January 08, 2005
A Different Take On The Jobless Recovery
That said, the rate of job growth continues to lag other recoveries. Over the year—December 2003 to December 2004—employment is up 2.2 million. While this is the best year for job growth since 1999, it still lags the 3.3 million jobs added at the analogous stage of the last recovery. On average over the past year, monthly employment grew by 186,000, compared to 272,000 at a similar stage of the 1990s recovery.
In fact, the average annual growth rate for payroll employment at this stage of a recovery is 2.8%, well above the 1.7% achieved this year. Had that average growth occurred this year, we would have added 3.6 million jobs since last December, 1.4 million more than the actual number. Thus, in historical context, the recent pace of job growth has been subpar.
Hard to argue with the facts, but the EPI then goes on to suggest that the historically low pace of job growth represents a significant failure of public (i.e. Bush) policy.
With the newly released payroll employment data for December 2004 it is now possible to assess whether the administration's tax cut strategy produced the employment growth that was projected (see table and figure below). The final verdict is grim. Job growth over the last 18 months has fallen short by 1,703,000—more than one-third less than the number of jobs the administration said would be created without the tax cuts. Given that the economy failed to produce the number of jobs expected with no policy change, it seems hard to argue that the tax cuts were a successful strategy in adding any jobs—the promised 1.4 million additional jobs never materialized. The announced revisions (up 236,000 in March 2004) to the payroll employment series (see Data Note below) do not materially change this assessment.
Without implying any particular position on what the effects of the administration's tax policies might, or might not, have been, the counterfactual experiment implied by the EPI's statement is, shall we say, heroic. Their interpretation clearly requires that tax policy has been the dominant difference between the recovery out of the 2001 recession and the early phases of previous U.S. expansions.
I'm not convinced. I too, of course, have been watching the pace of net job creation over the past several years, and scratching my head. But recently I've come to believe that the sluggish employment of this recovery may not be as anomalous as we have been thinking.
Here's a couple of observations that I find interesting. First, the slow jobs growth out of the recession is just a piece of the broader observation that economic activity has, in general, been less robust that historical experience would predict: Real (inflation-adjusted) GDP growth, though not terrible, has also been slower than the average for the first several years following post WWII recessions in the U.S. (at least if we focus on the period prior to the 1990-91 downturn).
And then there is picture.
The shaded bars are NBER recessions, and the arrows represent episodes when the relative price of energy (measured by the CPI) spiked by at least 10%. Although the timing, magnitude, and other details vary, the correlation of energy price jumps and economic contractions is striking. No recession after 1970 lacks an energy shock, no energy shock -- up to the most recent one -- lacks a recession.
A closer gander at the recent shock is instructive.
It doesn't take too much imagination to conjure up the following narrative: The surprisingly weak performance of the economy in 2002 -- real GDP growth for the year was only about 2-1/2 percent -- was followed by an equally disappointing first quarter of 2003. But maybe we shouldn't be surprised. Relative energy prices (at the consumer level) rose by 24 percent over that period.
Then things got better. Although still volatile, the average level of energy prices stabilized over the balance of the year. At the same time, the pace of the expansion improved dramatically. GDP growth was just over 4 percent in three of the next four quarters. The exception was third-quarter 2003 -- GDP grew at over 7 percent (annualized) during those three months.
Employment finally began the much anticipated turnaround in the first half of 2004: On net, the economy created just over 200,000 jobs per month during that period. Unfortunately, another sharp rise in energy prices had commenced -- the relative price rose by 15 percent through November -- and the momentum was not sustained in the second half of 2004.
Big surprise? Not obviously. It is very, very tempting to conclude that the struggles of the U.S. economy over the past three years amount to little more than the effects of good old-fashioned energy shocks, shocks of a magnitude that, in every other instance over the past thirty years, have been associated with recessions.
The case is not without its flaws, but the energy angle strikes me as a much stronger contender than the Bush tax cuts.
UPDATE: A clarification on this statement above: "Employment finally began the much anticipated turnaround in the first half of 2004: On net, the economy created just over 200,000 jobs per month during that period." Actually that 200,000 jobs per month figure applies to the second quarter. The figure was about 150,000 per month over the entire first half of 2004.
TrackBack URL for this entry:
Listed below are links to blogs that reference A Different Take On The Jobless Recovery: