Category: What We’re Reading

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 17 November 2024:

1. OpenAI Shifts Strategy as Rate of ‘GPT’ AI Improvements Slows – Stephanie Palazzolo, Erin Woo and Amir Efrati

In May, OpenAI CEO Sam Altman told staff he expected Orion, which the startup’s researchers were training, would likely be significantly better than the last flagship model, released a year earlier.

Though OpenAI had only completed 20% of the training process for Orion, it was already on par with GPT-4 in terms of intelligence and abilities to fulfill tasks and answer questions, Altman said, according to a person who heard the comment.

While Orion’s performance ended up exceeding that of prior models, the increase in quality was far smaller compared with the jump between GPT-3 and GPT-4, the last two flagship models the company released, according to some OpenAI employees who have used or tested Orion.

Some researchers at the company believe Orion isn’t reliably better than its predecessor in handling certain tasks, according to the employees. Orion performs better at language tasks but may not outperform previous models at tasks such as coding, according to an OpenAI employee. That could be a problem, as Orion may be more expensive for OpenAI to run in its data centers compared to other models it has recently released, one of those people said.

The Orion situation could test a core assumption of the AI field, known as scaling laws: that LLMs would continue to improve at the same pace as long as they had more data to learn from and additional computing power to facilitate that training process…

…However, OpenAI researcher Noam Brown said at the TEDAI conference last month that more-advanced models could become financially unfeasible to develop.

“After all, are we really going to train models that cost hundreds of billions of dollars or trillions of dollars?” Brown said. “At some point, the scaling paradigm breaks down.”…

…Orion was trained in part on AI-generated data, produced by other OpenAI models, including GPT-4 and recently released reasoning models, according to an OpenAI employee. However, such synthetic data, as it is known, is leading to a new problem in which Orion may end up resembling those older models in certain aspects, the employee said.

2. Prejudice And China – Louis-Vincent Gave

This led us to the comments made in September by Ford chief executive officer Jim Farley. Freshly returned from a visit to China, Farley told The Wall Street Journal that the growth of the Chinese auto sector poses an existential threat to his company, and that “executing to a Chinese standard is now going to be the most important priority.”

By any measure, this is an earth-shattering statement.

Making cars is complicated. Not as complicated as making airliners or nuclear power plants. But making cars is still the hallmark of an advanced industrial economy. So, the idea that China is suddenly setting the standards that others must now strive to meet is a sea-change compared with the world we lived in just five years ago…

…This brings me to the simplest, most obvious, and likeliest explanation why most CEOs and investors missed how China leapfrogged the West in industry after industry over the last five years: during that time, no one from the West bothered to visit China…

…Unlike Japan in the 1990s, China has not seen its banking system go bust and lose its ability to fund new projects. On the contrary, the surge in loans to industry over the past few years lies at the heart of China’s booming industrial productivity…

…This is another key difference between China today and Japan in the 1990s. China today is not only more efficient and more productive than a decade ago, it is probably more efficient and more productive than most other major industrial economies. And it boasts a very attractive cost structure. Until a few years ago, you would need to check your bank balance before going out for dinner in Tokyo. Today, you can stay in the Four Seasons in Beijing or Shanghai for less than US$250 a night. Perhaps the best illustration of how Japan’s past is a very poor guide to China’s present is the difference in their trade balances; a reflection of how different their competitiveness has been…

…This is not to understate the magnitude of the Chinese property bust. The rollover in real estate has been a massive drag on growth and on animal spirits over the past five years. But on this front, there is another key difference between China and Japan: in China, the contraction of real estate was the policy. It was not the unfortunate consequence of policies gone-wrong. Reallocating capital away from real estate and towards industry was a stated goal of the government…

…There seem to be at least three separate visions of China.

The first is the China you read about in much of the Western media: a place of despond and despair. It is permanently on the cusp of social disorder and revolution, or it would be, were it not an Orwellian nightmare of state surveillance, supervision and repression that strangles creativity and stifles progress. This is the place that Westerners who have never visited China typically imagine, because it is the place portrayed by the media…

…The second is the vision of China you get from talking to Chinese millennials in tier-one cities. This version of China recalls the “lost decades” of Japanese deflationary depression…

…This brings me to the third vision of China: that it is only just beginning to leapfrog the West in a whole range of industries. This vision is starting to show up itself in the perception of Western brands in China, and their sales. For example, Apple’s iPhones no longer figure in the five best-selling smartphone models in China. And Audi’s new electric cars made and sold in China will no longer carry the company’s iconic four-circle logo; the branding is now perceived to be more of a hindrance than a benefit.

To put it another way, following years of investment in transport infrastructure, education, industrial robots, the electricity grid and other areas, the Chinese economy today is a coiled spring. So far, the productivity gains engendered by these investments have manifested themselves in record trade surpluses and capital flight—into Sydney and Vancouver real estate, and Singapore and Hong Kong private banking.

This has mostly been because money earners’ confidence in their government has been low. From bursting the real estate bubble, through cracking down on big tech and private education, to the long Covid lockdowns, in recent years the Chinese government has done little to foster trust among China’s wealthy. It’s small surprise, then, that many rich Chinese have lost faith in their government’s ability to deliver a stable and predictable business environment.

This brings me to the recent stimulus announcements and the all-important question whether the measures rolled out will prove sufficient to revitalize domestic confidence in a meaningful way. Will it even be possible to lift confidence as long as the Damocles’ sword of a wider trade conflict with the US and yet more sanctions looms over the head of Chinese businesses?

From this perspective, perhaps the most bullish development for China would be for the new US administration (regardless who sits behind the Resolute desk) to come in and look to repair the damage done to relations by the 2018 semiconductor sanctions and the 2021 Anchorage meeting…

…When it comes to China’s relevance to investors, there are four ways of looking at things.

• China can be uninvestible and unimportant. This is the pool that most investors have been swimming in for the last few years. But this is akin to saying that China is like Africa. It simply doesn’t pass the smell test. Instead of sliding into irrelevance, China’s impact on the global economy only continues to grow.
• China can be uninvestible but important. This is essentially what Jim Farley, fresh back from his China trip, told The Wall Street Journal.
• China can be investible but unimportant. This is the space Japan inhabited for a couple of decades, and into which Europe seems to be gently sliding. However, the idea that China today is where Japan has been for the last three decades is grossly misplaced on many fronts, including the competitiveness of its economy, its overall cost structure, and its weight in global indexes.
• China can be investible and important. This is what David Tepper of Appaloosa Management argued on CNBC following the announcement of China’s stimulus (see Changing Narratives Around The World). For now, this is still a minority view, at least among Western investors. Not that Western investors matter all that much. What truly matters is whether Chinese investors themselves start rallying to this view. If they do, the unfolding bull markets in Chinese equities and the renminbi could really have legs.

3. $2 H100s: How the GPU Rental Bubble Burst – Eugene Cheah

ChatGPT was launched in November 2022, built on the A100 series. The H100s arrived in March 2023. The pitch to investors and founders was simple: Compared to A100s, the new H100s were 3x more powerful, but only 2x the sticker price.

If you were faster to ramp up on H100s, you too, can build a bigger, better model, and maybe even leapfrog OpenAI to Artificial General Intelligence – If you have the capital to match their wallet!

With this desire, $10-100’s billions of dollars were invested into GPU-rich AI startups to build this next revolution. Which lead to ….

The sudden surge in H100 demand

Market prices shot through the roof, the original rental rates of H100 started at approximately $4.70 an hour but were going for over $8. For all the desperate founders rushing to train their models to convince their investors for their next $100 million round…

…For most of 2023, the H100 prices felt like they would forever be above $4.70 (unless you were willing to do a huge upfront downpayment)

At the start of 2024, the H100 prices reached approximately $2.85 across multiple providers…

…In Aug 2024, if you’re willing to auction for a small slice of H100 time (days to weeks), you can start finding H100 GPUs for $1 to $2 an hour.

We are looking at a >= 40% price drop per year, especially for small clusters. NVIDIA’s marketing projection of $4 per GPU hour across 4 years, has evaporated away in under 1.5 years.

And that is horrifying because it means someone out there is potentially left holding the bag – especially so if they just bought it as a new GPUs…

…The average H100 SXM GPU in a data center costs $50k or more to set up, maintain, and operate (aka most of the CAPEX). Excluding electricity and cooling OPEX cost…

…If the price falls below $1.65/hour, you are doomed to make losses on the H100 over the 5 years, as an infra provider. Especially, if you just bought the nodes and cluster this year…

…Many infrastructure providers, especially the older ones – were not naive about this – Because they had been burnt firsthand by GPU massive rental price drops, after a major price pump, from the crypto days – they had seen this cycle before.

So for this cycle, last year, they pushed heavily for a 3-5 year upfront commitment and/or payment at the $4+ price range. (typically with 50% to 100% upfront). Today, they push the $2.85+ price range – locking in their profits…

…When a model creator is done training a model, you have no more use for the cluster. What would they do? – they resell and start recouping some of the costs…

…This ended up creating a triple whammy in reducing the demand for H100s!

1. Finetuning is significantly cheaper than training from scratch.

a. Because the demands for fine-tuning are significantly less in compute requirements (typically 4 nodes or less, usually a single node), compared to training from scratch (from 16 nodes, usually more, for 7B and up models).

b. This industry-wide switch essentially killed a large part of smaller cluster demands.

2. Scaling back on foundation model investment (at small/mid-tier)

a. In 2023, there was a huge wave of small and medium foundation models, within the text and image space.

b. Today, however, unless you are absolutely confident you can surpass llama3, or you are bringing something new to the table (eg. new architecture, 100x lower inference, 100+ languages, etc), there are ~no more foundation model cos being founded from scratch.

c. In general, the small & medium, open models created by the bigger players (Facebook, etc), make it hard for smaller players to justify training foundation models – unless they have a strong differentiator to do so (tech or data) – or have plans to scale to larger models.

d. And this has been reflected lately with investors as well, as there has been a sharp decline in new foundation model creators’ funding. With the vast majority of smaller groups having switched over to finetuning. (this sentiment is combined with the recent less than desired exits for multiple companies).

e. Presently today, there is approximately worldwide by my estimate:

<20 Large model creator teams (aka 70B++, may create small models as well)

<30 Small / Medium model creator teams (7B – 70B)

f. Collectively there are less than <50 teams worldwide who would be in the market for 16 nodes of H100s (or much more), at any point in time, to do foundation model training.

g. There are more than 50 clusters of H100 worldwide with more than 16 nodes.

3. Excess capacity from reserved nodes is coming online

a. For the cluster owners, especially the various foundation model startups and VCs, who made long reservations, in the initial “land grab” of the year 2023.

b. With the switch to finetuning, and the very long wait times of the H100’s
(it peaked at >= 6 months), it is very well possible that many of these groups had already made the upfront payment before they made the change, essentially making their prepaid hardware “obsolete on arrival”.

c. Alternatively, those who had the hardware arrive on time, to train their first few models, had come to the same realization it would be better to fine-tune their next iteration of models. Instead of building on their own.

d. In both cases, they would have unused capacity, which comes online via “Compute Resellers” joining the market supply…. 

…Both AMD and Intel may be late into the game with their MX300, and Gaudi 3 respectively.

This has been tested and verified by us, having used these systems. They are generally:

• Cheaper than a H100 in purchase cost
• Have more memory and compute than a H100, and outperforms on a single node.
• Overall, they are great hardware!

The catch? They have minor driver issues in training and are entirely unproven in large multi-node cluster training.

Which as we covered is largely irrelevant to the current landscape. To anyone but <50 teams. The market for H100 has been moving towards inference and single or small cluster fine-tuning.

All of which these GPUs have been proven to work at. For the use cases, the vast majority of the market is asking for.

These 2 competitors are full drop-in replacements. With working off-the-shelf inference code (eg. VLLM) or finetuning code for most common model architectures (primarily LLaMA3, followed by others)…

…Given that the open-weights model has entered the GPT-4 class arena. Falling H100 prices will be the multiplier unlock for open-weights AI adoption.

It will be more affordable, for hobbyists, AI developers, and engineers, to run, fine-tune, and tinker with these open models.

Especially if there is no major leap for GPT5++, because it will mean that the gap between open-weights and closed-source models will blur.

This is strongly needed, as the market is currently not sustainable. As there lacks the value capture on the application layer for paying users (which trickles down the platform, models, and infra layers)

In a way, if everyone is building shovels (including us), and applications with paying users are not being built (and collecting revenue and value).

But when AI inference and fine-tuning becomes cheaper than ever.

It can potentially kick off the AI application wave. If it has not already slowly started so.

4. Politics, Portfolios & Perspective: Investing in a Crazy Election Year – Alliance Wealth Advisors

How we feel about the economy is directly correlated to if the party we most closely identify with is in power or not. This is regardless of what the economic data actually tells us. In other words, our emotions get the best of us and cloud our ability to stay objective…

…In the past two presidential elections, there were many “expert” predictions claiming that electing both Donald Trump and Joe Biden would cause a significant stock market correction. Yet, both presided over stock market highs at various times. Anyone who made changes to their portfolio based on those election outcomes suffered a serious opportunity cost that will impact them for a long time…

…Politics aside, the stock market is a complex adaptive system, influenced by countless variables interacting with one another in constantly evolving ways. Companies are dynamic and run by smart people who learn to adapt to new environments. History has shown that companies can react to all kinds of changes and have always been able to grow their earnings over time. When they do stock prices tend to follow.

5. Writes and Write-Nots – Paul Graham

I’m usually reluctant to make predictions about technology, but I feel fairly confident about this one: in a couple decades there won’t be many people who can write…

…The reason so many people have trouble writing is that it’s fundamentally difficult. To write well you have to think clearly, and thinking clearly is hard…

…Till recently there was no convenient escape valve for the pressure created by these opposing forces. You could pay someone to write for you, like JFK, or plagiarize, like MLK, but if you couldn’t buy or steal words, you had to write them yourself. And as a result nearly everyone who was expected to write had to learn how.

Not anymore. AI has blown this world open. Almost all pressure to write has dissipated. You can have AI do it for you, both in school and at work.

The result will be a world divided into writes and write-nots…

…Is that so bad? Isn’t it common for skills to disappear when technology makes them obsolete? There aren’t many blacksmiths left, and it doesn’t seem to be a problem.

Yes, it’s bad. The reason is something I mentioned earlier: writing is thinking. In fact there’s a kind of thinking that can only be done by writing.

---

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 Apple. 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 10 November 2024:

1. Why I’m Leaving OpenAI and What I’m Doing Next – Miles Brundage

So how are OpenAI and the world doing on AGI readiness?

In short, neither OpenAI nor any other frontier lab is ready, and the world is also not ready.

To be clear, I don’t think this is a controversial statement among OpenAI’s leadership, and notably, that’s a different question from whether the company and the world are on track to be ready at the relevant time (though I think the gaps remaining are substantial enough that I’ll be working on AI policy for the rest of my career).

Whether the company and the world are on track for AGI readiness is a complex function of how safety and security culture play out over time (for which recent additions to the board are steps in the right direction), how regulation affects organizational incentives, how various facts about AI capabilities and the difficulty of safety play out, and various other factors.

As a sidenote, I think that AGI is an overloaded phrase that implies more of a binary way of thinking than actually makes sense. One of the things my team has been working on lately is fleshing out the “levels of AI” framework referenced here. I hope that OpenAI and I will be able to publish a related paper before long. But for now I’d just note that when I say “ready for AGI,” I am using this as shorthand for something like “readiness to safely, securely, and beneficially develop, deploy, and govern increasingly capable AI systems.”…

…I think the upsides of AI are already big and could be dramatically bigger, as are the downsides. As someone who has worked in this field for longer than most, it has been very sad to see increasing polarization along the lines of whether people focus on one side of the cost/benefit ledger or the other, or have different risk priorities, etc. My view is that there is a lot to worry about and a lot to be excited about, we don’t have to choose one thing to care about, and we should find common ground where it exists.

I think AI and AGI benefiting all of humanity is not automatic and requires deliberate choices to be made by decision-makers in governments, non-profits, civil society, and industry, and this needs to be informed by robust public discussion. Notably, this is true not just for risk mitigation but also for ensuring equitable distribution of the benefits, as is the case with, e.g., electricity and modern medicine as well. This is true for a few reasons, including, non-exhaustively, collective action problems, various unpriced negative externalities, and unequal starting positions of digital infrastructure access, wealth, etc. that affect who benefits and is harmed by default and to what degrees. As with railroads, electricity, etc., corporate and government policies will be critical to ensuring safe and fair outcomes.

I think AI capabilities are improving very quickly and policymakers need to act more urgently…

..I think quantitative evaluations of AI capabilities and extrapolations thereof, in combination with analysis of the impacts of certain policies, will be critical in truthfully and persuasively demonstrating that urgency. There’s great work happening on measuring frontier models from a safety perspective, measuring trends over time in AI, and a growing body of work assessing the labor market implications of AI, but more is definitely needed.

I think we don’t have all the AI policy ideas we need, and many of the ideas floating around are bad or too vague to be confidently judged. This is particularly true of international competition over AI, where I find the existing proposals to be especially bad (e.g. “race against [competing country] as quickly as possible”) and vague (e.g. “CERN for AI”), although it’s encouraging to see a growing trend towards more nuanced discussion of some of these ideas. There are also many aspects of frontier AI safety and security that will require creative solutions…

…I think that improving frontier AI safety and security is quite urgent, given the number of companies (dozens) that will soon (next few years at most) have systems capable of posing catastrophic risks. Given that that is not much time to set up entirely new institutions, I’m particularly interested in opportunities for action under existing legal authorities, as well as shaping the implementation of already-approved legislation such as the EU AI Act.

As noted above, and explained in more detail in this paper and similar work, companies and governments will not necessarily give AI safety and security the attention it deserves by default (this is not a comment specifically about OpenAI, as discussed above). There are many reasons for this, one of which is a misalignment between private and societal interests, which regulation can help reduce. There are also difficulties around credible commitments to and verification of safety levels, which further incentivize corner-cutting: people assume others are going to cut corners to gain an advantage and can’t tell what the ground truth is, or think they will change their minds later. Corner-cutting occurs across a range of areas, including prevention of harmfully biased and hallucinated outputs as well as investment in preventing the catastrophic risks on the horizon. There are, to be clear, some ways in which commercial incentives encourage safety, though I think it would be irresponsible to assume that those incentives will be sufficient, particularly for ambiguous, novel, diffuse, and/or low-probability/high-magnitude safety risks.

I’m excited about understanding how companies can credibly demonstrate safety while protecting valuable and potentially misusable IP. The difficulty of demonstrating compliance without compromising sensitive information is a major barrier to arms control agreements, which requires innovation to address. This issue is also at the core of effective domestic regulation. I’m excited to collaborate with people working on this and other related technical AI governance questions.

While some think that the right approach to the global AI situation is for democratic countries to race against autocratic countries, I think that having and fostering such a zero-sum mentality increases the likelihood of corner-cutting on safety and security, an attack on Taiwan (given its central role in the AI chip supply chain), and other very bad outcomes. I would like to see academics, companies, civil society, and policymakers work collaboratively to find a way to ensure that Western AI development is not seen as a threat to other countries’ safety or regime stability, so that we can work across borders to solve the very thorny safety and security challenges ahead.

Even if, as I think is very likely, Western countries continue to substantially outcompete China on AI, there is more than enough “gas in the tank” of computing hardware and algorithmic progress in autocratic countries for them to build very sophisticated capabilities, so cooperation will be essential. I realize many people think this sounds naive but I think those people haven’t thought through the situation fully or considered how frequently international cooperation (enabled by foresight, dialogue, and innovation) has been essential to managing catastrophic risks…

…I think it’s likely that in the coming years (not decades), AI could enable sufficient economic growth that an early retirement at a high standard of living is easily achievable (assuming appropriate policies to ensure fair distribution of that bounty). Before that, there will likely be a period in which it is easier to automate tasks that can be done remotely. In the near-term, I worry a lot about AI disrupting opportunities for people who desperately want work, but I think it’s simultaneously true that humanity should eventually remove the obligation to work for a living and that doing so is one of the strongest arguments for building AI and AGI in the first place. Likely some will continue to work in the long-term but the incentive to do so might be weaker than before (whether this is true depends on a variety of cultural and policy factors). That is not something we’re prepared for politically, culturally, or otherwise, and needs to be part of the policy conversation. A naive shift towards a post-work world risks civilizational stagnation (see: WALL-E), and much more thought and debate about this is needed…

…Compared to software, data, and talent, computing hardware has unique properties that make it an important focal point for AI policy: “it is detectable, excludable, and quantifiable, and is produced via an extremely concentrated supply chain” (quoted from this paper I worked on). This makes it worrying that the part of the US government responsible for overseeing what happens when that compute is shipped overseas is severely understaffed and underfunded, and that more generally there is little serious policy discussion of what the endgame is here (besides occasionally tightening export controls and requiring companies to report their big datacenters and training runs).

To the extent that there is serious analysis of compute governance happening in the academic literature, it generally lags behind developments in industry by a fair amount – e.g., to those within frontier AI companies, it has become increasingly clear in recent years that scaling up inference, not just training, can enable higher performance, but public analysis of the policy implications of this has only begun in earnest relatively recently. Ideas for distributing computing power (and the associated benefits of AI) more widely, such as via the government providing greater compute for academics, are generally too little too late and neglect issues specific to the developing world, which is in a quite different situation.

2. Industry Is Not Destiny – Greg Obenshain

We’d go as far as to argue that industry analysis generally is much less valuable than fundamental investors or strategy consultants might hope.

Mauboussin’s new study, Measuring the Moat: Assessing the Magnitude and Sustainability of Value Creation, grapples with this issue. Mauboussin’s study includes a chart that is difficult to unsee once you’ve seen it (h/t Edward Conard’s Macro Roundup for highlighting this)…

…This chart shows that profitability varies more within industry (the vertical bars) than across industries (the dots). Over the long run, the fate of a company is not primarily determined by its industry—a finding consistent with Chicago school research from the 1980s that dealt a death blow to structure-conduct-performance theory in antitrust law.

Mauboussin notes that while industry analysis matters when it comes to deciding where to compete, ultimately the right unit of analysis is not the industry level but the company level…

…Industries with higher overall profitability have more companies that are profitable, but even within industries with low profitability, there are still companies that have returns well above the cost of capital and some companies that have profitability substantially above.

Industry is not destiny. Great companies can emerge from mediocre industries.

3. Watch Out: Wall Street Is Finding New Ways to Slice and Dice Loans – Matt Wirz

Goldman Sachs GS 2.14%increase; green up pointing triangle this month sold $475 million of public asset-backed securitization, or ABS, bonds backed by loans the bank makes to fund managers that tide them over until cash from investors comes in. The first-of-its-kind deal is a lucrative byproduct of the New York bank’s push into loans to investment firms, such as these so-called capital-call lines.

Goldman’s new deal reflects two trends transforming financial markets. Increasingly large managers of private-debt and private-equity funds are moving up in the Wall Street pecking order, but they often need money fast. Banks, once again, are reinventing themselves to adapt…

…The transactions are relatively small for now. Still, they are intertwining banks (in Wall Street parlance, the sell side) with investors (the buy side) in ways that are new and difficult to parse for analysts, regulators and others…

…Capital-call loans function like credit cards for private-fund managers. The funds borrow money to invest quickly in private debt, private equity, real estate and infrastructure. They then “call up” cash commitments from clients in the funds, mostly institutions such as pensions and insurers, and repay the loans when the clients deliver.

Defaults on capital-call commitments from large institutions “have been historically close to 0%,” according to a marketing document for Goldman’s bond viewed by The Wall Street Journal. That makes the bonds extremely safe, said debt fund managers to whom Goldman offered the deal.

Even so, the shiny new products that banks are inventing have yet to be tested through market cycles…

…As Goldman and other banks make more capital-call loans to private-fund managers, they are also buying insurance from many of the same investment firms to protect against potential losses from corporate, consumer and real-estate loans. The so-called synthetic risk transfers, or SRTs, help banks reduce risk to meet new regulatory requirements and give fund managers investments to put into their wildly popular private-credit funds.

Some private-credit funds are developing another product that is similar to capital-call lines called net-asset-value, or NAV loans, made to private-equity fund managers. Rising interest rates have made it harder for private-equity funds to sell companies they own to repay their limited partners. NAV loans help them to start returning cash to clients until they can dispose of the companies. Many of the firms that manage private-equity funds also manage private-credit funds…

…The International Monetary Fund published a report in April warning that “interconnections and potential contagion risks many large financial institutions face from exposures to the asset class are poorly understood and highly opaque.”

4. Big Banks Cook Up New Way to Unload Risk – Matt Wirz

U.S. banks have found a new way to unload risk as they scramble to adapt to tighter regulations and rising interest rates…

…These so-called synthetic risk transfers are expensive for banks but less costly than taking the full capital charges on the underlying assets. They are lucrative for the investors, who can typically get returns of around 15% or more, according to the people familiar with the transactions.

U.S. banks mostly stayed out of the market until this autumn, when they issued a record quantity as a way to ease their mounting regulatory burden…

…In most of these risk transfers, investors pay cash for credit-linked notes or credit derivatives issued by the banks. The notes and derivatives amount to roughly 10% of the loan portfolios being de-risked. Investors collect interest in exchange for shouldering losses if borrowers of up to about 10% of the pooled loans default…

…The deals function somewhat like an insurance policy, with the banks paying interest instead of premiums. By lowering potential loss exposure, the transfers reduce the amount of capital banks are required to hold against their loans.

Banks globally will likely transfer risk tied to about $200 billion of loans this year, up from about $160 billion in 2022, according to a Wall Street Journal analysis of estimates by ArrowMark Partners, a Denver-based firm that invests in risk transfers…

…Banks started using synthetic risk transfers about 20 years ago, but they were rarely used in the U.S. after the 2008-09 financial crisis. Complex credit transactions became harder to get past U.S. bank regulators, in part because similar instruments called credit-default swaps amplified contagion when Lehman Brothers failed.

Regulators in Europe and Canada set clear guidelines for the use of synthetic risk transfers after the crisis. They also set higher capital charges in rules known as Basel III, prompting European and Canadian banks to start using synthetic risk transfers regularly.

U.S. regulations have been more conservative. Around 2020, the Federal Reserve declined requests for capital relief from U.S. banks that wanted to use a type of synthetic risk transfer commonly used in Europe. The Fed determined they didn’t meet the letter of its rules…

…The pressure began to ease this year when the Fed signaled a new stance. The regulator said it would review requests to approve the type of risk transfer on a case-by-case basis but stopped short of adopting the European approach.

5. Xi Stimulus Clues Found in Protest Data Showing Economic Stress – Rebecca Choong Wilkins

From a basement in Calgary, often accompanied by his pet cat, Lu Yuyu spends 10 hours a day scouring the internet to compile stats on social instability before they are scrubbed by China’s censors. The 47-year-old exile won’t reveal his exact method because it risks jeopardizing the overall goal of the project called “Yesterday,” which documents cases of group protests.

“These records provide an important basis for people to understand the truth of this period of history,” said Lu, who started the effort in January 2023 but didn’t make it public until he arrived in Canada a year ago. “I didn’t want to go to jail again,” he explained.

While Lu’s interests are political, his database — available for free — is among a growing number of metrics tracking dissent in China that investors are watching to figure out when Xi will open up the spigots to bolster growth. And some banks are now starting to develop similar products.

Morgan Stanley in September debuted a new gauge of distress that could be used to predict policy swings in China. Robin Xing, the bank’s chief China economist, says it’s nearing the low levels reached two other times in the past decade: in 2015, when Beijing took drastic steps to arrest a $7 trillion stock market rout, and in 2022 — the point at which the Communist Party abruptly dropped its strict Covid controls after simultaneous street protests in major cities…

…While China’s opaque political system makes it difficult to attribute policy moves to any single factor, investors and analysts who track instances of unrest say authorities may be especially sensitive to them when deciding on whether to roll out stimulus and how much to deploy. Economic protests have become more frequent in recent years as China’s youth unemployment rate soared and its housing crisis worsened…

…Getting a read on what’s happening on the ground is a challenge for academic researchers and finance professionals alike. Widespread censorship, heavy surveillance and suppression of dissent have made it hard to assess the depth of economic malaise in the country of 1.4 billion people…

…The rising prominence of dissent metrics is part of a blossoming industry of so-called alternative data aimed at decoding the state of the world’s second-biggest economy…

…Life has become tougher for many in recent years as pandemic lockdowns, a real estate crisis and trade tensions have slowed growth in China.

Incomes are still rising, but gains under Xi have been the weakest since the late 1980s. Faith in the country’s meritocracy also appears to be waning, leaving white-collar workers feeling increasingly disillusioned. Companies mired in fierce price wars are laying off employers, while college graduates are struggling to find work.

China Dissent Monitor’s data shows that cases of dissent rose 18% in the second quarter compared to same period last year, with the majority of events linked to financial issues.

“If you look at everything regarding social well-being — be it wage growth, urban unemployment rate, consumer confidence and even tracking labor incidents — I think it’s deteriorating,” Morgan Stanley’s Xing said.

Although protests aren’t particularly rare in China, they’re typically small scale, uncoordinated with other places and lacking in overt criticism of Beijing. Still, political criticism can bubble up, usually in cases linked to rural land actions where the local governments find themselves the target of discontent, according to China Dissent Monitor research…

…Even so, there are few signs that the unrest is coalescing around a particular instance of perceived injustice or a single issue. Unlike the Tiananmen Square protests and unrest in the late 1980s, current dissent doesn’t present an existential threat to the regime. A more likely response is therefore a dose of economic medicine that will keep the market guessing.

---

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 no vested interest in any companies mentioned. 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 27 October 2024:

1. China’s Fiscal Policy Update – Leonid Mironov

Ministry of Finance top brass spoke at a press briefing, and outlined the extent of the Fiscal policy support they can offer *within the confines of the current budget*. However, while the steps laid out suggest a cautious and structured approach, notable gaps in specific figures leave room for market speculation…

…China is set to enhance its strategy for managing local government debt, which remains a critical issue. The central government will issue large-scale debt swaps, a move aimed at addressing the opaque “hidden debt” local authorities have accumulated off the books. Local governments still hold 2.3 trillion yuan in available funds, providing some breathing room to manage obligations in the final quarter of 2024. These steps aim to steady the debt situation, though the path forward will undoubtedly be closely watched…

…There is commentary out there to say that this is not new spending, I would counter with that yes, its not new per se, but its spending that would go in to this gap (authorised/unspent) but won’t anymore. So this is stimulative…

…With property markets showing persistent weakness, local governments now have the authority to deploy funds from special bonds to purchase unsold homes. These homes will be converted into subsidized housing—a dual-purpose measure to both alleviate property inventory and address housing affordability. It signals a nuanced, albeit gradual, approach to propping up the beleaguered real estate sector.

This is likely where most of that 2.3trn RMB mentioned in (1) will go. Again since the the property market is such a significant drag on the economy, this is reasonable…

…In line with recent People’s Bank of China (PBOC) directives, four major state-owned banks announced forthcoming cuts to existing mortgage rates. These rate reductions, effective from October 25, are part of broader efforts to ease financial pressures on households and further stimulate economic activity. Again given the sheer amount of total mortgages outstanding (38 trn RMB at the end of ‘23, see chart), this is significant. PBOC expects an effective cut of about 50pbs on average…

…Perhaps the most telling aspect of the press conference was what remained unsaid. There were no specifics on the magnitude of additional fiscal stimulus or further bond issuances. Additionally, there was no precise indication of how much the fiscal deficit might increase—a critical piece of information many market participants were hoping for…

…The Ministry of Finance’s approach at this juncture reflects a cautious yet deliberate strategy. While existing resources are being leveraged, and flexibility is maintained, major new initiatives have not yet been unveiled. All eyes now turn to the late October NPC meeting, where the prospect of more significant fiscal interventions could reshape the economic landscape for the year ahead.

2. The Bitter Lesson – Rich Sutton

The biggest lesson that can be read from 70 years of AI research is that general methods that leverage computation are ultimately the most effective, and by a large margin. The ultimate reason for this is Moore’s law, or rather its generalization of continued exponentially falling cost per unit of computation. Most AI research has been conducted as if the computation available to the agent were constant (in which case leveraging human knowledge would be one of the only ways to improve performance) but, over a slightly longer time than a typical research project, massively more computation inevitably becomes available. Seeking an improvement that makes a difference in the shorter term, researchers seek to leverage their human knowledge of the domain, but the only thing that matters in the long run is the leveraging of computation. These two need not run counter to each other, but in practice they tend to. Time spent on one is time not spent on the other. There are psychological commitments to investment in one approach or the other. And the human-knowledge approach tends to complicate methods in ways that make them less suited to taking advantage of general methods leveraging computation…

…We have to learn the bitter lesson that building in how we think we think does not work in the long run. The bitter lesson is based on the historical observations that 1) AI researchers have often tried to build knowledge into their agents, 2) this always helps in the short term, and is personally satisfying to the researcher, but 3) in the long run it plateaus and even inhibits further progress, and 4) breakthrough progress eventually arrives by an opposing approach based on scaling computation by search and learning. The eventual success is tinged with bitterness, and often incompletely digested, because it is success over a favored, human-centric approach.

One thing that should be learned from the bitter lesson is the great power of general purpose methods, of methods that continue to scale with increased computation even as the available computation becomes very great. The two methods that seem to scale arbitrarily in this way are search and learning.

The second general point to be learned from the bitter lesson is that the actual contents of minds are tremendously, irredeemably complex; we should stop trying to find simple ways to think about the contents of minds, such as simple ways to think about space, objects, multiple agents, or symmetries. All these are part of the arbitrary, intrinsically-complex, outside world. They are not what should be built in, as their complexity is endless; instead we should build in only the meta-methods that can find and capture this arbitrary complexity. Essential to these methods is that they can find good approximations, but the search for them should be by our methods, not by us. We want AI agents that can discover like we can, not which contain what we have discovered. Building in our discoveries only makes it harder to see how the discovering process can be done.

3. Austan Goolsbee Explains the Fed’s Big Rate Cut – Tracy Alloway, Joe Weisenthal, and Austan Goolsbee

Joe (12:30):

You mentioned lags. I want to ask you a question about that. When the Fed started jacking up rates aggressively, one of the theories for why it didn’t have a sharper impact on the economy is that so many households and corporations had in, say, 2020, first half of 2021, termed out their debt and so there was not a lot of sensitivity to debt.

The flip side of that now — and people have been writing about this — is that even though the Fed has now commenced a cutting cycle, that the weighted average cost of debt is probably going to rise in 2025 basically just mathematically, right? Because eventually that’ll have to be refi-ed at higher rates and so forth. How do you think about that dynamic now when you’re thinking about these lags? You’re starting a cutting cycle, but at the same time probably cost of debt is actually going to rise for a fair number of economic actors in this economy.

Austan (13:19):

You have remarked on this subject and thought it through. In my world that goes into the economic conditions and there are many things that have made this a hairy, strange time for central banks because the business cycle, both down and up, looked almost nothing like historical precedents. This is one aspect of that.

We’ve analyzed this specifically thinking about mortgages. Okay, so if I had told you, the premise of your question, I a hundred percent agree with, six years ago, if you said ‘The Fed is going to raise 500 basis points in a single year, what is going to happen?’ I think most all economists would say ‘Yikes, there’s going to be a major, major contraction and it’s going to be concentrated. Autos down the tubes. Consumer durables, bye-bye. Business fixed investment construction, all going to collapse because they’re very interest rate sensitive.’

We didn’t really see the economy go into the steepness of collapse that you would’ve expected. And so that brings us back to this question. It’s kind of a twofold. Is there something about this unusual business cycle that makes economic activity less sensitive to the interest rate? Or is there something strange about this moment that the lag effect is longer. And it can be both and they can run together, but in the case of mortgages, one of the things that has made monetary policy transmission less direct, is the fact that a vastly higher share of mortgages are 30-year fixed mortgages now, than they were in 2005, 2009, whenever you want to look at.

And so when they change the interest rate — in some countries virtually all mortgages are adjustable rate mortgages. So when their central bank raises rates, they bring out parents onto TV, ‘The central bank is killing us. You know, our mortgage payment went up.’

In the US, if everybody’s on a 30-year fixed, in a way that’s just a delay, but it’s a 30-year delay. So I do think that notion that there are companies that don’t have a lot of debt so they aren’t as especially sensitive to the interest rate, that the term structure of their debt may be such that the average rates they’re paying might even be higher as the Fed cuts. I think that’s not a problem, that’s just a fact and we just need to understand it and see what the magnitude is…

…Tracy (16:29):

Yeah, but my question is going to be ultra simplistic. Can you explain to us in excruciating detail what exactly you expect happens in the economy now, as you cut interest rates? How does that cut get transmitted?

Austan (16:45):

Oof. Okay, as a general matter, the Fed has only one tool really, which is a screwdriver that can tighten or can loosen and I always say if your problem is, you know, a loose fender, that’s great. If your problem is can you make breakfast? No, you kind of can’t do that with a screwdriver.

So the main channels of monetary policy impact on the economy, I think are on the real economy side and they are on interest rate sensitive parts of the economy — like consumer durables, business fixed, investment construction, things like that.

Now there are other channels of monetary transmission where there’s a lot of argument. How important are they and they are, well if you change the value of assets, like the value of housing, the value of stocks, etc., is there a wealth effect so that consumer spending might go up as the asset values go up. Or if you contract and asset values go down, would that limit spending?

There’s a dollar channel that if rates in the US are moving relative to how rates are moving in other places, can affect the currency and that could affect imports and exports.

Those are probably a lot of the main channels and it’s always in the counterfactual. What would be happening if we didn’t do this? So to the extent that there’s already a debt structure or to the extent that we went through a business cycle that for the first time ever was not driven by cyclical industries, but was driven by services because nobody could spend money on that, and services aren’t especially interest rates sensitive, that’s another reason why you might think the monetary transmission mechanism, which is actually a whole bunch of different transmission mechanisms, just looks different this time than before.

Now everything that looks different is not bad. Okay, in a way this is frustrating that monetary policy doesn’t have the same impact, but at the same time in 2023 we hit what I called the golden path. Inflation came down almost as much as it ever came down in a single year, and there was no recession. And that never happened before. And so the unusualness of this thing, sometimes it’s good!…

…Austan (21:27):

Yes, does not necessarily. I agree with [that]. So let me finish two thoughts. One, did the Fed have anything to do with it? That’s kind of the question. If it was all supply shocks, then the Fed didn’t really, yes, the Fed can’t be blamed for the inflation going up, but then the Fed shouldn’t take credit for it coming down.

There is some component that as supply shocks heal, you get immaculate disinflation. I do think that the fundamentally different thing that happened this time than the last time we were getting supply shocks, like at the end of the 70s, is that the market expectations of inflation basically never went up. In the 70s, as actual inflation went up, the expectations went up. And part of what made the Volcker experience so hard is you didn’t have to just slay the inflation dragon. You had to go convince everyone that we will hold this thing underwater for as long as it takes until it surrenders and that’s a brutal process.

I do think that expectations stayed — even as actual inflation was almost double digits — stayed exactly at PCE 2% as the inflation target said, was fundamentally the Fed making a promise it may look bad but we’re going to get it back, and that the market de facto believed it. And that is to the Fed, is about Fed credibility, and I do think it made a big difference…

…Tracy (38:43):

That was perfect. Can I ask one more serious question before we wind it down? But you talked about restrictiveness earlier in this conversation and I get where that comes from and people look at things like real yields and stuff.

But if you look at stock market prices, we’re recording this on October 9th, I think stock indices are at records again. If you look at credit spreads, those are at multi-year lows. Where’s the restrictiveness? Because I don’t see it in parts of the financial market, let me put it that way.

Austan (39:14):

I’d say two things. I told you, my focus is primarily on the real side of the economy. I think those are the biggest, most impactful parts of the monetary policy transmission mechanism, historically.

So I’m less of a fan of interpreting financial conditions indices as a measure of monetary restrictiveness or what monetary policy should do because, in my view, it’s got a major reflection problem that, let’s say the market, which is forward-looking, decides they think it’s going to work, that there will be a soft landing, that rates are going to come down because inflation has been tamed and is at 2%. Then equity markets go up, long rates would come down and that would then be interpreted as a loosening of financial conditions and it would be like ‘Oh, you better stop cutting, you better raise.’ But that’s just self-referential. So I think that’s a little problematic.

And the inverted yield curve, for two years, which everybody has been saying is an indication that there’s about to be a recession, that’s not normal. If we go back to a regularly-shaped yield curve like we’re in more normal conditions, that’s not the end of the world.

My view of restrictiveness is we set the Fed funds rate, we set it high and held it there for more than a year and as inflation came down, the real Fed Funds rate just kept going up, passive tightening. That’s the highest the real Fed Funds rate had been in decades. And so to me that’s where the restrictiveness is.

4. Investing lessons from a mini-Berkshire Hathaway – Chin Hui Leong

Gayner believes mistakes of omission are far more costly than mistakes of commission.

He shared a personal example of passing on investing in Berkshire Hathaway (A shares) in 1984 when he first discovered the company.

At the start of 1984, shares were trading at around US$1,300. By the time he got around to buying some shares, the stock price had risen to US$5,750. Hence, he missed out on a gain of over 340 per cent.

I’ll add a second lesson to his point.

Shares of Berkshire Hathaway (A shares) closed at nearly US$694,000 per share last Friday. In other words, even though Gayner did not invest earlier, his shares are worth about 120 times more than what he paid.

While his returns could have been over 530 times if he invested earlier, I don’t think anyone would lose a smile with a 120-fold return.

So, here’s my take: if you find a great company with a promising future, it may not be too late to invest, even if the stock has already appreciated…

…When selecting stocks to invest, Gayner looks for four key factors.

The first is about finding a profitable business with minimal to no debt and a good return on capital. The reason is clear; starting with this pool of stocks increases your chances of finding a winner.

Secondly, he wants to have a talented management team with integrity.

Gayner may have taken a leaf out of Buffett’s playbook here. As Buffett once said, without integrity, the other positive management qualities, will work against you.

Interestingly, Gayner also connected the use of debt with management’s character.

For him, debt is a character marker.

In a podcast recorded earlier this year, Gayner recalled the advice of Shelby Davis, another legendary investor and mentor. Davis pointed out that in the absence of knowledge about a new business, the use of debt can be telltale sign.

Simply said, if a business is entirely equity-financed, the management team will have no incentive to steal from their own funds.

To be sure, this does not mean that a debt-laden company is fraudulent.

However, Gayner argued that leverage creates conditions for a dishonest management team to exploit since the money does not belong to them.

5. A Message From the Past (Thoughts on Nostalgia) – Morgan Housel

I was recently asked at a conference how investors should feel about the stock market given that it’s basically gone straight up over the last 15 years.

My first thought was: you’re right. If you started investing 15 years ago and checked your account for the first time, you would gasp. You’ve made a fortune.

Then I thought, wait a minute. Straight up for the last 15 years? To echo my wife: What are you talking about?

Are we going to pretend like the 22% crash in the summer of 2011 never happened?

Are we supposed to forget that stocks plunged more than 20% in 2016, and again in 2018?

Are we – hello? – now pretending that the worst economic calamity since the Great Depression didn’t happen in 2020?

That Europe’s banking system nearly collapsed?

That wages were stagnant?

That America’s national debt was downgraded?

Are we now forgetting that at virtually every moment of the last 15 years, smart people argued that the market was overvalued, recession was near, hyperinflation was around the corner, the country was bankrupt, the numbers were manipulated, the dollar was worthless, on and on?

I think we forget these things because we now know how the story ends: the stock market went up a lot. If you held on tight, none of those past events mattered. So it’s easy to discount – even ignore – how they felt at the time. You think back and say, “That was so easy, money was free, the market went straight up.” Even if few people actually felt that way during the last 15 years.

So much of what matters in investing – this is true for a lot of things in life – is how you manage the psychology of uncertainty. The problem with looking back with hindsight is that nothing is uncertain. You think no one had anything to worry about, because most of what they were worrying about eventually came to pass.

“You should have been happy and calm, given where things ended up,” you say to your past self. But your past self had no idea where things would end up. Uncertainty dictates nearly everything in the current moment, but looking back we pretend it never existed.

---

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 Markel (Tom Gayner is the CEO of Markel). 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 20 October 2024:

1. Actual Reform has materialised – Leonid Mironov

The Ministry of Justice and the NDRC have put out the draft of the Law on Private companies, or to give it’s full name, People’s Republic of China Private Economy Promotion Law. And it’s really good one…

…It emphasizes innovation, technological advancement, and participation in strategic industries, while also providing improved legal protections and equal treatment to address longstanding concerns. In return, private businesses are expected to follow Party leadership, contribute to national development, and operate in compliance with laws and regulations…

…The takeaway is that the private enterprises are now not discriminated against in the key project deployment. They will have similar cost of capital to the SOEs and will be able to supply most major national projects.

Chinese SOEs have been told to get more competitive earlier in the year, now the playing field is being somewhat levelled. The government, to my mind, is taking onboard the idea that employment is employment, whether SOE or not, and if a private enterprise can provide it, its fine.

I honestly think that this is the most consequential announcement, as it’s an example of a long-term reform that the government has committed to, and it is carrying out. This gives us hope for land and hukou reforms, as well as pension reform eventually. But also, this is a sign that there no decision to increase direct state participation in the economy but rather, assuming that companies follow guidance form the CCP, the more efficient actors, whether private or SOE, will drive the new policies.

2. Becoming Berkshire: 1969 – Illinois National Bank – The Weekend Investor

Around this time, Buffett and Munger sought a bank to purchase and found a candidate in Rockford, Illinois.

On April 3, 1969, Berskhire Hathaway, Inc. acquired 81,989 shares, out of a total of 100,000 shares outstanding, of the common stock of the Illinois National Bank and Trust Co. of Rockford, Illinois, at a cash price of $190.00 per share. They also have made a tender offer to acquire the remaining outstanding shares at the same cash price.

Buffett considered Rockford Bank one of the most well-run and profitable he had ever seen. It was managed by Eugene Abegg, who was 71 years old.

Abegg, who owned one-quarter of the shares, had been negotiating to sell the business to someone else before Buffett came along. The potential buyer had started criticizing the deal and wanted an audit. This affected Abegg, and he decided not to go ahead with the deal. Meanwhile, Buffett worked out what he was willing to pay, which turned out to be about $1 million less than the other buyers.

Abegg was so fed up with the other bidders that he pressured his fellow shareholders to accept Buffett’s offer, threatening to resign if they did not.

The crusty Abegg was just the type of fellow that Buffett liked.

In 1931, Eugene Abegg, a young man with only $250,000 of capital, formed a bank in Rockford, Illinois… It had $400,000 of deposits. Since then, no new capital had been added to the bank by its owners. Nevertheless, by 1969, Abegg had built, piece by piece, a bank with a net worth of $17m and $100m of deposits.

He carried thousands of dollars of cash in his pocket and cashed checks for people on the weekends. He carried a list of the number of unrented safe deposit boxes with him everywhere and would try to rent you a safe deposit box at a cocktail party. Mind you, this is the biggest bank in the second-largest city in Illinois at that time…

…The Illinois National Bank, which Buffett soon came to refer to by its colloquial name of Rockford Bank, had been chartered in the days before the U.S. Treasury assumed the exclusive right to coin money. Buffett was fascinated to discover that it still issued its own currency. The ten-dollar bills featured Abegg’s picture. Buffett, whose net worth was now more than $26 million, could have bought almost anything he wanted, but not this. Gene Abegg had done him one better. He and the United States Treasury had the privilege of issuing their own currency, but not the Buffett Partnership or Berkshire Hathaway. The idea of legal tender with your own picture on it captivated him. He began carrying a Rockford bill in his wallet…

…Berkshire paid $190 per share to acquire Illinois National, plus $2 per share to an investment bank for services rendered in the transaction…

…With total Assets of $117.3 million, shareholder equity of $16.8 million, and a net profit of $1.7 million in 1968, the bank had an ROE of 10% and ROA of 1.4%.

3. Jigar Shah on the Nuclear Power Revival in the US – Tracy Alloway, Joe Weisenthal, and Jigar Shah

Joe (04:30):

Thank you so much. So I’m going to start off with this question, which is: Okay, we went for a long time basically without building new nuclear power plants. It’s starting to pick up again. How much is it because something has changed policy-wise with subsidies and tax credits, et cetera, versus demand is back, therefore the economics of nuclear makes sense? Or would you say it’s not binary?

Jigar (04:53):

Well, look, I think that when you think about what happened through a historical context in the 1970s, we had high inflation and nuclear power was subject to high inflation. And so part of this is people were already worried about building new nuclear plants before the incident occurred because things were just getting more expensive. And when you think about the utility bankruptcies that occurred way back when, it was because they had cost overruns on nuclear power. And so I think that in general, it goes to when America stopped believing in itself and its ability to do big things and infrastructure. And I think this moment, with load growth and with the president saying we are going to build big things here, has gotten people thinking again, “Hey, what would it take to actually figure this out this time around?”

Tracy (05:48):

This is actually exactly what I wanted to ask you about because I was reading that the initial construction cost for Unit One of Three Mile was about $400 million. And I guess today the cost of building a nuclear plant would be like $5 billion, $10 billion. Obviously the $400 million isn’t adjusted for 1960s prices, but it does seem in general like it’s more expensive to build nuclear plants, certainly since the 1960s. Where did that additional cost increase actually come from?

Jigar (06:23):

So when you think about building things… like if you were to build multifamily housing and you would build one multifamily housing building versus building 12 throughout the city. You can imagine if you’re using the same design, it would be cheaper. The workers would get better. The first one would cost more, the second one would cost less, the third one would be even less.

You get faster. I mean, you see that when you go into a new home construction place. The first home takes it seems a lot longer and then suddenly the homes start popping up every week. This is the same with nuclear power. We trained 13,000 people to build the Vogtle nuclear plant in Georgia, and then we were done. And where did all those workers go? To other jobs. So now if we wanted to build Units Five and Six — we wanted to rebuild V.C. Summer [Nuclear Station] in South Carolina, which is like a hundred miles away — we’d have to go out and find another 13,000 workers. And so one of the things that we have to figure out how to do is to figure out how to build 10, right? And have those same workers that we trained, all those same EPC [engineering, procurement, and construction] contractors, all of those same suppliers, not have to stop and start, but we continue to do these one-off things…

…Jigar (09:42):

So when you restart a nuclear plant, the nuclear plant is viewed as new additional capacity, right? Because it was shut down. And so as a result, this technology agnostic credit that was created by Senator Wyden, right?

Because remember we always had the solar tax credit and the wind tax credit and all these other things. So over time the IRA moves us to a technology-neutral tax credit so that everything that is clean gets this technology-neutral tax credit. It’s a pretty lucrative tax credit. Depends on the technology, but let’s say 3 cents a kilowatt hour. And so now you’re in this place where you actually have a bonus production credit. Now you separately can choose to get an investment tax credit, but it happens to be that the production tax credit is more lucrative for these restarts of nuclear plants. But if you decide to do the investment tax credit, then you get the 30% tax credit, then there’s bonus tax credit.

So if it’s part of an energy community, you get an extra 10%, right? If you have a lot of domestic content, you get another 10%, right? So you could imagine that some of the folks who are building brand new nuclear plants might go that direction, but as a result of these incentives, nuclear power is now very cost effective.

Then the question becomes who actually wants to buy this power? Because wholesale market prices have been low. And so then the question becomes who wants to buy it? And it happened to be that two different utility groups in Michigan competed over wanting to buy all the output out of the Palisades restart. And so he picked one of the groups to buy that power and then that led to the project becoming financeable, right? And so once that succeeded, then Constellation was like, hell, maybe we could do this…

…Jigar (11:49):

So for a restart, you generally choose a production tax credit, not the investment tax credit. And that’s because the cost of restarting a reactor is a lot lower than the cost of building a brand new reactor. So you make more money by getting that extra 3 cents a kilowatt hour for the next 20 years. So the math there is you put up, it depends on where the final cost runs out, but let’s call it $1 [billion] to $2 billion to do the restart. And then you get this 3 cents a kilowatt hour multiplied by the number of kilowatt hours that plant creates. And remember, a nuclear power plant runs on average, in the United States, 92% of the time. So that’s a lot of kilowatt hours that comes out of that plant. Whereas with a solar farm, you might get 25% of the time production, with a tracking system. The math means that you could get almost all of your money back on the $1-to-$2 billion from the tax credits.

Then you’ve got the sale of the power that you’re signing a long-term contract for, and that’s where you make your return…

…Jigar (14:42):

Into the PJM [Interconnection.] And Microsoft says, depending on what happens with this power, we will make you whole on the payment. So if we said that we’re going to pay 9 cents a kilowatt hour and you end up getting 7 cents a kilowatt hour, we’ll pay that 2-cent difference. And that includes not just the kilowatt hour price, but also includes the capacity payment. So you may have heard that the PJM had a very large increase in the price that the capacity payment cleared and the capacity payment is essential, because it convinces the coal plants or the natural gas plants or others who are sort of at the end of life to make investments to last a little longer because they got paid a capacity payment to stay open. So the pieces that come here are both a capacity payment and the energy payment, and Microsoft is saying that we get all the attributes, so we get to call our usage green, but separately, if for whatever reason the wholesale market value for what the nuclear power plant is creating is less than the strike price that we agreed to, then we will make you whole…

…Jigar (19:34):

It really is an extraordinary thing. I think that most people view electricity like water. So you just put a bigger pump in, you put in more pipe, it gets to your house, you got hot water, that’s great. It’s not like that at all. There is this complex physics equation that you have to solve for.

Joe (19:53):

Because the grid has to be in perfect balance all the time, right?

Jigar (19:56):

Well, so there’s the perfect balance between supply and demand. But then there’s also figuring out what the constraints are of each individual segment on the transmission line.

So if you’re using power in New York City and you’re creating a lot of extra power out of the nuclear plants in Illinois, then that power has to go via Indiana and Ohio and then through Pennsylvania to New York City, and they may or may not be able to carry that much. And so they have to do these studies. So every time you try to add something to the grid, they have to do a study and they have to figure out whether that capacity is there, how often it’s there, whether it would continue to be balanced or whether it would be imbalanced.

And so the big fight there is that… so in Texas what they do is they just look at the safety part of it, but they don’t look at the capacity part of it. They just say, “You connect at your own risk and if we’re clogged, we’re just going to tell you to shut down, and that’s on you.” That’s why they’re approving people super fast. Whereas with the PJM and others, they’re saying, “Not only are we’re going to do a safety study, we’re also going to do a capacity study and we’re not going to let you connect until this other generator shuts down and frees up capacity for your generator.” And so that then makes the wait time much longer.

4. Invest Local? – Victaurs

Well, a community bank in the U.S. is generally defined as a depository or lending institution that primarily serves businesses and individuals in a small geographic area. These banks emphasize personal relationships with their customers and often have specialized knowledge of their local community and customers. They tend to base credit decisions on local knowledge and nonstandard data obtained through long-term relationships, rather than relying solely on models-based underwriting used by larger banks…

…As of a year or two ago there were roughly $25 trillion in assets in the entire U.S. Banking system.

And the entire amount of assets in the Community Banking system is … drumroll please … $4.8 trillion for a grand total of 19.2%. Only 1/5 of all the assets in the system are controlled and managed by these smaller banks…

...When people don’t bank locally, they inadvertently contribute to a cycle that can harm their local economies:

Capital Drain: Deposits in non-local banks are often invested in national or international ventures, rather than being reinvested in the local community

Reduced Access to Credit: As community banks disappear, so does their deep understanding of local economic conditions and business opportunities.

Loss of Personalized Service: Large banks often use standardized lending criteria that may not account for local economic conditions or individual circumstances.

Economic Homogenization: As local banks disappear, communities lose a key institution that helps maintain their unique economic character.

Decreased Local Decision-Making: When banking decisions are made in distant headquarters, local economic needs and opportunities may be overlooked.

I don’t want to over dramatize the situation, but do any of these things sound good to you? And given lots of us grew up in small towns, love where we came from, owe our position in life to the kindness of a HS coach or the first job at a local restaurant, do you want capital to move away from these people? I don’t think I do.

Banking is numbers, so here are some numbers because they paint a stark picture:

• For every $100 deposited in a local bank, $58 is reinvested locally. For large banks, that number drops to just $36. This isn’t to demonize big banks, only to point out the facts.
• Community banks make 60% of small business loans, despite holding only 12% of all banking assets. (I know their 12% doesn’t jive with my 19%).
• When a community bank closes, the local area experiences an average 33% reduction in small business lending for several years. I highly recommend checking out this study. This is an awful second level impact of losing community banks…

…As of 2023, the United States is home to a staggering 33.2 million small businesses. These enterprises employ 61.7 million people – that’s 46.4% of all U.S. employees. To put it in perspective, if small business employees formed a country, it would be the 23rd most populous nation on Earth, just behind Italy. That was pretty crazy to me. Imagine if all of the small businesses went away?

But it doesn’t stop there. Small businesses are the dynamos of American innovation and economic activity:

• They generate 44% of U.S. economic activity.
• They create 1.5 million jobs annually (64% of new jobs created) – that’s like creating a new city the size of Philadelphia every year, filled entirely with new job holders. And even for those of us who aren’t Eagles or Phillies fans, we can agree this is a massive deal.
• They contribute to 33.6% of known export value and represent 97.5% of all exporters in the United States.

“Small businesses are more than just economic units,” says Dr. Emily Chen, economist at the Small Business Administration. “They’re the innovation labs of America, constantly adapting and evolving to meet new challenges and opportunities.”…

…This is a repeat stat, but worth mentioning again. Community banks provide 60% of all small business loans, despite holding only 12% of all banking industry assets. It’s as if the local high school football team was outscoring all the pro teams combined!

They make 80% of agricultural loans, forming the financial backbone of rural America.

During the COVID-19 pandemic, community banks processed 57.5% of all Paycheck Protection Program (PPP) loans, saving countless small businesses.

Community banks have over 50,000 locations nationwide, compared to about 18,000 locations for the largest banks. That’s like having a friendly neighbor on every block, compared to a distant acquaintance every few neighborhoods…

…Community banks have consistently demonstrated resilience in the face of economic challenges:

During the 2008 financial crisis, community banks continued to lend when larger banks pulled back, increasing their small business lending by 5.2%.

In the recent COVID-19 pandemic, community banks were often the first to step up, offering forbearance and emergency loans to struggling local businesses.

66% of small businesses that received PPP loans from community banks said the process was “easy,” compared to 51% for large banks…

…In 2005, Hamdi Ulukaya bought a defunct yogurt factory in New Berlin, New York, with the help of a Small Business Administration loan backed by a local bank. From this modest beginning, Chobani has grown into a billion-dollar company, employing thousands and revolutionizing the yogurt industry.

“Without that initial loan and the trust of our local bank, Chobani might never have existed,” Ulukaya has said. “They believed in us when no one else would.”

5. The next tectonic shift in AI: Inference – Rihard Jarc

To simplify it, the o1 model has a backtracking ability. The model predicts something, realizes it did something wrong, goes back, erases that, and comes back and predicts again from that point.

The most significant implication of this kind of model is that inference workloads should grow substantially more than we were expecting in the pre-o1 period.

The calculation for Inference is now not just the number of users using it multiplied by the number of times they use it. The model can now take 10x or even more time on inference compute to come up with an answer. So inference also becomes part of the accuracy process.

The second big implication for investors is that inference computing is now becoming a new scaling paradigm. So, you not only scale the model with what is now known as data and training compute, but you can also scale them with more inference.

Noam Brown, an OpenAI researcher, has said that a study on the board game Hex using AI found that if you have 15x the inference compute, it equals 10x the training compute.

The fact that you can now scale LLMs via inference means that:

A. You can have smaller models that you dedicate more inference compute that can be as good as bigger parameter models with less inference compute

B. Inference computing is much cheaper than training computing, but the market for inference will be vastly bigger than training computing. In my discussion with Sunny, I asked Sunny how big he thinks, as an industry insider, the Inference market will be; Sunny revealed that he had the chance to preview an interview with Jensen Huang, the CEO of Nvidia, where Jensen said that Inference will be 1 billion times larger than Training. Sunny added that it makes sense to think that a model is going to be used billions of times before it is updated (trained) again.

It is also important to note that the Inference chip market has much more competition than the training market, where Nvidia dominates. From an industry expert:

»Training also is notoriously hard because you need special architectures and special cards and interconnects between the clusterand RDUs and stuff like that. It’s mostly dominated by NVIDIA because they’ve done the best work there. Inference is interesting because inference can be done anywhere. Inference is very, very easy to do on any hardware. Training is harder.«

This means that other companies will be able to reap the benefits of inference chips besides Nvidia. It also means margins on inference chips are not going close to Nvidia’s margins on its training GPUs, where it basically has a monopoly.

It also opens a path for some companies to lower some of their costs, and instead of going heavy on training GPUs and scaling there, they can split some of that on inference chips and still scale the models. Inference for customers is vastly cheaper than training…

…The thing that I also didn’t mention, but because of the o1 model release and the fact that we are coming to the start of Big Tech reporting earnings, I believe there is a high chance that the hyperscalers and companies like Meta, who are building these LLMs will increase their CapEx expectations now even more in the short-term than what they did before and much higher than analysts expect. The reason is that they now have to account for spending on Inference compute to improve these models. Inference was before a cost that they could gradually introduce and control more with users getting limited access to AI features, etc. This has changed, and you can use inference to scale the model. This might not be what investors will like in the short term. Still, it is something that, in the long run, brings us even more capable models, possibilities of easier agentic AI use cases, and SLMs that have a good enough accuracy to be used more often compared to bigger LLMs. There are already estimates on how much the inference costs are more expensive with an o1 model than with »pre o1 models«. This industry expert quantifies how much more expensive it is:

» Analyst: Strawberry o1, I’ve been told it’s 4X-5X more expensive than ChatGPT?

Industry Expert: Yeah. That’s the right level. That adds up given that it will essentially use 4X-5X more tokens on average. In the worst case, it will be 10X possibly. The 4X-5X is an average number of how much more expensive it is.«

---

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 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 13 October 2024:

1. An Interview with Meta CTO Andrew Bosworth About Orion and Reality Labs – Ben Thompson and Andrew Bosworth

Orion, Meta’s AR glasses, is spectacular. I must start with the caveat that this is not a shipping product; the glasses that I tried felt like a consumer-ready product, but they reportedly cost $10,000 each, and Meta has decided to hold off on shipping a consumer version until they can bring the price down. That will be a tall order, and that challenge should be kept in mind with everything that follows.

What follows is unadulterated praise. Orion makes every other VR or AR device I have tried feel like a mistake — including the Apple Vision Pro. It is incredibly comfortable to wear, for one. What was the most striking to me, however, is that the obvious limitations — particularly low resolution — felt immaterial. The difference from the Quest or Vision Pro is that actually looking at reality is so dramatically different from even the best-in-class pass-through capabilities of the Vision Pro, that the holographic video quality doesn’t really matter. Even the highest quality presentation layer will pale in comparison to reality; this, counter-intuitively, gives a lot more freedom of movement in terms of what constitutes “good enough”. Orion’s image quality — thanks in part to its shockingly large 70 degree field of view — is good enough. It’s awesome, actually. In fact — and I don’t say this lightly — it is good enough that, for the first time ever, I felt like I could envision a world where I don’t carry a smartphone.

Orion is a standalone product, at least in terms of needing a phone; instead there is a “puck”, an oblong unit that holds the compute for the operating system and connectivity, and which connects to the glasses wirelessly. The glasses themselves contain the compute necessary for low-latency calculations that pertain to the actual display. One challenge I see in this model is input: voice works well, and the wristband that detects the electrical signals in your arm worked flawlessly for me — you can control your glasses with your hand without anyone knowing — but I wouldn’t mind if that “puck” contained a Blackberry-style keyboard for extended text entry…

…Was there any aspect of the Vision Pro going so high end that also made you re-anchor on the low end, or do you think you would’ve ended up here regardless?

AB: No, I think we would’ve ended up here regardless. I mean, listen, I love that the Vision Pro — people won’t believe me — I love that they went maximalist. Just like, “What if we just take this dial and turn it to 11, and let the rest of the system fall where it does?”, and you see why we haven’t done that, just in terms of weight and cost. Like yep, that’s what it takes to take this dial and turn it to 11.

And this is why I think you and Mark right away seemed almost relieved by the Vision Pro.

AB: Your only real fear when a competitor launches a product is that they’ve had a breakthrough that you haven’t had. That there’s something that they’ve figured out, some technical thing that you haven’t figured out, because then they have a sustaining advantage potentially for some period of time until you can beat them on it. So I think whenever a device comes out, it’s like, “Oh good, this is all made with materials that we are aware of, this is all made with technologies that we have access to”.

“We understand why it costs this much, why it weighs this much.”

AB: We could have built this, we chose not to build this. It is both great for the world that there’s people exploring different quadrants of the space. By the way, if the Vision Pro had sold really well, of course we’d be changing our strategy. We’d be like, “Oh, okay, cool, there’s actually a bigger market than we’ve realized up there, let’s go do it”. And I think, by the way, I do actually think there will be a market there, when there’s software.

Are you surprised at how little content Apple’s released with Vision Pro?

AB: It’s how do you get the content? Before you have the devices and it’s a chicken-and-egg problem where it’s like, “Hey, okay cool, you have these devices out there, but there’s not enough for me to build my content for”.

Is part of the low cost a bet that if the egg is the end market, that’s the most important part?

AB: One hundred percent. We’ve talked about this all the time, you almost always hear me talk about the Quest ecosystem. I’m not talking about the Quest line of devices, I’m talking about building as big an audience as I can for developers to target so they can sell their software, so there’s more developers, that brings more consumers, and you have flywheel that way. Then at some point, that’s how you power your way up market. That’s how you power your way to, “Hey, we can now sell higher margin, higher end devices because there’s plenty of stuff”.

Well, to that point, Mark talks about, “In every market there’s the integrated version and the modular mass market one”, but if you go back to the PCs, Microsoft swept the market. Now one thing that’s important about that era that’s different from the smartphone era is in the smartphone era, Apple was first. In the PC era, DOS was first, so Microsoft was actually first, so they actually had developers first. At this point, seeing the Vision Pro, seeing what’s happened over the last six to nine months, are you shifting from a, “Yeah, we can both be winners here”, to, “We’re going to win the whole thing”?

AB: Man, I feel good about our position, if that’s what you’re asking.

I want to pretend I’m turning off the mic and getting your honest thought.

AB: With me, you’ll always get an honest thought, I have to make sure I’m phrasing this in a clever way.

The only reason I’m being careful here is I think — I don’t really want to be antagonistic with anybody, including Apple, I think it’s great that they’re investing, I want them to continue invest. Actually the Vision Pro has caused a surge of interest in investing in the entire space, including in us. I’ve gotten calls in the last couple of months especially that I would not have gotten, had Apple not launched the Vision Pro, and if they weren’t courting people to consider that there’s going to be a follow-on version. So it’s really, really healthy to have that competition. Good for consumers, good for us.

I also think that right now, if you’re a developer, you’d be an idiot not to build for us first, we have an audience that can actually go buy your software. It’s big enough to sustain you, and then yeah, no problem, bring it over to Apple Vision Pro after that.

Is your bigger concern losing to Apple, or that a market never materializes for these devices?

AB: Oh, good question. Yeah, my biggest concern is that the market gets capped somehow, like it doesn’t take off. The thing I worry about with Apple specifically is that they have their phones and devices so locked down that they can self-preference a ton. So they can easily, you look what our Orion glasses, these full AR glasses, incredible. We’ve got custom silicon in the glasses, we’ve got custom silicon in the puck, but Apple could build all that and just be like, “Oh, it only works with us,” which they’ve already done with the AirPods.

They don’t need a puck because they have a phone.

AB: They already have a phone, and they did this with Airpods.

Or the Apple Watch.

AB: Apple Watch. Those aren’t the best possible things you could build, but no one else is allowed to build those things, so it’s like, “Oh cool”, so if I have a concern about Apple, it’s not the competitiveness or non-competitiveness of their headsets, it’s that they’re going to bundle into their ecosystem in a way that really makes it hard for us to compete…

…This is the first device I’ve ever used that — I know you guys have been saying it — that genuinely feels post-phone.

AB: It could do it, right?…

…I’ll be totally honest, after using Orion, I’m excusing you all your billions of dollars a year spent, that’s how incredible it is, but I do think one of the critiques, and you talked about it when you went in, this was an entity that had two completely different camps that want to go in totally different directions. Then even a few years after you were there, you’re having an operating system bake-off for years instead of months and then it’s, “Should we do processors? Should we partner with someone doing processors?”. What is the forcing function that is getting you into, “Okay, we’re going to stop experimenting and actually start building”? What got you to that point? Was it the Year of Efficiency? Was there a bit where, “Look, we have to lay off half the team, so we’ve got to decide which half”?

AB: I love this question. It was before the Year of Efficiency hit. I think it’s not uncommon, you have these expansionary periods where you’re like, “We don’t know what matters yet, we truly don’t know what technology is the right technology, we don’t know what operating system is the right operating system, we don’t know what trade-offs matter yet”. So if you want to be successful with high confidence in a certain timeframe, it pays to parallel path a ton of stuff.

But how long do you parallel path it?

AB: We honestly turned the corner with Quest 2, especially when we had mixed reality in sight. That started the process, and now you’ve got to a point with a mixed reality with our metaverse division where it’s extremely focused, have a very clear vision of what good looks like, have a very clear ability to discern this is the path, this is not the path. As a consequence, you can be really, really much more efficient with your resourcing, your parallel pathing list, you’re just blitzing the things that matter more.

With augmented reality, Orion, a year ago, we actually hit this point where we’re like, “Okay, we believe in this, we see it, we have a really clear sense of where we’re going with this”, and you know what really helped a lot with that was the Ray-Ban Meta glasses as well. Cool, it’s not just that we have this distant AR thing, we actually have an entire family of devices coming before that that also matter.

Did AI save Reality Labs?

AB: Oh my gosh. So AI, because FAIR, the Fundamental AI Research group reported to me until this year. We just moved it over to join the rest of the AI stuff with Chris, and I don’t know if it saved us, but it’s a wonderful tailwind, it’s the first tailwind I can remember having. For us, it’s mostly just headwind after headwind after headwinds like, “Oh, guess what? This thermal performance is worse than you thought, this battery life is worse than you thought, the efficiency is worse than you thought”, and so we finally got a tailwind. We finally got a thing that showed up before it was expected, which was AI.

So I think to answer your first question, each of these devices has gone through an expansionary period and a contractionary period where it expands until you feel like you have a good understanding and intuition of what good looks like, and then you can start to prune and then you can get really good about pruning. Today our architecture is really tight, hand tracking, eye tracking, face tracking, Codec Avatars, these are shared technologies, they work in both VR and AR, and we have a single shared team building those technologies. Separately, the operating system for AR has to be its own operating system because it turns out the use cases, what you actually do, the interaction paradigm, completely different.

2. Xi Jinping is worried about the economy – what do Chinese people think? – Kelly Ng and Yi Ma

What is less clear is how the slowdown has affected ordinary Chinese people, whose expectations and frustrations are often heavily censored.

But two new pieces of research offer some insight. The first, a survey of Chinese attitudes towards the economy, found that people were growing pessimistic and disillusioned about their prospects. The second is a record of protests, both physical and online, that noted a rise in incidents driven by economic grievances.

Although far from complete, the picture neverthless provides a rare glimpse into the current economic climate, and how Chinese people feel about their future…

…The slowdown hit as the pandemic ended, partly driven by three years of sudden and complete lockdowns, which strangled economic activity.

And that contrast between the years before and after the pandemic is evident in the research by American professors Martin Whyte of Harvard University, Scott Rozelle of Stanford University’s Center on China’s Economy and Stanford masters student Michael Alisky.

They conducted their surveys in 2004 and 2009, before Xi Jinping became China’s leader, and during his rule in 2014 and 2023. The sample sizes varied, ranging between 3,000 and 7,500.

In 2004, nearly 60% of the respondents said their families’ economic situation had improved over the past five years – and just as many of them felt optimistic about the next five years.

The figures jumped in 2009 and 2014 – with 72.4% and 76.5% respectively saying things had improved, while 68.8% and 73% were hopeful about the future.

However in 2023, only 38.8% felt life had got better for their families. And less than half – about 47% – believed things would improve over the next five years.

Meanwhile, the proportion of those who felt pessimistic about the future rose, from just 2.3% in 2004 to 16% in 2023.

While the surveys were of a nationally representative sample aged 20 to 60, getting access to a broad range of opinions is a challenge in authoritarian China.

Respondents were from 26 Chinese provinces and administrative regions. The 2023 surveys excluded Xinjiang and parts of Tibet – Mr Whyte said it was “a combination of extra costs due to remote locations and political sensitivity”…

…In 2004, 2009 and 2014, more than six in 10 respondents agreed that “effort is always rewarded” in China. Those who disagreed hovered around 15%.

Come 2023, the sentiment flipped. Only 28.3% believed that their hard work would pay off, while a third of them disagreed. The disagreement was strongest among lower-income families, who earned less than 50,000 yuan ($6,989; £5,442) a year…

…There are other indicators of discontent, such as an 18% rise in protests in the second quarter of 2024, compared with the same period last year, according to the China Dissent Monitor (CDM).

The study defines protests as any instance when people voice grievances or advance their interests in ways that are in contention with authority – this could happen physically or online. Such episodes, however small, are still telling in China, where even lone protesters are swiftly tracked down and detained.

A least three in four cases are due to economic grievances, said Kevin Slaten, one of the CDM study’s four editors.

Starting in June 2022, the group has documented nearly 6,400 such events so far.

They saw a rise in protests led by rural residents and blue-collar workers over land grabs and low wages, but also noted middle-class citizens organising because of the real estate crisis. Protests by homeowners and construction workers made up 44% of the cases across more than 370 cities…

…Between August 2023 and Janaury 2024, Beijing stopped releasing youth unemployment figures after they hit a record high. At one point, officials coined the term “slow employment” to describe those who were taking time to find a job – a separate category, they said, from the jobless.

Censors have been cracking down on any source of financial frustration – vocal online posts are promptly scrubbed, while influencers have been blocked on social media for flaunting luxurious tastes. State media has defended the bans as part of the effort to create a “civilised, healthy and harmonious” environment. More alarming perhaps are reports last week that a top economist, Zhu Hengpeng, has been detained for criticising Xi’s handling of the economy.

3. Will Hurricane Helene Cause a Chip Shortage? What the Major Chipmakers Are Saying – Tae Kim

Hurricane Helene flooded and damaged the local infrastructure in Spruce Pine, making some roadways impassable, according to local news reports. Sibelco and The Quartz Corp., the two companies that manage the quartz mines in the town, have both temporarily shut down mining operations.

High-purity quartz found in Spruce Pine is a key material used in the production of silicon wafers that are used to make semiconductors. Quartz’s ability to withstand extreme temperatures is useful for making crucibles or containers that hold the melted polysilicon material used to produce wafers and solar cells.

According to Vince Beiser, author of The World in a Grain, the two companies’ Spruce Pine mines provide 70% to 90% of the world’s production of high-purity quartz used for the semiconductor industry…

…Ed Conway, author of Material World, a book about raw materials, posted that Spruce Pine quartz mines are unique in terms of purity and consistency, and finding another high-purity source would take months or possibly years.

But in a bad scenario, where the mines are offline for months, the chip industry may be insulated. “The significance of supply disruptions from the [Spruce Pine] mines is exaggerated,” Dylan Patel, chief analyst at SemiAnalysis said.

Patel added that the raw wafer companies had months of inventory, there are other countries that have high-purity quartz mines, and there are methods to purify lower-quality quartz or create synthetic quartz crucibles.

4. The Truth Behind the Highlight Reels – Thomas Chua

People are often shocked when seemingly perfect couples announce a divorce or breakup, even though their social media showed nothing but happiness just days earlier. What’s hidden from view are the realities that unfold behind the scenes—disagreements, financial pressures, or emotional distance.

The same kind of comparison happens in investing. We look at people like Warren Buffett, who delivered an incredible 30.4% annualized return during his partnership years (1957–1969), or Peter Lynch, who achieved 29.2% annualized returns at Fidelity’s Magellan Fund (1977–1990). Their results are awe-inspiring, but we rarely consider the personal price they paid to achieve them.

Warren Buffett’s biography, The Snowball, talks about how he spent his days working and his nights poring over Moody’s Manual. While his wife, Susie, took care of him wholeheartedly and assumed the responsibility of managing their household and raising the children, Buffett’s mind remained elsewhere. Even during family trips—like a visit to Disneyland—he would sit alone, engrossed in reading.

This single-minded focus on work created a widening distance between Buffett and his family. His children longed for his attention, and Susie craved a deeper connection. The strain eventually became too much, leading to Susie’s departure…

…Buffett and Lynch’s legendary results required intense focus and commitment, often at the expense of their relationships. This is not unlike the sacrifices elite athletes make, dedicating everything to training, diet, and recovery to reach the pinnacle of their sport…

…When it comes to investing, we need to ask ourselves: What’s the price we’re willing to pay? How much time, energy, and money are we truly prepared to invest? 

5. X (or Twitter) thread on China’s stimulus – Adam Wolfe

Why is China’s stock market booming if most economists, including me, don’t think the stimulus measures announced or reported go far enough to solve China’s economic problems or even its cyclical slump? I think it’s how the stimulus has been designed. 1/

The measures aimed at the real economy are mostly incremental, small, and inconsequential. But the measures that the PBoC announced to support the stock market are new, unlimited, and significant. 2/…

…Start with the 20-bps policy interest rate cut. 3/

PBoC Governor Pan was at pains to say this is as much as he could do. Rate cuts pass through to loan rates faster than deposit rates, so to keep banks profitable, he couldn’t offer any more. That also implies he won’t be cutting rates further soon. 4/…

…Existing mortgage rates were cut, too, saving some CNY150bn in interest payments per-year, according to Pan. But they would have adjusted in January anyway, so the actual savings are ¼ of that. They did the same thing last year. It had no macro impact. 7/

Lower down payment requirements for second home purchases? Done that before, too. It didn’t lead to higher sales of new homes. 8/…

…Another CNY1tn will be used to support consumption. About half of that would go toward extending the cash for clunkers programs. Those have helped specific industries but have had little macro impact. And the impact is getting smaller the longer these programs run. 11/

The only new thing for the real economy is the reported program that would give a monthly allowance to families with two more kids. I estimate that a bit over 10% of families would qualify, so it would cost about CNY500bn/year. 12/

This could have a big multiplier effect on growth, but CNY500bn is small beans. Plus, temporary support measures like this tend to be saved…

…But what about the stock market? The PBoC will set up a CNY500bn facility for institutional investors to higher risk assets for safe assets from the PBoC. This higher-quality collateral would then allow the investors to take on more leverage to buy more stocks. 14/

The PBoC will also open a CNY300bn re-lending window to encourage banks to finance the repurchase of stocks by listed companies. Pan made a point to say that both programs could be doubled or tripled if they work. The sky is the limit! 15/…

…Inflating a bubble in stock prices without doing much to boost earnings could end in tears. Alternatively, sucking liquidity out of safer assets like bonds could lead to another “redemption crisis” for WMPs/bond funds and losses for households. 19/

---

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 Apple, 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 06 October 2024:

1. My China’s Travelogue: China’s Super Apps, Western Brands, EV Boom and Challenges – Thomas Chua

When I arrived in Beijing, I activated my travel card directly through WeChat or Alipay, scanned a QR code, and that was it—smooth sailing. This seamless, mobile-first approach isn’t just a clever solution; it’s a necessity in a country of such a large scale. It offers an elegant, low-cost, and scalable system that is incredibly user-friendly and available in every city, eliminating the complications associated with traditional transport infrastructure.

In many ways, China’s lack of legacy infrastructure—often seen as a disadvantage—turned out to be a blessing. Without the entrenched banking systems and reliance on Mastercard or Visa that many developed countries face, China was free to reimagine its financial infrastructure from scratch. Where traditional payment systems come with a web of intermediaries and fees, China’s mobile payment platforms—WeChat Pay and Alipay—were built to be faster, cheaper, and more efficient.

However, a super app like WeChat does more than just make payments. It’s a super app, offering a vast range of services, from messaging and social networking to ride-hailing and food delivery. This super-app model emerged out of necessity, as many users in lower-tier cities had limited smartphone storage and data plans. WeChat’s mini-programs solved this problem, providing lightweight apps without requiring additional downloads or installations…

…I noticed all my rides were EVs, as were most cars on the road. I barely saw a dozen internal combustion engine (ICE) vehicles and didn’t come across a single petrol station in Beijing or Xi’an.

Curious about China’s rapid EV adoption, I began asking my drivers about the EV scene and learned that while many in Beijing drove 北汽 (BAIC) EVs due to their low price, the brand wasn’t highly regarded, with past models having plenty of problems. Surprisingly, I didn’t see many BYDs on the road in Beijing, but the drivers told me the brand was more prevalent in Shanghai and Hangzhou.

Tesla was perceived as superior but expensive. When I searched Baidu for the most popular EV, Tesla’s Model Y ranked as the most popular car in 2023.

How did China achieve a 54% electric vehicle (EV) penetration while other countries struggled? One key factor in Beijing is the stringent quota on car sales. To bypass the long wait for a vehicle (a local I spoke with has been waiting for over three years), consumers can jump the queue by opting for an EV instead of ICE or by having children.

When the policy of prioritizing EV purchases was first introduced in the mid-2010s, many people were hesitant to adopt EVs due to concerns over charging infrastructure. In those days, if you decided to buy an EV, you could have gotten the permit immediately. Today, those concerns have disappeared, with private companies installing charging points everywhere…

…During my trip, I visited several top consumer brands and observed foot traffic. Nike stores, to my surprise, were practically empty.

While not as deserted, Starbucks also saw far fewer customers than I expected. Li Ning had moderately more customers, while Lululemon was bustling. Even on busy streets, Starbucks was quiet, with consumers favoring milk tea brands like Chagee or 茶话弄 (Cha Hua Nong). Locals guessed this was due to the weak economy and Starbucks’ high prices.

However, this logic doesn’t fully explain Lululemon’s popularity despite its high prices. My guess is that Starbucks isn’t innovating fast enough for the fickle Chinese consumer. Competitors introduce new flavors at a much faster pace. And if Starbucks’ competitive advantage lies in its mobile app and loyalty program, it holds no advantage in China, where every food and beverage brand has a WeChat presence and can easily implement loyalty programs.

2. TIP661: Betting Big On China & Lessons From Bear Markets w/ Richard Lawrence – Clay Finck and Richard Lawrence

[00:12:23] Clay Finck: So is that a lesson of being more critical of the management teams you partner with? Did that help kind of hone in that side of the process?

[00:12:32] Richard Lawrence: Well, I think that number one, pyramid structures like that are out of bounds for any investor should never invest in a pyramid structure like that.

[00:12:42] Clay Finck: For those who aren’t familiar, what is the pyramid structure?

[00:12:45] Richard Lawrence: Well, the pyramid structure is the family owns 50. 1 percent of a public company that then in turn owns 50. 1 percent of another company and they own the asset. And the value of that asset is transferred to the owners of the top company and shareholders down below basically get nothing.

[00:13:03] Richard Lawrence: And they do it through any number of mechanisms. But paying dividends to minority shareholders, they’re generally not very big on. So you do it through excessive salaries, fees, costs. And so we learn, don’t invest in certain things.

[00:13:17] Clay Finck: Let’s transition here to chat about the Asian financial crisis. Many in our audience are likely aware of the crisis that occurred in Asia in 1997 and 98, but it’s not something we’ve covered too much on the show here.

[00:13:30] Clay Finck: I should also mention that many of our listeners are based in Asia as well. So you refer to this time period as nothing less than an economic nightmare, and there was really nowhere to hide for investors in these markets. How about you give us a sense of what was happening during this crisis that just made it so difficult for Asian economies?

[00:13:49] Richard Lawrence: Yeah, well, Clay, I was having a nice time here talking to you, and now you have to bring up memories of 1997 98. It churns my stomach still today with everything that went on. So, and really with the passage of time, there are fewer people that really understand the extent of the obliteration of value.

[00:14:09] Richard Lawrence: There was the inception of 97, 98 went back years that basically Asian economies were not able to self finance their growth. They were running what economists called current account deficits. And so they were having to borrow US dollars to fund their growth. And that US dollar debt built up, built up, built up.

[00:14:30] Richard Lawrence: And at the same time, they had fixed currencies. To the U. S. dollar. So all the companies were saying, well, if I can borrow dollars at the same price as my Korean won or my Thai baht or my Indonesian rupiah, I’m just going to do that. And so companies borrowed dollars, governments borrowed dollars, everybody was borrowing dollars and no one was hedging the currencies.

[00:14:54] Richard Lawrence: And then on July 2nd, 1997, the bot broke and it went from 25 eventually hitting 55 or 60 and it obliterated balance sheets. If you had 100 million of equity and 100 million of unhedged dollar debt, all of a sudden your debt grew 4 times in size and you needed 4 times the business to repay that debt. Of course, you didn’t have it.

[00:15:18] Richard Lawrence: Particularly when the economies were in recessions. To me, looking back, the dollar debt was really the biggest mistake. Current account deficits were a reflection of that. And the rupee in Indonesia went from 3000 to 15, 000. The one got obliterated within 6 months of July 2nd, 1997, the International Monetary Fund was called in to bail out Korea, Thailand, Malaysia, and Indonesia, which were the main markets at the time.

[00:15:49] Richard Lawrence: China was still a closed economy at the time. So they handled their own problems kind of internally out of sight the stock markets in various countries, like Thailand, Indonesia, went down nearly 90%. The real estate index in Thailand, which was a composite index, went down something like 98%. To do that, I calculated, you go down 80%, and then you go down another 80%, and then you go down another 80%.

[00:16:16] Richard Lawrence: That’s what going down 97, 98 percent is like. And so it was just a complete obliteration. We bottomed at Overlook. We went down from top to bottom about 65%. We really put a herd on our fledging investment management fund at the time, and we bottomed at about 4. 6 times earnings, about 0. 7 times book.

[00:16:37] Richard Lawrence: Interest rates rose because they had to protect their currencies. In Indonesia, the interest rates rose to 99%, and it didn’t stop the currency from declining. But the Indonesians couldn’t raise the interest rates higher because their banking system couldn’t accommodate three digits for interest rates.

[00:16:57] Richard Lawrence: And so interest rates meant nothing in Hong Kong, which had no dollar debt. It had a fixed currency to the dollar, but it was backed by the peg of the Hong Kong Monetary Authority and the government of Hong Kong. Interest rates went to 36 percent and they had no debt. So, you can just see the massive exodus.

[00:17:15] Richard Lawrence: It was a one way street out of Asia, and there I was, left. When eventually, 17 months later, I went out to Photon, and went in and met my friend Michael Chan, and went in and met the guys at Kingboard Chemical, and we survived by the chin of our chinny chin chin, as they say. But it was really something it was a lot of calls at night to my investors, you know, was pre zoom calls.

[00:17:39] Richard Lawrence: It was pre internet. I’d call them up and I’d say this is Richard Lawrence calling from Hong Kong and I’d get right through to the executive because calling from Hong Kong. It must be really expensive phone call. It was morning for the executives, the investors, and I totally wrecked their day. We went down 10 percent 5 straight months, maybe 6 straight months.

[00:17:58] Richard Lawrence: Now, you know why I feel so miserable answering, talking about 97, 98, but of course, like all bear markets, bear markets, self correct. And today, Asia runs current account surpluses. Today, Asia has really good balance of, with their government budgets. And we don’t have unhedged dollar debt. We have huge forex reserves.

[00:18:18] Richard Lawrence: All of that high savings rate, all of that’s come. The inception of that, all of it has come from 97, 98. We all, everybody. Any government official, corporate official, investor, we all learn those lessons. That’s why I think 97, 98 is such a pivotal time. We’ve had other bear markets, but they’re like water on ducks back…

…[00:22:33] Clay Finck: And you even mentioned in your book that you regretted ever listening to Buffett when it came to paying attention to macro. Is there anything else besides the current account deficits that you’ve implemented into your approach where you say you’re macro aware? Is there anything else to that process?

[00:22:50] Richard Lawrence: Yeah, there are. We had five, and then we added one, and then sort of in my Chinese way, I called them the five evils. And then the five evils plus one, but these are current account deficits, government deficits, fast loan growth. In my experience, banks can grow maybe 7 percent a year and not run into all kinds of trouble.

[00:23:08] Richard Lawrence: So, if they’re growing faster than that, they’re going to outgrow their ability to make good loans and know what they’re really lending. Loan to deposit ratios, forex reserves, those are all things that we’re very aware of. We track them religiously twice a year and we communicate them to our investors.

[00:23:24] Richard Lawrence: And in today’s world, we’re in another bear market in Asia, not as severe, but we don’t have a macroeconomic bear market. We have a geopolitical bear market, which is different. It has its severities. But Asia today is still running current account deficits, modest loan growth, very acceptable loan to deposit ratios, very acceptable government deficits, particularly compared to Europe or the U. S. So, it’s, Asia is still very, very competitive, I think…

…[00:46:49] Clay Finck: Since your team is looking at the macro situation or being macro aware and overlaying that on your very micro approach to picking stocks, how would you describe the current macro situation in China?

[00:47:02] Richard Lawrence: Well, let’s start off with the stuff that really matters, which is things like balance sheets. Okay, they got a 3 trillion of forex reserves. The household bank deposits are double the size of the market capitalization of the stock markets. And it nearly tripled the size of annual retail sales. So the individual Chinese consumer has a lot of firepower in their bank deposit.

[00:47:27] Richard Lawrence: Okay, so balance sheets are strong. Loan to deposit ratio is a conservative. The capital adequacy ratio at the banks is okay. So, you know, those balance sheet items are all in very good order. Current account, surplus, small government deficit. That’s not the problem. The problem is really a lack of confidence.

[00:47:45] Richard Lawrence: They’ve lost confidence. As you do in bear markets, as you do in recessions, you’ve lost confidence. And it was triggered by the declines in an overbuilt real estate market. The real estate guys had kind of a heads I win tails you lose kind of approach to real estate development, particularly the private guys.

[00:48:02] Richard Lawrence: They’ve all been gone bankrupt and all been flushed. But the residual is, is that real estate prices probably really have gone down 25 percent if you speak widely. You know, there are pockets where it’s stronger and pockets where it’s weaker. And that was the major asset of Chinese people. That’s what the citizens own.

[00:48:20] Richard Lawrence: They own some equities, but not a lot. And so they’re a bit shell shocked and they get sort of really mixed signals on capitalism from the government. And so their animal spirits have really been doubly repressed by lack of confidence and a concern over the commitment, both growth and capitalism in the country.

[00:48:40] Richard Lawrence: So that’s, that’s kind of where we are at the current time. And then you lever that on top with Geopolitical situation with the U. S. where both sides are at fault. Both sides have brought out the worst in each other and there’s a lot of sort of ganging on. It’s a bit like 10 year olds on a playground.

[00:48:59] Richard Lawrence: There’s kind of ganging up on each other. It’s not really great leadership for the world. This is probably the most important economic relationship in the world today. And the amount of discussion going on between governments is almost minimal. And we don’t have a big base in the United States of diplomats who are really well versed in China.

[00:49:21] Richard Lawrence: China for the last 40 years has not been the problem. And so the diplomats went to Afghanistan, went to Iraq, went to Syria, went to Ukraine and dealing with all those messes and largely sort of ignored China while China needed attention to address some of the fundamental problems. And so we don’t have.

[00:49:39] Richard Lawrence: The great outlook that we should have that we historically had starting with Kissinger on really creating a real relationship with China. And so it’s going to take the better part of the rest of this decade to turn that around. Now, having said that, my investments in China and in Asia are not predicated on US investors moving those stock prices up.

[00:49:58] Richard Lawrence: They’re just not going to come back. The sentiment towards Asia is so negative, but like I said, there’s plenty of, uh, gunpowder in banks and household bank deposits. And so I think that’s what will eventually turn it around, but we need more commitment to reform than we’ve had. There’ll probably be more rounds of stimulus.

[00:50:18] Richard Lawrence: You have to understand that the Chinese do stimulus differently. We don’t open the helicopter and throw the money out. They’re very tactical on how they stimulate, they’ll do tests, they’ll test it in a bunch of provinces, and if it’s successful, then they’ll roll it out. We’re seeing big reforms. They’re just offering refinance of all the mortgages, for example, because the interest rates have gone down.

[00:50:39] Richard Lawrence: Things like that will really help the Chinese citizen, and that’ll bring back animal spirits. It’s a long bear market. We’ve been three and a half years, almost three and three quarter years. And no upward momentum to speak of. So it’ll just take time. That’s the way life is sometimes.

[00:50:55] Clay Finck: China is certainly a very hot topic, both inside and outside the investing world.

[00:51:01] Clay Finck: Some like Overlook have been finding bargains within the Chinese market while others see China as uninvestable to some extent. What do you think is the biggest misconception when it comes to investing in China?

[00:51:14] Richard Lawrence: Well, if you think back to this eight year olds on the playground in the US, there’s a certain arrogance that China’s weak and has been brought to its knees and doesn’t have technology and is massively over levered and whatnot.

[00:51:28] Richard Lawrence: I think that’s not really realistic. If you look carefully at the semiconductor, which is something I’ve been tracking for nearly 24 years, we can try to restrict advanced semiconductors from China, but China takes a very long view of this stuff. And I guess in 8, 10 years, they’re going to have similar level of technology, and that will have happened faster than if we had really sat down and talked about what are the uses in China for the advanced technology, for the advanced chips, how to keep them out of the military.

3. Podcast: Eric Markowitz – Graham Rhodes and Eric Markowitz

I spoke with Eric Markowitz, Partner and Director of Research at Nightview Capital, about an essay he published this year titled, “How a brush with death shaped my long game” (link). We dive straight into the details of the health crisis which shook Eric’s world in early 2023 (link) and then have an open-ended conversation about what he’s taken from it as a husband, father, friend and investor…

…EM: Absolutely. I only survived because of a lot of other things—because I live in a country with access to good healthcare, because I have a wife who kept me alive, and because I had a great doctor and neurosurgeon. My brain surgeon was a friend of a friend, and the night they found the lesion in my brain, I called my best friend, Ben Jacobs, and said, “Take care of my wife and kids if things don’t work out.” He said, “Let’s call Zach.” Zach had just moved to Portland after finishing his neurosurgery residency at Stanford. He looked at my scan that evening and said, “I want to do your surgery.” So, I benefited from this social fabric that kept me alive.

These systems, like friendships, technology, and geography, all played a role. Without them, if I had been alone on an island, I wouldn’t have survived. There’s safety in numbers, but there’s also a danger in following the herd. Sometimes, you don’t want to be an outlier, but in other cases, you do…

…EM: Sure. “Tikkun Olam” is an ancient Jewish concept that essentially means making the world whole again. It’s about making the world a little bit better, and that can be interpreted broadly. It could mean doing charitable work, pursuing direct philanthropy, or finding other ways to contribute positively to the world…

…EM: Right. If your mission is to build something that can last 500 years, you should hope for crises. Crises create opportunities for reinvention within organizations. I’m starting to write about this for a column because it’s fascinating and not talked about enough. People often think crises are bad, but you should absolutely hope for one because that’s what will save you.

One of my favorite examples is a company from about 100 years ago called Kutol. They sold wallpaper cleaner, which was popular when homes were heated with coal and oil because walls would get dirty. But when homes started using cleaner energy like natural gas, demand for their product collapsed. The company was in turmoil.

The CEO’s sister-in-law, a schoolteacher, realized that kids liked playing with the wallpaper cleaner. They decided to repackage it as a children’s toy and, with nothing to lose, launched it as Play-Doh. Play-Doh became a phenomenal success, saving the company from bankruptcy. Kutol is still around today, over 120 years later. Had they not faced a crisis, I’m not sure they’d still exist. They were only willing to try something new because their backs were against the wall…

…EM: Yeah. I began my career as a business journalist and investigative reporter, and although I write a weekly newsletter, I hadn’t been doing much real writing. After I wrote the essay about my health crisis, I started writing a column for Big Think. It’s been fun, like an intellectual playground for me. It gives me something to focus on and a research project to dig into. I’m trying to make these columns really good, which forces me to cut out what doesn’t matter.

Writing is helping me think better, be a better investor, and find good companies. It’s helping me do better for my clients and my family because they’ll read these pieces one day and have their own insights. It’s compounding in a positive way, which I don’t think I was really doing before. Before, it was just about getting through the day—checking off calls, listening to earnings reports. Now, I’m more focused on doing one thing really well. 

4. Daniel Yergin – Oil Explains the Entire 20th Century – Dwarkesh Patel and Daniel Yergin

Daniel Yergin 00:06:43 

People think of John D. Rockefeller and Standard Oil and they go: gasoline. It had nothing to do with gasoline. John D. Rockefeller was a lighting merchant. What they did is that they rolled back the darkness with kerosene, with lighting. Before that, the number one source of lighting was candles and whaling. The whaling industry was delivering lighting. For the first 30 or 40 years of the oil industry it was a lighting business. Then came along this other guy named Thomas Edison. Suddenly you have electric lights and you say, “That’s going to be the end of the oil business.” But by the way, over here is Henry Ford and others. You’re creating this whole new market in the 20th century for gasoline. In the 19th century gasoline was a waste product. It went for like three cents a gallon.

Dwarkesh Patel 00:07:34

One of the things I learned from The Prize, I didn’t appreciate before. Before the car was invented, when Edison invented the light bulb, people were saying Standard Oil would go bankrupt because the light bulb was invented.

Daniel Yergin 00:07:47

John D. Rockefeller became the richest man in the United States as a merchant of lighting, not as a merchant of mobility.

Dwarkesh Patel 00:07:55

In some of the earlier chapters, you mention that Rockefeller was especially interested in controlling the refining business, not the land owning and drilling. A lot of the producer surplus went into refining. Why did the economics shape up such that the producer surplus went to refining?

Daniel Yergin 00:08:11

Because that was the control of the market. That was the access to the market. The producers needed John D. Rockefeller. There were a few other people but Rockefeller controlled about 90% of the business. He would either give you a good sweating—drive down prices and force you out of business—or force you to sell to him or amalgamate with him…

…Daniel Yergin 00:32:43

OPEC was setting prices, but then the market responds. Demand goes down. In fact, that’s exactly what OPEC did with its prices. It created incredible incentive to bring on new supplies and to be more efficient and undercut. It ended up undercutting its own price. Here’s one of the things I really carried away from The Prize. There are hundreds of really interesting characters in the book, but the two most important characters, one is named Supply and one is named Demand. That’s something that you’ve got to keep in mind with all the other drama that goes on.

Dwarkesh Patel 00:33:22

The interesting thing from the book is that oil did seem to be, at least until very recently, pretty different in that with other sorts of commodities you have strong elasticities of supply. If lithium gets more expensive, you’ll figure out substitutes for lithium and it’s not that big a deal.

Daniel Yergin 00:33:41

Or find more lithium.

Dwarkesh Patel 00:33:42

Yeah. Whereas, at least during the oil crises, it really felt like the entire world economy was just on hold…

…Dwarkesh Patel 00:59:51

How mad are the frackers that they basically solved America’s main geopolitical problem, but they were so successful that they’ve competed away their profits?

Daniel Yergin 01:00:03

That was a period up till about 2017, when it was growth for growth’s sake. Then basically the financial community said, “Hey guys, the party’s over. I’m not going to reward you for growth. I’m going to reward you for sending money back on my investment.” So in a sense, shale is almost a mature industry. I think people don’t understand how transformative it’s been. The US was the world’s largest importer of oil. We were only producing 5 million barrels a day of oil in 2008. Now we’re more than 13.2 million barrels a day. The US is energy independent. People thought it was a big joke. It could never be energy independent. Every president said, “We want energy independence.” Late night comedians could make fun of it. Actually, it’s happened and it’s had huge economic significance. Back in like 2008, the US was spending something like $400 billion a year to import oil. Now we basically spend nothing to import oil.

It’s been geopolitically very significant. That’s been a learning experience for the Biden administration. It turns out that if it wasn’t for shale gas made into what’s called LNG, liquefied natural gas, shipped to Europe, Putin could well have shattered the coalition supporting Ukraine by using the energy weapon with, not oil, but gas. Suddenly you had European politicians coming to the US to try and secure supplies of LNG because they were so worried about it. It really is a revolution that is playing out today. China imports 75% of its oil. It wishes it was in our position…

…Dwarkesh Patel 01:10:26

Let’s talk about solar and renewables. With oil, you have a commodity which is a flow. You can cut it off and you can turn it back on again. It gives the person who’s producing it a lot of leverage. Whereas with wind and solar, if you’re the people producing it, it’s just a capital stock. How does that change the geopolitical situation and the kind of leverage that the producer might have?

Daniel Yergin 01:10:51

It’s a question of scale. What I carried away, the basic premise of energy security goes back to Churchill. He said that safety lies in variety and variety alone, diversification. Wind and solar give you diversification. Electric vehicles diversify your fleet. Those are all there.

For China, wind and solar, electric cars, is very much a strategic issue because they see the vulnerability of importing 75% of their oil, much of it coming through the South China Sea. They know the story of what happened with World War II with Japan. For them, the shift to electric cars is less about air pollution and more about energy security. It’s also about knowing that they couldn’t compete in the global market with gasoline powered cars, but they can with electric cars. Those are the strategic things.

Wind and solar give you a more diversified system. Until you have batteries that can really deliver the storage, you have the intermittency problem. You take California today. People think wind and solar is advanced. It’s true. They are 25% of electric generation in California, but 43% of electric generation comes from natural gas. And that gets back to the data centers. You’re going to need to bolster your electricity power system. How much can you do with batteries and how much can you do with natural gas?

Wind and solar are also stories about entrepreneurship. In The Quest I asked myself, where did the wind and solar industries come from? The solar industry came from two émigrés who had left Europe, one of whom had driven his car out of Hungary in the 1956 revolution. In 1969, he’s a chemist working for the US government. He and his partner decided to go in the solar business. That became the first solar company. They started in 1973. With the wind business, I like to say the modern wind business is the result of the marriage between California tax credits and the sturdy Danish agricultural industry. It was driven by tax credits, but they needed to find wind power machines that could stand up when the wind blew in the Tehachapi Pass.

It took about 30 years for both those industries to become competitive. It only happened around 2010 that they actually became competitive. Now, of course, they’re very competitive but then guess what? Now,they’re all tied up. Renewables are also now tied up in geopolitics and, in what I call The New Map, the movement to the great power competition. The US just put 100% tariffs on Chinese electric cars, 25% tariffs on Chinese storage batteries. We recently had this bill, the Inflation Reduction Act. It’s huge, a trillion dollars the Treasury estimates when it’s done. It’s about climate and renewables, but it’s also about competing with China…

…I have the view that people have had too simple notions of how the energy transition will work.

That’s one of the things in The New Map. If people read one part of it, read the section on energy transition. It tells you that what we’re talking about today is not anything like any other energy transition. Every other energy transition we’ve had has been energy addition. Oil discovered in 1859 overtakes coal. Coal is the world’s number one energy source in the 1960s. Last year, the world used more coal than it’s ever used, three times as much as the 1960s. Now the idea is, can you change everything literally in 25 years?

Some of that thinking was developed during COVID, when demand went down and price collapsed. Part of it is people worrying about energy security. I was just reading last week the budget message from the finance minister in India. She talked about energy security and how they have to maintain economic growth. It’s very important to do that and energy security as well as energy transition. So it’s a different balance. There’s a difference between the North and South. Then there’s the constraints on minerals because as you make an energy transition, what people talk about, it’s more mineral intensive. An electric car uses two and a half times more copper than a conventional car.

We did the study and said, “Okay, let’s take the 2050 goals. And if you want to achieve them, copper supply has to double by about 2035.” What’s the chance of doing that? It takes 20 years to open a new mine in the US. We just did a study. It takes 29 years to open a new mine. Changing a $109 trillion world economy… it’s going to change. You said the development of solar is going to be really important. But things are not going to move in a straight line. We are in an energy transition, but it’s going to be a longer one. Here we are, as you mentioned, in Nantucket, which was a key part of the energy transition because it was a source of lighting in the 19th century from whaling.

5. The 27-Year-Old Billionaire Whose Army Does AI’s Dirty Work – Berber Jin

Sitting behind computers in cities across the world,  his startup Scale AI’s workers type out the stories, label the images, and craft the sentences that furnish chatbots with the text they need to better understand human speech patterns. Dubbed data labeling, their tasks range from composing haikus and summarizing news articles to writing stories in languages like Xhosa or Urdu.

The labor-intensive operation has become so in demand by businesses eager to enter the AI race that Scale’s revenue pace tripled last year, boosting its valuation to $14 billion…

…Meta’s code name is Flamingo—a stuffed version of which sat atop an employee’s desk on a recent visit to the startup’s headquarters. After Scale AI bungled a project last year for the tech giant, Wang declared a company emergency and launched an all-hands-on-deck effort to fix the job, called Flamingo Revival, according to former Scale employees.

Early last year, Meta Platforms asked the startup to create 27,000 question-and-answer pairs to help train its AI chatbots on Instagram and Facebook.

When Meta researchers received the data, they spotted something odd. Many answers sounded the same, or began with the phrase “as an AI language model…” It turns out the contractors had used ChatGPT to write-up their responses—a complete violation of Scale’s raison d’être.

The researchers communicated the disappointing results to Scale, prompting Wang to rally the entire company to try and save the contract.

He asked employees to drop everything and create new writing samples to send to Meta. An internal leaderboard showed who had completed the most labeling tasks. The prize for the winner: a paid vacation.

Later, Scale discovered that much of the bad data sent to Meta had come from Kenyans who had become experts in making a quick buck off the Remotasks platform, the former employees said. Scale restricted several new labeling projects to workers based in the U.S. and other wealthy, English-speaking countries. The change didn’t stem the fraud entirely: Some foreign workers found ways to skirt the new rules by buying labeling accounts registered to U.S. residents that they found for sale in group chats on WhatsApp and Facebook.

A Scale spokeswoman said that Scale has cracked down on such activity and the percentage of its freelancers exhibiting fraud fell to under 0.1% in July.

---

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 Meta Platforms, Starbucks, Tencent (parent of WeChat) and Tesla. 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 29 September 2024:

1. Digging Into The Coal Industry – Matt Franz

There are two main types of coal: thermal and metallurgical (“met”) or coking coal.

Met coal has more carbon, less ash, moisture, and sulfur than thermal coal. It is rarer and commands a higher price.

Met coal is a key ingredient in steel. To make steel, met coal is first turned into coke by heating it to 1,000ºC in the absence of oxygen. Without oxygen, the coal does not burn. The coal swells and then releases its gaseous volatile matter. Nearly pure crystalline carbon is all the remains. Coke is to coal what charcoal is to wood. Caking is a coal’s ability to be converted into coke. Thermal coal has no caking ability, which is why it is much cheaper.

Coke is mixed with iron ore, flux (e.g. limestone), and hot air in a blast furnace to create iron. Iron is put into a basic oxygen furnace where oxygen reduces the metal’s carbon content. It is further refined to remove impurities and alloys are added to make steel.

The quality of met coal influences the quality of the coke, iron, and steel produced. A blast furnace fed with higher quality coke will require less of it, lowering production costs. Steel makers have an economic incentive to pay more for higher quality met coal.

It takes 0.8-1.05 tons of met coal to produce one ton of steel. (1.3-1.5 tons of met coal make one ton of coke. 0.6-0.7 tons of coke make one ton of steel.) That’s a lot! 70% of the world’s steel is made this way…

… The major met coal exporting nations are the US, Canada, and Australia. The major importers are countries with large steel industries relative to their domestic met coal supplies – China, India, Japan, and South Korea.

The US usually exports ~70% of its met coal. Export contract prices are tied to international benchmark indices. Domestic contracts tend to specify a fixed price and a fixed volume for one year…

…Thermal coal has a lower calorific value (CV) and a lower cost than met coal. It is primarily used to generate electricity. It is also used to make cement. It takes 0.1-0.12 tons of coal to make one ton of cement.

Thermal coal has been in decline in Europe since the 1980s and in the US since the 2000s. It continues to grow in Asia. Worldwide, coal demand reached its highest level ever in 2022 and again in 2023…

…Today, coal remains the world’s largest energy source for electricity generation. Coal may be losing share as an energy source, but it continues to grow in absolute tons.

China is the world’s largest producer and consumer of coal. In the early 2000s coal produced 75-80% of its electricity. Today it’s more like 60-65%. Coal lost share but grew in absolute terms. Chinese electricity production rose eightfold and its coal consumption rose sevenfold…

…Coal’s peak share of U.S. energy occurred around 2007-2008 at 50% of electricity. Today it’s 20-22%. That’s still a meaningful amount. The decline in US coal was driven as much by fracking and its byproduct, cheap natural gas, as environmental considerations. Should US natural gas get expensive, we could see a shift back towards coal…

…Coal’s share of energy production is falling slower than the increase in total energy demand. Jevon’s Paradox describes this situation. Demand for energy is elastic. As energy costs decrease, demand for energy increases even more. On balance, energy demand increases, even as energy consumption becomes more efficient…

…In Energy and Civilization: A History (2017), Vaclav Smil explains that energy transitions often take more than a century. The transition from biomass (wood) to coal in Western Europe took 200 years. The transition from coal to oil began in the late 19th century, but oil didn’t overtake coal as America’s dominant energy source until the 1940s, approximately 50-60 years later. The transition is still far from complete, and that’s despite crude being more energy dense and easier to transport (via pipeline).

One of the factors affecting the speed of the transition is infrastructure. The transition from coal to oil in America was slowed by the need to replace steam engines with diesel engines. The modern analog is the cost and complexity of switching a power plant from coal to natural gas.

Price is another factor. If there’s a new fuel that is much cheaper than the legacy fuel, there’s an economic incentive to rebuild the infrastructure faster. A wide disparity between coal and natural gas prices that is expected to continue will drive more US coal plants to switch to gas. That’s less likely to happen in Asia, where gas is less plentiful and LNG infrastructure is more expensive.

Once again, this suggests that the last ton of coal will be very expensive, not very cheap. Thermal coal may be a sunset industry, but it is going to be a beautiful sunset.

2. Fed up with Fed Talk? Factchecking Central Banking Fairy Tales! – Aswath Damodaran

As I drove to the grocery story on Fed Cut Wednesday, I had the radio on, and in the news at the top of the hour, I was told that the Fed had just cut interest rates, and that consumers would soon see lower rates on their mortgages and businesses on their loans. That delusion is not restricted to newscasters, since it seems to be widely held among politicians, economists and even market watchers. The truth, though, is that the Fed sets only one interest rate, the Fed Funds rate, and that none of the rates that we face in our lives, either as consumers (on mortgages, credit cards or fixed deposits) or businesses (business loans and bonds),  are set by or even indexed to the Fed Funds Rate…

…While the Federal Open Market Committee controls the Fed Funds rate, there are a whole host of rates set by buyer and sellers in bond markets. These rates are dynamic and volatile, and you can see them play out in the movements of US treasury rates (with the 3-month and 10-year rates highlighted) and in corporate bond rates (with the Baa corporate bond rate shown)

There is a final set of rates, set by institutions, and sometimes indexed to market-set rates, and these are the rates that consumers are most likely to confront in their day-to-day lives. They include mortgage rates, set by lenders, credit card rates, specified by the credit card issuers, and fixed deposit rates on safety deposits at banks.  They are not as dynamic as market-set rates, but they change more often than the Fed Funds rate…

…To test whether changes in the Fed Funds rate are a precursor for shifts in market interest rates, I ran a simple (perhaps even simplistic) test. I looked at the 249 quarters that compose the 1962- 2024 time period, breaking down each quarter into whether the effective Fed Funds rate increased, decreased or remained unchanged during the quarter. I followed up by looking at the change in the 3-month and 10-year US treasury rates in the following quarter:

Looking at the key distributional metrics (the first quartile, the median, the third quartile), it seems undeniable that the “Fed as leader” hypothesis falls apart. In fact, in the quarters after the  Fed Funds rate increases, US treasury rates (short and long term) are more likely to decrease than increase, and the median change in rates is negative. In contrast, in the periods after the Fed Fund decreases, treasury rates are more likely to increase than decrease, and post small median increases…

…In the quarter after the Fed Funds rate increase, mortgage rates and fixed deposit rates are more likely to fall than rise, with the median change in the 15-year mortgage rate being -0.13% and the median change in the fixed deposit rate at -0.05%. In the quarter after the Fed Funds rate decreases, the mortgage rate does drop, but by less than it did during the Fed rate raising quarters. In short, those of us expecting our mortgage rates to decline in the next few months, just because the Fed lowered rates on Wednesday, are being set up for disappointment…

…How else can you explain why interest rates remained low for the last decades, other than the Fed? The answer is recognizing that market-set rates ultimately are composed of two elements: an expected inflation rate and an expected real interest rate, reflecting real economic growth…

…Interest rates were low in the last decade primarily because inflation stayed low (the lowest inflation decade in a century) and real growth was anemic. Interest rates rose in 2022, because inflation made a come back, and the Fed scrambled to catch up to markets, and most interesting, interest are down this year, because inflation is down and real growth has dropped…

…The Fed’s major signaling device remains the changes in the Fed Funds rate, and it is worth pondering what the signal the Fed is sending when it raises or lowers the Fed Funds rate. On the inflation front, an increase or decrease in the Fed Funds rate can be viewed as a signal that the Fed sees inflationary pressures picking up, with an increase, or declining, with a decrease. On the economic growth front, an increase or decrease in the Fed Funds rate, can be viewed as a signal that the Fed sees the economy growing too fast, with an increase, or slowing down too much, with a decrease…

…Viewed through this mix, you can see that there are two contrary reads of the Fed Funds rate cut of 50 basis points on Wednesdays. If you are an optimist, you could take the action to mean that the Fed is finally convinced that inflation has been vanquished, and that lower inflation is here to stay. If you are a pessimist, the fact that it was a fifty basis point decrease, rather than the expected twenty five basis points, can be construed as a sign that the Fed is seeing more worrying signs of an economic slowdown than have shown up in the public data on employment and growth…

…If you remove the Fed’s role in crisis, and focus on the effects of just its actions on the Fed Funds rate, the effect of the Fed on equity market becomes murkier…

…The S&P 500 did slightly better in quarters after the Fed Funds rate decreased than when the rate increased, but reserved its best performance for quarters after those where there was no change in the Fed Funds rate. At the risk of disagreeing with much of conventional wisdom, is it possible that the less activity there is on the part of the Fed, the better stocks do? I think so, and stock markets will be better served with fewer interviews and speeches from members of the FOMC and less political grandstanding (from senators, congresspeople and presidential candidates) on what the Federal Reserve should or should not do…

… The truth is that the Fed is acting in response to changes in markets rather than driving those actions, and it is thus more follower than leader. That said, there is the very real possibility that the Fed may start to believe its own hype, and that hubristic central bankers may decide that they set rates and drive stock markets, rather than the other way around…

…I believe that it is time for us to put the Fed delusion to rest. It has distracted us from talking about things that truly matter, which include growing government debt, inflation, growth and how globalization may be feeding into risk, and allowed us to believe that central bankers have the power to rescue us from whatever mistakes we may be making.

3. ‘There Are Real Issues in China Now,’ Ray Dalio Says (Transcript here) – Bloomberg Televsion

I think that there are real issues in China now, and they changed, really in the last four years, and that is that they need a restructuring. Individuals, 70% of their money was in real estate. Real estate has gone down. Stocks have gone down. Salaries have gone down. And so and as a result, they’re not spending and they’re concerned and they’re holding money in cash…

…At the same time, you have the government sector is a problem because most of the government spending – 83% of government spending – is spent by local governments. Those local governments got their money by selling land for real estate. Okay, there are no land sales and they borrowed a lot of money…

…It’s a situation that’s more challenging than Japan in 1990. It needs a restructuring in order to be able to do that. And then there’s also the question: Is the property ownership, is it respected? And Deng Xiaoping during his period said, “It’s glorious to be rich.” Is it still glorious to be rich?…

…Yes, there’s fantastic innovation in terms of technology, there’s nothing like it other than in the United States. Europe certainly isn’t a competitor in that. However, it’s very much government-directed. Can there still be entrepreneurship and that inventiveness? These are the big cosmic questions…

…I see investing in China as largely a very attractively-priced place that now has a lot of questions regarding the issues that I just referred to…

…There’s a small percentage of our portfolio which is in China, and we’ll stay in China, you know, through this process.

4. OpenAI’s New Model, How o1 Works, Scaling Inference – Ben Thompson

There are two important things happening: first, o1 is explicitly trained on how to solve problems, and second, o1 is designed to generate multiple problem-solving streams at inference time, choose the best one, and iterate through each step in the process when it realizes it made a mistake…

…There has been a lot of talk about the importance of scale in terms of LLM performance; for auto-regressive LLMs that has meant training scale. The more parameters you have, the larger the infrastructure you need, but the payoff is greater accuracy because the model is incorporating that much more information. That certainly still applies to o1, as the chart on the left indicates.

It’s the chart on the right that is the bigger deal: o1 gets more accurate the more time it spends on compute at inference time. This makes sense intuitively given what I laid out above: the more time spent on compute the more time o1 can spend spinning up multiple chains-of-thought, checking its answers, and iterating through different approaches and solutions.

It’s also a big departure from how we have thought about LLMs to date: one of the “benefits” of auto-regressive LLMs is that you’re only generating one answer in a serial manner. Yes, you can get that answer faster with beefier hardware, but that is another way of saying that the pay-off from more inference compute is getting the answer faster; the accuracy of the answer is a function of the underlying model, not the amount of compute brought to bear. Another way to think about it is that the more important question for inference is how much memory is available; the more memory there is, the larger the model, and therefore, the greater amount of accuracy.

In this o1 represents a new inference paradigm: yes, you need memory to load the model, but given the same model, answer quality does improve with more compute.

5. Learning from Richard Lawrence of Overlook – 14.3% CAGR for 30 years – Eugene Ng

The birth of Overlook’s Cap on Subscriptions originated when Richard Lawrence had lunch in 1992 in New York with Crosby Smith, a representative of the Dillon Family. Richard was asked why he would not just raise capital to generate fees like other investment managers. Crosby proposed that if Richard limited initial subscriptions into the fund at $30 million, the Dillon Family would invest $1 million. They shook hands, and Overlook had its first investor.

The Overlook Cap on Subscriptions was born in that spontaneous moment for Overlook. Richard Lawrence thought that the Cap on Subscriptions has proven to be the single most significant business decision in their 30-year history. In the early 1990s, Overlook decided to cap new subscriptions at 8-9% growth per year. This policy enabled the company to grow its AUM steadily…

…Effectively, the Cap on Subscriptions can smooth fund inflows, effectively lowering the cyclical volatility of AUM. One can then limit inflows of funds at the top of the market, have a ready queue of investors waiting to jump in during market declines, effectively making an investment fund more anti-fragile especially during market selloffs.

In addition, the Cap incentivizes investors to make a long-term commitment, which is aligned with a long-term investment horizon. Investors usually have to wait 6–12 months to gain access, so there are no short-term gains for investors trying to time the markets. The Cap effectively self-selects patient long-term investors.

What is the downside of having such a Cap? As you can imagine, such a Cap on Subscriptions is not for AUM/asset gatherers. The fund size grows much slower and takes longer to scale, and investment managers collect much lower fees…

…Time-weighted return (TWR): The TWR calculates the compound growth of a portfolio’s Net Asset Value on a per-share basis over a specified period of time. Fund managers most often disclose this number.

Capital-weighted return (CWR): CWR calculates the Internal Rate of Return (IRR) for an individual investor’s return and the return collectively earned by all investors in the fund. The CWR accounts for all cash flows into and out of the investor’s specific account and the fund since inception. Most fund managers do not report CWR, and CWRs typically underperform TWRs for most funds.

The Discount (the “Discount”) and the Premium: The discount is the difference between the TWR and the CWR for a specific fund. Discounts occur when CWRs are lower than and underperform TWRs. Peter Lynch was producing world-leading returns when he ran Magellan (high TWR), but the underlying investors performed far worse (low CWR).

Discounts typically happen for two reasons:

First, a fund manager can generate exceptional results as measured by TWRs at the fund’s inception when assets under management (AUM) are small. Then, the manager gets “discovered” and/or “promoted,” and an explosion of money enters the fund, to the great delight of the fund manager. However, with the larger asset base, the now-famous fund manager performs poorly, dragging down his TWR while crushing his CWR.

Second, CWRs are hurt when investments are poorly timed. Investors chase funds promoting hot themes, then bail out when markets turn down. This behavior inevitably decreases their CWRs. But even buying smartly and selling poorly, or buying poorly and selling smartly, can result in a Discount.

On average, the Discount increases when some of the following conditions prevail:

1. Funds experience fast growth of AUM: the Discount tends to increase as the absolute value of a fund increases.
2. Funds are invested in trendy asset classes.
3. Funds are exposed to excessive valuation risk.
4. Funds have excessive exposure to fund-of-funds’ investors. 
5. Funds are operated in higher volatility sectors…

…Richard realized that the elimination of Overlook’s Discount is overwhelmingly due to their legal Cap on Subscriptions. At first, he thought it was due to the success of their Investment Philosophy or the luck of their investors in timing their investments.

Overlook’s Investment Philosophy has helped them achieve outperformance of our TWR vs. the benchmark, but it did not impact CWR. The investors’ luck is not a factor either, as their investors have added funds consistently over time.

Instead, the answer lies exclusively in the legal Cap on Subscriptions because the Cap has allowed a limited amount of funds to enter Overlook steadily over the past 30 years. Control over the growth of AUM is the key to eliminating the Discount. The legal Cap on Subscriptions is the hero of Overlook’s story.

---

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 no vested interest in any company mentioned. 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 22 September 2024:

1. Mario Draghi outlines his plan to make Europe more competitive – Mario Draghi

Across different measures, a wide gap in GDP has opened up between the European Union and America. Europe’s households have paid the price in forgone living standards. On a per-person basis, real disposable income has grown almost twice as much in America as in the EU since 2000…

…Europe largely missed out on the digital revolution led by the internet and the productivity gains it brought: in fact, the productivity gap between the EU and America since 2000 is largely explained by the tech sector. The EU remains weak in the emerging technologies that will drive future growth. European companies specialise in mature technologies where the potential for breakthroughs is limited.

The problem is not that Europe lacks ideas or ambition. But innovation is blocked at the next stage: it is not translated into commercialisation, and innovative firms that want to scale up are hindered by inconsistent and restrictive regulations. Many European entrepreneurs prefer to seek financing from American venture capitalists and scale up in the American market…

…EU companies face electricity prices that are two to three times those in America. Natural-gas prices are four to five times higher. Over time, decarbonisation will help shift power generation towards secure, low-cost clean-energy sources. But fossil fuels will still set the energy price for most of the time for at least the remainder of this decade. Unless Europe better transfers the benefits of clean energy to end-users, energy prices will continue to dampen growth…

…As the era of geopolitical stability fades, the risk of rising insecurity becoming a threat to growth and freedom is increasing. Europe is particularly exposed. The EU relies on a handful of suppliers for critical raw materials and is heavily dependent on imports of digital technology.

2. When Chasing More Dividends Leaves You With Less – Jason Zweig

In July and August, as investors became more convinced interest rates will fall, exchange-traded funds specializing in dividend-paying stocks took in $4.5 billion in new money, estimates Ryan Issakainen, a strategist at First Trust, an ETF manager in Wheaton, Ill.

Although funds with big payouts sound safe, high income can lead to a poor outcome. You need to guard against needless tax bills, overexposure to narrow segments of the market and the chance of deep long-term losses…

…To see the potential downside of these funds, though, consider Global X SuperDividend, an ETF with $784 million in assets.

It yields nearly 11%.

That’s huge compared to the income returns of roughly 1.3% on the S&P 500, 2.1% on the Dow Jones Industrial Average and 5% on short-term U.S. Treasurys.

The SuperDividend fund’s supersized yield comes at a cost. Launched in June 2011 at $75, this week the shares traded around $22. That’s a 70% decline.

If you’d bought the ETF at its inception and held continuously through the end of August, you’d have lost 9%—after accounting for all those jumbo dividends along the way…

… A company that pays a steady stream of growing dividends is probably in robust financial health, but one that pays gigantic dividends is probably struggling and may be desperate to attract investors. Put a bunch of those into an ETF, and you get lots of income but even more risk…

…High-dividend funds often hold many more energy and financial stocks than broader portfolios do. That can raise risk.

In 2008, both First Trust’s Dow Jones Global Select Dividend and its Stoxx European Select Dividend had roughly 50% of their assets in financial stocks—right before the global financial crisis struck.

Over the 12 months ended March 31, 2009, as the MSCI World index lost 42.2% and European stocks overall sank 49.6%, First Trust’s Global Select fell 53.2% and European Select lost 63.9%—even after factoring in their dividends…

…Although a moderate dividend can be a sign of robust corporate health, a huge dividend can be a distress signal. A dividend four or five times greater than that of the overall market isn’t a green light; it’s a red flag.

3. Learning From Peter Keefe – John Garrett

The investment philosophy [at our new fund, Rockbridge Capital] is exactly the same: great businesses, great managers, bargain price. That remains unchanged.

The implementation has evolved over time. Great businesses, great managers, great price—it’s kind of like mom and apple pie. I mean, who’s opposed to it? It’s axiomatic that these things work, but I believe your approach to implementation should change over time…

…You don’t really know what makes a business great. You don’t really understand what contributes to compounding. You want a business with all the great characteristics—growth, rapid growth, sustainable growth—but you don’t know how to evaluate one business against another. You don’t know which businesses are mayflies and which are incredibly durable with multi-decade runways.

Learning how to discern and implement those three criteria does evolve over time. Another thing that evolves is the recognition that there are only a tiny number of businesses you will own over the course of a career that will compound and give you that 100-bagger effect or the 300-bagger effect—what Munger called the Lollapalooza effect. Those opportunities are incredibly rare.

But you spend your entire career looking for them. On day one, when you enter the business, you might think, ‘Well, maybe I’ll find it today,’ but you’re probably not going to find it today, tomorrow, or the day after. So what has evolved for me is the realization that when you find a compounder, don’t let it go…

…Every time I’ve trimmed a position and it involved a great business, it wound up being a huge mistake.

Now, we had this conundrum recently. We own a lot of Microsoft, which we bought back in the Balmer days. So it’s been in the portfolio over 10 years. We’ve made 10 times our money in the business, and it’s appreciated to have a very significant percentage of our portfolios.

Microsoft got a big bid recently because of the artificial intelligence stuff, and I don’t know enough about artificial intelligence to have a responsible opinion. But you can argue that there’s a trillion dollars’ worth of value in Microsoft attributable to AI. Do I trim the position? Well, based on the mistakes I’ve made in the past, no. But at the same time, is a 35 or 40 multiple sustainable for a company that’s already worth three trillion dollars? It’s hard to make that argument. And particularly when you’re managing both taxable and tax-exempt capital, you can make a pretty good argument that you should trim it. But again, that’s never worked out for me. So we are where we are.”…

…Every time we’ve had a business that’s compounded more than 10x—and we’ve had a couple that have compounded at 100x—there’s always been a leader and visionary who is a person of humility, thinking about their business in multi-decade timelines. Without exception, 10, 100, 200-baggers were always a person…

… They’re not thinking about an exit or the next thing; they’re thinking in 10, 20, 30-year time periods.

These people are artists. They’re focused on building something of great value—not just to accumulate wealth, but to create something valuable to society. To borrow from Tom Gayner, these are businesses that do something for people instead of to people. They are financially interested, but the finances are a means of keeping score rather than acquiring more things or a better jet. Those are the people I shy away from. The real artists see beauty in what they’re building and are focused on creating value for all stakeholders, especially the owners of the business.

When discussing people who want to serve all stakeholders, it’s not about rank-ordering which stakeholders to reward first. It’s about understanding that a business can do well for its employees, shareholders, and vendors. Munger talked about this all the time…

…People ask, ‘What makes you different?’ Well, it’s not my process. Everybody wants great businesses and great managers and to buy them at a bargain price. Nobody says they’re not a value investor or that they don’t like what Buffett does. So I think a major differentiator in this business is temperament. If I have an advantage, it’s that I don’t feel like I’m coming unglued when the world is coming unglued. I don’t know why that is; it’s just part of my makeup, but it’s an advantage because low prices are good for investors…

…The biggest compounder I’ve ever had in the investment business was American Tower. I was fortunate enough to figure out American Tower before it was even a public company. It was a footnote in the 10-K of a company called American Radio Systems. American Radio Systems was run by a brilliant, thoughtful capital allocator who fits into this liberal arts bucket I talked about earlier. Steve Dodge went to Yale and was an English major there.

Steve did cable transactions for one of the big New York banks. He got the idea that recurring revenue businesses or contractual revenue were great. So he moved into the cable business and then into the radio business. Around the time of the Telecom Act in the mid-1990s, digital networks for cell networks were beginning to roll out. Steve had people come to him and say, ‘We’d like to hang some of these digital antennas on your radio antennas.’ They also owned a portfolio of television broadcast antennas. They needed structures in suitable locations for these antennas.

That’s the genesis of American Tower, which was just a footnote. I remember calling Steve and asking about it. He basically hung up on me. I had a good relationship with him, so I knew I was onto something.

Long story short, American Tower was spun off and went to over $40 a share. Then came the dot-com bust. There had been a land rush in the tower business, and many companies had gotten levered up.

This was when I learned one of my early lessons about leverage, although it eventually helped me. American Tower dropped to under 80 cents a share from $44. Now that’s a drawdown.

I went up to Boston, where American Tower was headquartered. Chuck Akre was with me, and we met with Steve. He said, ‘I’ll tell you anything that I can legally tell you. I want you to know upfront that I don’t have much time. I have a business that needs my attention. It needs more attention than I can possibly give it because there’s only 24 hours in a day. I think that we can save this thing and I’m not sure that we can, but I also want to tell you, I am solely responsible. This is the worst thing that’s happened to me in my business career, but you’re looking at the guy who made the mistakes that got us in the pickle that we’re in.’

There was none of the usual excuses like ‘The dog ate my homework,’ or blaming the pandemic or the dot-com bust. Steve gave us none of that.

Steve figuratively raised his hand and said, ‘I messed it up, and I am sorry. I will do my best to get you and all the other shareholders out of this pickle.’

That kind of character in a moment of great crisis inspired me and others to make American Tower a more significant position, despite its distress.

We were convinced that the business wasn’t going to zero. It had one of the greatest business models in public companies’ history. A business where 100% of incremental revenue flows through to free cash flow and was growing by 20 to 30% a year. It was highly likely the business would be recapitalized. I can’t think of a financing environment where it wouldn’t be.

Steve’s character and willingness to accept responsibility were crucial in our decision to increase our position. It went up 300-fold from there.

4. Light-Based Chips Could Help Slake AI’s Ever-Growing Thirst for Energy – Amos Zeeberg

Recent results suggest that, for certain computational tasks fundamental to modern artificial intelligence, light-based “optical computers” may offer an advantage…

…In theory, light provides tantalizing potential benefits. For one, optical signals can carry more information than electrical ones—they have more bandwidth. Optical frequencies are also much higher than electrical ones, so optical systems can run more computing steps in less time and with less latency.

And then there’s the efficiency problem. In addition to the environmental and economic costs of relatively wasteful electronic chips, they also run so hot that only a tiny fraction of the transistors—the tiny switches at the heart of all computers—can be active at any moment. Optical computers could, in theory, run with more operations taking place simultaneously, churning through more data while using less energy…

…Seeing the potential advantages, researchers have long tried to use light for AI, a field with heavy computational needs. In the 1980s and 1990s, for instance, researchers used optical systems to build some of the earliest neural networks. Demetri Psaltis and two colleagues at the California Institute of Technology created a clever facial recognition system using one of these early optical neural networks (ONNs). They stored images of a subject—one of the researchers, in fact—as holograms in a photorefractive crystal. The researchers used the holograms to train an ONN, which could then recognize new images of the researcher and distinguish him from his colleagues.

But light also has shortcomings. Crucially, photons generally don’t interact with each other, so it’s hard for one input signal to control another signal, which is the essence of what ordinary transistors do. Transistors also work exceptionally well. They’re now laid down on coin-size chips by the billion, the products of decades of incremental improvements…

…The process of multiplying matrices, or arrays of numbers, undergirds a lot of heavy-duty computing. In neural networks, specifically, matrix multiplication is a fundamental step both in how networks are trained on old data and in how new data is processed in trained networks. And light just might be a better medium for matrix multiplication than electricity.

This approach to AI computation exploded in 2017, when a group led by Dirk Englund and Marin Soljačić of the Massachusetts Institute of Technology described how to make an optical neural network built on a silicon chip. The researchers encoded the various quantities they wanted to multiply into beams of light, then sent the beams through a series of components that altered the beam’s phase—the way its light waves oscillated—with each phase alteration representing a multiplication step. By repeatedly splitting the beams, changing their phase, and recombining them, they could make the light effectively carry out matrix multiplication. At the end of the chip, the researchers placed photo detectors that measured the light beams and revealed the result.

The researchers taught their experimental device to recognize spoken vowels, a common benchmark task for neural networks…

…Since that 2017 paper, the field has seen steady improvement, as various researchers have come up with new kinds of optical computers. Englund and several collaborators recently unveiled a new optical network they call HITOP, which combines multiple advances. Most importantly, it aims to scale up the computation throughput with time, space, and wavelength. Zaijun Chen, a former MIT postdoc now based at the University of Southern California, said this helps HITOP overcome one of the drawbacks of optical neural networks: It takes significant energy to transfer data from electronic components into optical ones, and vice versa. But by packing the information into three dimensions of light, Chen said, it shoves more data through the ONN faster and spreads the energy cost over many calculations. This drives down the cost per calculation. The researchers reported that HITOP could run machine-learning models 25,000 times larger than previous chip-based ONNs.

To be clear, the system is still far from matching its electronic predecessors; HITOP performs about 1 trillion operations per second, whereas sophisticated Nvidia chips can chug through 300 times as much data, said Chen, who hopes to scale up the technology to make it more competitive. But the optical chip’s efficiency is compelling. “The game here is that we lowered the energy cost 1,000 times,” Chen said…

…While optical computing has advanced quickly over the past several years, it’s still far from displacing the electronic chips that run neural networks outside of labs. Papers announce photonic systems that work better than electronic ones, but they generally run small models using old network designs and small workloads. And many of the reported figures about photonic supremacy don’t tell the whole story, said Bhavin Shastri of Queen’s University in Ontario. “It’s very hard to do an apples-to-apples comparison with electronics,” he said. “For instance, when they use lasers, they don’t really talk about the energy to power the lasers.”

Lab systems need to be scaled up before they can show competitive advantages. “How big do you have to make it to get a win?” McMahon asked. The answer: exceptionally big. That’s why no one can match a chip made by Nvidia, whose chips power many of the most advanced AI systems today. There is a huge list of engineering puzzles to figure out along the way—issues that the electronics side has solved over decades. “Electronics is starting with a big advantage,” said McMahon.

Some researchers think ONN-based AI systems will first find success in specialized applications where they provide unique advantages. Shastri said one promising use is in counteracting interference between different wireless transmissions, such as 5G cellular towers and the radar altimeters that help planes navigate. Early this year, Shastri and several colleagues created an ONN that can sort out different transmissions and pick out a signal of interest in real time and with a processing delay of under 15 picoseconds (15 trillionths of a second)—less than one-thousandth of the time an electronic system would take, while using less than 1/70 of the power.

5. Warren Buffett Case Study: Arbitrage – Dirtcheapstocks

By 1981, Arcata was the second largest printing services organization in the U.S. In addition, Arcata owned 77,500 acres of Northern California timberlands, which it used for timber harvesting, reforestation and milling.

Arcata was to be acquired by KKR. The stock was trading around $33/share at the time of the deal announcement. KKR’s $37 offer represented a reasonable premium over the current share price. But there was one other interesting bit of information.

“In 1978 the U.S. Government had taken title to 10,700 acres of Arcata timber, primarily old-growth redwood, to expand Redwood National Park. The government had paid $97.9 million, in several installments, for this acreage, a sum Arcata was contesting as grossly inadequate. The parties also disputed the interest rate that should apply to the period between the taking of the property and final payment for it. The enabling legislation stipulated 6% simple interest; Arcata argued for a much higher and compounded rate.” – Warren Buffett

“Buying a company with a highly speculated, large-sized claim in litigation creates a negotiating problem, whether the claim is on behalf of or against the company. To solve this problem, KKR offered $37.00 per Arcata share plus two-thirds of any additional amounts paid by the government for the redwood lands.” – Warren Buffett…

…“We started buying Arcata stock, then around $33.50, on September 30 and in eight weeks purchased about 400,000 shares, or 5% of the company. The initial announcement said that the $37.00 would be paid in January 1982. Therefore, if everything had gone perfectly, we would have achieved an annual rate of return of about 40% — not counting the redwood claim, which would have been frosting.” – Warren Buffett

“All did not go perfectly. In December it was announced that the closing would be delayed a bit. Nevertheless, a definitive agreement was signed on January 4. Encouraged, we raised our stake, buying at around $38.00 per share and increasing our holdings to 655,000 shares, or over 7% of the company. Our willingness to pay up – even though the closing had been postponed – reflected our leaning toward ‘a whole lot’ rather than ‘zero’ for the redwoods.” – Warren Buffett…

…“On March 12, KKR said its earlier deal wouldn’t work, first cutting its offer to $33.50, then two days later raising it to $35.00. On March 15, however, the directors turned this bid down and accepted another group’s offer of $37.50 plus one-half of any redwood recovery.” – Warren Buffett…

…“The trial judge appointed two commissions, one to look at the timber’s value, the other to consider the interest rate questions. In January 1987, the first commission said the redwoods were worth $275.7 million and the second commission recommended a compounded, blended rate of return working out to about 14%.” – Warren Buffett

“In August 1987 the judge upheld these conclusions, which meant a net amount of about $600 million would be due Arcata. The government then appealed. In 1988, though, before this appeal was heard, the claim was settled for $519 million. Consequently, we received an additional $29.48 per share, or about $19.3 million. We will get another $800,000 or so in 1989.” – Warren Buffett

The final result: 39% IRR…

…The greatest investor to ever live earns a 39% IRR in a low-risk arb deal. The most striking part of this case is not the return generated – but the lack of risk taken.

Arcata was a profitable, growing business. Take a look at its five-year history leading up to the deal.

Arcata had strong operating businesses that earned sufficient sums to cover its interest burden with plenty of comfort.

---

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 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 15 September 2024:

1. The ROI on Generative AI – Tanay Jaipuria

The poster child for this has been Klarna which leveraged AI to elevate their customer support. Their AI assistant has taken over the work of 700 employees, reducing resolution times from 11 minutes to just 2 minutes while maintaining high customer satisfaction levels…

...Microsoft casually dropped that they too are expecting to save hundreds of millions of dollars a year on call centers after adopting Generative AI.

“Dynamics with Gen AI built in is sort of really … the category that gets completely transformed with Gen AI, contact centers being a great example. We, ourselves, are on course to save hundreds of millions of dollars in our own Customer Support and Contact Center Operations. I think we can drive that value to our customers”…

…We’re also hearing examples of measurable, tangible benefits from enterprises, as Amazon shared about their software development assistant Q which has saved them over 4,500 developer years in a recent code migration task…

…”With Q’s code transformation capabilities, Amazon has migrated over 30,000 Java JDK applications in a few months, saving the company $260 million and 4,500 developer years compared to what it would have otherwise cost. That’s the game changer. And think about how this Q transformation capability might evolve to address other elusive but highly desired migrations.”…

…eBay launched a new AI-assisted selling flow, and are already seeing improvements in customer satisfaction as well as faster time to list and get value for Sellers…

…YUM Brands is enhancing customer experiences at Taco Bell by rolling out voice AI driven drive-through systems. This technology is not only improving customer satisfaction but also boosting team member productivity, and the results are so promising they are accelerating their roll-out timelines…

…Manulife is an example of a company already seeing large ROI in using AI to assist salespeople…

…”We’re using GenAI and machine learning models to make it really easy for agents to understand customer opportunities but also to generate these personalized communications at the click of a button to help them engage with more customers more often.

In our first 2 weeks live, about 68% of our agents had already used the new GenAI capabilities. And in July, we will be broadening that user base to about 2,000.

Based on our analysis in Singapore, we anticipate a 17% uplift and repurchase rates for our customer base, when this is fully rolled out to all of our agents.”…

…Rocket Mortgage is utilizing AI to automating the transcription of client calls and completing mortgage applications…

…”Now the Rocket Logic Assistant seamlessly generates over 300,000 detailed transcripts every week from outbound calls. It supports over 100 data points on mortgage applications saving our bankers from inputting tens of millions of data fields each week.”…

…Walmart has harnessed generative AI to enhance its product catalog, improving the quality of over 850 million pieces of data…

…”We’ve used multiple large language models to accurately create or improve over 850 million pieces of data in the catalog. Without the use of generative AI, this work would have required nearly 100x the current head count to complete in the same amount of time.”…

…Mastercard is leveraging the new advances in Generative AI to enhance fraud detection, achieving a 20% increase in accuracy. 

2. The Agent Era – Patrick O’Shaughnessy and Bret Taylor

Patrick

It’s such an interesting story because I think it becomes ultra relevant in today’s world. And you hear a lot about this, maybe the mythical 10x engineer, the 100x engineer, 1,000x engineer, the leverage available to one person with a growing tool kit.

And maybe that’s a great excuse to bridge the conversation into agents. I think everyone listening will have heard that term and maybe have thought about it a little bit, have gotten excited about the prospect of some sort of autonomous agent doing work on their behalf or their company’s behalf. But it would be great for you to ground us in your definition of what one of these things is, if this becomes a really critical part of the world of technology in the next year or two. I think it would be great for everyone just to have a level-set, simple definition from your perspective on what an agent is and does.

Bret

I’ll start with maybe the academic flavor of this, but then I’ll move into what I think is maybe the more — what I believe is the more relevant definition, but agent is like the word app. There’s not one definition, and I think it will be a noun that is quite meaningful in the age of AI. The word agent in the context of AI comes from the word agency and essentially is a system that can reason and take action autonomously is the way I think about it. And a system that is agentic is one where software and AI can reason and make decisions and take action without human intervention, which is really exciting but something that is relatively new though the idea is certainly not new.

I think the effectiveness of reasoning with AI systems has become so meaningfully better over the past couple of years that I think the concept is — like many parts of AI, the ideas are not new, but the effectiveness is, and so we’re living in an era of agents now.

In practice, I think the word agent, just like the word app or site in the age of the web, will become important to all of us. So one agent that I think is important is what my company Sierra does, which is your company’s conversational AI. And so just imagine you’re a retailer. I think you’ll put as much care and attention into your AI agent as you do your website or your mobile app. Or if you’re a bank, and you’ll put as much care and attention to your AI agent, which can help a customer look up the balance of their checking account or perhaps be an interface to your investment banking arm or wealth management arm. Or if you’re a streaming service, your agent might help people sign up for a plan or upgrade or downgrade their subscription, as an example.

In that case, an agent is something like website or mobile app that’s branded and it’s yours. And there are parts of it that are about agency and sort of the AI definition of the word. But more importantly, it’s your thing. It’s your digital asset. It becomes the digital manifestation of your brand.

And that’s what my company Sierra does. And we think that’s one really important part of an agent. Just like in 1995, the way you existed online was to have a website, we think in 2025, the way you will engage with your customers will be your AI agent, and we think it’s a really important new category.

But then taking, okay, what are the other types of agents out there? One will be, I’d like to think of them as persona-based agents. They’re internally facing. They do a job. You’ve talked about software engineering. I think there’ll be software engineering agents that will work to produce software. I was looking at a start-up called Harvey, I think, that’s making a legal LLM, which is super interesting. And I think across many job functions, there will be AI agents that produce the output of a — whether it’s a paralegal or a software engineer or an operations analyst, things like that. So that’s one.

So there’s your company’s agent, there’s a persona-based agent that does a job, and then the third one — category is probably personal agents. So this is the agent that will work on your behalf, whether it’s helping you plan a vacation or organize your calendar or perhaps triage your inbox and things like that. I think technically, they’re all similar, but my guess is they’re different enough in what job they accomplish for you that there’s — probably different companies will build those different categories of agent.

If you’re building a software to be a personal assistant agent, the breadth of systems you have to integrate with is infinite because different people use different calendars and different this and different that, and there’s lots of interesting investment into that. If you’re building a coding agent, it’s a much more narrow use case but very deep, and you’re probably evaluating it based on benchmarks of the effectiveness of the software produced and the robustness of the software it produces…

…Patrick

What do you think are the next most important unlocks for the power of these agents? You mentioned their access tools, access to the Internet. I’ve heard people talk about the ability to have some sort of stored memory about you, the customer or the specific customer or just memory in general that doesn’t just live inside of a context window that’s always re-fed in or something.

Are those the three things that we need to unlock the next tier of productivity out of agents? Are there other things that you and Sierra are focused on? I’d love to get down to the nitty-gritty capabilities and roadblocks that you’re thinking about and working on that might make these things as ubiquitous as you think they will be.

Bret

Yes. I’ll start with the vantage point of Sierra. We help companies build customer-facing AI agents. Today, if you’re setting up a new Sonos speaker, you can chat with an AI agent they’ve built on our platform to help you set it up. If you’re a SiriusXM subscriber, you can chat with Harmony, which is their AI agent they’ve built on our platform. And if you’re a WeightWatchers member, if you click on the 24/7 live coaching tab in their app, that’s an AI agent they’ve built on our platform.

One of the things that I think is a nuanced problem that is not strictly technical in nature is just the act of actually designing conversational customer experiences is a relatively new discipline. I remember in the early days of the Internet, most websites looked like DVD intro screens, like they’re very graphical, there’s four big buttons. It’s really interesting to go down the Wayback Machine and look at them.

And I would say it took a number of years to evolve into sort of the design idioms that we recognize with websites today. And now if you go to a retailer, they’ll have a hamburger menu on the top left, and the way you filter through items and these — they’re sort of emergent from people’s lived experiences, both designing and using websites.

And now you can talk to almost any web developer. And they’ll not only choose similar technologies to make a website, but even the design process and Photoshop or Figma to design a website, they’re sort of established practices, some of which are obvious and some of which are actually subtle, like why did this become the way these things are done, and it’s the cumulative experience we have building with them.

The difference between a website in a mobile app and an AI agent is both the breadth and non-determinism of AI agents. So if you have a menu on a website, you can control what links are there, and it’s essentially multiple choice, here’s the options available to you. If you have an AI agent with a free-form text box, people can type whatever they want into that. And so your concept of what your customer experience is defined by you, but it’s also defined by your customers, by what they write in there.

It reminds me — going back to my web analogies here, it reminds me of going from Yahoo Directory to Google Search. Rather than having a taxonomy of everything available, it’s just free form, and there’s a much longer tail of queries in Google than there was in Yahoo! because of the expressiveness of a search box versus a directory.

And I think that that’s one of the really interesting and, I think, exciting opportunities with conversational AI for customer experiences is it’s a really authentic way to actually hear from your customers what they want from you. And I think we’ve — and so it sort of stands to reason, your website was the rails on which your customers communicate with you. And this is a free form that I think it’s much more expressive. And we’ve had multiple customers learn things about their customers that they didn’t expect by providing this really free-form experience.

And then similarly, I think the other really interesting thing when I mentioned non-determinism is the word agent comes from agency, and it’s really how much creativity do you want to give your AI in interacting with your customers. I think if you start from a position of control, you can say, I want to put guardrails around everything, but then your conversational customer experience is somewhat robotic. You’ve essentially defined the multiple-choice options of your customers’ experience. If you give your agent too much agency, in the extreme case, it will hallucinate, but in the more practical case, it just might not protect your brand in the way that you want it to.

And I would say that design question is both a technology question, which obviously we’re quite invested in solving, and I’m really excited about some of the work we’ve done there, but there’s a deeper question here, too, that’s actually a philosophical branding and design question as well. And what we’re trying to do at Sierra is not necessarily predefining answers to those questions. I think every company and every brand will have a different perspective on what’s correct for their brand experience but provide a platform that’s powerful and expressive enough. Whatever your answers are personally to that question, you can build your agent on Sierra.

Patrick

It’s so interesting to think about the customer experience going to a website where I buy shoes or something. I think one of your first customers was flip-flops, and there was a funny story around that, but I’m going to buy a pair of sandals, let’s say, on a website. And rather than click around, I just describe what I want and I can imagine like another pane on the right just starts showing me stuff. And then maybe I check out through this same thing as well, and that’s a simple version of tooling or ability to take action.

I’m curious what the hardest parts for you have been to build. It’s quite technically daunting to even think about how to build something like this, let alone one that’s adjustable and tunable to my specific brand. So talk a little bit about how hard of a technical challenge this is for Sierra, like the degree of difficulty you’ve encountered relative to, say, your expectation.

Bret

Yes. It’s a really wonderful question. I think that generative AI broadly is a technology with which it’s very easy to make a demo and very hard to make an industrial-grade system. And I think that’s the area of technical challenge that we’re really trying to dive into. And I think it’s one thing to say that this system does the correct thing 90% of the time. And it’s really an inkblot test whether 90% is a really good number or a horrible number.

And it also depends on the process. And so if it’s a consumer application that was helping you with your homework, maybe 90% is decent. If it’s something operating revenue impacting part of your business or there’s a compliance concern, it’s absolutely unacceptable to be wrong 10% of the time.

And so a lot of the challenges that we’re facing are, we like to say that software systems are moving from rule-based to goals- and guardrails-based. And it’s a very different mental model for building software systems. Rule-based systems, if you think about just the software development life cycle that’s evolved over the past 20 years, it’s really about how you make more and more robust rule-based systems, how do you ensure that the same input produces the same output, that it’s reliable, that it’s stable, and there’s a lot of true innovation in the way we make software to make them more secure and robust.

Now if you have parts of your system that are built on large language models, those parts are really different than most of the software that we’ve built on in the past. Number one is they’re relatively slow compared — to generate a page view on a website takes nanoseconds at this point, might be slightly exaggerating, down to milliseconds, even with the fastest models, it’s quite slow in the way tokens are emitted.

Number two is it can be relatively expensive. And again, it really varies based on the number of parameters in the model. But again, the marginal cost of that page view is almost zero at this point. You don’t think about it. Your cost as a software platform is almost exclusively in your head count. With AI, you can see the margin pressure that a lot of companies face, particularly of their training models or even doing inference with high-parameter-count models.

Number three is they’re nondeterministic fundamentally, and you can tune certain models to more reliably have the same output for the same input. But by and large, it’s hard to reproduce behaviors on these systems. What gives them creativity also leads to non-determinism.

And so this combination of it, we’ve gone from cheap, deterministic, reliable systems to relatively slow, relatively expensive but very creative systems. And I think it violates a lot of the conventions that software engineers think about — have grown to think about when producing software, and it becomes almost a statistical problem rather than just a methodological problem.

And so that’s really what we’ve tried to solve. We shared on our website, but we have a process we call the agent development life cycle, which is the name comes from, say, in the software development life cycle, here’s what you should do with these agentic platforms. It’s also — we’ve developed a lot of unique technology to make these systems more robust with having one AI model supervise another AI model to layer different models on top of each other to produce statistically more robust results.

And then as importantly, we’ve developed ways that folks who aren’t experts in AI can express the behavior that they want in their agent. You shouldn’t have to be an AI expert to make an agent just like you shouldn’t have to have a PhD in computer science to make a website. I don’t think we’re there yet, but that’s really what we’re trying to solve.

And broadly speaking, I would say, on the spectrum of fundamental research institutions like OpenAI, we’re not that we’re applied. We’re really thinking about how do we engineer on top of these foundation and frontier models to produce robust or reliable agents for our customers.

Patrick

I love the title of this one Kevin Kelly book, What Technology Wants, and I’m curious what agents want. If I’m a customer, I’m a prospective customer, and I want to go work with Sierra to make the best possible version of a conversational agent for my customers to use, what can the companies provide that make the agent do the best job?

Bret

Yes, it’s a great question. I would say that there’s two types of knowledge that I think really produce a really robust agent. One is the factual knowledge of your company. This just grounds the agent so that it won’t just make something up.

There’s a pretty widely-used technique called retrieval augmented generation in AI right now that effectively means rather than relying on the knowledge encoded in the model to answer questions, you present the model with knowledge, maybe stored in a knowledge base or a database and say, “Hey, summarize the content from here. Don’t rely on the information you’ve been trained on.”

That has been an effective technique for two reasons. One is that it means that you don’t necessarily need to train or fine-tune a model to use it with proprietary data, which is a much cheaper deployment methodology. And it also can be effective at preventing hallucinations as well because you’re effectively — rather than relying on the AI to determine what it knows or doesn’t know, you present the AI with the knowledge that it’s allowed to know, a simple way of putting it.

And that’s factual knowledge. And I would say that’s necessary, but woefully incomplete because that would enable your AI agent to answer questions, but it wouldn’t necessarily enable it to orchestrate a complex process or take action on your customers’ behalf.

The other type of knowledge is procedural knowledge. We have a Sonos speaker. It stops working. What would the best Sonos engineer ask you and do to figure out whether it’s a problem with your hardware, a problem with your Sonos app or a problem with your Wi-Fi? Like what is the process by which you do that.

If you’re a subscription streaming service, what is the process of processing an upgrade or downgrade to your membership? Are there different offers available based on your membership level? Do you have a promotion running? What’s been the most effective technique to keep people, a subscriber for a long period of time?

This is all the stuff that if you are a person and expert in it, and so coming in with that knowledge of not only here’s the factual knowledge for our company, but here’s the processes that represent our greatest customer experience. What does the best salesperson do? What does the best customer service person do? What is — the most effective marketeer at your company, how do they describe your products? And that’s often there we work with our customers to improve when they deploy AI.

And then the third thing is just access to the underlying systems themselves. I think the AI agents shouldn’t just be about answering questions or having a conversation, they should actually be able to take action on your behalf, whether that’s a retailer processing a return or a subscription service, changing your level of membership or connecting to the telemetry system of a consumer electronics company. So we can say, “Hey, we know your device phoned home. You’re connected. We now figured out this other problem.”

Or even with something like SiriusXM sending a signal down from a satellite to refresh your radio if your radio stopped working. So three ingredients, factual knowledge, procedural knowledge and systems integrations, I think, are the three key ingredients. And then with the right methodology, your agent can do anything that a person could do on a computer, which is just an incredible opportunity for customer experiences.

3. Here’s What Happens When Credit Markets Go Dark – Joe Weisenthal, Tracy Alloway, Jared Ellias, and Elisabeth de Fontenay

Joe (11:53):

You spell out this evolution of the debt markets and the historical things you’re taught in law school about the dangers of single lenders. We’ve talked to people in the industry and they have their explanations for why this particular market has boomed. But from your research, what would you say are the drivers of this? Or when you talk to people, what problems does the private credit market solve for them?

Elisabeth (12:19):

The interesting thing about this is that there’s multiple stories going on at the same time. So one is that, this is just actually substituting for a lot of the activity that banks did because the banks, ever since the financial crisis, have been really constrained for a lot of reasons. One, they’ve primarily been constrained because of regulation, and sort of regulation designed to discourage them from making risky loans and from, you know, to have diversification in their portfolio, and so on. And just their evolving model of doing business, that they prefer to be sort of the middleman and get some fees rather than lend directly. [There are] all kinds of reasons why banks have retreated from particularly the lower middle market, but also all the way to the largest companies. A second story is just that there’s been too much bank regulation. So, I’m not going to take a position on whether that’s true or not, but that bank regulation is stifling the banks and they can’t really lend and so on.

A third story is one that we find really interesting and appealing, which is that, it may just be that it never really made all that much sense to fund loans using bank deposits. That essentially, you have a very short-term liability, which is customer deposits, and very long-term assets. So some of these loans, of course, are multi-year loans. And that’s just a fundamental mismatch that banks have always struggled with and that bank regulation has always struggled with. And this is a really nice, neat solution to that. And the reason it’s showing up now is that, thanks to sort of loosening of some of the securities laws and other things, it’s finally the case that you can get these investment funds that are big enough to actually take over the role of banks. And for them, the sort of positive side of private credit is that you now have a better match between the funding source, which is you have these big institutional investors putting capital into private credit funds that is locked in for a number of years, and you’re matching that really well against the loans that are also multi-year. So in some sense, it’s actually a better fit than banks for financing this type of loan… 

...Joe (20:43):

It sounds pretty good to me. Okay, so there’s less legal fees, less creditor on creditor violence, liability asset matching, the better user experience. So what’s the catch? I don’t see any problems.

Elisabeth (20:57):

One potential problem is, of course, these are, in some cases, absolutely massive loans. And so you do lose diversification benefits. These are very risky investments. I would say, the private credit structure has a partial solution to that problem, which is that, the investors themselves in a private credit fund oftentimes are so massive themselves that they really don’t lose diversification, which is to say, their portfolios are so large that they can make this enormous investment in one private credit fund because that’s a tiny piece of their portfolio. So that’s one downside of private credit. The other of course, is the absence of trading. So before. you had pretty good signals of what your position was worth. There were lots of syndicated loans that had pretty active trading and there were indices tracking all of this. The [Loan Syndications and Trading Association] LSTA provides lots of data on the loan market, and, of course, the bond market is public in terms of the pricing there. So exit is always going to be a concern in this market, and I don’t think this market really has been truly tested yet. So we’ll have to find out. But that illiquidity can be an issue depending on what kind of investor you are and what your expectation is for getting out of these things…

…Tracy (30:14):

Just to play devil’s advocate for a second, I think this is something you actually deal with in the paper, but one of the things you hear from people in the private credit industry is tha, ‘Oh, well, if you’re getting funding from a private entity, maybe a single lender or maybe a club of lenders but it’s a smaller group than you would have in the public market, maybe there’s greater potential for working out your issues if you get into trouble. So you can renegotiate your debt with a smaller group of creditors and maybe they know your business better than like a big fund that is buying pieces of all these different types of bonds and things like that.’ What’s your response to that argument? This idea that, well, private credit actually allows you to have more room for workouts or maybe even stave off bankruptcy for longer?

Jared (31:09):

So, I guess my answer is that, that all sounds great, but it’ll depend. And it’s hard to really understand which way any of these forces cut. The one thing that’s clear cut, that’s important is, we’re losing the claims trading markets. Like, that’s just going to look a lot different. Like, the active market and the claims of Chapter 11 debtors, when that debtor is a private credit funded firm. But, as to the question of, ‘Well, you know, aren’t these private credit lenders smarter, more versatile, more nimble, able to commit capital? And won’t that be good for companies?’ You know, at the end, it depends. So something you worry about is, well, maybe private credit lenders will have incentives, not to adjust their marks on their books and instead, just to do ‘amend and extend’s, and just keep loans going when the company really needed to liquidate or should have filed for bankruptcy sooner.

Think about how different the GM bankruptcy would’ve been had they filed for bankruptcy in like 2005 versus 2009 when their business had already eroded so much. So we think of that erosion as something that limits reorganization options. And it’s not necessarily obvious how private credit interacts with that. Because private credit lenders have their own incentives and maybe their incentives are to say, ‘Look, we make loans to sponsor backed companies and if the sponsor wants to continue, we’re going to keep doing that because we really want to participate in their next deals.’ Or they could say like, ‘Let’s pull the plug on these things earlier.’

So something that I’ve heard from lawyers working in this space is that when private credit lenders replace like your mid-market banks, like your Citizens and that kind of bank, when you have like a private credit lender with a $30 million loan that might have been done by a syndicate of two regional banks, the private credit lenders are much more aggressive and much more willing to pull the plug on the company and to own the asset then that bank might have been, but the world could look very different for larger companies where private credit lenders might be easier for companies to do workouts with. So it’s really hard to tell. But I’m certainly a bit skeptical of the idea that all of this is unidirectional and the private credit is just better in every way for everything. It’s different and there’ll be different pros and cons and we’ll learn more about them, and the law will adapt and hopefully deal with some of the ways in which the incentives of private credit lenders distort bankruptcy outcomes.

Tracy (33:28):

Since you mentioned GM, could you maybe talk about another specific example of a liquidation playing out a bit late, as you describe it? I’m still salty over the collapse of Red Lobster, which you mentioned in your paper. So could you talk a little bit about that one and what it tells us about private credit?

Jared (33:47):

Sure. So, something that has been the case over the past few years is you’ve had private equity owned restaurants and retailers that just ended up doing quick liquidations after stalling for a very long time. Red Lobster is really interesting. Red Lobster had been struggling for a little while and then Fortress Investment Group, which was its private credit lender, came in and took over the company and basically just owned the asset very quickly. And something that is so interesting about that is that, traditionally, other lenders would’ve been a lot more cautious about doing that, because other lenders are very cognizant of what we call ‘lender liability’ and this line of law that suggests that you shouldn’t, if you’re a lender, play too much of a role in business decisions of companies that you lend to.

And like, there’s an example of like a private credit lender just behaving in this really aggressive way, which is interesting. Like, again, it’s hard to tell exactly what’s going to happen, but certainly that example doesn’t fit well with the story of, well, you know, the private credit lender is just like the banker and you know, it’s your corner bank in 1925, who’s going to work with you on your farm. The answer is, maybe some of the time that’s the story, but other of the time, you’re dealing with a very sophisticated party who may have different incentives and be worried about different things than traditional bank lenders or investors in the broadly syndicated market.

4. Flash Crashes Are Getting Faster – Ben Carlson

In the spring of 1962, the stock market was already in the midst of a double-digit correction. Then on May 28, there was a flash crash, sending stocks down nearly 7% in a single day. It was the biggest one day sell-off since the Great Depression…

…It’s becoming clearer by the day that last Monday’s stock market swoon was also a flash crash. As of August 5, the S&P 500 was down more than 6% for the month. It’s now positive in August…

…Flash crashes happened in the 1920s, they happened in the 1960s and they happen today.

The biggest difference between now and then is the interconnected nature of the global markets. You have computer and algorithmic trading. Information flows at the speed of light. Every piece of economic data is parsed in real-time with a fine-tooth comb.

Overreactions can happen much faster now.

Just look at the biggest gap downs over the past 40+ years:

This chart shows the biggest difference between the opening price of the stock market and the prior day’s close. All of them have occurred this decade outside of the 1987 crash…

…We are likely to see more of these flash crashes in the future due to a combination of increased leverage in the system, globalized markets and computer trading.

The hard part for investors is that it’s now easier to lose control during these types of market events. You don’t have to call your broker on the phone to place a trade. You can change your entire portfolio on your phone with the push of a button.

Just because markets are getting faster does not mean your decisions must be made faster.

5. Gaining Currency – Rachel Cheung

In its effort to cement its role as an innovation powerhouse, China’s most ambitious technological debut was also its most controversial: The digital yuan was rolled out as the legal tender of choice for the Olympic games. Instead of cash or Visa (the corporate sponsor that had dominated the sports event for three decades), visitors were encouraged to exchange foreign currencies for digital yuan at automated teller machines and to pay digitally through the e-CNY app on their phones or through a card that can be used offline…

…Yet, despite all the attention, the launch of the digital yuan largely fell flat. The COVID-19 pandemic meant Olympic visitors were confined to “bubbles” with little opportunity to travel, shop and dine out, and very few foreigners chose to use the digital yuan over their credit cards. Beijing saw just $315,000 in digital yuan processed every day over the course of the games — a small fraction of the usual revenues at the Olympics. At the 2008 Olympics in Beijing, for instance, the city generated roughly $264 million per day…

…But while China acknowledged its Olympic failure, it has also quietly doubled down on the digital yuan, including a big push to drive adoption. Last year, several cities began paying civil servants and collecting taxes in digital yuan. Jiangsu province saw the most recorded transactions in the country after it gave away 30 million yuan ($4.18 million) in digital “red envelopes.” And this past May, the digital yuan expanded for the first time outside of mainland China when it became available for use in Hong Kong. Though there is no timeline for a nationwide launch yet, China has rolled out pilot schemes in 26 cities and 17 provinces since 2019.

The efforts have paid off. In a press briefing last week, the PBOC announced that total transactions reached $7 trillion yuan ($982 billion) in June — a four-fold jump since last June.

Digital yuan usage is still only a fraction of China’s $40-trillion payment market, of course. The total number of e-CNY wallets opened — 120 million as of last July — also trails behind that of Alipay, which had over a billion users by 2020 and recorded $118 trillion worth of transactions in one year alone.

But as Beijing continues to crackdown on its fintech giants, it is creating room for the digital yuan to rise. In fact, officials see the transition to digital currency as both necessary and inevitable. According to Yi Gang, former governor of the PBOC, the current moment of transition is not unlike that of the Ming Dynasty, when the government started taking tax payments in silver instead of labor and grains. China’s currency has evolved with time, he said during a speech at Fudan University in April, and “the digital yuan is no exception.”…

…Officials are also trying to expand the scope of e-CNY beyond consumer retail transactions. The Bank of China, for instance, has tested the use of “smart contracts” for afterschool programs in Chengdu of Sichuan province: Parents can pay a deposit in e-CNY to educational institutions, and the latter only receives the money after the lessons are taken.

These business-to-business and government programming applications could be a “game changer,” according to Warwick Powell, a senior fellow at Taihe Institute, a Beijing-based think tank, because they “ensure that the provision of certain funds can only be used for certain activities.”

Yet that same function triggers concern for others. For instance, although some local governments and banks have offered loans in e-CNY, companies are reluctant to take them, says Yang You, a finance professor at University of Hong Kong. “The nature of e-CNY is that a policymaker can generate a loan and see where it flows to,” says You. But companies, he notes, would much prefer non-traceable loans, despite repeated assurances from the People’s Bank of China that it will not hold information against them…

…Instead, the PBOC says the digital yuan follows a principle of “anonymity for small value and traceable for high value” as a way of striking a balance between privacy protection and combating criminal activities, such as tax evasion and money laundering. The e-CNY wallet, for instance, requires users to undergo a more complex verification process in order to unlock higher transaction limits…

… If anything, the search for an alternative to the U.S.-backed Swift, the global messaging network for the banking system, has gained momentum since the U.S.-led sanctions on Russia.

“China has used the sanctions as a reason to advance the cause of de-dollarization,” says Elizabeth Economy, a senior fellow at the Hoover Institution at Stanford University and recent advisor to the Department of Commerce. “It has made the case that the United States is weaponizing the dollar, hence other countries should begin to trade in their own currencies. It’s actually a deft diplomatic move on the part of China.”

According to the Bank of International Settlements (BIS), a survey of 86 central banks last year showed a sharp uptick in experiments with “wholesale CBDC” — transactions between banks and other financial institutions, rather than consumers and businesses. In October, for instance, the e-CNY set a new milestone: At the Shanghai Petroleum and Natural Gas Exchange, the state-owned PetroChina used digital yuan to purchase a million barrels of oil from an undisclosed seller.

“There’s still a conversation about the e-yuan [for domestic retail transactions], but there’s more discussion about a regional payment system,” says Victor Shih, an associate professor of political economy at the University of California. “An alternative to Swift potentially has more legs.”

The oil purchase seems to be a one-off so far, but a new project called mBridge hopes to make such transactions routine. It is a collaborative effort between the “innovation hub” of BIS and the central banks of five jurisdictions: China, Hong Kong, Thailand, United Arab Emirates, and most recently, Saudi Arabia. 

Underpinned by distributed ledger technology (which records transactions in multiple places at the same time), mBridge aims to be a multi-CBDC platform that can support instant cross-border payments. The idea is to make international settlement faster and cheaper than Swift. But it also means things are not dependent on the U.S. dollar.

---

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, Mastercard, 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 08 September 2024:

1. #360 Robert Kierlin: Founder of Fastenal – David Senra

What’s the most important part of Fastenal’s success that outsiders discovering the company for the first time don’t understand? The number one thing is the people aspect. The goal is to unleash entrepreneurial passion, a commitment that I will be self-driven to do better than what you can expect. It is a mindset.

This is what they’re telling their employees, “Run your business like you own it.” When you trust people to solve problems and make decisions and you let them go, that’s where the magic happens. That is the story of this company. Fastenal embraces a spirit of radical decentralization in autonomy.

“Each of its 2,700 stores operates as a stand-alone business with a clear leader and full P&L responsibility. We grow from the ground up based on the actions and decisions of thousands of people who run their businesses like they own it. I want those people to stay with us forever. They will never have to stop at a certain level.”…

…“We are now one giant organization. We are 2,700 small businesses wrapped up into one big company. Society tells us, you’re a big company, act like it. We say no, you don’t get to define us, we define ourselves. We’d go against the grain in almost everything that we do.” And so there’s just crazy stats about the company that, “More than 95% of our current batch of general managers have been promoted from within”.

“For nearly all of our senior leaders have worked their way up from entry-level positions.” So that leads into the second thing I want to tell you about, which is this interview with the CEO that had succeeded Kierlin or Bob — we call him Bob, when he stepped down. His name is Will Oberton.

Oberton started out at Fastenal as a store clerk. So he went all the way from — they mean business. He went all the way from store clerk to CEO…

…“By keeping operating costs very low, Fastenal is able to pay their employees incrementally higher wages, and thus, more effectively develop and retain talented salespeople. The quality of service and depth of knowledge that the employees have eventually brings in more revenue, which grows the business and allows it to further lower operating expenses as a percentage of revenue, thus allowing for more hiring of top-quality employees, which brings in more revenue. This is an overlooked virtuous circle of sorts.”…

…Why is it so important to have everybody working as a single cohesive team, to have everybody thinking that their role that they’re playing is just as important as the person next to them? “At Fastenal, we believe that you could be the best salesperson in the world. But if the order-picker doesn’t pick it right or the truck driver doesn’t get it there on time or the billing clerk doesn’t bill it correctly, you end up with an unhappy customer.”

“Everyone is key. You are better off working to make everyone equal so they stay focused” and he goes back, what do you think he’s about to say? I bet you can already finish his sentence for him. “You are better off working to make everybody equal so they stay focused on the common goal of pleasing the customer.”

[00:26:01] He’s going to give us more advice on how to do that. You need to install a reward system that keeps everyone focused on the common goal. He’s talking about incentives. If you have Fastenal’s common goal of growing our company through customer service, you will avoid any rewards that don’t fit that goal. And so when I got to this part of the book, I thought about Charlie Munger’s like three rules for incentives.

And so this is what he said, “Number one, everyone underestimates the power of incentives. Number two, never ever think about anything else before thinking about the power of incentives. And number three, which Bob is nailing, the most important rule in management, get the incentives right.”

And again, you have to be careful of these subgroups that are going to naturally develop in your company because his whole point is like “Listen, your incentives have to — they have to fit your overall common goal,” right, the common goal of pleasing the customer. And so he gives us an example, “If you do these incentives based on like separate groups, they can optimize for things that go against your common goal.”

So he gives an example that this is a really smart idea. “We do not reward production people for minimizing scrap. If some of that scrap you eliminate comes from the extra parts that guarantee you have a full order quantity ready when the customer wants it.” The incentive superpower that Munger talks about, you clearly see by picking up the book…

… I want to go back to that story of the CEO that was meeting with Buffett, the CEO that succeeded Kierlin. So this giant part of Fastenal’s business now after, this was invented after this book was written, okay, the first version in 1997, was the fact that they have these vending machines.

And the way I think about the vending machine is like think of anytime you’ve been in like a hardware store, right? You’ve got ACE Hardware or Home Depot or anywhere else. And think about how all the equipment and supplies are presented, kind of like searching through, it’s kind of like a chaotic mess.

So Will Oberton, which was the former CEO, but he’s no longer CEO now, but he’s the one that was CEO after Bob, okay? Oberton also developed an industrial vending machine system. There’s a video on YouTube that’s fascinating about this. It’s from Fastenal. Fastenal has their own YouTube channel. You can see the vending machine if you just type in Fastenal vending machine, if you’re interested in this, I thought it’s actually cool.

Oberton developed an industrial vending machine, and I searched for it after I read this because like I got to see what this looks like. Oberton had developed an industrial vending machine system, helping Bob realize a lifelong goal. In 1951, as a 12-year-old working in his father’s auto parts store, Bob was bothered by the fact that his dad had to send customers searching for nuts and bolts to someone else’s store.

He imagined that a vending machine installed at his father’s place might pop out fasteners like gumballs. Once on his own, he tried to convert a cigarette vending machine to this purpose. He couldn’t get it to work. So he started selling fasteners over the counter. Thus, Fastenal was born. 40 years later, working with a snack machine manufacturer and off-the-shelf software, Will Oberton got the job done.

Fastenal’s vending machines have been a big hit with customers. So their vending machines are actually installed in their customers’ locations. It cannot get simpler for this. You got to watch the video, I’m telling you. Oberton got the job done. Fastenal’s vending machines have been a hit with customers, generally helping them save 30% on supplies.

[00:46:05] The machines have cut down on theft and enabled automated reordering. That 4-year-old business, which I think now is like 15-years old, within 4-years old, this new idea already started contributing to 36% of the overall sales of Fastenal, I think it’s like over 40% now. 

2. A French Bank Like No Other in Europe Seeks to Export Its Model –  Phil Serafino and Albertina Torsoli

Bpifrance is a bank like no other in Europe.

The French lender has made more than €50 billion ($56 billion) in loans to small and mid-sized businesses and has €52 billion in stakes in almost 1,000 companies. It has backed everything from a startup wanting to take tourists to the edge of space in balloons and a chain of trendy Parisian nightspots to the automotive giant Stellantis NV. A force to reckon with on French deals for M&A advisers like Goldman Sachs Group Inc. and JPMorgan Chase & Co., it has lured away bankers from firms like UBS Group AG and Rothschild & Co…

…No other European country has an agency quite like Bpifrance: a for-profit, state-owned merchant bank with a mandate to foster national champions. Its wide-ranging lending activities are financed largely by borrowings guaranteed by its ultimate backer: the French taxpayer. And for all the political turmoil at the moment in France, its interventionist policies are likely to find favor no matter which coalition — from the left or the right — ends up forming a new government.

More than a decade after it was created under then-President François Hollande and his economic adviser — one Emmanuel Macron — Bpifrance exemplifies 21st-century French capitalism: Entrepreneurs build businesses with cash, nudges and nurturing from the state, which in turn wants them to create jobs at home and develop innovative technologies. Explicit in the deal: The government will fend off foreign interlopers if necessary…

…Bpifrance’s investment prowess and risk management haven’t really been tested because Dufourcq hasn’t faced a prolonged economic downturn, enjoying a favorable wind at his back almost from the start — even during the pandemic, when the French state opened the cash taps to prevent businesses from going under.

Its stock-picking bets also haven’t always paid off. A stake in train-car maker Alstom SA, for example, has lost about a third of its value since the investment early last year. Shares of Stellantis, in which the bank has a 6.4% holding, have slumped about 45% from their peak in March as the carmaker struggles to fix problems at its US and European operations.

Also, for much of the bank’s existence, it could finance itself at rock-bottom interest rates, something that’s no longer the case. A slowing economy and higher rates also may start to hurt companies that borrow from the bank: Bpifrance’s loans classified as doubtful stood at 4.7% at the end of 2022, up from less than 4% in recent years, according to the bank’s annual reports. It didn’t disclose the statistic in its 2023 report.

Dufourcq shrugs off such concerns, noting that the three decades-old agencies that combined to form Bpifrance survived some deep financial crises, and says his bank often says no to risky investment proposals.

While some European countries have national development banks, Bpifrance is unusual for the breadth of its offerings. It operates 50 offices around France, often sending representatives door to door to drum up business. In addition to debt and equity investments, it offers financing and credit insurance to exporters and training and consulting services to entrepreneurs — including on how to shrink their carbon footprint.

3. How Richmond Fed President Tom Barkin Sees The Economy Right Now – Cale Brooks, Tracy Alloway, Joe Weisenthal, and Tom Barkin

I talked to someone from Germany yesterday — this is going to make me interested but not your [audience]. Our savings rate went up at the beginning of Covid to about 15 or 16%. Same thing happened in Germany. Our savings rate has come down to about three and a half. Theirs is still at 17. So, why are German consumers not spending the way American consumers are? That’s an interesting topic. It’s something we spent some time on. It’s the kind of thing we spent some time on…

Tracy (08:01):

What’s the theory?

Tom (08:04):

Well, so the thing that really makes it crazy interesting is, there’s a whole social safety net in Europe that doesn’t exist here. And so, most of the time you think people are saving for retirement, they’re saving for a rainy day, or they’re saving because they’re worried about losing their job. Well, in Germany, they kept everyone’s job during the pandemic and you’ve got a pension. So why are they saving? And I think the best explanation I’ve gotten, it’s actually something on my list to study going forward, is there’s just a lot more precautionary feeling about the situation in Europe, the risk versus the Ukraine, and what’s happening over there. And it’s just a culture that maybe has just gotten a lot more cautious due to geopolitics if nothing else.

Joe (08:43):

That does sound really interesting. By and large, I mean obviously, the situation in the Argentina economy is radically different than it is here in the US. Germany is probably still, all things considered, similar cyclically to the us. Does it feel like, by and large, at least among developed countries’ central bankers, that there is a strong set of common mysteries perhaps? Or are they really like, everyone’s sort of seeing different things in their own country? I mean I’m sure it’s a mix of both, but how much of a global factor is there?

Tom (09:16):

Much more in common than different. The whole practice of central banking has been, I’d say, globalized over the years. And central bankers really do think about inflation targeting, for example, in the same ways. And there are banks, like New Zealand and Australia, that, back in 2000 or even before that, set inflation targets before the rest of us and we learned from them. And so there’s a lot of learning, there’s a lot of discussion. I think there’s very much a common framework. Now, the economies are very different.

I mean, the US economy has come through this unbelievably well, the European economies have not. And so we have a much stronger economy. So much of our economy is services, so much is supplied to ourselves. A lot of this deglobalization is felt much more on the European side. The challenges in China right now are felt much more on the European side. And then emerging market countries, they really just are worried we’re going to increase rates further and they’re going to end up offside. And so, they’re very dependent on our strength of our dollar and the weakness of our dollar…

…Tom (11:43):

I think the economy, since we were together three or four months ago, the economy’s moved in a very different way. First of all, on the inflation side, I might’ve even said four or five months ago I was looking for inflation to sustain and broaden. So, it’s sustained. We’ve got very low readings for four months in a row. And it’s now across the basket, whereas six months ago [or] eight months ago, it was really just in goods. And so the concern about inflation, reaccelerating has definitely come down significantly. At the same time, the labor market stats have also softened. And so, the phrase I’ve been using is, ‘people aren’t hiring but they’re not firing,’ and that’s just not a high likely sustainable outcome. Either demand will continue and people will start hiring again or you’ll start to see layoffs. And so I think there’s more concern on the labor market and less concern on inflation relatively…

…Tom (13:00):

So consumers, you hear a lot of talk about people saying that consumers [are] weak and people are running out of savings. That’s not what I’m hearing. What I’m hearing is consumers are still spending but they’re choosing. And, the way I think about it is, they now have the time when they go into a store and they see something that’s at a price they don’t like to say, ‘I think I’m going to do something else.’ And so if you look at Walmart’s results, they would talk about people trading down. If you look at Target’s results, they talked about the kind of reaction they’re getting to lower prices. McDonald’s results in the $5 value meal. I’ve talked to hotel chains that every room is booked, but they can’t raise price at all because the second they raise price, people just won’t buy it and won’t book it. I talked to a fast food leader who’s rolling out software actually to encourage their franchisees not to raise prices anymore…

…Tracy (15:15):

What’s the urgency, then, on supporting the labor market? And there’s obviously a debate going on right now about how fast deterioration in that market actually happens. We had Claudia Sahm on the podcast recently and she was talking about, ‘maybe it’s different this time,’ but how are you thinking about the pace or the rate of change in the labor market?

Tom (15:37):

So the other thing that’s happening in the labor market is a lot more supply of labor, and part of that is participation, prime age participation hitting 2025 year highs, and immigration, which is up significantly. And so the last jobs report where unemployment went up from 4.1 to 4.3, you actually added jobs, 114,000 jobs. We just added 420,000 people to the workforce. So the denominator got bigger. And so, you know, there’s some people who look at the unemployment rate and say, ‘Oh my gosh, the labor market’s about to fall off a cliff.’ That’s not how I see it. I see a loosening labor market being driven by a lot more supply. Now, what’s the urgency? We’re not in a situation, I don’t believe, where there is this big cliff there, but when we make policy, you’re trying to make it for a year from now, right? Because [of] the lags of monetary policy, you’re trying to meet a year from now.

And so you’ve got a labor market which is slowly cooled and you’ve got inflation which is now gradually cooled. And so, you sort of say, ‘Well, which do I worry most more about?’ And it’s been very clear for the last two and a half years that all you worry about is inflation. And now those are much more balanced…

…Tom (21:57):

Well, I see inflation upside risk in two places. First is, we’re at 2.5% for the last 12 months. Our target’s 2%. So while we’re doing great at bringing it down from when it was once 7.1%, core is still at 2.5%. And even the most optimistic forecast for the back half of this year don’t believe it’ll get to 2% because the numbers were so good on a 12 month basis…

Joe (22:20):

We’re talking we’re about on a year to year basis as opposed to like a three month though sequential, yeah okay.

Tom (22:22):

On a 12 month basis. Because the last half of last year was also very good. And so, we’re at least six months away, even with really good inflation data, from the inflation numbers hitting 2%. And if the numbers are just pretty good, not really good, there’s a risk that we plateau at some level over 2%. That’s one risk. The other risk is I do see medium term inflation pressures that are out there. We have a conflict in the Middle East that could spiral. Deglobalization is a very real risk and that means that the imports of goods could be more expensive going forward, or if we even reshore, more expensive. Housing’s a place where, if rates artists start coming down, one of the things I worry about is that will spool up demand for people who’ve been waiting to buy a house till mortgage rates come down, but there won’t be any new houses built. I mean that effect is two years, three years out.

And so what happens if you have more demand for houses with the same kind of supply? Or even if more houses come on the market, everyone who puts their house in the market is a buyer and a seller. So you’d still have this excess of demand over supply. So those things are potential inflationary risks. Now, good policy works against that, and if we do the right thing with rates we’ll work against, but that’s why I just want to make sure I understand it and see it before I declare victory.

Tracy (23:37):

What’s been the most surprising thing that you’ve heard at Jackson Hole this year? You talked about German savings rate, but beyond that, is there anything that caught your eye or your interest?

Tom (23:48):

Alan Blinder asked a question today that I thought was pretty interesting. He said, ‘When you think about monetary policy lags, why aren’t you talking about how to shorten them?’ And I’ve said, almost as it’s a given, that when we raise or lower rates, it takes 12 to 18 months for the full effect to go into the economy. Well, part of that is because the economy doesn’t behave in a way that would allow it to happen quicker. An example: I think the number is, in 2009, 60% of [the] mortgages in this country were adjustable rate. Today it’s 8%. And so when we raise or lower rates, it doesn’t flow through to mortgages quickly and certainly not even like it did 15 years ago. And I’m not saying we should change the mortgage market, but it does make you stop and think, how much of our policy, the effectiveness of our policy tools, is a given or how much could actually change over time as the economy changed?

4. No Priors Ep. 78 | With AWS CEO Matt Garman (Transcript here) – Sarah Guo, Elad Gil, and Matt Garman

…Now that we’re at a $100 billion run rate, I think 85% of workloads are still running on-prem today by most estimations, somewhere in that range. Pick your number, whether it’s 80 to 90, or whatever it is. That’s enormous. If there’s still 10x growth of just existing workloads – forget all the new genAI workloads that are being created every day – these are just existing workloads to move, there’s a 10x number in there, so that business is massive…

…Gil (12:40): You mentioned that 80% of workloads still haven’t migrated over. What do you think are the main blockers to that today? is it just momentum? Are there specific features? Are there big things still to build?

Garman (12: 48): There’s some technologies that I think… Look, if I had an easy button, and by the way we’re trying to build an easy button, but if I had an easy button that would just migrate mainframes to a modern cloud architecture today, almost everyone will push that button. But it doesn’t quite exist today and it’s not as simple as like, “Great, I’ll go run your mainframe in the cloud.” That’s not what customers want. They want to actually modernize those workloads and have them into microservices and containerized workloads and other things like that. So that’s one, is there’s just a bunch of workloads like that that are old and and their customer’s running a big SAP thing and they want to move it to the cloud but it just takes time because it’s tied to a bunch of other things like that. There’s also a bunch of workloads that as you get out of core IT workloads that are in line of business, that are the next set of things. Whether that’s say telco workloads that are running the 5G infrastructure around the world, we’ve slowly been moving those to the cloud and helping those customers get that flexibility and that agility of of running those in the cloud as well. But they’re slower to move.

If you think about all the compute that runs factories out there today on factory floors, most of those have not been modernized. And there’s a huge opportunity, by the way for AI, to totally revolutionize how you think about factory workflows and efficiency there. But a lot of that hasn’t moved. There’s on-prem infrastructure that people are still amortising, there’s still people whose jobs it is to run on-prem data centers, and so they’re resistant to moving things. There’s a bunch of factors in there and so some of it is just takes time, some of it is technology pieces, some of that is we still have stuff to go build and innovate and help make it easier for customers to do that.

Guo (14:37): I’d love to hear about just the initial investigation of generative AI as a technology change and how AWS began to react to it, invest in it, because to some degree it puts us all back in the on-prem co-lo era of the world, where to get one of these, if you’re doing any sort of real pre-training, to get your startup off the ground, you’re back to, “I’ll buy a bunch of DGX boxes somewhere and I need to think about the cost and management of that.”

Garman (15:07): Actually most people are still buying those but in the cloud. But it’s not a serverless type of thing. Most people are still not buying H100s and hosting them in a co-lo or anything like that. And increasingly, I think that’s going to get harder and harder as you move to liquid cooling and larger clusters. It is a super interesting space. I think we’ve been working on this space for how many years now – we’ve been investing in AI broadly for the last 10 years, and it’s why we started five or six years ago investing at the infrastructure layer and building our own processors, because we knew this was coming, we saw this path coming and we knew that that’s also not a short-term investment. It’s one of those things you got to invest way ahead. And then we were investing and building generative AI models, and then OpenAI made a generational leap forward with what they were able to do, what was possible, and many people have talked about this. But it really in some ways was a discovery as much as anything about just what was possible and unleashed the new set of capabilities.

So we actually as a business took half a step back and said, “These are going to be transformational abilities and assuming that this technology gets better and better and better over time, how do we make it so that every company out there can go build using those technologies?” Different than, “How can I go build a consumer application that people are going to be interested in?”, we took it from the point of view of AWS, “Just what are the building blocks that I can help all of our customers, whether they’re startups, whether they’re enterprises etc, go build interesting generative AI applications.” We started from first principles. Customers are going to care a ton about security. That’s not going to change. They’re not going to all of a sudden not care about securing their infrastructure.

We also had two more hypotheses. One, the idea that there wasn’t just going to be one model. We thought that there was going to be a lot of models for a lot of different purposes, and there’d be big models and small models, and people would want to combine them in new and interesting ways. I think the last two years have probably played that out. I think when OpenAI first launched, it wasn’t as obvious, but that was one of the bets that we made. The third one is that we view that every enterprise that was building on us, the interesting IP that they were going to bring to the table was mostly going to be their data, and they were going to care that their data didn’t leak back into a model or escape from their environment. So we built a bunch of what we did starting from those principles of how do we make sure that these things are secure, that their data is secure, that they can have access to every piece of technology that the customers need to go build interesting applications, and they can do it in a cost effective way. That’s how we approach the space.

I think we now have a platform in Bedrock, in Trainium chips and Inferentia chips, and then a bunch of the other capabilities around as well as the suite of models that we offer, both proprietary as well as open source ones – or open weights ones. I think we’re starting to see that momentum pick up and we’re seeing more and more customers really like that story. They like that platform to build from, and we’re seeing enterprises really lean in and want to build in that space because it gives them a lot of that control that they want as they go and build applications…

…Gil (26:25): The other place that a lot of people are spending time right now in terms of bottlenecks to utilization or usage or future-proofing, is actually more on the chip side or semiconductor or system side and in terms of DC capacity. Obviously you all have been building Trainium chips and other things which I think is really exciting to see that evolution. How do you think about future GPU shortages? Does that go away, when? I’m sort of curious about how you think about forward-looking capacity, and is the industry actually ready in terms of building out data centers, building out semiconductors, all the rest of it, packaging.

Garman (26:56): I think we’re probably going to be in a constrained world for the next little bit of time. Some of these things, they take time. Look how long it takes to build a semiconductor fab. It’s not a short lead time and that’s several years and TSMC is running fast to try to ramp up capacity, but it’s not just them. It’s the memory providers and frankly data centers that we’re building. There’s a lot of pieces in that value chain that I think as you look at the demand for AI which has been – exponential might be undershooting it – some of those components that support that I think are catching up and I think AWS is well positioned to try to do that better than others are.

We’ve spent a long time thinking about – in the last 18 years, learning how do we think about smart investing, how do we think about capital allocation. We’ve spent a bunch of time thinking about how do we acquire our own power, how do we ensure that it’s green and carbon neutral power, all super important things. We’re the largest purchaser of renewable energy over the last… new contracts, so actually going out and adding and supporting new renewable energy projects. We’re the largest provider I think, each of the last four or five years. So we’ve been leaning into that for a while to ramp up this and this is just a step up. So we’re thinking about how are we acquiring enough power. Our own chips is a way to support the growth of Nvidia chips, and so I think the more diversity there, the better off we are. We’re a huge partner of Nvidia’s. Nvidia actually runs their AI training clusters in AWS because we actually have the most stable infrastructure of anyone else, so they actually get the best performance from us. We love that partnership and we have a great and growing relationship with them. We think things like Trainum are a good diversification and I think there will be some workloads that run better on Trainium and are cheaper on Trainium over time, and as well as Inferentia.

I think inference is one of those workloads that – today it’s 50/50 maybe of training and inference. But in order for the math to work out, inference workloads have to dominate, otherwise all this investment in these big models isn’t really going to pay off, so hopefully for the industry that all happens. But I think we’re probably going to be tight for the next little bit of time, because the demand is almost infinite. I mean it seems infinite right now.

5. Timing the Stock Market Using Valuations – Ben Carlson

I’ve never found a legitimate way to utilize valuations to determine entry or exit points in the stock market. Maybe when things get to extremes but even then valuations can be unreliable.

In early 2017, I wrote a piece for Bloomberg about stock market valuations:…

...This was the lede:

Something happened in the stock market this week that has only occurred twice since 1871: Robert Shiller’s favorite valuation method for the S&P 500, the cyclically adjusted price-to-earnings ratio, reached 30. So, is it time to worry?

The only other times in history when the CAPE ratio reached 30 were in 1929 and 2000, right before massive market crashes. So it made sense that some investors were worried about the stock market being overvalued.

The S&P 500 is up nearly 170% since then, good enough for annual gains of roughly 14% per year.

Sometimes valuations matter, but other times, the market doesn’t care about your price-to-earnings ratios.

The same is true during bear markets. Sometimes stocks get downright cheap but not all the time…

…Three of the four bear markets this century didn’t see the CAPE ratio come close to previous bear market valuation levels. If your plan was to get more aggressive when the market got cheap enough, you would still be waiting.

The problem with using valuations as a timing indicator is that even if they do work on average, missing out on just one bull market can be devastating. You could be waiting a mighty long time to get back into the stock market and miss out on big gains in the meantime.

---

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). Holdings are subject to change at any time.