Peter Kelly-Detwiler: 617.875.6575 | Leighton Wolffe: 781.547.1193 pkd@northbridgeep.com, leighton@northbridgeep.com

The Advanced Research Projects Agency (ARPA-E) recently celebrated its seventh anniversary this year and held its seventh annual innovation summit at National Harbor outside Washington D.C. For those of you who are unfamiliar with ARPA-E, it is an agency within the U.S. Department of Energy that is modeled on the Defense Advanced Research Projects Agency (DARPA).

DARPA supports all kinds of defense technologies from unmanned underwater vehicles to cybersecurity. However, while DARPA is over a half-century old, ARPA-E only recently came into existence following an urgent warning in 2007 from the National Academies that the U.S. was losing its competitive edge in science and technology. That report specifically recommended the formation of the agency and in 2007 Congress passed – and President George W. Bush approved – the official creation of ARPA-E. Two years later, Congress appropriated  – and President Obama authorized – $400 million for the agency.

Seven years down the road, and during the current period of intense political debate concerning the role of the federal government in many areas, it would seem an appropriate time to evaluate the current strategy and get a sense as to what has been accomplished. To that end, I lined up a conversation with Director Ellen Williams WMB -3.02% to get her perspective on the organization she currently leads.

Image: ARPA-E - the innovation summit

Image: ARPA-E – the innovation summit

 The first thing to understand is just what ARPA-E is supposed to do. Its formal mandate – as stated on its website – is to advance “high-potential, high-impact energy technologies that are too early for private-sector investment. ARPA-E awardees are unique because they are developing entirely new ways to generate, store, and use energy.” In other words, its goal is to help new energy companies cross that difficult chasm from promising technology towards commercialization, by de-risking the technologies and improving the likelihood that they will eventually be supported by investors in the private sector.

Williams noted that this critical transition research to commercialization often doesn’t happen without government support.

“There are certain types of activities that the private sector will not do. The federal government funds a lot of blue sky research. The people who do it are amazingly brilliant, but they are basic researchers. Our job is to take these amazing early stage ideas and invest in them to see what has real legs and practical impacts in the commercial sector. They are too early and too high risk for the private sector to invest in. We are careful to invest in things that are too uncertain that industry just won’t do it. We are moving the technologies from early stage through a de-risking process, and then when industry is ready to invest we can hand off and they will invest.”

So how does that process of technical midwifery work in practice? Williams commented that it starts with the constant need to “be on the lookout for innovation, and we have a mandate to be on the lookout for new technology in every sector.” Each year ARPA-E’s program directors propose six or seven new program areas that they feel should be addressed. These are wide ranging, and could be anything from new approaches to generating wind, to new software tools, or novel technologies to store energy.

Williams observed that her agency is responsible for addressing the entire energy sector, and there is a rigorous assessment process involved. Once there is some agreement among the program directors on the specific topics,

“They go talk to experts in different areas, and once they are convinced there may be something interesting to do, they hold workshops with experts around the country to have detailed discussions about the needs and opportunities in a technology area.”

ARPA’s 2017 budget document (the agency requested $350 million for 2017, and received $291 million in 2016) states that new programs must “be based on significant potential for transformational technological innovation. The technical opportunity must be too early stage or currently too high-risk for commercial investment.” It must also have an impact on the areas specified in the legislation: “Improving Energy Efficiency, Reducing Dependence on Energy Imports, and Reducing Harmful Energy Emissions, specifically and critically including reduction of greenhouse gas emissions.”

In addition, it cannot duplicate work being done in other federal agencies and – critically – “There must be a pathway for advancing the technology toward hand-off to the private sector for commercial development and corresponding benefits to the economic and energy security of the U.S.”

At the end of the day, this is all about gestating and delivering technologies for ultimate adoption by the private sector. A good example of how this works in real practice is a recently-launched program run by Dr. Chris Atkinson related to driverless vehicles and automation. Although the general focus in this area had previously been on safety and efficiency, Williams noted,

“Chris took a look and realized if you have automation there is information coming to each vehicle telling it what the cars around it are doing, about slowdowns, and hills ahead.”

According to Williams, Atkinson realized that this data – if properly utilized – could actually be used to optimize how the vehicle itself was being propelled.

“You have a tremendous amount of information that will allow you to tailor your drivetrain to operate at a sweet spot. So the new program is ‘NEXT-Generation Energy Technologies for Connected and Automated on-Road-vehicles (NEXTCAR) and is tailored to address this question as to whether we can use automation information to modify how an engine runs and get extra efficiency output that way.”

Once the program themes are identified, the next step is to put out a call to companies who are interested in developing – or are currently developing – technologies in these areas that need assistance. To that end, the agency makes frequent funding opportunity announcements, inviting companies to apply for support. The average award is approximately $2.75 million (to date the agency has awarded $1.3 billion in 475 different projects).

Williams stated that once the recipient gets the money, the multi-year process is just getting started.

“All of our programs are set up from the beginning with really rigorous commercial and economic goals. At the beginning, we set technological targets and commercial targets. It’s not good enough to produce an outstanding widget. That widget has to perform a function better than any existing widget and has to be economically attractive.”

Each project team is assigned milestones and a program director with expertise in that sector, as well as a commercialization advisor. The ARPA-E advisors meet with their company teams on regular quarterly visits. Right from the beginning they are given these milestones and assigned a program director and commercialization advisor.

Sometimes it becomes obvious fairly early on in the process that the technology or the team isn’t going to achieve success. The key then is to recognize that fact, fail fast, and cut funding early. Objectivity is important.

“To prevent emotional involvement, we set out milestones at the very beginning and they have to meet milestones every quarter. We do find that in 5-10% of cases the technology that looked like it might be good to go is different and there are problems we realize that cannot be overcome. Then generally those terminations are done by mutual consent.”

In those cases, teams that terminate projects write up a report, highlighting lessons learned. Failure is not always a bad thing, though, and it’s a necessary part of the innovation and commercialization process. If the team were successful with 100% of the projects, it would mean they simply weren’t pushing the envelope far enough.

Overall, then, has the agency met its goals? Is it a good investment of the taxpayer’s resources? One way to answer the question is to look at simple obvious return on investment: To date, ARPA-E has invested $1.3 billion in 475 projects. Fewer than a tenth have gone on to success, with 45 projects securing $1.25 billion in follow-on financing for 36 new companies. So on the face of it, that would look like about a break-even. But is it? Only if you apply a relatively short-term focus on the money, and ignore a number of longer-view critical elements that are essential to the process of fostering innovation and commercialization of technology:

  • The agency has advanced applied science in ways that otherwise simply could not have been achieved.
  • It has educated thousands of our nations best tinkerers and entrepreneurs and made them smarter about what works commercially and what doesn’t, how to develop a business plan, and how to think like business people. Many now better understand the rigor of bringing technology to market.
  • The follow-on private sector funding is by no means complete at $1.25 billion. Additional funding rounds are likely to follow in the years to come.
  • The IP being developed and commercialized will also have positive knock-on benefits for years.
  • New companies and industries drive economic development and create jobs. Williams comments that the agency has not done a full-scale employment analysis, but anecdotally, they can connect the dots to some degree. For example, FastCap – a company that manufactures performance capacitors – started with two people in the company and the workforce has now burgeoned to 30 people.

As far as major successes, Williams points to 1366 technologies – a Massachusetts-based company that significantly reduces the cost of wafers in solar panels (and a company I wrote about in November of 2015). Williams comments “1366 is a real winner for the U.S. economy. It’s going to bring back to us a (solar) manufacturing sector that has long disappeared from the U.S.”

To find out what it looks like from the other side of the equation, I contacted Frank van Mierlo, CEO at 1366 Technologies. He was in an upbeat mood, fresh on the heels of an infusion of a $10 million investment from South Korea’s Hanwha Investment Corporation, to be allocated for construction of a commercial-scale factory which is scheduled to be producing solar wafers in the U.S. by 2017.

Van Mierlo commented that ARPA-E was “refreshing to deal with.” They had “a new team with excellent people that really set them apart. They were really top-notch individuals and they all had Ph.D.s from the best universities…they were commercially agile and a fast moving team.” Not a bad reference…

In addition to the ARPA-E advisory team and the $4 million in funding, the other aspect van Mierlo found helpful – and bracing – was that

“the competition was incredibly tough. Only one percent of the applicants received funding. That was tremendous validation. It was important to say ‘we were put through the wringer and examined by the best and brightest.’ It moved us from the process stage to an actual pilot…It was instrumental in securing our ($30 million) B round and put us on a real path.”

To date, 1366 Technologies has raised $80 million in private equity. It has also grown in size from a dozen employees to 60 at last count. When the 250-megawatt factory is built next year, it is expected to employ 150 workers (with additional positive employment multiplier effects for the supply chain). Long-term, the company has plans to employ thousands at its North American factories.

So on balance, the short-term numbers for ARPA-E may look like a close to break-even value proposition. But that perspective would be short-sighted. If one adopts a longer-term view and evaluates America’s position in the hyper-competitive race to assume the commanding heights of our global energy transformation, ARPA-E would seem to be a necessary and solid investment in our collective national energy future.