Shrinking dark matter watch

Or maybe shifting dark matter watch.

Via Barry Ritholtz comes news that increasing US investment (according to Mandel) in one key intangible asset - US brands - may be yielding diminishing returns.    Global surveys of teens suggest that US brands are slipping.  Danish brands probably are not much competition in certain parts of the world.   But, globally, the Finns (Nokia), Germans (Adidas) and Japanese (Sony) seem to be doing quite well.   The US is not the only country with intangible assets.

One of the great mysteries of the whole debate over dark matter is how the US will continue to create all the dark matter needed to offset annual $1 trillion current account deficits.  Hausmann and Sturzenegger have argued over at Martin Wolf's (youth-free) forum that the pace at which the US is creating dark matter has accelerated.  They now accept that it may not stay at this pace, but they are argue that the US can bank on adding 2-3% of GDP annually to its stock of dark matter.

Delong has put together a spreadsheet showing how this might happen.

The US makes $600 billion in foreign direct investment annually - and, when it does so, the US should add an additional $300 billion in intangible assets (dark matter) to its balance sheet.

I have to confess I initially found DeLong's spreadsheet confusing.

Largely because the numbers seemed a bit off.   US direct investment abroad is not likely to totally anything like $600 b in 2005.  In the first three quarters of 2005, total US FDI totaled only $20b.  That total reflects the Homeland Investment Act.  But even in 2004 - a peak year - the US direct investment abroad only totaled $250 b.   Between 2001 and 2004 US direct investment abroad averaged a bit under $135b annually (Foreign direct investment in the US averaged $75b).

And somehow, that (relatively small) flow of FDI - call it $150b -- is expected to generate huge amounts of dark matter -- $300 b - every year.

OK - there are other sources of dark matter.   Borrowing cheaply to buy higher yielding assets, whether emerging market debt or foreign stocks for example.  But the US doesn't own that much high-yielding emerging market debt.  And then there is the Peso problem Willem Buiter emphasizes.  Investing in risky assets pays off just until it doesn't.  Argentine bonds yielded high cash returns from 1994 to 2000.   But not in 2001, 2002, 2003 and 2004.

And even after the big rises in the value of foreign stock markets increased the value of US holdings of foreign stocks while a stalled US stock market kept the value of foreign holdings of US stocks down, US holdings of foreign stocks only exceeded foreign holdings of US stocks by $600 b in 2004 (that may be closer to $800b at the end of 2005).  So the US needs to eke out a lot of dark matter out of a relatively small low yielding debt for high return equity trade.

All in all, I don't quite see why Hausmann is so confident that the US will continue to add around 2-3% of its GDP to its stock of dark matter every year  ($250-$350b) -- see Mark Thoma's excerpts from the FT uber-blog over at Economist View.  Hausmann argues that the US historically has added between 2-3% of GDP in "dark matter" to its balance sheet every year, and  this increase would be well below the 5% of US GDP in dark matter that the US added to its balance sheet on average between 2000 and 2004:  

Messrs. Hausmann and Sturzenegger estimate that dark matter grew by $559 billion a year between 2000 and 2004 -- its fastest pace in the past 25 years. On average, annual dark-matter growth has been about $124 billion in the past quarter-century, according to the Hausmann-Sturzenegger study. (Source: Mark Gongloff of the Wall Street Journal)

Consequently, in the view of Hausmann and Sturzenegger, banking on a 2-3% of GDP annual increase in dark matter is quite conservative -- even if gross US FDI abroad is maybe $150-200 b and gross portfolio equity outflows are only $100b (the net flows are far smaller).    The somehow will eke out a dollar of dark matter out of each dollar of gross investment abroad.

I, by contrast, suspect that surge what Hausmann and Sturzenegger call dark matter between 2001-2004 stems not from the relatively small flow of FDI but a series of one off changes that reduced the United States external payments to unsustainably low levels, including:

1) The big fall in US interest rates, which reduced payments on what I have called the debt for future US exports (or current US imports) swap on the US balance sheet, as well as on the debt for equity swaps on the US balance sheet.

2) The tech bust, which reduced the value and profitability of many foreign investments in the US.

3)  The fall in the dollar, which increased the value of US firms European profits and led many European firms to accept lower profit margins to maintain their market share.  

And then there is the other source of dark matter, or maybe "green" matter.

Traditionally, foreign direct investment takes time to become profitable.  Not so with US investment in Ireland.   Microsoft's Irish investments are very profitable.  So too are Pepsi and Coke's Irish concentrate production.  And Pfizer' production of Lipitor in Ireland.  For some mysterious reason, US firms have decided to transfer the production of some their highest margin products to Ireland.  That increases the foreign profits of US firms - and increases dark matter.   I suspect that the low reported profitability of foreign direct investment in the US reflects the fact that European firms have a strong incentive to keep their highest-value added activities in low tax European jurisdictions as well.

Remember that the real source of dark matter is not the superb performance of US firms abroad, but rather the terrible performance of foreign firms in the US.

All in all, I suspect the "surge" in dark matter between 2001 and 2004 reflects low US interest rates, the tech bust, the euro/ dollar and a surge in corporate tax arbitrage.   Some of those changes have been reversed.  US interest rates are rising, and so to is the dollar.  So I doubt that the ongoing surge in "dark matter" that Hausmann and Sturzenegger forecast can be sustained -- indeed, I wouldn't be surprised if some of the dark matter the US discovered between 2002-2004 starts to disappear.

One last note: big moves in the RMB/ $ will yield far smaller balance sheet gains for the US than big changes in the Euro/$ and pound/ $, and I suspect far less dark matter as well.   The US doesn't actually have all that much direct investment in China, and China doesn't have much direct investment in the US.  And if there are not offsetting direct investment positions, it is hard for me to see where the dark matter comes from.