Global inequalities in food consumption, cropland demand and land-use efficiency: A decomposition analysis

Abstract

The world population is expected to rise to 9.7 billion by 2050 and to ~11 billion by 2100, and securing its healthy nutrition is a key concern. As global fertile land is limited, the question arises whether growth in food consumption associated with increased affluence surmounts increases in land-use efficiency (measured as food supply per cropland area) associated with technological progress. Furthermore, substantial inequalities prevail in the global food system: While overly rich diets represent a serious health issue for many of the world’s most affluent inhabitants and constitute a critical climate-change driver, undernourishment and hunger still threaten a considerable fraction of the world population, mostly in low-income countries. We here analyze trajectories in cropland demand and their main basic drivers food consumption (measured by a food index reflecting the share of animal products in diets) and land-use efficiency, for 123 countries (clustered in four income groups, covering 94% of the world population). We cover the period 1990–2013 and assess if these trajectories are associated with changes in inequality between countries. We find that while all groups of countries converged towards the high level of the per-capita food consumption of high-income countries, differences between income groups remained pronounced. Overall, cropland demand per capita declined over the entire period in all regions except low income countries, resulting in a tendency towards global convergence. However, the trend slowed in the last years. In contrast, land-use efficiency increased in all income groups with a similar trend, hence international inequalites in land-use efficiency remained almost unaltered. Because population and food requirements per capita are expected to grow in all income groups except the richest ones, failure to improve land efficiency sufficiently could lead to a less unequal but at the same time less ecologically sustainable world. Avoiding such outcomes may be possible by reducing the consumption of animal products in the richer countries and raising land-use efficiency in the poorer countries.

Full text

Contents lists available at ScienceDirect
Global Environmental Change
journal homepage: www.elsevier.com/locate/gloenvcha
Global inequalities in food consumption, cropland demand and land-use
efficiency: A decomposition analysis
Juan Antonio Duro
a,⁎
, Christian Lauk
b
, Thomas Kastner
b,c
, Karl-Heinz Erb
b
, Helmut Haberl
b
a
Economics Department and ECO-SOS, Universitat Rovira i Virgili (URV), Avinguda Universotat 2, Reus, Spain
b
Institute of Social Ecology, Department of Economics and Social Sciences (WiSo, University of Natural Resources and Life Science, Vienna, Schottenfeldgasse 29, 1070
Vienna, Austria
c
Senckenberg Biodiversity and Climate Research Centre (SBiK-F), Senckenberganlage 25, 60325 Frankfurt am Main, Germany
ARTICLE INFO
Keywords:
Food supply
Cropland demand
Inequality
Land-use intensity
Land-use efficiency
ABSTRACT
The world population is expected to rise to 9.7 billion by 2050 and to ~11 billion by 2100, and securing its
healthy nutrition is a key concern. As global fertile land is limited, the question arises whether growth in food
consumption associated with increased affluence surmounts increases in land-use efficiency (measured as food
supply per cropland area) associated with technological progress. Furthermore, substantial inequalities prevail
in the global food system: While overly rich diets represent a serious health issue for many of the world’s most
affluent inhabitants and constitute a critical climate-change driver, undernourishment and hunger still threaten
a considerable fraction of the world population, mostly in low-income countries. We here analyze trajectories in
cropland demand and their main basic drivers food consumption (measured by a food index reflecting the share
of animal products in diets) and land-use efficiency, for 123 countries (clustered in four income groups, covering
94% of the world population). We cover the period 1990–2013 and assess if these trajectories are associated with
changes in inequality between countries. We find that while all groups of countries converged towards the high
level of the per-capita food consumption of high-income countries, differences between income groups remained
pronounced. Overall, cropland demand per capita declined over the entire period in all regions except low
income countries, resulting in a tendency towards global convergence. However, the trend slowed in the last
years. In contrast, land-use efficiency increased in all income groups with a similar trend, hence international
inequalites in land-use efficiency remained almost unaltered. Because population and food requirements per
capita are expected to grow in all income groups except the richest ones, failure to improve land efficiency
sufficiently could lead to a less unequal but at the same time less ecologically sustainable world. Avoiding such
outcomes may be possible by reducing the consumption of animal products in the richer countries and raising
land-use efficiency in the poorer countries.
1. Introduction
The world population is expected to grow to ~9.7 billion by 2050
and ~10.9 billion by 2100 (UN, 2019). Adequately feeding this
growing number of humans is high on the agenda: while the number of
undernourished people steadily decreased for many years, it started to
increase again in 2016 in absolute numbers, with an estimated number
of undernourished persons of 812 million in 2017, i.e. 10.8% of the
world population (FAO et al., 2019). Reducing malnutrition and hunger
is high on the agenda (Sahn, 2015); e.g. it is one of the most important
“Sustainable Development Goals” or SDGs (Griggs et al., 2013). At the
same time, 640 million people representing 13% of all adults world-
wide are obese (FAO et al., 2017). Both hunger and obesity are well-
known threats to human health and wellbeing (Sahn, 2015; Swinburn
et al., 2015).
Producing enough food for all is not a sufficient condition for food
security, given prevailing inequalities, but it is still a necessary condi-
tion (Godfray et al., 2010). Increasing food production is a challenge
because fertile land is limited, and land area represents an important
planetary boundary (Rockströmet al., 2009; Haberl and Erb, 2017;
Steffen et al., 2015). Although cropland covers only ~12% of the
earth’s lands area except Greenland and Antarctica (Erb et al., 2007),
more than half of all biomass extracted and used globally by humans,
including livestock feed, is harvested on croplands (Krausmann et al.,
2008). Even considering the global potential to generate additional
cropland through conversion of grassland, e.g. spared by increasing
grazing intensities, many scenarios for 2050 are limited by cropland
expansion if deforestation is to be avoided (Erb et al., 2016).
https://doi.org/10.1016/j.gloenvcha.2020.102124
Received 12 February 2020; Received in revised form 31 May 2020; Accepted 20 June 2020
⁎
Corresponding author.
Global Environmental Change 64 (2020) 102124
0959-3780/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
T

Furthermore, the expansion of cropland is associated with many en-
vironmental pressures such as soil degradation, pesticide and nutrient
leaching, biodiversity loss, carbon emissions, and many more (IAASTD,
2009). This calls for strategies that allow to increase cropland pro-
duction without proportional increases in cropland area. Moreover, it
means that tracking the land ‘footprint’ (Wackernagel et al., 1999) of
supplying food is highly important. We do this by analysing data on the
actual cropland demand (Erb, 2004), i.e. global cropland demand re-
lated to each country’s food supply.
Inequalities of food provision are another key challenge. The fact
that the volume of food produced globally is sufficient to feed everyone
on the planet on the average, i.e. at a rate of 11.3–12.6 MJ/cap/day,
i.e. > 2700 kcal/cap/day (D’Odorico et al., 2014), illustrates that food
security is threatened by inequality at least as much as by insufficient
production of food. Better understanding how inequality of food con-
sumption between countries changes globally is hence an important
research area, even though important additional questions remain with
regard to within-country inequality of food access and distribution
(d’Odorico et al., 2019) that are beyond the scope of this paper
Experiences from the past suggest that rising incomes are strongly
related with higher demand for food energy in general, and higher
demand for more resource-demanding food such as protein-rich animal
products in particular (Tilman and Clark, 2014;Alexander et al., 2015).
In the past, wealthier nations were able to increase productivities and
efficiencies of the food system: higher yields and better feeding effi-
ciencies counteracted the rising volume and quality of food consumed
(Haberl et al., 2012; Krausmann et al., 2009). Beyond inequality, the
question hence arises whether limited available cropland (Erb et al.,
2016; Steffen et al., 2015) will allow producing enough food, which will
require sufficient efficiency improvements to cope with expectedly ri-
cher future diets.
Concerns about global inequality are in the center of current sus-
tainability debates (Motesharrei et al., 2016). It is widely held that not
only global resource use should be reduced, but also the international
inequalities of production and supply (Duro et al., 2018; United
Nations, 2015). Given widespread hunger and malnutrition, global re-
ductions in food consumption seem unacceptable and unoperative, but
Contraction and Convergence strategies are being discussed as a means
to reduce international inequalities (Meyer, 1999). Hence, it would
make sense to incorporate international inequality instruments into the
sustainabiliy debate and their analysis (Duro et al., 2018; Schaffartzik
et al., 2019).
In this context, the main objective of this article is to analyze the
international inequalities between all of the previous factors, that is,
food supply, land-use efficiency and cropland demand, for the period
1990–2013 at the country level, using an integrated approach. We use
the use the following terms in the text: food supply denotes the amount
of food available in each country. Cropland-use efficiency denotes food
provision per unit of global cropland area required to produce the food
consumed in each country. Cropland demand denotes the extent of
cropland area required within and outside a nations territory for the
amount of food consumed domestically. We analyze these phenomena
through a decomposition analysis that systematically links cropland
demand per capita with two basic and consistent factors: per-capita
food consumption, measured as a food index that reflects the share of
animal products in diets, and land-use efficiency (measured as cropland
demand per food index). For the international comparison, countries
are grouped in four groups, based on the per capita income (World
Bank, 2019).
2. Methods and data
We analyze the global relationship between food consumption,
land-use efficiency and cropland demand using factorial decomposition
instruments combined with an analysis by groups of countries (based on
income) and international inequality through the Theil index (Duro and
Padilla, 2006; Theil, 1967; Duro et al., 2018; Schaffartzik et al., 2019).
The dataset underlying this study includes 123 countries, covering
~94% of the world population. The analysis covers the period
1990–2013. The section below describes indicators, how the individual
data were assessed, what data sources were used and outlines the de-
composition approaches.
2.1. Indicators and data sources
2.1.1. The food index
Adequate food supply depends not only on sufficient digestible food
energy but also on the adequacy of macro- and micronutrients, vita-
mins, fibre and many other criteria. In this study, however, we focus on
the relation between food availability and cropland demand. Therefore
we choose a simplified approach: Given a country’s level of food supply,
the associated cropland demand (see below) is assumed to primarily
depend on the amount of dry-matter respectively energy of primary
crops related to this food, and on output per unit area (cropland yield)
of the cropland where the crops are grown. We constructed a simple
national food index that aims to indicate the resource costs of food
production in terms of the trophic level of the consumed food. This
implies that we give larger weight to food sourced from animals (meat,
milk and eggs) compared to plant-based food. The construction of the
index reflects that during the time period of our analysis, each unit of
animal products required roughly the fivefold amount of crops from
arable land in terms of energy (Bouwman et al., 2005; Krausmann et al.,
2008; FAOSTAT, 2019). Accordingly, we differentiate total food supply
into plant-based and animal-based products and weighted the latter by
a factor of 5, resulting in the following formula:
= +F r r kcal( 5 )
i t v a total,
with kcal
total
as average per capita food supply in each country in kcal/
cap/day (from ref FAOSTAT), r
v
the share of vegetal products and r
a
the
share of animal products in the overall supply (r
v
+ r
a
= 1). Thus, the
food index is aimed to express the amount of primary biomass required
to provide the food consumed in a country and relates to energetic
relationships; other dimensions, such as the supply of micronutrients,
vitamins or fibres and their relationsships, are for the sake of simplicity
neglected.
2.1.2. Cropland demand for food and feed
We estimate cropland demand for food and feed anywhere on the
planet required to produce food consumed in each country, based on
data that link countries of crop cultivation to countries where the re-
lated products are used. Total cropland demand therefore reflects dif-
ferences in productivity in the crops and in countries that supply the
cropland products demanded in a country. These data rely on national-
level crop production data and bilateral trade data for crops and pro-
ducts processed from them (including livestock products, FAOSTAT,
2019). Matrix algebra is applied to the data to eliminate transit coun-
tries and to establish consistency between countries of cultivation and
use (for details see, Kastner et al., 2011, 2014).
To exclude cropland not used for food or feed production, we
multiply the obtained overall consumption data by the shares for food
and feed use in overall crop use (i.e. food and feed use plus non-food/
feed uses). These shares are derived from FAOSTAT’s commodity bal-
ance accounts (http://www.fao.org/faostat/en/#data/BC). Data on
country- and crop-specific yields are then used to translate the values
into cropland area, which is expressed as the area of cropland harvested
in a given year.
2.1.3. Population and country grouping
Population data were derived from the UN’s population database
(UN, 2019). The country groups are based on per capita income and
come from World Bank (World Bank, 2019).
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
2

2.2. Decompositions and territorial differences
As analytical approach we use multiplicative factorial decomposi-
tions. Transforming the factors into logarithms allowed decomposing
the trajectory of the global indicator into a sum of factors. In a first step,
we decompose cropland demand of a country i in year t (C
i,t
) into:
= =C C P P c P/
i t i t i t i t i t i t, , , , , ,
(1)
where P
i,t
is population in country i in year t. The first explanatory
factor we analyze is per capita cropland demand, as an indicator of
intensity, and the second, the country's population, as a basic indicator
of scale. Calculating natural logarithms (abbreviated as ln, i.e. loga-
rithms with the basis e) on both sides of (1) and differentiating two
periods (t and t-1) yields:
= +ln C ln C ln c ln c ln P ln P( ) ( )
i t i t i t i t i t it, , 1 , , 1 , 1
(2)
Therefore, we can algebraically distinguish which part of the
changes in C
it
between two time points is attributable to changes in
average per capita cropland demand (variable intensity) and to changes
in population (variable scale). This simple decomposition allows an
initial characterization of growth patterns in cropland use for food and
feed.
In a second step, we decompose per-capita cropland demand (c
i,t
)
and aim to answer to what extent increased per-capita food consump-
tion is counteracted by the efficiency of cropland use:
= =c C F F P e f/ /
i t i t i t i t i t i t i t
, , , , , , ,
(3)
where F
i,t
is the food index explained in section 2.1 and all the other
variables are as in Eq. (1). In this formulation, e
i,t
is cropland demand
per unit of food which may be interpreted as the inverse of food pro-
duction efficiency. f
i,t
is food supply (measured with the food supply
index) per capita in the respective country and year. In logarithmic
form, Eq. (3) can be written as
= +ln c ln c ln e ln e ln f ln f( ) ( )
i t i t i t i t i t i t
, , 1 , , 1 , , 1
(4)
Beyond evaluating global patterns, this framework also allows to
investigate international differences in a synthetic way using inequality
indicators. We summarize inequality using a synthetic index for c
i,t
and
decomposing it into the sum of the previous two factors. As Duro and
Padilla (2006) have shown, this can be done using the Theil index
(Theil, 1967) as a reference indicator for measuring inequalities in
vector c. The Theil index T originates from the concept of entropy of
information and coincides with a measure of average logarithmic de-
viation that has many desirable properties (Bourguignon, 1979). While
there are other interesting inequality measures such as the Gini index,
the Atkinson indicator family or the variation coefficient (Duro, 2012),
the Theil index differs from those by being easily decomposable
(Shorrocks and Wan, 2005), in particular when the factors are multi-
plicative, as in our case in (3). The algebraic expression of the Theil
index for the analysis of cropland demand per capita by countries, is
given as
=
=
Tµ
(c , t) p ln (c )
c
i
i 1
n
i,t
i,t
i,t
(5)
In this formula, p
i,t
is the population share of country i (that is, its
population as a fraction of the population of all countries in the
sample). c
i,t
is the cropland demand per capita in country i and year t. μ
is the world average in the cropland demand per capita, and ln is the
natural logarithm. Lower index values imply lower inequalities. The
minimum value of the Theil index is zero and the maximum value is not
generally defined
1
.
It has been shown (Duro and Padilla, 2006) that the international
inequality index T, in the context of the two-factor decomposition in
(3), can be decomposed in an interesting way. On this regard, we would
define two ficticious vectors of cropland demand per capita (c
i
), where
in each case only one factor is allowed to vary (and the other would be
equalized to the world mean). Thus, we used the following hypothetical
factors:
=c e f
e i t i t t
, , , ,
(6)
=c e f
f i t t i t
, , , ,
(7)
At this point, if we calculate the inequality in terms of these two
ficticious vectors of cropland demand per capita (using the Theil index)
and we compare their sum with the global Theil index (following the
expression (5)) we get the following consistent decomposition:
= + + +T (c , t) T(c ) T (c ) log(1 c)
i e f
ei,t,fi, t
e
(8)
where c
e
is the fictitious vector of cropland demand per capita that
would emerge from the hypothetical assumption that f
i,t
were equal to
the world average for all countries. Similarly, c
f
is a second fictitious
vector of cropland demand per capita where it is assumed that effi-
ciency e
i,t
(cropland demand per food index) would hypothetically co-
incide with the world average. σ is the population-weighted covariance
between the two factors e
i,t
and f
i,t
.
In this formulation, therefore, the importance attributable to each
factor can be interpreted as the amount of inequality that would exist if
only the factor examined were allowed to vary between countries,
while the other factor is assumed to equal the global average. The total
cross-country inequality in cropland demand per capita can thereby be
decomposed in a perfect way into two indices that reflect the partial
contribution of each of the factors to the global inequality and a last
factor of interaction that is represented by the interfactorial correla-
tion
2
.
Another useful possibility to decompose T(c
i
, t) is by groups
(Shorrocks, 1984:Duro and Padilla, 2006). In this case, global in-
equality is divided into (1) a between-group inequality component
(assuming zero inequality within groups) and (2) a within-group com-
ponent (assuming no disparities between groups). In our case, the
groups of countries by income levels have been used for the descriptive
analysis, but there are other possibilities. In this regard, the Theil index
has been shown to be the best candidate for this type of decomposition
and its interpretation (Shorrocks and Wan, 2005). Thus, for example,
the algebraic expression of this decomposition for factor our factor c
would be given by the following formula:
= +
= =
Tµ
(c , t) p ln (c )
cp T(c )
i
k 1
m
k,t
i,t
k,t k 1
m
k,t k,t
(9)
where k is the number of groups; p
k,t
is the relative population of each
group; m
k
the mean of each group; and T(c
k
) the Theil index applied to
each group.
The first summatory in (9) would be the between-groups inequality
component and the second one the within-group component. In fact,
the role of the between-groups inequality component can be perceived
as an indicator about the empirical rellevance of the groups as de-
scriptive units.
1
Many inequality indices do not have a fixed maximum value, as the max-
imum will depend on the respective sample (like is the case for the Atkinson
family indices or the variation coefficient). Nevertheless, different illustrative
(footnote continued)
examples with our type of data (by countries) indicate that index values near to
1 indicate high values of inequality. Detailed results can be made available by
the authors upon reasonable request.
2
Note further that if
eit fit
ce
,
is sufficiently small, the decomposition could
approach
= + +T c t T c T c( , ) ( ) ( )
i e f
eit fit
ce
,
And that if the covariance term is
small, the expression is simplified to a sum of factorial Theils
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
3

3. Results
Global cropland demand for food production is shown in Fig. 1a,
broken down into the four income groups. Total cropland demand grew
unevenly over the period 1990–2013. It remained stable until 2002 and
grew faster thereafter. After 2002, we note a significant plus of ~1 mio.
km
2
of cropland for food (+9%). The trajectory of cropland demand is,
however, heterogeneous across the income groups. The increases are
small or negative in the two richer groups, while they amount to +25%
in the low-middle and +78% in the low-income group.
The first decomposition according to eq. (1) is shown in Fig. 1b. The
study period is broken down into two phases, 1990–2000 and
2000–2013, which are displayed separately for each of the income
groups as well as for the global aggregate. The decomposition dis-
aggregates total cropland demand into cropland demand per capita and
population growth. Cropland demand per capita alone, a function of
yields per unit area and food demand per capita (expression (3)), would
have reduced cropland demand almost universally, with the exception
of low-income countries in the second part of the period. Hence po-
pulation growth emerges as the factor driving overall increases of
cropland demand. Reductions in cropland per capita were a lot smaller
in the second part of the period in the total as well as in three of the four
regions (i.e., except the high-income group).
These results are further analysed in Fig. 2 (focusing on the per
capita values). Fig. 2a reveals that cropland demand per capita follows
a curvilinear pattern in two of the four world regions as well as the total
of all countries. The total falls from ~2000 m
2
/cap and seems to sta-
bilize around 1700 m
2
/cap towards the end of the study period. Upper-
middle countries trajectory closely matches the global aggregate tra-
jectory, lower-middle contries are below the average level and high
income as well as low income groups are higher than the global
average. The high-income trend shows a rather linear, decreasing trend.
In contrast, low-income countries have the highest per-capita cropland
demand which decreases only during the first part of the period and
increases thereafter, without reaching the same level as in 1990.
Inequalities analysed in Fig. 2b refer to inequalities between coun-
tries, not only between groups. Here the Theil index (our referential
inequality measure) refers to distances of cropland demand per capita
in each country to the mean of all countries. Results reveal that these
distances decline over the study period, suggesting a global con-
vergence in per capita cropland demand. However, the convergence
shows a decreasing rate after 2010, suggesting that in the last years of
analysis international inequality remains almost constant.
Following the identity (3), the changes in the food index can play a
role in driving up global cropland demand and, in particular, cropland
per capita. Fig. 3a shows its evolution by income groups. The evidence
suggests that the food index grows in all groups, except in high-income
countries. While the food index remains largely constant around 7
Mcal/cap/day throughout the study period in the latter group, it rises
rapidly in the higher-middle group and more slowly in the lower-
middle and, mainly, in the low-income group. Consistently, the inter-
national distances in food indices (using all the countries and their
differences regarding to the world mean), as represented by the Theil
index, have decreased considerably, but most of this trend towards
reduced inequality could be observed in the first few years, followed by
a slight increase until 2004, and then further slow decreases until the
end of the study period (Fig. 3b)
3
. Moreover, the income groups used
are very rellevant in explaining global food inequalities, given that the
between-groups inequality component, following formula (9), accounts
for a 80% in 2013.
The second identity factor that can explain the trajectory of crop-
land demand per capita, and also the globals, in our context is the land-
use efficiency in producing food (i.e. food index/cropland demand;
Fig. 4). Specifically, the reduction of this factor would indicate greater
efficiency in the production of food. If measured in this manner, land-
use efficiency has improved over the period, which has consequently
(a)
C (cropland demand), 1000 km2 and % growth
0
2000
4000
6000
8000
10000
12000
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Low-Income countries Low-Middle Upper Middle High income
+78,4%
+25,1%
+1,9%
-5,8%
+10,9%
Source: own elaboration based on FAOSTAT data
(b)
% growth
-0,3
-0,2
-0,1
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
cropland/capita population
low-income
high-income
lower-middle
upper-middle
world
.
.
..
.
.
.
.
...
.
.
.
Note: the dots show the resulting overall change in cropland per capita
Source: own elaboration based on FAOSTAT data
Fig. 1. Changes in cropland demand by income groups, 1990–2013. (a)
Changes in total cropland demand, (b) decomposition of changes in cropland
demand into population growth and cropland demand per capita according to
Eq. (1). The unit in (a) is 1000 km
2
of cropland area harvested and % growth
over the entire period 1990–2013, in (b) relative changes in % over each spe-
cified period. Note that global cropland is about 15.2 Mkm
2
in 2013, including
food and feed provisiponing areas reflected here as well as the cropland used for
the production of non-food goods and fallowed cropland areas. Note: the dots
show the resulting overall change in cropland per capita Source: own ela-
boration based on FAOSTAT data.
3
Previous literature on food inquality has used the Gini coefficient as a re-
ferential inequality mesure (Seekell et al, 2011;Carr et al, 2016). Both mea-
sures, the Gini coefficient and the Theil index are satisfactory relative indexes,
given that they meet some basic well-known desirable properties. Although its
evolution is often similar, we have selected the Theil index given its greatest
advantages to be decomposed by factors, in particular in a multiplicative way as
in identity (3). Additionally, the Theil index is better than the Gini index when
we want to break down global inequality by groups, for example, in our case by
countries’groups like in expression (9) (Shorrocks and Wan, 2005). In any case,
the utilization of both indexes in our work (like in Fig. 3b) doesn’t yield sig-
nificant differences. Calculations are available from the authors upon reason-
able request.
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
4

tended to reduce cropland demand. However, the almost linear
downward trend stops in 2010, in which year land-use efficiency sta-
bilizes. The clearest increase in efficiency (i.e. reduction of cropland
demand per food index) has occurred in the middle-income countries.
In the case of the low-income group, the increase in efficiency stops in
2009 and in regard to high-income the increase in efficiency was not
significant since 2000. Consequently, the saturation in efficiency im-
provements in recent years, together with the rise in food consumption
(as measured by the food index), has contributed to the resurgence of
growing cropland demand per capita since 2000, which, combined with
the increase in population, has resulted in the final recent large increase
in total cropland demand (Fig. 1a)
4
.
In fact, putting together both factors (food index and area effi-
ciency), and using the decomposition based on equation (3),Fig. 5a
reveals which part of the changes of cropland demand per capita in
each group results from changes in yields/area (cropland demand per
food supply index) and which part from changes in food demand
(diets), as assessed with the food index. At the level of all countries,
(a)
c (cropland demand per capita), m²/cap
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Low-Income L ow-Middle Upper Middle High income World
Source: own elaboration based on FAOSTAT data
(b)
T(c),Theil indices
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0.11
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Source: own elaboration
Note: in points the a trend line (polynomial specification degree 2)
Fig. 2. Trajectories for 1990–2013 of (a) cropland demand per capita and (b) differences of cropland demand per capita between countries in the global average.
Units: (a) m
2
/capita (b) Theil index (dimensionless). Income groups shown in (a) explain 17% of the national-level inequalities analysed in (b). Source: own
elaboration based on FAOSTAT data. Source: own elaboration. Note: in points the a trend line (polynomial specification degree 2).
4
Also in this case, the income groups are relevant in explaining global cross-
country inequalities in terms of the area efficieny indicator. Specifically, the
between-groups inequality component explains a 38% of total inequalities (see
formula 9). More details are available upon request
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
5

increased food supply alone would have resulted in stark increases in
cropland demand, which was overcompensated by even stronger re-
ductions in cropland demand per unit of food. It is noteworthy that the
growth of land-use efficiency was stronger during the first part of the
study period in the global aggregate and in all groups. Growth in food
supply (as assessed by the food index) was strong throughout the entire
period in both the lower-middle and the higher-middle group, whereas
it was more or less absent in the high-income group, where small
growth in the first period was compensated by a small reduction in the
second half and food supply remained stable at ~7 Mcal/cap/day. Low-
income countries show almost no increase in the food index in the first
part of the study period, followed by some increase in the second part,
while area-efficiency was practically stagnant in the second part of the
period.
In addition to the analysis in levels, its seems interesting to extend
this decomposition for the analysis of inequalities (using all countries
data). On this regard, the decomposition on international cropland
demand per capita inequalities (Fig. 2b) in terms of both factors
(cropland demand per unit of food, and as food per capita) reveals two
trends (Fig. 5b). First, the importance of the efficiency factor is cur-
rently larger than that of the food factor for explaining cross-country
differences in cropland demand per capita around the world. Second,
while the convergence in the food index per capita has reduced its role
in explaining global international differences in cropland demand, the
efficiencies have not been converging.
4. Discussion
We have analyzed the trajectories and the role of the intercountry
inequalities of cropland demand, decomposed to the parameters per-
2
3
4
5
6
7
8
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Low-Income Low-Middle Upper Middle High income World
Source: FAOSTAT and applying the food index formula, as explained in the method section.
a)
f (per capita food supply), Mcal/cap/day
b)
T(f), Theil index
0.0400
0.0450
0.0500
0.0550
0.0600
0.0650
0.0700
0.0750
0.0800
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Source: own elaboration
Note: in points the estimated trend line (polynomial specification degree 2)
Fig. 3. Food supply per capita. (a) Trajectories of per-capita food supply, as represented by the food index, in the study period 1990–2013 for country groups and the
total of all countries, (b) Theil indices of food consumption, given as food index per capita. Units: (a) food index [Mcal/cap/day], (b) dimensionless Theil index of
inequality in per-capita food index. Source: own elaboration. Note: in points the estimated trend line (polynomial specification degree 2).
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
6

capita food consumption (measured by the food index) and area effi-
ciency of food production which we estimated as global cropland de-
mand of each country per unit of food supplied (measured as food
index). We first discuss the trends in inequality of the food index in
various country groups and then possible future trajectories of global
cropland demand.
The global state of food security, in which under- and overnutrition
occur simultaneously, calls for a convergence towards healthy and
adequate diets (Beal et al., 2017;Tilman and Clark, 2014; Kummu
et al., 2017; Willett et al., 2019) in order to decrease environmental
pressures. Such a convergence would require the quantitative and
qualitative increase of food supply for many regions, mainly in the
Global South, concomitant with a reduction in overconsumption in
other world regions, mainly in the industrialized North. Indeed, we
observe a marked increase of the food index in the lowest and both
middle income groups of countries. The food index in the high-income
country group is not falling, however. Here the food index remains
remarkably stable over the entire period. Thus, while inequality related
to the food index is improving over time, it does so only at moderate
rates, and seems to converge towards the resource-intensive consump-
tion pattern of the richest country group. A short period around the
year 2000 even shows an increase in inequality, a trend that is reversed
again afterwards (see Figs. 3a and 3b).
The most pronounced upward trend of the food index is found in the
upper-middle country group, basically explained by trajectories in
China (note that the analysis was performed a the national level, but
a)
e (cropland are per food index), m2/kcal/day
200
300
400
500
600
700
800
900
1000
1100
Low-Income Low-Middle Upper Middle High income World
Source: own elaboration based on data from FAOSTAT
b)
T(e), Theil index
0.000
0.020
0.040
0.060
0.080
0.100
0.120
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Source: own elaboration based on data from FAOSTAT
Note: in points the a trend line (polynomial specification degree 2)
Fig. 4. Area efficiency (cropland area per food index) 1990–2013. (a) Trajectories of individual income group (b) Theil indices. Units: (a) cropland demand per food
supply m
2
/kcal/day; b) dimensionless Theil index of inequality in per-capita cropland demand. Source: own elaboration based on data from FAOSTAT. Note: in
points the a trend line (polynomial specification degree 2).
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
7

only data at the country-group-level are displayed). For this group of
countries, the food index increased almost linearly by 31% over the
entire period, reaching levels of 83% of the high-income group; the
food index of China even grows by 62%. If the current trends were to
continue, the upper-middle income group would reach the food index
of the highest income group in around 2030. The growth of the food
index of the lower middle income group was 21% over the entire
period, but their level is still only 52% of the high-income countries. In
this case, the increase has affected most countries in the group, al-
though large countries in this group have achieved higher increases,
e.g. Vietnam (95%), Myanmar (110%) and Ghana (65%).
The smallest growth in the food index occurred in the low-income
group of countries. Here, the food index grew on average by only 12%;
at that rate, their catchup to the highest-income group would require
170 years. This slow convergence of the two lower income groups, in
particular the low-income group, is a reason for concern: the poor re-
mained poor, and are bound to continue to do so for a long time if
current trajectories do not change. Significant growth of the per-capita
cropland demand would result from a continuation of the trends
observed in the three lower income groups towards the food index of
the rich countries. Cropland area efficiency would have to grow by 40%
over the level of 2013 until 2050 to compensate for a global con-
vergences towards the food index currently observed in the high-in-
come group of countries.
Food index trends in the observed period were only partly com-
pensated by increases in cropland use efficiency (food index per m
2
of
cropland) at the global scale. Overall, we found an increase of global
cropland demand in the last two decades of +11%, which progressed
most rapidly after 2002 (Fig. 1a). The decomposition shows that
growing cropland area efficiency and the stabilization of the food index
in the highest income group of countries was overwhelmed by popu-
lation growth and a growing food index in the lower income groups of
countries (Fig. 1b). In this context, efforts to close yield gaps without
raising ecological pressures have been proposed to supply an increasing
world population with sufficient food while requiring as little cropland
as possible and avoidiong the manifold socio-ecological detriments re-
lated to mainstream intensification (Loos et al., 2014; Garnett et al.,
2013). While crop yields seem to have reached a plateau in some
(a)
Percentages
-50.0%
-40.0%
-30.0%
-20.0%
-10.0%
0.0%
10.0%
20.0%
30.0%
40.0%
1990-2000
2000-2013
1990-2013
1990-2000
2000-2013
1990-2013
1990-2000
2000-2013
1990-2013
1990-2000
2000-2013
1990-2013
1990-2000
2000-2013
1990-2013
cropland/food
food/capita
Low-income
Lower-income
Upper-Middle
High-income
World
Source: own elaboration based on FAOSTAT data
(b)
Theil indices
-
0.020
0.040
0.060
0.080
0.100
0.120
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
cropland/capita inequalies
efficiency factor
food factor
Note: using Theils decomposition like expression (6)
Source: own elaboration
Fig. 5. Decomposition of changes in cropland demand per capita for the period 1990–2013. (a) Decomposition by income groups, (b) trajectory of the cross-country
differences as Theil indices of efficiency (cropland demand per food index) and diets (food index per capita). Units are (a) percent changes for each specified period of
cropland demand (m
2
) per food index (weighted kcal/cap/day) and food index/capita and (b) inequality in cropland per capita, the efficiency and the food factor,
dimensionless Theil indices. Note: using Theils decomposition like expression (6). Source: own elaboration.
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
8

countries with currently highest crop yields, large yield gaps prevail
especially in many of the countries with currently low yields (Lobell
et al., 2009). Our analysis reveals that the countries in the lowest in-
come group have the largest cropland demand per capita, while their
food index is lowest (Fig. 2a and 3a), implying very low land-use effi-
ciency. High yield gaps that probably exist in many of these countries
are the most likely explanation, and technology transfer may be a good
strategy to address that situation (Tilman et al., 2011). Of course, de-
creasing the inequalities beween low and high income regions in terms
of both diets and crop yields would also be highly beneficial in this
context. However, given current conditions of widely prevailing eco-
logically unequal exchange (Jorgenson, 2010). It seems uncertain
whether this will happen any time soon. An analysis of the extent to
which these current patterns are result of unequal exchange was beyond
the scope of our analysis.
Interestingly, our analysis reveals that the country-level inequality
in area efficiency is not changing significantly over time. Rather, all
regions show similar relative increases of yields (Fig. 4a and b). As crop
yields seem to be stagnating in some countries with the highest yields
(Ray et al., 2012), the question remains for how long this trend of
globally rather uniform increases in cropland efficiency can continue in
the next decades. Theoretical yield potentials are determined by cli-
matic factors and the light-use efficiency of photosynthesis. Some
agronomists argue that average yields in each country will not surpass
70% to 80% of maximum yields, beyond which costs of marginal yield
increments exceed incremental economic gain (Lobell et al. 2009). It is
therefore of paramount importance to preferentially raise the the low
crop yields in the poorer regions. However, although this strategy ad-
ditionally has been described as holding large potentials to increase
global nitrogen-efficiency and thus decrease global environmental
pressures (Niedertscheider et al., 2016; Mueller et al., 2012), it could
not be observed in our analysis. This not only indirectly points to trends
of decreased N-efficiency (Mueller et al., 2017), but also underlines that
closing of yield gaps in poorer regions will not be achieved by a con-
tinuation of past policies or strategies (van Ittersum et al., 2016). This is
so because when all regions show the same improvements, the yield-
gap frontier for countries is growing as well (Niedertscheider et al.,
2016;Erb et al., 2014).
In low-income countries, population increases drive total cropland
demand despite improvements in per-capita cropland demand. In the
lower-middle and upper middle countries, significant population
growth has been partially offset by a reduction of per-capita cropland
demand (Fig. 1b). If global trends do not change in terms of population
growth in lower income countries concomitant with relatively low re-
ductions in per-capita cropland demand, global cropland demand is
bound to rise substantially in the next decades. For example, if we
combine the population prospects for these country groups for 2050
and 2100 (UN, 2019) with per-capita levels of cropland demand in
2013 (i.e., assuming no improvements in per capita levels as the last
years trend), the total cropland demand in 2050 would grow by 38%
compared to 2013 levels, and by 58% until 2100. This is about the
extent of the estimated upper limits of well-suited global land well
suited for cropland (Ramankutty et al., 2002; Coelho et al., 2012), and
would likely result in severe problems of making sufficient grazing land
areas available without deforestation (Erb et al., 2016).
Our results show values on inequality, using the Theil index, near to
0.1 (Fig. 5b). These values are however much lower than e.g. in-
equalities for income and per capita CO
2
emissions, where Theil indices
are 0.4 and 0.6, respectively (own calculations using IEA data for
2013). This can be linked to the central role that food provision and
consumption holds for any society. Nevertheless, and in spite the pre-
vious argument, the inequality values are not irrelevant and they will
need to be reduced. Thus, if we compute alternatively the Gini coeffi-
cient for measuring the food index international inequalities (which
ranges from 0 to 1), the value for 2013 will indicate that these dis-
parities would account a 20% of a maximum inequality scenario (Theil,
1967; Duro and Padilla, 2006; Duro, 2012).
As a caveat, we are of course aware that the simple food index we
constructed by putting different weights to the share of plant an animal
products expresses only one qualitative aspect of food provision, i.e. the
primary biomass equivalent demand. Other aspects, such as the share of
sugar or highly processed food commodities in diets, or nutritional
values, are not grasped. It is difficult to judge if a food index that would
also represent other qualitative differences would change the general
insight of our study. The analysis by Tilman and Clark (2014) suggests
that the share of animal products and of highly refined food are posi-
tively interrelated, which suggests that – while such an index would
certainly be more informative – it would probably not drastically affect
our overall conclusions. Another caveat is that our methods did not
allow investigating within-country inequalities, which also need to be
considered to get a more complete perspective on food security
(d’Odorico et al., 2019).
5. Conclusion
To sum up, while the food index showed some moderate trends
towards a more equal distribution among countries, mainly explained
by the two middle income groups of countries, a similar trend could not
be found for cropland area efficiency (food index per unit of cropland).
For low-income countries, the trends of food demand and area effi-
ciency are of similar magnitude and cancel each other, resulting in a
stagnating per-capita cropland demand and hence rises in cropland
demand roughly in accordance with population growth. In the other
income groups, area efficiency gains overcompensated increases of the
food index (middle groups) respectively helped to reduce per capita
cropland demand at stagnating food index values. Thus, the relatively
reduction in inequality of per-capita cropland demand is not a result of
a global contract-and-converge trajectory, but strongly determined by
the middle and high income groups. A massive sustainability challenge
relates to low-income countries that face highly dynamic population
trajectories paired with an apparent difficulty in increase agricultural
yields in a situation of low purchasing power. Combating world hunger
and reducing pressures on the land system from rising cropland areas
will need to focus on that group if a more sustainable development of
the global food system is to be achieved.
CRediT authorship contribution statement
Juan Antonio Duro: Conceptualization, Methodology, Formal
analysis, Writing - review & editing. Christian Lauk:
Conceptualization, Methodology, Writing - review & editing. Thomas
Kastner: Conceptualization, Methodology, Writing - review & editing.
Karl-Heinz Erb: .Helmut Haberl: Conceptualization, Methodology,
Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
The research was funded by the Austrian Science Fund (FWF),
project GELUC, no. P29130-G27, ERA-NET SusAn project Animal-
Futures (BMNT 101243/1) and Ministerio de Economía y Competitidad
(Spain) ECO2016-79072-P (AEI/FEDER, UE).
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
9

Appendix
A1. Groups
Low-Income: Benin, Burkina, Central African, Chad, Gambia, Guinea, Guinea-Bissau, Haiti, Liberia, Madagascar, Malawi, Mali, Mozambique,
Nepal, Niger, Rwanda, Senegal, Sierra Leone, Togo, Uganda, Tanzania, Yemen and Zimbabwe.
Low-Middle: Angola, Bangla-D, Bolivia, Cape Verde, Cambodia, Cameroon, Congo, Cote d'Ivoire, Djibouti, Egypt, El Salvador, Ghana, Honduras,
India, Indonesia, Kenya, Lao, Mauritania, MOngolia, Morocco, Myanmar, NIcaragua, Nigeria, Pakistan, Phillipines, Sri Lanka, Swazilandia, Tunisia,
Vietnam and Zambia.
Upper-Middle: Albania, Algerie, Botswana, China, Colombia, Costa Rica, Dominican R., Ecuador, Fiji, Gabon, Grenade, Guatemala, Iran, IRak,
Jamaica, Jordan, Lebanon, Malaysia, Maldives, Mauritania, MExico, Romania, Saint Lucia, South Africa, Surinam, Thailandia, Turkey, USSR,
Venezuela and Yugoslavia.
Higher: Argentina, Australia, Austria, Barbados, Belgium-Lux, Brunei, Canada, Chile, Cyprus, Czecoslovakia, Denmark, Finland, France,
Germany, Greece, Hungary, Iceland, Ireland, Island, Italy, Japan, Kuwait, Malta, Netherlands, New Zealand, Norway, Panama, Poland, Portugal,
Korea (Rep), Saudi Arabia, Spain, Switzerland, Trinidad, United Arab E., Uk and USA.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.gloenvcha.2020.102124.
References
Alexander, P., Rounsevell, M.D., Dislich, C., Dodson, J.R., Engström, K., Moran, D., 2015.
Drivers for global agricultural land use change: The nexus of diet, population, yield
and bioenergy. Global Environ. Change 35, 138–147.
Beal, T., Massiot, E., Arsenault, J.E., Smith, M.R., Hijmans, R.J., 2017. Global trends in
dietary micronutrient supplies and estimated prevalence of inadequate intakes. PLoS
One San Franc. 12. https://doi.org/10.1371/journal.pone.0175554.
Bourguignon, F., 1979. Decomposable income inequality measures. Econometrica 47,
901–920.
Bouwman, A.F., Van der Hoek, K.W., Eickhout, B., Soenario, I., 2005. Exploring changes
in world ruminant production systems. Agric. Syst. 84, 121–153.
Carr, J.A., D’Odorico, P., Suweis, S., Seekell, D.A., 2016. What commodities and countries
impact inequality in the global food system? Environ. Res. Lett. 11.
Coelho, S., Agbenyega, O., Agostini, A., Erb, K.H., Haberl, H., Hoogwijk, M., Lal, R.,
Lucon, O., Masera, O., Moreira, J.R., 2012. Land and water: linkages to bioenergy. In:
Johansson, T., Patwardhan, A., Nakicenonivc, N., Gomez-Echeverri (Eds.), Global
Energy Assessment: Toward a Sustainable Future. International Institute of Applied
Systems Analysis (IIASA), Cambridge University Press, Cambridge, UK, pp.
1459–1525.
D’ D'Odorico, P., Carr, J.A., Davis, K.F., Dell’Angelo, J., Seekell, D.A., 2019. Food in-
equality, injustice, and rights. Bioscience 69, 180–190.
D’Odorico, P., Carr, J.A., Laio, F., Ridolfi, L., Vandoni, S., 2014. Feeding humanity
through global food trade. Earth’s Future 2, 458–469. https://doi.org/10.1002/
2014EF000250.
Duro, J.A., Padilla, E., 2006. International inequalities in per capita CO2 emissions: a
decomposition methodology by Kaya factors. Energy Econ. 28, 170–187. https://doi.
org/10.1016/j.eneco.2005.12.004.
Duro, J.A., 2012. On the automatic application of inequality indexes in the analysis of the
international distribution of environmental indicators. Ecol. Econ. 76, 1–7.
Duro, J.A., Schaffartzik, A., Krausmann, F., 2018. Metabolic inequality and its impact on
efficient contraction and convergence of international material resource use. Ecol.
Econ. 145, 430–440. https://doi.org/10.1016/j.ecolecon.2017.11.029.
Er Erb, K.-H., 2004. Actual land demand of Austria 1926–2000: a variation on Ecological
Footprint assessments. Land Use Policy 21, 247–259. https://doi.org/10.1016/j.
landusepol.2003.10.010.
Erb, K.-H., Gaube, V., Krausmann, F., Plutzar, C., Bondeau, A., Haberl, H., 2007. A
comprehensive global 5 min resolution land-use data set for the year 2000 consistent
with national census data. J. Land Use Sci. 2, 191–224. https://doi.org/10.1080/
17474230701622981.
Erb, K.-H., Niedertscheider, M., Dietrich, J.-P., Schmitz, C., Verburg, P.H., Jepsen, M.R.,
Haberl, H., 2014. Conceptual and Empirical Approaches to Mapping and Quantifying
Land-Use Intensity. In: Fischer-Kowalski, M., Reenberg, A., Schaffartzik, A., Mayer, A.
(Eds.), Ester Boserup’s Legacy on Sustainability. Orientations for Contemporary
Research, Springer, Dordrecht, pp. 61–86.
Erb, K.-H., Lauk, C., Kastner, T., Mayer, A., Theurl, M.C., Haberl, H., 2016. Exploring the
biophysical option space for feeding the world without deforestation. Nat. Commun.
7, 11382. https://doi.org/10.1038/ncomms11382.
FAO, IFAD, UNICEF, WFP, WHO, 2017. The State of Food Security and Nutrition in the
World 2017. Building resilience for peace and food security. Food and Agriculture
Organisation (FAO), Rome.
FAO, IFAD, UNICEF, WFP and WHO. 2019. The State of Food Security and Nutrition in
the World 2019. Safeguarding against economic slowdowns anddownturns (FAO),
Rome.
FAOSTAT, 2019. Statistical Database. Food and Agriculture Organization of the United
Nations, Rome.
Garnett, T., Appleby, M.C., Balmford, A., Bateman, I.J., Benton, T.G., Bloomer, P.,
Burlingame, B., Dawkins, M., Dolan, L., Fraser, D., Herrero, M., Hoffmann, I., Smith,
P., Thornton, P.K., Toulmin, C., Vermeulen, S.J., Godfray, H.C.J., 2013. Sustainable
intensification in agriculture: premises and policies. Science 341, 33–34.
Godfray, H.C.J., Beddington, J.R., Crute, I.R., Haddad, L., Lawrence, D., Muir, J.F., Pretty,
J., Robinson, S., Thomas, S.M., Toulmin, C., 2010. Food security: the challenge of
feeding 9 billion people. Science 327, 812–818. https://doi.org/10.1126/science.
1185383.
Griggs, D., Stafford-Smith, M., Gaffney, O., Rockström, J., Öhman, M.C., Shyamsundar,
P., Steffen, W., Glaser, G., Kanie, N., Noble, I., 2013. Policy: sustainable development
goals for people and planet. Nature 495, 305–307. https://doi.org/10.1038/
495305a.
Haberl, H., Erb, K.H., 2017. Land as a planetary boundary – a socioecological perspective.
In: Victor, P.A., Dolter, B. (Eds.), Handbook on Growth and Sustainability. Edward
Elgar, Cheltenham, UK and Northampton, MA, USA.
Haberl, H., Steinberger, J.K., Plutzar, C., Erb, K.-H., Gaube, V., Gingrich, S., Krausmann,
F., 2012. Natural and socioeconomic determinants of the embodied human appro-
priation of net primary production and its relation to other resource use indicators.
Ecol. Ind. 23, 222–231. https://doi.org/10.1016/j.ecolind.2012.03.027.
IAASTD, 2009. Agriculture at a Crossroads. International Assessment of Agricultural
Knowledge, Science and Technology for Development (IAASTD), Global Report.
Island Press, Washington, D.C.
Ittersum, M.K. van, Bussel, L.G.J. van, Wolf, J., Grassini, P., Wart, J. van, Guilpart, N.,
Table A1
Factors and income-groups.
1990 2000 2013
c e f c e f c e f
Low-Income 2568,4 963,2 2,7 2268,8 845,0 2,7 2461,9 823,4 3,0
Low-Middle 1733,4 599,9 2,9 1562,0 490,7 3,2 1459,3 404,0 3,6
Upper Middle 2039,5 461,8 4,4 1744,0 356,4 4,9 1706,1 294,9 5,8
High income 2339,0 335,3 7,0 2156,6 302,3 7,1 1877,4 268,4 7,0
World 2011,6 466,3 4,3 1775,1 388,2 4,6 1680,5 336,3 5,0
Note: c is cropland per capita (m
2
/cap); e is cropland area per food index (m
2
/kcal/day); f is food supply per capita (Mcal/cap/day).
Source: own elaboration based on data from FAOSTAT.
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
10

Claessens, L., Groot, H. de, Wiebe, K., Mason-D’Croz, D., Yang, H., Boogaard, H.,
Oort, P.A.J. van, Loon, M.P. van, Saito, K., Adimo, O., Adjei-Nsiah, S., Agali, A., Bala,
A., Chikowo, R., Kaizzi, K., Kouressy, M., Makoi, J.H.J.R., Ouattara, K., Tesfaye, K.,
Cassman, K.G., 2016. Can sub-Saharan Africa feed itself? PNAS 113, 14964–14969.
Jorgenson, A.K., 2010. World-economic integration, supply depots, and environmental
degradation: a study of ecologically unequal exchange, foreign investment depen-
dence, and deforestation in less developed countries. Crit. Sociol. 36, 453–477.
https://doi.org/10.1177/0896920510365204.
Kastner, Thomas, Erb, Karl-Heinz, Haberl, Helmut, 2014. Rapid growth in agricultural
trade: effects on global area efficiency and the role of management. Environ. Res.
Lett. 9 (3).
Kastner, Thomas, Kastner, Michael, Nonhebel, Sanderine, 2011. Tracing distant en-
vironmental impacts of agricultural products from a consumer perspective. Ecol.
Econ. 70 (6), 1032–1040.
Krausmann, F., Erb, K.-H., Gingrich, S., Lauk, C., Haberl, H., 2008. Global patterns of
socioeconomic biomass flows in the year 2000: a comprehensive assessment of
supply, consumption and constraints. Ecol. Econ. 65, 471–487 https://doi.org/16/
j.ecolecon.2007.07.012.
Krausmann, F., Haberl, H., Erb, K.-H., Wiesinger, M., Gaube, V., Gingrich, S., 2009. What
determines geographical patterns of the global human appropriation of net primary
production? J. Land Use Sci. 4, 15–33. https://doi.org/10.1080/
17474230802645568.
Kummu, M., Fader, M., Gerten, D., Guillaume, J.H.A., Jalava, M., Jägermeyr, J., Pfister,
S., Porkka, M., Siebert, S., Varis, O., 2017. Bringing it all together: Linking measures
to secure nations' food supply. Curr. Opin. Environ. Sustain. 29, 98–117. https://doi.
org/10.1016/j.cosust.2018.01.006.
Lobell, D.B., Cassman, K.G., Field, C.B., 2009. Crop yield gaps: their importance, mag-
nitudes, and causes. Annu. Rev. Environ. Resour. 34, 179–204.
Loos, J., Abson, D.J., Chappell, M.J., Hanspach, J., Mikulcak, F., Tichit, M., Fischer, J.,
2014. Putting meaning back into “sustainable intensification”. Front. Ecol. Environ.
12, 356–361.
Meyer, A., 1999. The Kyoto protocol and the emergence of “Contraction and
Convergence” as a framework for an international political solution to greenhouse
gas emissions abatement. In: Hohmeyer, O., Rennings, K. (Eds.), Man-Made Climate
Change, ZEW Economic Studies. Physica-Verlag HD, Heidelberg, pp. 291–345.
https://doi.org/10.1007/978-3-642-47035-6_15.
Motesharrei, S., Rivas, J., Kalnay, E., Asrar, G.R., Busalacchi, A.J., Cahalan, R.F., Cane,
M.A., Colwell, R.R., Feng, K., Franklin, R.S., Hubacek, K., Miralles-Wilhelm, F.,
Miyoshi, T., Ruth, M., Sagdeev, R., Shirmohammadi, A., Shukla, J., Srebric, J.,
Yakovenko, V.M., Zeng, N., 2016. Modeling sustainability: population, inequality,
consumption, and bidirectional coupling of the Earth and Human Systems. Natl. Sci.
Rev. 3, 470–494. https://doi.org/10.1093/nsr/nww081.
Mueller, N.D., Gerber, J.S., Johnston, M., Ray, D.K., Ramankutty, N., Foley, J.A., 2012.
Closing yield gaps through nutrient and water management. Nature 490, 254–257.
Mueller, N.D., Lassaletta, L., Runck, B.C., Billen, G., Garnier, J., Gerber, J.S., 2017.
Declining spatial efficiency of global cropland nitrogen allocation. Global
Biogeochem. Cycles 31 2016GB005515.
Niedertscheider, M., Kastner, T., Fetzel, T., Haberl, H., Kroisleitner, C., Plutzar, Christoph,
Erb, K.-H., 2016. Mapping and analysing cropland use intensity from a NPP per-
spective. Environ. Res. Lett. 11.
Ramankutty, N., Foley, J.A., Norman, J., McSweeney, K., 2002. The global distribution of
cultivable lands: current patterns and sensitivity to possible climate change. Glob.
Ecol. Biogeogr. 11, 377–392.
Ray, D.K., Ramankutty, N., Mueller, N.D., West, P.C., Foley, J.A., 2012. Recent patterns of
crop yield growth and stagnation. Nat. Commun. 3, 1293.
Rockström, J., Steffen, W., Noone, K., Persson, A.A., Chapin, F.S., Lambin, E.F., Lenton, T.
M., Scheffer, M., Folke, C., Schellnhuber, H.J., others, 2009. A safe operating space
for humanity. Nature 461, 472–475.
Sahn, D.E., 2015. The Fight Against Hunger and Malnutrition, The Role of Food,
Agriculture and Targeted Policies. Oxford University Press, Oxford.
Schaffartzik, A., Duro, J.A., Krausmann, F., 2019. Global appropriation of resources
causes high international material inequality – growth is not the solution. Ecol. Econ.
163, 9–19. https://doi.org/10.1016/j.ecolecon.2019.05.008.
Seekell, D.A., D’Odorico, P., Pace, M.L., 2011. Virtual water transfers unlikely to redress
inequality in global water use. Environ. Res. Lett. 6.
Shorrocks, A., 1984. Inequality decomposition by population subgroups. Econometrica
52, 1369–1385.
Shorrocks, A., Wan, G., 2005. Spatial decomposition of inequality. J. Econ. Geogr. 5,
59–81.
Steffen, W., Richardson, K., Rockstrom, J., Cornell, S.E., Fetzer, I., Bennett, E.M., Biggs,
R., Carpenter, S.R., de Vries, W., de Wit, C.A., Folke, C., Gerten, D., Heinke, J., Mace,
G.M., Persson, L.M., Ramanathan, V., Reyers, B., Sorlin, S., 2015. Planetary bound-
aries: Guiding human development on a changing planet. Science 347, doi: 10.1126/
science.1259855. https://doi.org/10.1126/science.1259855.
Swinburn, B., Kraak, V., Rutter, H., Vandevijvere, S., Lobstein, T., Sacks, G., Gomes, F.,
Marsh, T., Magnusson, R., 2015. Strengthening of accountability systems to create
healthy food environments and reduce global obesity. Lancet 385, 2534–2545.
https://doi.org/10.1016/S0140-6736(14)61747-5.
Theil, H., 1967. Economics and Information Theory. North-Holland.
Tilman, D., Balzer, C., Hill, J., Befort, B.L., 2011. Global food demand and the sustainable
intensification of agriculture. Proc. Natl. Acad. Sci. 108 (50), 20260–20264.
Tilman, D., Clark, M., 2014. Global diets link environmental sustainability and human
health. Nature 515, 518–522.
UN, 2019. World Population Prospects 2019 [WWW Document]. URL https://population.
un.org/wpp/ (accessed 12.17.19).
United Nations, 2015. Transforming our world: the 2030 agenda for sustainable devel-
opment. General Assembley 70 session.
Wackernagel, W., M., Onisto, L., Bello, P., Callejas Linares, A., Susana López Falfán, I.,
Méndez Garcı́a, J., Isabel Suárez Guerrero, A., Guadalupe Suárez Guerrero, M., 1999.
National natural capital accounting with the ecological footprint concept. Ecol. Econ.
29, 375–390.
Willett, W., Rockström, J., Loken, B., Springmann, M., Lang, T., Vermeulen, S., Garnett,
T., Tilman, D., DeClerck, F., Wood, A., Jonell, M., Clark, M., Gordon, L.J., Fanzo, J.,
Hawkes, C., Zurayk, R., Rivera, J.A., De Vries, W., Majele Sibanda, L., Afshin, A.,
Chaudhary, A., Herrero, M., Agustina, R., Branca, F., Lartey, A., Fan, S., Crona, B.,
Fox, E., Bignet, V., Troell, M., Lindahl, T., Singh, S., Cornell, S.E., Srinath Reddy, K.,
Narain, S., Nishtar, S., Murray, C.J.L., 2019. Food in the anthropocene: the
EAT–Lancet Commission on healthy diets from sustainable food systems. Lancet.
https://doi.org/10.1016/S0140-6736(18)31788-4.
World Bank, 2019. World Bank Country and Lending Groups [WWW Document]. URL
https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-
country-and-lending-groups (accessed 12.17.19).
J.A. Duro, et al. Global Environmental Change 64 (2020) 102124
11