By Graham Zabel <email@example.com> , 20 Jul 2008
Population and Energy 1
The Mortality Revolution 4
Punctuated Equilibrium 5
The Connection Between Population and Economics 6
The Missing Connection Between Population and Energy 6
An Energy Model of Population Growth 7
Biomass Population 10
Coal Population and the Industrial Revolution 12
Oil Population and the Twentieth Century 17
Natural Gas Population and the Twenty-first Century 19
The Sum-Of-Energies Population Model 21
Future Scenarios 21
Scenario 1: Continued Fossil Fuel Growth 22
Scenario 2: Fossil Fuel Decline with no Sufficient Substitute 22
Scenario 3: A New Energy Source 24
Figure 1: Percentage of Total Calculated Consumption Contributed by each Energy Source 9
Figure 2: Energy Share Dynamics 9
Figure 3: Sum-of-Energies model of World Population 10
Figure 4: Biomass Population - World Population 800-1850 compared to Exponential growth 11
Figure 5: Coal Population – World Population less Biomass Population 1850-1950 15
Figure 6: World Coal Consumption 1860-2000 16
Figure 7: Oil Population – World Population less Biomass and Coal Population 1950-2000 18
Figure 8: World Crude Oil Consumption 1900-2000 19
Figure 9: Natural Gas Population - post-2000 20
Figure 10: World Natural Gas Consumption 1900-2000 20
Figure 11: World Population vs. Sum-of-Energies Population 800-2000 21
Figure 12: Projected World Oil Production to 2050 23
This paper will argue that populations exhibit a behaviour that could be described as punctuated equilibrium. That is, populations generally exhibit long-term homeostasis. During brief and rare periods in history, population pressures lead to the commercialisation of a new source of energy – particularly a higher quality energy source – which in turn will raise the population ceiling, or the number of people the earth can support. At this stage, populations will grow quickly to approach the newly raised ceiling, then growth will slow and a new homeostasis will develop.
The planet could not support the six billion people that exist today without first the commercialisation of coal, then of oil and gas. If these energy sources were necessary for the historically rare and unprecedented population growth that has occurred over the last three hundred years, then this growth might be correlated (and modelled), in some way, after the pattern of consumption of these energy sources.
In 1750, the world’s population was approximately 720 million people. Over the previous 1000 years, this population had been growing very slowly at an average rate of about 0.13%. At this rate population doubles every 500 years and it would have taken over 1500 more years (sometime near the year 3250) to reach our current population of 6 billion people. But sometime in the 18th century, circumstances changed and population began growing rapidly.
The most common explanation for this change in circumstances is that a mortality revolution reduced the rate at which people died and that this mortality revolution was brought about by the Industrial Revolution. The Industrial Revolution changed everything. It was an economic revolution, which spawned revolutions in science, technology, transportation, communication and agriculture. As a consequence, humanity began to experience improvements in health, nutrition, food variety, medicine and quality of life. More people survived infancy and childhood and they carried on to live longer lives. Because people were dying less quickly, populations grew more quickly.
Large and sustained population growth is thus a contemporary phenomenon: until historically recent times it was rare to non-existent. Preindustrial populations grew when times were good (favourable climatic, agricultural, political and economic conditions) and shrank when times were bad (droughts, famines, wars, plagues, bad weather). Population growth was at all times restricted by the amount of land and food available. Land was needed to grow food for humans, fodder for animals and trees for building and fuel. As populations grew and occupied prime land, people were forced onto less productive land and the competing interests of food, fodder and fuel grew stronger.
This pressure on land led to a number of different consequences: rising prices, under-nourishment, hunger; migration, territorial expansion through aggression and war and internal revolt. Populations became more susceptible to famine, disease, plague and death. Thomas Malthus referred to these consequences as positive checks on population growth. Population pressure also lead to what Malthus referred to as preventive checks. Preventive checks consisted of celibacy, reducing fertility within marriage and through increased age at first marriage (i.e. marrying later). These observations – as populations grew survival became more difficult (populations experienced declining marginal returns), leading to positive or preventive checks on population growth - lead to Malthus’ famous Essay on Population (1798).
So according to Malthus, an initial population starts with few people. It then grows in an approximately exponential manner towards demographic saturation. This exponential growth then slows as the limit to population size, or the population ceiling, is reached. It is at this point that populations become homeostatic. This ceiling results when most available land has been used, and most productivity gains have been realised. Any further expansions into less productive land, or further productivity gains, suffer from declining marginal returns.
But events subsequent to Malthus have shown his ideas to be incomplete. Homeostasis can be disrupted. Growth may again accelerate if a population finds a way to shift upward its population ceiling, its demographic saturation point. The orthodox belief is that a population ceiling shifts upward due either to expansion into new frontiers (migration), improved productivity or technological breakthroughs. (Example: America’s population growth in the 19th century as frontiers expanded.) This paper hypothesises that there may be another, more fundamental, reason for upward shifts in a population ceiling: the commercialisation of a new source of energy.
Ester Boserup has argued that at the stage when population pressures are making life more miserable, food scarcer and prices higher, humans invent technologies to overcome these pressures. “This multiplication of world population would not have been possible without successive technological changes.” According to Boserup, technological improvements raise the population ceiling and allow populations to expand.
Roughly 10,000 years ago, increasing population pressure on wild food resources led to a shift from food gathering (hunter-gatherers) to food production (agriculturists) in several parts of the world. This lead to demand-induced technologies and sources of energy supply, such as “water power for flow irrigation, animal draft power, iron tools, and fire for land clearing and for improvement of hunting and pastoralism.”
Population pressures in many parts of Europe in the seventeenth and eighteenth centuries led to serious shortages of wood which in turn lead to many of the technological innovations that fuelled the Industrial Revolution. Coal’s replacement of wood as the most important source of energy in Western Europe is “a classic example of demand-induced innovation…promoted by population pressures on forested land in Western and Central Europe.”
Whereas Boserup argues that population pressures lead to technological advances, this paper argues that population pressures lead to the commercialisation of a new energy source (water and wind power, animal draft power, coal, oil and natural gas), which in turn lead to technical advances.
More recently several writers have attempted a synthesis of Malthus’ and Boserup’s ideas. Lee and Woods both show that population growth may exhibit a series of steps: populations grow to a Malthusian ceiling, and then a change in circumstance (be it technological advance or new energy source) raises the ceiling and population makes a Boserupian step to the next Malthusian ceiling.
Lee and Woods have argued that in general preindustrial populations were relatively stable, or homeostatic, over the long-term. They grew and shrank, but over the long-term “the human population has tended toward equilibria that have been tending upward” (Lee, 1987). A graph of preindustrial population growth (see Figure 4) depicts this ebb and flow of population growth and also shows the general slight long-term increase in the world’s population. Unlike today, many parts of the world were sparsely populated. There were frontier regions where land was sparsely populated and into which populations could migrate when domestic situations became too crowded or life too miserable.
It has been argued that there is no link between population growth and energy use. Examples are made of China (in the 1800s) and India and Africa (in the latter half of the 1900s). These regions witnessed explosive population growth without ever having become industrialised. Their use of coal and oil was minimal and not a significant factor in their population growth.
Yet coal powered the trains that criss-cross India. India’s extensive transportation network, powered by coal and later oil, becomes a great distribution network for food. It enables populations to live in marginal environments (like the deserts of Rahjastan). It probably explains why India has few of the problems Africa has in feeding its population. (Has India experienced many famines?).
In those regions, the argument goes, the greatest single factor in population growth was mortality decline. Modern medicine, health, hygiene, education, contraception are all given as reasons for mortality decline. Energy is never mentioned.
An argument can be made that every one of those reasons were a consequence of coal and oil use. The great Scientific Revolution was driven by men (mostly) who were afforded the leisure to experiment and research because machines and industry released them from the fields and provided the wealth for an educated class.
Vaccinations (smallpox in 1796, other dates? – 1855 London sewers modernised after cholera outbreak), developed and made in the first world, require fuel to be distributed in the third world. Most of sub-Saharan Africa has few oil resources yet very high population growth rates. How can there be a correlation between energy and population? While these African countries do not have the capital to develop vaccines, they benefit from oil purchased by the first world that fuels the aeroplanes and trucks that distribute vaccines (and contraceptives?) to them.
The majority of schools, hospitals and health clinics in Africa are built with first world capital, and with first world energy. Money does not build the schools and clinics; machines and tools, designed and built in the developed world, produced using fossil fuels, do.
A tractor, for example, is made of metal, rubber and plastic. Plastic is a direct by-product of oil. Metal is mined and forged using immense quantities of heat and energy. The designer of the tractor uses a computer, office space, lights and heating, drives to and from work in a gasoline-burning car. The tractor is built in highly sophisticated factory that in turn must be designed and built with metals and plastics, and required large amounts of energy. The chain of events always has another link, but the ultimate first cause is always inputs of energy.
This paper attempts to describe population behaviour as punctuated equilibrium: populations generally exhibit a long-term homeostasis but during brief, unusual periods in history certain events will dramatically raise (or lower) the population ceiling. At this stage, populations will grow quickly to approach the newly raised ceiling, then growth will slow substantially and a new homeostasis will develop.
In broad terms, the commercialisation of coal in the nineteenth century significantly raised the both national and global population ceilings. The commercialisation of oil in the twentieth century raised ceilings yet again. The commercialisation of natural gas may aid in squeezing ceilings up further still. And importantly, the decline of any or all of these energy sources may cause the population ceiling to return to lower levels.
Population ceilings are very difficult to determine. It is almost impossible to say how many people the earth could have supported had there been no commercialisation of coal and oil. It is possibly higher than 1.2 billion, the world’s population around 1850 a date arbitrarily chosen as the start of coal population. North and South America and Australia were all thinly populated at that time and could have sustained much higher populations based on the use of traditional renewable energy sources.
Similarly, without the commercialisation of oil, the world’s coal population could also have continued to grow substantially. Some estimate that there remain over 200 years of coal reserves. Admittedly, without oil and gas, coal consumption would have been much more rapid, but reserves and resources would likely have been large enough to support more than the 2.5 billion people existing at the start of oil population, beginning in the middle of the twentieth century.
Each commercialisation of a new source of energy, particularly if higher quality than the then dominant source, raises the population ceiling.
But contrarily, each new step up the energy ladder raises productivity per capita (assuming productivity growth can outpace population growth) and income levels. This tends to slow population growth through several negative effects on fertility rates.
The introduction of a new energy source affects population growth by:
- directly lowering mortality rates, through improvements in health and safety, medicine, vaccines, etc.
- indirectly affecting fertility rates. Fertility rates are directly affected by mortality rates, so a fall in mortality rates as described above will likely by followed by a fall in fertility rates.
A new energy source initially precipitates a dramatic increase in population growth due to a sharp increase in the population ceiling, then after the effects of the new energy source are assimilated into society, growth falls back to some steady state.
There is a substantial literature debating the relationship between economics and population. (Easterlin, Caldwell, etc.) Yet economics is driven by energy. Without energy, there is no work, and without work, no economy. In the genealogy of social forces, the top of the tree, the first cause, is always energy.
Most analysts in seeking causes for population behaviour, be it growth, fertility or mortality, look at socio-economic causes. Education, income, medical facilities and employment are all suspects when population behaviour is questioned. Yet…
Whereas the link between economics and demographics has been endlessly debated, it is very difficult to find anything in demographic literature relating population to energy supply or energy consumption.
In each index of three well-known books on British Population History (Anderson, Tranter, Wrigley and Schofield) there is not one entry for ‘coal’, ‘energy’ or ‘oil’. (There is one entry on ‘coalmining communities’ in Anderson regarding an issue not relevant to this study.) In the index of Wrigley and Schofield’s seminal 700-odd page ‘Population History of England’ there is not one listing for either coal or energy. Yet clearly without coal the population history of England would have been very different
The index of Livi-Bacci’s A Concise History of World Population contains no entries for ‘coal’ or ‘oil’. ‘Energy’ has two listings but in neither case is energy consumption or production linked to population growth. Livi-Bacci, like so many others, attempts to relate population growth to economic growth (ignoring energy) but even then abandons “any attempt to determine a casual relationship between population and economy”.
Vaclav Smil states, “Energy availability is … of limited usefulness in explaining population growth…Nor is it very helpful to see the rising use of fossil fuels, reflected in better housing, hygiene, and health care, as the key factor driving population growth. Undoubtedly, these changes were of major importance in Europe, but hardly so in contemporary China…Not surprisingly, energy considerations are also of limited help in trying to explain some of the greatest recurrent puzzles of history, the collapse of complex societies.”
Yet the world’s population would not be anything like the six billion that it is, if not for the discovery, commercialisation and mass use of coal and subsequently oil and gas. Vast inputs of energy into modern society have lead to vast increases in population.
Cohen (1995) examined several different methods of curve fitting – producing a mathematical equation describing a curve that mimics population growth. He examined several attempts: the exponential curve, the logistic curve, the doomsday curve and the sum-of-exponentials curve. All were found lacking the ability to describe population growth. Curve fitting has met with little success in describing any type of population behaviour.
But of the four models Cohen examined, the sum-of-exponentials had the most promise. This sum-of-exponentials model divides the total population into two or more subpopulations, fits an exponential curve to each subpopulation and then adds the curves together to get a global picture.
His example uses two populations growing exponentially: a large but slow growing population and a small but fast growing population. Summing these two equations of exponential growth produces a curve whose “visual similarity to the data on population history is not bad”.
Dividing the global population into subpopulations, and modelling each of these subpopulations separately makes common sense. A population model for Kenya will be very different from one for Sweden, or for California. Every country, every region, has different and unique demographic characteristics. Clearly one model cannot adequately represent such a wide variety of circumstances. On the other hand, a regional model approach would be so complex that any insights into the general causes of population growth would be lost. The sum-of-exponentials model was an attempt to find a middle way.
If there is a relationship between energy consumption and population growth, the different types of energy consumed may have different effects. If biomass is the only energy source, populations will not grow very fast. In such organically based economies, “the problem of expanding raw material supply, and especially the related problems associated with the very modest energy supply maxima…must curb growth with increasing severity as expansion takes place.” The emergence of coal as an energy source eliminated the ceiling on population growth that any organically based culture would eventually face. Similarly, the predominance of oil after the middle part of the twentieth century raised the ceiling even further – it allowed even faster population growth by contributing significantly to mortality decline through the expansion and distribution of food, trade and vaccines and as the input to fertilisers and pesticides.
Following this hypothesis, we develop a model of population growth that can be divided into three components: population growth due to biomass energy, population growth due to coal and population growth due to oil. (Natural gas will be examined briefly as its effects have only recently been felt). Each of these components can be modelled separately and then combined together to create a sum-of-energies model of population growth.
There is no reason to believe that all or any of these components will exhibit exponential growth. Organically based populations may grow at a slow exponential rate as long as there is frontier land to expand into. If frontiers are fixed, growth may stop or redirect itself through migration. Oil based populations may not grow exponentially at all. In fact, they may decline as higher standards of living lower fertility – a phenomenon observed throughout the industrialised world. Once frontiers are closed and productivity gains have been realised, population growth may begin to plateau, exhibiting the common S shape of the logistic curve.
The sum-of-energies model assumes that different energy sources dominate during different periods of history. For example, traditional renewables (wood, dung, etc.) were the world’s dominant sources of energy until almost 1900. Coal then was the dominant source of energy until the middle of the twentieth century, after which crude oil began to dominate. Oil remains the dominant source of energy to this day, but its share in the energy mix peaked in 1973 and has been declining since. The natural gas share of the energy mix has been steadily increasing and looks set to take over the number one position sometime early this century.
Sources: Jenkins (1989), WEC (1995), BP (2000)
Determining when the introduction of a new energy source can raise a population’s carrying capacity to a new level is not easy and is somewhat arbitrary. One method that seems to fit with empirical observation is that an energy source becomes globally important when it attains a 20% share of the world’s energy mix. For coal, this occurred in 1860, oil in 1950, natural gas attained that level very recently.
20% of Energy Share
Energy Share Peaks
20% off Peak
Source: WEC 1995, page 10
The newly ascending energy source typically reaches 20% of the energy mix at about the same time as the dominant source has fallen about 20% from its peak. Figure 2 shows how coal attained 20% of the energy mix in 1860, the same time that traditional renewables had lost 20% of their original 100% share. Oil attained a 20% share in 1948, shortly after 1940 when coal had dropped 20% from its peak of 1912.
This then will be the criterion used to separate the effects of each energy source on the world’s population: when a new energy source attains a 20% share of the global energy mix, it has reached a level where it can upwardly shift the population ceiling.
The model is constructed in the following manner:
The world’s population from the beginning of time until 1850, just as coal reached a 20% share of energy resources, will be referred to as Biomass Population. Coal Population reigns from 1850 until 1950, when oil reached a 20% share of energy resources. Coal Population is the population of the world not accounted for by the slowly growing Biomass Population. That is, it is represented by the population that remains when Biomass Population is subtracted from the world’s total population. Similarly, Oil Population is the population from 1950 until 2000 that is not accounted for by either Biomass Population or Coal Population. Natural Gas Population starts approximately now, within the last ten years, as it reaches a 20% share of energy resources. The behaviour of Natural Gas Population has yet to be determined. The diagram in Figure 3 illustrates these concepts.
An examination of each component follows.
Until 1850, most of the world’s population was still supported by traditional renewables (wood, dung, etc.) and animal power (with minor amounts of wind and hydropower). Admittedly Britain was already heavily influenced by coal, but very few other populations were. In 1850, Britain was producing more coal than the rest of Europe combined. In the same year, when the population of the United States was already 23 million, 90% of its energy requirements were still met from wood. So until the mid-1800s, energy from biomass was the main energy contributor to population growth. (It still contributes to population growth. It is estimated that 10% of the world’s energy in the year 2000 is provided by biomass and there are an estimated two billion people that still have no access to electricity.) Wrigley describes this preindustrial era as the Organic Economy, and in England’s case, the Advanced Organic Economy. In this model, it is called Biomass Population.
Biomass Population growth fluctuated in waves of feast and famine; economic growth and population checks. If populations grew too quickly, living standards declined, local carrying capacities were exceeded and food became more expensive. Malthusian population checks ensued: later age at first marriage, decreases in life expectancy and higher mortality. Biomass Population had been growing at a slow, exponential rate with some slight ups and downs for thousands of years. In other words, it exhibited homeostatic behaviour. Population pressures in Europe were relieved through the safety valves of migration. Settlers expanded into sparsely populated regions of the world such as North and South America, Australia and many African and Asian colonies. This enabled small upward shifts in the global population ceiling, or the population equilibrium.
If Biomass Population growth from 800 to 1850 were extrapolated to the year 2000, the value would be 1.09 billion people. This may or may not be an indication of how many people the planet would now be supporting if coal, oil and gas were never commercialised, assuming there were still frontiers to expand into.
Source: McEvedy and Jones (1978)
Figure 4 shows the world’s population from 800 to 1850. The plotted line represents pre-fossil fuel population, or the world population growth that occurred when biomass was the predominant source of energy. The black line is a fitted line representing exponential growth. Extrapolating the exponential trend line to the year 2000 gives a value of 1.09 billion people, as mentioned above. Hypothetically then, a world based solely on biomass may feasibly support a magnitude of one billion people. (While it was also just mentioned that today 2 billion people are without electricity, they do not necessary contribute to Biomass Population. It will be argued below that coal and oil are primary factors in the production and distribution of food and health care that support many of these people.)
“In ages past (pre-Industrial Revolution), better living standards had always been followed by a rise in population that eventually consumed the gains…Gone, Malthus’ positive checks and the stagnationist predictions of the ‘dismal science’; instead, one had an age of promise and great expectations.”
Both oil and coal have been used in small quantities for thousands of years. But until the Industrial Revolution, society’s energy requirements were fulfilled almost entirely by human and animal power and traditional biomass sources. For many years afterward, a large majority of the world remained dependent on traditional biomass. By 1850 population pressures led to the commercialisation of coal and the Industrial Revolution, and the energy derived from coal began to shape the forces that would raise the population ceiling. The world’s population entered a phase of disequilibrium.
In the early sixteenth century, Britain was heavily dependent on foreign suppliers for arms. The threat of war between England and the Catholic countries resulted in an embargo of Dutch manufactured arms to Britain. The embargo impressed upon King Henry VIII the need for self-sufficiency in arms manufacture so he proceeded to establish a domestic arms industry. In 1543, according to the Elizabethan chronicler Holinshed, “the first cast pieces of iron that ever were made in England.” Elizabeth continued the drive for self-sufficiency in many other manufactures: for example salt, copper and glass. All of these industries were heavily dependent on charcoal, which was made from wood. The increasing demands for wood in concert with an increasing population lead to an alarming rise in deforested lands first in England and then Ireland, as the search for timber widened. Although coal was dirty and smelly, the scarcity and rising costs of wood forced many people to resort to the burning of coal for heat. Even before 1600, “London and all other towns near the sea…are mostly driven to burn…coals, for most of the woods are consumed.”
Boserup argues that timber and charcoal became scarce, in response to population pressures and the growing demand for these products by nascent industrial sectors. The success of coal in the use of iron production toward the end of the eighteenth century meant that “the shortages of energy and raw materials were overcome and the Industrial Revolution became possible”.
Since the Industrial Revolution, populations have grown much more quickly. The countries that first experienced industrialisation were the first to grow more quickly. England, where the Industrial Revolution began, was also the first country to witness accelerated population growth. From the late 1700s, Britain’s population begun to grow at levels never seen before. There is little record of population growth before approximately 1500, but lacking the medical advances that are today taken for granted, it is unlikely historical mortality rates could ever have been as low as ours are now. Higher historical fertility rates were always more than compensated for by high mortality rates, putting a brake on population growth. Since 1541, the population of Great Britain never grew faster than it was growing by 1800. By the 1820s, England’s population was growing annually at approximately 1.6%, a rate never surpassed before or since. (Current population growth is negative – English population is declining for the first time since the early 1700s.)
“Between 1550 and 1820 the populations of France, Spain, Germany, Italy and The Netherlands all appear to have grown by between 50 and 80 per cent; in England over the same period the comparable figure was 280 per cent, a contrast so striking that by 1820 England, which had once been a small country by the standards of the larger European powers, though still less populous than France, Germany or Italy, was moving rapidly towards rough equality with them”. During the same period, Britain was mining coal in quantities unseen anywhere else in the world. “In 1800 the output of coal in Britain had reached about 15 million tons a year, at a time when the combined production of the whole of continental Europe probably did not exceed 3 million tons.”
Between 1800 and 1900, the Industrial Revolution crossed the Channel and spread to the rest of Europe. So did the importance of coal. The commercialisation of coal that occurred in Europe in the eighteenth and nineteenth centuries dramatically increased productivity through the use of steam engines that drove trains, boats and many other engines, and through the coking process used to produce steel. Coal made available twice as much heat as an equivalent amount of dry wood. Coal is much more productive than wood – it has a higher thermodynamic potential. By 1900, coal was powering the entire world’s major industrial processes, and powering the industrial nations’ population growth.
Between 1800 and 1900, Europe’s population more than doubled from about 187 million to 400 million. As a percentage of world population it climbed from 21% to roughly 25%. While this percentage increase does not seem very large, it doesn’t measure the roughly 35 million Europeans who immigrated elsewhere. These European immigrants and their descendants spawned large and often numerically dominant populations in many other parts of the world including the United States, Canada, Australia, New Zealand and many regions in Latin America. They also brought with them the European penchant for coal consumption. By 1865 coal had gained a 20% share of energy consumption in the United States. Shortly after 1880 coal became the main source of energy in the U.S. As a percentage of total consumption contributed by each energy source, coal consumption peak in 1910. The year of highest population growth in the U.S. in the twentieth century occurred in almost exactly the same year.
Hackett-Fischer (1996) explains population growth in the eighteen century as follows:
“There was also a modest improvement in life expectancy for infants and women during the eighteenth century, and a moderate stabilization of death-rates. But the primary cause of population growth in this period was a rise in fertility, not a fall in mortality.
Why did men and women choose to marry earlier and have more children? An improvement in material conditions was part of the answer, but not the whole of it. Husbands and wives decided to have more children because the world appeared to have become a better place in which to raise a family.”
That the world appeared to have become a better place was largely due to the commercialisation of coal and the subsequent technological innovations.
Later, in the twentieth century, as European populations at home and abroad began to grow more slowly, populations in other parts of the world began to reap some of the benefits of industrialisation. As these benefits filtered to the developing world, it too appeared to have become a better place and developing world populations started growing more quickly. In many cases these African, Asian and Latin American populations grew at rates never before witnessed, eclipsing even the unusually high rates of nineteenth century Europe. (Many started growing extremely fast after 1950 – Oil Population – due to advanced transportation and distribution facilities ??)
Coal greatly reduced pressures of land use. Wood for heating and fuel was replaced by coal, so the land needed to grow that wood could serve a new purpose. The large quantities of fodder for draught animals and horse transport were made redundant by coal and the machines driven on coal. This further reduced pressures on land use and freed large amounts for the increased use of agriculture for food for humans.
The commercialisation of coal eliminated the “dependence upon the products of the land whose quantity could not be expanded indefinitely…This ensured that the process of growth at a relatively high rate could be sustained over a very long period. The key change that ensured the latter was the tapping of a new store of energy capital, so abundant that its production could be expanded immensely without causing any immediate problems of exhaustion of the energy stock. Access to abundant energy stocks was initially of limited value because the new sources of energy could be used only to provide heat, but once a method had been devised for deriving mechanical work also from the new energy source the way was clear for individual productivity to make a quantum leap.”
1850 is chosen, somewhat arbitrarily, as the year that the world began to feel the effects of coal. Around 1850 the world’s population began to grow much faster. For the first time in history, annual world population growth exceeded 0.5%. Most demographers try to explain this growth in terms of economics or mortality decline. This model regards the phenomenon in terms of energy, in this case, coal consumption. Obviously, coal was important in British society much before 1850 and some continents would not feel the benefits of coal power until much later. But by 1850 coal was being mined extensively, canal transportation was growing quickly throughout Europe, the age of rail transport had begun and iron-works were commonplace. This lead to increases in wealth, prices, distribution of foodstuffs, and internal and external migration. The substitution of machines and engines for human and animate power fostered the improvement of material conditions and quality of life. This process began in Britain, but the effects spread far and wide.
If we subtract the slowly growing Biomass Population from the total population between 1850 and 1950, we are left with Coal Population (Figure 5). Coal Population’s contribution to world population is then extrapolated backward to 1750 and forward to the year 2000.
The increase in energy inputs into society from the use of coal drove the machines that freed up time for humans to make advances in medicine and health. Coal transported the machines that distributed these advances through European society. Coal has a higher thermodynamic energy potential than traditional biomass, and is able to perform more work.
Coal also played a large part in the development of electricity. With the establishment of the electricity industry in the 1880s following the remarkable achievements of Edison, Parsons, Stanley, Tesla, Westinghouse and their collaborators, electricity quickly expanded to power households, industry and railroads. Electricity was generated in power plants, and those power plants were fed with coal. Still today, 50% of America’s power is generated in coal-burning power plants.
These advances and productivity improvements aided (and may have brought about) the Mortality Revolution, Urbanisation and the Fertility Revolution (in Europe and America). During this time frontiers were still open but were shrinking fast. (Oklahoma, the 46th state of America, was founded in 1907).
The model predicts that Coal Population grows in a logistic manner. That is, population initially grows quickly but eventually a coal population ceiling is reached as coalfields diminish, coal becomes harder to extract from deeper mines, as the productivity of machines driven by coal begins to plateau and as new, cleaner, more productive energy sources begin to supplant coal.
Fairly reliable world coal consumption statistics are available from 1860 onwards. Current annual world coal consumption is approximately 2.2 Gtoe (giga-tonnes oil equivalent), or about 2/5 of one tonne per capita. Consumption has remained stable at this level for over a decade. Increases in coal consumption in developing countries are compensated by decreases in consumption in the developed world as these economies switch from coal to cleaner burner oil and natural gas technologies. Because coal emissions are the dirtiest of the fossil fuel emissions, pressure to reduce coal use grows with concern over the potential climate altering effects of increased carbon dioxide emissions.
Sources: Jenkins (1989), BP (2000)
This model assumes that annual coal consumption will peak at approximately 2.8 Gtoe. This value is midway between the World Energy Council (WEC) future energy scenarios B and C. Scenario B is a business as usual scenario which estimates coal use at 3.4 Gtoe in 2020 and 4.1 Gtoe in 2050. Scenario C is an ecologically driven scenario which estimates coal use at 2.3 Gtoe in 2020 and 1.5 Gtoe in 2050. Consumption declines in scenario C after 2020 as stricter emission controls take effect.
In the case of coal, a peak of 2.8 Gtoe has no relation to the amount of world coal reserves, which are estimated to last for over 200 years. Rather future coal consumption is seen as being limited by environmental concerns and cleaner alternatives.
If we assume that Coal Population grows in a similar manner to coal consumption then both logistically growing coal consumption (Figure 6) and logistically growing Coal Population (Figure 5) reach 80% of their limits in the year 2000. At this rate Coal Population reaches a plateau of approximately 2.3 billion people in the 21st century. In other words, at current and projected rates of coal consumption, coal supports just under 2 billion people in the year 2000 and can be expected to support as many as 2.3 billion people this century.
Before a coal population ceiling was reached, a new source of energy replaced coal’s dominance. Oil was the next source of energy to be commercialised.
In 1859, Colonel E. L. Drake struck oil in Pennsylvania. More oil was discovered in Texas in 1887. By 1900, oil was extracted in Baku on the Caspian Sea, in Romania, California and Sumatra. By World War I, production had expanded to Mexico, Trinidad, Venezuela and Iran.
Population growth rates in America up until the discovery of oil in 1859 were very high, around 3%. At these rates a population doubles in size in 23 years. By 1900, the United States numbered 45 states, most of the continent had been conquered and the high population growth rates of a country expanding territorially in 18th century America were falling. The population growth rate in the U.S. reached 2.11% in 1909, the highest rate ever reached in the United States in the twentieth century.
Oil is easier to handle than coal. It is cleaner burning and cheaper to transport and store, making it ideal as a transportation fuel. It has a higher thermodynamic potential than coal, and was able to further increase productivity and arguably lead to less land use demand (as oil is underground and puts few demands on land use, whereas open face coal mines use land that could be otherwise put to use).
It is clear that availability of fossil fuels, in particular crude oil, as had a profound effect on population growth. Population has grown because death rates have declined worldwide, but birth rates have remained at high levels in many parts of the world. Oil arguably plays a part in both phenomena.
Oil provides the energy needed to grow and distribute food, and to increase the nutritional content of agricultural produce. Extensive land, air and sea transportation networks enable easy distribution of food. This stimulates mortality decline by getting food to the people that need it, alleviating local food shortages, flying food aid to drought stricken regions and shipping grain to countries whose populations have grown larger than their output of food. As recently as the eighteenth century in Europe, food was typically transported no more than 15 kilometres. Today, jumbo jets transport fresh food around the world everyday.
Oil also plays a significant part in the so-called Green Revolution that has led to growth in agricultural output that has managed to keep up with or even exceed the number of mouths that require feeding. Green Revolution agriculture relies on large amounts of pesticides and fertilisers, products highly dependent on oil and gas. Intensification of agriculture leads to surplus production, enabling greater increases in population which in turn lead to still greater demands for food.
Water for agriculture is also highly dependent on fossil fuels. Pumping of aquifers and groundwater for irrigation “is a phenomenon of the late twentieth century, made possible by the availability of electricity and cheap pumps.”
From 1950 to 2000, Oil Population is derived by subtracting Biomass Population and Coal Population from the world’s total population. Oil Population is plotted in Figure 7 along with a fitted logistic curve. The graph shows that currently almost 3 billion people are supported by oil.
Sources: Jenkins (1989), BP (2000)
Figure 8 plots world crude oil consumption from 1900 to 2000. The dips in the oil consumption curve reflect the two oil shocks in the 1970s (1973 and 1979) and the consequences of the Gulf War in 1991. There was also a slowing of oil consumption growth in the late 1990s as a result of the economic slowdown in Asia but presently consumption growth is increasing again as Asia’s economy is recovering and a strong economy in America boosts demand.
A logistic curve is fitted to the oil consumption line which assumes a peak annual consumption of 3.8 Gigatonnes of oil (Gto). This is consistent with the WEC projected consumption in 2020 under their scenario B – business as usual. It is substantially higher than the decline to 3.0 Gto projected in their ecologically driven scenario C but lower than the average increase to 4.5 Gto projected in their high growth scenario A. Based on this assumed peak of 3.8 Gto per year, the world has reached 95% of that level in the year 2000.
Assuming a similar logistic curve could represent Oil Population as depicted in Figure 7, then Oil Population has reached 89% of its hypothetical ceiling of 3.2 billion people in the year 2000.
There is vociferous debate as to whether growth in oil reserves with continue to grow faster than growth in oil consumption, allowing the ceiling of oil consumption to move upward. Alternative future scenarios will be examined in a following section.
Although the history of natural gas consumption is short and trends are very recent, based on the above figures Natural Gas Population may raise the population ceiling by another 500 million people or so (Figure 9). This increase is much smaller than the increase due either to coal or oil.
It is only speculation, but it may be that the higher the thermodynamic potential of energy sources, the less impact they have on raising population ceilings. Higher quality energy sources also lead to improvements in mortality and in standards of living. Both these factors in turn lead to lower fertility levels and thus slower, or even negative, population growth.
This is a sum-of-energies Component view of the world’s population:
Biomass Population - Slow Exponential Growth - open frontiers - low thermodynamic energy – low contribution to world’s population
Coal Population - Fast Logistic Growth - forming frontiers - medium thermodynamic energy – high contribution to world’s population
Oil Population – Logistic Growth (so far) - fixed frontiers - high thermodynamic energy – high contribution to world’s population
Natural Gas Population - Logistic Growth (so far) - fixed frontiers - high thermodynamic energy – low contribution to world’s population
The current best method of population projection is the cohort-component method. But it is entirely unable to predict population discontinuities due to famine, war, etc. It is also unable to predict baby booms or baby busts. On a global scale, an energy-component method may be better able.
For example, famine is very rarely due to a lack of food, rather to a lack of food distribution. This requires energy - trains, planes and automobiles, and fuel.
Figure 11 shows sum-of-energies equation versus actual population growth.
There are three general scenarios that the world’s energy future may take. Their effects on population will be radically different. They are:
1. Continued fossil fuel growth
2. Fossil fuel decline with no sufficient substitute.
3. A new source of energy
Oil and gas resources continue to be found faster than we consume them and population grows as projected by the UN, for example, and for that matter almost all agencies, to between 9 and 10 billion people by 2050.
Or, based on the above sum-of-energies model, a different interpretation might be that Oil Population is very close to reaching a plateau of approximately 3.2 billion people, and the world’s population may already be slowing more quickly than most analysts realise. If so, the world’s population in 2050 may be substantially lower, closer to 7 billion people. The increased importance of natural gas in the 21st century may raise the population ceiling, as the introduction of new energy sources has done in the past. But based on current trends Natural Gas Population may play a smaller part in raising the population ceiling (it may raise the ceiling by about half a billion people).
Oil and Gas resources are beginning to peak, as a growing minority of experts believes.
The World Energy Council’s future energy projections posit six future energy scenarios. Of these six, three scenarios see oil consumption peaking at roughly current levels by 2020.
The International Energy Administration, in their most recent World Energy Outlook publication, sees oil supply peaking before 2020 and obliquely refers to an oil supply shortfall of 19.1 million barrels per day by 2020. Some industry experts (Campbell 1988 and Laherrere 1999) believe that the oil production peak will occur much sooner.
If oil and gas production does exhibit a bell curve shaped profile (that is production starts at zero and ends at zero, in between production rises to a peak and then declines back toward zero) then at some point humanity will reach the peak. After that time oil and gas will become much more ‘expensive’. A decline in production would mean a decline in energy inputs into society - less thermodynamic energy - a decline in productivity and, hypothetically, a decline in population. If population growth were in any way related to oil production, Oil Population may decline more quickly than most people anticipate.
Mortality rates may increase, as a population grown large through dependence on high quality energy sources now must allocate scarcer resources per person. This is evident in agriculture’s dependence on fossil fuel based fertilisers. Without them, agricultural productivity decreases and less people can be feed. Less fuel - more famines. Human carrying capacity decreases and the ceiling on population size lowers.
Source: Campbell, The Coming Oil Crisis
Figure 12 depicts projected world oil production to 2050, based on figures compiled by Colin Campbell in The Coming Oil Crises. These figures are based on conventional crude oil resources. They do not include natural gas liquids, shale oil, oil from tar sands, ultra-deep water oil or polar oil. These oil sources are not included because they are much more expensive to extract, in monetary terms but also in energy terms. In other words, a large amount of energy inputs are required to extract energy outputs from say, tar sands in Northern Canada. Hence the net energy gain is lower, and these energy sources may not be as important in raising productivity and thus population ceilings.
Based on Campbell’s oil production projections, the 3.2 billion people that are dependent on oil in the sum-of-energies population model are in serious jeopardy in the next fifty years as the world’s remaining oil resources are consumed, and world population could suffer a precipitous decline.
This scenario follows from Ester Boserup’s observations that many of humankind’s technological innovations have resulted from population pressures, or increased population densities. According to Boserup, demand-induced innovation led to the shift from hunter-gatherer societies to agricultural societies and from the use of wood to coal. One could speculate that a shortage of fossil fuels caused by population pressures would lead to yet more innovation and the discovery of newer and better sources of energy. From our vantage point, though, it is not clear what these innovations might be nor what new sources of energy would be capable of replacing fossil fuels.
A higher quality energy source, say fission, could lead to further productivity improvements, reducing the pressure on existing resources and further raising the ceiling on population size. But fission still lies closer to the realms of science fiction than science.
A lower quality energy source, like solar or wind, is less efficient. It has lower thermodynamic potential and has less ability to perform work and to raise productivity. For example, a recent study on renewable energy remarks that solar radiation is completely diffuse and contains no appreciable concentration of energy. “For this reason, the vastness of the resource base of solar radiation is not, in itself, an indication of the appropriateness of solar energy as a useful energy source for society.” Another problem with low quality energy sources is that their net energy is low - they require a large proportion of energy in, to get some energy out - in contradistinction to oil and gas, which have high net energy values. A switch to a lower quality energy source from fossil fuels will put further pressure on other remaining energy sources, such as wood and coal. This could lead to further pressures on land and other resources and hence lower the population ceiling. Low quality energy resources do not support large populations.
Nuclear power is not the answer. To replace diminishing oil and gas (which currently provides the world with 65% of its energy resources) with nuclear power (which currently provides 7.6%) would not only require vast amounts of capital but would require vast amounts of high thermodynamic energy. In a period of declining oil and gas resources, existing energy sources would be getting scarcer.
Perhaps a new source of energy will be found with a high thermodynamic potential. This would then add a new energy component to population growth. This may lead to a raised population ceiling and an initial burst of population growth as population grows to occupy the space between the previous ceiling and the new ceiling. Then growth may slow again as a new homeostatic situation is reached. Probably, higher productivity will have further negative effects on both mortality (higher life expectancy/lower mortality) and fertility.
As productivity growth outpaces population growth, fertility may decline. Typically this is an economic argument. I believe that the underlying argument is about energy, and the quality of energy.
Agriculture, medicine, health can all be viewed in terms of energy. In fact, even something like agriculture could be viewed as having three components: Biomass Agriculture, Coal Agriculture and Oil and Gas Agriculture.
Nitrogen prices have risen 25% since late
May, says Agriliance, with soaring natural gas prices taking their toll on this
essential farm input.
How much of the success of the Green Revolution can be claimed by science, and how much by cheap fossil fuels? Cheap fuel supplies water pumps, processing plants and field machines. It is a low cost raw material for fertilisers, pesticides and herbicides. Agriculture is the single largest user of fossil fuels in the U.S. (proof?).
“All the evidence suggests that we have consistently exaggerated the contributions of technological genius and underestimated the contributions of natural resources.”
“Industrialisation came about at a fast enough pace so that it enlarged per capita wealth and was not entirely devoted to enlarging population. In principle, any increase in carrying capacity-temporary or permanent-affords a choice between enabling a larger number of individuals to live at previous standards. When the enlargement of carrying capacity is modest and is spread over many generations, it tends to be used mainly to increase numbers; if it is enormous and comes so suddenly that human numbers just don’t rise at the same pace, it raises living standards. The European takeover of the New World had enlarged carrying capacity (for Europeans) just fast enough to begin having this salutary effect. By drawing down stores of exhaustible resources at an ever-quickening pace, industrialization (temporarily) augmented carrying capacity even faster, affording opportunity for quite a marked rise in prosperity and for a phenomenal acceleration of population increase. The welcome rise in prosperity reinforced the dangerous myth of limitlessness and obscured for a while the hazards inherent in the population increase.”
AIDS – not as a phenomenon unrelated to energy, that decreases population. But absolutely related to energy. In America where per capita energy consumption is very high, there is enough energy to fight the disease to stop its lethal effects. In Africa, they do not have the energy to fight the disease, that is the energy to develop, produce and distribute the drugs and medicines that would mitigate many of the disastrous effects. Per captia energy consumption is very low. (See Hackett-Fisher’s The Great Wave). So Africa’s potential reduction in population growth may be directly related to the world’s reduced ability to provide the necessary energy sources. In fact Africa does have substantial energy resources but the large majority of these resources are exported to the energy hungry, wealthy nations of the developed world.
Migration may be viewed in terms of energy. Populations migrate from energy poor regions to energy rich regions, either energy producers (the Middle East) or consumers (Europe, North America).
The Baby Boom might be partially explained by the large growth in oil consumption after World War II. (Whether the oil enabled the boom, or the boom fuelled the production of oil doesn’t really matter. It couldn’t have happened without large energy inputs.) It may be that populations grow quickly when first encountering a new energy source and then slow afterwards as productivity gains permeate society and improve education, health, etc. England’s fastest population growth ever was in 1826 just as Wrigley’s Advanced Organic Economy was being supplanted by his Mineral-Based Energy Economy. America’s fastest population growth after frontiers were fixed occurred in 1909, shortly after oil discoveries first in Pennsylvania (1859) and then Texas (1887).
It must be stressed that all of the above is merely hypothetical. Very little account has been taking of many variables - energy intensity, energy efficiency etc. Many figures are hypothetical – for example, limits of coal consumption could differ widely from the chosen value of 2.8 Gtoe.
Nuclear, hydroelectricity and renewables have not figure in this analysis because their contributions to the global energy mix are relatively minor. But they too must contribute somehow to population growth.
The main purpose of the paper has been to try and thrust the issue of energy into demography’s limelight. Energy is an issue that has been widely ignored when attempting to explain historical demography and it is widely ignored when attempting to project future demographic scenarios. Yet I hope this paper has shown that neither the past nor the future of demography can be adequately explained without also examining energy’s role (not economics’!) in the rise (and fall) of populations.
Anderson, M. 1996. British Population History: From the Black Death to the Present Day. Cambridge University Press.
Boserup, Ester. 1981. Population and Technology, Basil Blackwell.
Boserup, Ester. 1990. Economic and Demographic Relationships in Development, The John Hopkins University Press.
BP (BP Amoco). 2000. BP Amoco Statistical Review of World Energy, various editions 1979-2000.
Campbell, Colin. 1988. The Coming Oil Crisis. Multi-Science Publishing Company.
Catton, William. 1980. Overshoot: The Ecological Basis of Revolutionary Change. University of Illinois Press.
Cohen, Joel E. 1995. How Many People Can the Earth Support? W. W. Norton and Co.
Hackett-Fischer, D. 1996. The Great Wave. Oxford University Press.
IEA (International Energy Administration). 1999. World Energy Outlook.
Jackson, Tim. 1992. “Renewable Energy: Summary paper for the Renewable Series”, Energy Policy, September 1992.
Jenkins, G. 1989. Oil Economists’ Handbook, 5th Edition.
Landes, D. 1998. The Wealth and Poverty of Nations. W. W. Norton and Co.
Laherrere, J.H. 1999. “World oil supply-what does up must come down, but when will it peak?”, Oil and Gas Journal, Feb. 1, 1999, pages 57-64.
Lee, R. D. 1987. “Population dynamics of humans and other animals”, Demography 24:443-65.
Leslie, Jacques. 2000. “Running Dry: What happens when the world no longer has enough freshwater?”, Harper’s Magazine, July 2000, pages 37-52.
Livi-Bacci, M. 1997. A Concise History of World Population (Second Edition). Blackwell Publishers.
McEvedy, C. and R. Jones. 1978. Atlas of World Population History. Penguin Books.
Nakicenovic, N. 1987. “Technological substitution and long waves in the USA” in T. Vasko (ed.) The Long Wave Debate pages 76-104. Springer-Verlag.
Perlin, J. 1989. A Forest Journey: The role of wood in the development of civilisation. W.W. Norton & Company.
Schurr, S. H. and B. C. Netschert. 1960. Energy in the American Economy 1850-1975. John Hopkins Press.
Smil, V. 1994. Energy in World History. Westview Press.
Tainter, Joseph A. 1988. The Collapse of Complex Societies. Cambridge University Press.
Tranter, N. L. 1996, British Population in the Twentieth Century. MacMillan Press Ltd.
United States Bureau of the Census. 1960. Historical Statistics of the United States: Colonial Times to 1957.
United States Bureau of the Census. various editions. Statistical Abstract of the United States.
Watt, K.E.F and Colin Campbell. 2000. “World Population, Past and Future”, unpublished.
WEC (World Energy Council) 1995. Global Energy Perspectives to 2050 and Beyond. WEC, London.
Woods, J. W. 1998. “A theory of preindustrial population dynamics: demography, economy, and well-being in Malthusian systems”, Current Anthropology 39:99-133.
Wrigley, E. A. 1988. Continuity, chance and change: the character of the industrial revolution in England. Cambridge University Press.
Wrigley, E. A. 1994. “The classical economists, the stationary state, and the Industrial Revolution” in Snooks, ed., Was the Industrial Revolution Necessary? pages 27-42. Routledge.
Wrigley, E. A. and R. S. Schofield. 1981. The Population History of England 1541-1871. Edward Arnold.
Zabel, Graham. 2000. “U.S. Bureau of the Census population projections: Are they getting any better?”. Unpublished.
 a term borrowed from evolutionary biology, and coined by well-known scientist Steven Jay Gould who argues that evolution proceeds dramatically in short bursts of geological time rather than at a constant rate.
 Boserup, Population and Technology, p. 3.
 Boserup, Population and Technology, p. 46.
 Boserup, Population and Technology, p. 109.
 Smil, Energy in world history.
 See for example Boserup, Population and Technology, p. 125 or Cohen, How Many People Can the Earth Support?, p. 42.
 BP Amoco, Statistical Review of World Energy 2000, p. 30.
 Livi-Bacci, A Concise History of World Population. p. 147.
 Smil, Energy in World History, p. 253
 Cohen, How many people can the Earth support?, p. 91.
 Wrigley, “The classical economists, the stationary state, and the Industrial Revolution”, p. 27-28.
 The concept of energy eras, or energy long cycles (of approximately 50 years and sometimes related to Kondratieff cycles) has been noted by several observers. Nakicenovic (1987) lists the “age of canals” (1773-1840), the “age of railroads” (1840-1895), the “age of electricity” (1895-1945), the “age of oil” (1945-1995) and predicts a new energy era starting in the 1990s and suggests that natural gas will be the best candidate. In this case the first three “ages” are all based on the consumption of coal. See also Smil, Energy in World History, p. 240-241.
 Schurr and Netschert, Energy in the American Economy 1850-1975, p. 511.
 see Wrigley’s Continuity, chance and change.
 for a very cogently argued explanation of these fluctuations, see Hackett-Fischer, The Great Wave, p. 246-249.
 Landes, The Wealth and Poverty of Nations, pg. 187.
 quote from Perlin, A Forest Journey, p. 166.
 many contemporary English observers expressed great concern at the decimation of the forests. For an excellent account see Perlin’s A Forest Journey, Chapter 10.
 quote from Perlin, A Forest Journey, p. 186.
 Boserup, Economic and Demographic Relationships in Development, p. 35.
 According to extensive research carried out by the Cambridge Historical Demography group. See Wrigley and Schofield, The Population History of England 1541-1871.
 Wrigley, Continuity, change and change, p. 13.
 Wrigley, Continuity, Chance and Change, p. 54.
 Wrigley, Continuity, Chance and Change, p. 54.
 Schurr and Netschert, Energy in the American Economy, 1850-1975, p. 511
 highest growth occurred in 1909. Calculated from population figures in U.S. Bureau of the Census’ Historical Statistics of the United States:Colonial Times to 1957 and subsequently for the U.S. Bureau of the Census’ on-line International Database.
 Hackett-Fischer, The Great Wave, p. 125.
 Wrigley, Continuity, Chance and Change, p. 90.
 derived from McEvedy and Jones, Atlas of World Population History, p. 342.
 BP Amoco Statistical Review of World Energy, June 2000, page 30.
 Boserup, Population and Technology, p.70.
 Leslie, “Running Dry”, p. 40.
 Zabel, “U.S. Bureau of the Census Population Projections: Are they getting any better?”. Figures 1 and 2 on page 3 show that since 1975, future world population projections have been consistently too high and over time they show a marked downward trend. “It appears that the magnitude of the slowing of the world’s population growth rate has taken forecasters by surprise and continues to do so.” (p. 3).
 WEC, Global Energy Perspectives to 2050 and Beyond, p. C1.
 IEA, World Energy Outlook. Both table 7.12 (page 101) and table 7.18 (page 117) show a projected difference between oil demand (111.5 million barrels/day) and supply (92.3 million barrels/day) of 19.1 million barrels/day by 2020 which they account for as Unidentified Unconventional Oil – oil from currently unknown or uncertain projects.
 see Laherrere, “World Oil Supply-what goes up must come down, but when will it peak?”.
 85 percent of the cash cost of producing ammonia comes from natural gas
 Jackson, “Renewable Energy: Summary paper for the Renewable Series”, p. 867.
 Stewart L. Udall. 1980. in the forward to Catton, Overshoot: The Ecological Basis of Revolutionary Change, p. xv.
 Catton, Overshoot: The Ecological Basis of Revolutionary Change. p. 29-30.