Problem Definition
CityWater is the publicly owned drinking water utility of a mid-size Midwestern U.S. city of about 480,000 residents. It serves 165,000 service connections (150,000 residential, 15,000 commercial and industrial) through 1,600 miles of water main and two treatment plants. Annual revenue is $150 million, of which residential bills contribute $78 million (an average residential bill of $520 per year).
The network is old. About 45% of CityWater's mains are cast iron or asbestos cement, much of it laid before 1970. Last year the utility recorded 282 main breaks, about 17.6 breaks per 100 miles of pipe, well above the approximately 11 breaks per 100 miles reported as the U.S. and Canada average in the most recent Utah State University survey. Two of last winter's breaks were on a downtown trunk main and closed a hospital's main entrance road for three days.
At the same time, CityWater faces a federal deadline. Under the U.S. EPA's Lead and Copper Rule Improvements (LCRI), finalized in October 2024, community water systems must submit a baseline service-line inventory and a replacement plan by November 1, 2027 and replace all lead service lines by November 1, 2037. Nationally, EPA estimates there are approximately 4 million lead service lines still in the ground.
The city council has approved a capital envelope of $110 million per year for the next ten years (funded through revenue bonds, state revolving fund loans and rates). Today CityWater spends about $70 million per year on capital projects. The general manager has hired your team. How should CityWater prioritise its pipe-replacement capital over the next decade, and what does that mean for customer bills?
Additional InformationAsk for dataInterviewer’s data
If asked, please share that:
- CityWater's service-line inventory shows 21,000 confirmed lead service lines and 9,000 lines of unknown material. Where CityWater has dug up "unknown" lines so far, about one in three turned out to be lead
- Replacing a lead service line is a regulatory obligation with a hard deadline; it is not optional capital
- Treatment plants, pumping stations, meters and other non-pipe capital need about $25 million per year
- A common industry planning rule is to renew about 1% of water mains per year (implying a 100-year pipe life); AWWA notes systems that have fallen behind may need 2% or more
- The city has an AA- credit rating; the state public utility commission does not regulate municipal water rates, but the council must vote on any rate increase
Question 1Structuring
How would you structure the decision on how to allocate CityWater's $110 million annual capital budget?
Hint · Structuring
Build 3–4 branches that are specific to this client and question, not a generic framework. Check they don't overlap and together cover the problem.
Additional InformationAsk for dataInterviewer’s data
- Share the budget ($110M/year), the LCRI deadline (all lead service lines replaced by November 2037) and the break rate (17.6 per 100 miles vs. about 11 nationally)
- If the candidate asks about consequences of failure, confirm that not all breaks are equal: a break under a downtown arterial or near critical customers (hospital, fire service) costs far more than one on a residential cul-de-sac
Try it first, then checkCheck my answerModel answer
A strong structure has three layers:
1. What must be funded (mandatory)
- a) Lead service line replacement: size of the inventory (confirmed plus expected share of unknowns), unit cost, and the 2037 deadline
- b) Regulatory and safety-critical treatment and pumping capital (about $25M/year)
2. How to rank the discretionary main-replacement programme (risk-based)
- a) Likelihood of failure: pipe material, age, break history, soil corrosivity, pressure
- b) Consequence of failure: location (arterial roads, downtown), critical customers served (hospitals, fire flow), water lost, service interruption
- c) Cost and method: full replacement vs. rehabilitation (lining, cathodic protection), and coordination with road resurfacing to share trench costs
- d) Risk reduced per dollar as the ranking metric
3. How to pay for it (funding and affordability)
- a) Mix of rates, revenue bonds and low-cost state revolving fund loans or grants
- b) Resulting bill increase for the average household and for low-income households
- c) Phasing: what must happen in years 1-3 (inventory, lead lines in high-risk areas) vs. later
What the interviewer is looking forInterviewer’s viewInterviewer’s view
This question tests whether the candidate recognises that the budget has a mandatory layer and a discretionary layer, and that discretionary pipe spending should be ranked on risk (likelihood x consequence), not on age alone.
- Good candidates list the spending categories (lead service lines, mains, treatment) and propose ranking by condition
- Strong candidates separate non-negotiable spending (regulatory mandate, safety-critical treatment) from discretionary spending, and bring in a risk framework for mains
- Excellent candidates also include the funding side (rates, debt, affordability) and the timing constraint (the 2037 deadline compresses lead-line spending into roughly ten years), and ask what "success" means for the council: fewer breaks, compliance, stable bills, or all three
Question 2Numeracy
Using Exhibit 1 and the information provided, how much of the $110 million annual budget is left for main replacement once the lead-line mandate and other capital are funded? Is that enough?
Hint · Numeracy
Write the formula before you plug in numbers, keep units and zeros explicit, and sanity-check the order of magnitude at the end.
Exhibit 1CityWater Main Inventory and Break Rates
| Pipe material | Miles | Average age (years) | Breaks last year | CityWater breaks per 100 miles | U.S./Canada average breaks per 100 miles |
|---|---|---|---|---|---|
| Cast iron | 480 | 78 | 182 | 37.9 | 28.6 |
| Asbestos cement | 240 | 58 | 38 | 15.8 | 10.3 |
| Ductile iron | 420 | 35 | 42 | 10.0 | 5.1 |
| PVC | 400 | 20 | 14 | 3.5 | 2.9 |
| Steel and other | 60 | 50 | 6 | 10.0 | 9.2 |
| Total | 1,600 | -- | 282 | 17.6 | 11.1 |
U.S./Canada averages from the Utah State University water main break survey (2023 data, published 2024). CityWater figures from utility records.
Source: CityWater case file
Additional InformationAsk for dataInterviewer’s data
- Unit cost for lead service line replacement: EPA's earlier estimate was about $4,700 per line; an AWWA-commissioned study by CDM Smith found about $12,500 per line (2022 dollars) based on actual utility costs. CityWater's own recent contracts are close to the AWWA figure; use $12,500
- AWWA's planning estimate for water main renewal is about $500-600 per linear foot (2025 dollars); use $550 per foot, which is about $2.9 million per mile
- Assume the lead-line programme is spread evenly over ten years
Try it first, then checkCheck my answerModel answer
Step 1: Size the lead service line programme
- Confirmed lead lines: 21,000
- Expected lead among unknowns: 9,000 x 1/3 = 3,000
- Total lines to replace: 24,000
- Total cost: 24,000 x $12,500 = $300M; over ten years = $30M per year
- (At EPA's older $4,700 estimate it would be $112.8M, or about $11M per year. The gap between the two estimates is itself a risk worth flagging.)
Step 2: Subtract mandatory capital
| Item | $M per year |
|---|---|
| Capital budget | 110.0 |
| Lead service lines | -30.0 |
| Treatment, pumping, meters | -25.0 |
| Available for mains | 55.0 |
Step 3: Compare with the cost of renewal
- 1% renewal = 1% x 1,600 miles = 16 miles per year
- Cost per mile = $550 x 5,280 feet = $2,904,000, about $2.9M
- 1% renewal cost = 16 x $2.9M = $46.4M per year, which fits within $55M (about $8.6M to spare)
- But at 16 miles per year it would take 480 / 16 = 30 years to replace only the cast-iron mains
- A 1.5% catch-up rate (24 miles per year) costs 24 x $2.9M = $69.6M, which exceeds the $55M available by $14.6M per year
Conclusion: The budget covers compliance plus a "treading water" 1% renewal rate. It does not cover a catch-up programme. CityWater either needs about $15M more per year, or it must get much more risk reduction out of each dollar it spends on mains.
What the interviewer is looking forInterviewer’s viewInterviewer’s view
Look for a clean structure: size the mandatory layer first, then compare the remainder to what a sustainable renewal rate would cost. Common mistakes: forgetting the unknown-material lines, or confusing feet and miles.
"So What?" cascade:
- Level 1 (surface): The lead programme costs about $30M a year
- Level 2 (implication): After mandatory spending, the budget can fund only a 1% renewal rate, which would take about 30 years to replace just the cast-iron mains, whose average age is already 78 years
- Level 3 (actionable): The $110M envelope is not enough to both comply with the lead rule and reduce breaks. Either the budget rises (with a bill impact) or the utility must pick the highest-risk mains very carefully, which is Question 3
Question 3Judgement & Insights
The engineering team has proposed four main-replacement programmes (Exhibit 2). If CityWater can only fund some of them in the next five years, how would you rank them?
Hint · Judgement & Insights
Read the exhibit title, axes and units first. Lead with the ‘so what’, then back it with one or two numbers.
Exhibit 2Proposed Main Programmes
| Programme | Description | Miles | Current breaks per 100 miles | Cost per break (consequence) | Cost per mile | Total cost |
|---|---|---|---|---|---|---|
| A | Downtown cast-iron trunk mains (12-24 inch) | 20 | 45 | $250,000 | $4.5M | $90M |
| B | Residential cast-iron mains (6-8 inch) | 60 | 40 | $25,000 | $2.9M | $174M |
| C | Suburban asbestos-cement mains | 40 | 16 | $15,000 | $2.4M | $96M |
| D | Cathodic protection on ductile iron (corrosive soils) | 30 | 18 | $20,000 | $0.4M | $12M |
Source: CityWater engineering department. Costs in current dollars.
Source: CityWater case file
Additional InformationAsk for dataInterviewer’s data
- New PVC or ductile iron mains in CityWater's soils break at about 3 per 100 miles per year
- Programme D is not a replacement: it adds cathodic protection (corrosion control) to existing ductile iron and is expected to halve the break rate, from 18 to 9 per 100 miles
- "Cost per break" includes repair, lost water, road and traffic disruption, emergency response and business interruption. As a reference point, the USU survey implies an average direct repair cost of about $10,000 per break (about $2.6 billion of repair spend across about 260,000 breaks a year in the U.S. and Canada)
Try it first, then checkCheck my answerModel answer
Step 1: Breaks avoided per year
- A: 20 miles x (45 - 3) / 100 = 8.4
- B: 60 x (40 - 3) / 100 = 22.2
- C: 40 x (16 - 3) / 100 = 5.2
- D: 30 x (18 - 9) / 100 = 2.7
Step 2: Risk cost avoided per year and per $10M invested
| Programme | Breaks avoided per year | Risk cost avoided per year | Total cost | Breaks avoided per $10M | Risk cost avoided per $10M | Rank |
|---|---|---|---|---|---|---|
| A | 8.4 | $2.10M | $90M | 0.93 | ~$233K | 1 |
| D | 2.7 | $0.054M | $12M | 2.25 | ~$45K | 2 |
| B | 22.2 | $0.56M | $174M | 1.28 | ~$32K | 3 |
| C | 5.2 | $0.08M | $96M | 0.54 | ~$8K | 4 |
(A: 8.4 x $250K = $2.1M; B: 22.2 x $25K = $555K; C: 5.2 x $15K = $78K; D: 2.7 x $20K = $54K.)
Step 3: Recommendation
- Fund A first. It avoids about 5 times more risk per dollar than any other programme. Pair it with the city's downtown road resurfacing plan to share trench and paving costs
- Fund D in year 1. It is cheap ($12M), fast, and has the best breaks-per-dollar ratio. It is a no-regret move
- Phase B over the decade, starting with the blocks that have the most breaks and where lead service lines are also being replaced (one excavation, two jobs)
- Defer C. Asbestos-cement mains break at below-average rates here; monitor and revisit in five years
Key insight: No pipe programme "pays back" on avoided break costs alone. Even A saves only $2.1M a year on a $90M cost. The case for replacement is risk, service reliability and public safety. That is exactly why the ranking should use consequence, not just break counts.
What the interviewer is looking forInterviewer’s viewInterviewer’s view
The trap is ranking only by number of breaks avoided per dollar, which favours the residential cast iron (B) and the cheap corrosion fix (D). Once consequence is included, the downtown trunk main (A) dominates, even though it is the most expensive per mile.
"So What?" cascade:
- Level 1 (surface): Programme D avoids the most breaks per dollar; A is the most expensive per mile
- Level 2 (implication): A avoids about 7 times more risk cost per dollar than B because each downtown break is 10 times more costly
- Level 3 (actionable): Rank on risk avoided per dollar: fund A and D first, then B, and defer C. D is also a "no-regret" quick win because it costs little and can be done in year 1
Push back if the candidate ranks purely on pipe age or break counts. Ask: "Would the council accept another three-day hospital road closure to save money on a cul-de-sac?"
Question 4Numeracy
The general manager wants to fund the 1.5% catch-up plan. Finance estimates this needs about $24 million more revenue per year by year five (covering debt service and pay-as-you-go capital). What does this do to the average residential bill, and is it affordable?
Hint · Numeracy
Write the formula before you plug in numbers, keep units and zeros explicit, and sanity-check the order of magnitude at the end.
Additional InformationAsk for dataInterviewer’s data
- Residential customers pay 52% of CityWater's revenue ($78M of $150M); assume the increase is shared in the same proportion
- There are 150,000 residential accounts; the average bill is $520 per year
- Median household income in the city is about $61,000; about 25% of residential customers (about 37,500 of the 150,000 accounts) have household income under $25,000 per year
- A widely used affordability benchmark is that a household's drinking water bill should be no more than about 2.5% of household income (AWWA notes this benchmark is based on median household income)
- For reference, AWWA estimates the average U.S. household drinking water bill at about $429 per year in 2025
Try it first, then checkCheck my answerModel answer
Step 1: Residential share of the increase
- $24M x 52% = $12.48M
Step 2: Increase per account
- $12.48M / 150,000 = $83.20 per year, about $83
- New average bill: $520 + $83 = $603 (+16%)
Step 3: Affordability check
| Household income | Bill as % of income | Versus 2.5% benchmark |
|---|---|---|
| $61,000 (median) | $603 / $61,000 = 1.0% | Well below |
| $25,000 | $603 / $25,000 = 2.4% | At the limit |
| $15,000 | $603 / $15,000 = 4.0% | Well above |
Step 4: Implications
- The increase is affordable for the median household. CityWater's bill would still be above the national average of about $429, so the council will face pushback
- For roughly a quarter of households the bill reaches or exceeds the benchmark. A customer assistance programme is needed. For example, a $150 annual credit for 30,000 enrolled low-income households costs 30,000 x $150 = $4.5M, about 19% of the $24M increase
- Ways to reduce the rate impact: apply for lead-line funds through the state revolving fund (the Infrastructure Investment and Jobs Act set aside $15 billion for lead service line replacement), use low-interest state loans for mains, and coordinate with road projects to cut per-mile cost
What the interviewer is looking forInterviewer’s viewInterviewer’s view
The candidate should calculate the bill impact, then realise that "affordable on average" and "affordable for the poorest quarter" are different questions.
"So What?" cascade:
- Level 1 (surface): The average bill rises by about $83 (16%)
- Level 2 (implication): At the median it stays near 1% of income, but for households earning $25,000 or less it reaches or exceeds 2.4%, and for a household on $15,000 it is 4%
- Level 3 (actionable): The plan is fundable only if it comes with a targeted customer assistance programme, and the council vote should be framed around that. A good candidate also asks whether cheaper money (state revolving fund loans, federal lead-line funds) can reduce the $24M
Question 5Synthesis
The general manager presents to the city council next week. What is your recommendation?
Hint · Synthesis
Answer first: the recommendation, two or three reasons with numbers, then risks and next steps.
Try it first, then checkCheck my answerModel answer
Recommendation: Approve a ten-year, risk-based capital plan that treats lead-line compliance as fixed, raises main renewal from 1% to 1.5% per year, and pairs a 16% bill increase with a customer assistance programme.
1. Lock in compliance (about $30M per year)
- Finish the service-line inventory before the November 2027 deadline, prioritising the 9,000 unknown lines
- Replace about 2,400 lead lines per year to reach all 24,000 by November 2037, starting with schools, childcare centres and neighbourhoods with the highest lead readings
2. Spend main capital on risk, not age (about $70M per year)
- Year 1: cathodic protection on 30 miles of ductile iron ($12M), the cheapest risk reduction available
- Years 1-5: downtown trunk mains ($90M), which avoid about $2.1M a year of high-consequence break costs and remove the hospital-access risk
- Years 3-10: residential cast iron, sequenced block by block with lead-line work and road resurfacing
- Defer asbestos-cement replacement; monitor
3. Fund it fairly
- About $24M more revenue per year by year 5, meaning the average bill rises from $520 to about $603
- A $4.5M per year customer assistance programme for about 30,000 low-income households
- Aggressively pursue state revolving fund loans and lead-line grant money to reduce the increase
Expected outcomes by year 10: full lead compliance and a break rate moving from 17.6 toward the national average of about 11 per 100 miles. With the $150 credit, an enrolled household on $25,000 pays about $453 ($603 - $150), or 1.8% of income; the poorest households (about $15,000) would still pay about 3.0%, so the council should consider a larger, income-tiered credit for them.
Key risks and mitigants
- Lead-line unit costs come in above $12,500: bundle lines by street to cut mobilisation cost; re-tender annually
- More unknown lines turn out to be lead than 1 in 3: inventory first, and hold a contingency in years 1-3
- Political resistance to rate increases: phase the increase over five years and publish a break-rate and lead-line scorecard so residents see results
What the interviewer is looking forInterviewer’s viewInterviewer’s view
Look for an answer-first recommendation, backed by numbers from the case, with a clear sequence, and risks tied to mitigants. Excellent candidates frame it so the council can vote on it: a budget number, a bill number, and a protection for low-income households.
Data Sources
Market and regulatory facts in this case come from the public sources below. CityWater is fictional; its company-specific figures are illustrative.
- Lead and Copper Rule Improvements: rule effective December 30, 2024; systems must comply (including the inventory and replacement plan) by November 1, 2027; all lead service lines replaced within 10 years, i.e. by November 1, 2037 -> U.S. EPA, "National Primary Drinking Water Regulations for Lead and Copper: Improvements (LCRI)," Federal Register 89 FR 86418, October 30, 2024. https://www.federalregister.gov/documents/2024/10/30/2024-23549/national-primary-drinking-water-regulations-for-lead-and-copper-improvements ; overview: U.S. EPA, "Lead and Copper Rule Improvements," 2024. https://www.epa.gov/ground-water-and-drinking-water/lead-and-copper-rule-improvements
- About 4 million lead service lines in the U.S. (about 3 million reported plus about 1 million predicted among unknowns) -> U.S. EPA, "2025 Update to the 7th Drinking Water Infrastructure Needs Survey and Assessment" fact sheet, November 2025. https://www.epa.gov/system/files/documents/2025-11/fact-sheet-2025-7th-dwinsa-update.pdf
- Average break rate of about 11.1 per 100 miles per year; by material: cast iron 28.6, asbestos cement 10.3, steel 9.2, ductile iron 5.1, PVC 2.9 -> Utah State University / ASCE Civil Engineering Source, "Breaking water mains present US, Canada with $452B problem," April 2024. https://www.asce.org/publications-and-news/civil-engineering-source/civil-engineering-magazine/article/2024/04/breaking-water-mains-present-us-canada-with-$452b-problem
- About 260,000 main breaks per year in the U.S. and Canada and about $2.6 billion in annual repair costs (implying about $10,000 per break) -> Utah State University College of Engineering, "New Report Says Lack of Funding for Critical Water Mains is $452 Billion, Over 260,000 Breaks Annually," February 2024. https://engineering.usu.edu/news/main-feed/2024/new-report-says-lack-of-funding-for-critical-water-mains-is-452-billion-over-260000-breaks-annually
- Main renewal planning cost of about $500-600 per linear foot; typical renewal rate about 1% per year, 2% or more for systems that have fallen behind; average household drinking water bill about $429 in 2025; 2.5%-of-income affordability benchmark; $15 billion in IIJA funds for lead service line replacement -> American Water Works Association, "Beyond the Replacement Era: Balancing Compounding Infrastructure Needs with Household Affordability," draft report, November 12, 2025 (copy hosted by InRecap). https://www.inrecap.com/wp-content/uploads/2026/04/Beyond-the-Replacement-Era-awwa-2026-report.pdf
- Lead service line replacement cost: about $4,700 per line (EPA estimate) vs. about $12,500 per line (CDM Smith for AWWA, 2022 dollars) -> AWWA, "Beyond the Replacement Era," November 2025 (citing EPA 2019 and CDM Smith 2022), same URL as above
- U.S. drinking water infrastructure graded C- (a main break about every two minutes) -> American Society of Civil Engineers, "2025 Report Card for America's Infrastructure: Drinking Water," 2025. https://infrastructurereportcard.org/cat-item/drinking-water-infrastructure/
Free in the DrillCase case library · Editorial policy