As expected, GDP growth became even weaker in 2006:Q3.
The Bureau of Economic Analysis announced today that U.S. real GDP grew at an annual rate of 1.6% in the third quarter of 2006, down from an already modest 2.6% for 2006:Q2. The biggest change between the two quarters came from a resurgence of imports, which had only subtracted 0.2% from the 2006:Q2 GDP annual growth rate but contributed -1.3% to the 2006:Q3 figure. The other big story is of course housing, which is now judged to have contributed -0.7% to the 2006:Q2 GDP annual growth rate and -1.1% to 2006:Q3. Nonresidential fixed investment and exports were the two bright spots, both of which improved a bit from 2006:Q2.
Looking at GDP growth over each of the last four quarters, one performance (2006:Q1) was outstanding and the other three a bit sickly. Added together, they leave the level of chain-weighted real GDP for 2006:Q3 2.9% above the value of 2005:Q3– a bit slower growth than normal, but better than I was anticipating last year at this time and better than I’m expecting growth to come in over the next four quarters.
The weak performance over the last two quarters bumped the recession probability index up from 3.4% for 2006:Q1 to 9.3% for 2006:Q2. That’s the highest it has been for the last three years, though still at a level signaling clearly that a recession had not yet begun as of the second quarter of this year. This index is not a forecast of where the economy will be later this year, but is a backward-looking assessment of where the economy was as of 2006:Q2, using the latest data to form that assessment. It is intended as an alternative to the announcements made by the National Bureau of Economic Research, which, although highly authoritative and reliable, often fail to be made public until years after a recession has started. Background on how the recession probability index is constructed and a review of its historical performance are available here.
Technorati Tags: GDP,
recession probability,
recession probability index
Any comment on the possible misrecording of auto production that led to even this miserly figure being inflated?
Thanks for that very interesting link, Emmanuel. The claim that GDP 2006:Q3 has been overstated for this reason sounds plausible to me. These advance estimates can be revised for a number of other reasons as well.
Does your Recession Probability Index for Q2 only use data as of Q2, or does it use data also from Q3 to judge whether we were in a recession as of Q2? Is that why it is always behind? Or do you have to wait this long for some data about Q2 to be made available.
Hal, the index most definitely makes use of the 2006:Q3 data to form the inference for 2006:Q2. The reasons for waiting are (1) it really helps to see the 2006:Q3 value to be sure of what to make of 2006:Q2, and (2) the data are prone to enough revision that trying to make an inference about 2006:Q3 from the “advance”, “preliminary”, or even “final” 2006:Q3 numbers is prone to an unacceptable degree of error.
Come on. What’s so different about auto production from previous years, other than the decline in some brand production?
Where were the complaints about this issue last year?