Hard Money Loan Rates 2026: What 68,319 Investor Loans Actually Paid (Federal HMDA Data)

Hard money loan rates are a range, so here is the whole range

Search for hard money loan rates and you get lender pages with “starting at” numbers and broad ranges. Neither tells you what borrowers signed. The federal Home Mortgage Disclosure Act (HMDA) does: lenders above a size threshold report every mortgage they originate, with its note rate, amount, term, loan-to-value ratio, purpose, state and lender, and the FFIEC publishes the file loan by loan. We downloaded the nationwide 2024 and 2025 files on October 5, 2026, pulled out the short-term investor loans and computed everything below from them (our arithmetic; the scripts and the extracted loan rows are in the data folder).

US short-term (36 months or less) investor loans in HMDA: note-rate distribution

Share of loans by note-rate band

The record is tighter than a broad range suggests. In 2025, 93.5% of the reported loans carried a note rate under 12% and only 29 loans out of 67,636 with a rate were at 15% or more (our arithmetic). The middle of the market sat between 9% and 11%, and the premium over a 30-year investor rental loan narrowed from 3.125 points in 2024 to 2.615 points in 2025. Lending grew while rates fell: 15.2% more loans and 30.4% more dollars.

How the hard money loans were identified

HMDA has no “hard money”, “bridge” or “fix-and-flip” field, so we used fields that exist. A loan is in our group if it was originated, conventional, first lien, closed-end and not a reverse mortgage; secured by an investment property of 1 to 4 units; made primarily for a business or commercial purpose; reported with no debt-to-income ratio (the lender did not underwrite the borrower's income); and has a term of 36 months or less. The 30-year loans that pass the same filters are the DSCR-type rental loans we use as the comparison line.

What that proxy cannot see:

• Ground-up construction for sale is out. Regulation C excludes temporary financing, and the official interpretation says a construction-only loan is excluded when it is “extended to a person exclusively to construct a dwelling for sale.” New-build spec loans therefore barely appear here. Bridge loans planned from day one to be refinanced by the same borrower can fall under the same exclusion.
• The flip loan is in. The interpretation's own example: “Lender A originates a loan with a nine-month term to enable an investor to purchase a home, renovate it, and re-sell it before the term expires.” It concludes that such a loan is not temporary financing, so purchase-and-rehab loans are reportable.
• Only purchases, improvements and refinances. A business-purpose loan is excluded unless it is a home improvement loan, a home purchase loan or a refinancing (12 CFR 1003.3(c)(10)). Most hard money loans are purchases, so this mostly matters for loans taken for other reasons.
• Small lenders are missing. A lender that originated fewer than 25 closed-end mortgage loans in either of the two preceding years does not report (12 CFR 1003.3(c)(11)). Local private lenders and small funds are invisible.
• No points, no fees. Origination charges and discount points are reported only for loans subject to the TILA-RESPA closing disclosure (12 CFR 1003.4(a)(18)-(19)). In this group, 0 loans in either year report origination charges as a dollar amount; in 2025, 67,090 report NA and 1,229 report Exempt. Discount points are never a number, and the prepayment-penalty term is NA on 100% of the loans.
• One 2025 lender's terms look mis-keyed. EMPORIUM TPO LLC reported 2,290 loans, every one with a 30-month term, at a 7.25% median, which looks like 30-year loans entered as 30. Dropping all 30-month loans moves the 2025 median from 9.99% to 10% and the non-bank median from 10.25% to 10.45%. We keep them in the headline figures and show both.

Non-bank hard money lenders vs banks: the 2.75-point gap

The national median mixes two different products. Most of the loans come from non-bank lenders, the companies investors mean when they say hard money. A smaller share comes from banks and credit unions, mostly in the Midwest and South, that keep short-term investor loans on their own books. We classified each lender with at least 10 loans in a year using its federal agency code in the FFIEC institution file: the 2025 filing guide reads “If Federal Agency equals 7, indicating a non-depository institution”. Lenders under the CFPB code were split by name (with or without “bank”).

US short-term investor loans by lender type

The gap between the two medians was 2.75 points in 2025 (our arithmetic) and 2.49 points in 2024. Bank loans look different, too: 29.3% ran 36 months, against 0.1% of non-bank loans, and banks made 89.4% of all 36-month loans in the group. If a local bank lends on investment property where you buy, its short-term loan is the cheapest money in this record; expect it to ask more about you and your finances than a hard money lender does. Excluding the 30-month EMPORIUM loans, the non-bank middle half in 2025 was 9.75% to 11% and the median 10.45%.

Kiavi, Lima One, Anchor, RCN and 16 more: rates by lender

The 20 largest short-term investor lenders in 2025 (names as filed with HMDA)

Kiavi alone made 28.5% of the national loans, and the top 3 lenders made 41%; the top 20 made 76.3%. Three patterns stand out in the lender table:

• Big loans, lower rates. Leaving aside the EMPORIUM entries, the lowest medians belong to lenders whose median loan is large: Genesis Capital (8.75% on a $935,000 median loan), Conventus Lending (9.5%, $755,000) and Anchor Loans (9.5%, $475,000). Their median loans are 1.8 to 3.5 times the national median of $265,000 (our arithmetic).
• Fixed-price lenders. Easy Street Capital's middle half is a single rate, 9.9%, on mostly 6-month loans; 212 Loans priced 11.99% in both years. Dominion Financial's middle half is 10.88% to 11%. With these lenders the negotiation is over points, leverage and term, not the rate.
• Broker-heavy lenders sit higher. Loan Funder (70.1% through brokers), CV3 Financial Services (67.6%) and Temple View Capital (61.8%) had medians of 10.5% to 10.99%. Across all non-bank loans, broker loans had a 10.75% median and direct loans 10.24% (our arithmetic below).

RCN Capital files under two legal entity identifiers; the second one reported 693 loans in 2025 at a 10.89% median (11.24% in 2024). Investor Mortgage Finance LLC, one of the two entities Visio Lending names as originating its loans, reported 0 short-term loans in either year; Visio describes itself as a lender offering “long-term loans for SFR rental properties.” Velocity Commercial Capital reported 433 loans at a 10.99% median, the same as in 2024. Kiavi's history and securitizations are in our Kiavi review, Lima One's in our Lima One Capital review, and the lender-by-lender comparison from filings and rating-agency data in the best fix-and-flip lenders of 2026.

2025 vs 2024: the same lenders charged less

A falling national median can come from a shift in who lends, for example more bank loans. That is not what happened. We took every lender with at least 100 short-term investor loans in both years and compared each one's 2025 median with its own 2024 median.

Same-lender change in median note rate, 2024 to 2025 (lenders with 100+ loans in both years)

The typical lender cut its median by three-quarters of a point, about the same as the 0.76-point drop in the national median. 40.5% of the 2025 loans were sold to a private securitizer within the year, so these rates tend to move with what investors in those securities demand, not with a bank's deposit costs. Use the 2025 figures as a range for current quotes, not as a forecast.

Hard money loan rates by state

The 15 largest states for short-term investor loans in 2025

Read the state table in two columns. The all-lender median swings from 7.75% in Arkansas to 10.64% in Pennsylvania, but most of that swing is lender mix: banks made 89.4% of the Arkansas loans and 58.7% of the Missouri loans. The non-bank median, which is what an investor working with a national hard money lender would compare, stays between 9.75% (California) and 10.99% (Arkansas) in these 15 states. Among the 31 states with at least 500 loans, the lowest all-lender median was Wisconsin (7.5% on 969 loans, 47.6% from banks) and the highest Michigan (10.75%, 1.7% from banks). The top 5 states made 38.7% of the national loans. Florida's county-level detail, its usury limits and its closing taxes are on our Florida hard money page.

California's low non-bank median goes with its loan size ($735,000 median), which is where lenders such as Genesis Capital and Conventus compete. The same lender also prices differently by state, as Kiavi's own 2025 loans show:

Kiavi Funding's 2025 short-term loans in its six largest states

A 1.2-point spread between Kiavi's California and Pennsylvania medians (our arithmetic) is larger than the gap between many lenders. Part of it is loan size and borrower experience, which HMDA does not show, but it means a national lender's advertised range tells you little about the rate you will get in your market.

Does term, LTV or loan size change the rate?

Non-bank short-term investor loans in 2025: median note rate by segment

All short-term investor loans in 2025 by term

Leverage barely moves the rate. Non-bank loans at 60% LTV or less and loans at 70% to 80% had the same median, about 10.25%, and the median combined LTV across all lenders was 70% (middle half 64.52% to 74.99%). Many lenders report the loan against purchase price plus rehab or after-repair value, so treat LTV as reported, not as your down payment. Size matters more: non-bank loans of $1,000,000 or more had a 9.95% median against 10.5% under $150,000. Home improvement loans were the most expensive purpose at 11.25%. Up to 24 months, term hardly changes the median; the low medians at 25 months and longer come from bank loans and the 30-month EMPORIUM entries.

What the rate costs, and the half HMDA leaves out

The median 2025 loan was $265,000 and the non-bank median rate 10.25%. On an interest-only loan, the monthly payment is the loan times the rate divided by 12: $2,263.54 a month, $27,162.50 over 12 months (our arithmetic). The gap between the non-bank 25th and 75th percentile rates (9.5% and 11%) is $3,975 a year on the same loan.

Points are the other half of a hard money quote, and HMDA does not report them for these loans. Each point on the median loan is $2,650. A simple way to compare quotes is to divide the points by the term in years and add the result to the note rate:

Non-bank median rate plus points, simple annualized cost (our arithmetic, ignores other fees and early payoff)

Two points on a 6-month loan cost as much per year as four rate points. A lender with a lower note rate and higher points can be the more expensive one if your flip is short.

What a borrower can do with this

1. Place a quote on the distribution. In 2025, a non-bank note rate at or below 9.5% was in the cheapest quarter of the national market and one at or above 11% in the costliest quarter. Use your state's non-bank median from the table, not the national all-lender median, which includes bank loans you may not qualify for.
2. Ask a local bank first if you have one. Banks and credit unions made 14.3% of these loans at a 7.5% median, often with 36-month terms. They lend in fewer places and close slower, but the price gap was 2.75 points in 2025.
3. Convert points into a rate. Divide the points by the term in years and add them to the note rate before you compare. The shorter your project, the more the points weigh.
4. Compare more than one national lender in your state. The same lender's median moved 1.2 points across states, and lender medians ranged from 8.75% to 11.99% among the 20 largest. Ask each for the rate, points, extension fee and default rate in writing.
5. Plan the exit. Most of these loans are 12 months, interest-only, with a balloon. If you will hold the property, the 30-year DSCR-type median was 7.375% in 2025; our DSCR loan rates page has that record.

Kiavi pays us a referral fee when a loan closes through its button in the box below. The HMDA figures above are the same either way, and Kiavi's 2025 median (10.45% nationally) is in the lender table so you can compare. Visio Lending, also in the box, makes 30-year rental loans, the refinance step after a flip or BRRRR, not short-term loans.

FAQ

Loan-level data: FFIEC/CFPB HMDA Data Browser, nationwide originated conventional loans for 2024 and 2025, downloaded October 5, 2026 (request URLs in the data file; our scripts dl_nat.py, extract.py, hm_common.py and hm_us.py, and the extracted loan rows, are in the data folder); lender names and federal agency codes from the FFIEC filer lists and the FFIEC public institutions API, read October 6, 2026; agency-code definition from the FFIEC 2025 HMDA Filing Instructions Guide; coverage rules from 12 CFR 1003.3 and 1003.4 and the Supplement I official interpretations (eCFR); field definitions from the FFIEC public LAR data fields. All percentiles, medians, shares, changes and payment figures are our arithmetic. This is analysis of public documents, not investment, legal or tax advice, and not a loan offer.