Case Study 35.2 — When the After-Improved Value Does Not Materialize

A construction file that failed at completion, and a renovation file that worked — and the five differences between them.

Both files below are labeled composites. They are constructed from documented industry patterns — cost escalation during long builds, thin comparable sets on custom construction, change- order drift, and the re-qualification exposure in a two-close structure — and neither reports an individual borrower's records. Every figure is illustrative. This is Tier 3 material: the arithmetic is exact and the situation is representative, but no specific transaction is being described.

Part 1 — The build that came in short

The file at origination

A married couple owns a lot outright. They bought it eighteen months ago for cash and have spent that time with an architect. They have a fixed-price contract with a licensed builder, a complete specification list, and a line-item budget.

THE PROJECT AT MONTH 0                              [composite teaching file]

  Lot, owned free and clear                                 $95,000
  Construction budget (fixed-price contract)               $412,000
                                                          ─────────
  TOTAL PROJECT COST                                       $507,000

  Appraised value, SUBJECT TO COMPLETION per plans
    and specifications                                     $525,000

  Loan requested (construction only; the lot is
    the borrower's equity)                                 $412,000

  LTV against total project cost   $412,000 / $507,000       81.26%
  LTV against after-improved value $412,000 / $525,000       78.48%

This is a good file. The borrowers have equity in the ground — \$95,000 of lot value against a required investment of \$50,700 at a 90% cap, so the lot covers the down payment with room to spare. Both loan-to-value tests clear comfortably. Income and credit are strong. The builder is licensed, insured, and has four comparable completed projects with references that check out.

They chose a two-close structure. A local bank offered the interim construction loan at a floating rate that priced better than the single-close alternative, and the borrowers' plan was to refinance into a permanent loan at completion. Rates at the time supported a permanent loan modeled at an illustrative 6.500%.

  WHAT THEY MODELED
  Permanent loan $412,000 at 6.500%, 30-year fixed
    P&I                                                   $2,604.12
    taxes (illustrative)                                    $520.00
    insurance (illustrative)                                $185.00
                                                          ─────────
    PITI                                                  $3,309.12
  Household income $11,800/month; other debts $980/month
    back-end ratio  ($3,309.12 + $980) / $11,800             36.35%

Comfortable. Everyone signed.

What happened over twelve months

Four change orders, and a market.

  BUDGET DRIFT                                       [composite teaching file]

  Original fixed-price contract                            $412,000
    CO-1  upgraded window package (borrower-requested)      +$11,400
    CO-2  unforeseen rock excavation at the foundation      +$18,600
    CO-3  material price escalation, framing and
            mechanical, across the build                    +$14,200
    CO-4  kitchen upgrade (borrower-requested)               +$6,800
                                                          ──────────
  FINAL CONSTRUCTION COST                                  $463,000
                                                            +$51,000   (+12.38%)

Then the completion appraisal.

  Appraised value at origination, subject to completion    $525,000
  Appraised value at completion, as built                  $498,000
                                                          ──────────
                                                            -$27,000   (-5.14%)

Read those two boxes together, because that is the case: **they spent \$51,000 more than budgeted and the house appraised \$27,000 less than promised.** A \$78,000 swing against them, on a project that started with a 78.48% loan-to-value cushion.

Why the value fell is not mysterious, and all three reasons were visible in advance to anyone who looked:

1. The comparable set was thin. Custom construction in a market without much of it gives an appraiser few recent sales of genuinely similar finished homes. Reasonable appraisers disagree by more on these assignments, and the origination opinion sat toward the upper end of a wide range.
2. The market moved. Twelve months elapsed. The sales that supported \$525,000 were stale, and the market in this price band softened during the build.
3. Most of the overrun bought no value at all. This is the piece borrowers find hardest to accept. The \$18,600 of rock excavation created zero value — nobody pays more for a house because the hole was difficult to dig. The \$14,200 of material escalation created zero value for the same reason. Only the window package and the kitchen were even candidates for partial recovery. Roughly two-thirds of the overrun was, in valuation terms, money that vanished.

The problem at the closing table that never happened

  THE PERMANENT LOAN, MONTH 12                       [composite teaching file]

  Payoff required (final construction cost)                $463,000
  Completion appraised value                               $498,000
  LTV   $463,000 / $498,000                                  92.97%

  Program maximum, 90% LTV      $498,000 x 0.90            $448,200
                                                          ──────────
  SHORTFALL — cash required at closing                      $14,800
    (plus closing costs, plus prepaids)

  And rates moved during the build:  6.500% -> 7.250%

Now price the three ways out.

The modeled payment was \$3,309.12**. The achievable payment is **\$3,762.51 if they can find \$14,800 in cash, or **\$4,341.92 if they cannot — \$453.39** or **\$793.58 a month above the number they built their lives around, an increase of 17.41% or 30.47%**.

And the interim loan is maturing. It is a balloon. This is the situation Chapter 35 §35.2 named: the take-out that wasn't there.

How it resolved, and what each option cost

The borrowers found \$14,800 by liquidating a brokerage account and taking a modest tax consequence, and closed the permanent loan at 90% LTV. That is the good outcome. It is worth listing what the other doors led to:

• Request a reconsideration of value. Legitimate, and the loan officer should have tried it — Chapter 18 covers the process, and the Cypress Court file is its worked example. Here it would have needed genuinely comparable finished sales that the original appraisal had missed, and on a thin custom market those were not available. A reconsideration is a request for a second look at evidence, not an appeal to fairness.
• Take the 95% program. Available, and it costs \$239.22 a month of mortgage insurance and pushes the back-end to 45.10%. It would have closed. It would also have put the household at a ratio their own budget did not support.
• Extend the interim loan. The bank was willing, at a repricing and an extension fee, which buys time and solves nothing.
• Sell the house they had just built. Into the same softened market that produced the \$498,000 appraisal.

The five failures, in order of when they could have been caught

1. The two-close structure put a twelve-month rate-and-qualification risk on the borrower to save a visible up-front cost. Nobody quantified the risk being purchased. §35.2.
2. The change orders were not controlled. Two of the four were borrower-requested upgrades approved by the builder in the field. Nobody re-ran the value or the budget until completion. §35.3.
3. There was no contingency in the budget. A fixed-price contract is not a fixed price when the ground is unknown. The \$18,600 of rock had no reserve behind it.
4. Nobody re-underwrote the permanent loan mid-build. A month-six review at a stressed rate assumption and the actual running budget would have surfaced the entire problem while there was still time to reduce scope.
5. The after-improved value was treated as a fact rather than an opinion with a range. It was the upper end of a thin comparable set on a twelve-month horizon, which is exactly the situation in which it should have been stress-tested.

Part 2 — The renovation that worked

The contrast is instructive precisely because the underwriting concept is identical.

A HOMESTYLE RENOVATION PURCHASE                     [composite teaching file]

  Purchase price — a structurally sound 1970s house with
    an original kitchen, two dated baths, a failing HVAC
    system, and a 100-amp electrical panel                  $248,000

  RENOVATION SCOPE (no structural work, no additions)
    kitchen, two baths, HVAC replacement, 200-amp
    panel and rewiring of two circuits, flooring             $62,000
    contingency reserve @ 10%                                 $6,200
    fees: renovation specialist, permits, inspections,
      title update endorsements                               $2,400
                                                            ────────
    TOTAL RENOVATION ESCROW                                  $70,600

  TOTAL BASIS   $248,000 + $70,600                          $318,600
  AFTER-IMPROVED (as-completed) VALUE                       $352,000

  Loan at 95% of the LESSER of basis or as-completed value
    lesser = $318,600;  x 0.95 =                            $302,670
  Down payment   $318,600 - $302,670                         $15,930
  P&I at an illustrative 6.750%, 30-year fixed              $1,963.11

  AT COMPLETION: appraisal confirms $352,000.
    value above total basis   $352,000 - $318,600            $33,400

Note what this file did not claim. It did not claim that \$62,000 of renovation created \$62,000 of value — it created roughly \$33,400 above the all-in basis, a recovery of about 54%, which is an entirely ordinary result. The value was not created by the renovation's return on investment. It was created by buying a house at a price that reflected its condition and financing the repair of that condition in the same loan. That is the renovation loan's actual economic function, and stating it correctly is how you keep a borrower's expectations honest.

The five differences

The fifth difference is the one worth carrying: the renovation file's value opinion was easier to form and harder to be wrong about. An appraiser valuing a normal house in normal condition after normal work, ninety days out, in a neighborhood full of finished comparables, is doing a routine assignment. An appraiser valuing a custom home twelve months out in a market with few comparable sales is doing something much closer to a forecast.

A loan officer cannot change which assignment the appraiser has. They can absolutely change whether the file is structured as though the answer were certain.

Discussion questions

1. The construction borrowers saved a visible cost by choosing two-close and bought an unquantified risk. Design the one-page document you would hand a construction borrower at application that makes that trade explicit. What does it contain, and what does it deliberately leave out?

2. Roughly two-thirds of the \$51,000 overrun — rock excavation and material escalation — created no value at all. Explain to a borrower, in plain language, why money spent on a house does not automatically become value in a house. Then explain why the lender's value test is protecting them and not only the lender.

3. At month six, what specific review would have caught this problem while it was still fixable? Who would have had to run it, what would it have measured, and what would the borrower have been asked to decide?

4. Compare the reconsideration-of-value opportunity here with the Cypress Court file in Chapter 18. Name the most important way the two situations differ, and say what that difference implies about when an ROV is worth requesting.

5. The 95% option would have closed the loan at a 45.10% back-end ratio without requiring \$14,800 of cash. It was available and the borrowers did not take it. Construct the argument that they should have, then the argument that the loan officer was right to present it as the second choice.

6. The renovation file froze the scope at the work write-up and deferred borrower-requested upgrades to after closing, out of pocket. Draft the two sentences you would use at application to establish that rule with a borrower who is excited about the project — without sounding like you are saying no.