We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 02 November 2025:

1. Do We Want an Age of AI Robopets? – Jessica Roy

One August morning, Kaarage woke up and took the train from the Japanese countryside into the city. She went to a restaurant and enjoyed a lunch of vegetables and soup, as well as an iced coffee. Afterward, she studied a musical score on her iPad, then went home to relax.

Kaarage is not a person: She’s an internet-beloved Moflin, an AI-powered robopet made by Casio—yes, Casio—who shares a charming, bucolic life with her owner in rural Japan…

…Since Casio initially released Moflins last year in Japan, they’ve proven to be a surprise hit, with the company selling several million dollars’ worth of Moflins in a matter of months. Last month, Casio made them available in the U.S. too, offering the furry-haired critters on its website for $429 a pop in two colors: gold and silver…

…Casio markets them as an “AI companion and robot pet” that can offer “quiet reassurance,” “ease stress” and “bring comfort.” (Watch out, loneliness epidemic.) The wellness language here is purposeful—a very real attempt to imagine the softer, cuddlier side of AI…

…The fandom and internet subcultures devoted to the robopets include a Reddit board where Moflin owners share fur-care tips and celebrate their Moflins’ “50-day” birthdays, the point when Casio says the AI pet has fully studied its owners’ vocal tones and can best respond to them in a series of purrs and coos audible through a tiny, built-in speaker beneath their fur. In Japan, hardcore Moflin owners can even spend $49 annually to join Casio’s Moflin Membership Club, which gives them access to health checkups (for maintenance issues, like charging problems) and appointments at a salon to take care of their Moflins’ fur…

…The Moflin, which weighs as much as a small rabbit, comes equipped with an app, several auditory and touch-based sensors, and a battery that lasts about five hours. (It charges in a soup bowl–shaped bed.)…

… Its only facial feature is two beady eyes. This is by design.

“We intentionally avoided features like ears, tails or distinct facial characteristics because making them look like a real creature would only emphasize how they differ from one,” said Casio developer Daisuke Takeuchi. “The abstract design allows each person to interpret who Moflin is to them, which helps build a more personal bond.”

Moflins rely on what Casio calls “emotional AI” to learn and respond to their environments, developing different personalities based on their owners’ interactions. The companion app, MofLife, allows users to track a Moflin’s mood to see how affectionate and energetic it feels…

…Amy Wang, 27, of New York. Wang has bad allergies and a small apartment, so a real pet was out of the question, but she said her Moflin, which she named Roku, provides much of the same emotional support a pet would…

…Whether the Moflin appealing to a younger demographic is a good thing remains to be seen. Nataliya Kosmyna, a research scientist at Massachusetts Institute of Technology’s Media Lab who focuses on AI, said there’s not a huge amount of research into the effects of soft AI, like that used by the Moflin, on children’s brains, but that’s exactly the issue: Kosmyna argues there should be more research into the impact of emotional AI toys on kids before they hit the market. 

2. Argentina Could Be a Superpower – Tomas Pueyo

Argentina used to be rich.

Its capital, Buenos Aires, was “the Paris of South America”.

For decades, Argentina (which means “the country of silver”) was among the richest countries on Earth—richer than France, Germany, Japan, or Italy…

…Not only did the Western world leave Argentina behind. Traditionally poorer countries like Chile and China are now richer! And Brazil is catching up!

How is this possible?

Because, unlike most countries I write about, Argentina is poor despite its amazing geography. With better management, it could become the United States of Latin America…

…Argentina is basically the US of the Southern Hemisphere:

• Very similar defensibility, with oceans, mountains, and ice on three sides, and weak neighbors on the other
• The huge exception is Argentina’s neighbor, Brazil.
• Very similar land and climate, allowing for a world-class agriculture industry and cheap infrastructure.
• A very similar navigable river basin in the heartland, helping reduce transportation costs, and creating wealth and political harmony, all controlled from Buenos Aires.
• Huge, untapped mineral deposits.

Despite these striking advantages, Argentina has not been able to translate them into immigration and wealth. Geography is not destiny.

One way to put it: Geography is the hardware, our institutions are the software. When both work well, a country is unstoppable. With bad hardware but intelligent software, a country can go far. But it’s easy to waste good hardware with very bad software. This is what Argentina has done. Another way to put it: Geography is the chessboard: How you play on it determines your success, and Argentina hasn’t played very well.

3. The AI Boom’s Real Economy Problem – Bob Elliott

Meta’s release showed revenue grew 26% from the same quarter last year, or roughly 10bln, claiming that AI is now helping improve the way ads are being placed on the platform. The ads of course being the only source of revenue for the business…

…On the surface those numbers sound great for any company, but in context it’s a pretty mediocre outcome. For instance the rise of 26% y/y is only at a marginally faster rate than previous years 3Q reads which grew 19% and 23% respectively. All that AI investment for a few extra percentage points.

To achieve these goals Meta is spending upwards of $70bln on AI capex to say nothing of rising operational expenses all chasing the hope that it’ll drive increased income…

…Of course all the AI investment is driving more income, but at best it’s maybe 3-5bln more than they would have had relative to the underlying trends pre capex spend. I’m no individual company analyst, but investing 70bln/yr to get 3-5bln/yr of revenue seems like a pretty shitty ROI…

…The whole sector faces the same basic problem. Already they are spending upwards of 60% of their operating cash flow on CAPEX at this point…

…The math is pretty simple, unless there is a surge in revenues from these activities, big tech is going to pump nearly all their free cash flow into CAPEX in just a few years…

…Blowing all this cash on investment means that they need to start to generate significant incremental cash flow from their investment on real economy activities (not just self referential activities to each other on things like cloud, etc)…

…Cumulative investment has surged and yet actual revenues either direct or indirect from these activities has been, has been … lets call it subdued…

…But the reality is that there are already signs that the AI adoption curve for companies is starting to bend downward even as forward expectations are high, a real threat to the idea that revenues will surge ahead…

…Increasing revenue may not be the primary benefit of AI for the economy, because most of the benefit will come in the form of increased efficiencies…

…But higher margins do not come free of impact. Workers earnings by definition finance the vast majority of spending in the real economy. So the trouble is that if you fire a bunch of workers, they have less income to spend, and with less revenue earned. The real economy realities make what looks like a free lunch actually a drag.

4. Is AI Eating CSI? – Dragon Field

As the ChatGPT turns three in November 2025, the most popular recent riff is “AI is eating SaaS”, which has claimed countless victims of the once popular software companies such as Duolingo, Shutterstock, Coursera, Gartner, Adobe, and Constellation Software. Everyday we have hundreds of TikTok influencers and YouTubers hyping the notion that even people without any coding experience and no technical education can simply type a few prompts into ChatGPT, and the AI will automatically create a software application in a few minutes. We also have high-profile tech CEOs like Ali Ghodsi of Databricks and Satya Nadella of Microsoft all announcing that “AI is eating SaaS”.

In fact, the expression that “AI is eating software” was first mentioned by the Nvidia CEO Jensen Huang in a 2017 LinkedIn post…

…About 25 years ago, I became an IT operations manager for a metal stamping plant for one of the Detroit Big Three auto companies. The plant is 2.4 million square feet, sitting on 118 acres of land. It produces automotive parts like hoods, door panels, bumpers, floor pans, and hundreds of other smaller parts. The plant had about 1,600 employees working three 8-hour shifts for six days a week at the time. At its peak in the 1950s, the plant had several hundred presses and employed over 6,000 people…

…In stamping plants we don’t usually let the manufacturing execution system (MES) have full control of the production because if the IT system is down, we would not shutdown the press lines. This is a situation we call “running blind” and usually you want to restore the system as quickly as you can. In addition, our VMS system was integrated with our warehouse inventory and corporate ERP systems, so a sustained downtime can cause a lot of issues thus we consider it mission-critical. Accuracy and reliability are the most important for us.

At the core of our MES is a VMS for production monitoring. It was first developed in the 1980s by a small vendor in Michigan when the US Big Three auto companies started to automate and install IT systems in their manufacturing plants…

…This VMS is deeply imbedded in every aspect of our production and workflow as depicted in the chart below. It has integration with our ERP system that is running on IBM mainframe with blue screens. It’s used by most departments in the plant, even the Finance and HR people use the system regularly for production and labor hour reports…

…For many years, this small VMS vendor only had three employees: One hardware engineer who liked to hide in the workshop fiddling with all kinds of gadget, a software engineer who focused on the the software development and upgrades, and the third engineer who worked as the leader and the face of their company…

…For many years, we also tried to find a replacement for this VMS, either from another vendor or develop one by our own internal IT. Sometimes the pressure from my own IT headquarters was intense. Like most legacy VMS systems, this VMS was first procured by the business people and they did not confirm to our new IT standards. They called VMS like this “Shadow IT”. Our internal IT spent a few million dollar developed a replacement and it was pushed to many plants. It caused a lot of trouble to the business and headache to the manufacturing IT operations. Because the new system did not well, we had to keep the old vendor system running in a “passive mode” in case the new system broke. It was also needed to run data collection layer, the barcode system, and to provide data via SMS to the phones and emails. When our new system acted up (which happened a lot especially in the early years), we would quickly switch back to the old “passive” vendor system. We ended spending a lot of more money and manpower plus it tarnished IT’s reputation.

The last time (~7-8 years ago) I heard about the VMS and its vendor was when a friend mentioned to me my former company had decided they would retire the new corporate system and reverse back to the old vendor system (which was never truly replaced anyway). They announced the older vendor VMS the new corporate standard and called it “strategic”. The vendor had to hire a couple more engineers to support the added scope.

5. Stumbling Onto a Goldmine – Joe Raymond

One sunny afternoon in the late-80s (more than a decade after the Interstate Stores transaction), Larry and Nate were having lunch together on Long Island.

After eating, Nate asked Larry if they could swing by the bank so he could make a deposit. Larry was enjoying the good weather and friendly company. “Sure,” he said, “Let’s do it.”

They walked into the bank and up to the counter to grab a deposit slip.

Larry noticed on the counter a copy of the bank’s most recent quarterly balance sheet. It was one of the cleanest, most secure bank balance sheets he’d ever seen…

…He looked around the lobby and saw on the other side of the room a thick wood door with a big brass knob and the word “PRESIDENT” emblazoned across the front…

…A man in a suit opened the door and asked how he could help.

“You have a beautiful balance sheet,” Larry said. “I’d love to know how and why this came to be, and if there are any other banks out there like yours!”

The president invited Larry into his office and explained to him how the bank had recently converted from a mutual to a stock bank….

…Imagine a make-believe mutual bank with $1 million of tangible equity. Let’s say this bank wants to convert to a stock bank and offers 100,000 shares at $10 per share in an IPO. Only depositors are invited to participate in the offering.

On a pro forma basis, the converted bank will have $2 million of tangible equity (the original $1 million plus the $1 million of IPO proceeds), which equates to $20 per share of tangible book value ($2 million of equity divided by 100,000 shares).

As an IPO investor, you were able to purchase the shares at $10. You paid only 50% of tangible book value…

…The president explained all of this to Larry, including how he himself had made a killing on the bank’s conversion…

…”You should check out this little bank in Queens,” he said. “They are preparing for a conversion themselves, and I think it will be a good one.”

That little bank in Queens was called Jamaica Savings Bank. And JSB ended up being a killer investment…

…Less than two years later, on June 24, 1990, JSB Financial went public. Santa Monica bought 59,000 shares at $10 per share for an initial investment of $590,000. The pro-forma book value was $21 per share (0.48x P/TBV)…

…The shares shot up 30% to $13 right after the IPO. Many investors sold for a quick profit. Larry decided to hold on as he saw a bigger pot of gold down the line…

…BVPS could be north of $25 within three years and the company would be worth $35 per share to a strategic buyer at 1.4x TBV. This works out to a 51% annualized return over three years.

Given the nature of the balance sheet (liquid, overcapitalized, and invested primarily in short-term government securities), the downside was minimal.

Thus, Larry found himself in investment nirvana: low downside paired with big upside.

JSB became an avid repurchaser of its own stock, buying back 7% of its outstanding shares in 1991 and another 8% in 1992…

…The share count was further reduced by 10% in 1993, 9% in 1994, 2% in 1995, and 7% in 1996. Shares outstanding fell by a cumulative 38% from 1990 to 1998. And most of these buybacks were done at or below tangible book value…

…JSB entered a stock-for-stock merger with North Fork Bank (NFB) in 1999. Every one share of JSB received three shares of NFB…

…As for Larry, he held onto his stock until NFB sold to COF, at which point he elected to receive cash. The $590,000 investment in 1990 turned into more than $5.5 million in 2006.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Adobe, Mastercard, Meta Platforms, Microsoft, and Visa. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 26 October 2025:

1. Sanity Check – The Brooklyn Investor

When I say that if 10-year Treasury yields stay at around 4%, then the market should average a P/E of around 25x over time, it sounds crazy, but given that the market has traded at 22-23x P/E in the last 35 years, this is not so crazy to me.

Of course, one can come back to me and say, well, as much as 22-23x P/E was shocking in 1990, who’s to say that the market can’t shock us again, going back to 6-7% interest rates and 14x P/E ratio over the next 20? This is also true. I can’t say that can’t happen. But I’ve always said that I think 4% or so 10-year rate seems reasonable given 4% nominal GDP growth over time.

So, given that, how does the market look today? The market today looks like it is priced correctly. The 10-year Treasury rate is 4% today, and the S&P 500 index P/E is 25.5x, almost exactly where it should be according to the model. Next year’s estimate P/E is 22x.

In past bubbles, the rubber band was stretched. The table below is from an earlier post. Just before Black Monday, the rubber band was stretched as 10-year rates spiked to close to 10% while the earnings yield declined to 4.7%, creating a near 5% gap. On a price basis, the market was overvalued by 100%! During the internet bubble, the gap increased to 1.5% and the market was overpriced 40%. Today, there is no stretch in the rubber band…

…So, the other thing is all this talk about an AI bubble. It is really interesting and I have no idea what is going to happen. But there seems to be two extreme views that both can’t be right. On the one hand, some people fear all these trillions being invested into AI infrastructure (energy, data centers etc.) will not offer decent returns on investments as there is still very little revenue associated with many of these big AI models. On the other hand, there is a big fear that AI will wipe out entire industries. There are already reports that huge increases in productivity is being actualized in the coding world, so much that the word is that entry level computer science positions are completely gone, wiped out. Big tech have also been firing a lot of engineers as they are replaced by AI.

They both can’t be right.

Here’s some bogus math, just as a sanity check too. Let’s say AI replaces 10% of jobs in the U.S. There are 160 million workers in the U.S. Ridding 10% of them is 16 million jobs gone. Many of these replaced jobs will be office jobs (well, AI will eventually replace Uber and truck drivers, farmers, factory workers too). Let’s say office workers cost companies $100K / year, including benefits. I’ve heard this somewhere before. That’s $1.6 trillion in expenses that you can cut. How much would you be willing to invest to cut $1.6 trillion?

You are now talking about trillions of dollars in investments. Now, all those numbers people throw around don’t sound so silly anymore. Of course, you can’t just say spending $10 trillion to eliminate $1.6 trillion is a 16% return on investment, as AI costs money to keep running / maintaining and you need to replace servers every 3-5 years etc. But still, you start to see the magnitude of what can happen if AI really starts to replace workers.

2. How Silver Flooded the World – Tomas Pueyo

A century earlier, around 1350, the Black Death spread across Europe, killing 25-50% of its population.

Men were dying, but coins were not.—David Herlihy

For a century, there was more coin than people, so they didn’t notice when silver and gold production slowed down. But it did; first, because of fewer miners.

Second, because mines ran out of gold and silver.

Third, the supply of gold from Africa collapsed after the Mali Empire civil war in the 1360s and the Songhai Empire instability.

Fourth, mines in southeastern Europe, in Serbia and Bosnia, fell to the Ottoman Empire.

So new sources of silver and gold shrunk. Meanwhile, silver sinks continued. Europeans kept buying Chinese silks, Indian cotton cloth, dyes, and spices, Middle Eastern sugar and drugs… But Europeans had little to export: wine, slaves, wood, salt, and little more. Italian traders paid one third in merchandise and two thirds in precious metals.

As silver and gold became scarcer, people started debasing the currency: Diluting it with other metals, clipping its edges…

Between the Black Death, the scarcity of metals, the debasement of currency, the incessant warfare, and taxes, people did everything they could to hoard and hide their precious metals, whether through hidden coins, filled chests, plates, and any other conceivable way…

…When we say some resource is exhausted, what we generally mean is… with current technology. People abandoned mines when they couldn’t figure out how to reach more ore, or when they couldn’t get more metal out of them.

One of the most typical issues was that ore is in mountains, but mountains also have something else: rain. Mining shafts would get flooded, so mining was restricted to the surface…

…Romans knew about waterwheels and pumps, but they never used them for extracting water out of mines. Central Europeans put them together into ever more complex systems to dry up mines and extract more ore…

…But there were two significant innovations that allowed Europe to increase its silver production by 5x between the 1460s and the 1540s.

Both innovations were new processes to extract more silver from ore. The first one is called liquation, and was first discovered in southern Germany in the mid-1400s, just as the Great Bullion Famine was hitting hardest. Of course, that’s not a coincidence: It was the bullion famine that was spurring mining innovation. Within 15 years, it had spread throughout Germany, Poland and the Italian Alps…

…We’ve already explored how Portugal’s discovery of an alternative path to Asia was made to bypass the Ottomans, who had taken control of Istanbul and blocked Christian trade through the Silk Road. But that trade still required gold and silver, and Europe didn’t have any. So the Portuguese were also looking for gold and silver deposits to mine. They found some in Western Africa—remember the Mali Empire—but that was not enough.

Now you know why Spanish Conquistadors were so obsessed about finding gold and silver in the Americas. It was not just a matter of greed. It was an existential matter for Europeans after the Great Bullion Famine. This is why Columbus mentioned gold 65 times in his diaries!

Spaniards didn’t find much gold in the Americas, but they did find silver. Unfortunately, the high-quality silver ore quickly ran out, and Spaniards were left with ore that didn’t contain enough silver to be extracted.

That’s when they invented a new technique to get more silver from the lower quality ore: amalgamation, via the patio process.

3. Microsoft’s Cloud & AI Head on the AI Buildout’s Risks and ROI (Transcript Here) – Alex Kantrowitz and Scott Guthrie

Alex Kantrowitz: I totally understand that, but I have to go back to the diminishing returns of training question. Where do you stand on that?

Scott Guthrie: If you look at training broadly, I think you’re going to continue to see more value from the models by doing more training. But going back to my answer earlier, I don’t know if that’s always going to be pre-training. I think increasingly lots of post-training activities are going to significantly change the value of the model. By post-training I mean take the base model and how do you add financial data or healthcare data or something that’s very specific to an application or a use case.

What’s nice about post-training is that you don’t have to do it in one large data center in one location. Part of the technique that we’ve been focused on is how do we take this inferencing capacity around the world and a lot of it is idle at night as people go to sleep. How are we doing increasingly post-training in a distributed fashion across many different sites? Then when employees come to work in the morning, we serve the applications. Having that kind of flexibility and being able to dynamically schedule your AI infrastructure so that you’re maximizing revenue generation and training ideally in a very swappable dynamic way—I think is one of the things we’re investing in heavily and I think is one of the differentiators for Microsoft.

Alex Kantrowitz: Okay, but you’ll forgive me for going back to this scaling pre-training question. I’m just trying to see what you believe here. You haven’t said it outright, but from your answers, it does seem to me like you believe that spending wildly on scaling pre-training is a bad bet.

Scott Guthrie: I wouldn’t necessarily say that. I think we’ve definitely seen as the scale infrastructure for pre-training has gotten bigger, we are seeing the models continually improve and we’re investing in those types of pre-training sites and infrastructure. We recently, for example, announced our Fairwater data center regions around the US. We have multiple Fairwaters. We did a blog post recently of one of our new sites in Wisconsin. These are hundreds of megawatts, hundreds of thousands of the latest GB200s and GB300 GPUs. We think the largest contiguous block of GPUs anywhere in the world in one giant training infrastructure that can be used for pre-training. We’re investing heavily in that, as you could see from the photos from the sky in terms of massive infrastructure. We do continue to see the scaling laws improve.

Now will the scaling laws improve linearly? Will they improve at the rate that they have? I think that is a question that everyone right now in the AI space is still trying to calculate. But do I think they’ll improve? Yes. The question really around what’s the rate of improvement on pre-training? I do think with post-training, we’re going to continue to see dramatic improvements. That’s again why we’re trying to make sure we have a balanced investment both on pre-training and post-training infrastructure.

Alex Kantrowitz: Just to parse your words here, you can see improvement by doubling the data center, but that’s why I use the word bet—because are you going to get the same return if it doesn’t improve exponentially and just improves on the margins? That I think is the big question right now, right?

Scott Guthrie: It’s a big question. The thing that also makes it the big question is it’s not like a law of nature that’s immovable. There could be one breakthrough that actually changes the scaling laws for better, and there could be a lack of breakthroughs that means things will still improve but do they improve at the same rate that they historically did from a raw size and scale perspective? That is the trillion dollar question…

…Scott Guthrie: Yeah, going back to the comments we had earlier on balance, I think as you think about your GPU buildout, one of the things that we think about is the lifetime of the GPU and how we use it. What you use it for in year one or two might be very different than how you use it in year three, four, five, or six. So far we’ve always been able to use our GPUs, even ones that we deployed multiple years ago, for different use cases and get positive ROI from it. That’s why our depreciation cycle for GPUs is what it is…

…Scott Guthrie: If you are for example building one large data center that only does training and it’s not connected to a wide area network around the world that’s close to the users, it’s hard to use that same infrastructure for inferencing because you can’t go faster than the speed of light. Someone elsewhere around the world that wants to call that GPU—if you don’t have the network to support it, you can’t use it for those inferencing needs…

…Alex Kantrowitz: Okay. All right. It’s good to get something definitive on that. You mentioned your 39% Azure growth. I’m looking at your quarterly numbers every quarter and often talking about them on CNBC and the numbers are massive. The other side of it though is that’s spend coming from clients, right? There have been multiple studies that have come out recently that have talked about how enterprises aren’t getting the ROI that they’ve anticipated on their AI projects yet. When you see those studies, do they ring true to you? How do you react to them?

Scott Guthrie: I think when you say AI in general, it’s a very broad statement.

Alex Kantrowitz: This is in large part generative AI where companies everywhere have tried to adopt LLMs and try to put some version of that into play. It’s not recommender engines basically.

Scott Guthrie: But I think what you need to do is double-click even further from GenAI to GitHub Copilot or healthcare or Microsoft 365 Copilot or security products built with GenAI. I do think ultimately, the closer you can double-click on is this really delivering ROI, then you have much more precise data.

I do think a lot of companies have dabbled or done internal proof of concepts and some of them have paid off and some of them haven’t. But I think ultimately a lot of the solutions that are paying off that we continually hear from our clients and our customers are a bunch of the applications that we’ve built. Similarly, a bunch of the applications that our partners have built on top of us. Ultimately the Azure business is consumption-based, meaning if people aren’t actually running something, we don’t get paid. It’s not like they’re pre-buying a ton of stuff. We recognize our revenue based on when it’s used.

The good news is when you look at our revenue growth, it’s not a bookings number. It’s actually a consumption number. You can tell that people are consuming more. The last two quarters, our revenue growth has accelerated on a big number. That is a statement of the fact that I think people are getting a lot of ROI, at least with the projects that they’re running on top of our cloud…

…Scott Guthrie: I think increasing the number of tokens you can get per watt per dollar is going to be the game over the next couple years. Maximizing the ability of our cloud to deliver the best volume of tokens for every watt of power, for every dollar that’s spent—where the dollar is spent on energy, it’s spent on the GPUs, it’s spent on the data center infrastructure, it’s spent on the network, and it’s spent on everything else—is the thing that we’re laser-focused on. There’s a bunch of steps as part of that, GPUs being a critical component of it.

One of the things that our scale gives us the ability to do is to invest for nonlinear improvements in that type of productivity and that type of yield. If you’ve got a million dollars of revenue on a couple hundred GPUs, you’re not going to be investing in custom silicon. When you’re at our scale, you will be. You’re not just investing in custom silicon for GPUs for pre-training or for inferencing. You’re looking at what could we be doing for synthetic data generation with silicon. What can we be doing from a network compression perspective with custom silicon? What can we be doing from a security perspective?

We have bets across all of those, many of which are now in production and are actually powering a lot of these AI experiences. In fact, I think every GPU server that we’re running in the fleet right now is using custom silicon at the networking, compression, storage layer that we’ve built. The GPUs themselves are also going to be a prize that people are going to try to optimize—the actual instructions for doing the GPUs.

Nvidia is a fantastic partner of ours. We’re probably one of, if not the biggest customer in the world of theirs. We partner super deeply with Jensen and his team. At the same time, and partly why they’re so successful is they’re executing incredibly well. If you look at the history of silicon, it’s rare to have a silicon company that every single year is doing the absolute perfect work that’s differentiated. Kudos to Jensen for what he’s done, and I know he’s going to keep trying to do it going forward. But there will be other opportunities from other companies where people are going to look for a niche that’s going to be big enough in this AI space to be truly differentiated versus what Nvidia is delivering. Then we’re doing our own silicon investment in-house because we’re going to be going after those same opportunities.

Ultimately, the way we’ve tried to build our infrastructure, none of our customers know when they’re using Microsoft 365 or GitHub or any open models what silicon they’re running on. We’re going to be constantly tuning the use cases based on the applications. If we find ways that are breakthroughs, we’re absolutely going to be taking advantage of them for those use cases. At our balance of scale and our balance of use cases, I’m very confident that we’re going to find use cases where custom silicon will make a difference. I’m also very confident we’re going to continue to be a great partner to Nvidia and others in the world that are going to be selling us great solutions.

4. The coming debt deluge? – Abdullah Al-Rezwan

For example, last week Meta entered in a Joint Venture (JV) with Blue Owl Capital for their $27-Billion Hyperion Data Center campus, of which Meta will own 20% and the rest will be owned by funds managed by Blue Owl Capital. Meta is signing an “operating lease” with an initial term of only four years. They have the option to extend the lease every four years, but they are not obligated to.

To persuade the JV to accept the short four-year leases, Meta provided a “Residual Value Guarantee” (RVG) covering the first 16 years of operations. If Meta decides to leave (by not renewing or terminating the lease) within the first 16 years, they guarantee the campus will still be worth a certain amount of money (undisclosed). This payment is “capped” i.e. there is a pre-agreed maximum limit to how much Meta would have to pay. Again, we don’t know the exact capped limit in this deal.

The structure of this deal, featuring short 4-year leases combined with a long-term RVG on a highly specialized asset, closely resembles a financial tool known as a Synthetic Lease.

In a synthetic lease, the tenant (Meta) gains the flexibility of short commitments and favorable accounting treatment (keeping the debt off their balance sheet). However, to convince investors (Blue Owl Capital) to fund the construction, the tenant must assume the majority of the financial risks of ownership. The RVG achieves this risk transfer. To secure financing for such a massive, specialized asset, this cap must be set very high. While we don’t know the exact number, my guess is it’s likely somewhere between 80% to 90%. If we assume it to be 85%, for the $27 Billion Hyperion campus, Meta’s maximum possible exposure is $22.95 Billion.

If Meta decides to terminate the lease within the 16-year RVG period, the payout is determined by the following calculation:

Guaranteed Value at time of exit – Actual Market Value = Shortfall

Meta pays the shortfall, but only up to the agreed-upon cap (estimated at $22.95B)…

…Given Meta’s backing, the bonds issued to fund this investment received investment grade credit rating. However, the bonds were issued at 6.58% yield which is closer to junk bond yield.

Why is the yield so high? If the value of the data center catastrophically collapses due to obsolescence or for some other reasons, Meta’s RVG covers most of the loss, but the investors bear the portion exceeding the cap. Moreover, the debt belongs to the project entity, it is “structurally subordinated” to Meta’s own corporate debt. Investors demand a higher yield to compensate for this “tail risk”.

More importantly, the underlying collateral is a hyper-specialized AI data center. If Meta leaves, it’s likely that the facility cannot be easily repurposed. While the RVG mitigates the financial loss, the specialized nature of the underlying asset still influences the perceived risk and pushes the yield higher.

My guess is Meta (and other big tech) will do more of these deals going forward. In fact, just yesterday, Oracle appears to be raising debt even larger than Hyperion deal: $38 Billion for building data centers in Texas and Wisconsin. If the deal goes through, it would be the largest debt deal so far in AI infrastructure.

5. Thoughts on the AI buildout – Dwarkesh Patel and Romeo Dean

With a single year of earnings in 2025, Nvidia could cover the last 3 years of TSMC’s ENTIRE CapEx.

TSMC has done a total of $150B of CapEx over the last 5 years. This has gone towards many things, including building the entire 5nm and 3nm nodes (launched in 2020 and 2022 respectively) and the advanced packaging that Nvidia now uses to make datacenter chips. With only 20% of TSMC capacity1, Nvidia has generated $100B in earnings…

…Further up the supply chain, a single year of NVIDIA’s revenue almost matched the past 25 years of total R&D and capex from the five largest semiconductor equipment companies combined, including ASML, Applied Materials, Tokyo Electron…

…For the last two decades, datacenter construction basically co-opted the power infrastructure left over from US deindustrialization. One person we talked to in the industry said that until recently, every single data center had a story. Google’s first operated data center was across a former aluminum plant. The hyperscalers are used to repurposing the power equipment from old steel mills and automotive factories.

This is honestly a compelling ode to capitalism. As soon as one sector became more relevant, America was quickly and efficiently able to co-opt the previous one’s carcass. But now we are in a different regime. Not only are hyperscalers building new data centers at a much bigger scale than before, they are building them from scratch, and competing for the same inputs with each other – not least of which is skilled labor…

…Labor might actually end up being the most acute shortage – we can’t simply stamp out more workers (at least, not yet).

The 1.2 GW Stargate facility in Abilene has a workforce of over 5,000 people. Of course, there will be greater efficiencies as we scale this up, but naively that looks like 417,000 people to build 100 GW. And that’s on the low end of 2030 AI power consumption estimates. We’re gonna need stadiums full of electricians, heavy equipment operators, ironworkers, HVAC technicians,… you name it.

For reference, there’s 800K electricians and 8 million construction workers in the US…

…Anthropic and OpenAI’s combined AI CapEx per year (being done indirectly, mostly by Amazon and Microsoft in 2025) seems to be around $100B.

Revenues for OpenAI and Anthropic have been 3xing a year for the past 2 years. Together, they are on track to earn $20B in 2025.

This means they’re spending 5 times as much on CapEx as they’re earning in revenue. This will probably change over time – more mature industries usually have CapEx less than sales. But AI is really fast growing, so it makes sense to keep investing more than you’re making right now.

Currently, America’s AI CapEx is $400B/year. For AI to not be a bubble in the short term, the datacenters currently being built right now need to generate $400B in revenue over their lifetime. Will they?…

…Do you think that AI models will be able to do much of what a software engineer does by the end of a decade? If the 27M Software engineers worldwide are all on super charged $1000/month AI agent plans that double their productivity (for 10-20% of their salary), that would be $324B revenue already…

…A key question is whether datacenters will go “off-grid”—generating power on-site rather than connecting to the utility grid. Some of the largest datacenters are already doing this, e.g., Meta’s Orion or XAI’s Colossus.

Why would datacenters want to make power themselves rather than relying on the grid? They’re trying to get around interconnection delays. Connecting large new electricity sources to the grid now takes over 5 years…

…What will the distribution of individual datacenter sizes be? Here’s the argument for why we might end up seeing what looks like a thick sprinkle of 100 MW datacenters everywhere:

• If you can plop down a medium sized datacenter here and there, you can soak up any excess capacity in the grid. You can do this kind of arb with a 100 MW datacenter, but there’s no local excess capacity in the grid at the scale of 1 or 10 GW – that much power is on the scale of a whole grid itself.
• For pretraining like learning, you want to have large contiguous blobs of compute. But already we’re moving to a regime of RL and midtraining, where learning involves a lot of inference. And the ultimate vision here is some kind of continual learning, where models are widely deployed through the economy and learning on the job/from experience. This seems compatible with medium sized datacenters housing 10s of thousands of instances of AIs working, generating revenue, and learning from deployment.

Here’s the other vision. 1-10 GW datacenters, and then inference on device. Basically nothing in between.

• If we move to a world with vertically integrated industrial scale production of off-grid datacenters, maybe what you want to do is just buy a really big plot of land, build a big factory on site to stamp out as many individual compute halls and power/cooling/network blocks as possible. You can’t be bothered to build bespoke infrastructure for 100 MW here and there, when your company needs 50 GW total. A good analogy might be how a VC with billions to deploy won’t look at any deal smaller than deca millions…

…Why doesn’t China just win by default? For every component other than chips which is required for this industrial scale ramp up (solar panels, HV transformers, switchgear, new grid capacity), China is the dominant global manufacturer. China produces 1 TW of solar PV a year, whereas the US produces 20 GW (and even for those, the cells and wafers themselves are manufactured in China, and only the final module is assembled in the US).

Not only does China generate more than twice the electricity than the US, but that generation has been growing more than 10 times faster than in the US. The reason this is significant is that the power build out can be directed to new datacenter sites. China State Grid could collaborate with Alibaba, Tencent, and Baidu to build capacity where it is most helpful to the AI buildout, and avoid the zero-sum race in the US between different hyperscalers to take over capacity that already exists.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Alphabet (parent of Google), Amazon, ASML, Meta Platforms, Microsoft, and TSMC. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 19 October 2025:

1. Why AI Is Not a Bubble* – Derek Thompson

The dot-com bubble was genuinely insane. The internet companies didn’t have real revenue, and the telecom firms didn’t have real users. In its first fiscal year, Pets.com earned less than $700,000 in revenue and spent nearly $12 million on advertising. Telecom firms laid so much fiber-optic cable that as late as 2005, 85 percent of broadband capacity in the U.S. was going unused.

Today’s AI boom is nothing like that. The modern hyperscalers are among the most profitable enterprises ever. The eight biggest tech firms—the Magnificent Seven plus Broadcom—now account for 37 percent of the S&P 500 and are expected to grow profits by 21 percent this year. Return on equity for the S&P 500, around 18 percent, is the highest since at least 1991—and achieved with less leverage than the late 1990s. These are not fragile companies playing with borrowed money…

…So, you could boil down the whole “is AI a bubble?” thing to one simple question: Where’s the cash?

The answer is that three years ago, it was nowhere, and now it’s surging. According to Azhar, generative AI revenue has grown by ninefold in the last two years…

…What we’re really interested in is revenue that comes in from new businesses and customers. This comes from three sources.

• The first source is internal: Hyperscalers using their own AI to make money from their existing businesses, such as Meta using its AI to sell $1 billion more in ads (or, just as good for cash flow, using AI to save $1 billion).
• The second source is external: Companies like OpenAI and Microsoft getting money from companies that use their AI, such as a legal AI firm building a bespoke model off of GPT-5.
• The third source is novel: Using AI to create a business that doesn’t yet exist, like Tesla is trying (and mostly failing) to do with its fleet of Optimus robots.

The first two revenue sources seem to be firing on all cylinders. Microsoft and Amazon’s cloud divisions are surging as enterprise customers integrate generative-AI tools. Meta is using AI to sell more ads and to cut costs. Internal AI use (Meta’s ad tools, Microsoft’s Copilot, Amazon’s logistics optimization) plus external adoption (customers building on GPT-5 or Claude) means two of Azhar’s three revenue streams are already working.

The most important test for whether AI is a bubble is what you could call the “Triple-Digit Test.” The TDT says: If AI revenue grows more than 100 percent annually (or, even better, 200 percent) for the next few years, there probably won’t be a huge bubble pop. So far, that’s happening—or, at least, the biggest AI investors claim that it is happening.

• Microsoft says its AI business has surpassed a $13 billion annual run rate, up 175 percent year-over-year.
• Amazon claims its AI revenue is growing at “triple-digit percentages” year over year.
• The VC firm Menlo Ventures estimates OpenAI, Anthropic, Scale AI, and Perplexity are all doubling or tripling annual revenue, which means they’re growing at 100 percent or more…

…AI’s revenue problem is really a white-collar workforce creativity problem. It’s notable that AI’s profitability doesn’t just depend on how fast frontier labs innovate. It depends on how creatively their customers use the tools, as Smith points out. But most of the world’s companies are not prompt-engineering wizards. Researchers at Harvard and Stanford recently found that many firms are misusing AI so badly that workers spend more time fixing AI-generated “workslop” than doing their jobs. If that pattern persists, AI will increase frustration rather than productivity, and a slow adoption curve will blow up the Triple-Digit Test and make these companies vulnerable to a major correction in valuation or investment.

If everybody builds the same thing, where’s the moat? Something I still can’t figure out is how all of these companies are going to make money when they’re all building similar products. Anthropic’s Claude technology isn’t that different from OpenAI’s GPT technology, and when several companies build an interchangeable product, competition tends to drive down prices, which is great for consumers and bad for firms outlaying a trillion dollars to build that thing. Meanwhile, cheap Chinese or open-source models that are “99.8 percent as good for a tenth of the price,” as Smith writes, could commodify the industry overnight. In that world, the real winners wouldn’t be OpenAI or Anthropic but the downstream companies that use cheap AI to build their own moat to get rich.

2. Is AI eating Vertical Market Software? – Best Anchor Stocks

Much like some people already believed, AI has not “eaten” VMS software (at least not yet). Mark Leonard started the call with a powerful story that demonstrate his integrity and also the fact that some people might be running ahead of themselves with their forecasts:

In 2016, Geoff Hinton made a long-term forecast. For those of you who don’t know him, Geoff is known as the godfather of AI and is a Nobel Prize winner for his work in the field. And long-term forecasting is very difficult. I talked about this before, and I’m happy to send you some sources/information if you’d like to delve into that further.

Geoffs’s forecast in 2016 was that radiologists were going to be rapidly replaced by AI, and specifically, he said people should stop training radiologists. In the intervening nine years since he made that forecast, the number of radiologists in the US has increased from 26,000 (these are US board-certified radiologists) to 30,500 or a 17% increase. Now that’s outpaced the population growth in that period. So the number of radiologists per capita is up from 7.9 to 8.5. Now, Geoff wasn’t wrong about the applicability of AI to radiology. Where he was wrong was that the technology would replace people. Instead, it has augmented people. The quality of care delivered by radiologists has improved. And the number of practicing radiologists has increased.

So I told you this story to make two points. Firstly, you and I will never know a tiny fraction as much about AI as Geoff did. And secondly, despite his deep knowledge of AI, he was unable to predict how it would change the structure of the radiology profession…

…Management also discussed how the company leverages LLMs and how their strategy protects them from worsening unit economics. Both things are related, so let’s start first with how Constellation has structured its access to LLMs to avoid being “price-gouged:”

So we’ve essentially created our own centralized sort of platform that essentially removes the various factions that are currently going on where to a certain extent, you have to largely be within this cloud provider to have access natively to this LLM and so on and so forth. So there’s these turf wars being kind of created across the various cloud providers and whatnot.

And so with our strategy has been to really play a very neutral sort of Switzerland type role, where by centralizing things through strategic relationships, either directly with the model providers or with the platform providers and so on and so forth. We’ve managed to negotiate, I think some, some, some really aggressive deals and remove the element of these sort of factions. They’re all willing to kind of play nice with us in the sandbox. So that puts us in a very unique position where sort of technically we have access to 15,000 sort of unique models. And that’s because we’re essentially sort of coalescing sort of anything that otherwise couldn’t be or reside within other platforms. The other piece that sort of I had touched on very briefly, and Paul sort of alluded to as well, is sort of using a on prem based assets where and when possible.

So to the extent that the LLM needs to be or the AI model needs to be hyper specific or, you know, a specific trained one that it resides with a pre-existing best of breed provider, then sure, that may make sense to kind of tap into that one, but for basic, let’s say sort of translation service, summarization, service, and a myriad of other hosts of functionality and whatnot, you know, the on prem one is plenty sufficient and capable of doing it’s, you know, its own sort of thing.

This flexibility basically means that CSU will benefit from price wars across the different LLMs (which they expect will happen) and will also be able to take advantage of their on-prem infrastructure to lower costs for consumers (when able to).

3. An Interview with Gracelin Baskaran About Rare Earths – Ben Thompson and Gracelin Baskaran

GB: We have so much supply on the market right now, and that’s really coming from China, they keep overproducing and it’s actually forcing western companies out of operation. So to put this into context for you, in the last three years, lithium prices have fallen by 85%, nickel prices by 80%, and cobalt by 60%. So companies are struggling to operate at a time when we know we need a lot of these materials because the economics of it aren’t checking out.

And is this overproduction on purpose? Is it to knock out all these western companies to result in China dependence?

GB: I can tell you one thing is Chinese companies aren’t operating profitably by-and-large either, but they are willing to absorb long-term losses in order to gain a strategic monopoly on a lot of these sectors.

I’ll give you an example: Chinese companies in Indonesia five years ago were producing about 500,000 tons of nickel a year, now they’re producing over 2.5 million tons a year, and what’s happened is nickel prices have fallen so much that BHP, an Australian company, has closed their operations in Australia and Glencore, a Swiss company, has closed theirs in New Caledonia.

So that dominance, that willingness to absorb loss, has given them a dominance and the ability to weaponize minerals and cut us off…

…But I want to go back to your question about rare earths, there’s two things that are important. First of all, rare earths are not actually rare, they’re everywhere, but finding them in these large scale quantities that are again economically viable is actually much harder, number one.

Number two is you can mine rare earths in a lot of places, but we don’t actually process rare earths or we historically have not, which it means that no matter where the rare earths are mined — I mean, even this year until February, the rare earths that we mined in California still went to China for that processing phase, so that allowed them to build that dominance. But it’s a very small market. I need a ton of lithium, I need a ton of cobalt, rare earths are actually a small market…

…What is it about them that makes them so useful?

GB: They are the most powerful permanent magnet, which is actually really important. If you put a really good permanent magnet next to a fridge, you would basically pull the fridge off, so for defense technologies in particular, there’s nothing really that you can substitute at this point.

Got it. So is it just the magnetic properties or are there — for example, what’s their use in chips? I know particularly as chips have become more advanced, there’s questions about rare earths in there. Is that a magnetic thing or are there other properties as well?

GB: Rare earths are actually used in advanced semiconductors including memory chips and logic chips and actually when you look at the most recent export restrictions that China has applied, they’re actually reviewing the semiconductor end use on a case by case basis…

...Right. So what’s the trade-off there? Because you see numbers and you reference this before, I think China actually mines 60 to 70% of rare earths, but the actual processing is well over 90%. So what’s the bigger hole for us here? Is it the actual acquiring the rare earths or is it the processing/refining?

GB: Our big chokehold is processing because I can get rare earths from other places. So for example, now we are putting US government financial support, not just at mines, for example, Mountain Pass here in the United States, but we’re also providing financing to a project in Brazil that has rare earths. You can source feedstock from a variety of places, but it doesn’t actually matter when it goes back to China because then China can cut us off and we don’t have any of it. So we’ve got to build those processing capabilities here or else it’s like we never have access to them anyway.

So how does that happen? Is this an issue where basically there needs to be some combination of tariffs? Does China imports need to be blocked? There need to be a guaranteed price floor? How do you make the economics work? And you mentioned that these massive permitting issues when it comes to mines, is it better or worse when it comes to building these processing facilities?

GB: So really what we need is an all-of-the-above approach, and here’s what I mean by that. Again, it varies by commodity, the reason you need a price floor is this is when that US, the Department of Defense and MP Materials deal was signed earlier this year, which had a lot of support mechanisms. NdPr, neodymium-praseodymium oxide, which is one of our key rare earth compounds, was about $54 a kilogram. So at that price point, by 2030, there would only be eight projects outside of China that could even break even with their production costs because it was so commercially unviable. What you need in that case is you do need a price floor because what I don’t want is I don’t want my Western companies to go bankrupt or have to stop operating because prices are so low, and we’ve already seen it happen. In 2023, the United States opened its only cobalt mine and it closed it in the same year because prices had fallen so much. So we complained, we’re like, “Oh, I don’t want to lean into the Congo”, but we couldn’t keep our own mine open. We don’t want that to replicate for rare earths, so part of the story is a price floor story and the reality is you shouldn’t need a price floor forever.

What we saw after that deal, General Motors signed offtake, you saw Apple sign offtake, and already those prices have gone from $54 to about $84 or $85, the price floor is $110. So I’ve already closed what my fiscal responsibility by over 50%. As there’s more demand for a reliable supply chain and companies are now willing to pay that premium. I’m a Midwesterner, and what we saw after the rare earth export restrictions in April was that Ford actually had to stop manufacturing its Explorer model in Chicago because it couldn’t access these materials, so of course now we’re willing to pay a bit more to know that I won’t have to stop producing. Price floor is one part, but I need more than that.

So other mechanisms that become really important is I need concessional financing. Capital markets, because a lot of the risks that we’ve talked about, often tend to view this sector as too risky to lend to, but when the US government provides cheaper financing, we also see that banks are more willing to invest in that because they see it as a key de-risking mechanism.

The third thing I would add is government offtake helps because you’re not going to be able to sell everything to an American firm and so what we’ve seen this government do is say, “Okay, well we want a stockpile”. The recent budget in the US included $2 billion for a stockpile because if there is a supply chain disruption, I want to have enough to cover our national and economic security insurance, so they can backstop that by buying some of it…

…Given the massive risk that is here, let’s sketch out that risk. What happens if China actually just cut off rare earths tomorrow? What happens?

GB: Our manufacturing stops. Even in April 4th when those restrictions hit, US government officials said, “Maybe we get to June before we run out”…

…GB: I can manufacture for days, but the US at the end of the day has less than 1% of the world’s nickel, cobalt, we have about 2% of the world’s rare earths, we have less than 1% of graphite, we are not going to win this race alone no matter how we cut the cake. The question is how do we form — I mean, think about it — we used to form strategic alliances over oil, our relationship with Saudi was the defense for oil agreement that kept our economy open for a long time. The question is, “How do we work with our partners in a way that our supply chains are as close to us as possible?”, but we can’t do it alone, God didn’t give us the rocks.

Which mineral is the hardest problem to solve of these?

GB: I would say that the most complicated mineral is probably actually rare earths, and there’s a few reasons for that.

Is there one specific rare earth in particular?

GB: The United States Geological Survey just undertook its review of what is a critical mineral, and of the 55 or so minerals, samarium is ranked number one. The reason samarium is number one is when I take out a ton of rock from the ground, there’s a different percentage of every mineral in that ton, and samarium is such a small percentage of that rock that and I need more of it than that percentage is in there. So samarium is our most critical, which means that it is a high likelihood that there’s a failure of that supply chain. Niobium is right up there and rhodium is up there, and rhodium is a platinum group metal. So you pull it out with platinum, tiny percentage. So that’s what I mean by geology, I can’t will myself to have more Samarium in a ton of rock.

4. My friend became a millionaire at 17, and I got two book recommendations – Thomas Chua

“Actually… something huge happened.”

He told me slowly, almost reluctantly. His dad’s boss had given his father a red packet for Lunar New Year—a pretty standard gesture in Singapore. Bosses give employees red packets during the festive season as a bonus, usually cash.

Sometimes, though, they don’t give cash.

Sometimes they give hope.

In his dad’s case, his boss had given him a lottery ticket—the Singapore Sweep, with a top prize of over $2 million.

His dad won.

My jaw dropped. My teenage brain couldn’t comprehend that level of luck.

Over $2 million. The boss had fought to get the ticket back—or at least demanded a portion of it. I never learned exactly how it ended, but his dad quit his job not long after, so I assumed he kept everything. The family became estranged from relatives who’d expected generosity with the windfall, who’d wanted their own slice.

My friend and his siblings each received a tidy six-figure sum from their dad.

His family became millionaires overnight.

Looking back now, I realize that the Chinese New Year was probably their last normal one as a family. The last time money was just money, not a test of relationships. The last time people showed up because they wanted to, not because they wanted something…

…At seventeen, that kind of windfall looks like freedom. Looks like every door opening at once. No more worrying about tuition, about scholarships, about starting life in debt. Just pure possibility.

But freedom from what, exactly?

My friend hadn’t built anything yet. Hadn’t struggled for anything. Hadn’t earned the quiet confidence that comes from overcoming something you weren’t sure you could overcome. He hadn’t had the chance to discover what he was capable of when things got hard.

And here’s the thing about struggles: they don’t just test you. They build you.

5. National Bank of Detroit – Joe Raymond

Long story short, Buffett, Munger, and Guerin acquired control of Blue Chip Stamps in the late ’60s. The main appeal of the stock was the cheap price in relation to the large amount of deferred revenue from stamp sales. By taking control of the company, Buffett & friends could invest this “float” in securities.

Blue Chip had $89 million of stamp-related float in March 1972, $134 million of securities, and $74 million of common equities…

…Buffett needed to keep Blue Chip’s balance sheet liquid enough to handle stamp redemptions, but he knew he could do better than short-term debt instruments. Instead, he bought a group of solid companies at reasonable valuations.

Nearly two-thirds of the stock portfolio was made up of 10 banks…

…Blue Chip got out of most of these stocks within a decade. Nevertheless, I thought it would be fun to go through each of these banks and see how things played out over the long run.

It turns out this is a great way to learn about bank investing…

…This post will be dedicated to Blue Chip’s biggest position in 1972 – National Bank of Detroit – which is an interesting (and moderately successful) story…

…In March 1972, Blue Chip owned 218,380 shares of NBD (3.64% of the total outstanding) worth nearly $11 million. This equated to about 8% of the securities portfolio, 15% of the stock portfolio, and 24% of Blue Chip’s common equity.

All of this is to say this was a sizable bet.

The average price in 1971 (when Buffett was buying) was $50 per share…

…So, NBD was a dominant regional bank with a 12%+ ROE trading at a discount to book value. Loans to deposits was less than 60%, with the rest invested in conservative securities.

The 10-year track record was satisfactory…

…By the mid-80s, many states had passed reciprocity laws allowing bank holding companies from approved neighboring states to buy or merge across state lines.

Merger mania ensued; NBD did its fair share, making dozens of acquisitions from the mid-70s to mid-90s.

Despite the feverish M&A activity, results weren’t bad.

Book value per share grew from $57.24 in 1972 to $237.66 by 1995 (6.4% CAGR). The company also paid substantial and growing dividends over this period.

Annual BVPS growth adjusting for dividends came in around 11-12%…

…In 1995, NBD completed an all-stock merger of equals with First Chicago Corporation. The two banks had complementary business lines in adjacent geographies. The surviving entity operated under the combined name First Chicago NBD.

Then in 1998 First Chicago NBD merged with Banc One – a Columbus, Ohio based bank. Every one share of FCNBD received 1.62 shares of Banc One and the combined company was renamed Bank One (with a “k” instead of a “c”)…

…But Bank One’s fortunes started to turn south in the late ’90s shortly after the merger.

Earnings fell sharply in 1999 as growth slowed and anticipated cost savings failed to materialize. The credit card division from Banc One imploded due to bad loans and regulatory scrutiny. The stock fell by 50%. Analysts described Bank One as “the sick man of big banking.”

In 2000, a young executive by the name of Jamie Dimon was brought in to right the ship.

And right the ship he did.

Dimon wrote off billions in bad loans and goodwill. He centralized operations and established new risk controls. Tech systems were updated and unified. The credit card business was rebuilt.

By 2003, Bank One stock had tripled from its 2000 low.

In 2004, JPMorgan Chase and Bank One decided to merge…

…Each share of Bank One received 1.32 shares of JPM. Dimon was made President and COO for a year before taking the CEO title in 2005 and Chairman in 2006…

…Every one share of National Bank of Detroit Buffett purchased in 1971, if he had held for the next 54 years, would have turned into 14.58 shares of JPMorgan Chase today (as a result of multiple stock splits and stock-for-stock mergers).

NBD traded for an average price of $50 per share in 1971 whereas JPM trades for $310 per share today. As such, every $1,000 invested in NBD 54 years ago would be worth a little over $100,000 today.

Buffett’s $11 million stake would have grown to more than $1.1 billion.

Astute readers will note that this “only” equates to a 9% annual return. The buy-and-hold investor would have also received growing dividends over the decades, pushing the total annual return into the low-teens.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Amazon, Meta Platforms, and Microsoft. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 12 October 2025:

1. GDP’s absurdity – Abdullah Al-Rezwan

To understand the flimsy nature of such belief, we first need to understand the nuances around calculating GDP:

Discourse over GDP is frequently confused because there are actually three different calculation approaches: the income approach, the expenditures approach, and the value-added approach.

Each approach has its uses, but you have to be careful with which you use. What percent of GDP is healthcare? You get two different numbers depending on the approach. With the expenditures approach, healthcare is 17% of GDP, but for the value-added approach only 8%. Why? Because the value-added approach only counts expenditures on hospital and clinic workers toward the healthcare category. Money spent on manufacturing medical devices counts as manufacturing; money spent building hospitals counts as construction. For measuring healthcare’s share of the economy, it is probably better to use the expenditures approach because it is reasonable to include pharmaceutical production and hospital electricity bills as part of healthcare.

GDP is a very complicated statistical construct that is made by government bureaucrats behind closed doors without any ability of the public to replicate, audit, or verify assumptions. Sometimes, these kinds of constructs can be useful for accurately representing real-world phenomena, like manufacturing capacity. But a dive into how the sausage is made makes clear that GDP is not one of them.

So, how is the sausage made here? It is particularly striking to take a look at manufacturing:

If you want to see what percent of the economy is manufacturing, and how that has changed over time, you can only use the value-added approach. Only the value-added approach separates out each step in the economic chain: from mining the iron ore to transporting it to the factory to manufacturing the product to selling it at the store. The value-added approach categorizes each step, so you can sum together just the increase in price from the manufacturing step across all categories of spending.

2. This Is How the AI Bubble Will Pop – Derek Thompson and Paul Kedrosky

Thompson: How do you see AI spending already warping the 2025 economy?

Kedrosky: Looking back, the analogy I draw is this: massive capital spending in one narrow slice of the economy during the 1990s caused a diversion of capital away from manufacturing in the United States. This starved small manufacturers of capital and made it difficult for them to raise money cheaply. Their cost of capital increased, meaning their margins had to be higher. During that time, China had entered the World Trade Organization and tariffs were dropping. We’ve made it very difficult for domestic manufacturers to compete against China, in large part because of the rising cost of capital. It all got sucked into this “death star” of telecom.

So in a weird way, we can trace some of the loss of manufacturing jobs in the 1990s to what happened in telecom because it was the great sucking sound that sucked all the capital out of everywhere else in the economy.

The exact same thing is happening now. If I’m a large private equity firm, there is no reward for spending money anywhere else but in data centers. So it’s the same phenomenon. If I’m a small manufacturer and I’m hoping to benefit from the on-shoring of manufacturing as a result of tariffs, I go out trying to raise money with that as my thesis. The hurdle rate just got a lot higher, meaning that I have to generate much higher returns because they’re comparing me to this other part of the economy that will accept giant amounts of money. And it looks like the returns are going to be tremendous because look at what’s happening in AI and the massive uptake of OpenAI. So I end up inadvertently starving a huge slice of the economy yet again, much like what we did in the 1990s…

…Kedrosky: The market is rewarding [the big tech companies] for investing in AI even though it makes no economic sense to spend at this level because there’s no way they can recoup the value of the capital spending over the next three years. So they’ll be forced to do these kind of wacky shell games where they say, “Well, the building itself will actually be valuable in five years, because it’ll still have energy, it’ll still have water, it’ll still be able to cool things, the walls will still be standing, and I’ll just swap out the GPUs.” But the problem is the GPUs are the majority of the cost. The shell is the thing I’d like to write off, since I don’t want to have to write off GPUs every three years. But they’re the majority of the cost of what we call a data center.

Unlike telecom, unlike the fiber boom, unlike in railroads, there are actually two assets here. One that’s long-lived, a building, which is essentially a small fraction of the cost of the center; and one that’s very short-lived, which is the GPUs, which are the thing we’d like to have last and don’t, yet represent as much as 60 percent of the cost of the data center. So there’s the perversity.

Thompson: I want to talk about how some of this might go badly in the next few years, and I want to preface that discussion by saying that when I talk about AI as a bubble, I think some people see me as being pessimistic about the technology. The railroads were a bubble. There was a panic of 1857, of 1873, and of 1893. There were constant railroad depressions, and also the railroads changed the world. Broadband was a bubble, it also changed the world. Big infrastructure buildouts that changed the world often passed through a bubble phase. So it’s not pessimistic to say that AI is currently in a bubble. You could say it’s actually historically in tune to say that we are very likely in the middle of a bubble, because every industrial revolution passes through bubble phases.

Let’s start here. How close are the hyperscalers—Meta, Google, Microsoft, the big boys—to getting AI revenue to match AI spending?

Kedrosky: Nowhere near.

The hyperscalers are spending as much as 50 percent of income on capital expenditures, which is unprecedented. This doesn’t happen. Normally, if I did that as Microsoft or Amazon, I would be taken to the woodshed and beaten by investors because that’s such an incredible investment on one narrow slice of CapEx. They’re not being punished for that.

What I’m watching is how they’re moving the financing off their balance sheet. That for me is a reflection of not wanting the credit rating agencies to look at what they’re spending. What we’re seeing is these SPVs— special purpose vehicles—being created. Meta has a stake, some giant private debt provider has a stake, and the data center at the end is under Meta’s control, but they don’t “own” it. And so it doesn’t go on their balance sheet in terms of assessing creditworthiness. We’re seeing for the first time over the last six, seven months, the beginnings of a wave of these special purpose vehicles and other more exotic financing structures. We’re seeing the equivalence of some of the old collateralized debt obligations emerge. These are all, for me, the beginning of the sign that the bubble is becoming tired because the market is beginning to punish—at least there’s a perception that the market will punish—if I continue to keep this on my income statement. So I move it somewhere else. And that makes the entire process much more opaque. That’s the thing to watch. How hard are they trying to hide the expenditure?

3. Why Warm Countries Are Poorer – Tomas Pueyo

Societies that live closer to the equator are warmer. Why are they also poorer?…

…Here’s the kicker—I’m so excited about writing this, I have a huge grin on my face right now: We did not evolve in such warm places, and humans in warm countries don’t live where you think they live!…

…Lisbon, the capital of the first global empire of the West, actually gets warmer than Nairobi! Nairobi’s temperature is not that high, and is quite stable throughout the year…

…The answer is obvious when you think about it: The higher you are, the cooler the temperature. Normally, temperatures decrease by ~4–9ºC every 1000 meters higher (2 to 5 °F/1000 ft). Since Bogotá is at 2,600 m of altitude (8600 ft), its annual temperature is 14ºC (25ºF) cooler than Barranquilla, which is farther north from the equator but at sea level, on the coast.

Bogotá was created far inland in the mountains in 1538, only a few decades after the Spanish discovery of America. The colonizers had a much harder time with disease and conflict in coastal flatlands. It was worth traveling hundreds of miles inland and up thousands of meters to survive. That region is agriculturally much better than the sea-level flatlands too, because of the same lack of disease and the soil that doesn’t get leached as much. This logic is true of all three main Colombian cities: Bogotá (12.7M people), Medellín (4.4M) and Cali (4.2M) are all in the mountains…

…Arguably, civilization would have had a much harder time developing in the Americas if the land had been much flatter and low-lying. It’s not a coincidence that the Incan Empire was a mountain empire and was the only independent one in the world to form on the equator!

Even today, the Latin American population concentrates in the Andes!…

…So the trend is clear that, closer to the equator, people tend to live in higher altitudes. What are the consequences of that?…

…Mountains mean people need to travel up and down mountain passes and huge slopes to get anywhere. They mean no navigable rivers. They mean much higher costs of infrastructure, so there’s much less of it. This means transportation costs are much higher…

…This, in turn, means there’s dramatically less trade, and so less money is made, and less wealth accumulated. We’ve seen how these facts have dramatically impoverished countries like Mexico and Brazil, and the generic process in A Science of Cities…

…The other thing that happens with mountains is conflict. As transportation costs are so much higher, people don’t move as much from their valley. There’s substantially less regional integration, and people trust and like each other less. They develop their own independent customs and mistrust those of their neighbors. This leads to more conflict between valleys, regions, and countries.

This process is called Balkanization, for the mountainous Balkans in Europe. But we also see it in Mexico’s and Colombia’s cartels—in fact, nearly all cartels in Latin America are in the mountains. We saw it in Iran, a highly mountainous country that requires a very strong state suppressing dissent to keep the country together…

…The pattern, and its logic, is unmistakable:

• Humans evolved in the African highlands, where temperatures are stable throughout the year, and close to that of spring & fall in temperate regions. This is why we feel most comfortable there.
• Close to the equator, if we’re not in the mountains, the temperatures are too high for us. We can’t think or work properly because we overheat, and our sweat can’t cool us off because humidity is too high.
• We also suffer from many more diseases, more common in hot moist climates, but also because we didn’t evolve there.
• This also affects food, as agriculture is much harder in these hot moist climates, given the pests, the speed of rot, and the work required by crops.
• This prevented maladapted Westerners from efficiently transferring culture and institutions to these hot, humid, low-lying areas, yet another way these regions suffered.
• In order to avoid all that, people close to the equator tend to live higher up, in mountains, where temperatures are cooler and the dew point is lower, allowing people to cool down with sweat when necessary.
• The big tradeoff for this comfort though has been much higher transportation costs, so less trade, so less wealth.
• This also leads to much more ethnic diversity.
• This diversity breeds conflict, which makes everybody poorer.
• Ethnic diversity and conflict also mean institutions are much harder to make and keep.

This is how mountains are the most significant underdiscussed topic in economic development, and how they must be considered to better explain why warmer countries are poorer.

4. How Misleading Headlines Frame the Narrative – Michael Batnick

The Financial Times recently ran a story on pension funds and private credit with the headline, “US public pension funds pare allocations to private credit. Pullback highlights concerns about looser underwriting standards and rising credit risks.”

On the surface, it was about institutional investors growing cautious on the booming asset class. But look closer, and you’ll see something more telling about the way news gets written — and consumed.

The article opens with a small pension fund in Cincinnati that has tapped the brakes on private credit. The narrative builds around skepticism, risk, and pullback. Only at the very end do readers learn that the New York City pension fund — with over $300 billion under management — is fully committed to private credit. In other words, the story’s most significant character wasn’t just positive on the space, but “all in.”…

…For investors, policymakers, and the public, this matters. Media framing shapes how we understand markets, risk, and opportunity. When negativity consistently drowns out proportion, we risk making decisions based on skewed perceptions.

And for society at large, the same forces are at play. Politics, economics, health, culture — the most pessimistic interpretations tend to dominate. Not because they’re always right, but because they’re the most clickable.

5. A Sleepy 5x – Joe Raymond

In my experience, stocks with the following characteristics tend to do well on average over time:

1. Boring businesses with long histories of profitability
2. Clean balance sheets (more cash than debt)
3. Honest insiders (even if they aren’t terribly talented)
4. Trading cheaply (say, 5x EBIT or less)

Once in a while one of these sorts of stocks might do poorly. But in aggregate, this group does tremendously well – at least in my experience and based on conversations with many other investors…

…Bryan Steam Corporation (BSC) was founded in Peru, Indiana way back in 1916. The company started out making steam-powered cars and tractors…

…By the mid-1920s, it was clear that gasoline powered engines were winning out over steam in automobiles. BSC switched course and focused on boilers and related steam equipment, rather than vehicles.

And that’s basically what the company did for the next 80 years…

…1993 is the earliest year I have data, so that’s where we’ll start. This was around the time my friend was buying shares…

…Growth was modest and choppy, and the operating margin fluctuated between 5% and 10% depending on activity levels. ROE in most years came in somewhere between 7% and 12%.

These are extremely pedestrian numbers.

Most investors wouldn’t have been excited to sit on the bid and patiently build a stake in Bryan Steam. Sure, it was cheap, but it had single-digit margins and single-digit ROE most years. Growth was lackluster. The dividend yield was a mundane 3%…

…But, if you think about it, what was the risk buying BSC at $30 in 1993?

You were paying half of tangible book value. The balance sheet was net cash. The company had a multi-decade history of profitability. Earnings could be cut in half, and you’d still only be paying 10x profits…

…Bryan Steam grew revenue from $16.4 million in 1993 to $26.2 million in 1998 (9.8% CAGR). Cumulative earnings over the period were $6.5 million, which was more than the entire $5.7 million market cap in 1993.

Book value per share grew from $58.84 to $78.50 (5.9% CAGR). The company also paid $8.45 per share of total dividends over those five years.

These are “good, not great” numbers.

Yet the stock finished 1997 trading for $58.25 (18% CAGR before dividends from the 1993 price of $30). And it still traded for only 77% of TBV and less than 7x earnings…

…In September 1998, Bryan Steam entered into a merger agreement with Burnham Corporation (OTC: BURCA/B).

The price?

$152 per share…

…My friend who bought BSC in 1993 at $30 earned a 44% IRR, including dividends. More importantly, he did it without taking a whole lot of risk…

…What if Burnham hadn’t come in and offered $152 per share?

Remember, BSC had compounded at 18% over the prior four years before Burnham entered the picture. And the valuation was still sub-1x book value for a decent (7-12% ROE) business.

Let’s say there was no acquisition and Bryan Steam kept plugging along at its prevailing pace, compounding book value at 6% for the next 20 years.

By 2018, BVPS would have been north of $250 per share and annual earnings would have been around $25 per share. At 12x earnings, BSC would be worth $300 per share.

This results in a hypothetical 10% annualized return over the 25-year period from 1993 to 2018. Including dividends, the IRR would have been in the neighborhood of 12-13%.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Alphabet (parent of Google), Amazon, Meta Platforms, and Microsoft. Holdings are subject to change at any time.

We’ve constantly been sharing a list of our recent reads in our weekly emails for The Good Investors.

Do subscribe for our weekly updates through the orange box in the blog (it’s on the side if you’re using a computer, and all the way at the bottom if you’re using mobile) – it’s free!

But since our readership-audience for The Good Investors is wider than our subscriber base, we think sharing the reading list regularly on the blog itself can benefit even more people. The articles we share touch on a wide range of topics, including investing, business, and the world in general. 

Here are the articles for the week ending 05 October 2025:

1. Nature’s cruel lesson for bag holders – Thomas Chua

On the screen, a cheetah was stalking a gazelle through tall grass, and I found myself holding the stretch longer, mesmerized. My body frozen in tension—part from the stretch, part from watching this life-or-death chess match.

Then the gazelle’s head shot up. The cheetah immediately stopped. No hesitation. Just turned and walked away.

As I eased out of the stretch and settled onto the mat, I realized something: successful investors act like that cheetah, but most people? They do the exact opposite.

Here’s what fascinates me about predators: they never chase a losing cause. The moment their cover is blown—the second the distance starts widening—they quit. No ego. No “sunk cost” thinking. No “but I’ve already come this far.”

They just walk away.

But when you look at the markets, you see the opposite everywhere…

…The thing about us humans is that our prefrontal cortex—the part that makes us “smarter” than animals—actually makes us worse at knowing when to quit. We tell ourselves stories. We rationalize. We create elaborate justifications for why this time is different.

But the cheetah? It just knows: energy spent on a failed hunt is energy that can’t be used on the next opportunity.

This isn’t about giving up easily. It’s about recognizing when persistence becomes stupidity. When determination becomes delusion.

The sunk cost fallacy tricks you into thinking backward—valuing what you’ve already invested over what you could gain elsewhere. But successful investors, like predators, always look forward.

They ask: “From this moment, right now, is this my best opportunity?”

If the answer is no, they walk away. No drama. No hesitation.

Like nature’s hunters, the smartest investors know exactly when to abandon the hunt. They don’t chase dead ends. They stalk fresh opportunities.

2. America’s top companies keep talking about AI – but can’t explain the upsides – Melissa Heikkilä, Chris Cook and Clara Murray

The FT has used AI tools to identify these mentions of the technology in US Securities and Exchange Commission 10-K filings and earnings transcripts, then to categorise each mention. The results were then checked and analysed to help draw a nuanced picture about what companies were saying to different audiences about the technology.

SEC filings require companies to disclose risks to the businesses, and are necessarily more cautious than the sales pitches made by executives on earnings calls. But the increasing array of risks described in filings appears not to be weighing on executives in the public pronouncements.

374 of the S&P 500 mentioned AI on earnings calls in the past 12 months — with 87 per cent of the calls logged as wholly positive about the technology with no concerns expressed…

…The FT sought to categorise the expected positive benefits of the technology. Most of the anticipated benefits, such as increased productivity, were vaguely stated and harder to categorise than the risks.Companies anticipated being able to optimise workflows through automation, and hope to achieve market differentiation through their use of AI. Some hoped to be able to use the technology to improve the personalisation of their products.

Filings do reveal that the companies able to give clear AI upsides include those that serve the rising AI-driven data centre boom. Energy companies First Solar and Entergy cited AI as a demand driver.

Freeport-McMoran, which has a stockpile of copper, stated that “data centres and artificial intelligence developments” would support the metal’s price. The company also said the technology can help with material characterisation and mineral extraction.

Equipment manufacturer Caterpillar reported that its energy business was benefiting from supporting “data centre growth related to cloud computing and generative artificial intelligence”…

…As the number of companies discussing AI has grown, fewer businesses are expressing positive views about the technology than they did in 2022.

The most commonly cited concern was cyber security, which was mentioned as a risk by more than half of the S&P 500 in 2024.

3. The 100 faces of China’s retiree wallet – Nina Chen

The heterogeneity within generations is not unique to China, nor is it exclusive to the elderly. But what sets China’s current seniors apart is the unprecedented structural complexity born of rapid social upheaval and institutional ruptures. This has made them the most heterogeneous and fragmented generation in the country’s modern history.

A closer look shows that different age cohorts in China were shaped by profoundly different circumstances. Take the post-1950s cohort: as children, they endured the Great Famine, carrying lasting memories of hunger through their formative years. In adolescence, their schooling was interrupted by the Cultural Revolution. At an age when they should have been applying their knowledge and skills, they were sent to the countryside for “re-education” through manual labor. Many missed out on the economic dividends of China’s later boom simply because they lacked access to formal education.By contrast, the post-1960s cohort came of age in a very different world. They benefited from the reinstatement of the college entrance exam and the rapid expansion of education. Their youth coincided with the early years of reform and opening, full of energy and opportunities. By middle age, they had experienced soaring property prices, volatile stock markets, the rise of the internet, and widening wealth gaps. The difference between these two generations is not one of degree, but of kind—a qualitative rupture rather than a quantitative stretch…

…In many reports and media narratives, seniors are depicted as embracing new trends and eager to spend: the spotlight often falls on smiling tourists, silver-haired influencers in stylish outfits, and the curated lifestyles of upscale retirement communities. Such portrayals carry strong visual appeal but obscure the underlying consumption attitudes of the majority.

For most seniors, the guiding principle is frugality: “money must be spent where it matters.” Family resources are first directed toward projects deemed vital and long-term—children’s housing, weddings, or cars—seen both as investments in future security and as obligations rooted in traditional family responsibility. Frugality is regarded as a virtue, so spending always requires a clear justification. Tangible goods with lasting value are far more acceptable than abstract or experiential services, with subscription models in particular often dismissed as “non-essential.” Seniors are highly price-sensitive, prioritizing cost-effectiveness over brands or aesthetics…

…This generation lived through China’s abrupt transition from a production-oriented society to a consumer-driven one. Having grown up under scarcity and a planned economy, they later faced a sudden explosion of marketization and commercialization in adulthood. Without systematic guidance—whether from family or school—on new retail channels, advertising formats, rules, and risk awareness, most lack strong consumer judgment. Social isolation and emotional vulnerability further make them particularly susceptible to highly personalized, “caring” marketing.

This dynamic often leaves them swallowing losses in contradictory ways. Some engage in compensatory spending when finances allow, yet remain drawn to bargain hunting. They might skip a RMB 79.9 buffet but stockpile dozens of RMB 9.9 trinkets online. When warned about scams, they retort, “You just don’t understand.” They defend dubious products with, “It has a factory address.” And to children who question their “adopted sons and daughters” from livestreams, they reply, “At least they call me more than you do.” The cautionary phrase they once told their children—“Everything online is a scam”—has boomeranged back to them.

Seniors thus represent both an underserved market and a lucrative yet highly fragmented, information-poor, and emotionally fragile consumer base. Until high-quality elder-focused supply matures, low-cost, easily replicated, high-margin “elder exploitation” businesses will fill the gap. Often, all it takes is a livestreamer repeatedly calling viewers “grandpa” or “grandma” to trigger enthusiastic purchases…

…Within the senior consumer base, those with limited resources and capabilities make up a large share. Behind the silver economy narrative lies a stark truth: the majority of seniors remain low spenders, with consumption disproportionately shaped by a small, visible minority…

…Supplements: Over 80% of seniors do not take them. Among those who do, more than 80% spend less than RMB 3,000 annually. Although supplements are often viewed as the quintessential product for older consumers, a survey on the Living Conditions of Urban and Rural Seniors data shows otherwise: only 16.6% of seniors report using supplements. Of these, 60.6% spend less than RMB 1,000 per year, 24.0% spend RMB 1,000–2,999, 6.5% spend RMB 3,000–4,999, and just 8.9% spend over RMB 5,000…

…Elder care: More than 80% of seniors cannot afford standard retirement home costs. According to CEIC data, in 36 major cities, the average monthly fee for self-sufficient seniors exceeds RMB 2,600—including accommodation, meals, and basic care—and is even higher for semi-dependent or disabled residents. Survey data show that only 15.8% of seniors can afford RMB 3,000 or more per month, meaning over 80% remain priced out of institutional elder care…

…Over 80% of seniors did not travel in the past year. Among those who did, more than 80% spent less than RMB 5,000 annually. According to the 2021 Survey on the Living Conditions of Urban and Rural Seniors, only 9.1% traveled in 2020, and even with some growth in recent years, the share is still estimated at under 20%. Among the small group who do travel, the majority spend under RMB 5,000 a year—pointing to a market still dominated by short, budget-friendly trips…

…First, draw the finest slice, not the biggest circle. Before a single yuan is spent, nail down exactly who you’re serving: how many grandmothers and grandfathers within a ten-year birth band, how much they can actually pay, and how often they’ll open their wallets. Over-count the grey tide and you’ll build a palace for a village—then watch inventory rot and margins drown.

Second, once you’re inside the right yard, seniors likely stay. Familiarity beats flashy ads; trust is a lifelong contract. Win them once and they’ll keep the same travel agency, the same pill brand, the same breakfast stall—year after year—turning your customer-acquisition cost into a one-time entry fee.

Third, profits may also come from the quietest voices—but only if you’re willing to do the hard, on-the-ground work. Village grandpas without apps and grandmas without data plans don’t appear on dashboards, but their needs are vast and competition is thin. Reaching them is hard: you must squat on the lane curb, piggy-back existing clinics, and price for a pocket that holds only folded bills. Yet thin-margin, high-frequency sales—subsidised just enough by local government—add up quickly when idle assets are put to work. I searched online and found some uplifting examples:

• In Ningxia’s Tongyi village, a derelict fish pond and drying yard were simply re-leased; anglers’ tickets and night-market stall rents now fully finance an 18-bed nursing home that charges residents zero fees.
• In Gutian, Fujian, a ¥3 lunch canteen covers its costs with a tiny on-site grocery counter plus monthly on-site pharmacy sales, and the same micro-format has already been copied in 48 neighbouring villages.
• In Caoxian, Shandong, 200 shared e-tricycles (¥1 per 3 km) pay themselves off in 18 months through side ads and parcel deliveries, proving that even a village road can become a revenue-producing asset.

4. AI isn’t replacing radiologists – Works in Progress and Deena Mousa

Radiology is a field optimized for human replacement, where digital inputs, pattern recognition tasks, and clear benchmarks predominate. In 2016, Geoffrey Hinton – computer scientist and Turing Award winner – declared that ‘people should stop training radiologists now’. If the most extreme predictions about the effect of AI on employment and wages were true, then radiology should be the canary in the coal mine.

But demand for human labor is higher than ever. In 2025, American diagnostic radiology residency programs offered a record 1,208 positions across all radiology specialties, a four percent increase from 2024, and the field’s vacancy rates are at all-time highs. In 2025, radiology was the second-highest-paid medical specialty in the country, with an average income of $520,000, over 48 percent higher than the average salary in 2015.

Three things explain this. First, while models beat humans on benchmarks, the standardized tests designed to measure AI performance, they struggle to replicate this performance in hospital conditions. Most tools can only diagnose abnormalities that are common in training data, and models often don’t work as well outside of their test conditions. Second, attempts to give models more tasks have run into legal hurdles: regulators and medical insurers so far are reluctant to approve or cover fully autonomous radiology models. Third, even when they do diagnose accurately, models replace only a small share of a radiologist’s job. Human radiologists spend a minority of their time on diagnostics and the majority on other activities, like talking to patients and fellow clinicians…

…Over the past decade, improvements in image interpretation have run far ahead of their diffusion. Hundreds of models can spot bleeds, nodules, and clots, yet AI is often limited to assistive use on a small subset of scans in any given practice. And despite predictions to the contrary, head counts and salaries have continued to rise. The promise of AI in radiology is overstated by benchmarks alone.

Multi‑task foundation models may widen coverage, and different training sets could blunt data gaps. But many hurdles cannot be removed with better models alone: the need to counsel the patient, shoulder malpractice risk, and receive accreditation from regulators. Each hurdle makes full substitution the expensive, risky option and human plus machine the default. Sharp increases in AI capabilities could certainly alter this dynamic, but it is a useful model for the first years of AI models that benchmark well at tasks associated with a particular career.

There are industries where conditions are different. Large platforms rely heavily on AI systems to triage or remove harmful or policy-violating content. At Facebook and Instagram, 94 percent and 98 percent of moderation decisions respectively are made by machines. But many of the more sophisticated knowledge jobs look more like radiology.

In many jobs, tasks are diverse, stakes are high, and demand is elastic. When this is the case, we should expect software to initially lead to more human work, not less. The lesson from a decade of radiology models is neither optimism about increased output nor dread about replacement. Models can lift productivity, but their implementation depends on behavior, institutions and incentives. For now, the paradox has held: the better the machines, the busier radiologists have become.

5. China’s AWS of Manufacturing – Thomas Chua

Guangzhou and its neighbors—Shenzhen, Dongguan, Foshan—form the Pearl River Delta manufacturing cluster. Decades of development have created an ecosystem so dense that suppliers, manufacturers, and assemblers for almost any product sit within hours of each other.

I took this trip to visit some of these wholesalers and it’s amazing how much they can do.

Walking through the wholesale markets, I saw many products that retail in Singapore and on online platforms selling at a fraction of the price. Take compression boots, for example—under $200 here. A similar device with a different brand slapped on in a Singapore mall near my house? Around $1,000.

Five times the price. Similar product.

I’ve known friends who’ve come here with specifications for products, whether clothing retail or electronics, and they’re able to establish their products and start selling abroad very quickly. All without ever having to sink heavy investments into building a factory or dealing with hiring anyone to produce these items…

…The Laifen hair dryers in hotels across China cost around $50. Dyson? $600.

The performance difference? About as noticeable as the taste difference between Pepsi and Coke.

We’ve also seen DJI and Insta360 run laps around GoPro in their offerings. If businesses can’t innovate fast enough, they’re going to be left behind.

This changing landscape created lots of new value, with some accruing to these newer, more nimble businesses. Consumers capture a nice chunk of the value as competition intensifies…

…On the first day, I had to use DeepSeek for my daily tasks—research, responding to my tour guide in mandarin, planning.

I’d tested DeepSeek during its Sputnik moment in January 2025 and found it comparable to ChatGPT. But now, having to use it due to the firewall, I realized just how rapidly Claude and ChatGPT have advanced. These models improve incrementally day by day—you don’t notice until you’re forced to switch between them.

The pace of AI development is staggering.

I ended up getting LetsVPN to access Claude and ChatGPT again—reliable for short China trips if you need Western services…

…During my daily hour at cafes, coffee in hand, doing my reading, I noticed something.

Many people around me were perpetually on LLM tools. Not just occasionally checking—constantly working with them.

DeepSeek and ChatGPT being the two most common.

---

Disclaimer: The Good Investors is the personal investing blog of two simple guys who are passionate about educating Singaporeans about stock market investing. By using this Site, you specifically agree that none of the information provided constitutes financial, investment, or other professional advice. It is only intended to provide education. Speak with a professional before making important decisions about your money, your professional life, or even your personal life. We currently have a vested interest in Amazon (parent of AWS) and Meta Platforms (parent of Facebook and Instagram) Holdings are subject to change at any time.