buffet’s proposed import certificates ? Would you decrease the corporate income tax to increase the savings of corporations? Should we focus on the deficit by immediately raising taxes by about half? Would a huge jump in taxes have a negative effect on economic growth causing even worse problems? The trade deficit has been growing for almost 30 years. What does a good policy to correct it look like?


a 12.19.05 at 12:14 pm

Seb: A VAT and a tax on oil would do nicely, thank you.


Sebastian Holsclaw 12.19.05 at 12:28 pm

BTW, that sounded snarky and (for a change) I didn’t mean for it to.


y 12.19.05 at 12:29 pm

Another famous example is the neutrino, which turned out to actually exist–rather surprising, actually.


roger 12.19.05 at 12:47 pm

A naive question about U.S. and other overseas firms. Are such firms double-counted? Does the revenue of, say, Daimler-Chrysler count as a German asset overseas? Does the Ford share in the Daewoo Motor Co count as a U.S. asset overseas? I’m not sure I understand how this accounting works. Is the determining factor the country to which the asset pays taxes, or is there another factor.


steve kyle 12.19.05 at 1:32 pm

The problem is that the trade accounts only “see” the capital when it crosses a border. If, for ex. you invest in Germany, that exit of money will be recorded. But the US accounts have no way of knowing if your investment is bad, good, or wildly successful. The more wildly successful it is, the more “dark matter” there is out there – i.e. assets which we own but which arent “seen” by our cross border accounts.

Oh and as for your question, the revenue isnt an asset. Its the value of the share or the business thats the asset. And how to value assets in some other country? there’s the problem. as for what “determines” it the question is one of ownership. I can own shares of a factory in Germany and so pay taxes in Germany but I still own that share and the revenue stream that comes from it. If I repatriate the revenue stream then our cross border accounts will “see” it. If I reinvest it then they dont and I have added to “dark matter”

You will have noted that this works both ways. If we have “dark matter” overseas, then the foreigners have “dark antimatter” here. There is some netting out that has to happen before we know the true bottom line.


roger 12.19.05 at 2:23 pm

Steven, thanks.


John Quiggin 12.19.05 at 2:27 pm

Robert, thanks for the typo alert and other points. You’re obviously well-informed on history of science, but my recollection, backed by Wikipedia, is that the phlogiston theory was adjusted to incorporate negative mass when careful measurement showed that substances gained mass during combustion. Cosmic microwave background radiation is a nice example of a successful prediction, I think.

Sebastian, raising taxes seems like a pretty good response to me, as would removal of policy distortions that encourage investment in housing (tax deductibility of mortgage interest). It seems certain though, that any response that brings savings back into positive territory will reduce domestic consumption as well as imports.

Steve, what you say in your first para is correct which is why the first part of the dark matter story is not very controversial – OS assets are probably understated. The problem, as Daniel notes, is that big deficits lead to the need to sell off the dark matter.


abb1 12.19.05 at 2:35 pm

Thanks, Steve, I think I got it. So, the US consumers are buying cars produced in Germany -> they are exchanging their dollars for euros -> the dollar would have to start falling -> BUT -> these cars belong to Chrysler -> Chrysler’s stock goes up -> German investors convert their dollars into euros to buy Chrysler’s stock -> everything’s balanced.

Something like that?


abb1 12.19.05 at 2:39 pm

No, I mean: German investors convert their euros into dollars.


Tracy W 12.19.05 at 4:24 pm

Any theory of capital account deficits has to explain not only the US, but the UK, Australia and NZ’s deficits.

And the reason that governments don’t do anything about current account deficits is that all the options are unpleasant. The ways to reduce a current account deficit can be summed up as either increasing domestic savings or reducing domestic borrowing.

There are a number of options to address each part. Roughly, and in no particular order:
1. The government can increase its own savings (or reduce its deficit). The trouble with this is that it requires reducing government spending compared to the counter-factual, and governments like being able to offer voters new spending programmes or tax cuts (depending on ideological flavour). Voters apparently like new spending programmes and/or tax cuts more than they hate current account deficits. Please think of these spending programmes as healthcare and education before you advocate cutting them.

2. The government can force the rest of the country to increase savings. It can do this by creating incentives for savings (which takes away money from other spending programmes/tax cuts, see 1 above) or by mandating savings, which is politically unpopular as a fair amount of the population argues they need the money for food.

3. The government can try to reduce the rest of the country’s borrowing by raising interest rates. A trouble is that this raises interest rates for mortgage holders – and mortgage holders make up a number of voters. Plus substantially higher interest rates generally reduce economic growth by reducing capital investment (all other things being equal). Governments like economic growth because it brings in more taxes without the pain of having to raise tax rates.

4. The government can pass laws restricting foreign investment. This reduces the amount of capital available. See 3 for the downside.

5. The government can pass laws trying to directly restrict borrowing for consumption while allowing borrowing for investment. The trouble is that the government does not manage to hire all the smartest workers. Some are in the private sector. So any regulation that gets passed has some pretty smart cookies figuring out ways to get around it. And it’s pretty tough to come up with a definition that clearly distinguishes between consumption and investment spending despite any tricks an accountant can come up with.

For the following reasons, for the last couple of decades NZ governments have got elected promising to do something about the current deficit, and they have then not done anything because they have preferred to spend money on public healthcare and pensions. (NZ has moved to a government surplus in that time frame, but that was for other reasons). So I think blaming the Bush administration for failing to do anything about the current account deficit is a bit unfair as every other government seems equally as unresponsive.


Tom T. 12.19.05 at 9:53 pm

I have an ignorant question, and I’m asking honestly, without snark. Why isn’t a trade deficit indefinitely sustainable? As an individual, I run a trade deficit every year vis-a-vis my local clothing stores and electronics stores, and I assume that that situation will continue throughout my productive years. Why is that not scalable to the national level for the United States as to China and Japan?

Also, John Q, in your post you suggest that “increased international faith in the US” economy since 1999 is implausible. It’s not necessary for that faith to have increased in an absolute sense, though, is it, as long as investors’ faith in the US (or UK, or Australia) economy increases relative to investments in the rest of the world? Put another way, it may be rational for faith in the US economy to have declined since 1999, but can it be that faith in the non-Anglospheric economy has decreased even more? I.e., is it that Chinese investors are not betting in favor of the dollar but rather are betting against the euro and the yen?


John Quiggin 12.19.05 at 10:29 pm

Tom, there’s nothing to stop a bilateral deficit continuing as in your example. You run a deficit with the stores, but this is mostly or wholly offset by your surplus with your employer (or your own customers if you’re in business for yourself). You can run an aggregate deficit in a given year, financing by loans or credit cards, but this has to be repaid sooner or later.

Similarly, the US can run a deficit with China indefinitely, but it can’t run an aggregate deficit (correctly measured) indefinitely.

For the US vs Europe, it’s arguable that faith in both has declined, but not, I think that the US relative position has improved dramatically. Japan, OTOH, has recovered strongly, and looks much better relative to the US than it did in 1999.


Tom T. 12.19.05 at 11:48 pm

John Q, thanks; I see how the analogy breaks down. Is it a better question to ask, can the US indefinitely run an aggregate trade deficit as long as GDP continues to grow sufficiently (i.e., the domestic economy expanding enough to generate funds to service foreign debt)? Or am I confusing unlike concepts?


John Quiggin 12.20.05 at 12:02 am

Tom, this is closer to the mark. As a general rule, a country can run current account deficits indefinitely, with a stable debt/GDP ratio as long as the CAD/GDP ratio is equal to debt/GDP multiplied by the rate of nominal growth. For example, if debt is 60 per cent of GDP, nominal growth is 5 per cent and the CAD is 3 per cent, everything is stable.

But, in the long run most or all of the CAD will consist of interest payments on debt, so trade has to return to balance or surplus if debts are to remain sustainable.

The dark matter claim is that growth in US debt is (automatically?) offset by grwoth in the value of overseas assets.


derrida derider 12.20.05 at 12:21 am

“… blaming the Bush administration for failing to do anything about the current account deficit is a bit unfair as every other government seems equally as unresponsive.”

Firstly, not every other government has been as unresponsive – despite public propaganda that it’s for other reasons, it’s actually doubtful that both Australia and NZ would run a budget surplus if not for their current account problem. Secondly, other governments faced with a current account deficit have not elected to simultaneously slash taxes and increase their spending for imperial adventures and corporate lobbies.


abb1 12.20.05 at 3:02 am

Could someone explain why this can’t be analyzed as a function of the exchange rates? If a country imports more than it exports, the exchange rate should automatically correct it by making imports more expensive – no? If this isn’t happening, then there’s some non-obvious demand for dollars and the question is: what’s the nature of this demand, what is it based on? It shouldn’t be too difficult to analyze; and if this demand is the result of overseas assets, it should become apparent.

What am I missing here?


Robert Waldmann 12.20.05 at 9:37 am

Thanks for the Wikipedia link. I stand corrected.


steve kyle 12.20.05 at 10:45 am

dear abb1

Your reasoning underlies many an economic model of international trade. The problem is that too often the data dont support it. Exchange rates perversely move in directions different from what we forecast. If we had a model that accurately captured all of the influences on exchange rates I would be a much richer man than I am. Certainly one of the things missing from your argument is that trade in goods and services accounts for a distinct minority of currency transactions – trade in assets far outweighs the “visibles” for the US (though not for many small countries).

Bottom line – sounds easy but is much harder in practice. Please call when you discover the “true” model – but dont tell anyone else or you will ruin the game.


liberal 12.20.05 at 3:37 pm

abb1 wrote, If this isn’t happening, then there’s some non-obvious demand for dollars and the question is: what’s the nature of this demand, what is it based on?

I don’t know much about forex, but I thought at least one source of demand for dollars is Asian central banks purchase of Treasury notes. In an effort to keep their currencies undervalued.


abb1 12.20.05 at 3:58 pm

Exactly. I don’t think we need to trace every transaction and know all of the influences on exchange rates to figure out what kind of asset is being shipped outbound to balance that $700-billion annual trade deficit – it’s the treasury note. Now, is it being done ‘to keep their currencies undervalued’ or for other reasons – I don’t know, but obviously this is how the equilibrium is maintained and frankly I don’t see any connection here with Chrysler’s success in rapidly destroying Mercedes’ reputation and profitability.


Tracy W 12.20.05 at 4:18 pm

Derrida – having actually worked at the NZ Treasury, I can assure you the NZ surpluses aren’t being run due to concern about current account deficits. They’re being run because the government is concerned about its fiscal reputation, particularly with a National opposition, because the government wants to store up some money to pay pensions in the future, and stronger-than-expected tax payments. Current account deficits are rather low on the scale of concerns.

I cannot say anything for the Australian government.

And I think there are a lot of reasons to object to the US government cutting taxes while raising spending before you get to current account deficits.

Comments on this entry are closed.