Paper Bitcoin Summer

What is a bitcoin treasury company?

Bitcoin treasury companies are all the rage on Wall Street. What is going on? In today’s post, I will try to explain.

Asset Pricing

Discussions of asset pricing often sound quite different from discussions about the prices of goods and services. When we think about the price of goods, we know that people will pay a price that is less than or equal to their willingness to pay. A person’s valuation of a particular good is subjective and based on opportunity cost. If I purchase these bananas, I’m giving up my ability to purchase something else. The cost of purchasing bananas is the apples that I would have purchased otherwise.

But discussions of asset prices are different. On some level, this shouldn’t be surprising. After all, I purchase bananas in order to eat them. My own preferences determine whether I prefer bananas to apples and by how much. Given the relative market price and my preferences, I make a decision about whether to buy bananas or apples, and how many to purchase.

When it comes to assets, however, I am not buying an asset in order to consume it. An asset is a means to an end. I’m buying an asset to store value over time. Ideally, this asset will provide a rate of return so that I am rewarded for putting off current consumption. The return allows me to consume even more in the future. Asset pricing seems different.

People often say things like, “the price of this asset should equal its fundamental value.” I must admit, that sounds like a smart thing to say. It evokes a sort of expertise. But what is fundamental value?

The dirty little secret is that fundamental value is not something that can be discovered. There is no formal definition. What people typically mean by fundamental value — whether they realize it or not — is some sort of model-based estimate of what the asset is worth.

There are many different types of models. One way to model fundamental value is to extrapolate the value from a consumption-based model. If an asset is a means to an end, perhaps modeling the value of that end can tell us something about what we should be willing to pay for the asset.

Another way one might determine an asset’s fundamental value would be to compare one asset to another. If asset X is sells for price P, one can infer asset Y’s value by comparing their characteristics.

An asset that provides a cash flow could be valued as the present discounted value of those future cash flows.

Finally, if the asset is a stock, one way to estimate the fundamental value of the stock would be to determine its liquidation value. If the company sold off all of its assets and paid off all of its debt, what is left over would go to shareholders. If the value of the assets is expected to rise over time and the debt is expected to decline, one might think of the fundamental value as the present discounted value of the firm in some future liquidation period.

I’m sure that there are other ways to think about fundamental value, but this gives a general sense of the idea. Given this background, we can carry out a simple thought experiment.

A Thought Experiment

Suppose you run publicly-traded company. For simplicity, suppose that the company has $1 million worth of assets, all cash, and zero debt. There are 100,000 shares of the company. If we base fundamental value on liquidation value, then each share is worth $10. Let’s assume that initially the price of the stock is equal to this fundamental value.

The company is sitting on $1 million in cash. Suppose that it uses that $1 million to purchase an asset. After making the purchase, the share price rises (for whatever reason) to $20. The price is now above the firm’s liquidation value. If you run this company, you might be tempted to sell more stock. You issue 100,000 more shares and raise $2 million. You now have $3 million in assets and there are 200,000 shares outstanding. The liquidation value of the company is now $15 per share.

You realize that buying this other asset caused your stock price to go up, so you buy more of the asset in the hopes this will happen again. Suppose that it does. You use the $2 million of cash that you generated from selling more stock to buy this other asset. The price of your stock rises to $30. Again, this is double the liquidation value. So you sell another 100,000 shares. You now have $3 million of the asset and $3 million of cash. The liquidation value of the company is now $6 million, or $20 per share.

What I am describing sure sounds like a perpetual money pump. Every time you buy the asset, your stock price goes up. This higher stock price allows you to sell more stock and buy more of the asset. The liquidation value rises. When you buy more of the asset, the stock price goes up again. This allows you to issue more stock and use the proceeds to buy more of the asset. And on and on it goes.

But a perpetual money pump shouldn’t be possible. Arbitrage should put a stop to this. For example, anyone could replicate your exact same strategy by borrowing shares of your company, selling them for cash and using the cash to buy the same asset you are buying. If enough people do this, the short-selling should push the price of your company’s stock down and the purchases of the asset should push the price of the asset higher. This makes the strategy harder to execute. Why? Because arbitrage causes the stock price and liquidation value to converge. Once this happens, the money pump dries up.

Similarly, other companies could observe what you are doing and follow the same strategy. Again, doing so should result in convergence between the liquidation value and the price.

MicroStrategy

So how do we explain MicroStrategy? (The company recently re-branded to “Strategy.” I’m going to stick with the old name to avoid confusion.)

Most people think of MicroStrategy as a software company that owns a lot of bitcoin. However, the company has not only purchased bitcoin, but steadily increased the amount of bitcoin it holds per share. This sounds like a money pump, and people are naturally skeptical.

When MicroStrategy initially began purchasing bitcoin, it filled a niche. Some investors wanted exposure to bitcoin, but legally prohibited from doing so. As MicroStrategy accumulated bitcoin, this gave those investors an opportunity to legally gain that exposure. As the company purchased larger quantities of bitcoin, the value of the company (and thus its stock) would start to depend more on the value of its bitcoin than its underlying business. In fact, at a time when the Securities and Exchange Commission refused to approve bitcoin ETFs, I joked that MicroStrategy was becoming a bitcoin ETF with an option on a software company.

This created somewhat of an interesting dynamic. Bitcoin has experienced a great deal of volatility over its short lifespan, with the price rising from nothing to over $100,000 per bitcoin. Investors who believe that bitcoin is on its way to becoming a global reserve asset comparable to something like gold expect this price appreciation will continue. This creates an interesting problem for determining liquidation value.

Typically, in a complete market, one need not worry about the difference between future liquidation value and the current liquidation value. In a complete market, the absence of risk-free arbitrage means that the expected rate of return on an asset would just be the real interest rate. Discounting that back to the present using the real interest rate implies an equivalence between the current liquidation value and the present discounted value of a future liquidation.

However, if markets are segmented and there is an expectation that the price will continue to experience rapid appreciation, this makes the present discounted value of a future liquidation could exceed the current liquidation value. If the stock is selling at its current liquidation value, it is underpriced. Investors would then bid up the price and the stock would trade at a premium to its current liquidation value.

This creates the dynamics for the money pump. The company can sell stock and use the proceeds to purchase more bitcoin. Because it trades at a premium, this allows the company to accumulate more bitcoin per share.

In some sense, this is a self-fulfilling cycle. People believe that the present value of future liquidation is greater than the current liquidation value. This enables the company to purchase more bitcoin per share, which increases both the current and future liquidation value (all else equal).

However, it is wrong to suggest that this is a riskless money pump. The reason for this dynamic is the combination of expectations of a much higher bitcoin price in the future and the existence of segmented markets that prevents some investors from simply buying or selling bitcoin directly. If they are wrong about the future trajectory of bitcoin, the expectations would work in reverse and the stock would trade at a discount relative to its current liquidation value.

Michael Saylor, the co-founder and chairman of the company, has long described the objective of turning the company into a bitcoin bank. Most people completely misunderstood what this meant. They assumed that MicroStrategy wanted to become a typical financial intermediary, like a commercial bank, accepting bitcoin deposits and making bitcoin loans. This is not what Saylor meant. His objective was to accumulate bitcoin at scale to exploit regulatory arbitrage. There are countless other investors who would like exposure to bitcoin, but are prevented from doing so due to either their declared investment strategy or some sort of legal restriction on what they can purchase.

To that end, MicroStrategy has issued convertible bonds. These provide investors who are required to purchased fixed income securities with exposure to bitcoin. Since these bonds are convertible into shares of MicroStrategy, fixed income investors can gain exposure to any potential bitcoin price appreciation. If the price of bitcoin continues to rise, MicroStrategy’s liquidation value rises and thus raises the value of the bond conversion option. At the same time, these fixed income investors can hedge the downside risk by shorting the stock.

The company has also started issuing preferred stock. Some investors are required to purchase preferred stock given its lower risk profile in comparison to common stock. These preferred shares offer these investors exposure to bitcoin.

In short, what MicroStrategy has done is turn itself into a bitcoin bank by issuing dollar-denominated liabilities and purchasing bitcoin. The company is explicitly engaged in financial engineering to exploit regulatory arbitrage.

Again, this is not without risk. The ability to exploit this regulatory arbitrage is contingent upon investors maintaining this expectation that bitcoin is going to be worth considerably more in the future. If that turns out to be wrong, investors will who purchased MicroStrategy shares are likely to fare worse than those who simply bought bitcoin. But it is also important to consider the risks associated with preferred stock offerings.

MicroStrategy has created a number of different types of preferred stock offerings that pay investors dividends. Some of these offerings provide MicroStrategy with the option not to pay a dividend. However, others promise a guaranteed dividend based on the value at the time of issuance. Still another offers a promise to keep the price of the preferred stock constant by varying issuance and the dividend paid. These preferred stock offerings with dividends are a way of attracting a certain type of investor with bitcoin exposure while passing along some of the gains associated with a higher accumulation of bitcoin per share and the appreciation of bitcoin’s price. Because preferred shareholders rank above common shareholders, this comes with less risk than the common stock.

At the same time, these preferred shares introduce a new type of risk. By increasing its bitcoin per share, MicroStrategy is increasing its liquidation value. This increase is even greater if bitcoin’s price continues to rise. But MicroStrategy lacks the cash flow to pay these dividends on its preferred stock. Since its bitcoin per share is increasing, it could cash in on some of that profit, sell bitcoin, and use it to pay the dividend. Yet, the entire business model here is based on the premise of generating more bitcoin per share. If they company starts to sell bitcoin, it would be hard to predict how investors’ expectations would change — and Saylor has said that the company doesn’t plan to sell its bitcoin. That means that the dividends will have to be paid out of future fundraising (i.e. issuing more liabilities). Such a strategy (no pun intended) only works if the company’s liabilities continue to trade at a premium.

Furthermore, regulatory arbitrage only works while regulations remain in place.

Paper Bitcoin Summer

This brings me to the emergence of bitcoin treasury companies. These are trying attempting to copy MicroStrategy’s approach. Over the past few months, several new companies have launched with the deliberate strategy of exploiting the same regulatory arbitrage as MicroStrategy. Hence the recent internet meme: Paper Bitcoin Summer.

These companies aim to provide bitcoin exposure to investors in these segmented markets. Their success will depend on (1) the size of these segmented markets, (2) their ability to successfully use financial engineering, and (3) the level of competition that emerges.

It is important to note that history is replete financial engineering gone wrong.

It would be naive to think that this cannot happen in this case.

There is also a question about the end game. This strategy can only work if bitcoin becomes more firmly integrated into the global financial system and perhaps even the global trading system. If these treasury companies are correct about this, at some point major market participants will want to acquire bitcoin at scale. In that case, these treasury companies would become targets for acquisition since acquiring a treasury company would likely be cheaper than trying to acquire a large stock of bitcoin in the market.

Ultimately, these companies are betting on a joint hypothesis: (1) that bitcoin will play a major role in global finance, and (2) that the current regulatory and market segmentation will persist. For those interested in financial markets and financial history, it should be fascinating to watch.