Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 87
ANNEX A
WHAT DO WE REALLY KNOW?
METRICS FOR FOOD INSECURITY AND MALNUTRITION
Hartwig de Haen, Stephan Klasen and Matin Qaim1
Paper presented at the workshop on Measuring Food Insecurity and Assessing the
Sustainability of Global Food Systems, February 16-17, 2011, Keck Center of the National
Academies, Washington, DC.
1. INTRODUCTION
Recently, the International Food Policy Research Institute (IFPRI) published an estimate
of hunger in 12 sub-Saharan African countries. Based on an analysis of household surveys the
authors found that in the late 1990s 59 percent of the population was food energy deficient
(Smith et al., 2006). This result was in stark contrast to estimates by the Food and Agriculture
Organization of the United Nations (FAO), based on food balance sheets for the same countries,
the same period and using the same criterion of energy deficiency as an indicator of
undernourishment. The FAO prevalence estimate was 39 percent (Smith et. al., 2006, p 45),
hence significantly lower. Not only did the two methods differ with respect to the mean level of
undernourishment, the ranking of the 12 countries differed as well. In other words, there is not
even a close correlation between the two estimates. This example of divergent estimates of
hunger, measured with the same criterion, namely food energy deficiency, suffices to raise
interest in a thorough comparative assessment of the various methods used to estimate hunger.
Numerous statistics are published reporting on the food security and nutrition situation at
global, country, household and individual levels. A comprehensive overview of available or
conceivable indicators can be found under the FAO-led Food Insecurity and Vulnerability
Information and Mapping Systems initiative, FIVIMS, (http://www.fivims.org). FIVIMS was
established, initially as an inter-agency initiative following the World Food Summit (WFS) in
1
The authors are Professors at the Georg-August-University of Göttingen. H. de Haen and M. Qaim are,
respectively, Emeritus and Chair, International Food Economics and Rural Development, at the Department of
Agricultural Economics and Rural Development. S. Klasen is Chair, Theoretical Economics and Development
Economics at the Faculty of Economic Sciences. The authors would like to thank Mark Smulders , FAO, and
participants at the workshop for helpful comments and suggestions.
87
OCR for page 88
88 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
1996, which determined that food security exists when all people, at all times, have physical and
economic access to sufficient, safe and nutritious food to meet their dietary needs and food
preferences for an active and healthy life (WFS Plan of Action, 1996). This widely accepted
definition underlines the multidimensional nature of food security, comprising level and stability
of food access and availability, adequacy of food use and food consumption and nutritional
status. Conversely, it explains that food insecurity, i.e. the absence of food security, can be the
result of very diverse factors.
The equally broad and overlapping concept of nutrition security determined by
dimensions of food, care and health, can be assessed through a number indicators, including
those measuring undernutrition as well as overnutrition. Per Pinstrup-Andersen talks about the
triple burden of malnutrition, differentiating between (i) food energy deficiency, (ii) deficiency
in specific nutrients, especially micronutrients, which are also key for an active and healthy life,
and (iii) excessive net energy intake leading to overweight and obesity (Pinstrup-Andersen,
2007). In view of this multi-facetted character of food insecurity and malnutrition, it is not
surprising that—when indicators measure different dimensions—the conclusions may also be
different from one indicator to another. However, where different methods are used to measure
the same phenomenon, one would expect only little, if any differences. The comparative
assessment provided in this paper intends to discuss the reasons for differences between methods
and indicators in more detail. While we recognize that the obesity problem is increasing,
including in developing countries, we concentrate primarily on measures of food deprivation and
undernutrition.
Obviously, before a specific food insecurity information and mapping system is set up,
clarification is needed, as to which aspect of food insecurity is to be measured in each particular
situation and by which indicator. Expressed in simple terms, people are deemed food insecure
when their consumption of food is insufficient, insecure and/or unsustainable (Maxwell and
Frankenberger, 1995). They live in hunger or fear of starvation. Although hunger is commonly
understood as a sensation of not having enough to eat, its definition and measurement are not at
all trivial. On the one hand, the extent of hunger can be measured as a lack of essential nutrients
in the diet. A widely used indicator for this is food energy deficiency. On the other hand, hunger
may also be the result of humans’ inability to absorb and use food energy and specific nutrients
for body functions, implying that the overall nutritional status may also be affected by people’s
health. Accordingly, the combined effects of access to food and of food absorption and use are
best measured through outcome indicators that inform about people’s actual nutrition status such
as undernutrition or overnutrition.
Before proceeding to discuss the advantages and disadvantages of these various
approaches to the topic, it is necessary to clarify the purpose of measurement. Two quite
different purposes can be distinguished. One is to be informed about the extent and consequences
of an actual food emergency caused by a sudden drop in supply or access to food. In such
situations, indicators must provide information about people’s immediate needs of essential
nutrients to ensure survival. Indicators must be easy and quick to measure and useful for the
design of humanitarian aid action. The second purpose relates to chronic food insecurity, caused
by long term food deprivation linked to structural poverty and poor nutrition. One such indicator
is “undernourishment”, a measure of ‘chronic food insecurity, in which food intake is
insufficient to meet basic energy requirements on a continuing basis’ (FAO, (SOFI, 1999), p.
11). Information about chronic food insecurity is needed for an assessment of level, geographical
OCR for page 89
ANNEX A 89
distribution and trends of hunger and/or for the design and implementation of anti-hunger
policies, strategies and investment that seek to reverse undesirable trends. This paper focuses on
indicators of chronic food insecurity.
To be useful for a comprehensive assessment, indicators of food insecurity should
provide answers to at least three questions, namely: Who are the food-insecure? How many are
they? And where do they live? If the purpose of the measurement goes beyond assessment and
includes the design of policy responses, the indicators should also help answering the more
ambitious question: Why are people food insecure, what are the underlying causes and hence,
what should be done?
Numerous methods are in use to measure certain aspects of food insecurity. They can be
summarized as follows:
1. Indicators derived from food balance sheets
2. indicators based on household consumption surveys
3. indicators derived from anthropometric measurements
4. indicators derived from medical assessments
5. Composite indicators.
Methods (1) to (3) currently represent the principal tool kit. Both the first and the second
compare levels of nutrient consumption with levels of nutrient requirements. While both use
science-based nutritional norms as requirement standards, they differ in the source of
information about people’s food consumption. The first, used by FAO, calculates food available
for human consumption from national food balance sheets (FBS) and uses different information
sources for a statistical measure of dispersion to approximate the distribution of food
consumption levels within countries. The second derives the estimates of mean as well as
dispersion of food consumption from household surveys, asking respondents to recall food
consumption during a reference period. The third method relies on physical measurements of
people (principally weight and height measurement, often concentrating on children) as
indications of their nutritional status.
The fourth method provides additional data from medical analysis. This can include
clinical assessments, such as the observation of physical signs on the body that are symptomatic
of nutritional disorders (e.g., loss of skin pigment, edema) or biochemical assessment through the
examination of blood or urine. At the population level, health indicators such as child mortality
or low birth weight are also sometimes used as proxies for nutritional status. Finally, a number of
efforts have been undertaken recently to combine specific indicators into composite indicators
seeking to capture several critical dimensions of food insecurity and malnutrition at the same
time. The Global Hunger Index published jointly by IFPRI and the German Welthungerhilfe is
such an example designed for cross-country comparison. While they do not as such generate
additional measurements, composite indicators aim to facilitate communication of the
comprehensive nature of food insecurity and malnutrition.
To date, a consensus among experts on the reasons for discrepancies between the results
obtained from different methods is still elusive. The mentioned case of contradicting estimates of
undernourishment in countries of Sub-Sahara-Africa is one such example. The apparent
contradictions between only moderate estimates for the prevalence of undernourishment in the
overall population of India versus the much higher rates of undernutrition among India’s children
that result from anthropometric measurements are another example (see below). It is hoped that
OCR for page 90
90 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
the following discussion, which particularly examines the first three of the mentioned methods in
detail, will contribute to some progress towards a consensus regarding the most realistic
measurements.
We critically review the three principal methods with regard to their measurement
approach, the accuracy of the underlying data, and their usefulness for policy decisions,
including projections and the simulation of nutritional impacts of shocks. The intention is not to
describe the real food security situation but rather to describe and compare methods and
indicators and make some suggestions for improvement and future research.
As part of a focus on sustainable global food security, the National Research Council’s
Roundtable on Science and Technology for Sustainability is planning a workshop to examine
these indicators, reviewing the approaches used in developing the indicators and assessing their
strengths and weaknesses. This paper is a background paper for this workshop.
2. THE FAO INDICATOR OF UNDERNOURISHMENT
Definitions, Assumptions and Main Sources of Empirical Evidence
FAO estimates the prevalence of undernourishment, expressed as the share of people in a
national population not meeting their minimum food energy requirements. It is assumed that
food energy deficiency is the most critical indicator of hunger. The method is based on three key
parameters: the mean quantity of calories available for human consumption per person, the rate
of inequality in access to those calories within the population and the minimum amount of
kilocalories required by that population on average, based on the gender and age structure.
The graph in Figure I A-1 illustrates the methodological procedure for estimating the
proportion of the population whose food energy availability is below requirement, i.e., who are
undernourished. The function f(x) depicts the proportion of the population corresponding to
different dietary energy consumption levels (x), µx the mean dietary energy intake per person and
rl the minimum acceptable dietary energy requirement (MDER). The area under the curve left of
rl represents the proportion of the population not reaching the minimum level of dietary energy
requirement, i.e. the prevalence of undernourishment, pU. Multiplied with the size of the
population for the respective period it gives the number of undernourished.
OCR for page 91
ANNEX A 91
pU
µx
500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000
rL
Dietary energy consumption (Kcak/person/day)
FIGURE I A-1: The FAO method of estimating undernourishment
NOTE: Distribution of dietary energy consumption
The amount of food available for human consumption is calculated from national FBS,
compiled as the balancing item after considering production, trade, stock changes, non-food uses
and extra-household waste. The per capita Dietary Energy Supply (DES) is obtained by
aggregating all food items, converting the quantities into energy values and dividing the
aggregate volume by the total population. It is noted that the indicator measures food available
for human consumption at the household level, and not actual food intake. However, FAO
considers it to be a close enough approximation of actual dietary energy consumption.
The inequality of access to food is estimated assuming a log-normal distribution function
(Figure I A-1). This function with its short lower and longer upper tail was chosen because it
reflects ‘the fact that wastages, food fed to pets etc. are likely to be confined to the upper tail
representing the richer and more affluent households’ (FAO, 2003, p.12). The function is defined
by the mean level of dietary energy consumption per person and the Coefficient of Variation
(CV). The mean is assumed to equal the DES from the FBS. The CV is derived from the sample
distribution of kilocalorie consumption per person as measured from available household
surveys. Where food consumption information is not directly available from household surveys,
survey data on food expenditure or income are used to derive estimates of dispersion. Where no
survey data are available at all, data from comparable neighboring countries are used.
The third principal parameter needed for the FAO method is the aggregated Minimum
Dietary Energy Requirement per person (MDER). This is the amount of food energy needed to
balance energy expenditure in order to maintain an acceptable minimum body-weight, body
composition and a level of minimum (‘sedentary’) physical activity, consistent with long-term
good health. This includes the energy needs for optimal development of children, deposition of
tissues during pregnancy and secretion of milk during lactation consistent with the good health of
OCR for page 92
92 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
mother and child. The recommended level of dietary energy consumption for the average of a
population is the mean energy requirement of the healthy, well-nourished individuals who
constitute that population.2
Who Uses the FAO Indicator and for What?
FAO has been publishing estimates of undernourishment in irregular intervals in its
World Food Surveys since the 1960s. While scope and contents have been kept broadly similar,
the country coverage as well as details of the methodology have been gradually adjusted over the
years. Since 1999, the indicator is being published annually as a three-year average in the
flagship publication ‘The State of Food Insecurity in the World’ (SOFI). Beginning with the
2009 edition, SOFI is published jointly by FAO and the World Food Program (WFP).
According to the latest edition (SOFI, 2010), covering the period 2005-2007, the total
number of undernourished worldwide was estimated at 847 million people, of whom 835 million
were living in developing countries. The number has hardly changed since 1990-1992 (the base
year for the WFS goal and of the hunger target of the first Millennium Development Goal
(MDG-1) aimed at halving, respectively, the number and the percentage of undernourished by
2015. On the other hand, the prevalence of undernourishment declined from 16 to 13 percent
worldwide and from 20 to 16 percent in the developing countries. Since 2008, FAO has also
published a preliminary estimate of undernourishment for the respective current year, using a
simplified ex-post projection (see below). According to this method, the number of
undernourishment was estimated at 925 million in 2010, down from 1,02 billion that had been
estimated for 2009 using the same ex-post projection method (SOFI, 2009).
The main purpose of publishing the indicator regularly for a very large number of
countries is to inform the global community about levels and trends of undernourishment
(chronic hunger) in the world and thus facilitate global and regional governance of food security,
while also advocating for stepped up efforts in hunger reduction. The indicator measures chronic
food insecurity at national levels. It does not inform about the actual distribution of the number
of hungry within countries nor is it the intention to provide actionable information for policy
responses at sub-national levels. The estimates are therefore primarily of interest for international
comparisons and for assessments of changes over time.
The publication is receiving wide attention by the media and the wider public and clearly
fulfills its purpose to advocate action against hunger. The indicator is also used by food security
analysts. FAO and its governing bodies, in particular the Committee on World Food Security
(CFS) as well as many other international and national organizations concerned with
development cooperation, refer to the estimates regularly. Presumably, various donors use the
indicator also as one key information source for the ranking of priorities for aid allocations. The
FAO Undernourishment estimates also serve as one of the two official indicators of progress
towards target 2 of Millennium Development Goal One (“Halve, between 1990 and 2015, the
proportion of people who suffer from hunger”). The other indicator is ‘Prevalence of
underweight children under five years of age’ using anthropometric assessments (see below).
2
The norms have been defined by the FAO/WHO/UNU Expert Consultation on Human Energy Requirements in
2001, which established energy standards, published in 2004, for different sex and age groups performing sedentary
physical activity and with a minimum acceptable body-weight for attained heights.
OCR for page 93
ANNEX A 93
Since a number of years, IFPRI has used the FAO estimates of the prevalence of
undernourishment as one of three equally weighted indicators to construct its Global Hunger
Index (GHI), with the other two being the prevalence of underweight in children under the age of
five and the mortality rate of children under five years of age.
Governments of developing countries do take note of FAO’s undernourishment estimates
as an indicator of the extent of hunger and of progress or retreat over time. Severe levels of
undernourishment in any one country provide justification for appropriate policy measures to be
put in place to remedy the situation. However, the indicator does not, and is not meant to,
provide directly actionable information for policy design at sub-national level,
Sources of Funding – Past and Future
As informing the world about the scale, geographical distribution and implications of
food insecurity belongs to the core functions of FAO, work in basic statistics as well as the
preparation of the undernourishment indicator and its publication are normally funded from
FAO’s core budget. However, funding of FAO’s statistics program has been rather tight for a
number of years. The problem was recognized and a reform project launched, but its
implementation is still ongoing. This may have critical implications for the quality of the data
base and for the expert capacity in FAO to conduct the compilation of the undernourishment
indicator.
Strengths and Weaknesses of the FAO Indicator
Undoubtedly, the main strength of the FAO method is its world-wide coverage with
estimates for more than 100 countries, which enables the monitoring of national trends and
tracking of progress and setbacks using the same methodology and criteria for all. The main
weakness is the fact that it relies on national statistics compiled in FBS for the estimation of the
dietary energy supply, so that the accuracy of the method depends critically on the quality of the
statistical data obtained from member states and stored in FAOSTAT following a quality check.
One can therefore not rule out that both levels of undernourishment between countries as well as
changes in the indicator from one year to another within a country are determined by erroneous
data rather than a real change in the number of undernourished. The short (yearly) intervals
between publications of the indicator make such ‘over-interpretation’ more likely. Various
authors have also criticized methodological issues, including the focus on food energy, the
compilation of the dispersion of the intra-national distribution of food consumption and the
standards used for calculation of minimum dietary requirements. In the following, some of these
points will be discussed in more detail.
Mean Dietary Energy Supply per Person (DES)
FAO compiles the DES from FBS and uses it as an indicator of food energy
consumption. The quantities of food commodities available for human consumption are
calculated after deducting the net exports, stock increase, non-food use and extra-household
waste from domestic production. This raises several questions. The first is whether food energy
OCR for page 94
94 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
deficiency is an adequate indicator of food insecurity; the second is whether dietary energy
supply is a good approximation of food energy intake; the third concerns the accuracy of the
FAOSTAT data base.
(1) Is average food energy deficiency an adequate indicator of food insecurity?
Obviously, an adequate, healthy diet must satisfy human needs for energy and all essential
nutrients. In fact, according to the Report of a Joint FAO/WHO/UNU Expert Consultation on
Human energy requirements (FAO, 2001) “dietary energy needs and recommendations cannot be
considered in isolation of other nutrients in the diet, as the lack of one will influence the others.”
Adequate intake of food energy is essential for the metabolic and physiological functions of
humans, and in this sense FAO focuses on the key indicator. However, as very often other
nutrients are lacking in the diet, in particular micronutrients such as iron, vitamin A and zinc,
comprehensive assessments of people’s nutritional status should ideally not be limited to the
food energy deficiency indicator. In principle, the food balance sheet data can also be used to
assess the level of micronutrient consumption and adequacy (Wuehler et al., 2005), but the level
of commodity group aggregation is relatively high, which is a drawback that weighs more
heavily for micronutrients than for calories. The use of three year averages generates other
uncertainties. In a country where the fluctuations within the three years are very large, food
insecurity is arguably a much more serious problem than in a country where the three year
average is the same but caloric availability is much more stable.
(2)Is dietary energy supply a good approximation of dietary energy intake? As the
dietary energy supply includes foods, which are subsequently lost or wasted at the retail and
household levels, the method by definition overestimates the actual food energy intake.
(3) How accurate are the FAO food balance sheet data? FAO has been criticized by
various authors for a lack of accuracy of the data inputs used to calculate the mean per caput
DES. Svedberg suggests that “food availability is underestimated (by FAO) in most parts of the
developing world, although less so elsewhere than in Africa” (Svedberg, 2002). He suggests that
often the substantial share of food produced for subsistence tends to be underestimated in official
statistics, leading to an overestimation of undernourishment. Deviations from FAO’s estimates
have also been found in the IFPRI study of 12 African countries mentioned in the Introduction,
although in this case, the critique is that FAO’s measure underestimates hunger.
Testing the validity of these claims is not easy. Conceptually, the FAO method does
capture all components of supply and utilization, including subsistence production; however the
estimates are obviously subject to possible errors. In particular, assumptions regarding post-
harvest losses are often not transparent and there is very little hard data available on its level (let
alone its country-by-country distribution and trends over time). Moreover, it must be noted that,
although the FAO Statistics Division has to fill in missing data, in particular for stock changes,
non-food use and wastage, or use data from other sources, e.g., on trade, a major part of the data
input originates directly from countries. The case of India is worth mentioning here, as FAO’s
estimates of rising numbers of undernourished in spite of the country’s strong economic growth
are rather surprising. Whereas one would assume that economic growth in recent years should
have increased per capita food consumption significantly, the statistics used to estimate
undernourishment do not confirm this. FBS data show India’s per caput consumption stagnating
around 2300 kcal/person/day between 1999/01 and 2005/07. India’s own surveys even show a
OCR for page 95
ANNEX A 95
steady decline of per caput consumption during this period (Chattapadhyay and Chowdhury,
2010). A thorough analysis to explain this paradox is urgent, also because the development in
India is of great significance for global trends (see also Deaton and Drèze, 2008; SOFI, 2008;
and discussion below).
Inequality of Food Energy Consumption at National Level
FAO’s approach to compile the coefficient of variation (CV) of the intra-national
distribution of dietary energy supply has been subject to intensive debate among experts. The
debate has been centering around two questions, one regarding the realism of the CV estimate,
the other regarding the assumptions for changes of the distribution over time.
Is the CV parameter realistic? The critical arguments raised by experts are not all
consistent and partly contradictory. Svedberg suggests, for example, that “FAO must have
overestimated the variance in the calorie-availability distribution across households, because the
ensuing habitual intakes in the lower tail are impossibly low for living households” (Svedberg,
2003, p. 25). The mentioned IFPRI study of 12 African countries comes to the opposite
conclusion. Based on household expenditure surveys, Smith et al. (2006) estimate an average CV
of energy availability of 0.62 for the 12 African countries, whereas the FAO estimate for the
same countries is 0.3, hence much lower. Other household surveys result in similar high
dispersion parameters for food intake (Ecker et al., 2010).
FAO itself recognizes that the coefficient cannot be completely specified even without
considering problems associated with survey practices, measurement errors and sample design
(FAO, 2003, p. 23) The reason given relates to the (realistic) hypothesis that people’s food
consumption is not only influenced by income, but also by their age/sex specific energy
requirement. The following formula is used to calculate the CV:
where CV(x) is the total CV of the household per capita dietary energy consumption,
CV(x |v) is the component due to household per capita income (v), and CV(x |r) is the component
due to other sources of variation, in particular energy requirement (r). CV(x |r) is considered to
be a fixed component and is estimated to correspond to about 0.20. CV(x |v) is, however,
estimated on the basis of household survey data (FAO, 2003, p. 38).
According to FAO, the CV resulting from the analysis of survey data using the formula
above is occasionally further corrected to remove components of variation that are considered
not plausible. Moreover, as the log-normal distribution would not exclude energy intake levels
below the absolute minimum for survival or above possible maximum food intake levels, lower
OCR for page 96
96 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
and upper bounds for the range of the CV have been set at 0.2 and 0.35.3 The IFPRI study
suggests that due to these adjustments the FAO CVs may be biased downward (Smith et al.
2006, p. 50). However the empirical evidence for such a conclusion is limited.
Is it realistic to assume no change of the CV over time? FAO has so far kept most CVs
constant over time. Adjustments of the CVs have been limited to a few cases. FAO justifies this
by a lack of available survey data, but suggests also that “there has been little, if any, change in
the inequality of income/expenditure in most countries” (FAO, 2003, p. 16). The implications of
this procedure for the estimates of undernourishment could indeed be significant. Firstly,
empirical evidence suggests that, especially since the 1990s when structural adjustment programs
began to take effect in more developing countries, income distributions do change as economies
grow. In fact, there is evidence that income and expenditure inequality in a majority of
developing countries increased (at least slightly) between the early 1980s and the mid 1990s;
since then trends are more heterogeneous.4 Secondly, even if the relative income distribution
remains unchanged while average incomes grow, the food demand will grow faster in the lower
income brackets due to their higher demand elasticity. This alone would make it likely that the
CVs of food consumption would decline as average incomes and food consumption grow;
similarly, one would presume that drastic rise in global food prices, as witnessed in 2007/08 and
again 2010 would have a differential impact on food consumption patterns of different income
groups, thereby affecting the CV.
More generally, fixing the CV also means that changes in measured hunger across the
world will be driven by changes in the DES. This gives the erroneous impression that changes in
hunger over time are largely a problem of ‘food availability’, rather than changes in entitlements
(Sen, 1984) of different groups in the population to access to food. Situations where hunger in a
population goes up despite stable or rising DES (e.g., due to a regional national catastrophe,
rising food prices, conflicts, etc.) are ruled out by definition this way; any change in entitlements
across population groups would immediately imply a change the CV. Thus to study hunger, one
needs to examine entitlements of groups which can be affected as much by food prices,
employment, and wages as by food availability in the country; such assessments would lead to a
changing CV.
Lastly, a more technical issue is whether the CV is actually the best measure of
dispersion to estimate and apply in this case. As is well-known, the CV is particularly sensitive
to the distribution of calories in the upper parts of the caloric distribution. The use of the CV is
consequently problematic for two reasons. First, it is not ideal to use a distributional indicator
that will be heavily influenced by the distribution of calories among the ‘non-hungry’. Second, as
a result of the sensitivity to high levels of caloric consumption, any measurement error among
that group of ‘non-hungry’ will have an important impact on the resulting CV and the hungry.
There are ready alternatives to the CV, including the Atkinson inequality measure (see Atkinson,
1970) or the Theil family of inequality measures which are both sensitive to the distribution of
calories at the bottom end of the distribution, which is of interest here.
3
This ‘plausible range’ is based on the analysis of realistic variances of food intake levels within hypothetical
populations with highest and lowest food energy supplies per person.
4
See, for example, Gruen and Klasen (2003) for an analysis of these trends.
OCR for page 97
ANNEX A 97
Minimum Dietary Energy Requirement (MDER)
The third key parameter of the FAO method, the minimum requirement of dietary energy,
is defined as the consumption level that will balance energy expenditure. Questions raised
regarding the approach concern the assumptions determining the dietary energy requirements of
different age-sex groups and the rationale for a singular country-specific cutoff point.
(1) Are the assumptions determining the dietary energy requirements of different age-sex
groups correct? Components of energy expenditure comprise the basal metabolic rate (BMR),
i.e. the energy expended for the functioning of an individual in a state of complete rest; the
energy needed for digesting food, metabolizing food and storing an increased food intake; and
the energy required for performing light physical activities, both work and non-work. The BMR
ranges between 1300 and 1700 kcal/day for adults, depending on age, sex, height and body
weight, to which 55 or 56 percent are added for light activity of male and female adults,
respectively. For children, the energy required for growth is taken into account. An allowance is
also for children below age two from developing countries for the energy needed to recover from
frequent infections. For women during pregnancy and lactation, the energy required for the
deposition of tissue and secretion of milk is considered. As FAO specifies these dietary energy
requirements in accordance with the recommendations by the Joint FAO/WHO/UNU Expert
Consultation on Human energy (FAO, 2001), it is assumed that the assumptions are realistic.
However, further research is needed to examine the realism of the assumptions in the light of
various critical reviews.5
(2) Is a singular cutoff point a good approximation of a population’s minimum dietary
energy needs? The minimum per capita dietary energy requirement is derived by aggregating the
estimated sex-age-specific minimum dietary energy requirements, using the relative proportion
of the population in the corresponding sex-age groups as weights. As the sex-age distribution of
the population changes over time, this so-called cutoff point is being regularly adjusted to
demographic change (FAO, 2003). Svedberg suggested that this method has a “built-in flaw that
leads to biased estimates” (Svedberg, 2002, p. 6) because it fails to consider that even after
taking into account the effects of age, sex, activity and body weight, individuals differ in their
energy requirements due to differences in the efficiency of energy use. He suggests therefore
replacing the singular cutoff point by a bivariate distribution according to which the probability
of an individual not meeting the food energy requirements is not only determined by the
distribution of food intake but also by the covariance between food energy intake and
requirements. According to Svedberg, following this approach would as such result in a notably
higher incidence of undernourishment. Responding to this criticism, FAO experts showed that if
5
According to Svedberg, FAO uses a BMR that is unrealistically high for countries in the tropics, thus
overestimating undernourishment (Svedberg 2002). In contrast, Smith et. al. (2006, p. 48) use energy requirements
that are higher than those used by FAO, averaging 2050 kcal per day as compared to 1800 by FAO. This in itself
could explain why Smith et. al arrive at higher estimates of undernourishment than FAO. While both approaches
assume the same light activity level, they make different assumptions regarding the level of the requirements for
given age-sex groups. FAO classifies a person as undernourished that consumes less than the minimum dietary
energy requirement (MDER) for the respective age-sex group, whereas Smith et. al. classify all people as
undernourished who consume less than the average requirement.
OCR for page 114
114 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
As can be seen, none of the three approaches has a clear advantage in meeting all criteria.
In particular, there is no conclusive evidence regarding the accuracy of the three approaches in
assessing the prevalence of undernourishment or undernutrition. The consumption survey and
anthropometric approaches have clear advantages over the FAO method in several criteria,
especially in terms of measuring diversity and heterogeneity within countries. Although currently
hardly done, they could potentially serve to generate even more information that is important for
a complete assessment. For example, household consumption surveys are potentially very useful
to assess dietary diversity and micronutrient status.
While the analysis so far leads to the conclusion that there are several reasons to suspect
that the FAO method generates biased estimates, the direction of the likely biases cannot
currently be conclusively stated. On the other hand, the approach has the advantage of drawing a
regular picture, consistent with national aggregate food production and trade statistics, of
undernourishment for the population as a whole at national levels and above, which may be
useful for global governance and discussions on hunger and ways to combat it.
6. OPTIONS FOR IMPROVING FOOD SECURITY AND NUTRITION INDICATORS
In 2002, FAO organized an International Scientific Symposium on Measurement and
Assessment of Food Deprivation and Malnutrition, which considered all major methods
including qualitative methods for measuring people’s perception of hunger. There was wide
consensus that “no single method can capture all aspects of hunger while at the same time
providing policy-makers with relevant and timely information in a cost-effective manner” (FAO,
2003, p. XV). Accordingly, the Symposium concluded that a ‘suite’ of indicators was needed to
cover the different dimensions of food security. We believe that this conclusion is still valid.
One of our central suggestions for improving and broadening the empirical data base is
running more surveys. In the longer term, a regular availability of results from representative
surveys could greatly enhance the worldwide information on food insecurity and malnutrition
and reduce the need for additional indicators derived from macro food balance sheets. However,
even if this suggestion is followed, data availability would only improve gradually. Moreover,
the possibility of inconsistent and non-representative household data could still not be ruled out.
Therefore, our recommendations are twofold. First, the FAO method should be improved
through better data and more transparent science-based assumptions. Available data from
household consumption surveys can be used to improve the FAO parameter assumptions. In this
connection, lessons about how to deal with missing data through interpolation can be learned
from the World Bank, which uses living standard surveys to compile global poverty statistics and
updates in certain intervals. Second, household consumption survey and anthropometric
approaches should be further improved, both conceptually and through improved data bases.
These recommendations are further elaborated in the following.
OCR for page 115
ANNEX A 115
Improving the FAO Indicator
At the outset it should be noted that the experts in the FAO Statistics and Agricultural
Development Economics Divisions are currently fully aware of the weaknesses in the
methodology for estimating undernourishment. Work is underway to improve both the data base
and part of the methodology. Many critical points need to be addressed of which several are
currently under consideration in FAO. Of importance are in particular the following:
Food consumption: Continuing the overall reform of FAOSTAT with the view to
improve the quality and consistency of all data inputs for the FBS. The aim should be to improve
the estimates of daily dietary food energy supplies. Whereas consumption has so far mostly been
compiled as the balance of supply-utilization accounts (SUA) for the various commodities and
foods, efforts should be promoted, working with national authorities, to reconcile the estimates
with survey data. Improved estimates will also be needed for other significant components of the
FBS, namely food waste and losses inside households and commercial kitchens. Given the
overwhelming importance of the DES estimate for the FAO hunger measure (and many other
important uses of these data), improvements in the accuracy of these estimates is a top priority.
Intra-national inequality of food consumption: Examining the validity and updating of
the coefficients of variation (CV) of food consumption. Efforts currently underway in FAO to
improve the measures of inequality using household surveys are to be supported. Close
cooperation with national and international organizations conducting such surveys is
recommended so as to ensure consistency of the foods included and possibility of regular updates
to reflect change of CVs over time.
Minimum dietary energy requirements. Continuing inter-agency cooperation to determine
best science based estimates of MDERs. Regular updates are needed to reflect the effects of
changing age structures (in particular rising shares of adults in populations) and changing heights
and weight standards.
Moving Beyond the FAO Indicator
Even with the suggested improvements, the FAO method would not satisfy all
information needs with regard to food insecurity, nor would it suffice to provide policy makers
with actionable information needed to address the main obstacles to overcoming hunger and
malnutrition through effective food security strategies at country level. For these purposes,
consumption surveys and anthropometric measurements are much more useful, but there is also
substantial scope for improvement of these approaches. In our view, the most important ones are
the following:
Data availability: Improvements are required, especially with respect to nationally
representative household consumption surveys, which are usually integrated in more
comprehensive living standard surveys. Such living standard surveys with sufficiently
disaggregated food consumption modules should be carried out more frequently, and in a larger
number of countries, to improve the micro level information base. This will require additional
OCR for page 116
116 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
resources, but such data can be used for a variety of purposes. Beyond the food security context,
nationally representative household surveys can be used for tracking all sorts of developments at
the micro level and for planning and evaluating policy interventions. To the extent possible, the
survey formats should be standardized internationally.
Anthropometric measurements: Here the data base is much better, but anthropometric
surveys contain little other socioeconomic variables and no information on food consumption.
As nutrition, health, consumption, and income are so closely related, we propose linking
anthropometric surveys with household living standard surveys. This will not only help improve
the understanding of food security issues, but will also constitute a precious resource for broader
micro level research related to food, nutrition, health, demography and overall welfare.
Research: There are a variety of conceptual issues related to appropriate assumption for
minimum energy requirements, anthropometric standards etc., which need further research.
Integrated research that compares food intake and nutritional outcome indicators, controlling for
other health-related aspects, would be particularly useful to better understand the existing
contradictions and complementarities and improve the methodologies. This requires the
proposed link between (or integration of) anthropometric and household living standard surveys.
Dietary diversity: More research is also required beyond a calorie focus, to better
understand the role, determinants, seasonality and appropriate measurement of dietary quality
and diversity. A simple count of different food groups consumed by households (food variety
score) has been proposed as a good indicator of nutritional status and even of food security more
generally, but questions remain on advantages, drawbacks and limitations of such dietary
diversity measures in particular situations (Ruel, 2003). Such research would also benefit a lot
from a wider availability of nationally representative linked anthropometric and food
consumption data.
Depth of hunger: In SOFI (2000) FAO had published estimates of ‘depth of hunger’,
defined as the extent to which consumption levels of the undernourished fall below requirements.
For example, it was shown that, in 1996-1998, 46 percent of countries in Sub-Sahara Africa had
an average depth of more than 300 kcal per person per day, whereas this depth was only found in
16 percent of countries of Asia. Publication of this measure was not continued in later issues of
SOFI. Indeed, while information on the depth of hunger is of great interest for comprehensive
assessments of the state of food insecurity, compiling such an indicator can only add value if the
estimates are derived from empirical data with regard to people’s real consumption and not from
assumptions about the intra-national inequality of food consumption. More research is needed in
this field and a resumption of depth of hunger compilations could be considered once the
compilation and regular updating of empirically reliable CVs of food consumption has been
completed.
Policy impact simulations: Concerning the simulation of nutritional impacts of policies
and shocks at country level, household food consumption data currently seem to constitute the
best starting point. Since these surveys also contain information on food prices and household
incomes or total expenditures, calorie price and income elasticities can be estimated for the
population as a whole as well as for population subgroups. These elasticities, together with the
results on household food security, can then be used to predict changes in the prevalence of
undernourishment due to price and income changes. Ecker and Qaim (2010) have recently
developed such an approach, which beyond calories also captures micronutrient deficiencies and
OCR for page 117
ANNEX A 117
related price and income elasticities. Anriquez et al (2010a and b) have also used household
survey data to assess the possible effects of staple food price increases on households’ food
consumption and undernourishment. Of course, there is scope for improvement, but in general
the approach seems useful to simulate micro level nutrition effects of food price spikes or
economic crises to identify short-term problems that cannot wait for the assessment based on a
new household survey. If living standard surveys were linked with anthropometric surveys, as
proposed, then such analyses could be extended to also simulate impacts of policies and shocks
on the prevalence of child underweight, wasting, and stunting.
7. CONCLUSIONS
Improvements in the metrics of food insecurity and malnutrition are not only urgently
needed, but also possible. This assessment of available methods has shown various entry points
for improvement. A consistent and fully transparent process is recommended, comprising
additional research and stepwise updating of the various indicators. The aim would be to develop
an “Integrated Suite of Indicators,” in line with the recommendations already made by the
International Scientific Symposium on Measurement in 2002. Such a suite would eventually use
an improved data base to calculate a reformed FAO hunger indicator, combined with survey and
simulation-based estimates of hunger using household expenditure surveys, linked with
anthropometric surveys. The proposed process could involve the following three steps:
The agencies involved in the collection of relevant data and compilation of relevant
indicators should collaborate through active networking. An example for an appropriate
institutional framework for such networking could be the emerging global Food Security
Information Network (FSIN), which is currently being developed by FAO, WFP and IFPRI. It
links the past FAO FIVIMS work with the new FAO corporate strategy on Information Systems
for Food and Nutrition Security (ISFNS). This is a promising initiative which could eventually
be expanded to include agencies like WHO, UNICEF, World Bank and the EU Commission.
A complete inventory of estimates of relevant national indicators of food insecurity and
malnutrition should be established, published, and regularly updated. This could not only include
the main indicators, the food balance sheet approach of FAO, household surveys and
anthropometry, but also, as available, other useful indicators such as micronutrient deficiency
and dietary diversity. The inventory would contribute to greater transparency and enable more
comprehensive comparative assessments, identification of complementarities and contradictions
between indicators and areas requiring further research and data improvement. An appropriate
online portal would have to be found. One possible location of such a Portal for consideration by
the relevant agencies could be FSIN mentioned above.
A major and coordinated effort is also recommended towards enhancement of the
empirical data base, comprising all data relevant for measuring food insecurity and malnutrition.
More specifically, the following steps towards improving the quality and the accessibility of the
database are recommended:
• Continue the current review of the food balance sheet data base in FAO. This will play a
major role in efforts to improve the accuracy and reliability of aggregate food consumption
data. The FBS approach has the advantage of generating consumption data consistent with
OCR for page 118
118 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
national agricultural statistics. However, currently this important variable is normally not
derived from direct estimates, but calculated as a residual after accounting for all other
components of the equation. As several of these, for example stock change and post-harvest
losses, are subject to uncertain assumptions, the residual may actually accumulate a number of
statistical errors. FAO’s Statistics Division is aware of the underlying problems and seeking
to improve the approach.
• Reconcile data on food consumption derived from FBS and from household surveys. In
principle, mean levels of consumption per person derived, respectively, from national FBS on
and from representative household surveys should be similar. It is hoped that where both
types of estimates are available for the same countries, the proposed inventory of data may
serve to identify major reasons for differences and entry points for resolution of those
differences. Eventually, improved and representative household survey based estimates of
food consumption could substitute for the use of FBS in more and more countries.
• Enlarge the country coverage and frequency of household living standard and anthropometric
surveys so as to broaden the empirical base, including assessments of changes over time and
comparisons between countries.
• Harmonize the formats used for household living standard and anthropometric surveys so as
to facilitate comparisons over time and space.
• Coordinate the sampling frames of household living standard and anthropometric survey, or
even merge both survey types, so as to facilitate comparison of different food insecurity and
malnutrition indicators and enable important research on economic-nutrition-health linkages.
• Parallel to establishing an inventory and improving the data bases, a systematic improvement
of methods and assumptions is recommended for all three principal indicator methods, making
use of the latest findings in nutritional science and following up, inter alia, on the various
suggestions made in this paper. Such an assessment should also include a careful assessment
of the reliability and comparability of cut-offs currently used to determine inadequate access
to calories, nutrients, as well as to determine nutritional status. Ongoing improvement efforts
should be enhanced.
• More efforts should be made to use existing intergovernmental platforms for advocacy and
support of work with indicators of food insecurity and malnutrition. The most prominent
examples are the Committee on World Food Security (CFS) and the United Nations System
Standing Committee on Nutrition (SCN).
Effective action towards such improvements requires political will of governments and
governing bodies at regional and global levels, including the Committee on Word Food Security
(CFS), which has just been reformed “with the aim to become the central United Nations
political platform dealing with food security and nutrition”. Accurate knowledge about who is
food insecure, where they live and why they are food insecure and malnourished is a central
precondition for effective action at all levels.
Improving metrics of food insecurity is also a contribution to the monitoring of the
realization of human rights. As stated in the background document for the World Summit on
Food Security in 2009, “where the existence of hunger and malnutrition results from negligence
or ignorance by responsible policy-makers, it is also a violation of people’s basic human right to
adequate food and to a life in good health and dignity” (FAO, 2009).
OCR for page 119
ANNEX A 119
It is encouraging to note that the weaknesses of the current metrics of food insecurity and
malnutrition have been recognized by the agencies concerned, in particular the FAO, which is
currently undertaking a fundamental review of the data and methods with the aim to improve
them. We recommend that FAO interact closely with other relevant agencies, in particular WHO,
UNICEF and World Bank, as well as with national governments to ensure broad support,
consistency and mutual complementarities of the improvements.
With the current state of evidence it is safe to conclude that the available estimates of
chronic hunger are inaccurate, but it is not possible to conclude whether the real number of
undernourished is above or below the available estimates. It seems not even certain whether the
direction of change has been correctly assessed for the different countries. Even with revised
methods and more accurate data, estimates of food insecurity and malnutrition are bound to be
subject to measurement errors and projections will remain uncertain. Thus the conclusion by the
participants at the 2002 Symposium on Measurement and Assessment of Food Deprivation and
Undernutrition remains valid that if the magnitudes of food insecurity are uncertain, at least the
trends should be correctly captured. We believe that this is possible through improved data and
methodological approaches. While in the short run, an improved FAO method may be used, we
argue that, in the longer run, global measures of food insecurity and malnutrition should
increasingly be based on household surveys that combine food consumption and anthropometric
measurements.
REFERENCES
Adams, R. 2004. Economic growth, inequality and poverty: estimating the growth elasticity of poverty.
World Development 32(12):1989-2014.
Alexandratos, N. 2009. World Food and Aagriculture to 2030/50: Highlights and Views from Mid-2009.
Expert Meeting on How to Feed the World in 2050. Rome, Italy: FAO.
Alexandratos, N, ed. 1995. World Agriculture towards 2010. Chichester, UK: John Wiley and Sons, and
Rome, Italy: FAO.
Anand, S., and M. Ravallion. 1993. Human development in poor countries: on the role of private incomes
and public services. Journal of Economic Perspectives 7:133-150.
Anand, S., and Bärninghausen. 2004. Human resource and health outcomes: cross-country econometric
study. The Lancet 364:1603-1609.
Anriquez, G., S. Daidone, and E. Mane. 2010a. Rising food prices and undernourishment—a cross-
country inquiry. FAO/ESA (Agricultural Development Economics Division) Working Paper No. 10-
01. Rome, Italy: FAO.
Anriques, G., P. Karfakis, S. Daidone, and E. Mane. 2010b. Measuring dietary energy deficiency and the
impact of food price variations at the household level. FAO/ESA Working Paper No. 10-02. Rome:
FAO
Behrman, J. R., and A. B. Deolalikar. 1987. Will developing country nutrition improve with income? A
case study for rural South India. Journal of Political Economy 95:492-507.
Bouis, H., L. Haddad, and E. Kennedy. 1992. Does it matter how we survey demand for food? Evidence
from Kenya and the Philippines. Food Policy 17:349-360.
OCR for page 120
120 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
Bouis, H. E. 1994. The effect of income on demand for food in poor countries: Are our food consumption
databases giving us reliable estimates? Journal of Development Economics 44:199-226.
Bhandari, N., R. Bahl, S. Taneja, M. de Onis, M. K. Bhan. 2002. Growth Performance of affluent Indian
children is similar to that in developed countries. Bulletin of the World Health Organization
80(3):189-195.
Bogin, B. 1988. Patterns of Human Growth. Cambridge, UK: Cambridge University Press.
Bourguignon, F. 2003. The growth elasticity of poverty reduction: explaining heterogeneity across
countries and time periods. Growth and Inequality, T. Eichler and S. Turnovsky, eds. Cambridge,
UK: MIT Press.
Brinkmann, H., J. Drukker, and B. Slot. 1997. GDP per capita and the biological standard of living in
contemporary developing countries. The Biological Standard of Living, J. Baten and J. Komlos, eds.
Berlin, Germany: Springer.
Bruisma, J., ed. 2003. World Agriculture towards 2015/30. London: Earthscan.
Butte, N., C. Garza, and M. de Onis. 2007. Evaluation of the feasibility of international growth standards
for school-aged children and adolescents. Journal of Nutrition 137:153-157.
Chattapadhyay, G., and A. K. Chowdhury. 2010. Review of food insecurity assessment and trends in
India and the provinces. Analysis of India Consumer Expenditure Surveys 1987-2005 Draft Report.
FAO/ESS
Datt, G., and M. Ravallion. 1992. Growth and redistribution components of changes in poverty measures:
a decomposition with application to Brazil and India in the 1980s. Journal of Development
Economics 38(2):275-295.
Davies, D. P. 1988. The importance of genetic influences on growth in early childhood with particular
reference to children of Asiatic origin. Linear Growth Retardation in Developing Countries, J.
Waterlow (Hrsg). New York: Raven Press.
Deaton, A. 1997. The Analysis of Household Ssurveys. Baltimore, MD: The Johns Hopkins University
Press.
Deaton, A. and J. Dreze. 2008. Nutrition in India: Facts and Interpretations. Social Science Research
Network Working Paper.
de Onis, M. and J. P. Habicht. 1996. Anthropometric reference data for international use:
recommendations from a World Health Organization expert committee. American Journal of Clinical
Nutrition 64:650-658.
de Onis, M. and S. Garza. 2006. Foreword. Acta Pediatrica Supplement 450:5-6.
Devarajan, S. 2002. Goals for Development, Policy Research Working Paper No. 2819. Washington, DC:
The World Bank.
Drèze, J., and A. Sen. 1989. Hunger and Public Action. New York: Oxford University Press.
Ecker, O., and M. Qaim. 2010. Analyzing nutritional impacts of policies: an empirical study for Malawi.
World Development 39(3):412-428.
Ecker, O., K. Weinberger, and M. Qaim. 2010. Patterns and determinants of dietary micronutrient
deficiencies in rural areas of East Africa. African Journal of Agricultural and Resource Economics
4(2):175-194.
Eveleth, P., and J. Tanner. 1990. Worldwide Variation in Human Growth. Cambridge: Cambridge
University Press.
FAO (Food and Agriculture Organization of the United Nations). 1996. Sixth World Food Survey. Rome,
Italy: FAO.
OCR for page 121
ANNEX A 121
FAO, WHO (World Health Organization) and UNU (United Nations University). 1985. Energy and
protein requirements, report of a joint FAO/WHO/UNU expert consultation. WHO Technical Report
724. Geneva, Switzerland: WHO.
FAO, WHO, and UNU. 2001. Human energy requirement, report of a joint FAO/WHO/UNU expert
consultation. FAO Food and Nutrition Technical Report 1. Rome, Italy: FAO.
FAO. 2002. Summary of Proceeding: Measurement and Assessment of Food Deprivation and
Undernutrition. Internatinal Scientific Symposium. Rome, Italy: FAO.
FAO. 2003. Measurement and Assessment pf Food Deprivation and Undernutrition. Proceedings of
International Symposium. Rome, Italy: FAO.
FAO. 2006. World Agriculture towards 2030/2050–Interim Report. Rome, Italy: FAO.
FAO. 2009. Feeding the World, Eradicating Hunger. Background Document of World Summit on Food
Security. Rome, Italy: FAO.
FAO. 2010. The State of Food Insecurity in the World. Rome, Italy: FAO.
Foraita, R., S. Klasen and I. Pigeot. 2008. Using graphical chain models to analyze differences in
structural correlates of undernturtion in Benin and Bangladesh. Economics and Human Biology
6:398-419.
Gallup, J., and J. Sachs. 2001. The economic burden of malaria. American Journal of Tropiocalö
Medicine and Hygiene 64(1,2)S:85-96.
Gibson, R. S. 2005. Principles of Nutritional Assessment, 2nd ed. New York: Oxford University Press.
Gorstein, J., K. Sullivan, R. Yip, M. de Onis, F. Trowbridge, and P. Fajans. 1994. Issues in the
assessment of nutritional status using anthropometry. Bulletin of the World Health Organization
72:273-283.
Graitcer, P., and E. Gentry. 1981. Measuring children: one reference for All. The Lancet:297-299.
Guilkey, D., and R. Riphahn. 1998. The determinants of child mortality in the Philippines: Estimation of
a Structural Model, Journal of Development Economics 56:281-305.
Habicht, J. P., C. Yarbrough, R. Martorell, R. Malina, and R. Klein. 1974. Height and weight in pre-
school children: how relevant are ethnic differences in growth potential? The Lancet 303(7858):611-
615.
Haddad, L., and R. Kanbur. 1990. How serious is the neglect of intra-household inequality? Economic
Journal 100:866-881.
Harttgen, K., and M. Misselhorn. 2006. A Mulitlevel Approach to Explain Child Mortality and
Undernutrition in South Asia and Sub-Saharan Africa. Mimeographed. Göttingen, Germany:
University of Göttingen.
Kandala, N. B., L. Fahrmeir, S. Klasen, and J. Priebe. 2009. Geo-additive models of childhood
undernutrition in three Sub-Saharan African countries. Population, Space, and Place 15(5):461-473.
Kidd, K. K., J. A. L. Armour, T. Anttinen, C. A. May, E. E. Vega1, A. Sajantila, J. R. Kidd, J.
Bertranpetit, S. Pääbo, and A. J. Jeffreys. 1996. Minisatellite diversity supports a recent African
origin for modern humans. Natural Genetics 13:154-160.
Klasen, S., and M. Misselhorn. 2006. Determinants of the Growth Semielasticity of Poverty Reduction.
Mimeographed. Göttingen, Germany: University of Göttingen.
Klasen, S. 2000a. Measuring poverty and deprivation in South Africa. Review of Income and Wealth
46:33-58
Klasen, S. 2000b. Malnourished and surviving in South Asia, better nourished and dying young in Africa:
what can explain this puzzle? SFB Working Paper 214. Munich, Germany: Department of Statistics.
OCR for page 122
122 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
Klasen, S. 2003. Malnourished and surviving in South Asia, better nourished and dying young in Africa:
what can explain this puzzle? Pp. 283-287 in Measurement and Assessment of Food Deprivation and
Undernutrition. Rome, Italy: FAO.
Klasen, S. 2008. Poverty, undernutrition, and child mortality: Some inter-regional puzzles and their
implications for research and policy. Journal of Economic Inequality 6(1):89-115.
Klasen, S., and A. Moradi. 2000. The Nutritional Status of Elites in India, Kenya, and Zambia: An
appropriate guide for developing reference standards for undernutrition? SFB Discussion Paper No.
217. Munich, Germany: Department of Statistics.
Lancet. 1984. A measure of agreement on growth standards. The Lancet. January 21, 1984: Editorial.
Lence, S. 2003. Household expenditure survey methods. Pp. 91-96 in Measurement and Assessment of
Food Deprivation and Undernutrition. Rome, Italy: FAO.
Lipton, M., and M. Ravallion. 1995. Poverty and Policy. In Handbook of Development Economics 3b. J.
Behrman and T. N. Srinivasan, eds. Amsterdam: North Holland.
Maxwell, S., and T. R. Frankenberger. 1995. Household Food Security, Concepts, Indicators
Measurements: a Technical Review. New York: United Nations Children’s Fund and Rome, Italy:
International Fund for Agricultural Development.
Martorell, R., F. Mendoza, and R. Castillo. 1988. Poverty and Stature in Children. In Linear Growth
Retardation in Developing Countries. J. C. Waterlow (Hrsg.). New York: Raven Press.
McGillivray, M. 2005. Measuring non-economic well-being achievement. Review of Income and Wealth
51(2):337-365.
MacFarlane, S. B. 1995. A Universal Growth Reference of Fool’s Gold? European Journal of Clinical
Nutrition 49:745-753.
Misselhorn, M. 2010. Undernutrition and the Nutrition Transition. Courant Research Center Working
Paper No. 35. Göttingen, Germany: University of Göttingen.
Mohamed, A. J., A. W. Onyango, M. de Onis, N. Prakash, R. M. Mabry, and D. H. Alasfoor. 2004.
Socioeconomic predictors of unconstrained child growth in Muscat, Oman. Eastern Mediterranean
Health Journal 10:295-302.
Mosk, C. 1996. Making Health Work: Human Growth in Modern Japan. Berkeley, CA: University of
California Press.
Mosley, W., and L. Chen. 1984. An Analytical Framework for the Study of Child Survival in Developing
Countries. In Mosley, W. ed. Child Survival: Strategies for Research. New York: Population Council.
Naiken, L. 2003. FAO methodology for estimating the prevalence of food deprivation and undernutrition.
Pp. 7-42 in FAO. ed. Measurement and Assessment of Food Deprivation and Undernutrition. Rome,
Italy: FAO.
Nussbaum, M., and A. Sen. 1993. The Quality of Life. New York: Oxford University Press.
Osmani, S. R. 1990. Nutrition and the Economics of Food: Implications of Some Recent Controversies.
The Political Economy of Hunger 1, J. Dreze and A. Sen, eds. New York: Oxford University Press.
Osmani, S. R. 1997. Poverty and Nutrition in South Asia, in UN Sub-Committee on Nutrition eds.
Nutrition and Poverty. Geneva, Switzerland: WHO.
Osmani, S., and A. Sen. 2003. The hidden penalties of gender inequality: fetal origins of ill-health
Economics and Human Biology 1(1):105-121
Pelletier, D. 1994. The relationship between child anthropometry and mortality in developing countries.
Journal of Nutrition Supplement 124:2047S-2081S.
Pelletier, D. 1998. Malnutrition, morbidity, and child mortality in developing countries. In Too Young to
Die: Genes or Gender? New York: United Nations.
OCR for page 123
ANNEX A 123
Pelletier, D., E. Frongillo, and J. P. Habicht. 1993. Epidemiological Evidence for a potentiaing effect of
malnutrition on child mortality. American Journal of Public Health 83:1130-1133.
Pinstrup-Andersen, P. 2007. Agricultural research and policy for better health and nutrition in developing
countries: A food systems approach. Agricultural Economics 37(s1):189-198.
Ricardo, S., L. Naiken, and J. Mernies. 2007. The FAO Parametric Versus the IFPRI Non-Parametric
Approach to Estimating the Prevalence of Undernourishment. Statistics Division Working Paper
ESS/ESSG/012e. Rome, Italy: FAO.
Pritchett, L., and L. Summers. 1996. Wealthier is Healthier. Journal of Human Resources 31:841-868.
Ravallion, M. 1994. Poverty Comparisons. Chur, Switzerland: Harwood Academic Publishers.
Ramalingaswami, V., U. Jonsson, and J. Rohde. 1996. The South Asian Enigma. In The Progress of
Nations. New York: UNICEF.
Ruel, M. T. 2003. Operationalizing dietary diversity: a review of measurement issues and research
priorities. Journal of Nutrition 133(11):3911S-3926S.
Seckler, D. 1982. Small but healthy: a basic hypothesis in the theory, measurement and policy of
malnutrition. In Newer Concepts in Nutrition and their Implication for Policy. D. Sukhatme (Hrsg).
Pune, India: Maharashtra Association for the Cultivation of Science.
Sen, A. 1984. Poverty and Famines. Oxford, UK: Oxford University Press.
Sen, A. 1998. Development as Freedom. New York: Knopf.
Smith, L., H. Alderman, and D. Aduayom. 2006. Food Insecurity in Sub-Saharan Africa: New Estimates
from Household Expenditure Surveys, IFPRI (International Food Policy Research Institute) Research
Report 146. Washington, DC: IFPRI.
Smith, L., and L. Haddad. 2002. How potent is economic growth in reducing undernutrition? What are the
pathways of impact? New cross-country evidence. Economic Development and Cultural Change
51:55-76.
Smith. L., and L. Haddad. 1999. Explaining Child Malnutrition in Developing Countries: A Cross-
Country Analysis. IFPRI Discussion Paper No. 60. Washington DC: IFPRI.
Smith, L., U. Ramakrishnan, L. Haddad, R. Martorell, and A. Ndyae. 2003. The Importance of Women’s
Status for Child Nutrition in Developing Countries. IFPRI Research Report No. 131. Washington DC:
IFPRI.
Smith, L. 2003. The use of household expenditure surveys for the assessment of food insecurity. Pp. 57-
88 in Measurement and Assessment of Food Deprivation and Undernutrition. Rome, Italy: FAO.
Stein, A. 2006. Micronutrient Malnutrition and the Impact of Modern Plant Breeding on Public Health in
India: How Cost-Effective is Biofortification? Göttingen, Germany: Cuvillier Verlag.
Svedberg, P. 1991. Poverty and Undernutrition in Sub-Saharan Africa. Stockholm, Sweden: Swedish
Institute for International Economics.
Svedberg, P. 1999. 841 million undernourished? World Development 27:2081-2098.
Svedberg, P. 2000. Poverty and Undernutrition: Theory, Measurement, and Policy. Oxford, UK: Oxford
University Press.
Svedberg, P. 2002. Undernutrition Overestimated. Economic Development and Cultural Change 51(1):5-
36.
Sukhatme, D. 1982. Measurement of Undernutrition. Economic and Political Weekly 17.
The World Bank. 1993. World Development Report 1993: Investing in Health. New York: Oxford
University Press.
The World Bank. 1997. Confronting AIDS. Washington DC: The World Bank.
OCR for page 124
124 A SUSTAINABILITY CHALLENGE: FOOD SECURITY FOR ALL
The World Bank. 2000. World Development Report 2000/2001: Attacking Poverty. New York: Oxford
University Press.
The World Bank. 2005. World Development Indicators. Washington DC: The World Bank.
UNDP (United Nations Development Programme). 1990. Human Development Report. New York:
Oxford University Press.
UNDP. 2000. Human Development Report. New York: Oxford University Press.
UNICEF (United Nations Children’s Fund). 1998. The State of World’s Children: Focus on Nutrition.
New York: UNICEF.
WHO (World Health Organization). 1983. Measuring Change in Nutritional Status. Geneva, Switzerland:
WHO.
WHO. 1995. Physical Status: The Use and Interpretation of Anthropometry. WHO Technical Report
Series No. 854. Geneva, Switzerland: WHO.
WHO. 1999. Infant and young child nutrition: the WHO multicentre growth reference study. WHO
Implementations and Resolutions EB105/inf.doc/I. Geneva, Switzerland: WHO.
WHO Multicentre Growth Reference Study Group. 2006a. WHO child growth standards based on
length/height, weight, and age. Acta Paediatrica Suppl 450:76-85.
WHO Multicentre Growth Reference Study Group. 2006b. Enrolment and baseline characteristics in the
WHO multicentre growth reference study. Acta Paediatrica Suppl 450:5-15.
WHO Multicentre Growth Reference Study Group. 2006c. Assessment of differences in linear growth
among populations in the WHO Multicentre Growth Reference Study. Acta Paediatrica Suppl
450:56-65.
Wuehler, S. E., J. M. Peerson, and K. H. Brown. 2005. Use of national food balance data to estimate the
adequacy of zinc in national food supplies: Methodology and regional estimates. Public Health
Nutrition 8:812-819.
