This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Seeking Consensus on the Externalized Costs of Solar Power

Schalk Cloete's picture
Research Scientist, Independent

My work on the Energy Collective is focused on the great 21st century sustainability challenge: quadrupling the size of the global economy, while reducing CO2 emissions to zero. I seek to...

  • Member since 2018
  • 1,014 items added with 421,375 views
  • Feb 20, 2018

What is meant by “externalized costs”?

Externalized costs are real costs that are not quantified within the levelized cost calculations presented in the internalized cost articles. These costs are directly or indirectly paid by various sectors of the economy in forms such as pollution-related health costs, grid integration costs of intermittent renewables, and a reduction in the free services rendered by the biosphere.

Externalized costs of solar power

While solar power has a low environmental impact, it is not quite as low as the $1/MWh calculated for wind power in the previous article. As shown in the review figure below, the environmental impact of PV is generally 3-10 times larger than wind, while CSP has a 2-5 times larger impact. We will therefore assume an externalized environmental cost of $3/MWh for PV and $2/MWh for CSP.

Key: AP = Acidification potential, EP = Eutrophication potential, GWP = Global warming potential, POCP = Photochemical ozone creation potential.

Despite the slightly higher environmental cost relative to wind, the integration costs (see previous article) of solar PV should be lower due to the lower current level of deployment (global solar power output is about a third of wind power output) and the fact that solar output generally matches better with demand.

Balancing costs (shown below) appear to be similar to wind at about €2/MWh.

Grid-related cost estimates for utility-scale solar are not yet available, but should should be similar to the €5/MWh estimate for wind.

At very low market penetrations, the profile costs for solar PV can actually be negative, i.e. produced solar power often displaces expensive electricity from peaker plants. Due to the concentrated nature of solar PV output, however, profile costs quickly rise with increasing penetration. The graph below shows that solar commands a price premium up to a market share of about 3%. Given that PV represents about 1% of global electricity production, it is likely that the average solar profile cost is still negative. Here, we will assume that it cancels out balancing and grid-related costs discussed above, yielding a total integration cost of $0/MWh.

The total solar PV externality should therefore amount to only $3/MWh. This number can also be used for distributed solar PV even though integration costs will be different. In particular, grid related costs may even be negative at very low market shares, although this is not applicable if the grid capacity is already built. However, profile costs will be higher because distributed PV generally does not use utility PV strategies like higher inverter loading ratios, tracking and westward orientation to increase market value. We will therefore assume that lower grid-related costs are cancelled out by higher profile costs for distributed solar. For perspective, the internalized costs of utility and distributed solar PV were estimated as $101/MWh and $214/MWh respectively.

As for CSP with thermal storage, integration costs may actually turn out to be negative. Relative to solar PV, there will not be any balancing costs (CSP with storage is dispatchable) and thermal storage should allow the plant to maintain negative profile costs at higher market shares. We will assume a total externalized cost of -$5/MWh. The internalized cost of CSP was previously estimated as $175/MWh. It should be fair to assume that the externalized costs for solar thermal heating applications like water heaters is zero.


If you have a number that differs significantly from the estimates given above, please add it in the comments section below together with an explanation and a reference. 

Wayne Lusvardi's picture
Wayne Lusvardi on Feb 21, 2018

I would like to leave data but perhaps not everything can be captured in data. Driving coal fueled power out of the market with RE has unintended, albeit foreseeable, consequences. Cheap coal power keeps the price of Nat Gas power low by competition. Same at times with hydropower. But despite the claims that RE is cheaper than all and the fairest of them all, this doesn’t factor in subsidies and cost shifting. How would we provide data for the rise in Nat Gas power prices (rates) due to coal being displaced from the market? And what about the unintended consequences of shutting down Nuke and Coal power in California replaced by RE? Stranded assets of two mega billion dollar Nuke plants. The rewiring of the grid because those Nuke plants are no longer voltage hubs. And the shut down of Nuke plants has led to a rise in C02. This has put more pressure on Nat Gas to back up RE leading to such incidents as the Aliso Canyon gas leak resulting in spillage of huge amounts of C02. And even perhaps the San Bruno gas pipeline explosion can be tied to the shift to RE and accompanying greater stress on nat gas facilities. So arguably California has not reduced C02 at all even with its shift to RE because nuclear and Nat Gas facilities became stressed beyond design capacities. Of course, these incidents are perceived y those in power as more justification for shutting down any competition to RE. How do we data-fy perceptions?

Schalk Cloete's picture
Schalk Cloete on Feb 21, 2018

DATA: Current externalized costs of solar PV and solar thermal: 3 and -5 $/MWh respectively according to the estimations in the article.

Willem Post's picture
Willem Post on Feb 21, 2018


The below economic is for an ACTUAL, 2000 kW, filed-mounted solar system in Vermont.

The economics shows solar is nowhere near competitive with fossil, hydro and nuclear.

The owners receive 13.036 c/kWh for the production, as well as federal and state ITCs, rapid write offs, low-cost loans, etc. The project has an IRR = 9%/y

PV solar Electricity Cost During 25 Years of Operation

Large-scale solar projects usually are financially structured in three phases:

Phase 1, years 1 – 6, the subsidy and write off phase; short term loan being paid off.
Phase 2, years 7 – 18, remaining subsidy phase; long-term loan being paid off.
Phase 3, years 19 – 25, remaining subsidy phase; loans paid off

Phase 1, year 1 – 6: Dividing the “Subsidy costs” and the “Excess paid above Ne England midday wholesale prices” by the 6-y production yields 14.0 c/kWh and 7.0 c/kWh, respectively, and adding the 6.0 c/kWh for NE midday wholesale, yields a total production cost of 27.1 c/kWh, during the first 6 years of operation.

Phase 2, year 7 – 18: The remaining subsidy is the “Excess paid above NE midday wholesale prices” of about 7 c/kWh. Adding other costs, such as paying off long-term loans, the production cost decreases to about 9.8 c/kWh

Phase 3, year 19 – 25: The remaining subsidy is the “Excess paid above NE midday wholesale prices” of about 7 c/kWh. Adding other costs, the production cost decreases to about 8.2 c/kWh.

The weighted average price of all 3 phases is about 13.036 c/kWh for this sample SO project.

This price is similar to the average price of the 4 auctioned Standard Offer solar projects. See table 3 and URL.

NOTE: The 27.1 c/kWh is less in sunnier areas of the US, such as Texas and the Great Plains.

The above EXCLUDES the costs of EXTERNALITIES, such as:

– Any peaking, filling-in and balancing performed by the other generators
– Any battery systems to stabilize grids with many solar systems.
– Any measures to deal with DUCK curves
– Any grid expansions and augmentations to connect distributed solar systems

Nathan Wilson's picture
Nathan Wilson on Feb 22, 2018

What about subsidies? In the article on internal costs, it said, “...these costs typically consist of capital costs, financing costs, operation and maintenance costs, and exploration costs.” So where do subsidies fit?

We are constantly being told that renewables are often competitive without subsidies; at the same time, subsidies are more generous than ever in the US, with a 30% investment tax credit (ITC), plus accelerated depreciation that allows owners to write-off the entire plant over 5 years, with half in the first year (before the US tax rate cut, this would have refunded another 30% of the plant value to the owner).

Meanwhile advocates claim the ITC is being phased out for wind, starting in 2017. In fact, the rules allow any project which was started by 2016 (however slowly) to receive the full 30% ITC as long as construction completes by 2020, this includes the addition of new and un-related project phases.

The 30% ITC for US solar projects is for any project that starts construction before the end of 2019, with completion before the end of 2023.

The 5 year accelerated depreciation for solar and wind never expires, but the 1-year partial bonus depreciation was 50% for 2017, 40% for 2018, 30% for 2019, and zero thereafter.

Of course when we state cost without subsidies, this does not matter, but so often we hear only the power purchases agreement price, which obviously includes not ony the Federal subsidies listed above, but also state incentives and insulation from the same market pressure that has hurt many nuclear plants.

Willem Post's picture
Willem Post on Feb 22, 2018

Hi Nathan,

This article examines in detail the effects of subsidies on c/kWh in New England.

Competitively bid, large-scale, field-mounted solar costs about 13.5 c/kWh in New England, HEAVILY SUBSIDIZED.

Excluded Costs: The above EXCLUDES the costs of EXTERNALITIES, such as:

– Any peaking, filling-in and balancing performed by the other generators
– Any battery systems to stabilize distribution grids with many solar systems.
– Any measures to deal with DUCK curves, such as grid-scale storage and demand management
– Any grid expansions and augmentations to connect distributed solar systems

Those costs, as c/kWh, are not easy to quantify, and as a result they are charged to ratepayers via rate schedules, and to taxpayers. Ultimately, they are reflected in the increased costs of goods and services, which usually acts as a headwind to economic growth. There is no free lunch.

Schalk Cloete's picture
Schalk Cloete on Feb 23, 2018

Yes, one could argue that subsidies impose a significant externalized cost on the economy via the lasting effects of economically suboptimal capital deployment. But the benefit of subsidies is of course the accelerated development and cost reductions of the subsidized technologies. Personally, I think the cost started outweighing the benefit already in 2013 when solar PV truly established a global value chain, but this point is highly contentious and difficult to quantify, so I ignored it.

It is true though that the ability of wind and solar to attract and keep their strong technology-forcing policies is quite remarkable. The ideological attractiveness of these technologies truly is their biggest asset.

I’ve never fully understood the impact of accelerated depreciation on overall project economics. As I understand it, accelerated depreciation gives a tax saving in the initial years of project operation, but this saving is cancelled out by larger taxes in later years when no more depreciation is possible. The benefit is therefore dependent on the discount rate employed and may not be so large in developed economies where the discount rate is typically quite small. Am I understanding it correctly?

Also, what is the depreciation pathway of wind/solar compared to a conventional power plant in the US?

Willem Post's picture
Willem Post on Feb 23, 2018

Hi Schalk,

Early money is always better than late money.

The Standard Offer projects are structured to provide 9%/y internal rate of return, which is identical to an electrical utility in Vermont, and quite generous compared to long term interest rates.

I suggest you read my article.
After I completed it, the results were an eye opener for me.

During the first 6 years of a solar project, the economic cost of solar is 27.1 c/kWh (using only 3 of the 6 subsidies to simplify the analysis).

Because of rapid solar build-outs, a substantial part of the installed solar capacity, MW, is less than 6 years old, and therefore in 27.1 c/kWh mode, in New England.

The same is true for wind projects.

This acts as a major drag on economic growth.

All that is separate from the list of externalities at the bottom of my earlier comment.

For example, to bring wind from the Panhandle to east Texas, $7 billion of transmission was built. That entire cost was “socialized”, i.e., charged to mostly residential bills.

Schalk Cloete's picture
Schalk Cloete on Feb 28, 2018

At $3000/kW, the solar farm used in your example appears quite expensive. The capacity factor of 16% is also rather low. The numbers in your article are therefore not representative of the broader solar PV industry in the US. Current nation-wide average utility PV costs and capacity factors are about $2000/kW and 25%, returning about half the levelized cost in your article.

In any case, the higher cost of PV and wind are accounted under their internalized cost in this series of articles. I’ll soon complete the final externalized cost article, followed by a summary article adding internalized and externalized costs for all technologies to show the full picture.

As stated in this article, the externality related to solar PV intermittency is small at present because of the very low current market share. I fully agree that this cost will grow substantially in the future, but this article is about current externalized costs.

Schalk Cloete's picture
Thank Schalk for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Get Published - Build a Following

The Energy Central Power Industry Network® is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »