How to read this case study. The engagement summary reflects the published record. Client names, actual costs, loan terms and proprietary data are confidential, so the project budget, market sizing and financial projections below are a modeled example built from public data for this project type and location. They show the analysis a lender-grade study performs, not the client's figures.
In 2024 this practice prepared a feasibility study for a drinking water treatment plant in Glasgow, Montana, under the USDA Water and Waste Disposal (WWD) program in the under $5 million tier. USDA was the lender, and the study was prepared with rate and affordability analysis. This page works through the questions that kind of study answers, using a modeled treatment plant upgrade, a modeled water utility of 1,780 equivalent dwelling units, and public data for Glasgow and Valley County.
A WWD application rests on two documents. The preliminary engineering report (PER) says what the plant needs and what it will cost; it belongs to the community's licensed engineer, and this practice does not prepare it. The rate and affordability analysis says whether the people on the system can pay for it, how much of the cost a loan can carry, and whether any of it qualifies for a grant. Rural Development sets the loan, the grant and the interest rate from that second set of numbers [1][8]. The model below shows a result many small towns do not expect: a plant upgrade under $5 million, spread over about 1,800 users, is affordable on a loan alone and does not reach the grant test, unless the system already carries heavy debt. Data current as of November 2026.
Engagement summary
| Published record | Detail |
|---|---|
| Program | USDA Water and Waste Disposal direct loan and grant (filed under WEP on the engagement index) |
| Project | Drinking water treatment plant |
| Location | Glasgow, Montana (Valley County, Mountain West region) |
| Loan size tier | Under $5 million |
| Year | 2024 |
| Lender | USDA direct |
| Outcome | Prepared with rate and affordability analysis |
Source: this site's engagement index [11]. The plant scope, users, budget, rates and projections below are a model, not the city's figures, rates or debt.
The engagement card on the index carried a longer summary. This page keeps what the regulations and data on file support, corrects what they contradict, and marks the rest as the card's own statement.
| Statement in the engagement card | Status on this page |
|---|---|
| USDA Rural Development committed $12.3 million to Montana water infrastructure in one 2021 cycle, including a $3.91 million loan and $7.36 million grant to Harlowton (population about 1,000) for a regional system serving six communities | Kept as the card's statement; the dollar figures were not checked. The population is consistent with the Census file: Harlowton had 959 residents in the April 2020 base and 976 in July 2025 [4] |
| The Central Montana Regional Water Authority's Musselshell-Judith project: a $54 million Phase 2 with 65 miles of pipeline, and $28.6 million of 2023 Bipartisan Infrastructure Law funding for Phases 3, 3A and 4 | Kept as the card's statement; not checked |
| Rural connection costs in Montana regional pipelines of about $171,165 per connection, "the baseline for rate and affordability modeling" | Kept as the card's figure for regional pipelines, but not used as this plant's baseline. A treatment plant upgrade in an existing town costs a small fraction of that per user; in this model, $2,679 per EDU |
| Montana DEQ regulates public water systems serving 25 or more people or 15 or more connections under the Safe Drinking Water Act | Kept as the card's statement; the definition and the state's primacy were not read for this page |
| Eligible borrowers are public entities and nonprofits in communities under 10,000 "with median HHI at or below 100% of the state non-metropolitan median" | Corrected. The 10,000 population test and the borrower types are right (1780.3, 1780.7) [1]. The 100% income line limits grants and sets the interest tier; it does not limit loan eligibility (1780.10(b), 1780.13) [1] |
| Long-term direct loans up to 40 years, with a grant if needed to keep user rates reasonable | Kept: the term is useful life, state statute or 40 years, whichever is less, and the grant is sized by the 1780.35(b) tests [1] |
| "Monthly user charge should not exceed 1.5% of median household income for hardship communities" | Corrected. Part 1780 tests the debt service share of the annual cost per EDU against 0.5% or 1.0% of median household income, then the cost of similar systems (1780.35(b)) [1]. The 1.5% whole-bill figure is shown below as a screen only |
| All projects require a PER to RUS bulletin standards | Kept in substance: a PER is required with the application and must conform to customary professional standards and the Agency's guidelines (1780.33(c), 1780.55) [1]. The bulletin number was not checked |
| Deliverables: deficiency documentation, design capacity selection, capital cost benchmarked to Mountain West pricing, 40-year amortization, rate per residential equivalent connection, affordability test against Montana non-metro median income | Consistent with the published outcome. The deficiency findings, capacity and cost estimate come from the engineer's PER; the amortization, rate and affordability work is the study's. This page shows each in model form |
The question USDA needed answered
Rural Development's question is not whether the plant is a good idea. The engineer and the State staff engineer settle that. The Agency's question is whether the users can pay: whether taxes, assessments, revenues or fees will cover operation and maintenance, reasonable reserves and debt payment (1780.7(f)), with rates set on a realistic number of users rather than the most the system could serve (1780.39(c)(1)), and how much of the cost a grant must carry so that the rate stays reasonable (1780.35(b)) [1]. Those answers set the loan amount, the grant, the interest rate tier and a large share of the priority score.
Part 1780 places the work on the owner and treats each piece as a separate professional job [1][8]:
| Document or task | Rule | Who prepares it | What the Agency checks |
|---|---|---|---|
| Preliminary engineering report (two copies with the application) | 1780.33(c), 1780.55 | The owner's engineer, licensed in Montana (1780.54) | Customary professional standards; State staff engineer review |
| Environmental report | 1780.33(f), 1780.55 | Environmental professionals under 7 CFR part 1970 | No action that limits alternatives before review ends (1780.57(a)) |
| Operating budget (Form RD 442-7) | 1780.33(h)(1), 1780.41(a)(2) | Applicant, with its accountant or feasibility consultant | Revenue covers O&M, reserves and debt service (1780.7(f)) |
| User estimates and rate schedule | 1780.39(c), 1780.35(c) | Applicant and its financial advisor or feasibility consultant | Realistic users; reasonable charges covering all costs |
| Grant determination | 1780.35(b) | Processing office, from the applicant's numbers | Debt service share of median household income; similar system cost |
| Credit elsewhere certification | 1780.7(d), 1780.33(d) | Applicant | Inability to finance at reasonable rates and terms |
The rate and affordability study is the bridge between the engineer's cost estimate and the city's rate ordinance. If the PER's design population and connection count differ from the numbers in the operating budget, a reviewer stops there [8].
Eligibility and program fit
| Test | Rule | Glasgow result |
|---|---|---|
| Rural area | Not in a city or town over 10,000 people under the latest decennial census (1780.3) | Eligible: 3,206 residents in the April 2020 estimates base, 3,145 in July 2025 [4] |
| Applicant | Public body, community-controlled nonprofit or tribe (1780.7(a)) | A city is a public body |
| Credit elsewhere | Cannot finance on reasonable rates and terms from its own resources or commercial credit (1780.7(d)) | Certified in the application; the study's rate comparison supports it |
| Legal authority | Authority to build, borrow, pledge and repay (1780.7(e)) | City ordinances and counsel's opinion |
| Project need and design | Will not decline below design population; capacity for present needs and reasonable growth (1780.7(c)) | Population trend tested below |
| Economic feasibility | Revenues cover O&M, reserves and debt (1780.7(f)) | The pro forma below |
| County setting | Non-metro proxy | Valley County lies outside any metropolitan or micropolitan area in the OMB July 2023 delineation [6] |
WWD guaranteed loans exist as well, under part 5001 with a 90% guarantee, a 1.0% fee and no annual retention fee in fiscal 2026, capped at $50 million per borrower [2][3]. For a small city, the direct program's 40-year term and income-based rate tiers usually decide the choice, and the guaranteed program's lender rate would have to compete with them.
Montana's towns sit almost entirely inside the direct program's reach. Of the 127 incorporated places in the Census file, 119 (93.7%) had 10,000 or fewer residents in the April 2020 base [4]:
| Montana incorporated places, April 2020 base | Places | Residents, 2020 base | Residents, July 2025 | Places that lost population | WWD population priority points |
|---|---|---|---|---|---|
| 1,000 or fewer | 74 | 30,392 | 31,788 | 31 | 25 |
| 1,001 to 2,500 | 25 | 42,436 | 45,272 | 10 | 15 |
| 2,501 to 5,500 (Glasgow) | 12 | 45,639 | 47,824 | 6 | 5 |
| 5,501 to 10,000 | 8 | 63,036 | 65,672 | 3 | 0 |
| 10,001 to 50,000 | 4 | 102,152 | 115,565 | 0 | Outside the direct rural test |
| Over 50,000 | 4 | 305,414 | 319,164 | 1 | Outside the direct rural test |
Source: Census Vintage 2025 subcounty estimates [4]; points from 1780.17(a) [1]. Glasgow is one of the 6 places in its band that lost population.
Glasgow earns 5 population points. A town of 900 earns 25 for the same project. That gap is one reason a Glasgow application has to win its score on health priorities, income and other funds, which the feasibility package documents.
Project description (modeled)
The model is an upgrade of an existing plant on its current site, not a new plant: a new filter train sized at 1.5 million gallons a day (MGD) of firm capacity to replace the oldest filters, chemical feed and disinfection improvements, recovery of backwash water, new pumps with variable frequency drives, controls and standby power. That is the kind of project that fits under $5 million for a town of about 3,100. The engineer's PER would choose the process, the capacity and the cost; the figures here are assumptions that let the rate work be shown.
Modeled users (equivalent dwelling units, EDUs)
| User class | Connections | EDUs | Basis |
|---|---|---|---|
| Residential | 1,470 | 1,470 | About 2.1 residents per connection on the 2025 population of 3,145 (assumed) |
| Commercial, institutional and public | 185 | 310 | Converted to EDUs by metered use (assumed) |
| Total realistic users | 1,655 | 1,780 | 1780.39(c)(1): realistic, not maximum |
Modeled water demand
| Measure | Value | Basis |
|---|---|---|
| Average billed use per EDU | 6,000 gallons a month | Assumed; summer irrigation raises the average |
| Water billed a year | 128.2 million gallons | EDUs x use |
| Non-revenue water | 12% | Assumed; 1780.57(f) asks that losses not exceed reasonable levels |
| Water produced a year | 145.6 million gallons | Billed / (1 - loss) |
| Average day production | 0.40 MGD | Arithmetic |
| Maximum day at a 2.6 peaking factor | 1.04 MGD | Assumed peaking factor |
| Firm treatment capacity in the model | 1.50 MGD | The engineer's PER selects it; assumed here |
Capacity above the maximum day is not waste in the rule's terms: 1780.57(d) and (e) ask for reasonable fire protection and capacity for growth, while 1780.10(a)(1) asks that facilities be modest in size, design and cost [1]. The study's job is to make sure the users pay for the capacity the engineer can justify, and that the rate is built on today's users, not on the growth the extra capacity allows.
Modeled schedule
| Phase | Months from engagement | What happens |
|---|---|---|
| PER and rate study | 0 to 4 | Engineer's PER; operating budget, users and rate analysis alongside |
| Application and review | 4 to 9 | Completeness check within 15 federal working days; eligibility and funding advice within 45 days of a complete file (1780.32) |
| Letter of conditions and design | 9 to 15 | Rate ordinance adopted; design and environmental review completed |
| Bidding and construction | 15 to 27 | Twelve-month build with interim financing for a loan over $500,000 (1780.39(d)) |
| Year 1 of operations | 27 to 39 | New rates in force; first full installment |
Project cost and sources and uses
Every line is an assumption with a range. A real budget comes from the engineer's PER estimate at the time of application and from bids at the end.
Modeled project budget
| Line item | Amount | Assumed range |
|---|---|---|
| Filtration: replace the oldest filter train with a 1.5 MGD train | $1,450,000 | $1,150,000 to $1,800,000 |
| Chemical feed and storage improvements | $310,000 | $220,000 to $420,000 |
| Disinfection and clearwell baffling for contact time | $260,000 | $180,000 to $360,000 |
| Backwash water recovery and residuals handling | $340,000 | $240,000 to $460,000 |
| Raw water pumping improvements and variable frequency drives | $280,000 | $200,000 to $380,000 |
| High-service pumps and variable frequency drives | $240,000 | $170,000 to $330,000 |
| Electrical, SCADA and instrumentation | $420,000 | $320,000 to $560,000 |
| Standby generator and transfer switch | $185,000 | $140,000 to $240,000 |
| Building envelope, heating and ventilation, laboratory | $220,000 | $150,000 to $320,000 |
| Construction subtotal | $3,705,000 | |
| Construction contingency, 10% at PER stage | $370,500 | 5% to 15%, falling as design advances |
| Engineering: PER, design, bidding, construction administration and inspection (13%) | $481,650 | Engineer's contract |
| Environmental report and review (7 CFR part 1970) | $25,000 | Environmental consultant |
| Legal and bond counsel | $35,000 | Counsel's engagement letter |
| Fiscal advisory: rate and affordability analysis | $12,500 | Inside the site's published $5,000 to $15,000+ range for rate analysis [9]; not this engagement's fee |
| Administration, permits and publication | $15,000 | Assumed |
| Interest on interim construction financing (12 months, 50% average draw, 5.50%) | $124,270 | Interim lender; 1780.39(d) |
| Total project cost | $4,768,920 |
Modeled uses of funds
| Use | Amount | Share |
|---|---|---|
| Construction and contingency | $4,075,500 | 85.5% |
| Engineering | $481,650 | 10.1% |
| Environmental, legal, fiscal and administration | $87,500 | 1.8% |
| Interim interest | $124,270 | 2.6% |
| Total | $4,768,920 | 100.0% |
Sources of funds (base case: loan and local cash, intermediate rate)
| Source | Amount | Share | Terms |
|---|---|---|---|
| USDA Water and Waste Disposal direct loan | $4,519,000 | 94.8% | 40 years at an illustrative 3.25% intermediate rate |
| City water fund reserves | $249,920 | 5.2% | Deposited before loan funds |
| USDA grant | $0 | 0.0% | Not triggered in the base case (see the grant test) |
| Total | $4,768,920 | 100.0% |
Every cost category in the budget is eligible under part 1780: construction, legal, engineering, fiscal advisory, environmental and administrative fees can be paid from loan or grant funds, and construction-period interest from loan funds (1780.9(e)) [1]. Fees paid before approval are reimbursed only on a written request and a finding of compelling reasons (1780.9(f)) [1], so a city that pays its engineer and rate consultant early carries that risk.
There is no equity percentage in WWD. The city's $249,920 from water fund reserves does two other jobs: it is 5.2% of the project in other-than-USDA money, which earns 5 priority points (1780.17(d)(4)), and it shows the State Office that the utility has kept reserves [1]. Raising other funds to 20% would earn 10 points.
At $4,768,920 for 1,780 EDUs, the project costs $2,679 per EDU. The card's regional pipeline figure of $171,165 per connection is about 64 times that. The difference explains most of what follows: Part 1780's grant tests are built for projects whose cost per user is high, such as long pipelines to scattered users, and a treatment upgrade in an established town rarely is.
Market analysis: users, population and the commercial base
In a WWD study the "market" is the user base: how many users there are, whether they will stay, how much water they buy, and which of them are businesses whose fortunes the rate depends on.
Population and trend
Glasgow had 3,206 residents in the April 2020 base and 3,145 in July 2025, a decline of 1.9%, or about 0.37% a year [4]. Valley County fell 2.6%, from 7,584 to 7,385, because deaths exceeded births by 155 and net domestic migration took away 68 more people than it brought [5]. The county's other incorporated places are small: Nashua (292 in 2025), Fort Peck (232) and Opheim (73); 3,643 people live outside incorporated places [4].
Regional comparison: northeastern Montana
| County | County seat | OMB 2023 status | April 2020 base | July 2025 | Change | Births less deaths | Net international | Net domestic |
|---|---|---|---|---|---|---|---|---|
| Valley | Glasgow | Outside any CBSA | 7,584 | 7,385 | -2.6% | -155 | +25 | -68 |
| Phillips | Malta | Outside any CBSA | 4,216 | 4,203 | -0.3% | -78 | +13 | +52 |
| Roosevelt | Wolf Point | Outside any CBSA | 10,804 | 10,242 | -5.2% | -25 | +20 | -575 |
| Daniels | Scobey | Outside any CBSA | 1,653 | 1,612 | -2.5% | -30 | +2 | -13 |
| McCone | Circle | Outside any CBSA | 1,728 | 1,720 | -0.5% | +24 | +0 | -31 |
| Garfield | Jordan | Outside any CBSA | 1,170 | 1,148 | -1.9% | +0 | -1 | -19 |
| Sheridan | Plentywood | Outside any CBSA | 3,547 | 3,436 | -3.1% | -109 | +40 | -37 |
Sources: Census Vintage 2025 county estimates, components summed from April 2020 to July 2025 [5]; OMB July 2023 delineation [6].
Every county in the table lost population, and in five of the seven deaths outnumbered births. Montana as a whole grew 5.6% over the same period, entirely through migration: the state had 3,631 more deaths than births, while net domestic migration added 59,510 people and international migration 4,792 [5]. In 34 of Montana's 46 non-metro counties deaths exceeded births [5]. Northeastern Montana is not sharing in the state's in-migration. For a forty-year loan that sets the base case: a slowly shrinking user base, with the rate study showing what each remaining user pays as the count falls.
Realistic users over the 40-year loan
| Scenario | Year 10 | Year 20 | Year 40 | Installment per EDU in Year 40 (flat installment) |
|---|---|---|---|---|
| Observed Glasgow trend, -0.37% a year | 1,716 | 1,654 | 1,538 | $132 |
| Faster decline, -1.0% a year | 1,610 | 1,456 | 1,191 | $171 |
| Stable | 1,780 | 1,780 | 1,780 | $114 |
Part 1780 asks that the area served not be likely to fall below the design population (1780.7(c)(1)) [1]. On the observed trend the user count falls about 14% over the loan, and the installment per EDU rises from $114 to $132 a year in today's dollars. That is manageable because the debt share of the bill is small. It also argues against building capacity for growth the trend does not show; the PER's growth allowance should rest on evidence, such as fire flow or a known commercial user, not on a hoped-for reversal.
The commercial base
| Valley County, 2023 | Establishments | Employees |
|---|---|---|
| Hotels and motels | 6 | 165 |
| Full-service restaurants | 11 | 102 |
| Limited-service restaurants | 6 | 70 |
| Gasoline stations with convenience stores | 5 | 38 |
| Offices of physicians | 4 | 55 |
| Outpatient care centers | 3 | 23 |
| Manufacturing (all) | 5 | 9 |
Source: Census County Business Patterns 2023, private establishments with paid employees [7]. County-wide; most but not all are likely on the city system.
The commercial users are service businesses, not industry: lodging, food service, fuel and health care. No single user dominates, which means the 1780.7(f) test for a system that depends on one business does not arise here [1]. The commercial EDUs still matter, because 310 of the model's 1,780 EDUs come from 185 connections, and lodging and restaurant use follows the summer travel season. The rate study bills them by meter, and the sensitivity cases test lower volumes.
Financial feasibility
Income tier: what it sets
Median household income decides the interest tier, the grant ceiling, the grant threshold and the income priority points. Neither Glasgow's income nor Montana's statewide non-metro median is in this site's data, so the model uses stated assumptions and tests a range.
| Input | Value | Effect |
|---|---|---|
| Service-area median household income | $56,000 | Assumed; five-year ACS in a real study (1780.1(b)) |
| Montana statewide non-metro median household income | $63,000 | Assumed; set by the Agency |
| Ratio | 88.9% | Between 80% and 100% |
| Interest rate tier | Intermediate: 80% of market, not over 7% | 1780.13(c) |
| Illustrative rates | Market 4.125%, intermediate 3.25%, poverty 2.50% | Rounded to the nearest one-eighth (1780.13(a)) |
| Maximum grant | 45% of eligible project cost | 1780.10(c) |
| Debt service threshold for grant consideration | 1.0% of MHI = $560 per EDU a year | 1780.35(b)(2) |
| Income priority points | 15 | 1780.17(c)(3) |
The illustrative market rate is the one this site's national WWD page uses; Rural Development publishes the actual rates each quarter [8].
Modeled operating budget, Year 1 (assumptions, not Glasgow's budget)
| Item | Amount | Per EDU a year |
|---|---|---|
| Personnel: 4.0 FTE (certified treatment and distribution operators, share of billing) at $78,000 loaded | $312,000 | $175 |
| Electricity: 2,000 kWh per million gallons at $0.11, plus buildings | $48,040 | $27 |
| Treatment chemicals: $0.25 per 1,000 gallons produced | $36,409 | $20 |
| Laboratory and compliance monitoring | $22,000 | $12 |
| Repairs and maintenance, plant and distribution | $110,000 | $62 |
| Insurance | $28,000 | $16 |
| Billing, administration, audit and allocated overhead | $95,000 | $53 |
| Operation and maintenance before the project | $651,449 | $366 |
| Net added O&M from the upgrade (chemicals and residuals +$42,000, pump energy savings -$12,000) | $30,000 | $17 |
| Short-lived asset replacement reserve (1780.35(c), 1780.39(e)) | $85,000 | $48 |
Personnel is almost half the operating budget, as it is in most small water systems; certified operators are the fixed cost a shrinking user base cannot reduce. A real study builds this table from three years of audited statements and the city's adopted budget, and the PER's estimate of the new plant's chemical, power and residuals costs.
Grant determination under 1780.35(b), step by step (base case)
| Step | Calculation | Result |
|---|---|---|
| 1. Debt service threshold per EDU | 1.0% x $56,000 | $560 a year |
| 2. Proposed debt service | $4,519,000 at 3.25% for 40 years | $203,481 a year |
| 3. Debt service per EDU | $203,481 / 1,780 EDUs | $114 a year (0.20% of MHI) |
| 4. Is debt service above the threshold? | $114 vs $560 | No: no grant is considered |
| 5. Debt the users could carry at the threshold | $560 x 1,780 / 0.04503 | about $22,137,000 |
| 6. Similar system cost test | Applies only if a grant is made | Not reached |
The order of the tests is fixed by the rule: compare incomes, set the ceiling and threshold, consider a grant only where debt service per EDU exceeds the threshold, cut it back if it would push the annual EDU cost below similar systems, and put the rest on the loan at the maximum term [1][8]. In this model the test stops at step 4.
How large the project must be before a grant is considered (1,780 EDUs, 40 years)
| Case | Threshold per EDU a year | Users' debt service at the threshold | Loan at which the grant test starts | Debt per EDU at that point |
|---|---|---|---|---|
| Intermediate tier, MHI $56,000, threshold 1.0% | $560 | $996,800 | $22,137,000 | $12,437 |
| Health project, MHI $49,000 (77.8%), poverty rate, threshold 0.5% | $245 | $436,100 | $10,947,000 | $6,150 |
Even if Glasgow's income fell below 80% of the state non-metro median and the plant were needed to meet health standards, the city could carry about $10.9 million of new debt before a grant came into play. For a sponsor this is the most useful number in the study. It tells the council early that the application should be built as a loan, that the grant line in the funding plan should be zero unless existing debt changes the picture, and that the score must be earned elsewhere.
If the system already carries debt: when the grant test starts (illustrative)
| Existing annual debt service (assumed) | Debt service per EDU, intermediate case | Above $560 (1.0% of $56,000)? | Debt service per EDU, health project at the poverty rate | Above $245 (0.5% of $49,000)? | Grant to bring debt service to the threshold (health case, 75% cap) | Grant share |
|---|---|---|---|---|---|---|
| $0 | $114 | No | $101 | No | $0 | 0.0% |
| $150,000 | $199 | No | $185 | No | $0 | 0.0% |
| $300,000 | $283 | No | $270 | Yes | $1,102,512 | 23.1% |
| $500,000 | $395 | No | $382 | Yes | $3,576,690 | 75.0% |
The model assumes no existing water debt, because this site does not know the city's. The table shows why a real study starts from the debt schedule: if the threshold test counts existing debt service, a system already paying $300,000 a year could qualify for a grant of about $1.1 million in the health and low-income case, and none in the intermediate case. How existing debt enters the calculation should be confirmed with the processing office before the funding plan is written.
Five-year pro forma (modeled)
| Line | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 |
|---|---|---|---|---|---|
| Realistic users (EDUs), trend -0.37% a year | 1,780 | 1,773 | 1,767 | 1,761 | 1,754 |
| Base charge per EDU a month (+3% a year) | $21.70 | $22.35 | $23.02 | $23.71 | $24.42 |
| Volume charge per 1,000 gallons (+3% a year) | $4.42 | $4.55 | $4.69 | $4.83 | $4.97 |
| Water revenue | $1,029,890 | $1,056,912 | $1,084,643 | $1,113,102 | $1,142,308 |
| Operation and maintenance (+3% a year) | $681,449 | $701,635 | $722,420 | $743,821 | $765,858 |
| Net revenue before reserves and debt | $348,441 | $355,277 | $362,223 | $369,281 | $376,450 |
| Short-lived asset reserve | $85,000 | $87,550 | $90,176 | $92,882 | $95,668 |
| WWD loan installment | $203,481 | $203,481 | $203,481 | $203,481 | $203,481 |
| Debt reserve deposit (one tenth of an installment) | $20,348 | $20,348 | $20,348 | $20,348 | $20,348 |
| Surplus | $39,611 | $43,897 | $48,217 | $52,570 | $56,952 |
| Coverage of the installment, before reserves | 1.71x | 1.75x | 1.78x | 1.81x | 1.85x |
| Coverage after the short-lived asset reserve | 1.29x | 1.32x | 1.34x | 1.36x | 1.38x |
| Average monthly bill per EDU (6,000 gallons) | $48.22 | $49.66 | $51.15 | $52.69 | $54.27 |
The rate schedule is designed, not observed: 45% of revenue from a base charge and 55% from volume, set to collect about 4% more than the full revenue requirement in Year 1 and indexed at 3% a year. Part 1780 asks for rates that cover all costs and reserves without building a substantial surplus [1][8], so the cushion is deliberately thin. The protection is the annual rate review, not a large margin.
Year 1 cost per EDU and the affordability measures
| Component | Per EDU a year | Per EDU a month | Share of assumed MHI |
|---|---|---|---|
| Operation and maintenance | $383 | $31.90 | 0.68% |
| Short-lived asset reserve | $48 | $3.98 | 0.09% |
| Debt service | $114 | $9.53 | 0.20% |
| Debt reserve | $11 | $0.95 | 0.02% |
| Cushion (surplus) | $22 | $1.85 | 0.04% |
| Total bill | $579 | $48.22 | 1.03% |
The rule's measure is the debt service row: $114 a year, 0.20% of the assumed income, against a 1.0% threshold. The whole bill is 1.03% of income, under the 1.5% screen the engagement card cites. Operation and maintenance is two thirds of the bill; the project's debt is a fifth.
What the project alone adds to each EDU
| Item | Per year | Per EDU a month |
|---|---|---|
| Loan installment | $203,481 | $9.53 |
| Debt reserve | $20,348 | $0.95 |
| Net added O&M | $30,000 | $1.40 |
| Total added by the project | $253,829 | $11.88 |
This is the number a council can take to a public meeting: on these assumptions the upgrade adds about $11.88 a month to a typical bill. The public information meeting that 1780.19(b) requires before approval, unless a vote authorizes the debt, is where it will be asked [1].
Interest rate tier sensitivity (rates are illustrative)
| Rate | Annual installment | Debt service per EDU a year | Debt and reserve per EDU a month | Coverage after reserve, Year 1, at model rates |
|---|---|---|---|---|
| Poverty rate (health project, income below 80%): 2.50% | $180,020 | $101 | $9.27 | 1.46x |
| Intermediate rate (model): 3.25% | $203,481 | $114 | $10.48 | 1.29x |
| Market rate: 4.125% | $232,580 | $131 | $11.98 | 1.13x |
| Market rate a full point higher: 5.125% | $267,882 | $150 | $13.80 | 0.98x |
The tier is worth about $1.20 a month per EDU between poverty and intermediate, and $1.50 more from intermediate to market. The rule gives the borrower the lower of the rate at approval or closing unless it chooses otherwise (1780.13(a)) [1]. If the loan closed at the market tier rather than the intermediate one, the model's rates would need about 3.1% more revenue; a market rate a full point higher would need about 6.9%, which is a rate decision the annual review can carry.
User and volume sensitivity at model rates
| Case | EDUs, Year 5 | Revenue, Year 1 | Coverage after reserve, Year 1 | Coverage after reserve, Year 5 | Surplus, Year 5 |
|---|---|---|---|---|---|
| Model: 1,780 EDUs, trend decline | 1,754 | $1,029,890 | 1.29x | 1.38x | $56,952 |
| 150 fewer EDUs connect or stay (-8.4%) | 1,606 | $943,101 | 0.90x | 0.94x | -$32,912 |
| Faster decline: -1.0% a year | 1,710 | $1,029,890 | 1.29x | 1.25x | $30,036 |
| Use falls to 5,000 gallons a month | 1,754 | $935,483 | 0.89x | 0.93x | -$35,106 |
Volume is the sharper risk. A wet summer or conservation that cuts average use to 5,000 gallons a month takes about $94,000 off Year 1 revenue, and 150 fewer EDUs take about $87,000. In both cases the fix is a rate increase of a few percent, about 4.4% for the lost users, because most costs are fixed. On the model's assumptions the plan covers all costs and reserves as long as at least about 1,712 EDUs pay, 3.8% fewer than modeled. That is why the study puts more weight on the base charge than a conservation-minded rate design might, and why 1780.39(c) asks for realistic users rather than hoped-for ones [1].
Cost overrun and delay, financed with added loan funds
| Case | Added cost | Loan | Installment | Debt and reserve per EDU a month | Coverage after reserve, Year 1, at model rates |
|---|---|---|---|---|---|
| Model | $0 | $4,519,000 | $203,481 | $10.48 | 1.29x |
| Bids 10% over the construction estimate | $370,500 | $4,889,500 | $220,164 | $11.34 | 1.20x |
| Bids 20% over | $741,000 | $5,260,000 | $236,847 | $12.20 | 1.11x |
| Twelve-month delay at 4% escalation | $163,020 | $4,682,020 | $210,822 | $10.86 | 1.25x |
The rate effect of an overrun is small: a 20% overrun adds $1.72 a month per EDU. The tier effect is larger: the same overrun takes the loan past $5 million. The 10% contingency covers the first case; the second needs bid alternates, a smaller first phase or additional funds, and the Agency's review of any amended loan.
Water loss: what non-revenue water costs
| Non-revenue water | Million gallons produced a year | Year 1 O&M | Coverage after reserve, Year 1 |
|---|---|---|---|
| 12% (model) | 145.6 | $681,449 | 1.29x |
| 20% | 160.2 | $688,294 | 1.26x |
| 30% | 183.1 | $699,050 | 1.21x |
At this plant's chemical and power costs, losing more water costs relatively little in operating money. The larger cost is capacity: water lost in the distribution system has to be treated, and a plant sized for 30% loss is a plant sized for water nobody buys. Part 1780 asks that losses in an existing system not exceed reasonable levels (1780.57(f)) [1], so a high loss figure belongs in the PER's alternatives, and possibly in the project, before the treatment capacity is fixed.
Median household income scenarios (bill held at the Year 1 model level)
| Service-area MHI | Share of assumed state non-metro MHI | Rate tier | Maximum grant | Debt service threshold | Year 1 bill as share of MHI |
|---|---|---|---|---|---|
| $45,000 | 71.4% | Poverty if a health or sanitary project; otherwise intermediate | 75% (health project) or 45% | $225 per EDU | 1.29% |
| $49,000 | 77.8% | Poverty if a health or sanitary project; otherwise intermediate | 75% (health project) or 45% | $245 per EDU | 1.18% |
| $56,000 (model) | 88.9% | Intermediate | 45% | $560 per EDU | 1.03% |
| $63,000 | 100.0% | Intermediate | 45% | $630 per EDU | 0.92% |
| $70,000 | 111.1% | Market | No grant | n/a | 0.83% |
Income moves the tier and the threshold more than the bill. In every case the project's debt service of about $101 to $131 per EDU stays under the threshold, so the grant answer does not change; the interest rate and the priority points do. Where five-year survey data look wrong for a town this size, the rule allows a reliable income survey instead (1780.1(b)) [1], and that decision belongs before the application, not after the grant has been calculated.
Combined stress case
| Stress | Setting |
|---|---|
| Construction bids | +15%, financed |
| Rate tier | Market 4.125% instead of intermediate |
| Users | 120 fewer EDUs |
| Water loss | 20% instead of 12% |
| Fixed O&M | +5% |
| Loan | $5,074,750 |
| Coverage after reserve, Year 1 / Year 5 | 0.62x / 0.65x |
| Rate increase needed in Year 1 to cover all costs and reserves | 12.9% |
Everything going wrong at once requires a 12.9% rate increase in Year 1. That is a hard council vote, but it is a rate decision, not a solvency event, and it would still leave the bill well under 1.5% of the assumed income. The study's recommendation in a case like this is to adopt the rate ordinance with an annual review clause before closing, so that the increase is a scheduled step rather than an emergency.
Technical and environmental feasibility
- The PER decides the technical case. Process selection, capacity, the alternatives considered and the cost estimate are the engineer's, prepared to customary professional standards and reviewed by the State staff engineer (1780.33(c), 1780.55) [1]. The study adopts the PER's numbers and checks that its connection count and design population match the operating budget.
- Safe Drinking Water Act progress. A water project must show how it moves the system toward compliance (1780.57(g)), and a documented health requirement is worth 25 priority points and opens the poverty rate tier [1].
- Metering and losses. Each connection is metered unless an exception is justified (1780.57(m)), and losses must be reasonable (1780.57(f)) [1].
- Cold-weather design. A northern plant needs heated process space, freeze protection on chemical and backwash lines, and standby power for winter outages; the model carries the building envelope and the generator as their own lines.
- Residuals. Backwash water recovery reduces water lost to the process and the volume of residuals to manage; the PER sets the method and its permit path.
- Environmental review. No action that limits alternatives may be taken before the review under 7 CFR part 1970 ends (1780.57(a)) [1].
Management and governance feasibility
A city water utility is managed by its council and run by its certified operators. The study looks for the things part 1780 asks of every borrower: financial, technical and managerial capability, and evidence that alternatives of ownership, design and shared services were considered (1780.1(j)) [1]. In practice that means audited statements and accrual-basis records (1780.47), an enterprise fund that keeps water revenue for water costs, a rate ordinance the council has shown it will update, operator certification at the level the new plant requires, and a succession plan for the chief operator. In a county losing population, the study also asks whether the plant could serve neighbouring systems; merging facilities earns 15 points (1780.17(d)(1)) and spreads the operator cost over more users [1].
Priority points the feasibility package documents (modeled)
| Criterion | Points | Section |
|---|---|---|
| Population 2,501 to 5,500 (Glasgow 3,206 in the 2020 base) | 5 | 1780.17(a)(3) |
| Safe Drinking Water Act requirement, if the PER documents one | 25 | 1780.17(b)(1) |
| Income 80% to 100% of state non-metro median (assumed) | 15 | 1780.17(c)(3) |
| Public body | 5 | 1780.17(d)(3) |
| Other than RUS funds: city reserves of $249,920 (5.2%) | 5 | 1780.17(d)(4) |
| Unreliable drinking water quality or supply, if documented | 10 | 1780.17(d)(7) |
| Modeled total before discretionary points | 65 | |
| State program official and National Office points | up to 15 each | 1780.17(e), (g) |
Applications are funded in priority order when money is short (1780.36) [1]. The two lines the city controls most directly are the health documentation in the PER and the other-funds share; moving from 5.2% to 20% of non-USDA money adds 5 points.
Process and documents
| Stage | What happens | Rule |
|---|---|---|
| Early contact | Processing office explains eligibility and technical assistance on income surveys and rates | 1780.31(a), (b) |
| PER preliminary review (optional) | PER submitted before the rest of the application | 1780.33(c)(1) |
| Application | SF 424.2, two PER copies, environmental report, credit elsewhere certification, eligibility documents | 1780.33 |
| Completeness and eligibility | Missing items within 15 federal working days; eligibility and funding advice within 45 days of a complete file | 1780.32 |
| Public notice and meeting | Notice within 60 days of filing; public information meeting with 10 days' notice unless a vote authorizes the debt | 1780.19 |
| Full application | Operating budget on Form RD 442-7, user agreements where required, professional service contracts | 1780.39 |
| Letter of conditions and obligation | Issued when funds are available; sets the rate | 1780.39(i), 1780.41 |
| Construction | Interim financing for loans over $500,000 where available | 1780.39(d) |
Sources: [1], summarized from this site's national WWD page [8]. Documents the rate study needs: the PER or its draft and the engineer's estimate; three years of audits; the current rate ordinance; the customer list by class with metered use; existing debt instruments; any income survey; and letters for other funds. This site's engagements typically run 4 to 8 weeks, with a 3 to 4 week rush option [12].
Economic and community impact (modeled)
| Measure | Value | Basis |
|---|---|---|
| Residents served | 3,145 | Census Vintage 2025 [4] |
| Realistic users | 1,780 EDUs on 1,655 connections | Model |
| Construction spending (construction and contingency) | $4,075,500 | Budget |
| On-site job-years (35% labor share at $70,000 loaded, assumed) | 20 | Arithmetic, not a multiplier model |
| Monthly bill added by the project | $11.88 per EDU | Model |
| Commercial establishments in the county that depend on public water (lodging, food service, fuel, health care) | 35 | Census CBP 2023 [7] |
The impact a State Office reads first is public health: a treatment plant that keeps the town's water within federal standards and its supply reliable through a northern winter. The economic case follows from it. The county's hotels, restaurants and clinics cannot operate on a boil order, and a town that is losing residents cannot afford to lose a reason for businesses to stay.
Risks and mitigants
| Risk | How it shows up in the model | Mitigant the study tests |
|---|---|---|
| Lower water sales | 5,000 gallons a month takes coverage to 0.89x | Base charge share; annual rate review clause |
| Fewer users | 150 fewer EDUs need a 4.4% increase | Realistic count; customer list reconciled to meter records |
| Long-run decline | 1,538 EDUs in Year 40 on the observed trend | No unproven growth capacity; regional service options |
| Construction cost | +20% bids take the loan past $5 million | 10% contingency; bid alternates; phasing |
| Interest tier | A full point above market cuts coverage to 0.98x | Lower of approval or closing rate; income survey if data look wrong |
| Existing debt | Changes the grant answer | Debt schedule first; confirm treatment with the processing office |
| Water loss | Capacity built for lost water | Loss audit in the PER; distribution repairs in scope |
| Operator capacity | A new process needs a higher certification | Training budget; shared services with neighbours |
What the five Part 5001 components showed
WWD direct is governed by part 1780, and the engineering case belongs to the PER. This site still organizes the rate and affordability work around the five components it uses for every USDA study:
| Component | What the model shows | What a reviewer still asks |
|---|---|---|
| Economic | A service town of 3,145 in a county losing 2.6% since 2020 | The city's own customer trend and commercial base |
| Market (users) | 1,780 realistic EDUs; 1,538 in Year 40 on trend | Customer list by class, metered use, vacant connections |
| Technical | 1.5 MGD upgrade within $4.77 million | The PER, State staff engineer review, compliance status |
| Financial | Debt service $114 per EDU against a $560 threshold; bill 1.03% of assumed income | Audited budget, existing debt, actual income data |
| Management | Council rate decisions and certified operators | Audits, rate review history, operator certification |
Why WWD rate studies are sent back
- The rate is built on the maximum users the system could serve rather than a realistic count (1780.39(c)(1)) [1].
- The PER and the operating budget use different connection counts or design populations.
- The operating budget leaves out the debt reserve or short-lived asset replacement (1780.39(e)) [1].
- A grant is assumed in the funding plan before the threshold test shows one.
- Income data are disputed after the grant has been calculated (1780.1(b)) [1].
- Existing debt is left out of the debt service picture.
- Fees were paid before approval without a written request (1780.9(f)) [1].
Program terms that applied
| Term | WWD direct rule | Source |
|---|---|---|
| Rural area | Not in a city or town over 10,000 under the latest decennial census | 1780.3 [1] |
| Rate tiers | Poverty (60% of market, not over 5%), intermediate (80% of market, not over 7%), market (Bond Buyer index); set quarterly, rounded to one-eighth | 1780.13 [1] |
| Term | Useful life, state statute or 40 years, whichever is less; principal deferral up to 36 months | 1780.13(e) [1] |
| Grant ceiling | 75% (income below the higher of the poverty line or 80% of state non-metro median, health project) or 45% (up to 100%) | 1780.10(c) [1] |
| Grant threshold | Debt service per EDU above 0.5% or 1.0% of median household income; reduced to the similar system cost | 1780.35(b) [1] |
| Reserves | Debt reserve of one average installment, built at one tenth a year; short-lived asset reserves | 1780.39(e), 1780.35(c) [1] |
| Interim financing | Commercial interim financing for loans over $500,000 where available | 1780.39(d) [1] |
| Guaranteed alternative (FY2026) | 90% guarantee, 1.0% fee, no retention fee, 0.50% if issued before construction completion; $50 million cap | 91 FR 11272; part 5001 [2][3] |
Lessons for sponsors of similar projects
- Run the grant test first. In this model the project would have to cost about $22 million, or $10.9 million in the low-income health case, before a grant came into play. Knowing that early changes the funding plan.
- Start from the debt schedule. Existing debt service can change the grant answer; the rate study needs it on day one.
- Bill on realistic users and real volumes. Water sales, not the interest rate, are the sharpest risk to coverage.
- Match the PER and the budget. One connection count, one design population, one cost estimate.
- Earn the points the city controls. Health documentation and a 20% non-USDA share are worth more than the population band.
- Adopt a rate review clause. A thin cushion is what the rule wants; an annual review is what makes it safe.
Related pages
- The Water and Waste Disposal engagement page lists this program's engagements, including this one.
- The USDA Water and Waste Disposal feasibility study page sets out part 1780, the grant tests, priority points and a worked rate example [8], and the Water and Waste Disposal program article explains the program and who leads each piece [9].
- The Montana state page carries the state's data, eligibility notes and project guides [10].
- The case studies page lists the site's long-form work, including the water and waste disposal case study for Mississippi.
- The USDA rural definition article explains why the WWD population line is 10,000 while other programs use 20,000 or 50,000.
- The USDA Feasibility Study Framework and the cost and timeline page set out the method and fees [12].
If your city, district or water association is preparing a WWD application and needs a USDA feasibility study consultant for the rate and affordability work alongside your engineer's PER, the engagement index describes the process: a fifteen-minute scoping call, a written proposal inside one business day and a firm fixed fee before you sign [11]. Request a proposal for your study [13].
Data current as of November 2026.
USDA Feasibility, Inc.