Ricardo Hausmann defends dark matter in Martin Wolf's FT Economist's Forum (the section of the Hausmann and Sturzenegger paper on page 5 called "What is Behind Dark Matter" is a good introduction if you are unfamiliar with the dark matter concept):
FT Forum, by Ricardo Hausmann: Dear all: Martin Wolf is quick to dismiss the discussion of dark matter and side with Willem Buiter. However, Federico Sturzenegger and I still believe they miss the point big time. ... At this point of the debate I hope we all agree that there is a puzzle between the story that the current account deficit describes and what the financial income flows reflect. A correct descriptive explanation of this puzzle is that the rates of return on US liabilities is significantly smaller that the return on its assets. ...
OK, so the question is why US assets abroad earn more than US liabilities at home. ... The profession must offer an interpretation of this puzzle. Our interpretation is that the US has a comparative advantage in providing insurance, liquidity and innovation to the rest of the World, and that the rest of the world is happy to pay for this. Willem Buiter and Martin Wolf buy the seignorage argument but point out that the magnitude is too small to compensate for the US deficit. They jump from this conclusion to the belief that therefore the US current account imbalance is unsustainable. But this does not solve the puzzle. They still owe us an explanation for why the US accumulated a deficit of over 5,000 billion dollars and is not paying for it. ...
The key point is that we believe that for the purpose of assessing global imbalances this differential return ... should be considered as a regular income flow, the capitalized value of which we call the stock of dark matter. ... If the reasons for this return differential are and will be there, you need to take them into account when thinking about the future. One way to do so is to realize that this is like an asset and you may want to give it a value to get a better sense of your current asset position. In the technical version of the paper we show how this should be done in a complete market setup but we don’t need to worry about this here.
Let me explain two additional reasons why our numbers for the stock of dark matter are large and correct. First, it is clear to us that BEA´s adjustment of the value of FDI abroad is very arbitrary and may seriously underestimate the problem. ... Second, it has been argued that our measure incorrectly uses the accrued income flows, rather than what is actually paid. ... What is clear is that at the end of the day the measure will be fine ... over a period of 25 years our measure stands and is really big.
This takes us to our final point. Since we’ve circulated our paper many people have volunteered to us a prediction on the evolution of US net foreign debt obtained with conventional accounting exercises. They were trying to point to us that in spite of a large stock of dark matter the dynamics still look bleak. Fair enough. This may turn out to be true. But this misses our point. We show that the US has historically been exporting (and quite robust statistically at that) over 2% of its GDP in FDI know-how, insurance or liquidity services, call it as you wish. This number has gone up to around 5 percent over the last five years. So when we look forward, can we assume that from now on nothing of this will continue to happen? Again, that may very well be true, but you need a hell of an argument to say that what has been a steady trend over the last 30 years suddenly is about to disappear in 2006.
What is our personal take on the question of whether dark matters exports may continue at their recent pace of close to 5% of GDP? Probably not... In any case, it may very well be the case that the US settles for its long run average of 2-3% of GDP of dark matter exports. No enough to cover a 7% of GDP deficit, but something that helps quite a bit. The point is that you cannot just assume it away without a pretty strong explanation.