Contents

Nothing in this post is financial advice, and as always, Do Your Own Research. This post is intended for educational purposes only, and to show my latest thinking on a category (which may change!). I look forward to feedback and corrections.

Do DATs actually make any sense?

I had mostly ignored Digital Asset Treasury companies (“DATs”) — the original being Strategy, née MicroStrategy — until a couple months ago. But when dozens (hundreds?) of DATs for every conceivable cryptoasset started flooding the scene, I started to wonder “do these actually make any sense?”

My answer, after wading through a wash of mind-numbing debates, op-eds, and other materials, is a solid “sometimes.”

This is in stark contrast to the consensus views, which seem to be either “they are literal ponzi schemes” or “they are the greatest financial instrument ever created.” As usual, when such extremes exist, the middle generally wins the day.

And after talking to many friends about DATs, I figured I would publish my current understanding.

This post assumes a solid foundational understanding of DATs and financial markets. Here is a good primer on the former from Galaxy.

What’s the real question?

In general, the question you are asking yourself when considering buying ordinary DAT sharesAssuming buying at market — we’ll revisit the PIPEs and other instruments later. is:

Will this purchase outperform the underlying asset over my holding period?

i.e. if I buy MSTR, will it outperform Bitcoin over my holding period?

That question is a more rigorous version of “do these actually make sense?”

The conclusion, from my perspective:

Well-managed DATs can outperform their underlying asset, if that asset itself appreciates over time.

Both bolded phrases are key, as is the framing of the question itself. On the question: we are not asking “is this a good investment on an absolute return basis”; we are asking “is this a good investment on a relative return basis, as compared to the underlying asset?”

If you don’t think Bitcoin will go up in value, there is no reasonAgain, here I am talking about an ordinary investor debating going long the asset — not actors with extrinsic motivations like pair trades, option strategies, etc. to buy a Bitcoin DAT like Strategy.

But, clearly, there are people who are long Bitcoin (or Ethereum, or Solana, or other assets). So for them, the question becomes whether they should consider a DAT slice within their allocation to that asset. Should they buy a certain DAT for the asset, in addition to spot or other holdings?

Growth can outpace premia

The reason DATs “don’t make sense” intuitively to most people is because, in the market, you are almost always buying them at a premium to NAV. So our task here becomes: is it sometimes sensible to buy into a DAT at a premium to its NAV? It feels like it shouldn’t be. You’re paying more for less underlying asset exposure than if you bought in the spot market.

Going forward, we’ll talk about this for Bitcoin DATs, but it could be any asset.

Here’s what needs to be proven: that Bitcoin DATs can increase their Bitcoin Per Share (BPS) over the investor’s holding period enough to compensate for the initial premium to NAV. Any remaining premium at exit is icing on the cake; we will address discounts in a moment.

I will show why that is the burden. Imagine:

  • BTC is at $100k
  • There is a DAT that owns 1,000 BTC
  • It is trading at a 25% premium to NAV
  • So: its NAV is $100MM ($100k / BTC * 1,000 BTC), and its market cap is $125MM ($100MM * 1.25 premium)
  • It has 10MM shares outstanding
  • Which means a current Price Per Share (PPS) of $12.5 / share, and current BPS of 0.0001 BTC / share

In this case, if you had $1MM, you could:

  1. Buy 10 BTC
  2. Or buy 80,000 shares of the DAT, which represent 8 BTC

This is the part where people stop and say “how could that make sense, I’d so much rather have the 10 BTC.”

But if the company can increase BPS by 10% annually:

  • End of year 1, your 80k shares * 0.00011 BPS represent 8.8 BTC
  • End of year 2, your shares represent 9.68 BTC
  • End of year 3, your shares represent 10.648 BTC — more exposure than if you had bought spot!

Note that nothing in this math explicitly assumes the premium increases or even persists above 0 and it doesn’t assume share count stays the same. All it requires is that Bitcoin Per Share goes up.

Your entry premium to NAV affects how long it’ll take before you’re outperforming a spot BTC purchase. If there’s a premium at exit, then you are even more in the money than this model shows!

This all does assume the DAT doesn’t move to a discount — if the stock is discounted below NAV at the time you want to exit, then your outperformance hurdle is higher. However, a well-managed BTC DAT should not trade meaningfully below NAV (if it does, the DAT should do buybacks). DATs should — in general — not trade meaningfully south of NAV for prolonged periods.

Aside from buybacks as a mechanism to get back to near-parity, I also believe that discounted DATs will be acquired by the large DATs — it’s incredibly accretive for their BPS to be able to buy Bitcoin at a discount.

So: what allows DAT to increase Bitcoin Per Share?

  1. If trading at a premium, they can perform at-the-market (“ATM”) offerings, selling stock at a premium to NAV and buying more Bitcoin. This is commonly referred to as “accretive dilution,” because despite causing more share issuance, the key BPS metric goes up.
  2. Convertible bond issuances. DATs have been able to sell low-coupon convertible bonds to fund further Bitcoin purchases. (Mostly) convert-arb investors buy these bonds and often short the equity to gamma trade around the (very volatile) stock price. Long-BTC investors also buy these bonds to get a senior/principal-protected way to own a long-dated call on BTC.
  3. Option selling and related strategies. DATs can create other yield-generation strategies to capture vol (inherent to the underlying asset, and amplified by the effectively-leveraged DAT structured), including simply selling covered calls. They sometimes layer this with the convert issuances, putting a warrant call-spread on top (i.e. buy calls, sell higher-strike calls) to push the convert conversion premiums higher and pull coupon down.
  4. If trading at a discount to NAV, DATs can buy back stock to increase BPS.

Not all sunshine and rainbows

It’s not all sunshine and rainbows. There are two categories of risk to be aware of within the construct we’re operating (assuming you’re looking at this as part of a Bitcoin-focused portfolio allocation):

  1. Risks that cause the investment to underperform Bitcoin (i.e. BPS not growing fast enough)
  2. Catastrophic risks

In the first category: premium depression, expense drag, and poor capital allocation by management could drag on BPS appreciation that could cause the investment to underperform the underlying Bitcoin. In general, these are resolved by effective management and allocation (including across Bitcoin bull and bear markets), and further mitigated by operating income.

But they are a headwind for all DATs, and I would expect that most DATs will fail to outperform their underlying on the basis of these issuesThere are also a variety of other risks and factors here — governance limitations, borrow pressure, regulatory or capital-structure overhangs, and more. These can constrain management’s ability to execute on desired strategies. For example, Rule 10b-18 can limit their ability to conduct buybacks, which could cause a discount to persist longer than it would otherwise, and endanger the company or drag down BPS. Again: this point is meant to give a general overview of DATs and their “sensibility,” not break down a complete diligence approach to an individual investment..

The second category is what skeptics jump to when they talk about DATs like Strategy — the “unwind” risk.

At the end of the day, the primary cause of such a scenario is convertibles coming due during a prolonged bear market, where the company has been unable to refinance its debt. This could lead to either insolvency on a cash basis, or a huge amount of new share issuance to satisfy the converts, depending on how they are structured.

A bear market by itself is not enough to kill a well-managed DAT — because if the convertibles are not callable, the company does not have an obligation to pay off the debt when prices are down. Between NAV going down from the underlying assets falling and the premium depressing (or even flipping to a discount), the company could end up in a situation where the debt load is large relative to equity value — but as long as the Bitcoin market returns, it is possible to manage through this.

Again: we are making an assumption throughout that Bitcoin will continue to appreciate over the long run. If that doesn’t happen, all bets are off, and the fundamental assumption underlying the investors’ bet in the DAT (that the underlying asset will appreciate, and they are considering the DAT allocation as part of their overall Bitcoin exposure) disappears.

This all goes back to the management quality point: success hinges on capital-markets execution, including sizing and laddering debt so maturities do not cluster in prolonged bear markets.

And it is what I mean when I say well-managed DATs can outperform the underling asset. It is not enough to be a DAT and for your underlying asset to perform — the company itself has to be well-managed.

Not a crazy concept! It is the same as literally any company. No matter how compelling a product or service, or how strong the secular tailwinds are, or how strong a position the company is in, if not well-managed, it will underperform (most companies do!).

Of course, different companies can be easier or harder to manage. I think DATs are on the “harder” side, but not even close to the hardest or most complex. I expect most DATs will fail to outperform their underlying asset, and eventually either liquidate and wind down, or be sold off to larger DATs — the cycle of life that occurs in so much of the equities market, but perhaps on accelerated timelines.

Where your work starts

If you’ve gotten this far, then DATs should make sense conceptually, and the question becomes: can the team manage the company well? Can they appropriately manage debt load and timing to avoid unwind risk? Can they ensure they stay ahead of their expenses to increase BPS?

This then goes to individual company diligence, beyond the scope of this post. All I set out to argue is that:

  1. To answer the question “do DATs actually make any sense?”
  2. One should consider them within the context of an overall allocation to the underlying asset, which means asking more precisely “Will this purchase outperform the underlying asset over my holding period?”
  3. And that the answer to that question is “Well-managed DATs can outperform their underlying asset, if that asset itself appreciates over time.”

And so you’ve arrived at my decisive answer to the original question: Do DATs actually make any sense? Sometimes.

Your task, once you’ve decided you’re long an underlying asset and are considered a more leveraged bet on it, is to decide whether a DAT is the best way to achieve that, and then diligence the relevant DAT companies, their management, your target holding period, and market conditions (same as any other investment), to assess whether they will be able to outperform.

In 2007, you could have decided that “mobile” as a category was set to explode. And if you diligenced and then bet on Apple or Samsung, you would have won big. If you bet on BlackBerry or Nokia or HTC or LG, you would have lost.

In the 2000s, you could have decided that “retail to ecommerce” as a trend was going big. If you bet on Walmart, you would have won. If you bet on Bed Bath & Beyond or Big Lots, you would have lost.

And in 2008, you could have decided that “solar stocks” were going to continue to rip. But by 2012, it didn’t matter which one you had bought — they were all down. The category blew up (in that timeframe).

And it is just so with DATs. Your first step is to correctly pick an underlying asset (i.e. “category” or “trend”) that is going to appreciate. If you get that wrong, all bets are off.

But then, being right on the asset is not enough — you then need to diligence and select the right company within that DAT category, and only if you select correctly do you have a chance of beating the underlying asset.

But hopefully I have convinced you that, even if only in a minority of cases, it is in fact possible.

Additional notes

Outside the core argument I’m making (that DATs sometimes make sense), here are some assorted notes, in no particular sequence.

Reflexivity

DATs, like most leveraged bets, are highly reflexive. A DAT that is outperforming and expected to continue to outperform (i.e. increase BPS faster than their competitors), will likely benefit from increased buying interest, and thus a higher premium to NAV, and thus an increased ability to issue ATMs at substantial premia (and converts on good terms, etc.) and increase their BPS even faster.

This is a positive feedback cycle, and a classic example of reflexivity.

DATs also suffer from negative reflexivity — if they are underperforming or expected to underperform, their premium will shrink, which will narrow management’s optionality for ATM offerings, converts, and other strategies. This means they will grow BPS more slowly, and the cycle will continue.

George Soros wrote about reflexive market dynamics like this in The Alchemy of Finance in 1987, before Bitcoin was even a twinkle in Satoshi’s eye. It’s not new to DATs. But it is something you must be mindful of.

One other consequence of the reflexivity of DATs is that the ability of the management team to market and promote the company plays a significant role in its success. Again, this is not unique to DATs, but it is perhaps amplified over more traditional types of companies.

Michael Saylor has done an exceptional job promoting Strategy and building its brand power. That has led to a higher NAV premium, which has allowed MSTR access to cheaper cost of capital and a positive reflexive loop. As Soros outlined in The Alchemy of Finance:

“I am not concerned with forecasting what will happen, but with understanding the reflexive interaction between perception and reality.”

He describes how prices are both cognitive (helping market participants understand) and participative (influencing the situation they are supposed to reflect). DATs are an excellent case study for this, in large part due to their “purity” as financial instruments.

As Benjamin Graham said:

“In the short run, the market is a voting machine but in the long run, it is a weighing machine.”

Valuation frameworks / “how much premium”

The best traditional-markets analogies I have found for DATs are agency mREITs and gold-royalty / streaming companies (both categories probably worth studying if you’re looking seriously at this space) — each are commonly considered to have an asset-value component and a premium component of their overall valuation.

For a DAT, the asset-value component is obvious, but what drives a premium? The heart of the premium is a combination of expectations for the underlying asset plus the issuer-specific question of whether BPS is anticipated to grow. And the latter is itself a combination of, at least:

  • Does the company have operating income?
  • Does the company manage expenses well?
  • Does the company have a strong brand (as in Helmer’s Brand Power)? If so, it will be able to access capital markets favorably compared to another issuer, ceteris paribus. This is a durable competitive advantage currently enjoyed by MSTR.
  • Does the company manage risk and debt load/timing appropriately?
  • Does the company execute ATMs, converts, and other strategies effectively to maximize BPS growth?

And, of course, this is all considered within the context of the underlying asset. No matter how good the management is, if the underlying asset is not something you want exposure to, then buying a DAT for that asset is almost certainly a bad idea.

Why are the PIPEs so big?

You often see huge dollar figures for new DATs — they go out with an “initial raise” from lots of big names. Some people have asked me “are all those people investing at a big premium to NAV?”

The answer is no. These initial investments are done via a PIPE (not an at-the-market offering), typically right around NAV. That means these investors — assuming the DAT launches with a premium — are immediately “in the money” on their investment due to the premium arbitrage. The companies need these PIPEs in order to build up an initial treasury to begin growing, and to pay expenses and build momentum. So they are willing to take on capital at or around NAV, with no premium, especially from noteworthy investors. A lot of these investors are benefiting from an immediate markup as the entity launches at a premium to NAV.

Additionally, in many cases, the DATs accept “in-kind” PIPE contributions of the underlying asset. Depending on the circumstances, these contributions can be tax-free typically via Section 351 exchanges if structured correctly. This means that an investor with a large pile of Bitcoin can contribute the Bitcoin in-kind for shares in the new DAT, pay no taxes on the contribution, and then as soon as the DAT goes live with a premium, be profitable (on paper, at least) on the trade.

The PIPE structures can get quite complicated with warrants and ratchets and other devices and incentives. But the key point to understand is that the PIPE investors (and thus the publicly-announced “initial financing”) is typically done in a way that minimizes NAV premium.

These dynamics, along with the hot market for DATs leading to consistent and meaningful premia, are what is driving the large initial contributions.

All DATs are not created equal

Aside from the considerations above about valuation frameworks, the relative quality of the underlying asset, and the quality of management, there are other differential qualities of DATs that you should consider when evaluating an investment. Two (of many) that I will highlight here:

  1. Structure. In order to get to the public markets quickly, DATs are taking a variety of paths, each with their own set of tradeoffs (which are beyond the scope of this post, but I at least wanted to mention this as a consideration). Some DATs are merging with a SPAC; some are doing a reverse takeover (RTO) and replacing management of an existing public company; some are negotiating a control transaction with an existing company and retaining management. These each come with wildly different upsides and downsides, and should be studied closely if considering an investment. Some of the considerations include: speed to market, Well-Known Seasoned Issuer (WKSI) status, management alignment risk, change-of-business considerations, trailing liabilities from the company’s prior operations, and more. I predict some messiness in the DAT market as a result of poorly-executed or -diligenced structures.
  2. Income. DATs have expenses (management, legal, accounting, etc.), which are a drag on BPS growth. If they can generate meaningful income in some way, that income can offset or exceed expenses, which then results in BPS growth. Significant income from operations can be a real de-risker and give the company more flexibility to service debt or pay expenses without eating into its asset treasury. Many new DATs have a staking or yield component, if the underlying asset allows it, and this can be appealing as well, because those income lines scale naturally with NAV.

But there’s no product

One objection I hear sometimes is “but these companies are mostly just financial instruments, there’s no meaningful product or service!” At some level, this is true (while they do have operating components, successful DATs will have their treasury value outpace operations value quickly). And if you only want to invest in products or services that are very directly serving some non-financial use cases, then that’s great.

But there are many “purely financial” assets out there, far beyond DATs — and there’s nothing wrong with investing in them if you want to. Financial products serve very real financial needs, and in this case, the needs it is serving include a form of leveraged Bitcoin exposure for holders, long-dated cheap senior exposure for convert holders, volatility arbitrage opportunities for traders, and more. Markets are coordination mechanisms, and DATs meet the needs of a range of participants.

For people who don’t like investing in “purely financial products,” DATs are probably not a fit. But again, this is not a new concept to DATs — there are many such companies and funds out there.

Formalism & Modigliani-Miller

An interesting paper came out recently: Valuing MicroStrategy. The key takeaway from their abstract:

assuming the firm is able to issue new debt at a premium to fair value during a “hype state”, thereby violating Modigliani-Miller conditions. This violation creates a “financing franchise”’ owned by shareholders that can be valuable enough to push the market value of equity above the market value of the assets.

Modigliani-Miller is an economic theorem which suggests that in the absence of a few factors (tax, bankruptcy costs, principal-agent divergence costs, asymmetric information, and market inefficiency), a company’s value should be unaffected by how it is financed.

What Andrade, Coomes, and Duarte (the authors of the paper) argue is roughly that so long as Strategy (and presumably other DATs) can issue convertible bonds at a premium to “fair price” in order to invest in Bitcoin and increase Bitcoin-per-share, they have actually created an underlying business (a “financing franchise”) that deserves to trade at a premium to NAV (and this is what violates Modigliani-Miller — which suggests that having a “financing franchise” should not affect the market value of the company, as financing structure should not affect value). They caution that the ability to do so is likely capped and limited temporally, and that a sudden evaporation could cause the premium to collapse.

Soros, as noted above, would likely argue that these markets slide away from efficiency in the first place, and so this is why it’s possible to violate Modigliani-Miller criteria and sell the converts above “fair price” — and so perhaps “fair price” is what the market clears at in the first place.

Looking for more to read?

Want to hear about new essays? Subscribe to my roughly-monthly newsletter recapping my recent writing and things I'm enjoying:

And I'd love to hear from you directly: andy@andybromberg.com