Berkeley’s council has two ways to handle an item. One gets a debate, a public comment period, and a recorded vote. The other gets adopted in a single motion with the item numbers unread. The choice between them is not made by how much money is at stake — and moving an item from the second pile to the first takes one member, no second, and no majority.
This page is not about any one decision. It is about a sorting mechanism, and about what happens to money that is sorted into the second pile over and over for years.
It is the general case. For a single program followed all the way through — one building, four appropriations, thirty-four years, and an operating question that was never once answered — see the companion page, Buy once, pay forever.
Every city council uses a consent calendar. It exists so that routine business — minutes, proclamations, contract extensions nobody disputes — does not consume floor time that substantive matters need. Used properly it is good government. The question this page asks is narrower: what has Berkeley actually put in each pile, and does the sorting track the size of the commitment?
Berkeley knows how to deliberate. The record contains a clear demonstration.
On April 28, 2025, the Council convened a special meeting whose entire agenda was a resolution on a ceasefire in Gaza. The annotated agenda records what the body was willing to spend on it:
| Roll call | 5:04 p.m. |
| Public comment | Rules suspended to establish a three-hour comment period, in person and online |
| Then extended | Rules suspended again for an additional hour |
| Meeting extended to | 12:00 a.m. |
| Speakers | 197 |
| Competing resolutions | Two. One failed 4–1–4; a substitute carried 7–1–1 as Resolution No. 71,726–N.S. |
Seven hours. Four hours of public comment. Two competing resolutions, both requiring members to vote on the record, with abstentions and dissents recorded by name. Every procedural tool the Council possesses was deployed.
Whatever a reader thinks of the merits, this is the machinery of deliberation working exactly as designed. Members took positions. The public was heard at length. The disagreement was visible, and it was resolved by a vote that cost each member something to cast.
The resolution committed no money and bound no party. It was addressed, in part, to the President of the United States.
Now consider the same council handling items that commit substantial money or contribute to obligations lasting decades.
Every follow-on cost in Berkeley’s affordable housing portfolio traced for these pages arrived on consent:
| Date | Item | Amount | Calendar | Speakers |
|---|---|---|---|---|
| 2025-01-21 | 2207 Haste Street rehabilitation | $500,000 | Consent | 0 |
| 2025-06-24 | HTF RFP — Adeline Street and Ashby Lofts | $5,914,829 | Consent | 0 |
| 2025-06-24 | Maudelle Miller Shirek reservation | $500,000 | Consent | 0 |
| 2025-12-02 | Loan modifications, three HTF projects | terms revised | Consent | 0 |
| 2026-01-20 | Capacity building contract | per year | Consent | 0 |
Roughly $6.9 million, adopted under the same formula each time: “to adopt the Consent Calendar in one motion except as indicated. Vote: All Ayes.” None of these items was among those indicated. None was pulled. Zero speakers on each.
Housing is not the only place this happens. The Berkeley pier-ferry project has been assembled almost entirely on the consent calendar over five years.
| Date | Step | Amount | Calendar |
|---|---|---|---|
| 2019-03-12 | City–WETA Memorandum of Understanding, planning phase | — | Consent |
| 2023-07-11 | Accepting Alameda CTC grant | $5,138,685 | Consent |
| 2023-12-05 | Resolution renaming the pier on reopening | — | Consent |
| 2024-05-14 | MOU Amendment #1, accessing WETA funding | up to $3,000,000 | Consent |
| 2024-05-14 | COWI North America, engineering and design | up to $6,500,000 | Consent |
| 2024-07-09 | Rincon Consultants, environmental services | up to $3,080,000 | Consent |
The 2021 preferred-concept discussion was a worksession, which takes no vote. The 2025 project update was a City Manager report, which takes no vote either. Across the project’s documented history, the archive records no action-calendar vote at all.
What that sequence has produced, per the City’s own October 2025 project update: a preliminary construction cost estimate of $103,639,467, excluding ferry vessels, and the sentence “construction funding has not been identified.”
The original 2019 MOU set an explicit staged gate: subsequent phases “shall be subject to approval by the City Council and the WETA Board, contingent upon the availability of regional transportation funding and City funding.”
That is a well-drafted exit provision. Each phase was to be a fresh decision. In practice each of those approvals arrived on consent, in one motion, alongside unrelated business — which satisfies the letter of the gate while removing the fresh deliberation that a staged approval process makes possible.
None of the individual amounts above would justify a special meeting. That is precisely what makes the pattern durable. A $3,080,000 environmental services contract is a routine professional services agreement. A $500,000 building rehabilitation is a normal housing item. Each is defensible in isolation, and each was properly noticed and lawfully adopted.
The aggregate is never presented. There is no document in Berkeley’s budget process that assembles a project’s consent-calendar history into a single number, and no point in the annual cycle at which a member is shown what the accumulated commitments now total. A councilmember wanting that figure for the housing portfolio would have to reconstruct it from consent items across four meetings spanning thirteen months. For the ferry, across five years.
The result is a commitment that grows without ever being decided. Not one large vote anyone can be held to, but a long series of small ones nobody remembers casting.
Every Berkeley staff report carries a section headed Alternative Actions Considered. It is where the report tells Council what else could have been done — a smaller version, a different site, a delay, doing nothing. It is the part of the record that shows a choice was a choice.
Across the 652 staff reports extracted from meeting packets for these pages, 425 contain that section. In 163 of them — 38 percent — the answer is “None.”
Those reports are not spread evenly across the agenda. Of the 163, 144 arrived on the consent calendar — 88 percent. Setting aside procedural recurrences where “None” is genuinely the honest answer, such as the scheduled bid solicitations and the annual appropriations limit, 27 consent-calendar items of $1,000,000 or more reached Council with staff reports considering no alternative at all.
Sometimes “None” is correct. A statute may compel the action. A grant deadline may genuinely foreclose the options. The November 2025 contract for the CareBridge reentry program said “None. Delay in approving this action would risk program implementation and jeopardize grant compliance.” — and that is a real constraint, on restricted money, for a program with a real constituency and no General Fund exposure. It was item 7 on the consent calendar, at $7,510,842 over three years.
But an answer given in more than a third of cases has stopped carrying information. Whatever analysis of alternatives happened, it happened somewhere a voter cannot read. The section that exists to show the choice was a choice now mostly records that the question was not put.
Berkeley’s own City Auditor measured the adjacent question and published the answer. In City Contracts: Outdated Process Led to Non-Competitive Contracts, reviewed by Council on March 10, 2026, the Auditor examined every contract executed in fiscal year 2024.
Ninety-four of 218 contracts had no documented evidence of competition — about $43 million, against $102 million that did. Of those 94: twenty-six were authorized by Council resolution, fifty-three went through waivers approved by the City Manager, and fifteen went through “other or unclear means.”
The same audit found that in fiscal year 2022 Berkeley executed recycling contracts worth nearly $85 million over ten years without seeking bids. The Purchasing Manual requires formal competition for service contracts above $50,000. The Auditor noted that neither the City Charter nor the Municipal Code actually required it for service contracts like those — a rule stated in a manual and mandated by nothing.
This is the oversight function working. The Auditor found the problem, counted it, named the dollar figures, and recommended that Finance begin tracking what kind of competition each contract received. What the record does not yet show is what changed afterward.
This site’s budget framework sets out what should be true before a public function continues. It is explicit that continuation is not automatic:
“Continue is not the default. It means the function meets a defined service level, its outcomes are visible, its total cost remains defensible, its delivery model compares favorably with alternatives, and it still fits the City’s priorities.”
Berkeley Budget Framework — A Strategy for Strategy
And that stopping has to remain genuinely available:
“STOP must be a real management option. Organizations often preserve programs because employees, contracts, facilities, commissions, grants, or constituencies have grown around them. That makes STOP politically painful, but not optional.”
Berkeley Budget Framework — A Strategy for Strategy
A consent calendar cannot deliver either test, and it is not built to. An en-bloc motion has no mechanism for asking whether total cost remains defensible, because it does not surface a total. It has no mechanism for comparing delivery models, because it presents one option. And it cannot make stopping a real choice, because nothing in the motion frames continuation as a choice at all.
Each successive approval can also make stopping harder. Once $6,500,000 of engineering and $3,080,000 of environmental work are complete, declining the next phase means accepting that some prior spending may not lead to construction. That is the ordinary logic of sunk cost, and the consent calendar is an efficient machine for accumulating it out of view.
There is an obvious objection to everything above: the consent calendar is set by staff, so this is an administrative artifact rather than a council choice.
It is not, and the record is unambiguous. Any single member can move an item from consent to action. It takes no second, no majority, and no permission — one member saying so is sufficient. It is the most accessible procedural power on the dais.
And Berkeley’s councilmembers use it. The annotated agendas record it at least ten times in this period:
| Date | Item moved to Action | Result |
|---|---|---|
| 2024-12-10 | Restaurant Inspection Audit Status Report | Discussion held |
| 2025-04-15 | Resolution on state arrest-quota legislation | Discussion held |
| 2025-04-29 | Adding ADA expertise to the Safe Streets oversight committee | 7 speakers; resolution amended and adopted |
| 2025-07-22 | Surveillance technology ordinance, city jail | 2 speakers; resolution adopted |
| 2025-10-28 | Authorization to encrypt police radio channels | 31 speakers; resolution amended and adopted |
| 2026-02-24 | A Guide to Measuring Performance in the City of Berkeley | Presentation and discussion; 2 speakers |
| 2026-03-24 | Mills Act contract, 2845 Woolsey Street | 3 speakers; item continued |
So the pull is not a theoretical power. It is a live, exercised practice, applied to a single-address historic preservation contract and to a committee’s membership criteria.
Which means leaving the items in this page’s first two tables on consent was not something Councilmembers were powerless to change. At every meeting, each member individually had the ability to require discussion. Nobody had to persuade a colleague. Nobody needed a majority. On $6.9 million of housing obligation and $17.7 million of ferry authorizations, no member said the words.
The traffic also runs the other way, and the most recent example is the clearest. On July 7, 2026, Item 9 — Waterfront Specific Plan Environmental Review Authorization — reached the Council on the action calendar. It would authorize environmental review of the Waterfront plan, refer $310,000 to the November 2026 appropriations process to conduct the EIR, and defer a standing Council direction issued on November 18, 2024. The annotated agenda records its disposition in six words:
“Moved to Consent Calendar. Approved the recommendation.”
Annotated agenda, July 7, 2026, Item 9 — recorded speakers: 0
An item that had been placed where it would be discussed was moved to where it would not, and adopted. The same procedural lever that can create scrutiny can remove it.
What went unremarked is worth stating precisely, because it is not the ferry. The authorization commissions a Program EIR on the Waterfront Specific Plan, scoped to a maximum development alternative of roughly 1,215,000 gross square feet — a land-use review for the whole waterfront, separate from the project-level environmental work contracted for the pier and ferry in 2024. Its second recommendation also defers a standing Council direction from November 18, 2024 to prepare a nature-and-recreation conceptual plan for César E. Chávez Park, folding those priorities into zoning provisions instead. Both are consequential decisions; both were disposed of in six words.
Every item in the table above shares a trait: it was already contested, or visibly touched a constituency, at the moment it appeared. Police radio encryption drew 31 speakers. Surveillance technology, arrest quotas and ADA representation are live disputes in Berkeley.
The examples reveal a pattern: pulls tend to occur where an item is already contested or visibly touches an organized constituency. They are not being used to create scrutiny where none exists yet — which is precisely the case an accumulating cost presents. A ninth engineering contract on a five-year-old project divides nobody in the room on the night it appears.
Berkeley’s council is capable of sustained, difficult, public deliberation. April 28, 2025 proves it: seven hours, four of them public comment, competing motions, recorded dissent. The capacity is there.
On this record, the strongest predictor of whether that capacity gets used is not the size of the commitment. It is whether an item is contested when it appears. A resolution about a war divides the room, so it goes to action and gets the full apparatus. A $3,080,000 environmental services contract divides nobody, so it goes to consent and gets a shared motion.
Over five years that sorting rule produces a $103 million project with no construction funding identified and no action-calendar vote in its record, and a housing portfolio whose recurring costs return every few years as fresh capital requests nobody connects to each other.
The fix is not to abolish the consent calendar. It is to notice that Berkeley has no procedural moment for the question the framework says should govern everything: is this still worth continuing, and can we afford to own it? That question does not divide the room on the night it should be asked, which is exactly why it never gets asked.
And the absence of that moment is not an excuse, because the remedy required no reform. Nine members each held, at every one of these meetings, the unilateral power to stand the item up and make the body look at it. They have used that power for a historic-preservation contract on one address and for the membership rules of an oversight committee. They did not use it once for $17.7 million of ferry authorizations or $6.9 million of accumulating housing obligation.
What that suggests is not indifference to cost. It is that the deliberative instinct fires on contention and not on magnitude.
The first question has an answer available in the room. The second requires someone to have thought past the decision in front of them, and on this record, across five years and two projects, no member did.