Source of Trade Deficit Reductions: Exports, Imports or Prices?

Yesterday's July trade release was a little bit of a surprise, due to oil [1];  Haver covers the numbers. Calculated Risk discussed the release, and actually took the outcome as a fairly positive, albeit with some anxiety about whether exports will keep up the robust growth necessary to continue shrinking the deficit.

I want to focus on a couple of other aspects of the release which seem to make me worry a bit more.

First, let's confirm that the non-oil deficit is shrinking rapidly in nominal terms, and certainly as a share of GDP.

Figure 1: Monthly trade balance (blue) and trade balance ex oil imports, in millions of USD, seasonally adjusted. NBER defined recession dates shaded gray. Source: BEA/Census, July Trade release, September 11, 2008, and NBER.

Where is the gain coming from? As Figure 2 illustrates, goods exports are rising rapidly, as are total goods imports, while goods imports ex oil are trending upward much more slowly.

Figure 2: Log monthly goods exports (blue) and goods imports (red), and goods imports ex-oil (green), in millions of USD, seasonally adjusted. Source: BEA/Census, July Trade release, September 11, 2008.

The series are graphed in log terms, so that a constant growth rate shows up as a straight line.

Nominal values are important, particularly in regard to the trade balance. That's because the trade balance, which roughly equals the current account balance, is what has to be financed by capital inflows. Always something to keep in mind when one ponders the desirability of dollar denominated assets as the US financial system is buffeted by shock after shock. For now, flight to safety is supporting flows into risk-free dollar assets. What will happen in the future as other dollar assets begin to look riskier and riskier is something else.[2]

But what I want to stress here is the real magnitudes. And that's because these are what matter in terms of contributions to US GDP growth, in an accounting sense.

Figure 3: Log monthly real goods exports (blue) and goods imports (red), and goods imports ex-oil (green), in millions of Ch.2000$, seasonally adjusted. Linear trend for 2003.01-08.01 (black) Source: BEA/Census, July Trade release, September 11, 2008, and author's calculations.

What is interesting to me is that exports are growing at a pretty constant rate of 8% per annum (log terms), which is only slightly above the rate 7.5% recorded since the end of the last recession. The jump in the total value of export is to some degree a function of the prices of US exports. Imports on the other hand are coming down. Of course, this is exactly why GDP and Gross Domestic Purchases have diverged. (And declining import quantities reminds us that the 3.1% contribution of net exports to the 3.3% GDP growth in 2008Q2 was about half and half driven by import compression and export expansion [3]).

Figure 4: Log price of goods exports (blue) and goods imports (red), goods imports ex-oil (green), and goods exports ex-agricultural commodities. Source: BLS, August import/export price release, September 11, 2008.

What Figure 4 demonstrates is that the jump in the nominal value of exports is substantially due to price effects particularly over the past year (remember, these are logged prices). The fact that the total goods export price index is below the ex-agricultural commodities index reflects the commodity boom. Hence, my key question: What happens if food prices follow energy prices? Then it's a race, with both the value of total goods and value of total goods exports coming down (one tends to think oil imports must be overwhelmingly larger than agricultural exports and other commodities the US exports. Through July, combined US imports of food and industrial supplies is about $530 billion, and combined exports about $300 billion). In any event, the ex-oil goods trade deficit will probably evidence less improvement over time if commodity prices come down rapidly. Figure 5 shows IMF forecasts (left panel) and futures prices (right panel) for some commodity categories and commodities, respectively.

Figure 5: Commodity prices and projections/futures prices. Source: IMF DMD Lipsky presentation (9/9/08).

While futures do not indicate a sharp drop in agricultural commodity prices, I think the sharp drop in oil prices should give us all pause for thought.

A final observation is that the recent disjuncture between GDP and gross domestic purchase price deflators [4] is not entirely due to oil. Prices of manufactured imports are also rising -- in particular in prices of imports sourced from the NICs and China. Here, we seem to have achieved what our policymakers have been asking for.

Figure 6: Log price of goods imports from Newly Industrializing Countries (blue) and goods imports from China (red). Source: BLS, August import/export price release, September 11, 2008.

Is there any bottom line from this survey? If there is, it is that there are two reasons to worry about the sustainability of improvement in the ex-oil trade deficit. The first is the one Calculated Risk highlights (will faltering rest-of-world growth hamper export growth?). The second one involves the trajectory of prices of some of our exports, in particular agricultural commodities.

Comments

[TheRabble:]

A highly academic way of saying the dollar is overvalued.

2008 Sep 12 05:20 PM