Recession Watch: February real GDP falls sharply from January peak but Q1 still on track to be positive

Monthly real GDP, one of the five primary economic indicators that the NBER Business Cycle Dating Committee uses to judge recession start and end dates, fell very sharply in February from the January peak. The government does not publish GDP data at a monthly level, but the NBER BCDC refers to sources such as Macroeconomic Advisers (MA) and their MGDP data series. This was a significant decline, and in fact below the August level, but a one-month decline is not sufficient to "call" a recession. As as Macroeconomic Advisers puts it:

Monthly GDP declined 1.2% in February.  This was the second largest one-month decline in the nearly 16-year history of the index; it declined 1.6% in September 2001.  The sharp drop in monthly GDP in February was led by a sharp decline in nonfarm inventory investment.  Also subtracting from monthly GDP were declines in net exports, capital goods, and construction.  Averaged over January and February, monthly GDP was 0.8% above the fourth-quarter average at an annual rate.  Our latest tracking forecast of a 0.4% increase of GDP in the first quarter assumes a 0.3% increase in monthly GDP in March.

If the NBER BCDC is the definitive expert on marking of recessions, MA is the definitive expert on measuring real GDP at the monthly level with their MGDP data series. So, for now, I will accept their forecast that in a month we will see a rebound in monthly GDP for March so that Q1 GDP is likely to come in with a modest +0.4% gain. Such a gain would mean that a recession did not start in Q1 using the two-negative-quarter rule, but the NBER BCDC does not use the two-negative-quarter rule. The NBER BCDC measures from the peak month, so it could take a few months or more for monthly GDP to climb back up to the January peak level.

In short, February was a bad month and suggested a plunge into a recession, but one bad month does not constitute a recession. Note that even is March and April bounce back somewhat, that may still represent a decline from the peak in January. The strongest statement I can make is that we are almost there, but not yet clearly there.

-- Jack Krupansky