The question I get about bitcoin treasury companies is always some version of the same puzzle. The company holds a stack of bitcoin worth some amount, the market values the whole company at two or three times that amount, and the person asking wants to know what they are missing. Sometimes the answer is that the market is wrong and they are watching a dollar sell for two. But the premium has a real structure underneath it, and once you can see the structure you can at least decide for yourself when it is justified and when it is a bubble with a ticker.
The first step is to stop reading these things as tech companies. Most of them have some legacy operating business, software or mining or media, and in almost every case you can value that business generously and it still explains none of the gap. The stock is a wrapper around a pile of bitcoin plus a financing machine, so price the wrapper and the machine, and mostly ignore the product pages.
Building the mNAV number step by step
mNAV is just the market price of the company divided by the value of what it holds, but the details are where people get burned, so here is how I actually run it.
Start with the coins. Companies in this trade tend to disclose holdings aggressively, in filings and often on their own dashboards, because the whole pitch depends on you knowing the number. Multiply coins by spot price and you have gross bitcoin NAV. Add cash, and add whatever operating business exists at a sensible standalone value, which for most of these names is small enough that skipping it would not change your conclusion.
Then deal with the debt, which is where the calculation earns its keep. Most treasury companies fund purchases with convertible notes, and a convert is two different things depending on where the stock trades. If the notes are deep in the money, treat them as shares and put them in the diluted count. If they are out of the money, treat them as debt that has to be repaid or refinanced in cash. Getting this wrong flatters the number in exactly the scenario, a falling stock, where you most need it to be honest.
Apply the same discipline on the equity side. Use a fully diluted share count, including warrants and in-the-money converts, and check the size of any at-the-market issuance program, because the share count you see today is often a snapshot of a number that grows every week by design.
Now compute it two ways. The simple version is market cap divided by bitcoin NAV. The stricter version is market cap plus debt minus cash, all divided by bitcoin NAV, which is closer to asking what the market is paying for the whole capital structure per coin. The gap between the two versions tells you how much leverage is baked into the wrapper. At an mNAV of 1.0 you are paying spot price for the coins with extra steps. At 2.0 you are paying double, and the entire investment case rests on why that could ever make sense.
Where the premium actually comes from
Three mechanisms explain most of it, and operations are conspicuously absent from the list.
The first is leverage you cannot get yourself. These companies issue convertible debt at coupons an individual could never borrow at, sometimes near zero, because convertible arbitrage funds will pay up for the embedded option on a volatile stock. The equity holder ends up with levered bitcoin exposure and no liquidation engine attached, no margin calls at 3am. Debt maturities do function like margin calls in slow motion, though, and it pays to remember that.
The second is the issuance flywheel. When the stock trades above NAV, the company can sell new shares and buy more bitcoin per share than the sale dilutes. Sell a dollar of stock at an mNAV of 2 and the buyer receives a claim on roughly fifty cents of underlying bitcoin while the company adds a full dollar of new bitcoin to the pool, so coins per share rise for everyone who was already in. Companies report this as a yield, and a large part of the premium is a bet that the machine keeps running.
The third is access and index mechanics. An equity wrapper can go places the asset cannot. Index inclusion brings passive flows that buy without asking about valuation, and there are options chains, margin eligibility, and mandates or retirement accounts that can hold a listed stock but not spot bitcoin. Scarce routes to an exposure carry a toll, and part of the premium is that toll.
The useful thing about this list is that every item on it is checkable. You can read the debt terms, watch the issuance filings, and look up which indices a name sits in. None of it requires a view on anyone's vision.
What happens when the flywheel stalls
The mechanism is circular, and that cuts both ways. The premium enables accretive issuance, the issuance grows coins per share, and the growth in coins per share is the main justification for the premium. Run it backwards and the same loop drains itself. As mNAV falls toward 1, each share sold adds less and less bitcoin per share, the reported yield decays, and the reason to pay a premium decays with it.
Below 1 the logic inverts entirely. Issuing shares to buy bitcoin destroys value per share, and the accretive move becomes selling bitcoin to buy back stock, which is the opposite of everything the company promised. Meanwhile any converts that never converted are now just bonds, and they come due in cash. A company with no meaningful operating cash flow has three ways out at maturity, refinance the paper, issue equity at a discount to NAV, or sell coins, and all three hurt, so the market starts pricing them in well before the maturity date arrives.
There is a historical rhyme worth keeping in mind here. GBTC traded at a persistent premium for years while it was one of the few convenient ways to hold bitcoin in a brokerage account, and plenty of smart people treated that premium as structural. Then better wrappers arrived, the premium flipped to a deep discount, and it stayed there for a long time. Premiums on wrappers survive only as long as the wrapper is the scarce path to the exposure, and scarce paths have a habit of getting crowded.
How I actually price one
My working rule is to treat every one of these as a leveraged bitcoin position with a fee attached, then ask whether the fee is worth paying. In practice that looks like this.
- Compute NAV per share fully diluted, treating in-the-money converts as shares and everything else as debt.
- Compute mNAV both ways, simple and debt-adjusted, and note the gap as the leverage inside the structure.
- Translate the premium into a hurdle. Paying 1.5x means coins per share need to grow roughly 50 percent just for you to break even against having bought spot, so ask whether issuance capacity and premium durability can plausibly get there, and over what horizon.
- Map the debt maturities against a flat or down bitcoin scenario and decide who refinances that paper, and on what terms.
- Compare the whole package against the boring alternative, spot or an ETF, with modest leverage on top if leverage is the point. The treasury company has to beat that after the premium you paid, or there was no reason to own it.
The failure mode I see most often is people treating the reported bitcoin yield like an operating growth rate and putting a compounding multiple on it. That yield is manufactured by selling stock above NAV to new buyers, so it exists only while the premium exists, and using it to justify the premium is the loop eating its own tail. I keep a scorecard of a few of these names against spot bitcoin in Blockcircle, and splitting the return into coin performance and wrapper performance is usually the fastest way to see what I am actually paying for.
None of this says the premium has to collapse tomorrow, or ever. Flywheels can run for a long time, and the leverage and access arguments are real. It just means you should know the mNAV you are paying, know which of the three mechanisms is doing the work, and have a plan for the day the issuance machine slows down. If the only sentence you can write about why the wrapper deserves its premium ends up including the premium itself, buy the coin instead.