Why the Huge Discrepancy Between ISM and the S&P Manufacturing PMI?

Yesterday, a reader asked about negative ISM readings and positive PMI readings.

Reader TexasTime: Mish, is there a reason you use the ISM PMI instead of the S&P PMI? It would be interesting to know why the two PMIs are differing in their opinions on manufacturing.

Mish: I tend to look at ISM because that is what GDPNow looks at. GDPNow does not consider S&P at all. And the market does not seem to respond to the PMI reports either.

Finally, ISM manufacturing employment does seem to track BLS reports on manufacturing jobs.

But yeah, that’s a fair question. Clearly there is a sampling issue of some sort. They are both diffusion indexes with big problems. I will look at PMI now to see what they have to say.

S&P Global US Manufacturing PMI^®

Please consider the S&P Global US Manufacturing PMI^® for November.

Operating conditions in the US manufacturing sector improved for a fourth successive month in November. A solid rise in production and a further increase in employment was reported as confidence in the outlook strengthened.

However, current sector performance was limited by a considerable slowdown in demand growth, in part due to weak sales, which in turn contributed to an unprecedented rise in stock of finished goods for a second month in a row.

Inflationary pressures meanwhile remained historically elevated. Tariffs were again frequently mentioned by panelists as having driven input costs higher, although the degree of pass through by manufacturers was weaker than in November with selling price inflation amongst the lowest of the year so far amid intense competition and weak demand.

Are the reports stunningly different? Not if you focus on production, inflation, and S&P comments.

Chris Williamson, Chief Business Economist at S&P Comments (Emphasis Mine)

• “Although the headline PMI signalled a further expansion of factory activity in November, the health of the US manufacturing sector gets more worrying the more you scratch under the surface. The main impetus came from a strong rise in factory production, but growth in new order inflows slowed sharply, hinting at a marked weakening of demand growth.
• “In short, manufacturers are making more goods but often not finding buyers for these products. This combination of sustained robust production growth alongside weaker than expected sales led to a worryingly steep rise in unsold inventories. For two successive months now, warehouses have filled with unsold stock to a degree not previously seen since comparable data were available in 2007. This unplanned accumulation of stock is usually a precursor to reduced production in the coming months.
• “Profit margins are meanwhile coming under pressure from a combination of disappointing sales, stiff competition and rising input costs, the latter widely linked to tariffs.
• “Encouragingly, manufacturers have grown more optimistic about the year ahead, with the ending of the government shutdown helping lift confidence from the sharp drop suffered in October. Optimism is being fueled by hopes of improved policy support, including lower interest rates, as well as greater political stability, though it is clear that uncertainty remains elevated and a drag on business growth in many firms, holding confidence well below levels seen at the start of the year.”

ISM Manufacturing Contracts for the 35th Time in 37 Months

Let’s take another look at my report ISM Manufacturing Contracts for the 35th Time in 37 Months

Please consider the November 2025 Manufacturing ISM® Report On Business® .

“Decreases in two of the four demand indicators (Backlog of Orders and New Orders) overwhelmed the gains posted by the New Export Orders and Customers’ Inventories indexes. 

“Regarding output, production jumped into expansion, but employment contracted at a faster pace, as 67 percent of panelists (the same as October) indicated that managing head counts is still the norm at their companies, as opposed to hiring.

Q: So, are the reports really that much different?
A: If you look at headline numbers, perhaps. But the details are very similar.

Q: How does this happen?
A: They both measure the same industries but diffusion indexes have major issues.

In a diffusion index, direction matters not the amount. For example: An employer adding 3 workers will offset an employer laying off 50 workers.

There are sampling issues and survivor bias issues. A struggling company may not bother responding. A business that has gone out of business certainly won’t respond.

I do not know the number of businesses sampled or the response rates for either company.

Month-Over-Month Change in Manufacturing Jobs

Manufacturing jobs have been anemic since January of 2023. In 2022, the economy added 357,000 manufacturing jobs!

The above chart aligns nicely with ISM’s reported manufacturing contraction for the 35th time in 37 months. It does not align well with the S&P PMI.

Since January 2023, manufacturing lost 183,000 jobs. In Biden’s last two years, manufacturing shed 129,000 jobs.

You cannot blame Trump for that. But you can blame trump for his policies and his results since 2025, and especially his bragging about what he is doing.

It’s Trump’s Economy Now

Since January 2025, Manufacturing has lost 54,000.

If you wish to call -54,000 in 9 months a slight annual improvement since Biden’s last two years, be my guest. But it will not appease anyone who lost their job. Nor will it appease anyone who put a lot of faith in Trump’s exaggerated promises.

Trump repeatedly overpromises and underdelivers.

Tariffs Destroy Jobs

Are tariffs bad for growth? Yes, say five decades of data from 150 countries

We embarked on a data collection exercise, covering over 150 countries, and more than a half-century of data to tackle our basic question: does an increase in import tariffs boost the size and growth of the aggregate pie (GDP) or shrink it, and if so by how much?

Using aggregated annual data for 151 countries (34 advanced and 117 developing) over 1963–2014, we find that tariffs have economically- and statistically-significant adverse effects on output growth. The impact is persistent and increases with the magnitude of the tariff change. Our baseline econometric model suggests that a one standard deviation increase in the tariff rate (corresponding to a 3.6 percentage points) leads to about a 0.4% decline of output five years later.

We Pay the Tariffs Anecdotes, Lost Sales and Going Bankrupt

A reader brought this story to my attention in response to We Pay the Tariffs Anecdotes, Lost Sales and Going Bankrupt

Reader Anecdote: “The furniture tariffs decimated my wife’s employer, VCF, a 75-year furniture chain. Entered Chapter 11 last week. 3,500 about to lose their jobs in January. Ironically, this hits their furniture manufacturers in the Carolinas. It will be a net loss of manufacturing jobs in the end.”

Trump’s tariffs, allegedly designed to bring jobs back to the US, will destroy more jobs than it creates.

This is no surprise. The same applies to steel and aluminum. It’s how tariffs work.

No jobs will return to the US over this nonsense. Tariffs are a net destroyer of jobs while driving up prices in the short term.

Related Posts

February 11, 2025: Trump’s Steel Tariffs Now Will Work as Good as the First Time

Q: How’s that? A: Very poorly.

March 13, 2025: The Amazing “Success” of Trump’s 2018 Aluminum Tariffs in One Picture

I hope you can take a bit of headline sarcasm because the true story follows.

September 6, 2025: Trump’s Aluminum Tariffs Seriously Backfire Already

Tariffs did not and will not bring production back to the US.

Here’s a pertinent comment from  ISM Manufacturing Contracts for the 35th Time in 37 Months

“We are starting to institute more permanent changes due to the tariff environment. This includes reduction of staff, new guidance to shareholders, and development of additional offshore manufacturing that would have otherwise been for U.S. export.” (Transportation Equipment)

Hoot of the Day: That company is developing more offshore manufacturing and firing US staff, not bringing more manufacturing back to the US.

Understand the Obvious

One should not need any studies to understand the obvious.

Steel and aluminum tariffs are a great example. For every steel or aluminum job gain, there are tens-of-thousands of users and buyers of steel and aluminum, all of which lose from tariffs.

So, if you are rooting for more tariffs, you are a brainwashed, tariff-loving economic illiterate (like Trump and Biden) rooting for net job losses. Alternatively, you are someone in an industry spouting economic nonsense for personal gain.

To repeat: Tariffs did not and will not bring production back to the US. They are a guaranteed net job destroyer to everyone but the protected industry (and sometimes even the protected industry).

I am tired of arguing this time-proven and obvious economic point.

Addendum – National Security

Reader: You’ve stated previously that America must do more to manufacture strategic goods. I’ve asked many times, but I don’t think you’ve ever taken up my challenge to list those 10 or so things that fit that bill.

And then, more importantly, what does a well thought out & implemented national plan look like to achieve that goal?

Mish: I have indeed mentioned areas where I support government interference, multiple times – in general, on matters of genuine national security.

I have named two.

1. Chips
2. Rare Earth minerals

There could be more, but not aluminum, not steel, not clothes, toys, or even cars. And I mentioned what I would do.

• I praised Biden for getting getting TSMC to build in the US (only one of 2 things Biden did that I agree with)
• I have repeatedly called for cooperation between the US, Canada, and Mexico on rare earths. That’s been my stated position for years.

Neither is a matter of tariffs. Both are matters of subsidies. We need to be independent of China on both.

Getting Taiwan Semiconductor TSMC to build a plant here was a great win by Biden. The cost was a mere $6 billion + $5 billion in loans. What a great deal.

My idea on Rare Earths applies to Greenland, Canada, and Mexico. Greenland and Canada have lots of minerals and land. Mexico has cheap labor.

Instead of cooperating with Canada, Trump engaged in stupid tariffs and repeatedly made idiotic comments about Canada becoming the 51st state.

Trump’s taunts were foolish for multiple reasons: Canada is quite liberal and would vote with Democrats, and second it is counterproductive to taunt genuine allies. Canada is now in discussion with China over oil exports. How stupid is that besides extremely?

Trump would not rule out invading Greenland. The result was Greenland opting to stay closer to Denmark than go independent in the last election. Again, Trump’s bluster was blatant stupidity.

Another possible strategic interest is AI, but it does not need any financial help.

While typing, I will add aircraft, aerospace, and defense. However, Elon Musk has done an amazing job with aerospace. The US is the world leader in need of no assistance. Our defense contractors need no assistance either, except for rare earth minerals.

If you count those separately that’s six. Chips, Rare Earth Minerals, Aerospace, Aircraft, Defense, and AI. Of those, only two need assistance, and I have been on those bandwagons for years.

For obvious defense reasons, we cannot let China dominate in any of those areas. The big problem now is rare earths.