What to Watch on Jobs Day: Furloughs and month-to-month volatility

The government shutdown adds several layers of complexity to interpreting January employment figures set for release on Friday. While I still plan to look at wage growth and the prime-age employment-to-population ratio to measure the slack that remains in the labor market, the effect of the extended partial government shutdown on the top-line employment numbers will get a lot of attention too. In this post, I will explore how the shutdown likely will and will not affect the numbers, along with a reminder about typical month-to-month volatility.

The reference period for the upcoming Current Employment Statistics (CES) and the Current Population Survey (CPS) is the pay period or week, respectively, including January 12, 2019. The 12th of January marks roughly the middle of the 35-day partial government shutdown. The shutdown directly affected three types of workers and their paychecks: government employees working without pay, government employees not working, and government contractors not working. Indirectly, the shutdown also affected workers who typically service government employees who were not working, but who are not directly paid for by government contracts—think restaurants, taxi cabs, and other services purchased by the government workforce.

Let’s start with the CES. Government workers, furloughed or not, will be counted as employed in the CES. As soon as the Government Fair Treatment Act of 2019 was signed into law, those furloughed workers were guaranteed their back pay and therefore count as employed workers. In other words, the effects of the furlough will not affect the count of federal jobs in the CES. The loss of private sector work, on the other hand, either as direct federal contractors or indirect employment that services government work, could register as a fall in private-sector employment or private-sector hours, or both.

The CPS will be counting working non-paid federal employees as working and non-working furloughed federal employees as being on temporary layoff, thus counted among the unemployed. Anyone else not working because of the lapse in appropriation will also be counted as on temporarily layoff. Those with reductions in hours will be counted as part-time for economic reasons.

On net, the partial government shutdown could register as a mild drag on private payroll employment growth as well as a slight increase in the unemployment rate and the number of involuntary part-time workers. In the Washington, DC area, the drop in economic activity was palpable, but the extent to which the shutdown moved any of the national numbers is probably limited.

There is also a fair amount of month-to-month volatility in all the data series released in the BLS employment report, even without any shutdown-type disruptions. In December, payroll employment grew 312,000, a full 100,000 jobs faster than the average of the previous 11 months in 2018. Every month, payroll employment gets revised and December may see relatively large revisions. In the monthly report, the BLS reports an approximate confidence interval around the monthly changes in CES employment on the order of plus or minus 115,000. This means the point estimate of 312,000 for December has a 90 percent confidence interval of 197,000 to 427,000—a large range. A similar confidence interval can be drawn around the point estimate for the unemployment rate—a range of plus or minus 0.2 percentage points.

This is all to say that even if payroll employment comes in low for January, this may simply be because of a reversion to the trend from the unusually large number last month and may not necessarily be indicative of huge labor market effects from the partial government shutdown. However, even if the government shutdown does not move the national employment statistics, it has had real and sometimes devastating consequences for furloughed employees, federal contractors, and other related workers, particularly those living paycheck to paycheck.