GDP Data Tells You Nothing About Recession Risk
On the verge of the 2001 recession, in February 2001, the real-time data showed GDP growth declining to 1.1% (green line). But with the Fed cutting rates, the S&P 500 saw a one-and-a-half-month 19% spurt in April-May, even as the recession tightened its grip.
And in November 2007, on the cusp of the Great Recession, real-time data showed GDP growth surging to 4.9% (red line). Only in the years following the recession did revisions cut it down to less than half that initial reading.
The point is that, at least in real time, the strength of GDP growth does not tell us whether a recession is about to hit. It is only long after the fact, following repeated revisions, that the GDP data becomes more informative about the timing of the recession.
While the current cycle’s recent 2.6% GDP growth print – tracking right in-between the 2001 and 2007 trajectories – was well received by many economic prognosticators (blue line), by no means does it rule out a looming recession.
In principle, GDP is a coincident indicator of the economy, with no real predictive value. But because more than half of the initial GDP estimate is based on survey data and the extrapolation of recent trends, the initial vintages of GDP are often misleading, especially around business cycle turning points.
This helps explain why it is often only well after a recession has begun that revisions to GDP data show an economic contraction in progress. Until then, the consensus may be wrongly persuaded that the coast is clear.
This is a key reason why good leading indexes are so valuable. Unlike GDP, ECRI’s leading indexes avoid major revisions over time, most crucially with regard to their cyclical timing and directional calls. Consequently, we were able to call the 2001 recession and Great Recession on a timely basis.
* * *