Solar park loan DSCR in Hungary: how a bank sizes debt, the DSRA and the negative-price tail

The solar park loan DSCR (debt service coverage ratio) sets the loan size. A bank divides forecast cash flow after operating costs by the target DSCR and lends only what that smaller sum repays over the tenor. Lenders test a conservative yield case, discount merchant revenue, shorten the tenor and require a debt service reserve account (DSRA).

In Hungary the revenue source adds a second question: which scheme pays. KÁT (the legacy feed-in tariff), METÁR (the premium scheme) and merchant income carry different price risk, so the same park can support different amounts of debt depending on its revenue mix.

How does a lender size a solar park loan?

A lender works backwards from cash flow. It estimates the cash available for debt service (CFADS), which is revenue minus operating costs and taxes, in each year of the loan. It then divides that figure by the DSCR it requires and treats the result as the most the borrower can pay in that year.

Fitch Ratings defined DSCR in a 2012 presentation as cash available for debt service divided by debt service. A ratio above 1.00x means a cash cushion. The loan is the present value of those permitted payments, discounted at the loan's interest rate.

In steps:

1. Forecast energy output under a conservative yield case.
2. Multiply by expected prices, from contracts, support schemes or market curves.
3. Subtract operating costs and taxes to get CFADS.
4. Divide each year's CFADS by the target DSCR to get the maximum debt service.
5. Discount those payments at the loan rate over the tenor. The total is the debt capacity.

Debt capacity then gets compared with the project cost, so the sponsor funds the rest with equity.

What do P50 and P90 mean for debt sizing?

P50 is the yearly energy output expected to be reached or beaten in half of all years. P90 is the output expected to be reached or beaten in nine years out of ten, so it is lower. A research chapter from the Economic Research Institute for ASEAN and East Asia (ERIA) notes that project lenders and credit rating agencies often require developers to estimate the P50, P90 and even P99 of annual generation. It adds that banks want a higher level of confidence in actual output to judge whether a project can service its debt.

Fitch's 2012 presentation shows one use. Its rating case used "one-year P90 energy production" with further adjustments to availability and operating costs, while its base case applied a 0% to 10% adjustment to the energy forecast. The same slides stress the independent engineer's resource assessment. The exact method varies from lender to lender, and the loan documents set it.

What DSCR do lenders and rating agencies look for?

Banks do not publish their DSCR covenants, and the right number depends on contract cover, tenor, country and the lender. The public benchmarks come from rating agencies, and they are dated.

Read the table with care. These are rating-agency views, not bank loan covenants. They date from 2012 and 2017, and Fitch's current criteria may differ. They also describe contracted revenue: Fitch's 2012 list of indicative attributes for an investment-grade renewable project included "no merchant tail".

Banks do not publish a comparable figure for merchant revenue. The direction is clear, though. Bracewell, a law firm, wrote in 2022 that projected revenues from the merchant element of the offtake are typically discounted for debt sizing. It added that the more merchant risk in the offtake, the shorter the tenors.

What are sculpting and tenor?

Sculpting means shaping repayments so that debt service in each year equals that year's CFADS divided by the target DSCR. The coverage ratio then stays level across the loan. The alternative, a flat annuity, gives uneven coverage when cash flow moves.

Tenor is the length of the loan. Longer tenors lower each year's payment and raise debt capacity, but they extend exposure to later years when prices are less certain.

What is a debt service reserve account (DSRA)?

A DSRA is a cash reserve that the borrower must hold so that future debt service can be paid even if cash flow falls short. Forvis Mazars says it is generally sized to a number of months of projected debt service, "commonly six or twelve".

The term sheet sets how it is funded. Forvis Mazars lists three typical routes:

• Funded in full on the last day of construction.
• Partly funded then, and built up from project cash flows.
• Built up entirely from project cash flows.

For a solar park, the DSRA buys time. A bad quarter, a delayed payment from a buyer or a repair does not trigger a missed instalment while the problem is fixed. Its cost is cash tied up in the account, which lowers equity returns.

What is a cash sweep?

A cash sweep sends part of surplus cash to prepay the loan instead of paying it out as dividends. Bracewell describes sweeps as typically applied above a negotiated threshold in the base-case model, with surplus cash used to prepay the outstanding loans.

Sweeps matter most where the loan outlives the contract. Bracewell says a sweep helps reduce the exposure in the tail and the associated refinancing risk when a financing has a large bullet payment at maturity. A sweep lets good years shrink the balance before the merchant years arrive.

How do lenders treat KÁT, METÁR and merchant revenue in Hungary?

No Hungarian lender publishes its sizing rules, so the public evidence is the design of each revenue scheme plus a few attributed statements. The schemes differ in how much price risk the owner keeps, which is the logical basis for different treatment in a sizing model. The comparison below is an analysis of the decree terms, not a lender survey.

Negative prices. No premium is paid for output in the negative-price units that section 15(4) of Decree 299/2017 counts, which depends on a run of at least six consecutive negative units. Check the unit length in the offtaker's rules. The KÁT decree text has no such clause. In a sizing model, the negative-price tail is therefore a separate input for METÁR and merchant revenue. The detail of each scheme is in METÁR vs METÁROLÓ vs KÁT.

Merchant revenue. Montel reports that Hungarian solar capture prices fell from 90% of the average price in 2020/21 to 55% in the first half of 2025 (Montel), and the capture-rate article has the series, the method and the negative-price counts.

What market participants say. In July 2026, Dr Gábor Juhász-Farkas, a SolServices managing director (ügyvezető), said that most KÁT projects were built with commercial bank financing (SolServices interview). SolServices' experience in 2023 was that market-based financing still seemed too risky to many in Hungary, that financiers preferred PPA-backed projects and that the PPA market was still young (SolServices, 2023). The Voice of Renewables wrote in August 2026 that European lenders stress-test merchant exposure more conservatively than in 2025. That is trade-press commentary, mostly about Spain, France and Italy, not a lender survey.

"Tail" has two meanings here: the merchant tail, the years after a contract ends, and the price tail, the hours at or below zero. A model can treat those hours as zero revenue or as curtailed volume. Either way, they pull down the capture rate that feeds CFADS.

How much does a lower capture rate cut debt capacity?

By more than the revenue falls, because operating costs stay the same. The table below is a worked example with assumed inputs. It is not market data, a lender quote or a SolServices figure.

Assumptions: 100 MWp park with a P50 specific yield of 1,250 kWh/kWp, or 125 GWh a year. PVGIS returns 1,252 kWh/kWp for a south-facing fixed-tilt system (35 degrees tilt, 14% system losses) near Szolnok. The P90 case is set at 92% of P50, or 115 GWh a year, flat for the whole loan. Baseload price EUR 90/MWh. Operating costs EUR 1.2 million a year. Fully merchant revenue. Target DSCR 1.50x on P90. Level 15-year tenor, annual payments sculpted to the DSCR, interest rate 6.0%. No tax, degradation, inflation, fees or hedges. The capture rate is net of negative-price hours. The DSRA is six months of debt service.

From 85% to 45%, revenue falls 47% but debt capacity falls 54%. A lender that also wants more cover or a shorter loan will lend less again. The next table holds capture at 55% and changes the lender's terms, with a contracted case for comparison.

The DSCR values above 1.50x and the EUR 70/MWh price are assumptions chosen to show the mechanics. In the PPA case the price is fixed per MWh produced, so the capture rate drops out and only volume risk remains. The article on pay-as-produced and pay-as-nominated PPAs explains who carries the shape risk in each contract.

How were SolServices' own parks financed?

SolServices financed Lumen Park Szolnok with a bilateral forint loan from UniCredit Bank Hungary at 80% credit intensity with an interest-rate swap, and Lumen Park Szászberek with a euro loan at 70%, as set out in SolServices' 2023 interview on the financing. The DSCR, tenor, margin and reserve terms are not public, and this article does not infer them.

The Lumen Park Szolnok portfolio page has the park's background.

UniCredit's 2022 annual report calls the Szolnok financing the largest corporate green loan agreement of 2022 in Hungary, and the deal won a Transaction of the Year award that year (SolServices, 2023). UniCredit Bank Hungary's release of 13 October 2023 names it as an example of the bank's green lending.

The SolServices fleet has three KÁT plants, one METÁR plant and two market-based plants (interview with Dr Gábor Juhász-Farkas, July 2026). The 150 MW / 300 MWh battery programme is financed purely on market terms.

Lumen Park Mezőtúr is one of the market-based plants: it was the group's first fully free-market project in Hungary (Polat Holding, 2023 annual report).

What this means for solar park owners, investors and buyers

Debt capacity follows the revenue that a lender will accept, not the revenue an owner hopes for. As of October 2026, the public evidence points one way: merchant solar revenue is discounted and covered more thinly than contracted revenue.

• Owners of merchant or part-merchant parks should track capture rate and negative-price hours as separate inputs, since both feed CFADS.
• KÁT and METÁR owners should know the remaining years and energy in the MEKH decision, because supported revenue counts only for that period. METÁR owners should also map which of their settlement units fall under the negative-price rule.
• Owners planning a refinancing face a market where the same park may support less debt than it did at the original financing, if capture rates have fallen since. The sweep and DSRA terms matter as much as the headline margin.
• Investors should read the DSCR together with the tenor, the contracted share and the reserve. A low DSCR on a short contracted period is a different risk from a low DSCR on a long merchant tail. The monthly operating pack that lenders and investors can ask for is set out in running a solar park as an independent power producer.
• Buyers of a financed park inherit loan terms or must refinance. The guide on how to buy a solar park in Hungary covers the due diligence steps around this.