Accelerating the Pace of Energy Change

The U.S. energy challenges are separable into two largely independent issues. The first is energy security, which is mostly about imported oil and transportation. The United States imports roughly 60% of its oil, sending about a $1 billion per day offshore. Those imports also make us dependent on the actions and fates of a small number of countries. Recognizing this situation, the Obama administration has set a goal of reducing crude oil use by 3.5 million barrels a day within a decade, which corresponds to about 25% of what we use every day for transportation.

The second energy challenge is to reduce greenhouse gas emissions, which is mostly about carbon dioxide, most of which emanates from stationary sources (heat and power). There is an urgency to demonstrate U.S. leadership in dealing with what is intrinsically a global problem, requiring action from the developing world as much as from the developed world. Further, because energy infrastructure lasts for decades, it is important not to lock in future emissions. The conventional coal plants we build today will still be emitting greenhouse gases in the middle of the century. If we can develop the low-emission technologies that the world is going to want in the next several decades, the United States will benefit economically. For all of these reasons, the administration has set a goal of reducing U.S. energy-related greenhouse gas emissions by 17% in the next decade and 83% by 2050 (relative to a 2005 baseline).

Meeting these national energy goals will require significant changes in the ways we produce, deliver, and use energy. The challenge is to identify and implement the most material, timely, and cost-effective solutions. Neither time nor resources are abundant to effect these changes.

Why has energy supply changed so slowly?

History gives some guidance as to how and how rapidly our energy system could change. Figure 1 shows U.S. primary energy supply during the past 160 years [units are quadrillion BTUs (Quads)]. The rise of coal consumption to power the industrial revolution and of oil consumption to fuel the mobility revolution are evident, as is the more recent growth of natural gas and nuclear energy in the latter half of the past century. One lesson from this chart is that virtually all energy consumption has increased as the country has developed. Another is that new energy sources have largely supplemented existing sources, not displaced them.

Figure 1 also shows that the most rapid change in any source was the drop in oil consumption in the late 1970s and early 1980s in response to the oil embargo; the slope is about one Quad per annum. The country then used a total of about 80 Quads per year (now slightly less than 100), so the historical drop in oil consumption was an ~1% annual shift in the energy supply. This continued for about five years before reversing in subsequent decades. Because we are aspiring to reduction goals that amount to 17 to 20% over the next decade, the oil embargo analog gives a sense of the scale of change necessary, which must also be accomplished in a way that enhances our prosperity and quality of life.

The proportions of primary energy sources (see Figure 2) show that nonhydro renewable energy technologies, including wind, biomass, and solar power, are currently a small fraction of the total. The good news is that the energy mix changes because of technology innovation, economics, and politics. The sobering news is that change has taken place historically only on decadal time scales. We must learn how to accelerate the pace of change if we are to meaningfully and promptly address energy security and greenhouse gas emissions.

The rapid evolution of information technologies during the past several decades is often taken as a demonstration of just how rapidly energy could change. For example, the Department of Energy’s (DOE’s) world-leading computing systems have become 10,000 times more powerful during the past 15 years. And personal audio and video equipment went from tapes and CDs a decade ago to flash memory and mp3, during which time the mix of deployed energy technologies remained virtually unchanged. Effectively addressing energy challenges must start with an understanding of why energy has changed so slowly, and, of course, what we can do to accelerate that change.

The energy demand and supply sectors are really quite different in how they change. For example, until very recently, we were purchasing roughly 12 million lightweight road vehicles each year, each expected to last 10 to 15 years. That scale and turnover provide ample opportunity to practice engineering, to optimize the fuel/vehicle/ infrastructure system, and to come down learning curves. Moreover, most people do not buy those automobiles to make money, nor do they buy them with multidecade horizons. The supply-side contrast is stark: In 2008, the United States built only five new coal plants, 94 new gas plants, and about 100 new wind farms, and those supply technologies were purchased with the expectation that they would make money and last for decades. As a consequence, evolution on the supply side is much more difficult and slower than it is on the demand side.

There are four barriers to transforming our energy supply. The first is the scale of things. Large power plants, refineries, and transmission lines are each multibillion-dollar capital investments that must work with the existing infrastructure.

The second is the ubiquity of energy for heat, light, and mobility. That ubiquity means that many people are interested in it, and those interests don’t always align.

The third factor is incumbency. Energy is a commodity; electrons on the grid are anonymous, and fuel molecules are pretty much all the same. Any new technology is then going to have to compete on cost, and the margins are going to be thin.

Finally there is the longevity factor. Large energy assets last for decades (for example, nuclear power plants are being licensed to operate for 60 years or more in the United States). Businesses must bet a large amount of capital on what will happen 30 or 40 years out, with the expectation of commodity-level returns. All four of these factors combine to make the energy supply change very slowly under current circumstances.

The role of the private sector

To accelerate energy change, we must recognize the central role of the private sector and its interaction with the government. The U.S. energy system is almost entirely in the hands of the private sector, which therefore must be the executive agent for any change. Although the government has played a prominent role in technology development through R&D, pilot facilities, and commercial-scale demonstrations, scaling to commercial deployment and operations has traditionally been the responsibility of industry. Further, government alone cannot finance large-scale energy transformation; the total annual investment in the U.S. energy system is some $200 billion, whereas the DOE’s entire annual budget, including basic research, energy research, waste cleanup, and nuclear security, is $25 billion, which is no more than the capital budget for a single large energy company.

Energy transformation requires deploying innovative technologies, but it is not industry’s goal to deploy the most innovative or greenest technologies, although sometimes innovative or green approaches can be useful tactics toward profit. The multidecade time horizons for capital investments in the energy supply business highlight the centrality of predictable return on investment.

Combining the central role of the private sector in energy transformation with the priorities of industry leads to the conclusion that the energy supply will change only when business finds it profitable or it is mandated. President Obama clearly stated as much in his 2010 State of the Union address when he spoke of the need to “finally make clean energy the profitable kind of energy in America.”

Yet despite private-sector ownership, regulatory authority over the energy sector is held by a host of federal, state, and local governments, a composition that often inhibits changes in energy supply. For example, the some 3,000 public utility commissions and governing bodies that oversee the electrical grid all have different motivations, most of which do not include the mandate of energy change or issues germane to transformation. Examples include focusing on low rates rather than total consumer cost and a disregard for the issues of system efficiency or carbon externalities.

To illustrate the impact of regulation, tax policy, and profit on energy transformation, one need look no further than the wind and ethanol industries during the past decade. Since 1999, the relationship between production tax credits to induce profitability in these industries and the pace of change has been irrefutable (see Figure 3).

Although it is obvious that regulation and economics influence the actions of the private sector, if we are to rise to the president’s challenge of making clean energy profitable, it is important to understand how these forces drive business to act. At the heart of business is the allocation of capital, either the capital that you have or the capital that you might borrow. The goal of a business is to invest that capital for a legal and predictable return by balancing risk against reward. More risky investments must have higher reward than more conservative investments. Historically, electricity provides a modest but stable return on investment of around 5%, with a variance on annual return of only 1%.

The magnitude and stability of return are central to business goals, and many elements make up the assessment of any single big energy project. One element is technology risk: If a new technology is deployed, is it certain to work? Another is construction risk: Can the project be built on time and on budget? Another is supply risk: If the project converts switchgrass into ethanol, what is the confidence in switchgrass prices two decades into the future? Operations risk: Can the facility be run efficiently and reliably enough in order to hit profitability goals? Policy risk: Will shifts in policy during a few decades affect the economic viability of the business model? Market risk: How will the prices of the product or services sold fluctuate in the future?

Technology is therefore only one element in judging the viability of a project, and it is often the least influential in moving a project along. Operations and marketing issues are often much bigger factors in deciding whether to do something. So when new technologies are introduced in large projects, conservatism is the norm. To spend billions of dollars on a technology, the performance must be certain and the financial advantage must be significant, otherwise the technology will not be deployed.

Although our near-term goals require rapid deployment of new technologies, the time and expense associated with doing so in energy supply are great. A technology must be taken from a laboratory, where one might invest millions of dollars producing a fuel at a hundredth of a barrel a day (about half a gallon), and moved through pilot facilities, demonstration, and ultimately to full-scale production and deployment, where billions of dollars are spent for daily production capacity on the order of 100,000 barrels. This is a decade-long process because of the confidence required to take each step. For example, the SOx and NOx scrubbers used on coal plants today required 40 years and four generations of technology to reach their current state of maturity.

Many in the venture capital world are optimistic about their ability to transform the energy system. That could well be true on the demand side, where unit sales volumes can allow more efficient technologies to be more easily refined and scaled. But the average venture capital fund is about $150 million, with individual investments of $3 million to $5 million (and unusually $10 million to $20 million). These sums are more than 100 to 1,000 times smaller than the cost of a single power plant. Additionally, in energy supply one invests with a multidecadal perspective, whereas the average venture firm exit time is five or seven years. Also discordant with the venture capital business model are the high illiquidity of big infrastructure assets and the small and conservative returns noted above. All of this need not and will not discourage venture capital activity in energy supply, but it is important to understand the character and nuances of the big energy business. It is often said that big energy companies don’t understand risk, but the truth is precisely the opposite: The job of a large energy firm is to manage risk, but on a 30-year time scale and with much larger sums of money than other businesses usually do.

How can government accelerate change?

The discussion above shows that to effect significant changes in our energy supply, we in government must fully engage with the private sector, understanding that mindset and thinking like a business. Thus, the first and most important consideration is predictability. Given the long time horizons, policies that are uncertain or fluctuate every few years will not get us anywhere and in fact are often counterproductive. And of course the policies need to be well considered; we should not be starting out on paths that are suboptimal in terms of technology, economics, or environmental impact.

A second consideration is to play to business’s risk/reward calculus and do what we can to mitigate risks for the early movers. The DOE’s National Laboratories (large, multidisciplinary, mission-oriented research organizations) could play a much greater role here. The provision of user facilities as full-scale test beds for new technologies could also be very important for mitigating technology risk. Much as the DOE’s scientific user facilities have greatly advanced understanding of the physical world, technology user facilities could provide significant insight into the operation of innovative technologies and components at scale. For example, a microgrid test bed where various demand-side technologies can be tested in real environments, or combustion facilities where technologies for gas treatment, CO2 absorption, and other components could be tested at commercial scale and operational conditions would reduce uncertainties and risks surrounding the deployment of new technologies.

Market risk can be mitigated by establishing renewable or low-carbon power standards. Although various states are starting to do that, we lack national standards that would create a market for new technologies. Renewable fuel standards do exist and function in much the same way, although it is important to set these standards carefully. Longer-term power purchasing agreements similarly damp out market risk for the deployment of innovative technologies. These are all good things that the government can and should do to mitigate market risk.

To address capital risk, the DOE has been executing loan guarantees. Earlier in 2010, the department issued guarantees for the first new U.S. nuclear plants in 30 years. Other programs are stimulating technological energy innovation in this country, as well as the demonstration and deployment of new energy technologies at commercial scales.

The DOE is also working to accelerate the transition of energy technologies out of the laboratory. The Secretary of Energy and the leadership team have made connections across organizational units to bring the department’s resources to bear more productively on the energy problem. We have set up a number of new mechanisms, including Energy Research Frontier Centers that focus on the basic research side of energy technologies and the scientific barriers to technological progress, and Energy Innovation Hubs that span basic research and technology demonstration and concentrate resources where opportunity exists for rapid commercialization of discovery. The Advanced Research Projects Agency–Energy (ARPA-E), which funds high-risk projects in the search for new technical platforms, with the expectation of soon hitting a few home runs, has been inaugurated in the first year of the administration. Finally, the DOE is executing demonstration projects with industry to accelerate technologies out of the development cycle and into commercial viability. In doing so, we must do a better job of capturing the knowledge from each demonstration and making it widely available to the community trying to solve these problems.

Simulation is an important tool the research community can use to mitigate risk. The United States currently leads the world in applying high-performance computing to realistic physical models. That capability is largely resident in the DOE’s science and weapons programs and principally stems from a deliberate effort started in 1995 to understand the nuclear stockpile in the absence of underground nuclear testing. That program accelerated computing capabilities by a factor of 10,000 in little more than a decade. More importantly, it taught us how to combine data from integral systems with the results of laboratory experiments into the codes that are truly predictive for complex systems. Simultaneously during the past decade, the DOE’s Office of Science has been making that same class of machines available to the open scientific community for problems ranging from climate modeling to protein folding to materials science.

It is now time to focus simulation capabilities on energy systems. Combustion devices (internal combustion engines, boilers, and gasifiers), fission reactors, carbon capture and storage facilities, and the electrical grid are all applications where validated high-performance simulation would help to optimize designs, compress design cycles, and facilitate the transition of technologies to scale. Right now, this is a unique U.S. capability and so is a competitive advantage that must be seized. Recent simulation successes of Cummins in diesel engines and Goodyear in tires give some sense of what is possible. But real impact will require more deliberate efforts to bring today’s simulation capabilities into broader commercial practice and to improve them for future applications. Foremost among the latter would be accelerating computing capability by another factor of 1,000 within the next decade.

We hope to have made a convincing case that first, the energy supply business is not simple—there are nuances and aspects that may not be readily apparent to most people not directly involved with the industry—and second, that the government’s key role in catalyzing a transformation of the energy system is to mitigate risk for the private sector. Setting a predictable and well-considered playing field of policies and economics is the most important thing government can do. Beyond that, the DOE should facilitate large-scale demonstration projects and support precompetitive research, as well as the technology transfer necessary to move new technologies into the private sector.