There is no elephant in the room

This brief study estimates the relative strengths of population and economic growth as two key drivers of total environmental pressure. In the imagined absence of planetary limits to growth, aggregate consumption in the global South catches up with consumption in the global North by 2073—adding up to almost 29 times today’s environmental pressure. However, this pressure is primarily driven by economic growth per capita, which exceeds by a factor 4.6 the effect of population growth.

Since ‘The Population Bomb’ (Ehrlich, 1968), many continue to blame environmental scarcity on overpopulation. Since ‘The Limits of Growth’ (Meadows et al., 1972), many have added that the culprit is not just overpopulation, but also ‘its terrible twin – overconsumption’ (Kopnina and Washington, 2016: 140; Bradshaw et al., 2021; Czech, 2021). The current environmental crisis supposedly escalates as a result of the impact of a growing number of poor taking on the same habits of resource exploitation as the rich. Growing levels of biophysical throughput, especially in highly populated low and lower-middle income (LLMI) countries are regarded as a writing on the wall. For example, looking at the total Ecological Footprint (EF) per national income band, LLMI countries were responsible for 33% of the total EF in 2017, compared to 29% for high income (HI) countries. LLMI countries have also increased their EF by 345% compared to 116% for HI countries (Table 1).

National income bands (based on World Bank categories)Ecological Footprint of Consumption (global hectares) in 2017Ecological Footprint of Consumption (as % of total) in 2017Change in Ecological Footprint of Consumption (as % increase) from 1961 to 2017Population in 2017Ecological Footprint of Consumption (global hectares per person) in 2017
High income73501862282911612250310386.0
Upper middle income98622301613927228177800463.5
Low & lower-middle income83190321483334534662633952.4
Table 1. Ecological Footprint (EF) per national income band (based on GFN, 2021 and WB, 2021).[i]

The argument suffers from several flaws and limitations. First, with differences in population size, the per capita EF is higher in HI countries with 6.0 global hectares per person compared to 2.4 global hectares per person in LLMI countries. Second, the rising throughput in LLMI results largely from industrial and agricultural production to supply a disproportionate amount of goods to HI countries—and to high-income areas within both HI and LLMI countries. ‘[C]onsumption of affluent households worldwide is by far the strongest determinant and the strongest accelerator of increases of global environmental and social impacts’ (Wiedmann et al., 2020: 1). This leads to an outsourcing of ecological impacts into LLMI countries and their own peripheries (Akizu-Gardoki et al., 2020; Rammelt and Gupta, 2021). EF data such as in Table 1 goes a long way but fails to account for the full range of those externalities (Van den Bergh and Grazi, 2014) which means that the allocation of resources, risks and responsibilities is likely to be even more unequal. Finally, the argument conflates the processes of population and economic growth (Hartmann 1998). The present paper therefore aims to provide a straightforward estimate of the relative strength of these drivers of the environmental crisis.

The drivers of population and consumption

The average population growth rate in the LLMI countries was 1.79% (over the period 2002-2017), which is higher than the average 0.63% rate in the HI countries (Table 2). In 2017, LLMI countries also accounted for 46% of the global population; HI countries for only 16%. Not only is population relatively greater in LLMI, so is the population growth rate.

Consumption will be measured as Gross National Income (GNI) in constant 2010$, which accounts for total income regardless of citizens and businesses are located, and for inflation to express growth in ‘real’ terms. The average economic growth rate (GNI per capita) in LLMI countries was 3.97% (2002-2017), which is also higher than the average 1.22% rate in HI countries. However, the absolute levels of consumption reveal something different. In 2017, LLMI countries accounted for only 8% of total consumption; HI countries for 65%.

We therefore see an inverse relationship between population and consumption levels for these two income bands. In 2017, an average ‘rich’ person claimed 24 times more than an average ‘poor’ person [=(65/16)/(8/46)].

National income bands (based on World Bank categories)Average % growth population 2002-2017Total population in 2017% of total population in 2017Average % growth GNI per capita 2002-2017GNI (constant 2010 US$) in 2017GNI (% of total) in 2017
High income0.63%1.23E+0916%1.22%5.22E+1365%
Upper middle income0.80%2.82E+0938%4.96%2.21E+1327%
Low & lower-middle income1.79%3.47E+0946%3.97%6.16E+128%
TOTAL 7.51E+09100% 8.05E+13100%
Table 2. Changes and levels of population and consumption (based on WB, 2021).

Comparing the drivers’ relative strengths

We see that population and economic growth rates in LLMI exceed those in HI countries. At some point, and in the imagined absence of planetary limits to growth, aggregate LLMI consumption therefore catches up with aggregate HI consumption. A relatively simple back-of-the-envelope calculation gives us the year when this happens. I assume a continuation of past population and economic growth rates from Table 2. I use these rates to extrapolate the growth of total population and average consumption for all three income bands. For each year after 2017, GNI per capita is multiplied by population size to get the projected consumption level. I find that aggregate consumption by LLMI countries catches up with consumption by HI countries by 2073 (Figure 1).

Figure 1. GNI growth for all three income bands, actual (until 2017) and projected (2017 onwards) (based on World Bank 2021).

So, which one is the main driver: population or economic growth? Both are contributing to LLMI catching up with HI by 2073, but not with an equal weight. Since LLMI comprise almost half of the world population and its population growth rate is almost three times that of HI, one might assume that population is the culprit. This would be incorrect.

Let’s take 2017 as a base year with:

            consumption = population × GNI/capita = 100

After 56 years (2017–2073):

            consumption = 100 × (1+0.0179)^56 × (1+0.0397)^56 = 100 × 2.7001 × 8.8481 = 2389.6641

The population factor therefore contributes to aggregate consumption growth by 170.01% (= 270.01%-100%). On the other hand, the average consumption factor contributes by 784.81% (= 884.81%-100%)—4.6 times the contribution from population growth. To this conclusion, I should immediately add that GNI per capita levels hide further inequalities within the income bands, between countries and between individuals (UNU-WIDER, 2021). For example, from the early 1990s to the late 2000s, household income inequality as measured by the population-weighted average level of the Gini index increased by 9% for HI countries and by 11% for LLMI countries (UNDP, 2013). This further undermines attributions of the environmental crisis to the consumption of a large and growing number of ‘poor’ people. Of course, the proposed calculation does not account for the actual biophysical claims and impacts of consumption, which I have measured using the monetary index of GNI. However, throughput and GNI are strongly correlated (Rammelt and Gupta, 2021). Based on Ecological Footprints (EF) rather than GNI, others have also concluded that consumption is the main driver, not population (Galli et al., 2012; Toth and Szigeti, 2016).

Finally, let me propose the following thought-experiment as another way to approach this: say average consumption magically stopped growing from now on (i.e., the GNI per capita growth rate drops to zero after 2017, and stays at that level), aggregate consumption by a relatively faster growing population in LLMI countries would catch up with HI consumption only by 2198. Relative differences in the population growth rates of the two income bands are therefore virtually irrelevant for the ecological crisis in the coming decades. The projected cut-off point would probably occur even later with a more realistic slowing down of population growth rates (Vollset et al., 2020). Similarly, if populations stopped growing from now on (zero growth rate), LLMI consumption would still catch up with HI consumption by 2093 (20 years later than with population growth). Again, per capita consumption growth drives the crisis.


Through a relatively straightforward analysis, this study supports the finding that the main driver for environmental disruption is not population growth, but average consumption growth—especially in HI countries, and to a lesser extent in LLMI countries (and to an even lesser extent by the peripheries in those countries). Commentators often refer to overpopulation as the ‘elephant in the room’ that scholars tend to ignore, or cannot see (Kopnina and Washington, 2016; Czech, 2021). However, as far as global environmental impacts are concerned, it seems we are not dealing with an elephant but with something smaller. Moreover, a focus on overpopulation, or even on the conflated forces of overpopulation plus overconsumption, brings about a not-so-subtle shift of responsibility for environmental problems to the global South (Norton, 2000; Rammelt and Boes, 2013; Fletcher et al., 2014). Resources are not primarily consumed by a fast-growing population in the global South for the satisfaction of basic needs, but by a slow-growing population in the global North for the satisfaction of boundless demands: ‘Instead of a Population Bomb we should speak about an Over-Consumption Detonator of environmental disaster’ (Toth and Szigeti, 2016: 290).


Akizu-Gardoki O, Wakiyama T, Wiedmann T, Bueno G, Arto I, Lenzen M et al. (2020) Hidden energy flow indicator to reflect the outsourced energy requirements of countries. Journal of Cleaner Production 278: 123827.

Bradshaw CJA, Ehrlich PR, Beattie A, Ceballos G, Crist E, Diamond J. et al. (2021) Underestimating the challenges of avoiding a ghastly future. Frontiers in Conservation Science 1(615419): 1-9.

Czech B (2021) Population Growth: The Ironic Vexer. Available at:

Ehrlich PR (1968) The population bomb. New York: Ballantine Books.

Fletcher R, Breitling J and Puleo V (2014) Barbarian hordes: the overpopulation scapegoat in international development discourse. Third World Quarterly 35(7): 1195-1215.

Galli A, Kitzes J, Niccolucci V, Wackernagel M, Wada Y and Marchettini N (2012) Assessing the global environmental consequences of economic growth through the ecological footprint: a focus on China and India. Ecological Indicators 17: 99-107.

Global Footprint Network (GFN) (2021). National Footprint and Biocapacity Accounts, 2021 Edition. Available at:

Hartmann B (1998) Population, environment and security: a new trinity. Environment and urbanization 10(2): 113-128.

Kopnina H and Washington H (2016) Discussing why population growth is still ignored or denied. Chinese Journal of Population Resources and Environment 14(2): 133-143.

Meadows DH, Meadows DL, Randers J and Behrens WW (1972) The Limits to Growth. London: Earth Island.

Norton BG (2000). Population and consumption: Environmental problems as problems of scale. Ethics & the Environment 5(1): 23-45.

Rammelt CF and Boes J (2013) Galtung meets Daly: A framework for addressing inequity in ecological economics. Ecological economics 93: 269-277.

Rammelt CF and Gupta J (2021) Inclusive is not an adjective, it transforms development: A post-growth interpretation of Inclusive Development. Environmental Science & Policy 124: 144-155.

Toth G and Szigeti C (2016) The historical ecological footprint: From over-population to over-consumption. Ecological Indicators 60: 283-291.

United Nations Development Programme (UNDP) (2013). Humanity Divided: Confronting Inequality in Developing Countries. UNDP: New York

United Nations University World Institute for Development Economics Research (UNU-WIDER) (2021). World Income Inequality Database – WIID. Available at:

Van den Bergh JCJM and Grazi F (2014) Ecological footprint policy? Land use as an environmental indicator. Journal of Industrial Ecology 18(1): 10-19.

Vollset SE, Goren E, Yuan C-W, Cao J, Smith AE, Hsiao T et al. (2020) Fertility, mortality, migration, and population scenarios for 195 countries and territories from 2017 to 2100: a forecasting analysis for the Global Burden of Disease Study. The Lancet 396(10258): 1285-1306.

Wiedmann T, Lenzen M, Keyßer LT and Steinberger JK (2020) Scientists’ warning on affluence. Nature communications 11(1): 1-10.

World Bank (WB) (2021). World Development Indicators. Available at:

[i] Table 1 also shows data for upper-middle income countries for the sake of completeness, but I will simplify the argument by leaving this category out.

Blogs Talks

A Degrowth Movement in The Netherlands

During the 8th international degrowth conference in The Hague next summer we will pay attention to the potential for a degrowth movement in the Netherlands. In this Key Conversation we will seek to facilitate a space for exchange between initiatives (on food, housing, energy, mobility and money) that are aligned with degrowth principles, to generate momentum for degrowth thinking, practice and political mobilisation in The Netherlands. Here’s a short video message about our aims for this Key Conversation. For more details, please see below.

The emphasis for this Key Conversation is on networking and generating momentum for degrowth thinking and practice in The Netherlands. We plan to approach and invite existing initiatives and networks (local, provincial, national, and in different sectors) that possibly share and (implicitly) pursue degrowth principles, such as redistributing, caring, de-commodifying and commoning. We will seek these potential alliances in different sectors or spheres of work (food, energy, mobility, housing, money, etc.). Initiatives and networks would be invited to (1) discuss if and how they adopt/pursue these principles, in order to generate common ground, and (2) explore shared experiences of the challenges of a transition away from current growth-based social, cultural and economic forces. Our observation is that there is a myriad of relevant initiatives, but that these initiatives are not always able to find each other or work within their respective silos. We think that by opening a space for these initiatives to meet and connect under the umbrella of degrowth, we can foster exchange, learning and collaboration between them to acknowledge their common struggles, consider alternatives solutions and increase their collective power and impact. To contribute to these endeavors, we have come up with a range of possible activities, for which we will collaborate with the Dutch Degrowth platform Ontgroei. Activities could include:

  • Constructing a simple overview of degrowth objectives, principles and policies to help interested initiatives to recognize degrowth in their activities . 
  • Building a database of initiatives and networks in the Netherlands that could potentially participate and/or help organise the conference, as well as in long-term alliances within a wider degrowth movement.
  • A pre-conference session in early 2021 to start discussing degrowth with interested initiatives/networks (per sector).
  • Further recurring and informal networking opportunities throughout the conference rather than one single event, e.g., through ‘Dialogue Walks’ to discuss shared visions and challenges (experiences would be gathered through video/audio fragments), and/or through local/decentral meeting places outside of The Hague.
  • Thematic sessions around different degrowth principles, and then scalar or sectoral break-out sessions . Initiatives would be encouraged to contribute around relevant themes, not just to participate.
  • Create a collaborative map of the initiatives and networks.
  • Keynote, e.g., in the form of a carousel-type event (5 or 6 ‘portraits’ of initiatives in a sequence of short dialogues, facilitated by a moderator).

We see these activities as stepping stones towards building a degrowth movement in the Netherlands. Post-conference follow-ups include:

  • Create a community of practice page on, with quotes from the initiatives on “degrowth”, or as signatories to a degrowth manifest.
  • Digitalise and share the collaborative map (see above).
  • And much more… Stay tuned via Ontgroei, the Dutch degrowth platform.

Degrowth in the Netherlands: An Overview

Through a degrowth-course at the University of Amsterdam and meetings with like-minded people, the Dutch degrowth-platform Ontgroei was established. Co-founder Crelis Rammelt reports about the state of degrowth thinking and its movement in the Netherlands.

“Just act normal, as that’s insane enough as it is,” we tell each other here in the Netherlands. We are quick to judge those who do not follow normal behaviour. But when normalcy has led us to social injustice and environmental destruction, it would be insane not to act abnormal!“ To understand the state of de-/post-growth thinking and its movement in the Netherlands, we need a bit of context. Our dominant social imaginary has long been firmly rooted in socially unjust and ecologically destructive growth fetishism. Read more…


Strong correlations between economic growth and environmental impacts

Perpetual economic growth is incompatible with a finite planet. To some extent, innovation can reduce throughput, thus ‘decoupling’ economic growth from environmental impacts. However, the level of decoupling has been ‘relative’, which means that impacts still grow, just not as fast as the Gross World Product (GWP) (Figure 1).

Figure 1. Gross World Product (GWP), World primary energy consumption and World consumption emissions (based on WB 2020; BP 2020; GCA 2020).

An indication of a strong coupling between economic growth and environmental impacts is suggested by a strong long-term correlation between the two (Figure 2). The GWP growth rate is strongly correlated with both world primary energy consumption growth (Pearson correlation coefficient: 0,88) and world consumption emissions growth (0,75).

Figure 2. Growth rates of GWP, primary energy consumption and consumption emissions (based on WB 2020; BP 2020; GCA 2020).


British Petroleum (BP) (2020). Statistical Review of World Energy. Retrieved 21 Aug, 2020, from

Global Carbon Atlas (GCA) (2020). CO2 Emissions. Retrieved 21 Aug, 2020, from

World Bank (WB) (2020). World Development Indicators. Retrieved 19 Aug, 2020, from


Is degrowth about bringing down GDP?

In a thought-provoking piece in CASSE’s Steady State Herald, Brian Czech worries that many degrowthers seem agnostic about GDP degrowth. But should degrowth be about bringing down GDP? My answer: not primarily. We had an interesting exchange of ideas on CASSE’s website (Below, I have tried to summarise some of the points. For full text see here).

Me: GDP is just a blunt aggregate weighing scale. You can make yourself sick by eating too much apple pie, but it also matters whether the pie contains too much dough and too few apples: a pie of the right size can still be unhealthy. Degrowth efforts are likely to bring down GDP because socially and ecologically undesirable economic activities that count towards GDP are deliberately replaced by socially and ecologically desirable economic activities that don’t count towards GDP. In other words, not everything that can be counted counts, and not everything that counts can be counted. Other activities that currently count towards GDP, e.g., in health care, education, welfare and other public services, should be preserved—like apples in an apple pie. Degrowth is therefore not primarily about bringing down GDP; it’s about downscaling what is unhealthy, unfair and unsafe. GDP-degrowth cannot guide us there.

Brian: … GDP growth with healthy and happy services proliferating everywhere would still entail an expanding trophic base … In seafaring terms, when she’s loaded to the Plimsoll line it doesn’t matter if you add a puppy or a skunk; either way your ship is sunk! …

Me: I agree with you that ‘healthy and happy’ services also contribute to throughput, and therefore to our ‘sinking ship’. But I’m still not 100% convinced about GDP-degrowth, because it lumps together what should be downscaled and what should be preserved. Another metaphor: just as GDP growth is sailing by a wrong compass, so too could GDP-degrowth lead us astray. So for me, we are left not with GDP-degrowth, but with biophysical degrowth towards a steady-state (AND the ‘degrowth’ efforts I have noted).

Brian: I am very tempted to say, acknowledging but the merest tinge of exaggeration, that there is nothing more biophysical than GDP! …

Me: The growth rates of GDP and material throughput are indeed very strongly correlated. No disagreement with you there.
However, we are primarily concerned with lowering material throughput, not the indicator of GDP per se. Material throughput must shrink, which will undoubtedly bring down GDP. But we should not put the cart before the horse. For me, as a Degrowth scholar, the matter of how we shrink material throughput is key, and GDP-degrowth as a policy goal doesn’t help—especially as we begin to partly decommodify economic life, reclaim the commons and work towards other ambitions linked to the degrowth movement.
I already referred to preserving (or even growing) essential basic services (more apples and less dough in the cake). Adding to that, it would be beneficial if the material standards of living of the ‘poor’ go up (on a side note: this type of growth is not equivalent to GDP growth in the global South, which tends to happen at the expense of the poor). In a world of limits, growth of throughput by the ‘under-consumers’ demands more degrowth of the throughput by the ‘over-consumers’.
These are just some examples of why we need more targeted degrowth strategies (as opposed to GDP-degrowth in the aggregate). Of course, in the aggregate, we also need to make sure that we downscale to and remain in a safe steady state. But I doubt that World GDP is the best indicator to help us make that assessment (especially if we start to transform the economy as mentioned above). To track our aggregate biophysical impacts, we might want to look at the planetary boundaries framework, for example.

Brian: I don’t view the choice as either/or. I am all for targeted throughput caps, starting with oil at the wellhead… The main point of my article is that it is misleading and counterproductive to call for degrowth while wavering on the implications for GDP. When we cap oil at the wellhead, we will (ceteris paribus) be lowering the rate of GDP growth… On the other side of that exact same coin, … efforts to lower the rate of GDP growth … will absolutely lower throughput. Take just three examples:
– Removing tax code incentives for having more children than two.
– Sectoral salary caps (see today’s Steady State Herald (later today)).
– Keeping the Federal Reserve (or whatever monetary authority) out of the growth game and purely onto its original mission of fighting inflation.
These and many other fiscal and monetary reforms are throughput droppers as surely as wellhead capping, and with crystal clarity of intent—vis-à-vis GDP—for all to see.

Crelis: Thank you for this clarification Brian. I fully agree with you on the first point. It would indeed be counterproductive not to recognise that a reduction in throughput brings down GDP. I believe we are on the same page there. For me, the discussion was more about the second point, regarding the policy focus on GDP-degrowth.
There is a difference between saying “we need to reduce GDP” and “we need to understand that GDP will be reduced as a consequence of all this targeted degrowth stuff we‘ve been talking about.” It seems to me that the monetary/fiscal reforms you mentioned intend to do much more than to lower the indicator of GDP (even if that is going to be one of the outcomes). But maybe I am splitting hairs.
In any case, your article is in part about why some degrowthers are not engaging with a GDP-degrowth policy, or failing to take a clear stance. While I don’t speak for them of course, I was trying to provide examples of some of the ambitions in the degrowth movement for which GDP-degrowth doesn’t provide a very useful compass. At the same time, I do see the importance of debunking the GDP fetish. So of course it is not an either/or choice. We need to tackle the problem on many different fronts. But that is exactly the point I think; there are different fronts.


Recovering from COVID-19? Let’s do it without GDP growth

COVID-19 has infected both our bodies and economies. On March 12, Mark Rutte compared the Dutch economy to ‘a patient’ requiring treatment. The next day, his government was ready with a rescue plan for major firms, such as Air France-KLM and Schiphol. Just as our bodies need oxygen, the economy needs money – so the theory goes. Other European governments are also preparing to inject credit, and a lot of it, to ensure their economies’ survival.  Yet, this cure is limited. It presupposes that our economies need growth in order to flourish. If we do not steer away from this limited paradigm, then the cure we apply now will become a liability by giving rise to potential crises in the future. Read more…


This isn’t the type of downscaling that degrowth thinkers have in mind!

In the early 17th century, the bubonic plague is said to have played a crucial role in popping the tulip bubble in the Netherlands. Today, the coronavirus (COVID-19) is leading not only to a health crisis, but also an economic one. The outbreak is sparking realistic fears of a deep global downturn. Our globalised, just-in-time, cost-cutting, risk-taking and profit-maximising economy has shown a rather limited ability to absorb shocks. In a time of crisis, the instability and fragility, but also the inequality of the economic system becomes painfully obvious. Read more…


Capitalism inevitably leads to crises

Capitalism typically goes through cycles of expansion to contraction. We commonly refer to these as ‘business cycles’ or ‘economic cycles’. Every now and then, however, these cycles go off the hinges. They become unstable and can lead to recessions, crises and depressions. This inherent instability of the capitalist system cannot be explained by standard economic models. Instead, those models blame instability on excessive regulation, government interventions or other factors outside the market-economic system. Some unorthodox economists see it differently. Steve Keen, for example, developed an alternative kind of economic model—one that can mimic instability. The model provides compelling and urgent insights into how economic crises arise from within the structures of capitalism. Another crisis seems inevitable as long as the causes are misdiagnosed. Read more…


Good infinity

Cycles of growth and decline in ecology were famously revealed in so-called predator-prey relationships. As foxes eat chickens, chicken reproduction drops, which leaves fewer chickens for foxes to eat. Consequently, the fox population drops, which then allows the chicken population to rebuild. As its food source becomes available again, the fox population also rebuilds and the cycle starts again. Read more…


Decoupling: a key fantasy of the SDG agenda

At the core of the new Sustainable Development Goals (SDGs) is the idea that economic growth (defined as money flow or market value) can be ‘decoupled’ from the physical growth of the economy (resource consumption) and the associated environmental pressures (degradation, pollution). Paradoxically, however, the concept’s main promoters admit that there is virtually no evidence that decoupling works, that the conceptual basis for it is weak, and that even if it were possible it is not politically feasible. Decoupling is, thus, a dangerous fantasy sustained by disavowal – the simultaneous admission and denial – of its impossibility in practice. Read more…