Content uploaded by Jason Hickel
Author content
All content in this area was uploaded by Jason Hickel on Dec 09, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ctwq20
Third World Quarterly
ISSN: 0143-6597 (Print) 1360-2241 (Online) Journal homepage: http://www.tandfonline.com/loi/ctwq20
Is it possible to achieve a good life for all within
planetary boundaries?
Jason Hickel
To cite this article: Jason Hickel (2018): Is it possible to achieve a good life for all within planetary
boundaries?, Third World Quarterly, DOI: 10.1080/01436597.2018.1535895
To link to this article: https://doi.org/10.1080/01436597.2018.1535895
Published online: 10 Dec 2018.
Submit your article to this journal
View Crossmark data
THIRD WORLD QUARTERLY
Is it possible to achieve a good life for all within planetary
boundaries?
Jason Hickel
Department of Anthropology, Goldsmiths, University of London, London, UK
ABSTRACT
The safe and just space framework devised by Raworth calls for the
world’s nations to achieve key minimum thresholds in social welfare
while remaining within planetary boundaries. Using data on social and
biophysical indicators provided by O’Neill etal., this paper argues that
it is theoretically possible to achieve a good life for all within planetary
boundaries in poor nations by building on existing exemplary models
and by adopting fairer distributive policies. However, the additional
biophysical pressure that this entails at a global level requires that rich
nations dramatically reduce their biophysical footprints by 40–50%.
Extant empirical studies suggest that this degree of reduction is unlikely
to be achieved solely through efforts to decouple GDP growth from
environmental impact, even under highly optimistic conditions.
Therefore, for rich nations to fit within the boundaries of the safe and
just space will require that they abandon growth as a policy objective
and shift to post-capitalist economic models.
Introduction
Over the past few years there have been a number of research findings and conceptual
innovations that pose significant challenges to development theory. The first and most sig-
nificant of these is the research on planetary boundaries. Drawing on data from Earth System
science, Rockstrom etal. and Steffen etal. have identified a number of critical boundaries
that are essential to observe in order to maintain the planetary biosphere – boundaries on
climate change, biodiversity loss, ocean acidification, land-system change, nitrogen loading,
phosphorous loading, freshwater use, atmospheric aerosol loading, chemical pollution and
stratospheric ozone depletion.1 The researchers concluded that five of these boundaries
have been overshot: climate change, biodiversity loss, nitrogen loading, phosphorous load-
ing and land-system change. For two of the others (ocean acidification and freshwater use),
the process of degradation is two-thirds of the way toward the boundary, relative to pre-in-
dustrial levels. Ozone depletion is the only process that has been brought under control,
thanks to a successful campaign in the 1980s. For chemical pollution and aerosol loading,
the data are not yet robust enough to yield conclusions.
ARTICLE HISTORY
Received 05 May 2018
Accepted 25 September
2018
KEYWORDS
Sustainable development
planetary boundaries
de-growth
ecology
© 2018 Global South Ltd
CONTACT Jason Hickel j.hickel@gold.ac.uk Department of Anthropology, Goldsmiths, University of London,
London, UK
https://doi.org/10.1080/01436597.2018.1535895
2 J. HICKEL
Table 1. Social and biophysical indicators covered in the dataset provided by O’Neill et al. (2018).
Social indicator Threshold
Life Satisfaction 6.5 on 0–10 Cantril Scale (Gallup World Poll)
Healthy Life Expectancy 65 healthy life years
Nutrition 2700 kcal per person per day
Sanitation 95% with access to improved sanitation facilities
Income 95% living on more than US$1.90 per day
Access to Energy 95% with access to electricity
Education 95% enrolment in secondary school
Social Support 90% say they have relatives or friends they can depend on
Democratic Quality 0.8 on –2.5–2.5 scale, average of Worldwide Governance Indicators on
voice, accountability and political stability
Equality 70 on 0–100 scale, based on Gini coecient of household disposable
income
Employment 94% of the labour force employed
Biophysical indicator Boundary
CO2 Emissions 1.6 tonnes of CO2 per person per year
Phosphorous 0.9 kilograms P per person year
Nitrogen 8.9 kilograms N per person per year
Blue Water 574 cubic meters H20 per person per year
eHANPP 2.6 tonnes C per person per year
Ecological Footprint 1.7 global hectares (gha) per person per year
Material Footprint 7.2 tonnes per person per year
Building on this framework, Kate Raworth has argued that any vision for development
would have to somehow fit within planetary boundaries: in other words, resources should
be mobilised to improve social indicators toward certain minimum thresholds, but without
exceeding ecological limits.
2
Raworth termed this the ‘safe and just space’ and conceptualised
the objective as a matter of fitting within a ‘doughnut’, with the outer border of the doughnut
represented by planetary boundaries and the inner border represented by social foundations.
Drawing on the most popular submissions of national governments to the Rio +20 Conference
on Sustainable Development, Raworth identified 12 social priorities: health, education,
income and work, water and sanitation, energy, networks, housing, gender equality, social
equity, political voice, and peace and justice.
Building on Raworth’s intervention, a team of researchers led by Daniel O’Neill at the
University of Leeds published the ground-breaking paper ‘A good life for all within
planetary boundaries’.3 The study is the first attempt to determine whether fitting inside
the proverbial doughnut is possible, given existing relationships between social perfor-
mance and resource use for 151 nations. For each nation, the study looks at progress with
respect to 11 social thresholds, which overlap substantially with those identified by Raworth
(see Table 1).
The study also looks at each nation’s per capita resource use over seven biophysical
categories and compares these against global planetary boundaries rendered in per capita
equivalents (see Table 1).4 Five of the categories are derived directly from the planetary
boundaries framework (climate change, phosphorous loading, nitrogen loading, freshwater
use and land-use change), while two other commonly used indicators have been added:
ecological footprint and material footprint. The biophysical indicators are rendered in con-
sumption-based terms, so that the ecological impact of goods and services is attributed
to the nations in which they are consumed, regardless of where in the world they are
produced.
THIRD WORLD QUARTERLY 3
O’Neill etal. state that ‘no country meets basic needs for its citizens at a globally sus-
tainable level of resource use’.5 Indeed, the general trend shows that ‘the more social thresh-
olds a country achieves, the more biophysical boundaries it transgresses, and vice versa’.6
Many wealthy nations do well on social indicators, but significantly transgress biophysical
boundaries. Meanwhile, many poorer nations remain within biophysical boundaries, but
perform poorly on social indicators. The paper illustrates this result in a graph that plots each
country according to the number of social thresholds it achieves and the number of bio-
physical boundaries it has transgressed.7 The only countries that have achieved all of the
social thresholds have transgressed at least five of the biophysical boundaries. And the only
countries that remain entirely within all of the biophysical boundaries have achieved at most
three of the social thresholds. The most promising outlier is Vietnam, which has achieved
six of 11 social thresholds while transgressing only one biophysical boundary.
This study brings us to the cutting edge of inquiry in sustainable development. The
results are not encouraging: ‘It shows that meeting the basic needs of all people on the
planet would result in humanity transgressing multiple environmental limits, based on
current relationships between resource use and human well-being’.
8
O’Neill etal. conclude:
‘If all people are going to lead a good life within planetary boundaries, then our results
suggest that provisioning systems must be fundamentally restructured to enable basic
needs to be met at a much lower level of resource use’.9 In what follows I will build on O’Neill
etal.’s data to argue that it is theoretically possible to achieve a good life for all within
planetary boundaries in poor nations by using existing policy options. However, the safe
and just space framework requires de-growth strategies among rich nations and at an
aggregate global level.
Some nations come close to achieving a good life for all within planetary
boundaries
According to O’Neill etal.’s paper, the best-performing country is Vietnam, which has achieved
six of the social thresholds while transgressing only one biophysical boundary (CO2 emis-
sions). Vietnam falls short on life satisfaction, sanitation, equality and democratic quality
(with no data for education). While Vietnam provides an interesting case, it fails to achieve
the basic aspirations laid out in the SDGs and therefore cannot be held up as an ideal model.
There are other interesting cases to investigate, however, although these tend to be obscured
in the original results. O’Neill etal. have chosen to represent social thresholds and biophysical
boundaries in binary terms. In other words, a country has either achieved the social threshold
or it has not; and it has either overshot a biophysical boundary or it has not. This is a sensible
approach, but it can give a misleading impression of how badly a country performs. For instance,
if a country is just 1% over all seven biophysical boundaries, it would register at the highest
possible level of biophysical overshoot (seven out of seven). Similarly, if a countr y is just 1% under
all 11 social thresholds, it would register as a complete failure on the social scale (0 out of 11). In
such cases, the binary measure may end up obscuring countries that are quite promising.
We can correct for this problem with a different approach. O’Neill etal.’s data is available on
the project’s website.10 Biophysical indicators are all standardised to the same scale, with the
boundary for each indicator rendered as 1 and current values rendered as a ratio of the bound-
ary. For instance, the boundary for CO2 emissions is 1.6 tonnes of CO2 per person per year. The
UK emits 12.1 tonnes of CO2 per person per year, while Bangladesh emits 0.4 tonnes of CO2 per
4 J. HICKEL
Figure 1. Ecological eciency of nations: scatterplot of average biophysical impact with respect to
planetary boundaries (horizontal, with boundary as 1) versus their achievement with respect to social
thresholds (vertical, with threshold as 1)
person per year. Since the boundary is rendered as 1, then the UK’s emissions are rendered as
7.48 (over the boundary) while Bangladesh’s emissions are rendered as 0.28 (under the bound-
ary). Social indicators are also standardised to a single scale, with the lowest actually-existing
value for a given indicator rendered as 0, the threshold rendered as 1 and current values ren-
dered as a ratio of the threshold.11 UK life satisfaction is 1.1 (over the threshold) while Bangladesh
is 0.58 (under the threshold). Given how the data are normalised, we can determine a country’s
average distance from the biophysical boundaries and average distance from the social thresh-
olds, such that each country has a single biophysical score and a single social score. For example
the UK’s average biophysical score is 4.10, while its average social score is 1.10. These results are
useful in that they allow us to compare countries’ performance vis-à-vis biophysical boundaries
and social thresholds, but because the underlying indicators are normalised from different scales
and therefore weighted differently in the average, they should not be taken as standalone
figures.12
Figure 1 renders all nations’ average scores on a scatter plot along social and biophysical axes
(excluding nations for which fewer than half of the social or biophysical data points are available).
The result shows that achievement on social indicators rises rapidly as biophysical pressure
increases, but reaches a kink-point at or near the average biophysical boundary, after which it
begins to flatten off. If we focus on this kink-point, we see that a number of countries manage
to come quite close to the social threshold while remaining within biophysical boundaries on
average (most notably Moldova, Algeria and Vietnam, all of which exceed 0.9 on the average
social scale), while other countries achieve the social threshold while only slightly overshooting
average biophysical boundaries (most notably Costa Rica and Cuba).
Cuba is the obvious exceptional performer, but as data for Cuba is only available for five
of the 11 social indicators, the result is not robust; I have kept Cuba in the scatterplot for
THIRD WORLD QUARTERLY 5
reference only. Of the remaining high-performers on the social scale, Costa Rica is perhaps
the most promising. Costa Rica has a high average social score of 1.03, with an average
biophysical score of 1.25. Costa Rica’s strong performance is reflected also in its top ranking
in the Happy Planet Index, which weighs life expectancy and life satisfaction against eco-
logical footprint. Costa Rica’s success on social indicators like healthy life expectancy, edu-
cation and life satisfaction is due largely to its strong commitment to universalism, with a
public health and education system that delivers impressive outcomes with relatively little
money.13 Costa Rica matches or even exceeds most high-income nations in these indicators
with a fraction of their GDP per capita (US$11,800).
While this approach gives us a sense for the general biophysical efficiency of nations in
generating social outcomes, it is not consistent with the logic of the planetary boundary frame-
work as it allows good performance on some biophysical indicators to compensate for poor
performance on others. Just because a country lives within the boundary for one biophysical
indicator (e.g. nitrogen) does not mean it can then exceed the boundary for another (e.g. CO2
emissions). Overshoot of any boundary is potentially catastrophic; this is why O’Neill at al. have
adopted the binary approach. A similar caution applies to the social indicators: just because a
country exceeds the threshold for life expectancy does not mean it can ignore education. Also,
for the nutrition indicator, a score of more than 1 does not necessarily mean better, since addi-
tional calorie intake could indicate obesity and related problems. It is possible to refine the
methodology in order to prevent the compensation effect and adhere to the principles of the
planetary boundary framework. To do so, Figure 2 uses the following settings: (1) For nations
that have scores less than 1 for all biophysical indicators, the scores are simply averaged.
Therefore, in Figure 2, nations that do not transgress any planetary boundaries are depicted as
Figure 2. Ecological eciency of nations, excluding the compensation eect: scatterplot of average
biophysical impact with respect to planetary boundaries (horizontal, with boundary as 1) versus their
achievement with respect to social thresholds (vertical, with threshold as 1).
6 J. HICKEL
having total biophysical scores less than 1; (2) For nations with overshoot on one or more bio-
physical indicators, the overshoot (i.e. aggregate distance over 1) is added up and then divided
by seven (the total number of biophysical indicators) to yield the extent of average biophysical
overshoot. Therefore, in Figure 2, nations that overshoot even one planetary boundary have
total biophysical scores greater than 1, even if they are under the boundary on all of the other
indicators; (3) For the social scale, the compensation effect is removed by ignoring values that
exceed threshold levels, so that no value exceeds 1. For example, the UK’s life satisfaction score
of 1.1 is rendered as 1. The scores are then simply averaged. This method will be used in all of
what follows.
The results show that of nations that remain within all biophysical boundaries, Moldova
does the best on social indicators, with an average social shortfall of 10% (or a score of 0.9).
14
Of the social high-performers, Costa Rica is the most efficient with average social shortfall
of 4% (a score of 0.96) and average biophysical overshoot of 33% (a score of 1.33). Other
notable nations include Vietnam, Sri Lanka and El Salvador. Again, the result for Cuba is not
robust and is included only for reference.
Some social thresholds can be achieved with little additional biophysical
pressure
Some social indicators are more resource intensive than others. O’Neill etal. demonstrate
that ‘the social indicators most tightly coupled to resource use are secondary education,
sanitation, access to energy, income and nutrition’, as these are related to physical needs and
all clearly require resource inputs (including education, which, at least in its modern institu-
tional form, requires material infrastructure and supplies).15 All of these social indicators have
a coefficient of determination (R2) greater than 0.5 with respect to most of the biophysical
indicators.16 In other words, more than 50% of performance on these indicators is explained
by resource use. The more ‘qualitative’ social indicators are not as tightly coupled to resource
use: life satisfaction, equality, social support, democratic quality and employment. All of
these have R2 values less than 0.5 for most biophysical indicators. Social support has R2 of
less than 0.4 for most biophysical indicators. Equality has R
2
of less than 0.3 for most biophys-
ical indicators. Employment has no statistical relationship with resource use at all. According
to O’Neill etal.’s data, nations that achieve the thresholds of these qualitative indicators also
have very high levels of resource use. But this needn’t be the case. Given the weak relationship
between qualitative indicators and resource use, it is theoretically possible to achieve the
thresholds with relatively little additional biophysical pressure. In fact, some of them can be
achieved without any additional biophysical pressure at all. For instance, a country could
achieve the equality threshold by simply shifting income from rich households to poor house-
holds (through higher wages, progressive taxation or direct transfers, for instance) and could
achieve high employment by, say, shortening the working week and sharing necessary labour.
Such policy moves would require no additional resource use, in and of themselves (although
they may entail shifts in the composition of resource consumption).
If this is the case, there may be an argument for removing the five qualitative indicators
from the aggregate analysis, to clarify the challenge when it comes to social thresholds that
do require more intensive resource use to achieve. As Figure 3 illustrates, rendering the data
this way makes the kink-point significantly sharper. Under these parameters, a number of
other promising countries come into view. As with Figure 2, Figure 3 uses the refined method
THIRD WORLD QUARTERLY 7
to correct for the compensation effect. The results show that Sri Lanka and Moldova have
average social shortfall of only 6% (a score of 0.94) while remaining entirely within biophysical
boundaries. Costa Rica comes close to a social score of 1 (0.99) with average biophysical
overshoot of 33% (1.33). Cuba also achieves a social score of 0.99, with a biophysical score
of 1.27, and this time the result is more robust (with four of six data points). Tunisia is the
best overall performer according to this approach, with a social score of 0.98 and a biophys-
ical score of 1.17.
We can take this a step further. Even for some of the indicators that are tightly coupled
with resource use, minimal thresholds can be achieved without any additional biophysical
pressure. For example, the income threshold calls for 95% of the population to be living
on more than US$1.90 per day. According to research by Chris Hoy and Andy Sumner,
three- quarters of the global poverty gap at US$1.90 could be covered through nation-
al-level redistribution (without any growth at all), simply by reallocating public resources
from fossil fuel subsidies and surplus military spending and using the money to fund
direct transfers to the poor, assisted by a modest rate of progressive taxation.17 Nine of
the 10 middle-income countries with the highest numbers of people in poverty (including
India, China, Brazil, Indonesia and the Philippines) could use redistribution to achieve
the income indicator without any additional biophysical pressure. Indeed, the redistrib-
utive policies suggested by Hoy and Sumner may even reduce biophysical pressure.18
Low-income nations, however, can only cover about 37% of the poverty gap on average.
It may be possible to use the same method of national redistribution to achieve other
resource-intensive social indicators – for example, by investing in universal social services
for healthcare, education and electricity provision. Hoy and Sumner’s data shows that the
nine high-poverty middle-income countries that can cover the poverty gap through redis-
tribution can do so an average of more than 20 times over. In other words, even after achiev-
ing the social threshold for income they would have plenty of financial resources left over
Figure 3. Ecological eciency of nations, excluding the compensation eect and excluding “qualita-
tive” social indicators: scatterplot of average biophysical impact with respect to planetary boundaries
(horizontal, with boundary as 1) versus their achievement with respect to social thresholds (vertical,
with threshold as 1).
8 J. HICKEL
for other social investments. This would not be possible for most low-income nations, how-
ever, which lack the aggregate resources even to cover shortfall on the income threshold.
All of the promising outliers identified in the graphs above fall into the category of mid-
dle-income countries: Costa Rica, Tunisia, Algeria, Vietnam, etc. In light of the above, it should
theoretically be possible for these nations to cover their remaining (minimal) social shortfall
through redistribution without any additional biophysical pressure. Even more interestingly,
countries like Sri Lanka and Moldova, which are presently well within all biophysical bound-
aries (with an average score of 0.49 and 0.52, respectively) and which have minimal shortfall
on non-qualitative social indicators (with a score of 0.94), should be able to cover their social
shortfall while still remaining within all biophysical boundaries.
Higher poverty lines make the challenge more dicult
For the income indicator, O’Neill etal. rely on a poverty line of US$1.90 per day ( 2011 PPP).
US$1.90 is the standard international poverty line (IPL) used by the World Bank and the UN’s
Sustainable Development Goals (SDGs). But the IPL has been widely criticised by scholars
as too low to be meaningful.19 The IPL is based on the national poverty lines of the world’s
low-income countries, many of which have been set using poor data. What is more, they tell
us little about what poverty is like in even slightly better-off countries. In most developing
countries, the IPL underestimates poverty when compared to national lines. In India, for
example, national data shows that absolute poverty is twice as high as the IPL suggests.20
In Mexico, the figure is about 10 times as high.21 Moreover, research shows that in many
countries the IPL is not sufficient for basic human health. In India, a child living just above
the IPL has a 60% risk of being malnourished. In Niger, babies born just above the IPL face
an infant mortality risk of nearly 16%, which is five times the world average.22
If US$1.90 is not sufficient to guarantee basic nutrition or infant survival, then we cannot
claim that lifting people above this line means bringing them out of poverty – to say nothing
of achieving a good life. Rahul Lahoti and Sanjay Reddy argue that people need about
US$5.04 per day in order to achieve minimum basic nutrition alone, aside from other require-
ments.23 The New Economics Foundation argues that people need about US$7.20 per day
to reduce infant mortality down to the world average of 30/1000, which is still five times
higher than in developed countries.24 Research by Peter Edward shows that in order to
achieve normal human life expectancy of just over 70 years, people need between 2.7 and
3.9 times more than the IPL, or about US$7.40 per day.
25
This is what Edward calls the ‘ethical
poverty line’. Longitudinal studies show that, in many regions, something closer to US$10
per day is necessary for a permanent escape from poverty.26 This conclusion is in keeping
with arguments by Charles Kenny and Lant Pritchett, who suggest that the global poverty
line should be as high as US$12.50 or even US$15 per day.27
If we are to be serious about achieving a good life for all, we cannot rely on the US$1.90 line.
Raworth suggests US$3.10 per day for the safe and just space framework.
28
An ethical poverty
line of US$7.40 (2011 PPP) per day would be a more reasonable minimum, as it allows for
meaningful achievement on the key indicators of nutrition, life expectancy and infant mor-
tality. For most low- and middle-income nations, using this poverty line significantly worsens
their average social shortfall. The best performers identified in Figure 3 all drop by as much
as 10%. Costa Rica falls from 0.99 to 0.96 (with 21% of the population living on less than
US$7.40), while Vietnam drops from 0.95 to 0.83 (with 58% of the population living on less
THIRD WORLD QUARTERLY 9
than US$7.40). Note that one problem with this approach to the income indicator is that it
does not account for the poverty ‘gap’ – the extent of shortfall below the poverty line. In
other words, nation A and nation B might have the same percentage of people living in
poverty, but the people of nation B might be much deeper in poverty than the people of
nation A and therefore require more resources to bridge the gap. This nuance is obscured
by the income indicator.
Adopting a higher poverty line makes it more difficult to end poverty while remaining
within planetary boundaries. At the US$7.40 line, Belarus is the most promising, with minimal
social shortfall (a score of 0.98) excluding qualitative indicators, but its average biophysical
score is 1.64. Of the nations that achieve all non-qualitative social thresholds, the most
biophysically efficient is Oman, which has an average biophysical score of 2.66. In other
words, given the existing best-case relationship between resource use and income, achieving
a good life for all with an income threshold of US$7.40 per day would require that poor
nations overshoot planetary boundaries by at least 64% to 166%.
Of course, some of this could be covered by national redistribution. But at US$7.40 per
day, it would be much more difficult for global South countries to end poverty by this method
alone. Hoy and Sumner show that three-quarters of the global poverty gap could be ended
at US$5 per day with national redistribution along the lines that they propose. At US$10 (the
next highest poverty line they examine), only 17% of the global poverty gap could be cov-
ered. Only six of the high-poverty middle-income nations they examine could end poverty
at US$5 per day with national redistribution, while none could end poverty at US$10 per
day. One way to end poverty at these higher thresholds is by growing the domestic econo-
mies of global South countries, so that they generate new resources that could be redistrib-
uted toward poverty eradication; but this would exacerbate the transgression of planetary
boundaries. Alternatively, income could be better distributed globally. There are two ways
to accomplish the latter: (1) change the rules of the global economy – on trade, debt, tax
evasion, capital flows, global governance, etc – to make it fairer to global South countries,
thus allowing them to claim a greater share of global GDP (and, hopefully, use the additional
resources to achieve social thresholds); or (2) redistribute income through, say, a universal
basic income or universal social services funded by a financial transaction tax, a carbon tax,
a resource extraction tax, a wealth tax and so on.29
The above analysis illustrates how sensitive the income indicator is to the definition of
poverty. That said, there may be reasons to question the utility of relying too heavily on
income as a key indicator of a good life. If the purpose of setting an income threshold is to
allow for meaningful achievement on indicators like nutrition, life expectancy and infant
mortality, then it is not clear that income needs to be included in the analysis if these other
indicators are already represented. Indeed, doing so may penalise countries that are able to
deliver high levels of human well-being without high levels of income.
Achieving a good life for all will exacerbate global ecological overshoot
As I have argued above, it is theoretically possible – under already-existing conditions and
with known policy measures – for nations to achieve all key social thresholds without exceed-
ing biophysical boundaries. For the low-income countries clustered toward the vertical axis
in the graphs, which are well under biophysical boundaries, this will entail at least some
increase in their biophysical footprint. This in turn means increasing the global aggregate
10 J. HICKEL
Table 2. Per capita consumption of resources relative to planetary boundaries, with development
according to Boundary model.
CO2Phosphorus Nitrogen
Ecological
Footprint
Material
Footprint
Boundary model 1 1 1 1 1
World 3.21 2.22 2.31 1.32 1.41
World with development according to Boundary model 3.33 2.46 2.51 1.50 1.64
Additional overshoot 12% 24% 20% 18% 23%
Table 3. Per capita consumption of resources relative to planetary boundaries, with development
according to Sri Lanka model.
CO2Phosphorus Nitrogen
Ecological
Footprint
Material
Footprint
Sri Lanka 0.65 0.17 0.21 0.68 0.45
World 3.21 2.22 2.31 1.32 1.41
World with development according to Sri Lanka model 3.27 2.23 2.32 1.37 1.44
Additional overshoot 2% 1% 1% 4% 2%
biophysical footprint as well, which is problematic given that planetary boundaries are
already being overshot on a global level. The O’Neill data shows global overshoot on CO2
emissions, phosphorous, nitrogen, ecological footprint and material footprint. If this is the
case, then the only way for all global South nations to achieve all social thresholds without
triggering further overshoot is for rich nations to significantly diminish their biophysical
footprints.
We can use the O’Neill dataset to quantify this. Let us assume that there is a ‘boundary
model’ of efficient development whereby poor nations can achieve all social thresholds without
overshooting planetary boundaries (as I have argued, is theoretically possible). If poor nations
implement this model and achieve all social thresholds while increasing their biophysical
footprints up to the boundary for each indicator, how much additional biophysical pressure
would this represent on a global scale? Tables 2–5 show existing global scores on five biophys-
ical indicators. They exclude indicators for land and water, as the O’Neill data show that it is
possible to achieve a good life for all with relatively low per capita use of land and water, as
many nations with full social achievement are within the boundaries for these indicators.
Table 2 shows what global resource use would be if poor nations (those with average
social scores less than 1 and average biophysical scores less than 1) implemented the bound-
ary model of development and increased their biophysical footprints to 1 for each indicator.
The final row in Table 2 shows the additional overshoot that this would entail, rendered as
a percentage of the planetary boundary. The results show that development according to
the boundary model entails exacerbating global overshoot by an average of 19%. This
assumes no additional resource use by rich nations.
We can test a more optimistic scenario using what we might call the Sri Lanka model,
whereby we assume that poor nations can achieve all social thresholds while increasing
their biophysical footprints up to the level of Sri Lanka for each indicator. Sri Lanka’s average
social shortfall is minimal, with a score of 0.94 (excluding the five qualitative social indicators),
and as a middle-income country it should have the capacity to cover this shortfall through
national redistribution, without any additional biophysical pressure. Table 3 shows that if
poor nations implement the Sri Lanka model, it would entail minimal additional biophysical
pressure, exacerbating global overshoot by an average of only 2%.
THIRD WORLD QUARTERLY 11
Sri Lanka is an outlier, however (along with Moldova). We can be more conservative
by using the Tunisia model. Tunisia’s average social shortfall is even less than Sri Lanka’s,
with an overall score of 0.98 (excluding the five qualitative social indicators). Its biophys-
ical footprint is in overshoot, but minimally so. As a middle-income country, it should be
able to cover its social shortfall through national redistribution. Table 4 shows that if poor
nations implement the Tunisia model, it would entail exacerbating global overshoot by
an average of 29%.
Relying on the above models means making assumptions about the political feasibility
of national redistribution, and about the ability of nations to achieve qualitative social thresh-
olds without any additional biophysical pressure. While these assumptions are theoretically
valid, we have no evidence for them among existing national examples. We can be more
realistic by using the Costa Rica model. Costa Rica has minimal average social shortfall (with
a score of 0.99) across all 11 social indicators, including the qualitative ones, and therefore
requires no speculation on the possibility of achieving qualitative social indicators without
any additional biophysical pressure, and there are no concerns about the feasibility of
national redistribution. Table 5 shows that if poor nations implement the Costa Rica model,
it would entail worsening global overshoot by an average of 35%.
Rich countries will need to adopt de-growth strategies
Tables 2–5 demonstrate that while it is possible for poor nations to achieve a good life for all
within planetary boundaries, the additional resource use that this entails would significantly
exacerbate global overshoot of planetary boundaries, given the high degree of overshoot
that presently characterises rich economies. This conclusion holds true for all four develop-
ment models explored above. The only way to achieve a good life for all within planetary
boundaries is for overshoot nations to significantly reduce their biophysical footprints.
Table 6 quantifies the average biophysical reductions required of overshoot nations under
three different scenarios. Row 1 assumes that all poor nations achieve social thresholds by
increasing their biophysical footprints to planetary boundaries for each indicator (the ‘bound-
ary model’ of development), and shows the average reductions required of overshoot nations
Table 4. Per capita consumption of resources relative to planetary boundaries, with development
according to Tunisia model.
CO2Phosphorus Nitrogen
Ecological
Footprint
Material
Footprint
Tunisia 1.7 1.14 0.96 1.02 1.24
World 3.21 2.22 2.31 1.32 1.41
World with development according to Tunisia model 3.61 2.53 2.49 1.51 1.76
Additional overshoot 40% 31% 18% 19% 35%
Table 5. Per capita consumption of resources relative to planetary boundaries, with development
according to Costa Rica model.
CO2Phosphorus Nitrogen
Ecological
Footprint
Material
Footprint
Costa Rica 1.72 1.2 1.14 1.29 1.41
World 3.21 2.22 2.31 1.32 1.41
World with development according to Costa Rica model 3.62 2.55 2.57 1.65 1.84
Additional overshoot 41% 33% 26% 33% 43%
12 J. HICKEL
Table 6. Average biophysical footprint reduction from current levels required of overshoot nations.
CO2Phosphorus Nitrogen
Ecological
Footprint
Material
Footprint
Boundary model 70% 59% 60% 33% 39%
Tunisia model 53% 55% 61% 32% 30%
Costa Rica model 52% 53% 56% 22% 23%
to get global biophysical footprints down to the level of planetary boundaries (average
reductions of 52% from current levels required). Row 2 assumes that poor nations achieve
social thresholds by implementing the Tunisia model, and shows the average reductions
required of overshoot nations to get global biophysical footprints down to the level of Tunisia
(average reductions of 46% required). Row 3 repeats the exercise for convergence at the
biophysical scores of Costa Rica (average reductions of 41% required). Note that convergence
at the Tunisia and Costa Rica models would still entail overshooting planetary boundaries
(except for Nitrogen in the case of the Tunisia model).
Theoretically, it should be possible for overshoot nations to reduce their biophysical con-
sumption down to planetary boundaries without falling below social thresholds even while
improving performance on social indicators. Indeed, this is what O’Neill etal. argue. But it
may not be possible for them to do so while at the same time pursuing continuous
GDP growth.
There is strong evidence for this in relation to the material footprint indicator, which
measures extraction and use of biomass, minerals, fossil fuels and construction materials.
Material footprint is a key indicator in that it pertains to a broad range of ecological concerns,
including deforestation, meat consumption, overfishing, greenhouse gas emissions and
environmental damage due to mining. To reduce material footprint while at the same time
pursuing GDP growth requires absolute decoupling of GDP from material use. Three recent
studies (Dittrich etal., Schandl etal., UNEP) have explored whether aggressive policy mea-
sures and gains in technological efficiency can drive absolute decoupling in the decades to
2050; all of them conclude that relative decoupling can be achieved, but they find no evi-
dence that absolute decoupling will happen – even under highly optimistic assumptions.30
Models that incorporate the ‘rebound effect’ yield particularly discouraging results.31
These studies look at material footprint trends at a global level, but the same general
conclusion holds for rich nations. While one well-known model (Hatfield-Dodds etal.) sug-
gests absolute decoupling may be possible (in Australia), it assumes a rate of efficiency
improvement that lacks empirical basis and is in any case out of scope.32 Moreover, Ward
etal. demonstrate that the result holds only in the short term. As efficiency improvements
reach physical limits, the scale effect of growth drives total resource use up. Ward etal.
conclude that this implies a ‘robust rebuttal to the claim of absolute decoupling’: ‘decoupling
of GDP growth from resource use, whether relative or absolute, is at best only temporary.
Permanent decoupling (absolute or relative) is impossible for essential, non-substitutable
resources because the efficiency gains are ultimately governed by physical limits. Growth
in GDP ultimately cannot plausibly be decoupled from growth in material and energy use,
demonstrating categorically that GDP growth cannot be sustained indefinitely’.33
Similar concerns apply to CO2 emissions. It is possible to achieve absolute decoupling of
GDP from CO2 emissions; the question is whether it can be achieved at a rate rapid enough
to respect the carbon budget for 2C. Anderson and Bows have modelled the emissions
reductions necessary for achieving a 50% chance of staying under 2C (assuming the principle
THIRD WORLD QUARTERLY 13
of common but differentiated responsibility, whereby high-income nations need to lead on
emissions reductions). They conclude that high-income nations (Annex 1 nations) need to
reduce emissions by 10% per year, beginning in 2015.34 At existing rates of economic growth
in Annex 1 nations (i.e. 1.86% per year, the average from 2010–2014), decoupling must occur
at a rate of 13.18% per year.35 This is seven times faster than existing rates of decoupling in
Annex 1 nations (viz., 1.9% per year from 1970 to 2013).
36
It also exceeds the decoupling rate
implied by the average G20 Nationally Determined Contributions under the Paris Agreement
(viz., 3% per year) by a factor of more than four.
It is theoretically possible to achieve the emissions reductions required for 2C by relying
on negative emissions technologies. Most IPCC pathways for 2C rely on BECCS (bioenergy
with carbon capture and storage) in particular. However, an emerging consensus among
climate scientists rejects the use of BECCS in climate models on the grounds that it is a
speculative technology and there is no evidence that it can be scaled fast enough and to
the extent required; moreover, it would require such expansive land use that it would make
it impossible to meet minimum food requirements for the world’s population (violating the
nutrition threshold) and would significantly exacerbate biodiversity loss, which is one of the
key biophysical boundaries.
37
Thus, relying on BECCS for negative emissions is not acceptable
as part of a strategy for achieving a good life for all within planetary boundaries.
In sum, there is no empirical evidence to support the notion that rich nations can make
sufficiently dramatic reductions in resource use and emissions while at the same time pur-
suing economic growth. The reason for this is that the scale effect of growth eats the gains
that can be feasibly achieved through decoupling. In light of this, achieving a good life for
all within planetary boundaries will require that rich nations begin to gradually downscale
their aggregate economic activity, embarking on a trajectory of planned de-growth. One
approach would be to gradually reduce the size of the population (in an equitable, progres-
sive and non-coercive way), so that GDP per capita can be maintained even while total
economic activity shrinks. But if we assume that the population grows according to existing
projections and stabilises at 9–11 billion, this will require de-growth in both absolute and
per capita terms. Scholars argue that de-growth can be achieved without any loss to social
indicators, and could further enhance human well-being if done equitably.38 This can be
accomplished by downscaling socially unnecessary and ecologically destructive industries,
while covering any employment shortfalls by shortening the working week, by distributing
existing income and resources more fairly through progressive taxation and reallocation
into social spending (i.e. on healthcare, education, etc) and/or by improving wages.
It is clear, however, that any prolonged, planned reduction of aggregate economic activity
is not compatible with capitalism, which fundamentally depends on ever-increasing growth
of production and consumption. De-growth strategies will therefore require evolving beyond
the strictures of capitalism toward a post-growth system.
Implications for the development agenda
When it comes to achieving a good life for all within planetary boundaries, poor nations are
the ‘easy’ part. It is rich nations that present the real challenge. For poor nations, achieving
a good life for all within planetary boundaries requires improving the development model
to make it more efficient at converting resources into well-being. In some cases, this can be
accomplished largely through redistribution of existing domestic resources; in others, it
14 J. HICKEL
requires moderate increases in resource use, up to the level of planetary boundaries. For
rich nations, it requires reductions of resource use that are so significant as to require the
adoption of de-growth strategies, and therefore a shift toward post-capitalist economic
models. This requires a fundamental reorientation of development theory, from focusing
primarily on the deficiencies of poor countries to focusing on the excesses of rich countries.
Much of the existing literature on the safe and just space framework sidesteps this
conclusion. For example, Raworth does note the crucial role of national-level redistribu-
tion (advocating for stronger minimum wages and the introduction of maximum wages,
land-value taxes, resource taxes, more egalitarian distribution of finance, a shift to coop-
erative models of business, etc), and highlights various strategies for recycling and regen-
eration that businesses and governments can use.
39
However, on the question of growth
she is ‘agnostic’ and argues for the need to design ‘an economy that promotes human
prosperity whether GDP is going up, down, or holding steady’.40 At most, she promotes
an ‘S Curve’ for growth. Just as plants and animals grow to the point of maturity and then
remain at an equilibrium, so too nations should seek to achieve ‘arrival’ at an adequate
level of economic development, with GDP reaching a steady, zero-growth level.41 But
she sidesteps the question of how much GDP is actually sufficient for a good life for all,
and sidesteps the question of whether de-growth will be necessary for countries that
dramatically overshoot planetary boundaries. Indeed, the S Curve implies that rich
nations can safely continue their existing levels of economic activity so long as they do
not grow any further. In an era of dangerous ecological overshoot, and given the absence
of empirical evidence for sufficient absolute decoupling of GDP from environmental
impact, this is not a defensible position.
Like Raworth, O’Neill etal. identify key strategies that nations can use to reduce their
biophysical footprints: switching to renewable energy, producing products with longer
lifetimes, reducing unnecessary waste, shifting from animal to crop products, investing in
new technologies and moving beyond GDP to embrace new measures of progress. Unlike
Raworth, however, they indicate that ‘It could also involve the pursuit of “degrowth” in
wealthy nations, and the shift towards alternative economic models such as a steady-state
economy’.42 Yet it would seem that the O’Neill data requires a stronger conclusion here.
Achieving a good life for all within planetary boundaries will require overshoot nations to
reduce their biophysical footprints by at least 40–50% on average from current levels
(assuming poor nations can achieve social thresholds within planetary boundaries). Extant
empirical evidence indicates that this is highly unlikely to be possible without de-growth
strategies.
This has radical implications for our approach to international development. The
Sustainable Development Goals, for instance, will need to be rethought. At present they
include a demand for exponential global GDP growth. Target 8.1 reads: ‘Sustain per capita
economic growth in accordance with national circumstances and, in particular, at least 7 per
cent gross domestic product growth per annum in the least developed countries’, as mea-
sured by ‘annual growth rate of real GDP per capita’. The assumption is that global growth
will facilitate the achievement of key social goals, such as on poverty, hunger and education.
But achieving the aggregate rate of growth required by Goal 8 will violate the sustainability
goals (i.e. Goal 12.2: ‘By 2030, achieve sustainable management and efficient use of natural
resources’; Goal 13: ‘Take urgent action to combat climate change and its impacts’). In other
words, the SDGs, as presently written, are internally contradictory.
THIRD WORLD QUARTERLY 15
In order for the SDGs to succeed, they will need to allow for growth in poor nations (for
the sake of defined social goals, with high levels of efficiency in converting resources into
well-being, and with heavy pro-poor bias), while calling for rich nations to reduce their
biophysical footprints down to sustainable levels, with specific targets. As noted above, one
way to achieve this would be through redistributing global GDP from rich nations to poor
nations, either by making the rules of the global economy fairer, or through direct transfers
of income. But the only way to ensure that planetary boundaries are not violated on a global
level would be to impose caps on resource use and pollution for every biophysical process
identified in the planetary boundary framework, so that we never extract more than the
Earth can safely regenerate, and never pollute more than it can absorb. The ‘budgets’ for
each biophysical process could then be distributed equitably among nations, on the basis
of ‘common but differentiated responsibility’ (as in UNFCCC Article 3.1), taking account of
development needs and historical responsibility for overshoot.
Acknowledgments
I would like to thank Dan O’Neill for his comments on an earlier draft and for insights he
shared through conversations and correspondence. Thanks also to the anonymous reviewers,
whose comments significantly improved the argument. Any shortcomings are my own.
Notes on contributor
Jason Hickel is an anthropologist at Goldsmiths, University of London. He is the author of a
number of books, including most recently The Divide: A Brief Guide to Global Inequality and
its Solutions (Penguin, 2017). He has published widely on international development, political
economy and ecological economics, contributes regularly to The Guardian and Al Jazeera,
serves on the Task Force on International Development for the Labour Party, and sits on the
executive board of Academics Stand Against Poverty. He is a Fellow of the Royal Society
of Arts.
Notes
1. Rockström etal., “Planetary Boundaries”; Steen etal., “Planetary Boundaries.”
2. Raworth, “A Safe and Just Space”; Raworth, Doughnut Economics.
3. O’Neill etal., “A Good Life for All.”
4. Per capita equivalents are applied equally across countries. O’Neill et al. do not account for
whether some countries require more resource use than others for reasons beyond their control;
for instance, Arctic nations may require more energy to heat their homes than tropical nations.
5. O’Neill etal., “A Good Life for All,” 88.
6. Ibid., 90.
7. Ibid., 90.
8. Leeds University, “A Good Life for All.”
9. O’Neill etal., “A Good Life for All,” 92.
10. Leeds University, “A Good Life for All.”
11. In mathematical terms, the normalised data are given by ynorm= (y − ymin) ÷ (y* − ymin), where y is
the social indicator, y* is the social threshold and yminis the lowest value for the social indicator.
12. For instance, changing the lowest actually-existing value for one of the social indicators would
change its weighting in the average.
16 J. HICKEL
13. Franzoni and Sanchez-Ancochea, Quest for Universal Social Policy.
14. The Moldova result is dubious; there are some concerns about the validity of the biophysical
data for Moldova, as it is a very small country and information on trade across its borders is
dicult to verify.
15. O’Neill etal., “A Good Life for All,” 91.
16. O’Neill etal., “Supplementary Materials,” Table 3.
17. Hoy and Sumner, “Gasoline, Guns and Giveaways.”
18. Higher levels of inequality tend to increase ecological degradation. For instance, Holland etal.,
“CrossNational Analysis”, nd that countries with higher levels of inequality have higher levels
of biodiversity loss. It is reasonable to expect that removing subsidies for fossil fuels and reallo-
cating surplus military spending would probably reduce CO2 emissions. Redistributing income
downward might have a similar eect, given that CO2 emissions are disproportionately high
among the richest 10% of each nation; Chancel and Piketty, Carbon and Inequality.
19. Hickel, “True Extent of Global Poverty and Hunger.”
20. NDTV, “Poverty in India”; Prashad, “Making Poverty History.”
21. Cimadamore etal., Poverty and Millennium Development Goals.
22. Wagsta, “Child Health on a Dollar a Day.”
23. Lahoti and Reddy, “$1.90 per Day.”
24. New Economics Foundation, “How Poor is Poor?”
25. Edward, “The Ethical Poverty Line.”
26. López-Calva and Ortiz-Juarez, “A Vulnerability Approach”; Sumner etal., “Prospects of the Poor.”
27. Kenny, “Why Ending Extreme Poverty”; Pritchett, “Who is Not Poor?”
28. Raworth, Doughnut Economics.
29. Hickel, The Divide, 253–278.
30. Dittrich etal., Green Economies; Schandl etal., “Decoupling Global Environmental Pressure”;
UNEP, “Resource Eciency.”
31. UNEP, “Resource Eciency,” 106 .
32. Alexander etal., “A Critique of Decoupling.”
33. Ward etal., “Is Decoupling Possible?”
34. Anderson and Bows, “Beyond ‘Dangerous’ Climate Change.”
35. Using the equation: Rate of necessary decoupling = GDP growth rate / (1 – Rate of necessary
emissions reductions).
36. Decoupling slowed from an average of 2.3% per year in the rst half of the period to an average
of 1.6% in the second half, according to World Bank, World Development Report 1999/2000,
Databank, CO2 emissions (kg per 2010 US$GDP).
37. For concern about the viability of BECCS, see: Anderson and Peters, “The Trouble with Negative
Emissions”; Larkin etal., “What if Negative Emissions Technologies Fail?”; Fuss etal., “Betting on
Negative Emissions.” For concern about the ecological consequences of implementing BECCS,
see: Smith et al., “Biophysical and Economic Limits”; Heck etal., “Biomass-Based Negative
Emissions.”
38. Alier, “Socially Sustainable Economic De-growth”; Jackson, Prosperity without Growth; K allis, “In
Defense of Degrowth.”
39. Raworth, Doughnut Economics.
40. Ibid., 245.
41. Ibid., 251.
42. O’Neill etal., “A Good Life for All,” 92.
Bibliography
Alexander, S., J. Rutherford, and J. Floyd. “A Critique of the Australian National Outlook Decoupling
Strategy: A Limits to Growth Perspective.” Ecological Economics 145 (2018): 10–17.
Alier, J. M. “Socially Sustainable Economic De-growth.” Development and Change 40, no. 6 (2009):
1099–1119.
THIRD WORLD QUARTERLY 17
Anderson, K., and A. Bows. “Beyond ‘Dangerous’ Climate Change: Emission Scenarios for a New World.”
Philosophical Transactions of the Royal Society of London A 369, no. 1934 (2011): 20–44.
Anderson, K., and G. Peters. “The Trouble with Negative Emissions.” Science 354, no. 6309 (2016):
182–183.
Chancel, L., and T. Piketty. Carbon and Inequality: From Kyoto to Paris. Paris: Paris School of Economics,
2015.
Cimadamore, A., G. Koehler., and T. Pogge, eds. Poverty and the Millennium Development Goals:
ACritical Look Forward. London: Zed Books, 2016.
Dittrich, M., S. Giljum, S. Lutter, and C. Polzin. Green Economies around the World. Implications of
Resource Use for Development and the Environment. Vienna: SERI, 2012.
Edward, P. “The Ethical Poverty Line: A Moral Quantication of Absolute Poverty,” Third World Quarterly
37, no. 2 (2006): 377–393.
Franzoni, J. M., and D. Sánchez-Ancochea. The Quest for Universal Social Policy in the South: Actors, Ideas
and Architectures. Cambridge: Cambridge University Press, 2016.
Fuss, S., etal. “Betting on Negative Emissions”. Nature Climate Change 4, no. 10 (2014): 850–853.
Heck, V., etal. “Biomass-Based Negative Emissions Dicult to Reconcile with Planetary Boundaries,”
Nature Climate Change 8 (2018): 151–155.
Hickel, J. “The Divide: A Brief Guide to Global Inequality and its Solutions.” Cornerstone: William
Heinemann, 2017.
Hickel, J. “The True Extent of Global Poverty and Hunger: Questioning the Good News Narrative of the
Millennium Development Goals.” Third World Quarterly 37, no. 5 (2016): 749–767.
Holland, T. G., G. D. Peterson, and A. Gonzalez. “A CrossNational Analysis of How Economic Inequality
Predicts Biodiversity Loss.” Conservation Biology 23, no. 5 (2009): 1304–1313.
Hoy, C., and A. Sumner. “Gasoline, Guns, and Giveaways: Is There New Capacity for Redistribution to
End Three Quarters of Global Poverty?” Center for Global Development Working Paper No. 433
(2016).
Jackson, T. Prosperity without Growth: Economics for a Finite Planet. Abingdon: Routledge, 2009.
Kallis, G. “In Defence of Degrowth.” Ecological Economics, 70, no. 5 (2011): 873–880.
Kenny, C. “Why Ending Extreme Poverty Isn’t Good Enough,” Bloomsburg Businessweek, April 29 (2013).
Lahoti, R., and S. Reddy. “$1.90 per Day: What Does It Say?” Institute for New Economic Thinking, 2015.
Accessed August 1, 2016. https://www.ineteconomics.org/ideas-papers/blog/1-90-per-day-what-
does-it-say?p=ideas-papers/blog/1-90-per-day-what-does-it-say.
Larkin, A., etal. “What if Negative Emissions Technologies Fail at Scale?” Climate Policy (2017).
Leeds University. “A Good Life for All Within Planetary Boundaries.” Accessed May 1, 2018. http://good-
life.leeds.ac.uk
López-Calva, L. F., and E. Ortiz-Juarez. “A Vulnerability Approach to the Denition of the Middle Class.”
Journal of Economic Inequality 12, no. 1 (2014): 23–47.
NDTV, “Poverty in India 2.5 Times The Ocial Figure: Study,” February 20, 2014. http://www.ndtv.com/
india-news/poverty-in-india-2-5-times-the-ocial-gure-study-551445. Accessed August 1, 2016.
New Economics Foundation. “How Poor Is Poor? Toward a Rights-Based Poverty Line,” 2010.
O’Neill etal., “Supplementary Materials,” Downloaded From http:goodlife.leeds.ac.uk
O’Neill, D. W., A. L. Fanning, W. F. Lamb, and J. K. Steinberger. “A Good Life for All Within Planetary
Boundaries.” Nature Sustainability 1, no. 2 (2018).
Prashad, V. “Making Poverty History,” Jacobin, November 10, 2014. Accessed August 1, 2016. https://
www.jacobinmag.com/2014/11/making-poverty-history/.
Pritchett, L. “Who is Not Poor? Proposing a Higher International Standard for Poverty,” Working Paper
33, Center for Global Development (2003).
Raworth, K. “A Safe and Just Space for Humanity: Can We Live within the Doughnut.” Oxfam Policy and
Practice: Climate Change and Resilience 8, no. 1 (2012): 1–26.
Raworth, K. Doughnut Economics: Seven Ways to Think Like a 21st-Century Economist. White River
Junction, Vermont: Chelsea Green Publishing, 2017.
Rockström, J., W. Steen, K. Noone, Å. Persson, F. Stuart Chapin III, E. Lambin, T. M. Lenton, et al.
“Planetary Boundaries: Exploring the Safe Operating Space for Humanity.” Ecology and Society 14,
no. 2 (2009).
18 J. HICKEL
Schandl, H., et al. “Decoupling Global Environmental Pressure and Economic Growth.” Journal of
Cleaner Production 132 (2016): 45–56.
Smith, P. etal. “Biophysical and Economic Limits to Negative CO2 Emissions.” Nature Climate Change 6,
no. 1 (2016): 42–50.
Steen, W., K. Richardson, J. Rockström, S. E. Cornell, I. Fetzer, E. M. Bennett, R. Biggs, etal. “Planetary
Boundaries: Guiding Human Development on a Changing Planet.” Science 347, no. 6223 (2015).
Sumner, A., A. Anshory Yusuf, and Y. Suara. “The Prospects of the Poor: A Set of Poverty Measures
Based on the Probability of Remaining Poor (or Not) in Indonesia.” Center for Economics and
Development Studies (CEDS) Working Paper. Bandung: Padjadjaran University, Indonesia, 2014.
United Nations Environment Programme (UNEP). Resource Eciency: Potential and Economic
Implications. A report from the International Resource Panel, 2017.
Wagsta, A. “Child Health on a Dollar a Day: Some Tentative Cross-country Comparisons,” Social
Science and Medicine 57, no. 9 (2003): 1529–1538.
Ward, J., etal. “Is Decoupling GDP Growth from Environmental Impact Possible?” PloS One 11, no. 10
(2016).
World Bank. World Development Report 1999/2000. Washington, DC: World Bank, 2000.
