← All fair lending scans

Scan 04 · Pricing

Run your own pricing disparity scan

A pricing scan looks at the loans that closed and asks whether the terms differed. Higher-priced loan incidence, rate spread, points and fees by borrower group and tract type, for your institution against peers and the market, built from loan-level HMDA.

Free tier is three answers a day, fifteen a month, no card.

Start here

What a higher-priced loan is

A first-lien mortgage whose annual percentage rate exceeds the average prime offer rate for a comparable loan by 1.5 percentage points or more, or 3.5 points for a subordinate lien.

Regulation Z, 12 CFR 1026.35; reported in HMDA as rate spread

Pricing is the fair lending risk that closes rather than denies. The borrower gets the loan, the file shows an approval, and the disparity sits in the rate spread, the points and the fees. HMDA carries all three at the loan level, so the scan runs on every lender in the market.

The red flags a scan is looking for

01

An HPL rate above the market

For the same loan purpose and period, a higher share of your originations carrying the spread than the market's.

02

Spread concentrated by tract

Higher-priced loans landing disproportionately in majority-minority tracts, which is where reverse redlining appears.

03

Points and fees that differ by group

Similar loan amounts and purposes, different discount points or origination charges by borrower group.

04

One purpose carrying the spread

Cash-out refinance or home improvement running several times the purchase HPL rate, then splitting by group.

05

Channel doing the pricing

Broker or correspondent loans priced differently from retail for the same group.

06

A pricing index above peers

The pricing disparity index for majority-minority tracts higher than lenders of similar size show.

The scan itself

HPL originations and odds, market first, then yours

The HPL analysis reports originations, higher-priced originations, the HPL rate and the odds of HPL by loan purpose. Odds of HPL is the ratio of the HPL rate to the non-HPL rate. Read the market line first, then select your institution and compare row by row.

HPL analysis, market, 2025
Loan purposeOriginationsHPL originationsOdds of HPL
Purchase8,5701672.4%
Home improvement3,479185.2%
Cash-out refinance1,96716312.8%
Refinance1,828111.9%
Other purpose1,590196.6%
All purposes17,4363784.0%
Where it feeds the summary
IndexYour institutionPeers
Pricing disparity, majority-minority tracts—0.6
Pricing disparity, majority-Black tracts—0.5

One metro, all lenders, 2025 HMDA. The pricing disparity index is the share of HPL originations in the protected group over the share in the control group. A dash means no higher-priced originations in that cell, which is itself a finding.

What the scan says

Cash-out refinance carries a higher-priced rate more than five times the purchase rate in this market, and 43 percent of all HPL originations. A lender whose cash-out book concentrates in majority-minority tracts has a pricing question before anyone asks it.

Read with care

Rate spread reflects the loan's APR against APOR on the lock date. Loan size, lien and term move it before any borrower characteristic does. Compare like purpose with like purpose.

Then check

Discount points and origination charges for the same group across peers. That separates a product-mix effect from a pricing decision.

The workflow

Six steps, fully reproducible

Each step is a set of filters in HMDAVision. Keep them and the scan reruns on the next release in minutes.

1

Define the market, period and loan population

Originations only, first liens and subordinate liens read separately, purchased loans excluded.

Output: a defined population with a documented rationale.

2

Read the market HPL line by purpose

This is the baseline. Every purpose has its own normal, and cash-out refinance is rarely close to purchase.

Output: the market rate your book will be read against.

3

Select your institution and compare

Same table, your originations. Any purpose where your odds of HPL sit above the market is the next filter.

Output: the purposes where your pricing is above the line.

4

Split by borrower group and tract type

Minority status, Minority Detail, majority-minority tract, LMI tract. The spread should not follow any of them.

Output: whether the spread follows a protected characteristic.

5

Open rate spread, points and fees

Averages and medians by group for the segment in question, against peers of similar volume.

Output: which pricing component carries the difference.

6

Read the index and record

The pricing disparity index by tract type, yours beside peers, in the Fair Lending Summary. Keep the filters.

Output: the board figure, and an audit trail.

From finding to action plan

The review starts with the file already filtered

A pricing finding arrives with the purpose, the group, the tract type and the pricing component attached. That is the scope of the review, and it is narrower than a whole book.

Price review

Pull the segment where your odds of HPL sit above the market and the spread follows a group. Pricing exceptions, lock policy and lender credits are the first places to look.

Product mix

When one purpose carries the spread, ask whether the product is the right one for the borrowers receiving it. A cash-out book in majority-minority tracts may be a home equity or renovation product question.

Channel review

Where broker or correspondent loans price differently from retail for the same group, the compensation and pricing rules for that channel are the finding.

What to measure next quarter

Rerun the same filters. An HPL rate that moves toward the market line in the segment you acted on is evidence the change worked.

Start scanning

See your pricing against the market before the market sees it for you.

Ask one question at mortgagedata.ai, or open the HPL analysis on your own institution with a guided seven-day trial of HMDAVision.

Polygon Research is a research and analytics firm, not a law firm. Nothing here is legal advice, and a fair lending program should be built with counsel.