themortgagellm^™

Response

Your client’s current LLPA position — FICO 695 / LTV 88% / Fannie conventional purchase, $400,000 loan amount on owner-occupied single-family residence:

• FICO bucket: 680-699
• LTV bucket: 85.01-90.00%
• Current LLPA: 1.500% = 150 bps = $6,000 upfront fee (Fannie May 2023 redesign grid)

Two levers worth testing — FICO improvement (via debt paydown, account aging, error disputes) and LTV reduction (via increased down payment) — with a sometimes-counter-intuitive interaction at this profile.

Scenario A: FICO improvement (LTV stays at 88%)

Each FICO bucket crossing saves ~25 bps = $1,000 on this loan. Realistic short-term boosts (paying down credit cards, removing collections, disputing errors) tend to land in the 10-40 pt range, so 720 is plausible within 30-60 days for many borrowers; 740 typically needs sustained work.

Scenario B: LTV reduction (FICO stays at 695)

Critical insight: under the May 2023 redesigned grid, dropping LTV from 88% to 80% or 75% actually raises your client’s LLPA. The 75-85% LTV cells at sub-720 FICO were specifically priced up to fund the lower-FICO subsidies. The borrower needs to drop materially below 75% (and ideally below 70%) to save money.

*Extra cash to close assumes a $454,000 home price and adjusts down payment accordingly while holding the LTV target.

To put numbers on capital efficiency: bringing $63,600 of additional cash to closing to hit 74% LTV saves $1,500 of LLPA — a 2.4% one-time return on the cash deployed. Compare to Scenario A’s FICO route, where paying down ~$5,000 of credit card debt to boost FICO 30 points saves $2,000 in LLPA — a 40% one-time return on that paydown. FICO improvement is roughly 15× more capital-efficient at this profile.

Scenario C: Both levers combined

What about DTI?

The current Fannie LLPA grid does not price DTI directly. A DTI-based LLPA was announced as part of the May 2023 redesign but was withdrawn before going live (Lender Letter LL-2023-06, May 17, 2023). So your client’s 45% DTI doesn’t move the LLPA needle on its own.

But DTI matters for two indirect reasons:

• Approval. 45% DTI is at the edge of Fannie’s standard DU approval window for non-Home Possible / non-HomeReady loans. A small DTI improvement (paying off an auto loan, for example) can move a marginal file from refer/caution to approve/eligible.
• FICO interaction. Paying down revolving balances has a double benefit: it improves DTI and typically boosts FICO 10-30 points via lower credit utilization. So the same debt paydown that’s recommended for DTI relief also delivers the FICO-side LLPA savings shown in Scenario A.

Loan-officer recommendation

1. Lead with FICO improvement. $2,000 of LLPA savings on this loan from a 30-point FICO boost is the highest-ROI play. Realistic levers: pay down credit card balances below 10% utilization (fast, 30-60 day effect), dispute any recent errors on the credit report, and avoid any new credit pulls in the 90 days before closing.
2. Treat LTV reduction as the second lever, not the first. At this client’s 695 FICO, moving from 88% to 80% LTV is actively counterproductive — cash is better deployed against FICO-improving debt paydown. LTV reduction only helps once you can hit below 75% LTV.
3. Don’t forget the time-value math. $80K of additional cash at closing is a permanent allocation to equity rather than a one-time fee saving. The LLPA savings are real but small relative to the cash deployed.
4. Cross-check with rate-spread. If this loan is on the bubble of HPML status (rate_spread within 25 bps of 1.5 pp first-lien threshold), reducing LLPA by improving FICO also pulls the note rate down, which can keep the loan out of HPML territory and avoid the related rate-spread disclosures.

Caveats

• FICO changes from debt paydown are not guaranteed. Effect varies by current utilization, account age, and credit-history depth. Tradeline-rapid-rescore can confirm before final lock.
• This analysis is the base FICO × LTV LLPA grid only. Add-ons apply if your client’s loan is a second home (+112.5 bps at this LTV), investment property (+112.5 bps), condo (+75 bps), or high-balance ($766,550+ in 2024 / 2025 standard limit) — check the relevant cards on this page for those scenarios.
• The savings figures use Fannie’s May 2023 grid (the current regime). Freddie pricing is byte-identical on the base FICO × LTV cells per FHFA alignment, so the same savings table applies to a Freddie execution.
• Underwriting eligibility is a separate gate. A 695 FICO / 88% LTV / 45% DTI conventional purchase may be a tight DU file; improving any of the three inputs also broadens the AUS approval envelope.

Reference: Fannie LLPA Matrix May 2023 redesign, base grid cells for FICO 680-699 / 700-719 / 720-739 / 740-759 / 760-779 at LTV 60-95% range. Bucket boundaries: FICO buckets divide at 700 / 720 / 740 / 760 / 780. LTV buckets divide at 60% / 70% / 75% / 80% / 85% / 90% / 95%. Each bucket crossing shifts the LLPA cell.