Amplification and Convexity, in Plain English
The three numbers behind the share price
Start with the coins. If a company holds 848,000 bitcoin and has 384.23m shares, then each share owns 0.0022 bitcoin. That is bitcoin per share, and multiplying it by the bitcoin price gives the value of the coins behind one share. With bitcoin at $85,623.5 that is $188.97. Call it net asset value per share.
The share trades at whatever the market pays. If it trades at $162.54, the market is paying 86% of the value of the coins. That ratio, share price divided by net asset value per share, is the multiple to net asset value, written mNAV. Above 1.0 the shares are a premium to the coins; below 1.0 they are a discount. That is the whole vocabulary. Everything else in this lesson is what those three numbers do to one another.
Why it matters: those three numbers decide what the company can do. A premium lets the company sell new shares and buy more coins without short-changing existing holders. A discount makes new share issuance sting, and makes retiring older claims the better move.
Amplification is leverage, said plainly
Amplification is how much the share price moves when bitcoin moves. Suppose a company's coins are worth $100 and the common stock is worth $60. The coins are bigger than the equity behind them, because the rest of the balance sheet is funded by claims that rank ahead of the common. Now bitcoin rises 10%. The coins gain $10. The common, which started at $60, gains $10 before the senior claims are paid - a 17% move from a 10% move in the asset.
That is the whole idea. Senior claims - preferred stock, debt - are fixed. The common is the residual, and the residual absorbs all of the change. The more fixed claims sit ahead of the common, the sharper the common's swing, in both directions. Amplification is not a promise of more return; it is a statement that the common is the last in line and therefore the first to feel a move.
The trap for a new reader is to read amplification as safety. It is the opposite. Amplification means the share price is a geared claim on the coin, and gearing cuts both ways.
Convexity is a loop, not a curve
Convexity has a technical meaning in option pricing. In this structure it has a simpler, practical meaning: the company's ability to raise cheap money improves as its equity grows. That makes four steps feed one another.
A larger common equity means a deeper cushion under the preferred, so the preferred is safer and can be sold at a lower yield. A lower yield means more money raised for the same promise, so more coins are bought. More coins raise the amplification of the common, so a given bitcoin move lifts the share price more. A higher share price is a larger cushion, and the loop begins again. Each turn makes the next turn easier. That is what makes it convex rather than merely geared.
The loop is why a company in this structure wants its shares to be expensive rather than cheap. Under the older model the shares were the currency being spent, so a high price was a chance to issue. Under this model the shares are the collateral being advertised, so a high price is what makes the next credit sale cheap.
Worked example: the same company, both engines
Take a company with $100 of bitcoin and 10 shares, so $10 of coin per share. The share trades at $8, a 0.8 multiple. Now it wants to buy $20 more bitcoin.
Under the equity engine it sells $20 of new shares. That is 2.5 shares at $8. The company now holds $120 of coin across 12.5 shares: $9.60 per share, down from $10. The balance sheet grew 20% and each holder's claim on the coins fell 4%. That is the dilution the old scoreboard was designed to outrun.
Under the credit engine it sells $20 of preferred stock that pays a dividend and ranks ahead of the common. The company now holds $120 of coin across the same 10 shares: $12.00 per share, up 20%. The common keeps all of the coin. What it gave up is the front of the queue: the preferred is now entitled to its dividend before the common sees anything, and it is entitled to it whether bitcoin rose or fell. The common traded safety of claim for size of claim.
The cushion, and why the credit buyer cares
A buyer of the preferred is lending against the coins. Their question is not how high bitcoin can go; it is what is left for them if it falls. That residue is the cushion: the value of the common equity relative to the senior claims sitting ahead of it.
A thick cushion means the preferred is covered many times over. A thin cushion means the preferred is close to the collateral, and a large drawdown reaches it. This is why the same preferred dividend can look safe on one balance sheet and speculative on another, and why the size of the common is not a vanity metric in this structure. It is the collateral that makes the credit cheap.
The practical test a credit buyer runs: take the coins, subtract every claim that ranks ahead of the preferred, and ask how far bitcoin could fall before the preferred is impaired. The larger that number, the lower the yield the buyer will accept - and the cheaper the next round of coins.
What this model cannot do
It cannot make the preferred safe in the way a bank deposit is safe. A perpetual preferred is credit: cumulative, junior to debt, with no maturity at which the company must hand the money back. It is repaid only if the company chooses to buy it in or keep paying it. A price below face value is the market pricing that.
It cannot create demand for the credit by itself. The loop runs only while there are buyers for the yield. If demand thins, the company falls back on the equity engine and its dilution, or on cash flow, and the flywheel slows to whatever those can fund.
And it cannot repeal arithmetic. Senior claims are paid first. In a large enough drawdown the common is the instrument that goes to zero, and the preferreds are the instruments that get whatever is left. That is the price of the model, and it is worth stating plainly rather than discovering.
Next: Credit Creation