How the World Really Works: The Science Behind How We Got Here and Where We're Going

by Vaclav Smil

Electricity is only 18 percent of total final global energy consumption, and the decarbonization of more than 80 percent of final energy uses—by industries, households, commerce, and transportation—will be even more challenging than the decarbonization of electricity generation. Expanded electricity generation can be used for space heating and by many industrial processes now relying on fossil fuels, but the course of decarbonizing modern long-distance transportation remains unclear.

And while the US has greatly reduced its dependence on coal—replaced by natural gas in electricity generation—the country’s share of fossil fuels in primary energy supply was still 80 percent fossil in 2019. Meanwhile China’s share of fossil fuels fell from 93 percent in the year 2000 to 85 percent in 2019—but this relative decline was accompanied by a near tripling of the country’s fossil fuel demand. The economic rise of China was the main reason why the global consumption of fossil fuels rose by about 45 percent during the first two decades of the 21st century, and why, despite extensive and expensive expansion of renewable energies, the share of fossil fuels in the world’s primary energy supply fell only marginally, from 87 percent to about 84 percent.[83]

The only foraging societies with high population densities were coastal groups (most notably in the Pacific Northwest), who had access to annual fish migrations and plentiful opportunities to hunt aquatic mammals: reliable supply of high-protein, high-fat food allowed some of them to switch to sedentary lives in large communal wooden homes, and left them with spare time to carve impressive totem poles. In contrast, early agriculture, where the just-domesticated crops were harvested, meant that more than one person per hectare of cultivated land could be fed. Unlike the foragers who might have gathered scores of wild species, practitioners of early agriculture had to narrow the variety of the plants they cultivated, as a few staple crops (wheat, barley, rice, corn, legumes, potatoes) dominated typical, overwhelmingly plant-based, diets—but these crops could support population densities that were two or three orders of magnitude higher than in foraging societies.

Now most people in affluent and middle-income countries worry about what (and how much) is best to eat in order to maintain or improve their health and extend their longevity, not whether they will have enough to survive.

The United Nations’ Food and Agricultural Organization (FAO) estimates that the worldwide share of undernourished people decreased from about 65 percent in 1950 to 25 percent by 1970, and to about 15 percent by the year 2000. Continued improvements (with fluctuations caused by temporary national or regional setbacks due to natural disasters or armed conflicts) lowered the rate to 8.9 percent by 2019—which means that rising food production reduced the malnutrition rate from 2 in 3 people in 1950 to 1 in 11 by 2019.[4]

The steep reduction in global undernutrition means that in 1950 the world was able to supply adequate food to about 890 million people, but by 2019 that had risen to just over 7 billion: a nearly eight-fold increase in absolute terms!

Saying that the increase has been the combined effect of better crop varieties, agricultural mechanization, fertilization, irrigation, and crop protection correctly describes the changes in key inputs—but it still misses the fundamental explanation. Modern food production, be it field cultivation of crops or the capture of wild marine species, is a peculiar hybrid dependent on two different kinds of energy. The first, and most obvious, is the Sun. But we also need the now indispensable input of fossil fuels, and the electricity produced and generated by humans.

People everywhere will point out the combustion of liquid fuels that power most of our transportation but the modern world’s most important—and fundamentally existential—dependence on fossil fuels is their direct and indirect use in the production of our food.

Indirect use is much broader, taking into account the fuels and electricity used to produce agricultural machinery, fertilizers, and agrochemicals (herbicides, insecticides, fungicides), and other inputs ranging from glass and plastic sheets for greenhouses, to global positioning devices that enable precision farming.

What has changed is the intensity of our crop, and animal, production: we could not harvest such abundance, and in such a highly predictable manner, without the still-rising inputs of fossil fuels and electricity. Without these anthropogenic energy subsidies, we could not have supplied 90 percent of humanity with adequate nutrition and we could not have reduced global malnutrition to such a degree, while simultaneously steadily decreasing the amount of time and the area of cropland needed to feed one person.

This has always been so, from the very beginnings of settled cultivation going back some 10 millennia—but two centuries ago the addition of non-solar forms of energy began to affect the crop production and later also the capture of wild marine species. Initially this impact was marginal, and it became notable only in the early decades of the 20th century. To trace the evolution of this epochal shift, we’ll look next at the past two centuries of American wheat production. However, I could quite easily have chosen English or French wheat yields, or Chinese or Japanese rice yields; while agricultural advances may have happened at different times in cultivated parts of North America, Western Europe, and East Asia, there is nothing unique about this comparative sequence that is based on US data.

Many people nowadays admiringly quote the performance gains of modern computing (“so much data”) or telecommunication (“so much cheaper”)—but what about harvests? In two centuries, the human labor to produce a kilogram of American wheat was reduced from 10 minutes to less than two seconds. This is how our modern world really works. And as mentioned, I could have done similarly stunning reconstructions of falling labor inputs, rising yields, and soaring productivity for Chinese or Indian rice. The time frames would be different but the relative gains would be similar.

(the share of the US population who were farmers in 1800 was 83 percent) or their daily bowl of rice (in Japan, close to 90 percent of people lived in villages in 1800). The road to the modern world began with inexpensive steel plows and inorganic fertilizers, and a closer look is needed to explain these indispensable inputs that have made us take a well-fed civilization for granted.

Lightning will do it: it produces nitrogen oxides, which dissolve in rain and form nitrates, and then forests, fields, and grasslands get fertilizer from above—but obviously this natural input is too small to produce crop harvests to feed the world’s nearly 8 billion people. What lightning can do with tremendous temperatures and pressures, an enzyme (nitrogenase) can do in normal conditions: it is produced by bacteria associated with the roots of leguminous plants (pulses, as well as some trees) or that live freely in soil or in plants. Bacteria attached to the roots of leguminous plants are responsible for most natural nitrogen fixation—that is, for the cleavage of non-reactive N2 and for the incorporation of nitrogen into ammonia (NH3), a highly reactive compound that is readily converted into soluble nitrates and can supply plants with their nitrogen needs in return for organic acids synthesized by the plants.

As a result, leguminous food crops, including soybeans, beans, peas, lentils, and peanuts, are able to provide (fix) their own nitrogen supply, as can such leguminous cover crops as alfalfa, clovers, and vetches. But no staple grains, no oil crops (except for soybeans and peanuts), and no tubers can do that. The only way for them to benefit from the nitrogen-fixing abilities of legumes is to rotate them with alfalfa, clovers, or vetches, grow these nitrogen fixers for a few months, and then plow them under so the soils are replenished with reactive nitrogen to be picked up by the succeeding wheat, rice, or potatoes.[18]

But at what cost of human toil! This great nitrogen barrier to higher crop yields was nudged only during the 19th century with the mining and export of Chilean nitrates, the first inorganic nitrogenous fertilizer. The barrier was then broken decisively with the invention of ammonia synthesis by Fritz Haber in 1909 and with its rapid commercialization (ammonia was first shipped in 1913), but subsequent production grew slowly and the widespread application of nitrogenous fertilizers had to wait until after the Second World War.[20]

New high-yielding varieties of wheat and rice introduced during the 1960s could not express their full yield potential without synthetic nitrogenous fertilizers. And the great productivity shift known as the Green Revolution could not have taken place without this combination of better crops and higher nitrogen applications.[21]

Bread has been the staple of European civilization for millennia. Given the religious proscriptions on the consumption of pork and beef, chicken is the only universally favored meat. And no other vegetable (although botanically a fruit) surpasses the annual production of tomatoes, now grown not only as a field crop but increasingly in plastic or glass greenhouses. Each of these foodstuffs has a different nutritional role (bread is eaten for its carbohydrates, chicken for its perfect protein, tomatoes for their vitamin C content) but none of them could be produced so abundantly, so reliably, and so affordably without considerable fossil fuel subsidies. Eventually, our food production will change, but for now, and for the foreseeable future, we cannot feed the world without relying on fossil fuels.

Basic sourdough bread is the simplest kind of a leavened bread, the staple of European civilization: it contains just bread flour, water, and salt, and the leavening is made, of course, from flour and water. A kilogram of this bread will be about 580 grams of flour, 410 grams of water, and 10 grams of salt.[25]

Growing the grain, milling it, and baking a 1-kilogram sourdough loaf thus requires an energy input equivalent of at least 250 milliliters of diesel fuel, a volume slightly larger than the American measuring cup. For a standard baguette (250 grams), the embedded energy equivalent is about 2 tablespoons of diesel fuel; for a large German Bauernbrot (2 kilograms), it would be about 2 cups of diesel fuel (less for a wholewheat loaf).

Even in France, neighborhood boulangeries have been disappearing and baguettes are distributed from large bakeries: energy savings from industrial-scale efficiency are negated by increased transportation costs, and the total cost (from growing and milling grain to baking in a large bakery and distributing bread to distant consumers) may have an equivalent energy consumption as high as 600 mL/kg!

But if the bread’s typical (roughly 5:1) ratio of edible mass to the mass of embedded energy (1 kilogram of bread compared to about 210 grams of diesel fuel) seems uncomfortably high, recall that I have already noted that grains—even grains after processing and conversion into our favorite foods—are at the bottom of our food energy subsidy ladder.

The US Department of Agriculture publishes statistics on the annual feeding efficiency of domestic animals, and over the past five decades these ratios (units of feed expressed in terms of corn grain per unit of live weight) show no downward trends for either beef or pork, but impressive gains for chicken.[29]

Producing one American chicken (whose average edible weight is now almost exactly 1 kilogram) needs 3 kilograms of grain corn.[31]

Corn’s efficient, rain-fed cultivation has high yields and relatively low energy costs—equivalent to about 50 milliliters of diesel fuel per kilogram of grain—but the energy cost of irrigated corn may be twice as high as that of rain-fed feed, and typical corn yields and feeding efficiencies around the world are lower than in the US. As a result, feed costs alone can be as low as 150 milliliters of diesel fuel per kilogram of edible meat, and as high as 750 mL/kg.

Growing broilers to slaughter weight also requires energy for heating, air conditioning, and maintaining the poultry houses, for supplying water and sawdust, and for removing and composting waste. These requirements vary widely with location (above all, due to summer air conditioning and winter heating), and hence when combined with the energy cost of delivered feed a wide range of volumes is produced—from 50 to 300 milliliters per kilogram of edible meat.[33]

Adding the energy needed for slaughtering and processing the birds (chicken meat is now overwhelmingly marketed as parts, not as whole broilers), retailing, storing and home refrigeration, and eventual cooking raises the total energy requirement for putting a kilogram of roasted chicken on dinner plates to at least 300–350 milliliters of crude oil: a volume equal to almost half a bottle of wine (and for the least efficient producers, to more than a liter).

Tomatoes can be grown anywhere with at least 90 days of warm weather, including the deck of a seaside cottage near Stockholm or in a garden on the Canadian Prairies (in both cases, from plants started indoors). Commercial cultivation is a different matter, however. As with all but a small share of the fruits and vegetables that are consumed in modern societies, tomato cultivation is a highly specialized affair and most of the varieties available in North American and European supermarkets come from only a few places. In the US it is California; in Europe it is Italy and Spain. 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—not only in Canada and the Netherlands but also in Mexico, China, Spain, and Italy.

Plastics are a less expensive alternative to constructing multi-tunnel glass greenhouses, and the cultivation of tomatoes also requires plastic clips, wedges, and gutter arrangements. Where the plants are grown in the open, plastic sheets are used to cover the soil in order to reduce water evaporation and prevent weeds. The synthesis of plastic compounds relies on hydrocarbons (crude oil and natural gas), both for raw materials (feedstocks) and for the energy needed to produce them. Feedstocks include ethane and other natural gas liquids, and naphtha produced during the refining of crude oil. Natural gas is also used to fuel plastic production, and it is (as already noted) the most important feedstock—the source of hydrogen—for the synthesis of ammonia. Other hydrocarbons serve as feedstocks to produce protective compounds (insecticides and fungicides), because even plants inside glass or plastic greenhouses are not immune to pests and infections.

Direct energy inputs are easy to quantify on the basis of electricity bills and gasoline or diesel fuel purchases, but calculating the indirect flows into the production of materials requires some specialized accounting, and usually some assumptions. Detailed studies have quantified these inputs and multiplied them by their typical energy costs: for example, the synthesis, formulation, and packaging of 1 kilogram of nitrogenous fertilizer requires an equivalent of nearly 1.5 liters of diesel fuel. Not surprisingly, these studies show a wide range of totals, but one study—perhaps the most meticulous study of tomato cultivation in the heated and unheated multi-tunnel greenhouses of Almería in Spain—concluded that the cumulative energy demand of net production is more than 500 milliliters of diesel fuel (more than two cups) per kilogram for the former (heated) and only 150 mL/kg for the latter harvest.[37]

Plastic greenhouses located in the southernmost part of Almería province are the world’s largest covered area of commercial cultivation of produce: about 40,000 hectares (think of a 20 km × 20 km square) and easily identifiable on satellite images—look for yourself on Google Earth. You can even take a ride on Google Street View, which offers an otherworldly experience of these low-elevation, plastic-covered structures. Under this sea of plastic, the Spanish growers and their local and immigrant African laborers produce annually (in temperatures often surpassing 40°C) nearly 3 million tons of early and out-of-season vegetables (tomatoes, peppers, green beans, zucchini, eggplant, melons) and some fruit, and export about 80 percent of it to EU countries.[39]

That works out to nearly 90 milliliters per kilogram of tomatoes, and transport, storage, and packing at the regional distribution centers as well as deliveries to stores raises that to nearly 130 mL/kg. This means that when bought in a Scandinavian supermarket, tomatoes from Almería’s heated plastic greenhouses have a stunningly high embedded production and transportation energy cost. Its total is equivalent to about 650 mL/kg, or more than five tablespoons (each containing 14.8 milliliters) of diesel fuel per medium-sized (125 gram) tomato!

High agricultural productivities of modern societies have made hunting on land (the seasonal shooting of some wild mammals and birds) a marginal source of nutrition in all affluent societies. Wild meat, mostly illegally hunted, is still more common throughout sub-Saharan Africa, but with rapidly growing populations even there it has ceased to be a major source of animal protein. By contrast, marine hunting has never been practiced more widely and more intensively than it is today, as huge fleets of ships—ranging from large modern floating factories to decrepit small boats—scour the world’s oceans for wild fish and crustaceans.[41]

If you want to eat wild fish with the lowest-possible fossil carbon footprint, stick to sardines. The mean for all seafood is stunningly high—700 mL/kg (nearly a full wine bottle of diesel fuel)—and the maxima for some wild shrimp and lobsters are, incredibly, more than 10 L/kg (and that includes a great deal of inedible shells!).[43]

This means that just two skewers of medium-sized wild shrimp (total weight of 100 grams) may require 0.5–1 liters of diesel fuel to catch—the equivalent of 2–4 cups of fuel.

Broilers are herbivores, and when in confinement their energy expenditure on activity is limited. Therefore feeding them suitable plant matter—now mostly a combination of corn- and soybean-based mixtures—will make them grow fast. Unfortunately, the marine species that people prefer to eat (salmon, sea bass, tuna) are carnivorous, and for their proper growth they need to be fed protein-rich fish meals and fish oil derived from catches of wild species such as anchovies, pilchards, capelin, herring, and mackerel. Expanding aquaculture—whose total global output, freshwater and marine, is now closing in on the worldwide wild catch (in 2018 it was 82 million tons compared to 96 million tons of wild-caught species)—has eased the pressure on some overfished wild stocks of preferred carnivorous fishes, but it has intensified the exploitation of smaller herbivorous species whose growing harvests are needed to feed expanding aquaculture.[44]

So, the evidence is inescapable: our food supply—be it staple grains, clucking birds, favorite vegetables, or seafood praised for its nutritious quality—has become increasingly dependent on fossil fuels. This fundamental reality is commonly ignored by those who do not try to understand how our world really works and who are now predicting rapid decarbonization. Those same people would be shocked to know that our present situation cannot be changed easily or rapidly: as we saw in the preceding chapter, the ubiquity and the scale of the dependence are too large for that.

This may be a surprisingly small share, but it must be remembered that the Sun will always do most of the work of growing food, and that external energy subsidies target those components of the food system where the greatest returns can be expected by reducing or removing natural constraints—be it by fertilizing, irrigating, providing protection against insects, fungi and competing plants, or by promptly harvesting mature crops. The low share may also be seen as yet another convincing example of small inputs having disproportionately large consequences, not an uncommon finding in the behavior of complex systems: think of vitamins and minerals, needed daily in just milligrams (vitamin B6 or copper) or micrograms (vitamin D, vitamin B12) to keep bodies weighing tens of kilograms in good shape.

There are many reasons why we should not continue many of today’s food-producing practices. Agriculture’s major contribution to the generation of greenhouse gases is now the most-often cited justification for following a different path. But modern crop cultivation, animal husbandry, and aquaculture have many other undesirable environmental impacts, ranging from the loss of biodiversity to the creation of dead zones in coastal waters (for more on this see chapter 6)—and there are no good reasons for maintaining our excessive food production with its attendant food waste. So, many changes are clearly desirable, but how fast can they actually happen, and how radically can we reform our current ways in reality?

Could we return to purely organic cropping, relying on recycled organic wastes and natural pest controls, and could we do without engine-powered irrigation and without field machinery by bringing back draft animals? We could, but purely organic farming would require most of us to abandon cities, resettle villages, dismantle central animal feeding operations, and bring all animals back to farms to use them for labor and as sources of manure.

The numbers to confirm all of the above are not difficult to marshal. The decline of human labor required to produce American wheat outlined earlier in this chapter is an excellent proxy for the overall impact that mechanization and agrochemicals have had on the size of the country’s agricultural labor force. Between 1800 and 2020, we reduced the labor needed to produce a kilogram of grain by more than 98 percent—and we reduced the share of the country’s population engaged in agriculture by the same large margin.[50]

This provides a useful guide to the profound economic transformations that would have to take place with any retreat of agricultural mechanization and reduction in the use of synthetic agrochemicals.

In contrast, urea, now the world’s dominant solid nitrogenous fertilizer, contains 46 percent nitrogen, ammonium nitrate has 33 percent, and commonly used liquid solutions contain 28–32 percent, at least an order of magnitude more nitrogen-dense than recyclable wastes.[52] This means that to supply the same amount of the nutrient to growing crops, a farmer would have to apply anywhere between 10 and 40 times as much manure by mass—and in reality even more of it would be needed, as significant shares of nitrogenous compounds are lost due to volatilization, or dissolved in water and carried below the root level, with the aggregate losses of nitrogen from organic matter being almost always higher than those from a synthetic liquid or solid. Moreover, there would be a more than commensurate claim on labor, as the handling, transporting, and spreading of manure is far more difficult than dealing with small, free-flowing granules that can be easily applied by mechanical spreaders or (as is done with urea in small Asian rice fields) simply by sowing it by hand.

Atmospheric deposition—mainly as rain and snow containing dissolved nitrates—and recycled crop residues (straws and plant stalks that are not removed from fields to feed animals or burned onsite) each contribute about 20 megatons of nitrogen per year. Animal manures applied to fields, mainly from cattle, pigs, and chickens, contain almost 30 megatons; a similar total is introduced by leguminous crops (green manure cover crops, as well as soybeans, beans, peas, and chickpeas); and irrigation water brings about 5 megatons—for a total of about 105 megatons of nitrogen per year. Synthetic fertilizers supply 110 megatons of nitrogen per year, or slightly more than half of the 210–220 megatons used in total. This means that at least half of recent global crop harvests have been produced thanks to the application of synthetic nitrogenous compounds, and without them it would be impossible to produce the prevailing diets for even half of today’s nearly 8 billion people.

Global crop cultivation supported solely by the laborious recycling of organic wastes and by more common rotations is conceivable for a global population of 3 billion people consuming largely plant-based diets, but not for nearly 8 billion people on mixed diets: recall that synthetic fertilizers now supply more than twice as much nitrogen as all recycled crop residues and manures (and given the higher losses from organic applications, the effective multiple is actually closer to three!).

But none of this means that major shifts in our dependence on fossil fuel subsidies in food production are impossible. Most obviously, we could reduce our crop and animal production—and the attendant energy subsidies—if we wasted less food.

Even so, the well-documented global food losses have been excessively high, mostly because of an indefensible difference between output and actual needs: daily average per capita requirements of adults in largely sedentary affluent populations are no more than 2,000–2,100 kilocalories, far below the actual supplies of 3,200–4,000 kilocalories.[63] According to the FAO, the world loses almost half of all root crops, fruits, and vegetables, about a third of all fish, 30 percent of cereals, and a fifth of all oilseeds, meat, and dairy products—or at least one-third of the overall food supply.[64]

Eliminating waste that takes place all along the long and complex production-processing-distribution-wholesaling-retailing-consumption chain (from fields and barns to plates) is extremely challenging.

In well-off societies, a better way to reduce agriculture’s dependence on fossil fuel subsidies is to make appeals for adopting healthy and satisfactory alternatives to today’s excessively rich and meaty diets—the easiest choices being moderate meat consumption, and favoring meat that can be grown with lower environmental impact.

Full expression of human growth potential on a population basis can take place only when diets in childhood and adolescence contain sufficient quantities of animal protein, first in milk and later in other dairy products, eggs, and meat: rising post-1950 body heights in Japan, South Korea, and China, as a result of increased intake of animal products, are unmistakable testimonies to this reality.[71]