Scratching the Surface


A 2013 retrospective on ‘Saving Gigabucks with Negawatts’ (1985)

Fortnightly Magazine - March 2013

Slowly, ponderously, but with gathering speed, the competitive forces my NARUC talk described 29 years ago are inexorably transforming the electricity business.

Negawatts aren’t yet allowed to compete head-to-head with supply in a third of the United States, utilities in 35 states are still rewarded for selling more electricity and penalized for cutting customers’ bills, and electric generation remains far more subsidized than efficient use. Yet as predicted, electricity’s share of end-use energy is rising while total electricity use is falling. Rocky Mountain Institute’s 2011 “grand synthesis,” Reinventing Fire,1 found about 1 percent per year average demand shrinkage plausible to 2050, despite extra use for electric autos and average GDP growth of 2.4 percent per year.

Of course, a great many details have changed. Reinventing Fire found in 2011 that electric productivity could rise even more with plausible adoption during 2010 through ’50 than my 1984 NARUC talk and 1985 article, “Saving Gigabucks with Negawatts,”2 claimed for full long-run adoption—but at just one-third the average cost, because today’s technologies and designs are far better than 1984’s. Even in such enormous quantities, modern efficiency techniques cost less than just operating any thermal power plant.

Not every technology mentioned in the article got to market, but so many others did that even today, “we have barely scratched the surface of how much efficiency is available and worth buying.” And far from petering out, negawatts are getting bigger and cheaper—not only by better technology but also by smarter design. The integrative design techniques3 that Reinventing Fire reported for buildings, factories, and vehicles let good designers, with consistency and impunity, achieve expanding returns to efficiency investments.

Efficiency techniques keep evolving quickly. The compact fluorescent lamp described is now ubiquitous, durable (just last year I retired my last one at age 30), but nearly obsolete. My home’s latest lighting retrofit, its fifth since 1984, uses all LEDs, which continue to improve weekly. Modern LED luminaires make about as many lumens per watt as CFLs but can deliver them many times more effectively. Likewise, the center-of-glass insulating value of my home’s best windows today is nearly four times the “best commercially available” in 1984—and should again be surpassed later this year.

Amory B. Lovins

Even in heavy industry, our practice’s retrofit designs can often save around 40 to 60 percent of energy use and pay back in a few years, while new designs usually save even more, generally with lower capital cost. After that will come the next revolutions, like additive manufacturing, biomimetic design,4 and atomically precise manufacturing.5

Customers’ steadier purchases of efficiency help create the “new normal” of stagnant or falling demand described in these pages last December.6 Yet a huge “overhang” of unbought efficiency remains. How fast it’s bought still “depends on customers’ access to information, capital, and freedom of choice”—and on how hassle-free the delivery channels can make the negawatt-buying experience. Emerging innovations even include instrumented-van multispectral drive-by surveys linked with sophisticated automatic bill analysis. On-bill financing, PACE (property assessed clean energy) bonds, and efficiency service agreements (power purchase agreements for negawatts) are starting to scale too, removing the capital burden so the customer has a positive cash flow from day one. Such implementation advances are just as important as those in technology and design.

New Economies of Scale

The supply side is shifting so quickly that even such bedrock assumptions as the permanent competitiveness of operating old nuclear plants can no longer be taken for granted.7 Analytic changes are important too: the “dozens of neglected disecono­mies of scale” mentioned in the article’s note 7 turned out to be 207 effects that collectively can often shift value by an order of magnitude8—enough to flip almost any investment decision.

Thanks largely to utility and state leaders who steered by a tall star while federal policy weathervaned, the distributed and renewable generation revolution emergent in 1984 is now in full swing. It’s just two decades behind where it could have been with consistent and supportive policies, so the renewable fraction of U.S. electricity rose from 11.9 percent in 1985 to just 12.1 percent in 2012. Thus during 2005 through ’10, while renewables crawled from 9 percent to 10 percent of U.S. electricity, the mighty industrial power of Portugal raised its share from 17 percent to 45 percent. The United States’ economic losses from its lagging exploitation of the renewable energy resources that it largely invented are self-inflicted.

The new renewable competitors are even cheaper than expected. Many keep getting cheaper so steadily that over the next few decades, sometimes much sooner, unsubsidized renewables will beat still-subsidized non-renewables. New 2012 wind-belt PPAs sold 25-year power for about $25 to $40 per megawatt-hour levelized, and falling.9 Photovoltaic modules got another 34 percent cheaper just in 2012, and 65 percent cheaper in China during 2011 and ’12. When the three big California shareholder-owned utilities in 2011 solicited PV bids, they were offered over 50 GW, roughly the state’s total peak load. By spring 2012, the winning PV bids10 undercut a new combined-cycle gas plant even though U.S. installed PV system costs averaged twice those of cloudy Germany. German installers use the same equipment but have streamlined value chains by scaling to about 8 GW per year, so in each of the past two years, Germany has added more PV capacity in one peak month than the U.S. installed all year.

The article’s basic conclusion that big thermal plants are obsolete has proven true, as has its vigorous call for flexibility and strategic risk management. But the big issues now are no longer about marginal costs; they’re about the very nature of the electricity enterprise. Let’s recall some history. In the 1880s, Thomas Edison didn’t sell kWh; he sold lighting services. As motors became popular, New York Edison Co. wanted to sell kWh for lighting at night and motors in the daytime. Edison objected that if the utility sold kWh, then more-efficient lamps, which would surely be invented, would enrich lamp-makers. But the same lamps would enrich the utility if it sold lighting services. He was overruled in 1892, and we’ve been making the same mistake ever since—except in street-lighting, where Edison’s service-tariff concept survives. Invention continues too: today’s best LED street lights can deliver superior service with just 1 to 2 percent of ASHRAE’s prescribed power density.

The radical bypass scenarios that give utility executives nightmares also give venture capitalists sweet dreams. Many of the important new products are unregulated and probably unregulatable. Utilities’ pricing, business, and regulatory models are all up for grabs. And however they turn out, the grid’s existential cyber- and physical-security challenges must be quickly overcome before superstorm or solar storm, accident or malice, catch up with us and shatter the economy.

Skill, Agility, Imagination

The U.S. electricity system—aging, dirty, insecure—must be modernized and essentially rebuilt by 2050. Reinventing Fire found this will cost around $6 trillion in present value no matter what we build—whether it’s more of the same, or new nuclear and so-called “clean coal,” or centralized renewables, or half-distributed renewables. But these four futures, with virtually identical costs, differ profoundly in risks related to security, technology, finance, fuel, water, climate, environment, and health. The distributed-renewables scenario, with high efficiency and microgrid architecture, can eliminate the cascading-blackout risk and best manage all the other risks. Yet this future requires the most challenging shifts in the industry’s mental models, business models, culture, regulation—everything that’s even harder and more important than hardware. And ready or not, here comes that future.

The outcome is not certain. “Change,” says Jack Riggs, “is not absolutely required because survival is not a mandatory condition.” But as market actors come and go, merge and surge, navigating through unprecedented turbulence will require all our skill, agility, and imagination.11

The challenge I laid down 29 years ago is now squarely before us, and we must rise to the occasion. Having grown up in a relatively staid and stable business, we are now fated to live in interesting times.


1. A.B. Lovins and Rocky Mountain Institute, Reinventing Fire: Bold Business Solutions for the New Energy Era, Chelsea Green (White River Junction VT), 2011,, summarized at and

2. Amory B. Lovins, “Saving Gigabucks with Negawatts,” Public Utilities Fortnightly, March 21, 1985.

3. A.B. Lovins, “Integrative Design: A Disruptive Source of Expanding Returns to Investments in Energy Efficiency,” RMI Publication #X10-09, 2010, and “Factor Ten Engineering Design Principles, Version 1.0.”

4. Janine Benyus, Biomimicry, Wm. Morrow, 1997;

5. Eric Drexler, “Radical Abundance,” Public Affairs, in press, 2013.

6. Ahmad Faruqui and Shultz, Eric, “Demand Growth and the New Normal,” Public Utilities Fortnightly, December 2012.

7. A.B. Lovins, “The economics of a U.S. civilian nuclear phase-out,” Bulletin of the Atomic Scientists, March/April 2013, in press.

8. A.B. Lovins et al., Small Is Profitable: The Hidden Economic Benefits of Making Electrical Resources the Right Size, a 2002 Economist book of the year, Rocky Mountain Institute.

9. Those prices (LBNL-5559e, 2012, p. 52), averaging $32 and trending downwards, are net of new wind power’s production tax credit, whose levelized value in 2011 dollars is about $18/MWh at a 3 percent per year societal real discount rate, or $28 from an investor perspective with a 15 percent per year real hurdle rate. New nuclear plants get larger operating subsidies, plus construction subsidies rivaling or exceeding their construction cost: D. Koplow, “Nuclear Power: Still Not Viable Without Subsidies,” 2011.

10. Averaging $89/MWh busbar, net of the 30 percent federal solar tax credit, which is generally less than subsidies to new nonrenewable power (id.).

11. To help figure out many of the industry’s thorniest issues, my RMI colleagues, led by James Newcomb, have launched an Electricity Innovation Laboratory (e-Lab)—a multi-year, multi-stakeholder exercise in rapid mutual learning.