It's not a popular thing to argue right now, but yes, there certainly is a vital role for government in support of cleantech innovation.
Let me start by acknowledging that I absolutely understand and quite often agree with the sentiment that government shouldn't be in the business of venture capital. Market-based policies are, to my liking, almost always preferable to policies where government employees select and fund specific innovators or companies. Thus, the most effective way for governments to support cleantech innovation would be to price in externalities like climate change and dependence on imported energy, and then let the market sort things out.
But that's not likely to happen anytime soon in the U.S.
So in the meantime, how should we view policymakers' efforts to promote innovation and startups that would otherwise languish in today's skewed pricing environment? It's a very complex problem, which I'll illustrate below.
I would propose that an appropriate government effort to promote cleantech innovation will follow three core principles:
Within the realm of desirable outcomes for the U.S. economy and energy mix, some are already being tackled by the private sector, and some are not. As a rule, government efforts even in economically beneficial innovation areas shouldn't duplicate the existing efforts of the private sector. In this case, the government should be aiming to fill capital gaps.
This is a lot easier said than done, however. First of all, how do you define a capital gap? It's hard enough to get reliable information about venture funding overall, much less get reliable data about what specific stages and sectors and business models are hard to get funded. Secondly, these capital gaps also wax, wane, and shift as investor enthusiasm for a sector ebbs and flows. Thirdly, in some of these capital gaps, there might be non-institutional investors who are interested in putting in money where the institutional VCs won't, and the capital gap is one of check size (angels writing $25,000 checks when $250,000 is what's needed), not one that's sector- or stage-driven. And finally, there might even be gaps across various efforts within an individual company -- for example, a research project at a venture-backed startup, where the research wouldn't have happened with VC dollars alone.
You also want to avoid either having these government-directed dollars flowing into the hot new startup (because the program is under pressure to show some 'successes' early on), or a negative selection bias where government dollars are being directed away from the best innovative startups because VCs might become interested. But you do want to see leverage resulting from these efforts, where the government program helps to unlock follow-on capital from the private sector. In other words, VC dollars following government dollars is one indicative metric of a successful program. But government dollars following VC dollars is less so (with the caveat of the additional research project scenario as described above).
How to prove additionality under such complex conditions? It doesn't make sense to attempt to codify a definition legislatively. Rather, the best approach is to to develop flexible processes that are outcome-oriented and provide opportunities for an evolution of the program over time.
One of the less-reported but damaging trends in U.S. policy right now is the shift in tactical control from the executive branch to the legislative branch. In other words, lawmakers are writing in too many detailed directives as to how programs should work, instead of stepping back. Because if there's one group of people you don't want designing your management process, it's Congress. Yet that's exactly what's going on. Many of the more public examples of failed energy policy recently were, at heart, driven by byzantine program designs dictated from Capitol Hill, not in the White House or inside the DOE. It's the equivalent of a startup's board of directors going around the CEO to specify how junior engineers in the R&D group should organize their daily activities. Lawmakers should establish goals, provide boundaries and oversight, and then let the implementers on the firing lines figure out the best way to accomplish these objectives.
If lawmakers are to surrender some control over process, however, there does need to be sufficient oversight, as well as a forcing function to make sure the program really does evolve over time as market conditions change.
3. The Voice of the Market
To reiterate, I strongly prefer market-based policies to any effort to have government groups select some specific recipients over others. But if there are some areas where economic imperatives necessitate this latter type of policy, the next best choice is to at least give the voice of the market a major role.
Peer review is one commonly used way to get at this. And if done the right way, it can be quite valuable. But if done wrong, of course, it's useless. A good peer review process will work hard to get a wide variety of knowledgeable technical and market viewpoints, control for potential conflicts of interest, and include force-ranking or other ways of making sure 'grade inflation' doesn't creep in.
What I haven't seen as much of, but what would also be valuable, would be to have additional reviews done by groups of end-users or customers. This is tougher, and won't be definitive, because quite often customers don't know what they've been missing until they see something in action. But still, for building energy efficiency technology (for example), there should be collective reviews by large building owners who could eventually be purchasing the technology, both within the government system (e.g., let's get the Navy helping ARPA-E select what technologies they'd love to have commercialized for use on their bases) and, of course, really including the private sector owners of buildings. It will be important to avoid inclusion of channel partners who seem like purchasers but really are biased distributors (sorry, ESCOs). It won't be a perfect process, but these programs shouldn't be "build it and they will come" by design -- they should be bringing in the voice of the customer right from the point of selection. And the input and perspective will help all participating startups, even those not selected by the program.
Outside perspectives from market participants can, at least in a wisdom of crowds format, also help identify which areas are truly capital gaps and which aren't. Perhaps doing this on a case-by-case basis would be unwieldy, but at the very least, a large panel of advisors from the private sector could help evaluate the program staff's own identification and definition of capital gaps on an annual basis, helping to validate the strategy while keeping it somewhat flexible. It would have to be a large and diverse panel -- each advisor would absolutely walk in with their own biases in certain sectors and categories, so it will be important to flood out each bias with enough of a crowd of perspectives.
Bringing this all together, I'll throw out there a vision of what this might look like:
- A program given a concrete and ambitious but relatively broad set of mandates in terms of goals (such as $1.00 per watt installed solar cost), with an authorized budget and department home, and some basic parameters around types of activities (grants vs. investments, etc.) and a requirement to demonstrate additionality.
- An oversight committee to review the program's activities on a quarterly basis and strategy/program design on an annual basis.
- An auditor established elsewhere in the government to evaluate adherence and effectiveness of the program on an annual basis.
- A strong manager out of the private sector with a mandate to hire a top-notch team, tasked with designing and running the program.
- Peer review and market input processes so that as much as possible of the program's activities are guided by the voice of the market, while still giving the internal team latitude to use their experience and judgment toward meeting the goals of the program.
Obviously, a fourth unstated principle, therefore, is that this requires a really sharp team. Not a bureaucrat-laden team, but a team of experienced managers with private-sector experience, leavened with smart young technologists and market analyst types. This is the single most important determinant, in my mind, between a successful effort like this and an unsuccessful effort. Groups like ARPA-E and the Massachusetts Clean Energy Center have had leadership like this, and it's a major reason why (in my mind, at least) they've been pretty successful to date. Effective, mission-oriented people have been brought on board by the leadership of each program, and they've found a way to make a positive impact.
But will that last? I do worry that the crushing nature of politics means that such people will eventually be driven out by partisanship-inspired attacks. I'm not sure that can be avoided (unless anyone has any brilliant ideas on how to remove party politics from energy policy in this country?). So we can only hope that effective businesspeople with tough hides will continue to agree to lead programs like this out of a sense of mission, and that they will be able to continue to inspire strong managers and analysts to come join these teams even though they will have to expect that it becomes unpleasant at times.
In the meantime, we can hope to continue to see broader recognition of the need for market-based energy policies that will establish a more even playing field, so that all of the above becomes less necessary. Because right now, it's absolutely critical.