How are rising CO2 emissions linked to a rising world population?

Many discussions on sustainability rarely mention the world’s growing population and whether current or projected future levels are – or can be made – compatible with living within the limits set by the Earth’s regenerative capacity. As far as fossil fuel emissions are concerned, however, the growing population is not incompatible with lower levels of energy use but, as David Knight‘s paper below shows, the rising levels of consumption in rich countries and “emerging” ones like Brazil, India and China certainly are.

—————
Population, Wealth, and Climate Change
by David Knight

Introduction

According to the International Programs Center of the US Census Bureau, the world population was projected to be 6.913 billion at 17 April 2011[1]. Current projections show a continued increase of population, but a steady decline in the population growth rate, with the population expected to reach between 7.5 and 10.5 billion by the year 2050[2]. Although this increase is a cause for concern, these figures are only projected increases and are far from set in stone. The actual size of the population in 2050 will be influenced by a combination of socioeconomic factors, the behaviour of individuals, and to a much lesser extent, national population policies. More importantly unforeseen factors such as widespread food and water shortages, pandemics, and the effects of Peak Oil could result in a considerable undershoot on the projected figures. This article discusses the effect of population growth rate, population size and wealth on total energy use, greenhouse gas emissions and environmental impact. [3] It also considers some of the relationships between population growth, wealth, peak oil and climate change.

In summary it provides evidence that rapid population growth in the least developed countries has not contributed significantly to increased atmospheric concentrations of greenhouse gases (GHG).

It argues that global population growth is only a secondary cause of increased GHG emissions and that the primary cause is an increase in wealth in the developed countries and those with recently advanced economic development (Brazil, Russia, India and China, known collectively as the BRIC countries). The paper also attempts to draw out the implications of its analysis for population and climate change policies.

How does population growth influence GHG, emission rates, cumulative emissions and environmental impact?

Figure 1. Growth in population and CO2-equivalent emissions (production figures) over the period 1980–2005 by groups of nations classified according to their average per capita income levels. This illustrates the lack of correlation between population growth and emissions growth. Source: Satterthwaite, D., 2009 The implications of population growth and urbanization for climate change. Environ Urban 2009; 21: 545–67.

Figure 2. Contribution to the growth in population and CO2-equivalent emissions 1980–2005 by the nations with the largest emissions growth. Source: As Figure 1.

Figure 3. Cumulative CO2 emissions (megatonnes) derived from fossil fuels 1850–2005 and from land use change 1950–2000 by country. Global top 20 (left) and bottom 20 (right) nations. Yellow marked rows are developed (Kyoto Annex 1) countries. Source: cait.wri.org/ cait.php?page=cumul& mode=view

Figure 1 enables the growth in population and GHG emissions (expressed as CO2 equivalent emissions) to be compared over the period 1980-2005 for countries grouped according to their income per capita. There is clearly no relationship between rate of population growth and rate of emissions growth over this period. Similarly Figure 2 shows that there was no direct relationship between the rate of population growth and emissions growth for the five countries with the highest emissions growth rates. Taken together, Figures 1 and 2 provide strong evidence that rapid population growth is not the principal cause of cause of growth in emissions since 1980.

A lack of relationship can also be seen when population growth rates are compared with cumulative emissions (1850 to 2000) (Figure 3) rather than emission rates (Figures 1 and 2). It is clear from Figure 3 that the low income, high fertility countries forming the bottom 20 emitters have not made significant contributions to cumulative CO2 emissions compared with the top 20 emitters, many of which have low population growth rates. This is not surprising as the size of cumulative emissions chiefly depends on history of industrialisation and economic development, processes that are long established in the top 20 emitters but have not even begun in the bottom 20. CO2 is in effect a cumulative poison in both the atmosphere and sea because it is only very slowly removed by geological processes. It follows that cumulative emissions rather than current emission rates should be used to compare the relative culpability of different countries for today’s atmospheric CO2 concentrations and CO2 –induced marine acidification. Although to date most of the cumulative emissions have been made by developed countries they will soon be overtaken by the BRIC countries in this respect if current trends continue.

Further evidence that population growth rate and population size are not the key determinants of environmental impact can be seen from the work of Bradshaw et al. (2010). [4] These authors ranked 228 countries for their population growth rate and for their proportional environmental impact, presenting the results in a scatter plot (Figure 4 c). This shows a weak negative correlation rather than the strong positive one expected if population growth rate was determining environmental impact. The weak negative correlation may stem from the negative effect of high population growth rate on economic development. When these authors ranked the 228 countries by total population size instead of population growth rate and plotted this against environmental impact rank there was only a weak positive correlation (Figure 4 a) indicating that that population size is not the key determinant of environmental impact. The stronger positive correlation for population density (Figure 4 b) is probably explained by the close relationship between urbanisation and high population density and the heavy impact of urbanisation on the environment.

Taken together Figures 1 to 4 strongly indicate that population growth rate does not strongly correlate with, and is therefore not the principle cause of either cumulative GHG emissions , or the recent increase in emissions or Proportional Environmental Impact.

So if the population growth isn’t the primary cause of cumulative emissions and high current emission rates, what is?

Here is the evidence that the primary cause of cumulative emissions and high current emission rates is wealth-led consumption[5]. Figure 5, a log/log scatter diagram, shows that for low to moderate income countries (GDP’s less than 20,000$) there is a close relationship between their total national GDP and total energy use. The relationship breaks down at high income levels (not shown) largely because cheap oil prices in oil exporting countries give rise to exceptionally high energy use. As would be expected, the relationship between wealth and energy use can also be seen in a graph from a different source when energy use per capita is plotted against GDP per capita (Figure 6).

Figure 5. The close correlation between total national GDP and total primary energy use. Each circle represents data for a different country, with the size of the circle proportional to the country’s population size. The graph is truncated at GDPs greater than 20,000 (see text). The reader is advised to visit www.bit.ly/lSKoG1 to see the full range of GDP’ and follow changes in energy use and GDP from 1960 to 2007 by pressing the play button. Source: http://www.gapminder.org/

Figure 6. Correlation between energy use per capita (measured in tonnes of oil equivalent) and GDP per capita (measured in US$ with equivalent purchasing power in 2005). The trajectories show how individual countries’ energy use has changed with wealth over time. Source: http://stochastictrend.blogspot.com/2009_09_01_archive.html

Figure 7. Correlation between GDP per capita and CO2 emissions (tonnes per person). Each circle represents data for a different country, with the size of the circle proportional to the country’s population size. http://www.gapminder.org/

This correlation stems from a circular relationship between energy and wealth. The availability of cheap fossil fuel has been the primary driver of wealth generation while national wealth in turn largely determines how much energy each nation uses. This relationship is also true when energy use per town or household (not shown) are compared. The close relationship between wealth and energy consumption underlies the clear correlation between CO2 emissions per capita and GDP per capita when different countries are plotted in a scatter diagram (Figure 7). Put simply, in the past and to date, wealth determines fossil fuel utilisation which in turn drives GHG emissions.

A further conclusion can be drawn from this data. The curves in Figures 5-7 show no evidence of the decrease in both energy use and emissions at high levels of national wealth that would be predicted on the basis of the Environmental Kuznets Curve theory.[6] This now largely discredited theory has been used to provide an environmental justification for continued economic growth in the rich countries. It postulates that progress to the highest levels of economic development results in reduced inequality and a concurrent switch from physical to human capital thereby reducing energy use and emissions. Not only do Figures 5-7 show that energy use and emissions fail to decrease at high levels of wealth as would be expected by Kuznets, they do not even show good evidence for a decline in the rate of increase of these parameters.

Wealth and population growth.

As wealth appears to be the primary determinant of GHG emissions, can we dismiss the impact of population growth on climate change? Certainly not, global population growth does have a secondary effect on emissions. This can be best seen from an equation known as the Kaya identity[7] developed by Japanese energy economist Yoichi Kaya. This equation describes total global carbon emissions as the product of the global population, GDP (i.e. wealth) per capita, energy intensity of wealth generation and carbon intensity of energy generation as follows:

Total anthropogenic carbon emissions = population * (global GDP / population) * (global primary energy consumption/ global GDP) * (carbon emissions/ primary energy use)

where the third and fourth terms in this expression are respectively the energy intensity of the world GDP and the carbon intensity of global primary energy use.

As we have seen above, increase in wealth in the rich and emerging countries appears to be the most significant factor behind recent emissions growth; the world’s poorest people have little impact on emissions. Thus there is a strong argument that population growth has only a secondary and weaker impact on emissions compared with economic growth. The Kaya function suggests that this secondary impact of population growth on global emissions is brought about by increasing the product of the first (population) and second (average per capita GDP) term of the identity or in other words by increasing the number and average wealth of global consumers. Thus the problem underlying growing emissions isn’t population growth per se and definitely not the rapid population growth in the poorest countries.

Although the Kaya identity strictly only applies to global emissions it can be used as a rough guide to inform emission reductions policy at the national level if allowances are made for the effects of embedded emissions in imported goods and services and their impact on the calculation of both energy intensity and carbon intensity.

The poorest countries and individuals within them are not significant consumers, but this could possibly change in the future. Fairness and pragmatic considerations require that underdeveloped countries be encouraged to develop. However this is potentially a problem from the perspective of emissions as the Kaya identity suggests that economic development measured by an increase in GDP would lead to an increase in global GHG emissions. This is because it is very unlikely that economic development can be achieved while reducing the other two terms of the Kaya function energy sufficiently to offset the impact of economic growth (and any population growth). While a reduction in energy intensity has been achieved in some developed countries by a switch from manufacturing to financial services and retailing this is not possible in the predominantly rural economies of the poorest countries. Similarly significant reductions in the carbon intensity of energy generation are unlikely as the poorest countries lack the wherewithal for a renewable energy programme and are for the most part, reliant on imported fossil fuels.

Moreover, there is recent evidence that recent economic development in some developing countries may be having an appreciable effect on global emissions. Peters et al. (2011)[8] showed that many developed countries have increased their imports from developing countries to the possible benefit of the latters economic growth but with a significant increase in global emissions. However fairness dictates that this increase in emissions should be charged to the developed countries as the increase is caused by their outsourcing of manufacturing, food production, and need for raw materials. However these emissions are currently charged to the developing countries under the Kyoto Protocol used for example to calculate Britain’s emissions.

Even though several of the wealthier underdeveloped countries have shown modest increases in GDP in the last five years there are doubts that this will continue into the future. Climate change and the peaking of oil, water, phosphate fertilizers, food and other resources threat threaten economic growth in the underdeveloped countries at least as much and possibly more in than the developed and BRIC ones. Also, many other factors including neo-colonialist exploitation, corruption, elitist dictators, military expenditure and armed conflict, and lack of good governance are likely to continue to have negative impacts on their economic development.

The leaders of the low income, high fertility countries are very aware that their countries already, and will increasingly, bear the brunt of climate change caused largely by the developed and BRIC countries. Events at the UN Climate Change Conferences in Copenhagen and Cancun make it manifestly clear that international agreement on mitigation of climate change will not be achieved unless the rich and emergent nations accede to demands of the poor countries for financial assistance to help them adapt to climate change, to limit emissions increases, and to reverse the destruction of forest carbon sinks.

The impact of population policy.

We have seen above that reductions in the rate of population growth could play a part in reducing global emissions growth by helping to reduce the number of consumers. At present the developed and BRIC countries produce practically all the global emissions so reduction in the number of consumers in these countries could in theory greatly reduce emissions. An equivalent reduction in the population growth rate in the poor countries would have a much lower impact on emissions. The latest data (March 2011) from the Population Reference Bureau [9,10] shows twenty countries currently with negative or zero population growth (Europe Ukraine, Russia, Belarus, Bulgaria, Latvia, Lithuania, Hungary, Romania, Estonia, Moldova, Croatia, Germany, Czech Republic, Japan, Poland, Slovakia, Austria, Italy, Slovenia, and Greece). The Population Reference Bureau’s predictions of future population rates do not account for changes in immigration and emigration but even when these are taken into consideration only Austria’s population is predicted to grow. All of the negative or zero population growth countries except Russia are classified as developed and all except Japan are located in Europe. This is hardly surprising as the demographic transition required for reduced population growth rates is dependent on mainly expensive socioeconomic improvements predominantly the prerogative of rich countries: reduced child mortality, widespread adult female literacy, good employment prospects for women and child care while they are at work, good public health and child welfare, and the provision of care and financial security for the elderly.

All this is encouraging from the perspective of emissions reduction but there is clearly scope for other developed countries with high current emission rates including USA, Canada, France and the Scandinavian countries to join this list. However, the aging population that accompanies the demographic transition is not without its problems.
The BRIC countries currently account for some 40% of the world population[11] and have the following population growth rates: Brazil 1.26, Russia -0.51, India 1.46, China 0.47 according to UN figures for 2005-2010.[12] There is therefore considerable scope for Brazil, Russia and India to reduce emissions by reducing the population growth rate.
However population control measures cannot be the first line of action to reduce global GHG emissions for the following reasons:

First, while China’s “One Child” policy with near compulsory abortions and selective female infanticide was effective in producing a rapid reduction in population growth, it is almost inconceivable that such policies could be introduced in the developed countries or in Brazil, Russia, India, or even in today’s China.

Second, the slowness of the removal of carbon dioxide from the atmosphere has the consequence that emissions need start to fall in the next few years and decline to practically zero by 2050. More seriously a growing body of evidence suggests that carbon negativity is probably required to give a reasonable chance of avoiding disastrous warming. However the slow response time of acceptable population policies means that they are likely to have only a small impact within this time frame.

Third, there is no easy and cheap method for reducing birth rates; the universal availability of effective and affordable artificial methods of contraception for both men and women is only a small factor in accelerating the demographic transition. Effective policies depend on far reaching and expensive socio-economic changes as outlined above.
Fourth, it follows from our argument above that population reduction has only a smaller and less predictable effect on emissions than a reduction in GDP. While it is at present unthinkable that countries will voluntarily reduce their GDP’s, Peak Oil and other resource scarcities might do this very effectively to our debt-based and currently oil-dependent economy.

Summary and Policy implications

1. Climate change is not primarily caused by population growth, least of all in the low income, high birth rate countries.
2. Economic growth fuelled by cheap fossil fuel prices, easy credit and consumerist policies has been the prime driver of increasing greenhouse gas emissions and the destruction of natural carbon dioxide sinks.
3. Little progress on emissions reductions is likely while economic growth continues.
4. Population policies in the least developed countries with high birth rates are unlikely to produce a significant reduction in global emissions.
5. Vigorous population policies in the developed and BRIC countries could play a part in reducing emissions but cannot be the first line of action as they are too slow to produce the fall in global emissions needed in the next five years.
6. Evidence discussed here does not support the Kuznet theory which justifies economic growth on the grounds that it would eventually lead to emissions reduction.
7. While it is unthinkable that countries would introduce policies to reduce economic growth , this might be achieved by Peak oil and other resource constraints within the time frame required to prevent catastrophic temperature rises.
8. Both climate change and Peak oil dictate a rapid but planned transition to a carbon-neutral or carbon-negative global society. Such a society could have considerable social and health benefits. The transition necessitates a rapid switch to low-carbon energy sources.
9. A chaotic transition rather than a planned one might have unspeakably dreadful consequences.

References and notes

1 http://www.census.gov/ipc/www/popclockworld.html
2 World Population Prospects:The 2008 Revision”. Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat. June 2009. http://www.un.org/esa/population/publications/popnews/Newsltr_87.pdf
3 This paper uses a range of different measures for environmental impact including Proportional Environmental Impact, CO2 and CO2 equivalent. This is because no single source of data shows the effect of all different socioeconomic factors of interest on CO2 equivalent emissions, the preferred measure for climate impact. The figures and tables reproduced here are based on production emissions (Kyoto protocol) and do not include consumption figures (embedded emissions). The primary sources of the data should be consulted for further information on how values used have been calculated.
4 Bradshaw, C. J. A., Giam, X., Sodhi, N. S. (2010) Evaluating the Relative Environmental Impact of Countries. PLoS ONE 5(5): e10440. doi:10.1371/journal.pone.0010440
5 Stephenson, J., Newman, K., Mayhew, S. Population dynamics and climate change: what are the links? Journal of Public Health Vol. 32, No. 2, pp. 150–156
6 http://en.wikipedia.org/wiki/Kuznets_curve
7 http://en.wikipedia.org/wiki/Kaya_identity
8 Peters, GP1, Minx, JC, CL and Edenhofer, O. , 2011. Proc. Nat. Acad. Sci. Wash. Doi 2011.1073, pnas.1006388108 Growth in emission transfers via international trade from 1990 to 2008
9 http://geography.about.com/od/populationgeography/a/zero.htm
10 http://www.prb.org/
11 http://www.investordaily.com/cps/rde/xchg/id/style/801.htm?rdeCOQ=SID-3F579BCE-819F182C
12 http://en.wikipedia.org/wiki/List_of_countries_by_population_growth_rate
DPK 25/05/2011

Featured image: Europe. Author: Lusi. Source: http://www.sxc.hu/photo/1170737 .

——————————-

Dr David Knight has parallel existences as: an academic scientist (Biophysics, Biochemistry, Ultrastructural Analysis); an activist with interests in climate change and civil and military nuclear power; an inventor of surgical devices based on silk; teacher of medical sciences and ecology; and sculptor. He is currently an Honorary Research Associate in Zoology at Oxford University and has held three visiting professorships and a visiting fellowship at MRC MBL Cambridge. He is interested in economics, theology and Jungian psychotherapy but has no expertise in these. He is has a lovely immediate family: wife, two children and five grandchildren.

Related posts:

1. Cap and Share – A fair way to cut greenhouse emissions
2. Emissions Rationing and the Oil Price Crisis
3. Feasta’s response to the European Commission’s proposal concerning emissions of greenhouse gases from aircraft
4. Slideshow on the EU Emissions Trading Scheme from the UNSCD meeting
5. Radical overhaul of emissions allocation required