(This was a research note prepared as part of the preparation for an article written by George Monbiot, Mark Lynas and me and published on the Guardian web site. Full article here. )

The revival of the nuclear industry in Europe started in the frozen winter months of 2005 on an island off the Finnish coast. Alongside two existing nuclear plants, the ground was prepared for a new power station to be built to the latest design. Completion was scheduled for about four years later in the first half of 2009,

Construction went wrong from the start. The numbers of workers on the site ballooned, peaking at almost five thousand people from all around northern Europe. As time passed, the completion date was pushed back. In early 2009, the plant was supposed to start in 2012. By 2010, the date was late 2013. The press releases carrying details of delays came out with predictable regularity. Today, the latest estimate is that the power station will begin generating electricity at the end of 2018, almost fourteen years from the first shovels in the ground. Ten years after the start, a project that was meant to take about fifty months is still forty months away from completion.

The owner is suing the contractor for nearly €3bn and the contractor counter-claims for even more. Costs are probably quadruple what was expected when construction started. This new nuclear power station has nearly bankrupted Areva, the French nuclear construction company running the site.

There’s one other nuclear construction project going on in Europe and it uses exactly the same design as in Finland. This time it’s on the Normandy coast and is under the direct control of EdF, the French company behind the proposed nuclear plant at Hinkley Point in Somerset. Work was begun at the Normandy power station in late 2007 with completion promised at the start of 2012. Press releases with wording eerily similar to the Finnish texts come out almost as frequently. Each time EdF claims rapid progress on the site while pushing back the finish date a few more months. Full power from the plant is now expected in 2019, twelve years after the start of the project. Costs are over three times what was predicted in 2007.

EdF is the only company in Europe that still believes that the design it is using in Normandy - and prospectively at Hinkley Point - can provide electricity at a reasonable price. Other major international businesses have quietly slunk away from nuclear. Siemens withdrew from the Finnish project as soon as it could, the huge Italian utility ENEL withdrew from the Normandy plant in 2012 (as well as from a commitment to the other five reactors it had intended to build in partnership with EdF) and British Gas owner Centrica wisely withdrew from the Hinkley Point consortium in early 2013.

EdF soldiers on. It now says that Hinkley will start sometime ‘after 2023’. Given that the Finnish and Normandy plants will take at least twelve and fourteen years respectively the lack of specificity is understandable.

Why has this new design of nuclear power station proved so difficult to build? Tony Roulstone, who runs the Master’s programme in Nuclear Engineering at Cambridge University, gave his view in a public lecture late last year. This type of new power station was ‘unconstructable’, he said, adding a comment that the Hinkley Point design was like ‘building a cathedral within a cathedral’. Huge numbers of inexperienced workers were crowded into a limited area, each unsure of exactly what they were doing or how it fitted into the master plan. The power station is over-complex and construction is unmanageable, he concluded.

Just for interest, we looked at exactly how long a cathedral might take to construct. Salisbury Cathedral, one of the biggest, took forty six years to complete in the thirteenth century. Hinkley Point probably isn’t going to take as long as this, but the difference is less than you might imagine.

Despite the evidence from other countries and the views of an increasing number of experts like Tony Roulstone, the government ploughs on with its unquestioning support for the EdF plan. And unfortunately, the main competing design also vying for permission to construct nuclear plants in the UK is also experiencing huge construction problems in China and the US. Electricity consumers in the state of Georgia have just had another 6% added to their bills to pay for the delays in the completion of the power station at Vogtle. As in Finland, the contractors and the owner are scrapping over who is to blame for the overruns.

All this might be acceptable if this generation of new nuclear plants was eventually going to reduce the costs of the transition to a fossil-free future. The chances of this look remote in the extreme. Hinkley will be paid at least double the current wholesale price of electricity if it is ever completed. This means it will receive a subsidy from UK electricity bill payers of about £1.1bn a year, more than the total cost of the Feed-In Tariffs for PV and wind that the government recently curtailed because of a shortage of money. This subsidy will continue for thirty five years, far longer than the support for any other technology. The UK is saddling itself with a billion pound burden each year for more than a generation. If the project takes until 2025 to finish, a baby borne today will be forty five years old when the subsidy ceases.

Against our pessimism the government argues that Hinkley Point is needed because of its ability to deliver large amounts of power reliably every hour of the day. Other technologies such as PV and wind cannot offer this security. Today, that conclusion is correct. But with sufficient R&D and government encouragement, by the time Hinkley is ready the problems of storage of energy will be solved.

Other countries – less bewitched by the allure of nuclear – are making fast progress on the road to energy systems that can cope well with daily, and seasonal, swings in power production from renewables. And in many parts of the world, solar and wind are now costing little more than half what the UK government is promising EdF for its risky Somerset plan. Solar, in particular, is now priced at less than a quarter of five years ago and the cost reductions are continuing. Construction is 50 times faster; a large solar farm takes 12 weeks to build compared to the 12 years for the Normandy reactor.

UK Government R&D support for all alternative energy technologies is probably running at about £250m a year, a quarter of what will be spent on eventually subsidising Hinkley Point. The rational choice today is for the UK to back away from this generation of nuclear power and invest properly both in next generation of atomic energy and in renewable energy technologies that can shift the UK rapidly to a green future.