Just an Oily Patch on the Road to Recovery?
The Bureau of Economic Analysis estimates that real GDP grew at an annual equivalent rate of 1.8 percent in the first quarter of 2011, down from a pace of 3.1 percent in the fourth quarter of 2010. On the surface, this substantial deceleration owes much to reductions in defense spending, nonresidential structures, as well as to increases in imports.
The question we explore here is whether this slowdown is likely to be temporary. From a purely statistical perspective the answer is yes. During the average post-WWII recovery, output has grown 5.5 percent annually (from the trough) in terms of GDP, but in the latest recovery the growth rate has been a comparatively paltry 2.8 percent. If you place your faith in statistical regularities, you would say the current recovery is overdue for a little pick-me-up.
One problem with that conclusion is the fact that there have not been a lot of recessions, at least not enough to make a meaningful statistical inference. The economic contexts under which the “average post-WWII recovery” occurred are likely to be very different from the one we find ourselves in right now, not to mention the fact that the average recession was not as large as the latest one.
The slowdown would also certainly seem temporary if one focused on the components that constituted the drag on GDP growth in the first quarter of 2011. National defense spending tends to be very erratic on a quarter-to-quarter basis, but eventually we would expect it to increase to levels that are consistent with the appropriation spending outlined in the budget. The increase in imports is also likely to be short-lived, given the continued weakness in the U.S. dollar. The problem with nonresidential structures might be more, well… structural, but even there one can point to short-term factors like the expiration of the renewable-energy tax incentive at the end of last year, which caused a substantial pullback in power-generating structure construction.
The problem with this type of analysis is that it ignores categories that were not a drag on GDP growth but could have grown more had economic circumstances been different. Here we are thinking of the effect of high energy and commodity prices on private consumption and investment. Such prices have direct effects on production, as they are an important component of the cost of intermediate products and services.
There are also discretionary income effects that come about because disposable incomes have fallen across many households (after accounting for low elasticity spending like transportation and heating costs). Moreover, there are other, perhaps less obvious, indirect effects. Associated with energy price increases is usually an increase in their volatility, which typically leads households and businesses to postpone purchases of durables and investment goods, respectively. Finally, there is something economists refer to as resource temporary unemployment from sectoral shifts: as resources get reallocated from more to less energy-intensive activities over time, a fraction of these resources will go unemployed if there are frictions impeding the reallocation.
Private consumption and private investment—the two main components of GDP—are both underperforming relative to the average recovery. However, investment’s underperformance only started in the fourth quarter of 2010. While there is no way to prove that it was an increase in the cost of energy that derailed investment, oil prices did go up from an average of $76 a barrel in the third quarter of 2010 to an average $86 in the fourth. They have not looked back since and currently stand at $114.
To the extent that energy and commodity price increases are temporary, the drag they put on the recovery will be, too. To know whether the increases are temporary, one needs to investigate why these prices are increasing. In a 2009 paper Lutz Killian proposed modeling the behavior of oil prices as a function of real world economic activity (a measure of oil demand), world oil production (a measure of oil supply), and the past behavior of the average oil price itself. His results allow him to distinguish between changes in the price of oil that come about because of oil supply shocks, demand shocks that are specific to the oil market, and demand shocks that occur because of changes in aggregate economic activity. Although his study only runs until the end of 2007, we have updated his results to the end of 2010 here at the Cleveland Fed and found that the vast majority of the increase that Killian found in his analysis was due to positive innovations in aggregate economic activity. This suggests that unless increases in aggregate demand abate, energy prices will continue to be elevated and will constitute yet another headwind this recovery will have to face.