Cowden's Take: Why Nucor's unchanged CSP felt like big news

Case in point: Nucor officially held CSP at $875/ton from Aug. 25 to Oct. 20, 2025 – so for two months, according to SMU price records. SMU’s spot price for HR, in contrast, dipped from the low $800s/ton into the $700s/ton in September 2025. And we didn’t get back to $875/ton until just after Thanksgiving.

Coincidentally, just after Thanksgiving is roughly when a sheet price rally – one that has endured pretty much uninterrupted ever since – started. And Nucor’s CSP reflected that. It has been up nearly every week since Jan. 27. The last time it was flat was Jan. 12, or nearly six months ago. (The company didn’t post a CSP on Jan. 19 because of MLK day.)

We’ve gotten so used to the slow, steady increases by now that $5/ton feels like the new flat. And flat, after so many consecutive weeks of gains, feels like a decrease.

We saw an immediate reaction on futures markets to the announcement. It was a sea of red on the CME. And steel equities also fell. But has anything really changed in the physical market? Lead times remain long, service center inventories remain lean, and SMU is not aware of any sudden deterioration in demand.

Maybe Nucor was concerned about import competition and so wanted to put the brakes on further price increases? That would make sense. Sure, prices continue to rise in the US. But they’ve been drifting lower in the rest of the world.

Meanwhile, import volumes have been ticking up from one month to the next. The US imported 1.92 million metric tons (mt) of steel in May, according to preliminary figures from the US Commerce Department. (June figures are not yet available.) That’s up 11% from April and up 34% from a recent low of 1.43 million mt in December. We haven’t been above 2 million mt since July of last year.

I’ve noted this before. And I think it bears repeating. In the past, mills typically talked up prices and lead times. We’re in a situation now where the reverse seems to be the case. Steel buyers often say they’re not able to find spot material at Nucor’s CSP. Nor can they find it within the company’s published three- to five-week lead time. I’m focusing on Nucor here simply because the steelmaker lists a price and lead time range. But it’s a similar story at other mills. (Another EAF mill published sheet lead times with no lead times – unless you consider “closed” to be a lead time.)

With lead times still long and supplies still tight, how will other mills and the rest of the market react to Nucor’s move? Service centers who’ve bought at current high levels won’t want to see the value of their inventory devalued or new orders underwater by the time they arrive. And if you’re buying from certain domestic mills, that might not be until September. (Some of you think it’s actually more like October given how late certain producers have been running.) Also, will other mills follow along if they think they can continue to push prices higher and? (Especially in the case of producers who might need higher spot prices to offset low fixed-price contracts.)

A lot to consider. And maybe it will be a nothing burger if this is just a pause before the slow, steady gains resume after July 4. As it stands, we got some fireworks before Independence Day that caught me a little bit by surprise.

Here are some potential fireworks I’m also keeping an eye out for: Any holiday adjustments to Section 232 tariffs. That falls into the category of “Yes, I’m paranoid. But am I paranoid enough?” I think back to the summer of 2017, when there were widespread rumors President Trump (then in his first term) might announce Section 232 tariffs on or around Independence Day. Speculation had been brewing since Trump announced Section 232 investigations steel imports in April 2017. I was told at the time the rumors themselves were enough to impact buying patterns. Basically, that people bought ahead in anticipation of the tariffs.

SMU service center inventory data doesn’t stretch back quite that far. (Our data series starts in 2019.) But our pricing archives do. We saw HR prices jump from $580/st on average in early June 4, 2018, to $610/st in on July 4, 2018. A $30/st gain is quaint by post-pandemic standards. But it was notable at the time. (Trump didn’t actually announce the tariffs until March 2018 – when it caught even well-connected folks in the industry by surprise.)

I also bring up the tariff issue because we’ve had some questions from some of you about whether S232 might go to 75%. I’ve learned over the last decade it’s not wise to try to predict tariff policy when it comes to this president. That said, Trump 2.0 does seem to like 25% increments.

So here’s a fun fact: US HR bottomed out an average of ~$800/ton from Aug.-Oct. of last year. That was just before the impact of the 50% S232 became apparent with a plunge in imports in late Q4 that continued into Q1. Add 50% to that $800, and you’re at $1,200, where the high end of SMU’s HR price range is as of this week.

Speaking of tariffs, we got some interesting responses to a question last week in our steel market survey about whether S232 rates for Canada and Mexico might change as part of the review of the US-Mexico-Canada Agreement (USMCA). The mandatory review process starts on July 1. We’ll share those results with you in our Thursday issue.

SMU Steel Summit and Final Thoughts

Work tends to slow down this time of year. If you find yourself with a little extra time, now is a great time to reserve your for SMU Steel Summit on Aug. 24-26 at the Georgia International Convention Center in Atlanta. You can see the full agenda here, the companies attending here, and you can reserve your spot here.

Speaking of Final Thoughts, they’re taking a pause for the holidays. We’ll resume them again, as a separate newsletter delivered on Friday, next week.