How the World Really Works by Vaclav Smil
Energy supply, food production, and the material world
I began the year not knowing a single thing about energy and commodities. But then Russia invaded Ukraine, the West imposed sanctions, and the impact to energy and commodities has been hard to miss. Here’s a smattering:
Russia sent the rest of Europe into an energy crisis after shutting off its gas exports.
Britain implemented price caps on sky rocketing energy and gas bills. The European Union is expected to follow suit with its own package to tame high energy bills.
California barely avoided blackouts amid a record heat wave.
The chief economist at the United Nations Food and Agriculture Organization expressed concern that a prolonged Ukraine war could trigger food insecurity because Russia’s a leading exporter of fertilizers, and both Russia and Ukraine are major exporters of cereals.
We’ve seen some impact of this already: Sri Lanka’s catastrophic organic farming transition was exacerbated by soaring grain, fertilizer, and fuel prices.
This energy crisis and the Ukraine war animate the latest discussions on the transition to clean energy. To learn more on the subject, I was in search of a book that’d offer a concise, just-the-facts approach on how the world works through the lens of energy. And in my search, one person’s name kept coming up: Vaclav Smil — a highly respected scientist whose work spans energy, food production, environmental and population studies, and public policy.
I think I first heard about Smil because he’s Bill Gates’ favorite author:
When I saw this tweet, I rolled my eyes pretty hard. Who looks forward to a new textbook more than a movie premiere? Have you even seen Star Wars, Bill? I found it so inconceivable that I went down a Google rabbit hole into Smil’s work.
Smil’s written a lot of books diving deep (like really deep) on topics like global food production, energy transitions, the Earth’s biosphere, manufacturing, and much much more. He condenses these subjects into single chapter treatments that flow well with one another in his newest book, How the World Really Works.
I found Smil’s approach refreshing. He describes himself as neither a pessimist nor an optimist, but a scientist. His aim is to “reduce comprehension deficit” by describing the “fundamental ruling realities governing our survival and our prosperities.”1 He just wants to explain how the world actually works.
I’m going to highlight three chapters from this book:
Fuels and electricity
Food production
The material world and the four pillars of modern civilization
Fuels and Electricity
Usually when I hear people describe “how the world works," it’s from the perspective of international relations, foreign policy, global trade, and world history. That's one aspect of it, but Smil has a different angle. He points out that energy conversions “are the very basis of life and evolution” and that “history can be seen as an unusually rapid sequence of transitions to new energy sources.”2 He writes:
Physicist Robert Ayres has repeatedly stressed in his writings the central notion of energy in all economies: “the economic system is essentially a system for extracting, processing, and transforming energy as resources into energy embodied in products and services.” Simply put, energy is the only true universal currency, and nothing (from galactic rotations to ephemeral insect lives) can take place without its transformations.
I’m receptive to that, but what is energy? The most common definition is the “capacity for doing work,” but there’s more to it than that. Energy has many different forms, and each form has its own formula. There’s gravitational energy, kinetic energy, heat energy, electrical energy, nuclear energy, and more. To make energy useful, we have to convert it from one form to another. We take for granted what it takes to substitute between different forms of energy — swap out chemical energy here for electrical energy there. Sometimes replacing energy is cheap, reliable, quick and cost effective. But often times it’s not.
For example, the transition from coal to crude oil (which needs to be refined to create products like gasoline, aviation kerosene, and diesel) took generations to accomplish. Crude oil extraction first started in 1850 and it has distinct advantages over coal and firewood (the prevailing fuel sources at the time). Crude oil has much higher energy density, meaning it takes up a lot less space for the same energy production, compared to coal or firewood. Crude oil is also much easier to produce than coal and easier to transport. But despite these advantages, it wasn’t until the 1950s — when new markets for refined oil products started to take off, like gasoline fueled vehicles, jetliners, diesel fueled trucks, ships, and heavy machinery — that crude oil was widely adopted.
This chart from Our World in Data3 helps visualize this in the context of all other primary energy sources.
Historically, energy transitions have taken a really long time. Think about this in light of calls to flip the switch from fossil fuels to renewables within a couple decades. There's no doubt we should be working towards decarbonizing our energy generation, but Smil argues that the world will be reliant on fossil fuels for a lot longer than most think4:
Reliance on fossil fuels has created the modern world, but concerns about the relatively rapid rate of global warming have led to widespread calls for doing away with fossil carbon as expeditiously as possible. Ideally, the decarbonization of global energy supply should proceed fast enough to limit average global warming to no more than 1.5°C (at worst 2°C). That, according to most climate models, would mean reducing net global CO2 emissions to zero by 2050 and keeping them negative for the remainder of the century.
Using wind and solar to decarbonize electricity generation a viable strategy because “installation costs per unit of solar or wind capacity can now compete with the least expensive fossil-fueled choices.”5 But Smil draws a distinction between a country that derives 20-40% of electricity from renewables vs. a “national electricity supply that relies completely on these renewable flows.”6 Take Germany for example.7
In 2021, Germany generated 42% of its electricity from renewables, but they still have to keep fossil fuel capacity on reserve to handle demand when it’s cloudy or not windy. That’s because we don’t yet have a cheap way to store solar and wind electricity — batteries “have capacities orders of magnitude lower than needed by large cities, even for a single day’s worth of storage,”8 and at least in Germany and the US, construction lines to transmit wind electricity have lagged as wind generation has made up an increasing share of electricity mix.
But what about Denmark? They generate 75% of their electricity from renewables. Why can't Germany do the same? It's because Denmark’s energy demand is 1/20th of Germany’s9, so Danes don’t need the level of reserve capacity that Germans do. And that reserve capacity is enough to contribute meaningfully to fossil fuels' share of total energy generation.
One way to help with large scale electric storage is pumped hydro storage (PHS), but it has its own drawbacks. Smil explains10:
We need very large (multi-gigawatt-hour) storage for big cities and megacities, but so far the only viable option to serve them is pumped hydro storage (PHS): it uses cheaper nighttime electricity to pump water from a low-lying to high-lying storage, and its discharge provides instantly available generation. With renewably generated electricity, the pumping could be done whenever surplus solar or wind capacity is available, but obviously PHS can work only in places with suitable elevation differences and the operation consumes about a quarter of generated electricity for the uphill pumping of water.
Nuclear energy is likely a better option. Nuclear reactors are safe, reliable, and they last 40+ years. But Smil explains why “the future of nuclear generation remains uncertain.”11:
America’s new small, modular, and inherently safe reactors (first proposed in the 1980s) have yet to be commercialized, and Germany, with its decision to abandon all nuclear generation by 2022, is only the most obvious example of Europe’s widely shared, deeply anti-nuclear sentiment.
Let’s also recognize that decarbonizing electricity is just one part of net-zero emissions efforts. We have to decarbonize our entire primary energy consumption, of which electricity is only 18%.12 Worldwide demand for fossilized carbon is north of 10 billion tons a year annually, which in 2019, represented 80% of total global demand. Personally, I'm long on human ingenuity and I'm certainly more of an optimist than Smil, but I'm inclined to agree with him that the next few decades likely won't be characterized by "a sudden abandonment of fossil carbon, nor even its rapid demise -- but rather its gradual decline."13
Food Production
What does it take to feed the world? In 1950, we could only feed 35% of the world’s population, but by 2000, that number rose to 85%14. Pretty impressive considering the world’s population grew from 2.5 billion to 7.7 billion in the same period.15 Today, it takes less than two seconds of human labor to produce a kilogram of wheat, whereas in 1801 it took 150 hours. How were we able to achieve such a feat? We couldn't have had efficient harvests or high crop yields without fossil fuels.
On one hand, coal is used to make steel for farm machinery — think 4-wheel drive tractors that can get up to nearly 700 horsepower, steel plows, drills, seeders, and fertilizer applicators. And of course, these machines use diesel or gas for day-to-day operations. But I was surprised to learn that agrochemicals (e.g., fungicides, insecticides, herbicides, and fertilizers) require far more energy than farm machinery.
Fertilizers provide plants with three macronutrients: potassium, phosphorous, and nitrogen. Before the advent of synthetic fertilizer, the most common way to to enrich soil's nitrogen stores was to use human and animal waste. But organic matter has really low nitrogen content, so you need a lot of it, — like 10-30 tons of waste/hectare16 (literally a shit ton) — to produce enough nitrogen. It was both very labor intensive and inefficient.
There were concerns at the beginning of the twentieth century that we might not have enough food to feed the world’s fast growing population. But then two German scientists, Fritz Haber and Carl Bosch, invented ammonia out of basically thin air. It wouldn’t be an understatement to argue that Haber-Bosch was the greatest invention in the last 100 years: ammonia is the basis for synthetic nitrogen fertilizer, which Smil estimates17 is responsible for feeding about half of the world’s population today. This lines up with Our World in Data's finding as well18:
Smil takes this a step further and conducts an accounting of exactly how much fossil energy is required as input to produce food. How much fuel is required to produce a loaf of bread? A kilogram of chicken? How about a tomato? Seafood?
When you consider the fuel and electricity it takes to grow the grain, mill it and bake it, Smil estimates that it takes 210-250 mL of diesel fuel per kilogram of bread.19 That’s around a cup of diesel oil for every loaf of bread.
Chicken’s a little more energy intensive. Considering what it takes to produce chicken feed, run a poultry house, slaughter the birds, store, refrigerate, and distribute, Smil estimates a cost of 300-350 mL per kilogram (a single chicken typically weighs around 1 kg).
Tomatoes (or as Smil calls them, “appealingly shaped containers of water”) seem environmentally friendly. Tomatoes are vegan and they're a fixture in the Mediterranean diet. But “in order to increase their yield, improve their quality, and reduce the intensity of energy inputs, tomatoes are increasingly grown in plastic-covered single- or multi-tunnel enclosures or in greenhouses…”20 Cultivating tomatoes require a lot of plastic, and plastic is made from crude oil and natural gas. Tomatoes are also "among the world's most heavily fertilized crops: per unit area they receive up to 10x as much nitrogen (and also phosphorus) as is used to produce grain corn, America's leading field crop."21 After considering production and energy costs, Smil estimates it takes 650 mL of diesel per kilogram to produce a medium-sized tomato. That's a staggeringly high energy cost relative to bread and chicken.
Seafood isn't any better, with an average energy expenditure of 700 mL per kilogram.
There's no two ways about it, "our food supply...has become increasingly dependent on fossil fuels" and it's not something that can "be changed easily or rapidly...the ubiquity and the scale of the dependence are too large for that.22
It’s not feasible to go back to organic farming, but we could try to waste less food. The United Nations Food and Agriculture Organization estimates that “the world loses almost half of all root crops, fruits, and vegetables, about a third of all fish, 30% of cereals, and a fifth of all oilseeds, meat, and dairy products — or at least one-third of the overall food supply.”23 We could also change our diets -- choose meat with a lower carbon footprint and eat a moderate amount of it. But I'm not sure that's an appropriate universal recommendation -- as Smil notes, there are people in many countries who would stand to benefit from having more meat in their diet. We could make nitrogen fertilizers more efficient -- about two-thirds of it is lost from plant uptake. And we could also transition our tractors and farm machinery to electric. But these changes (which are by no means an exhaustive list) are all a long way off.
The Material World: the Four Pillars of Modern Civilization
How much of the stuff in our world could we comfortably live without? It’s hard to imagine life without computers or mobile devices. But when you boil it down, Smil argues there are four materials that have proved critical to the modern world: ammonia, plastics, steel, and cement. And (you guessed it!) they all require significant use of fossil fuels. Smil explains24:
Global production of these four indispensable materials claims about 17 percent of the world’s primary energy supply, and 25% of all CO2 emissions originating in the combustion of fossil fuels — and currently there are no commercially available and readily deployable mass-scale alternatives to displace these established processes. Although there is no shortage of proposals and experimental techniques to produce these materials without relying on fossil carbon…none of those alternatives has been commercialized..it would obviously take decades to displace the existing capacities that are producing, at affordable prices, at annual rates of hundreds of millions to billions of tons.
I covered ammonia and the Haber-Bosch process in the prior section, so I’ll just say a bit about plastic, steel, and cement.
It’s not revelatory to point out that plastic is everywhere. Everyone gets that. But why? What’s so special about plastic that we use it for everything? Plastic is highly malleable, so it can be molded into virtually any shape. It’s a light, durable material that’s stronger than wood and glass and can handle a fair amount of stress before breaking. There’s also a wide range of plastics that all have slightly varying physical properties, each of which make them uniquely suitable for different kinds of products.
Steel has physical properties that lend itself to being the backbone of modern infrastructure. It's used in everything from skyscrapers, bridges, cars, planes, public transportation, TV/radio towers, wind turbine towers, ship hulls of oil and liquefied-gas tankers, and more. Container ships transport “cargo in steel crates of standardized dimensions…[and] chances are that everything you wear was carried to its final point of sale in a steel container that started its journey in a factory in Asia.”25 These tools and machines are assembled by machines that are also made of steel, and the electricity generated to power these machines wouldn't be possible without steel either.
Even though it's recyclable, steel’s impact on the environment is nontrivial: steelmaking requires coal and natural gas, and the World Steel Association estimates that steelmaking contributes 7-9% of the world’s carbon emissions.26
Concrete (of which cement is an ingredient) is the most “massively deployed material of modern civilization…particularly when reinforced with steel.”27 Though “much less energy intensive,” cement's “global output is nearly three times that of steel, [and] its production is responsible for a very similar share (about 8 percent) of emitted carbon.”28 Concrete is typically reinforced with steel, and reinforced concrete is everywhere from highways, airports, and buildings to the world's biggest bridges and dams.
We use a lot of cement in the US. We consumed 30 million tons of cement in 1928 building our highway system and airports, and in recent years we’ve consumed around 100 million tons per year.29 But that doesn’t hold a candle to China, who in 2019 produced 2.2 billion tons of cement. Here’s the environmental impact of cement by country:
That works out to a little over half the world’s carbon emissions from cement that can be attributed to China.
How much of this can be assuaged with transitioning to renewable electricity generation (i.e., using wind and solar)? Today, much of renewable electricity generation is highly dependent on steel, cement, and plastics30:
No structures are more obvious symbols of “green” electricity generation than large wind turbines — but these enormous accumulations of steel, cement, plastics are also embodiments of fossil fuels. Their foundations are reinforced concrete, their towers, nacelles, and rotors are steel (altogether nearly 200 tons of it for every megawatt of installed generating capacity), and their massive blades are energy-intensive — and difficult to recycle — plastic resins (about 15 tons of them for a midsize turbine). All of these giant parts must be brought to the installation sites by outsized trucks and erected by large steel cranes, and turbine gearboxes must be repeatedly lubricated with oil.
Even electric cars require a high amount of steel, aluminum, plastics, and a bevy of metals (lithium, cobalt, nickel, copper, graphite). As we ramp up electric car production in the coming decades, we’re going to need to find new ways to extract and process lithium, cobalt, and nickel to match growing EV adoption, and that’ll likely require a mix of both fossil fuel and electricity generation.
In short, we're going to be dependent on on these four materials -- ammonia, plastic, steel, and cement -- for a long time. Smil notes that "in 2019, the world consumed about 4.5 billion tons of cement, 1.8 billion tons of steel, 370 million tons of plastics, and 150 million tons of ammonia, and they are not readily replaceable by other materials." 31
Where does that leave us?
Apparently, some EU researchers developed a scenario we can halve our global per capita energy demand by 2050. I'm not sure that's entirely realistic. It's not clear, for example, how that squares with the pace of China's growth32:
In 1999 the country had just 0.34 cars per 100 urban households, in 2019 that number surpassed 40. That is a more than 100-fold relative increase in only two decades. In 1990, 1 out of every 300 urban households had an air-conditioning window unit; by 2018, there were 142.2 units per 100 households: a more than 400-fold rise in less than three decades.
And what does it mean for developing nations who are trying to increase their standard of living and grow their economies, even if it's at a fraction of China's growth?
That’s not to suggest that the world is doomed or that we’re hopeless. There's a lot we can do: expand our nuclear generation, transition electricity generation to solar and wind, continue to ramp up electric cars, and much more.
But instead of arbitrarily setting targets for years ending in 0s or 5s, we should be honest about where we are today and what it'll actually take to achieve net-zero global emissions. It’s not fatalistic to “admit the limits of our understanding, approach all planetary challenges with humility, and recognize that advances, setbacks, and failures will all continue…as long as we use our accumulated understanding with determination and perseverance, there will also not be an early end of days.”33
Accordingly, the takeaway isn't that we should be pessimistic. It's that if we're truly serious about tackling this crisis, we need to understand the problem space thoroughly and be clear-eyed about what we're going up against. People aren't great at predicting the future (Smil has a whole chapter on that), but I think we can reasonably claim that the future holds “a mixture of progress and setbacks, of seemingly insurmountable difficulties and near-miraculous advances. The future, as ever, is not predetermined. Its outcome depends on our actions.”34
Pg. 6
Pg. 20
Hannah Ritchie, Max Roser and Pablo Rosado (2020) - "Energy". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/energy' [Online Resource]
Pg. 34
Pg. 38
Pg. 38
Hannah Ritchie, Max Roser and Pablo Rosado (2020) - "Energy". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/energy' [Online Resource]
Pg. 40
Pg. 39
Pg. 40
Pg. 40
Pg. 41
Pg. 43
“Feeding the world” is based on an estimate from the United Nations’ Food and Agricultural Organization on the fraction of the world’s undernourished people. Pg. 46
Pg. 46
Pg. 54
Pg. 68
Hannah Ritchie and Max Roser (2013) - "Fertilizers". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/fertilizers' [Online Resource]
Pg. 58
Pg. 59
Pg. 61
Pg. 63-64
Pg. 71
Pg. 78-79
Pg. 91
Pg. 94
Pg. 95
Pg. 94
Pg. 98
Pg. 101
Pg. 78
Pg. 194
Pg. 226
Pg. 229
Qué impacto crees que tiene el trabajo de Smil en cómo entendemos los sistemas y las complejidades del mundo que nos rodea?
https://giligroup.com/