Count me among those who were surprised by the size of the fall in the q4 current account deficit. I was expecting a fall, but not as big a fall as the BEA reported this morning.
The q4 deficit came in at $195.8b, down from $229.4b in q3. Add in a $213.8b deficit in q1 and $217.7b deficit in q2, and the annual deficit came in at $856.7b. That is about $200b more than the 2004 deficit.
A significant fall in the q4 deficit was expected. The big fall in oil prices in q4 -- together with decent export growth and relatively subdued non-oil import growth -- brought down the trade balance. But the size of the fall was bigger than I expected. I thought the income balance would continue to deteriorate. Instead, it improved -- a q3 deficit of $5.3b turned into a q4 surplus of $3.0b. I was expecting something more like a deficit of $8b -- or was until the flow of funds data came out (it also showed an improvement in the income balance in q4, see table F107)
The q4 income balance likely will be revised downward over time. That has been the pattern this year (note the rise in the estimated current account deficit for q1 and q2). And some of the details of the q4 data seem a bit strange. The earnings of foreign direct investors in the US fell by over $5b in q4. Sure, the US slowed, but I thought Toyota still was doing rather well. The microdata indicates the fall comes entirely from a huge fall in reinvested earnings, which more than offset a rise in distributed earnings.
Seven other things jumped out at me.
1. The US borrows to consume, not to invest. That at least is the story of the flow data. US equity investment abroad exceeded foreign equity investment in the US (counting portfolio equity as well as FDI) by about $80b. The current account deficit by contrast was around $860b. Think of it this way: the ratio between borrowing to invest ($80b) and borrowing to consume ($860b) was about 1: 10. The true global financial intermediaries of today's world are found in Europe, as London and a host of eurozone financial capitals take in large sums from Asian central banks and oil exporters looking for alternatives to the dollar and use those inflows predominantly to finance their external investment. So long as Euroepeans didn't use central bank inflows to buy too many CDOs backed by subprime US mortgages, that presumably is a good business.
2. The rise in the US oil import bill accounted for most of the deterioration in the US current account deficit in 2006. As Menzie Chinn has highlighted, the non-oil trade deficit has stabilized on the back of strong export growth and slowing non-oil import growth. The transfers deficit fell in 2006, as US government grants fell by about $10b (from $30b to $20b) -- think a fall off in budgeted economic aid to Iraq. And the deterioration in the income balance -- $20b -- was smaller than might be expected. The US afterall needed to take on $850b of debt at say 5% to cover its current account deficit, which implies a roughly $40b increase in its interest bill ...
3. Optimists will no doubt argue that the q4 data shows that the US deficit has peaked. I am not so sure. If the US avoids a recession, I doubt the US non-oil trade balance will improve significantly. Oil now looks to be in the $60 range -- so it is hard to bank on further improvements. And I don't expect the q4 improvement in the income balance to be sustained -- so I still expect a deterioration in the income balance to push the 07 deficit toward $900b, absent a recession or a big fall in the price of oil.
4. Believers in dark matter will take comfort in the fact that the US earned almost $300b on its direct investment abroad, while foreigners earned only $150b on their direct investment in the US -- and increase of $15b over 2005. (The actual numbers are $295.9 and $145.6b). Thanks to the strong performance of foreign equity markets, the market value of US FDI abroad exceeds the market value of foreign FDI in the US -- but not by a ratio of 2 to 1. The reported rate of return on US FDI is still higher than the (very low) reported return on FDI in the US. Skeptics will note that $130b of the $150b difference comes from "reinvested" earnings; the gap is a lot smaller when you just look at actual distributed profits. US firms reinvested $213b abroad, foreign firms reinvested only $80b in the US. My take: the very low reported return on FDI in the US suggests that the US data is undercounting the real profits of foreign firms operating in the US.
5. Interest payments on US debt rose by $137b in 2006. But the interest receipts from US lending rose by $102b, almost as much. Both totals include dividend payments on portfolio equity, but those are small. Rising debt and rising rates increased US interest payments, as one would expect. But the rise in US interest receipts was more impressive -- and seems to reflect the very short-term structure of US external lending. As short-term rates rose, so did US receipts. My forecast for 2007 assumes the rise in debt payments will continue, but, not that the Fed is on hold, the rise in US interest receipts will stall. The result: a big deterioration in the income balance.
6. Recorded net debt inflows of around $800b fell short of the $940b the US needed to finance its $860b current account deficit and its $80b equity investment abroad. In other words, the errors term was large ...
7. Recorded official inflows rebounded from their 2005 low, reaching $300b. Think $120b in treasuries, $125b in agencies, $20b in bank deposits (onshore) and $30b in other (mostly corporate bonds, but some stock). However, $300b in recorded inflows still seem small relative to the roughly $900b increase in foreign official assets ($750b of reserves, $150b roughly in oil investment funds, including Norway). I think official actors put more than a third of their assets into US dollars. Some of the gap is explained by the growth of central banks' offshore dollar deposits (though that growth stalled in q3, according to the BIS data). But some of the gap is also explained by the use of custodians for central bank purchases, and the fact that some central banks and oil funds use private managers for a share of their portfolios.
Update -- the BIS data (table 5c) suggests that offshore dollar deposits are on track to rise by between $125-135b this year (they are up $100b ytd), offshore euro deposits are on track to rise by $75b, and offshore pound deposits to fall (ball park) by $25 -- after a huge surge in pound deposits in 05, central banks have been paring back, though it isn't clear whether they are shifting into other pound assets or not. I don't think the COFER data suggests a comparable fall off in pound reserve growth, just a slower increase than in 05 -- but the COFER data has gaps. Sum up the $300b in US inflows and $125b in offshore dollar deposits -- the $425b total (which may have some double counting) is still on the small side relative to the $900b growth in official assets; hence my belief that there are some official flows masquerading as private flows in the US data.