Nouriel and I have just posted our updated analysis of US external sustainability, i.e. the United States capacity to continue to run large trade and current account deficits.
The paper is certainly long and perhaps a bit technical, though hopefully it serves as more than just a cure for insomnia. Our conclusion is fairly stark: the US cannot continue on anything like its current path, and won’t be able to continue to place enough debt abroad to be able to finance the trade and current account deficts assocaited with exporting 10-10.5% of GDP and importing 16% of GDP for long. We tried hard to document the analysis behind our conclusion clearly.
If any of you read the earlier version of the paper, we did not change much. Certainly the conclusions are the same. We did update our numbers -- notably by increasing out estimate of the 2004 current account deficit to $670 billion, or 5.7% of GDP -- and make our oil price assumptions explicit. But otherwise, the changes were modest -- a footnote here and there, a critique of Richard Cooper’s thesis, now picked up by some at Morgan Stanley, that the US will be able to import 10% of global savings for a long time, so the current account deficit is sustainable, etc.The key points that emerge from our analysis are:
1) We expect strong import growth in 2004, (15.5%) driven in part by higher oil prices, while export growth will tail off slightly from its pace in the first three quarters, but still come in at a strong 12%. That implies a trade deficit of about $610 billion -- over $100 billion larger than the deficit in 2003.
2) Higher oil explains only about 1/2 of the change in the trade deficit, it is certainly not the whole story. Non-oil imports are also growing at a very fast clip. The forecast assumes the 2004 average oil price will be $42 a barrel, which is consistent with oil at around $50 a barrel through the end of the year. An 04 trade deficit of $610 billion, combined with a growing deficit in transfers and a slight deterioration in investment income translates into an 04 current account deficit of $670 billion, or 5.7% of GDP.
3) We expect both import and export growth to cool a bit in 2005, and for the sake of simplicity, we assumed that oil prices would fall a bit over the course of 2005 and average $42 a barrel -- their expected 2004 average. Our baseline also assumes average export and import growth rates -- 5.5% for exports, a bit above 7% for imports. Implicitly, we are assuming that US and world GDP growth continues, though at a slightly more subdued pace in 2004 and that the recent fall in the dollar against some industrialized economy (which brought the Morgan real dollar down to 90 from a bit above 92) is too small to trigger a major increase in US export growth, or even to prompt US exports to grow faster then US imports. These assumptions generate a trade deficit of $670 billion and an estimated current account deficit of $780 billion, or 6.4% of estimated 2005 GDP.
4) If oil is above $50 in 2005 and averages say $52 a barrel without significantly slowing the US economy, and any increase in US exports to oil exporters is offset by slower growth in US exports to other major oil importers, the mechanical impact of higher oil prices would increase the trade and current account deficits by about $50 billion, or 0.4% of GDP.
5) Even in a more optimistic scenario where world growth remains strong and the small current fall in the JP Morgan real dollar (perhaps augmented by some additional depreciation during the course of 2005) leads exports to grow more rapidly than imports -- say export growth at 7% and import growth of 5%, the 2005 current account deficit would widen to around $725 billion, or 5.9% of GDP. In this scenario, the trade deficit would stay roughly constant in nominal terms and fall a bit in real terms. However, we are forecasting that growing external debt will lead the U.S. to make net interest payments to the rest of the world in 2005, and hence some deterioration in the broader current account even if the trade deficit stays constant.
6) We did not spend a whole lot of time trying to get our 2005 forecast exactly right. It does not matter too much for the long-term, and the core point of the paper is that deficits of this magnitude are not sustainable over the medium term, let alone the long-term. No matter how you look at it, a trade deficit of 5% of GDP or more is way too high to be sustainable in the long-run. The expansion of the 2004 trade deficit did have one interesting long-term effect: it raised the peak US external debt level in our sustained adjustment scenario from around 50% of GDP to around 55% of GDP. The reason is simple: we did not accelerate the pace of adjustment (we believe any faster adjustment would not be consistent with continued US and world economic growth), so higher deficits in 2004 implied slightly higher deficits in 2005, 2006 and so on.
That is one reason why it is important for the U.S. to put in place policies that bring about some reduction in our need to borrow externally, and thus start to slow the growth in the trade deficit. Alan Meltzer mocked Ken Rogoff and others who have long predicted the dollar will fall at a recent AEI conference, and also mocked those who think the US economy will ever be driven by anything other than consumption.
He should be careful. Just because the US was able to increase its net external debt from 15% to 30% over the past four years does not mean it will be able to increase its external debt from 30% to 50% over the next four (remember, US external deficits are rising, so the pace of external debt accumulation is accelerating), at least not without paying much higher rates to compensate creditors for the growing risk.