Nuclear power is dying. Can radical innovation save it?

Among experts who study energy policy, there's remarkably widespread agreement that tackling climate change would be much, much easier if we could quickly and cheaply build lots of new nuclear power plants. It's not always a popular argument, but it's a compelling one on its face.

This paper from Jesse Jenkins and Samuel Thernstrom offers a great overview of the relevant research. Yes, it's entirely possible to imagine a carbon-free grid powered 100 percent by renewables like wind, solar, and hydropower. But it's also undeniably hard to juggle intermittent sources of electricity. Mixing in nuclear plants (or fossil plants that bury their CO2) that can run at all hours alongside renewables could greatly lower the cost of deep decarbonization — at least in theory.

The practical flaw in any pro-nuclear argument, however, is that very few countries are actually building new reactors anymore — and many are downsizing their existing fleets. In Germany and Japan, public opinion has turned sharply against nuclear. In the United States and Britain, meanwhile, nuclear programs have been plagued by high upfront costs, construction overruns, and regulatory roadblocks. Today, the world isn't even building enough new reactors to offset coming retirements (see chart here).

More from Vox:
Ivanka Trump once organized a campaign encouraging women to share their job titles
On health reform, Donald Trump followed Republican leaders into a ditch
Donald Trump and the rise of tribal epistemology

So what would it take to turn this situation around and bring back nuclear power as a vital weapon against climate change? Broadly speaking, there are two options here. Neither is guaranteed to succeed, but for people worried about the herculean efforts needed to stop global warming, they're both worth considering seriously:

1) The first option would be for governments and industry to clear away the (many!) obstacles hindering construction of new light-water reactors, the most common type of reactor built in the 20th century. This technology is proven to work, after all — it supplies one-fifth of America's electricity — and South Korea has figured out how to build such reactors affordably. So other nations would basically need to mimic South Korea, relying on standardized designs and economies of scale to drive down costs and expand their nuclear fleets. Michael Shellenberger, a pro-nuclear advocate with Environmental Progress, has made this case eloquently here and here.

But this first option is easier said than done. In many countries, it would require serious market and policy reforms, and perhaps streamlining regulations around light-water reactors. It would also likely require a tidal change in public opinion about nuclear, allaying concerns about waste and safety. Green groups would probably have to soften their opposition to the technology.

2) Increasingly, many nuclear advocates have come to believe those political hurdles are just too formidable. So they've placed their faith in a second option: radical innovation. There are dozens of startups in the US working on clever alternatives to the traditional, hulking light-water reactor — advanced nuclear designs that, ideally, could prove smaller, safer, more flexible, and ultimately cheaper, perhaps even with less waste. These ideas would need an initial jolt of government support to come to market, but the hope is that new, small, advanced reactors, by virtue of design, could overcome the social and economic barriers crippling the nuclear industry.

This second option is gaining popularity among both parties in Congress, and it's well-articulated in a new report from the Breakthrough Institute: "How to Make Nuclear Innovative." There, authors Jessica Lovering, Loren King, and Ted Nordhaus lay out a series of policies that could help bring advanced reactors to fruition, from licensing reform to targeted aid from the Department of Energy. They draw analogies to federal support for fracking, drug research, even the private spaceflight industry.

The catch? There are no guarantees that advanced nuclear tech will pan out anytime soon. And, as Nordhaus told me in an interview, it's quite possible that even these newer reactor designs could run into the same pitfalls facing existing nuclear, like public opposition or stifling new regulation.

"None of that goes away overnight," he says. "But if we want to reset public perception of nuclear, I think starting with technologies that are quite different from today's designs is our best bet."

So those are two big ideas on offer for reviving nuclear power: Figure out how to widely deploy a tested 20th-century technology that's run into serious trouble, or invent something entirely new. They are, in essence, two very different visions of the current political landscape. Below, I'll flesh out each option, probing some pros and cons — and then take a brief look at what happens if they both fail.

The best argument for doubling down on 20th-century nuclear power technology is that we know it works. France and Sweden built some of the cleanest grids in the developed world by scaling up nuclear power very rapidly in the 1970s and '80s. If the rest of the world merely copied what Sweden did, one recent study in PLOS One found, we could eliminate all fossil fuels from electricity in just 25 to 34 years.

But right now, nuclear power faces several massive roadblocks to such an expansion. First, the meltdowns at Three Mile Island in 1979 and Fukushima in 2011 have turned public opinion sharply against nuclear. While no one died in either accident — and while nuclear isvastly safer than coal or natural gas — these meltdowns intensified opposition to nuclear, and led to strict regulations making new reactors more costly. Some countries, like Germany, are actively trying to phase out all nuclear plants.

Second, current nuclear reactors are very large by design, requiring huge upfront costs, which means they're tricky to finance. In recent decades, they've also been plagued by delays and construction overruns — a terrifying prospect to investors thinking about borrowing billions of dollars upfront to build these things. In many places, it's just easier and safer to invest in smaller natural gas, wind, or solar plants (especially given public subsidies for the latter).

Finally, nuclear's woes seem to be getting worse over time, not better. To allay public fears of meltdowns, the industry developed new "Generation III reactors" that featured intricate new safety systems. That includes four new AP1000 reactors currently being built in Georgia and South Carolina.

But like many first-of-a-kind projects, these new reactors have faced early missteps and delays, a disaster given the size of the investments at stake. Toshiba's Westinghouse, the company building these AP1000s, is now filing for bankruptcy over its struggles. And, even if these four units get built, it's unlikely we'll see more AP1000s built in the United States anytime soon.

There is, however, an intriguing exception to nuclear's current woes. South Korea has figured out how to build light-water reactors on time and under budget — and they've gotten quite good at it. The secret? The country's state-owned utility initially settled on a single Generation II reactor design, the OPR-1000, that was based on proven technology and then built it over and over and over, learning from experience as it went. The country also had stable regulations and a highly skilled workforce. Now that South Korea is adept at building OPR-1000s, it's gradually moving on to a newer Generation III design, with plans to build seven in Korea and the United Arab Emirates.

"We can learn from the Koreans," writes Michael Shellenberger in a long recent essay on Toshiba/Westinghouse's woes. To revive nuclear, countries and industry would need to likewise settle on a proven light-water reactor design and build it again and again to drive down costs — as opposed to the current situation, where countries like the United Kingdom are pursuing a welter of new designs. Only once a global supply chain is reestablished will it be time to experiment with incremental new models.

"Nations must work together to develop a long-term plan for new nuclear plant construction to achieve economies of scale," Shellenberger writes. "Such a plan would allow for certainty, learning-by-doing, cost declines and lower financing costs."

It's a forceful call to action. Yet it's also clear that this would require major policy and political shifts in many countries. The US, for example, doesn't have a single state-owned utility like South Korea does — it has balkanized state electricity markets and deregulated utilities, which has made standardization and coordination on nuclear power far more difficult. This vision would also likely require significant government investment in nuclear, at least early on. France's nuclear build-out was guided by the heavy hand of the state.

Just as importantly, such a build-out would likely require changing public opinion about nuclear power in the United States, Europe, Japan, and elsewhere — overcoming long-standing (but often unfounded) concerns around radiation and waste. It would require persuading regulators that existing light-water reactor technology is already safe enough and shouldn't be bogged down by shifting requirements that drive up costs.

It'd take too long to delve into all those issues here, and I won't hazard a guess as to whether these changes are feasible. But they're certainly daunting, which is why many nuclear advocates now think radical innovation is a better way forward…

In their paper, "How to Make Nuclear Innovative," the Breakthrough Institute authors argue that a massive scale-up of existing light-water reactors is unlikely today (outside of a few countries like South Korea or maybe China). Such a push, they write, would require "reversing robust political and economic trends" in places like the US or Europe, such as liberalized electricity markets and declining public investment.

"If we treated climate change like it was an incoming asteroid, then we might embark on a crash program of public investment in nuclear the way France and Sweden did," says Nordhaus. "But I just don't think we're going to do that."

So, instead of trying to fight against the tide of neoliberalism, the Breakthrough authors argue that bold new technology is the best path forward — specifically, disruptive new reactor designs that, if they panned out, could slide more easily into today's energy world.

Right now, there are at least 50 nuclear startups across North America working on radical alternatives to the traditional large light-water reactor. Samuel Brinton did a great overviewfor the think tank Third Way in 2015: Some firms are exploring molten salt reactors that can operate more efficiently at higher temperatures and are inherently safer. Others are working on small modular reactors that come in all shapes and sizes and could, conceivably, be prefabricated in a factory. Down the road, we may see reactors that can consume nuclear waste (see diagram here).

Unfortunately, these clever ideas all face daunting regulatory hurdles. "The Nuclear Regulatory Commission is presently unprepared to license them," the Breakthrough report notes. "The Department of Energy and the national laboratories are not well set up to assist them. Nor are all but the largest incumbent nuclear firms in a position to acquire the large amounts of capital that would be required to navigate the current licensing process, develop advanced materials and fuels, or build first-of-kind reactors."

These radical new designs, advocates argue, would have several major advantages over existing light-water reactors. First, newer reactor designs are unlikely to melt down due to automatic shut-down features — which could help them gain public acceptance, and would also eliminate the need for costly concrete containment domes and other safety features that drive up the cost of today's nuclear plants.

Second, small modular reactors could prove much cheaper to build over the long run. Because they're smaller than traditional reactors, they're less daunting to invest in. Not only are the upfront costs lower, but they could be used in wider array of situations. And, because they're small, manufacturers could potentially mass-produce them, perhaps in a factory, and drive costs down that way. (Note that one company, NuScale, is on track to build the first small modular reactor by 2024.)

Admittedly, these arguments are all unproven. It's still unclear how well any of these advanced designs will work — note that Transatomic, a Peter Thiel-backed startup working on a molten salt reactor, recently had to walk back some of its lavish claims about efficiency. Likewise, small modular reactors might ultimately prove cheaper, but that's speculation at this point.

And advanced designs might be inherently safer, but it's uncertain if the public will feel any better about them — or if regulators will treat them with a lighter touch. Plus, many of these technologies are still decades from commercial development.

"There are no guarantees that any of these designs will get us cheap, competitive nuclear power," says Nordhaus. "Technological change doesn't work that way. But, again, if we're not going to move to a French-style state-led [nuclear push], then this is the only place I see a future for the nuclear industry."

To that end, the Breakthrough authors outline some US policy changes that they see as giving the advanced nuclear industry the best chance of showing what it can do. That includes reforming the Nuclear Regulatory Commission so that it's better able to help startups license new reactor designs; reorienting the Department of Energy's National Laboratories to provide these startups with technical support; building publicly funded test reactors; and having the government offer some sort of cost-sharing for the initial wave of reactors. Think around $2 to $3 billion per year in public spending, says Nordhaus. Certainly more modest than a French-style state-led nuclear buildout.

The authors draw on lessons from past public-private partnerships that have led to innovation, such as NASA's efforts to nurture a private spaceflight industry. In 2005, NASA asked private firms to develop a replacement for the agency's Space Shuttle to supply the International Space Station. Rather than preselect a favored technology, NASA basically dangled prizes (in the form of contracts) to any company that could figure out how to solve the problem. Startups like SpaceX and Blue Origin stepped up, and today commercial spaceflight is a $1.1 billion industry.

In essence, this is a vision of a nuclear future that's led by private industry and only secondarily supported by the state, rather than the deep government involvement that marked the 20th century's nuclear push. This option feels more in line with the present-day political zeitgeist — which is why Congress is showing serious interest in advanced nuclear. But we also won't know for years or decades if the technology is viable at large scale.

The two options above are hardly mutually exclusive. It's possible to imagine a world where more governments make a serious push to build dozens or even hundreds of new light-water reactors based on existing technology, while at the same time the US and others are nurturing startups that could develop advanced nuclear designs down the road.

But you can also imagine a world where both options falter. That is, few countries outside of China and South Korea make a serious push to build many more light-water reactors, the industry can't get its costs under control, and advanced reactor designs struggle to make it to market. If that happens, the nuclear industry will wither in the coming decades. Plain and simple.

If nuclear withers, efforts to combat climate change wouldn't necessarily be doomed. Wind and solar power are still galloping ahead at a rapid clip, after all, and it's entirely conceivable that renewable technology keeps improving significantly, and that innovators, engineers, utilities, and policymakers solve the obstacles in the way of a 100 percent renewable grid. (Dirt-cheap batteries would certainly help.) Or perhaps some other energy technology, like carbon capture and storage for coal or natural gas, comes along to save the day.

It's hard to say what the grid of 2050 will look like. But note that decarbonization is hardly a certain prospect. The scale of changes to our energy system needed to stop global warming are utterly mind-boggling, and it's far from clear we'll be able to pull it off in a timely fashion. Doing it without a technology that has historically been a vital source of carbon-free power would make that challenge all the more difficult.

• This Third Way report by Josh Freed, Samuel Brinton, Erin Burns, and Amber Ronson offers a handy guide to the various advanced nuclear technologies currently being researched.
• A longer look at why America largely abandoned building new nuclear power plants. And here's a look at how New York is trying to save its existing fleet from retiring prematurely in the face of low natural gas prices.
• Note that a key Senate panel recently approved the Nuclear Energy Innovation and Modernization Act, which would streamline the licensing process at the Nuclear Regulatory Commission. While this bill has bipartisan support, it doesn't go nearly as far as the policy recommendations outlined by the Breakthrough Institute._