Insights From Warren Buffett’s 2024 Shareholder’s Letter

There’s much to learn from Warren Buffett’s latest letter, including his thoughts on the P/C (property and casualty) insurance industry, and how to think about shares in public-listed as well as private companies.

One document I always look forward to reading around this time of the year is Warren Buffett’s annual Berkshire Hathaway shareholder’s letter. Over the weekend, Buffett published the 2024 edition and here are some of my favourite insights from it that I wish to document and share. 

Without further ado (emphases are Buffett’s)…

Could the use of the word “mistakes” frequently in company reports be a signal to find great investing opportunities?

During the 2019-23 period, I have used the words “mistake” or “error” 16 times in my letters to you. Many other huge companies have never used either word over that span. Amazon, I should acknowledge, made some brutally candid observations in its 2021 letter. Elsewhere, it has generally been happy talk and pictures.

I have also been a director of large public companies at which “mistake” or “wrong” were forbidden words at board meetings or analyst calls. That taboo, implying managerial perfection, always made me nervous (though, at times, there could be legal issues that make limited discussion advisable. We live in a very litigious society.)   

It’s hard to strike a bad deal when you’re dealing with a great person, even when the deal is vaguely-worded

Let me pause to tell you the remarkable story of Pete Liegl, a man unknown to most Berkshire shareholders but one who contributed many billions to their aggregate wealth. Pete died in November, still working at 80.

 I first heard of Forest River – the Indiana company Pete founded and managed – on June 21, 2005. On that day I received a letter from an intermediary detailing relevant data about the company, a recreational vehicle (“RV”) manufacturer…

…I did some checking with RV dealers, liked what I learned and arranged a June 28th meeting in Omaha…

…Pete next mentioned that he owned some real estate that was leased to Forest River and had not been covered in the June 21 letter. Within a few minutes, we arrived at a price for those assets as I expressed no need for appraisal by Berkshire but would simply accept his valuation.

Then we arrived at the other point that needed clarity. I asked Pete what his compensation should be, adding that whatever he said, I would accept. (This, I should add, is not an approach I recommend for general use.)

Pete paused as his wife, daughter and I leaned forward. Then he surprised us: “Well, I looked at Berkshire’s proxy statement and I wouldn’t want to make more than my boss, so pay me $100,000 per year.” After I picked myself off the floor, Pete added: “But we will earn X (he named a number) this year, and I would like an annual bonus of 10% of any earnings above what the company is now delivering.” I replied: “OK Pete, but if Forest River makes any significant acquisitions we will make an appropriate adjustment for the additional capital thus employed.” I didn’t define “appropriate” or “significant,” but those vague terms never caused a problem.

The four of us then went to dinner at Omaha’s Happy Hollow Club and lived happily ever after. During the next 19 years, Pete shot the lights out. No competitor came close to his performance.   

A handful of great decisions can wash away a multitude of mistakes, and then some

Our experience is that a single winning decision can make a breathtaking difference over time. (Think GEICO as a business decision, Ajit Jain as a managerial decision and my luck in finding Charlie Munger as a one-of-a-kind partner, personal advisor and steadfast friend.) Mistakes fade away; winners can forever blossom. 

A person’s educational background has no bearing on his/her ability

One further point in our CEO selections: I never look at where a candidate has gone to school. Never!…

…Not long ago, I met – by phone – Jessica Toonkel, whose step-grandfather, Ben Rosner, long ago ran a business for Charlie and me. Ben was a retailing genius and, in preparing for this report, I checked with Jessica to confirm Ben’s schooling, which I remembered as limited. Jessica’s reply: “Ben never went past 6th grade.”    

Insurance companies are going to face a staggering environmental catastrophe event someday, but insurance companies can still do well if they price their policies appropriately

In general, property-casualty (“P/C”) insurance pricing strengthened during 2024, reflecting a major increase in damage from convective storms. Climate change may have been announcing its arrival. However, no “monster” event occurred during 2024. Someday, any day, a truly staggering insurance loss will occur – and there is no guarantee that there will be only one per annum…

…We are not deterred by the dramatic and growing loss payments sustained by our activities. (As I write this, think wildfires.) It’s our job to price to absorb these and unemotionally take our lumps when surprises develop. It’s also our job to contest “runaway” verdicts, spurious litigation and outright fraudulent behavior.  

EBITDA (earnings before interest, taxes, depreciation, and amortisation) is not a good measure of a company’s profitability

Here’s a breakdown of the 2023-24 earnings as we see them. All calculations are after depreciation, amortization and income tax. EBITDA, a flawed favorite of Wall Street, is not for us.  

A company can pay massive tax bills and yet be immensely valuable

To be precise, Berkshire last year made four payments to the IRS that totaled $26.8 billion. That’s about 5% of what all of corporate America paid. (In addition, we paid sizable amounts for income taxes to foreign governments and to 44 states.)…

…For sixty years, Berkshire shareholders endorsed continuous reinvestment and that enabled the company to build its taxable income. Cash income-tax payments to the U.S. Treasury, miniscule in the first decade, now aggregate more than $101 billion . . . and counting.   

Shares of public-listed companies and private companies should be seen as the same kind of asset class – ownership stakes in businesses

Berkshire’s equity activity is ambidextrous. In one hand we own control of many businesses, holding at least 80% of the investee’s shares. Generally, we own 100%. These 189 subsidiaries have similarities to marketable common stocks but are far from identical. The collection is worth many hundreds of billions and includes a few rare gems, many good-but-far-from-fabulous businesses and some laggards that have been disappointments… 

…In the other hand, we own a small percentage of a dozen or so very large and highly profitable businesses with household names such as Apple, American Express, Coca-Cola and Moody’s. Many of these companies earn very high returns on the net tangible equity required for their operations…

…We are impartial in our choice of equity vehicles, investing in either variety based upon where we can best deploy your (and my family’s) savings…

…Despite what some commentators currently view as an extraordinary cash position at Berkshire, the great majority of your money remains in equities. That preference won’t change. While our ownership in marketable equities moved downward last year from $354 billion to $272 billion, the value of our non-quoted controlled equities increased somewhat and remains far greater than the value of the marketable portfolio. 

Berkshire Hathaway has a great majority of its capital invested in equities, despite what may seem to be otherwise on the surface (this is related to the point above on how shares of public-listed companies and private companies should be both seen as equities); Berkshire will always be deploying the lion’s share of its capital in equities

Despite what some commentators currently view as an extraordinary cash position at Berkshire, the great majority of your money remains in equities. That preference won’t change. While our ownership in marketable equities moved downward last year from $354 billion to $272 billion, the value of our non-quoted controlled equities increased somewhat and remains far greater than the value of the marketable portfolio.

Berkshire shareholders can rest assured that we will forever deploy a substantial majority of their money in equities – mostly American equities although many of these will have international operations of significance. Berkshire will never prefer ownership of cash-equivalent assets over the ownership of good businesses, whether controlled or only partially owned. 

Good businesses will still succeed even if a government bungles its fiscal policy (i.e. government spending), but it’s still really important for a government to maintain a stable currency 

Paper money can see its value evaporate if fiscal folly prevails. In some countries, this reckless practice has become habitual, and, in our country’s short history, the U.S. has come close to the edge. Fixed-coupon bonds provide no protection against runaway currency.

Businesses, as well as individuals with desired talents, however, will usually find a way to cope with monetary instability as long as their goods or services are desired by the country’s citizenry…

…So thank you, Uncle Sam. Someday your nieces and nephews at Berkshire hope to send you even larger payments than we did in 2024. Spend it wisely. Take care of the many who, for no fault of their own, get the short straws in life. They deserve better. And never forget that we need you to maintain a stable currency and that result requires both wisdom and vigilance on your part. 

Capitalism is still a force for good

One way or another, the sensible – better yet imaginative – deployment of savings by citizens is required to propel an ever-growing societal output of desired goods and services. This system is called capitalism. It has its faults and abuses – in certain respects more egregious now than ever – but it also can work wonders unmatched by other economic systems.

P/C (property and casualty) insurance companies often do not know the true cost of providing their services until much later; the act of pricing insurance policies is part art and part science, and requires a cautious (pessimistic?) mindset

When writing P/C insurance, we receive payment upfront and much later learn what our product has cost us – sometimes a moment of truth that is delayed as much as 30 or more years. (We are still making substantial payments on asbestos exposures that occurred 50 or more years ago.)

This mode of operations has the desirable effect of giving P/C insurers cash before they incur most expenses but carries with it the risk that the company can be losing money – sometimes mountains of money – before the CEO and directors realize what is happening.

Certain lines of insurance minimize this mismatch, such as crop insurance or hail damage in which losses are quickly reported, evaluated and paid. Other lines, however, can lead to executive and shareholder bliss as the company is going broke. Think coverages such as medical malpractice or product liability. In “long-tail” lines, a P/C insurer may report large but fictitious profits to its owners and regulators for many years – even decades…

…Properly pricing P/C insurance is part art, part science and is definitely not a business for optimists. Mike Goldberg, the Berkshire executive who recruited Ajit, said it best: “We want our underwriters to daily come to work nervous, but not paralyzed.”   

There are forms of executive compensation that are one-sided in favour of the executives, which can create distorted incentives

Greg, our directors and I all have a very large investment in Berkshire in relation to any compensation we receive. We do not use options or other one-sided forms of compensation; if you lose money, so do we. This approach encourages caution but does not ensure foresight.

Economic growth in a country is a necessary ingredient for the P/C insurance industry to grow

P/C insurance growth is dependent on increased economic risk. No risk – no need for insurance.

Think back only 135 years when the world had no autos, trucks or airplanes. Now there are 300 million vehicles in the U.S. alone, a massive fleet causing huge damage daily

It’s important for an insurance company to know when to shrink its business (when policies are priced inadequately)

No private insurer has the willingness to take on the amount of risk that Berkshire can provide. At times, this advantage can be important. But we also need to shrink when prices are inadequate. We must never write inadequately-priced policies in order to stay in the game. That policy is corporate suicide. 

A P/C insurance company that does not depend on reinsurers has a material cost advantage

All things considered, we like the P/C insurance business. Berkshire can financially and psychologically handle extreme losses without blinking. We are also not dependent on reinsurers and that gives us a material and enduring cost advantage. 

Good investments can be made by just assessing a company’s financial records and buying when prices are low; it can make sense to invest in foreign countries even without having a view on future foreign exchange rates 

It’s been almost six years since Berkshire began purchasing shares in five Japanese companies that very successfully operate in a manner somewhat similar to Berkshire itself. The five are (alphabetically) ITOCHU, Marubeni, Mitsubishi, Mitsui and Sumitomo…

…Berkshire made its first purchases involving the five in July 2019. We simply looked at their financial records and were amazed at the low prices of their stocks…

…At yearend, Berkshire’s aggregate cost (in dollars) was $13.8 billion and the market value of our holdings totaled $23.5 billion…

…Greg and I have no view on future foreign exchange rates and therefore seek a position approximating currency-neutrality.

Berkshire Hathaway will be investing in Japan for a very long time

A small but important exception to our U.S.-based focus is our growing investment in Japan…

…Our holdings of the five are for the very long term, and we are committed to supporting their boards of directors. From the start, we also agreed to keep Berkshire’s holdings below 10% of each company’s shares. But, as we approached this limit, the five companies agreed to moderately relax the ceiling. Over time, you will likely see Berkshire’s ownership of all five increase somewhat. 

It can make sense to borrow to invest for the long haul in foreign countries if you can fix your interest payments at low rates

Meanwhile, Berkshire has consistently – but not pursuant to any formula – increased its yen-denominated borrowings. All are at fixed rates, no “floaters.” Greg and I have no view on future foreign exchange rates and therefore seek a position approximating currency-neutrality…

…We like the current math of our yen-balanced strategy as well. As I write this, the annual dividend income expected from the Japanese investments in 2025 will total about $812 million and the interest cost of our yen-denominated debt will be about $135 million. 


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. I currently have a vested interest in Amazon and Apple. Holdings are subject to change at any time.

What Do Job Cuts Mean For Shareholders?

Job cuts can have both positive and negative consequences for a company

Recently, Meta Platforms Inc (NASDAQ: META) announced that it would cut around 5% of its global workforce. I was discussing this with a friend of mine, who is also currently working for Meta and we talked about some of the pros and cons of job cuts from the perspective of shareholders.

Let’s start with some of the pros.

Canceling unvested RSUs

When Meta cuts jobs, it also cancels all unvested restricted stock units (RSUs) that would have vested over time had the employee stayed on. The cancellation of unvested RSUs reduces the dilution from stock-based compensation.

Bear in mind, the number of RSUs granted is based on the stock price back when the RSUs were granted, and not when they vest. Back in 2022, Meta granted a huge number of RSUs as refreshers because of its lower stock price. For context, Meta granted 59 million RSUs in 2021 (when its stock price was high) but because of the refreshers and low stock prices in 2022 and 2023, Meta granted 107 million and 109 million RSUs in 2022 and 2023, respectively. 

Cancelling some of these unvested RSUs will reduce dilution. In addition, hiring new employees and granting new RSUs will not result in as much dilution because Meta’s stock price is now around 7 times higher from the lows seen in 2022.

Getting better talent/ motivate existing employees

Meta cut jobs based on performance. By cutting low performers and hiring new employees, Meta could potentially improve the quality of its talent.

It also keeps current employees on their guard and creates an environment where employees work hard to ensure that performance reviews are good. This prevents employees from simply coasting through work and collecting wages without adding much value to the company.

Reducing the wage bill

Wages are one of the largest expenses for a company such as Meta. Although it is likely that Meta will eventually replace the employees that were removed, the company seems intent on keeping the team lean.

In 2022, Meta cut 11,000 employees, or 13% of its workforce and in 2023, the company cut an additional 10,000 employees as it strived for a “year of efficiency”.

For perspective, Meta’s head count declined from 86,482 in 2022 to 74,067 in 2024, despite revenue climbing 41% in two years from US$116.6 billion to US$164.5 billion. This, together with operational leverage, resulted in net profit margins rising from 20% in 2022 to 38% in 2024. 

But, employee cuts could potentially end up with undesirable side effects. Here are the cons.

Lower risk taking

Cutting staff based on performance can lead to less risk-taking and innovation. This is because if the employee embarks on a more innovative but risky project that ends up failing, his or her performance may be considered poor.

This may lead employees to be less innovative or to take a safe approach when it comes to projects, creating an environment of lower innovation.

Internal competition

Another potential side effect is employees may start competing with each other. This may result in less collaboration and senior staff may be less willing to train new employees as they view them as competitors to their job.

This can create a toxic work environment. 

Final thoughts

Job cuts are difficult for those impacted. However, it may also be a necessary way for companies to reduce expenses and to ensure that the company remains competitive.

Looking from the lens of a shareholder, I believe job cuts can be a good thing if done correctly and can also lead to more efficiency, more profits and eventually more dividends.

However, my discussion with my friend has also opened my eyes to some of the negative impacts of workforce reduction. Companies that do layoffs need to consider these factors and try to ensure that some of these potential negative side effects do not have a huge impact on the company.


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. I have a vested interest in Meta Platforms Inc. Holdings are subject to change at any time.

The Buyback Endemic

Buying back stock at unreasonably high valuations is not a good use of capital and can destroy shareholder value.

Buybacks can be a good way for companies to enhance shareholder value. Share buybacks reduce the number of shares outstanding, allowing companies to pay a higher dividend per share in the future.

But not all buybacks are good. Done at the wrong price, buybacks can actually be a bad use of capital. In fact, I have seen so many companies do buybacks recklessly and without consideration of the share price.

The problem probably arises from a few reasons. 

Wrong mindset

First, some executives do not have a good grasp of what buybacks are. Take this statement from Tractor Supply’s management in its 2024 second-quarter earnings report for example:

“The Company repurchased approximately 0.5 million shares of its common stock for $139.2 million and paid quarterly cash dividends totaling $118.5 million, returning a total of $257.7 million of capital to shareholders in the second quarter of 2024.”

The issue with this statement is that it lumps dividends and share repurchases in the same bracket. It also implies that share repurchases are a form of returning capital to shareholders. The truth is that share repurchases is not returning cash to long-term shareholders but only to exiting shareholders. If management mistakes repurchases as capital return, it may lead them to do buybacks regularly, instead of opportunistically.

Although I am singling out Tractor Supply’s management, they are just one out of many management teams that seem to have the wrong mindset when it comes to buybacks.

Incentives

Additionally, executive compensation schemes may encourage management to buy back stock even if it is not the best use of capital. 

For instance, Adobe’s executives have an annual cash remuneration plan that is determined in part by them achieving certain earnings per share goals. This may lead management to buy back stock simply to boost the company’s earnings per share. But doing so when prices are high is not a good use of capital. When Adobe’s stock price is high, it would be better for management to simply return dividends to shareholders – but management may not want to pay dividends as it does not increase the company’s earnings per share.

Again, while I am singling out Adobe’s management, there are numerous other companies that have the same incentive problem.

Tax avoidance

I have noticed that the buyback phenomena is more prevalent in countries where dividends are taxed. 

The US, for instance, seems to have a buyback endemic where companies buy back stock regardless of the price. This may be due to the fact that US investors have to pay a tax on dividends, which makes buybacks a more tax-efficient use of capital for shareholders. On the contrary, Singapore investors do not need to pay taxes on dividends. As such, Singapore companies do not do buybacks as often.

However, simply doing buybacks for tax efficiency reasons without considering the share price can still harm shareholders. Again, management teams need to weigh both the pros and cons of buybacks before conducting them.

Final thoughts

There is no quick fix to this problem but there are some starting points that I believe companies can do to address the issue. 

First, fix the incentives problem. A company’s board of directors need to recognise that incentives that are not structured thoughtfully can encourage reckless buybacks of shares regardless of the share price.

Second, management teams need to educate themselves on how to increase long-term value for shareholders and to understand the difference between buybacks and dividends.

Third, management teams need to understand the implications of taxes properly. Although it is true that taxes can affect shareholders’ total returns when a company pays a dividend, it is only one factor when it comes to shareholder returns. Executive teams need to be coached on these aspects of capital allocation.

Only through proper education and incentives, will the buyback endemic be solved.


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. I have a vested interest in Adobe and Tractor Supply. Holdings are subject to change at any time.

Shorting Stocks Is Hard, Really Hard

It’s far easier to recognise poor underlying business fundamentals in a stock and simply avoid investing in it.

In investing parlance, to “short a stock” is to make an investment with the view that a stock’s price will decline. On the surface, shorting seems like a fairly easy thing to do for an investor who has skill in “going long”, which is to invest with the view that a stock’s price will rise – you just have to do the opposite of what’s working.

But if you peer beneath the hood, shorting can be a really difficult way to invest in the stock market. Nearly four years ago in April 2020, I wrote Why It’s So Difficult To Short Stocks, where I used the story of Luckin Coffee to illustrate just how gnarly shorting stocks can be:

In one of our gatherings in June 2019, a well-respected member and deeply accomplished investor in the club gave a presentation on Luckin Coffee (NASDAQ: LK)…

…At the time of my club mate’s presentation, Luckin’s share price was around US$20, roughly the same level from the close of its IPO in May 2019. He sold his Luckin shares in January 2020, around the time when Luckin’s share price peaked at US$50. Today, Luckin’s share price is around US$4. The coffee chain’s share price tanked by 76% from US$26 in one day on 2 April 2020 and continued falling before stock exchange operator NASDAQ ordered a trading halt for Luckin shares…

…The wheels came off the bus only on 2 April 2020. On that day, Luckin announced that the company’s board of directors is conducting an internal investigation. There are fraudulent transactions – occurring from the second quarter of 2019 to the fourth quarter of 2019 – that are believed to amount to RMB 2.2 billion (around US$300 million). For perspective, Luckin’s reported revenue for the 12 months ended 30 September 2019 was US$470 million, according to Ycharts. The exact extent of the fraudulent transactions has yet to be finalised. 

Luckin also said that investors can no longer rely on its previous financial statements for the nine months ended 30 September 2019. The company’s chief operating officer, Liu Jian, was named as the primary culprit for the misconduct. He has been suspended from his role…

…it turns out that fraudulent transactions at Luckin could have happened as early as April 2019. From 1 April 2019 to 31 January 2020, Luckin’s share price actually increased by 59%. At one point, it was even up by nearly 150%.

If you had shorted Luckin’s shares back in April 2019, you would have faced a massive loss – more than what you had put in – even if you had been right on Luckin committing fraud. This shows how tough it is to short stocks. Not only must your analysis on the fundamentals of the business be right, but your timing must also be right because you could easily lose more than you have if you’re shorting. 

Recent developments at a company named Herbalife (NYSE: HLF) present another similar illustration of the onerous task of shorting. High-profile investor Bill Ackman first disclosed that he was short Herbalife in December 2012. Back then, the company was a “global network marketing company that sells weight management, nutritional supplement, energy, sports & fitness products and personal care products” in 79 countries, according to its 2011 annual report. Today, Herbalife is a “global nutrition company that provides health and wellness products to consumers in 95 markets,” based on a description given in its 2023 annual report. So the company has been in pretty much the same line of business over this span of time.

Ackman’s short-thesis centred on his view that Herbalife was a company running an illegal pyramid scheme, and so the business model was simply not sustainable. When Ackman announced that he was short Herbalife’s shares, the company was reporting consistent and strong growth in its business. From 2006 to 2011, Herbalife’s revenue compounded at an annualised rate of 13% from US$1.9 billion to US$3.5 billion while its profit grew from US$143 million to US$415 million, representing a compounded annual growth rate of 24%.

Although Herbalife has to-date never officially been found to be operating an illegal pyramid scheme, its business results since Ackman came public with his short has been poor. The table below shows Herbalife’s revenue, net income, and net income margins from 2011 to 2023. What’s notable is the clear downward trend in both Herbalife’s net income and net income margin in that time frame. 

Source: Tikr

According to a Bloomberg article published at the end of February 2018, Ackman had effectively ended his short position on Herbalife by the time the piece came to print. I think most investors who are made to guess Ackman’s returns from his Herbalife short by looking only at the trajectory of the company’s financials from 2011 to 2017 would have noted the stark deterioration – the company’s net income declined by nearly 40% and its net income margin shrank from 12.0% to 4.8% – and conclude that Ackman had probably made a decent gain. 

But the stock market had other ideas. Herbalife’s stock price closed at US$23.16 on the day just prior to Ackman’s first public declaration of his short position. It closed at US$46.05 – a double from US$23.16 – when the aforementioned Bloomberg article was published. From December 2012 to today, the highest close for Herbalife’s stock price was US$61.47, which was reached on 4 February 2019. Right now, Herbalife’s stock price is at US$8.07. This comes after Herbalife’s stock price fell by 32% to US$8.03 on 15 February 2024 after the company reported its 2023 fourth-quarter results. Following the sharp decline, Ackman proclaimed on X (previously known as Twitter) that “it is a very good day for my psychological short on Herbalife.” 

The market eventually reflected the deterioration in Herbalife’s fundamentals, but the interim journey was a wild ride. In a similar manner to Luckin’ Coffee (and borrowing the prose from the last paragraph of the excerpts above from Why It’s So Difficult To Short Stocks), if you had shorted Herbalife’s shares back in December 2012 and held onto the position till now, you would have faced a massive loss in the interim – more than what you had put in – even if you were right on Herbalife’s collapsing fundamentals and eventual stock price decline.  

The investing sage Philip Fisher once wrote that “it is often easier to tell what will happen to the price of a stock than how much time will elapse before it happens.” This explains why shorting stocks is hard – really hard. To be successful at shorting, you need to correctly read both the stock’s underlying business fundamentals and the timing of the stock’s price movement. In contrast, it’s far easier to recognise poor underlying business fundamentals in a stock and simply avoid investing in it.


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. I currently have no vested interest in any company mentioned. Holdings are subject to change at any time.

The Everlasting Things In Human Affairs

Knowing the things that are stable over time can be incredibly useful in all areas of life.

Morgan Housel is one of my favourite writers in finance. In November 2023, he published his second book, Same as Ever: A Guide to What Never Changes. As the title suggests, the book is about mankind’s behavioural patterns and ways of thinking that do not seem to change over time.

Jeff Bezos, Amazon’s founder, once said: 

“I very frequently get the question: “What’s going to change in the next 10 years?” And that is a very interesting question; it’s a very common one. I almost never get the 14 question: “What’s not going to change in the next 10 years?” And I submit to you that that second question is actually the more important of the two — because you can build a business strategy around the things that are stable in time. … [I]n our retail business, we know that customers want low prices, and I know that’s going to be true 10 years from now. They want fast delivery; they want vast selection.”

Similarly, I believe that knowing the things that are stable over time can be incredibly useful in all areas of life – business, investing, relationships, and more. While reading Same as Ever, I made notes of the striking things I learnt from the book. I thought it would be useful to share this with a wider audience, so here they are:

The USA could have lost the Revolutionary War to Britain were it not for something as capricious as the wind

The Battle of Long Island was a disaster for George Washington’s army. His ten thousand troops were crushed by the British and its four-hundred-ship fleet. But it could have been so much worse. It could have been the end of the Revolutionary War. All the British had to do was sail up the East River and Washington’s cornered troops would have been wiped out. But it never happened, because the wind wasn’t blowing in the right direction and sailing up the river became impossible.

Historian David McCullough once told interviewer Charlie Rose that “if the wind had been in the other direction on the night of August twenty-eighth [1776], I think it would have all been over.”

“No United States of America if that had happened?” Rose asked.

“I don’t think so,” said McCullough.

“Just because of the wind, history was changed?” asked Rose.

“Absolutely,” said McCullough. 

Risk is what you don’t see

As financial advisor Carl Richards says, “Risk is what’s left over after you think you’ve thought of everything.” That’s the real definition of risk—what’s left over after you’ve prepared for the risks you can imagine. Risk is what you don’t see.

When a past event looks inevitable to us today, we may be fooled by hindsight bias

Two things can explain something that looks inevitable but wasn’t predicted by those who experienced it at the time: 

  • Either everyone in the past was blinded by delusion.
  • Or everyone in the present is fooled by hindsight.

We are crazy to think it’s all the former and none of the latter.

The level of uncertainty in the economy rarely fluctuates, just people’s perceptions

There is rarely more or less economic uncertainty; just changes in how ignorant people are to potential risks. Asking what the biggest risks are is like asking what you expect to be surprised about. If you knew what the biggest risk was you would do something about it, and doing something about it would make it less risky. What your imagination can’t fathom is the dangerous stuff, and it’s why risk can never be mastered

Even when the Great Depression of the 1930s happened, unemployment was not thought to be an issue by people with high posts

The Depression, as we know today, began in 1929. But when the well-informed members of the National Economic League were polled in 1930 as to what they considered the biggest problem of the United States, they listed, in order:

1. Administration of justice

2. Prohibition

3. Disrespect for law

4. Crime

5. Law enforcement

6. World peace

And in eighteenth place . . . unemployment.

A year later, in 1931—a full two years into what we now consider the Great Depression—unemployment had moved to just fourth place, behind prohibition, justice, and law enforcement. That’s what made the Great Depression so awful: No one was prepared for it because no one saw it coming. So people couldn’t deal with it financially (paying their debts) and mentally (the shock and grief of sudden loss).

Having expectations instead of forecasts is important when trying to manage risk

It’s impossible to plan for what you can’t imagine, and the more you think you’ve imagined everything the more shocked you’ll be when something happens that you hadn’t considered. But two things can push you in a more helpful direction.

One, think of risk the way the State of California thinks of earthquakes. It knows a major earthquake will happen. But it has no idea when, where, or of what magnitude. Emergency crews are prepared despite no specific forecast. Buildings are designed to withstand earthquakes that may not occur for a century or more. Nassim Taleb says, “Invest in preparedness, not in prediction.” That gets to the heart of it. Risk is dangerous when you think it requires a specific forecast before you start preparing for it. It’s better to have expectations that risk will arrive, though you don’t know when or where, than to rely exclusively on forecasts— almost all of which are either nonsense or about things that are well-known. Expectations and forecasts are two different things, and in a world where risk is what you don’t see, the former is more valuable than the latter.

Two, realize that if you’re only preparing for the risks you can envision, you’ll be unprepared for the risks you can’t see every single time. So, in personal finance, the right amount of savings is when it feels like it’s a little too much. It should feel excessive; it should make you wince a little. The same goes for how much debt you think you should handle—whatever you think it is, the reality is probably a little less. Your preparation shouldn’t make sense in a world where the biggest historical events all would have sounded absurd before they happened.

Geniuses are unique in BOTH good and bad ways

Something that’s built into the human condition is that people who think about the world in unique ways you like almost certainly also think about the world in unique ways you won’t like…

…John Maynard Keynes once purchased a trove of Isaac Newton’s original papers at auction. Many had never been seen before, as they had been stashed away at Cambridge for centuries. Newton is probably the smartest human to ever live. But Keynes was astonished to find that much of the work was devoted to alchemy, sorcery, and trying to find a potion for eternal life. Keynes wrote:

I have glanced through a great quantity of this at least 100,000 words, I should say. It is utterly impossible to deny that it is wholly magical and wholly devoid of scientific value; and also impossible not to admit that Newton devoted years of work to it.

I wonder: Was Newton a genius in spite of being addicted to magic, or was being curious about things that seemed impossible part of what made him so successful? I think it’s impossible to know. But the idea that crazy geniuses sometimes just look straight-up crazy is nearly unavoidable…

…Take Elon Musk. What kind of thirty-two-year-old thinks they can take on GM, Ford, and NASA at the same time? An utter maniac. The kind of person who thinks normal constraints don’t apply to them—not in an egotistical way, but in a genuine, believe-it-in-your-bones way. Which is also the kind of person who doesn’t worry about, say, Twitter etiquette.

A mindset that can dump a personal fortune into colonizing Mars is not the kind of mindset that worries about the downsides of hyperbole. And the kind of person who proposes making Mars habitable by constantly dropping nuclear bombs in its atmosphere is not the kind of person worried about overstepping the boundaries of reality.

The kind of person who says there’s a 99.9999 percent chance humanity is a computer simulation is not the kind of person worried about making untenable promises to shareholders. The kind of person who promises to solve the water problems in Flint, Michigan, within days of trying to save a Thai children’s soccer team stuck in a cave, within days of rebuilding the Tesla Model 3 assembly line in a tent, is not the kind of person who views his lawyers signing off as a critical step.

People love the visionary genius side of Musk, but want it to come without the side that operates in his distorted I-don’t-care-about-your-customs version of reality. But I don’t think those two things can be separated. They’re the risk-reward trade-offs of the same personality trait.

What gets you to the top also brings you down

What kind of person makes their way to the top of a successful company, or a big country? Someone who is determined, optimistic, doesn’t take no for an answer, and is relentlessly confident in their own abilities. What kind of person is likely to go overboard, bite off more than they can chew, and discount risks that are blindingly obvious to others? Someone who is determined, optimistic, doesn’t take no for an answer, and is relentlessly confident in their own abilities. Reversion to the mean is one of the most common stories in history. It’s the main character in economies, markets, countries, companies, careers—everything. Part of the reason it happens is because the same personality traits that push people to the top also increase the odds of pushing them over the edge.

Outrageous things can easily happen if the sample size is big enough

Evelyn Marie Adams won $3.9 million in the New Jersey lottery in 1986. Four months later she won again, collecting another $1.4 million. ‘‘I’m going to quit playing,’’ she told The New York Times. ‘‘I’m going to give everyone else a chance.’’ It was a big story at the time, because number crunchers put the odds of her double win at a staggering 1 in 17 trillion.

Three years later two mathematicians, Persi Diaconis and Frederick Mosteller, threw cold water on the excitement. If one person plays the lottery, the odds of picking the winning numbers twice are indeed 1 in 17 trillion. But if one hundred million people play the lottery week after week— which is the case in America—the odds that someone will win twice are actually quite good. Diaconis and Mosteller figured it was 1 in 30. That number didn’t make many headlines. ‘‘With a large enough sample, any outrageous thing is apt to happen,” Mosteller said

Why something bad happens nearly every year

If next year there’s a 1 percent chance of a new disastrous pandemic, a 1 percent chance of a crippling depression, a 1 percent chance of a catastrophic flood, a 1 percent chance of political collapse, and on and on, then the odds that something bad will happen next year—or any year—are . . . not bad.

The demise of local news, because of the internet, altered our perception on the frequency of bad news

The decline of local news has all kinds of implications. One that doesn’t get much attention is that the wider the news becomes the more likely it is to be pessimistic. Two things make that so: 

  • Bad news gets more attention than good news because pessimism is seductive and feels more urgent than optimism.
  • The odds of a bad news story—a fraud, a corruption, a disaster—occurring in your local town at any given moment is low. When you expand your attention nationally, the odds increase. When they expand globally, the odds of something terrible happening in any given moment are 100 percent.

To exaggerate only a little: Local news reports on softball tournaments. Global news reports on plane crashes and genocides. 

The internet’s existence means we’re more aware of bad things happening – but bad things are not necessarily happening more today

In modern times our horizons cover every nation, culture, political regime, and economy in the world. There are so many good things that come from that. But we shouldn’t be surprised that the world feels historically broken in recent years and will continue that way going forward. It’s not—we just see more of the bad stuff that’s always happened than we ever saw before.

A contemporary of Ben Graham seemed to know more about investing but was not as good a writer, so he is today much more obscure than Graham

Professor John Burr Williams had more profound insight on the topic of valuing stocks than Benjamin Graham. But Graham knew how to write a good paragraph, so he became the legend and sold millions of books.

US forces suffered against German forces during WWII because American leaders failed to account for Hitler going mad

Historian Stephen Ambrose notes that Eisenhower and General Omar Bradley got all the war-planning reasoning and logic right in late 1944, except for one detail—the extent to which Hitler had lost his mind. An aide to Bradley mentioned during the war: “If we were fighting reasonable people they would have surrendered long ago.” But they weren’t, and it—the one thing that was hard to measure with logic—mattered more than anything.

Lehman Brothers actually had strong financial ratios – better than even Goldman Sachs and Bank of America – in 2008 just before it went bankrupt; what went wrong for Lehman was that investors lost faith in the bank

A few examples of how powerful this can be: Lehman Brothers was in great shape on September 10, 2008. Its tier 1 capital ratio—a measure of a bank’s ability to endure loss—was 11.7 percent. That was higher than the previous quarter. Higher than Goldman Sachs. Higher than Bank of America. It was more capital than Lehman had in 2007, when the banking industry was about as strong as it had ever been. 

Seventy-two hours later Lehman was bankrupt. The only thing that changed during those three days was investors’ faith in the company. One day they believed in the company and bought its debt. The next day that belief stopped, and so did its funding. That faith is the only thing that mattered. But it was the one thing that was hard to quantify, hard to model, hard to predict, and didn’t compute in a traditional valuation model. GameStop

Hyman Minsky’s economic theory of stability leading to instability can be found in nature too

California was hit with an epic drought in the mid-2010s. Then 2017 came, dropping a preposterous amount of moisture. Parts of Lake Tahoe received—I’m not making this up—more than sixty-five feet of snow in a few months. The six-year drought was declared over.

You’d think that would be great. But it backfired in an unexpected way. Record rain in 2017 led to record vegetation growth that summer. It was called a superbloom, and it caused even desert towns to be covered in green. A dry 2018 meant all that vegetation died and became dry kindling. That led to some of the biggest wildfires California had ever seen.

So record rain directly led to record fires. There’s a long history of this, verified by looking at tree rings, which inscribe both heavy rainfall and subsequent fire scars. The two go hand in hand. “A wet year reduces fires while increasing vegetation growth, but then the increased vegetation dries out in subsequent dry years, thereby increasing the fire fuel,” the National Oceanic and Atmospheric Administration wrote. That’s hardly intuitive, but here again—calm plants the seeds of crazy. 

Why financial markets will always overshoot on both ends of the optimism and pessimism spectrum

The only way to know we’ve exhausted all potential opportunity from markets—the only way to identify the top —is to push them not only past the point where the numbers stop making sense, but beyond the stories people believe about those numbers. When a tire company develops a new tire and wants to know its limitations, the process is simple. They put it on a car and run it until it blows up. Markets, desperate to know the limits of what other investors can endure, do the same thing. Always been the case, always will be.

Markets going beyond the point of crazy is a normal thing 

One is accepting that crazy doesn’t mean broken. Crazy is normal; beyond the point of crazy is normal. Every few years there seems to be a declaration that markets don’t work anymore—that they’re all speculation or detached from fundamentals. But it’s always been that way. People haven’t lost their minds; they’re just searching for the boundaries of what other investors are willing to believe

Many things in life have a “most convenient size”

“For every type of animal there is a most convenient size, and a change in size inevitably carries with it a change of form,” Haldane wrote. A most convenient size. A proper state where things work well but break when you try to scale them to a different size or speed. It applies to so many things in life…

…Starbucks had 425 stores in 1994, its twenty-third year in existence. In 1999 it opened 625 new stores. By 2007 it was opening 2,500 stores per year—a new coffee shop every four hours. One thing led to another. The need to hit growth targets eventually elbowed out rational analysis. Examples of Starbucks saturation became a joke. Same-store sales growth fell by half as the rest of the economy boomed. 

Howard Schultz wrote to senior management in 2007: “In order to go from less than 1,000 stores to 13,000 stores we have had to make a series of decisions that, in retrospect, have led to the watering down of the Starbucks experience.” Starbucks closed six hundred stores in 2008 and laid off twelve thousand employees. Its stock fell 73 percent, which was dreadful even by 2008 standards.

Schultz wrote in his 2011 book Onward: “Growth, we now know all too well, is not a strategy. It is a tactic. And when undisciplined growth became a strategy, we lost our way.” There was a most convenient size for Starbucks—there is for all businesses. Push past it and you realize that revenue might scale but disappointed customers scale faster, in the same way Robert Wadlow became a giant but struggled to walk.

Different management skills are needed as a company changes in size

A management style that works brilliantly at a ten-person company can destroy a thousand-person company, which is a hard lesson to learn when some companies grow that fast in a few short years. Travis Kalanick, the former CEO of Uber, is a great example. No one but him was capable of growing the company early on, and anyone but him was needed as the company matured. I don’t think that’s a flaw, just a reflection that some things don’t scale. 

Militaries are really good at innovating because the problems they deal with are so important

Militaries are engines of innovation because they occasionally deal with problems so important—so urgent, so vital—that money and manpower are removed as obstacles, and those involved collaborate in ways that are hard to emulate during calm times. You cannot compare the incentives of Silicon Valley coders trying to get you to click on ads to Manhattan Project physicists trying to end a war that threatened the country’s existence. You can’t even compare their capabilities. The same people with the same intelligence have wildly different potential under different circumstances.

How the harsh conditions of the 1930s forced USA to innovate

The 1930s were a disaster, one of the darkest periods in American history. Almost a quarter of Americans were out of work in 1932. The stock market fell 89 percent. Those two economic stories dominate the decade’s attention, and they should. But there’s another story about the 1930s that rarely gets mentioned: it was, by far, the most productive and technologically progressive decade in U.S. history.

The number of problems people solved, and the ways they discovered how to build stuff more efficiently, is a forgotten story of the ’30s that helps explain a lot of why the rest of the twentieth century was so prosperous. Here are the numbers: total factor productivity—that’s economic output relative to the number of hours people worked and the amount of money invested in the economy—hit levels not seen before or since. Economist Alex Field wrote that by 1941 the U.S. economy was producing 40 percent more output than it had in 1929, with virtually no increase in the total number of hours worked. Everyone simply became staggeringly more productive.

A couple of things happened during this period that are worth paying attention to, because they explain why this happened when it did. Take cars. The 1920s was the era of the automobile. The number of cars on the road in America jumped from one million in 1912 to twenty-nine million by 1929. But roads were a different story. Cars were sold in the 1920s faster than roads were built. That changed in the 1930s when road construction, driven by the New Deal’s Public Works Administration, took off. Spending on road construction went from 2 percent of GDP in 1920 to over 6 percent in 1933 (versus less than 1 percent today). The Department of Highway Transportation tells a story of how quickly projects began: 

Construction began on August 5, 1933, in Utah on the first highway project under the act. By August 1934, 16,330 miles of new roadway projects were completed.

What this did to productivity is hard to overstate. The Pennsylvania Turnpike, as one example, cut travel times between Pittsburgh and Harrisburg by 70 percent. The Golden Gate Bridge, built in 1933, opened up Marin County, which had previously been accessible from San Francisco only by ferryboat. Multiply those kinds of leaps across the nation and the 1930s was the decade that transportation blossomed in the United States. It was the last link that made the century-old railroad network truly efficient, creating last-mile service that connected the world.

Electrification also surged in the 1930s, particularly to rural Americans left out of the urban electrification of the 1920s. The New Deal’s Rural Electrification Administration (REA) brought power to farms in what may have been the decade’s only positive development in regions that were economically devastated. The number of rural American homes with electricity rose from less than 10 percent in 1935 to nearly 50 percent by 1945. It is hard to fathom, but it was not long ago—during some of our lifetimes and most of our grandparents’—that a substantial portion of America was literally dark.

Franklin Roosevelt said in a speech on the REA:

Electricity is no longer a luxury. It is a definite necessity. . . . In our homes it serves not only for light, but it can become the willing servant of the family in countless ways. It can relieve the drudgery of the housewife and lift the great burden off the shoulders of the hardworking farmer.

Electricity becoming a “willing servant”—introducing washing machines, vacuum cleaners, and refrigerators—freed up hours of household labor in a way that let female workforce participation rise. It’s a trend that lasted more than half a century and is a key driver of both twentieth-century growth and gender equality.

Another productivity surge of the 1930s came from everyday people forced by necessity to find more bang for their buck. The first supermarket opened in 1930. The traditional way of purchasing food was to walk from your butcher, who served you from behind a counter, to the bakery, who served you from behind a counter, to a produce stand, who took your order. Combining everything under one roof and making customers pick it from the shelves themselves was a way to make the economics of selling food work during a time when a quarter of the nation was unemployed.

Laundromats were also invented in the 1930s after sales of individual washing machines fell; they marketed themselves as washing machine rentals.

Factories of all kinds looked at bludgeoned sales and said, “What must we do to survive?” The answer often was to build the kind of assembly line Henry Ford introduced to the world in the previous decade. Output per hour in factories had grown 21 percent during the 1920s. “During the Depression decade of 1930–1940— when many plants were shut down or working part time,” Frederick Lewis Allen wrote, “there was intense pressure for efficiency and economy—it had increased by an amazing 41 per cent.”

“The trauma of the Great Depression did not slow down the American invention machine,” economist Robert Gordon wrote. “If anything, the pace of innovation picked up.” Driving knowledge work in the ’30s was the fact that more young people stayed in school because they had nothing else to do. High school graduation surged during the Depression to levels not seen again until the 1960s.

All of this—the better factories, the new ideas, the educated workers— became vital in 1941 when America entered the war and became the Allied manufacturing engine. The big question is whether the technical leap of the 1930s could have happened without the devastation of the Depression. And I think the answer is no—at least not to the extent that it occurred. You could never push through something like the New Deal without an economy so wrecked that people were desperate to try anything to fix it.

Innovation takes time to be recognised, so it’s easy for people to think that innovation is lacking 

A lot of pessimism is fueled by the fact that it often looks like we haven’t innovated in years—but that’s usually because it takes years to notice a new innovation.

Economic progress has been incredible over long periods of time, but is unnoticeable over short periods

Real GDP per capita increased eightfold in the last hundred years. America of the 1920s had the same real per capita GDP as Turkmenistan does today. Our growth over the last century has been unbelievable. But GDP growth averages about 3 percent per year, which is easy to ignore in any given year, decade, or lifetime. Americans over age fifty have seen real GDP per person at least double since they were born. But people don’t remember the world when they were born. They remember the last few months, when progress is always invisible. Same for careers, social progress, brands, companies, and relationships. Progress always takes time, often too much time to even notice it’s happened.

Why progress happens slowly but bad news comes quickly 

Growth always fights against competition that slows its rise. New ideas fight for attention, business models fight incumbents, skyscrapers fight gravity. There’s always a headwind. But everyone gets out of the way of decline. Some might try to step in and slow the fall, but it doesn’t attract masses of outsiders who rush in to push back in the other direction the way progress does…

…The irony is that growth and progress are way more powerful than setbacks. But setbacks will always get more attention because of how fast they occur. So slow progress amid a drumbeat of bad news is the normal state of affairs. It’s not an easy thing to get used to, but it’ll always be with us. 

Good news is what did NOT happen whereas bad news is what did happen

A lot of progress and good news concerns things that didn’t happen, whereas virtually all bad news is about what did occur. Good news is the deaths that didn’t take place, the diseases you didn’t get, the wars that never happened, the tragedies avoided, and the injustices prevented. That’s hard for people to contextualize or even imagine, let alone measure. But bad news is visible. More than visible, it’s in your face. It’s the terrorist attack, the war, the car accident, the pandemic, the stock market crash, and the political battle you can’t look away from.

Why we underestimate big risks

Big risks are easy to overlook because they’re just a chain reaction of small events, each of which is easy to shrug off. So people always underestimate the odds of big risks…

…The Tenerife airport disaster in 1977 is the deadliest aircraft accident in history. The error was stunning. One plane took off while another was still on the runway, and the two Boeing 747s collided, killing 583 people on a runway on the Spanish island. In the aftermath authorities wondered how such an egregious catastrophe could occur. One postmortem study explained exactly how: “Eleven separate coincidences and mistakes, most of them minor . . . had to fall precisely into place” for the crash to occur. Lots of tiny mistakes added up to a huge one. It’s good to always assume the world will break about once per decade, because historically it has. The breakages feel like low-probability events, so it’s common to think they won’t keep happening. But they do, again and again, because they’re actually just smaller high-probability events compounding off one another. That isn’t intuitive, so we’ll discount big risks like we always have.

The fascinating history behind the phrase, “The American Dream”

“The American dream” was a phrase first used by author James Truslow Adams in his 1931 book The Epic of America. The timing is interesting, isn’t it? It’s hard to think of a year when the dream looked more broken than in 1931.

When Adams wrote that “a man by applying himself, by using the talents he has, by acquiring the necessary skills, can rise from lower to higher status, and that his family can rise with him,” the unemployment rate was nearly 25 percent and wealth inequality was near the highest it had been in American history.

When he wrote of “that American dream of a better, richer, and happier life for all our citizens of every rank,” food riots were breaking out across the country as the Great Depression ripped the economy to shreds.

When he wrote of “being able to grow to fullest development as men and women, unhampered by the barriers which had slowly been erected in older civilizations,” schools were segregated and some states required literacy tests to vote.

At few points in American history had the idea of the American dream looked so false, so out of touch with the reality everyone faced. Yet Adams’s book surged in popularity. An optimistic phrase born during a dark period in American history became an overnight household motto.

One quarter of Americans being out of work in 1931 didn’t ruin the idea of the American Dream. The stock market falling 89 percent—and bread lines across the country—didn’t, either. The American Dream actually may have gained popularity because things were so dire. You didn’t have to see the American Dream to believe in it—and thank goodness, because in 1931 there was nothing to see. You just had to believe it was possible and then, boom, you felt a little better.

In nature, species are never perfect in any one trait because perfection involves compromising in other areas

There is no perfect species, one adapted to everything at all times. The best any species can do is to be good at some things until the things it’s not good at suddenly matter more. And then it dies.

A century ago a Russian biologist named Ivan Schmalhausen described how this works. A species that evolves to become very good at one thing tends to become vulnerable at another. A bigger lion can kill more prey, but it’s also a larger target for hunters to shoot at. A taller tree captures more sunlight, but becomes vulnerable to wind damage. There is always some inefficiency. So species rarely evolve to become perfect at anything, because perfecting one skill comes at the expense of another skill that will eventually be critical to survival. The lion could be bigger and catch more prey; the tree could be taller and get more sun. But they’re not, because it would backfire. So they’re all a little imperfect. Nature’s answer is a lot of good enough, below-potential traits across all species.

Biologist Anthony Bradshaw says that evolution’s successes get all the attention, but its failures are equally important. And that’s how it should be: Not maximizing your potential is actually the sweet spot in a world where perfecting one skill compromises another.

The probability of a species going extinct is independent of its age

Leigh Van Valen was a crazy-looking evolutionary biologist who came up with a theory so wild no academic journal would publish it. So he created his own journal and published it, and the idea eventually became accepted wisdom. Those kinds of ideas—counterintuitive, but ultimately true—are the ones worth paying most attention to, because they’re easiest to overlook.

For decades, scientists assumed that the longer a species had been around, the more likely it was to stick around, because age proved a strength that was likely to endure. Longevity was seen as both a trophy and a forecast. In the early 1970s, Van Valen set out to prove that the conventional wisdom was right. But he couldn’t. The data just didn’t fit.

He began to wonder whether evolution was such a relentless and unforgiving force that long-lived species were just lucky. The data fit that theory better. You’d think a new species discovering its niche would be fragile and susceptible to extinction—let’s say a 10 percent chance of extinction in a given period—while an old species had proven its might, and has, say, a 0.01 percent chance of extinction.

But when Van Valen plotted extinctions by a species’ age, the trend looked more like a straight line. Some species survived a long time. But among groups of species, the probability of extinction was roughly the same whether it was 10,000 years old or 10 million years old.

In a 1973 paper titled “A New Evolutionary Law,” Van Valen wrote that “the probability of extinction of a taxon is effectively independent of its age.” If you take a thousand marbles and remove 2 percent of them each year, some marbles will remain in the jar after twenty years. But the odds of being picked out are the same every year (2 percent). Marbles don’t get better at staying in the jar. Species are the same. Some happen to live a long time, but the odds of surviving don’t improve over time. Van Valen argued that’s the case mainly because competition isn’t like a football game that ends with a winner who can then take a break. Competition never stops. A species that gains an advantage over a competitor instantly incentivizes the competitor to improve. It’s an arms race.

Evolution is the study of advantages. Van Valen’s idea is simply that there are no permanent advantages. Everyone is madly scrambling all the time, but no one gets so far ahead that they become extinction-proof.

An example of the unpredictable path of innovations: how planes made nuclear power plants possible

When the airplane came into practical use in the early 1900s, one of the first tasks was trying to foresee what benefits would come from it. A few obvious ones were mail delivery and sky racing. No one predicted nuclear power plants. But they wouldn’t have been possible without the plane. Without the plane we wouldn’t have had the aerial bomb. Without the aerial bomb we wouldn’t have had the nuclear bomb. And without the nuclear bomb we wouldn’t have discovered the peaceful use of nuclear power. Same thing today. Google Maps, TurboTax, and Instagram wouldn’t be possible without ARPANET, a 1960s Department of Defense project linking computers to manage Cold War secrets, which became the foundation for the internet. That’s how you go from the threat of nuclear war to filing your taxes from your couch—a link that was unthinkable fifty years ago, but there it is

The fascinating backstory behind the invention of Polaroid film

Author Safi Bahcall notes that Polaroid film was discovered when sick dogs that were fed quinine to treat parasites showed an unusual type of crystal in their urine. Those crystals turned out to be the best polarizers ever discovered. Who predicts that? Who sees that coming? Nobody. Absolutely nobody. 

The power of incentives can explain extreme events, unsustainable events occuring for prolonged periods of time, and warped beliefs

When good and honest people can be incentivized into crazy behavior, it’s easy to underestimate the odds of the world going off the rails. Everything from wars to recessions to frauds to business failures to market bubbles happen more often than people think because the moral boundaries of what people are willing to do can be extended with certain incentives. That goes both ways. It’s easy to underestimate how much good people can do, how talented they can become, and what they can accomplish when they operate in a world where their incentives are aligned toward progress.

Extremes are the norm. Unsustainable things can last longer than you anticipate. Incentives can keep crazy, unsustainable trends going longer than seems reasonable because there are social and financial reasons preventing people from accepting reality for as long as they can. A good question to ask is, “Which of my current views would change if my incentives were different?” If you answer “none,” you are likely not only persuaded but blinded by your incentives.

It’s hard to predict our behaviour during downturns because the environment changes so much

In investing, saying “I will be greedy when others are fearful” is easier said than done, because people underestimate how much their views and goals can change when markets break. The reason you may embrace ideas and goals you once thought unthinkable during a downturn is because more changes during downturns than just asset prices.

If I, today, imagine how I’d respond to stocks falling 30 percent, I picture a world where everything is like it is today except stock valuations, which are 30 percent cheaper. But that’s not how the world works. Downturns don’t happen in isolation. The reason stocks might fall 30 percent is because big groups of people, companies, and politicians screwed something up, and their screwups might sap my confidence in our ability to recover. So my investment priorities might shift from growth to preservation. It’s difficult to contextualize this mental shift when the economy is booming. And even though Warren Buffett says to be greedy when others are fearful, far more people agree with that quote than actually act on it. The same idea holds true for companies, careers, and relationships. Hard times make people do and think things they’d never imagine when things are calm.

Why humans prefer complexity over simplicity

The question then is: Why? Why are complexity and length so appealing when simplicity and brevity will do? A few reasons: 

Complexity gives a comforting impression of control, while simplicity is hard to distinguish from cluelessness. 

In most fields a handful of variables dictate the majority of outcomes. But paying attention to only those few variables can feel like you’re leaving too much of the outcome to fate. The more knobs you can fiddle with—the hundred-tab spreadsheet, or the Big Data analysis—the more control you feel you have over the situation, if only because the impression of knowledge increases. The flip side is that paying attention to only a few variables while ignoring the majority of others can make you look ignorant. If a client says, “What about this, what’s happening here?” and you respond, “Oh, I have no idea, I don’t even look at that,” the odds that you’ll sound uninformed are greater than the odds you’ll sound like you’ve mastered simplicity.

Things you don’t understand create a mystique around people who do. 

If you say something I didn’t know but can understand, I might think you’re smart. If you say something I can’t understand, I might think you have an ability to think about a topic in ways I can’t, which is a whole different species of admiration. When you understand things I don’t, I have a hard time judging the limits of your knowledge in that field, which makes me more prone to taking your views at face value.

Length is often the only thing that can signal effort and thoughtfulness. 

A typical nonfiction book covering a single topic is perhaps 250 pages, or something like 65,000 words. The funny thing is the average reader does not come close to finishing most books they buy. Even among bestsellers, average readers quit after a few dozen pages. Length, then, has to serve a purpose other than providing more material.

My theory is that length indicates the author has spent more time thinking about a topic than you have, which can be the only data point signaling they might have insights you don’t. It doesn’t mean their thinking is right. And you may understand their point after two chapters. But the purpose of chapters 3–16 is often to show that the author has done so much work that chapters 1 and 2 might have some insight. Same goes for research reports and white papers.

Simplicity feels like an easy walk. Complexity feels like a mental marathon.

If the reps don’t hurt when you’re exercising, you’re not really exercising. Pain is the sign of progress that tells you you’re paying the unavoidable cost of admission. Short and simple communication is different. Richard Feynman and Stephen Hawking could teach math with simple language that didn’t hurt your head, not because they dumbed down the topics but because they knew how to get from A to Z in as few steps as possible. An effective rule of thumb doesn’t bypass complexity; it wraps things you don’t understand into things you do. Like a baseball player who—by keeping a ball level in his gaze—knows where the ball will land as well as a physicist calculating the ball’s flight with precision.

The problem with simplicity is that the reps don’t hurt, so you don’t feel like you’re getting a mental workout. It can create a preference for laborious learning that students are actually okay with because it feels like a cognitive bench press, with all the assumed benefits.

Why people will always disagree

The question “Why don’t you agree with me?” can have infinite answers. Sometimes one side is selfish, or stupid, or blind, or uninformed. But usually a better question is, “What have you experienced that I haven’t that makes you believe what you do? And would I think about the world like you do if I experienced what you have?”

It’s the question that contains the most answers about why people don’t agree with one another. But it’s such a hard question to ask. It’s uncomfortable to think that what you haven’t experienced might change what you believe, because it’s admitting your own ignorance. It’s much easier to assume that those who disagree with you aren’t thinking as hard as you are.

So people will disagree, even as access to information explodes. They may disagree more than ever because, as Benedict Evans says, “The more the Internet exposes people to new points of view, the angrier people get that different views exist.” Disagreement has less to do with what people know and more to do with what they’ve experienced. And since experiences will always be different, disagreement will be constant. Same as it’s ever been. Same as it will always be. Same as it ever was


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. I currently have a vested interest in Amazon. Holdings are subject to change at any time.

Lessons From The Immortal Charlie Munger

Neuroscientist David Eagleman once wrote: “There are three deaths: the first is when the body ceases to function. The second is when the body is consigned to the grave. The third is that moment, sometime in the future, when your name is spoken for the last time.”

Along Eagleman’s line of reasoning, Charlie Munger, who passed away peacefully last night, would be immortal since he would never experience the third death – his accomplishments, and the wisdom he has shared throughout his life, would see to it. 

Munger is one of my investing heroes. In remembrance of his life, I would like to share my favourite lessons from him.

On the importance of thinking in reverse, or inverting

“Another idea that I discovered was encapsulated by that story Dean McCaffery recounted earlier about the rustic who wanted to know where he was going to die, so he wouldn’t go there. The rustic who had that ridiculous sounding idea had a profound truth in his possession. The way complex adaptive systems work, and the way mental constructs work, problems frequently become easier to solve through inversion. If you turn problems around into reverse, you often think better. For instance, if you want to help India, the question you should consider asking is not: How can I help India? Instead, you should ask: How can I hurt India? You find what will do the worst damage, and then try to avoid it. Perhaps the two approaches seem logically the same thing. But those who have mastered algebra know that inversion will often and easily solve problems that otherwise resist solution. And in life, just as in algebra, inversion will help you solve problems that you can’t otherwise handle.”

On the importance of being equanimous when investing

“If you’re not willing to react with equanimity to a market price decline of 50% two or three times a century you’re not fit to be a common shareholder and you deserve the mediocre result you’re going to get compared to people who do have the temperament, who can be more philosophical about these market fluctuations.”

On the importance of incentives

“From all business, my favourite case on incentives is Federal Express. The heart and soul of their system – which creates the integrity of the product – is having all their airplanes come to one place in the middle of the night and shift all the packages from plane to plane. If there are delays, the whole operation can’t deliver a product full of integrity to Federal Express customers. And it was always screwed up. They could never get it done on time. They tried everything – moral suasion, threats, you name it. And nothing worked. Finally, somebody got the idea to pay all these people not so much an hour, but so much a shift – and when it’s all done, they can go home. Well, their problems cleared up overnight.”

On great career advice

“Three rules for a career: (1) Don’t sell anything you wouldn’t buy yourself; (2) Don’t work for anyone you don’t respect and admire; and (3) Work only with people you enjoy.”

On the importance of admitting mistakes

“There’s no way that you can live an adequate life without many mistakes. In fact, one trick in life is to get so you can handle mistakes. Failure to handle psychological denial is a common way for people to go broke.”

On the importance of not letting rare events completely shape how you approach life

“Ben Graham had a lot to learn as an investor. His ideas of how to value companies were all shaped by how the Great Crash and the Depression almost destroyed him… It left him with an aftermath of fear for the rest of his life, and all his methods were designed to keep that at bay.”

On the importance of handling problems from many different angles

“Most people are trained in one model – economics, for example – and try to solve all problems in one way. You know the saying: “To the man with a hammer, the world looks like a nail.” This is a dumb way of handling problems.”

On the importance of getting a little wiser each day

“I constantly see people rise in life who are not the smartest, sometimes not even the most diligent, but they are learning machines. They go to bed every night a little wiser than they were when they got up, and boy, does that help, particularly when you have a long run ahead of you.”

On how to invest

Over the long term, it’s hard for a stock to earn a much better return than the business which underlies it earns. If the business earns 6% on capital over 40 years and you hold it for that 40 years, you’re not going to make much different than a 6% return—even if you originally buy it at a huge discount. Conversely, if a business earns 18% on capital over 20 or 30 years, even if you pay an expensive looking price, you’ll end up with a fine result. So the trick is getting into better businesses. And that involves all of these advantages of scale that you could consider momentum effects.”

On how to get others to agree with you

“Well, you’ll end up agreeing with me because you’re smart and I’m right.”

On the secret to a happy life

“I always say the same thing: realistic expectations, which is low expectations. If you have unreasonable demands on life, you’re like a bird that’s trying to destroy himself by bashing his wings on the edge of the cage. And you really can’t get out of the cage. It’s stupid. You want to have reasonable expectations and take life’s results good and bad as they happen with a certain amount of stoicism.”

On courage and perseverance

I saved the most poignant lesson I’ve learned from Munger for the last. Not many may know this, but the first decade-plus of Munger’s adulthood was tragic. 

Munger got married when he was 21, but the marriage ended when he was 29. He “lost everything in the divorce”, according to his daughter Molly Munger. Shortly after the divorce, Munger’s son, Teddy Munger, was diagnosed with leukaemia. “In those days, there was no medical insurance – I just paid all the expenses” Munger once said. But more importantly, there was absolutely nothing doctors back then could do for leukaemia. When Munger was 31, Teddy passed on. Munger recounted the heart-wrenching episode: “I can’t imagine any experience in life worse than losing a child inch by inch. By the time he died, my weight was down 10 to 15 pounds from normal.” One of Munger’s friends, Rick Guerin, said that “when his [Munger’s] son was in the bed and slowly dying, he’d go in and hold him for awhile, then go out walking the streets of Pasadena crying.”

So by the time Munger was 31, he had already gone through a divorce, experienced the painful death of his son from an incurable disease, and was broke. 

But when Munger left the world last night, he was a billionaire, and was widely revered around the world for his wit, wisdom, and character. He taught me that with courage and perseverance, we can eventually build a better life for ourselves. “You should never, when faced with one unbelievable tragedy, let one tragedy increase into two or three because of a failure of will,” he admonished. 

See you on the other side, Mr Munger.

Thoughts on Artificial Intelligence

Artificial intelligence has the potential to reshape the world.

The way Jeremy and I see it, artificial intelligence (AI) really leapt into the zeitgeist in late-2022 or early-2023 with the public introduction of DALL-E2 and ChatGPT. Both are provided by OpenAI and are known as generative AI products – they are software that use AI to generate art and text, respectively (and often at astounding quality), hence the term “generative”. Since then, developments in AI have progressed at a breathtaking pace. One striking observation I’ve found with AI is the much higher level of enthusiasm that company-leaders have for the technology compared to the two other recent “hot things”, namely, blockchain/cryptocurrencies and the metaverse. Put another way, AI could be a real game changer for societies and economies.

I thought it would be useful to write down some of my current thoughts on AI and its potential impact. Putting pen to paper (or fingers to the keyboard) helps me make sense of what’s in my mind. Do note that my thoughts are fragile because the field of AI is developing rapidly and there are many unknowns at the moment. In no order of merit:

  • While companies such as OpenAI and Alphabet have released generative AI products, they have yet to release open-source versions of their foundational AI models that power the products. Meta Platforms, meanwhile, has been open sourcing its foundational AI models in earnest. During Meta’s latest earnings conference call in April this year, management explained that open sourcing allows Meta to benefit from improvements to its foundational models that are made by software developers, outside of Meta, all over the world. Around the same time, there was a purportedly leaked document from an Alphabet employee that discussed the advantages in the development of AI that Meta has over both Alphabet and OpenAI by virtue of it open sourcing its foundational models. There’s a tug-of-war now between what’s better – proprietary or open-sourced foundational AI models – but it remains to be seen which will prevail or if there will even be a clear winner. 
  • During Amazon’s latest earnings conference call (in April 2023), the company’s management team shared their observation that most companies that want to utilise AI have no interest in building their own foundational AI models because it takes tremendous amounts of time and capital. Instead, they merely want to customise foundational models with their own proprietary data. On the other hand, Tencent’s leaders commented in the company’s May 2023 earnings conference call that they see a proliferation of foundational AI models from both established companies as well as startups. I’m watching to find out which point of view is closer to the truth. I also want to point out that the frenzy to develop foundational AI models may be specific to China. Rui Ma, an astute observer of and writer on China’s technology sector, mentioned in a recent tweet that “everyone in China is building their own foundational model.” Meanwhile, the management of online travel platform Airbnb (which is based in the US, works deeply with technology, and is clearly a large company) shared in May 2023 that they have no interest in building foundational AI models – they’re only interested in designing the interface and tuning the models. 
  • A database is a platform to store data. Each piece of software requires a database to store, organize, and process data. The database has a direct impact on the software’s performance, scalability, flexibility, and reliability, so its selection is a highly strategic decision for companies. In the 1970s, relational databases were first developed and they used a programming language known as Structured Query Language (SQL). Relational databases store and organise data points that are related to one another in table form (picture an Excel spreadsheet) and were useful from the 1980s to the late 1990s. But because they were used to store structured data, they began to lose relevance with the rise of the internet. Relational databases were too rigid for the internet era and were not built to support the volume, velocity, and variety of data in the internet era. This is where non-relational databases – also known as NoSQL, which stands for either “non SQL” or “not only SQL” – come into play. NoSQL databases are not constrained to relational databases’ tabular format of data storage and can work with unstructured data such as audio, video, and photos. As a result, they are more flexible and better suited for the internet age. AI appears to require different database architectures. The management of MongoDB, a company that specialises in NoSQL databases, talked about the need for a vector database to store the training results of large language models during the company’s June 2023 earnings conference call. Simply put, a vector database stores data in a way that allows users to easily find data, say, an image (or text), that is related to a given image (or text) – this feature is very useful for generative AI products. This said, MongoDB’s management also commented in the same earnings conference call that NoSQL databases will still be very useful in the AI era. I’m aware that MongoDB’s management could be biased, but I do agree with their point of view. Vector databases appear to be well-suited (to my untrained technical eye!) for a narrow AI-related use case, whereas NoSQL databases are useful in much broader ways. Moreover, AI is likely to increase the volume of software developed for all kinds of software – not just AI software – and they need modern databases. MongoDB’s management also explained in a separate June 2023 conference that a typical generative AI workflow will include both vector databases and other kinds of databases (during the conference, management also revealed MongoDB’s own vector database service). I’m keeping a keen eye on how the landscape of database architectures evolve over time as AI technologies develop.
  • Keeping up with the theme of new architectures, the AI age could also usher in a new architecture for data centres. This new architecture is named accelerated computing by Nvidia. In the traditional architecture of data centres, CPUs (central processing units) are the main source of computing power. In accelerated computing, the entire data centre – consisting of GPUs (graphic processing units), CPUs, DPUs (data processing units), data switches, networking hardware, and more – provides the computing power. Put another way, instead of thinking about the chip as the computer, the data centre becomes the computer under the accelerated computing framework. During Nvidia’s May 2023 earnings conference call, management shared that the company had been working on accelerated computing for many years but it was the introduction of generative AI – with its massive computing requirements – that “triggered a killer app” for this new data centre architecture. The economic opportunity could be immense. Nvidia’s management estimated that US$1 trillion of data centre infrastructure was installed over the last four years and nearly all of it was based on the traditional CPU-focused architecture. But as generative AI gains importance in society, data centre infrastructure would need to shift heavily towards the accelerated computing variety, according to Nvidia’s management.
  • And keeping with the theme of something new, AI could also bring about novel and better consumer experiences. Airbnb’s co-founder and CEO, Brian Chesky, laid out a tantalising view on this potential future during the company’s latest May 2023 earnings conference call. Chesky mentioned that search queries in the travel context are matching questions and the answers depend on who the questioner is and what his/her preferences are. With the help of AI, Airbnb could build “the ultimate AI concierge that could understand you,” thereby providing a highly personalised travel experience. Meanwhile, in a recent interview with Wired, Microsoft’s CEO Satya Nadella shared his dream that “every one of Earth’s 8 billion people can have an AI tutor, an AI doctor, a programmer, maybe a consultant!” 
  • Embedded AI is the concept of AI software that is built into a device itself. This device can be a robot. And if robots with embedded AI can be mass-produced, the economic implications could be tremendous, beyond the impact that AI could have as just software. Tesla is perhaps the most high profile company in the world today that is developing robots with embedded AI. The company’s goal for the Tesla Bot (also known as Optimus) is for it to be “a general purpose, bi-pedal, autonomous humanoid robot capable of performing unsafe, repetitive or boring tasks.” There are other important companies that are working on embedded AI. For example, earlier this year, Nvidia acquired OmniML, a startup whose software shrinks AI models, making it easier for the models to be run on devices rather than on the cloud.
  • Currently, humans are behind the content trained on by foundational AI models underpinning the likes of ChatGPT and other generative AI products. But according to a recently-published paper from UK and Canadian researchers titled The Curse of Recursion: Training on Generated Data Makes Models Forget, the quality of foundational AI models degrades significantly as the proportion of content they are trained on shifts toward an AI-generated corpus. This could be a serious problem in the future if there’s an explosion in the volume of generative AI content, which seems likely; for context, Adobe’s management shared in mid-June this year that the company’s generative AI feature, Firefly, had already powered 500 million content-generations since its launch in March 2023. The degradation, termed “model collapse” by the researchers, happens because content created by humans are a more accurate reflection of the world since they would contain improbable data. Even after training on man-made data, AI models tend to generate content that understates the improbable data. If subsequent AI models train primarily on AI-generated content, the end result is that the improbable data become even less represented. The researchers describe model collapse as “a degenerative process whereby, over time, models forget the true underlying data distribution, even in the absence of a shift in the distribution over time.” Model collapse could have serious societal consequences; one of the researchers, Ilia Shumailov, told Venture Beat that “there are many other aspects that will lead to more serious implications, such as discrimination based on gender, ethnicity or other sensitive attributes.” Ross Anderson, another author of the paper, wrote in a blog post that with model collapse, advantages could accrue to companies that “control access to human interfaces at scale” or that have already trained AI models by scraping the web when human-generated content was still overwhelmingly dominant. 

There’s one other fragile thought I have about AI that we think is more important than what I’ve shared above, and it is related to the concept of emergence. Emergence is a natural phenomenon where sophisticated outcomes spontaneously “emerge” from the interactions of agents in a system, even when these agents were not instructed to produce these outcomes. The following passages from the book, Complexity: The Emerging Science at the Edge of Order and Chaos by Mitch Waldrop, help shed some light on emergence:

“These agents might be molecules or neurons or species or consumers or even corporations. But whatever their nature, the agents were constantly organizing and reorganizing themselves into larger structures through the clash of mutual accommodation and mutual rivalry. Thus, molecules would form cells, neurons would form brains, species would form ecosystems, consumers and corporations would form economies, and so on. At each level, new emergent structures would form and engage in new emergent behaviors. Complexity, in other words, was really a science of emergence… 

…Cells make tissues, tissues make organs, organs make organisms, organisms make ecosystems – on and on. Indeed, thought Holland, that’s what this business of “emergence” was all about: building blocks at one level combining into new building blocks at a higher level. It seemed to be one of the fundamental organizing principles of the world. It certainly seemed to appear in every complex, adaptive system that you looked at…

…Arthur was fascinated by the thing. Reynolds had billed the program as an attempt to capture the essence of flocking behavior in birds, or herding behavior in sheep, or schooling behavior in fish. And as far as Arthur could tell, he had succeeded beautifully. Reynolds’ basic idea was to place a large collection of autonomous, birdlike agents—“boids”—into an onscreen environment full of walls and obstacles. Each boid followed three simple rules of behavior: 

1. It tried to maintain a minimum distance from other objects in the environment, including other boids.

2. It tried to match velocities with boids in its neighborhood.

3. It tried to move toward the perceived center of mass of boids in its neighborhood.

What was striking about these rules was that none of them said, “Form a flock.” Quite the opposite: the rules were entirely local, referring only to what an individual boid could see and do in its own vicinity. If a flock was going to form at all, it would have to do so from the bottom up, as an emergent phenomenon. And yet flocks did form, every time. Reynolds could start his simulation with boids scattered around the computer screen completely at random, and they would spontaneously collect themselves into a flock that could fly around obstacles in a very fluid and natural manner. Sometimes the flock would even break into subflocks that flowed around both sides of an obstacle, rejoining on the other side as if the boids had planned it all along. In one of the runs, in fact, a boid accidentally hit a pole, fluttered around for a moment as though stunned and lost—then darted forward to rejoin the flock as it moved on.”

In our view, the concept of emergence is important in AI because at least some of the capabilities of ChatGPT seen today were not explicitly programmed for – they emerged. Satya Nadella said in his aforementioned interview with Wired that “when we went from GPT 2.5 to 3, we all started seeing these emergent capabilities.” Nadella was referring to the foundational AI models built by OpenAI in his Wired interview. One of the key differences between GPT 2.5 and GPT 3 is that the former contains 1.5 billion parameters, whereas the latter contains 175 billion, more than 100 times more. The basic computational unit within an AI model is known as a node, and parameters are a measure of the strength of a connection between two nodes. The number of parameters can thus be loosely associated with the number of nodes, as well as the number of connections between nodes, in an AI model. With GPT 3’s much higher number of parameters compared to GPT 2.5, the number of nodes and number of connections (or interactions) between nodes in GPT 3 thus far outweigh those of GPT 2.5. Nadella’s observation matches those of David Ha, an expert on AI whose most recent role was the head of research at Stability AI. During a February 2023 podcast hosted by investor Jim O’Shaughnessy, Ha shared the following (emphasis is mine):

Then the interesting thing is, sure, you can train things on prediction or even things like translation. If you have paired English to French samples, you can do that. But what if you train a model to predict itself without any labels? So that’s really interesting because one of the limitations we have is labeling data is a daunting task and it requires a lot of thought, but self-labeling is free. Like anything on the internet, the label is itself, right? So what you can do is there’s two broad types of models that are popular now. There’s language models that generate sequences of data and there’s things like image models, Stable Diffusion you generate an image. These operate on a very similar principle, but for things like language model, you can have a large corpus of text on the internet. And the interesting thing here is all you need to do is train the model to simply predict what the next character is going to be or what the next word is going to be, predict the probability distribution of the next word.

And such a very simple objective as you scale the model, as you scale the size and the number of neurons, you get interesting emerging capabilities as well. So before, maybe back in 2015, ’16, when I was playing around with language models, you can feed it, auto Shakespeare, and it will blab out something that sounds like Shakespeare.

But in the next few years, once people scaled up the number of parameters from 5 million, to a hundred million, to a billion parameters, to a hundred billion parameters, this simple objective, you can now interact with the model. You can actually feed in, “This is what I’m going to say,” and the model takes that as an input as if it said that and predict the next character and give you some feedback on that. And I think this is very interesting, because this is an emergent phenomenon. We didn’t design the model to have these chat functions. It’s just like this capability has emerged from scale.

And the same for image side as well. I think for images, there are data sets that will map the description of that image to that image itself and text to image models can do things like go from a text input into some representation of that text input and its objective is to generate an image that encapsulates what the text prompt is. And once we have enough images, I remember when I started, everyone was just generating tiny images of 10 classes of cats, dogs, airplanes, cars, digits and so on. And they’re not very general. You can only generate so much.

But once you have a large enough data distribution, you can start generating novel things like for example, a Formula 1 race car that looks like a strawberry and it’ll do that. This understanding of concepts are emergent. So I think that’s what I want to get at. You start off with very simple statistical models, but as you increase the scale of the model and you keep the objectives quite simple, you get these emergent capabilities that were not planned but simply emerge from training on that objective.

Emergence occurred in AI models as their number of parameters (i.e. the number of interactions between nodes) grew. This is a crucial point because emergence requires a certain amount of complexity in the interactions between agents, which can only happen if there are large numbers of agents as well as interactions between agents. It’s highly likely, in my view, that more emergent phenomena could develop as AI models become even more powerful over time via an increase in their parameters. It’s also difficult – perhaps impossible – to predict what these emergent phenomena could be, as specific emergent phenomena in any particular complex system are inherently unpredictable. So, any new emergent phenomena from AI that springs up in the future could be anywhere on the spectrum of being wildly positive to destructive for society. Let’s see!


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

How Warren Buffett Analysed Lehman Brothers

Warren Buffett rejected Lehman Brothers’ request for financing help during the Great Financial Crisis – here’s how he analysed Lehman back then

A friend of Jeremy and I recently shared a tweet with us from Andrew Kuhn, a partner at the investment firm Focused Compounding, that I found fascinating. It mentioned an interview Warren Buffett did with the Wall Street Journal in 2018 where he explained his reasons for rejecting Lehman Brothers’ plea for financing during the 2008-09 Great Financial Crisis.

Prior to its bankruptcy in September 2008, Lehman Brothers was a storied investment bank that was founded more than a century ago in 1847. Lehman first approached Buffett for help in March 2008. After studying Lehman’s then-latest 10-K – for the financial year ended 30 November 2007 (the 10-K is the annual report that US-listed companies have to file) – Buffett discovered multiple red flags and decided to turn the bank down.

During his interview with the Wall Street Journal, Buffett showed a physical copy of Lehman’s 10-K that he read and made notes on. I managed to find a copy of Buffett’s 10-K and thought it would be an interesting exercise to run through all the pages he marked out as red flags to understand how he analysed Lehman Brothers in 2008.

Before I continue, here are some important things to note:

  • What I’m about to share are merely my interpretations and I make no claim that they accurately portray Buffett’s actual thought process.
  • This is the most complex set of financial statements that I’ve seen since I started investing in 2010 so I might be getting some of the details wrong (it’s also a great reminder for me to proceed with extreme caution when investing in banks!).
  • Kuhn created a video with his colleague, Geoff Gannon, that featured their analysis of Buffett’s copy of Lehman’s 10-K. I watched it while reading the document and it was really helpful for my own understanding of the red flags that Buffett noted.

Buffett’s mark ups: Page 106 & 107

These pages contain Figure 1 below, which shows the high yield bonds held by Lehman in FY2007 and FY2006.

Figure 2; Source: Buffett’s Lehman 10-K

Three things stood out to me: 

  • The high yield bond positions increased significantly by 137% from US$12.8 billion in FY2006 to US$30.4 billion in FY2007. 
  • The increase was a result of Lehman being unable to offload these positions, an indication that perhaps these assets were of poor quality. Per Lehman’s 10-K (emphasis is mine): “The increase in high-yield positions from 2006 to 2007 is primarily from funded lending commitments that have not been syndicated.”
  • The high yield bond positions need to be seen in relation to Lehman’s shareholder’s equity of merely US$22.5 billion in FY2007. If these high yield positions – US$30.4 billion – were to decline sharply in value, Lehman’s shareholder’s equity, and thus financial health, would be in serious trouble.

Pages 106 and 107 also mentioned that Lehman was authorised to buy back up to 100 million shares of itself “for the management of our equity capital, including offsetting dilution due to employee stock awards.” I’m guessing this did not sit well with Buffett from a capital allocation perspective because buying back shares merely to offset dilution is not an intelligent nor prudent use of capital.

Buffett’s mark ups: Page 115

This page is linked to the following passages (empahses are mine): 

We enter into various transactions with special purpose entities (“SPEs”). SPEs may be corporations, trusts or partnerships that are established for a limited purpose. There are two types of SPEs— QSPEs and VIEs.

A QSPE generally can be described as an entity whose permitted activities are limited to passively holding financial assets and distributing cash flows to investors based on pre-set terms. Our primary involvement with QSPEs relates to securitization transactions in which transferred assets, including mortgages, loans, receivables and other financial assets, are sold to an SPE that qualifies as a QSPE under SFAS 140. In accordance with SFAS 140 and FIN-46(R), we do not consolidate QSPEs. We recognize at fair value the interests we hold in the QSPEs. We derecognize financial assets transferred to QSPEs, provided we have surrendered control over the assets.”

What these passages effectively mean is that Lehman had off-balance sheet entities (the QSPEs) that housed certain assets so that they would not show up on Lehman’s own balance sheet. But it was exceedingly difficult to know (1) the value of these assets, (2) what these assets were, and (3) Lehman’s liabilities that were associated with these assets. Buffett might have been worried about the damage these unknowns could wrought on Lehman if trouble manifested in them.

Buffett’s mark ups: Page 125

This page is linked to the following passages (emphases are mine):

Derivatives are exchange traded or privately negotiated contracts that derive their value from an underlying asset. Derivatives are useful for risk management because the fair values or cash flows of derivatives can be used to offset the changes in fair values or cash flows of other financial instruments. In addition to risk management, we enter into derivative transactions for purposes of client transactions or establishing trading positions. The presentation of derivatives in our Consolidated Statement of Financial Position is net of payments and receipts and, in instances where management determines a legal right of offset exists as a result of a netting agreement, net-by-counterparty. Risk for an OTC derivative includes credit risk associated with the counterparty in the negotiated contract and continues for the duration of that contract.

The fair value of our OTC derivative assets at November 30, 2007 and 2006, was $41.3 billion and $19.5 billion, respectively; however, we view our net credit exposure to have been $34.6 billion and $15.6 billion at November 30, 2007 and 2006, respectively, representing the fair value of OTC derivative contracts in a net receivable position after consideration of collateral.”

Lehman had OTC (over-the-counter) derivative assets of US$41.3 billion in FY2007. These assets were problematic because (1) it’s hard to tell what’s in them and thus if Lehman had any counterparty risk, (2) it’s hard to tell what their actual values were since they were traded over-the-counter, and (3) they had more than doubled in value from FY2006 to FY2007. Moreover, Lehman’s shareholder’s equity in FY2007 was just US$22.5 billion, as mentioned earlier. This meant the investment bank did not have much cushion to absorb any significant declines in the value of its OTC derivative assets if they were to occur. 

Buffett’s mark ups: Page 173 & 175

These pages are linked to Figure 2, which shows all the financial instruments and inventory owned by Lehman in FY2007 and FY2006:

Figure 2; Source: Buffett’s Lehman 10-K

I think what troubled Buffett here would be the owned derivatives and other contractual agreements of US$44.6 billion in FY2007. The number was double that of FY2006 and as Figure 3 below illustrates, all of these assets were traded over-the-counter and thus had values that could not be easily determined. Let’s not forget too, that Lehman’s shareholder’s equity – US$22.5 billion in FY2007 – would provide only a thin buffer if any large decline in value for the owned derivatives and other contractual agreements happened. 

Figure 3; Source: Buffett’s Lehman 10-K

Buffett’s mark ups: Page 180

This page is linked to a description of the way Lehman groups its assets based on how their values are derived. Per the 10-K (emphases are mine):

“Level I – Inputs are unadjusted, quoted prices in active markets for identical assets or liabilities at the measurement date. The types of assets and liabilities carried at Level I fair value generally are G-7 government and agency securities, equities listed in active markets, investments in publicly traded mutual funds with quoted market prices and listed derivatives.

Level II – Inputs (other than quoted prices included in Level I) are either directly or indirectly observable for the asset or liability through correlation with market data at the measurement date and for the duration of the instrument’s anticipated life. Fair valued assets and liabilities that are generally included in this category are non-G-7 government securities, municipal bonds, certain hybrid financial instruments, certain mortgage and asset backed securities, certain corporate debt, certain commitments and guarantees, certain private equity investments and certain derivatives.

Level III – Inputs reflect management’s best estimate of what market participants would use in pricing the asset or liability at the measurement date. Consideration is given to the risk inherent in the valuation technique and the risk inherent in the inputs to the model. Generally, assets and liabilities carried at fair value and included in this category are certain mortgage and asset-backed securities, certain corporate debt, certain private equity investments, certain commitments and guarantees and certain derivatives.”

Put simply, Lehman had three types of assets: Level I assets had values that were determined simply by publicly-available prices while Level II and Level III assets had values that were determined using management’s inputs. Page 180 is also linked to Figure 4 below:

Figure 4; Source: Buffett’s Lehman 10-K

What Figure 4 shows is that one of Lehman’s single-largest asset categories – mortgage and asset-backed securities – were nearly all Level II and Level III assets. They are thus assets whose prices were not easily determinable by third-parties at that point in time. And their collective value was US$89.1 billion, four times higher than Lehman’s shareholder equity of US$22.5 billion. Buffett might have been worried that Lehman would be wiped out if these assets were to fall by just 25% in value – a distinct possibility given that the US housing market was already shaky back then.

Another aspect of Lehman’s financials linked to Page 180 of its 10-K that might have troubled Buffett is shown in Figure 5: Lehman’s Level III mortgage and asset-backed positions had surged threefold from just US$8.6 billion in FY2006 to US$25.2 billion in FY2007. 

Figure 5; Source: Buffett’s Lehman 10-K

Buffett’s mark ups: Page 184

This page is linked to the following paragraphs (emphases are mine):

“The Company uses fair value measurements on a nonrecurring basis in its assessment of assets classified as Goodwill and other inventory positions classified as Real estate held for sale. These assets and inventory positions are recorded at fair value initially and assessed for impairment periodically thereafter. During the fiscal year ended November 30, 2007, the carrying amount of Goodwill assets were compared to their fair value. No change in carrying amount resulted in accordance with the provisions of SFAS No. 142, Goodwill and Other Intangible Assets

Additionally and on a nonrecurring basis during the fiscal year ended November 30, 2007, the carrying amount of Real estate held for sale positions were compared to their fair value less cost to sell. No change in carrying amount resulted in accordance with the provisions of SFAS No. 66, Accounting for Sales of Real Estate, SFAS No. 144, Accounting for Impairment or Disposal of Long Lived Assets, and other relevant accounting guidance. The lowest level of inputs for fair value measurements for Goodwill and Real estate held for sale are Level III.

It turns out that Lehman’s real estate held for sale of US$21.9 billion in FY2007 – first shown in “Buffett’s mark ups: Page 173 & 175” – could have been Level III assets. So the stated value of US$21.9 billion may not have been anywhere close to what these assets could fetch in an open transaction, since the US housing market was already in trouble at that point in time.

Final word

Buffett’s marked-up copy of Lehman’s 10-K contained more pages that he noted down as red flags, such as pages 188, 199, 209, and more. But when I read them, there was nothing that jumped out at me as being highly unusual so I’ve not included them in this article.

Again, everything I’ve shared earlier are merely my interpretations and I make no claim that they accurately portray Buffett’s actual thought process when he studied Lehman’s 10-K. Nonetheless, I found it to be an interesting exercise for myself and I hope you find my takeaways useful too. The biggest lesson I have is that if I were to research a bank, I need to study its footnotes and I should be extremely wary of banks with complex balance sheets that contain a significant amount of assets with questionable values.


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. I currently do not have a vested interest in any companies mentioned. Holdings are subject to change at any time. Holdings are subject to change at any time.

Lessons From The Ongoing Bear Market

Surviving long-term, the importance of cash, and good management teams are some of the key lessons from this bear market.

This year demonstrated the cruel realities of investing in stocks.

Year-to-date, the widely followed US stock market benchmark, the S&P 500, is down 14%. Meanwhile, the NASDAQ Composite, a tech-heavier benchmark for US stocks, has lost 22% of its value.

But that’s just the tip of the iceberg. Many fast-growing companies have had it worse. For instance, the ARK Innovation ETF, an exchange-traded fund that invests in high-growth tech companies, is down by more than 50%.

Multiple stocks that were big winners during the COVID-induced lockdowns have also since returned all their gains; some are even trading well below their pre-COVID prices.

In my nine years as an investor, I’ve never seen such sharp and steep drawdowns across such a wide array of companies. But this likely won’t be the last time either.

With this in mind, I’ve penned down a list of investing thoughts to prepare myself for future downturns.

Don’t celebrate when prices go up

Stock prices gyrate wildly. During the booming market of 2020, there were many investors who celebrated when prices went up. Today, many of the stocks that rose in 2020 have returned all those gains – and then some.

2022 has so far reinforced the fact that stock prices really don’t matter in the short run. If prices run up without fundamentals, they will come back down eventually. Similarly, if stock prices fall below intrinsic values, don’t panic. Prices will eventually return to their underlying values.

As a long-term investor, I have learned to ignore near-term price movements and focus on business fundamentals and valuations. 

Cash matters!

When stock prices were rising, companies could raise capital easily by issuing new shares at inflated prices. This increased their cash balances with minimal dilution to existing shareholders.

But now that stock prices have fallen, this source of capital has evaporated. Debt has also become more expensive due to rising interest rates.

It is in times of crisis that companies with strong balance sheets survive, while those with weak financials struggle. Companies that are burning cash and have insufficient cash may end up in a liquidity crisis or end up having to raise more capital at depressed valuations, which could severely impact existing shareholders. If these companies are unable to raise money, their debt holders may end up taking over them, leaving equity holders with scraps.

Invest in strong managers!

With asset prices low, this is a time for companies with the financial muscle to double down on investing for their future. This is a time when prudent managers shine through.

If a company has a great capital allocator at the helm, the company can come out of this bear market stronger than before.

Berkshire, for instance, has started to become more aggressive with its investments in terms of both buybacks and acquiring stakes in other businesses. I believe Warren Buffett’s recent decisions will pay off handsomely for Berkshire shareholders in the future.

Diversify

When stock prices were going up, there was a lot of discussion about concentrating one’s portfolio into just a few stocks.

But this is a risky strategy. Every company has its own set of risks that could result in long-term underperformance of its stock. Companies that are still growing fast and burning cash bear even more risk.

When investing for the long run, we are placing a bet on a company performing well for many years. This doesn’t always pan out. In fact, most companies don’t do well over time and the strong performances of market indexes are driven by just a small handful of companies. When investing, we never deal with absolutes. We are always playing the probability game.

As a long-term investor, survival and long-term steady returns are more important to me than simply maximising earnings. While having a diversified portfolio might reduce my expected returns, it increases my odds of long-term survival and stable returns.

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. I currently do not have a vested interest in any companies mentioned. Holdings are subject to change at any time. Holdings are subject to change at any time.

Why Shareholders Shouldn’t Fret Over Short-term Fluctuations in Business Growth

Businesses can have good years and bad years. But the good ones will eventually keep growing.

As a long-term investor, business fundamentals matter more to me than the near-term fluctuations in stock price. That’s because if a company can grow its free cash flow per share every year, the share price will likely follow suit over the long term.

But this does not mean that a company which has a bad year will be a bad investment.

The truth is that businesses don’t grow in straight lines. Even the fastest growing companies have periods of time when growth decelerated or even turned negative. Business growth depends on a host of factors, some of which are not within the control of companies. 

Let’s take Apple for example. Today, Apple is the largest listed company in the world but its business experienced ups and downs along the way.

The table below shows Apple’s revenue and revenue growth from 2007 to 2021

Source: Apple annual reports

From 2008 to 2021, Apple managed to grow its revenue almost tenfold. But from the right-most column, we can see that the growth rates were very inconsistent. Apple even saw its revenue contract year-on-year in 2016 and 2019. Those declines in revenue did not make Apple a bad company overnight. The iPhone maker managed to bounce back to post much stronger results each time. 

As shown, even one of the most innovative companies in the world can experience inconsistent business growth.

Ultimately, a company that has a capable and innovative management team, great products, and a great value proposition to customers will be able to accelerate growth in the future.

This year, in the current challenging economic environment, many companies that previously had stellar records of growth are either growing more slowly or are experiencing contractions in revenue.

Although this is unpleasant to witness, I think shareholders should focus on what’s causing the deceleration in growth and whether the company can post a rebound. A bad year does not make a trend.

It is in times like this that we need to remember what being a long-term shareholder truly is. Your portfolio companies will not always grow at the same rate each year. There will be some good years and some challenging years. 

So be patient. Focus on the metrics that matter and the quality of the business. Don’t be too quick to write off a company and don’t get too caught up with Wall Street’s obsession with near-term results. 

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. I have a vested interest in Apple. Holdings are subject to change at any time