Last week in the column "GDP Puzzles Even the Best Economists," I discussed the issue of "residual seasonality," which may be exacerbating a pattern of lower first-quarter GDP increasingly seen over the past 30 years, and especially in the past 15-, 10- and five-year periods.
I'm going to address this issue further here because it's quickly becoming very important to the capital markets, the timing of Fed rate hikes, and will be critical in determining what the Bureau of Economic Analysis (BEA) announces for second-quarter GDP on July 30.
There is nothing new about first-quarter GDP being lower than the other three quarters. However, it didn't start to become a concern for monetary policymakers until 2014 when first-quarter GDP was reported as negative.
The repeat of that this year, with current expectations for revised GDP figures to show a contraction in Q1 and the increasing potential for the same in Q2, presents the Federal Open Market Committee (FOMC) with a substantial problem. This would almost certainly require The National Bureau of Economic Research (NBER) to determine that the U.S, economy began a recession during the first half of 2015, and maybe earlier, just as the Fed is attempting to begin the process of raising interest rates.
A May 14 research report by the Federal Reserve Bank of Philadelphia, "First Quarters in the National Income and Product Accounts," discussed what appears to be a growing seasonal pattern of Q1 GDP underperformance that is not accounted for by the BEA and which may indicate a residual seasonality issue.
The report received almost no media coverage.
On May 18, four days later, the Federal Reserve Bank of San Francisco released an Economic Letter, "The Puzzle of Weak First-Quarter GDP Growth," which quantified residual seasonality to produce an annualized rate of GDP growth for Q1 2015 of 1.8% vs. the 0.2% announced by the BEA.
The letter received a lot of media attention.
On May 22, the BEA announced it would apply a residual seasonality adjustment to the GDP figures for each quarter of 2012, 2013, and 2014 when it releases the advance estimate of second-quarter 2015 GDP on July 30. What was missing from that announcement is whether or not the BEA will apply those adjustments to the GDP estimate for Q1 2015 and the advanced estimate for Q2 2015. (I will post to the comments section of this column and to columnist conversations if and when the BEA clarifies this.)
The second and third estimates of GDP for Q1 are currently scheduled for release by the BEA on May 29 and June 24, respectively. Right now, those estimates are expected to show that GDP contracted in Q1. If the BEA revises those estimates on July 30 for residual seasonality, it is probable that Q1 will end up in positive territory.
The importance of this is that now that the BEA has announced its intentions to adjust data, if it doesn't do so for the second and third estimates of Q1 GDP, those estimates are now irrelevant, even if they show, as expected, that GDP contracted in Q1 and may do so in Q2.
There is another very important issue with how the BEA may adjust the data, however. The Philadelphia Fed paper concluded that the residual seasonality is in the production component of GDP, not in income calculations. But the BEA announced its intention to adjust personal consumption expenditures (PCE) in the application of residual seasonality.
The implication is that the BEA is using the residual seasonality issue as blanket justification for adjusting inputs to GDP that the Philadelphia Fed has concluded are not wrong and shouldn't be adjusted.
The bottom line is that the implication of this issue, in its totality, appears to be an almost complete and abrupt reversal of the operating mandate of transparency at the Fed and within the BEA, and a move toward a more opaque reporting system for government-supplied economic data.
How the capital markets will respond to this is unclear and I will address the potential scenarios tomorrow.