I have a love-hate relationship with publicly held research and development companies because outside the biotech space, very few investors understand them. To make matters worse you can't know what a particular innovation will be worth until the money's all been spent and a cruel and heartless end-user market casts the only vote that counts. The entire process is like having sex with a gorilla. Once you start down the path you can't stop until the market is satisfied with your innovation or it decides to kill you. Most of the signals and clues are subtle and very difficult to interpret. Success is incredibly gratifying and rewarding. Failure is not.
Most of my readers are familiar with the Hype Cycle. While I typically use the following graph to caution stockholders of companies that are approaching the peak of inflated expectations, it can be very helpful for other investors who want to understand how the future is likely to unfold.
The critical requirement if you want to use the hype cycle as an investment tool is to understand where a particular company is situated on the curve at a particular moment in time.
I always get a kick out of the Hype Cycle graph because it shows R&D as a discrete point in the lower left-hand corner without offering any insight into the complexity of the R&D process. While experts have written volumes on the innovation and product development process, the three essential stages every innovation must navigate and traverse in sequential order are:
- Developing a comprehensive understanding of an innovation;
- Learning how to build devices that make the innovation useful; and
- Refining production methods and developing products for specific markets.
Each of these stages involves years of work and huge amounts of capital. There are no short-cuts or certain outcomes, and the only thing an R&D investor can do is monitor progress toward the ultimate goal of a product that end-users will buy in sufficient volumes to justify the time, effort and expense. Until the process is completed, R&D companies simply don't have anything they can sell in meaningful volume. They can generate lumpy bits of revenue by selling prototypes and services to potential end-users, but predictable revenue and meaningful growth can't happen until a company exits the R&D stage and begins commercialization.
In my experience, the most important character traits for managers of R&D companies are conviction that their project is worth pursuing, confidence that the multitude of challenges can be overcome, and a dogged determination to see the project through to a bitter or brilliant end. Failures are far more common than successes, but the payoff of successful innovation makes the entire game worthwhile, and more than a little addictive.
I've been involved with Axion Power International (NASDAQ:AXPW) since 2004 because a public shell I controlled was used in the reverse merger that made Axion a public company. I served as a director of Axion for three years and as legal counsel for four. Since early 2008 I've only been a stockholder, but lawyers who work in my field develop deep bonds of friendship with former clients and tend to follow their evolution closely for years after the attorney-client relationship comes to an end, particularly if they're still stockholders.
The following graph shows my assessment of the research, development and commercialization phases Axion has successfully traversed over the last ten years. It also shows the 10-year growth in Axion's share count, the 10-year growth in its trading volume and the 10-year evolution of its market capitalization. The discussions after the graph add more detail on the challenges Axion faced and overcame during each of the principal development stages.
The First Three Years
The first three years of Axion's existence (2004 through 2006) were a period that I've come to think of as the basic scientific research and technology litigation years. Axion's main focus was developing a better understanding of the how the PbC electrochemistry worked, discovering its strengths and weaknesses and finding ways to improve the performance, charge acceptance and cycle-life of single cell laboratory prototypes that bore only a passing resemblance to a battery. The rest of its time was spent with lawyers who fought long and hard to eliminate the clouds on Axion's title to the core patents and intellectual property. From inception through September 2006, Axion raised and spent a total of $9.9 million in cash.
Axion won the critical lawsuits in the fall of 2006 and those victories set the stage for its first big financing, an $8.1 million placement of convertible preferred stock that closed in Q4-06. With that funding in hand, Axion was ready to move into the second phase of R&D work on the PbC technology, figuring out how to build a multi-cell battery that had the charge-acceptance, cycle life and other performance attributes the scientific staff demonstrated during the scientific research and technology litigation years.
The Second Three Years
The second three years of Axion's existence (2007 through 2009) were a period that I've come to think of as the basic product design and process development years. We knew from the outset that the easiest and most effective development path for the PbC technology would be to find a way to integrate our carbon electrode assemblies with the manufacturing equipment and methods used by every lead-acid battery manufacturer in the world. If we could make carbon electrodes in a form factor that could be used in existing battery plants, the path forward would be far easier than one that would the construction of new battery plants. The challenge was developing an electrode assembly that could work as a drop-in replacement for conventional electrodes without requiring major changes in established battery manufacturing equipment and methods.
Exiting the scientific research stage was a frightening time because the first prototypes built by the scientific staff required lateral compression that was an order of magnitude higher than we could get with standard battery cases and manufacturing technology. We knew how we wanted the electrode assemblies to work and had a sensible path forward with a concept for laminated electrode assemblies, but that path required Axion to develop new methods for processing activated carbon and binding powdered carbon into uniform electrode sheets with appropriate structural integrity, electrical conductivity and other physical characteristics. Then we had to develop new methods for bonding corrosion barriers to current collectors, bonding electrode surfaces to corrosion barriers and sealing everything tightly enough to survive years in a sulfuric acid bath. The challenge of minimizing internal resistance between the electrode assembly layers was particularly daunting.
The second three years would have been impossible without the New Castle battery plant that Axion bought for a song in early 2006. As battery plants go it was old and not heavily automated, which would have made it hard to make money producing conventional lead-acid batteries. While New Castle wasn't a great manufacturing facility, it was the perfect prototyping facility because it had the same equipment as every lead-acid battery plant in the world. That meant the scientific team could make electrode prototypes in the lab and then immediately build battery prototypes using company-owned facilities that didn't have to be scheduled weeks or months ahead of time. While most battery technology developers have to rely on third parties to build prototypes, Axion had a unique ability to do everything in house at a lower cost and with better security for its rapidly evolving intellectual property estate.
The second three years was an expensive time. By the end of 2007, Axion had spent most of the $8.1 million in proceeds from the 2006 offering and it needed additional capital to design and build an automated fabrication line for the carbon electrode assemblies the scientific team had previously built by hand. In January 2008, Axion closed the first $4 million tranche of an $18 million investment from the Quercus Trust. The two remaining tranches came in April and July of 2008, which was a darned good thing given the market meltdown that hit in the fall of 2008.
By the end of 2009, Axion had finished its work on a first generation automated electrode assembly fabrication line. The carbon sheeting for electrode surfaces was still being made with a labor-intensive manual process, but the lamination and sealing of the electrode assemblies was fully automated and Axion was beginning to get favorable reviews from potential customers who were impressed with the performance of its pre-commercial prototypes.
The successful development of a basic manufacturing process for PbC batteries set the stage for Axion's largest financing ever, a $26 milion placement of common stock that closed in Q4-09. With that funding in place, Axion was ready to move into the third phase of its development work on the PbC technology, finding ways to refine its manufacturing methods and begin the difficult and costly process of proving product performance developing battery products for specific applications.
The Third Three Years
The third three years of Axion's existence (2010 through 2012) were a period that I've come to think of as the process refinement and application development years. Axion quickly learned that its first generation automated electrode assembly line wasn't good enough to satisfy first tier customers. It also needed a less labor-intensive process for manufacturing carbon sheeting.
The first prong of the manufacturing process refinement was simple. Axion hired a respected industrial engineering firm to redesign the electrode fabrication line, significantly augment the robotics and add better quality control throughout. That upgrade took about a year.
The second prong was more challenging because there was a delicate balance between the amount of binder used in an electrode sheet and the electrical performance of the finished sheet. The basic problem was that gluing small particles together blocked pores in the carbon and reduced the accessible surface area while increasing internal resistance. In the original carbon sheeting process particles were mixed with binder and the carbon dough was passed through rollers, folded over and passed trough the rollers again. The rolling and folding was repeated over and over again until the desired structural integrity was achieved. The closest analog was a French bakery making croissants. It was a mind-numbing repetitive process that left a lot of room for variability, but it used a very low binder to carbon ratio and resulted in electrode surfaces that performed well in pre-commercial prototypes.
While Axion usually maintains tight control over information, word that they'd found an automated solution for the carbon sheeting spread quickly through the grapevine in 2011. It took another 18 months, however, to prove that the automated sheeting process would work at scale. The big advantage of automated sheeting was that it stripped 80% to 90% of the labor cost out of a PbC battery.
When I first heard about the automated sheeting process, a nagging concern in the back of my mind was that increasing the binder to carbon ratio would impair electrode performance. I was pleasantly surprised when Axion announced that it had commissioned an automated sheeting line and the new sheets performed better than the originals. While I haven't been able to confirm my suspicions, I think the reason the new electrode sheets perform better is that repeatedly rolling, folding and re-rolling the carbon resulted in a stratification or layering like you see in a fine French pastry while a single pass process doesn't.
The Last Year
With the successful commissioning of the automated carbon sheeting line in 2013, Axion finally arrived at the lower left-hand corner of the Hype Cycle graph. The PbC had survived the R&D process without significant performance degradation and it was time for the battery market to consider the question of whether the PbC technology was worth the ten years of effort.
The early results are encouraging. Until November of last year, Axion had never sold a PbC to a customer that planned to use the battery in a commercial activity. There were lumpy sales to outfits like BMW, Norfolk Southern, the DC Naval Yard and ePower who planned to test the PbC to find out whether it would serve their needs, but all those purchases were experimental in nature and non-recurring by definition.
Last November's PowerCube sale was very different. It was a sale of 600 batteries that the buyer is interconnecting with a solar panel array and using for frequency regulation, renewables integration, demand charge management and emergency power backup at a commercial business facility. The customer didn't buy the batteries with the intention of testing them. It bought the batteries for the express purpose of putting them to work in a day-to-day commercial activity. The recent follow-on order from the same customer for four more systems is confirmation that Axion has its first commercial battery customer. These are working batteries, not testing batteries.
In my view, these recent sales mark the beginning of Axion's journey up the Hype Cycle.
While most of us are looking forward to the launch of the second generation NS 999 and the expansion of ePower's demonstration fleet, these activities are still experimental. NS won't know whether an electric switcher will satisfy its needs until the locomotive has been put to work moving boxcars around a rail yard. Likewise, ePower won't know whether its engine dominant series hybrid drivetrain will meet the needs of the trucking industry until it builds a small fleet of tractors, puts them to work hauling freight and accumulates several million miles of experience with industry partners who will make their own decisions on the value of ePower's drivetrain. The same is true for ongoing testing by automakers and other developers of systems that need the PbC's charge acceptance and cycle life. While there's no doubt in my mind that some of these experimental projects will fall by the wayside, it's increasingly clear to me that some are likely to succeed and add impetus to the PbC's success in stationary applications.
What The Hell Happened in the Stock Market?
At this point I'd like to revisit my development stages graph. For the first six years of its existence Axion's market capitalization trend was moderately positive, which is exactly what I'd expect from an early stage R&D company that still faced huge challenges of developing a product, refining the product and introducing the product to potential customers. Axion's spending during each phase ramped slowly but steadily and each investment resulted in a proportional added value that was reflected in Axion's market capitalization.
Beginning in 2010 there was an extraordinary breakdown in the market capitalization trend. Axion's value adding activities accelerated rapidly but its market capitalization started as steady downward trend. The only explanation that makes sense to me is that a handful of very large investors who had no business selling their shares did so anyway, and they did so in the most savage manner possible. For better or worse, nano-cap markets are like babies. If you try to feed them too much at a sitting or try to force-feed them when they're feeling finicky, they'll puke all over your shirt. That's exactly what happened with Axion's stock price.
I believe the Axion that exists today is far more valuable than the Axion that existed in early 2010. The old Axion didn't have
- Fully developed products;
- Cost effective manufacturing processes:
- Identifiable potential customers;
- Successful testing by potential customers; or
- Expectations that anyone would buy PbC batteries for use in a commercial activity.
Axion may not have had any of those things in 2010, but it has them all today.
The PbC is a fine but expensive battery that's certain to get cheaper as Axion builds scale and progresses along the learning curve that impacts every manufacturing business.
While Axion is little more than an asteroid in the universe of public companies, several planet-sized giants have fallen into orbit around Axion and a pair of them publicly embraced the PbC technology before there was a product. I can't remember another nano-cap that had disclosed relationships with two first-tier companies and discovered relationships with several more. That fact alone tells me the PbC is every bit as valuable as Axion's founders hoped it would be.
It makes me crazy when I read criticisms that Axion's sales haven't ramped quickly enough. The simple fact is Axion didn't have anything to sell during the first six years and all it could offer from 2010 through 2013 was pre-commercial prototypes that were suitable for testing and demonstration but were not suitable for use in a commercial activity and couldn't be produced in sufficient volume to support a commercial activity.
Over the last few months that business dynamic has changed. Last November a customer bought a PowerCube for use in a commercial activity. It recently ordered four more comparable systems and Management has told the stockholders that several additional PowerCube projects are in the sales funnel and expected to mature into purchase orders later this year.
In addition to the commercial sales for stationary applications, several experimental applications of the PbC including the NS 999 and ePower's series hybrid tractor are approaching transition points where we'll get the first clear indication of the PbC's value in rail and heavy trucking applications. Only the gods know when we'll hear from the automakers While there are still no guarantees, there are many good reasons to be optimistic that these additional markets will develop in due course.
The PIPE investors have just left the stage and for the first time in four years Axion doesn't have several huge investors pushing and shoving at the pay window trying to force their shares into the market ahead of the other guy. In fact the only possible sellers at this point are the old guard and the retail investors who did all the buying over the last four years. I have a hard time imagining a better dynamic for investors.
Disclosure: I am long AXPW.