Digital Asset Treasury (DAT) Companies

Updated 4 July 2026

Digital Asset Treasury (DAT) companies are publicly traded firms that hold, manage, and refinance significant crypto portfolios using clear thresholds like 700 BTC or market cap criteria.
They are evaluated using key metrics such as mNAV, tokens-per-share, and BTC intensity to gauge leverage, treasury composition, and risk exposure.
Research emphasizes dynamic treasury construction, risk hedging, and custody strategies to sustain operations during volatile crypto market conditions.

Digital Asset Treasury (DAT) companies are, in the recent literature, publicly traded firms whose primary corporate strategy is to hold significant cryptocurrency reserves on the balance sheet, or, more broadly, entities—often publicly listed—whose primary business is to hold, manage and (re-)finance a portfolio of crypto-tokens on the corporate balance sheet (Wen, 3 Nov 2025). Their equity commonly functions as a levered proxy on underlying token price movements through valuation premia over token net asset value, while treasury construction, refinancing structure, custody design, and disclosure policy determine whether that leverage remains durable across market regimes (Meister, 20 Jul 2025).

1. Definitions, scope, and typologies

The research literature uses closely related but not identical operational definitions of DAT companies. One strand defines a DAT company as a publicly listed firm whose corporate treasury holds a material amount of Bitcoin. To ensure economic significance, one empirical study includes any company meeting at least one of three thresholds as of April 11, 2025: it holds more than 700 BTC; it has a market capitalization at the latest BTC acquisition date exceeding \$1 billion; or its BTC holdings exceed 50% of its market capitalization (Aufiero et al., 20 May 2025). Another strand adopts a broader formulation: a DAT company is an entity—often publicly listed—whose primary business is to hold, manage and (re-)finance a portfolio of crypto-tokens on its corporate balance sheet (Meister, 20 Jul 2025).

Concept	Definition in the literature	Illustrative values
DAT company	Public firm with significant crypto reserves; or entity holding, managing, and re-financing crypto-tokens on balance sheet	39-firm BTC sample; 110-company illustrative DAT summary
Inclusion thresholds	Filters for economically material BTC exposure	$>700$ BTC; market cap $>\$1 $billion; or BTC holdings$ >50\% $of market cap</td> </tr> <tr> <td>Core valuation KPIs</td> <td>$ mNAV $, Tokens-per-Share, implied leverage when$ mNAV>1 $</td> <td>Median$ mNAV \approx 1.2\times $; range$ 0.8\times $to$ 4\times$</td> </tr> </tbody></table></div> <p>Within that broad category, the literature distinguishes pure-play BTC treasurers, multi-token treasurers, and hybrid corporates with an operating business plus a material crypto treasury. Meister’s illustrative classification defines pure-play BTC treasurers as firms holding at least 95% of assets in BTC, while multi-token treasurers may hold BTC together with <a href="https://www.emergentmind.com/topics/eigenstate-thermalization-hypothesis-eth" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">ETH</a>, stablecoins, or select alt-coins (<a href="/papers/2507.14910" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">Meister, 20 Jul 2025</a>). This broader framing matters because it shifts analysis away from a single-asset proxy model toward treasury composition, capital structure, and governance.</p> <p>Available descriptive datasets indicate both concentration and heterogeneity. One illustrative summary covers 110 DAT companies, of which 94 are in North America, 9 in Europe, and 7 in Asia. Aggregate Q2 2024 estimates in that summary are a combined market capitalization of approximately \$50 billion, total BTC held of approximately 110,000 BTC, and median $mNAV$ of approximately $1.2\times$ , with a range from $>\$1$0 to $>\$1$1. Token allocation is also concentrated: on average 88% BTC, 10% ETH or stablecoins, and 2% other alt-coins, with the top 5 tokens accounting for 96% of NAV (Meister, 20 Jul 2025). 2. Balance-sheet mechanics, leverage, and valuation A DAT company’s canonical business model is simple in outline but highly nonlinear in market effect: raise capital through equity or debt, convert proceeds into crypto, and allow the listed equity to trade as a levered proxy on asset price movements. In one formulation, $>\$1$2; in another, $>\$1$3. Both are used to capture how far public-market valuation exceeds the marked value of treasury tokens (Wen, 3 Nov 2025). When $>\$1$4, new equity issuance can be accretive in the narrow sense used by the literature: issue shares at the market price, buy additional tokens, expand NAV, and potentially reinforce the share price. Meister describes this as an equity-issuance loop, and extends the analysis to shareholder-level borrowing and company-level debt. In the reported sample, investors may pledge DAT shares as collateral with typical loan-to-value ratios of approximately 30–50%, haircuts of 20–40% due to concentration risk, and margin-facility interest rates of 5–8% per annum. At the company level, some firms issue unsecured or token-backed debt such as convertible notes; reported debt/equity ratios range from 0 to 0.3 with a median of approximately 0.1, and the leverage ratio defined as $>\$1$5 has a sample mean of approximately 0.12, skewed right with a few firms up to 0.4 (Meister, 20 Jul 2025). The same mechanism is fragile when public-market premia compress. In bear markets, $>\$1$6 can move toward or below 1, at which point equity issuance at depressed prices dilutes existing shareholders, debt facilities can become under-collateralized, and operating losses combined with mark-to-market declines can produce a vicious cycle of price decline, credit withdrawal, and forced sales (Wen, 3 Nov 2025). The literature therefore treats DAT valuation not merely as a statement about token holdings, but as a joint function of market access, credit conditions, refinancing optionality, and investor belief in the persistence of the premium. Illustrative time-series evidence reinforces that interpretation. A “typical pure-play” sample firm in Meister’s summary grows from 10 million shares and 1,500 BTC in 2020 at $>\$1$7 $>\$1$8 to 40 million shares and 20,000 BTC in 2024 at $>\$1$9 $>50\%$0, with an intervening peak of $>50\%$1 in 2023 (Meister, 20 Jul 2025). This does not establish a universal law of DAT growth, but it does show how share-count expansion and treasury accumulation can proceed together. 3. Market exposure, information flow, and hedging Empirical work on DAT equities treats them as assets with measurable crypto sensitivity rather than as purely idiosyncratic corporate claims. In a dataset of 39 publicly listed BTC-holding firms, daily logarithmic return regressions against BTC returns produce an average BTC beta of approximately 0.62, and 12 companies exhibit $>50\%$2 (Aufiero et al., 20 May 2025). The same study reports significant positive co-movements via Pearson correlations and identifies BTC as the dominant information driver through transfer entropy analysis, with only brief, announcement-driven feedback from stocks to BTC during major financial events. The exposure is not homogeneous across firms. Using estimated $>50\%$3, BTC intensity $>50\%$4, and trading-liquidity proxies based on the Amihud ratio, the 39-firm sample is partitioned into three groups: high exposure with $>50\%$5, medium intensity, and high liquidity; low exposure with low intensity and high liquidity; and low exposure with high intensity and low liquidity. Strategy, MARA, Riot, CleanSpark, and Hut 8 are listed in the first group, while Tesla, Coinbase, and Galaxy Digital are listed in the second, and Fold Holdings, Semler Scientific, and HIVE Digital in the third (Aufiero et al., 20 May 2025). This grouping has direct treasury and risk-management implications. The cited work argues against one-size-fits-all hedge ratios and recommends time-varying $>50\%$6 estimated via rolling OLS or Kalman filtering. It also recommends monitoring directional information flows using rolling transfer entropy and increasing hedges when $>50\%$7 is significantly above baseline, especially around major crypto-market events such as the January 2024 ETF approval or the April 2024 BTC halving (Aufiero et al., 20 May 2025). Firms with high treasury intensity and low liquidity are identified as facing the greatest price-impact risk and are advised to cap BTC exposure well below 50% of market capitalization. A recurrent misconception is that DAT equities are interchangeable expressions of crypto beta. The empirical literature instead shows that beta, treasury intensity, and liquidity jointly matter. This suggests that DAT analysis belongs as much to market microstructure and dynamic hedging as to simple balance-sheet accounting. 4. Bear-market survival and operating cash flow The bear-market literature reframes DAT sustainability as a cash-flow problem rather than a pure NAV problem. A sustainable DAT is said to require three elements: a conservative treasury policy, an operating line that monetizes crypto independent of asset prices, and satisfaction of a no-forced-sale runway condition over a target horizon $>50\%$8–24 months (Wen, 3 Nov 2025). The formal condition is $>50\%$9 where $mNAV$0 is initial cash and stablecoin reserves, $mNAV$1 is fee-based cash inflow from payment operations, and $mNAV$2 is total monthly outflow including OPEX, interest, and maintenance capex. Conservative policy in this framework has explicit operational content. The DAT should maintain 12–24 months of OPEX in cash or stablecoins, separate from the crypto treasury; favor long-dated bonds over short-term or margin loans; use optional hedging such as modest put positions or futures to insure up to a 50% drop; and cap the value-at-risk of any monetization sleeve at no more than 20% of the total cash reserve (Wen, 3 Nov 2025). These recommendations are intended to prevent treasury liquidation from becoming mechanically coupled to spot-price decline. The most developed proposed operating line is a “BTC-to-sats” payments rail on the Lightning Network. In that design, a DAT allocates 2–5% of BTC holdings to a liquidity sleeve used to open Lightning channels, while the core treasury of 95–98% remains untouched. Revenue streams are specified as an acquiring take-rate of 10–50 basis points of gross payment volume for merchant checkout services, routing fees on Lightning hops, and a few basis points of hedge or FX spread when sat-denominated settlements are converted into fiat on a $mNAV$3 basis. Settlement inventory is hedged back-to-back through futures or spot markets so that net payment P&L is largely independent of BTC mark-to-market (Wen, 3 Nov 2025). The same paper proposes an implementable disclosure regime so that investors can test whether operating cash flows bridge a 1.5–2 year bear. Suggested KPIs include monthly GMV or AUM, average take rate in basis points, total fee revenue $mNAV$4, payment <a href="https://www.emergentmind.com/topics/success-rate-sr" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">success rate</a> or loan repayment rate, routing revenue per 100,000 transactions or yield per asset unit, rebalancing cost in basis points, merchant or supplier churn, and the percentage of OPEX covered by non-mark-to-market revenue. The numerical example gives quarterly GMV of \$mNAV$51.25 million of fee revenue, routing revenue of \$mNAV$66 million, and OPEX coverage of 24% (Wen, 3 Nov 2025). Objectively stated, the controversy here is not whether bear markets matter—they do—but whether DATs should be evaluated primarily as balance-sheet vehicles or as operating businesses with treasury exposure. The cited framework argues for the latter without denying the former. 5. Custody, proof systems, and accountable treasury infrastructure As DAT balance sheets scale, custody and verifiability become research topics in their own right. One line of work introduces the Treasury Proof Ledger (TPL), a Bitcoin-anchored logging framework for multi-domain Bitcoin treasuries. TPL models the treasury as a conserved multi-domain exposure vector $mNAV$7 over domains such as cold wallets, exchanges, lending books, derivatives, and a distinguished fee-sink domain. It records proof-of-reserves snapshots, proof-of-transit receipts for movements between domains, and policy metadata, while supporting restricted views defined by stakeholder permissions (Puente et al., 3 Dec 2025). The TPL formulation is explicitly state-machine based. It treats append-only records, snapshots, and Bitcoin anchors as ledger transitions, and it introduces a balance-update rule with a fee sink. Under the stated update rule, and ignoring flows to or from external non-modelled parties, the sum of exposures across all non-fee domains is conserved over time. The deployment-level security notions are exposure soundness, policy completeness, non-equivocation, and privacy-compatible policy views. Importantly, the authors characterize the results as existence-type statements rather than claims that any current system already provides these guarantees (Puente et al., 3 Dec 2025). That qualification addresses a common misconception that proof-of-reserves alone is sufficient for public accountability. The practical deployment guidance in the same framework is also specific. It calls for domain decomposition so that all economically material exposures are assigned to some domain; independent proof-of-reserves providers, proof-of-transit operators, and anchoring key operators; stakeholder classes such as internal, board, auditors, regulators, and public; and infrastructure components including a TPL Engine, PoR Adapter, PoT Adapter, Anchor Service, Policy Engine and API, and append-only storage. Anchoring cadence can be daily or monthly; one anchor transaction embeds one 32-byte hash; and at 200 vbytes per transaction and 5 sat/vbyte, the cost is approximately 1,000 sats per transaction (Puente et al., 3 Dec 2025). A complementary line of work addresses custody under a rational-adversary model through decentralized custodian groups. Instead of one large multisig, a DAT can partition $mNAV$ 8 custodians into many small overlapping groups of size $mNAV$ 9, each controlling $mNAV>1$ 0 of total assets under an $mNAV>1$ 1-of- $mNAV>1$ 2 threshold policy. Reported constructions include a symmetric design with $mNAV>1$ 3, $mNAV>1$ 4, $mNAV>1$ 5; a polynomial design with $mNAV>1$ 6, $mNAV>1$ 7, $mNAV>1$ 8, $mNAV>1$ 9; and a $mNAV \approx 1.2\times$ 0 block design with $mNAV \approx 1.2\times$ 1. Security is parameterized through an efficiency factor $mNAV \approx 1.2\times$ 2, the maximal $mNAV \approx 1.2\times$ 3 ratio for which rational theft is unprofitable, where $mNAV \approx 1.2\times$ 4 denotes customer funds and $mNAV \approx 1.2\times$ 5 the uniform custodian deposit. Reported examples include $mNAV \approx 1.2\times$ 6 for the polynomial design and $mNAV \approx 1.2\times$ 7 for the block design (Chen et al., 2020). From an operational perspective, that decentralized custody scheme keeps per-transaction work local to one group. Assuming a threshold signature such as BLS, signing cost is $mNAV \approx 1.2\times$ 8, communication cost is $mNAV \approx 1.2\times$ 9 broadcasts, verification is $0.8\times$ 0 with aggregate signatures, and with $0.8\times$ 1 a threshold-signature round can finish in approximately 1–2 seconds on modern hardware (Chen et al., 2020). For DAT companies, the general implication is that verifiability and custody resilience can be engineered as first-class treasury properties rather than treated as back-office afterthoughts. 6. Reserve engineering, asset-backed tokens, and liability management Research adjacent to DAT balance-sheet management increasingly addresses treasury engineering for tokenized reserves and on-chain liabilities. One example is the Fungible Reserve Standard (FRS), proposed for asset-backed tokens representing real-world assets with structurally positive and predictable holding costs. FRS defines an asset-per-token variable $0.8\times$ 2, with continuous-time evolution $0.8\times$ 3 where $0.8\times$ 4 is the annualized carrying-cost rate. For gas efficiency it adopts a first-order daily approximation. The core accounting identity is $0.8\times$ 5 where $0.8\times$ 6 is total on-chain token supply and $0.8\times$ 7 is the physical reserve in custody. When $0.8\times$ 8 declines and $0.8\times$ 9 is unchanged, new tokens $4\times$ 0 are minted to a transparent CarryCollector address so that holder balances never change, every token’s economic backing is uniformly diluted, ERC-20 composability is preserved, and CarryCollector mint events remain fully on-chain and auditable (Tan et al., 25 Jun 2026). FRS is explicitly designed to avoid rebasing while bringing negative carry on-chain. Calibration of $4\times$ 1 proceeds by identifying recurring costs such as storage or custody fees, insurance premiums, audit and attestation costs, and financing or warehouse-line costs; expressing each as a percentage of asset value; verifying the data with custodians, insurers, and auditors; optionally incorporating a buffer such as +10 bps; and parameterizing the rate through governance for the next daily checkpoint at 00:00 UTC. The paper’s vaulted-gold example sets storage at 0.50% p.a., insurance at 0.20% p.a., audits at 0.05% p.a., yielding $4\times$ 2 p.a. It also gives examples of warehoused grains with $4\times$ 3 and carbon credits with $4\times$ 4 (Tan et al., 25 Jun 2026). For DAT operators, the significance is twofold. First, FRS links on-chain reserve accounting to off-chain expense accrual: each $4\times$ 5 mint is treated as an on-chain expense, $4\times$ 6 and $4\times$ 7 are queryable in real time, and auditors can match on-chain $4\times$ 8 and $4\times$ 9 to off-chain $mNAV$ 0 reports. Second, it is designed to integrate with DeFi infrastructure: AMMs use standard pair creation with oracles multiplying token price by $mNAV$ 1; lending markets update collateral valuations through oracle feeds incorporating $mNAV$ 2; staking or structured vaults require no special handling because balances remain steady; and compliance layering can be added through ERC-3643 or custom KYC modules (Tan et al., 25 Jun 2026). A related but distinct treasury problem arises for pegged digital currencies. In that setting, reserve management is modeled as a continuous-time control problem over cash reserves, short-dated government bills, peg deviation, and Hawkes intensities for redemptions and mints. Controls are the reallocation rate $mNAV$ 3 between cash and bills and the mint or burn spread fee $mNAV$ 4. The resulting stochastic MPC problem, analyzed with Pontryagin’s Maximum Principle, yields a bang-off-bang or soft-threshold structure: reallocation is inactive when the switching function is below threshold, becomes negative to build cash under rising stress, and saturates at $mNAV$ 5 in the limit of severe outflow forecasts (Hammerl et al., 13 Aug 2025). Simulation evidence in that reserve-control framework is precise. Under 92-day stress scenarios with three 8-hour windows per day and 100 Monte Carlo runs, the optimal policy produces total revenue of $mNAV$ 6 in the single-shock, prolonged-clustering, and false-alarm scenarios respectively, compared with $mNAV$ 7 for the max-yield benchmark and zero for the max-liquidity benchmark. Reported depeg performance is 1% frequency with response lag of approximately 1.8 days under the optimal policy, immediate 100% depeg under max-yield, and zero depeg but zero carry under max-liquidity (Hammerl et al., 13 Aug 2025). A plausible implication is that DAT companies issuing asset-backed or pegged liabilities require treasury rules that are state-dependent, auditable, and explicitly stress-responsive, rather than static reserve-allocation heuristics. Markdown Report Issue Upgrade to Chat References (7) 1. How Digital Asset Treasury Companies Can Survive Bear Markets: The Case of the Strategy and Bitcoin (2025) 2. Through the Looking Glass: Bitcoin Treasury Companies (2025) 3. Cryptocurrencies in the Balance Sheet: Insights from (Micro)Strategy -- Bitcoin Interactions (2025) 4. The Treasury Proof Ledger: A Cryptographic Framework for Accountable Bitcoin Treasuries (2025) 5. Decentralized Asset Custody Scheme with Security against Rational Adversary (2020) 6. The Fungible Reserve Standard: A Deterministic Framework for Encoding Carrying Costs in Asset-Backed Tokens (2026) 7. Optimal Control of Reserve Asset Portfolios for Pegged Digital Currencies (2025) Topic to Video (Beta) No one has generated a video about this topic yet. Sign Up to Generate All Videos Subscribe on YouTube Whiteboard No one has generated a whiteboard explanation for this topic yet. Sign Up to Generate Follow Topic Get notified by email when new papers are published related to Digital Asset Treasury (DAT) Companies. Sign Up to Follow Topic by Email Continue Learning How do DAT companies determine the thresholds for significant crypto holdings? What role does mNAV play in assessing the financial leverage of DAT companies? How do treasury composition and token diversification impact the risk profiles of DAT firms? What strategies are employed by DAT companies to hedge against market volatility and bear-market pressures? Find recent papers about decentralized custody mechanisms in crypto treasury management. Related Topics Real & Digital World Assets (RDWAs) Protocols for Loanable Funds (PLFs) Flow-Debt Trade-Off Dynamics Vault Credit Risk Metrics Overview Vault Credit Score Overview Hyperliquid Blockchain: Microstructure & Perpetuals FinTSFMs: Financial Time-Series Models Monetary Policy Expectations (MPE) Index Non-atomic MEV: Mechanisms & Implications No-Forced-Sale Condition in Market Mechanisms Content Overview References Topic to Video Whiteboard Follow Topic Continue Learning Related Topics Stay informed about trending AI papers: About Labs API Email Digest Chrome Extension RSS Terms Privacy Contact Twitter Discord

Concept

Definition in the literature

Illustrative values

DAT company

Public firm with significant crypto reserves; or entity holding, managing, and re-financing crypto-tokens on balance sheet

39-firm BTC sample; 110-company illustrative DAT summary

Inclusion thresholds

Filters for economically material BTC exposure

>700

BTC; market cap $>\$1

billion; or BTC holdings

>50\%

of market cap</td> </tr> <tr> <td>Core valuation KPIs</td> <td>

mNAV

, Tokens-per-Share, implied leverage when

mNAV>1

</td> <td>Median

mNAV \approx 1.2\times

; range

0.8\times

to

4\times$</td> </tr> </tbody></table></div> <p>Within that broad category, the literature distinguishes pure-play BTC treasurers, multi-token treasurers, and hybrid corporates with an operating business plus a material crypto treasury. Meister’s illustrative classification defines pure-play BTC treasurers as firms holding at least 95% of assets in BTC, while multi-token treasurers may hold BTC together with <a href="https://www.emergentmind.com/topics/eigenstate-thermalization-hypothesis-eth" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">ETH</a>, stablecoins, or select alt-coins (<a href="/papers/2507.14910" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">Meister, 20 Jul 2025</a>). This broader framing matters because it shifts analysis away from a single-asset proxy model toward treasury composition, capital structure, and governance.</p> <p>Available descriptive datasets indicate both concentration and heterogeneity. One illustrative summary covers 110 DAT companies, of which 94 are in North America, 9 in Europe, and 7 in Asia. Aggregate Q2 2024 estimates in that summary are a combined market capitalization of approximately \$50 billion, total BTC held of approximately 110,000 BTC, and median

mNAV

of approximately

1.2\times

, with a range from $>\$1$0 to $>\$1$1. Token allocation is also concentrated: on average 88% BTC, 10% ETH or stablecoins, and 2% other alt-coins, with the top 5 tokens accounting for 96% of NAV (Meister, 20 Jul 2025).

2. Balance-sheet mechanics, leverage, and valuation

A DAT company’s canonical business model is simple in outline but highly nonlinear in market effect: raise capital through equity or debt, convert proceeds into crypto, and allow the listed equity to trade as a levered proxy on asset price movements. In one formulation, $>\$1$2; in another, $>\$1$3. Both are used to capture how far public-market valuation exceeds the marked value of treasury tokens (Wen, 3 Nov 2025).

When $>\$1$4, new equity issuance can be accretive in the narrow sense used by the literature: issue shares at the market price, buy additional tokens, expand NAV, and potentially reinforce the share price. Meister describes this as an equity-issuance loop, and extends the analysis to shareholder-level borrowing and company-level debt. In the reported sample, investors may pledge DAT shares as collateral with typical loan-to-value ratios of approximately 30–50%, haircuts of 20–40% due to concentration risk, and margin-facility interest rates of 5–8% per annum. At the company level, some firms issue unsecured or token-backed debt such as convertible notes; reported debt/equity ratios range from 0 to 0.3 with a median of approximately 0.1, and the leverage ratio defined as $>\$1$5 has a sample mean of approximately 0.12, skewed right with a few firms up to 0.4 (Meister, 20 Jul 2025).

The same mechanism is fragile when public-market premia compress. In bear markets, $>\$1$6 can move toward or below 1, at which point equity issuance at depressed prices dilutes existing shareholders, debt facilities can become under-collateralized, and operating losses combined with mark-to-market declines can produce a vicious cycle of price decline, credit withdrawal, and forced sales (Wen, 3 Nov 2025). The literature therefore treats DAT valuation not merely as a statement about token holdings, but as a joint function of market access, credit conditions, refinancing optionality, and investor belief in the persistence of the premium.

Illustrative time-series evidence reinforces that interpretation. A “typical pure-play” sample firm in Meister’s summary grows from 10 million shares and 1,500 BTC in 2020 at $>\$1$7 $>\$1$8 to 40 million shares and 20,000 BTC in 2024 at $>\$1$9 $>50\%$0, with an intervening peak of $>50\%$1 in 2023 (Meister, 20 Jul 2025). This does not establish a universal law of DAT growth, but it does show how share-count expansion and treasury accumulation can proceed together.

3. Market exposure, information flow, and hedging

Empirical work on DAT equities treats them as assets with measurable crypto sensitivity rather than as purely idiosyncratic corporate claims. In a dataset of 39 publicly listed BTC-holding firms, daily logarithmic return regressions against BTC returns produce an average BTC beta of approximately 0.62, and 12 companies exhibit $>50\%$2 (Aufiero et al., 20 May 2025). The same study reports significant positive co-movements via Pearson correlations and identifies BTC as the dominant information driver through transfer entropy analysis, with only brief, announcement-driven feedback from stocks to BTC during major financial events.

The exposure is not homogeneous across firms. Using estimated $>50\%$3, BTC intensity $>50\%$4, and trading-liquidity proxies based on the Amihud ratio, the 39-firm sample is partitioned into three groups: high exposure with $>50\%$5, medium intensity, and high liquidity; low exposure with low intensity and high liquidity; and low exposure with high intensity and low liquidity. Strategy, MARA, Riot, CleanSpark, and Hut 8 are listed in the first group, while Tesla, Coinbase, and Galaxy Digital are listed in the second, and Fold Holdings, Semler Scientific, and HIVE Digital in the third (Aufiero et al., 20 May 2025).

This grouping has direct treasury and risk-management implications. The cited work argues against one-size-fits-all hedge ratios and recommends time-varying $>50\%$6 estimated via rolling OLS or Kalman filtering. It also recommends monitoring directional information flows using rolling transfer entropy and increasing hedges when $>50\%$7 is significantly above baseline, especially around major crypto-market events such as the January 2024 ETF approval or the April 2024 BTC halving (Aufiero et al., 20 May 2025). Firms with high treasury intensity and low liquidity are identified as facing the greatest price-impact risk and are advised to cap BTC exposure well below 50% of market capitalization.

A recurrent misconception is that DAT equities are interchangeable expressions of crypto beta. The empirical literature instead shows that beta, treasury intensity, and liquidity jointly matter. This suggests that DAT analysis belongs as much to market microstructure and dynamic hedging as to simple balance-sheet accounting.

4. Bear-market survival and operating cash flow

The bear-market literature reframes DAT sustainability as a cash-flow problem rather than a pure NAV problem. A sustainable DAT is said to require three elements: a conservative treasury policy, an operating line that monetizes crypto independent of asset prices, and satisfaction of a no-forced-sale runway condition over a target horizon $>50\%$8–24 months (Wen, 3 Nov 2025). The formal condition is

$>50\%$9

where $mNAV$0 is initial cash and stablecoin reserves, $mNAV$1 is fee-based cash inflow from payment operations, and $mNAV$2 is total monthly outflow including OPEX, interest, and maintenance capex.

Conservative policy in this framework has explicit operational content. The DAT should maintain 12–24 months of OPEX in cash or stablecoins, separate from the crypto treasury; favor long-dated bonds over short-term or margin loans; use optional hedging such as modest put positions or futures to insure up to a 50% drop; and cap the value-at-risk of any monetization sleeve at no more than 20% of the total cash reserve (Wen, 3 Nov 2025). These recommendations are intended to prevent treasury liquidation from becoming mechanically coupled to spot-price decline.

The most developed proposed operating line is a “BTC-to-sats” payments rail on the Lightning Network. In that design, a DAT allocates 2–5% of BTC holdings to a liquidity sleeve used to open Lightning channels, while the core treasury of 95–98% remains untouched. Revenue streams are specified as an acquiring take-rate of 10–50 basis points of gross payment volume for merchant checkout services, routing fees on Lightning hops, and a few basis points of hedge or FX spread when sat-denominated settlements are converted into fiat on a $mNAV$3 basis. Settlement inventory is hedged back-to-back through futures or spot markets so that net payment P&L is largely independent of BTC mark-to-market (Wen, 3 Nov 2025).

The same paper proposes an implementable disclosure regime so that investors can test whether operating cash flows bridge a 1.5–2 year bear. Suggested KPIs include monthly GMV or AUM, average take rate in basis points, total fee revenue $mNAV$4, payment <a href="https://www.emergentmind.com/topics/success-rate-sr" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">success rate</a> or loan repayment rate, routing revenue per 100,000 transactions or yield per asset unit, rebalancing cost in basis points, merchant or supplier churn, and the percentage of OPEX covered by non-mark-to-market revenue. The numerical example gives quarterly GMV of \$mNAV$51.25 million of fee revenue, routing revenue of \$mNAV$66 million, and OPEX coverage of 24% (Wen, 3 Nov 2025).

Objectively stated, the controversy here is not whether bear markets matter—they do—but whether DATs should be evaluated primarily as balance-sheet vehicles or as operating businesses with treasury exposure. The cited framework argues for the latter without denying the former.

5. Custody, proof systems, and accountable treasury infrastructure

As DAT balance sheets scale, custody and verifiability become research topics in their own right. One line of work introduces the Treasury Proof Ledger (TPL), a Bitcoin-anchored logging framework for multi-domain Bitcoin treasuries. TPL models the treasury as a conserved multi-domain exposure vector $mNAV$7 over domains such as cold wallets, exchanges, lending books, derivatives, and a distinguished fee-sink domain. It records proof-of-reserves snapshots, proof-of-transit receipts for movements between domains, and policy metadata, while supporting restricted views defined by stakeholder permissions (Puente et al., 3 Dec 2025).

The TPL formulation is explicitly state-machine based. It treats append-only records, snapshots, and Bitcoin anchors as ledger transitions, and it introduces a balance-update rule with a fee sink. Under the stated update rule, and ignoring flows to or from external non-modelled parties, the sum of exposures across all non-fee domains is conserved over time. The deployment-level security notions are exposure soundness, policy completeness, non-equivocation, and privacy-compatible policy views. Importantly, the authors characterize the results as existence-type statements rather than claims that any current system already provides these guarantees (Puente et al., 3 Dec 2025). That qualification addresses a common misconception that proof-of-reserves alone is sufficient for public accountability.

The practical deployment guidance in the same framework is also specific. It calls for domain decomposition so that all economically material exposures are assigned to some domain; independent proof-of-reserves providers, proof-of-transit operators, and anchoring key operators; stakeholder classes such as internal, board, auditors, regulators, and public; and infrastructure components including a TPL Engine, PoR Adapter, PoT Adapter, Anchor Service, Policy Engine and API, and append-only storage. Anchoring cadence can be daily or monthly; one anchor transaction embeds one 32-byte hash; and at 200 vbytes per transaction and 5 sat/vbyte, the cost is approximately 1,000 sats per transaction (Puente et al., 3 Dec 2025).

A complementary line of work addresses custody under a rational-adversary model through decentralized custodian groups. Instead of one large multisig, a DAT can partition $mNAV$ 8 custodians into many small overlapping groups of size $mNAV$ 9, each controlling $mNAV>1$ 0 of total assets under an $mNAV>1$ 1-of- $mNAV>1$ 2 threshold policy. Reported constructions include a symmetric design with $mNAV>1$ 3, $mNAV>1$ 4, $mNAV>1$ 5; a polynomial design with $mNAV>1$ 6, $mNAV>1$ 7, $mNAV>1$ 8, $mNAV>1$ 9; and a $mNAV \approx 1.2\times$ 0 block design with $mNAV \approx 1.2\times$ 1. Security is parameterized through an efficiency factor $mNAV \approx 1.2\times$ 2, the maximal $mNAV \approx 1.2\times$ 3 ratio for which rational theft is unprofitable, where $mNAV \approx 1.2\times$ 4 denotes customer funds and $mNAV \approx 1.2\times$ 5 the uniform custodian deposit. Reported examples include $mNAV \approx 1.2\times$ 6 for the polynomial design and $mNAV \approx 1.2\times$ 7 for the block design (Chen et al., 2020).

From an operational perspective, that decentralized custody scheme keeps per-transaction work local to one group. Assuming a threshold signature such as BLS, signing cost is $mNAV \approx 1.2\times$ 8, communication cost is $mNAV \approx 1.2\times$ 9 broadcasts, verification is $0.8\times$ 0 with aggregate signatures, and with $0.8\times$ 1 a threshold-signature round can finish in approximately 1–2 seconds on modern hardware (Chen et al., 2020). For DAT companies, the general implication is that verifiability and custody resilience can be engineered as first-class treasury properties rather than treated as back-office afterthoughts.

6. Reserve engineering, asset-backed tokens, and liability management

Research adjacent to DAT balance-sheet management increasingly addresses treasury engineering for tokenized reserves and on-chain liabilities. One example is the Fungible Reserve Standard (FRS), proposed for asset-backed tokens representing real-world assets with structurally positive and predictable holding costs. FRS defines an asset-per-token variable $0.8\times$ 2, with continuous-time evolution

$0.8\times$ 3

where $0.8\times$ 4 is the annualized carrying-cost rate. For gas efficiency it adopts a first-order daily approximation. The core accounting identity is

$0.8\times$ 5

where $0.8\times$ 6 is total on-chain token supply and $0.8\times$ 7 is the physical reserve in custody. When $0.8\times$ 8 declines and $0.8\times$ 9 is unchanged, new tokens $4\times$ 0 are minted to a transparent CarryCollector address so that holder balances never change, every token’s economic backing is uniformly diluted, ERC-20 composability is preserved, and CarryCollector mint events remain fully on-chain and auditable (Tan et al., 25 Jun 2026).

FRS is explicitly designed to avoid rebasing while bringing negative carry on-chain. Calibration of $4\times$ 1 proceeds by identifying recurring costs such as storage or custody fees, insurance premiums, audit and attestation costs, and financing or warehouse-line costs; expressing each as a percentage of asset value; verifying the data with custodians, insurers, and auditors; optionally incorporating a buffer such as +10 bps; and parameterizing the rate through governance for the next daily checkpoint at 00:00 UTC. The paper’s vaulted-gold example sets storage at 0.50% p.a., insurance at 0.20% p.a., audits at 0.05% p.a., yielding $4\times$ 2 p.a. It also gives examples of warehoused grains with $4\times$ 3 and carbon credits with $4\times$ 4 (Tan et al., 25 Jun 2026).

For DAT operators, the significance is twofold. First, FRS links on-chain reserve accounting to off-chain expense accrual: each $4\times$ 5 mint is treated as an on-chain expense, $4\times$ 6 and $4\times$ 7 are queryable in real time, and auditors can match on-chain $4\times$ 8 and $4\times$ 9 to off-chain $mNAV$ 0 reports. Second, it is designed to integrate with DeFi infrastructure: AMMs use standard pair creation with oracles multiplying token price by $mNAV$ 1; lending markets update collateral valuations through oracle feeds incorporating $mNAV$ 2; staking or structured vaults require no special handling because balances remain steady; and compliance layering can be added through ERC-3643 or custom KYC modules (Tan et al., 25 Jun 2026).

A related but distinct treasury problem arises for pegged digital currencies. In that setting, reserve management is modeled as a continuous-time control problem over cash reserves, short-dated government bills, peg deviation, and Hawkes intensities for redemptions and mints. Controls are the reallocation rate $mNAV$ 3 between cash and bills and the mint or burn spread fee $mNAV$ 4. The resulting stochastic MPC problem, analyzed with Pontryagin’s Maximum Principle, yields a bang-off-bang or soft-threshold structure: reallocation is inactive when the switching function is below threshold, becomes negative to build cash under rising stress, and saturates at $mNAV$ 5 in the limit of severe outflow forecasts (Hammerl et al., 13 Aug 2025).

Simulation evidence in that reserve-control framework is precise. Under 92-day stress scenarios with three 8-hour windows per day and 100 Monte Carlo runs, the optimal policy produces total revenue of $mNAV$ 6 in the single-shock, prolonged-clustering, and false-alarm scenarios respectively, compared with $mNAV$ 7 for the max-yield benchmark and zero for the max-liquidity benchmark. Reported depeg performance is 1% frequency with response lag of approximately 1.8 days under the optimal policy, immediate 100% depeg under max-yield, and zero depeg but zero carry under max-liquidity (Hammerl et al., 13 Aug 2025). A plausible implication is that DAT companies issuing asset-backed or pegged liabilities require treasury rules that are state-dependent, auditable, and explicitly stress-responsive, rather than static reserve-allocation heuristics.