Elon Musk’s April announcement that Tesla will manufacture a $35,000 electric vehicle prompted a huge amount of discussion in the investment community regarding the coming bull market in renewables and its impact on global oil markets.
As many of you know, oil’s primary use today is as a transportation fuel, and the introduction of an affordable electric vehicle powered by renewable energy sources has long been the dream of both environmentalists and geopolitical analysts who hope to see the importance of the Middle East de-emphasized. Following Mr. Musk’s announcement, both CNN and Bloomberg interviewed analyst after analyst, who all stated without a doubt, that the new vehicle would provide the final missing piece necessary to jump-start the so-called “renewable energy revolution.” Up to now, the renewable energy bull-market has been riddled with frustration and setbacks. With the introduction of Mr. Musk’s new lower-cost vehicle, analysts believe the global economy is now finally ready to make the switch away from hydrocarbons towards an economy powered by electricity generated from renewable sources.
Before going into an analysis of this complex subject, we should point out that the subject of renewable energy is something we have thought long and hard about, and one we have studied at great length for over a decade. Back in January 2007, in the energy section of the Chilton Global Natural Resource Fund quarterly letter, I wrote an essay on the oil bull market. In that essay, I predicted the bull market would take place as a three-act play, with each distinct “act” featuring a different set of characters. In re-reading that essay (written nearly a decade ago), I am fascinated by how much I got right and how much I missed. For example, I never predicted the huge impact the development of the oil-shale plays would have on North American supply. However, I did discuss how the third and final act of the great bull market in crude oil would revolve around the adoption of renewables. I wrote: “It is at this point [the third act] that investors will come to the realization that the problems of growing future oil supply is long-term and structural in nature. As many of you have heard me say, it will be investments in alternative energy producers that will be leaders of the energy bull market when the third leg, or “act”, of the bull market in energy plays out. Although the third act is still years away, we already doing our research into which companies will be the winners.”
Over the last 10 years, we have watched almost all of the alternative energy companies struggle (many have even gone bankrupt). The few alternative energy projects that have been successful have been heavily subsidized by the US government through investment tax credits or by utilities, which have been forced by state regulators to buy produced power at artificially high rates. If this is only the start of the “third act,” it stands to reason that the renewable industry has struggled. The time for renewables has not yet arrived. The oil price just isn’t high enough for renewables to be competitive. However, if our modeling of oil demand and supply is correct, we believe that the oil market’s “third act” has indeed begun. In this cycle, oil prices will experience a huge rise, and fears of supply reliability will re-emerge. The time for renewables has arrived. It’s now finally time to seriously think about the best ways to participate in the upcoming renewable-energy bull market.
While we are firm believers that renewable energy has a bright future, we think most people are wrong about what will force its adoption. For example, while most analysts predict that climate change initiatives, or technology breakthroughs will be the major factors behind renewable adoption (with resulting downward pressure on oil prices), we instead believe that it will ultimately be a tight oil market (and the resulting high prices) that will push consumers toward renewables.
As many of you know, we believe that energy and, more specifically oil, is becoming more and more supplied constrained. Although nearly all of the analysts today argue that the shale plays have left us awash in crude oil, we have a different perspective: the increasing scarcity of conventionally sourced oil has forced global energy companies to find and exploit more and more complicated and expensive sources of crude oil. For example, back in 2000, we calculated that the global oil industry needed $15 oil to bring on the marginal barrel of production (back then the marginal barrel was found in the offshore deep-water). Today things are very different. Even though huge advancements have been made with the oil shales here in the US, we nevertheless calculate the oil industry now needs a $60 oil price to make these plays work. In other words, despite the belief that we are awash in oil, the price needed to bring on additional supply has increased four-fold over the last 15 years -- not something you would expect with a traditional over-supplied market.
Also, everyone is under the assumption that the high-quality nature of the US and Canadian shales (and the resulting surge in production) will soon be replicated across the rest of the world. This is something that we do not agree with. We have completed a huge research project that ranks the prospectivity of every major shale basin in the world, and our research suggests that for the most part, the North American shales are unique in their high quality and productivity. We ranked all 450 of the major global shales according to a series of geological and geochemical properties, and what we found is that the vast majority of the world’s best shale plays are located here in the US and Canada. Once you get outside of North America, most shales do not have the properties necessary to make them ultimately productive: many contain too much clay, while others lack either the right pressure, or the necessary total organic content to make them high quality. If you believe our research, then it becomes obvious that the great shale revolution we have experienced here in the US and Canada, simply cannot and will not be replicated in the rest of the world, except for in a very few select places. In other words, we continue to believe that structural supply issues will re-emerge -- even despite the US shale oil revolution currently underway. As a result, oil prices will once again be pushed significantly higher in the years to come. For those that are interested, please look for our next quarterly letter where we will outline our conclusions about the world’s shale plays. We will go into great detail, rank the shales, and explain why most prospects outside of North America will likely not be productive. Please stay tuned.
It will be neither climate change issues, nor the falling costs of solar or wind energy that will push the renewable bull market forward, but rather high oil prices (pushed significantly higher by supply constraints and extremely strong demand) that will kick off the bull market in renewables. Also, we do not think that any of these developments will lead to the end of the “age of oil,” as many market prognosticators claim at present. In fact, we believe quite the opposite: as we mentioned, we believe the adoption of renewables will come about because of the bullish fundamentals in the crude oil market. Renewables will certainly have an ever-greater role in providing energy to the global economy; however, it will be the bull market in oil (and a renewal of scarcity fears) that will cause the bull market in renewables to take place. Please read on to the oil section of this letter to see exactly why a bull market renewable will not at all be at the expense of an equally strong bull market in crude oil.
Under this framework, what will emerge in the next several years as the best way to play the upcoming bull market in renewable energy? Our research tells us the answer is surprisingly simple: it will not be complicated new technology companies, or solar panel makers, or massive wind or solar projects themselves. Instead, the best way to play the bull market in renewables will be a very straight-forward and conventional product that needs little explanation: copper. Why? The dynamics currently affecting global copper supply; the continued strong growth of “old” sources of conventional copper demand; and most important, the emergence of “new” sources of copper demand related to the build-out of renewable energy sources. All of these factors will unite to produce a huge copper bull market, not dissimilar to the one that started in 2001, which lasted a decade, and which ultimately carried copper up higher by nearly seven-fold in price.
Let’s start with the growth in “conventional” sources of copper demand. In our July 2014 quarterly letter, we outlined our research that indicated why analysts are significantly underestimating traditional copper demand. We predicted that China’s copper consumption (which at the time represented over 50% of world demand), would continue to grow strongly, ultimately reaching upwards of 70% of total world demand sometime in the next 10 years. Our forecast was based upon a predictable and stable relationship between any given economy’s installed copper base, and its level of real per capita GDP. For example, in order for an economy to support a per capita GDP level of $10,000 (we use 2005 real dollars), it requires an installed base of copper of between 80 and 150 lbs. per person. When a country hits $15,000 of real per capita GDP, the total invested copper stock is tightly clustered around 200 lbs. per person. And finally, once an economy reaches $20,000 and $30,000 of per capita GDP, those economies require 400 lbs. and 600 lbs. of copper respectively.
To demonstrate just how stable this relationship really is, we calculated that Japan, South Korea, and Germany had an installed copper base of 600, 550, and 650 lbs. per person respectively by the time each economy reached $30,000 of per capita real GDP (and remember, these three economies were all very different in terms of industries, geographies and eras when they hit $30,000 of per capita GDP). By the end of 2013, with China’s real per capita GDP at $8,500, our models told us that an economy of its size should require 120 lbs. of installed copper per person. In order to check our models, we calculated China’s installed copper base at the end of 2013 and it worked out to be 123 lbs. per person—only three lbs. off from our models. We estimated at that time that given a 6.5% annualized rate of growth, China’s real GDP would reach $17,000 by 2024, which would require an installed copper base of 290 lbs. per person. In order to reach this installed copper base, and assuming global copper mine supply was able to grow by 3% per year (a level that we believe is unrealistically optimistic), China’s copper consumption would have to grow from 54% of world demand to a staggering 70% by 2024.
So, how well has this model performed in the two years since we made our first predictions? We had estimated that Chinese real GDP would grow by 6.5% per year and that by 2015, the installed base of copper would need to reach 144 lbs. per person. According to official figures, Chinese real GDP actually grew by 7% per annum and as a result, the total installed copper stock in China has now reached 153 lbs. per person – exactly in line with our models once you adjust for the higher-than-expected growth in real GDP. Back in 2014 we predicted that China’s copper consumption would double by 2024, and now that projection looks to be too conservative. China consumed 9.8 mm tonnes of copper in 2013, and we modelled that China should consume 11 mm tonnes of copper in 2016; however, based upon data from the World Bureau of Metal Statistic (WBMS), for the first four months of 2016, it now looks as though China will consume 12.6 tonnes of copper this year. And, please remember, China is only one of the many countries that is going through the stage in its development where it must add to its installed base of copper. Indonesia, the Philippines, Malaysia, Thailand, Vietnam, and India all need to see huge increases in their installed copper base if their economies are ever to reach middle income GDP levels.
We should point out the copper demand story we have described above only includes the historical “traditional” sources of copper demand that have existed over the last 70 years in the post-World War II environment. This story is notable in and of itself, if only because investors are completely underestimating the global demand for copper, based upon the historical relationships between per capita GDP and the installed copper stock. However, if our prediction regarding the coming bull-market in renewables is correct, then we are in the process of adding a massive source of “new” copper demand as well.
Why should a bull-market in renewables lead to surging copper demand? As most of you know, copper’s physical properties and price make it the ideal metal to be used for the generation, transportation and distribution of electricity. In fact, more than 50% of copper’s demand is related to electricity generation and transmission. However, what you may not know is that generating electricity from renewable sources such as wind and solar is much more copper-intensive than from traditional sources, such as coal, natural gas or nuclear. While a conventional power plant requires approximately one tonne of copper to generate one Mw of electricity, an offshore wind farm requires 5.7 to 9.6 tonnes of copper to produce the same amount of power (given the aesthetics and the greater load-factor reliability, it now looks like offshore wind farms will win the race to be the most favored method to generate renewable power). Photovoltaic solar can be better, but not by much: one Mw of power from a PV solar farm requires between 2.5 and seven tonnes of installed copper. On-shore wind farms meanwhile require 2.5 to 6.8 tonnes of installed copper per megawatt of power generated.
The reason for renewable energy’s high copper intensity is that power-generating units in either a solar or wind farm are widely dispersed, as opposed to a conventional power plant where all of the generation is centralized. For example, to replicate the electricity output of a single-turbine 600 Mw coal-fired steam generator requires a 200-unit wind farm. Since each unit in the wind farm requires its own copper-rich turbine, the total amount of installed copper is much higher with a wind farm than with a coal-fired power plant. Furthermore, whereas the electricity generated from a coal-plant is all in one location, the wind farm’s 200 units are widely spaced from each other and copper wire is required to gather electricity from each generating unit. Generating electricity at solar farms is also incredibly copper intensive. Although, solar farms don’t require copper-laden turbines, all of the electricity collected from hundreds of solar panels needs to be transformed and “stepped-up” using transformers, which are very copper intensive. As a result, the total required copper to generate electricity from a solar-farm is again much greater than from a comparable coal-fired power plant.
The other factor driving the huge copper intensity of renewable energy is related to the reliability and the load-factor of renewable sources. Other than standard maintenance downtime, a coal-fired or nuclear power plant can theoretically always operate. In practice, the average up-time or “load-factor” of a coal-fired power plant is greater than 60% while a nuclear power plant can exceed 70%. On the other hand, the average solar farm has a load-factor of only 10%, while the average windfarm has a load-factor of 20-30% (on-shore wind farms are at the low end of this range and off-shore wind farms are the high end). The difference is simply due to the fact that the sun may not always shine and the wind may not always blow whereas a conventional electricity producer is almost always able to run (other than during period of normally scheduled maintenance). Given the extremely low load-factors of renewable energy sources, more installed generation capacity is required to replicate the equivalent amount of electricity generated from a conventional power plant.
The diseconomies of scale that are associated with generating electricity from widely dispersed power units, combined with the extremely low load-factors, means that the copper intensity needed to generate electricity from renewable sources is orders of magnitude higher than from conventional sources. In fact, when adjusted for the additional capacity needed to compensate for renewable’s miserably low load-factors, the copper intensity to actually produce electricity from renewable sources is estimated to be 30 to 40 times greater than what is needed to generate electricity from a conventional power-plant.
Using estimates from the International Energy Agency, (and given our assumption that the next leg of the oil bull market has started) we project renewable-sourced power will grow from 5% of total world electricity output in 2015 (1,160 trillion watt hours), to 9% by 2025 (2,600 trillion watt hours), which represents an 8.5% compounded rate of growth. Given the various copper intensities and load-factors mentioned above, and assuming that a large amount of the renewable power is generated by off-shore wind farms (by far the most efficient), we can begin to make a rough estimate of the incremental copper demand needed to supply this amount of electricity. Between now and 2020, we project that renewables will represent 400,000 tonnes per year of entirely “new” copper demand and that by 2025, this will accelerate to 500,000 tonnes per year.
To put these figures in perspective, over the last 15 years, copper demand has increased by an average of approximately 500,000 tonnes per year – so renewables have the potential to nearly double annual copper demand growth. And remember, these demand numbers only include the additional copper needed to generate electricity from renewable sources. If the demand for electric cars takes off (which goes hand in hand with a growth in renewables), then this will add a whole new dimension to the copper demand story, something that we discuss in the oil section of this letter.
It is unclear how the world will meet these entirely new sources of copper demand. Given that copper mine-supply has only grown by about 400,000 tonnes per year over the last 15 years, it is hard to imagine how the copper market can avoid shifting into a large structural deficit over the next several years. Adding to the copper shortfall going forward, we believe that copper supply will continue to disappoint relative to expectations over the next five years. In our third quarter 2014 letter we wrote about our outlook for copper mine supply over the next ten years. In that essay, we discussed how almost all of the increases in copper reserves over the previous ten years had been the result of higher copper prices, which in turn allowed companies to lower their “cut-off” grades when calculating their reserve base. In fact, our models told us that only a tiny fraction of the industry’s reserve additions had come from any new discoveries. Also, our research told us that because of the “log-normal” shape of a typical porphyry’s reserve grade distribution, the copper industry’s ability to further increase reserves by lowering the “cut-off” grade had come to an end. We concluded that copper mine supply would significantly disappoint as we progressed through the second half of the decade.
Our analysis has proved to be correct: in the two years since we wrote that essay, copper mine supply has, in fact, severely disappointed relative to analysts’ expectations. For example, consensus estimates for copper mine supply growth for both 2014 and 2015 were approximately 1.3 mm tonnes for both years (7% growth). Actual copper mine supply growth for 2014 came in at only 200 tonnes (1.2% growth) and 2015 copper supply growth was 850 tonnes (4.6%). Given mine disruptions and project delays, actual production in both years fell far short of consensus estimates. Despite these shortfalls, analysts today are nevertheless betting that 2016 copper supply growth will be very strong—1.2 mm tonnes. While we admit that mine supply growth in 2016 has so far been strong (according to the WBMS), it has once again lagged relative to consensus. Furthermore, our models tell us that whatever mine supply growth we do experience in 2016 will end up being the exception as we progress through the rest of the decade. The reason is that while Peru will see production continue to surge into 2017 (led by the start-up of four massive new mines – Las Bambas, Toromocho, Constanzia and the Cerro Verde Expansion), Peruvian production will definitely slow materially after that. Similarly, in Kazakhstan, production is surging right now as two new mines are being brought on line (Aktogay and Bazshakol); however, after these mines ramp up production growth will grind to a halt after 2017. Outside of these two bright-spots, global copper mine supply will hardly grow, according to our models. As a result, after 2016, global copper mine supply growth should decelerate materially, and will remain muted all the way into the next decade.
However, we will not have to wait until next decade for the bull market in copper to begin. In fact, because of the very strong demand growth we outlined above, the global copper market may have shifted into structural deficit already. Most investors have been preoccupied with a potential severe slowdown in China’s economic growth and few investors have paid any attention to this year’s copper market fundamentals. If investors had looked, they would be surprised by the strength in 2016 copper demand growth. According to WBMS global copper demand for the first four months of the year has been up an extremely strong 7% versus the first four months of 2015. China’s demand, driven by all of the factors already discussed, was up 12% and continues to surprise to the upside. Non-OECD copper demand excluding China was also up a very strong 7%. Even OECD demand was up 2% in the first four months of 2016 versus last year. The only negative fundamental data point in the copper market surrounds supply. As we already discussed, copper mine supply grew by 5% in the first four months of 2016, according to the WBMS. However, even with this strong level of supply growth (which we think will slow materially), the copper market is in deficit today according to the WBMS figures. Global copper inventories on the listed exchanges have been steadily drawing down this year, which confirms the deficit suggested by the WBMS figures. Since the end of 2015, the combined copper inventories on the COMEX, LME, and Shanghai Metal Exchange Exchanges have been drawn down from over 600,000 tonnes to 400,000 tonnes at present. This is exactly what you would expect if global copper demand were indeed exceeding supply.
To summarize: we believe that copper demand is going to continue to surprise to the upside over the next five to 10 years. “Old” sources of traditional demand will continue to grow strongly while “new” sources of demand related to the build-out of renewable energy sources are set to surge. Our modelling also tells us that copper mine supply will continue to disappoint as we progress through the end of this decade. Strong demand and disappointing supply means that the copper market has now slipped into structural long-term deficit. Look for a potential massive bull market to develop in copper very soon. Given the extremely bearish sentiment towards copper (COMEX speculators are once again approaching record net-short positions) and given the emerging deficit we described, we think copper is setting up for a huge upward move. Investors are bearish, the stocks are at rock-bottom valuations, and there is an incredibly bullish fundamental story developing in the metal, both from the supply and demand side.