Maple Finance is an on-chain credit business that runs as a token-holder DAO. Where a peer-to-pool money market lends against volatile crypto collateral, Maple originates institutional loans – over-collateralised digital-asset credit, and tokenised exposure to real-world private credit – and packages the yield into stablecoin products anyone can hold. The protocol has described itself as building “onchain asset management since 2019” (maple.finance/about); it launched on Ethereum mainnet in May 2021 and, by its own reporting and DeFiLlama, grew assets under management into the multi-billion range through 2025–2026 as the “RWA on-chain” thesis matured.
What the DAO actually governs
Maple is a governance-over-a-business, not a governance-of-a-primitive. Holders of SYRUP do not vote loan-by-loan; they set the policy around an operating credit desk: how protocol revenue is used, what the token’s emission and buyback schedule looks like, which strategic initiatives the treasury funds, and how the two product surfaces – permissionless Syrup and permissioned Maple Institutional – are resourced. Day-to-day underwriting, borrower diligence and pool management sit with the Maple team and its delegates; the DAO governs the economic frame they operate inside. That split – a professional credit operation wrapped in token governance – puts Maple closer to Ondo and other real-world-asset issuers than to a fully permissionless lending protocol.
From MPL to SYRUP: collapsing to one token
Maple originally ran a two-token design (MPL and staked xMPL). The rebuild happened through a numbered, publicly recorded sequence of votes on the DAO’s Snapshot space: MIP-009 (closed 12 September 2023) upgraded the token design and set a defined inflation schedule; MIP-010 (9 September 2024) launched SYRUP and opened the one-time conversion of 1 MPL → 100 SYRUP, structured so migrating holders were not diluted; MIP-011 set the conversion deadline at 30 April 2025; and MIP-017 (18 May 2025) closed a final window and directed the unconverted remainder to strategic use. Documentation of the mechanics is in the migration docs.
The vote that settled governance rather than supply was MIP-012, “stSYRUP as the Sole Governance Token”, which closed alongside MIP-011 on 30 December 2024. In practice the live Snapshot space counts SYRUP and stSYRUP one-for-one as two independent balance strategies, so staking earns yield rather than voting power; “sole” refers to MPL’s retirement, not to a staking requirement.
Supply is best read from the chain rather than from a schedule. The Ethereum SYRUP contract (0x643C4E15d7d62Ad0aBeC4a9BD4b001aA3Ef52d66) reported a totalSupply() of 1,244,677,492.51 SYRUP at Ethereum block 25,900,561 (4 September 2026, 01:07 UTC). Of that, 163,969,389.67 SYRUP – 13.2% of supply – sat inside the stSYRUP staking contract (0xc7E8b36E0766D9B04c93De68A9D47dD11f260B45) at the same block, at an exchange rate of 1.0510 SYRUP per stSYRUP.
The arc from quarterly votes to a standing rule
Maple’s most instructive governance feature for the wider DAO industry is what it did to its revenue and buyback policy. For eighteen months the DAO re-legislated the same question every quarter, and the vote record is the clearest statement of the cost:
- MIP-013, “Fee Buyback Mechanism” (closed 27 January 2025), created the buyback; MIP-016 (5 May 2025) continued it for Q2 and MIP-018 (29 July 2025) raised it for Q3. Three votes, one recurring decision.
- MIP-019, “Activate the SSF and Sunset Staking” (31 October 2025), stood up the Syrup Strategic Fund and ended the old staking-rewards program, tying value to revenue instead of inflation. It passed 66,270,651.54 For to 488,845.21 Against on 26 votes. Coverage: The Defiant.
- MIP-020, “H1 2026 SSF Allocation” (26 January 2026), was the next discretionary allocation vote – 82,090,483.60 For, nothing against, on ten votes.
- MIP-021, “A Rules-Based Buyback That Scales With Revenue” (13–17 July 2026), finally replaced the per-quarter sizing decision with a formula. 61,300,404.59 For to 20,256.90 Against on 17 votes.
That arc – discretionary, then quarterly, then rules-based – is a live example of a treasury DAO deliberately narrowing the surface where a vote (or a signer) can act, to shrink the gap between a passed policy and its execution.
A caveat on where the arc currently stops. A forum thread titled “MIP-023: Hyper-Programmatic Revenue Model and Governance Parity” circulates as a proposal for a roughly 90% fully programmatic SYRUP buyback. As of 4 September 2026 it is a draft: the mapledao.eth vote record runs contiguously from MIP-009 to MIP-021 with no MIP-022 or MIP-023 on it, and no version of it has been archived. Until it is voted, the standing policy is MIP-021’s rule, not the programmatic model. This page previously stated otherwise, which is a reminder that a governance-forum thread is a proposal, not an outcome – see verifiability.
The electorate: a quorum sized on supply, met by six addresses
Maple sets its quorum as a share of total supply, and raises it as supply inflates: the bar was 34,913,160.5 tokens through the MIP-012–MIP-016 era, 53,600,000 for MIP-017 and MIP-018, 56,000,000 for MIP-019, 57,500,000 for MIP-020 and 58,200,000 for MIP-021. Against the chain reading above, the current bar is about 4.7% of SYRUP supply.
The electorate that actually turns out is a much smaller pool, which is what makes the ratio worth stating. Only 13.2% of SYRUP is staked, so 58.2 million is roughly 36% of every staked token. And turnout has fallen as the bar has risen: 56 votes on MIP-017 (May 2025), 42 on MIP-018, 26 on MIP-019, 10 on MIP-020 and 17 on MIP-021.
MIP-021’s per-voter record makes the concentration explicit. Of 61,320,661.49 tokens cast by 17 addresses, the largest single voter carried 22,137,278.03 – 36.1% of the whole vote; the top three carried 72.5%, and the top five 91.8%. The top five together hold 56.26 million tokens, just under the 58.2 million bar, so six addresses are sufficient to constitute quorum on this DAO. The proposal cleared quorum by 5.4%, and 99.97% of the weight cast was For.
This is the ordinary shape of a supply-weighted token-weighted vote with a supply-denominated quorum rather than a Maple-specific failing – but it is the number a governance operator should read before treating a quorum as evidence of a broad mandate. Compare the participation figures on GnosisDAO and Uniswap, and the delegation-based answers surveyed in voting and delegation.
Products: Syrup and Maple Institutional
Syrup is the permissionless surface: depositors mint yield-bearing syrupUSDC / syrupUSDT, earning yield sourced from Maple’s lending activity plus a points-style Drips rewards program that multiplies with longer capital commitments. Both are ERC-4626 vaults, so their size can be read directly rather than quoted. At Ethereum block 25,900,561 (4 September 2026, 01:07 UTC), totalAssets() on syrupUSDC (0x80ac24aA929eaF5013f6436cdA2a7ba190f5Cc0b) was $964.29 million against 816.22 million shares, and on syrupUSDT (0x356B8d89c1e1239Cbbb9dE4815c39A1474d5BA7D) $395.26 million against 346.21 million shares – $1.360 billion across the two Ethereum vaults, at share prices of 1.1814 and 1.1417 respectively. Maple also runs the products on other chains, so this is the Ethereum figure, not the protocol total.
Maple Institutional is the permissioned surface, where vetted borrowers access over-collateralised digital-asset credit and where Maple has built facilities such as an on-chain warehouse line for asset-backed loans. Both surfaces feed the same revenue engine that the SSF taps for buybacks. Protocol-wide AUM moves with subscriptions and credit demand and is best read from DeFiLlama rather than a fixed number.
Where Maple sits in the DAO landscape
Maple is one of the clearest cases of a DAO governing a credit institution rather than an automated protocol. It shares the real-world-asset frame with Ondo Finance, the yield-bearing-stablecoin frame with Sky (formerly MakerDAO), and the on-chain-lending frame with Aave and Morpho – but differs from all of them in how explicitly it has converted governance from a discretionary body into a rules-writing one. For the lending directory, it is the reference point for “institutional credit, run as a DAO.”
How Caper approaches this
Maple’s progression is really about one thing: shrinking the distance between a policy the token approves and the action that carries it out. Caper attacks the same distance, but it does not close it by making the ballot self-executing. A Caper proposal is settled in two phases. The ballot picks a winning option; then a price window opens, and the proposal passes only if the time-weighted average price over that window is at or above the baseline frozen when the proposal was triggered (resolve_proposal in the protocol logic sets passed = window_twap >= twap_baseline). A ballot majority is necessary and never sufficient – the market has to ratify it. Execution is then a further, separate call that asserts the proposal is both settled and passed before it will move anything.
What Caper does remove is the discretionary gap Maple had to legislate away: nothing between the settled verdict and the transfer is a human decision, because execution is an on-chain function on the proposal itself rather than a multisig honouring a signal. And where Maple routes revenue to a strategic fund, a Caper’s value flows to its own on-chain treasury; a member’s exit claim on that treasury is computed by the same function as their vote weight, so influence and payout come from one figure rather than two negotiated ones.
References
- mapledao.eth on Snapshot – the DAO’s complete vote record (13 proposals, MIP-009 to MIP-021), including per-proposal quorum, tallies and per-voter weights. Queried keylessly via the Snapshot GraphQL API on 4 September 2026.
- Maple Governance Forum – where MIPs are drafted and discussed before a vote. The forum is Cloudflare-walled to automated requests; the Snapshot record above is the citable outcome.
- Ethereum contracts, read at block 25,900,561 (4 September 2026): SYRUP
0x643C4E15d7d62Ad0aBeC4a9BD4b001aA3Ef52d66, stSYRUP0xc7E8b36E0766D9B04c93De68A9D47dD11f260B45, syrupUSDC0x80ac24aA929eaF5013f6436cdA2a7ba190f5Cc0b, syrupUSDT0x356B8d89c1e1239Cbbb9dE4815c39A1474d5BA7D. - Maple Finance – protocol site; About (“onchain asset management since 2019”). Documentation; MPL→SYRUP migration.
- DeFiLlama – Maple – live TVL/AUM across all chains.