What We’re Reading (Week Ending 05 March 2023) - 05 Mar 2023
Reading helps us learn about the world and it is a really important aspect of investing. The legendary Charlie Munger even goes so far as to say that “I don’t think you can get to be a really good investor over a broad range without doing a massive amount of reading.” We (the co-founders of Compounder Fund) read widely across a range of topics, including investing, business, technology, and the world in general. We want to regularly share the best articles we’ve come across recently. Here they are (for the week ending 05 March 2023):
1. Planning for AGI and beyond – Sam Altman
If AGI is successfully created, this technology could help us elevate humanity by increasing abundance, turbocharging the global economy, and aiding in the discovery of new scientific knowledge that changes the limits of possibility.
AGI has the potential to give everyone incredible new capabilities; we can imagine a world where all of us have access to help with almost any cognitive task, providing a great force multiplier for human ingenuity and creativity.
On the other hand, AGI would also come with serious risk of misuse, drastic accidents, and societal disruption. Because the upside of AGI is so great, we do not believe it is possible or desirable for society to stop its development forever; instead, society and the developers of AGI have to figure out how to get it right… …Although we cannot predict exactly what will happen, and of course our current progress could hit a wall, we can articulate the principles we care about most:
- We want AGI to empower humanity to maximally flourish in the universe. We don’t expect the future to be an unqualified utopia, but we want to maximize the good and minimize the bad, and for AGI to be an amplifier of humanity.
- We want the benefits of, access to, and governance of AGI to be widely and fairly shared. We want to successfully navigate massive risks. In confronting these risks, we acknowledge that what seems right in theory often plays out more strangely than expected in practice.
- We believe we have to continuously learn and adapt by deploying less powerful versions of the technology in order to minimize “one shot to get it right” scenarios…
…As our systems get closer to AGI, we are becoming increasingly cautious with the creation and deployment of our models. Our decisions will require much more caution than society usually applies to new technologies, and more caution than many users would like. Some people in the AI field think the risks of AGI (and successor systems) are fictitious; we would be delighted if they turn out to be right, but we are going to operate as if these risks are existential.
At some point, the balance between the upsides and downsides of deployments (such as empowering malicious actors, creating social and economic disruptions, and accelerating an unsafe race) could shift, in which case we would significantly change our plans around continuous deployment…
…The first AGI will be just a point along the continuum of intelligence. We think it’s likely that progress will continue from there, possibly sustaining the rate of progress we’ve seen over the past decade for a long period of time. If this is true, the world could become extremely different from how it is today, and the risks could be extraordinary. A misaligned superintelligent AGI could cause grievous harm to the world; an autocratic regime with a decisive superintelligence lead could do that too.
AI that can accelerate science is a special case worth thinking about, and perhaps more impactful than everything else. It’s possible that AGI capable enough to accelerate its own progress could cause major changes to happen surprisingly quickly (and even if the transition starts slowly, we expect it to happen pretty quickly in the final stages). We think a slower takeoff is easier to make safe, and coordination among AGI efforts to slow down at critical junctures will likely be important (even in a world where we don’t need to do this to solve technical alignment problems, slowing down may be important to give society enough time to adapt).
Successfully transitioning to a world with superintelligence is perhaps the most important—and hopeful, and scary—project in human history. Success is far from guaranteed, and the stakes (boundless downside and boundless upside) will hopefully unite all of us.
2. Berkshire Hathaway 2022 Shareholder Letter – Warren Buffett
A common belief is that people choose to save when young, expecting thereby to maintain their living standards after retirement. Any assets that remain at death, this theory says, will usually be left to their families or, possibly, to friends and philanthropy.
Our experience has differed. We believe Berkshire’s individual holders largely to be of the once-a-saver, always-a-saver variety. Though these people live well, they eventually dispense most of their funds to philanthropic organizations. These, in turn, redistribute the funds by expenditures intended to improve the lives of a great many people who are unrelated to the original benefactor. Sometimes, the results have been spectacular.
The disposition of money unmasks humans. Charlie and I watch with pleasure the vast flow of Berkshire-generated funds to public needs and, alongside, the infrequency with which our shareholders opt for look-at-me assets and dynasty-building.
Who wouldn’t enjoy working for shareholders like ours?…
…Charlie and I allocate your savings at Berkshire between two related forms of ownership. First, we invest in businesses that we control, usually buying 100% of each. Berkshire directs capital allocation at these subsidiaries and selects the CEOs who make day-by-day operating decisions. When large enterprises are being managed, both trust and rules are essential. Berkshire emphasizes the former to an unusual – some would say extreme – degree. Disappointments are inevitable. We are understanding about business mistakes; our tolerance for personal misconduct is zero.
In our second category of ownership, we buy publicly-traded stocks through which we passively own pieces of businesses. Holding these investments, we have no say in management.
Our goal in both forms of ownership is to make meaningful investments in businesses with both long-lasting favorable economic characteristics and trustworthy managers. Please note particularly that we own publicly-traded stocks based on our expectations about their long-term business performance, not because we view them as vehicles for adroit purchases and sales. That point is crucial: Charlie and I are not stock-pickers; we are business-pickers.
Over the years, I have made many mistakes. Consequently, our extensive collection of businesses currently consists of a few enterprises that have truly extraordinary economics, many that enjoy very good economic characteristics, and a large group that are marginal. Along the way, other businesses in which I have invested have died, their products unwanted by the public. Capitalism has two sides: The system creates an ever-growing pile of losers while concurrently delivering a gusher of improved goods and services. Schumpeter called this phenomenon “creative destruction.”…
…The math isn’t complicated: When the share count goes down, your interest in our many businesses goes up. Every small bit helps if repurchases are made at value-accretive prices. Just as surely, when a company overpays for repurchases, the continuing shareholders lose. At such times, gains flow only to the selling shareholders and to the friendly, but expensive, investment banker who recommended the foolish purchases.
Gains from value-accretive repurchases, it should be emphasized, benefit all owners – in every respect. Imagine, if you will, three fully-informed shareholders of a local auto dealership, one of whom manages the business. Imagine, further, that one of the passive owners wishes to sell his interest back to the company at a price attractive to the two continuing shareholders. When completed, has this transaction harmed anyone? Is the manager somehow favored over the continuing passive owners? Has the public been hurt?
When you are told that all repurchases are harmful to shareholders or to the country, or particularly beneficial to CEOs, you are listening to either an economic illiterate or a silver-tongued demagogue (characters that are not mutually exclusive)…
…I have been investing for 80 years – more than one-third of our country’s lifetime. Despite our citizens’ penchant – almost enthusiasm – for self-criticism and self-doubt, I have yet to see a time when it made sense to make a long-term bet against America. And I doubt very much that any reader of this letter will have a different experience in the future.
3. Does Long-Term Investing Work Outside of the United States? – Ben Carlson
Elroy Dimson, Paul Marsh and Mike Staunton published a book the early-2000s called Triumph of the Optimists: 101 Years of Global Investment Returns that looked at the historical record of equity markets around the globe since the year 1900…
..And lucky for us, the authors update the data on an annual basis for the Credit Suisse Global Investment Returns Yearbook. The latest edition was just released and it’s filled with data and charts about the long-run returns in stock markets around the globe…
…The U.S. is near the top but it’s not like they’re running away with it like Secretariat… Sure, there have been some complete washouts over the years (Russia’s stock market was basically shut down for 75 years following World War I) but returns in other countries have been anywhere from OK to respectable to strong…
…The MSCI World ex-USA dates back to 1970. These were the annual returns1 from 1970 through January 2023:
- S&P 500: 10.5%
- MSCI ex-USA: 8.4%
That’s a pretty good lead for the old US of A but it’s not like the rest of the world has been chopped liver over the past 50+ years. And the majority of the U.S. outperformance has come since the 2008 financial crisis. These were the annual return through the end of 2007:
- S&P 500: 11.1%
- MSCI ex-USA: 10.9%
It was pretty darn close before the most recent cycle saw U.S. stocks slaughter the rest of the world. And it’s not like U.S. stocks have outperformed always and everywhere.
4. AI-generated comic artwork loses US Copyright protection – Benj Edwards
On Tuesday, the US Copyright Office declared that images created using the AI-powered Midjourney image generator for the comic book Zarya of the Dawn should not have been granted copyright protection, and the images’ copyright protection will be revoked…
…Last September, in a story that first appeared on Ars Technica, Kashtanova publicly announced that Zarya of the Dawn, which includes comic-style illustrations generated from prompts using the latent diffusion AI process, had been granted copyright registration. At the time, we considered it a precedent-setting case for registering artwork created by latent diffusion.
However, as the letter explains, after the Copyright Office learned that the work included AI-generated images through Kashtanova’s social media posts, it issued a notice to Kashtanova in October stating that it intended to cancel the registration unless she provided additional information showing why the registration should not be canceled. Kashtanova’s attorney responded to the letter in November with an argument that Kashtanova authored every aspect of the work, with Midjourney serving merely as an assistive tool.
That argument wasn’t good enough for the Copyright Office, which describes in detail why it believes AI-generated artwork should not be granted copyright protection. In a key excerpt provided below, the Office emphasizes the images’ machine-generated origins:
Based on the record before it, the Office concludes that the images generated by Midjourney contained within the Work are not original works of authorship protected by copyright. See COMPENDIUM (THIRD ) § 313.2 (explaining that “the Office will not register works produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author”). Though she claims to have “guided” the structure and content of each image, the process described in the Kashtanova Letter makes clear that it was Midjourney—not Kashtanova—that originated the “traditional elements of authorship” in the images…
…It’s possible that the ruling may eventually be reconsidered as the result of a cultural shift in how society perceives AI-generated art—one that may allow for a new interpretation by different members of the US Copyright Office in the decade ahead. For now, AI-powered artwork is still a novel and poorly understood technology, but it may eventually become the standard way visual arts emerge. Not allowing for copyright protection would potentially preclude its use by large and powerful media conglomerates in the future. So the story of AI and copyrights is not over yet.
5. Even a Brain-Eating Amoeba Can’t Hide From This Cutting-Edge Diagnosis Tech – Ron Winslow
When a middle-aged man who had suffered a seizure was admitted to the University of California San Francisco Medical Center in 2021, doctors seeking the cause for his condition quickly became stumped.
After pathologists spent two weeks peering through microscopes and monitoring petri dishes, doctors knew something serious was harming the patient’s brain; they had no idea what it was or how to treat it.
They turned to an emerging strategy known as unbiased diagnosis. It ultimately confirmed an illness so rare and so deadly, few doctors have ever seen it: brain-eating amoeba disease. The patient’s brain had been invaded by a single-cell critter called Balamuthia mandrillaris, one of at least three types of amoebas known to infect human brains.
The unbiased approach is called metagenomic next-generation sequencing, a powerful technology that analyzes all of the genetic material in a patient’s tissue sample and as a result can screen for a wide range of disease-causing microbes in a single test…
…The conventional search for the cause of an infection involves examining patient tissue under a microscope or culturing samples in a petri dish to see if bacteria or other microbes grow. But doctors have to be looking for a particular bug to find it. Such tests are typically ordered after doctors weigh the details of a case and form a hunch about the cause of the infection—a biased approach…
… Metagenomics is the future of medical diagnostics, said Eric Topol, director of Scripps Research Translational Institute, La Jolla, Calif. “It should be the present,” he said, but not many hospitals are equipped to do it.
A metagenomics test spells out the order of the four letters that make up the genetic code in all the DNA and RNA in a patient sample and compares the result against human and nonhuman genome sequences stored in databases such as the National Institutes of Health’s GenBank.
A typical sample might yield 100 million snippets of genetic material, Dr. DeRisi said. Some 99% would be human. Those sequences are computationally stripped away and the remaining 1 million pieces are screened against all the sequences in GenBank in an effort to find a match…
…A biased diagnosis can be likened to the card game Go Fish, said Natasha Spottiswoode, an infectious disease physician at UCSF who has overseen care of the Balamuthia patient. A player holding a green fish card asks another, “Do you have any green fish?” If the answer is no, the question on the next turn may be, “Do you have any red fish?”
For an unbiased query, “What you really want to ask is, ‘Do you have any fish at all?’” Dr. Spottiswoode said. “And then figure out what color they are.”
In 2014, Dr. DeRisi was among a team of researchers and clinicians at UCSF who reported on one of the first patients to be successfully treated based on metagenomics sequencing—a 14-year-old boy whose treatable, but potentially fatal Leptospirosis bacterial brain infection went undiagnosed for several months until the test was performed.
The case convinced Dr. DeRisi and his colleagues that a metagenomics test should be deployed as a clinical tool for diagnosing brain infections and eventually led UCSF to offer the tests to other hospitals. Innovation in semiconductor technology is helping make the service possible, Dr. DeRisi said. “If we dial back 10 or 12 years ago, we couldn’t do this,” he said. “If we didn’t have increases in computer storage, memory and speed, we’d be sunk.”…
…Metagenomics has limitations. The test can pick up dormant or otherwise clinically irrelevant microbes, making it difficult to interpret results. It can miss pathogens that are detected by conventional means. UCSF’s brain infection test costs about $2,000, far less than the cost of a day in the ICU, but still a potential impediment to regular use. Insurance reimbursement is spotty. Turnaround time can be as long as six or seven days, Dr. DeRisi said.
6. James Revell – Wise: Moving Money Around the World – Zack Fuss and James Revell
James: [00:11:24] Maybe I’ll start with just providing a bit more context and background on the cross-border money transfer market, and that can set up this counter-position Wise has. So if you think about cross-border money transfers, it goes back thousands of years. There are history books written on this. Ultimately, back in the day, it would have been gold bullion or precious metal spices being loaded on to ships and transferred across borders.
Obviously, that is quite impractical, causes security concerns, costs and all sorts. Over time, we’re talking 11th, 12th century now, the bill of exchange was created, whereby you could essentially create an IOU which meant you didn’t need to transfer actual money or currency across borders, it’s more a paper-based exchange of value that could be redeemed at a bank. These were posted at the time.
And then that has grown into telex messaging when cross-Atlantic cables are laid and this electronic means that of communication between bank arrive. And so the history of cross-border transfers, if you think about it, money stopped moving across borders a long time ago. It was a lot of credits and debits of accounts with each banks held with each other, moving numbers around on ledgers as opposed to money actually being sent on a ship or through this kind of pipe. It’s just a series of relationships between banks.
That’s known as correspondent banking, and we can come back to some of the problems with that model. But maybe just to set up then how big this market is. So I think there’s about GBP 100 trillion of volume transferred across borders every year. If you cut out, say, the really big enterprise government or interbank transfers, you’re left with about GBP 2 trillion of personal cross-border transfers and about GBP 9 trillion of small businesses transferring money across borders. So this is a colossal market.
And that flows through to, say, a revenue line of anywhere between GBP 100 million and GBP 200 billion paid in fees by customers. In terms of the split of that market, about 2/3 still sits with banks, about 10% to 20% with money transfer operators like the Western Unions and the MoneyGrams and the rest is split up between the remaining players. The market is growing. So on average, growing about 5% per annum over the past decade. But interestingly, fees are reducing.
There’s some pressure on fee down was largely caused by Wise, but also by regulatory attention. But 2008, 2009, it was about 9% in fees. It’s now like, say, close to 6%, 7%. The tailwinds behind this market is largely been driven by globalization. So international trade and supply chains, global e-commerce, international traveler migration. And it’s received, like I said, a lot of regulatory attention. So the G20 and the FSB are really focused on this right now.
UN have set of goal that by 2030, cross-border transfer fees should be close to 3%. And why that is, is because it’s a very inefficient way of doing things. These are very high. And largely, unfortunately, the people that suffer are normally immigrants, trying to work overseas and transfer money every month back home to support their families. And so that’s one reason I think the regulatory attention has come. The other reason is because financial crime and the proceeds of crime transferring across borders, if that’s in cash, that causes a problem.
So underpinning this market is what’s known as correspondent banking which is the relationship between banks around the world. So if you’re, say, Commonwealth Bank of Australia, you need to transfer money to Barclays in the U.K., you may not have a direct relationship or say, you’re a credit union in Australia, it would very unlikely to have a relationship with the bank in the U.K. And so you’ll contact a bank in Australia who will contact a bank in, say, the U.K. who will then contact the recipients bank.
This chain of communication between banks is called correspondent banking. And what they’re doing is basically transferring the request and the amount of money they need to spend as well as the customer information. Now the way the banks do this communication is reliant upon an organization called SWIFT. So SWIFT is owned by 200 banks. It’s used by over 11,000 of them and it stands for the Society of Worldwide Interbank Financial Telecommunications.
So it’s essentially an electronic communication network that tries to put in place a common language and some standards around how a cross-border transfers work, what format data needs to come in. So you may have heard of SWIFT code or an IBAN number, these are identifiers that banks use globally to help them manage this complex network of communication, which underpins these transfers.
That doesn’t take away the individual effort put on to banks in order to complete transfers. And so maybe just to run through the process quickly. First of all, you’ve got to find a bank so you want to transfer money to Thailand, as a bank, you may need to go through three or four different banks to get to the recipients banking and so you’ve got to find a way of getting to the recipient bank.
Every step of the way you need to validate the data, you needed to complete the regulatory checks, you then need to potentially transform the data for the next bank along in the chain, you need to transmit it. You then need to sort out funding on the back end. So settling crediting and debiting accounts, this could be like a $50 transfer. And these six banks all need to communicate, they all need to settle funds and they all need to reconcile and make sure that all ticks were completed, all data has been transferred accurately and everyone’s got the money they need.
That’s set up, you can see that there’s a ton of problems caused by that. The first and the most obvious one, which has been growing over the past 10 years, particularly since the GFC, the regulatory and compliance burden along that chain is huge. So making sure that there’s no anti-money laundering going on. There’s no counterterrorism financing going on. You’ve got your sanctions checks, you’ve got to protect data.
So you’ve got be mindful of privacy, you’ve got prudential worries about liquidity and bank regulations and consumer protections like if something goes wrong, you got to solve disputes. So there’s an enormous amount of complexity just on the regulation and compliance side. There are interesting things like opening hours. So banks communicating across the world, they have different opening hours.
The way banks transfer money domestically between each other is normally done in a batch process model, whereby it’s only operating five days a week and at certain times of that day. So if you span a weekend, your money is not moving. There’s also a lot of paper and legacy technology involved. There’s liquidity and foreign exchange risks.
So if transfers between banks aren’t done simultaneously, which often they are, that creates liabilities between banks, which is problems and banks charge fees for that. So correspondent banking has a lot of inefficiency built in, and that flows through. So those are supply side problems flow through to problems for customers like our poor Kristo and Taavet back in 2008. It’s incredibly expensive. Every hand off along the chain, they need to be paid. It’s slow.
Each step along the chain, obviously, there’s some waiting time involved and it’s opaque. Often upfront where the transfer starts, you have no idea how much it’s going to cost when it comes out the other end. And ultimately, it’s inconvenient…
…Zack: [01:03:30] And then as you kind of reflect upon what you’ve learned about this business as you studied it and the broader payment space, what is a lesson that you can take from this business and apply to others from an investor’s perspective? And then some of the other early stage or late-stage growth companies that you look at, what are lessons you’d like to see them borrow from Wise and apply to their own business as operators?
James: [01:03:48] The one thing we haven’t really touched upon, which I think Wise has, and being an operator myself is something that I could have learned from. The way they organize their people and prioritize their culture, I think, is relatively unique. I attended a talk with someone from TransferWise back in 2018. You talked about the way they prioritize resource allocation internally.
So what they have is they have small teams, so they have over — I think it’s over 100 teams, small, highly autonomous empowered, cross-functional teams each working on a solution or a feature. And that team is empowered to create its own vision, its own mission, its own objectives. To the extent it can even go and seek its own legal advice. So it’s fully — each team is fully autonomous. And they basically vie for prioritization amongst each other.
So it’s a lot of mini businesses within a business. Why I think that is fascinating. It has some problems, which potentially we can touch on. But why that resonates with me is the thesis around having very highly aligned but very loosely coupled people within your organization to overcome this inertia and this slowdown that occurs as the business gets bigger.
So if you have low alignment and high autonomy, you have very empowered silos, but they’re doing their own thing. They might not be pulling in the same direction. There’s probably some duplication or overlap or potentially even pulling in opposite directions. On the flip side, you can have very high alignment but low autonomy, which is basically a command and control type structure. Both of these can exist and do exist quite regularly, but they really struggle to scale.
Reed Hastings talks about this high alignment loose coupling where you tell people what to do, not how to do it. And I think this is what these small many businesses within the business is what Wise has managed to organize as they’ve grown up. And key to it is having that super clear mission, super clear vision, you’ve got your customer at the center of everything, you know why you exist, and that creates the alignment.
There’s a huge amount of trust and transparency within the business. And I think this culture focusing on one thing, you see in other scale economy shared businesses. There’s been great breakdowns on Floor & Decor or Costco, where it’s a relentless focus on one thing, retaining operational improvements and the benefit of unit cost efficiencies back to the customer. But it is that one thing focus and just doing that one thing really well and organizing a culture behind it. For us, as people and culture investors, that is so powerful, and I think it’s one of the untold stories of this business.
7. Will High Risk-Free Rates Derail the Stock Market? – Ben Carlson
Because of the Fed’s interest rate hikes, investors are being offered a gift right now in the form of relatively high yields on essentially risk-free securities (if such a thing exists). You don’t have to go further out on the risk curve to find yield right now.
Short-term bonds with little-to-no interest rate or duration risk are offering 5% yields.
The big question for asset allocators is this: Will higher risk-free rates impact the demand for stocks and other risk assets which leads to poor returns?
This makes sense in theory. Why take more risk when that 5% guaranteed yield is sitting there for the taking?
The relationship between risk-free rates and stock market returns is not as sound as it would seem in theory…
…The highest average yields occurred in the 1980s, which was also one of the best decades ever for stocks. Yields were similarly elevated in the 1970s and 1990s but one of those decades experienced subpar returns while the other saw lights-out performance…
…I also looked at the performance of the stock market when 3-month T-bill yields averaged 5% for the entirety of a year (which could happen this year). That’s been the case in 25 of the last 89 years.
The annualized return for the S&P 500 in those 25 years was 11%. So in years with above-average risk-free rates, the stock market has actually seen above-average returns.
Disclaimer: None of the information or analysis presented is intended to form the basis for any offer or recommendation. We currently have a vested interest in Costco and Wise. Holdings are subject to change at any time.
