September Added 29,000 Jobs. The Summer Revisions Deserve a Look, Too.

September’s jobs report has me curious about the numbers behind the number. The headline payroll gain is modest, and the summer revisions make the story more interesting. Let’s dig in before anyone decides one report gets to boss around an entire retirement portfolio.

U.S. nonfarm payroll employment rose by an estimated 29,000 in September, while the unemployment rate was 4.2%, according to the Bureau of Labor Statistics release on October 2. BLS described both measures as little changed. July and August together now show 60,000 fewer payroll jobs than previously estimated.

For investors who also rely on a paycheck, that brings up a practical question: how much breathing room does the household have if income gets interrupted? A national report can nudge us to check. Our own bills get the final say.

The revision changes the three-month picture

Here’s where I’d linger for a minute. July’s payroll change was revised from a gain of 21,000 to a loss of 10,000. August’s gain moved from 162,000 to 133,000. Add September’s 29,000 and the latest estimates show a combined increase of 152,000 jobs over those three months.

That works out to about 50,700 jobs per month. I’d keep that calculation beside September’s headline. It smooths the monthly swings a little and shows how much August contributed to the July-through-September net gain.

One tempting calculation needs a firm little nope: subtracting the 60,000 downward revision from September’s 29,000 and declaring that September lost jobs. The revision belongs to July and August. Mixing the periods gives us a tidy answer to the wrong question.

These estimates can change again. BLS routinely revises the previous two months as additional business and government reports arrive and seasonal factors are updated. Its October 2 release explains the latest revisions. Useful evidence, yes. Final word, no.

Paychecks need a closer look, too

Average hourly earnings for private nonfarm employees rose 0.1% in September and 3.0% over the year. Their average workweek held at 34.4 hours. If you’re mentally comparing that with your own paycheck, fair enough. These broad averages can’t tell us who got a raise, lost overtime or changed jobs.

This is where I’d get a little inquisitive on a consumer-facing company’s next earnings call. Are customers buying less? Are promotions helping sales at the expense of margins? Does management’s outlook assume demand improves? I want answers backed by company disclosures. The jobs report hasn’t answered those questions for us.

And “jobs” needs a quick word check. Payroll figures count jobs on employer payrolls, while the unemployment rate comes from a separate household survey. People with nonfarm payroll jobs at different establishments are counted once at each establishment. The monthly payroll change is a net change in payroll jobs, not a count of all people hired during the month.

Thursday’s claims report adds context

A day earlier, the Labor Department reported an advance estimate of 197,000 initial unemployment insurance claims for the week ending September 26, seasonally adjusted. That was 1,000 below the previous week’s revised figure.

Initial claims are new applications for unemployment-insurance benefits, not a count of every layoff or every unemployed worker. They cover a different period and measure a different activity from the monthly payroll survey.

So I’d give that weekly decline some attention without asking it to do the entire report’s job. It doesn’t wave away weak payroll growth, and weak payroll growth doesn’t establish that a wave of layoffs has already arrived. Both deserve a look.

A five-minute check closer to home

Let’s bring out a calculator. Mercifully, this part doesn’t require an economics degree.

Take a hypothetical household with $18,000 in accessible cash set aside for essential expenses. At $3,000 a month, that covers six months. If expenses would rise to $3,600 after an income interruption, the same balance covers five months. No investment return is assumed, and this is an illustration rather than a recommended savings target.

That missing month is worth a second look. Replacement health coverage, debt payments or another obligation could change the calculation. I’d write down the costs that would actually remain and any income that would still arrive. The account balance is only half the little math problem.

Then comes the investment question: would paying those bills require selling during a decline? Someone with a long investing horizon and reliable cash reserves faces a different decision from someone who expects to need portfolio money soon.

I’d also check whether employment income and investments depend heavily on the same industry. A slowdown could affect both at once. That’s worth examining thoughtfully, without treating one release as an instruction to trade.

September’s modest payroll increase and the weaker revised summer figures give us something concrete to monitor. I’d update the three-month calculation with the next release. Meanwhile, we can check whether the household cash calculation still works. It’s a small bit of homework with a much more useful payoff than arguing with a headline.

For general education. This analysis is not personalized investment advice.