The Energy Collective Group

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.

9,988 Members


Southeast Europe Needs More Nuclear Power to Head Off Energy Crisis

Cernavoda nuclear plant in Romania.

Southeast Europe is headed for an energy crisis. The region has an energy infrastructure that is unreliable, inefficient, and unsustainable, while at the same time it is faced with the need to reduce dependence on external sources and conform to EU climate and air quality regulations. The best way out, argues Tim Yeo, Chairman of the New Nuclear Watch Institute, is to invest in new nuclear capacity.

The energy system in Southeast Europe – especially in the Balkan peninsula – has been suffering from chronic underinvestment since the dissolution of the Eastern Bloc and the breakup of the Yugoslav state. Now, a combination of geopolitical and environmental factors has made it extremely urgent to take measures to modernise the sector.

Tensions between Europe and Russia – stemming from the annexation of Crimea and the conflict in Donbass – have magnified concerns about energy security and dependence. Moreover, the European Union, post-Paris Agreement, has adopted a raft of rigorous, binding environmental legislation, designed to reduce carbon emissions and enhance sustainability.

A reliance on imported gas represents an ultimate reliance on a limited group of nations

As such, the reform and the renewal of the energy sector has taken on a fresh impetus. The governments of the region have become acutely aware that doing so is crucial to safeguarding future development – while avoiding the potential uproar that higher energy prices could provoke – while Brussels, with a mind to the imminent EU accession of countries such as Serbia and Montenegro, is keen to allay their dependence on imported, carbon-intensive power generation.

Coal phase-out

An upcoming publication by The New Nuclear Watch Institute (NNWI) – The Electricity Market of Southeast Europe: The Impact of New Trends and Policies – which will be presented next month in Sofia, makes clear the necessity and urgency of radical action. Should governments fail to sufficiently expand generation capacity in the immediate future, Southeast Europe (Albania, Bosnia and Herzegovina, Bulgaria, Kosovo, Greece, Hungary, Macedonia FYR, Montenegro, Romania, and Serbia) could become a net importer of electricity as soon as the early-2030s.

If EU accession is in part contingent upon the decommission of the region’s oldest coal-fired thermal plants – over 11 GW of installed coal-fired capacity was commissioned before 1980 – this could occur as early as 2027.

Reform to Phase 4 of the EU Emissions Trading Scheme (ETS) is another factor that will quicken the arrival of electricity deficits. The reinvigoration of the market stabilisation reserve (MSR) in conjunction with an acceleration in the rate at which the total number of allowances is reduced will cause the price of a European Emission Allowance (EUA) to rise above its current depressed level.

The undeniable precariousness of the situation is only worsened when upcoming demand-side disruptions – such as the electric vehicle revolution and the widespread adoption of digital technologies (such as Blockchain) are factored in.

In the absence of an enforced coal phase-out and these demand-increasing disruptions, NNWI forecast an electricity deficit of 40 TWh in 2035 and, upon their growth, the deficit widens to 132 TWh, slightly in excess of 40% of current electricity use.

To import this deficit would be expensive. The EU Reference Scenario 2016 (EURS16) indicates that the EU-wide electricity price could reach €165 per MWh (2013 prices) by 2035, substantially in excess of what nations in the regions are forecast to pay (€148 and €144 in Romania and Bulgaria in 2040 respectively).

Diversifying gas imports

Thus far, the policy response from governments and Brussels has been to accelerate plans to diversify natural gas supplies and to invest heavily in the related infrastructure (as embodied in the aims of the Central and South Eastern Europe Energy Connectivity (CESEC) Working Group).

Such efforts are at odds with headline EU energy strategy on two fronts.

The integration cost – the sum of grid, balancing, and interaction costs – of renewable energy increases as its share of final electricity generation expands

Firstly,  a reliance on the imported fuel represents an ultimate reliance on a limited group of nations. Norway and Russia supplied 57% of the EU-28 gas imports in 2015 (UN Comtrade) and, taken as an aggregate, the group’s five largest trade partners – the two mentioned thus far in addition to Algeria, Qatar, and Libya – accounted for 90% of total imports in the same year (UN Comtrade). This then seemingly undermines the EU’s commitment to reducing its dependence on external, highly concentrated supplier markets.

Secondly, to meet the total deficit forecast by the NNWI with natural gas would entail emissions of 1.3 billion tonnes of CO2 equivalent; in addition to the financial cost of such emissions – €40 billion at President Macron’s proposed €30 price per tonne – such a policy would undermine the EU’s commitment to sustainable consumption.

Renewable energy and integration cost

As the EURS16 illustrates, the contribution of renewable energy sources to system generation capacity will continue to grow – impressively so – up to 2040. Net renewable capacity (including hydropower) in southeast Europe is forecast to increase by 17,115 MW between 2020 and 2040, with 57% of the growth (9,774 MW) provided by additions to wind capacity alone.

In 2040, more than half (61%) of installed capacity and 44% of actual electricity generation in southeast Europe will be powered by renewable energy sources. The EU’s Energy roadmap 2050 projects that the latter figure will rise to 55% by 2050. This is to be applauded, the deployment of renewable energy sources in the EU post-2005 resulted in a 7% decrease in greenhouse gas emissions than would otherwise have been the case, as noted by the European Environment Agency (EEA).

An incremental expansion of renewable energy sources – beyond that described above – would be costly.

Certainly, the headline cost – the levelised cost of electricity (LCOE) – of renewable energy source generation is falling (IRENA) and has approached the generation cost of fossil fuel-fired power generation cost in some regions of the world.

However, research by the Potsdam Institute of Climate Impact Research (PICIR) has shown that the integration cost – the sum of grid, balancing, and interaction costs – of renewable energy increases as its share of final electricity generation expands.

For example, an increase in the final electricity share held by wind power from 10% to 30% leads to an increase in system cost – the sum of generation and integration costs – of €20 per MWh.

As the report concludes, “System LCOE and integration costs significantly increase with [variable renewable energy] penetration and can thus become an economic barrier to further deployment of wind and solar power.”

Moreover, the International Hydropower Association (IHA) makes clear that Southeast Europe is approaching its technical capacity of hydropower. Several nations in the region generate a significant amount of electricity from the resource – Albania, for instance, produces almost all of its electricity in this manner – but potential expansion is restrained by availability as well as environmental concerns.

New nuclear capacity

Therefore, it seems clear that significant investment in new nuclear capacity must be seriously considered.

At present, nuclear capacity in Southeast Europe slightly exceeds 5,000 MW, with Bulgaria, Hungary, and Romania accounting for 36%, 37%, and 27% of regional capacity respectively (EURS16). In the region, this capacity currently accounts for 13% of total electricity generation.

Capacity is forecast to rise to 8,440 MW by 2040 (EURS16), an increase of 3,164 MW, split between Romania (1,414 MW) and Hungary (1,732 MW).

Romania plans to add two new 720 MW reactors to the nuclear power plant in Cernavoda (to be operational in the early 2020s). In Hungary, two 1,200 MW units are to be added to the Paks plant and connected to the grid by 2030 (net capacity expansion is only 1,732 MW due to the planned retirement of existing Paks units).

Nuclear power offers Southeast Europe a means to preserve the security of its electricity supply, to meet its environmental commitments, and to support the future social and economic development of the region

The expansion of existing site (‘brownfield’) capacity – due to its lower cost relative to greenfield development –  appears to be the most economical solution for nations with existing nuclear facilities.

In Bulgaria, both the expansion of Kozloduy (Unit 7) and the proposed plant at Belene present such an opportunity.

Today, the status of the two projects is uncertain (neither is yet under construction), due to concerns over their financial structure and wider industry concerns (the bankruptcy of Westinghouse in particular).

The project at Belene is not strictly brownfield – Sofia suspended the new build project back in 2012 – but the site preparation works were underway and the equipment is already owned by the National Electric Company (NEK) as a result of a compensation agreement with Russia’s Rosatom. To restart the project would mean making use of it as an investment as opposed to treating it as a sunk cost.

An expansion of nuclear capacity of 4.6 GW between 2020 and 2030 – implying the timely completion of all new build units in Hungary and Romania, as reflected in the EURS16, and a commitment to new build of at least 1.5 GW in Bulgaria to be completed by 2027 – would prevent an electricity shortfall in southeast Europe before 2030.

An overall expansion of 10 GW by 2040 would meet 50% of the deficit projected by the NNWI.


Nuclear power offers Southeast Europe a means to preserve the security of its electricity supply, to meet its environmental commitments, and to support the future social and economic development of the region.

The NNWI report makes clear how urgent the issue of electricity generation is and so, acknowledging the time between the planning stage and the completion of a nuclear project, the time to commit politically and financially to new nuclear builds is now.

Editor’s Note

The New Nuclear Watch Institute (NNWI), the research arm of New Nuclear Watch Europe (NNWE),  is a think-tank focused purely on the international development of nuclear energy.

Original Post

Tim Yeo's picture

Thank Tim 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.


Bob Meinetz's picture
Bob Meinetz on March 26, 2018

Tim, thanks for your perspective on the extremely important and under-reported topic.

In a political/power crisis, it’s conceivable Ukraine would consider restarting Chernobyl Units 2 & 3, which operated well into the 1990s and haven’t yet been decommissioned. Falling back on these obsolete and unsafe RBMK-1000 units might be a wake-up call to Germany and other EU countries: ideological antinuclear policies can have unintended nuclear consequences.

David Nickerson's picture
David Nickerson on March 26, 2018

Back in the early 1990s post Chernobyl, we (along with Siemens Westinghouse) offered to deliver up to 1000 MW in combined cycle gas turbine power barges with the cooperation of the US government (Nunn Luger Act) into Ukraine. Unfortunately we were unable to conclude a project. This offer was based on leaving only the VVER reactors in operation.

Bas Gresnigt's picture
Bas Gresnigt on March 27, 2018

It is faster and cheaper and will become even far more cheaper in next decades to create electricity production via more wind+solar+storage (incl. PtG+storage in deep earth caverns). It’s the path Germany and other EU countries now follow.

Even the old EU reference scenario indicates already that renewable will produce >50% of the electricity in the EU after ~2040. As that will occur mainly with wind & solar, the other 50% have to be flexible as wind & solar will then produce >100% during substantial part of the time (~25%). Market prices will then be 2x) than wind+solar+storage (incl. PtG-S-GtP), it will then become even far more expensive…

Hence nuclear electricity in these countries will then become a drain to their economies via its extreme high costs. It becomes a burden for them in the competitive EU market.

Bob Meinetz's picture
Bob Meinetz on March 27, 2018

Bas, telling that the word “will” appears seven times in your brief forecast. Wind “will” do this, solar “will” do that.

Noteworthy of any discussion on renewables, since the 1970s, is the absence of the verb “to have” in present perfect tense: biomass “has” done this, solar and wind “have” done that. Here, it appears zero times.

You’ll have to forgive a veteran environmentalist if I can’t help but point out the countless promises and visions of glory for renewables which have gone bust. All of them, in fact, making it hard to believe any of the hype coming from that determined group of individuals who believe they can’t take “no” for an answer.

“Believe they can’t”, but will. There’s too much at stake.

Herr U's picture
Herr U on March 27, 2018

Care to cite what the projected costs are?

Here are the values to beat – the thing we actually care about the most about is the cost per TWh for 30 years in relation to the capacity (W).

(Do pay attention to when usd, eur, GWe, kW, MW, kWh and MWh are used)

To take examples from current nuclear projects we have stuff like Barakah (UAE) for 20bn USD (5.6GWe, about 39-44TWh/year), the standard Rosatom offering of 2x1200VVER for 10bn EUR (2.3GWe, about 16-18TWh/year), and even the atrociously mismanaged Olkiluoto 3 (1.6GWe, about 11-12TWh/year) for 9bn EUR. Those projects pan out in the range of about 3600$/kW, 4200€/kW, 5600€/kW.

Do note that modern nuclear had an expected capacity factor of at least 80% and a lifetime of between 60 and 120 years (barring politics)

That capacity factor and lifetime matters quite a bit, just for fun let’s assume the plants only run for 30 years at 80% cf average that means the plants themselves will land at about 17.1$/MWh, 20€/MWh, 27€/MWh respective.
At 90% cf it will land at about 15.2$/MWh, 18.6€/MWh, 25€/MWh
At 60 years those are cut in half.
(Probably worth pointing out that O&M and fuel together makes up about half of the running cost of nuclear – so add about 50% for that, and add up to 25% more if the country in question has questionable politicians, and add 25% if the money for the project was borrowed. Upgrades are a part of O&M)).

Just to give a sense of range that puts the OL3 in at between 12.5 and 55€/MWh, and properly managed projects in at between 7.6 and 40$/MWh
While nuclear might be expensive to build the power and electricity output from it is insanely high and it averages out pretty nicely over time.
(In cold countries (except sweden due to politics) one also could use the waste heat for district heating and thus increase the value of the services further)

(And in case you want to point only at future tech I should point out that the cost an EPR once they learn how to build it is expected at about 3bn eur per unit, which will put it at 1900€/kW, 9€/MWh, 8.4€/MWh, and 4.2€/MWh (60yrs@90%). So this is what future non-nuclear has to beat if you are going to claim that nuclear is expensive over time.
Probably should point out that China and South Korea aldready builds some units at about 2000$/kW (and south korea has salaries comparable to the western world) so the figures passes sanity checks as well))

And regarding speed – well – most nuclear is built in about 40 to 80 months, to take the latest completed reactor (Barakah-1 UAE) it was officially designated complete today and they started to build it in 2012-Jul-19 (unit 2 is expected to come online later this year, unit 3 in 2019 and 4 in 2020. That puts it in at about 8 years to build 40TWh of generation, six years for first reactor). (Do note the staggering of the builds, which means that from now and until 2020 UAE will bring about 10TWh/year online per year of nuclear power (until the complete 40TWh is reached)). So if you consider it fast or not depends on how you count really.
(If we where to cherrypick I would have picked the KK-6 (ABWR, 1315MWe) since it was built in 37.5months from first concrete to first critical (KK-7 took 40 months, it started up seven months after KK-6 (46months from first concrete for first reactor to first criticality of the second second reactor, that is 2.6GWe in total – so the GEH and Toshiba team could basically build all the needed 40TWh in about 58 months (just shy of six years), but in fairness add about a year for site preparation))
What really kills nuclear is red tape.

Bas Gresnigt's picture
Bas Gresnigt on March 28, 2018

Bob, may be you didn’t notice but the price decrease of wind (~5%/a) and solar (8%/a) is widely predicted to continue during next decade while those (unsubsidized) prices are at many places already at ~3cnt/KWh. No promises but contracted.

For solar those price levels are already shown at sunny regions such as at the Arabic peninsula, Chili, Mexico ( US$20.57/MWh), etc. Even SW-USA is coming near.

For wind those price levels are shown at the coast of Morocco, the plains in USA and for new wind farms in the North Sea (Germany contracted 1200MW, Netherlands recently 700MW to be operational in 2022. All without subsidy while Dutch govt stated to expect av. whole sale prices in 2035 (=~halfway the 30 years lifespan) are ~€28/MWh and those were ~€31/MWh in 2017. Note that these wind farms are expected to have CF’s of more than 55%, thanks to their high and big (10-15MW) wind turbines together with smart management controls.

With the continued price decreases such low price levels will spread to more and more regions.

Bas Gresnigt's picture
Bas Gresnigt on March 28, 2018

No nuclear plant became older than 50 years.
Your predictions of the lifespan of new nuclear is unrealistic optimistic considering the past and especial the future where nuclear will be faced with even lower whole sale prices.

Your cost predictions of new nuclear are invalidated by the guaranteed prices which are needed to cover the costs of Hinkley C (the fifth and sixth EPR!):
You should add to that price the grants & subsidies due to:
– loan guarantees of UK govt. Value £13/MWh.
– liability limitation subsidies (accident and waste). Value £10/MWh .
So the costs are £123/MWh or $173/MWh to be inflation corrected until 2062.

Despite those high tariffs EDF (the owner) needed and got additional major subsidies (billions) from the French state, and the participation of Chinese nuclear…
So it’s clear that they don’t expect that those high tariffs and subsidies deliver high profits.

Though less relevant: Your construction periods are also too optimistic.
Hinkley C is now scheduled to take >11years.
Barakah-1 UAE is still not operational.
Referring to short construction periods before 2000 is not relevant as those NPP’s were unsafe. Due to the experience with a.o. Chernobyl, Fukushima and 9/11 (aircraft rule) new NPP’s have to be far more safe which increases construction periods and costs greatly.
*) The guaranteed price is £92.50/MWh in 2012 £’s to be inflation corrected which delivers now £100/MWh.
The price correction for inflation will continue until 35years after the start of the NPP which is scheduled now at 2027.

Herr U's picture
Herr U on March 29, 2018

It is true that currently no nuclear power reactor are 50 years (a few research reactors on the other hand (NRU did 60 years (1957-2018))) however there a couple that started operating in 1969 (so they will turn 50 next year) (and a couple of dozen came online between 1970 and 1974 are still operating).
(Considering that Calder Hall started in 1956 that is pretty impressive).

Those “even lower whole sale prices”, care to post a link? or an estimate of just what those actually will be?

You are invalidating current data in Finland, UAE and a couple of active tenders from Russia based on a project in UK? By that logic dams in holland doesn’t work since the three gorges in china had a catastrophic failure (see just how insane that method is?).

But HPC in and of itself is mainly a political project that just had time run away from it, but even so, that project is based on giving an 8% (iirc) profit for investors over 25 years.
For the loan guarantees, without adding in what timespan those covers the value in MWh is completly useless, please supply timespan.
The “accident and waste” actually is included in the “questionable polticians” quote I mentioned (quite frankly, I took the percentage from last year’s financial report from Forsmark, just added upp all “other”-percentages).
Also, the strike prices in UK has almost nothing to do with actual costs (regardless of technology, take a look at what the strike price for seabased windpower in the UK was the same year as the HPC contract was negotiated), referring to those for costs to build a project is a bit like referring to taxi prices over the lifetime of a car to try to estimate the cost of building the car.
A quick search seems to place the cost to build HPC at about £20bn and the strike price at £50bn.

But if you so much insist on the guaranteed prices for projects should we then use the guaranteed prices* for OL-3 (EPR) or Hanhikivi-1 (VVER-1200)?
(Probably should point out that Hanhikivi-1/Fennovoima aims at “below 50€/MWh”, and that cost is expected to fall over time plant life).

My constructions periods are not optimistic, they are actual data (look it up in IAEA PRIS for yourself in case you don’t belive me). And do try to read what I actually write rather than what you think I write. I did specify what points in the project I use the date for. For instance Barakah-I is _completed_ but not operational (the holdup is that FANR needs to issue a nuclear operator license for them to be allowed to load fuel in the reactors). (The precise meaning of the different stages of nuclear builds matter quite a bit – if I wanted to try to pawn it off as for when most other industries would consider a build to be done I would have taken the date for the start of the cold hydrostatic testing)

The new rules, well, doesn’t really matter that much for some designs, so let’s have a bit fun and take a project that was completed back in 2003, Qinshan Phase-III. That project was 2xCANDU 6. It took them 80 monthts from _contract signed_ to _second unit in commercial operation_. And those units conform to current western regulation as well (well, in the countries based on actual knowledge of nuclear power).
But even beyond that VVER-1200 (all variations of it) actually conform to the post-fukushima requirements, US ABWR is found to conform to the new standards as well (look it up at US NRC, they did the assesment a few years ago), also the KK-6 and KK-7 units currently are being upgraded to conform to japan’s new standards (expected restart in 2020) (due to how the ABWR is designed that doesn’t really add that much to the cost or buildtime).

* there are no strike prices in Finland, which means that if you base costs of build on strike price those units are at the very least free.

Susan Roaf's picture
Susan Roaf on March 30, 2018

The great Fire of London started early on the 12th of September the east wind drove the flames before it – burning much of London and spreading smoke as far as Oxford. In 2007 a nuclear power station site was still proposed at Didcot, central England, on the Thames that dried to a ‘trickle’ in 1666. Even when in ‘down mode’ a nuclear power station requires around 3MW of cooling to keep the fuel rods stable. In August 2011 Dominique Bestion, Research Director of the French Atomic Energy and Alternative Energies Commission told an international audience that he foresaw that no new inland nuclear power stations being ever built again because of climate change. He cited the experience of France when they increasingly have to put their nuclear plants into sleep mode on rivers like the Loire. In the summer of 2003, when 15,000 people died in the heatwave in France, three of the Loire plants were shut down and blackouts ensued. In response France turned to installing over 200MW of distributed photovoltaics to meet that extreme summer peak demands when such inland nuclear plants fail to generate and investing heavily in dispatchable PV and battery supported micro-grids like the ‘Nice Grid’ to shave the killer summer peaks. Nuclear energy in a rapidly warming climate with more extreme weather is unsafe whether inland or by the sea but proposing it for a slew of landlocked countries in S.E Europe, as Tim Yeo did, show half baked, dangerous thinking that will lead inevitable to a fully cooked region when rivers do dry up and grids fail making nuclear plants un-coolable.

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 »