Boeing 787 battery fires underscore complexity of energy storage technology

Energy storage is still the rage in cleantech.  But after the collapse of A123 and Beacon, and the spectacular failure on the Fisker Karma in its Consumer Reports tests, fire  in Hawaii with Xtreme Power’s lead acid grid storage system and with NGK’s sodium sulphur system, and now battery problems grounding the Boeing Dreamliners, investors in batteries are again divided into the jaded camp, and the koolaid drinker camp.   Not a perjorative, just reality.  New batteries and energy storage is still one of the juiciest promised lands in energy.  And still undeniably hard.  Basically, investors are relearning lessons we learned a decade ago.

Batteries are just hard.  Investing in them is hard.  Commercialization of batteries is hard. So why is it so difficult to make money in new battery technology?

Above and beyond the numbers, there are a number of commonalities related to the commercialization and venture financing life cycle of battery technologies that seem to differ to some degree from other venture investments in IT or even other energy technologies.  Having looked at probably 100+ deals over the years, and on the back of an deep study we did a couple of years ago on benchmarking valuations in energy storage, here’s our take on the why.

Timing – Battery technology commercializations have historically tended to be one of the slower commercialization cycles from lab stage to market.  Startups and investors in batteries have a long history of underestimating both the development cycle, capital required, and the commercialization cycle, as well as underestimating the competitiveness of the market. 

Special chemistry risk – There is significant risk in launching a technology in newer battery chemistry.  There have been only a limited number of new chemistries succeed, and when they do, as in the case of NiMH and Energy Conversion Devices, they are typically either co-opted by larger competitors obviating a first mover advantage (that advantage is typically much weaker in this field than others) or requiring expensive patent suits.  Also as in the case of NiMH, there is no guarantee the chemistry will have legs (just when it is hitting its stride, NiMH is already becoming eclipsed by Li-On.  This risk has proven to be especially high for new chemistries (like Zn type) that are not as widely researched, as the supply chain development does not keep pace.  In addition, the battery field is highly crowded, and research is old enough that and despite new chemistry in most cases truly defensible patent positions are extremely hard to come by, or provide only discrete advantages (ability to supply a range of quality product cheaply in high volumes (or with value add to the product) seems to be the primary competitive advantage).  Few battery technologies of any chemistry end up their commercialization cycle with anywhere near as sustained an advantage as their inventors expected.

High capital costs – In any case, almost all battery startups will require extremely large amounts of capital (on the order of US$50 to 100 mm+) to achieve commercialization (much higher for real manufacturing scale), and the end product margins tend not to be particularly high.  Even with stage gate, a very large portion of this investment (US$10-50 mm+), is generally required to be spent while the risk of technical and economic failure is still high.  In addition, during the manufacturing scale up phase post R&D, capital investment required per $1 of revenue growth tends to be linear, making these technologies capital intensive to grow.

Degradation of initial technical advantage – In many technology areas one can expect the performance of the final manufactured product to improve over the performance in initial lab results, In part because of the low cost target, high reliability, high volume requirements of this product type however, promising battery technologies, are often forced to make compromises in the scale up, manufacturing, and commercialization stages that mean the performance of actual product might be expected to fall from levels or rates seen in lab scale experiments (though cost may go the other way).    At the same time, battery performance of standard technologies, while mature, is a moving target, and during the time frame for commercialization, will often improve enough to obviate the need for the remaining technical advantages.

Size matters – Most battery products (whether batteries or components like anode or cathode materials or electrolyte), are sold to large customers with very large volume requirements, and highly competitive quality and performance requirements.  As a result, breaking into new markets generally is extremely hard to do in niche markets, and means a battery startup must prove itself and its technology farther and for a longer period of time than other technology areas (see capital costs, timing and down rounds).  Many battery components technology developers as a result will be relegated for early adopters to emerging customers with high risks in their own commercialization path.

Lack of superior economics from licensing – As a result of these size, capital cost, timing, and commercialization risk issues most battery technologies will command much lower and more short-lived economics than anticipated from licensing (or require expensive patent lawsuits to achieve), and will require almost as late a stage of development (ie manufacturing operating at scale with proof of volume customers) and commensurate capital requirements, as taking the product to market directly.

Propensity for down rounds – In addition, battery technology companies tend to have down rounds in much larger numbers in the post A rounds (Series B through D+) than other venture investment areas, as these challenges catch-up to investors and management teams who overestimated the scope of work, capital and timing required in the seed, A and B rounds.  In particular, battery investors have tended to invest in seed, A and B stage battery technologies (pre-scaled up manufacturing process or even lab and prototype scale) with expectations of typical venture style timing and economics.  Quite often instead, it is the B, C, or D investor group that post cram-down rounds achieve the Series A economics (even when the technology IS successful), and the seed, A and B investors suffer losses or subpar IRRs.