Ferg’s Finds (Tin Special)

This is a short weekly email that covers a few things I’ve found interesting during the week.

Article(s)

President Prabowo Orders Closure of 1,000 Illegal Tin Mines

Speaking before political party leaders in Jakarta on Monday, September 29, 2025, President Prabowo stated that the crackdown is expected to prevent potential state losses of up to IDR 22 trillion (approximately USD 1.47 billion) between September and December 2025, with the figure projected to reach IDR 45 trillion by the end of 2026.

“Bangka Belitung has long been a leading global tin mining center. It hosts 1,000 illegal mines. Starting September 1, I ordered the TNI, Police, and Customs to launch a massive operation to shut down what has allowed nearly 80 percent of our tin to be smuggled out. We’re closing it down, 80 percent of our tin!” the President declared.

For those that have been following this little market the saviour was supposed to be the restart of mines in Wa State in Myanmar, which to date has been a tickle at best.

Tin market still beholden to the fortunes of Myanmar mine

It’s been two years since the mine, one of the world’s largest, was closed for a resource audit. It’s been six months since the authorities in the semi-autonomous Wa State invited applications for new mining permits.

Yet to date there is no evidence of any ramp-up in activity. Indeed, the flow of tin concentrates from Myanmar to neighbouring China has almost dried up completely.

• May 2025: Less than 700 metric tons exported.

• July 2025: Just 933 metric tons exported (94% reduction from pre-suspension monthly averages of 15,000 tons).

• August 2025: data not yet available (but SHFE drawing down…)

Putting this in context, Myanmar was supplying an average of 15,000tons per month between 2018 and 2024 (as high as 40,000tons per month in the 2015 to 2017 period).

I don’t see production bouncing back for all the reasons I ran through in Revisiting Tin

Podcast/Video(s)

Yes, it’s nine months old, but it’s still packed with insight relevant for today: Investing in Tin 2024: The Investment Case for Tin - Tom Langston, ITA

Crazy, only 632 people have taken the time to watch this….

Quote(s)

I’ll get to the consequences for tin…

“Other recent AI news: Oracle’s stock jumped by 25% after being promised $60 billion a year from OpenAI, an amount of money OpenAI doesn’t earn yet, to provide cloud computing facilities that Oracle hasn’t built yet, and which will require 4.5 GW of power (the equivalent of 2.25 Hoover Dams or four nuclear plants), as well as increased borrowing by Oracle whose debt to equity ratio is already 500% compared to 50% for Amazon, 30% for Microsoft and even less at Meta and Google.”

- Michael Cembalest

Tweet/notes

I started beating the drum on how bullish this was in this Ferg’s Finds in June.

Now we have SHFE stocks having drawn down a further -8% and LME -32%…

Charts

The Shanghai Futures Exchange (SHFE) tin warehouse stock is what you need to keep an eye on here, as I fully expect it to keep rolling over with what’s happening in Indonesia and Myanmar, which should result in prices taking out prior highs, barring a nasty recession.

Something I’m Pondering

I’m pondering what a bust in AI CAPEX would do to tin demand?

AI applications have represented an extraordinary 93.3% of all global tin demand growth over the last 5 years.

Therefore, it’s only fair that investors are questioning whether things could end up resembling the 2022-2023 period again in the cyclical sector of semiconductors, which accounts for roughly half of tin demand through solder (51% of tin consumption is solder/electronics, 17% is tinplate, 12% lead acid batteries, 15% chemicals/alloys and 5% electroplating).

This is unlikely due to the 2020-2022 period being unique, during which COVID lockdowns drove global electronic demand to unprecedented heights. When lockdowns ended, electronics sales plummeted.

AI representing 93.3% of all global tin demand growth over the last 5 years amounted to 28.0 kilotons (the tin demand grew from 350kt to 380 kt).

When you break it down, AI represents ~9% global tin consumption.

• AI Chips and GPUs account for 4.13% of global tin usage (15.73 kilotons).

• AI Data Centre Infrastructure accounts for 3.01% (11.47 kilotons).

• If you include AI Edge Devices, which contribute 1.36% (5.18 kilotons), and AI Training Hardware, which represents 0.73% (2.78 kilotons).

China and the “rest of Asia-Pacific” represent 46% of AI tin consumption. Asian demand isn’t as frothy as US (see previous semis chart) while being strategic (government backed) and having a long growth runway(I’m invested in it).

So, even if you lump the likes of Taiwan, South Korea, Japan, and Europe in with the US, you are talking about ~4.8% of total global tin demand.

Tin Growth

Similar to energy, tin is ultimately an Asia story (The Coming Asian Boom)

Over the past 20 years, global tin consumption has grown at a compound annual rate of ~2%.

Asia-Pacific: The regional concentration, especially China, powered by electronics, green technology, and EV growth, consistently outperformed the global average, posting regional CAGRs up to 3.2% annually

Tin Supply

I’ve dug into the tin supply situation here and here (and reinforced the view with the latest announcements out of Indonesia and Myanmar). Needless to say, I’m a massive tin bull and see the size of the supply issues far outweighing any demand issues arising from an AI Capex bust or even a recession.

Cheers,

Ferg

P.S. I continue to be impressed with how well Perplexity Pro performs, particularly in its ability to select the data source, choose the appropriate model, and provide a clear breakdown of sources. I went over how I’m getting the most out of Perplexity and Koyfin in my recent piece.

This function is awesome