If you’re new to the publication, I recommend checking out the quick start guide. It’ll tell you what to expect from these articles, and get you excited for what’s to come.

👋 Introduction

Welcome back! Today we’ve got a special article, one that I think will be really helpful to decipher the macro investing climate. We’re not going to look at a single company; instead, we’ll look at the AI movement as a whole. In particular, we’re going to try to make sense of whether we’re in an AI bubble right now.

Don’t know what a bubble is? Not a problem. I’m going to walk you through the definition of a bubble, why it matters, some historical examples, and then finally assess today’s climate.

Let’s dive in!

What is a bubble?

At its most simple, a bubble occurs when there’s a huge disconnect between the price of assets and their underlying value. The larger a bubble is, the worse it is. The size of the bubble is directly proportional to two factors:

• The gap between the price of the company stock and its underlying value

• The number of companies within an industry/sector/entire market that are exhibiting this disconnect

For example, a small bubble would be if 5 companies were overinflated in value by 50%. In fact, a bubble of this magnitude happens so regularly that it might not even be noteworthy. Maybe you’ll see a few talking heads predicting an impending crash, but for the most part this is business as normal.

On the other hand, a large bubble would mean that hundreds of companies are overinflated by 100+%. In this scenario, there’s a huge amount of implied value in the market that doesn’t actually exist. This is dangerous, for reasons we’ll cover in a second.

Before we move on, there are a few other important things to know about a bubble:

1. The creation of a bubble is typically attributed to a change in investor behavior, not a change in the underlying fundamental value of the asset. It’s a human-driven occurrence, not a market- or financial-driven occurrence.

2. Bubbles are characterized by excessive speculation. When people are buying stocks left, right and center, with no basis for their investments, that’s considered speculation. ¹

3. Economist Robert Shiller once said “irrational exuberance is the psychological basis of a speculative bubble”. When everyone thinks that the market can go nowhere but up, and there are no risks in sight, that may be the surest indicator that we’re in a bubble. This ties back to point #1, where it’s the human behavior that drives the creation of bubbles.

Now that you know what a bubble is, why should you care?

Why should we care about a bubble?

There’s one undeniable, inevitable truth about a speculative bubble. When one exists, it will eventually burst. 100%, no exceptions.

The “bursting” of a bubble refers to when asset prices sharply and rapidly decrease, returning closer to their underlying fundamental value. Herein lies the danger I referenced earlier. If you’re invested in a company that is affected by the bubble, you will lose a lot of money.

When a bubble bursts, it’s worse than you think. I’m not talking a 10-20% drawdown. Even a 50% loss isn’t the worst of the horror stories. There are innumerable examples of people losing 100% of their investment when companies simply go bankrupt and fold up shop.

It’s really, really important to know why you’ve decided to invest in a company. Are you following the crowd and simply speculating because everyone else seems to be making a ton of money? If so, you’re playing a risky game. You could end up losing it all. It’s something that has happened to investors time and time again.

What are the five stages of a bubble?

Bubbles are pretty consistent in following a pattern. They usually grow and collapse the same way:

1. Displacement - This is when a new paradigm is noticed. A small group of people identifies the new paradigm as an investment opportunity. Think: the internet, or AI.

2. Boom - This is when prices start to rise. Usually, the new paradigm has impacted the traditional way of doing things. And when the status quo is affected, the mainstream/majority take notice and hop aboard the hype train. As more and more investors pile into the market, it causes the asset prices to rise. Those who initially exhibit restraint are often overcome with FOMO - a fear of missing out - so they jump into the market too. You can see the positive feedback loop now… prices rise, people jump in, that causes prices to rise further, so more people jump in, rinse and repeat. ⁷

3. Euphoria - Prices are positively skyrocketing. There’s absolutely no caution in investors anymore. Nothing can go wrong and everyone will be a billionaire by tomorrow, guaranteed. ⁵

4. Profit Taking - The sell off begins. “Smart” money ² , also known as institutional investors, exit their positions. At this stage, the prices aren’t going down, but they’re no longer skyrocketing. A lot of people have recently entered the market (in the euphoria phase) and are desperately hoping for massive returns to continue.

5. Panic - Usually triggered by one piece of bad news, the sell off accelerates into a death spiral. Most investors were riding irrational exuberance when they entered the market, so they don’t have the conviction to hold their positions. Now they’re trying to offload stocks at any price. This causes share prices to plummet.

While the pattern of bubbles is consistent, it’s almost impossible to identify what stage, if any, is currently in effect. These stages are easily marked in retrospect, but it’s hard to tell while it’s happening.

Even if you’re certain that the market is in a bubble, no one can know how long that bubble will last. Maybe it’ll take one month to burst… but maybe it’ll take 2 years. Would you be able to sit out of the market for 2 years while everyone around you is making incredible returns? Again, it’s easy to say yes in hindsight. It’s near impossible to say yes while it’s happening.

Economist John Maynard Keynes put it best: “The markets can stay irrational longer than you can stay solvent.”

What happened during previous bubbles?

1600s - Tulip Mania

The tulip mania bubble that took place in the 1600s is the first recorded bubble in history. A scientist brought tulips from Constantinople to the Netherlands for an experiment. Tulips weren’t previously available in the Netherlands, and their discovery by the scientist’s neighbors made them a sensation. The flowers were scarce and novel, so of course everyone wanted to get their hands on some. Almost overnight, tulips became a luxury commodity.

Because of people’s craze to buy tulips, the flowers started selling for exorbitant prices. Things got so crazy that a futures market was created. In layman’s terms, people were betting on the future prices of tulips!

A missed delivery of tulips cascaded to the market-wide realization that prices were unsustainable, and the bubble burst. People lost fortunes.

2001 - Dot Com Crash

The 2001 dot com bubble began because of the growth of the internet, the next big thing. People rightly assumed that the internet would change everything about society and business. So they started investing in any company that had even a tangential relationship to the internet.

Arguably the most damaging aspect of the bubble was driven by venture capital (VC) firms. VCs would raise tons of money for startups that had no product and no profit. Because interest rates were low, and it was easy to loan money, VCs didn’t do their due diligence. Anyone who asked for money was showered with it.

Changed your name from Fertilizer Co to Fertilizer.com Co? Here’s a hundred million dollars, you’re now a tech startup. Created a website landing page? Here’s another hundred million dollars. Company valuations were 10x-ing overnight for no good reason.

Unsurprisingly, this didn’t last. People started to question the value of these startups, especially as the companies had their initial public offerings (IPO ⁶ ). It was a house of cards. As soon as the first card fell, suddenly everyone realized their folly and the entire house came crumbling down. Most of the so-called internet startups were no longer in business less than a year after raising hundreds of millions of dollars.

2008 - Housing Crisis / Great Recession

The most recent bubble, the 2008 housing crisis that led into the Great Recession, was driven by real estate and mortgages. You’ve probably heard the term “subprime mortgages”. Here’s simply how it worked:

1. Banks gave prospective home buyers variable-rate mortgages, which allowed the monthly payments on expensive houses to be very cheap at first. Imagine getting a $1 million house for $100 a month ⁸ . Who wouldn’t take that deal?

2. Banks would then bundle up a bunch of these mortgages, and sell it as an investment vehicle.

3. This process would repeat - the bundles would get bundled into even bigger bundles, and those bundles would get turned around and sold.

There was one critical flaw in the chain. After the introductory low-payment period for the homebuyer, when interest rates increased, suddenly they needed to pay $2500 a month instead of $100. Unsurprisingly, a lot of people didn’t have that money, so they defaulted on their mortgage payments.

Here’s where the chain began to unravel. The bundles of mortgages that the banks sold - they were pitched as safe investments with guaranteed returns. But they weren’t safe if homeowners didn’t make their payments on time. So when homeowners began to default, the big institutions at the top of the food chain - the ones holding the largest bundles that were supposedly guaranteed - now had huge deficits on their books. What looked like a thriving business when the payments were coming in, was now a bankrupt investment firm. Billions of dollars of wealth was lost, and the country was plunged into one of the worst economic downturns in its history.

Let’s apply these lessons to today’s situation and see if we can make sense of the macro environment.

What indicates that we are in an AI bubble today?

• The initial stages of a bubble are unfolding predictably. Displacement, check. Boom, check. Euphoria, check. Profit taking? Possibly. If you look at the moves made by Warren Buffett and other “smart money” investors, a lot of them have sold positions and are holding cash. They clearly believe that the market is overheated and that we’re due for a correction.

• AI has created irrational exuberance. I’ve said this before, but people typically overestimate the impact of new technology in the short term, and underestimate its impact in the long term. We’re in the short term phase, where people are over-excited and that has led to irrational exuberance. When you read constant proclamations about how a company’s slight variation on existing technology is going to change the world, you know people are no longer assessing products objectively.

• Investor behavior is driving rapid increases in stock prices. A company has AI in its name? Time to buy shares! I have no idea what it does, but it’s the future!

• There’s an explosion of AI startups with questionable moats. Swaths of companies exist and raised money for no reason other than they “do AI”. How much value are they actually creating? Are VCs giving them money because there’s a problem to be solved, or because they want to invest in AI? Do their products have any kind of moat at all?

• AI offerings lack clear paths to monetization. This is starting to attract more attention, which is good. When Google, Apple, Microsoft and Meta spend billions of dollars training AI models, how do they profit? They’re adding the AI-powered features to existing tools, like Search, Siri, Bing, etc. Consumers haven’t indicated a willingness to pay additional monthly fees for the AI features. At what point does the bottomless money pit become untenable for these tech giants?

At this point you might be thinking, “Crap, we’re definitely in an AI bubble!” but it’s not so obvious. Let’s look at the other side of the coin…

What indicates that we’re not in an AI bubble today?

• We’re not in an era of cheap money and low interest rates. This AI craze is happening against the backdrop of some of the tightest fiscal policy in the last two decades. That indicates a general attitude of greater fiscal attentiveness, where investors and institutions are being more careful with where their money goes and how it’s used. It’s usually during low interest rate periods that investors have the least amount of caution.

• Following that lead, VC investments are much smaller for AI compared to those made during the dot-com bubble. Between 1999 and 2001, approximately $141 billion of venture capital money was funneled to startups, primarily to so-called “internet” companies. In the last two years, there’s been a total of $55 billion invested ³ . That’s a much smaller slice of the pie, especially when you account for economic growth during the 20 years between.

• Likewise, IPOs are not rampant in the same way as the dot com bubble. There were 892 (!!) IPOs between 1999 and 2001. In the last two years, there’s been a total of 154 IPOs, of which 50 are AI-related companies. That’s a meaningfully smaller number. With fewer reckless IPOs, and less fake wealth generation, the potential risk exposure is much lower. This is a good sign for the average investor.

• AI passes the common-sense test. It truly will be a huge part of society. It’s a natural evolution of our modern technology. It’ll increase automation, improve efficiency on existing tasks, create new jobs, and unlock opportunities we can’t even fathom yet. I have no doubt that it’ll change the world ⁴ , so it makes sense to spend on it. It may take time to figure out how to use AI more efficiently, but that’s ok. It’s part of the process.

• The biggest AI spenders (Google, Meta, Apple, Microsoft) are not propped-up companies. They have incredible financials and create real value for their customers. The products they sell are deeply ingrained in society and are highly profitable. Even if AI were to end up as a massive flop, and all of these companies simply abandoned their quest for next-gen AI tech, it wouldn’t cause their bankruptcies. If anything, these companies might have better looking financials because the huge R&D expense of AI would have been eliminated.

• Because of the financial health of the aforementioned biggest spenders, there’s unlikely to be a cascading effect if an AI bubble bursts. These companies aren’t levered up to their eyeballs and relying on debt-financed expansion. They have robust cash flow that is being reinvested in R&D. The hypothetical abandonment of AI wouldn’t cause a chain reaction of debt defaults, bank bailouts, and institution bankruptcies. ⁹

Applying the lessons from previous bubbles should make it clear that today’s AI situation does not exhibit the same harbingers of impending doom and collapse.

What should we take away from this?

It’s hard to tell whether we’re in a bubble while the bubble is happening. In all of the previous examples we looked at, the bubble was identified in retrospect. The stages were clearly delineated in retrospect. The identification and cause of the irrational exuberance was pinpointed in retrospect.

That unfortunately applies to present day as well. There are a number of arguments that indicate we’re in an AI bubble, and a number of arguments that indicate we’re not in an AI bubble.

So as investors, how should we use this information? How should it guide our investment strategies? There are three key takeaways:

1. Be very careful with where you invest your money during a period of irrational exuberance. If you do your due diligence, and find high quality companies with stellar financials, then you’re minimizing your downside. You can never know with 100% certainty if you’re making a good investment, but you can tilt the odds in your favor. At a time like this, lean on your thorough research and discerning investment decisions. Avoid, at all costs, investing from FOMO and jumping on a bandwagon that others are ebullient about.

2. Maintain conviction in your bets even if a bubble bursts and there’s a 50% correction. If you’ve done the proper research, and you know that you’ve found a winner, don’t panic sell. The worst thing you can do is sell on the way down. Instead, treat it as a buying opportunity. If a house you loved was 50% off, you’d be delighted to buy it at a bargain. Think of a correction the same way - it’s an opportunity to buy a wonderful business at a bargain price. As Warren Buffett says, “Be fearful when others are greedy, and be greedy when others are fearful”. During a market downturn or bubble burst, others will be very fearful. That’ll be your opportunity to buy.

3. Don’t try to time the market. It’s not clear that we’re in a bubble, so are you really willing to sell everything and sit on the sidelines? What if the stock market keeps its astronomical growth for another 2 years? Unless you’re a professional investor (and even then, most of the time they’re wrong), you shouldn’t try to time the market. Stay invested and focus on #2 when the opportunity arises.

It’s an interesting time in the market! Because of how expensive everything is, it’s hard to find good deals. But we’ll keep analyzing companies and staying ready when an opportunity presents itself.

Hope you found today’s special article to be interesting! Next week, I’ll be taking a break, so there won’t be an article. I want to wait for ASML and TSMC earnings calls the following week (10/15 and 10/17, respectively) before releasing a final wrap up of the semiconductor industry. The wrap up will have a recommended path forward for each of the three companies (Nvidia, TSMC and ASML) that I analyzed.

Then we’ll be on to a new series of companies… Stay tuned!

Agree? Disagree? Would love to hear your thoughts - leave a comment.

1  A famous story about John D Rockefeller encapsulates this lesson. Once, a boy giving him a shoeshine started giving him stock tips (not knowing that Rockefeller was a successful investor and businessman). Rockefeller decided that if his shoeshiner was giving him investment advice, then too many people were excited about the market, and it was likely in a bubble. He sold a lot of his holdings, which helped the Rockefeller family take advantage of the Great Depression.

2  The term “smart money” might be generous. Institutions make risky investment decisions, over leverage themselves, and then wait for a government bailout. Not sure how that’s considered smart. But I digress.

3  $14 billion of that has gone to OpenAI, which has proven the value of its product and technology, so it’s safe to say that investment was not a waste.

4  AI’s internal monologue at the moment: https://www.youtube.com/watch?v=2B3slX6-_20.

5  This is the stage at which shoeshiners are giving stock tips.

6  An initial public offering, or IPO, is when a private company first goes public. Once a company is public, then the average person can buy shares of that company through their brokerage account. When a company is private, only certain people - accredited investors - can buy into the ownership group. A company going public typically indicates that it has reached a level of maturity and stability that will allow public investors to (more or less) safely invest their money.

7  This explanation reinforces the earlier message about bubbles being driven by human behavior. It’s the psychology of investors, who don’t want to be left out of massive returns, that causes them to jump on a bandwagon without knowing whether it’s a good investment or not. This cycle drives the market to astronomical heights until the bubble bursts.

8  This is an exaggeration. I haven’t run the numbers, it’s simply to make a point.

9  Granted, the same can’t be said for some of the AI startups that are popping up. That’s where you have to be careful. Google will still be around if the bubble bursts, but Fertilizer.AI might not be.