Smart Energy Stimulus Means Thinking Small

Liam Denning
·7-min read

(Bloomberg Opinion) -- In 1968, a crisis loomed for the Pacific Northwest. FDR-era hydroelectric dams that had powered the postwar boom were reaching their limit, threatening blackouts and economic stagnation. Big challenges beget big plans: The Hydro-Thermal Power Plan called for 26 new nuclear and coal-fired plants to keep the region humming through the mid-1990s. But 15 years later, a public power agency tasked with building five nuclear power plants called WPPSS — dubbed “Whoops!” by the wags — triggered the then-largest municipal bond default in U.S. history.(1)Of the five plants, only one went into operation, while four were partly built at enormous expense and abandoned.

In 2020, the whole country faces a crisis — a public-health and economic debacle rolled into one. Once we get past this one, another still looms: climate change. The two are comparable in several respects, not least their utter indifference to politicization. They are also linked by the question of stimulus money: How much of any program will go toward climate-friendly initiatives such as electrification and renewable energy? Part of the answer must be another question: How can such resources be deployed most usefully?

The fiasco of WPPSS offers some pointers. At the core of its many failures were wildly overoptimistic forecasts. While the plan envisioned more than 40,000 megawatts of new capacity, regional demand increased by only about 9,000 average megawatts.(2) First proposed in the early 1970s, the only one of the WPPSS plants to ever generate electricity didn’t switch on until 1984, a year after the bond default. The estimated cost of all five quintupled to roughly $24 billion.(4)

It didn’t help that WPPSS was a governance black hole and poorly managed, nor that its grandiose plans coincided with mounting opposition to nuclear power (the Three Mile Island accident happened in 1979). But the underlying failure was committing to large-scale solutions to a perceived problem rather than scalable ones. In attempting to build several large, complex reactors at once, WPPSS left ratepayers and bondholders exposed to the risks of delays, overruns and, as it turned out, the possibility of the electricity not being needed in the first place. A more measured approach aligned with actual needs would have saved a lot of heartache (and money).

Bent Flyvbjerg is a professor at Oxford’s Saïd Business School who specializes in mega-projects, from nuclear power stations to staging the Olympics. In a recent article, he drew a line between Covid-19 and climate change, phenomena where the impact of the next disaster will likely be worse than the last — “regression to the tail,” rather than the mean, as he calls it. In managing such risks, our aim should be to cut the tail of extreme outcomes — by, say, stockpiling medical equipment or developing zero-carbon energy — and err on the side of caution.

That we would eventually face a pandemic was a known known, so to speak. Not only were we unprepared, though, but our response also lacked urgency and coordination, leading to tens of thousands of unnecessary deaths and the resort to mass lockdowns with huge economic costs. Similarly, we know climate change is happening and, if unaddressed, its impacts will compound over time. Flattening (and then crushing) the emissions curve is an immediate issue, not one to be deferred to the next generation.

When planners in the Pacific Northwest feared blackouts, they responded with a vast, unwieldy program that left the region vulnerable to other tail risks of cost and delay. You might say they didn’t have a choice back then: If you needed more power, you built giant power stations. On the other hand, maybe they could have started with one nuclear plant and seen how it went. In any case, we no longer have that excuse today. We have options.

For a climate-centered energy stimulus to achieve its objectives, it must create jobs and generate competitive power — fast. “What you want to do is projects that are COTS: Commercial off-the-shelf. That is the opposite of bespoke,” says Flyvbjerg, adding, “We don’t have 10 or 15 years to wait for the effects; effects on the economy, effects on employment but also effects on climate.”

If recent history is anything to go by, then nuclear projects in the U.S. are still prone to overoptimistic forecasting, multibillion-dollar cost overruns and long delays (see this). So-called small modular reactors have long been touted as a way to reduce costs and shorten construction schedules. If they can be shown to work, then they may yet compete. But like other traditional generating technologies, they face ever stiffer competition from renewable energy.

This isn’t to single out nuclear (or any other) generation. Rather, it’s to highlight renewable energy’s edge. Not only have the costs of solar and wind power dropped dramatically over the past decade. They are also, crucially, scalable rather than large-scale. Wind and solar farms can be built quickly using standardized equipment, and added to (or not) over time, keeping supply aligned with demand. As capacity grows, and manufacturers and developers keep learning, costs fall further — the industry has been revolutionized in the space of a decade, less time than it can take to develop a nuclear plant. And these assets can be dispersed, spreading the jobs and other benefits more widely than traditional stations. It also helps that almost 1 in 2 jobs in wind and solar power, and the associated areas of energy storage and smart grids, are in stimulus-friendly construction. That’s the highest proportion for any sector of America’s energy industry (see this).

Any stimulus effort comes with a built-in backlash, which, if the past decade has taught us anything, is a fat-tail risk itself. Cutting that off means mitigating the chances of a Whoops!-like fiasco that ends up being a business-school case study or, more pertinently, fodder for an attack ad. On that front, Flyvbjerg finds an average budget overrun for nuclear projects of 122% versus just 13% for wind farms.(3)

It might not make much sense that Covid-19, of all things, could be the event that pushes America to finally fix its decadent energy system. Regardless, if it does, renewables provide a way of dealing with our climate risk without creating new ones along the way.

(1) Washington Public Power Supply System was a public power joint operating agency formed in 1957 to procure power for regional utilities. Or, as Jeremiah D. Lambert writes: "WPPSS was a vehicle designed to minimize oversight and allowed individual utilities to escape accountability and criticism. Using tax-free revenue bonds, it had the legal right to borrow money without voter approval by action of its board of directors. The public would not feel the costs until much later." ("The Power Brokers", MIT Press, 2015).

(2) An average megawatt is one million watts supplied continuously 24 hours a day for a full year (essentially, the output of a one megawatt plant operating at 100% utilization).

(3) Source: "Nuclear Implosions" (Cambridge University Press, 2008).

(4) Data are drawn from a sample of 194 nuclear power projects and 53 wind farms. In addition, Flyvbjerg finds the average delayfor a sample of 177 nuclear projects is 65% in excess of the original schedule versus 22% for a sample of 28 wind farms.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Liam Denning is a Bloomberg Opinion columnist covering energy, mining and commodities. He previously was editor of the Wall Street Journal's Heard on the Street column and wrote for the Financial Times' Lex column. He was also an investment banker.

For more articles like this, please visit us at

Subscribe now to stay ahead with the most trusted business news source.

©2020 Bloomberg L.P.