Governance ReviewNo. 11Epochs 548 to 55325 Mar 2025 to 24 Apr 2025
The two year ₳300M and ₳250M spending limit closes at 34.7%
The proposal to cap treasury withdrawals at ₳300M in 2025 and ₳250M in 2026 closed with 34.7% DRep support. The Eternl DRep Committee and SASA opposed fixing the second year's limit in advance, while committee members challenged the proposal's documentation and approval threshold.
The ₳300M and ₳250M spending limit closes at 34.7%
At the start of epoch 553 the two year limit, ₳300M for 2025 and ₳250M for 2026, closed with 34.7% of DRep power behind it, on 138 yes votes, 112 no and 14 abstain. Its document gives no names for its authors, only that they are community members and not IOG, the Cardano Foundation, EMURGO, PRAGMA, Intersect or their staff. As its own bar it named more than 67% of the power delegated to DReps, the bar a treasury withdrawal needs, while quoting the more than 50% of the active voting stake the constitution sets. The tally stayed below both. The review of epochs 545 to 547 sets out how the figures of both limits were reached.
On the two year limit, ₳1.69B of DRep power voted yes and ₳1.62B no, but ₳1.38B of delegated power did not vote and ₳169.2M was delegated to the always no confidence option, and both count on the no side.
DRep power cast on the two year limit, by epoch of the vote
In millions of ada, replaced votes included
YUTA, the largest DRep, voted yes, writing that the treasury’s income this year is clearly lower than last year’s and that the limit for 2026 should be set at the same time. YUTA also voted yes on Intersect’s ₳350M limit, so that at least one limit would exist. SIPO called the numbers reasonable and the proposal’s community origin the governance culture it hopes for. EMURGO, the Eternl DRep Committee and Yoroi W₳llet voted no in epoch 552 and the Cardano Foundation in epoch 551, the four DReps with the most power among those voting no.
Eternl and SASA object to fixing 2026 in advance
The Eternl DRep Committee wrote that setting the limit for two years at once was not the right approach even though it agreed more with the ₳300M figure, and that there was “no need to rush” a limit for 2026. SASA_nagamaru_ながまる found the basis of the numbers clear and voted no, writing that fixing 2026 in advance risks making it “difficult to respond flexibly to unexpected future changes”. Tempo.Vote wrote only that it was too early to set a limit for 2026. Dave voted no for a different reason, finding the limits “excessively high” for a governance system still early in its maturity.
Four committee members find the two year limit unconstitutional
Input Output, Emurgo, the Intersect Constitutional Council and the Cardano Atlantic Council found the two year limit unconstitutional. The Cardano Foundation, which voted no as a DRep, voted yes as a committee member, the Eastern Cardano Council abstained and the Cardano Japan Council did not vote, a committee share of 16.7%. Emurgo and the Cardano Atlantic Council objected to the 67% bar, since the constitution sets more than 50% of the active voting stake, and the Cardano Atlantic Council read the proposal’s wording as an attempt “to modify the ratification thresholds dictated by the Cardano Constitution”. Hephaestus Stake Pool, a DRep that voted no for another reason, disagreed with that reading, writing that a proposal holding itself to a higher bar still meets the one in the constitution.
Input Output wrote that the proposal did not hash the contents of its two references, a spreadsheet and a forum post, so they could change without a verifiable record. The anchor document itself was hosted on GitHub, and the version served there when the proposal was filed did not match the hash recorded on chain. A version uploaded to the same address later that day removed only the final line break, which brings the file in line with that hash, and it is still the one served. The Intersect Constitutional Council wrote that the change within the first 48 hours left voters unsure whether the hashes matched even days after the correction, which a majority of its members found enough to breach the right to a process “protected from undue influence and manipulation”. Hephaestus Stake Pool voted no over the change while preferring these figures to ₳350M, writing that the error was corrected quickly but showed the risk of references that can be changed.
Intersect’s ₳350M limit stands at 73.4% at the close
2025 Net Change Limit, Intersect’s ceiling of ₳350M on treasury withdrawals from the start of epoch 532 to the end of epoch 604, stood at 73.4% of DRep power as of the close of epoch 553, the last epoch of its vote, on 201 yes votes, 123 no and 13 abstain. Its document names more than 50% of the active voting DRep stake as its bar and says the period was agreed with the Cardano Budget Committee.
The Cardano Vision and Roadmap closes at 63.8%
Defining the Cardano Vision and Roadmap for 2025 and beyond, the development priorities of the product committee of Intersect that the review of epochs 542 to 544 describes, closed at the start of epoch 549 with 63.8% of DRep power behind it, on 243 yes votes, 9 no and 21 abstain. Yoroi W₳llet voted yes in epoch 548, writing that the roadmap’s work on programmable assets and layer two scaling is in line with Yoroi’s own development plans. AdaStat voted yes in the same epoch despite its doubts about a reference link “that may change at any time”, writing that an information action only takes the temperature.
On the committee, Emurgo, the Cardano Foundation and the Cardano Japan Council voted yes, the Intersect Constitutional Council no, and the Eastern Cardano Council, the Cardano Atlantic Council and Input Output abstained, a committee share of 75.0%. The Intersect Constitutional Council found the action “not constitutional” by four votes to none with one abstention, writing that its main reference is a website that can change, although the content it reviewed “was not found to contain anything unconstitutional”. The Eastern Cardano Council abstained, writing that the committee should not rule on information actions that measure no sentiment for an on chain action. The Cardano Atlantic Council wrote that “imprecise language” split its members on whether the action was a poll, a request for feedback or something else, and Input Output that the constitution sets no direct limits on this kind of initiative.
Delegation to DReps reaches ₳4.82B
The power delegated to DReps rose from ₳4.18B at epoch 548 to ₳4.82B at epoch 553, the most up to then, ₳298.9M of it between epochs 549 and 550. The ten largest DReps held 41.3% of it at epoch 548 and 47.3% at epoch 553. At epoch 548, 409 DReps had cast a vote in the previous twelve epochs, the most up to then, and 352 at epoch 553, while the final votes cast in the window came from 300 DReps. The window recorded 684 votes, 527 from DReps, 137 from pools and 20 from committee members, replaced votes included. The treasury read ₳1.73B at epoch 553, ₳22.4M above epoch 548 and the most up to then, with nothing yet withdrawn from it through governance.
Decided in this window
| Action | Outcome | DRep yes |
|---|---|---|
| Set 2025 Net Change Limit of 300M ADA, 2026 Net Change Limit of 250M ADA | Closed 553 | 34.7% |
| Defining the Cardano Vision and Roadmap for 2025 and beyond | Closed 549 | 63.8% |
Open at the close of the window
| Action | Status | DRep yes |
|---|---|---|
| 2025 Net Change Limit | Undecided at the close of epoch 553 | 73.3% |
The numbers behind the window
- Delegated to DReps, at the start of epoch 548
- ₳4.18B
- Delegated to DReps, at the start of epoch 553
- ₳4.82B
- Votes cast in the window, superseded votes included
- 684
- DReps whose final vote fell in the window
- 300
- Treasury, at the start of epoch 548
- ₳1,712.3M
- Treasury, at the start of epoch 553
- ₳1,734.7M
- DRep shares are the share of counted power: abstaining power is left out, while power that did not vote and power delegated to the always no confidence option are folded into the no side. That is why the two year limit could carry more yes power than no power and still read 34.7%. Committee shares are yes votes over the members active at the decision minus those who abstained, so a member who does not vote lowers the share without being an abstainer.
- An information action has no bar in the protocol. The 67% and the more than 50% compared with the tallies here are the bars the two limits name for themselves in their own documents.
- Intersect's limit closes at the boundary that ends this window: voting on it ended at the start of epoch 554, so its tally is the state at the close of epoch 553, and it is quoted without an outcome.
- The reasons quoted are the published rationales of the DReps and committee members named, cited under the sources below, and say nothing about voters who published none. The four largest no votes are ranked by delegated power at epoch 553, not at the epoch of the vote.
- What the two limits ask for and who filed them is read from their own documents and the sources named below. Intersect is named for the single year limit because it was filed from an address the site attributes to Intersect. The replacement of the two year limit's anchor document is read from the file's history on GitHub: each version was hashed and compared with the anchor hash recorded on chain.
- The count of DReps with a recent vote is a rolling twelve epoch measure that includes superseded votes and excludes votes without a block time. Votes cast is every vote of epochs 548 to 553 across all three roles, superseded votes included, and vote power per epoch is DRep power at the epoch of the vote. Delegated power and the top ten share exclude the predefined delegation options.
Sources and further reading
23 sources, open the list
- The two year proposal asks for ₳300M for 2025 and ₳250M for 2026, pegged to forecast treasury inflows. It sets its own bar at more than 67% of delegated ada voting yes, aligned with the treasury withdrawal threshold, while noting that the constitution asks for more than 50% of the active voting stake. Its authors give no names and describe themselves as community members who are not IOG, the Cardano Foundation, EMURGO, PRAGMA, Intersect or their staff. Its two references, a spreadsheet of models and a forum post, carry no hash. the document behind the two year limit
- The file history of the two year limit's anchor document on GitHub shows a version committed minutes before the proposal was submitted and a replacement uploaded later the same day. The earlier version, hashed the way the chain records it, does not match the anchor hash recorded on chain, while the replacement, which is still the version served, does. The replacement removes only the final line break of the earlier version and changes nothing in its content. the file history of the two year limit's anchor document
- The positions page of the two year limit counts ₳1.69B of DRep power yes, ₳1.62B no, ₳1.38B of delegated power that did not vote and ₳169.17M delegated to the always no confidence option, and lists the Cardano Japan Council among the committee members as not voted. the tally and committee positions on the two year limit
- YUTA voted yes on the two year limit, writing that it preferred the community led figures to the ₳350M alternative it calls led by Intersect because income in the current year is clearly lower than the year before and a limit for the following year should be set at the same time, but that it would vote yes on both so that a limit exists. the rationale of YUTA on the two year limit
- SIPO voted yes on the two year limit, writing that the numbers are reasonable, that the structure emphasises long term sustainability, and that its community driven origin embodies the governance culture it hopes for. the rationale of SIPO on the two year limit
- The Eternl DRep Committee voted no on the two year limit, writing that setting a limit for two years at once is not the right approach even though it agrees more with the ₳300M figure, that the community will refine the process in the coming months, and that there is no need to rush a limit for 2026. the rationale of the Eternl DRep Committee on the two year limit
- SASA voted no on the two year limit, writing that aiming to operate within the roughly 300,000,000 ada of annual inflow is understandable and clearly reasoned, but that fixing the ceiling for 2026 in advance risks making it hard to respond to unexpected changes. the rationale of SASA on the two year limit
- Tempo.Vote voted no on the two year limit, writing that it is too early to set a limit for 2026. the rationale of Tempo.Vote on the two year limit
- Dave voted no on the two year limit, writing that the proposed limits are excessively high for a governance system still early in its maturity and could set a precedent that undermines fiscal responsibility. the rationale of Dave on the two year limit
- Hephaestus Stake Pool voted no on the two year limit while preferring its figures to the ₳350M alternative, writing that it was submitted with a mutable reference that was altered within 48 hours of submission, causing a mismatching hash and confusion among voters, that the error was corrected quickly but showed the risk of references that can change, and that it has no issue with a proposal setting a higher bar for itself, since more than 50% is still met. the rationale of Hephaestus Stake Pool on the two year limit
- Input Output published a rationale finding the two year proposal procedurally unconstitutional: its rationale did not hash the contents of two of its references, so the information at those addresses could change without a verifiable record, which the constitution requires of the off chain content of a governance action. The same rationale notes that the constitution places no restrictions on the content of a net change limit, applauds the initiative, and says Input Output is eager to see DReps agree on a limit. the published rationale of Input Output on the two year limit
- The Intersect Constitutional Council found the two year limit unconstitutional, writing that the metadata anchor file was changed within the first 48 hours after submission, causing confusion among voters about whether the hashes matched even days after it had been corrected, which a majority of its members deemed enough to violate the principle that voting be protected from undue influence and manipulation, while agreeing that the wording of the proposal itself was constitutional and that multi year limits are permissible. the rationale of the Intersect Constitutional Council on the two year limit
- Emurgo found the two year limit unconstitutional as a committee member because the proposal sets its own threshold at 67% while the constitution defines the threshold for a net change limit as more than 50% of the active voting stake. the rationale of Emurgo on the two year limit
- The Cardano Atlantic Council found the two year limit unconstitutional because its rationale uses language proposing to modify the ratification threshold the constitution dictates, stating that the action shall be considered ratified at more than 67% where the constitution sets more than 50% of the active voting stake. the rationale of the Cardano Atlantic Council on the two year limit
- The single year proposal asks for ₳350M for a period it describes as 72 epochs, from the start of epoch 532 to the end of epoch 604, which inclusive is 73 epochs, pegged to the previous year of actual treasury income. It states that the period was agreed with the Cardano Budget Committee, that actual treasury income over epochs 459 to 531 was 335,957,093 ada, within 5% of the proposed figure, and that it counts as agreed at more than 50% of the active voting DRep stake, the figure the constitution sets. the document behind the single year limit
- A Lido Nation post describes the ₳350M limit for 2025 as a governance action proposed by Intersect, and the two year alternative as submitted by anonymous community members. Lido Nation, on who proposed the two limits
- The roadmap put to the DReps came from the product committee of Intersect and set out priorities for scaling the main chain, bringing in liquidity from other chains, expanding layer two, developer tooling, programmable assets and running several node implementations. the product committee on the 2025 vision and roadmap
- Yoroi voted yes on the roadmap signal, writing that as an open source infrastructure product focused on usability it found the roadmap's initiatives on programmable assets and layer two scaling in line with its own development plans. the rationale of Yoroi on the roadmap signal
- AdaStat voted yes on the roadmap signal, writing that its doubts came from a reference link to a website that may change at any time, but that this matters less for an information action that only takes the temperature. the rationale of AdaStat on the roadmap signal
- The Intersect Constitutional Council found the roadmap signal not constitutional on a split of four to none with one abstention, writing that not all critical components of the roadmap are captured in immutable metadata since the primary reference is a website that may change, that the content reviewed contained nothing unconstitutional, and that proposers should anchor all materials in decentralised storage. the rationale of the Intersect Constitutional Council on the roadmap signal
- The Eastern Cardano Council abstained on the roadmap signal, writing that the committee should abstain on information actions that do not measure sentiment for future governance actions to be enacted on chain. the rationale of the Eastern Cardano Council on the roadmap signal
- The Cardano Atlantic Council abstained on the roadmap signal, writing that imprecise language left it unclear whether the action was an informal poll, a solicitation of feedback or something else, which split its members. the rationale of the Cardano Atlantic Council on the roadmap signal
- Input Output abstained on the roadmap signal, writing that most uses of an information action are not a matter of constitutional adjudication, that the constitution creates no direct limitations on this kind of initiative, and that it abstains to keep the neutrality appropriate to it. the rationale of Input Output on the roadmap signal