Which Mortgage Lenders Have the Lowest Closing Costs?

The HMDA Modified Loan Application Register records total loan costs as reported on Page 2 of the Closing Disclosure for every originated loan. This captures origination charges (the fees your lender sets), required third-party services, and services you can shop for — everything except prepaid expenses like property taxes and homeowner's insurance, which are not lender-controlled.

We filtered to first-lien purchase mortgages on owner-occupied principal residences where both cost and rate fields were reported (not exempt or missing). We use medians, not averages, to reduce the influence of outlier loans. Lenders need at least 500 purchase originations nationally to appear in national rankings — small sample sizes make medians unreliable.

It's worth noting what this data doesn't capture: lender credits that offset costs, discount points paid for a lower rate, or how loan mix (jumbo vs. conforming, urban vs. rural) might affect a lender's median. Rankings tell you where to look, not what you'll personally pay.

The spread between the lowest-cost and highest-cost lenders with 500+ originations nationally is substantial — often more than $3,000 in median total loan costs on a standard purchase mortgage. That's not a rounding error; it's a meaningful difference that compounds when you consider the transaction size.

The lenders that consistently rank lowest tend to share a few characteristics: they originate high volumes in competitive markets, they operate direct-to-consumer (fewer intermediaries taking a cut), and they have technology-efficient processes. But these are generalizations — there are expensive credit unions and cheap correspondent lenders.

Equally important: a lender's national rank may not reflect their pricing in your state. Cost rankings shift significantly at the state level. A lender that ranks in the top 10 nationally may rank 30th in California, where title insurance costs and market dynamics differ from the national average.

For most borrowers, state-level rankings are more useful than national ones. Closing costs are inherently local — they depend on your state's transfer tax structure, whether your state requires an attorney at closing, local title insurance rates, and the competitive landscape of lenders operating in your market.

Our lender rankings break out cost data by state and by loan type (conventional, FHA, VA, USDA). A lender that ranks well nationally for conventional loans may not even participate in the FHA or VA market, or may price those products very differently. If you're using a government-backed loan, check the rankings for your specific loan type.

County-level data is the most granular view available in HMDA. For counties with enough loan volume (at least 10 qualifying originations per lender), you can see which lenders were actually lending in your specific market and at what cost. This is especially useful in high-volume counties like Los Angeles, Cook County, or Harris County, where dozens of lenders compete and pricing can differ from statewide averages.

The lowest-cost lenders in the HMDA data generally achieve low costs in one of three ways: (1) they originate at high enough volume to spread overhead, (2) they price the interest rate higher and reduce upfront fees (a trade-off worth understanding), or (3) they have genuinely efficient processes with lower operating costs per loan.

The second point is critical. A lender offering zero origination fees is still getting paid — they're just collecting their margin through a higher rate over the life of the loan instead of upfront. Whether that's better or worse depends entirely on how long you keep the loan. If you plan to sell or refinance within 5 years, low upfront costs often beat a lower rate with fees. If you're staying 30 years, the opposite may be true.

HMDA data captures total loan costs and rate independently, so you can see both sides of this trade-off for each lender. A lender with low median costs but a higher median rate may not be cheaper overall than one with higher costs and a lower rate — run the math for your holding period before deciding.

The right way to use HMDA rankings is as a screening tool, not a final answer. Start by identifying 5-10 lenders that rank well in your state for your loan type. Then apply to 3-5 of them within the same week (multiple mortgage credit inquiries within a 45-day window count as a single inquiry on your credit report).

Once you have Loan Estimates, compare them using the standardized format — specifically Section A (origination charges), the total closing costs on Page 3, and the APR. The Loan Estimate is the document that matters for your specific scenario. HMDA data tells you about median borrowers; your Loan Estimate tells you about you.

If a lender's Loan Estimate is higher than their HMDA ranking would suggest, that's your negotiation opening. Their own federal data shows they can price loans more cheaply. Use it.

• Use state + loan type rankings to build your shortlist of 5-10 lenders
• Apply to 3-5 lenders in the same week to avoid multiple credit score hits
• Compare Section A (origination) and total closing costs on Page 3 of Loan Estimates
• Use our Fair Loan Check to audit any Loan Estimate for inflated fees
• Reference a lender's HMDA rank when negotiating — it's public data they can't dispute

HMDA data shows year-over-year changes in median costs by lender and geography. In 2024, the median total loan costs nationally reflected the continued high-rate environment — borrowers were less likely to pay discount points when rates were elevated, which can reduce reported total costs even as origination fees held steady or rose.

Lenders that showed the largest year-over-year cost decreases in 2024 were often those that reduced origination charges to compete for a smaller pool of purchase borrowers. Purchase volume nationally declined from its pandemic peak, and lenders competing for market share had more incentive to reduce fees.

Year-over-year comparisons should be interpreted carefully — changes in a lender's median can reflect changes in their loan mix (average loan size, geographic distribution, loan type) as much as actual pricing changes.