The returns on waste stocks have been attractive for many, many years. I wanted to write a bit about what I’ve learned about the industry over the past few months of studying it. The model is straightforward: companies collect waste from residential, industrial, or commercial customers and dispose of it at landfills, transfer stations, or recycling facilities (including the increasingly popular renewable natural gas facilities). They earn revenue through hauling fees, tipping fees at disposal sites, and various charges tied to handling.

Moats + Opportunities

There are obvious moats in waste management, including economies of scale, high barriers to entry from the cost of equipment and disposal assets, and customer stickiness. The benefits of scale for a waste hauler are clear. Larger companies can spread the costs of expensive infrastructure over more customers, giving them a cost advantage that smaller players cannot replicate—unless, of course, they are acquired, which is exceedingly common in the space. As the major players grow, they also benefit from improved route density and optimization, making collection routes shorter, cheaper, and more efficient, which lowers operating costs and improves margins.

Vertical integration is central to value creation in the industry. A vertically integrated operator that owns both the hauling routes and the landfill can internalize the waste it collects, dramatically reducing tipping costs and capturing the economics at every step of the chain. To illustrate, imagine a company paying $30 per ton in tipping fees at a landfill while charging customers a total haul cost of $500, which includes transportation and tipping. If that same company owned the landfill, it would eliminate those tipping fees entirely. In other words, the cash that would have been paid out flows straight to the bottom line, improving margins significantly.

There also exists a valuation gap between vertically integrated and their non-integrated counterparts to reflects this dynamic. According to an industry expert, you can expect the private market valuations of these large vertically integrated haulers to be in the high single digits, with non-integrated players seeing their multiples half of that. Landfill ownership creates a moat; the regulatory, environmental, and zoning requirements to build new disposal capacity cannot be understated. Couple this with landfill scarcity increasing further demonstrates the value of these assets. Unexpected volume increases can see the useful life of landfill assets drop faster than expected, creating a need for alternative methods of disposal (e.g., incineration, waste-to-fuel, recycling).

This is driving investment into recycling infrastructure, material recovery facilities, renewable natural gas projects, and waste separation technologies. These initiatives are not just about environmental compliance; they are increasingly tied to economics. Rising values for recycled materials, such as OCC (old corrugated containers), and regulatory incentives like renewable identification numbers (RINs) generated through RNG projects are creating new avenues for margin expansion. The waste majors understand this and have already begun taking steps in this direction, as seen with WCN’s New Ridge RNG plant.

Valuations

What’s been particularly interesting is how valuation multiples for the majors have steadily expanded over time, creating a powerful tailwind for returns (see image below on 2015 earnings multiples versus 2025). The market has rewarded their scale, predictability, improved margins, and balance sheets with progressively higher multiples, making the passage of time almost as important a contributor as earnings growth. But it also raises the question of how much room is left for further multiple expansion. When valuations are already elevated relative to history, it becomes harder to rely on them as a consistent driver of returns. Instead, you wonder whether multiples might ultimately normalize and turn to a headwind not tailwind, placing greater emphasis on underlying execution and cash flow durability, which these businesses surely have.

M&A

M&A is rife in the space and ultimately functions as a strategic tool that allows management teams to reshape their competitive position more quickly than organic growth alone would permit. When executed well, it lets companies consolidate fragmented markets, acquire capabilities they don’t currently possess, or accelerate entry into adjacent categories where the economics are more favourable.

I do not foresee a change in M&A strategy from the waste majors who seem to be broadening their scope through acquisitions of diversified services. In the past, waste companies just used to buy other waste firms (see Brad Jacobs @ UWS & Ron Mittelstaedt @ WCN) but now we’re seeing them venture into hazardous waste companies and specialty waste companies to diversify their portfolios. A few examples would be Republic buying out U.S. Ecology ($2.8bn, 14x EBITDA), Waste Management buying out Stericycle ($7.2bn, 13x), Clean Harbor & HEPACO ($400mn, 11x), and Secure’s forced divestiture of Tervita assets to WCN (C$1.1bn, 7.5x ). From what I read most of the smaller tuck-ins are taking place in the United States South East, SW & MW. And of course, internally these waste majors are improving margins through price/volume improvements. While in the costly Northeast there is a push for using rail to move waste.

These companies clearly possess enduring competitive advantages inherent in their business models (i.e., economics resembles that of a natural monopoly). They operate with scale that compounds efficiency, sustain pricing power, and distribution networks that create formidable barriers to entry. It’s no wonder they’ve been able to compound earnings for long periods of time. The M&A pipeline remains strong as well, giving management teams yet another lever to deepen the moat and grow earnings. Altogether, it paints a picture of businesses built to compound steadily, supported by both structural tailwinds and disciplined execution.

Private banking institutions are one of the most important institutions in society. The banking system is responsible for acting as the key intermediary between credit-worthy borrowers and savers, efficiently allocating capital towards productive uses. In addition, banks are essential in maintaining financial stability and facilitating the transmission of monetary policy. Given their role in the economy, I find that banks are mostly overlooked–after all, credit is a commoditized good with little competitive moats existing. This quick write-up looks to address how I think an equity investor should evaluate banks at a high level.

Business Model

The business model of a bank is relatively simple to understand. Banks take in customer deposits (liabilities on their balance sheet) and loan (assets for the bank) those funds out — charging an interest rate to act as compensation for the risk of default. To entice customers to deposit at a bank, they also pay interest on customer deposits. The difference between the interest paid on customer deposits and loans is known as the net interest margin (NIM). The NIM is arguably the most important metric in personal and commercial (P&C) banking. Though, banks don’t always have to loan out their deposits. Instead, they could invest those funds into–typically fixed rate–marketable securities (i.e., mortgage backed securities, corporate bonds, treasuries). To help soften the potential impacts of a defaults on loaned funds a bank investor would want equity capital to be built up in anticipated for unexpected (above provisioned) credit losses.

Banks of course have other divisions beyond retail banking that make them money. The most important to consider are a bank’s business in wealth management, capital markets, and asset management. Realistically, only of these strikes me as having the potential to be an above average business – wealth management. First, the capital markets business has its pros and cons. On the pros, the business can be strategic to corporate relationships and during economic expansions can be a source of revenue growth. On the flip side, the capital markets business has the negatives of cyclicality, high regulatory burden, and can have higher compensation costs (excluding the cost of employee turnover). The asset management business has the same issues though it is a bit more defensible in that it generates recurring management fees, however, fierce competition leads to fee compression. This leaves us with the wealth management business which tends to be stickier and creates opportunities for cross-selling (e.g., estate/tax planning). The recurring nature of the wealth business increases its attractiveness which is only further supported by demographic tailwinds that will keep the business growing through cycles.

Value Creation & Banking Analysis

Using a DuPont model to breakdown ROE, we know it is the product of return on assets (ROA = Net Income Assets) and asset leverage (assets/equity). Asset leverage is an inherent feature of the business model and cannot really be messed with–juiced up–because of regulation and the unease depositors would have if credit risk in loan book is underwritten solely with their deposits.

First, looking at ROA we consider that it is derived by net income divided by assets (primarily loans). The key drivers of net income for a bank then is the bank’s NIM and their fee yield. The fee yield refers to income earned from non-interest sources (e.g., wealth management fees, M&A fees, etc…). With respect to expenses, the most prominent are vanilla operating expenses and credit costs. As the asset leverage is hard to grow, for a bank to have a moat it must be better than competitors along one of the four elements (NIM+Fee Yield-OPEX-Credit costs) determining net income. You would imagine that all businesses have minimized their operating expenses so only three of the elements can really be differentiated.

Here’s how each of the three elements can achieve differentiation. A bank with a strong franchise may have stickier deposits that require lower interest payments–as they are only trusted intermediary–reducing the cost of funds and growing the NIM. They could also have a highly sought after wealth management or capital market franchise that juices up their fee yield. Finally, they could possess a clairvoyant ability to see which borrowers are actually likely to default, enabling them to minimize their credit losses (this is where most fintech institutions claim they can leverage technology to get an advantage). It is quite difficult to get some of these advantages to oneself but there are cases where a bank is able to defend a moat.

Ultimately, banks are in the business of allocating credit to customers and that credit is a commodity. The consumer doesn’t care where it comes (TD, Scotia, etc…) as long as its cost is cheap and the friction in transaction minimal. For this reason, banks in my opinion are inherently poor businesses. The business models necessitate leverage–in the form of customer deposits–to sell commodity products, I couldn’t think of anything worse. Yet, there are examples of banks that have delivered impressive long-term results (e.g., Handelsbanken in Sweden, DBS in Singapore, and a few North American names as well).

Choosing Between Banks

I’ve discussed what matters in banking and shareholder returns, next is to examine the differences in banks and which you’d rather own. The first dimension to analyze should be deposits. There are two important differentiators when considering deposits: their stickiness and their cost. An investor would like to see a bank’s deposits be low interest bearing–zero being the best–and sticky. Simply, you wouldn’t want the largest liability you possess to have a high cost associated with it and you certainly wouldn’t want deposits to disappear (your short term liabilities being called early) in a whim. Were this to happen then you would be forced into selling your assets to allow for customer withdrawals. That could become a problem if the asset side doesn’t reflect risks taken on the liabilities of that bank.

Next, you’d want to consider the quality of the bank’s assets (primarily, its loan book and investment securities). As an equity investor you’d hope the loans are (1) to a diverse group of creditworthy borrowers (2) the borrowers are charged an interest rate that covers the risk of default, (3) and the composition of the loan portfolio is reflective of the nature of deposits (do the liabilities & assets match, aka. asset-liability matching).

A Case Study: Silicon Valley Bank (How Not to Asset-Liability Match)

The 2023 Silicon Valley Bank (SVB) fiasco provides an interesting case study of when liabilities and assets don’t match. SVB”s depositors were mainly emerging technology companies who had been flush in part due to a boom in venture capital. Seeing an increase in deposits, SVB parked their cash in long-term government debt and mortgage backed securities. SVB opted for longer term assets to park cash because the yield they would receive in shorter term options wasn’t satisfactory as they approached zero. SVB’s management were in my opinion, reaching for yield. When the Fed in 2022-23 aggressively raised rates to temper runaway inflation, the value of these assets dropped markedly. SVB’s failure to envision a world where short-term rates weren’t at zero would cost it dearly. When deposits started to leave the bank (a negative of having a concentrated deposit base), the bank was forced to sell their long-term assets (which had lost value) and faced a liquidity crunch. SVB’s liquidity crunch resulted in a classic bank run that left the bank with nowhere left to go. The root cause of SVB’s collapse was their exposure to interest rate risk (due to long-term duration on investment securities), poor asset-liability matching, and flighty depositor base. Great banks do not have these characteristics and are thus less likely to meet complete loss in the same fashion as SVB did.

**As an aside, the accounting treatment of HTM (held-to-maturity securities differs to available for sale [AFS] securities which are marked to market, so one easily notices changes in values.

The banking business is competitive, regulatory complex (I didn’t even go over some of the metrics of importance like CET ratios and RWA), and inherently levered. I would never call it a good industry for shareholder returns, however, I believe it is important to understand it as opportunities can arise regardless. When dislocations happen, whether due to credit fears, regulatory overhang, or misunderstood balance sheets, banks can trade at deep discounts to intrinsic value and present compelling setups for the patient investor.

Google recently reported a decent first quarter earnings, here’s what I took away.

Highlights

Consolidated revenues came in 12% higher YoY at $90B, driven primarily by Search & YT ads (up 10% respectively), offsetting declines in Google’s network revenues. This is a good thing for Google – Network ads are lower margin (higher [TAC] traffic acquisition cost) so as they make up less of advertising the company’s earnings will improve. Google Cloud revenues were also robust growing 28% YoY (this was a slight miss versus consensus of 29%). EBIT for the quarter came in 20% higher than last year in part due to margins expanding by 230 bps to 34%. The expansion we see on margins is partly explained by increases in the number of years infrastructure is depreciated against, revenue mix, and compensation recognition. Needless to say, this is a cash generating machine even when it seems to be counter-positioned.

Google Search

No commentary on Google is worth writing without consideration of Search’s position relative to ChatGPT and other chatbots. I continue to hold the belief that Google’s distribution advantage will protect the company from new entrants stealing their lunch in Search. In the earnings call there were a couple lines that stuck out to me:

Search revenues were up 10% YoY – not exactly the rate of growth you’d come to expect of a business in terminal decline. I believe this is because ChatGPT and other competitors haven’t found a way yet to get users prompting with commercial intent consistently. I believe that users still tend to use Google Search for searches where it matters which is suggested by the first quote above. According to Needham, only 20-25% of queries on Search are actually monetized with the remainder not generating Google any revenues (as of mid 2024).

Since GPT, Google has defended Search with AI Overviews, Lens shopping, and Circle-to-Search. Lens shopping is completely incremental revenue, respectably growing (10%), and is indicative of the opportunities advancements in ML can have for the business.

Monetization Metrics

It is worth speaking on monetization metrics of the search business. There are only a few ways in which Search & YouTube can grow revenues: (1) increase ad loads (2) increase time spent per user (3) increase user count (4) increase advertiser spend with improved targeting. I found this graph (see below) interesting on the historical path of growth in paid clicks and cost-per-click.

Since GPT, paid clicks have been falling with cost-per-click rising. The volume of paid clicks plateauing is cause for some concern. However, if we go by Schindler’s words (earlier quotes), then this could be reasoned to be the result of better ad-targeting resulting in higher costs for advertisers (cost-per-click). To me, fewer clicks aren’t necessarily a bad thing (would still want it going up) if those clicks are more valuable (but we don’t know if that is the case or if this is Google squeezing advertisers). If this trend continues to persist (even in an during an unkind macro backdrop) then yes – it would be fair to say targeting has improved. After all, advertisers won’t continue spending if the are not getting a satisfactory ROAS.

I’m certain there will be some cannibalization on commercial queries as ChatGPT and other services improve. The question is will the new revenues from AI offset both the cannibalization and the increased costs of serving queries with AI?

Size of Google + Wiz

In other news, Google announced their intent to purchase Wiz a cybersecurity company for $32B (or ~65x sales)! From what I’ve read Wiz seems to be ran by competent individuals and could be a good differentiator for Google’s Cloud business. These big tech companies command such large market caps that what should be a large purchase is not all that much. Google trades for ~$2 trillion so the Wiz acquisition is ~1.6% of market cap. To compare the size of the acquisition you could imagine a $2 billion company acquiring a $32 million business. I don’t have a clue if the nosebleed price they seem to be paying will be worth it but it is important to note this won’t be the end of the world should it go wrong.

Disclaimer: I own shares in Alphabet.

This is a newly written pitch on Medical Facilities, a small-cap Canadian company with operations in the U.S. that has been selling their non-core assets and sending that capital back to shareholders.

Link

It is all but confirmed that Gogo will be purchasing one of its only two real competitors in the race for business aviation global broadband supremacy – Satcom Direct. Here is what Oak Thorne (CEO) had to say about the company back as a competitor in 2021 and 2023:

“Satcom is a reseller of other satellite companies products. So they are a very good service organization and we have a lot of respect for them. But again, they’re sort of at the high end of the market… and we would cohabitate sometimes with Satcom Direct installs.” (2021)

“Satcom Direct has a deal with OneWeb as well as we do, we don’t see a lot of progress from them on that front. They seem to be pushing some of their other projects more heavily. We always have to take them seriously, they are a good competitor.” (2023)

Revisiting Gogo

As a refresher, Gogo is the leading provider of in-flight connectivity for business aviation customers. Founded in the 90s, the company provided air-to-ground (ATG) connectivity for both commercial and business end-markets. The company sold off its commercial aviation (CA) business in 2020, opting to focus on its more profitable business aviation segment and deleverage its balance sheet.

Frankly, ATG is no true competition to satellite-based internet in the CA space – ATG has limited bandwidth relative to satellite solutions, so as more passengers opt to use the internet an ATG provider works to limit their usage in the form of higher (absurd) prices. There was and continues to be no long-term viability in ATG as a solution to commercial IFC because of this key misalignment.

Starlink

The first question I’m always asked is: “What about Starlink?” and rightfully so. Without a doubt, Starlink is the only true competitor for Gogo in business aviation. Despite management brushing off the company as consumer-grade equipment repurposed for aircrafts, the business continues to make headlines in the commercial and business aviation segments. 

I must note the difficulty it is to attempt to objectively analyze this business due to some irrational share price fluctuations. Elon Musk has this ability to make markets move at will when it comes to the companies he runs. Unlike Tesla, SpaceX is not publicly listed. However, that hasn’t stopped competitors from moving off his successes. For example, last month Hawaiian Airlines & Air France – remember, these are commercial airlines – announced that they would be installing Starlink antennas to service their flights. Shares in Gogo were down 5%+ on release of the news about a business they sold off four years ago. While any aviation win for Starlink can be viewed negatively for Gogo, I believe some of the daily moves witnessed in the past few months have made little sense. Nonetheless, Starlink is still a formidable competitor in heavies today and will be in the small-mid segment when they do eventually come up with an antenna to service them.

So then, why do I continue to believe the stock has a high probability of being mispriced? First, the organizational structure and depth of support matters to what is a demanding end-customer. Gogo first and foremost is a service organization that serves business aviation customers and has done so even before the Starlink project was announced to the world in 2015. This extends to Gogo’s relationships with OEMs and dealers. The latter of which was largely ignored by Starlink until it realized that to succeed the company required a strong after-market presence (after all, MROs have capacity for 500-600 jets per year; in comparison to estimated new jet shipments in the 700s). 

Starlink sales & support staff numbers in the dozens not the hundreds of a traditional player like Viasat. The result was Viasat keeping Starlink at bay in commercial aviation – until it wasn’t (partly due to faults in Viasat’s own satellite solution). You don’t require all too deep a relationship in the commercial segment relative to the business segment, especially once you get to small/mid jets. You can imagine that these fliers will demand support for their >$200k product. Gogo knows this and is structured to do so; Starlink on the other hand lags behind on this front.

Additionally at present, Starlink is not line fit at any of the major OEMs and perhaps the most well-known of business aviation OEMs (Gulfstream) has actively stated antagonism to the Starlink STCs that have been approved by the FAA (for the G650 model). In a Maintenance & Operations to all Gulfstream operators the company stated its reservations noting concerns over the structural integrity with respect to antenna installation on the model’s fuselage. As a result, Gulfstream may not be able to provide in-service support with the installation without additional costs. Gulfstream also noted that its structural warranty may not apply to affected areas of an installation based on the STC. While Gulfstream’s disapproval of the STC does not mean Starlink can’t be installed on any of its aircraft, it does bring some level of validation to Oak’s criticisms of the Starlink product for general aviation purposes.

I also believe that Gogo isn’t dead money due to Starlink due to the size of the total market. Per Jetnet there were 24,123 business jets in the world in 2023, of which I estimate 70%¹ were North American (16,886 jets). Importantly, these numbers exclude an estimated ~8,500 turboprops that Gogo often includes in its estimate of its total addressable market.

As you can see the majority of the private jets are either light (9,792) or heavy (9,463), as medium sized jets (4,868) grew at a rate of 1.37% over the last 18 years versus the total fleet ‘s growth of 3.33%. Using these numbers, I attempted to estimate the total size of the business aviation market – specifically, the recurring services revenue. Here’s what I found using some conservative estimates:

In the U.S. alone I estimate the total TAM to be ~$800 million. I make the following assumptions to get to that number: (1) The ARPA of a light jet is $3k, $4k on a medium, and $5k on heavy jets (2) All business jets are assumed to have some IFC system installed. Given the level of conservatism embedded into these numbers, I believe the market is large enough to accommodate both Starlink and Gogo. Even if the industry splits 60-40 in favour of Starlink’s IFC solution there would still be ~$320 million of service sales flowing to Gogo’s top line and this excludes the equipment revenue made along the way. This is only an example to illustrate the level of safety I see in the terminal value of Gogo’s business.

I spend a lot of time thinking about Starlink purely because I believe that is the only impediment to the success of Gogo. If Oak and team are able to execute and not cede the entire business aviation market to Musk, I have no doubt that the stock will eventually work. 

Execution

Winston Churchill once remarked, “no matter how beautiful the strategy you should occasionally look at the results.” Each passing day the importance of execution to the story continues to grow. The company noted that a few customers churned to Starlink in the second-quarter and it’s fair to say the market was unamused. Fortunately, Gogo’s HDX antenna (designed for small-to-medium sized jets) will be the first to be shipped at some point this year with the FDX antenna expected to launch in early-to-mid 2025. I wouldn’t be too surprised if Starlink does end up swaying some more customers until the FDX antenna is commercially available.

With that being said, onto the Satcom acquisition…

The Satcom Acquisition

There are a few things I like and dislike about this deal. I’ll start with what I like. The company bought a competitor that compliments its existing business well. This allows Gogo to truly get a foothold in the larger jet market and global presence through Satcom’s sales and support staff. To compete against Starlink, Gogo must really focus on what Starlink is unable to do as well as it does (given its size and focus). Despite Satcom being a reseller of GEO satellite capacity – what Oak himself has referred to as a dying technology – its current footprint (of ~5k jets, with ~1.3k being heavy jets), will enable Gogo to have an edge versus Starlink in the inevitable switch from GEO solutions to LEO solutions.

In the M&A announcement call, Oak noted that at the high-end of the heavy jet market there is a need for redundancy (having alternative service providers in case of some failure). This probably holds true in the short-term but over time, non-geostationary satellites will dominate the market.

Now for the negatives…  At closing Gogo’s net leverage ratio will shoot back up to 4x, the company will issue 5 million shares and will have to deal with integration. This comes off the back of Oak stating earlier this very year that “organic growth is the right place for [Gogo] to invest capital… and that [it] is less risky than M&A. To me this suggests that Oak feels it is necessary to arm the company with extra arsenal to compete with Starlink in the heavy-jet market despite his remarks over the product direction and quality of Starlink. 

Some other areas of concern that I have with the deal include the possibility of integration diverting attention away from execution. Gogo is not in a position in which it can continue to be off the ball with Starlink already getting installed on heavy jets. 

Overall, I am begrudgingly in favor of the deal despite its highlighted drawbacks as I believe it will allow Gogo to keep a sizable number of aircraft away from churning to Starlink. Speaking of Gogo as an investment, I think it makes sense for me to continue holding shares as the launch of its LEO products are about to hit the market. There are still questions to be answered such as the company’s 5G antenna – which I now view as optionality for the company that is likely to be cannibalized by its own LEO offering. This deal will only magnify the downside and doesn’t really make a dent in the upside in my opinion.

The inflection point in free cash flow I wrote about is slowly approaching for the company. I believe it only makes sense to watch how the company’s products are received and whether its competitive position is capable of creating value for shareholders. My analysis based on the incentives and motivations of industry participants (see below for an industry map) suggests the company can do as much, it is now whether or not the market or I have misjudged the competitive dynamics.

1. I estimate 70% to be North American jets using globally wholly-owned jets in operation per geography as a percent of its total – a different number to historical fleet size, reported by Jetnet.

EDIT (Dec, 10): I have since reduced my position in Gogo by ~35% in part due to the change in the company’s management team (incentives cannot line up as strong as they were with Oak although he still remains in the fold as Exec Chair + GCTR) and deterioration in fundamentals of the business (as noted in the write-up above). The chief reason is the probability the company overpaid in the transaction given the uncertainty of achieving what they’ve guided for. From what I’ve seen and read of the company – that probability of an inflection in FCF is very real and likely (why I still hold as much as I do) but how much of the benefits they’ll reap for spending $375mm in what could be a very expensive distribution deal relatively? That, I do not know…

I would like to share with you a business I particularly like and one that is part of my personal portfolio.

Click here for the pdf.

Disclaimer: I do own shares of the company and bought those earlier this month (July, 2024). This is not investment advice or even education for that matter – just a window into my thinking.

Here’s the link to a write-up on a business I found to be interestingly positioned over the coming years/decades: https://drive.google.com/file/d/1eif4P7HcszHbU8nDIsTnBD_GiX2VKv5i/view?usp=sharing

Disclaimer: I owned shares in the company discussed at the time of writing.

I do not share these introspective pieces but I felt this piece of journaling might be a good one to put into the open.

The Start (2019)

The story starts at the age of fifteen, I had saved up money selling sweets to classmates and completing homework/projects on the side. Generating income for myself felt good, I was emboldened to turn that income into a greater pool of capital. I was aware of the market at the time and acutely so of the great moneymakers – Buffett and Soros, in particular. I took what was then the equivalent of ~100 euros and put it into a brokerage account. My first investment was Amazon, I bought it ahead of earnings (unknowingly) and it went up 2.5% within two days of ownership. I held it ever since and have since doubled the money I invested in the business for a total return of 126%.

Unfortunately, the previous sentence is untrue – I sold the very next morning.

Looking around the brokerage I used, I noticed I was able to put my money in other assets & instruments. I noticed there was an option to trade binary options. Or as I call it now – gambling – and put some of my money on them. My poison of choice back then was forex options. I had spent my summer and weekends learning about Bollinger bands, RSI, and MACD. Through the venture I made some money and then lost magnitudes more. I didn’t like losing money. I abandoned the trading experiment months later, permanently impairing well over half of my capital. I realized then, multiplying more capital in minutes not months/years was simply not my game. I left trading for good (my bank account thanks me) and decided it fit me best to learn from other greats.

Dividend & Value Investing (The Covid Years: 2020-2021)

Losing money is never fun, I didn’t want to experience it again. In my senior year, I stumbled across the idea of dividend investing. I thought it made some sense – it’s pretty hard to lose money when you get paid a portion of it consistently. As long as the company didn’t cut its dividend or go bust you were guaranteed income. However, I didn’t commit any capital to this investment style, simply because further research revealed to me that I could earn higher rates of return on an index fund. I want(ed) to beat the market and compound my capital at attractive rates of return, so I wasn’t too impressed by these findings. So while behaviorally at the time dividend investing was of interest to me, I avoided it. I’m fortunate I didn’t get drawn in by dividend crowd, I believe those type of investors are more susceptible to the endowment effect and end up accepting below average results.

Value Investing

I then decided to step up my consumption of Buffett-related YouTube videos and writing. I figured he knew quite a bit about the investing world looking at his net worth. I actually still remember the first few videos I watched. In true Buffett fashion he continued exclaiming how it’d be much easier to invest in a low-cost index fund and wait. I thought it would be possible to beat the market and get rich faster. To gain depth in my understanding, I then began reading the Intelligent Investor. Controversially, I believe the book is a bit overrated. I did manage to get some key takeaways that kept me in the loop: (1) Investment is most successful when it is most business like (2) An investment operation is one with thorough analysis that promises safety of principal and a satisfactory return (3) A look at the successful investors over the past and their methodologies pointed to there being something special about the value investing approach.

I then read Security Analysis which was almost sort of a revelation to me. I immediately got it. The early pages (3rd Edition) included examples demonstrating the concept of margins of safety and the appreciation that true business value is something one cannot pinpoint. It was a series of examples between pages 18-30 that confirmed my affinity to the approach. Luckily, I’ve kept notes and the book, here a two that I found poignant:

The JI Case Company (pg. 20-25, 1933): Graham laid out the example of JI Case which had a market price of $30/share and asset value of $176/share. JI averaged EPS of $9.5/share over the 10 years prior to 1933 – an unreliable figure given the prevailing economic environment. Graham’s key takeaway was the difficulty in forming a conclusion on the relation of intrinsic value to market value. Was JI’s true value closer to $30 or $176? Contrast this to his next example:

Wright Aeronautical Corporation: Its shares were trading for $8, it paid a $1 dividend, averaged EPS of >$2/share, and had >$8/share in cash assets. In this case it is simple to determine that the intrinsic value per share was substantially above market price. Unlike JI, there is no stress figuring out if you’re getting a bargain or not nor do you need to find the exact price. All one needs to know is that price is well above the market’s offer.

Anyways, I became anchored to finding companies that were trading for ‘cheap’ multiples on current earnings. I opened a paper trading account and a screener on Finviz and began my search. My idea was to form a portfolio of 20 stocks trading at <5x earnings. I built out the paper portfolio and the results were lackluster given the market’s prevailing euphoria. Results did improve relative to the market in 2022 but importantly I also kept learning. I realized I didn’t like buying cheap stocks for the sake of it. How can one put meaningful sums of capital at risk with faith the market will reprice a security before it is too late? Do you only buy cheap when catalysts exist? I don’t think so; Burry was correct when he noted that catalysts are unnecessary in the presence of sheer outrageous value. I learned that the best margin of safety is not just price but quality – and the latter grows in importance as one’s time horizon grows.

Quality Tilt (2021-2022)

It was around the time of beginning this blog I realized for me to commit meaningful amounts of my capital, I had to ensure that the quality of the business would allow me remain unworried by my decision to buy the stock. I began studying businesses, industries, and strategy through history. I familiarized myself with key characteristics and mental models behind enduring competitive advantage. I was lucky enough that the market sold off in 2022/23 and there were many buying opportunities. One that I bought and wrote on this site about was Google. The quality of the businesses within Alphabet are well known – in fact the DoJ at the time brought an antitrust suit against the company illustrating those advantages. With Alphabet you were buying a business that require(d) little to no capital investments for incremental earnings growth. The core product, Search, exhibited all the hallmarks of a wonderful business and it was trading relatively inexpensive, too.

Management Tilt (2022-Present)

I never paid this dimension of investment analysis its due until I got burned by it. I passed on the opportunity to buy Meta for $95/share – an expensive mistake – thankfully, one of omission. I was able to correctly identify the acquisitive and adaptive nous of Zuckerberg but failed to gain conviction in his ability to show operational discipline. He hadn’t need to show this side of his arsenal in an era of easy money. The company was hiring freely, investing speculatively, and winning. I was terribly off the mark on his ability to show operational discipline. I should’ve bought on the uncertainty and trusted that his prior victories in acquisition and mobile refocusing to be translatable to cost discipline. It was an expensive lesson, but one that taught me early just how management can quickly change perception and the importance of trusting the track record of great managers.

Through this I learned it is not enough to have an exceptional business at a cheap valuation. The last thing you want is management actively pursuing value destructive activities. One’s investment returns will vary depending on management’s quality – an quality even Graham wrote on in Security Analysis. Now I dedicate time to evaluating management’s incentives, capital allocation preferences, and operational discipline into analyses. Having trustworthy and capable management turns uncertainty into opportunity and increases the margin of safety in investment opportunities.

Today:

I’ve combined all the learnings I’ve accumulated in the past years to my present strategy. I look for companies with durable competitive advantages, ran by intelligent (and ideally committed) management teams, and all at prices at a discount to a range of estimates of intrinsic value. I continue to improve upon my analysis in each of the three. For example, realizing how operating leverage can result in a business looking expensive today (on a multiples basis) but cheap tomorrow, once you assume look-through margins. Or in the case where management is clever enough to divest away from underperforming business lines and refocus on those in which they have attractive economics. Or even the administrator-type managers with the wisdom to return capital to shareholders rather than let it sit idly by. As Joel Tillinghast put it: “Crisply put, you want low P/E stocks that are also high quality and growing, with a high degree of certainty about the long-term outlook.” The focus should lay in the degree of certainty and time horizon. The best businesses fend off change well and may even surprise on the upside if high-quality managers are present.

With that being said, I absolutely do believe that despite gaining familiarity with ‘compounder’ type stocks one should not forget the cheaper ones entirely. Being ‘short-term’ greedy as Munger would put it and aiming for cigar butts continues to make sense to me too. Paradoxically, as I’ve focused on compounder-type securities I’ve become more comfortable looking for the cheaper (often smaller in size) stocks. The problem with cigar butts is the issue of scalability & time. My capital base allows me to ignore the former, unfortunately. The only real way to avoid the issue of time with these investments are to position size appropriately and perhaps look for near-term catalysts. As I noted earlier, I don’t like relying on catalysts for if they fail to muster a reaction… what next?

I now find myself looking for companies that satisfy both investment styles. I prefer to keep my universe as large as possible and companies as small. I’m not ideological on this though, opportunities can exist anywhere from household names to esoteric industries in foreign countries. Peter Lynch said it best: “The person that turns over the most rocks, wins the game.” I intend to follow his advice.

“The only way you’ll have success in the stock market is if you have a variant perception, something different from the masses.” – Ted Weschler

On May 13th, OpenAI demo’d real-time translation capabilities of GPT-4o. You can watch the video using this link.

Needless to say, it is quite remarkable what OpenAI has been able to develop with the technology they have at hand. I would much rather talk about the reaction to the announcement. Upon announcement, shares in Duolingo were down ~3.8%, with a noticeable decline at 1:00 PM EST (time of announcement). Obviously correlation is not causation but let’s assume so for illustrative purposes. Let’s imagine the entire decline in Duolingo’s market value for the day can be attributed to the OpenAI announcement. In effect, OpenAI’s real time translator would’ve erased ~$250 in Duolingo’s market value. But does that make sense?

I have little knowledge of the economics of Duolingo. Of what I do know, the business has 90 million MAUs of which 8% subscribe for additional quirks. The primary motivation of Duolingo subscribers is to learn the language. Surprising, right?! So why would the market have decided to reprice the stock? The most common opinions I’ve seen is that GPT’s real-time translator wipes out the need to learn languages. I believe this perception still misses the mark on why people use Duolingo and why those who pay for it do.

There has been live translation on Google for years (albeit, not as seamless as the demo). So while impressive, it is not novel. I hold the opinion, the majority of Duolingo subscribers pay because of a necessity (English is the most learned language) or intrigue. Live translation does not reduce necessity nor does it impact intrigue. If you must learn a second language, GPT’s live translation will only enhance not disrupt the need. I suspect this would be due to relocating to a foreign country. Real-time translation would undoubtedly help, but, learning the language would most definitely occur in concert.

Additionally, should you become fascinated with a new culture/language, you’ll learn as you wish. These consumers are either driven by trends or general interest. To satisfy the thirst of these consumers, real-time translation isn’t enough. They are more likely to hope for immersion into the culture of the language and probably willing to enter foreign countries with the aim of understanding the language better. So the Duolingo value proposition is left unchanged for them regardless of GPT-4o capabilities.

In my opinion, real-time translation isn’t really in the mind of these consumer groups, as it is does not serve their ambition. These groups of consumers will only be satisfied once assimilated – the constant need to have an AI speak for you is not assimilation. If it does not assimilate, it does not threaten Duolingo either in my view.

I haven’t began to factor in other reasons users subscribe to Duolingo either(e.g., the games, test administration, etc.). The thing with all great public equity investing is that to generate above-average returns you must be willing to take idiosyncratic positions. If we all follow the crowd, then you’ll earn the returns of the crowd – and that’s just not what we’re all here to do. So, despite the wisdom of the crowd, I heed to Benjamin Franklin’s advice when he said: “If we all think alike, then no one must be thinking.”

“Having opinions about facts is what makes markets.” – Jim Chanos

*Just to reiterate, no one can for certain know why Duolingo shares were down on the day or even in the past month (if you zoomed out further). I did, however, think this would make an excellent example of variant perceptions.*