A fairly recent piece by Hausmann and Sturzenegger challenges this view.
Their approach is not unlike that of astrophysicists trying to explain the stability of galactical systems; the physicists have had to rely on the gravitational pull of something called "dark matter" for their explanations. Hausmann and Sturzenegger argue, analogously, that the international exchange market is much more stable than one might think, based on looking at only current account deficits. They posit that there must be something else, call it dark matter, holding the system together and imparting considerable stability to the system of international exchange.
The nature of this dark matter, according to them, is trade in know-how and specific information. When goods and services are sold (and especially licensed) to foreigners by US firms, there is almost always a part of the agreement specifying that the licensee will also purchase technical know-how and implementation consulting services. These are long-term contracts, and they are assets for the firms selling them, but they don't show up in the current account as sales of assets. Including them in the data provides a different picture of the U.S. dollar situation. My reproduction of their graph below is far from clear, but the two lines show a dramatic difference over the years from 1980 - 2004 (the teal-coloured line is the cumulative current account, as measured; the brown line is the cumulative current account, including a measure of "dark matter" and showing a stable balance over the past two and a half decades).

It is enough to make me hesitate, that's for sure. To the extent this is true, foreigners are importing a lot of know-how from the U.S. and exporting a lot of promises to pay in the future, and neither of these items is showing up on the books.
Brad Setser has a lengthy analysis of their argument; summary: it's the result of corporate tax arbitrage.
I don't think it is wise to downplay the role of licensing know-how in technology transfer. Numerous folks at UWO's Ivey Bizskool have done research on this topic and, as I recall, thought it was pretty important.
I'm afraid it may be too late to submit a paper on this topic to this conference.




