The Inevitable Future of Energy

Let’s dig into some of the technical details behind Reason Number 1 for the inevitability of the energy transition.

When thinking about the future of energy it is very important to become comfortable with units like Watt-Hours (Wh) and multiples there-of (eg. a thousand Wh becomes kWh pronounced kilo-watt-hours). Rather than thinking of energy in the traditional SI unit ‘joules’, it is useful to think of energy as power generated or consumed for some amount of time. For example, a microwave runs at about 1.2kW, so if you microwave a frozen burrito for 3 minutes, you use 1200W*3/60h = 60Wh of energy (this also happens to be the amount of energy a typical laptop battery can store). 

Now suppose you lived in a state where electric power was produced entirely through the combustion of coal. Coal used for this purpose in the US contains on average about 6Wh of chemical energy per gram. The average coal power plant in the US converts about 30% of that chemical energy into electricity. This means that each frozen burrito that gets microwaved in this hypothetical coal state requires 33g of coal to be burned. While that doesn’t sound like much we use electricity for a lot more than microwaving burritos. The average coal plant in the US has the capacity to produce 400MW (mega or million watts). Running at 400MW for an hour, that average plant burns more than 200,000kg of coal. That’s 200 metric tonnes per hour. To feed one average coal plant in the US, we dig up 200 tonnes, process 200 tonnes, transport 200 tonnes, and then deal with about 20 tonnes of ash left over - EVERY HOUR. 

There are quite a few ways for this complex fuel supply chain to go wrong and it has to keep running to produce every additional unit of energy. There are similar issues for natural gas and oil supply chains, with the added complexity that those commodities are traded on a global market (this wasn’t true for natural gas until recently, we now export about 10% of the NG we produce and we are building the capacity to export closer to 20%). Global commodity markets for fuel expose all participating nations to significant geopolitical risk and economic shocks. 

Time for the good news. Wind and solar power plants are now incredibly cheap and getting cheaper at a remarkable pace. Let’s make some rough comparisons of upfront costs for power plants in the US today just to get a sense for what I mean by incredibly cheap (estimates from NREL ATB):

• A typical coal plant costs $4-7/W (our average 400MW would plant cost around $1.6-2.8B to build today)

• A typical natural gas plant costs $1.3-2.4/W

• Onshore wind plants cost around $1.7/W

• Utility scale solar PV plants are around $1.5/W

This means that solar and wind are comparable or cheaper than fossil power plants in terms of upfront cost and they have zero fuel cost and zero fuel supply chain risk. The standard rebuttal to this is, “the sun doesn’t always shine and the wind doesn’t always blow.” Of course the sun is always shining and the wind is always blowing somewhere, connecting wind and solar generators to a grid spanning hundreds or thousands of miles largely solves the variability issue with batteries increasingly shifting solar power into the evening and night. 

Local variability in sunshine and wind means that over a year the average power output from a solar or wind plant in the US is only 20% and 35% of their peaks respectively. These percentages are known in the business as “capacity factors” (CF) and if we want a more fair comparison of cost we should divide our dollars per watt estimates by the typical capacity factors for each type of power plant:

• The average CF of the coal fleet in the US in 2023 was only 43%. While in principle these power plants could operate closer to 80% CF, in practice they are not competitive in power markets due to their high fuel costs and as such only get scheduled to run when they are needed to meet high demand. This effect drives their effective upfront cost to $9-16/<W> (where <W> means average power over a year)

• Natural gas plants also have high fuel costs; despite their ability to respond more quickly to changes in demand, their average CF in 2023 was only 36%! This brings their effective upfront cost to $3.6-6.7/<W>. 

• Onshore wind comes up to an effective $4.9/<W>

• Utility scale solar is around $7.5/<W>

It’s worth noting that these are averages and there is quite a lot of variability! These rough averages are useful for context and understanding the basic considerations for utility scale power generation investments. 

Now let us consider fuel costs; the primary reason coal and gas plants do not run near 80% CF in today’s market. Natural gas and coal (after accounting for generator efficiency) cost around $30/MWh with significant volatility around those average prices. In 2022 for example, the average NG price was more than double the average in 2023. Over a year our typical coal and NG plants will pay something like $0.10/<W> for fuel, adding $1/<W> for every decade they operate at this average CF. 

Meanwhile, solar, wind and batteries are the most expensive they will ever be, getting cheaper due to technological advancements and learnings from deployment every year with no end in sight. There are many examples around the world of grids operating at or close to 100% renewable energy today. For example, Iceland is almost entirely powered by renewables.

Wikipedia has a nice table ranking countries by their renewable power fraction, most of the ones with high fractions are primarily powered by hydropower and geothermal power. Getting to high fractions with solar and wind has only been achieved in small island nations on an annual basis, but many places reach high solar and wind fractions for significant portions of the year. For example, on May 8th 2022 California produced enough power from wind and solar alone to meet 103% of demand (can be greater than 100% because some power is exported from the state). Already in 2024, California has had 100 days where at least part of the day was 100% clean power. Batteries are being rapidly deployed and allowing California to replace evening natural gas peaker plant power with stored solar power. 

This is a very exciting time for the energy transition! There is a lot of work to be done and still a lot of questions to be answered. I am very excited about the potential for a world of true energy abundance enabled by extremely cheap variable renewable generation and energy storage. This will fundamentally reshape the global economy for the better. The transition will be messy and far too slow, but the end goal is beautiful.

EIA Data:

US generator power capacity by energy source: https://www.eia.gov/electricity/annual/html/epa_04_03.html

US generator output by energy source: https://www.eia.gov/electricity/annual/html/epa_03_01_a.html

Heat rates: https://www.eia.gov/electricity/annual/html/epa_08_01.html 

Fuel costs: https://www.eia.gov/electricity/annual/html/epa_07_04.html