Perusing Private September Data Mid-Shutdown

Washington’s ongoing shutdown has interrupted the usual flow of government-produced economic data, leading some to wonder what the delayed reports may actually show. But worry not: Private companies produce data aplenty, which can shed some light. While these aren’t perfect, neither are official government sets. Here we will dive into available September figures from an investor’s point of view to get a least a rough update on developments.

ADP’s Latest Report Mirrored Ongoing Labor Market Trends

The payroll provider’s data showed private US employers shed -32,000 jobs in September, well below analysts’ expectations for a 45,000 increase.^[i] Though it also warns its preliminary annual “re-benchmarking” against the 2024 Quarterly Census of Employment and Wages extended losses to -43,000 and flipped August’s initially reported 54,000 private payroll increase to a -3,000 job loss. This hints at downward revisions across the board when the full-year benchmarking happens next February—old news to stocks, but perhaps noteworthy.

Industry trends probably won’t shock, either: All shed workers except Information Services (up 3,000) and Education/Health Services, which added 33,000 jobs. That extends longer-running labor market trends, which have caused snide remarks about AI and illness being America’s only economic growth drivers. So we doubt anything here improves economic sentiment. It probably just extends stocks’ wall of worry.

Wages, meanwhile, painted a slightly rosier picture, growing 4.5% y/y in September—little changed from August’s growth and continuing to outpace inflation, helping households continue regaining purchasing power after 2022’s pain.^[ii]

So overall, this report lacked big shifts—no sudden turns for good or ill in the labor market. It wasn’t a great report, but employment data are always late-lagging. Thus, private job losses are likely an aftereffect of weakness earlier this year. They won’t tell you what is coming, making them largely inconsequential for forward-looking stocks.

Mixed September Purchasing Managers’ Indexes (PMIs) Revealed Little New Information

Starting with the headline figures, America’s services sector moderated but remained mostly expansionary. S&P Global’s PMI came in at 54.2, down slightly from August’s 54.5 but still above early 2025’s readings and over 50.0, indicating expansion.^[iii] Forward-looking new orders also slowed, but overseas orders rose for the first time since Liberation Day. Today’s orders are tomorrow’s production, so that augurs well for the US’s mighty services sector to keep trudging along.

The Institute for Supply Management’s (ISM’s) Services PMI, which slipped from 52 to 50.0, looks iffier.^[iv] Production contracted slightly, while new orders fell sharply from 56.0 in August to 50.4—a negative sign for future output growth. Inventories also flipped to contraction, but respondents’ comments indicate this isn’t because firms had trouble keeping up with demand. Mostly, they seemed content to let pre-tariff stockpiles run down and mitigate the risk of overstocking as demand weakens.

Manufacturing PMIs were also mixed. S&P’s gauge remained in expansion, moderating slightly to 52.0 from August’s 53.0.^[v] New orders expanded for a ninth straight month but slowed, and export orders fell for a third straight month. So while services’ international customers have returned after the tariff spook, factories’ overseas clients are still either uncertain or working through goods stockpiled earlier this year.

ISM’s PMI didn’t shed significantly more light. It contracted at a slower rate, inching from August’s 48.7 to 49.1 in September, but the underlying figures weren’t encouraging.^[vi] While production flipped back to expansion at 51.0, new orders reverted to contraction at 48.9, with export orders tumbling from 47.6 to 43.0. Manufacturers’ and customers’ inventories fell, but this is always open to interpretation and there were no comments revealing whether demand outpaced supply or businesses ran leaner on purpose, fearing weaker demand as their services counterparts were.

Beyond these meh macro readings, PMIs help sketch the employment and inflation landscape. On employment, PMIs reaffirmed what ADP and earlier BLS data already showed. ISM reported manufacturing and services employment contracting, while S&P global showed negligible services hiring and stronger factory job growth—the last bit a noteworthy ADP divergence. This isn’t shocking, considering PMIs measure only how many businesses are hiring, not how many workers they are adding. The majority of firms’ adding jobs might have been insufficient to offset deeper cuts elsewhere. Meanwhile, on inflation, S&P reported slower selling price growth for services and manufacturing despite tariffs’ raising costs.

Foggy Private Spending Data Pointed to Slower Growth

As a refresher, Census Bureau data showed retail sales rose 0.6% m/m in August—the last government reading before the shutdown began.^[vii] Most private sector sources estimated spending slowed slightly from here in September. For instance, credit card data from Bloomberg’s Second Measure and Bank of America pointed to slower spending, particularly for discretionary items like furniture and appliances.^[viii] Elsewhere, the Fed’s Beige Book noted retail spending “inched down in recent weeks,” while the Chicago Fed’s Advance Retail Trade Summary projected 0.5% retail sales growth excluding autos.^[ix]

Mind you, these are imperfect projections. Public spending data aren’t perfect either, but they are seasonally adjusted and employ a survey of around 5,000 establishments.^[x] Conversely, many private sector sources rely on formulas and omit services spending—the majority of Americans’ spending—and cash transactions. These are some major blind spots worth considering. The Fed’s Beige Book is a helpful but mostly qualitative tool, based on anecdotal reports from regional Fed banks’ industry contacts. Still, nothing here suggests consumers are suddenly pinching pennies. Not that spending is an economic swing factor, but we have seen a lot of talk about missed auto payments signaling troubled US consumers. These reports, all told, seem to disagree.

So overall, the latest data show pockets of strength and weakness, with no big shocks on either side. And we won’t even have to wait long for our first official confirmation of private-sector reports! As a special one-off, the BLS is releasing September’s CPI report on Friday to hit Social Security’s Cost of Living Adjustment deadline. But this—as with any economic series—has its limitations. Not only do they experience collection issues, but they input tens of thousands of items to spit out one or a few outputs. Assuming this final product provides a precise read on any part of the economy is off base. Economic data hint at broad trends at best—worth keeping in mind if headlines remain focused on data accuracy or delays.