Part 2: Ireland’s Cost of Energy Delusion: Wind and Solar
Wind and Solar are cheap but not when integrated on our grid
Executive Summary (this summary written by an LLM)
This article argues that Ireland's rising electricity prices, the highest in Europe, stem from hidden costs associated with integrating variable renewable energy (VRE) sources of wind and solar, despite their low Levelised Cost of Energy (LCOE) on paper. While wind and solar appear cheap in isolation, their intermittency necessitates expensive backup from thermal (gas, coal, oil) plants, which then set the marginal price for the entire market, a backup supply that Ireland can never abolish without huge storage.
The author explains that LCOE, a metric for individual project costs, doesn't account for system-wide expenses. The key hidden costs include:
Reduced Capacity Factors of Thermal Plants: As VRE penetration increases, thermal plants are dispatched less frequently, reducing their capacity factors. To remain viable, these essential backup plants must increase their per-MWh selling price at peak times, further driving up overall billpayer costs.
Curtailment and Cannibalisation: When there's an oversupply of VRE, output is curtailed, but investors still get paid, increasing consumer burden. Furthermore, increased VRE penetration leads to "cannibalisation," where VRE sources compete with each other, reducing their capacity factors and increasing their effective LCOE.
Grid Upgrades: Integrating more VRE requires significant investment in transmission and distribution infrastructure, costs that are passed on to consumers.
Storage Costs: A 100% VRE grid is not feasible without massive, expensive energy storage solutions (LDES), for which Ireland currently lacks a commercial roadmap or sufficient financial incentives. The existing storage capacity is negligible compared to what's needed, and scaling it up would cost billions.
The article concludes that current policies promoting VRE dominance are misleading the public about potential cost reductions. It calls for a more integrated approach among Irish energy institutions, economists, and engineers to accurately quantify retail costs, challenge existing EU policy & legislative restrictions, and avoid the continued escalation of electricity prices.
Introduction
Ireland's electricity prices are the highest in Europe, placing a significant burden on households and businesses, and deterring industries like data centres from establishing a presence here. This situation raises a crucial question, if "renewables are cheap" then why won’t they lower our costs, and why has growth in them stalled?
Let's examine two key facts:
Solar and wind energy sources have the lowest Levelised Cost of Energy (LCOE) for Ireland (IRENA, 2023)
Thermal generators (gas, coal, oil) are Ireland's most expensive marginal generating units, setting the market's clearing price as the last dispatched units of energy (IEA, 2023).
If solar and/or wind (also known as Variable Renewable Energy sources, or VRE) are so cheap, and expensive gas plants dictate market prices, why then does Ireland still face sky-high electricity costs as we increase the capacity of VRE?
Perhaps the institutions mentioned in Part 1 (SEAI, ESB, Eirgrid, MAREI) can be forgiven, since the situation is somewhat paradoxical. Wind and solar are cheap. The problem is they are only cheap on their own and not when integrated without storage, onto a modern 21st century grid. Lets unpack the hidden expenses driving Ireland's energy prices, including the costs of intermittency, grid upgrades, cost of storage, and market dynamics such as cannibalisation (where wind eats wind, solar eats solar, wind eats solar!)
Levelised Cost of Energy (LCOE)
First, to understand the cost of producing a unit of electrical energy, we must introduce the metric of Levelised Cost of Energy (LCOE). Simply put, this metric represents the cost per unit of electrical energy that any generator type can produce, averaged or "levelised" over the full lifetime of the energy project.
Intuitively, a project’s LCOE increases with higher capital costs (CapEx) and higher annual operational O&M costs (OpEx). While the more units of energy that can be produced annually, the lower the LCOE. This cost of energy figure is a key metric for project-specific investors and developers of all generation types, as it enables them to establish their return on investment (ROI) and assess the financial viability of a new project in question for any price of energy obtained (i.e. for each MWh produced and sold on the market or by artificial subsidy)
Often a large financial institution or fund is checking such LCOE in a due diligence on a proposed project, which also requires financial risk analyses in tandem: LCOE analyses are where economics and engineering meet.
The LCOE metric is typically expressed in €/MWh, but for familiarity with energy costs, we will list them in euro cent/kWh here. LCOE for any project is not constant: it is geographically dependent, project-size specific, and highly reliant on interest rates and time value of money (discount rate). For example, a Solar PV project's LCOE in China might be 2.5 cent/kWh, whereas an identically-sized project in Ireland with high land prices could be 5.5 cent/kWh. Similarly, an offshore wind project in Scotland might have an LCOE of 9 cent/kWh, but a sister project off the coast of Waterford could be 13cent /kWh due to more expensive port operations here. Another example for a solar farm: a higher annual MWh yield in Wexford would result in a slightly lower LCOE, compared to if a same-sized solar project were located at the more northerly latitude of Donegal.
Therefore, the following LCOE figures can only be approximate typical values for new Irish projects, i.e. there can never be a definitive list of LCOE for a general type of generation tabulated against LCOE’s c/kWh. It is true that solar PV consistently has the lowest LCOE, while industrially heavy projects like new coal, new gas, or new nuclear having the highest for Ireland. This order is established and not debated, although it has been argued that the new technology of modular (i.e. modular = built elsewhere and shipped here) Nuclear SMR could1 be low enough to compete with offshore wind.
Here is the general list of costs of various technologies:
Solar PV ≈ 5.5 c/kWh
Onshore Wind ≈ 7.5 c/kWh
Offshore Wind ≈ 10.5 c/kWh
Gas, OCGT ≈ 13 c/kWh
Oil: ≈ 14 c/kWh
Coal ≈ 15 c/kWh
Nuclear Power ≈ 16 c/kWh
These figures demonstrate that solar and wind are the lowest-cost forms of generation: a kWh of solar energy is three times cheaper than a kWh from gas or coal, and offshore wind is expected to cost half that of conventional nuclear. This is partly due to wind and solar having no fuel costs. These figures are a snapshot in time, and it is likely that solar PV cost will drop further, and learning rates for offshore wind here mean offshore wind LCOE should drop further (get cheaper over time). For reasons that will be explain here in further detail the LCOE of the gas, coal, oil thermal plant will likely increase (get more expensive over time)
Note: This also suggests that the cheapest way to generate energy is to install solar PV on your own roof, especially since you are the investor/owner in that case. Given a grant and/or a generous feed-in tariff such an investment could lead to excellent payback and ROI.
So, could we rest assured, safe in the knowledge that Ireland's Wind and Solar (VRE) policy will result in the lowest electricity costs for Irish billpayers?! Alas, no.
There is a significant disconnect between this metric for individual tech and what billpayers actually pay. Crucially, these snapshot LCOE figures do not account for situations where the capacity factor (C.F.) of a new project is reduced due to supply from other sources, particularly if they share the same rung on the Merit Order of dispatch. The Merit Order is how Eirgrid decide what generators to dispatch to meet demand: They star dispatching the cheapest, and end with the highest cost (marginal). When there is too much wind and/or solar on our grid, greater than demand at the time, this has an effect that causes wholesale prices2 to go very low and even to €0/MWh. This has the effect of bringing down the average wholesale price, but a wholesale price that is disconnected from the reality of the retail price. When this “free wholesale electricity” scenario happens, the developer still receives the full RESS price (e.g. €100/MWh) which means the actual selling price is capped at €100/MWh, not zero. In other words ,having low or zero intraday wholesale prices could not possibly have any effect on the price downstream at the retail level3
For any generation source (VRE or firm Thermal) any reduction in C.F. means that fewer MWh are produced by the project over the project's lifetime. Any reduction of a plant’s C.F, leads to a revenue reduction for the plant in question, and effectively increases the LCOE for the project in question.
A closer-to-home example of that revenue reduction is if EirGrid or the ESB needed to switch off (or curtail) any oversupplied domestic energy exporting from your PV panels. As the investor in these panels, you would subsequently feel the pinch, as you would not receive the forecasted export tariff that was factored into your own panel’s payback period estimate. The solar energy generated from your panels in that case are wasted. With oversupply of VRE, this C.F. reduction happens to all forms of generation that are available to be dispatched on the grid.
VRE Oversupply: Unforeseen Costs
Irish households face Europe's highest electricity prices, despite the supposed cheapness of renewables at utility scale. The LCOE only informs us of the cost to the investor, not what the billpayer ultimately pays at the retail level.
Investor/Owners in VRE receive their own special feed-in tariff, known as the Renewable Energy Support Scheme (RESS), sometimes with generous CFD arrangements. This provides investors with an “adequate” fixed price for each MWh produced. This is the government paying for renewable energy, but not at wholesale prices. This price the government pays is between €90 and €120/MWh (between 9c and 12c/kWh. This contracted price with the government needs to be set at least higher than the project's LCOE to ensure viability, meaning a positive & attractive ROI, and a reasonably short payback period for the investor. Newer RESS prices are also increasingly, and controversially, index-linked, meaning they rise with Ireland's inflation, guaranteeing the investor's payback.
Interestingly, when energy is switched off due to oversupply and bottlenecking, a process known as curtailment or dispatch down occurs as discussed in Part 1. In such a bizarre case the investor/developer/owner still gets paid for any unrealised energy at the same agreed RESS fixed price/MWh.
This RESS price therefore represents the absolute minimum cost to the consumer, not the average wholesale price. The real costs to consumer arise from ancillary services, transmission cable upsizing, substation upgrade, SNSP support services, energy storage, and export capabilities required with any continued VRE grid penetration, and more importantly costs from the backup plant(s).
VRE Oversupply: Cannibilisation
Now, what happens when Ireland plans even further installed capacity of wind and solar, as discussed in Part 1? For any new offshore wind with tall 10MW units, a high capacity factor of ~45% is expected. For the LCOE calculation, it was easy to determine the yield of MWh produced over the lifetime of any offshore wind farm and even determine if this ~45% range was the correct range for offshore wind.
Traditionally, estimating the yield was a straightforward engineering calculation, as there was nothing else to compete with wind output MWh during its initial growth stage in the 2010s. At that time, during the initial boom period for wind growth, there was no need to forecast a model of other VRE acting on the wider grid. Nowadays, the situation for LCOE calculations becomes muddied, as other forms of VRE energy are already on the system and planned to be on the system at a greater capacity, overlapping a developer's own project during its lifetime. Suddenly competing directly for its market share when such a project is grid connected. This can lead to even the hint or possibility of an unforecastable and uncalculatable situation of oversupply. Will the curtailment claim mechanism be available to the developer of the wind farm, even after 10 years? This phenomenon is known in the industry as cannibalisation, as mentioned in Part 1 of this article. So we have cannibalisation and even “unforecastable cannibalisation” over the financial lifetime of the wind farm (20 years). Suddenly, the actual C.F. forecast over that lifetime is no longer fixed. Instead, it reduces over time.. See Figure 1.
The C.F. for wind was, in the past, simply based on forecasted average maximum wind speed taken from a wind rose for a particular location. It allowed the investor to be certain of the annual yield in MWh. In the case of solar, it is simply based on latitude’s gross kWh/sq.meter of total panel area. Now, it becomes very difficult for the developer to know for certain if the C.F. they will use for their investor ROI calculations are correct. This introduces too much risk, causing investors to look elsewhere—perhaps to an emerging market, as Ireland was in the 2010s, where such oversupply unknowns did not exist. No wonder growth in the industry is stalling.
True Costs of Intermittency: Ancillaries
Let's revisit our (somewhat notional) LCOE table once again:
Solar PV ≈ 5.5 c/kWh
Onshore Wind ≈ 7.5 c/kWh
Offshore Wind ≈ 10.5 c/kWh
Gas, OCGT ≈ 13 c/kWh
Oil: ≈ 14 c/kWh
Coal ≈ 15 c/kWh
Nuclear Power ≈ 16 c/kWh
While solar and wind are clearly the cheapest individual forms of generation on paper. unfortunately these are not “full system” costs of energy needed to generate enough MWh to meet demand at any time. There is a “temporal mismatch for power" to be delivered to when it is needed. When cost is put in terms of energy (c/kWh or €/MWh) like this, do they appear to be the “cheapest”. As discussed, these are costs that the investor cares for (such that profit can be determined from revenue generated from the sale of each MWh produced). A more realistic cost comparison for a generated MWh is one that includes enough storage to make that intermittent source a dispatchable one too. Comparing the cost of power, not just the cost of energy, might be an academic exercise which we could explore in a Part 3.
If enough storage is not included, then the more expensive gas generator always sets marginal cost of generation because no grid can be designed to run on 100% variable wind and solar electricity alone.
100% RES-E Not Physically Possible
Bluntly speaking, a 100% Wind and Solar grid is not possible* without the inclusion of storage, which means that the firm gas/oil/nuclear/coal plant will always set the marginal price.
*It is not even mathematically possible for a grid to operate at annual average of 75% RES-E as long as the grid SNSP upper bound is at 75%, as it is currently. Eirgrid are well aware of the challenges in this respect, and are pushing (at greater expense to consumer) to higher levels as discussed in Part 1.
The obvious remedy for this is dispatchable long duration energy storage (LDES) of course, which is not without astronomical cost either. No current commercial roadmap exists for LDES in Ireland. No financial incentive exist to sell excess VRE into storage under the current RESS. This lack of roadmap for LDES and lack of incentive to commercialise exists for all storage types:
(PHES, CAES, large arrays of lithium BESS, or Iron-air batteries)
Either the public will have to pay for these huge systems, or the RESS system itself will have to change completely (to favour LDES rather than curtailment). In the meantime, the public will have to pay a very high price until these systems are technically ready. Only one of these LDES systems is at a high TRL, and that is PHES like our very own Turlough Hill built in the 1960s.
Fossil Fuel (Thermal Plant) Required Forever
If it is true that LDES is not coming soon due to astronomical cost of low TRL4, then firm thermal plants like Moneypoint (915MW) are needed to fill in the gaps "forever,". Not even a reduction can be permitted. “Forever” in quotation marks because there will come a time, perhaps in the next decade or two, that local North Sea fossil fuels will become more scarce and therefore too expensive to run our economy competitively with. Without LDES storage, which is not coming anyway, our thermal gas burning plants will need to be fueled and available for dispatch on the grid. They must have the ability to supply the full peak GW capacity of Ireland's winter demand in common ‘dunkelflaute’ conditions. Not just in dunkelflaute weather, but on any windless night at any time of year.
The current record for our demand is 7.5 GW. Allowing for redundancy, we need a further 0.5 GW of additional standby firm capacity. So, as of 2025, we currently have, and need, 8 GW of thermal generation to deliver secure firm dispatchable supply from thermal plants such as Poolbeg, Aghada, Whitegate, and Moneypoint forever. [until the time LDES materialises into the final commercial readiness level at sufficiently low cost]
No amount of additional solar or wind installed capacity can reduce the need for any of this 8 GW “firm power safety net”. Not even scaling wind and solar 8x, 9x, 10x, or more times. Carpeting the entire country with panels and turbines still couldn’t do it.
It is a bottomless pit of temporally mismatched demand for VRE, as discussed in Part 1. With astronomical public investment we may get to 60% to 70% RES-E levels and consider that to be more “energy independent”.. but not if we are fully reliant on Russian gas piped through Europe with fully depleted North Sea gas reserves.
Furthermore, any increase in demand due to the electrification of everything (i.e. data centres, EVs, heat pumps, induction hobs) means this peak 7.5 GW demand is set to rise further, increasing a supply deficit. Once again; no new amount of wind or solar on their own can replace this future need for the full GW backup capacity required, without storage which means Irish billpayer will always pay for two systems of transmission in parallel. As of 2025, Ireland is already in a power deficit situation, needing to increase the quantity of less financially viable, expensive thermal plant, which is being rebranded in the renewable sector as "Peaker Plant." Plans are in place to build new gas plants in Ireland at an aggregated level of approximately 1.5 GW in various counties in Ireland, subject to planning permission. There is no question as to whether 1.5 GW of much cheaper new solar could achieve the same function, because it simply cannot meet continuous power demand. Instead, we will be forced to build high cost of energy new gas plants (at an ever-increasing LCOE)
Oversupplied Wind & Solar Increase Costs - for all
With that in mind, let's also apply the market cannibalisation effect, discussed previously, to this full 8 GW backup thermal generation as well – the very plants that provide us with security of supply and are the only generation type filling that increasing deficit.
The exact same "CF reduction effect" happens with this type of thermal plant as well, but to a much greater degree. The only means for such increasingly expensive thermal plant to generate revenue is by selling every MWh of energy they produce. If wind and solar are dominating and producing record amounts of MWh annually to the grid, this means that the CF for the thermal plant must also drop every year. This raises the cost of energy, or LCOE, for the gas/oil/coal plant – which is already the most expensive plant on the grid.
Every MWh that wind or solar generates is a MWh that the gas plant cannot generate. The merit order for Ireland’s supply is to always put the cheaper VRE supply on the grid first, and the gas/coal plants last ("marginal"). This strategy is not producing the desired outcome.
If an engineer in the office of any of our thermal plant were to run new cost of energy calculations for their plant, they would find a new increase in the cost of energy. They do this calculation regularly. The reason for this increase in cost is mainly due to their C.F. being reduced, irrespective of any rise in international gas prices or increase in carbon taxes. This is already observed in California, where gas plant production offset by solar, with reduced capacity factors, drives price spikes during evening peaks (CAISO, 2024). This situation is worsening there. All 8 GW of our own backup thermal plant must increase their own selling price as a result of reduced market share.
In a NetZero dream world, fully replacing our thermal plants, e.g., shutting down Moneypoint, Aghada, Whitegate, and Poolbeg, will require much more than just turbines and panels. Ireland’s grid requires billions in upgrades to manage new variable renewable output, with costs passed to consumers via network tariffs (EirGrid, 2023), as well as the squeezed MWh output of the marginal last-resort generator costs, as discussed.
Other costs associated with renewable integration are also well accepted in the wider industry. In Ireland, solar projects face grid connection fees more than double those in Spain, further inflating costs (ESRI, 2022). Interconnectors in Ireland are Public-Private Partnership (PPP) funded, with the public (taxpayer / billpayer) footing the bill for huge billion-euro amounts. Even in the case of remote onshore wind, local grid connections from remote locations, say in Donegal, need to be increased in capacity to allow for higher MW outputs to areas further south where the demand is, and these remote onshore projects require multi-million euro PPP investment. Importantly, these are ancillary, distribution wide CapEx investments, that are not factored into the individual project LCOE calculations tabulated in the article. i.e. Such ancillary costs would not appear in the CapEx columns for any new proposed wind farm in the Inishowen peninsula, but they will appear on customers' bills as network charges or be absorbed into the final 30-40 c/kWh that customers throughout the island of Ireland pay.
Storage is another major financial burden, not borne out in individual project-based LCOE calculations. Ireland’s current 1 GW of storage (750 MW battery + 250 MW Turlough Hill) is vastly insufficient to back up any of Moneypoint’s output (CRU, 2024). ESB has already invested €300,000,000 in batteries that contribute less than 0.25% of demand at any time. (Yes, a quarter of a percent of our demand for only €0.3 billion). Scaling to just 2 GWh of battery storage could cost €600–800 million at €300–400/kWh, with maintenance and replacement adding billions over decades (NREL, 2022). 2GWh is only a drop in the ocean compared to what is required for Ireland, so there are enormous costs to come with no guarantee of success. Please refer to
’s tool to estimate storage requirements here:Wind & Solar & Storage Requirement App, where he estimates the levels installed capacity of wind and solar for primary energy scenarios as well as the levels of storage required in a fully post-fossil-fuel Ireland.
Conclusion
Ireland’s pursuit of variable renewable energy dominance, while environmentally laudable, comes at a steep hidden cost to consumers. Costs that appear off sheet of any individual project financial analysis, instead borne out by the public.
Lobby groups like Wind Energy Ireland are misleading the public when they proclaim that intermittent renewables will bring down our costs; worse still, they know they are misleading. The aforementioned institutions are guilty of operating in silos. What is required is a coming together and joined-up thinking with economists and engineers in those institutions. The intermittency of wind and solar, coupled with grid upgrades, storage needs, and the market dynamics discussed here, clearly drives electricity prices higher for the reasons given, burdening consumers and severely damaging the competitiveness of our industry.
References
CAISO. (2024). Annual Market Performance Report. California Independent System Operator. Available at: [CAISO website] (Accessed: 8 June 2025).
CRU. (2024). Electricity Security of Supply Report 2024. Commission for Regulation of Utilities, Dublin.
EirGrid. (2023). All-Island Generation Capacity Statement 2023–2032. EirGrid, Dublin.
ESRI. (2022). Renewable Energy Grid Connection Costs in Ireland. Economic and Social Research Institute, Dublin.
IEA. (2023). Electricity Market Report 2023. International Energy Agency, Paris.
IRENA. (2023). Renewable Power Generation Costs in 2022. International Renewable Energy Agency, Abu Dhabi.
NREL. (2022). Cost Projections for Utility-Scale Battery Storage: 2022 Update. National Renewable Energy Laboratory, Golden, CO.
What’s Next
We invite EirGrid, ESB Networks, ESB International, SEAI, MAREI, the EPA to either invalidate the arguments outlined in this article’s Part 1 & 2, or if they unable to do so, to set up skunkworks-style meetings with EU law experts, to challenge EU policy restrictions that are on us and take any next positive policy steps. Those within these organisations with vested interests, or any lobbying group can not be included (O&G, biogas, nuclear, solar or the wind lobby)
It is also the responsibility of these organisations to accurately quantify the retail costs that the customer and industry is paying in very detailed breakdown with a comparison to fully fossil-fuel, fully nuclear or fully firm renewable countries. Any lobbyist claim of “reduced cost due to Wind & Solar” need to be justified. This article outlines a reason why electricity should expect to rise, if we remain on the current path.
(This remains to be seen for any real Nuclear SMR project)
Both Intraday market prices, and day-ahead prices
It is this lie in particular that Wind Energy Ireland perpetuate: It is seen in the FAQ of their website, heard at their conferences, and parroted in their regular news reports: Average wholesale electricity prices drop for third straight month. It is also prevalent in news media in Ireland. It is the worst kind of lie from this lobby group, because the opposite is in fact true, for the other reasons outlined in this article.
TRL = Technological Readiness Level (level of maturity or commercial readiness)