# The Fungible Reserve Standard: A Deterministic Framework for Encoding Carrying Costs in Asset-Backed Tokens

Crypto.world

> Tokenizing physical assets such as vaulted gold, warehoused commodities, or insured fine art runs into a problem that tokenized treasuries do not have: these assets cost money to hold.

**arXiv**: 2606.26704

**Published**: 2026-06-25

**Authors**: JJ Jia Jing Tan, Eva Meng, Josh Ng, Zack Zhang, September Liu, Teelet Wang, Ludwig Zhang, Seth Yan

## AI summary

## Overview

**Research area:** Blockchain and crypto-asset research — specifically real-world asset (RWA) tokenization, token mechanism design, and smart contract economics. The paper is filed under Cryptography and Security (cs.CR) on arXiv (arXiv:2606.26704v1, 25 June 2026), authored by JJ Jia Jing Tan, Eva Meng, Josh Ng, Zack Zhang, September Liu, Teelet Wang, Ludwig Zhang and Seth Yan, with Matrixdock listed as an affiliation.

**Technical level:** Intermediate. Readers need basic familiarity with ERC-20 tokens, DeFi composability (DEXs, lending markets, oracles), and the idea of a fund expense ratio, but the mathematics is elementary algebra rather than advanced theory.

**Scope:** The paper proposes and analyzes the Fungible Reserve Standard (FRS), a token design in which a monotonically decreasing "asset-per-token" variable `q(t)` encodes the physical carrying costs of an asset-backed token while holder balances stay numerically fixed, preserving ERC-20 fungibility.

## What This Paper Is About

Tokenizing physical assets such as vaulted gold, warehoused commodities, or insured fine art runs into a problem that tokenized treasuries do not have: these assets cost money to hold. Custody, insurance, and audit fees create structural *negative carry*. Most existing tokenization projects either absorb those costs at the issuer level (making the token's economics opaque), rebase holder balances (breaking DeFi integration), or wrap the token in a fee-charging contract. The paper's goal is to specify a token standard that puts those carrying costs directly on-chain in a deterministic, publicly verifiable way without any of those compromises.

## Key Contributions

1. **A formal design principle — "economic purity."** The authors define the principle that a tokenized real-world asset should reflect the underlying asset's fundamental economics, including structural carrying costs, "without distortion, subsidization, or obfuscation," and decompose it into four requirements: R1 Reserve Integrity, R2 Deterministic Accounting Identity (`S(t) · q(t) = R(t)`), R3 Economic Isolation, and R4 On-Chain Cost Encoding.
2. **The Fungible Reserve Standard mechanism.** A token design where `q(t) = R(t)/S(t)` starts at `q(t₀) = 1` with `S(t₀) = R(t₀)`, and declines according to a predefined annualized carry rate `c` using a linear deduction `q(t_k + nΔt) = q(t_k) · (1 − c_k · n/N)` with `N = 365`. A supply reconciliation step mints new tokens (`ΔS`) to a designated `CarryCollector` address so that the accounting identity holds while existing holder balances stay unchanged.
3. **A proof that the linear formulation approximates ETF-style daily compounding.** Proposition 1 shows the linear FRS formula equals the first two terms of the binomial expansion of the ETF decay formula `a(t) = a₀ · (1 − r/365)^t`, with higher-order terms negligible for typical carry rates (`c ≤ 0.02`), while being cheaper to execute on-chain.
4. **A comparative and structural analysis.** Table 2 scores the FRS against issuer-subsidized, rebasing, and wrapper-token approaches across eight properties; Table 3 maps FRS terms onto traditional asset-management accounting (asset per share vs. asset per token, expense ratio vs. pure carrying cost rate, daily NAV accrual vs. daily `q(t)` update, periodic audit vs. on-chain verification plus reserve audit).

## Main Findings

- **The FRS is the only mechanism in the comparison marked as satisfying all eight evaluated properties.** Table 2 covers on-chain cost transparency, holder balance stability, ERC-20 DeFi composability, economic fidelity to the underlying, institutional accounting clarity, deterministic cost schedule, long-term sustainability, and reserve integrity. Issuer-subsidized models fail on cost transparency and economic fidelity; rebasing fails on holder balance stability and ERC-20 composability; wrapper tokens are marked "Partial" or "Varies" on most rows, with a cross on on-chain cost transparency, economic fidelity, and institutional accounting clarity.
- **Balance stability is the core differentiator for DeFi.** Because holder balances do not change numerically, the paper argues the token needs no custom integration with DEXs, lending protocols, price oracles, structured vaults, or cross-chain bridges. The paper's concrete illustration of the rebasing failure mode: a lending protocol records a deposit of 100 tokens, and after rebasing the balance falls to 99.97, creating accounting discrepancies.
- **Cost changes require strict checkpointing.** When the rate moves from `c_k` to `c_{k+1}`, all outstanding carry must first be applied at a boundary checkpoint — typically 12:00 AM (00:00:00 UTC) of the day immediately preceding the effective date — locking in `q(t_{k+1})` before the new rate applies only to subsequent days.
- **Precision must be handled deliberately.** Because the EVM uses integer math, the paper recommends 9 decimal places of scaling (compared with the 18 decimals standard for general ERC-20 tokens), arguing this keeps rounding truncation at "nominal fractions of a billionth of an asset unit" and avoids leaving unassigned "dust" when minting `ΔS` to the `CarryCollector`.
- **Reserve additions and redemptions are priced off the live `q(t)`.** Minting uses `ΔS_mint = ΔR₊ / q(t)` and burning uses `ΔS_burn = ΔR₋ / q(t)`, with the identity `(S(t) ± ΔS) · q(t) = R(t) ± ΔR` holding afterward.
- **The framework is asset-agnostic but conditional.** It applies to assets whose holding costs are structurally positive, time-dependent, and operationally verifiable — the paper lists vaulted precious metals, warehoused commodities, carbon instruments, and fine art/collectibles. It is explicitly *not* designed for positive-carry instruments such as tokenized Treasuries.
- **No empirical results are reported.** The paper contains no dataset, no benchmark, no simulation, no gas measurements, and no deployed-system metrics. All evaluation is conceptual and comparative; the authors list empirical analysis (simulations or pilots) as future work.

## Methodology in Plain English

The work is a design-and-argument paper rather than an experimental one. The authors first survey how traditional finance already handles negative carry — ETF expense ratios that accrue daily and are deducted from NAV so that gold backing per share declines while share counts stay fixed, futures contango reflecting cost of carry, and REITs passing through at least 90% of taxable income by regulation. They then survey on-chain precedents for encoding economics in protocol logic: Bitcoin's 21 million cap and 210,000-block halving schedule, and Ethereum's EIP-1559 base fee that adjusts and is burned. From these they extract a principle ("economic purity") and four requirements, then build a mechanism to satisfy them: shrink the asset-per-token number instead of touching balances. They check the mechanism's internal consistency with algebra — deriving the supply reconciliation formula and proving the linear decay is a tight first-order match to ETF compounding — and then argue its merits through a structured side-by-side comparison table against three rival designs. Finally, they walk through engineering edge cases (rate changes, integer rounding, mint/burn at the current `q(t)`) and enumerate limitations.

## Why This Matters

**Impact on research.** The paper reframes a tokenization question as an economic-accounting question: it argues that no prior framework formally specifies how negative carry should be encoded in a fungible token, and it supplies both a named design principle (economic purity) and a testable mechanism. That gives future work a vocabulary and a set of invariants — chiefly `S(t) · q(t) = R(t)` — that could be formally verified, simulated, or challenged.

**Real-world applications (as identified by the paper):**

- Vaulted precious metals — gold, silver, platinum, palladium in allocated custody with defined storage and insurance costs.
- Warehoused commodities — industrial metals, agricultural products, and raw materials in regulated warehouses.
- Carbon instruments — carbon credits and emissions allowances with registry, verification, and audit costs on externally defined schedules.
- Fine art and collectibles — insured, climate-controlled storage with periodic appraisal.

**Industry relevance.** The paper cites Boston Consulting Group's estimate of a $16.1 trillion tokenized asset market by 2030 and McKinsey & Company's projection of roughly $2 trillion by 2030 (with an optimistic scenario of $4 trillion), excluding cryptocurrencies and stablecoins, and notes that tokenized money market funds backed by U.S. Treasury securities are already live on public blockchains. Its pitch is to institutions: the mechanism is mathematically familiar from daily expense-ratio accrual but unbundles pure custody cost from management fees, and keeps `q(t)`, `S(t)`, and `c` queryable at any time instead of relying on a once-daily NAV publication and periodic audits. It also claims compatibility with RWA-specific infrastructure such as Plume Network, Polymesh, Canton Network, Centrifuge, and Provenance Blockchain, and with the ERC-1400 and ERC-3643 (T-REX) standards, by keeping economic design separate from compliance and identity layers.

## Future Directions

- **Governed dynamic cost rates.** Permitting `c` to be updated through governance while preserving transparency guarantees and giving advance notice — the current formulation assumes a fixed rate adjustable only at discrete intervals.
- **Continuous rather than epoch-based accrual.** Investigating mathematical models for charging carrying costs continuously instead of in discrete daily epochs.
- **Multi-asset reserves and cross-chain consistency.** Extending the framework to baskets of physical assets with heterogeneous carrying costs, and examining how the accounting identity can be maintained across multiple blockchains via bridges.
- **Formal verification and empirical work.** Applying formal methods to prove invariant preservation of `S(t) · q(t) = R(t)` in an implementation, and running simulations or pilots across varying market conditions, carry rates, and DeFi integration scenarios.

## Target Audience

Tokenization and RWA product designers, smart contract engineers building asset-backed tokens, and crypto researchers working on token mechanism design will get the most from the mechanism itself. The paper is equally aimed at institutional finance readers — compliance, fund accounting, and custody professionals — who need a token structure whose cost accrual maps onto expense-ratio logic and whose parameters are continuously auditable. DeFi protocol developers assessing whether an asset-backed token can be listed without custom rebasing adapters are a third natural audience.

## Authors’ abstract

The tokenization of real-world assets (RWAs) has emerged as a transformative application of blockchain technology, with market projections estimating trillions of dollars in tokenized assets within the coming decade. However, a fundamental challenge remains unaddressed: physical assets such as precious metals, stored commodities, and warehoused goods incur structural negative carry -- custody, insurance, and audit costs that accumulate over time. While existing tokenization…

[The full abstract on arXiv](https://arxiv.org/abs/2606.26704)

[Read the original paper](https://arxiv.org/abs/2606.26704)

---

Canonical: https://crypto.world/paper/the-fungible-reserve-standard-a-deterministic-framework-for-encoding-2606.26704
