One week ago, the BEA admitted that it had “found a problem” when it comes to calculating GDP numbers. Specifically it blamed “residual seasonality” adjustments for giving historical GDP numbers a persistent optimistic bias. This came in the aftermath of last week’s shocking Q2 GDP report which printed at 1.2%, less than half of Wall Street’s consensus.
Today, seasonality made another appearance, this time however in the much anticipated July jobs number, which unlike the woeful Q2 GDP number, was the opposite, coming in far higher than expected. In fact it was higher than the top Wall Street estimate.
And, just like in the case of GDP, it appears that seasonal adjustments were the culprit for today’s blowout headline print which excluding the Arima X 13 contribution to the headline number, would have been notably weaker.
As Mitsubishi UFJ strategist John Herrmann wrote in a note shortly after the report, the “jobs headline overstates” strength of payrolls. He adds that the unadjusted data show a “middling report” that’s “nowhere as strong as the headline” and adds that private payrolls unadjusted +85k in July vs seasonally adjusted +217k.
In Herrmann’s view, the government applied a “very benign seasonal adjustment factor upon private payrolls to transform a soft private payroll gain into a strong gain.”
He did not provide a reason why the government would do that.
Courtesy of Southbay Research, which also blasted today’s seasonal adjustment factor, this is how the seasonal adjustments look like relative to history.
We leave it up to readers to decide just why the government may want to represent what would otherwise have been a far weaker than expected report, into a blowout number, one which merely adds to the economic “recovery” narrative, which incidentally will come in very useful to Hillary’s presidential campaign.
Yet even assuming the market has no doubts about the seasonally adjusted headline number, as appears to be the case, the other problem that has emerged for the Fed is how to ignore this strong number. As Bank of Tokyo’s Chris Rupkey writes, “Let’s see Yellen get out of this one and find something in the data to once again not raise rates in September.” (We assume he did not see the unadujsted numbers.)
As he adds, slowing 2Q GDP growth of 1.2% took Sept. rate hike “off the table” and now “the million dollar question” is whether 255k payroll jobs in July, 292k in June put it back on. As a reminder, Yellen speaks exactly in three weeks time at Jackson Hole on Aug. 26; “let’s see if she provides some guidance.” But while rate hike odds may have spiked after today’s report, it is almost certain that, as we said last night, the Fed will not dare to hike the rate in September and potentially unleash market turmoil in the most sensitive part of the presidential race.
As for a December rate hike, there are 4 months until then, and much can happen: who knows, maybe the BLS will even undo the significant seasonal adjustment boost that send July jobs soaring.
* * *