A bizarre series of events has caused havoc for recent economic data and a headache for forecasters

Advertisement
A bizarre series of events has caused havoc for recent economic data and a headache for forecasters

trader

Advertisement
  • Forecasts for the number of jobs created each month haven't been this off in more than two and a half years, a Business Insider analysis found.
  • The government shutdown and weather appear to have muddled the outlook.
  • But economists also acknowledge there are risks ahead.

So, uh, about the February jobs report.

And the January one. December, too.

Forecasts for the number of jobs created each month haven't been this off in more than two and a half years, a Business Insider analysis found. Here are the number of nonfarm payrolls economists surveyed by Bloomberg expected and what was reported by the Bureau of Labor Statistics over the past three months:

  • February: 180,000 estimate versus 20,000 reported
  • January: 165,000 estimate versus 311,000 reported
  • December: 183,500 estimate versus 227,000 reported

The aggregate three-month spread between the survey consensus and actual number of nonfarm payrolls added was 349,500 through February, the largest it's been since July 2016.

Advertisement

NFP history vs. survey

Economists partly blamed discrepancies on the partial government shutdown that ended in January, which lasted 35 days to make it the longest in history. Nine of 15 cabinet-level departments were shuttered, delaying data collection and economic releases.

Some of the 800,000 workers who were furloughed or forced to work unpaid during that period took up second jobs, raising the possibility of double counting in the payroll survey. The number of workers employed part-time due to economic reasons fell sharply in February, after jumping by more than half a million the month before.

"This is a confirmation that the government shutdown was a big deal," said Josh Wright, chief economist at iCIMS, a recruiting-software company. "This is why it was such a big concern in the business community. It does lead to this kind of volatility, this start and stop."

NFP 3 month trailing

Advertisement

Storms cause trouble

"Forecasting monthly employment is a humbling experience, particularly early in the year when weather can play a significant role," said Ryan Sweet, an economist at Moody's Analytics. "There were a number of moving parts that are difficult to quantify over the past few months."

Sweet pointed out that there was a nationwide winter storm during the jobs report survey week, which occurs during the week that contains the 12th day of the month. In fact, the February storm affected more than 100 million people across the US.

To Sweet's point, some of the industries most sensitive to weather changes showed particular weakness in February's report. For instance, the construction sector lost 31,000 jobs for the month.

Adding weather weirdness to the distortions caused by the shutdown, it should be no surprise that forecasters were so far off the mark.

Not a worry, yet

This is far from the first time there has been noise in the month-to-month employment numbers. But against a backdrop of surprisingly soft economic data, some acknowledge that it could portend a sooner-than-expected slowdown.

Advertisement

Weak reports for other key economic indicators like retail sales have raised concerns about a possible soft patch. And GDP models show that first-quarter economic growth could be anemic.

But some economists say it's too early to tell if distortions caused by the shutdown are simply causing the appearance of a slowdown or if there is an actual downward trend forming.

"If we had 3 months in a row like this, everyone would conclude it just meant that people are terrible at predicting the start of a recession," Austan Goolsbee, who was chairman of the Council of Economic Advisers in the Obama administration, said in an email. "If we have a few months of big positive numbers then people will conclude it was the unexpected impact of the shutdown."

{{}}