Policy
“Soon every PV system will require batteries”
Valerio Covicchio, country manager for WHES, says Italy is unlikely to be dogged by the widespread occurrence of zero and negative electricity price hours which have happened in other European nations.
By
Sergio Matalucci
Oct 22, 2025
The Italian flag. | Image: Sabrinabelle/Pixabay
pv magazine Italia has discussed Italian electricity market prospects with Valerio Covicchio, country manager for WHES. He believes the full impact of Italy’s newest solar capacity will be felt in April and there will be zero negative electricity price periods thanks to the ubiquitous presence of battery energy storage systems (BESS). Covicchio also discussed zonal pricing, the granularity of the day-ahead market, and data centers.
pv magazine: Away from the coverage of daily newspapers, everything is happening in the world of electricity – in Italy and in Europe. Granularity is increasing on the day-ahead market, from hourly to 15-minute prices. Can you explain what this means for the captured price of PV?
Valerio Covicchio: As is well known, the electricity system must at all times guarantee … consumption equal to production, under penalty of blackouts. However, the physics of the system must be coupled with commercial reality. The electron consumed by your PC or smartphone while reading this article has been bought and sold many times in different market sessions. One of these is the day-ahead market (MGP), which closes one day before the production and consumption of this electron. In the subsequent markets, a complex machine works so that the commercial agreements, the result of a prediction, can be connected to the physical reality of the instant in which the aforementioned match takes place.
The MGP, since Oct. 1, has allowed market participants to describe these production and consumption expectations with four times more granularity (from 1 h to 15 min). This involves what? Laying bare all intra-hourly events that were previously completely time-mediated: short increases in wind; sudden rainfall; absorption peaks; and, above all, clouds. All these are examples that can occur with characteristic times of less than 60 minutes and can impact the captured price, especially if the photovoltaic [system] is coupled with … storage.
Can you also explain the background to this change? What is the role of EntsoE [the European Network of Transmission System Operators for Electricity] and the various algorithms?
The transition to the four-hour electricity market is not a technical whim but the final piece of a European puzzle that began over 10 years ago. It all stems from the Cacm (capacity allocation and congestion management) regulation written by EntsoE, the network of European transmission operators – including Terna (later approved by Acer [the EU Agency for the Co-operation of Energy regulators] and the European Commission). The goal? Speaking the same “market language” across Europe, with common rules for allocating capacity and managing congestion. Today, prices and flows are no longer calculated by individual country but through a single brain: the Euphemia (Pan-European Hybrid Electricity Market Integration) algorithm of … SDAC (single day-ahead coupling) which, every day, thanks to the work of the Nemo [nominated electricity market operators] – of which GME [the Gestore dei Mercati Energetici] is the Italian representative – finds the optimal market solution. The adoption of the 15-minute MTU [market time unit] serves precisely to synchronize Italy even more with this European orchestra [for] more granularity and more physical coherence in a truly unique market. With this change, the possibility of marketing, and providing for the exchange of energy in a way that is more in line with the dynamism of the current renewable energy park and consumption habits is unlocked. [Renewables parks and consumers] will increasingly be active players in the market and not just passive spectators.
In a certain sense, the electricity market is, therefore, deterministic, easily predictable, unless there are external shocks, correct? What can be said about the volatility of captured prices? This also implies a greater focus on BESS solutions, right?
Having worked in the TSO , having produced production forecasting models (in GSE [the Gestore dei Servizi Energetici) and demand forecasting models (in Enel Energia) I cannot say that everything is “easily predictable” but I can certainly say that, to certain changes in input, certain consequences on the price of energy correspond deterministically. Specifically, I refer to the massive entry of many gigawatts of photovoltaic power in the coming months and years which will significantly impact the price at which the energy produced during peak photovoltaic hours will be remunerated. In this sense, the BESS is one of the strategic elements of the energy development of our country. To date, I have dealt with BESS precisely to help investors, IPPs [independent power producers], industrialists, and the system in the broadest sense, to overcome this techo-economic challenge.
As mentioned before, the transition to the quarter of an hour now allows BESS to extract even greater value, which we do by having integrated the MGP (quarter-of-an-hour) data into the arbitrage software from the very day this historic transition took place. This novelty has led to an increase in volatility, with price variations between two quarters of an hour of different hours of up to €99 [115] per megawatt-hour, and between quarters of an hour of the same hour of €15/MWh, on average, with peaks of €40/MWh with an increase in revenues obtainable on the quarter-hourly profile of 20% greater than on the hourly profile.
At the same time, we are in a transitional period, in the midst of the change from PUN [prezzo unico nazionale] to zonal prices. Can you explain where we are and what differentials could be created between areas? What effects could they then have on installations and batteries?
Those who produce in Italy are paid according to the area in which they produce while those who consume pay a national average price (PUN Index GME). This completely Italian anomaly was supposed to be remedied on Jan. 1, 2026 but [that planned change has] overrun and we are still waiting for Arera [the Autorità di Regolazione per Energia Reti e Ambiente] to [announce a decision]. The desired effects are to make the results of the generation park of the individual area more evident by tracing this price information back to the final consumer as well. If the change were to take place today, given the current generation parks, the areas of the CSUD [centro Sud]; South; SICI[lia]; CAL[abria]; [and] SARD[inia] would pay less for energy than The North and northern areas due to the higher ratio of renewable power to zonal demand. We are talking about values that are negligible for ordinary citizens but less negligible for industrialists. Batteries will be able to generate greater value on the spreads obtainable in The South and work well on maximizing self-consumption in The North.
You have also suggested that, with the addition of more than 5 GW of solar and the decrease in electricity demand typical of the spring months, we could find ourselves in a situation where almost every day, energy prices – and this should be mostly in The South – will steadily reach zero between 12 and 16 times, correct? Can you explain why?
The Italian system absorbs between 20 GW and 55 GW of electrical power almost all of the hours. In some extreme cases, a minimum of 17 GW and a maximum of 60 GW have been reached. The 5 GW I referred to referred only to the spring months of low demand. In fact, we should keep in mind that, to date … PV [capacity] of 41 GW has been installed. While 1 GW of nuclear power can actually produce a constant 1 GW every hour, the gigawatt[-capacity] reported for PV plants is the one produced at peak hours under standard conditions. All this is to say that when we talk about the gigawatts needed to bring energy to low or zero values, we must keep in mind what period we are in, both for the photovoltaic seasonality just explained and for the seasonality of demand. The mid-seasons, [at the] end of April and beginning of May and [the] end of September and beginning of October, are the times when there is the highest irradiation and demand ratio of the year, [and] consequently the times when zero prices occur, especially in The South where there is less industrial demand and greater irradiation. In order for zero prices to occur frequently, it would be necessary, [based] on current demand, to install more than [the] 10 GW to 15 GW already well covered by the recent results of the Fer [Fonti di Energia Rinnovabile] X, to which all the other incentive mechanisms will be added (NZIA [the Net-Zero Industry Act], CER [the Comunità Energetiche Rinnovabili], Thermal Account 3.0, various tenders, etc.
If you were to make a prediction on zero prices, what numbers do you expect, starting from 2024? In how many hours was the PUN zero in 2024 and 2025? During how many 15 minute periods are prices expected to reach zero in the next few years?
To date, since the beginning of 2025, there have been only 10 hours of zero GME PUN index [prices] on the purchase side while, on the production side, for example in The South area, we have reached 35 hours of zero zonal prices. In 2024, there were only 11. Not much, then, but be careful because it is not a linear phenomenon. By 2026, I expect that the number of events will exceed triple figures. Everything will depend on the PV-installed curve and how quickly these plants will go into production. Free energy is causing a sensation but what I think will be interesting to monitor is the fall in the value [of electricity] captured by photovoltaics. My view is that we are entering a phase of the market in which the euro invested in PV, if exposed to price risk (i.e. without PPA [power purchase agreement], CfD [contract-for-difference] in Fer X etc…) will be increasingly difficult to repay without BESS.
In your opinion, what is the probability of reaching negative prices in Italy?
I rule it out. There are no negative quantities of offers or changes in the generation fleet or, above all, increases in foreign connections such as to be able to expect phenomena similar to those observed in other European nations in the short term.
If I’m not mistaken, reading between the lines, it seems that you claim that, soon, every photovoltaic system will have a battery. Am I wrong? If not, why?
Already today, €1 invested in a photovoltaic system returns more slowly than €1 invested in a well-sized BESS that flanks it. In addition, that euro invested in BESS is protecting the euro invested in PV. In short, whether you look at the revenue or the risk profile, it makes no sense not to evaluate a BESS within a PV plant. The unprogrammability of PV can no longer hide behind the shadow of the marginal price mechanism. As it matures in terms of technology and size, its role within the markets and the electricity system must also mature. Today we have an economically profitable way to make the sun programmable, it is good for everyone (distributors, TSOs, investors and consumers) to start knowing and implementing it.
What is the role of data centers in your forecasts? Could they really significantly increase the demand for electricity in Italy? What are the consequences for photovoltaic capture prices?
The higher the electricity demand, the higher the price, with the same generation park: it’s a fact. What is often underestimated is that not all connected gigawatts have the same weight (both on the demand side and on the production side). The real effect of 1 GW on zonal prices depends on its “equivalent” use, i.e. how much that power is actually used … Asking the distributor for a few gigawatts does not automatically imply an additional demand of the same value. In addition, the impact depends on the quarter-hourly profile and geographical concentration. Currently, the vast majority of projects seem to be in The North area. These gigawatts that are making headlines today, in the wake of the AI [artificial intelligence] boom, must then, actually be grounded. There are 50 GW of data center requests ([according to] Terna’s June data), but there are also 152 GW of photovoltaic and 180 GW of wind power [requests] ([according to] Terna’s October data) which, however, do not make the headlines. How many will actually come into operation, on both sides? We will see.
What other externalities might change your assessments and predictions?
The other factors that are not purely energy[-related] are linked to political and economic macro-phenomena that many talk about and that we hope we will not have to experience. I am not an expert so I do not comment on the probability that these two phenomena may or may not occur but it is reasonable to assume that, if they were to occur, they would lead to an economic contraction and therefore also to [reduced] industrial consumption. But if something goes wrong on demand, not all prices drop linearly with it. Unfortunately, or fortunately some hours are more sensitive than others. Looking at the supply curve on the MGP, a drop in demand would lead to much lower prices captured by photovoltaics during the day and slightly lower evening prices, since gas would remain to determine the marginal price, behaving insignificantly with respect to the level of demand.
From pv magazine Italia.
ess-news.com |
