Shades of Nuclear Green

In a not particularly sunny December day, I find myself by the gas station. As well as suffering from existential anxiety, triggered by calculating how many times I have consumed my weight in petrol this year, I am also surprised at the current gas prices. The problem is solving itself I guess?

The long term, Dyson sphere powered view is clear. What is less clear is how we can get there.

I must admit, I find something romantic about powering civilisation by splitting the atom, like a “coming of age” for humanity, the Victorian dream of taming nature coming to fruition. The advantages are clear: a continuous, large energy output, as well as a very small carbon footprint during operation.

With the goal of decarbonising our energy grid being a lot more present in my mind, I am now however less certain of the proportion of energy that should be produced using fissionable material. Not because of the typical safety and environmental arguments (which many people have put forward very succinctly), but because of simple economic pragmatism.

Energy costs into the 21st century

If we were in the 1950s, I’d very happily go full Dr Strangelove and support the development of a mainly nuclear power grid. Fast forward to today, the economics of renewable energy make it between 50% and 80% cheaper than nuclear energy in most cases. The relatively high cost of nuclear compared to other options surprised me when I first encountered it. There are two reasons for this:

• I thought renewables were expensive. Due to massive technological and manufacturing advances, mostly in China for solar and Europe for wind, the cost of most renewables has plummeted in recent years. Also, decades of anti-renewables propaganda don’t help.
• I thought nuclear was cheap. Reality is, nuclear was cheap

Mispricing of nuclear energy - how and why

I am going to put forward the argument that nuclear energy has been systematically underpriced through the 20th century, due to the nature of building a nuclear power plant, as well as the incentives of the key players involved in it. Looking at the graph above, one may ask how specifically have these costs increased

• As a society, we have become remarkably bad at executing very large project. Nuclear powerplants, like Olympic stadiums and high speed rail projects, are subject to the “Iron Law of Megaprojects”: overtime, overbudget, over and over again. Startup date and construction costs are hard to check parameters that greatly influence viability of project and investment opportunity. That makes it very vulnerable to the project management equivalent of p-hacking. The 2.4MW Vogtle plant in Georgia, US, deserves special mention, with a 1200% cost overrun from $660M to $8.7Bn in 1990s Dollars ($16.2Bn in 2019 Dollars).
• Nuclear powerplants are extremely capital intensive (~80% of total costs). As governments in the west privatised the energy generation network, these large capital costs were transferred to the private sector. Dieter Helm estimates that the costs of the Hinckley C power station in west England would have been halved if it had been financed at the 2% borrowing rate enjoyed by the government, instead of EDF’s 9% cost of capital. You can make the argument instead that the 2% financing costs are a consequence of leveraging the risk of the project with the rest of the governments operations, offloading the risk to the public
• Experience building and maintaining nuclear power plants has taught us that they are very expensive to decommission, a process that can take decades and billions of dollars. Decommissioning costs were not appropriately taken into account in nuclear projects until the 1970s, and then they were estimated to be of the order of ~3% of total costs. Current experience decommissioning these assets have seen figures of ~10% of total costs in Europe and the US. There are strong incentives not to explore these costs at design stage, and let the taxpayer foot the bill 60 years down the line.

All these issues, while complex, are not extremely far off in the "unknown unknowns" territory. Why did the three points above occur for half a century without anyone noticing? The overarching pattern to these failures, not unique to nuclear, is:

1. Large costs over a large period of time
2. "All or nothing" nature - building half of a nuclear plant costs significantly more than half of the total costs, and it will in most cases produce little to no returns
3. Decisions being made by small number of people that are relatively insulated of the consequences of the mispricing.
   Companies (EPC contractors) will make money if the project goes ahead and fail, but won’t if the project is never started. Politicians involved will most likely be out of office and retired, having reaped the benefits of job creation in their local constituencies.

Then what next?

It is clear to me that we will not solve the carbon issue of our energy sector just by producing barium and krypton without significant societal costs. Beyond the simple economic argument however, there are various reasons why state actors would like to include nuclear in their power mix:

• To complement renewables when they are not at their peak power production (although there are alternatives — see the European Supergrid project)
• To develop nuclear engineering within the country. It is very difficult to sustain a military nuclear industry without a civil nuclear industry (with its network of universities engaging in nuclear engineering and research)
• To obtain energy independence in the international stage, if the country possesses uranium but not fossil fuels
• …

However, all of the above are political questions, worth engaging in, but political nevertheless. This means there are no clear cut yes or no answers, but a question about how we want to run our society and distribute scarce resources:

• How much we are willing to pay to obtain a reliable energy network, and how OK we are with the impact to such network to our environment
• How much nuclear deterrence we want our country to have (if at all)
• The role of our country in the international community
• …

And if we choose, as a society, to pursue this option, combining public-private initiatives, guaranteed sale prices and 4 years election cycles has to be one least effective structures incentives-wise. We will need to rethink how to drive accountability for projects spanning not years, but centuries.

The good thing is, figuring out how to appropriately organise these complex web of incentives in major projects will not just help us build better energy networks, but open the door for actually sustainable, equitable human societies into the 21st century.

Some other minor thoughts

The analysis outlined is a simplification, as it does not cover the complexity of some of the subsidy mechanisms used by western liberal democracies (UK and EU in particular). Fossil fuels, renewables and nuclear all receive generous subsidies from different governments, which only compounds with the mispricing exposed above to hide true costs to society.

This only makes the above more poignant: our energy mix is determined both by the material reality of the time and place we live in, and what we choose to do about them. What do you want?

Me? I’m going to hibernate until the Chinese crack fusion. Which turns out is going to happen a lot sooner than I would have predicted.