Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA,  ENVIRONMENTAL SCIENCE ( (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 29 October 2020

Agricultural Energy Demand and Usefree

  • David Roland-HolstDavid Roland-HolstUniversity of California, Berkeley


This overview article examines the historical and technical relationship between agrifood supply chains and energy services. Because agriculture is the original environmental science, all technological change in food production has environmental implications, but these are especially serious in the context of conventional energy use. Agrifood sustainability is of paramount importance to us all, and this will require lower carbon pathways for agriculture.


  • Agriculture and Farming


Food is the primary source of our subsistence as it provides most of humanity’s biological energy. Thus, it is hardly surprising that most of the energy expended by humans during the course of their history was directed toward food gathering, production, processing, and consumption. Even early-21st-century low-income societies still dedicate 60% to 80% of their human, animal, and other energy resources to food.1 Although it was a potent catalyst for later transitions, energy resources played a relatively neutral role in humanity’s shift from nomadism to more sedentary agrarian life. Historians cite a variety of decisive factors influencing this process, beginning at least 10,000 years ago and lost in prehistory, but political economy was generally considered to be the most salient.2

For over 95% of humanity’s agrarian history, agriculture relied on renewable sources for energy services.3 Animal traction (including homo sapiens) was the primary factor or mode of production in field agronomy, processing, and transport. Gravity played an essential role in irrigation and other water services including transport. Wind was also harnessed in many societies for water conveyance, food processing, and also transport. Solar energy and biomass fuels played important roles in agricultural processing, especially for desiccating agrifood products to limit perishability and transport costs. To the extent that animals shared living space with their keepers, biomass heating also contributed to husbandry and the many goods and services that livestock provided to their keepers.4

While this description of the rural energy economy might sound picturesque, early agriculture was characterized by low productivity and limited means for managing risks associated with weather, pestilence, and spoilage. The result was a high degree of food insecurity, manifesting itself in chronic cycles of hunger and malnutrition. Among the oldest historical records in China and Egypt are descriptions of governments distributing food stores in times of famine. Although these circumstances were often aggravated by conflict, low productivity in agriculture was a primary threat to human security and sustained development, exerting an onerous constraint on public health, population growth, and economic growth and diversification. For most of the ten millennia comprising the Agrarian period, the majority of humanity was essentially trapped in subsistence production, with small agricultural surpluses handed over to markets or despots to support urban minorities.

This overview article examines the historical and technical relationship between agrifood supply chains and energy services. Because agriculture is the original environmental science, all technological change in food production has environmental implications; however, these are especially serious in the context of conventional energy use. Agrifood sustainability is of paramount importance to us all, so we make an effort in this discussion to highlight lower carbon pathways for agriculture.

Industrial Transformation and Demographic Transition

Agricultural technology and practices progressed slowly but steadily during the early agrarian period, until the emergence of new energy technologies in 18th-century Europe began a profound and ultimately global transformation of food production and distribution systems. The aptly named “industrial revolution,” an epic technology transition impelled by fossil fuel resources, has conferred living standards upon us that were unimaginable to our ancestors, touching on every dimension of life including our food system. By domesticating fossilized solar energy, in the form of petrified biomass (oil, coal, natural gas, etc.), humanity has been able to amplify its physical power and leverage its ingenuity as never before, achieving unprecedented productivity growth in resource development, production, and all manner of services. Farming, food production, processing, and distribution were all revolutionized in the confluence of this global innovation process, leading us more or less continuously to the 21st century’s modern industrial and (in some societies) post-industrial global farm and food economy.

At the same time that fossil fuel energy services were revolutionizing technology, their application to agriculture and industry set in motion the drivers of a pervasive and lasting demographic transition. For the overwhelming majority of human history, most of society produced its own food or lived in relatively immediate proximity to those who did. The appearance of advanced civilizations meant that many people were freed from farming by status, wealth, or specialized occupations. Even when these people lived in urban areas, however, farms and farmers were never far away. A select few could indulge in exotic treats such as the Roman shaved ice relayed from Alpine glaciers. For most people in even the largest ancient cities, food came from local farmer’s markets, delivered by walking people or animals. And the majority populations of most countries comprised subsistence farming households.

All this began to change with the parallel advance of three new energy-intensive technology systems: agricultural mechanization, urban industrialization, and steam- powered transport linking them together. Each played an essential part in the rise of the modern 21st-century food economy and in the process changed societies forever. This process is the subject of an extensive economic history literature, generally referred to as “structural transformation research.” Here we can only summarize salient features relevant to agriculture and energy, but this large body of work rewards closer inspection.5

Agricultural Mechanization

The technical details of agricultural mechanization are important to fully understand how it has transformed our food system, but its overarching contribution to agriculture was increasing the productivity of other factors, including land—and especially labor. Mechanical innovations at all stages of farm production, especially those that were energy fueled, led to dramatic increases in yield per acre and per worker (Da Rocha & Restuccia, 2006). The former sharply improved investment incentives for farming, supporting a virtuous cycle of rural enterprise development that made peasant society a distant memory in most Organization for Economic Cooperation and Development (OECD) member country agricultural communities. Parallel increases in labor productivity had even more dramatic social consequences, liberating workers and family members for migration to urban labor markets, where employment was growing rapidly with new, labor intensive industrial growth. Mechanizing agriculture allowed these countries to fuel their urban development with a steadily increasing supply of both workers and food, without the trade-offs that would have occurred in a static technology setting.

Urban Industrialization

The urban narrative of the industrial revolution has been told many times, but its significance for agriculture deserves emphasis in the present discussion. Simply put, the rise of urban working populations, with concurrent productivity and wage growth, created a new food economy in every industrializing country. Not only were more people living in cities where they relied on others to produce their food, they were generally enjoying higher incomes that could support a larger and more diversified agrifood economy. Over time, and sometimes only intermittently because of property, labor, and other rights issues, this trend improved living standards at both ends of agrifood supply chains, sharing the unprecedented productivity gains of industrial innovation between both food producers and consumers.

Steam-Powered Transportation

For farmers everywhere, the primary barrier to market access is transport margins. Farmers are in a kind of ice business—they leave for market with a relatively massive, perishable product whose net value begins to decline from the moment they pass the farm gate. For this reason, transport efficiency ultimately delineates the profitable horizon for investing in marketable agricultural products. With steam- and eventually liquid fuel– powered transportation, the industrial revolution redefined the geography of agrifood supply. By dramatically reducing travel time and cost, motorized rail and road transport opened vast areas, formerly confined to subsistence production or completely undeveloped, to marketable agriculture. This triggered significant new settlement activity in some regions (e.g., United States and Latin America), but generally higher labor productivity limited the growth of new rural populations and continued to reduce it in established rural areas.

Just as profound as this new agricultural expansion were companion trends of specialization in production and differentiation in consumption. More efficient transport transformed traditional, localized agrifood supply chains, promoting specialization by individual producers, offering much more product variety to final consumers and weaving a complex web of intermediate links through food processors and aggregators. Today’s international brands and supermarket systems have become a service economy summation of this mobility revolution, moving farm products from everywhere to everywhere else, adding value and energy services along the way.

Thus, the industrial revolution begat an agrifood revolution, with dramatic consequences for food, technology, energy, and above all for human populations. The resulting structural transition to modern society is dramatically revealed in Figure 1, a rural–urban demographic shift that has inexorably followed from agricultural mechanization, urban industrial growth, and a transport revolution. Although populations in all these economies have grown over the long term, rural shares in most have dropped from significant majorities to single-digit percentages.6

Figure 1. Employment in agriculture since 1800.

Source: Herrendorf et al. (2014).

Energy Services to Modern Agriculture

Today’s energy system provides direct services to agriculture that are in most cases familiar from Agrarian-era analogs, including support for tillage, water conveyance, transport, and others. Because of the energy density of fossil fuels, however, these services are much more potent, and as a consequence of sustained technology innovation, they are much more productive. The term horsepower (hp), for example, was coined in the late 18th century by Scottish engineer James Watt to compare the output of steam engines with the power of draft horses. Watt determined that an average farm horse could pull with a force of 180 pounds. By 1937 John Deere was producing tractors averaging 10 hp, while today their product line ranges from 100 hp to 600 hp.7 In addition to conventional farming applications, energy supports many more aspects of 20th- and 21st- century agriculture and the food system, with profound implications for its scope, complexity, reliability, and sustainability. In addition to so-called direct services from fuel and electricity, the majority of energy requirements in modern agriculture are now indirect, used in the production of yield-enhancing agrochemical inputs. For a modern agricultural economy such as the United States, we see the dominance of indirect energy services clearly for so-called commodity crops (Figure 2), staples of global food production. Downstream, Figure 3 shows how this energy dependence is passed through to livestock production via feed for meat, egg, and dairy production. Some differences are also apparent across species, due to higher direct energy services for livestock that are more likely to be enclosed (swine, poultry, and dairy cattle).

Figure 2. Total expenditures and cost for direct and indirect energy services to U.S. agriculture (2011).

Source: USDA (2012).

Figure 3. Average energy expenses per farm for livestock producers (2014).

Source: USDA (2012).

A more complete technical discussion could elucidate the role of mechanization or in other agrifood processes but suffice to note that energy use technologies have now infiltrated every task and stage of production. To facilitate understanding of energy’s complex role in the agrifood system, we review energy services to agriculture in four generic categories.


Perhaps the most recognizable impact of the industrial revolution has been mechanization, a transformation of material technology made possible by harnessing fossilized solar energy, in the form of solid, liquid, and gaseous carbon fuels, to power machinery and the recursive succession of industrial production—making mechanical products with machinery. We formerly lived in a world of animal, wind, water, and solar power, but in the 21st century our world is dominated by fuel- and electric-powered devices that serve our needs in drastically different ways. In agriculture, machines help us complete every step of crop and livestock production. We discuss only the most prominent aspects of mechanization in agricultural production, but in all its dimensions the agency of carbon fuel energy and machinery has transformed the sector.

Table 1. Energy Inputs for Tilling 1ha of Soil by Technology Type

Tilling Unit

Required Hours

Machinery Input (kcal)

Petroleum Input (kcal)

Human Power Input (kcal)

Oxen Power Input (kcal)

Total Input (kcal)

Human power






Oxen (pair)







6-HP tractor






50-HP tractor






a Each ox is assumed to consume 20,000 kcal of feed per day.

b An estimated 23.5 L of gasoline used.

c An estimated 30.3 L of gasoline used.

Source: Pimentel and Pimentel (2008).


Tillage is generally the most arduous task of traditional agriculture, making it the primary target of early mechanization and labor substitution. As Table 1 clearly shows, even relatively small tractors are up to 100 times more productive than human laborers. Machines wholly displaced animal traction in modern economies, and humans navigate their fields from air-conditioned cabins during the main stages of fieldwork for extensive crops. In the early 21st century some are experimenting with autonomous vehicles, and drones ply the air monitoring conditions and even applying agrochemicals. These innovations have not only reduced the physical burden for farmers but have also vastly increased labor productivity and the economic scale of individual farm production. Average farm size in Vietnam, for example, is about half a hectare, while in the United States the figure is closer to 100 hectares. This fact, largely a result of mechanization, has profound implications for livelihoods, demographics, and food security.

Farm size is one of the primary livelihood constraints across much of the developing world. Unless you have a goose that lays golden eggs, you are unlikely to attain middle-class status with the proceeds from operating a one-acre farm. As noted above, productivity growth from mechanization has also freed the overwhelming majority of OECD populations for higher-wage urban employment, supporting essential skill-intensive growth and economic diversification (Figure 4 to 6). Ironically perhaps, mechanization has in many instances reduced agricultural diversification, promoting large-scale crop and animal specialization. In the aggregate, however, this has increased output and incomes, permitted farm families to contribute more to society’s food security, while purchasing most of their own food with the proceeds.8

Figure 4. Percentage of power for agricultural traction by technology.

Source: Pimentel and Pimentel (2008).

Figure 5. Agricultural employment in OECD economies, 1980–2000.

Source: Herrendorf et al. (2014).

Figure 6. Share of total labor force effort in agriculture.

Source: Herrendorf et al. (2014).

Mechanization is progressing across the developing world, too, but faces significant constraints, including financial and technical capacity, marketing and other supporting services, and “goodness of fit” with respect to local conditions and institutions. Because of differences in land tenure patterns Western technologies have been of limited use outside the OECD and some parts of Latin America, but Japan, Korea, and emerging Asian industrial economies have been leaders in developing smaller-scale technologies to fill this gap. The result has been rapid acceleration of smallholder mechanization and per capita energy use in Asia and (albeit in less dramatic form) in Africa as well.

In terms of low carbon energy transition, renewable electrification in agriculture is spreading at very different rates across the OECD, mainly due to varying public commitments to recognition of the social cost of carbon.9 Vehicle electrification has thus far been limited, without much uptake by farm equipment manufacturers. Having said this, most farming is essentially solar powered, so there is considerable potential in this sector. One promising area is enclosed production, which has been an early adopter of both PV solar and biogas recycling for on-site electricity generation. Finally, for community- and household-level electrification in low-income countries, PV solar is more cost effective than conventional large-scale generation and grid investment. In these important countries (30% of the world’s households still lack electricity) solar energy offers the promise of leapfrogging over legacy commitments to carbon-intensive electric power infrastructure.


An existential resource for terrestrial plants and animals, renewable fresh water is relatively scarce globally and unequally distributed over space and time. To sustain global production of crops and livestock requires about 80% of the water used by human populations (3,600 km3 of 4,500 km3 per year, respectively). Making this water available requires not only rainfall but determined investments in water storage and conveyance around the world, further compounding existing patterns of resource inequality between higher- and lower-income populations.

Because it is relatively massive, moving water requires a significant amount of energy and fortunately most of that is provided by nature through atmospheric convection and gravity. In one spectacular example, the so-called Pacific Atmospheric River evaporates and transports the volumetric equivalent of five Mississippi Rivers across the Pacific Ocean to northern California every winter.10 Once the water arrives, however, it has to be re-distributed to the rest of California’s valuable agricultural and other activities/communities. This process of moving water presently consumes about 20% of the energy used in the world’s fifth largest economy (CEC, 2005). While California is not typical—over 90% of agricultural water globally is still delivered by weather and gravity—it foretells future patterns of water use that will be dictated by economic and environmental necessity. To allocate water more equitably and secure world food production, much more determined investments will be needed in storage and conveyance, probably requiring much more intensive and extensive application of electric and other motive energy to the water supply.

It should be emphasized, however, that irrigation investments have historically been quite profitable thanks to significant risk reduction and yield benefits. About 16% of global agricultural land area in the 21st century is irrigated, yielding roughly 40% of total agricultural production. Thus, the average productivity of irrigated land is 3.6 times that of unirrigated land, while irrigated land has also produced crops with much higher average market values (about seven times higher return per harvest unit).

New attention is turning to water infrastructure because of energy and environmental considerations. Water storage has historically delivered longer and more predictable growing seasons but also flood control and hydro-electric energy services. Dams for these purposes encountered controversy in the past for other reasons, mainly local environmental impacts and community/habitat displacement. With the global recession of ice and snow (the second largest freshwater reserve after groundwater) caused by climate change, however, dams are being reconsidered for their potential to stabilize seasonal water allocation and support renewable power development. In addition to direct hydropower potential, impounded water resources can also store daytime solar energy, acting as “natural batteries” that can transfer renewable energy services from peak supply to peak demand periods.


Enclosing crop and livestock production has always been practiced in smallholder agriculture, not only to delineate and protect property rights but to restrain one’s own livestock and deter intruding animals. The rights aspect of enclosure enhances incentives to invest in higher productivity, since these benefits can more effectively be captured. And such investments—including irrigation, intensive feeding, and environmental control—have increased energy use in agricultural operations going back to prehistoric times. The industrial revolution accelerated this process, with metal frame glazing permitting much wider use of greenhouses and steam heating improving climate control for plants and animals. Growing seasons were extended for higher-value plants and animals, and agriculture was brought into peri-urban and even urban areas.11 Finally, declining bulk transport costs made feedlot production more economical in relation to free-range production.

The trends established with early industrialization have expanded substantially in the last century, both extensively and intensively. For plants, greenhouse innovations in Holland and elsewhere have achieved extraordinary benefits in terms of productivity, quality control, and value creation. Including hydroponic technologies, companies such as Eurofresh Farms have developed completely enclosed turnkey operations that can be “dropped in” to nearly any climatic location, requiring only a steady supply of electricity and water to produce all year around.12 Value creation has advanced so far that affordability of energy for these systems is much less important than environmental sustainability, leading to an industrywide shift toward renewable electric power.

Thus, enclosure has significantly increased output and value-added energy intensity of agrifood production. In the process, it has also created an important opportunity for environmental mitigation, especially with respect to greenhouse gases. Ironically, the term gave scientific meaning to the underlying cause of global warming holds the key recycling one of our most dangerous GHG’s. Methane (CH6) has 27 times the radiative forcing potential of CO2, and its primary anthropogenic source is agriculture. A large portion of what makes up methane comes from open field biomass decomposition, and this can be better managed with less water-intensive agronomic practices. The other two substantial shares come from livestock and secondary agricultural biomass, both of which could be managed more effectively, recycling methane from enclosed animal/plant production and biomass aggregation. The alternative uses of these “waste” gases include onsite electric power generation (displacing other emissions from utilities) and bioplastic production, and a host of other innovations that can stabilize methane in durable or reusable products (Roland-Holst et al., 2013).


Energy services play two main roles at harvest time. First of these can be called micro-climate control, any measures to influence temperature and humidity during the latter stages of crop ripening. Of course, climate control is fundamental to all stages of enclosed (plant or animal) production, the more energy-intensive but higher-value approach. Even in open field production, however, when colder temperatures threaten to delay ripening or even to destroy crops, high BTU heating systems can be deployed. When early rain and/or humidity threatens to promote fungal or bacterial infection, heating systems may also be activated. Conversely, excessive heat can exhaust late season crops, requiring aerosol spraying and/or high-velocity ventilation to cool plants/animals and promote transpiration.

The second and even more widely used category of energy-intensive activity at this production stage is mechanized harvesting. With advances in precision and dexterity, nearly all crops have some degree of mechanization in their harvest process, from the taking of plants/fruits/vegetables directly in the field to conveying them off-field for initial processing, “combining” basic harvest stages from the ripe standing crop to initial sorting, stripping, and cleaning. Using a wheat analogy, the basic steps of cutting, topping, threshing, winnowing, and storing are all mediated by machines on a modern farm enterprise, and each of these machines uses energy. Certain specialty fruits (some grapes, berries, stone fruits, etc.) still require a human’s discerning eye and dexterous hand, but more of these crop categories are being automated with each passing year.


After genetic improvements and mechanization, the most important contributor to modern agricultural productivity has been application of agrochemicals—mainly fertilizers and pesticides. These products require energy to produce, distribute, and apply, but fossil fuels and their derivatives are also important inputs to their manufacture. In this way, modern agriculture relies not only on energy services but energy carriers (fuels). Finding lower carbon alternatives presents very different challenges in these two contexts.


Soil amendment to maintain or enhance plant yields has long been integral to farming, but for most of agricultural history this was confined to recycling biomass from harvests and other organic waste streams. One hundred years ago, however, a German chemist made one of the most important technical discoveries of the 20th century. Compared by some to the light bulb, aviation, and the automobile, synthetic ammonia became the primary source of manufactured fertilizer and is estimated to have increased global food production by up to 40% (Figures 7 and 8). This saved the lives of hundreds of millions of people in the process.13

Ammonia is the second largest chemical product produced in the world, and the primary feedstock for its production (99% of ammonia tonnage) is natural gas. Other hydrocarbon fuels can be used, but natural gas is preferred. About 80% of ammonia production is converted to urea, which in turn is 90% converted into nitrogen fertilizer. By the first decade of the 21st century, between 3% and 5% of the world’s annual natural gas production—roughly 1% to 2% of the world’s annual energy supply—was being converted using the Haber-Bosch process to produce more than 500 million tons of nitrogen fertilizer globally.

Figure 7. The Haber-Bosch process.

Source: Erisman et al. (2008).

Figure 8. Trends in human population and nitrogen use throughout the 20th century.

Source: Erisman et al. (2008).

Because of its hydrocarbon fuel intensity, synthetic fertilizer is a potent source of greenhouse gases, responsible for over 3% of global annual emissions. Each ton of nitrogen fertilizer corresponds to more than 10 tons of CO2e. Emission intensity of fertilizer production varies with technology and feedstock, with EU averages of 9.7MTCO2e per ton of fertilizer while China’s average over the period 2000–2015 was over 13MTCO2e (Zhang et al., 2016). Thus, the innovation that now sustains about half of humanity’s food supply may itself be unsustainable. Fortunately, this paradoxical threat is driving today’s best agrochemical scientists to look for lower carbon fertilizer (Brown et al., 2016; Zhang, 2016).


Petroleum products are also an important constituent of modern pesticides. A pesticide consists of an active ingredient coupled with inert ingredients that facilitate storage, delivery, and other ancillary product characteristics. The active ingredient kills the pests, while the inert ingredients facilitate spraying and coating the target plant. Active ingredients were once distilled from natural substances; now they are largely synthesized in laboratories, and almost all are based on hydrocarbons derived from petroleum. Most pesticides contain other chemicals: Chlorine, oxygen, sulfur, phosphorus, nitrogen, and bromine are among the most common. Inert ingredients can be many substances, but there was a historical preference for kerosene and other petroleum distillates because of their volatility and relatively low cost. Research over the period 2005–2015 has targeted these for substitution by water-based chemicals, but market shares for these new varieties remain small.

Figure 9. Energy consumption by stage of agrifood production (United States, 1992 and 2002 compared).


Production of raw agricultural products is only the first of many steps in our food supply, especially in modern times, as agrifood supply chains span the globe through a myriad of intricate channels, using energy at every stage. The topic of energy in agrifood supply chains would easily justify book-length treatment, but a few salient insights are worth highlighting from the simplified breakdown in Figure 10 (see also Canning, 2011; Canning et al., 2010). First, energy use during the decade considered (1992–2002) increased at every stage except wholesale and retail marketing, where efficiency improvements apparently overcame larger sales volumes. Secondly, household food purchasing (including transport), storage, and preparation represent the largest energy use category. Third, food processing and marketing each use more energy than farming itself. Finally, food transportation is the lowest energy cost category in the food supply chain. This is especially remarkable in a country with the size and climatic diversity of the United States.

Despite the complexity of 21st-century agrifood supply chains, the second main step from “farm to fork,” food processing, has always addressed three fundamental objectives: preservation, value-added, and product differentiation. Energy services have played essential and ever-growing roles in fulfilling all three goals.


After the many uncertainties of agricultural production, the greatest risk to our food security is perishability, threatening both producer livelihoods and consumer sustenance. Because raw farm products are organic material, they are highly susceptible to spoilage and attack by pests of all sizes, from bacteria to bears. Moreover, the seasonality of farm production introduces the need for stockpiling over extended cycles of variable temperature and humidity. For most of our history, we relied on desiccation and simple treatments such as smoke and salt to preserve meat, vegetables, and fruits. The main energy sources in this context were solar radiation and biomass combustion. As cereal production developed, grains were milled to eliminate moisture that could support bacteria and attract pests. Rice could be fully dried as hulled kernels, but wheat and its nearer relatives were more often milled and stored as compressed noodles/sheets, ground directly, or reconstituted with water and baked into loaves.14 All these cereal processing techniques have been modernized continuously to arrive at the vast and energy-intensive baking and packaging industries of the 21st century.

More sophisticated approaches to food preservation evolved in parallel, including worldwide domestication of legions of microorganisms that help us ferment, pickle, and otherwise preserve fresh food products of every imaginable variety. One leading advantage of these “natural” preservation techniques is relatively low energy intensity that is achieved through the agency of living organisms. Fermentation may require heating, especially for initial sterilization, but most traditional fermented food processes (kimchi, miso, cheese, tofu, etc.) have evolved to achieve stabilization without much additional energy. This approach works because it is powered by living organisms consuming a fraction of the initial feedstock. A new generation of food producers and processors15, as well as scientists in many fields of biology16, are awakening to this experience as an important model for sustainable innovation in future food production and many other types of bioresource processing.

Although more recognized for its social and medicinal roles, alcohol was also a valuable preservative. With the help of countless varieties of endemic and domesticated yeasts, beer and wine have been produced since prehistoric times to stabilize agrifood nutrients from cereals (beer is essentially aqueous bread) and fruit, respectively. In doing so, these beverages offer important carbohydrates, calories, vitamins, minerals, with a little relief from physical and psychic discomfort in the bargain.17 Higher proof alcohol was also extensively used to preserve decoctions of herbal and other traditional medicines. Displaced by modern pharmaceuticals, this role has diminished in modern times, but we still see traces of it in a delightful array of European aperitifs that remain popular commercial beverages.

Other approaches to preservation, especially cooling and heating, have long historical precedence but only became prominent with the advent of more energy-intensive refrigeration, pasteurization, and other technologies. In the 21st-century global agrifood economy, refrigeration is one of the largest energy-using technologies, familiar to all in retailing but also instrumental at most stages of transit through modern food supply chains.

Value Added

Economically, the primary incentive to process raw agricultural products is value creation, making them more attractive to consumers by investing in quality characteristics that command higher prices. Wine is one of the most spectacular examples of this: Depending on investments in real and perceived quality, the same quantity of ripe grapes can yield 750ml of fermented juice worth $5 or $500. More mundane examples of value-added processing now dominate 21st-century global food production, especially as households migrate from rural areas to urban formal sector employment and require more stable, ready-to-cook, or ready-to-eat food products.

As part of the larger demographic transition discussed in Section “Industrial Transformation and Demographic Transition” above, food must also travel farther between rural and urban markets within and between countries. These longer transits usually require some origin processing to reduce bulk and prolong shelf life. They may also require more destination processing to sort, grade, and cull damaged inventory and incorporate it into aggregate products (e.g., fruit preserves, canned sauces, etc.) where fresh features are less discernable or important. All this added value relies on energy services, and as Figure 10 reveals, we are far from the model of cooking raw food products obtained directly from the field. Energy used in intermediate processing is now roughly equal to and growing faster than total energy used in primary agricultural production.

Product Differentiation

Some agrifood products are necessities and others are luxuries. After basic needs are met, consumers also become more interested in product variety. As incomes rise, their tastes shift toward goods perceived to be of higher quality, usually including higher value-added and higher energy inputs (see, e.g., Dennis & Iscan, 2009; Foellmi & Zweimüller, 2008). Catering to these evolving tastes is a primary challenge for agrifood globally, and thanks to the dynamic of middle-class emergence in economies like China, global food product diversity is expanding more rapidly (Dekle & Vandenbroucke, 2012) than at any time in history. Firms generally pursue product variety to strengthen individual consumer identification with their products, thereby increasing willingness to pay. This kind of pricing power is effective even when products are functionally similar but identifiably different, such as toothpaste or breakfast cereal brands—countless variations on a small number of ingredients. For national firms, more varieties require more organizational complexity, more processing, and more energy use. For firms sourcing or marketing globally, product variety means mixing more originating varieties and sending them to more different destinations—requiring more energy use in transport that would be needed to produce and sell everything within each national market.


The combined perishability and relatively low value/mass ratio for many food products makes transport efficiency essential to profitable market access for agriculture. Conversely, significant innovations in transport and logistics have been historically associated with expansion and diversification of agrifood markets. This can be facilitated with simple infrastructure investments such as road and bridge building, as well as institutional reforms that facilitate trade (Eaton & Kortum, 2002), like more efficient customs clearance. All these measures reduce trade and transport margins and expand the profitable horizon for investment in food production, processing, and marketing.

The most exotic and valuable agricultural products, such as silk and opium, have supported profitable long-distance trade for centuries. With dramatic improvements in naval architecture and navigation technologies, a broader spectrum of products could be shipped over long distances. Higher unit values were still essential, however, as we saw when 17th-century Dutch traders focused their energies on precious spices and ingeniously transplanted addictive foodstuffs (sugar, tobacco, coffee, chocolate) to New World production for global marketing.

The most salient contributions to market access in modern times, however, have been energy-driven innovations in transport services, especially carbon-fueled transport mechanization. As outlined in Section “Industrial Transformation and Demographic Transition”, these innovations first linked higher-income domestic urban markets to the countryside and then extended agrifood trade across a modern global web of international logistics and multinational branded marketing. Since the late 18th century, rail transport steadily expanded the scope of agrifood shipment and with it the amount of arable land converted to farming. By the mid-19th century, rail shipment of agrifood products, especially extensive commodity crops such as meat was a primary driver of midwestern agricultural settlement in the United States, with much of Latin America, eastern Europe, and Russia following suit over the next 50 years. After World War II, long-distance trucking was complementing or competing with the same agrifood freight services across the United States, Europe, Australia, and many other emerging markets. This process of geographic specialization between rural and urban areas was facilitated by rising labor productivity in agriculture, already discussed above, and resulted in pervasive and lasting restructuring of modern societies. An analogous but more intermittent process is still underway in developing countries and more advanced in China and emerging east and Southeast Asia.18

In recent decades, containerized and refrigerated shipping have globalized trade in agrifood beyond anything envisioned by the European colonial powers and their trading companies. In the early 21st century we have Alaskan salmon and crabs flown to China for deboning and “shaking,” only to be flown back to U.S. and other markets as “fresh” filets and whole crabmeat.19 While these examples are extreme, food mobility has increased dramatically with the advent of the WTO, with global agrifood trade increasing over 400% since around the year 2000. This growth was strongly supported by carbon fuel energy, as shipping and air transport have dominated a process of international specialization in production and consumption of agrifood products. As indicated in the following figure, China (self-sufficient as recently as 1990) is the world’s largest food importer, with most of its imported food demand being satisfied by countries thousands of miles away.

Figure 10. Embodied cropland in non-animal agricultural trade among major partners (2015). Flows illustrated are those between the 10 largest producer and consumer nations of embodied land in global trade in non-livestock agriculture. This is 32% of the total embodied land traded. Figure reproduced with the permission of Chatham House, the Royal Institute of International Affairs.

These trends have led to critical reliance by hundreds of millions of consumers on international food mobility, supported by elaborate networks of conventional energy fueled transportation (and refrigeration) technologies. The sustainability of these trends is open to question, and there are several important but countervailing perspectives influencing the research and policy dialogue in this area.

From one generally higher-income perspective, significant public attention has been given in OECD economies to the concept of “food miles,” a proxy of transport services embodied in food products (Coley, Howard, & Winter, 2009; Saunders et al., 2006). Internationally air-shipped salmon and crab are extreme examples, in contrast to herbs in the kitchen window box or backyard eggs. In some countries, an emerging “locavore” community has begun advocating substitution of local production for more distant food sources. This movement also appeals to concerns about food self-sufficiency and a number of qualitative considerations, including greenspace development, legacy product variety, among other factors. Given that transport energy costs are in the low single-digit percentages of farm-to-fork food supply costs, however, it seems unlikely that transport energy cost will be independently decisive in the evolution of efficient food supply chains. Twenty-first-century food trade patterns remain significantly influenced by political and social policies, but factor costs and productivity appear to be the salient economic drivers.

Another perspective on food globalization emphasizes neoclassical efficiency and market-determined patterns of specialization. In these cases, lower market costs of local resources and transport confer comparative advantages on some countries as agrifood exporters, even though they may be a long way from their most important markets. Countering this efficiency view are arguments that such countries may have weak property rights or environmental standards governing factor use (e.g., labor and especially land). Others argue that current transportation technologies are exploiting a divergence between the market price and social cost of carbon fuels. The political economy arguments can only be evaluated on a case-by-case basis, but there is no denying that global food trade is not pricing in the future cost of climate change.

Indeed, low lifecycle transport costs in the food sector (Figure 11) could be masking significant environmental risks, and this is particularly true of the fastest growing (international) category of food trade. Over 95% of most long-distance international food product trade uses maritime transport. Granted, shipping has always been the most cost-effective technology for moving mass around the planet, but for most of human history boats relied only on renewable wind and human energy. Twenty-first-century cargo ships, however, are by far the most pollution-intensive transport vehicles. Shipping fuel (to cite only one criteria pollutant) contains sulfur concentrations more than 3,500 times greater than the diesel emissions that triggered the Volkswagen scandal. Overall, the shipping sector is responsible for 3% of the world’s total greenhouse gas emissions such as carbon dioxide.20 While these emissions make significant long-term contributions to global warming and extreme weather events, they already have a long history of damaging public health. Independent research (Sofiev et al, 2018) indicates that along developing country coastlines and ports in Asia where shipping lanes are concentrated, ship pollution contributes to millions of cases of childhood asthma and premature deaths every year. This highlights an essential distinction between economic and social costs of food trade. Fortunately, commercial shipping interests are negotiating voluntary adoption of lower carbon fuels and pollution mitigation technologies. The timetable for this is vague at the moment, however.

Sustainable Energy Balances in the Food System

A final perspective on agriculture and energy takes us back the introductory discussion about energy balances—food is the primary source of energy for humanity and its fellow animals, and energy is essential to produce food. Thus, the challenge of sustainability in the long term will depend on energy balances in this process—using energy to create the energy that sustains us. This brief survey article can only summarize what we know about this profoundly important issue, but certainly it deserves diligent study now and in the foreseeable future.

As Figure 11 indicates, even though some food products are more transport and energy intensive than others, agrifood still only relies on transportation for about 5% of its average lifecycle energy use. This means we are unlikely to see much adjustment in patterns of food trade and specialization driven by energy prices. Far larger shares for food lifecycle energy needs are in farming, processing, marketing, and final consumption (including household transport for food purchase and refrigeration/preparation at home). These activities can be expected to be more responsive to fuel prices in seeking efficiency and other induced innovations.

Figure 11. Share of energy in U.S. farm business cash expenses (2014). USDA, National Agricultural Statistics Service and USDA, Economic Research Service, 2014 Tenure, Ownership, and Transition of Agricultural Land Survey.

From a different perspective, comparing food product types reveals important disparities of energy intensity, depending on technology choice and factor requirements. This picture is clouded, however, by varied energy intensity of different production stages. For example, Figure 12 suggests that commodity crops have the highest energy intensity in farming, with primary energy use dominated by indirect energy services from agrochemicals and mechanization. On the other hand, after processing and marketing energy are considered, livestock products are substantially more energy intensive (Figure 12) and different consumer choices within food groups (meat, vegetables, beverages, etc.) can also induce very different energy use patterns in the food system. Finally, (Figure 12 again), energy requirements are loosely correlated with dietary energy yield, but this is an imperfect predictor.

Figure 12. Meal components, dietary energy, and lifecycle energy inputs for two different dinners.

Source: Carlsson-Kanyama, Ekstrom, and Shanahan (2003).

It should also be emphasized that different food groups have different dietary energy yields. Of course, taste remains a primary determinant of willingness to pay for food in higher-income countries, but from a sustainability perspective it can be noted (Figure 13) that some sources have input energy requirements that are more than 10 times greater than they yield in nutritional energy.

Figure 13. Input versus dietary energy by food group (MJ per standard portion).

Source: Carlsson-Kanyama et al. (2003).

It is clear from this simple evidence that human tastes, especially when unfettered by poverty, are not only omnivorous but actually strive for satisfaction with characteristics that reflect ever-higher levels of value-added (or factor) inputs. To the extent that these inputs come from labor, society is promoting a virtuous cycle of post-industrial, service-intensive employment. To the extent that additional value comes from more intensive use of natural resources, for the sake of future generations we probably need to be more attentive to opportunity cost and environmental consequences. For example, it will certainly be difficult to indulge in ever more land-intensive eating habits for a population that is expected to double by 2050 (e.g., Ahmed, 2008).

The situation with energy, however, is fundamentally different. Sustainability in this context is much more about the source of energy than its total usage because conventional and renewable energy are subject to completely different demand and supply paradigms. Processing and using fossil fuels liberates prehistoric carbon that has been sequestered for millions of years, endangering the delicate atmospheric chemistry that protects all living organisms. PV solar and windmill technologies, in contrast, harvest primary solar energy from direct radiation and atmospheric convection, with negligible impact on atmospheric equilibrium. On the supply side, fossil fuels may still be abundant, but they are essentially (i.e., within a rational human planning horizon) finite and subject to ever-increasing economic scarcity (rising exploration costs, risks, and prices). Wind and solar power, by contrast, are essentially boundless resources, available to us subject only to technology constraints. If we can nearly or fully decarbonize the supply of renewables, it really doesn’t matter how much energy humans want.

Further Reading

  • Braudel, F. (1992). Civilization and capitalism, 15th-18th century: The perspective of the world. (Vol. 3). Berkeley: University of California Press.
  • Levenstein, H. A. (2003). Revolution at the table: The transformation of the American diet (Vol. 7). Berkeley: University of California Press.
  • Pimentel, D., & Pimentel, M. H. (2008). Food, energy, and society. CRC Press.
  • Roland-Holst, D., Otte, J., Pfeiffer, D., & Zilberman, D. (2012). Health and animal agriculture in developing countries. New York: Springer Verlag.
  • Scott, J. C. (2017). Against the grain: A deep history of the earliest states. New Haven, CT: Yale University Press.


  • Ahmed, A. I., Uddin, M., & Global Forest Coalition. (2008) Underlying causes of deforestation and forest degradation in Bangladesh.
  • Braudel, F. (1992). Civilization and capitalism, 15th-18th century: The perspective of the world (Vol. 3). Berkeley: University of California Press.
  • Brown, K. A., Harris, D. F., & Molly. (2016). Light-driven dinitrogen reduction catalyzed by a CdS: Nitrogenase MoFe protein biohybrid. Science, 352(6284), 448–450.
  • California Energy Commission (CEC). (2005). California’s water-energy relationship. Final Staff Rep., CEC 700- 2005-011-SF.
  • Canning P., Charles, A., Huang, S., Polenske, K., & Waters, A. (2010). Energy use in the U.S. food system. United States Department of Agriculture, Economic Research Service, Economic Research Report 94.
  • Canning, P. (2011). Energy use in the US food system. Diane Publishing.
  • Carlsson-Kanyama, A., Ekstrom, M., & Shanahan, H. (2003). Food and life cycle energy inputs: Consequences of diet and ways to increase efficiency. Ecological Economics, 44, 293–307.
  • Carty, R. (2012). Casualty of war. Chemical Heritage, 30(2).
  • Coley, D., Howard, M., & Winter, M. (2009). Local food, food miles and carbon emissions: A comparison of farm shop and mass distribution approaches. Food Policy, 34, 150–155.
  • Da Rocha, J. M., & Restuccia, D. (2006). The role of agriculture in aggregate business cycles. Review of Economic Dynamics, 9(3), 455–482.
  • Dekle, R., & Vandenbroucke, G. (2012). A quantitative analysis of China’s structural transformation. Journal of Economic Dynamics and Control, 36, 119–135.
  • Dennis, B. N., & Iscan, T. B. (2009). Engle versus Baumol: Accounting for structural change using two centuries of U.S. data. Explorations in Economics History, 46, 186–202.
  • Diamond, J. M. (1997). Guns, germs, and steel: The fates of human societies. New York, NY: W.W. Norton.
  • Duarte, M., & Restuccia, D. (2010). The role of the structural transformation in aggregate productivity. Quarterly Journal of Economics, 125, 129–173.
  • Eaton, J., & Kortum, S. (2002). Technology, geography, and trade. Econometrica, 70, 1741–1879.
  • Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z., & Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1(10), 636.
  • Foellmi, R., & Zweimüller, J. (2008). Structural change, Engel’s consumption cycles, and Kaldor’s facts of economic growth. Journal of Monetary Economics, 55, 1317–1328.
  • Haber, L. F. (1986). The poisonous cloud: Chemical warfare in the First World War. Oxford, U.K.: Oxford University Press.
  • Herrendorf, B., Rogerson, R., & Valentinyi, A. (2014). Growth and structural transformation. In Handbook of economic growth Vol. 2B.
  • Herrendorf, B., Rogerson, R., & Valentinyi, Á. (2009). Two perspectives on preferences and structural transformation. American Economic Review.
  • Herrendorf, B., Schmitz, J. Jr., & Teixeira, A. (2012).The role of transportation in U.S. economic development: 1840–1860. International Economic Review, 53, 693–715.
  • Johnston, B. F., & Kilby, P. (1975). Agriculture and structural transformation: Economic strategies in late-developing countries. Oxford, U.K.: Oxford University Press.
  • Johnston, B. F., & Mellor, J. W. (1961). The role of agriculture in economic development. American Economic Review, 51, 566–593.
  • Jorgenson, D. W., & Timmer, M. P. (2011). Structural change in advanced nations: A new set of stylised facts. Scandinavian Journal of Economics, 113, 1–29.
  • Levenstein, H. A. (2003). Revolution at the table: The transformation of the American diet (Vol. 7). Berkeley: University of California Press.
  • Kim, D. I., & Topel, R. H. (1995). Labor markets and economic growth: Lessons from Korea's industrialization, 1970-1990. In Differences and changes in wage structures (pp. 227–264). University of Chicago Press.
  • Pelletier, N., Audsley, E., Brodt, S., Garnett, T., Henriksson, P., Kendall, A., . . . Troell, M. (2011). Energy intensity of agriculture and food systems. Annual Review of Environment and Resources, 36 (2011), 223–246.
  • Pimentel, D., & Pimentel, M. H. (2008). Food, energy, and society. Boca Raton: CRC Press.
  • Redzepi, R., & Zilber, D. (2019). The Noma Guide to Fermentation. New York: Artisan Press.
  • Roland-Holst, D., Otte, J., Pfeiffer, D., & Zilberman, D. (2012). Health and animal agriculture in developing countries. New York: Springer Verlag.
  • Roland-Holst, D., Triolo, R., Heft-Neal, S., & Bayrami, B. (2013). Bioplastics in California. Sacramento: California Environmental Protection Agency.
  • Saunders, C. M., Barber, A., & Taylor, G. (2006). Food miles-comparative energy/emissions performance of New Zealand’s agriculture industry.
  • Scott, J. C. (2017). Against the grain: A deep history of the earliest states. New Haven, CT: Yale University Press.
  • Simmons, B., McClendon, S. D., Baath, T., Petzold, C. J., Adams, P. D., & Singer, S. W. (2012). Thermoascus aurantiacus is a promising source of enzymes for biomass deconstruction under thermophilic conditions. Biotechnology for Biofuels, 5, 54.
  • Simmons, C., Reddy, A., Simmons, B. Singer, S., & Van der Gheynst, J. (2014). Effect of inoculum source on the enrichment of microbial communities on two lignocellulosic bioenergy crops under thermophilic and high-solids conditions. Journal of Applied Microbiology, 117(4), 1025–1034.
  • Smil, V. (2004). Enriching the earth: Fritz Haber, Carl Bosch, and the transformation of world food production. Cambridge, MA: MIT Press.
  • Sofiev, M., Winebrake, J. J., Johansson, L., Carr, E. W., Prank, M., Soares, J., . . . Corbett, J. J. (2018). Cleaner fuels for ships provide public health benefits with climate tradeoffs. Nature Communications, 9(1), 40.
  • Sørensen, B. (1991). A history of renewable energy technology. Energy Policy, 19(1), 8–12.
  • Stewart, W., Dibb, D., Johnston, A., & Smyth, T. (2005). The contribution of commercial fertilizer nutrients to food production. Agron, 97(1), 1–6.
  • Sundquist, B. (2007). Irrigation overview. In The earth’s carrying capacity, some related reviews and analysis.
  • Swedish National Food Administration (SNFA). (1996). Livsmedelstabell. Energi och na ̈ringsa ̈mnen. Uppsala, Sweden.
  • Tyedmers, P., Watson, R., & Pauly, D. (2005). Fueling global fishing fleets. AMBIO, 34(8), 635–638.
  • Tyedmers, P. (2010). Energy and seafood: Challenges and opportunities. Presented at Energy Use Fish. Symp., Seattle.
  • Woods, J., Williams, A., Hughes, J. K., & Black, M., & Murphy, R. (2010). Energy and the food system. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 2991–3006.
  • Zhang, W., Dou, Z., He, P., Ju, X.-T., Powlson, D., Chadwick, D., . . . Norse, D. (2013). New technologies reduce greenhouse gas emissions from nitrogenous fertilizer in China. Proceedings of the National Academy of Sciences, 110(21), 8375–8380.
  • Zilberman, D., Zhao, J., & Heiman, A. (2012). Adoption versus adaptation, with emphasis on climate change. Annual Review of Resource Economics, 4(1), 27–53.


  • 1. For readers interested in more detailed discussion along these lines, the best entry point is Pimentel and Pimentel (2008), which offers an exhaustive and largely definitive technical review of energy in agrifood systems.

  • 2. The most incisive recent perspective on this issue is offered by Scott (2017), which cites precedence generously.

  • 3. Sørensen (1991) offers some background but is more focused on the current transition back to renewables.

  • 4. Roland-Holst et al. (2012) provided an overview of this topic, proposing a set of sustainable development goals associated with livestock.

  • 5. Good entry points on the larger historical process are Diamond (1997) and Herrendorf et al. (2014).

  • 6. For comparison, see also Johnston, Kilby, and Mellor (1961, 1975) and Duarte and Restuccia (2010).

  • 7. Beyond farming, transportation provides even more compelling examples of the energy revolution. Wind enabled us to travel more conveniently on water, traveling up to 30 mph and even traversing oceans in a matter of months. The energy density of jet fuel allows us to make the same passage in a matter of hours, traveling 600 mph 30,000 feet above the Earth’s surface.

  • 8. See, for example, Levenstein (2003).

  • 9. Even California, a climate policy leader, originally exempted agriculture from its flagship GHG Cap & Trade system (Roland-Holst et al., 2012).

  • 10. NOAA provides a dramatic radar video of this phenomenon here: NOAA.

  • 11. In an extreme example, indoor dairies operated in major U.S. cities in the 19th century, with cows fed on the lees of breweries, until health and food safety codes prohibited the so-called Swill Milk industry (Rathge, 2016).

  • 12. See, for example, EuroFresh Farms Introduces Innovative Food-Safety, EnviroLockTM.

  • 13. Haber’s invention earned him a Nobel (Haber, 1920), but a darker side of his talent had emerged a few years earlier when he weaponized chorine and other toxic gases, leading to the deaths of tens of thousands on European battlefields (Haber, 1986) and a bitter cycle of personal tragedy (Carty, 2012).

  • 14. Fresh bread is a relatively recent luxury. For most of agrarian history, bread was baked at harvest time, broken or cut with an ax, and covered with a sauce to make it palatable (Braudel, 1992).

  • 15. Emergent global knowledge resources and more cosmopolitan tastes are fueling a renaissance of interest, adoption, and innovation in fermentation processes, from artisanal breweries to leading restaurants (e.g. Rezepi and Zilber: 2019).

  • 16. This is a dynamic area of sustainability research. Good entry points are Simmons et al. (2012, 2014).

  • 17. Until the invention of aspirin in 1910, alcohol was the only analgesic universally available to humanity. For most of human history, chronic pain was also a fact of adult life, with most of us dying by our late 40s from combinations of disease and disabilities large and small. These facts help explain the ubiquity of alcohol (even where sanctioned) in human society.

  • 18. The is a long and interesting literature on this process, for example, Jorgensen and Timmer (2011), Kim and Topel (1995), and Herrendorf, Rogerson, and Valentinyi. (2009).

  • 19. Seattle Times, NW salmon sent to China before reaching U.S. tables, July 16, 2005.

  • 20. This figure is three times the contribution of all commercial aviation, a sector that has already agreed to securitize 100% of its GHG emissions by 2025. The estimated cost, just 1% of revenue, will be passed on to passengers in ticket prices.