US Job Growth Surges on the Strongest Seasonal Adjustment Effect on Record

• US payrolls smashed consensus including positive revisions

• Seasonal adjustment factors played a major role and should be faded…

• ...which counsels smoothing monthly job gains over time

• Wage growth is hot and needs sustained productivity gains

• Markets largely wiped out pricing for another 50bps cut...

• ...as the FOMC may well have committed policy error by going big out of the gates

• Skip November?

• Markets also trimmed BoC cut pricing

• All the negativity pushing big cuts ignores the positives of a strong US economy

• US nonfarm payrolls, m/m 000s // UR %, September:
  
• Actual: 254 / 4.1
• Scotia: 165 / 4.2
• Consensus: 150 / 4.2
• Prior: 159 / 4.2 (revised from 142 / 4.2)

US job growth surged ahead and crushed market pricing for the Federal Reserve to deliver another outsized cut in keeping with our forecast for two quarter-point cuts in November and December. Most details were strong, but my argument on how seasonal adjustment factor changes would drive a good report worked (here and here) almost too well for comfort. Clients are advised to smooth what’s going on with US job growth figures over the months and to look at the year in totality while fading bouts of excess pessimism and excess optimism.

Nonfarm payrolls powered forward by posting a quarter-million gain (+254k) alongside a net positive revision of 72k extra jobs over the prior two months. When the September beat is combined with the revisions it translates into more than double the number of extra jobs in the US economy than consensus expected.

MARKET REACTION FADES 50-POINT MOVES

Stocks loved solid job growth more than dampened rate cut expectations as the S&P is up ½% today. The US 2-year Treasury yield climbed 12bps in the aftermath. The 10-year yield jumped 8bps higher to 3.95%. If this keeps up I’m going to have to raise the 4% 10s target for next year. Cumulative pricing for rate cuts by year-end was shaved by 8bps to about 58bps in fed funds futures which mostly takes out market pricing for one of them being a 50 move. The dollar was broadly stronger, but an offsetting influence in FX land came via repriced expectations for foreign central banks. For example, market pricing for cumulative rate cuts by the BoC over the remaining two meetings this year was shaved by about 7bps to 63bps that will be further informed by next Friday’s Canadian jobs report.

SMOOTH THOSE WONKY SEASONAL ADJUSTMENTS

September 2024 had the biggest seasonal adjustment factor on record compared to like months of September in history (chart 1). I had expected a strong seasonal adjustment factor to be a turning point in favour of a strong gain in jobs, but not this strong! 

Chart 2 shows why this mattered. The 460k gain in seasonally unadjusted nonfarm payrolls last month (red bar) translated into a seasonally adjusted gain of 254k (dark blue bar) because of the unusually strong SA factor. If, instead, the SA factor had been one of the slightly weaker ones then the light blue bars show what would have happened to the reported gain in seasonally adjusted payrolls.

We’re in a different era for how the BLS seasonally adjusts jobs data as evidenced by the fact that the biggest September SA factors in history have all been in recent years. I’m not entirely sure why. They are mum on the topic. It could be because of the unusual timing of the pandemic shock, followed by subsequent loosening and tightening of restrictions and then the timing of the rebound. That doesn’t explain why even SA factors just before the pandemic were higher than prior ones.

Regardless, this is a marked turning point compared to the summer months when seasonally unadjusted payrolls were being haircut by unusually low seasonal adjustment factors compared to like months of August in history, like months of July, and so forth.

What this means is that we should be looking at reported job growth and smoothing the figures over time. The start of the year wasn’t as strong as the 310k gain in March indicated because SA factors back then were too liberal, and the summertime reports were stronger than they appeared because the SA factors invoked too much of a haircut to job growth. Now we are quite possible back to overstating US job growth. On average, the US economy has created 200k jobs per month in the first nine months of 2024 and that remains strong. It also remains around popular estimates of the breakeven rate for payrolls growth that would be compatible with a reasonably stable unemployment rate, assuming that the post-election outcome doesn’t invoke a draconian retrenchment in terms of immigration flows.

Lastly (on seasonal adjustments) we might expect the same thing to play out next month (chart 3).

SOLID BREADTH

Chart 4 demonstrates that there was significant breadth to the gain in payrolls. The private sector added 223k jobs with government adding 31k and almost all at the state and local levels.

The goods sector was up 21k but services led the way with a gain of 202k. Construction added 25k with manufacturing little changed (-7k). Within services, the biggest gainers were education/health (+81k), leisure/hospitality (+78k), business services (+17k) and retail (+16k).

WAGES RIPPED

Wage growth skyrocketed (chart 5). Average hourly wages were up 0.4% m/m SA in September and revised up a tick to +0.5% m/m in August. That means +4.5% m/m SAAR wage growth in September and 5.6% in August for a three-month moving average gain of 4.3%.

Americans had better secure ongoing gains in productivity to justify such wage gains that are in excess of inflation. I have my doubts.

A LOWER UNEMPLOYMENT RATE

The unemployment rate edged down a tick to 4.1%. That’s because it is drawn from the companion household survey that registered a job gain of 430,000 that exceeded the 169k expansion of the labour pool.

This UR is roughly in line with estimates of the NAIRU equilibrium rate of unemployment for the US.

Be careful with these numbers as the household survey has even more statistical noise than the payrolls report. The 90% confidence band around monthly changes in jobs reported by the household survey is +/-600k which means that 90 times out of 100 in repeated household sampling all we can say is that job growth was somewhere between a loss of 170k and a gain of over one million. Righto.

By comparison, the monthly change in payrolls has a 90% confidence band of +/- 130k which means that in 90 times out of 100 all we can say is that payrolls were up by somewhere between 124k and 384k—any of which would have been solid.

HOURS WORKED WERE WEAK; GDP GROWTH NEEDS PRODUCTIVITY GAINS

A fly in the ointment lies in the fact that hours worked were down by –0.1% m/m SA. That’s the second drop in three months. The overall third quarter saw very little growth in hours worked (chart 6). 

Since GDP is an identity defined as aggregate hours worked times labour productivity with the latter defined as real GDP divided by hours worked, US GDP growth tracking in Q3 of around 2½% requires a solid gain in labour productivity.

IMPLICATIONS FOR THE FEDERAL RESERVE

Scotiabank Economics continues to forecast quarter point rate cuts at the November and December meetings. Quite honestly, if I were Powell, then barring a shock and unusual effects stemming from the US election, I would be very open to skipping the November meeting after going big to start.

I continue to believe that the FOMC committed policy error by cutting 50bps last month. The only other times in modern memory when they have cut by 50bps to start an easing cycle were in the exigent circumstances of the dot com and GFC. This is not a crisis. The US economy remains deep into excess aggregate demand with a positive output gap. There is no move toward creating disinflationary slack and therefore the US economy hasn’t even begun the lagging out effects of slack into sustainably soft inflation.

Chair Powell & Co lit up fed funds futures as I had argued they would by going 50bps out of the gate and handed the keys to monetary policy over to markets. That was a mistake and now the FOMC has to rein in markets.

The bigger issue is that all this negativity that is trying to push the FOMC toward aggressively easing risks backfiring by creating the false sense of a crisis brewing. The US economy is strong and quite possibly in a no-landing scenario. That should be celebrated from a growth standpoint, but treated very carefully from a monetary policy standpoint that risks unleashing all the same forces that policy tightening sought to bottle up.