Solar Loan Dealer Fees: Why the Cash Price Is Lower

In a complaint the State of Minnesota filed in Hennepin County District Court on 8 March 2024, a homeowner is offered one solar system at two prices. $28,700 if they pay the installer's cash price. $38,818 if they finance it through the installer's lender. Same system, same installer. The complaint says the difference "accounts for GoodLeap's upfront fee" (State of Minnesota v. GoodLeap LLC et al., No. 27-CV-24-3558, complaint ¶ 55, read 19 September 2026).

That $10,118 is a dealer fee. It is not interest, so it does not appear in the loan's interest rate. It is not a line on the proposal, so it does not appear in the price breakdown. And it is not in the APR on the federal disclosure either, which is the part most people find hardest to believe until they see how the paperwork is built.

A quick note on what follows. Everything quoted from the Minnesota complaint is the state's allegation, not a court's finding. The lenders' responses are part of the court record, and no ruling on the merits turned up in the searches run for this piece; the Minnesota court docket under that case number is where to check the current status. The legal outcome is not what matters here. The arithmetic the complaint lays out is, because it matches what the federal consumer regulator found nationally, and it is arithmetic you can run on your own quote.

One fee, two percentages, and which one you will be quoted

Start with the $10,118, because it already contains the first trap.

cash price                        $28,700
financed price / loan principal   $38,818
difference                        $10,118

as a share of the loan            10,118 / 38,818 = 26.1%
as a share of the cash price      10,118 / 28,700 = 35.3%

Both percentages describe the same dollars. The first is how the lenders in the Minnesota case measured it: the complaint gives GoodLeap's average fee as 19.32% "of each loan" (¶ 51), and puts Sunlight Financial's at 21.4% (¶ 92), Solar Mosaic's at 17.6% (¶ 141) and Dividend Solar Finance's at 18.8% (¶ 192), each stated as a share of the borrower's loan amount. Across the four, the state alleges about $35 million in fees on nearly 5,000 Minnesota loans since 2017 (Minnesota Attorney General press release, 8 March 2024, read 19 September 2026).

The CFPB counted the other way. Its Issue Spotlight: Solar Financing (August 2024, read 19 September 2026) says the fees "typically range from between 10 to 30 percent of the cash price but can exceed 50 percent."

The conversion is one line. If the fee is f as a share of the loan, it is f / (1 − f) as a share of the cash price:

So when someone tells you a fee is "about 20 percent," ask twenty percent of what. The loan-based figure is always the smaller one, and it is the figure a lender's product sheet will naturally use, because the lender's fee is calculated on the loan it writes.

Where the $10,118 goes

The mechanics are simple once you see who gets paid. The CFPB's own worked example uses a $30,000 cash price and a $9,000 fee: "The loan documents would show a loan principal of $39,000 plus interest. The lender would remit the $30,000 cash price to the installer and keep the $9,000 hidden fee."

The installer is paid what it would have charged a cash buyer. The lender keeps the difference out of the loan proceeds on day one. You repay the full principal, with interest, on the whole thing, including the part nobody installed.

The complaint describes installers who priced it exactly that way. One former partner of GoodLeap is quoted as saying that on a $30,000 cash price the lender "would extend a loan of approximately $37,650-$37,950 to the consumer but retain between $7,650-7,950 of the loan proceeds" (¶ 58). Another, Sun Badger Solar, is described as having told its sales staff in a 2020 meeting to present Sunlight Financial loans "at the 10% higher price" and to offer the lower cash price only if the customer indicated they could pay some other way.

The fee size is not individual, either. The complaint says the fee on a given loan depends on the term and interest rate the installer selects from the lender's menu, not on the borrower's credit (¶ 52–53). The lower the advertised rate, the more money up front it takes to buy it down, and the fee is that money.

That last point is the whole economic engine, and it is why a dealer fee is not automatically a bad deal. More on that below.

Why the APR on the disclosure leaves it out

If you have signed a solar loan, you received a Truth in Lending disclosure with a box labelled Annual Percentage Rate and another labelled Finance Charge. The natural assumption is that anything you pay because you borrowed shows up in one of them. Regulation Z's own definition leans that way: the finance charge "includes any charge payable directly or indirectly by the consumer and imposed directly or indirectly by the creditor as an incident to or a condition of the extension of credit" (12 CFR 1026.4(a), text pulled from the eCFR versioner API as of 15 September 2026).

Two other passages of the same regulation pull in opposite directions, and the fight over solar dealer fees sits between them.

The passage lenders can point to. Section 1026.4(c)(5) excludes "seller's points" from the finance charge. The official commentary to that paragraph defines them as "any charges imposed by the creditor upon the noncreditor seller of property for providing credit to the buyer or for providing credit on certain terms," and adds: "These charges are excluded from the finance charge even if they are passed on to the buyer, for example, in the form of a higher sales price." If the fee is a charge on the installer, and the installer passes it on through the price, it stays outside the APR.

The passage consumers can point to. Section 1026.4(b)(9) lists, as an example of a finance charge, "Discounts for the purpose of inducing payment by a means other than the use of credit." The commentary gives an illustration that reads uncomfortably close to a solar quote: land offered at $10,000 per tract, $9,000 if the buyer pays cash. "The $1,000 difference is a finance charge for those who buy the tracts on credit."

That land example involves a seller who is also the lender, which a solar installer is not, and that distinction is part of what is being argued. Minnesota's complaint alleges that GoodLeap "devised contracts with Minnesota solar companies that disguise the charge as an overhead cost incurred by the Minnesota solar company" (¶ 61), while the fee is in fact "paid by the borrower, at Solar Mosaic's direction, as a condition of financing" (¶ 158). It also quotes the small print that appeared on some disclosures, for example Solar Mosaic's: "The Installation Contractor may provide us a fee or discount against the Amount Financed noted herein" and "[t]he Amount Financed shown is not reduced by any such fee or discount" (¶ 163).

You do not need to resolve the legal question to use the practical answer. On a solar loan with a dealer fee, the APR box measures the interest rate on an inflated principal. It does not measure what borrowing costs you relative to paying cash. Comparing two solar loans by APR alone, which is exactly what the payment-focused sales pitch invites, compares the wrong thing.

When a fee loan is cheaper, and the month it stops being cheaper

Here is where most write-ups on dealer fees stop, with the conclusion that fee loans are a rip-off. The arithmetic does not support that as a blanket statement.

A dealer fee is prepaid interest. You pay a lump up front (inside the principal) in return for a lower rate for the life of the loan. Whether that trade pays depends almost entirely on how long you keep the loan, and the CFPB reports that "solar loans are typically repaid in 7 to 9 years due to prepayments," against stated terms of 8 to 25 years.

To see where the line falls, put the ¶ 55 prices next to a hypothetical no-fee loan. The 2.99% and 7.99% rates below are assumptions for illustration, not figures from the complaint. The complaint does not state the rate on that loan, and 2026 rates vary by lender and credit; the point is the method, so substitute the two quotes in front of you.

Loan A (with dealer fee)      $38,818 at 2.99%, 300 months
  monthly payment               $183.88
Loan B (no fee, e.g. a credit union)
                              $28,700 at 7.99%, 300 months
  monthly payment               $221.32

Payment formula:  M = P × r / (1 − (1 + r)^−n),  r = annual rate / 12

Loan A has the lower payment by $37.44 a month, and that is the number the pitch leads with. Now ask what each loan really costs if you pay it off early. The effective rate below is the annual rate that makes your actual payments, plus the payoff balance in the exit month, equal the $28,700 you would otherwise have handed the installer in cash.

The crossover sits at month 94, just under eight years. Pay Loan A off before then and the dealer fee made the money more expensive than the 7.99% loan. Keep it longer and it made the money cheaper. Held to term, the "2.99%" loan actually costs about 5.9% on the cash price, which is still below 7.99%.

Two things follow.

First, the CFPB's 7-to-9-year typical repayment window lands right on top of the crossover. The average borrower, in other words, is close to the point where the fee is a wash, and a borrower who sells the house in year four is well on the wrong side of it. The complaint alleges that for 88 Minnesota GoodLeap borrowers who paid off early, the effective rate once the fee was counted exceeded the state's legal limit, by an average of $3,417.14 in finance charges each (¶ 66). That is the same curve as the table, with a statute drawn across it.

Second, the decision is not really about the fee. It is about your exit date. People planning a move within five years, or expecting to refinance the balance into a mortgage, are paying for a low rate they will not keep. People who intend to stay put for fifteen years may be getting a decent price on credit. Neither group can tell which case they are in from an APR.

The effective-rate column is an internal rate of return on monthly cash flows. Any spreadsheet does it: put −28,700 in the first cell, the monthly payment in each following cell up to your exit month, add the payoff balance to the last one, and use =RATE(n, payment, -28700, -balance) * 12, or IRR on the column times 12.

The payment step that was designed around a credit that is gone

Many fee loans had a second feature the CFPB spent a whole section on. The payment you were shown assumed you would prepay a large lump, and if you did not, the loan re-amortized to a higher payment. In the spotlight's words: "It is commonplace for solar-specific loans to re-amortize at a higher monthly payment amount at the 19th month of the loan term if the consumer does not make a substantial prepayment before then. Frequently, the necessary prepayment to avoid that re-amortization is 30 percent of the loan principal, which is the current size of the federal Investment Tax Credit."

Two details matter for the fee.

The 30% was of the loan principal, fee included. On the ¶ 55 loan that is $11,645, while 30% of the cash price is $8,610. The cash price is the defensible basis for the credit, because the IRS page tells filers not to include "interest paid including loan origination fees" in the cost, and Minnesota's complaint argues the fee is exactly that kind of financing cost (¶ 18). Even a household that received the full credit would have been $3,035 short of the prepayment. The fee pushed up the target the credit was supposed to hit.

And the credit is gone. The IRS page for the Residential Clean Energy Credit now states that it "is not available for any property placed in service after December 31, 2025" (IRS, Residential Clean Energy Credit, read 19 September 2026). What that repeal does and does not reach is covered in the post on the 25D expiry. For this subject the consequence is narrow. Any 2026 loan whose payment schedule still has a step at month 18 or 19 keyed to a percentage of principal is assuming a lump sum that no federal credit will supply for a system installed this year. Whether a given lender still structures loans that way is not something I could verify across the market. It is something you can verify on your own disclosure, because the payment schedule has to show every scheduled payment amount and when it changes.

Folding the fee into a payback number

A payback figure is cost divided by annual saving, and the dealer fee sits on the cost side. How you count it depends on which question you are asking.

If the question is whether the system pays back, use the cash price. $28,700, not $38,818. The fee is not part of the system; it is part of the financing, and using the financed price overstates what the equipment costs. That is the figure to set against the annual bill saving you build from the four numbers on your own electric bill.

If the question is whether this purchase, financed this way pays back, use the total you will actually pay by the year you expect to exit, from the table above. At a seven-year exit that is $46,131 on Loan A against $43,903 on Loan B, a $2,228 gap. At two years it is $7,850. The gap belongs in the cost line of the financed payback, not in a footnote.

What you should not do is the thing proposals often do, which is compare the monthly loan payment with the monthly bill saving and call the difference a profit. That is a cash-flow comparison. A $184 payment against a $200 saving looks like a win in month one whether the principal behind it is $28,700 or $38,818. The same quiet problem runs through the rate escalator and degradation assumptions on the savings side, which get their own treatment in the three dials that shorten payback on paper.

The detail that makes this worse after 2025 is simple. With a 30% credit, a proposal could hide a $10,000 fee behind a $9,000 tax benefit and the "net cost" headline barely moved. The CFPB flagged that exact display, the real loan principal "in a small, light font" and the net-of-credit figure "in a large, bright font." With no credit to subtract, the fee now lands on the payback figure at full size.

What to ask, and whom to ask

The installer and the lender each know half of this, and the proposal is written by the one with less reason to volunteer it. Ask in writing, before any financing paperwork is opened on a tablet.

To the installer:

1. What is the cash price for the identical system, with the same equipment list, the same warranties and the same scope of work? Get it on the proposal itself, not in a text message. Item 5 of a complete solar proposal is exactly this number.
2. If I bring my own financing from a bank or credit union, is that the price I pay?
3. Which loan product did you select for this quote, by name and term?

To the lender:

1. What fee applies to that product, as a percentage and in dollars, and is it calculated on the loan amount or the cash price?
2. Do you offer the same term with no fee at a higher rate? The Minnesota complaint describes lenders that came to offer low- or 0%-fee loans with higher disclosed rates, which installers "generally reject as unattractive" (¶¶ 124, 208), and the CFPB describes lenders that "price all costs of credit into the APR" instead of marking up the price.
3. Does the payment schedule change at any point before maturity, and what prepayment would prevent that?

Regulation Z also gives you something to request directly. A creditor must provide a written itemization of the amount financed, including "any amounts paid to other persons by the creditor on the consumer's behalf," or tell you that you have the right to receive one (12 CFR 1026.18(c)). If you are offered the tick box, tick it. Depending on how the fee is booked, the itemization may or may not show the gap between what the installer receives and what you owe. It costs nothing to find out.

Then do one subtraction: financed price minus cash price. Put the answer next to your expected years in the house, run it through the effective-rate formula above with the no-fee rate your own bank quotes you, and see which side of the crossover month you land on. That single number is the one the APR box was never going to tell you.