Case Study 2-2 — Larkspur Flats: Reading the Owner's Pro Forma, and Why Six Months Costs More Than You Think
Larkspur Flats, Ridgeline Development Partners, and every figure below are a Tier-3 composite. Rents, cap rates, and interest rates vary enormously by market and change constantly; treat these as illustrative arithmetic, not market data. The structure of the arithmetic is what transfers.
Setup
Case Study 2-1 looked at a contractor's balance sheet. This one looks across the table.
Avis Thorne is a development partner at Ridgeline Development Partners. She has a 3.4-acre site three miles from the Rivermont city line, entitled for 96 market-rate apartment units in a four-story wood-frame building over a partial parking level. Ridgeline is putting up part of the equity; the rest comes from Fennimore Capital, whose asset manager, Kirby Nunes, sits on the investment committee that says yes or no.
The general contractor bidding it is Brannock Builders, and their chief estimator is Dez Whittaker.
Dez has been building for twenty-two years and has never seen a developer's pro forma. He is about to find out that the construction budget he has been asked to hit was not derived from what the building costs to build.
What Happens
Part 1 — The pro forma, built backwards
Start where Avis starts: with what the finished building will earn.
Income
| Line | Calculation | Amount |
|---|---|---|
| Units | 96 units × 840 net SF avg | 80,640 net SF |
| Average rent | $2.05 per net SF per month | $1,722 per unit/month | |
| Gross potential rent | 96 × $1,722 × 12 | $1,984,000 | |
| Other income (parking, pet, fees) | $92,000 | |
| Gross income | $2,076,000 | |
| Vacancy and credit loss @ 6% | ($124,560) | |
| Effective gross income | $1,951,440 | |
| Operating expenses | $7,900 per unit × 96 | ($758,400) | |
| NET OPERATING INCOME (NOI) | $1,193,000 |
Value
The building is worth its NOI divided by the capitalization rate — the yield a buyer of stabilized apartments in this market will accept.
Value = NOI ÷ cap rate = $1,193,000 ÷ 0.0525 = $22,724,000
Cost — and here is the move that surprises contractors
Ridgeline does not build to cost. They build to a spread: a yield on cost meaningfully above the exit cap rate. Fennimore's investment committee requires at least 100 basis points. So the target yield on cost is 6.25 percent, and the maximum total development cost the deal can bear is:
Maximum total cost = NOI ÷ target yield = $1,193,000 ÷ 0.0625 = $19,088,000
Now subtract everything that is not construction.
| Line | Amount |
|---|---|
| Land | $2,400,000 |
| Soft costs (design, permits, legal, insurance, marketing, taxes during construction) | $2,050,000 |
| Developer fee | $620,000 |
| Financing and interest reserve | $1,180,000 |
| Owner contingency | $560,000 |
| Subtotal, everything but construction | $6,810,000 |
| HARD COST — the construction contract | $12,278,000 |
$12,278,000 ÷ 104,000 gross SF = $118.06 per gross SF
$12,278,000 ÷ 96 units = $127,896 per unit
Read that table one more time. The construction budget was not built up from quantities, crews, and unit costs. It is a remainder. It is what is left when you take what the finished asset is worth, apply a required return, and subtract land, soft costs, fee, financing, and contingency. The rents set the value; the value sets the budget; the budget sets Dez's number.
Part 2 — The bid comes in
Brannock's number: $13,400,000. Over by $1,122,000 — 9.1 percent.
Dez Whittaker: "It's a good number, Avis. I've got three subs in every trade and I'm not carrying anything I don't need. If you need to get to twelve-three, tell me what to take out."
Avis Thorne: "I don't need you to take anything out. I need you to be wrong."
Dez: "I'm not wrong. Can't your equity partner just put in another million?"
Avis: "They can. And then the deal returns eleven percent instead of nineteen, and Kirby's committee funds somebody else's deal instead of mine. The million isn't the problem. The return on the million is the problem."
That exchange is the whole case. Contractors routinely assume an over-budget number is a negotiating position or a funding inconvenience. It is neither. The value of the finished asset caps the budget, and no amount of goodwill moves that ceiling.
Part 3 — Value engineering, and the trap inside it
Avis, Dez, and the architect put five options on the table.
| Option | Capital saved | Effect on income |
|---|---|---|
| A — Delete the below-grade parking level; surface and tuck-under instead | $610,000 | Loses 14 spaces; reduces achievable rent by about $0.04/SF/month | |
| B — Fiber-cement lap siding in lieu of brick veneer on three elevations | $268,000 | Appraisal comparables risk; no direct rent effect assumed |
| C — Standard-efficiency rooftop units in lieu of high-efficiency | $141,000 | Raises operating expense about $18 per unit per year | |
| D — Vinyl plank in lieu of engineered wood flooring in units | $96,000 | Modest rent-premium risk |
| E — Reduce landscaping and amenity deck scope | $173,000 | Lease-up velocity risk |
| Total identified | $1,288,000 |
The identified savings exceed the $1,122,000 gap. Problem solved — except that three of the five options reduce income, and income is what created the budget in the first place. A value-engineering item that reduces NOI is not saving money. It is shrinking the deal. You have to run each one both ways.
Option C, worked. Capital saved: $141,000. Operating expense increase: `$18 × 96 units = $1,728 per year`, which reduces NOI by $1,728. Value lost: $1,728 ÷ 0.0525 = $32,914.
Net effect: +$141,000 − $32,914 = +$108,086. Take it. The capital saving dwarfs the value impact.
Option A, worked. Capital saved: $610,000. Rent lost: `80,640 net SF × $0.04 × 12 = $38,707` of gross potential rent per year. After 6 percent vacancy, NOI falls by about `$36,385. Value lost:$36,385 ÷ 0.0525 = $693,048`.
Net effect: +$610,000 − $693,048 = −$83,048. Reject it. The biggest single "saving" on the list destroys more value than it creates.
This is the most useful thing a construction manager can learn about owners: the cheapest option and the best option are frequently not the same option, and the owner is doing arithmetic you cannot see. When Avis rejects the largest VE item on your list and accepts three smaller ones, she is not being sentimental about parking. She is running the numbers above. Chapter 11 builds this into a formal value-engineering log with priced options — and this is why that log has a column for the income effect, not just the cost effect.
Ridgeline takes B, C, D, and E — $268,000 + $141,000 + $96,000 + $173,000 = $678,000 — reduces the unit count by two to shrink the building slightly, and Fennimore contributes the balance. The contract is signed. Construction starts.
Part 4 — The six-month delay
A dry utility relocation is discovered in the wrong place, the permit revision takes eleven weeks, and the long-span steel joists over the amenity space were released late. Substantial completion slips six months.
Here is what that costs Ridgeline. Note that only the first two lines ever appear on a construction cost report.
| Impact | Calculation | Amount |
|---|---|---|
| Additional construction loan interest | ~$11,000,000 avg. balance × 8.0% × 0.5 yr | $440,000 | |
| Additional carry — taxes, insurance, admin | 6 months | $96,000 |
| Six months of NOI never earned | $1,193,000 × 0.5 | $597,000 |
| Quantifiable delay cost | $1,133,000 | |
| Plus: market risk on the exit | Cap rate moves 25 bp, 5.25% → 5.50% | (see below) |
The third line is the one contractors miss. A six-month delay does not push six months of income to the right. It deletes six months of income from the life of the deal, permanently. The building will still operate for decades; it will simply have operated for six fewer months under Ridgeline's ownership.
And then the exit. Ridgeline planned to sell at stabilization. Six months later the market has moved 25 basis points against them:
Value at 5.50% cap = $1,193,000 ÷ 0.055 = $21,691,000 — down $1,033,000 from $22,724,000. Nobody caused that. It is simply what happens when you are in the market six months longer than you planned.
The return, both ways:
| Base case | With six-month delay | |
|---|---|---|
| Total development cost | $19,088,000 | $19,624,000 | |
| Construction loan (62% of cost) | $11,835,000 | $11,835,000 | |
| Equity | $7,253,000 | $7,253,000 | |
| Stabilized value at exit | $22,724,000 | $21,691,000 | |
| Profit on sale | $3,636,000 | $2,067,000 | |
| Return on equity | 50.1% | 28.5% |
| Hold period | 30 months | 36 months |
| Simple annualized return | ~20.0% / yr | ~9.5% / yr |
Six months cut the developer's annualized return roughly in half — and that is before counting the $597,000 of NOI the deal will never earn.
Part 5 — The liquidated damages don't come close
Brannock's contract carries liquidated damages of $2,400 per calendar day.
180 CD × $2,400/CD = $432,000
$432,000 ÷ $1,133,000 = 38.1%
Liquidated damages recover about 38 percent of the owner's quantifiable delay cost, and nothing at all of the cap-rate movement.
That number explains a great deal of owner behavior that contractors read as unreasonable. LDs are usually the owner's exclusive remedy for delay — which means they also cap the contractor's exposure, protecting Brannock from the full $1,133,000. The trade is deliberate and it is fair. But it means the owner knows, from day one, that if you are late they will eat most of it. So they push. They call three times a week. They resist every time extension. They want the schedule updated monthly and they read it.
Analysis
Why the owner behaves the way they do. Every apparently irrational owner behavior in this case has a line in the pro forma behind it.
| Owner behavior | The line in the pro forma |
|---|---|
| "The budget is $12,278,000 and it cannot move." | The budget is a remainder: value minus land, softs, fee, financing, contingency |
| Rejects the biggest VE saving | It destroyed $693,048 of value to save $610,000 of cost |
| Will not accept "just add equity" | Added equity dilutes return on equity, and the committee funds returns, not buildings |
| Obsesses over the completion date | Six months costs $1,133,000 of hard money plus exposure to a moving market |
| Fights every time extension | Every granted day is an owner-owned day, and LDs recover only 38 percent |
| Rigid about lien waivers and draw paperwork | The construction lender requires it before funding; the owner is relaying, not inventing |
What the contractor should do with this. Three concrete things.
- Ask for the schedule of the owner's obligations, in dollars. "What is your daily cost of being late?" is a legitimate question in a preconstruction meeting, and the answer tells you exactly how much acceleration the owner will fund and how hard they will fight a time extension. On Northgate, Kestrel knows the answer is $10,650 a day. On Larkspur, it is roughly $6,300 a day of hard carry — and much more if the market moves.
- Price the income effect of every value-engineering option, not just the cost. Even a rough estimate. A contractor who hands Avis a VE log with an income column is worth double a contractor who hands her a cheaper list.
- Treat long-lead procurement as a financial control, not a paperwork task. The steel joists that arrived late cost Ridgeline more than the joists cost. Chapter 16 and Chapter 17 are, from the owner's chair, financial chapters.
Discussion Questions
- Dez Whittaker's $13,400,000 was a fair, well-covered number. Avis's ceiling was $12,278,000. Neither of them was wrong. Whose problem is the gap, and what should have happened before the bid to prevent it?
- Option A saved $610,000 of construction cost and destroyed $693,048 of asset value. Would a general contractor, given only the drawings and the specifications, ever be able to see that? What information would you have to ask for, and would a developer give it to you?
- Liquidated damages of $2,400 per day recover 38 percent of the owner's quantifiable delay cost. Argue both sides: why should LDs be the exclusive remedy, and why might an owner want them not to be?
- If the cap rate had moved 25 basis points in Ridgeline's favor, the delay would have been nearly costless. Does that change your view of how much of an owner's schedule anxiety is genuinely about your performance?
- Suppose Ridgeline had used a CM-at-Risk delivery with Brannock engaged during design, as Meridian did on Northgate. Which of the failures in this case would that have prevented, and which would it not?
Your Turn
Rebuild the Larkspur pro forma with one change: average rent falls from $2.05 to $1.92 per net SF per month — a 6.3 percent drop, entirely plausible if two competing projects deliver into the same submarket during your construction period.
Work it through in order: new gross potential rent, new effective gross income, new NOI, new value at a 5.25 percent cap, new maximum total development cost at a 6.25 percent yield on cost, and the new hard-cost ceiling once you subtract the same $6,810,000 of land, softs, fee, financing, and contingency.
Then answer the only question that matters: is Larkspur Flats still a project? And write one sentence explaining, to a contractor who does not read pro formas, why a thirteen-cent change in monthly rent per square foot can cancel a $19,000,000 development.