Wednesday, June 01, 2011

Supply-Side Factors and the Soft Patch

It seems to be official; while those immersed in the data have been noting ominous signs about the strength of the US's economic recovery for a couple of months, the business press has now grabbed onto the theme wholeheartedly:
The weakness in the US economy over the past few months has been disappointing, to say the least. But I've been wondering whether it's possible that the palpable slowdown in growth could be directly the result of two potentially significant events that have certainly had a negative impact on the US economy in recent months: the sharp rise in oil prices, and the earthquake in Japan. Could this spring's sudden slowdown in the US economy simply be due to these suppply-side shocks?

So I did a little back-of-the-envelope estimation to see whether we can explain the "soft patch" as simply the result of expensive oil and horrible earthquakes. The table below shows my calculations. I make a couple of assumptions to arrive at these estimates. First, I use as a rule-of-thumb the formulation that a $10/bbl increase in the price of oil translates into a roughly 0.3% fall in GDP over the course of a year. (That figure is based on a number of discussions about the impact of the price of oil on the US economy, including by Dean Baker and Roubini and Setser (pdf).)

Second, I construct an estimate of the impact of the earthquake in Japan on manufacturing output in the US, based in part on some of the narrative accompanying the Fed's most recent release of Industrial Production data, which specifically noted how earthquake disruption significantly impacted US auto production in April. Third, I assume that US exports to Japan fell as Japan's economy slowed sharply in the wake of the earthquake. (Note that I also assume that both of these effects will be slighly reversed in future months.)

The results are not trivial, but neither are they as large as I had hoped, to be honest. The calculation above suggests that these two effects reduced payroll employment by perhaps 70 thousand in both April and May. The negative impact on GDP growth was likely only around 0.2% in Q1( from the headline annualized rate), though it may be closer to 0.5% in Q2.

So unfortunately, these effects are simply not large enough to completely account for all of the slowdown in economic activity in the US in recent months. They certainly haven't been helping things, and are responsible for some part of the soft patch we're in... but it also seems that there must be some demand-side contributors as well. And that's the worrying part, because while the negative supply-side factors may reverse themselves in coming months on their own, it's considerably more difficult to imagine where a new boost to demand is going to come from.

1 comment:

  1. Ken Houghton11:14 AM

    references tht follow are for standard calendar quarters.

    Finger exercise: adjust the 2000-2011 GDP trend as follows:  Keep Q1-Q3 data stable.  Choose one of (1) replace each Q4 with the mean of the previous Q3, the following Q1 and the average of the two Q2s on either side or (2) take the monthly data for October and November and either (a) average or (b) extrapolate linearly for December. (Remember, this is a finger exercise; the distinction between linear and cubic splines is not significant.)

    Recast the GDP data for 2000, and bootstrap the rest of the way using the real change for non-Q4 quarters.  Lather, rinse, repeat.

    Call the result "The Christmas Miracle." Not especially what such changes do to implied recession datings.