Well, coming into today's payrolls report we said that the number would be a beat and the only question was how big, as a result of millions of Americans seeing their emergency benefits expiring. Boy were we wrong: moments ago the BLS reported that with expectations of a 500K print and whisper numbers sharply higher, in September the US added just 194K jobs (in fact, the number came below the lowest of all but one of the 71 economist forecasts, with just Banque Pictet's Thomas Costberg forecasting a 0 print). Sept payrolls were down by nearly half from the upward revised 366K in August, the first back-to-back monthly drop in payrolls this year and the lowest print of 2021 (even as then umber of Household survey showed an increase of 526K jobs in the month!), likely impairing the Fed's tapering schedule as this number was so bad as to pass as a "major shock" from the Fed's perspective, even if as the BLS itself admits a big reason for the drop was a seasonal adjustment in local government education, which may have pulled the number down by some 150K.
Putting the woeful September number in context, thus far this year, monthly job growth has averaged 561,000, roughly half of what Fed officials promised at the start of the year. Nonfarm employment has increased by 17.4 million since a recent trough in April 2020 but is down by 5.0 million, or 3.3 percent, from its pre-pandemic level in February 2020. In September, notable job gains occurred in leisure and hospitality, in professional and business services, in retail trade, and in transportation and warehousing. Employment in public education plunged over the month (more below).
Despite the dismal headline print, there were several mitigating factors:
- First, the number of employed workers tracked by the Household Survey actually rose by 526K, sharply better than the headline Establishment Survey print.
- Second, both prior months were revised higher, with the change in employment for July revised up by 38,000, from +1,053,000 to +1,091,000, and the change for August was revised up by 131,000, from +235,000 to +366,000.
- A third mitigating factor to the horrendous headline number is that the miss seems like it’s mostly in the government sector as opposed to the private sector: private payrolls added 317K, missing the expectation of 450K far less than the headline print. Meanwhile, government workers tumbled by a whopping 123K, led by the loss of 144K government education jobs. As shown below, this was the biggest drop in government workers since 2020.
- Commenting on the plunge in local government teacher, the BLS said that "hiring this September was lower than usual, resulting in a decline after seasonal adjustment. Recent employment changes are challenging to interpret, as pandemic- related staffing fluctuations in public and private education have distorted the normal seasonal hiring and layoff patterns. Since February 2020, employment is down by 310,000 in local government education, by 194,000 in state government education, and by 172,000 in private education."
- A fourth silver lining: the unemployment rate actually plunged to just 4.8%, from 5.2% and below the 5.1% expectation, dropping for all races as the number of Unemployed workers plunged by 710K from 8.384MM to 7.674MM.
- Curiously, even as the unemployment rate dropped, the participation rate also declined from 61.7% to 61.6% as the civilian labor force somehow shrank in September by 183K to 161.354 million. Adding to the confusion, the employment-population ratio rose to 58.7% from 58.5%, if still a far cry from the pre-pandemic 61.1%.
- A fifth positive offset: the average hourly earnings actually ticked up again, rising from 0.4% to 0.6%, beating the consensus of 0.4%. On a Y/Y basis, earnings rose 4.6%, in line with expectations.
The rise in hourly earnings indicates that wage growth remains solid despite some expectations that as millions of Americans move away from emergency benefits they will flood the labor force with new supply. Clearly that has not happened yet.
Some more details from the report: workers unable to work due to bad weather was 94K, slightly higher than the historical average for Sept. which is 83k. Tied to this, 404k workers who usually work full-time could only work part-time due to the weather last month.
Looking at the youth component, some more ominous trends emerge:
- The underemployment rate for young adults aged 16 to 24 in Sept. is 14.2%, according to Bloomberg calculations of Bureau of Labor Statistics data.
- Total unemployed aged 16-24 at 1,772,000
- Marginally attached workers aged 16-24 at 450,000
- Employed part-time for economic reasons aged 16-24 at 717,000
- Civilian labor force aged 16-24 at 20,318,000
Looking at major industries, employment in leisure and hospitality increased by 74,000 in September, with continued job growth in arts, entertainment, and recreation (+43,000). Employment in food services and drinking places changed little for the second consecutive month, compared with an average monthly gain of 197,000 from January through July. Employment in leisure and hospitality is down by 1.6 million, or 9.4 percent, since February 2020.
- Professional and business services added 60,000 jobs in September. Employment continued to increase in architectural and engineering services (+15,000), management and technical consulting services (+15,000), and computer systems design and related services (+9,000). Employment in professional and business services is 385,000 below its level in February 2020.
- Employment in retail trade rose by 56,000, following 2 months of little change. Over the month, employment gains occurred in clothing and clothing accessories stores (+27,000), general merchandise stores (+16,000), and building material and garden supply stores (+16,000). These gains were partially offset by a loss in food and beverage stores (-12,000). Retail trade employment is 202,000 lower than its level in February 2020.
- Employment in transportation and warehousing increased by 47,000 in September, in line with gains in the prior 2 months. In September, job gains continued in warehousing and storage (+16,000), couriers and messengers (+13,000), and air transportation (+10,000). Employment in transportation and warehousing is 72,000 above its pre-pandemic level in February 2020.
- Employment in the information industry increased by 32,000 in September. Gains occurred in motion picture and sound recording industries (+14,000); in publishing industries, except Internet (+11,000); and in data processing, hosting, and related services (+6,000). Employment in information is down by 108,000 since February 2020.
- In September, social assistance added 30,000 jobs, led by a gain in child day care services (+18,000). Employment in social assistance is 204,000 lower than in February 2020.
- Employment in manufacturing increased by 26,000 in September, with gains in fabricated metal products (+8,000), machinery (+6,000), and printing and related support activities (+4,000). These gains were partially offset by a decline of 6,000 in motor vehicles and parts. Manufacturing employment is down by 353,000 since February 2020.
- Construction employment rose by 22,000 in September but has shown little net change thus far this year. Employment in construction is 201,000 below its February 2020 level.
- In September, employment in wholesale trade increased by 17,000, almost entirely in the durable goods component (+16,000). Employment in wholesale trade is down by 159,000 since February 2020.
- Mining employment continued to trend up in September (+5,000), reflecting growth in support activities for mining (+4,000). Mining employment has risen by 59,000 since a trough in August 2020 but is 93,000 below a peak in January 2019.
- Employment in local government education decreased by 144,000 and by 17,000 in state government education. Employment changed little in private education (-19,000). Most back-to-school hiring typically occurs in September. Hiring this September was lower than usual, resulting in a decline after seasonal adjustment. Recent employment changes are challenging to interpret, as pandemic- related staffing fluctuations in public and private education have distorted the normal seasonal hiring and layoff patterns. Since February 2020, employment is down by 310,000 in local government education, by 194,000 in state government education, and by 172,000 in private education.
- Employment in health care changed little in September (-18,000). Job losses occurred in nursing and residential care facilities (-38,000) and hospitals (-8,000), while ambulatory health care services added jobs (+28,000). Employment in health care is down by 524,000 since February 2020, with nursing and residential care facilities accounting for about four-fifths of the loss.
Some other notes: the leisure and hospitality sector added 38,000 jobs in August instead of the net zero gain that was first reported for that sector last month, after revisions. And then another 74,000 in September. While still below the run rate of the previous several months, the number is not as bad as previously feared.
Commenting on the report, Blackrock PM Jeffrey Rosenberg said that “It certainly has a delta-variant characteristic to it, just like we had last month.The key disappointing industry sectors that point to the delta variant are leisure and hospitality and the education numbers....Delayed school reopenings relative to normal seasonal factors."
So the big question is whether the Fed will delay tapering after this report, and here analysts are confident that nothing much changes: Priya Misra, global head of rates strategies at TD Securities, was saying on Bloomberg TV before the data came out that there was a “really low bar” for the Fed to proceed with tapering. “Anything above zero, the Fed’s going to taper.” Although with the number not too far off from zero, she may reasses.
Bloomberg's Ira Jersey agrees, writing that “the payrolls miss (even after revisions) may mean Treasury yields will not break some near-term technical levels and may re-test the 1.47% area. We don’t think this changes the Fed’s path to taper next month, but pricing for hikes could get pushed further into 2023.”
BMO's Ian Lyngen is also in the no change to the taper camp, writing that “this should be more than sufficient to keep tapering on schedule for the November announcement and we’ll look to the incoming Fedspeak to reinforce this notion. The decline of labor participation rate suggests that as unemployment benefits expired, some workers fell out of the labor participation category. Average Hourly Earnings were better than expected at +0.6% vs. +0.4% consensus -- wage inflation only adds to the Fed’s argument for tapering. Overall, a mixed report that does little to shift the macro narrative."
Glenmede CIO Jason Pride echoed the consensus view, saying that “the result, in aggregate, is likely not enough to knock the Fed off its path to begin the tapering process of its balance sheet later this year. The Fed has already outwardly stated that they believe the employment test for tapering has largely already been met, and today’s report may not meaningfully change that.”
Perhaps the most balanced take came from Cornerstone's Robert Perli who wrote that “This is a difficult employment report from the point of view of the Fed. Bottom line, I think the Fed would think it prudent to keep tapering on schedule and start the process next month. But it will probably add caveats to the language to leave itself room to stop tapering if things get worse.”
That said, as Bloomberg notes, the number "is a real conundrum for the Fed. Not at all what they’d like to see. Inflation pressures remaining acute, but job growth weakening."
In other words: does the Fed dare tighten into a stagflation?