CrowdfundedWealth

Rental Property Refinance Rates 2025: Cash-Out vs Rate-and-Term

ResearchBy Jorge19 min read

Some links pay us a referral fee; each one says so. Disclosure

Quick answer

As of November 19, 2026, the federal mortgage record for 2025 (HMDA) shows 36,501 originated refinances on one-to-four-unit rental property in Florida, Texas and Georgia: 14,454 rate-and-term at a median note rate of 7.059% and 22,047 cash-out at 7.25% (30-year fixed loans, our arithmetic from HMDA 2025 records). The gap between the two was 0.191 points, against 0.625 points for homeowners refinancing their own house (6.0% and 6.625%). 29.6% of the investor applications that ended in a loan or a denial were denied (15,363 of 51,864), 29.5% of rate-and-term and 29.7% of cash-out, against 33.0% for the same refinances on a principal residence. Collateral was the most common denial reason for a rental in all three states and both loan types, and denial rates for cash-out rose from 28.3% at a combined loan-to-value (CLTV) of 75% to under 80% to 49.1% at 80% to under 85%.

Update alert · free

An email when the rental property loans numbers change

When a rate, rule or filing behind this page changes: what changed, the one number that matters, and the source to check it yourself.

Key takeaways

  • Three states, 36,501 rental refinances originated in 2025: Florida 14,892 from 538 lenders, Texas 15,396 from 527 lenders, Georgia 6,213 from 364 lenders. 60.4% were cash-out (Florida 67.0%, Georgia 58.3%, Texas 54.8%), against 46.7% for refinances of a principal residence (our arithmetic).
  • The cash-out penalty is small on a rental and large on a home. Pooled median note rate on 30-year fixed loans: rental rate-and-term 7.059%, rental cash-out 7.25% (0.191 points); principal-residence rate-and-term 6.0%, cash-out 6.625% (0.625 points). The investor premium over a homeowner was 1.059 points on rate-and-term and 0.625 on cash-out (our arithmetic).
  • 29.6% of investor refinance applications were denied (rate-and-term 29.5%, cash-out 29.7%). Homeowners were denied 27.4% on rate-and-term and 38.4% on cash-out. In Florida the rental rate-and-term denial rate (34.5%) was above the cash-out rate (29.6%).
  • Collateral led the denial reasons on rentals in all six state and purpose groups (26.9% to 32.8% of denials named it). For homeowners the leading reason was debt-to-income ratio or credit history on cash-out and an incomplete application on rate-and-term.
  • Leverage: 13.8% of rental rate-and-term loans with a CLTV were at 80% or more, against 3.3% of cash-out loans. Denial rates for cash-out jumped from 28.3% (75 to under 80) to 49.1% (80 to under 85) and 70.9% (85 and above); for rate-and-term the jump came later, 27.7% at 80 to under 85 and 55.6% at 85 and above.
  • Who lends: United Wholesale Mortgage was first in all three states (1,412 in Florida, 1,560 in Texas, 551 in Georgia). The ten largest lenders made 40.8% of rental refinances in Florida, 34.6% in Texas and 36.1% in Georgia. Only 3.7% to 8.5% of rental refinances were sold to Fannie Mae or Freddie Mac in the year, against 36.0% to 54.1% for owner-occupied ones.
  • On a rental valued at $300,000 with $180,000 owed, a rate-and-term refinance at the pooled 7.059% is $1,204.69 a month and a cash-out to 70% ($210,000 at 7.25%) is $1,432.57: $227.88 a month more, $82,038.55 over 360 payments for $30,000 of cash before costs (our arithmetic).

CSV · 659 rows

Rental-property refinances in Florida, Texas and Georgia in federal HMDA data, 2025: rate-and-term against cash-out, with owner-occupied loans, denial reasons, CLTV bands and lenders

659 rows: for Florida, Texas and Georgia, investment property and principal residence, rate-and-term and cash-out refinances: originated, denied, denial rate, median rate and points, CLTV, loan size, channel, sales to Fannie Mae and Freddie Mac, denial reasons, denial by CLTV band, the largest lenders, two Fannie Mae Selling Guide rules and the worked example.

What the federal record adds

Lender pages describe this choice in ranges and general terms. The federal record counts the loans. Lenders above a size threshold report every mortgage they originate and every application they decide, with its purpose, occupancy, note rate, combined loan-to-value ratio (CLTV), amount, lender and, for denials, the reasons. The CFPB announced that the 2025 data were available on March 31, 2026.

We downloaded the 2025 records for three large investor states, Florida, Texas and Georgia (all three are in the ten largest states of our national count; see our national cash-out page for the national counts, the agency loan-to-value limits and the DSCR lenders' own terms, which we do not repeat). This page asks the question a landlord with an existing loan asks: rate-and-term or cash-out, and what did the market do in 2025 for each. HMDA labels a loan as cash-out when the lender “considered it to be a cash-out refinancing in processing the application or setting the terms” (Regulation C, Supplement I, comment 4(a)(3)-2), so the split is the lender's own label.

Rate-and-term against cash-out, rental against home

Refinances originated in 2025 in Florida, Texas and Georgia combined (first lien, closed-end, one to four units, not reverse)

Measure (2025)Rental, rate-and-termRental, cash-outHome, rate-and-termHome, cash-out
Loans originated14,45422,047106,36593,358
Applications denied6,0629,30140,05558,229
Denial rate (denied / originated plus denied)29.5%29.7%27.4%38.4%
Median note rate, 30-year fixed7.059%7.25%6.0%6.625%
Loans in that rate median9,83018,95875,44671,794
Median CLTV70%69.784%79.213%67.52%
Median loan amount$235,000$215,000$355,000$245,000

Three readings. First, a rental is mostly a cash-out business: 60.4% of the 36,501 rental refinances took cash out, against 46.7% of home refinances (our arithmetic). Second, the cash-out premium shrinks on a rental. A homeowner who takes cash out paid 0.625 points more than one who did not; a landlord paid 0.191 points more. The reason is not that cash-out is cheap for landlords but that rate-and-term is dear: 7.059% against a homeowner's 6.0%, a premium of 1.059 points (our arithmetic). Third, the denial pattern differs. A homeowner's cash-out was denied 38.4% of the time against 27.4% for a rate-and-term; a landlord's, 29.7% against 29.5%. HMDA cannot say why, but one candidate is that many rental loans are underwritten on the property's rent rather than the borrower's income, so a debt-to-income test is less often the reason (our reading; HMDA has no field for the underwriting method).

Points are thinly reported. Discount points were present on only 935 of 9,830 rental rate-and-term loans in the rate median and 1,886 of 18,958 cash-out loans (median $2,657.31 and $2,608), against 35,604 and 56,293 for home loans, so the rental rates above are not directly comparable with a quote that includes points.

State by state: Florida, Texas and Georgia

Rental refinances (investment property), 2025

StateRate-and-term loansCash-out loansCash-out shareMedian rate, rate-and-termMedian rate, cash-outGap (points)Denial rate, rate-and-termDenial rate, cash-out
Florida4,9119,98167.0%7.125%7.25%0.12534.5%29.6%
Texas6,9528,44454.8%7.0%7.375%0.37527.3%29.6%
Georgia2,5913,62258.3%7.0%7.34%0.3425.2%30.0%

The same refinances on a principal residence, and what a landlord paid over a homeowner

StateHome median rate, rate-and-termHome median rate, cash-outRental premium, rate-and-term (points)Rental premium, cash-out (points)Rental share sold to Fannie Mae or Freddie MacHome share sold to Fannie Mae or Freddie Mac
Florida5.999%6.5%1.1260.757.0% and 3.7%52.7% and 47.5%
Texas6.125%6.999%0.8750.3768.5% and 5.1%50.6% and 36.0%
Georgia5.999%6.49%1.0010.858.2% and 6.4%54.1% and 54.0%

In the last two columns the first figure is rate-and-term and the second is cash-out. The premium of a landlord over a homeowner was largest on rate-and-term in every state and was 0.376 points on a Texas cash-out, where homeowners paid 6.999%. Between 3.7% and 8.5% of rental refinances were reported as sold to Fannie Mae or Freddie Mac in the year, against 36.0% to 54.1% of home refinances: most rental refinances stay on a lender's balance sheet or go to private investors, and 80.6% to 87.2% were reported as made for a business or commercial purpose (our arithmetic). Where the loan goes after closing decides the underwriting rules that apply to it.

Channel also differs. The share of rental applications that did not come directly to the lender (the HMDA field marks those submitted through a broker or correspondent) was 47.3% on Florida rate-and-term and 65.7% on Florida cash-out, 46.1% and 54.0% in Texas and 33.1% and 53.5% in Georgia.

Leverage: where the denial rate turns

Rental refinances in the three states by combined loan-to-value (CLTV) at application, 2025

CLTVRate-and-term originatedRate-and-term deniedRate-and-term denial rateCash-out originatedCash-out deniedCash-out denial rate
Under 602,87983422.5%5,2071,85426.3%
60 to under 703,01693423.6%4,9591,65325.0%
70 to under 752,62891125.7%4,4111,65127.2%
75 to under 803,0011,11027.0%4,7721,88128.3%
80 to under 851,44155227.7%44643149.1%
85 and above39849855.6%21853270.9%

Only loans and applications with a CLTV on the record are in this table. Close to half of rental loans sit at 70% to under 80% (42.1% of rate-and-term and 45.9% of cash-out loans with a CLTV). Above 80% a cash-out is rare, 664 of 20,013 loans with a CLTV (3.3%), and mostly refused: 963 of the 1,627 applications at 80% and above were denied, 59.2% (our arithmetic). A rate-and-term at 80% to under 85% was denied at 27.7%, close to the 27.0% at 75% to under 80%, and 1,441 were made. The share of rental loans above 75% was 16.8% (Florida), 23.3% (Texas) and 16.6% (Georgia) on rate-and-term, and 3.2%, 5.6% and 4.4% on cash-out (our arithmetic). In this record, the loans above 75% of value were mostly rate-and-term.

Why rentals get denied

Share of denied rental refinance applications naming each reason, 2025 (up to four reasons per application, so the shares add to more than 100%)

State and typeDeniedCollateralDebt-to-income ratioCredit historyIncomplete applicationOther
Florida, rate-and-term2,58328.5%17.2%15.1%15.5%16.1%
Florida, cash-out4,20332.8%18.7%20.4%17.1%16.2%
Texas, rate-and-term2,60526.9%17.2%14.5%20.3%18.6%
Texas, cash-out3,54431.0%20.9%20.6%16.8%17.3%
Georgia, rate-and-term87430.1%17.8%14.9%12.4%15.9%
Georgia, cash-out1,55427.2%19.4%23.9%14.9%13.3%

Collateral, the HMDA reason code for problems with the property that secures the loan, was the most common reason in all six groups. For homeowners the leading reason on cash-out was debt-to-income ratio (33.1% in Florida, 31.8% in Texas) or credit history (31.8% in Georgia), and on rate-and-term an incomplete application (26.2%, 25.7% and 24.1%). The practical reading for a landlord: a problem with the property or its value was the most common single reason named on a refused rental refinance, ahead of the borrower's credit history. Denial reasons are the lender's recorded reasons, not findings of fault, and a denial may name several.

Who made the loans

Largest lenders of rental refinances in each state, 2025 (rate-and-term plus cash-out originations; the ten largest made 40.8%, 34.6% and 36.1%)

StateLenderOriginatedRate-and-termCash-outShare of stateDenial rate
FloridaUnited Wholesale Mortgage1,4123911,0219.5%34.5%
FloridaA&D Mortgage LLC9271817466.2%16.8%
FloridaChampions Funding, LLC7241775474.9%25.8%
FloridaHomeXpress Mortgage Corp.545804653.7%47.6%
FloridaKiavi Funding, Inc.52252203.5%26.4%
TexasUnited Wholesale Mortgage1,56074581510.1%29.5%
TexasThe Loan Store, Inc.5541893653.6%13.7%
TexasChampions Funding, LLC5101613493.3%33.8%
TexasHomeXpress Mortgage Corp.468813873.0%52.3%
TexasRF Renovo Management Company, LLC421963252.7%0.0%
GeorgiaUnited Wholesale Mortgage5511873648.9%22.3%
GeorgiaKiavi Funding, Inc.39239206.3%18.5%
GeorgiaRocket Mortgage, LLC263991644.2%34.2%
GeorgiaChampions Funding, LLC173481252.8%30.2%
GeorgiaHomeXpress Mortgage Corp.159251342.6%48.2%

These are rankings by originations only; each state's list has 364 to 538 lenders and the five shown are not a recommendation. A lender's denial rate here depends on its channel and product (a lender that takes applications through brokers and pre-screens has a different mix from one that takes walk-in applications), and the lender's name is no judgment on how it treats borrowers. Two reporting points matter. Kiavi Funding reports no cash-out loans at all in these states, so all its loans sit in the rate-and-term column; because the lender chooses the label, that column may include loans where the borrower took cash. Our rate medians therefore describe what the lenders reported, not what each borrower did. Second, Velocity Commercial Capital appears with 382 loans in Florida (346 cash-out), 327 in Texas and 151 in Georgia, with 2, 3 and 9 denials: a lender with so few recorded denials is not comparable on denial rate with one that records many.

A worked example: what $30,000 of cash costs (our arithmetic)

Take a rental valued at $300,000 with $180,000 owed on a loan at 4.25% with 300 months left. The three outcomes below use the pooled 2025 medians for 30-year fixed loans on a rental (our arithmetic; closing costs, points and the lender's own quote are not included).

ChoiceNew loanRateMonthly principal and interestChange from today's $975.13
Keep the existing loan$180,0004.25% (stated)$975.13none
Rate-and-term refinance$180,0007.059%$1,204.69+$229.56
Cash-out to 70% CLTV$210,0007.25%$1,432.57+$457.44

The cash-out costs $227.88 a month more than the rate-and-term, $82,038.55 over 360 payments, for $30,000 of cash before costs. Of the $227.88, $204.65 a month is the extra $30,000 borrowed for 30 years at 7.25% and $23.23 is the 0.191-point higher rate on the first $180,000 (our arithmetic). When the existing rate is far below the market, as here, the refinance question is mostly whether the cash is worth giving up the old rate on the first $180,000; that depends on what the cash earns and is for you or an adviser to weigh.

What Fannie Mae's guide says about the two types

Two rules in the Selling Guide change the choice. A limited cash-out refinance, Fannie Mae's name for a rate-and-term, may include “providing cash back to the borrower (or any other party) in an amount that, in aggregate, does not exceed the greater of 1% of the new refinance loan amount or $2,000” (Selling Guide B2-1.3-02, version dated 10/08/2025); on the $180,000 loan above that is $2,000, since 1% would be $1,800. For a cash-out refinance, “At least one borrower must have been on title for at least for six months prior to the disbursement date of the new loan” (B2-1.3-03, version dated 12/10/2025), with exceptions such as inheritance. Nationally, the loan-to-value limits for rental property and Freddie Mac's matching rules are on our cash-out page, and the investor premium in purchase loans is on our investment property mortgage rates page. HMDA shows that most rental refinances were not sold to either company, so these rules bind a minority of the loans above; lenders that keep or sell to private investors publish their own.

What these numbers cannot tell you

  • The rate is the rate at origination in 2025. It is not a quote for today, and rates moved during the year.
  • No credit score and no debt-service coverage. HMDA has neither, and a lower median can reflect a stronger borrower or more points paid. Points were reported on a small share of rental loans.
  • Rate medians cover only 30-year fixed loans with no introductory period. Adjustable and interest-only loans, 40-year loans and short-term loans are not in the rate figures.
  • CLTV was missing on 25,592 originated loans across all occupancies and purposes in the three files; they are left out of CLTV figures and bands. Dollars and counts of originated loans and denials are complete for the filter.
  • Business-purpose loans: 80.6% to 87.2% of rental refinances were reported as made for a business or commercial purpose; we did not exclude them, because many rental loans are labeled that way.
  • Only three states. Florida, Texas and Georgia are not the country; the national comparison is on our cash-out page.
  • We counted 36,501 loans from the three state files and checked the method against the official aggregation service: Florida cash-out refinances originated, all occupancies, were 68,738 in our stream and 68,738 in the Data Browser's aggregations API.

What a landlord can do with this

  1. Decide which of the two you are asking for before you collect quotes. If you only want a lower rate or a different term, ask for a rate-and-term and for the lender's cash-back limit in writing; if you want money out, ask for the cash-out price separately, because in 2025 the median spread between the two was between 0.125 and 0.375 points in these states.
  2. Get the appraisal question answered early. Collateral led the reasons for denial in all six groups, so ask how the lender orders the appraisal, who pays if it comes in low, and whether you can walk away.
  3. Ask for the cost sheet, not only the rate. Points were reported on a minority of rental loans; a low note rate may come with points or a prepayment charge that HMDA does not show. For DSCR-specific timing and penalties, see how to refinance a DSCR loan.
  4. Check who the lender is for your state and your product. The five largest lenders in each state made different mixes of rate-and-term and cash-out loans; a lender with hundreds of loans like yours has done the work.
  5. If the plan is to buy, fix, rent and refinance, the timing rules matter more than the rate; our BRRRR page has the seasoning rules and national numbers.

FAQ

This is analysis of public documents, not investment, legal or tax advice. Sources: Federal Financial Institutions Examination Council, HMDA Data Browser, 2025 loan-level records for Florida, Texas and Georgia (https://ffiec.cfpb.gov/v2/data-browser-api/view/csv, downloaded November 19, 2026) and its aggregations API; CFPB, HMDA data page (2025 data announced March 31, 2026); Regulation C, 12 CFR part 1003, Supplement I, comment 4(a)(3)-2 (eCFR, November 2026); Fannie Mae Selling Guide B2-1.3-02 (version dated 10/08/2025) and B2-1.3-03 (version dated 12/10/2025), accessed November 19, 2026. Medians, shares, denial rates and the worked example are our arithmetic from the HMDA records.

Affiliate Disclosure: Some links are affiliate links. We may earn a commission at no extra cost to you. This does not affect our ratings. Learn more.