Chapter 13 — Key Takeaways

The core claims

1. A loan is six decisions, not one. Program, down payment, term, rate and point position, mortgage insurance structure, concessions and buydowns. The first three mostly determine whether the loan is possible; the last three mostly determine what it costs. New originators spend their energy on the first group, which has few options, and skip the second, which has many.

2. The borrower owns all six decisions. You own the analysis. That is not a formality. A borrower who was told rather than shown will re-litigate the decision with a relative three days before closing.

3. Program selection is a statement about a borrower, never about a product. Any originator with a favorite program has stopped doing the analysis.

4. The whole down-payment question is one word: feasibility or convenience? If the borrower cannot close without the lower down payment, the program that provides it is the transaction. If they can close either way, it is one number in a comparison. Harlow Street is feasibility. Linden Street is convenience.

5. Every down-payment relief is purchased with something — an upfront premium, a higher rate, mortgage insurance that never cancels, or a second lien. Find the price tag.

6. A structure the borrower cannot fund is not a structure. Delete it and say why you deleted it.

7. Break-even is arithmetic; the horizon is the input you cannot document. Ask for it three ways, and ask before you recommend a point.

8. Present the ARM's worst case before its best case, and know that under Ability-to-Repay the low introductory payment does not improve qualification — on the Linden Street illustration it made the back-end ratio worse, 43.53% against 42.66%.

9. A temporary buydown is an escrow account, not a rate. The note rate never changes and the borrower is qualified at it.

10. Show every concession three ways — temporary buydown, permanent buydown, price reduction — because the buyer is being offered the shape that suits whoever is paying for it.

11. A loan can be perfectly underwritten, fully disclosed, correctly qualified, and still be the wrong structure. Nothing catches that but you.

12. Recommend, then hand it back. Refusing to recommend is not neutrality; it is abandonment with a compliance rationale.

The key formulas

POINTS / CREDIT BREAK-EVEN
    months = cost of points (or credit received) / monthly payment saved (or added)

TEMPORARY BUYDOWN ESCROW
    sum over each subsidized period of:
       (note-rate P&I  -  subsidized P&I)  x  months in that period

TOTAL COST OF CREDIT, over a stated horizon
    upfront charges (points, financed premiums)
  + all interest paid
  + all mortgage insurance paid
    ... EXCLUDING the down payment, which is equity, not cost

QUALIFYING RATE ON AN ARM (Ability-to-Repay)
    the GREATER of the fully indexed rate or the introductory rate,
    on a fully amortizing schedule

RATE SHEET PRICE
    100.000 = par.  Above 100 = credit to borrower.  Below 100 = cost.
    One point = 1% of the LOAN AMOUNT.

The rules of thumb

• On this grid, points recover in about 60 months and credits repay in about 45. They are not symmetric, because it costs 0.500 points to buy the first eighth of rate and pays only 0.375 to sell one. A borrower short on cash with an uncertain horizon has an easy answer; a borrower with cash and an uncertain horizon should probably sit at par.
• Never quote without the four pricing facts: representative score, LTV, occupancy and property type, lock period.
• Hold every assumption identical across the columns — same price, same closing date, same taxes, same lock — or it is not a comparison.
• Quantify in four units the borrower can hold: monthly payment, cash to close, reserves left afterward, total cost over a stated horizon.
• Name the load-bearing assumption out loud and ask them to correct it.
• Put the number that hurts your recommendation on the same page, in the same size type, as the number that helps it.

The Linden Street numbers to remember

Break-even on the half point: \$1,828.75 ÷ \$30.31 = 60.3 months. 2-1 buydown escrow on this loan, had anyone funded one: \$8,375.40.

All figures constructed and illustrative. Verify current pricing, MI and MIP factors, and guideline values at their sources.

Key terms

Loan structure · program fit · permanent buydown · temporary buydown · 2-1 buydown · total cost of credit · loan comparison

Monday morning

You should be able to take any live file and, in under thirty minutes, produce a one-page comparison of every structure the borrower can actually fund — identical assumptions, four units, the horizon assumption named — make a recommendation with a reason and a reversal condition, and hand the decision to the borrower.

If you can do that, most of §13.9's compliance obligations are satisfied incidentally, and you will close files against competitors quoting a lower rate. If you cannot, you are quoting rates, and somebody's website will always say a smaller number than yours.