Markets
When, not how much: what sets the value of an Italian solar MWh
On 1 May 2026 the Italian PUN — the national reference power price — sat at zero euros per MWh for about six hours, across the whole country, in the middle of the day. The notable part is not that it happened. It is that it happened again: the same thing occurred on 1 May 2025. The midday zero is no longer a curiosity. It is becoming an annual fixture, and the window widens every spring.
For a solar plant that reframes the question. The value of a MWh is set less by how much you produce than by when you inject it — and peak production now lands in the hours the price is heading to zero. The chart everyone draws is the production curve. The one that pays the bills is the price curve.
Intraday zonal price shape
| Hour | EUR/MWh |
|---|---|
| 00:00 | 95 |
| 01:00 | 88 |
| 02:00 | 82 |
| 03:00 | 80 |
| 04:00 | 82 |
| 05:00 | 88 |
| 06:00 | 96 |
| 07:00 | 100 |
| 08:00 | 84 |
| 09:00 | 52 |
| 10:00 | 18 |
| 11:00 | 3 |
| 12:00 | 0 |
| 13:00 | 0 |
| 14:00 | 1 |
| 15:00 | 6 |
| 16:00 | 20 |
| 17:00 | 48 |
| 18:00 | 86 |
| 19:00 | 120 |
| 20:00 | 138 |
| 21:00 | 140 |
| 22:00 | 119 |
| 23:00 | 104 |
The midday zero is the new baseline
On a sunny, low-demand day the zonal price traces a canyon: a deep trough at midday, when solar floods the market, and a sharp peak after sunset, when it leaves. On 1 May 2026 the floor of that canyon touched zero for six hours and the evening climbed back above 130 EUR/MWh.
Italy still has a regulatory floor at zero. Unlike Spain, Germany or France, the day-ahead price cannot yet go negative — the floor is the only thing standing between «free» and «you pay to inject». It is under review, not removed. So the Italian version of cannibalisation does not show up as negative prices. It shows up as hours that are simply worth nothing, clustered exactly when the panels work hardest.
The mechanism: capture-price decay
The mechanism has a name: capture-price decay. As solar penetration rises, solar pushes down its own selling price in the very hours it produces — it cannibalises itself. Two numbers track it: the capture price (the average price a solar MWh actually earns) and the capture factor (that price as a share of the baseload average). Both fall as more panels come online.
Solar capture price and capture factor
| Market/year | Capture price (EUR/MWh) | Capture factor (%) |
|---|---|---|
| ES 2023 | 61 | 83% |
| ES 2025 | 16.8 | 54% |
| IT 2026 | 97 | 86% |
Spain is the precedent. In two years the Spanish solar capture price fell from 61 to 16.8 EUR/MWh, and the capture factor from 83% to 54%. It is not only Spain: BCG puts the 2025 hit to EU solar and wind revenue at over 14 billion dollars, with negative-price hours in the most affected markets rising from around 200 in 2020 to over 500 in 2025 (Germany about 575, Spain about 570, France above 510). Italy logged zero negative hours — but only because of the floor.
Where Italy actually sits
Italy is not Spain — yet. It still has Europe's highest solar capture rate, around 86%, and a captured price near 97 EUR/MWh, because gas usually still sets the marginal price during Italian solar hours. That is the buffer, and it is real.
Two things are thinning it. First, the PUN was formally superseded on 1 January 2025; a transitory mechanism still keeps demand-side prices uniform nationwide, but supply is already priced zonally, and full zonal pricing is a question of when, not if. Second, every new GW of solar pushes more midday hours down toward the floor. The annual average still looks comfortable. The marginal midday MWh does not — and that is the one a new plant adds.
Four ways to sell the same MWh
Here is the operational point. The same physical MWh is worth strikingly different amounts depending on how you route it to market.
Gross value captured per MWh, by route to market
exposed to decay; PMG 47.5 EUR/MWh only <=1 MW and first 2 GWh
price set by profile risk, not yield; >=500 kWp, >=5 years
energy + TIP premium on shared energy; max 1 MW/unit, one primary substation
avoided variable retail cost; covered by iperammortamento even at a distance
| Route | Min (EUR/MWh) | Max (EUR/MWh) | Detail |
|---|---|---|---|
| Merchant / RID | 45 | 90 | — |
| Fixed PPA | 70 | 90 | — |
| CER + TIP | 130 | 200 | energy 70 + TIP 90 |
| Self-consumption | 150 | 220 | — |
- Merchant / RID follows the zonal price. Below 1 MW a guaranteed minimum price (PMG) of 47.5 EUR/MWh covers only the first 2 GWh; above 1 MW there is no floor at all. It is the route most exposed to decay, because it earns precisely in the hours that go to zero.
- Fixed PPA de-risks the price: roughly 70–90 EUR/MWh in the North for plants above ~500 kWp over five years or more. The level is set by profile risk, not by yield.
- CER + TIP: the energy-community premium (tariffa incentivante, DM 414/2023) pays 60–120 EUR/MWh on top of the energy value of the shared energy, for 20 years, with a Northern uplift — but only inside hard constraints (max 1 MW per incentivised unit, one cabina primaria / primary-substation area, a single soggetto referente). It moves value to where there is contemporaneous consumption. We covered the operating reality in CER after the GSE rules.
- Self-consumption (on-site and at a distance) is worth the avoided variable retail cost: roughly 150–220 EUR/MWh, against ~70–100 if the same kWh were sold. It is the route that preserves the most value, and the 2026 iperammortamento covers it even «at a distance».
The routes do not differ by a few points. The structured routes — CER and self-consumption — are worth a multiple of the commodity ones, and only merchant slides toward zero over time. A fifth, regulated route is now arriving: FER X, whose definitive decree was signed in June 2026, fixes price through a two-way contract-for-difference over 20 years (37.15 GW in total, a PV reference around 85–95 EUR/MWh, first auction in autumn 2026). It is largely utility-scale and does not replace the profile logic for anyone building around sharing or self-consumption — but it is one more way the «when» gets fixed by contract instead of left to the market.
Storage is the bridge
If the problem is timing, storage is the most direct answer.
Storage captures the hourly spread
spread ~137 EUR/MWh
| Moment | EUR/MWh |
|---|---|
| Midday | 3 |
| Evening (peak) | 140 |
Shifting a MWh from the midday trough toward the evening peak captures a spread of well over 100 EUR/MWh. That is not a speculative upside; it is what makes the merchant share of a project investable in the first place. The market is already moving: Italian batteries reached 18.8 GWh and 7.73 GW by end-March 2026, with utility-scale capacity roughly doubling in 2025 and behind-the-meter C&I expected to grow by about a third in 2026. The IREX 2026 report counts 51 billion euros invested in Italian renewables and storage over 2025. Storage is becoming part of the revenue thesis, not a line item bolted on at the end.
Where we see it go wrong
From the builder's side, the recurring mistakes are about structure, not yield:
- modelling solar at an annual average price instead of an hourly profile;
- assuming the capture factor holds while penetration keeps rising;
- choosing the route to market after sizing the plant, instead of before;
- treating storage as an option rather than as part of the revenue case;
- underpricing profile risk in PPAs, and merchant risk in the post-incentive tail.
It is a structuring problem before it is a yield problem. Which plant you build matters less than how you decide to sell what it makes — a point that sits right next to the due-diligence gaps we wrote about in bankability of Italian PV.
The «how much» is an engineering question, and it is largely solved. The «when» is a financial one, and it is where the value now lives. If you are sizing a project and want a second read on the route to market before you commit, write to us early.
Sources
Power prices and the midday zero: GME (zonal/PUN), Il Sole 24 Ore, Italia Solare. Capture-price decay: Modo Energy (Spain and Europe) and BCG, «Flexibility, Not Capacity, Will Decide Renewable Energy's Future» (2026). Merchant/PMG and the PUN-to-zonal transition: ARERA, GSE. PPA benchmarks: Pexapark and trade press. CER premium: DM MASE 414/2023, GSE, MASE FAQ. Self-consumption value: ARERA, Eurostat. Iperammortamento and FER X: L. 199/2025 and the 4 May 2026 decree, and the FER X decree (June 2026), MASE. Storage and investment: Italia Solare on Terna Gaudi data, Althesys IREX 2026. Figures are as of June 2026.