John Petersen
Since last week’s article, Common Sense in Energy Storage Investing, was well-received by readers who’ve recently discovered this blog and want to better understand the energy storage sector, I’ve decided to continue with the theme and drill down deeper into some broad issues. Most of today’s material is pretty basic stuff, but when the hype machine starts spinning a firm grasp on economic reality and investment fundamentals can be important to investors that want to avoid a boom and bust bubble like we had in corn ethanol.
In the fall of 2008 I confessed to being a shameless early adopter of cutting edge technology. I bought the first portable electronic calculator in 1971; bought word processing, laser printing, videotape, compact disks and satellite TV in the early ‘80s; bought a cell phone and established an Internet domain in the early ‘90s; and established a paperless office and a global law practice by the late ‘90s. If it was new and a major advance, I had to have it first regardless of cost. When I look back at the last 40 years, I’m amazed at how quickly the latest and greatest technologies became obsolete when newer, better and cheaper products emerged. The most recent example of how quickly technologies can rise and fall happened just last week when Sony (SNE) announced that will quit making floppy disks next March. As an investor, I’m horrified by the idea that a technology as important as the floppy disk can rise to global dominance and decline to insignificance in forty years.
In most cases, I’ve adapted well to changing conditions. My only line in the sand has been an almost religious devotion to the Macintosh operating system, which I switched to in the fall of 1989 based on the personal advice of Dr. Wilson K. Talley of the Lawrence Livermore National Laboratory. While I’ve never questioned my choice in computers, a graph comparing the long-term stock price performance of Microsoft (MSFT) and Apple (AAPL) serves as a stark reminder of how a sound technical decision played out in the equity market.
Today I can sit back and bask in the glow of being right about Apple’s inherent technical superiority, but that doesn’t change the fact that I was right too early. If I’d been a truly prescient investor, I would have owned Microsoft for the first dozen years and then switched to Apple for the long term.
Last week I re-printed a table from a July 2008 Sandia National Laboratories report that estimated the current and 10-year projected cost of stationary energy storage systems for solar power installations, including the storage devices and power conditioning equipment necessary for turning DC output into 60-Hz AC power suitable for delivery to the grid. The following chart puts the projected future cost of systems using the ten battery technologies included in the Sandia study in graphic form. While the media is enthralled with lithium-ion batteries because of effective PR and the oh so alluring promise of electric cars, my experience as a Mac user tells me that the vast majority of likely buyers will obey the laws of economic gravity and buy the cheapest system that can do the work.
The bottom-line is that major innovations take decades to evolve and work their way through the markets. The process was first explained in the technology adoption lifecycle, a model that emerged in the ’50s and has since been refined by contributions from Geoffrey Moore and others who explain the process with graphs like this one from Crossing the Chasm.
We are living in the first days of the Age of Cleantech, the sixth industrial revolution. The media is chock full of stories about how wind and solar power will change the way we generate electricity, the smart grid will change the way we distribute and use electricity, vehicle electrification will free us from pollution and the tyranny of imported oil, and energy storage will be the keystone – an enabling technology that makes all the other advances possible. What the news stories don’t tell us, because frankly nobody knows, is when these technological marvels will hit their stride and make a meaningful difference in the way we live. To help put things into perspective I’ve used data from a press release teaser for the American Wind Energy Association’s annual market report for 2009 to create a graph of the annual and cumulative changes in U.S. wind power capacity over the last 15 years.
The first use of a large windmill to generate electricity was a system built in Cleveland, Ohio, in 1888 by Charles F. Brush. If you only consider the cumulative values since 1995 the growth seems pretty stable. If you think about the hundred and twenty year history of wind and study the annual additions and other data from the teaser, it becomes clear that wind power didn’t transition out of the innovators stage until 2004, and then it took another three years to reach the early majority stage.
A similar trend is clear in the 10-year history of the HEV market, as shown by the following graph from hybridcars.com.
Viewed in isolation, HEVs have built an impressive growth history. Viewed as a segment of the larger market, they’re just beginning to scratch the surface with 2009 numbers that represented 2.8% of light duty vehicle sales. Returning to the technology adoption lifecycle, HEVs are just now transitioning out of the innovators stage and into the early adopters stage. Plug-in vehicles, in comparison, are at the earliest possible point on the curve. I’m very optimistic about the future of HEVs because they’ve already demonstrated a decade of consistent growth and built a solid core of satisfied consumers. I’m less sanguine about plug-in vehicles because they have no track record and even their strongest advocates acknowledge insurmountable obstacles to widespread vehicle electrification over the next decade including:
- The high cost of batteries;
- The lack of recharging infrastructure;
- Capacity, regulatory and coordination problems in the electric power sector; and
- Consumer acceptance issues.
While I’m not willing to go out on a limb and predict what future penetration rates will be for powertrain electrification technolo
gies, Roland Berger Strategy Consultants has predicted that full or partial powertrain electrification will be a key automotive efficiency technology by 2020 and forecast high scenario market penetration rates as follows:
| Plug-in | HEV | Stop-start | ICE | |
| Western Europe | 20% | 7% | 67% | 6% |
| United States | 13% | 13% | 51% | 23% |
| Japan | 8% | 15% | 60% | 17% |
| China | 16% | 6% | 30% | 48% |
If we study the Berger forecast and think back to the technology adoption lifecycle graph, it’s pretty clear that HEVs are expected to follow a natural growth path over the next decade as their market share quadruples. It’s also clear that something beyond normal market forces is expected to drive the adoption of plug-ins and stop-start systems. In the case of plug-ins the main driver of growth will be subsidies and incentives as governments around the world tax Peter to pay for Paul’s new car. In the case of stop-start systems, the main driver will be new CO2 emissions and fuel economy regulations that require automakers to reach increasingly stringent targets. The first approach relies on incentives to create demand that wouldn’t otherwise exist. The second approach relies on penalties to force automakers to implement efficiency technologies without regard to consumer preferences. In my experience, government is not very effective when offering a carrot but it’s darned good at using a stick. Under the circumstances, I’m inclined to believe the stop-start penetration rates are a sure thing while the plug-in penetration rates include a hefty dose of wishful thinking.
Over the next five years manufacturers of inexpensive energy storage systems for stop-start applications are certain to report major revenue gains from C02 emissions and fuel efficiency regulations that are now fait accompli. The main publicly traded beneficiaries include Johnson Controls (JCI), Exide Technologies (XIDE), Maxwell Technologies (MXWL) and Axion Power International (AXPW.OB). If the planned introductions of plug-in vehicles later this year proceed as planned, the government incentives are successful and innovator class purchasers don’t experience too many problems with battery pack failures, range limitations, poor cold weather performance and limited charging infrastructure, battery manufacturers like Ener1 (HEV) and A123 Systems (AONE) may begin realizing revenues that justify their market capitalizations in the second half of the decade.
I’ve already had my Apple vs. Microsoft experience and don’t intend to repeat it. I’ll continue to buy green bananas, but my days of trying to carve a new plantation out of the jungle are over.
Disclosure: Author is a former director of Axion Power International (AXPW.OB) and holds a substantial long position in its common stock.
Even though wind energy was not the primary point of this article, it was used as an important illustration of trends. Thus, a bit of further clarification may be appropriate.
Present installed USA wind capacity is indicated to be 35,000 MW. This is “nameplate” capacity. Actual capacity is variable depending primarily on wind availability. It is probably not unreasonable to say that the capacity factor for installed US wind generation is on the order of 25% to 30%(at some sites it is somewhat better and at some it is much worse). Assuming 30%cf, the USA installed wind capacity is actually around 12 GW. In comparison, overall USA electric generating capacity (based on nameplate) is on the order of 1000 GW (1 Terrawatt). This overall capacity has a much higher capacity factor than wind, say on the order of 85%. (This figure too can be, and has been, debated.) Still, if we use these as indicators, “available” US wind generation capacity (12GW) is only about 1.4% of “available” overall US electric generation capacity.
On the other hand, if we look at this growth of wind capacity, adjusted for cf, compared to overall adjusted capacity growth we see a much different picture. In 2008, US electric generating capacity grew by 19.1 GW (x0.85=16.2GW). In 2008 wind grew by 8.3GW(x0.3=2.5GW) So wind’s contribution to overall effective generating capacity growth during 2008 was 15% (2.5GW/16.2GW).
The preceding does not consider the effects of plant retirements or conversions. Still it demonstrates that wind is still in the “early adopters” phase of technology implementation. It also highlights that, strong concerns to the contrary, unless there are other changes to US electrical generation policy wind is likely to be “several decades” from achieving a meaningful percent of overall generating capacity in the USA.
This analysis, when applied to automotive electrification would tend to further support your argument that electrification will likely be slower than many people expect.
Having said that, I suspect that most people who are “betting” on more rapid technology adoption also have at least an implicit belief in a dramatic increase in fuel prices in the medium term. If, for example, the price of US gasoline exceeds 5$ during the next 5 years, it seems likely that overall auto production will once again take a hit and that fuel efficiency will be given much higher importance as a criterion in vehicle purchase.
In such a price spike scenario, the price of electricity will not increase as dramatically (and perhaps not much at all). Payback of cost premium for hybrid, or other fuel efficiency, technologies will then be faster than at present. Perhaps as important, there will likely be an increasing willingness to pay a premium as insurance against future increases and against volatility itself.
Your observation about the capacity factor for wind is very accurate and if we focus on capacity adjusted nameplate, then wind is still in the upper end of the early adopters stage. The ultimate point is that changes of the magnitude people talk about take a long time and are generally unprofitable until scale is reached.
While there seems to be a consensus that oil prices will skyrocket while electricity prices will remain stable, I don’t understand how that can happen as we spend more money to get less electricity. Everybody focuses on nameplate capacity price parity for wind and solar, when real parity is still decades down the road. We talk about the smart grid as the next big thing, but ignore the fact that the efficiency gains alone will not pay the capital costs. That cost will ultimately be paid by consumers. The same is true for cap and trade proposals that may improve air quality, but will be a major driver of electricity prices. When I put it all together, I see stable future electricity prices as an impossible dream.
John,
Thank you for your response. It appears we are in general agreement that the adoption of technology can take longer than anticipated. However, even though we agree on this, it could be interesting to consider the nuances further…some other time.
Right now I’d like to comment further on the concept of electric price stability.
As history, both recent and longer term, has shown, the price of oil can be extremely volatile. During the last decade, the spot price per barrel of oil ranged from roughly $25 to roughly $145 (nearly a 600% increase) and then back below $40. It is presently trading around $80. By any measure this is a very volatile price. Not well hidden within that volatility is the fact that the price peak represented a new peak in both nominal and real terms. Further, the most recent low price was well above, roughly 50%, the most recent low. In other words, the pricing for oil during the last decade demonstrates higher highs and higher lows, an uptrend, contained within high volatility.
In contrast, during that same period the price of electricity demonstrated far less volatility and range. In 2000 the average US retail price for electricity was $0.068 per kWh; it peaked in 2008 at $0.107 and is now at roughly $0.099. During the decade, the price of electricity demonstrates an uptrend. However, the peak price for electricity was only 60% higher than the low, a substantially subdued increase compared to oil’s 600% min/max range.
Why is there such a difference between the two price patterns and what can be said about the future?
With regard to oil, the price at the start of the decade was low, perhaps unreasonably so, due to in large part to excess oil production capacity. By mid-decade this excess was substantially consumed. Simultaneously demand spiked, particularly from developing countries, resulting in very limited excess capacity and the ability for unusual price speculation (also thanks in part to new investment vehicles and wider use of futures). The “Great Recession” combined with new production capacity has substantially restored the oil production “buffer”. However, it is increasingly evident that this is likely a temporary condition. It now seems likely that the world will repeatedly “bump into” peak oil production capacity, until it dramatically reduces its need for oil and/or develops effective (technically, economically and environmentally) alternatives. These “re-tests” of liquid fuel production capacity will inevitably result in repeated “tests’ of higher prices for oil with the likelihood that oil prices will experience new highs. The most likely timeframe for this to occur is during this decade.
In contrast to oil, electricity pricing has several very different characteristics, including for example:
1. electricity is produced from a variety of fuels and so is less susceptible to fuel capacity constraints.
2. electricity is a regional commodity, not priced on the world market
3. electricity prices are generally regulated and not subject to high market volatility
4. electricity consumption intensity in developed countries has been decreasing for decades
5. it increasingly appears that future electricity consumption intensity may decrease even more. For example:
a. lighting efficiency has improved during the last decade but is likely to demonstrate a very substantial improvement over the next two decades
b. there is increased emphasis on energy efficiency of buildings
c. energy standards are being applied to a wider range of appliances
d. newer technology and market saturation is probably reducing the overall power consumption of consumer items such as televisions.
e. Industrial processes are increasingly energy efficient
6. electricity production, transmission and distribution itself offers substantial opportunities for efficiency improvements. And there is growing awareness and demand for such improvements.
7. a key element of electricity cost is the fixed cost of the generating plants. Since these are long lived assets, once they are “paid for”, electricity cost can drop substantially. This is particularly true of generation technologies with low/no fuel costs such as nuclear, wind, geothermal, hydro. Use of such technologies will tend to “front-load” production costs due to construction capital requirements. However, they offer in return low volatility pricing.
8. etc.
The price of electricity in the USA (and most likely “everywhere”) will increase during the next decade. But the maximum increase, even with cap & trade seems unlikely to exceed 100%. And, for the next 5 years it will probably remain “fairly” stable, say an increase of no more than 20%. In contrast, oil could spike to double its present price within the next 5 years. And, forecasts identify the possibility of price trebling, $200+/bbl, this decade.
With electricity prices providing more stability and lower fuel cost per mile, it is reasonable to expect that, as the decade and the scenario sketched above develop, a growing but still small minority of drivers will accept the cost premium associated with electrified cars.
Axion crashed below the buck today, presumably on news they might have to relocate to Russia. How do you see their chances for a U.S. rescue so they can stay in this country? I had read about the funding problem in the Youngstown Vindicator, but I guess the Inky story got more people’s attention. I have learned a lot reading your columns, I realize this is not your usual query, but should I sell what remains of my holding or hold on or buy more shares? Tx, Pat
My view of oil prices has a bit more stability than yours. The chart I’ve put together for my long term expectations is here:
http://www.altenergystocks.com/archives/2010/01/plugin_vehicles_waist_deep_in_the_big_muddy_1.html
We did have a spike way above trend in 08 and a matching strike below trend in 09, but the 10 year path is pretty well in the channel.
I generally agree that oil will be a good deal more volatile and probably increase in price more quickly than electricity, I don’t believe, however, that expecting price increases of a couple percent a year is reasonable in light of capital spending plans. Cap and trade impacts are a complete wild card for me, but they won’t be good for consumers.
Axion is very well funded and can expand comfortably at a modest rate using the $23 million in cash it had at year end. The only reason to go to a plant in another country would be developing demand that exceeds available expansion capital by a wide margin. Any time a company needs to expand rationally, it has to put its plant where the terms will maximize the value of its capital. I think all Granville is saying is “we’d rather build in PA, but we’ll go elsewhere with the facility if we can’t match the levels of support.”
I know the Axion management team very well from several years of work as a team member. The recent stories are very out of character unless there are prospective customers lined up around the block. It may unsettle the market a bit, but I think this may be very good news. One thing’s certain, the biggest volume day in Axion’s history before today was about 360,000 shares. Since sellers and buyers are always balanced, big volume days impress me.
John,
I understand the logic behind your “big muddy” chart. As a note, extrapolating your chart to 2020 produces price results roughly in keeping with the reference case for oil prices in the EIA AEO2010 report, about $120 /bbl. As a comparison, the low price for 2020 in that report is about $50 and the high price is about $190. Unless the world economy “falls off a cliff” or there is a truly remarkable technological development sometime soon, $50 seems very unlikely. An increasing amount of oil production is from more expensive sources. This marginal production will set the price floor in the $50+ area. On the other hand, a sustained price of $190 also seems unlikely given its effect on the economy. Having said that, a variety of “black swan” events can produce such sustained high prices.
What seems more likely to me is a continued up-trend (perhaps as you and the EIA forecast) with high and unpredictable volatility. It is this combination that, I believe, will drive consumer psychology regarding vehicular fuel economy.
In contrast, I believe that 2% per year price increases for electricity are possible, particularly for the next 5 years in the USA. Given the present state of the US economy and the excess capacity for electric generation in much of the US, it seems unlikely that the price of fuel for electric generation (primarily coal, nat gas and uranium) will increase much or that the investment in new electricity infrastructure will be robust. Rather, it seems that investment in infrastructure will be more economically cautious. As an example in Florida, FPL had planned substantial replacement, upgrading and additions to their generation, transmission and distribution assets. They recently announced they will continue with certain “smart grid” projects and with the replacement of two medium sized oil burners with large combined cycle natural gas plants. However, they have discontinued indefinitely some $8+ billion in projects including two new nuclear reactors and a new natural gas pipeline. They claim the $2 billion they will invest in the new CC plants will save their customers money due to the use of less expensive fuel and much higher generating fuel efficiency.
I believe there are numerous large projects with similar potential for good payback. In an era of tight money, these will receive the attention. (Two of the more exotic that I appreciate in particular are: replacement of pressure reduction valves with large fuel cells at natural gas city gates and; implementation of Silex technology by GE for uranium enrichment. The first can provide substantial new electric power at “zero” fuel cost while the second may dramatically reduce electricity consumption for the very energy intense enrichment process in an era of increased global demand for enriched uranium.) The combination of the subdued increase in prices for fuels used to generate electricity and the judicious improvement to infrastructure will, I contend, result in constrained price increases for electricity in the US during this decade. Interestingly, this is substantially in agreement with the EIA electricity price forecast in their aeo2010r.d111809a.
And, until more is known, I agree that the effect of cap and trade is largely conjecture. However, due to politics, I would be surprised if it had much effect in the US prior to 2015.