Regional in practice. Two-county in law.

Most regional consolidations begin with a shared debt to divide. This one does not. The 2014 revenue bonds are gone, the original construction financing is paid, and the jail arrives at this negotiation owning itself outright.

The Authority is not without borrowings. It drew $1.71 million on a line of credit in April 2026 to replace door locks and cameras — a short-term borrowing against a capital need rather than construction debt. Section 07 returns to why a debt-free authority ended up financing a lock replacement on credit.

That single fact rules out the conventional answer. A county joining a partnership normally buys in by assuming a proportionate share of the remaining debt service — the price is already written down, and the negotiation is about percentages rather than dollars. With the debt retired, there is nothing to assume. The question becomes what an ownership interest in a debt-free, twenty-year-old correctional facility is actually worth, and what a new member should pay for it.

It is a harder question than it looks, because the answer has three parts that are easy to conflate: what the asset is worth, what the founding counties have actually put in, and what a new member pays under whichever structure the counties choose. Those are three different numbers, and only the third one is real.

The building has also begun to age into its own capital cycle, which puts a second question alongside the first: how a twenty-year-old facility gets paid for over the next twenty years, and how many counties are standing behind it when the roof comes due.

The Lincoln and Sagadahoc Multicounty Jail Authority was created by special act on 21 May 2003, codified at 30-A M.R.S.A. chapter 17. The counties executed their operating and governance agreement in 2004. The facility opened in 2006.

It is not a general regional-jail-authority law that a third county can be added to, but a single-purpose statute written for exactly two counties, which it names throughout. The territory clause is explicit:

Section 1955, which authorizes general obligation bonds in aid of the Authority, likewise names those two counties and no one else. Knox cannot be admitted by agreement, by board vote, or by contract. That is consistent with the RFP contemplating legislative support, and with the counties already holding draft legislation.

Ten provisions appear to be engaged. What each currently says is a matter of record; what a bill should do about any of them is a question for the counties and their counsel. Set out that way, the scope of the legislative work looks larger than “add Knox to the list of counties.”

Some of the work is available under the statute as it stands and some is not, and the two are worth separating early. Terms in the first column can be settled and operated under while any statutory change follows.

The counties have been here once already. LD 1950 of the 132nd Legislature would have authorized the inclusion of additional counties in the Authority. It was introduced on 7 May 2025, sponsored by a senator from Waldo and co-sponsored by senators from Knox, Lincoln and Sagadahoc — a bill with the whole district behind it. It was placed in legislative files, and it was not defeated on its merits. Sagadahoc's minutes record what happened: the Jail Authority “had no knowledge of the bill until it was posted.”

That is a solvable problem, and the counties have already started solving it. The standing committee that has met monthly since June 2026 brings the administrators, the commissioners and the sheriffs into the same room — the parties whose agreement a bill needs and did not have. What remains is the work of turning a room into a settled position: a framework the Authority, three boards of commissioners, three sheriffs and the legislative delegations have each seen, marked up, and agreed to before anything is filed.

Sagadahoc's minutes record that the 2025 language is being used as a template for a new bill “that will be submitted after the November elections.” The incoming Legislature convenes in December 2026, and the deadline for bill requests falls in its opening weeks. On the current schedule the bill is filed before the framework exists — the counties’ own decision point, December 2026 or January 2027 per the working group, lands two to three months into a six-month engagement rather than at the end of it.

Maine's joint rules contain a tool built for this position. A concept draft may be filed with only an enacting clause and a summary, holding the cloture slot while the substance is still being settled — with the trade that any committee amendment must be germane to that summary. The rules are re-adopted each Legislature and concept drafts are not universally welcome in Augusta, so the applicable rule and the cloture date both need confirming against the incoming session.

The question usually asked about a three-county board is how big it gets. Total size may not be the most informative number — eighteen directors would not be an outlier for a multi-jurisdiction public body. The figure that differs most from the other authorities located for comparison is the per-county delegation.

Section 1853(1) gives each county six directors — three public members, two commissioners and a sheriff. Six per county is what produces eighteen at three counties, and it sits above the delegations used by the multi-county jail authorities located for comparison.

None of this establishes what the right delegation is for these three counties, and none of it suggests the present structure is inappropriate. Chapter 17 was written for two counties and has governed them for twenty years. What the comparison shows is narrower: delegation size is a design variable rather than a fixed convention, it has been set differently in most other places, and a third member makes it a live question. What works here is a matter for the three counties to settle.

One practical point does follow from the comparison, and it is already the counties' own practice. Chapter 17 lets each county appoint alternates without limit — Lincoln's March 2026 minutes state the rule plainly, and directors have moved between full and alternate status through the year. Where alternates vote and count toward quorum, as they do under Virginia's statute, attendance depends on the alternate bench rather than on the number of seats. That separates the two things a discussion about delegation size usually runs together: how many people a county has in the room, and how many votes it holds.

Two Bridges has operated for years as a regional facility in fact well before it becomes one in law. On a typical day in 2026, the two counties that own the jail hold fewer beds between them than the two counties that contract for beds in it.

Knox alone runs roughly twice Lincoln's count and close to four times Sagadahoc's. Penobscot, which is not a party to any of this, is consistently among the largest single users in the building.

The counties are also being asked to design a permanent structure at a moment when demand is not growing. The 2026 facility study projects average daily population flat to slightly declining through 2039, absent legislative change — driven by slowed population growth, a rising median age, and low, stable crime rates.

That changes what regionalization is for here. The argument is not capacity but cost-sharing, and the durability of a formula over a period in which the denominator may shrink. A formula driven purely by average daily population, on a declining population, produces a rising cost per inmate-day — which is one of the things a formula would have to account for.

Two other trends change what the facility costs to run without changing how many people are in it: bookings have fallen by roughly half since 2012, average length of stay has nearly doubled, and pre-trial detainees have gone from two-thirds of the population to well over four-fifths.

Knox, Lincoln and Sagadahoc are within six percentage points of one another on population. They are not close at all on taxable value, and they are nowhere near each other on use of the jail.

That mismatch is the whole negotiation. Every candidate basis for allocating cost — population, state valuation, jail use, equal shares — is defensible on its own terms, and every one of them produces a different winner. Each county's preference follows from arithmetic alone.

There is a second layer beneath the choice of basis. Operating cost and capital cost need not follow the same rule, and there is a reasonable argument that they should not: a county's use of the beds in a given year is a different thing from its stake in the building. A formula that answers both questions with one number has made a choice between them, whether or not the choice was discussed.

One further variable sits underneath every basis. The County Jail Operations Fund now covers under 15% of statewide jail operating costs; counties sought an $8 million annual increase for FY2026 and received an emergency $4 million. Any formula anchored to today’s state contribution will not survive its first legislative session, which raises the question of how a change in state funding would be shared. Whatever the counties settle on reaches taxpayers through assessments apportioned among 47 municipalities on state valuation.

Section 1952 is a rate covenant: it sets out six things the Authority’s rates must be sufficient to do. One of them is funding repairs, replacements and renewals. Two of them name reserves — and they are not the same two.

The chapter obliges the Authority to fund repairs, replacements and renewals, then names only an operating contingency capped at 3.5% and reserves securing bonds and notes. Neither is a reserve for replacement. That is silence in the drafting rather than a prohibition — the 3.5% cap attaches by its own words to one reserve for one purpose, and the Authority does report a capital fund balance. Nothing here says it holds too much or too little. What the chapter does not do is name a reserve for replacement, or say how one would be sized, funded and governed.

The scale is what makes the silence worth raising. Leveled across twenty years, the renewal program in the 2026 facility study runs to roughly $2.39 million a year, about 27% of the operating budget. The first five-year tier alone is $10.26 million — roofing at end of life and already leaking, mechanical and electrical replacement, detention plumbing, security controls, fire alarm — and the counties have already drawn $1.71 million on a line of credit for a door lock and camera replacement. That is the reserve gap in a single transaction: an authority carrying no bonded debt, financing a foreseeable renewal on short-term credit because no replacement reserve was sized to absorb it.

On 19 May 2026 the Lincoln County Commissioners accepted the 2026 facility study three votes to nil. All three governance options in that study are four-county — Knox, Lincoln, Sagadahoc and Waldo, the counties of the Sixth Judicial Circuit — and the most comprehensive is described there as “the most effective governance model.” The four-county premise on this page is the counties’ own, taken from advice they commissioned and adopted.

Waldo County is not a party to this procurement, has no contract, has made no application, and nothing here argues that it should be admitted. It appears for one reason: the framework these three counties settle will either accommodate a fourth member or it will not, and that is decided when the framework is drafted rather than when a fourth county asks.

Waldo is not incidental to the building's history either. It boarded inmates at Two Bridges until 31 December 2025 — the day before Knox's agreement began — and across the 2010–2024 booking data in the 2026 facility study it is the largest external contributor by a wide margin — 3,589 bookings, or 53% of the identified out-of-jurisdiction admissions. A further 18% of all bookings in that period have no originating agency recorded, so Waldo’s share of external admissions is a share of what the data identifies rather than of everything outside the two owner counties.

The question is narrow: whether the admission mechanism, the cost-allocation formula and the founding-county protections are drawn around a named county or around a member generally. Those are not the same document.

No figure on this page is estimated, and none comes from a private source. Where a number is derived — the allocation shares, the leveled twenty-year renewal figure — the derivation is stated alongside it.