$132 million. Then, in the same disclosure, a second clause: XRP per share could decrease.
Most readers will skim past the second clause. It reads as a caveat. It is not a caveat. It is the load-bearing sentence of the entire transaction, and the arithmetic that reconciles it with the first clause is the only thing in this story worth quantifying.
I have spent sixteen years reading treasury structures, collateral schedules, and custody agreements, and the pattern is consistent. The headline tells you what the issuer wants you to feel. The per-share line tells you what the issuer expects to happen. Those two texts rarely agree. Here they openly disagree โ and the disagreement is not a contradiction. "Buying more XRP" and "XRP per share decreases" are compatible facts. They are, in fact, the same fact, observed from opposite ends of the same equation.
What follows is an attempt to reconstruct that equation from the fragments available. This is not a defense of Evernorth and not an attack. It is a specification of what must be true for the transaction to create value, and what must be true for it to destroy it. Four variables decide the outcome: ownership, exercise timing, purchase price, and share count. Three of the four are undisclosed.
Ledger integrity precedes market sentiment. So does arithmetic.
The Treasury Vehicle Problem
Evernorth is a digital asset treasury โ a DAT. The category is simple to describe and miserable to operate. A DAT is a listed equity shell whose balance sheet is dominated by a single crypto asset. It raises capital in public markets, converts that capital into the asset, and reports a per-share metric that functions as a substitute for earnings. For a Bitcoin treasury, the metric is BTC per share. For Evernorth, it is XRP per share.
The model has one source of genuine value creation and one source of illusion. The genuine source is accretion. When the company trades at a premium to the net asset value of its holdings, issuing new shares at that premium and buying more of the underlying asset raises the asset backing of every existing share. The illusion is that this process can continue indefinitely. It cannot. Accretion is a function of the premium, the premium is a function of market sentiment, and market sentiment is exogenous to the company's operations. A DAT does not create XRP. It buys XRP with other people's money and hopes the arithmetic holds.
Evernorth carries two complications that a Bitcoin treasury does not. First, its asset, XRP, occupies a residual regulatory category โ neither a clean commodity nor a settled security, a status that constrains how institutional capital can treat it and how the issuer can structure its own securities. Second, and more decisive, a spot XRP ETF now exists in the United States. That instrument delivers XRP exposure at a low fee, with no premium risk, no management discretion, no dilution mechanism, and no need to trust a treasury committee. The ETF did not merely compete with DATs. It removed the structural justification for paying a premium to a wrapper that performs the same function worse.
Set those two facts beside the $132 million funding option and the full context appears. A premium-dependent vehicle, holding a residual-category asset, in a market where a cheaper substitute already trades, is raising capital through a mechanism whose dilution effects depend on variables it has not disclosed. Everything that follows is an attempt to price that.
The Arithmetic of a 70% Threshold
The single most useful fragment is the reported 70% XRP allocation threshold. It has not been defined publicly, and definitions matter, but the number is too specific to be arbitrary. It is almost certainly a derived quantity, and there is one derivation that fits it exactly.
Build the model. A treasury holds X tokens of an asset and has S shares outstanding. X per share equals X divided by S. The asset trades at price P, so net asset value per share equals (X/S) times P. The stock trades at V, and the ratio mNAV equals V divided by NAV per share. That ratio โ market price over net asset value โ is the entire engine of the model, because it determines whether raising money helps or hurts the existing holder.
Now suppose the company raises F dollars by issuing new shares at the market price V. It must issue ฮS equals F divided by V new shares. Suppose a fraction p of the proceeds is used to buy the asset at price P. The company acquires p times F divided by P new tokens.
The existing holder is made whole only if asset backing per share does not fall. That condition is: new tokens acquired, divided by the new share count, must be at least the old tokens per share. Working through the algebra โ the intermediate steps are standard and the result is what matters โ the condition collapses to one inequality:
p must exceed 1 divided by mNAV.
Put the reported threshold into that slot. If p equals 0.70, then accretion requires mNAV to exceed 1 divided by 0.70, which is 1.43.
The 70% threshold is not a policy preference. It is the algebraic shadow of a required premium of roughly 43% over net asset value. When the stock trades above 1.43 times its XRP backing, issuing at market and deploying 70% of proceeds into XRP is accretive. Below that level, the same transaction dilutes. At mNAV of 1.10, the break-even allocation is not 70%. It is 91%. No realistic underwriting lets 91% of gross proceeds reach the asset, because fees, legal costs, audit costs, and any redemption obligations consume the difference first.
This is why the headline is not a contradiction. "Buying more XRP" describes the gross use of proceeds. "XRP per share decreases" describes the net effect on the existing holder. The transaction can do both at once, and it does both precisely when the premium is insufficient. The company's own language โ "could decrease," not "will decrease" โ tells you the model itself lands inside the dilution range. A company confident of accretion writes "will increase." A company that has run the numbers and found them marginal writes "could decrease."
I have encountered this exact hedge before. In 2024 I reviewed a trust conversion where the sponsor's technical brief described custody risk as "could be material under certain conditions." That phrasing existed because the sponsor's own stress test showed the conditions were not remote. Hedge words in disclosure documents are not neutral. They are the residue of an internal model that did not produce the desired answer. Hype evaporates; solvency remains โ and so does the word "could."
Why "Option" Is a Load-Bearing Word
The second fragment is the word "option." It is doing more work than the $132 million figure.
A funding option is not a funding commitment. It is the right, not the obligation, to raise capital on specified terms within a specified window. The category contains at least four instruments with radically different dilution profiles: a warrant, a convertible note, an at-the-market issuance program, and a contingent consideration arrangement. Each produces a different per-share outcome. None of the four is disclosed.
The accompanying wording โ "exercises," and the three-variable dependency on "ownership, exercise, and purchase" โ points toward a warrant or a convertible rather than a plain equity raise. That matters because the dilution mechanics invert between those instruments.
In a plain equity raise, the company controls the timing. It issues when the premium is favorable. Dilution is active and discretionary.
In a warrant, the holder controls the timing. The company grants a right to buy shares at a fixed strike for a fixed period, and the holder exercises when that right is in the money โ which is to say, when the company's stock has risen. The company cannot choose the moment. Dilution is passive and forced. This is the structural inefficiency no investor relations effort can remove: the warrant holder's optimal exercise point is the company's worst issuance point. Arbitrage exists only in structural inefficiency, and a warrant is a structural inefficiency by construction.
In a convertible, the dynamic is worse. The holder exercises into weakness as well as strength, because the conversion feature pays in volatility, not direction. A convertible holder who can short the underlying against the conversion right extracts value from the equity regardless of which way the stock moves. The company has sold optionality it cannot hedge.
The $132 million figure, standing alone, cannot be priced. Without the strike, the term, the anti-dilution provisions, and the lock-up, the number is a label, not a quantity. It tells you the size of a potential liability and nothing about the probability of it being drawn. A $132 million at-the-money warrant on a stock trading at 1.10 times NAV is a very different instrument from a $132 million deeply out-of-the-money warrant on a stock trading at 1.60 times NAV. One is near-certain dilution. The other may never be exercised.
This is where most coverage fails. The reporting treats "option" as a synonym for "raise." It is not. A raise is a completed transfer. An option is a transfer of decision rights from the company to a counterparty. The difference between those two things is the difference between a known dilution and an open-ended one.
The Flywheel's Critical Speed
Step back from the instrument and examine the machine it sits inside.
The DAT flywheel runs on five links. The stock trades at a premium. The company issues at that premium. It buys the asset. Asset backing per share rises. The premium is sustained by the accretion narrative, and the cycle repeats. Every link is a necessary condition. Break one and the machine stops.
The 70% threshold is the critical speed of that flywheel. It defines the minimum premium at which the machine turns forward instead of backward. Below 1.43, the same sequence runs in reverse: issuing at a discount to NAV, buying the asset, and watching asset backing per share fall. The machine does not fail gracefully. It fails by doing exactly what it was built to do, at a premium that is no longer sufficient.
This is a mechanical defect, not a moral one. It is worth stating plainly because most commentary on DATs oscillates between cheerleading and accusations of fraud, and neither is analytically useful. A flywheel that reverses below a threshold premium is not a Ponzi scheme. It is a leveraged vehicle with a variable cost of capital. The problem is that the vehicle's marketing presents the accretion case as the base case, when the algebra presents it as the conditional case. Accretion is not the default. Dilution is.
The recent history of the category confirms the mechanism. Through the second half of 2025, a large share of newly listed DATs traded below net asset value. For those vehicles, the flywheel had already reversed. Every issuance was dilutive. The rational response โ buy back shares below NAV, which is accretive in the opposite direction โ was rarely taken, because buybacks shrink the balance sheet and the narrative. The category chose scale over per-share value, which is the predictable choice when management compensation is tied to assets under management rather than to NAV per share.
That last point deserves emphasis because it is where incentives and shareholder interest diverge. A treasury manager paid on assets under management will always choose to grow the asset base, even when growth destroys per-share value. The 70% threshold, if it is genuinely binding, is evidence that someone inside Evernorth built the model and understood the difference. Whether the compensation structure respects that model is a question the disclosure does not answer.
Zero Operating Cash Flow
Here is the part of the DAT model that receives the least attention and deserves the most.
Evernorth has no operating business. It holds XRP. It may earn a small yield by lending or staking, but XRP's native staking yield is effectively zero. The company does not collect protocol fees. It does not run validators in any economically meaningful sense. It does not sell a service. Its entire return to shareholders is the price change of XRP multiplied by the number of XRP per share.
That means the sustainable-revenue test fails at the first hurdle. A business that derives approximately none of its value from operations and all of its value from the appreciation of a held asset is a closed-end fund with extra steps. The steps are: it pays management fees, it pays underwriting fees when it raises, and it accepts a premium or a discount in the market. Each step subtracts from the shareholder.
Compare the two ways to own XRP. The first is an ETF: a defined fee, daily creation and redemption at net asset value, no premium, no discount, no management discretion over capital structure, full liquidity in a brokerage account. The second is the DAT: a management fee, an issuance mechanism that dilutes when the premium is insufficient, a governance layer that can change strategy, and a market price that can diverge from NAV in either direction. The only reason to prefer the second is a belief that the management team can reliably issue above 1.43 times NAV and compound the difference. That belief is a bet on the team's market timing, not on XRP.
Stability is a calculated illusion. The DAT structure is stable only while the premium holds. The ETF structure is stable because it has no premium to lose.
The ETF Already Ate the Moat, and the 1940 Act Asymmetry
Two structural facts cap the upside of any XRP treasury vehicle, and both are absent from the coverage.
The first is substitution. Before a spot XRP ETF existed, a DAT offered something an investor could not easily obtain: regulated, brokerage-accessible XRP exposure in a familiar wrapper. That gap is gone. The ETF delivers the same exposure more cheaply and without the dilution mechanism. The DAT's remaining claim is "active accumulation," the promise that management will buy more XRP when conditions are favorable. But the disclosure at the center of this story โ XRP per share could decrease โ is an admission that active accumulation is not reliably accretive. Remove the accretion story and the DAT has no differentiated product. It is an expensive ETF with a governance discount.
The second is regulatory. A Bitcoin treasury can plausibly argue that it holds a commodity and is therefore not an investment company under the Investment Company Act of 1940. That argument is the load-bearing wall of the entire Bitcoin DAT category, and it is already contested. An XRP treasury has a weaker version of the same wall, because XRP's status is not settled. The 2023 ruling in the SEC's case against Ripple distinguished programmatic secondary sales from institutional sales. Programmatic sales were held not to be securities offerings. Institutional sales were not given that treatment. XRP therefore sits in a residual category: not a clean commodity, not a fully settled security, subject to a distinction that depends on how the tokens were sold.
For a company whose principal activity is acquiring and holding XRP, that residual status is a direct exposure. If the company's holdings were acquired through channels a regulator characterizes as institutional, the 1940 Act analysis becomes materially harder. A vehicle that fails the investment-company test faces registration obligations it was never designed to meet. This is not a hypothetical risk. It is the specific risk that distinguishes an XRP treasury from a Bitcoin treasury, and it is the risk the source material does not mention at all.
I flagged a version of this in 2024, when I reviewed a trust conversion for a competitor and found fourteen gaps in the custody and surveillance-sharing arrangements. The trust was approved anyway. The gaps did not disappear; they were priced by the market as negligible until the day they were not. Audits reveal what code conceals โ and what disclosure omits. Here the omission is the 1940 Act question, and it is not a small one.
What the Bulls Got Right
The bearish reading is easy and, in one respect, unfair. There is a bullish case, and it is stronger than the tone of the preceding analysis suggests.
The first point in its favor is that an option is optionality, and optionality has real value to the issuer when structured correctly. If the warrant strike is set above the level at which accretion begins โ above an mNAV of roughly 1.43 โ then the company has sold dilution only in states of the world where dilution is offset by the premium received. That is a well-designed instrument. It transfers risk to the counterparty who believes the premium will hold. A company that structures its option this way is not reckless; it is disciplined about the price at which it will accept new capital. The disclosure of a 70% threshold is consistent with that discipline, and discipline in a category that has mostly abandoned it is worth acknowledging.
The second point is that the demand-side signal for XRP is real, even if small. A $132 million purchase, if fully deployed, is a rounding error against XRP's daily spot volume, which routinely runs in the billions. But the signal is not the size of the purchase. The signal is the existence of an institutional buyer willing to publish a threshold, accept a per-share reporting obligation, and submit to the disclosure regime that comes with a listed vehicle. That is a different kind of demand than retail flow. It is slower, more durable, and less reflexive. Ledger integrity precedes market sentiment โ and so does institutional adoption. The order matters: infrastructure first, price second.
The third point is the one bears most often miss. The company chose a structure that preserves its downside. If the premium collapses, the option is simply not exercised. The company does not have to issue at a discount. The dilution the headline warns about is conditional on management choosing to draw the option in a bad state โ and a board that has published a threshold has given itself a reason not to. The instrument's optionality cuts both ways.
None of this rescues the equity from the arithmetic. But it does mean the correct judgment is "conditional and unverifiable," not "structurally doomed." The distinction is the difference between a forensic conclusion and a guess.
The Document That Isn't There
The only document that can settle this is the one not in front of us. A funding option of this size, attached to a listed vehicle holding XRP, must appear somewhere in the SEC filing record โ an 8-K, an S-4, a 424B prospectus supplement, or the warrant exhibits attached to one of them. Those documents will state the strike, the term, the anti-dilution provisions, the share count, and the allocation covenant. Until they are read, the $132 million is a label on an unopened box.
So the question is not whether Evernorth's XRP per share will rise or fall. The question is whether anyone publishing on this transaction has opened the box. The per-share line is the only number that matters, and it is the only number the coverage did not compute. That is not a gap in reporting. It is the reporting.
Precision is the only risk mitigation. Go find the filing.