2004 and 2005 current account deficits

Now that we have three of the four data points for 2004, it is worth trying to make a (well) informed guess about the end of the year current account deficit -- and to peer into 2005. This post of chock full of numbers and is quite dense: My goal is to lay the basis for my subsequent, less quantitative, commentary.

2004. Quarterly current account deficits this year have been $147 billion, $164 billion, and $165 billion. Add another $165 for q4, and you get an annual deficit of $641 billion. That is probably too low. The October trade deficit was around $55.5 billion, and simple extrapolation based on current growth rates of imports and exports (adjusted for falling oil prices) suggests monthly trade deficits of $53 billion in November and $54.5 billion in December, for a quarterly deficit of $163 billion (v. $155 billion in q3).

Let’s adjust that down by $2 billion to $161 billion (Last December’s imports were quite high, making it harder to sustain current y/y growth rates). That adds $6 billion to the q4 current account deficit. Then assume that the outflow on transfers -- which was unusually low in q3 because of unusually high transfers inflows: it seems US insurance companies had bought policies from European reinsurers that protected against hurricane losses -- revert back to their q1 and q2 average of around $19.5 billion. That would lead to a $5 billion increase in the current account deficit, and results in a q4 current account deficit of $176 billion. The final tally: an annual 2004 deficit of around $650 billion (5.6% of GDP), with a trade balance of -606, an income balance is + 27, and a transfers balance of -73 billion.

For the record, that is a bit better than I estimated earlier in the year, largely because the "income" surplus now seems larger than initial q2 data suggested.

2005. Forecast out a $175 billion quarterly deficit for a full year and you get an annual current account deficit of $700 billion in 2005, or 5.9% of GDP. That translates into a trade deficit of $640 billion, an income balance of +20, and a transfers balance of -80 billion. One small point: keeping the quarterly trade deficit constant at q4 2004 levels is consistent with y/y import growth of 5% and y/y export growth of 4% -- that kind of year over year growth is just what is needed to keep monthly exports at their end 2004 levels through the entire year (remember, December 2004 exports and imports are higher than January 2004 exports and imports). A $640 billion trade deficit also is consistent with say 8% import and export growth.

However, just forecasting out a continuation of the q4 2004 current account deficit is a bit optimistic. The income balance is likely to be zero or maybe slightly negative rather than positive. Rising short-term interest rates are already leading to rising US interest payments on its external debt stock and higher levels of debt and rising US interest rates are likely to soon overwhlem small increases in the returns on US assets abroad. The transfers deficit also is likely to grow a bit, maybe by $5 billion to $85 billion.

So even if the trade deficit stays at roughly the same level as in the fourth quarter of 2004, the US looks on track to run a current account deficit of at least $725 billion (6.1% of GDP).

However, that too seems a bit optimistic. There is no evidence as of now that the torrid pace of US (non-oil) import growth is slowing -- Morgan Stanley is forecasting a deficit of 6.5% of GDP, or around $800 billion, ouch.

If nothing changes, that forecast is quite plausible. Our oil import bill will keep on rising because of growing volumes (about 5% a year, on current trends) even if oil prices stay at their average 2004 levels, and right now non-oil imports are growing much faster than non-oil exports. If exports grow at a respectable 8% in 2005 (a bit faster than nominal GDP), as a falling dollar offsets a slowing global economy, and if import growth slows to 10%, our trade deficit grows to about $700 billion. Transfers of -85 and income of -15 billion get you to - $800 billion.

And all this assumes a significant slowdown in non-oil import growth. If it continues at its current 15% or so y/y pace, watch out.

This all makes sense from a savings - investment perspective as well: strong import growth implies falling private savings/ growing US domestic investment and a widening gap between US domestic savings and domestic investment that has to be financed by importing foreign savings.

The interesting question is whether the financing the US needs to run deficits in the $725-800 billion plus range will be there.

There was one good sign in the q3 data. Foreign firms invested more in the US in q3 than US firms invested abroad, for the first time in a long time. The net inflow was not large -- only + $10 billion -- but it is better than the -$30 billion or so average quarterly outflow over the previous six quarters. It is easier to raise $725-800 billion abroad than to raise $850-920 billion.

On the other hand, if US investors keep on buying foreign stocks and bonds and other securities at the October pace ($15 billion), the US will need to raise far more than $725 billion abroad next year.

Consider a dark financing scenario. Suppose that quarterly FDI outflows are at their average pace of the last four quarters (-$20 billion), and US purchases of foreign securities (mostly equities) fall back a bit from their October pace, but still average $10 billion a month, or $30 billion a quarter. That would generate an additional capital outflow of $50 billion a quarter, or $200 billion in the year. To fund both its current account deficit and its purchase of overseas assets, the US would need to raise between $925 billion and $1 trillion abroad.

I am cherry picking a bit to get that large an estimate, but I did not have to work very hard to generate it either ... it is -- sadly -- not entirely implausible that the US might need to "export" $1 trillion in debt next year. We live in a debt society, not an ownership society. Still, my overall scenario is probably implausible: if equity investors are taking (net) $200 billion out of the US next year, I doubt the US will be able to sell the one trillion in debt necessary to finance both that equity outflow and a $800 billion current account deficit. Interest rates would have to rise a bit more than is forecast to attract that kind of funding, slowing the US economy, slowing consumption growth, slowing import demand and reducing the trade and current account deficits.

But no matter how you cut it, the US will need to place an enormous amount of debt abroad next year.