I thought it would be a good "New Year" exercise to update the FEST map and add some thoughts on each climate tech sector.

As a reminder, FEST is a framework, inspired by Bill Gates' book, that illustrates the impact of climate tech on the different parts of how we live. Emissions can broadly be divided into Food (F), Energy (E), Stuff (S), and Transport (T).

I use this chart as a reference for looking at the different climate technologies in the context of the broader picture. It's by no means comprehensive...at least not yet! This current version has ~170 nodes, with each node representing a niche of climate technologies.

Updating this has generated some interesting ideas for me of what to write about this year. I hope it's as helpful for you as it is for me.

ETM-FEST-Map-2-correctedDownload

Transport

Stuff
Energy

Ag/Food

Carbon

Happy New Year!

I spent some time over the holidays making a fun little reference timeline for the IRA credits. The various effective dates can be confusing and I figure it would be helpful to have a birds eye view of when each provision comes into play.

From this angle, we can observe that there are several critical deadlines in which a good chunk of the new IRA credits come into effect or go out of effect: the beginning of this year, 1/1/2023, 1/1/2025, and 1/1/2033.

There are still certain provisions that rely on federal guidance before they are in effect, something that can affect actual deadlines. For example, the guidance for the critical minerals and components requirements for clean vehicles was supposed to be issued by 12/31/2022. Due to a delay in the guidance until March 2023, certain vehicles that qualify for the credit without taking this requirement into consideration can now qualify for the credit until the guidance is issued (like Tesla).

Check out the pdf or spreadsheet versions below and as always, let me know if you have any feedback.

ETM-IRAcalendar1.19.23Download

View full sized spreadsheet

Full tax code with the IRA adjustments available from Tax Notes

Past IRA analysis / summaries:

There’s been a LOT of money poured into climatetech over the last few years and even with the recent downturn, new money keeps coming in. According to CTVC, 2022 has seen an uptick in the number of new climatetech VC investors announced (45 -> 47), despite rising interest rates and a broader market slowdown for VC. It seems that the LPs of the world – endowments, pension funds, sovereign wealth funds, mega-family offices – are still quite bullish on climate and hungry for new early stage funds in this space. And as climatetech deal activity (as measured by deal count) also hasn’t slowed, new fund managers continue to have opportunities to deploy capital in.

So the environment for being an investor in climatetech seems to be as good as ever these days…but has it evolved at all in other ways? That’s the question I was asking myself after I read this update. I suspected that the VCs that are raising money today in climatetech are of a slightly different flavor than the VCs raising money in climatetech a few years ago.

Examining the climatetech-oriented or related funds that have been raised over the last 6 years, I found that there’s been an interesting evolution that’s been happening when it comes to sector specialization. While most of the funds raised in the early days of climatetech / cleantech were generalist in nature, recent funds seem to have erred to choosing specific niches within climatetech like alternative proteins, buildings, or circular economy.

A taste of some of the announcements this year:

A total of 29 specialist climate funds have been launched this year, up 45% from last year. That’s out of a total of 106 specialist climate funds that are actively investing. These 106 funds span 14 specialties which can be aggregated into 4 main categories. All of this is shown in the chart below.

Some observations:

  1. Early specialist funds were concentrated on new ag/food and mobility. In the last two years, specialist funds cover a more diverse set of specialties.
  2. These funds don’t necessarily have to be completely new funds. 11% of these new specialist funds come from existing VCs with a more generalist or broader focus.
  3. This dataset only shows institutional specialist funds. Other specialist funds naturally exist through corporate venture capital, where strategics typically aim to invest in specialties that can complement or overlap with their existing business. Having more institutional specialist funds blurs the line a bit between the value add from a CVC and the value add from a VC, creating competitive pressure for the CVCs to add outsized strategic value or offer better terms in a transaction.

Overall, specialization is a positive development for the climatetech world. It’s at the very least a signal that climatetech is maturing as an investment category and fund managers are getting more sophisticated. It’s also a good sign that investors are seeing ample opportunities in climatetech for the long term.

The consequences of specialization are net positive. Having more specialized VCs places additional pressure on other VCs to think about finding a differentiated angle to offer to startups. This can push the investor world to developing better, more value-added relationships with their investments. For startups, having more options from specialist VCs increases competition in their favor, while taking money from specialist VCs can offer more strategic channels to the right industry network.

The other side of specialization is that specialist funds have limited options when it comes to pivoting in a challenging macro environment, leaving them vulnerable to industry-level macro shifts. Choosing to specialize is taking a risk that your chosen specialty will be around for years to come. Having an increasing proportion of specialist VCs in the climatetech ecosystem means that the overall capital stack for climatetech has more points of vulnerability.

But all in all, there are benefits of specialization to both the investors and startups.

For investors, specializing allows for:  

  1. Easier deal screening. Less deals need to be vetted in a smaller market, allowing for more time to be spent with fewer companies. Having a public sector-specific mandate also means startups working in that area are more likely to know about you and proactively reach out.
  2. More thoughtful deal screening. Certain sectors may have sector-specific metrics that may be more relevant than traditional VC metrics like revenue or EBITDA. Things like quality of patents, technical benchmarks, partnership agreements, customer indications, etc. can be assessed in more depth in deal screening. For many VCs, especially ones coming in from tech to climatetech, using criteria that is too generalist may exclude startups that are expressing momentum in other ways.
  3. More leverage in attracting “hot” startups. In competitive, oversubscribed rounds, investors with clear differentiation and strategic value often have more appeal to startups than other investors, even those with more favorable valuation/terms. Similarly, a niche focus can help investors without a substantial track record argue their value to startups.
  4. Portfolio synergies. Investors that specialize are better able to aggregate resources particular to that specialty, like technical help, advisors, corporate partners, customer relationships, etc., for the broader portfolio, allowing startups in that portfolio to access those resources at lower cost. A specialized portfolio that brings together companies working around the same sector can also potentially find ways for companies in the portfolio to work together, officially or unofficially.

For startups, specialist funds can allow for:

  1. Easier investor screening. Most capital raises involve conversations with a hundred or so investors, with half of those being immediate dead ends because of lack of transparency by the investor on what problem areas they are interested in. Specialist VCs are very clear on this aspect, allowing startups to schedule those conversations without fear that time will be significantly wasted.
  2. Reduced competition with deals in other industries. Pitching to a generalist fund or a fund with a broad climate mandate can mean that your startup is being compared to other startups that have a totally different growth model. Investing in a materials science company selling into the auto market is worlds apart from investing in a carbon credit software company selling to consumers. Evaluating these deals together often time favors a certain industry or growth model instead of favoring the better company.
  3. More value-added partnerships. Specialist VCs often have more relationships within industry, which means that startups that partner with specialist VCs can access more relevant resources that can help them scale. Specialist VCs are likely also better able to contribute meaningfully to accelerating the startup’s growth as board members or advisors, making them better partners in the long run.
  4. Easier access to subject matter experts. Specialist VCs attract other specialist service providers and specialist partners, which means taking on a specialist VC can naturally allow a startup access to a wide array of subject matter experts in their sector. This can make the build process on a niche problem a lot less painful.

TLDR; we’re seeing more specialist VCs in climatetech. And that’s a good thing for the ecosystem.

Last time I looked at the history of wave energy and what lessons we can draw from past cycles that could be relevant to today’s climatetech challenges.

This week I wanted to create the actual wave energy market map, in part to get a better sense of the variety of wave energy converters (WEC) that are being developed today and in part to better know and appreciate all of the companies that are working in such a difficult but rewarding space.

Though there have been hundreds of WEC designs proposed in the last few decades (130+ are included in The Liquid Grid’s database and 94 in Net Zero Insights' database), only a small fraction of those are still being developed today. I found 25 that are actively on the path towards commercialization (loosely defined to be: 1- not just a project by an academic institution and 2- having at least one announced milestone in the last year). Those 25 land in six different WEC categories that I could identify.

Included in this map:

A few observations on this set:

It's also worth noting that I made some purposeful exclusions in this landscape.

  1. I excluded businesses that combined their WEC design with a specific use case. Ocean Oasis and Oneka Water are using wave energy for desalination, E-Wave is developing WEC systems for aquaculture, and Ocean Motion is developing its WEC for ocean observation buoys. These are worth noting but because of their focused markets, I didn’t think they should be compared to companies harvesting wave energy for general electricity.
  2. A portion of ocean energy to be harvested includes non-wave kinetic energy from currents and tides. Companies like Current Power Energy Solutions are targeting ocean currents while companies like Nova Innovation are targeting tides. These are worth their own deep dive at some point.

After looking at the breadth of ocean-based climate solutions last week and having a better understanding around the potential of the ocean, I thought it might be informative to look at wave energy more specifically.

Four things that are worth noting about the wave energy space:

  1. The resource is huge. As mentioned last week, wave energy has the potential to be a critical asset for clean power generation. Wave energy alone can fulfill 67% of current US electricity demand (2,640 TWh / 3,930 TWh) and 140% of current world electricity demand (32,000 TWh / 22,848 TWh). By 2050, those numbers will be 51% (2,640 TWh / 5,138 TWh) and 71% (32,000 TWh / 45,000 TWh). If looking at just the population of people along coastlines, wave energy can potentially fill all of that power demand.

  2. It’s incredibly technical. Because waves have both an amplitude and period, the capture of their associated kinetic energy requires much more than simply having the waves pass through a turbine. Designs need to transfer the motion of a wave, something that is non-linear in multiple directions, into movement that is linear/unidirectional for a generating system like a turbine, all with energy loss minimized as much as possible. This is very difficult with fluids that move as chaotically as the water in the open ocean. Minute Earth explains it well here.

    In addition to the technical challenges of actually capturing the energy, there’s the design challenge of building a device that is resilient against storms, biofouling, and corrosion. A device that is optimized for kinetic energy capture may not be optimized for the occasional violent wave or high wind event. A device that is optimized to do both may have an expensive and impractical maintenance schedule. (Luckily, automation and IoT have helped wave energy converter developers set up systems that respond immediately to potentially disastrous circumstances and alert those onshore of any components that need maintenance.)

    All of these engineering challenges need to also be addressed with a design that is cost efficient, further increasing the difficulty.

  3. The combination of the above has created a large ecosystem of creative and diverse wave energy converter designs. The Liquid Grid lists out 130+ different wave energy converters, the vast majority of which are being developed by separate companies. Most fall into seven categories (point adsorbers, terminators, oscillating water columns, attenuators, oscillating wave surge, submerged pressure differential, and rotating mass) but designs can be very different even within each category. For example, though both ExoWave and AW-Energy are classified as oscillating wave surge, one uses moving cones while the other uses a large moving panel. I’ll take some time next week to map out the different wave energy startups with active designs, but for now, know that they are quite diverse in their approaches.

  4. History has been unkind. Wave energy actually has been around for as long as other renewable technologies like solar and wind, with its first patent tracing back to the late 18th century. Its failure to commercialize as quickly as other technologies can be attributed in part to unfortunate historical dynamics.

    First, the competition with wind energy placed pressure on wave to commercialize too quickly during the 1970s oil embargo. Prior to the 1970s, wave energy had actually been on a steady rise. Hundreds of patents had been filed for new wave energy converters throughout the early 20th century and significant advances were made by Yoshio Masuda, the father of modern wave power, in the 1940s and 50s.

    The oil embargo of the 1970s spurred a step change in interest in renewables, including wave energy. On the one hand, it incentivized many countries like the UK and Norway and academic institutions like the University of Edinburgh and MIT to start seriously participating in wave energy research. On the other hand, it led to rapid investment in technologies that achieved first commercial, leaving slower-developing technologies behind. For wind, these commercialization milestones were hit in a short period of time. Between the oil embargo in 1973 and the end of high oil prices in the early 1980s, the first US wind farm was put online (1975), the first multi-megawatt turbine was produced (1978), the first commercial wind farm was deployed in New Hampshire (1980), the levelized cost of wind reached $0.38/kWh (1980), several commercial turbine manufacturers were founded (1980 – 1986), and more than a dozen other wind farms were deployed in the US.

    For wave, the story played out a big differently. Though exciting developments occurred during this time period, including the invention of famous wave energy converter Salter’s Duck in 1974, the emergence of critical studies for wave energy design, and the establishment of many wave energy-dedicated international conferences in the late 1970s/early 1980s, no technology ever made it beyond the small-scale prototyping stage (even Salter’s Duck). By the mid-1980s, lower and lower oil prices and competing nuclear energy programs led to wave energy programs being shut down and funding being pulled from key technologies. Unfortunately, because wave never hit first commercial, technology development stopped once macro tailwinds disappeared. And because wind had already hit its stride, those that were keen on continuing renewables investment flocked to the winning horse. For wave energy, the window of opportunity vanished before momentum hit.

    Wave energy got its second run in the new millennium, but again, market forces led to pre-mature conclusion. 2000 saw the world’s first commercial wave power device in Scotland by a company called Wavegen. This was followed by a number of other key milestones for wave by Pelamis and Aquamarine Power. In 2004, Pelamis tested a full-scale prototype at a wave test site in Scotland and later, in 2008, deployed a 2.25MW commercial farm in Portugal, the first of its kind. Similarly, Aquamarine Power deployed two full-scale prototypes (315kW and 800kW) in 2009 and 2012, respectively. But again, both companies eventually failed because of market timing. Pelamis’ farm was shut down due to technical problems with the generators and lack of funding from the financial crisis to redeploy them. Pelamis later went under in 2014 after failing to acquire more funding (a product of the cleantech 1.0 bust). Aquamarine, despite two successful prototypes and a recently won government award, failed in its 2014/15 raise, resulting in the shutdown of the company. Thus a combination of unfortunate timings around two major macro busts (recession and cleantech 1.0 blowup) cooled down developments in wave energy yet again.

These four characteristics of wave energy – the fact that the market is huge, the problem is hard, the technologies are diverse, and the history is long – make it an interesting case study for other hard tech climatetech spaces, especially those that are going through a similar Cambrian explosion of technology. There are lessons that we can take from looking at this area for not only the future of wave energy development but for the future of other climatetech sectors with a similar profile.

Here's what I observed:

  1. The wave energy industry depended too much on government funding. One issue with why the first cycle for wave failed was that technologies were dependent on ephemeral wave energy programs set up by governments whose agendas eventually pivoted. While government funding has led the way on development thus far, history shows us that in order to keep this current cycle going, we need investment from private dollars not subject to capricious political budgets.
     
  2. The industry was balanced on too few demonstration projects, leading to large losses for the entire industry when those projects went bust. This is especially evident in the second cycle, when wave energy’s renaissance was led by only two major flagship projects: Pelamis and Aquamarine Power. The loss of both of those in a short time period was a huge blow to the wave energy industry. In the future, we need to make sure funding is distributed across many different flagship projects for the overall health of the industry.

    The other takeaway from this point is that licensing out the technologies to developing entities can help reduce the risk that a technology will die when a company dies. Perhaps we need to be pushing for more licensing models across R&D-heavy climatetech sectors to better the chances of an industry’s longevity.

  3. Having access to sufficient testing sites with infrastructure that made up the balance of plant was critical for companies being able to iterate their technologies before first commercial. There are more testing sites for wave energy than ever these days, thanks to entities like the Oregon State University, University of New Hampshire, EMEC, The Marine Institute, and many more. But there is still a need for more testing sites, especially ones that can be near potential demand sources for wave energy, like aquaculture, desalination plants, or hydrogen production, in order to properly test load requirements. Going to public or philanthropic sources of funding for these testing sites can be a long journey (PacWave took 10+ years). Is there a way we can incentivize private parties and corporates with pre-permitted sites to offer more testing facilities?

  4. Revival of old IP is a worthwhile strategy, especially in industries like this that require significant R&D. Long science can sometimes pay off but can also result in lost IP when market forces prevent commercialization. Our technology / VC-heavy ecosystem places a significant premium on new IP, but perhaps it’s worth checking to see what IP may have been unfairly lost in the past and creating an early stage investment ecosystem that supports old IP revival.

  5. Lack of momentum can have outsized impacts on funding potential. As we saw with wind after the oil embargo, a series of milestones achieved in a short period of time is what allowed it to persist through a competitive oil price environment in the 80s. Lack of that momentum is what led to wave cycle 1’s demise.

    There’s a micro and macro lesson here. On an individual company level, demonstrating consistent momentum can make or break your future funding rounds. In practical terms, that means making sure to announce wins – big and small — on a regular basis.

    Similarly, on an industry level, demonstrating consistent momentum can lead to more funding and longer support for everyone in the space. That’s why celebrating wins from other companies in the industry, even competitors, can be net positive.

Would love to hear if there are other lessons from those that lived and breathed this cycle or that have seen this play out in other sectors.

Look forward to diving into the wave energy market map next week. Stay tuned!

Most of what we talk about in climatetech happens on land – hydrogen production, power generation, carbon sequestration, grid optimization, industrial decarbonization, residential energy efficiency, etc. This is because the majority of technologies get deployed around people’s homes, in city environments, or industrial complexes, all of which largely occur on land.

But what about the ocean?

I became intrigued by the ocean after observing a renaissance of wave energy recently: the DOE announcement on funding for wave energy technologies and a smattering of news for wave energy companies like Atargis, CalWave, Eco Wave, and Mocean.

But wave energy is just one part of a larger suite of ocean-based climate solutions that have emerged over the last few years. The bigger category includes things like tidal energy, mining, regenerative aquaculture, carbon sequestration, low carbon shipping and maritime transport, hydrogen production, offshore solar, and of course, offshore wind.

All of these technologies are in varying stages of development, but most are early stage and require high amounts of capital going forward. Most also face a bevy of challenges being first-of-a-kind in the ocean: long permitting cycles and evolving regulatory requirements, limited testing sites, potential seawater corrosion or storm damage, lack of interconnections, power supply, and other infrastructure, difficult and expensive maintenance, and digital connectivity issues.

Many of these challenges can be tackled with funding, but ocean technologies consistently are deprioritized compared to land-based technologies. Even offshore wind, arguably the golden child for ocean-based technology deployment, is not growing fast enough. Net zero pathways like Princeton’s Net Zero America, BNEF’s Energy Outlook, and John Doerr’s Speed and Scale often build in limited to no allocation for marine solutions outside of offshore wind and ocean protection. Project Drawdown does include additional categories like ocean power, ocean shipping, and improved fisheries, but ranks them all fairly low on the potential impact scale (even offshore wind ranks 38 in a list of 84 ranked solutions, well below its renewables counterparts on land, which take the top 2).

IEA is the only one that seems to give the blue economy some credit. IEA’s ETP list includes 12 different types of overtly (some categories like CO2 storage lump offshore and land-based together) ocean tech and lists the large majority of them as “High” or “Very high” impact (though it’s worth noting that almost 40% of the list of 503 technologies is either “High” or “Very high” impact).

So to summarize, these are early stage technologies that face lots of challenges, require more funding, and are frequently ignored in net zero pathways. What’s the point in putting dollars to work in this area? Why in the world do we need to go through the trouble of funding these ocean-based climate solutions?

Technologies like regenerative / low carbon aquaculture or low carbon shipping are needed to decarbonize industries that we assume will persist in a net zero scenario. So that’s an easy answer.

But for technologies like offshore wind, offshore solar, wave energy, tidal energy, carbon sequestration, seawater and seabed mining, and hydrogen production, there are comparable land-based alternatives that one can argue obviate the need for more complex ocean solutions. A few thoughts on this category:

TLDR; ocean-based climate solutions present a valuable but underappreciated solution set to the world’s climate challenges. The ocean’s plentiful resources, proximity to demand centers, and high co-location potential present a compelling opportunity for both builders and investors in climatetech.

Companies shown: Sweetch Energy, Heimdal, Alora, Orsted, Makai, Atargis, SolarDuck, T-Omega, Lhyfe, Nauticus Robotics, Running Tide, Boundary Layer Technologies, Impossible Metals

This week I had a grand plan to write about the economics of first commercial facility / FOAK commercial and why it makes complete sense to invest in these projects. But it was much more difficult to make the numbers work than anticipated. Even if a company gets over the FOAK hurdle, the lengthy time to exit coupled with the higher capital need results in returns for venture investors far from competitive to regular way venture investments.

There are paths to achieving competitive returns to regular VC – some existing ways include having philanthropic or government capital come in at FOAK stage, which can ‘lever’ up a project with non-dilutive financing, pursuing a licensing model, which, with the right partner, can allow a company to scale faster with less capital requirements, or recruiting an evergreen fund or fund-like operating entity that can provide flexible financing at multiple stages to accelerate the scale up.

But the path that is most murky (or at least was to me when I decided on this topic) is how to achieve competitive returns with a traditional institutional capital cycle.

After experimenting with a rough model, I found that one way you can achieve competitive returns is by creating a FOAK-focused fund and encouraging companies to early exit to this fund. The early exit helps early stage venture investors find liquidity sooner, increasing IRRs and the likelihood of a successful investment. For the FOAK-focused fund, acquiring the company before FOAK means having the ability to capture all future facility economics vs. competing for 2nd facility and beyond economics with very low cost of capital. The likely levered returns past first facility substantially reward the FOAK investor for taking the FOAK risk.

This FOAK fund is not unlike a private equity firm that aims to acquire a company and lever it up before exiting in a few years. The difference is that this FOAK fund would aim for infra-like returns (10 - 15%, or the higher end of infra) at the portfolio level and private equity-like returns (20 - 30%) at the individual investment level, building in some expected failure rate (in the example below, 2 out of 3 investments can fail and the portfolio will be fine) similar to a venture fund to go from the individual investment to portfolio level returns.

The FOAK fund is a novel concept because 1) infra investors are generally extremely risk averse and don’t think about their portfolio in failure terms, 2) private equity investors do balance their portfolio according to expected return but also don’t necessarily build in a failure rate and don’t aim for as modest of a portfolio return, and 3) venture investors do think about failure rate but don’t typically look for opportunities to optimize for cash flow or lever up an investment. A vehicle that combines elements of private equity, infrastructure, and venture capital can address the imperfect match of each one of these traditional vehicles with the FOAK problem.

I don’t think there is something out there in the market today that looks like this…the closest is Generate Capital and their strategy of acquiring, operating, and scaling sustainable infrastructure companies. But I haven’t seen an institutional investor that addresses the FOAK problem. Perhaps someone can correct me here!

Here’s how the returns stack up in this illustrative example:

A typical venture cycle runs 4 - 5 rounds of funding before a company exit. Because of the high failure rate of startups, VC investors target a homerun exit – $1B or more in less than 10 years. Putting in some reasonable assumptions for round sizes and valuations up until that point, we get that, for this individual investment, a Pre-Seed investor can expect a return of 76% IRR / 92x MOIC, a Seed investor can expect a 68% IRR / 39x MOIC, a Series A investor can expect a 69% IRR / 23x MOIC, a Series B investor can expect a 68% IRR / 10x MOIC, and a Series C investor can expect an 88% IRR / 6.7x MOIC.

If we assume these returns for the successful homerun investment, that means that, in order to achieve a minimum of 20% portfolio IRRs, a Pre-Seed investor can have 20 failures to one successful investment, a Seed investor can have 10 failures, a Series A investor can have 7 failures, a Series B investor can have 4 failures, and finally a Series C investor has the least amount of wiggle room with 3 failures.

Moving on to the hard tech example, we assume that the round sizes increase due to greater capital intensity, two more rounds of capital are added to fund additional development, and exit gets prolonged 3 years until the 12th year of the startup’s existence. In this case, the returns are lowered to 40% / 39x for a Pre-Seed, 32% / 16x for seed, 30% / 10x for Series A, 25% / 5.2x for Series B, and 26% / 3.9x for Series C. Series D and E investors get a 20 – 21% or 1.7 – 2.6x return. The failure tolerances are significantly decreased, with Pre-Seed investors only allowed 4 failures to maintain a 20% portfolio return, Seed investors 2 failures, Series A investors 1 failure, and all other later stage investors no failures (i.e. every investment must be a success, a tall order for this type of risk capital).

Now we come onto the early exit case. In this scenario, the company raises a few rounds of capital to develop the technology to the point where it’s ready for FOAK commercial. Then, the company exits early to a FOAK fund. The FOAK fund acquires the company at a lower discount rate than what the VC investors would normally target, therefore being able to pay more than what a VC would have valued the company. After the FOAK fund acquires the company, it helps the company get past FOAK and move onto being a levered fully functioning infrastructure owner-operator, after which it can be exited.

Exiting early allows the Pre-Seed investors to realize a 77% / 18x return, the Seed investors a 64% / 7.3x return, the Series A investors a 67% / 4.7x return, and the Series B investors a 76% / 2.3x return. Though the MOIC is lower, the IRRs compare since the capital is returned a lot sooner. Exiting early also allows for a lower failure threshold, since the liquidity event for these investors is no longer dependent on getting past FOAK. Thus the 6 failures for Pre-Seed, 3 for Seed, 2 for Series A, and 1 for Series B, while lower than in traditional VC, reflect an easier success case.

For the FOAK fund investor that invested with a premium valuation to a VC, they would still be able to realize an exit at nearly 30% IRR and 5.7x MOIC, a very healthy return for any private equity or infrastructure investment. Because this investor is taking FOAK risk, there is a chance that the investment fails in a binary fashion (vs. half-failing or generating a partial return), similar to a VC investment. In this case, with a 28% individual investment return, the FOAK fund investor can tolerate a reasonable 2 failures for every success case to achieve a “risky infra” return of 12%.

TLDR;

Investors, new or existing in climatetech, can create a functioning FOAK strategy that can make investing in FOAK a financially attractive proposition.

Founders in hard tech climatetech can look to early exits as a means of scaling and providing liquidity to early investors. Also early exits to risk-taking infra investors can be more lucrative than continuing with the venture cycle.

Service providers can consider encouraging and supporting the creation of these FOAK funds.

After discussing types of FOAK last week, this week I wanted to see if I could find a good estimate for the amount of FOAK funding we need.

One way to look at this is to try to pare down estimated infrastructure spend to what will be spent on FOAK technologies.

Most infra spend estimates land at ~$4 to $9 trillion / year. These numbers are never detailed enough to understand specific technologies they’re actually building in, but we can kind of glean from the breakouts an upper-end limit for what spend might be:

So all of the estimates roughly point to $1.5 - 2T / year being spent on technologies that have not yet scaled. Over the next 28 years, that’s $42 – 56T in total. Let’s just call it $50T as a midpoint.

To estimate how much we’d need for FOAK, we can just ballpark how much funding we can assign to deployed facilities (2nd facility and beyond) for every FOAK – or, assuming FOAK cost is around the commercial cost, basically how many deployed facilities for every FOAK facility. As one bookend, we can say that 1 out of every 100 facilities requires a FOAK. This is very high level but not crazy to assume. There are only 109 biodiesel plants in the US, 31 years after the first biodiesel plant in Kansas. There are only 18 renewable diesel plants (5 - 6 current renewable diesel plants with another 12 under construction), 12 years after the first renewable diesel plant in Louisiana. Using 1/100, that means we’d need to spend $500B for FOAK assets.

Another bookend would be looking at solar farms as a comp. There are 2,500 solar farms in the US, 39 years after the first solar farm in California. Let’s assume there’s 4 different types of PV (monocrystalline, polycrystalline, multijunction, and thin film) and that these 2,500 only needed 4 FOAKs (probably a very wrong assumption). If we assume 4 out of every 2,500 facilities requires a FOAK, we’d need $80B for FOAK assets.

So somewhere between $80B - $500B is probably the right number here…

To put this into context, PE funding in cleantech is at around $20 – 25B / year, while late stage VC funding is at ~$24B / year. If all of this funding was directed towards FOAK, we’d have enough…but part of this funding is used for digital technologies and for stages other than FOAK. If we assume that we have to put all of the FOAK in place by the halfway mark, we only really have $700B in place for potential FOAK funding. At the low end, 11% of our late stage – PE capital should be directed to FOAK. At the high end, 70% of that capital should be directed to FOAK.

I can tell you anecdotally that that true percentage of capital being directed to FOAK is nowhere near 11%, much less 70%.

If you’re not convinced by this top-down method of estimating potential FOAK spend, we can try to do a bottoms-up approach. We can assume that each startup that survives to FOAK needs FOAK funding…so we need an estimate the number of climatetech startups that are out there that will survive to FOAK.

Using Net Zero Insights’ database, I isolated 32 key areas that will likely need chunky FOAK funding, separated by the approximate types they’ll be in:

  1. Large plants: DAC, CCUS, sustainable aviation fuels, nuclear, steel, alternative proteins, water treatment, cement, biofuels, hydrogen, recycling, chemical, petchem/plastic
  2. Large assets: aviation, CSP, geothermal, wave, hydro, offshore, energy generation, energy storage
  3. Large manufacturing factories: aquaculture, algae, maritime, textiles, EVs, micromobility, PVs, robotics, energy efficiency, built environment, and food/ag

18,182 startups were identified as being in these categories. Let’s assume that 25% of the large plants, 10% of large assets, and 5% of large manufacturing faciltiies make it to FOAK. At $100mm per FOAK, that would mean we need $193B in funding for FOAK with just the startups that exist today. If we attempt to get more granular on individual category funding needs and assume the large plants need an average of $100mm, assets need $20mm, and manufacturing facilities need $50mm, we get $150B. At $150 – 193B of FOAK need, we’d need 21 - 28% of late stage VC + PE capital directed to FOAK, implying that one out of every 4 of these late stage deals need to be funding a FOAK.

I realize that a lot of this exercise is very hand-wavey…but it does at a high level illustrate an important need to direct more funding to FOAK facilities. We either need to be deliberately setting targets for the proportion of institutional deals that are FOAK OR we need to be bringing more capital in the door to help fund FOAK. In that latter case, the government, non-profits, and non-traditional capital sources like family offices and sovereign wealth funds will play an important role.

TLDR; FOAK funding will cost us billions, likely hundreds of billions, of dollars. We need more funding directed to it.

After last week’s look at commercial facilities that have been successfully funded, I wanted to better understand what projects in the future will need large first commercial facility (or large first-of-a-kind / FOAK for short) funding.

I went through the list of hard-tech climatetech technologies and think there's ultimately 3 types of startups that will need FOAK funding.

In order from most scale up risk to least scale up risk, assuming technology risk is equal:

  1. Those that need a plant to commercialize a set of large-scale chemical reactions – in other words, startups which have chemical processes at the core of the plant. Because these reactions require feedstock and offtake, the scale up risk is compounded by project-specific characteristics like site location, vendor access, and commercial agreements. Many of these projects find ways to attach themselves to an existing plant or asset to reduce this risk. There is also generally more uncertainty with a chemistry-based scale up as many things are environmentally sensitive and small changes can produce no product or unwanted product.

    This category typically includes renewable fuels, carbon capture, oxycombustion, pyrolysis, low carbon concrete, low carbon steel, chemical recycling, nuclear, new fertilizers

  2. Those that produce and install chunky assets – in other words, startups that rely on deploying big installations. The FOAK project is a single asset produced at a scale large enough to be commercially viable. The scale up risk in this case is less dependent on integration with the supply chain / commercial agreements around the project and more on the pure engineering of scaling up the technology itself. This simplifies the scale up a little more compared to #1 but also puts the onus of whether a project works on the technology. Since there are usually fewer third party companies involved in this type of project than in #1, third party validation of the technology holds more weight in derisking the project for an investor.

    This category typically includes utility scale energy storage, geothermal, carbon storage, clean aviation, concentrated solar power, wave energy, automated mining, automated waste sorting, hydrogen electrolyzers, hydrogen fuel cells

  3. Those that need a manufacturing / assembly / processing plant designed to handle volume – in other words, startups that have a need to produce product in large volumes. This is by far the largest category in terms of number of startups that could grow to need this kind of project. BUT this is also the category that may not require complete FOAK funding if the startups 1) successfully use contract manufacturing, 2) outsource most of the manufacturing and only need a small facility for assembly, or 3) acquire an existing manufacturing facility that produces a similar or adjacent product. Manufacturing is also arguably the easiest to scale up out of the three categories since there is less a scale up of technology and more a scale up of process, something which also has plenty of precedence in other industries.

    This category typically includes EV batteries and battery components, EVs, EV retrofits, EV chargers, heat pumps, smart thermostats, smart glass, soil sensors, e-scooters, e-bikes, green textiles, algae farms, new photovoltaics, alternative proteins

These categories aren’t necessarily mutually exclusive. A company that builds an automated waste sorting facility may also need a manufacturing facility for robots (see AMP Robotics). A company that that installs large flared gas-to-datacenter systems may also want to manufacture its data centers (see Crusoe). (By the way, these two were not included in last week’s list because their FOAKs were <$50mm.) But most startups that are still commercializing their technologies are only contemplating FOAK in one of the three categories.

So what’s the point in knowing these categories? Understanding which category a startup lands in when planning out a FOAK commercial project can help identify a peer group with a similar scale up risk profile. Perhaps there are milestones and timelines that can be informative for early stage project planning, best practices that can be used between companies in each category, or benchmarks that can be used to help pitch the project risk profile to investors. Since the universe of FOAK commercial is so limited in climatetech, being able to creatively find a peer group to help tell the story is more important than in other industries.

There are also different recommendations I would make for capital raising in each category:

Those that need a plant:

Those that need a large installation:

Those that need a manufacturing facility:

Would love to hear:

If you're a startup - does this framework make sense to you? Or is there another category that's missing? If you're looking for FOAK funding or thinking about your FOAK plans in the future, I would love to connect.

If you're an investor - are there different or additional recommendations you would give for each of these categories?

If you're part of a corporate - are you or have you contemplated the types of partnerships described above?

Icon source

The scaling problem in hard asset climatetech is well-known and well-documented…valleys of death, unfit capital, project development challenges, etc. etc. Technologies that require some kind of plant, facility, or large chunky infrastructure to be built struggle the most with scaling. Here's how the ease of funding curve looks across a company's maturity (thanks, Lanzatech):

Initial funding for these technologies, if in small dollars, is relatively plentiful. For R&D and prototyping, startups can access grant funding and traditional VC capital (as well as capital from family offices, corporate VCs, incubators & accelerators, and other entities that surround the VC ecosystem).

After the technology has been prototyped and shown to work at lab scale, engineering work can take it to the next level and show that it can be used in the real world. Engineering work can mean expanding the team to include more engineers and/or contracting third parties like EPCs or labs to perform feasibility studies. Larger VC dollars can fund engineering work, though the pool of VCs that can write a later stage check for a pre-commercial tech is more limited than at the earlier stage.

Pilots and first demonstration assets / facilities are where the capital stacks start to detract from the normal VC ecosystem. Funding for pilots and demo facilities can edge into really late stage VC to growth equity levels of capital. Activities like permitting, buying construction materials, hiring a construction agency, adding plant personnel, etc. are expensive. Since the goal of this stage is to make sure the technology works and to prove out the engineering work, the plants are smaller, limited in connectivity to commercial outflows like roads or the grid or customers, and operate only for a fraction of the time. I’ve found that most of the “mature” hard tech climatetech companies with big plants to build are at this stage. This is especially true for the battery industry. Several of the names that have gone public via SPAC over the last few years – Solid Power, Quantumscape, FREYR, SES – are still undergoing pilots.

After the technology has been validated at smaller scale in the real world, both the technology and business model need to be proven together in building the first commercial version of the asset or facility. As we reach the trough of the growth curve, this is the hardest step to get funding for but the one that derisks the company the most. Building a commercial plant requires the capital of a pilot or demo facility scaled up to the point where the economics make sense + capital for additional personnel to run the plant or asset full time + capital for processes or certifications to enable the product to be sold commercially + capital for logistical infrastructure like trucks, roads, pipes, or other conveyances + capital for contingencies, unplanned downtime, regular maintenance….the list goes on. This step attracts investors looking for cash flow, which means that the technology has to be derisked, the financial model has to be airtight, and there has to be a high degree of certainty that revenue will occur.

Post-first commercial, companies can access the much larger pool of capital: private equity, infrastructure capital, and project finance equity and debt. These capital pools don’t have enough climatetech opportunities at their desired maturity levels and check sizes. A startup that reaches this stage usually has enough leverage to get pretty advantageous economics.

I wanted to better understand how a startup can get past that first commercial trough – are there any learnings that we can glean from past projects? To do this, I set out to collect examples of large commercial facilities (requiring >$50mm in capital) that have been funded climatetech.

Unfortunately, they were very difficult to find. My ideal dataset would be 30+…but I could only find 10.

Companies: CoolPlanet, Enerkem, LanzaTech, Monolith, Fulcrum Bioenergy, Lucid, Lanzajet, Arbor Renewable Gas, Carbon Engineering, Lithion

Some observations:

  1. Corporate involvement helps a lot. Half the projects on this list had some type of corporate sponsor or corporate partnership backing. For startups, working with a corporate on first commercial projects makes a lot of sense. Corporates often have already-permitted, already-functioning sites with access to logistical infrastructure, vendors, and/or customers. Corporates also likely have much lower cost of capital and may ask for less aggressive financial returns than a comparable institutional equity player, which translates to more favorable economics for the startup and other investors in the project. And finally, startups may be able to inherit best practices from a corporate partner – things like preventative maintenance, project management, safety regulations, and corporate governance – that can be difficult for a startup to learn on its own and prevent some painful lessons further down the road.

  2. Private and non-traditional equity can also play a major role in funding these projects. Most private equity firms don’t want to touch technology risk, which leaves a huge gap in equity offerings in the first commercial space. Tech-leaning private equity firms and non-traditional equity firms like family offices or sovereign wealth funds can play a big part in filling this gap.

    I was a little surprised to see a complete lack of infrastructure equity on the list. I know there has been a big push as of late to redefine infrastructure and push capital deployment up the risk spectrum…but maybe we just haven’t seen these projects manifest publicly yet.

  3. The new fuels space has enjoyed the most number of successfully funded large projects. 7 out of the 10 projects on this list produce some sort of fuel. Two reasons why:

    1) The technologies that underlie these facilities are familiar and have precedence. Though the IP can be new, the reactions are most likely a revival of chemical processes that have been studied and understood for years (e.g. Monolith’s pyrolysis). As a result, these projects probably have found a better ecosystem of firms and consultants with the ability to diligence the technology for them.

    2) Demand for the product exists today in large quantities. Most of these facilities aim to produce a drop-in fuel, which means that the market is easier to define and customers easier to sell to. Most other sectors of climatetech don’t have that luxury.  

  4. Other projects that didn’t make the list bypassed the first commercial hurdle by repurposing an older facility. You might look at the list and wonder why Lucid is the only EV manufacturer on it. That’s because other EV manufacturers have taken a different approach: purchasing existing factories. Tesla’s Fremont, CA factory, which it acquired in 2010 when it started mass-producing Model S’, used to be a GM/Toyota factory. Rivian’s first plant was also purchased – it used to be a Mitsubishi plant until it closed down 2 years before Rivian bought it. Hyzon purchased an old GM fuel cell facility in New York as its first plant.  

    This strategy hasn’t only been adopted by carmakers. Bolder Industries, a tire recycling startup, acquired an old tire recycling plant as its first plant.

    Acquiring an existing facility can save money (especially if the asset is distressed) and, like working with a corporate, provide access to strategic benefits like logistical infrastructure, best practices, and even labor. But of course, this strategy only works for technologies that can share a lot of assets with an already-scaled operation, which limits its usefulness to areas like EVs or chemical plants.

TLDR; funding for large first commercial facilities has limited precedence. As climatetech scales to need a greater number of large first commercial facilities, companies should look to the fuels space for learnings, corporate and tech-forward private / non-traditional equity for sources of capital, and acquisition of existing facilities as an alternative strategy.

HOW IT WORKSPRICINGABOUTCONTACT
© 2022 All rights reserved.