- When will renewables be cheaper than fossil fuels?
According to the annual Lazard LCOE (levelised cost of energy) study, renewables are already cheaper without any subsidies than fossil fuels in most parts of the world. Indeed, in some areas, the most expensive form of solar energy is cheaper than the cheapest form of conventional energy on an LCOE basis. In addition, fuel-price inflation or volatility is less of a risk for renewables because wind and sunshine is relatively predictable compared to fossil fuel prices, making it easier to lock in long-term prices. Finally, renewables projects are much simpler to undertake from an engineering perspective. As such, the projected cost at the start of a renewables project is much more likely to be accurate than a conventional energy project. According to the Union of Concerned Scientists, the average nuclear power station project runs over budget by approximately +107%, whilst wind and solar project costs overrun by less than 10% and less than 1% respectively.
The average cost of energy (unsubsidised)
2. Will we be able to entirely replace fossil fuels with renewable energy?
Yes. We believe this is not a question of if but when. The reality is that it will take decades to completely wean ourselves off fossil fuels, even if we invest heavily now. But renewables should continue to get cheaper due to relative technological immaturity and scaling up of the industry. In the future, the areas with the best wind and sunshine will probably export renewable energy, much like they do fossil fuels today. Below is a thought experiment from Quora which was referenced by Forbes:
If we cover 43,000 square miles of the Earth with solar panels, even with moderate efficiencies achievable easily today, it will provide more than 17.4 TW power. The Great Saharan Desert in Africa is 3.6 million square miles and is prime for solar power (more than twelve hours per day). That means 1.2% of the Sahara desert is sufficient to cover all of the energy needs of the world from solar energy. The cost of the project would be about five trillion USD, a one-time cost at today’s prices without assuming any economy of scale savings.1
If $5 trillion is anywhere near correct, it is very cheap indeed. A recent estimate (see chart below) estimates $2.1 trillion of capex on renewables by 2050. We view this as a conservative estimate. For context:
- Estimated $7.9 trillion capex has been spent on the oil and gas sector since 1998.
- Estimated $1.5 trillion has been spent on the recent US tax cut
- World GDP is about $50 trillion annually.
- The equivalent estimated cost 17.4 TW of nuclear power would be $52 trillion cost (pre cost overrun)
The IMF estimates that we are actually subsidising fossil fuels indirectly to the tune of 300bn euros per year if environmental damage is taken into account. According to Munich Re, losses from natural disasters are on the rise; they estimate that the cost in 2017 alone was $330 billion. Like I said, a $5 trillion one time cost is cheap.
Source: DNV GL as of 30/09/17. Data from 2015-2050 is estimated or forecast
3. When will electric vehicles with decent range be cheaper than the equivalent oil burner?
If you can afford to (and want to) spend over £50k on a car, then you can get an electric vehicle today that is equivalent to, or cheaper on a three-year cost of ownership basis, than the equivalent oil burner (i.e. the Tesla Model S is cheaper than the BMW 7 Series). In 2019, with the release of the Tesla Model 3 in the UK, drivers will be able to get an electric vehicle for under £40k that is equivalent to a BMW 3 series. Whilst this serves a broader portion of the economy, and is a potential game changer for the electric vehicle adoption, it will probably take a further 2-3 years before affordable cars become available for mass adoption at the sub £20k price range. However, there are multiple car manufacturers launching such cars in the near future. Remember, the largest proportion of and EV cost is the battery, which continues to decline in cost every year.
4. Will we have the resources and energy grid to cope with electric vehicle demand?
The disruptive adoption of a new technology requires innovation. The prevailing assumption that the electricity grid of today will not adapt to renewables is flawed in our view. Utility-scale storage is becoming economical for storing excess supply and it can be fed into the grid during peak demand. New ways to offer on-street charging, like lamppost charging points, combined with bi-directional charging (i.e. using consumer’s electric vehicles as a battery which absorbs excess power and then returns it to the grid during peak demand) are examples of innovations that enable charging and disrupt the traditional utility model.
Interestingly, peak demand for electric vehicle charging will be at night, when demand for other electricity is at its lowest. Morgan Stanley has estimated $1.7 trillion capital expenditure is required on electric vehicle infrastructure by 2040. As referenced earlier, this is just 20% of the $7.9 trillion spent on oil & gas capex since 1998. The resources required to support the building of batteries and renewables are generally abundant and, as the industry scales, the vast majority of materials used will become recyclable and reusable.
5. What companies should we invest in to benefit from these disruptive trends?
The companies most easily linked to a particular trend are rarely the best companies to invest in. Disruptive companies tend to come from the ‘left field’ and offer something new and innovative that the incumbents either haven’t thought of, have found too difficult or have tried to prevent due to the threat it poses to their existing business. Mature incumbents are therefore the last place we look for investment ideas to play disruptive trends.
Furthermore, companies that become synonymous with a particular trend (i.e. electric vehicles = Tesla) might be good investments but suffer from two main issues:
1) They are very well known and intensely analysed by market participants, and
2) Tend to quickly attract elevated valuations which can be very difficult to live up to.
So where do we look? We have three key preferences when investing in disruptive trends:
1) We prefer innovative and disruptive, typically not mega-cap, companies that are in growth mode.
2) We prefer companies that are unheard of by the wider public and have little sell-side coverage.
3) We demand that they have a powerful enough position in the value chain to consistently capture a return above their cost of capital as they grow.
The last point is important because too often investors make the mistake of chasing a trend and investing in commoditised products and services which are temporarily benefiting from a trend but which have no discernible competitive advantage.
More questions? Perhaps we’ve already answered them in previous soapbox articles:
Just one thing (most common questions about electric vehicles) here
Joined up thinking (Cross sector inefficiencies when investing in disruptive trends) here
Statistically significantly wrong (difficultly forecasting disruptive trends) here
America’s new pollution king (cars now emit more CO2 than coal) here.
Opinions and views from the Equities team at Kames Capital are not an investment recommendation, research or advice and should not be considered as such. Content discussing investment strategies and stocks is derived from and solely relates to the investment management activities of Kames Capital.
About the author
Craig Bonthron is an investment manager in the Equities team, responsible for co-managing global equities portfolios. He joined us in 2014 from SWIP, where he was investment director in global equities. In addition Craig also had analysis responsibilities for the tech, energy and utility sectors. Prior to SWIP, he was a portfolio manager at Kleinwort Benson Investors, a member of the global environmental equity team. Craig has a 1st Class honours degree in Building Surveying, an MSc with Distinction in Business Information Technology Systems from Strathclyde Business School. He has 16 years’ industry experience (as at 31 October 2017).