Tomorrow's hurricane-affected September jobs report will be... confusing. That is the (lack of) consensus from Wall Street analysts, who expect an average print of 80,000 (down from the 3-month average of 185K), however with huge variance on either side, with 4 economists predicting a loss of jobs, three expecting a print higher than 150K and one optimistic forecaster going as high as 260,000.
The amusing breakdown by bank is as follows:
- Hugh Johnson 153k
- Standard Chartered 150k
- UBS 125k
- JPMorgan 100k
- Bank of America 80k
- Citi 70k
- Wells Fargo 55k
- Goldman Sachs 50k
- Deutsche Bank 50k
- SocGen -25k
- Jefferies -45k
Of course, the numbers are so scattered (as to make tomorrow's report meaningless) due to the negative distortions from hurricanes Harvey, Irma, and Maria. The payroll survey is designed to ask employers about their number of employees for the pay period that includes the 12th of the month. Hurricane Irma made landfall just prior to the workweek containing the 12th of September. For workers who are paid on a weekly basis, if they were not able to work that week following Irma’s landfall, they may not be counted in the payroll report.
Still, while jobs may be depressed due to the hurricanes, the impact on average hourly earnings would be the opposite. As Citi's desk notes, the hurricanes themselves may put upward pressure on earnings given overtime paid as companies have a hard time finding workers. The street is fairly split on this fact, with 23 of the 57 economist submissions so far with an estimate of 0.2 or below (2 estimates at 0.1%) and the rest calling for 0.3% or above (4 estimates at 0.4%), although this may understate risk for a positive surprise.
With that in mind, here is a summary of what to expects tomorrow at 8:30am, courtesy of RanSquawk
US NONFARM PAYROLL PREVIEW – SEP 2017, ANALYST FORECASTS (forecast, range, previous):
- Non-farm Payrolls: 80k (-45k to 209k, Prev. 156k)
- Unemployment Rate: 4.4% (4.3% to 4.6%, Prev. 4.4%)
- Average Earnings Y/Y: 2.5% (2.4% to 2.7%, Prev. 2.5%)
- Average Earnings M/M: 0.3% (0.1% to 0.4%, Prev. 0.1%)
- Average Work Week Hours: 34.4 hrs (34.2 to 34.6 hrs, Prev. 34.4 hrs)
- Private Payrolls: 83k (-25k to 199k Prev. 165k)
- Manufacturing Payrolls: 10k (-20k to 40k, Prev. 36k)
- Government Payrolls: No forecasts (Prev. -9k)
- U6 Unemployment Rate: No forecasts (Prev. 8.6%)
- Labour Force Participation: No forecasts (Prev. 62.9%)
Headline non-farm payrolls have averaged 176k in the first eight months of 2017, slightly lower than the 187k 2017 average. But the trend rate of payroll growth has picked up in recent months; on a three-month rolling average basis, payroll growth averaged 185k in August, which is roughly in-line with the trend over the last three-months, and above the 175k twelve-month rolling average. Over the last five years, payrolls in September has averaged 204k, though it is worth noting that this month’s data will be distorted by the impact of Hurricanes Harvey, Irma, and Maria.
“Expectations are heavily discounted due to the recent hurricanes,” write analysts at Lloyds. “The natural disaster is going to leave the data unclean for a while and the release would have to be significantly weaker than anticipated for the market to start discounting a December rate hike from the Fed.”
Analysts at Capital Economics estimate that the hurricanes may have cut payrolls growth in half. “We already know that Hurricane Harvey caused jobless claims to spike in Texas. The rise in claims was smaller than that seen post-Katrina in 2005, but was larger than the impact from Ike in 2008. The rise in claims in Florida following Hurricane Irma so far has been negligible.”
State-Level Jobless Claims Data Suggests a Larger-than-Normal Payrolls Impact from Hurricanes Harvey and Irma
In 2005, Hurricane Katrina resulted in two months where payroll growth languished beneath 100k (versus the six-month trend of around 250k being added on a monthly basis).
Hurricane Ike is a more troublesome example to look for clues about how payrolls react, given that Lehman Brothers collapsed two days after the hurricane hit, CapEco says.
Looking further back, Capital Economics notes that payroll growth fell by around 100k in the aftermath of Hurricane Andrew in 1992, which was the largest storm on record to hit Florida at the time.
In its latest policy statement, the FOMC noted that “job gains have remained solid in recent months, and the unemployment rate has stayed low”, but “market-based measures of inflation compensation remain low”. Accordingly, despite the expected distortions to this month’s data, analysts will be looking to see if the wage inflation story remains intact, reinforcing the narrative behind a December rate hike.
ING’s analysts believe “a decent wage number should help cement market pricing of a December hike,” and are more optimistic than consensus, forecasting annualised wage growth of 2.6%, arguing that calendar quirks may drive a strong print.
Last month’s data was disappointing, ING says, but can be attributed to a change in the number of work days between the calendar months. The bank notes that where there were two additional workdays in the August, there were two fewer in the September month.
“Of course, these statistical quirks tell us nothing about the economics. The tight labour market and increasing job-to-job flows should drive up the pace of pay rises over coming months. But it is a slow-moving picture and we may struggle to see growth above 3% this year.”
CLAIMS: The most recent data shows initial jobless claims at 268,250 on a four-week moving-average basis, higher than the 250,250 going into last month’s Employment Report. However, it seems that this jump was more mild than analysts had foreseen. “We feared a big spike in claims as a result of Hurricane Maria hitting Puerto Rico, but it didn't happen,” writes Pantheon Macroeconomics. “Given the extent of the devastation and the subsequent drop in economic activity, this is odd.” Pantheon notes that the lingering impact of Harvey and Irma kept claims above the pre-storm trend around 240k, and believes that a reversion to these levels will be seen in a matter of weeks. “We have no reason to think that the underlying labour market has changed much since August,” Pantheon says.
ADP: The ADP’s measure of payroll growth was more-or-less in-line with expectations, showing the addition of 135k payrolls in September. But the tone of the commentary was generally positive; Moody’s chief economist Mark Zandi said “Hurricanes Harvey and Irma hurt the job market in September, but looking through the storms, the job market remains sturdy and strong.”
CHALLENGER JOB CUTS: Challenger reported announced job cuts by US employers fell 4.4% MM, while they lodged a 27% YY fall. Looking at Q3 as a whole, job cuts were down by 6.2% QQ and were 22.5% lower than Q3 2016. The data continues to paint a picture of a healthy labour market. “Job cuts have remained low since the second half of last year. As companies grapple with potential deregulation and changes to health care costs in a tight labour market, employers are holding on to their existing workforces while many positions requiring skilled labour go unfilled,” Challenger noted.
ISM SURVEYS: The employment sub-indices for both the manufacturing and non-manufacturing ISM surveys rose in September by 0.4pts and 0.6pts respectively. That marked the 12th straight month of employment growth in the manufacturing sector, and the 43rd consecutive month of growth in the non-manufacturing sector. And this is similar to the findings from the IHS Markit PMIs; for the manufacturing sector, employment growth was the fastest in nine months, though Markit noted that growth was running hotter than output, which signals that productivity may be slipping. The services PMI, however, saw employment growth slip to a three-month low, though Markit said it was still “solid” in September.
* * *
WHAT BANK DESKS ARE SAYING:
Barclays: We look for nonfarm payrolls to expand by 75k, down from 156k in August. Informing our view are initial and continuing jobless claims, which have risen following the landfall of Hurricanes Harvey and Irma. Offsetting this to some degree are other factors like part-time employment for economic reasons and the employment diffusion index, which have shown improvement in recent months. Elsewhere in the report, we look for private payrolls to rise by 70k. We also expect average hourly earnings to rise by 0.3% m/m and 2.6% y/y, while average weekly hours remain unchanged at 34.4. With the slowdown in employment growth on the month, we expect an unchanged unemployment rate at 4.4%.
Deutsche Bank: we expect only a 50k gain on headline nonfarm payrolls (+50k private). However, this may be enough to keep the unemployment rate steady at 4.4%. In fact, the “weather workers” series within the Household survey should provide a reasonable sense of the magnitude of the hurricane related disruptions to the payroll data.
Goldman Sachs: We estimate that nonfarm payroll growth slowed to +50k in September, below consensus of +80k and the 3- month average pace of +185k. Our forecast reflects the widespread flooding and power outages caused by Hurricanes Harvey and Irma, which affected over 10% of the population and caused over $100bn in damages. The impact on tomorrow’s report is highly uncertain, but our base-case assumes a significant impact of -125k that partially offsets continued job growth in the rest of the country. We also expect the hurricanes to weigh on the household survey results, and given the high unrounded level of last month’s unemployment rate (4.442%), we believe the September jobless rate is more likely to round up than down (we estimate a rise to 4.5%). Finally, we estimate average hourly earnings increased 0.4% month over month and 2.7% year over year, reflecting positive calendar effects.
Citi: Hurricane-related distortions to September payrolls imply market sensitivity to this month’s jobs tally will be lower than usual. Focus will be on wage growth for signs of inflation given last week’s soft PCE deflator print. We expect growth in average hourly earnings of 0.3% MoM (on-consensus) and 2.6% YoY (above-consensus), but see risks of a 0.4% print from a possible hurricane boost. Our forecast is for nonfarm payrolls growth of 70K (consensus: 80K) based on an analysis of initial claims data around past hurricanes. It would likely take a substantially below-consensus print (around flat or negative payrolls) to induce a large market reaction. Instead, the market may be more sensitive to upside surprises as a strong reading would imply either a smaller-than-expected impact from the hurricanes or stronger ex-hurricane job growth. Any hurricane-related softness in the September report will likely subsequently revert and boost the October payrolls report, as activity in the hurricane-affected areas returns to normal. Hurricane-related boosts to average hourly earnings will be temporary. Still, boosts to GDP growth from reconstruction will contribute to further labor market tightening, and with solid underlying growth, should lead to stronger wages over time.
ING: With this month's payrolls number likely to be "written off" given the effect of recent hurricanes, a decent wage number should help cement market pricing of a December hike. But in reality, if the Fed decides not to move at the end of the year, it's unlikely to be because of economics. Instead, the new debt ceiling deadline falls almost right on top of the December meeting and given the divisive nature of US politics, it could go down to the wire. A substantial pick-up in market volatility on the possibility of a government shutdown could conceivably see the Fed delay until 2018 - although that's not our base case.
RBC: The survey period for the September employment report (the week that includes the 12th of the month) began as Hurricane Irma was making landfall in Florida and as Texas was still reeling from the effects of Hurricane Harvey. Thus, we will be shocked if we don’t see a significant slowing in nonfarm payroll growth on the month. Our best guess (given what Texas and Florida have added to payroll growth recently and trends in unemployment insurance claims in these states in the wake of the Hurricanes) is that headline and nonfarm payroll growth will slow to around 50K on the month, against a 6-month trend of around 160K. There is incredible uncertainty in this forecast and this is also reflected in the range of Street estimates (from -25K to +145K, if we throw out the highest and lowest guesstimates). Given the probability we get a very messy report is high, the market is likely to fade any weakness.
SocGen: Hurricane Harvey may have led to a decline in nonfarm payrolls in September, which would mark the first negative reading in seven years. Quantifying the hurricane’s impact on job growth is fraught with uncertainty, but we suspect that Harvey’s impact was similar to the drag on payrolls seen in the wake of Hurricane Katrina in 2005.