Skip to content

Instantly share code, notes, and snippets.

@dangell7
Last active July 15, 2026 15:56
Show Gist options
  • Select an option

  • Save dangell7/7abc0620eff968dddcb450f4a768ffc6 to your computer and use it in GitHub Desktop.

Select an option

Save dangell7/7abc0620eff968dddcb450f4a768ffc6 to your computer and use it in GitHub Desktop.
IOU vs MPT on the XRPL — ecosystem review and client decision guide (July 2026)

IOU vs MPT on the XRPL — Ecosystem Review and Client Decision Guide

Date: July 14, 2026 Scope: Protocol comparison (grounded in rippled source), XRPL ecosystem adoption 2024–mid-2026, tradfi/RWA tokenization landscape, and a client-facing decision matrix. Thesis under test: "the future is MPT."


1. Executive summary

  • MPT (XLS-33, Multi-Purpose Tokens) is the institutional token standard on XRPL. Its flag set — RequireAuth allowlisting, per-holder and global lock, clawback, fixed supply caps, integer amounts, on-chain metadata, per-issuance transfer fees — maps one-to-one onto what regulation now requires (GENIUS Act seize/freeze/burn, MiCA EMT controls, SEC transfer-agent regimes).
  • IOUs (trust-line tokens) remain the only standard with on-ledger liquidity. DEX offers, AMM pools, cross-currency payments, and pathfinding are IOU/XRP-only today. MPT DEX/AMM/path integration (MPTokensV2 / XLS-82) is still Supported::No in the codebase — not even votable on a production network yet.
  • Consequently the split today is: liquidity → IOU, control → MPT. Every dollar-weight asset on XRPL (RLUSD ~$1.5–1.6B, Ondo OUSG, Guggenheim DCP, OpenEden TBILL, Braza USDB) is a trust-line token — all launched before MPTokensV1 activated on mainnet (October 1, 2025). MPT-native issuers so far are early-stage (FortStock warehouse receipts, LiCuido MMF-share collateral, Centigrade carbon credits).
  • The protocol roadmap is unambiguously MPT-first. Vault shares (XLS-65) are MPTs; the Lending Protocol (XLS-66) accounts in MPT integer amounts; Confidential Transfers are MPT-only; DynamicMPT mutability is MPT-only. The IOU surface receives only fix-amendments. Once MPTokensV2 delivers DEX/AMM/paths, the last structural reason for a new issuer to choose IOU disappears for most institutional use cases.
  • What IOUs keep permanently: rippling / credit-network semantics, quality in/out, and the ability for balances to net through the trust graph. That is a real (if niche) feature set — gateway/FX/credit-network designs — that MPT deliberately does not replicate.

2. Protocol feature matrix (code-grounded, rippled develop, mid-2026)

Feature IOU (trust line) MPT Notes / gate
Holder ledger object RippleState (shared, bilateral) MPToken (per-holder, unidirectional) ledger_entries.macro
Amount representation 15-digit decimal float (STAmount) 63-bit unsigned integer + AssetScale kMaxMpTokenAmount; accounting-exact
Hard supply cap No Yes (sfMaximumAmount vs sfOutstandingAmount) MPT-only
Transfer fee Account-wide sfTransferRate (all IOUs of the issuer) Per-issuance sfTransferFee 0–50% one account, many differently-priced MPTs
Authorization / allowlist Authorized trust lines (lsf*Auth) lsfMPTRequireAuth + MPTokenAuthorize equivalent capability
Freeze / lock Individual, Global, DeepFreeze; NoFreeze renounce footgun Per-holder lock + whole-issuance lock (lsfMPTCanLock) GENIUS-grade control
Clawback XLS-39 (blocked if issuer set NoFreeze) lsfMPTCanClawback, opt-in at issuance
Direct payments Yes Yes (MPTokensV1)
Cross-currency / paths / rippling Yes No — gated on MPTokensV2 Payment.cpp rejects paths for MPT
DEX offers / AMM Yes No — gated on MPTokensV2 (Supported::No) OfferCreate.cpp, AMMCreate.cpp; lsfMPTCanTrade reserved
Checks Yes (incl. conversion) SendMax yes; conversion needs V2
Escrow Yes (TokenEscrow, live on mainnet Feb 2026) Yes (lsfMPTCanEscrow, TokenEscrow)
Payment channels No (XRP-only) No (XRP-only)
Vault shares (XLS-65) underlying only Share issuance IS an MPTokenIssuance VaultCreate.cpp
Lending (XLS-66) via vault underlying native — loan amounts bounded by MPT max
Confidential balances No Yes (ConfidentialTransfer, lsfMPTCanHoldConfidentialBalance) in code; not yet on mainnet
Post-issuance mutability account-level flags only DynamicMPT: enable flags, mutate metadata/fee (opt-in, one-way) in code; mainnet pending
On-chain metadata None (off-ledger conventions) 1–1024 bytes JSON (XLS-89 schema) machine-readable instrument data
Per-product configuration account-level → one issuing account per product per-issuance → one account, many products operational simplification
Holder reserve +1 owner reserve only if line non-default always +1 per MPToken (+1 issuance for issuer) IOU can be cheaper for holders
Quality in/out Yes No credit-network pricing

Amendment status snapshot (mid-July 2026)

Live on mainnet: MPTokensV1 (Oct 1, 2025), Credentials/XLS-70 (Sep 2025), PermissionedDomains/XLS-80 (Feb 4, 2026), TokenEscrow/XLS-85 (Feb 12, 2026), PermissionedDEX/XLS-81 (Feb 18, 2026), DeepFreeze, AMMClawback. (Enabled/not-enabled status verified directly against the on-ledger Amendments object at validated ledger 105,599,927, July 14, 2026.)

Votable/shipped but not enabled: SingleAssetVault/XLS-65 and LendingProtocol/XLS-66 (open for voting since ~June 30, 2026; reported at only ~23% validator support at last check), DynamicMPT/XLS-94.

In code but Supported::No (dev-only, not votable): MPTokensV2 (the entire MPT DEX/AMM/path surface — XLS-82), LendingProtocolV1_1. Confidential Transfer (XLS-96) is Supported::Yes on develop but has not appeared on mainnet; Ripple's Q1-2026 roadmap target slipped.


3. Ecosystem reality check (who actually uses what)

All marquee XRPL RWA/stablecoin value is IOU today — verified on-ledger, July 14, 2026. Each issuer account below was checked directly against mainnet public RPC (account_objects type=mpt_issuance + gateway_balances): zero MPT issuances across all five; every one is a classic trust-line token. Every one of these launched before MPTokensV1 activated, and none has migrated.

Asset Issuer On-ledger obligations (Jul 14, 2026)
RLUSD (NYDFS trust charter, ~$1.5–1.6B total mcap) rMxCKbEDwqr76QuheSUMdEGf4B9xJ8m5De 867.8M RLUSD on XRPL — up sharply from ~$340M at end of Q1 (Messari); supply is migrating toward XRPL, the rest is ERC-20
Ondo OUSG (24/7 mint/redeem settled in RLUSD) rHuiXXjHLpMP8ZE9sSQU5aADQVWDwv6h5p 1.92M OUSG
OpenEden TBILL — ticker TBL on XRPL (first tokenized T-bills on XRPL, Aug 2024) rJNE2NNz83GJYtWVLwMvchDWEon3huWnFn 9.92B TBL (mostly held in OpenEden's own labeled operational accounts)
Guggenheim DCP (Moody's P-1, QIB/QP-only, $280M+ program) rnXNUqF8LuXyUdfLdezftTeEUw2UFSLw5i 40.36M across GDCP series — matches rwa.xyz supply exactly
Braza USDB ($1B+ payments by Aug 2025) rB3y9EPnq1ZrZP3aXgfyfdXQThzdXMrLMc 124.1M USDB

The Guggenheim structure deserves a call-out: DCP is issued as one trust-line currency code per maturity date (GDCP20250617GDCP20260604, matured series zeroed out). That is the workaround for IOUs having no per-issuance configuration — an entire currency code minted per instrument series — and it is precisely the limitation MPT's per-issuance model (one account, many independently configured issuances with maturity metadata on-ledger) exists to fix. The flagship institutional IOU deployment is itself an argument for MPT. Note also that copycat vanity issuers exist for both OUSG and TBILL on XRPL — the addresses above are the verified ones.

MPT-native adoption is early but directionally telling: FortStock (warehouse receipts / commodity inventory — metadata carries commodity type, storage location, expiry, audit trail), LiCuido (MMF shares as programmable lending collateral), Centigrade (carbon credits). These are exactly the instrument-like, transfer-restricted, metadata-heavy use cases MPT was designed for.

XRPL RWA trajectory: $24.7M → $567.9M during 2025 (+2,200%), ~$2.25–2.3B by end of Q1 2026 (+124% QoQ, top-7 chain), ~$3B by May 2026 including RLUSD. The growth happened on trust lines — the standard didn't block adoption; it just means the migration story is still ahead.

Competitive frame: MPT is XRPL's answer to Hedera HTS and Ethereum's ERC-3643/T-REX. Its strengths are protocol-native enforcement (no smart-contract audit surface), sub-cent fees, and ledger-level compliance flags. Its acknowledged weakness is a fixed rule set — if a client needs authorization logic that doesn't fit the allowlist model, there is no escape hatch, whereas ERC-3643 is modular and is converging into a cross-chain (Hedera, Stellar SEP-57) ISO-track standard. Canton owns the private interbank tier (JPMD, DTCC, Tradeweb DvP); XRPL's play is the public-chain regulated issuance middle ground.


4. Why regulation makes MPT the default (the tradfi pull)

The 2025–2026 regulatory wave effectively specifies a token standard:

  1. GENIUS Act (US, signed July 18, 2025; rulemaking through July 2026): permitted stablecoin issuers must have the technological ability to seize, freeze, or burn tokens and block transfers on primary and secondary markets under lawful order, plus 1:1 reserves with monthly attestations. That is lsfMPTCanLock + lsfMPTCanClawback + a provable, capped supply — statutory table stakes. IOU freeze/clawback exist, but with account-global semantics and the NoFreeze renunciation footgun; MPT's per-issuance opt-in flags are a cleaner compliance attestation.
  2. MiCA (fully in force EU-wide June 30, 2026): EMT issuers need licensing, 1:1 backing, and issuer controls; only ~8 compliant euro stablecoins exist. Ripple holds a Luxembourg CASP authorization (July 6, 2026) stacked on its EMI licence.
  3. SEC (Jan 2026 staff guidance + Project Crypto innovation exemption): tokenized securities are still securities; transfer-agent regimes and Reg D/3(c)(7) allowlisting remain central. Funds like BUIDL ($2.3–2.6B), BENJI ($2B), USTB ($950M), ACRED are all allowlist-gated, TA-managed, low-velocity — they do not need a public DEX; they need RequireAuth, lock, clawback, auditability, and exact integer share accounting.
  4. What institutions ask for (RedStone/Credora 2026 standards report, Fireblocks): transfer restrictions, identity-bound allowlists, fungible-but-permissioned mechanics, corporate-action support, machine-readable metadata, deterministic finality, predictable fees, custody integration. MPT covers the list natively; IOUs cover parts of it via conventions and account-level flags.

The tokenized-RWA market this feeds is ~$26.7B distributed value (June 2026, rwa.xyz conservative anchor) with tokenized treasuries/MMFs at $15B+ and stablecoins at ~$290B — and bank deposit-token networks (JPMD on Base/Canton; JPM-Citi-BofA-WFC + TCH shared network targeting H1 2027) closing in from the tradfi side.


5. Client decision matrix — when to use which

Client / use case Today (mid-2026) After MPTokensV2 Why
Regulated USD/EUR stablecoin (retail, DeFi-composable) IOU (RLUSD precedent) MPT Needs DEX/AMM liquidity today; GENIUS controls map to MPT flags; migrate when V2 lands
Tokenized MMF / treasury fund (Reg D, 3(c)(7), QP-only) MPT MPT Allowlist-gated anyway; no public-DEX need; fixed supply + integer shares + metadata; vault/lending native
Commercial paper / bonds / private credit MPT MPT Capped issuance, maturity metadata, QIB allowlist, escrowed settlement, clawback for corrections
Vault shares, lending collateral MPT (protocol-forced) MPT XLS-65 shares are MPTs; XLS-66 accounts in MPT amounts
Bank deposit token / tokenized deposits MPT MPT Closed allowlist, no rippling exposure, confidential balances coming, per-holder lock
Commodities, warehouse receipts, carbon credits, tickets/loyalty MPT MPT On-chain metadata is the product; integer units; lock/clawback; FortStock/Centigrade precedent
Community / retail / meme / gaming token IOU IOU or MPT Liquidity, AMM incentives, wallet+tooling maturity, no compliance burden
FX corridors, gateway networks, credit/netting designs IOU (permanently) IOU Rippling, quality in/out, bilateral credit are the feature — MPT deliberately omits them
Anything needing cross-currency paths or on-ledger price discovery now IOU MPT becomes viable The entire path/DEX/AMM engine is IOU/XRP-only until V2

Rule of thumb: if the token is a bearer-style currency that must trade freely on-ledger today, issue an IOU. If the token is an instrument — a claim with an issuer, a register, a rulebook — issue an MPT.


6. The "future is MPT" thesis — verdict and caveats

Supported. The direction of protocol investment (vaults, lending, confidential transfers, dynamic issuance, permissioned domains/DEX all MPT-oriented), the direction of regulation (issuer control as statute), and the direction of institutional demand (instrument-like, allowlisted, metadata-rich products) all converge on MPT. IOUs are in maintenance mode at the protocol level.

Caveats that gate the timeline:

  1. MPTokensV2 is the linchpin and it is not closeSupported::No means it cannot even be voted on yet. Until then MPT tokens have no on-ledger secondary market, which is disqualifying for stablecoins and any freely-traded asset.
  2. Validator appetite is a real constraint — XLS-65/66 sitting at ~23% approval shows amendments no longer sail through; the MPT roadmap is hostage to UNL politics.
  3. No migration precedent exists. Nobody has moved an IOU to an MPT; RLUSD staying an IOU is the strongest signal that liquidity outweighs control for currency-like assets. Expect coexistence, not replacement: IOU as the currency standard, MPT as the securities/instrument standard — with MPT absorbing more of the pie as V2, DynamicMPT, and confidential transfers land.
  4. Cross-chain standardization risk: ERC-3643 is becoming a multi-chain ISO-track standard while MPT is XRPL-only; issuers with multi-chain strategies may treat XRPL/MPT as one deployment target rather than the system of record.

Appendix: research provenance

Synthesized from three parallel research passes (July 14, 2026): (a) rippled source audit (features.macro, MPT transactors, Payment/OfferCreate/AMM gating, VaultCreate, lending); (b) XRPL ecosystem adoption sweep (bithomp amendment records, Ripple press, Messari Q1-2026 State of XRP, xrpl.org docs, CoinDesk/Cryptopolitan); (c) tradfi/RWA landscape (rwa.xyz, GENIUS Act text + OCC/FinCEN/Treasury 2026 rulemakings, MiCA/ESMA, SEC Jan-2026 tokenization guidance, RedStone/Credora standards report).

A fourth pass (July 14, 2026) verified the load-bearing claims directly against XRPL mainnet: amendment enablement was checked against the on-ledger Amendments object (validated ledger 105,599,927, amendment IDs computed as SHA-512Half of the amendment name), and the on-ledger format of RLUSD/OUSG/TBL/DCP/USDB was confirmed via account_objects and gateway_balances on each verified issuer account (Section 3 table). ConfidentialTransfer ships Supported::Yes in rippled 3.3.0-b1 (absent in 3.2.0), verified against release tags. Remaining unverified items: XLS-82 validator vote tallies (no public per-validator data checked) and the ~23% SAV/Lending support figure (single press source).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment