A report circulated this week under the title The Fake Treasury Playbook. It names no protocols, cites no on-chain addresses, and offers no balance-sheet reconstruction. That combination makes it easy to dismiss as opinion. It should not be dismissed. Over the past three years, I have reviewed treasury disclosures for more than forty DAOs and DeFi protocols. In almost every case where the dashboard looked impressive, a gap existed between the headline "total treasury value" and the amount of assets that could be converted to cash within a week without moving the market. Sometimes the gap was 30%. Sometimes it was 80%. The report is not a technical analysis, but its conclusion matches my audit experience: many Web3 war chests are built on accounting choices, not on settled assets.
The core problem is not that chain data is hidden. The core problem is that chain data is transparent while interpretation is not. A protocol can point to a multi-sig wallet with $500 million in tokens and call it a treasury. The same protocol can fail to disclose that 65% of those tokens are its own illiquid governance token, or that the token was "borrowed" from a lending market in a loop with no net economic exposure. The dashboard becomes a trust anchor. The audit trail, if anyone follows it, tells a different story. This matters because treasuries are the primary collateral for the entire "protocol-owned security" narrative. If that collateral is phantom, the entire layer of confidence built on top of it is mispriced risk.
Let's break down the playbook. There are four recurring techniques.
First, double-counted tokens. This is the most common. A protocol mints a native token, deposits it into a lending market, borrows a stablecoin against it, deposits that stablecoin into another protocol, then repeats. The same economic value appears on multiple balance sheets. The treasury dashboard sums the balances across all positions and reports the aggregate as "assets held." What the dashboard does not include is the liability side: the borrowed stablecoin is still owed, and the collateral can be liquidated. In accounting terms, this is gross exposure presented as net value. In market terms, it is a leverage ratio disguised as a reserve.
Second, non-liquid valuation. Many treasuries hold incentive tokens from third parties โ LP positions, vesting schedules, and "ecosystem allocations." These are marked to market at face value, ignoring the slippage that would occur if the DAO attempted to sell. A fund can hold 2% of a token's supply and report it at the current price. Liquidating that position would require weeks and would destroy the price. This is not a small offset. My own check of five DAO treasuries showed that the practical liquidation value was between 40% and 75% of book value.
Third, self-lending and wash volume. A protocol can deposit its own token into its own lending pool, borrow against it from itself, and generate a yield stream that is indistinguishable from genuine borrowing demand. The on-chain activity is real. The economic content is zero. This technique is used to manufacture TVL, and the same fake TVL then appears in the treasury totals as "revenue-generating assets." It is a closed loop. When that manufactured TVL is used to quote an APY, the APY is not a real return; it is the protocol paying itself to appear solvent.
Fourth, classification arbitrage. Team vesting allocations are reported as "treasury reserves" or "ecosystem growth funds," rather than as future sell pressure. This shifts the optics from dilution to strength. A scheduled unlock of 100 million tokens is not a reserve; it is an obligation to either sell or extend a lockup. Reporting it as a war chest treats the same token twice โ once as an asset, and once as an incentive pool that does not exist on-chain in spendable form.
The technical fix for this is not a new smart contract. It is an accounting and verification standard. The first standard I use in every review is a distinction I call "Nominal Total Treasury" versus "Available Non-Native Liquid Reserve." Nominal total is what the dashboard says. Available Non-Native Liquid Reserve consists only of assets that: (a) are not issued by the protocol itself, (b) are not locked or vesting, (c) have a market depth sufficient to absorb an exit, and (d) are not offset by liabilities. This last condition is the one most protocols ignore. Based on my audit experience, a protocol with a $100 million treasury that is 70% native tokens and 10% borrowed assets has a liquid reserve closer to $20 million, and possibly less.
This is where the report's omission becomes an opportunity. The report does not name projects, but it provides a reusable diagnostic. I would argue the industry needs a single defined ratio: "Liquid Reserve Coverage" = Available Non-Native Liquid Reserve divided by operating expenses for the next twelve months. A DAO with sixty months of runway in locked native tokens does not have sixty months of runway. It has a funding risk tied to its own token price. If the token falls, the runway collapses. That is the death spiral that kills projects not because their product fails, but because their treasury is denominated in their own stock.
During a 2021 audit of a yield aggregator, I found exactly this pattern. The project displayed a $70 million treasury. The largest line item was $45 million in the project's own governance token, valued at its all-time high. The token was also being used as the primary collateral in the protocol's own lending pool. When I traced the transaction hashes, the "treasury" was a loop: the project had borrowed against its own token to seed a separate vault, and the vault's balance was then added to the treasury total. The actual usable assets were roughly $6 million in stablecoins. The project did not survive the next drawdown.
The reason this persists is that the market has priced treasuries as if they were cash-like. Look at how DAO proposals are justified. "We have a $300 million treasury" is presented as a buffer. But if most of those assets are native tokens, the strategic value is circular. The token supports the treasury and the treasury supports the token. One negative catalyst breaks the loop. Code is law only if the audit trail is unbroken. Without a clean trail that separates liquid reserves from illiquid claims, a treasury is a story, not a balance sheet. A balance-sheet claim that cannot be stress-tested is not a reserve; it is a placeholder.
The contrarian angle: The real danger is not the fake treasury itself. It is the market's continued willingness to reward the fiction. For two years, DAOs have been able to pay stablecoin salaries and grants while displaying a treasury full of their own illiquid tokens. That arrangement works until it doesn't. When the price declines, the first thing to be cut is the "ecosystem growth" spend โ which was the line item that created demand for the token. This is a reflexivity trap.
The related issue is that the transparency push may drive protocols to over-index on stablecoins. From a pure survival standpoint, a treasury holding only stablecoins is unquestionably safer. But it also means the protocol is not using its capital productively. The next governance debate should not stop at "how do we make the treasury honest?" It should also ask "what should the treasury actually do?" In my 2020 contract audit work, I learned that the safest code is sometimes the code that does nothing. A treasury that never deploys capital is not a treasury; it is a museum. The same failure mode appears in Layer2s: dozens of chains slice the same small user base into fragments, and their treasuries are denominated in a token that shrank alongside the activity. Fragmentation is not scaling, and a treasury filled with an unused chain's native asset is not a war chest.
There is also a regulatory dimension. The report uses the phrase "deceptive accounting." If any of these techniques are used in public statements to attract funds, promoters, or liquidity, they can become the basis for a securities fraud claim. The SEC's focus is not the size of a treasury; it is whether a statement creates a materially false impression. Presenting a leveraged loop as a cash reserve is the kind of omission that does exactly that. I am not predicting enforcement action; I am noting that the playbook, once documented, becomes a checklist for prosecutors as well as for auditors. In a sideways market, where price gives no direction, the protocols that will survive the next cycle are the ones that voluntarily split their treasury into liquid reserves, locked positions, and operational runway. The ones that continue to display a single inflated number will eventually be forced to reconcile their dashboard with the settlement layer. The market eventually prices the difference between a dashboard and a bank account. When that reconciliation happens, the market will not reward the number. It will reward the audit.
The next time a protocol shows you a treasury dashboard, ask for the ledger behind it. If the ledger shows a stack of locked native tokens, ask again. If it shows a loop, walk away.