Speeches

ARPA-E Energy Innovation Summit

Wednesday, February 29, 2012
Accelerating Energy Innovation, Five Steps at a Time

Almost two weeks ago, I announced that I would be stepping down later this year as president of MIT. That decision has led me to reflect a bit on the beginning of my service at MIT, more than seven years ago, and one memory stands out from all the rest: the clear, unambiguous message from the Institute’s entire community that it was time for MIT to step up and do something serious about energy. After laying the groundwork, in the fall of 2006 we launched the MIT Energy Initiative, or MITEI, and MIT joined the emerging movement to invent a sustainable energy future. All of you here are part of that movement, representing every sector, from founders to funders.

Today, we are here to ask ourselves how we can accelerate energy innovation. Before I try to offer a response to that critical question, let me take a moment to reflect with you on just how wildly the world has changed in the last few years; so many of our fixed assumptions have come unfixed. Consider four changes in the global energy marketplace, in the last five years alone:

First, the economy. In 2006, global energy demand was soaring, with no upper limit in sight. However, three years later, in 2009, we found ourselves in the midst of a continuing economic downturn in which global energy consumption decreased for the first time in 30 years. Yet, despite the most serious recession since the Great Depression, oil prices remained stubbornly high, underscoring concerns about the security of oil supply and the need to develop affordable, scalable alternative transportation fuels. Today we see yet another face of energy economics: with the economy only very gradually gathering steam, gas prices are rising so fast that they have become fodder in the presidential race.

Policy represents a second area of marked change since 2006, and its effects on our prospects for game-changes. When MIT began its new energy push in 2006, we hoped and, candidly, we anticipated, that the US and other key nations would take concrete policy action to reduce greenhouse gas emissions. Yet, as we all know, progress under the Kyoto Protocol has stalled, climate legislation failed to pass the US Congress, and we now hold little hope of new energy and climate policy any time soon. That means that low-carbon technologies must meet economic objectives early and cannot succeed if they only mitigate greenhouse gases.

A third, related change: the economic recession has exacerbated political opposition to spending, mandates or subsidies, putting added pressure on the existing renewable energy technologies. While we are delighted to have allies like Representative Chaka Fattah and Senator Chris Coons, their dedication to these issues is all too rare in Congress. The fact is that, whether through new materials, economies of scale or new distribution methods, renewable energy research must focus, above all, on bringing cost down. New entrant technologies must be able to compete on cost when they enter the market, or they will not enter it at all.

A fourth, unexpected change in the energy marketplace is that the landscape for conventional fuels has turned out to be anything but conventional. In 2006 the US was a net natural gas importer, and most people thought North America was running out of natural gas. Today, largely thanks to the development of affordable shale gas, the US is now the number-one gas producer in the world, and recent assessments indicate that global shale reserves could significantly alter the geopolitics of gas. As a result, the US is closer to achieving energy independence than at any time since 1992. By some estimates, by 2020 we could surpass Russia as the world’s number-one energy producer. As MIT Institute Professor and former CIA director John Deutch recently observed, “The past image of the United States as helplessly dependent on imported oil and gas from politically unstable and unfriendly regions of the world no longer holds.” At the same time, we still feel the repercussions from the Macondo oil spill. In addition, in the wake of Fukushima, the long-term future of nuclear energy is uncertain. The upshot, as we all know, is that conventional fuels will be used for decades to come, meaning that the world still needs ambitious research on how to mitigate their environmental impacts.

I present these familiar facts to say that if the past five years have taught us one thing, it is that just about the only permanent fact in the energy landscape is the persistent and pressing need to design a sustainable energy future. The question then becomes, without policy or subsidies on our side, how can we reach that future, faster?

Let me interrupt my doleful recitation to say, emphatically, that I am extremely optimistic, because America’s innovation system has proven, over decades, to be remarkably resilient and adaptable. However, if we want our innovation system to serve the clean tech revolution as effectively as it served the IT and biotech revolutions before it, we need to give it a tune up, stem to stern. I would like to suggest five areas where we could make adjustments right away.

The first comes directly from our experience at MIT: University researchers pursuing sustainable energy alternatives should partner more actively with industry. We all know that the energy incumbents and the aspiring energy newcomers often seem to speak entirely different languages. If we want clean tech to succeed at scale in our lifetimes, we must draw on the market expertise of the companies that already deliver energy to billions. At MIT, we wanted our energy work to have rapid impact on the energy marketplace, so from the start, we sought long-term strategic industry partnerships. We needed the energy industries’ deep knowledge of global energy markets. At the same time, our industry partners quickly saw that they could gain much from MIT’s research capabilities, as well as from our record of developing game-changing technologies and transferring them to the marketplace.

As a result, we have partnered with more than 60 industry sponsors who together have provided more than $350 million in funding to support nearly 300 faculty and senior researchers working to invent a sustainable energy future. This openness to industry has implications for federal research, as well. DARPA has long emphasized a “hybrid” model that brings breakthrough university research into collaboration with industry research, particularly through smaller firms and startups that can work on implementation and engineering. ARPA-E understands this model, and we need to propagate it broadly.

Second, government funding agencies also need to take seriously the market dimensions of sustainable energy. I know that Dr. Majumdar and Secretary Chu both understand this well. But for decades, DOE operated as a collection of silos, some focused on frontier science, others focused on applications, but neither paying sufficient attention to the demands of the marketplace. This may be one of the reasons that 30 years of 20th century energy research produced so few scaleable technologies. Could we start to see connections between the lonely silos?

I am delighted to see ARPA-E actively changing the long-standing DOE model. DOE is pushing aggressively to put the best university talent on the task of advancing the science through 46 new Energy Frontier Research Centers. DOE understands that to take these technologies to scale will require the resources of more than any single research group, and so the agency has created three, and we hope soon five, Energy Innovation Hubs. Ideally, ARPA-E would function as a connector: spotting the best breakthrough ideas coming out of the Frontier Centers, then accelerating them for three to five years to get them into development and well on the road to commercialization, and then handing them off to the Hubs and the applied programs that work with industry. Importantly, they now expressly reward research teams for pursuing the real-world problem of going to market, with cost defined as part of the technical problem to be solved. Together, these programs constitute an entirely new model in government energy funding, one that should be further encouraged across all the offices and agencies that fund energy research, and DOE’s leadership deserves the credit for getting it started.

Third, we should seize the unexploited opportunities to use the sheer scale of government itself as a mechanism for testing new ideas. As one example, consider building technologies. We all know that buildings account for about 40% of total energy consumption. However the construction business is overwhelmingly composed of small operators who lack the capital and the research capacity to develop new techniques or technologies. However, the Department of Defense controls the largest block of buildings in the nation, with 507 installations, in every kind of geography and climate. In a period of cost cutting, improving the energy efficiency of the military’s 300,000 buildings represents a huge potential cost savings. What’s more, unlike DOE, DOD has experience with test beds; it routinely creates initial markets for technologies it wants to develop.

Each year, DOD receives $18-to-$20 billion in appropriations for new and rehabbed construction. Could we leverage this funding to pioneer technology solutions that DOD urgently wants? Could we have a cross-agency program between DOD and DOE? We should all celebrate the fact that, today, DOD and DOE are indeed actually talking to each other. Let us press for more talking and for even more shared action.

Fourth, if we want new energy technologies that can serve millions or even billions of customers, we must drive down the cost of production, and that means tackling the problem of manufacturing. We are all familiar with America’s innovation system: the federal government funds breakthrough university research, which migrates to innovative companies, which produce breakthrough products. Even now, no one in the world is better at inventing the new. However there is another area, one that receives scant public attention but that is ripe for exciting gains through innovation: our remarkably unsystematic system for remaking how we make things. This system, mainly in the hands of small- and mid-sized manufacturers, focuses not on breakthroughs but on incremental engineering advances in manufacturing technologies and processes, geared to driving down costs and steadily increasing product efficiency and quality.

ARPA-E and our research universities are busy leading us to energy breakthroughs. Yet given that cost has become the do-or-die gateway to market success, we need manufacturing breakthroughs, too, and we cannot afford to assume that the current incremental system will somehow take care of itself. Suppose, however, we brought the talent of the breakthrough system to bear on the incremental system? Suppose we linked the two? Even beyond energy, I am convinced that our nation’s economic prospects hinge on our ability to achieve a new generation of production breakthroughs; the good news is that energy technologies would be an obvious early beneficiary. We simply will not get the benefit of breakthrough energy technologies without breakthroughs in production, so we must see the two tasks as one.

Fifth and finally, we need to attend to the back end of the innovation system: the financing necessary for new technologies to succeed outside the lab. If you had asked me a year ago, I might have been pessimistic, and some people persist in describing clean tech as an investment bust. However, recently released 2011 data show that clean tech simply stumbled in parallel with the whole economy; with the economy now gaining strength, dollars are streaming back to the promise of clean tech, too. In 2011, venture capital funding to clean tech companies not only recovered from the trough of 2009, it reached an all-time high of $4.3 billion. Clean tech has also maintained a relatively steady share of total investments, accounting for 15% of total VC dollars in 2011. That is not nearly enough, but we are not in the midst of an investment crash, either. Nevertheless, it is pretty clear that the VC industry will have to adjust its expectations to adapt to the realities of clean tech. The model that worked for IT clearly does not directly apply: energy technology will never be a five-years-to-IPO business, so the market needs to evolve viable investment models for long-cycle, capital-intensive industries. I am, however, extremely encouraged to see the resurgence of interest and dollars. At the same time, we also need to consider whole new models for financing the energy technologies we need. MIT Professor Richard Lester’s and George Mason University Professor David Hart’s new book, Unlocking Energy Innovation, describes a regionally based financing approach, with multiple new stakeholders. MIT Professor Andrew Lo has envisioned a new financing model for advanced technologies that would bring in a much larger group of investors, making modest investments in wide-ranging technology portfolios, to better manage the inherent risk. With ideas like these in play, I have great faith in the adaptive creativity of the marketplace.

On the stage where we confront the great and growing global energy challenge, the scenery and the players are always changing. The backdrop, however, remains the same: to respond to the pressing concerns of national security, growing global demand, environmental degradation, and economic growth, we must design a sustainable energy future – quickly.

I have outlined some immediate first steps: First, universities and government should shake off their trepidations about working with both the incumbent and emerging energy industry. Second, the rest of DOE, and frankly the rest of government, should follow ARPA-E’s lead in defining competitive cost as central to the clean-tech innovation equation. Third, we should figure out how to use the scale of government energy consumption, especially through DOD, to provide test beds for important new technologies. Fourth, we should seize the opportunities of advanced manufacturing to help change the cost equation for new energy technologies. And fifth, we need to develop a range of financing models that will allow energy technologies to develop and thrive in the market.

This represents the beginning of our shared to-do list, and I actually omitted a number of areas also crying out for attention, like the need to develop multi-disciplinary, highly analytical curricula to educate the next generation of energy pioneers to supplement a rapidly graying workforce. Some may also find it surprising that this list does not include securing a dramatic increase in federal funding for energy research. We would all welcome that, of course. In a time of fiscal austerity, however, we should focus on making the very best use of the resources we currently control, and I have confidence that success will catalyze amplified support.

With all the drive and talent in this room, and in all the groups you represent, we have the power to invent the future we need. If we can only advance these new models, with unrelenting creativity and, in Dr. Majumdar’s phrase, with fierce urgency, I know we can change the game.