Name the decision and version.
State the exact action, sponsor, authority, deadline, fund, period, amount, alternatives, and evidence owner.
Journey 5 · Finance Committee decision method · eight gates
Trace the decision. Test the whole picture. Make the advice reviewable.
Use one sequence for every material financial proposal. Define the action before testing numbers. Establish today's baseline before claiming savings. Reconcile money before judging affordability. Compare complete alternatives before recommending.
Governing question: What should Town Meeting authorize, given the complete service, financial, household, implementation, and long-term consequences—and what conditions should travel with the advice?
Quick start
One proposal. One sequence. One controlled record.
This method can be used for an operating request, position, capital project, building scenario, borrowing, reserve use, grant-funded proposal, or other material commitment.
Move gate by gate.
Record pass, conditional pass, or stop. A missing required input stays missing; it does not become $0.
Refresh the record.
Confirm the recommendation matches the current scope, amount, funding, motion, assumptions, and unresolved conditions.
Follow the result.
Connect contracts, spending, milestones, outcomes, later obligations, changes, and closeout to the original decision.
Worked-example context
The FY2027 final-voted General Fund is $57.018 million; HWRSD assessment and debt are $33.095 million, or 58.042% of that denominator. The separate $66.372 million Expense Control total is a multi-fund scope. This is why Gate 3 requires measure identity before arithmetic or advice.
The review sequence
Eight gates before advice.
Each gate produces a concrete output. A stop condition means the committee should request evidence, narrow the question, or make uncertainty explicit before proceeding.
01
Define the decision.
Question: What exactly may be authorized, by whom, under what authority, by what date, and for what public objective?
Required: sponsor, scope, current version, amount or range, funding path, decision body, deadline, and intended outcome.
OutputOne-sentence decision statement and authority map.
Stop ifThe action, decision-maker, scope, or deadline is ambiguous.
Stop ifThe action, decision-maker, scope, or deadline is ambiguous.
02
Establish the baseline.
Question: What service, demand, staffing, asset, contract, debt, reserve, and risk exists before this proposal?
Required: current actuals and budget, service level, demand/utilization, condition, obligations, known changes, and status-quo cost/risk.
OutputControlled current-state baseline and status-quo alternative.
Stop ifSavings or need is claimed against an undocumented baseline.
Stop ifSavings or need is claimed against an undocumented baseline.
03
Control every measure.
Question: What does each headline number mean, and can the compared figures legitimately share a denominator?
Required: value, unit, entity, period, fund, accounting basis, lifecycle stage, source, evidence state, formula, and freshness.
OutputMeasure register with compatible comparison groups.
Stop ifBudget, actual, levy, debt, estimate, District total, or household figure is being substituted.
Stop ifBudget, actual, levy, debt, estimate, District total, or household figure is being substituted.
04
Reconcile money and future obligations.
Question: Where does each dollar originate, where does it land, when does cash move, and what follows after the opening action?
Required: sources/uses, restrictions, match/reimbursement, debt, contracts, staffing/benefits, operations, maintenance, renewal, transition, exit, and non-duplication controls.
OutputBalanced funding bridge and annual obligation schedule.
Stop ifSources do not equal uses, hoped-for proceeds are treated as cash, or continuing tails are absent.
Stop ifSources do not equal uses, hoped-for proceeds are treated as cash, or continuing tails are absent.
05
Test affordability.
Question: Can the whole system carry the choice—not merely the first appropriation?
Required: recurring forecast, annual cash timing, complete capital/debt portfolio, reserves, baseline services, downside cases, household distribution, and capacity remaining.
OutputFour-part affordability finding: recurring, timing, capital, household/service.
Stop ifOne-time money funds a permanent cost without a replacement plan or no portfolio forecast exists.
Stop ifOne-time money funds a permanent cost without a replacement plan or no portfolio forecast exists.
06
Compare complete alternatives.
Question: What other feasible ways can meet the objective, including the current path, repair, phase, share, lease, consolidate, redesign, delay, or stop?
Required: same scope, period, assumptions, outcome measures, cost categories, funding rules, and risk treatment for every option.
OutputComparable option matrix and reasons any alternative was screened out.
Stop ifOne option is complete while another is only a headline or the status quo is priced at zero.
Stop ifOne option is complete while another is only a headline or the status quo is priced at zero.
07
Show consequences, value, and risk.
Question: Who gains, pays, waits, loses service, or assumes risk—and how reversible is the choice?
Required: service outcome, beneficiaries, access, household bands, opportunity cost, community/environment effects, legal/implementation risk, uncertainty, and triggers.
OutputConsequence and risk statement for each alternative.
Stop ifCost is shown without public value or a downside is hidden outside the chosen fund or time horizon.
Stop ifCost is shown without public value or a downside is hidden outside the chosen fund or time horizon.
08
Make advice accountable.
Question: What should happen, why, despite what counterargument, under what conditions, and with what follow-through?
Required: recommendation, evidence/reasoning, strongest opposing view, conditions, dissent, owner, milestones, measures, due dates, and reconsideration trigger.
OutputReviewable recommendation tied to the current motion and later result.
Stop ifThe conclusion cannot be reconstructed from the evidence and tradeoffs.
Stop ifThe conclusion cannot be reconstructed from the evidence and tradeoffs.
Gate 5 in detail
Affordability is four tests—not one ratio or available balance.
A proposal can pass one test and fail another. The finding should name each result and the most important assumption.
Recurring balance
Do dependable recurring revenues cover continuing baseline services and the recurring proposal requirement without unidentified tax/fee increases, repeated reserve draws, cuts, or deferral?
Cash-flow timing
When do appropriations, borrowing, reimbursements, construction draws, debt service, transition, and operations occur? Can each year—not only the average—be paid?
Capital sustainability
Does the choice fit the complete asset, maintenance, replacement, debt, and reserve portfolio without hiding displacement or double-counting retiring capacity?
Household and service
Who pays and when? What do residents receive? What service, access, reserve, or asset condition changes if assumptions fail?
Minimum conclusion language
State whether each test is supported, supported with conditions, not supported, or not determinable from current evidence. “Repayable,” “fits within unused levy capacity,” “cash is available,” and “grant eligible” do not by themselves mean affordable.
External-force register
Name how outside conditions reach the decision.
Do not blend a real cash change, an accounting measurement change, and a legal-capacity change into one “economic impact” number.
Money paid or received changes.
Wages, health costs, materials, construction, energy, insurance, borrowing, investment income, state aid, and receipts can move annual cash.
Show: source assumption, base/downside/upside, year, owner, and refresh trigger.
Present value changes.
Discount or actuarial assumptions can change a reported pension/OPEB or project present value without automatically changing the obligation or current-year cash.
Show: nominal cash flows, rate, sensitivity, and separate annual affordability.
A limit or allowable path changes.
Assessed value, new growth, exclusions, overrides, or statutory rules can affect ceilings and levy paths without creating cash by themselves.
Show: ceiling, operative limit, actual levy, and unused levy capacity separately.
Minimum downside case
Test compensation, health/insurance, general inflation, construction escalation, interest rates, state aid, new growth, taxable value, HWRSD allocation, schedule delay, grant/reimbursement timing, and service demand as separate inputs where material.
New state-aid baseline: CSKE now preserves final Wilbraham and HWRSD Cherry Sheet estimates for FY2000–FY2027. Use that history to frame a range and ask what changed; do not substitute it for the current budget stage, actual payment records, or a complete Town/HWRSD funding bridge. Open the state-aid and charges guide.
The useful “+?”
Add a Future Financial Obligation Statement to every material request.
This proposed record bridges the annual workflow and the eight-gate decision review. It should follow a proposal from submission through recommendation, vote, implementation, and closeout.
Keep the same proposal traceable.
ID, sponsor, purpose, alternative, version, authority path, status, current amount, recommendation, and final vote.
Name what is authorized now.
First appropriation or authorization, source, date, restrictions, match, contingency, and procurement/financing dependencies.
Show the stabilized annual effect.
Service, staffing, compensation, benefits, contracts, utilities, maintenance, revenue, and offsets.
Carry the full tail.
Debt, leases, subscriptions, pension/OPEB cash effects, renewal, replacement, remediation, regional obligations, and exit/closeout cost.
Say what the proposal really changes.
Cash-funded improvement, structural change, risk reduction, measurement-only change, deferral/cost shift, or unresolved effect.
Assign accountability.
Implementation owner, milestones, actual cost, outcome, remaining balance, review cadence, correction point, and closeout.
Framework visible; live calculation not ready.
The required proposal, payroll, benefit, contract, asset, debt, HWRSD, funding, household, and service records are not yet connected at a complete decision level. Do not populate a live obligation total from broad budget categories or assumptions.
Recommendation record
A strong recommendation preserves the judgment—not just the verdict.
The final language should be understandable to residents and reconstructable by a later committee.
| Field | Complete this sentence | Required support |
|---|---|---|
| Action | We recommend approve / reject / defer / amend… | Exact current proposal, amount, source, authority, and motion relationship. |
| Because | The evidence shows… | Baseline, reconciled money, affordability, alternatives, value, and risk. |
| Even though | The strongest opposing consideration is… | Good-faith counterargument and remaining uncertainty. |
| Provided that | Support is conditioned on… | Measurable conditions, owners, due dates, and consequence of noncompliance. |
| Review when | Reconsider if… | Cost, schedule, revenue, demand, service, legal, or risk trigger. |
| Follow through | The responsible owner will report… | Milestones, actual spending, outcome, remaining balance, and closeout date. |
Query evidence at the review gates
Use the safest available layer for the question in front of the committee.
More discoverable evidence improves due diligence only when the query keeps meaning, provenance, coverage, and double-count controls attached.
01Typed canonical fact
02Accepted normalized observation
03Candidate evidence for review
04Original source and locator
Keep identity and missingness.
Entity, period, scope, stage or basis, unit, scale, availability, source, assurance, and relationship travel with the value.
Test hierarchy and lifecycle.
Do not add issue and security principal, a note and its takeout bond, a final Cherry stage and its canonical fact, or an embedded audit and the original audit.
Legal room is one input.
Tax and DOR history, current services, recurring revenue, debt service, HWRSD burden, reserves, capital, household distribution, and downside conditions still need one compatible baseline.
A larger evidence inventory can still support “not determinable from current evidence.”
The disclosure layer's 6,701 source-linked rows, DOR's 17,722 observation states, and Affordability 0303's reconciled tax-base layer make questions easier to investigate. They do not automatically resolve mappings, comparability, currentness, or decision sufficiency.
Affordability evidence test
Ask what the new data proves before using it in advice.
The Finance Committee manual's whole-picture test still governs. Historical depth improves due diligence; it does not replace the forecast, portfolio, household, and service evidence required for a current choice.
| Capacity question | Evidence now available | Committee finding today |
|---|---|---|
| Legal · what Proposition 2½ permits | FY2007–FY2026 levy-capacity series; FY2026 actual levy $48,746,259 and unused levy capacity $1,322,755. | Supported as legal-capacity evidence |
| Recurring · what can continue | Historical revenue/expenditure/local-receipt series; incomplete FY2027 source bridge and no governed multi-year baseline. | Not determinable |
| Cash · what can be paid when due | Historical free-cash/reserve measures with controlled missing states; no detailed cash forecast. | Not determinable |
| Debt · what the portfolio can carry | Historical series, FY2025 position evidence, and accepted existing Town/HWRSD payment schedules through FY2040/FY2041; future borrowing, later member allocations, and a consolidated capital scenario remain absent. | Partial; current capacity not determinable |
| Household and service · who pays and what changes | Tax mechanics and assessment-band formulas; proposal-specific distribution and service consequences absent. | Not determinable |
Safe recommendation language
“The current evidence supports the stated historical, legal-capacity, existing scheduled-payment, or standardized distribution finding. It does not establish dependable remaining recurring capacity or the household/service affordability of this proposal.”
Review the Town and HWRSD annual debt-service horizon → · Open Affordability Evidence v2 and its controls →
Publication and recommendation gate
Do not call a decision ready until the record passes these checks.
A committee can still act under uncertainty. It should name the uncertainty, its consequence, the reason action cannot wait, and any condition or review trigger.
Ready to deliberate
- Authority, deadline, scope, sponsor, version, and objective confirmed.
- Current baseline and status quo documented.
- Every headline number scoped, dated, sourced, and evidence-labeled.
- Sources and uses reconcile; overlaps and missing values are controlled.
Ready to compare
- Feasible alternatives use the same assumptions and cost categories.
- All four affordability tests are complete or explicitly not determinable.
- Five-, ten-, and twenty-year effects and downside cases are visible.
- Household, service, opportunity-cost, and risk consequences are named.
Ready to recommend
- Reasoning, strongest opposing view, dissent, and conditions are preserved.
- Implementation owners, milestones, measures, due dates, and triggers are assigned.
- Recommendation matches the current article, motion, amount, and funding source.
- Later actuals and closeout can trace back to this decision.
See what connected data could add to the review
Follow one question across the available Town evidence, then see which source, relationship, or operating context would make the analysis more complete.
Continue through the full committee conversation or read the framework notes.
New decision test · August 20
A levy-affecting recommendation can now show its distribution assumption explicitly.
For a material request, state the additional annual levy or annual debt service by fiscal year, show the uniform-allocation sensitivity across assessed values, and then test recurring balance, debt timing, reserves and cash, capital needs, household burden, services, alternatives, and uncertainty.
Do not confuse sensitivity with advice.
The accepted FY2026 denominator improves transparency; it does not certify a parcel bill, predict classification policy, establish legal capacity, prove affordability, or choose an alternative. Dependable remaining recurring capacity is not determinable from current evidence.