Practice Questions: Health Issues in Developing Countries
1. Present bias is the tendency to favor immediate rewards at the expense of our long-term
goals. In which way present bias might explain the pattern of high expenditure in curative
health care and low adoption of preventive health care technologies in developing
countries?
2. Disease burden in developing countries has two main features: i) affects people at much
younger ages; ii) main channels of morbidity and mortality are infectious and parasitic
diseases– generate important public health externalities. What does these imply for the
child and adult health outcomes (mortality and life expectancy) in the developing
countries?
3. Standard public finance analysis implies that health goods generating positive
externalities should be publicly funded, or even subsidized at more than 100% if the
private non-monetary costs (such as side effects) are high. Why do you think that
consumption of health goods that generate positive externalities will not be optimal at
market prices?
4. As per Standard public finance analysis,
a) What type of health good (with positive externalities) not necessarily be provided
completely free (100% subsidized) for improving the efficacy of public subsidies?
b) How does cost-sharing (rather than free distribution) may lead to higher usage intensity
of those health products? Or alternatively, why is argued that charging a positive price for
a commodity is necessary to ensure that it is effectively used?
5. Which of the following can be possible outcome(s) of a partial subsidy in the context of a
poor developing economy?
a) While cost-sharing may lead to higher usage intensity than free distribution, it may also
reduce program coverage by dampening demand.
b) If people who cannot afford to pay a positive price are more likely to be sick and need the
good, then charging a positive price would screen out the neediest and could significantly
reduce the health benefits.
c) Selection effects are not straightforward in the context of credit and cash constraints.
d) Both a and b
e) a, b and c
6. What was randomized in the paper “Free Distribution or Cost-Sharing? Evidence from a
Malaria Prevention Experiment” by Jessica Cohen and Pascaline Dupas?
7. Why do the authors of “Free Distribution or Cost-Sharing? Evidence from a Malaria
Prevention Experiment” argue that they do not find any evidence that cost-sharing
reduces wastage by sifting out those who would not use the ITN (Insecticide-treated bed
nets)?
8. Which of the following is FALSE as per the findings of the paper “Free Distribution or
Cost-Sharing? Evidence from a Malaria Prevention Experiment” by Jessica Cohen and
Pascaline Dupas?
a) As per the finding of the experiment, while it doesn’t increase usage intensity, costsharing does considerably dampen demand
b) The authors find that the usage of ITNs is insensitive to the price paid to acquire them
c) The authors find evidence that cost-sharing induces selection of women who need the net
more: those who pay higher prices appear sicker than the average prenatal client in the
area in terms of measured anemia (an important indicator of malaria).
d) Given the large positive externality associated with widespread usage of ITNs Free
distribution of ITNs could save many more lives than cost-sharing programs have
achieved so far.
9. As per the paper “Worms: Identifying Impacts on Education and Health in the Presence
of Treatment Externalities”, why randomized provision of deworming treatment within
schools to treatment and placebo groups of students, and then examining the impact of
deworming on cognitive outcomes would understate the actual impact of deworming on
the treatment group?
10. Can you show the chains through which a school wide deworming treatment would lead
to better educational outcomes?
11. Why the authors of “Worms: Identifying Impacts on Education and Health in the
Presence of Treatment Externalities” argue that their study improves upon previous
researches in terms of measuring the externality effects of deworming?
12. What intervention was given to the treatment units in the paper “Worms: Identifying
Impacts on Education and Health in the Presence of Treatment Externalities”?
13. The paper “Worms: Identifying Impacts on Education and Health in the Presence of
Treatment Externalities” uses
a) instrumental variable method to measure the within-school externalities effect of the
deworming intervention.
b) experimental method to measure the within-school externalities effect of the
deworming intervention.
c) non-experimental approach to measure the within-school externalities effect of the
deworming intervention.
d) Both a and b
14. The following table is from the paper “Worms: Identifying Impacts on Education and
Health in the Presence of Treatment Externalities”.
What do the coefficients -0.25 and -0.26 in column (1) mean? Explain the coefficients.
15. Why sometimes it is argued that in the context of developing countries (where resources
are very limited) reducing the burden of sickness and death associated with HIV/Aids
through prevention would be more cost effective that treatment?
16. In the context of HIV and risky sexual behaviors preventive measures, what do risk
avoidance and risk reduction behaviors mean? Give an example of each type.
17. What might be the problem of using only Self-Reported Sexual Behavior Survey to
examine the impact of an HIV education campaign? How does the paper “Do Teenagers
Respond to HIV Risk Information? Evidence from a Field Experiment in Kenya” attempt
to solve this problem (do they collect any additional information to get better idea of the
overall picture)?
18. As per the findings of the paper “Do Teenagers Respond to HIV Risk Information?
Evidence from a Field Experiment in Kenya”, what is the best way to widen HIV
education campaign impacts among group of population at risk – emphasizing on risk
avoidance information or risk reduction information?
Chapter 8
Poverty and Undernutrition
8.1. Introduction
There is no more visible characteristic of economic underdevelopment than poverty. It is also the most
shocking characteristic—the outgrowth of layer upon layer of inequality. There is, first, the inequality of world
income distribution. As if this were not enough, there is the inequality of income distribution within a country.
The outcome, for many millions of people, is destitution, squalor, and lack of hope.
It is all too easy to provide “illustrative” examples of the development process: there are many in this book
and in every textbook on economic development, but it is not easy to describe, head on, the horrors of
poverty and its attendant correlates: illiteracy, undernutrition, ill health, and the utter bleakness of the future.
Poverty strikes not only at the core of ongoing existence. By effectively taking away the rights of a human
being to live in good health, to obtain an education, and to enjoy adequate nutrition, poverty destroys the
aspirations, hopes, and enjoyment of the future as well. Poverty was a medieval scourge for a good reason:
the world was generally poor then. There is little excuse for living with poverty today.
Considering that the world has generated significant growth in per capita income, its track record on
poverty is pretty dismal. Over the period 1965–75, consumption per capita in developing countries grew by
32%, and then by another 26% over the period 1975–85.1 However, by fairly conservative estimates that we
will discuss subsequently, the number of poor people in the world in 1990 was over a billion (in a total of well
under six billion). The figure alone is staggering.
Just as in the case of inequality, poverty is both of intrinsic and functional significance. Most people would
say that the removal of poverty is a fundamental goal of economic development. Hence, the characteristics of
the poor and the appropriate measure of poverty are important considerations in policies that must be sharply
targeted toward the poor. However, poverty is not only of intrinsic interest: it has enormous implications for
the way in which entire economies function. Some of these functional implications were tied up in our
discussion of inequality, but there are others that are specific to poverty itself.
This chapter is divided into four parts. First, we discuss the concept of poverty, and—something that’s
obviously related—how to go about measuring it. Next, we apply some of these measures to obtain a sense
of the extent of poverty in the world today. In addition to these quantitative estimates, we also describe the
correlates of poverty: characteristics that are widely shared by poor individuals. Not only does an
understanding of these characteristics help to identify the poor, but it may also serve as a focal point for
policies that are geared toward ending poverty. Third, we analyze the functional impact of poverty. At many
stages this issue links up with material in other chapters of this text, and we will point at this material to avoid
repetition. Finally, we discuss policies for poverty alleviation.
8.2. Poverty: First principles
8.2.1. Conceptual issues
At the heart of all discourses on poverty is the notion of a poverty line: a critical threshold of income,
consumption, or, more generally, access to goods and services below which individuals are declared to be
poor. The poverty line, then, represents a minimum level of “acceptable” economic participation in a given
society at a given point in time. For instance, we could collect data on minimum nutrient levels that make up an
adequate diet, on the prices of foodstuffs that contain such nutrients, and on the costs of shelter and clothing,
and then add up the consumption expenditures needed to obtain these basic requirements to obtain an
estimate of the poverty line for a particular society. We could use the prevailing legally decreed minimum wage
in a country as an estimate for the poverty line of that country. Alternatively, we could fix some other norm,
183
say, 60% of the mean income of a country, to arrive at an estimate of its poverty line.
Nutrition-based poverty lines are not uncommon. The poverty line used in the United States is based on
Orshansky’s [1963, 1965] estimates, which scale by three a minimum-budget estimate for food requirements
(the scaling proxies for other requirements such as rent and clothing). Indian poverty lines have traditionally
been drawn by using estimates of expenditure necessary to guarantee a minimum consumption of calories. Of
course, such poverty lines (and probably all poverty lines) should be approached with some caution and
scepticism: the poorer the country, the better the nutrition- based approximation. Issues of scaling become
more problematic as the average standard of living rises.
The following subsections explain some of the fundamental concerns that surround poverty measurement.
Overall expenditure or item-by-item consumption?
Should we declare a person to be poor when her actual, observed consumption basket falls below certain
prespecified thresholds or when her expenditure (or overall income) falls below the minimum required to
obtain these consumption standards? Certainly, we could conjure up examples where the two approaches
yield different results; for instance, what are we to make of the wealthy ascetic who starves himself on an
ongoing basis? At a more serious level, nutrition levels may not unambiguously rise with income.2 For instance,
canned foods may become quite popular at certain levels of income, even though their nutritive value is
questionable. Thus, even through elasticities may be high with respect to changes in income, nutrient elasticities
may not be correspondingly high. Income represents the capacity to consume, not consumption itself.
Nevertheless, income- or (aggregate) expenditure-based poverty lines are far easier to use, given the scarcity
of available data.
Absolute or relative?
Clearly, there is something absolute about the notion of poverty. Regardless of the society we live in, people
need adequate levels of food, clothing, and shelter. Whereas it is certainly the case that there are variations in
what might be considered “adequate” (shelter, in particular, might be subject to varying society-specific
interpretations), nobody would deny the biological imperative of nutrition, for instance, or the near-universal
norms of adequate clothing. At the same time, it is unclear that the phrase “acceptable levels of participation in
society” can be given absolute meaning, independent of the contours of the society under consideration. In
some societies, the ownership of a television may be deemed socially necessary for living a “full” life; in others
it is not. Likewise, minimal standards of leisure, access to scientific education, ownership of private means of
transportation, and so on, are all concerns that must be evaluated relative to the prevailing socioeconomic
standards. These considerations quite naturally give rise to the need for poverty lines that share certain
common components, but vary (perhaps widely) from country to country.
Note carefully that although poverty lines should (and do) incorporate relative notions of what constitutes
“necessity” or “basic needs,” we must still think of them as fulfilling some absolute notion of the ability to
function in a society. The previous paragraph chooses our examples carefully to make this point. 3 For
instance, it would be foolish to define poverty by, say, the percentage of the population earning less than half
the average income of society. Such a measure confuses poverty with inequality. For instance, the measure
would remain completely unchanged if all incomes were scaled down by the same proportion, plunging half
the population into famine!
Temporary or chronic?
As we will see, people who live in (or close to) a state of poverty, however that state is measured, often
experience significant fluctuations in their income and consumption. This is especially true for the poor or nearpoor in developing countries, where a large fraction of the population may depend on a quirky, weather184
dependent agriculture. Expressed as fractions of their average earned income, these fluctuations are large. As
Morduch [1994] pointed out, notions of “structural” or chronic poverty must therefore be complemented by a
study of “temporary poverty.” The latter occurs when, because of bad economic shocks (such as poor rainfall
or low prices for one’s production), individuals temporarily enter a poverty sample. The distinction is not just
for the sake of a distinction: the policies required to combat temporary as opposed to chronic poverty may be
very different.
The temporary versus chronic distinction is closely related to Friedman’s [1957] famous distinction
between temporary and permanent income. Income in a given year may be far from capturing the smoothed
or “permanent” stream of consumption that an individual or household enjoys over time. For this reason,
household or individual expenditures are often thought of as a more reliable way to assess chronic poverty.
Households or individuals?
Often household-level data on expenditure and income are all that is available. It is tempting, then, to simply
express household consumption as individual averages (so that household size can be accounted for), and then
apply one’s favorite measure of poverty. However, this neglects an exceedingly important issue: that the
allocation of expenditures within the house-hold are often significantly skewed. Among the potential victims
are females and the elderly. There is some evidence that such discrimination grows sharper with the overall
level of destitution of the household. Macroestimates of poverty should therefore be complemented by
“microstudies” that study intrahousehold allocation. We will study some examples in the subsequent text.
Neglecting altogether the problems of distribution, a second set of concerns arises from the fact that larger
households typically have more children. Some correction for the presence of children is desirable, because
they consume somewhat less than adults. The construction of adult equivalence scales—conversion factors
that express the consumption of children as a fraction of a representative adult—would get around this
problem.4
Finally, there are fixed costs in setting up and running a household. Smaller households cannot spread
these fixed costs over several household members. They are therefore at a disadvantage. We return to this
and related issues later.
Why a poverty line, anyway?
It is possible to argue that a fixed notion of the poverty line is untenable. In part this is because of some issues
raised earlier; for example, the relativity of poverty or its fluctuating nature. Even if we stick to chronic,
nutrition-based measures of poverty, we still are unable to find some magic level of nutrition below which
people abruptly go up in little puffs of smoke (in which case there would probably be no poverty to speak of,
anyway). As we shall see later in this chapter, undernutrition is not the same as immediate and obvious
disaster, and therefore it is more insidious. The world can indefinitely carry a stock of undernourished people,
living and breeding under impaired circumstances. Although more will be said presently on such issues, it is
important to realize that poverty lines are always approximations to a threshold that is truly fuzzy, more
because the effects of sustained deprivation are often felt at a later point in time. There is really little to be
done about this criticism except to realize that quantitative estimates of poverty lines are not to be memorized
all the way down to the third decimal place and that they are basically (important) pointers to a deeper and
less quantifiable concept.
8.2.2. Poverty measures
With the preceding qualifications in mind, then, we will consider a poverty line to be an expenditure
threshold that is regarded as minimally necessary for “adequate” participation in economic life. People below
this threshold will be said to be poor.
A little notation will be useful. As in Chapter 6, y denotes income (or expenditure) and subscripts i,
185
j, . . . , refer to individuals. Let’s denote by p the poverty line5 and by m the mean income of the economy.
One natural measure that comes to mind is simply to count the number of people below the poverty line.
We might be interested in the numbers per se or in the relative incidence of the poor. In the latter case,
divide by the total population of the country or region under consideration. The first measure is known as the
head count, and the latter as the head-count ratio, which is just head count as a fraction of population. In
part because they don’t place great strains on available data, these measures are widely used. In our notation,
the head count (HC) is given by the number of individuals i such that yi < p, whereas the head-count ratio
(HCR) is just
where n is the total population.
An obvious problem with the head-count ratio is that it fails to capture the extent to which individual
income (or expenditure) falls below the poverty line. This is related, of course, to observation 5 (Why a
poverty line, anyway?) in the previous section that poverty is not a “zero–one” concept. People further below
the poverty line are “poorer” than people closer to it, and the head count is insensitive to this observation.
However, matters are worse than plain insensitivity: use of the head count can lead to problematic policy
decisions, as the following example suggests.
Example 1: You are a planner in Ping, a poor land, where the poverty line is set at 1000 pah a year. It turns
out that in Ping there are two equal-sized groups below the poverty line. One group consists of 100
individuals: they have equal earnings of 500 pah a year each. The second group also has 100 people: they
earn 900 pah a year each. Of course, there are also people who are above the poverty line. You have been
allocated a budget of 20,000 pah a year. You must allocate this budget among the 200 poor people.
(i) Suppose you were to forget about the poverty line. Who would you give the money to?
(ii) Now suppose that you are firmly told by the President of Ping to use this money to minimize, as far as
possible, the head count. Who would you give the money to?
The point of the example is very simple. The use of the head count as a measure of poverty systematically
biases policy in favor of individuals who are very close to the poverty line. Statistically, these people offer the
biggest bang for the buck, because they are most easily taken above the poverty line. Yet of all the poor, they
are relatively in the least need of help. A benevolent government that is perfectly secure and without fear of
losing the next elections may ignore the problem and act in the best interests of the people, but most
governments, like most people, are more interested in maximizing the observable and seemingly objective
measures of their success.
One way to partially offset this bias, and more fundamentally take account of the extent of poverty, is to
use a measure of the average income shortfall from the poverty line. An example is the poverty gap ratio,
defined as the ratio of the average of income (or extra consumption) needed to get all poor people to the
poverty line, divided by the mean income (or consumption) of the society. The reason for dividing by the
average for society as a whole is that this gives us an idea of how large the gap is relative to resources that
potentially may be used to close the gap. In this sense, the poverty gap ratio is not really a measure of poverty
itself, but a measure of resources required to eradicate it.
In terms of our notation, the poverty gap ratio (PGR) is given by
where m, you will recall, is mean income.
Dividing by average economywide income might give a misleading impression of poverty in highly unequal
186
(but overall wealthy) societies with a large number of poor people. The poverty gap ratio in such societies
may look pretty small, even though the plight of the poor is made no less acute by this maneuver. Therefore, a
close relative of this measure, called the income gap ratio, is often used. This is exactly the same measure of
total shortfall of the poor from the poverty line, except that we divide the shortfall by the total income required
to bring all the poor people to the poverty line. This places a slightly different perspective on things. It captures
more directly the acuteness of poverty, because it measures it relative to the total income needed to make that
poverty go away.6 Thus the income gap ratio (IGR) is given by the formula
where we recall that HC is just the number (head count) of the poor.
The PGR or the IGR is not susceptible to the same kind of policy distortion as the head count, as the
following example shows.
Example 2: Return to the problem of Example 1. Now imagine that you are told to minimize (as far as
possible) the PGR or the IGR. Does the way you now spend your money necessarily contrast with the
intuitive reactions you noted in part (i) of Example 1?
It should be clear from the discussion that the PGR or IGR avoids the “bang for the buck” problem by
deliberately neglecting numbers or fractions of people that are below the poverty line. In a sense, PGR and
IGR only capture the “per capita intensity” of poverty. The head count (or HCR), whatever its other failings,
does not suffer from this problem. For this reason, it is a good idea to use measures of each type jointly,
where possible, to evaluate the extent of poverty.
Finally, we note that both the head count and the poverty gap class of measures share an additional
drawback relating to the fact that both these measures ignore the important issue of relative deprivation
among the poor. 7 Relative deprivation is just another phrase for inequality among the poor. The new phrase
is used to capture the fact that we are concerned only with the inequality among the deprived, or poor. The
main concern is captured by the following example.
Example 3: Return to Example 1, where, as you will recall, there are 200 people below the poverty line; half
of them have an income of 500 pah and the rest have an income of 900 pah.
(i) Suppose that each person who earns 500 pah gave 50 pah to each person who earns 900 pah. The
new income levels are then 450 and 950 pah. What do you think would happen to the intensity of poverty in
this new situation relative to the old? Now compute the HCR and PGR (or IGR) in both situations. Compare
what the measures say with what you feel intuitively.
(ii) To make the point even more starkly, transfer 110 pah each (instead of 50 pah) between the same
groups and redo the exercise.
Even if we were to take the head count and the gap-ratio measures together, there are other aspects of
poverty that may be left out. This observation leads to more sophisticated measures of poverty that have been
proposed by economists such as Sen [1976] and Foster, Greer, and Thorbecke [1984]. With better data,
these more demanding measures can be easily applied. The Appendix to this chapter contains a discussion of
the Foster–Greer–Thorbecke index.
8.3. Poverty: Empirical observations
We now turn to the data to get a sense of the extent of poverty and the characteristics of the poor. We begin
with a universal poverty line to facilitate cross-country comparison. Be aware that this is a tricky business. We
already discussed the fact that poverty has relative as well as absolute components. The choice of some
187
“universal” poverty line creates overly high “real poverty” in some countries and too little poverty in others. To
partly circumvent this problem, the World Development Report (World Bank [1990]), which represents a
landmark study on poverty in developing countries, experimented with a choice of two poverty lines: $275
and $370 per person per year, expressed in 1985 PPP prices. The range was chosen to reflect that fact that
the poverty lines of some of the poorest nations fall between these two limits.8
Table 8.1. Poverty in developing countries, 1985 and 1990, using “universal” poverty lines.
Source: World Development Report (World Bank [1990, 1992]).
Note: Poverty lines are at 1985 PPP prices. The 1992 report updates and changes head-count information for 1985 and provides 1990
data. The PGRs for 1985 are unaltered from the 1990 report.
Table 8.1 puts together poverty data from two World Development Reports. Keeping in mind that these
poverty lines were chosen quite conservatively, the results are staggering, to say the least. In 1990, well over
one billion individuals were estimated to earn less than $370 per year (or $420 per year at 1990 PPP
prices). The time trend does not look very hopeful either. Except for East Asia, which experienced very high
rates of growth, the absolute numbers of the poor rose significantly between 1985 and 1990. The overall
percentage of people in poverty (at the $370 line) was roughly constant over this period at 30% of the
population of all developing countries.
Even if we were to use the extra-conservative poverty line of $275 per year per person, we would see
that in 1985, over 600 million people were poor even by these unexacting standards. The overall figures for
poverty would be significantly higher were we to use country-specific poverty lines.
We now turn to the characteristics of the poor.
8.3.1. Demographic features
It is not surprising that those households whose members fall below the poverty line also tend to be large
relative to the average family. For Brazil, Fishlow [1972] reported that 29% of all families had a size of six or
more individuals, and over half of such families fell below the poverty line. Similarly, for Malaysia, Anand
[1977] noted that the incidence of poverty rises with family size, ranging from 24% in a household of one to
46% in households with ten or more people. The World Development Report (World Bank [1990])
observed that in Pakistan in 1984, the poorest 10% of households had an average of 7.7 members; the
corresponding national average was 6.1.
Not surprisingly, these larger, poor families often have a high ratio of dependent members, often children.
In all the examples cited, the number of children per family was significantly correlated with their poverty. This
is of great concern, because it suggests that the burden of poverty often falls disproportionately on the young.
Given the immensely important role that childhood nutrition and education play, this is a double tragedy that
overall head counts and poverty gap ratios cannot fully capture.
Clearly family size may be both a cause of poverty as well as an effect. Larger families, especially those
188
with larger numbers of children, are likely to have lower per capita income simply because of the higher
dependency ratio. To be sure, some of this dependency is eroded by institutions such as child labor, but
children are not paid much. More significantly, poverty may actually feed on itself by creating the incentive to
have a large number of children. Why this might be the case is a topic for Chapter 9. Suffice it to say that we
speak of a correlation here, but as always, we cannot establish causality without more careful study.
There are two reasons, however, to doubt the high degree of observed correlation between household
size and poverty. First, there is the problem of using per capita expenditure (or income) of the household as
the relevant indicator, as most studies do. Larger households have a greater fraction of children, as we’ve
already noted, and to the extent that children consume less than adults, the use of per capita expenditure
overstates the amount of poverty. Second, some allowance should be made for the fact that larger households
enjoy significant economies of scale. Once again, per capita measures generally overstate the extent of their
poverty.
Correcting for these factors in a way that is conceptually satisfactory is not an easy task, but some
allowance for adult equivalence is better than none. For instance, one could use a weight of 0.5 for children
(although some variation here is also desirable, depending on age and sex). This weighting will certainly lower
the estimates of poverty for large households. Correcting for increasing returns to scale—the fixed costs of
setting up and running a household—has its own share of conceptual problems as well. One way out of this is
to try different parametric values for returns to scale and see if “reasonable” values overturn the observed
correlation between poverty and household size.9
It should also be noted that women are disproportionately represented as heads of poor households. The
Fishlow study on Brazil cited earlier noted that there are twice as many female-headed households among the
poor as among the nonpoor. This trend is widespread, being reflected in Africa, other parts of Latin America,
and South and East Asia.10 The absence of a principal male earner appears to be closely related to poverty.
For more discussion on the connections between gender bias and poverty, see the concluding section of
this chapter.
8.3.2. Rural and urban poverty
Even if we take into account the differences in rural and urban cost of living, poverty in rural areas is
significantly higher. Even countries with substantial advances in creating an equitable agriculture display higher
rural poverty than their national averages. Table 8.2 summarizes rural–urban disparities in poverty, as well as
in two major indicators of well-being, for selected countries.
8.3.3. Assets
A natural characteristic of poverty is that it is correlated with the lack of ownership of productive assets.
As usual, we must be careful not to establish a one-way causal relationship between the lack of ownership of
assets and poverty. Just as the paucity of assets leads to poverty, a condition of poverty leads to the sale of
assets. In a word, the scarcity of assets and poverty must be viewed as closely related phenomena.
Given that poverty is so closely related with location in rural areas, it is not surprising that the bulk of the
poor are found among the landless or near landless. Poverty and small-scale agriculture are especially strongly
correlated in Africa: most of the poor in countries such as Botswana, Ghana, Kenya, and Nigeria are small
farmers or pastoralists (World Development Report, World Bank [1990]). Apart from southern Africa,
where the rural poor hire out their labor, the poor are largely self-employed. In contrast, in South Asia,
landless labor is more widely represented among the poor. India, Pakistan, and Bangladesh all display a mix
of poverty that is borne as much by landless labor as by small holders. Note, however, that after a point, the
distinction between small landowners and landless laborers is blurred or meaningless: we are talking about
pitifully low quantities of land in any case.
Table 8.2. Rural and urban poverty in the 1980s.
189
Source: World Development Report (World Bank [1990]).
Nevertheless, it is true that there is a significant difference in poverty once we move from negligible or
near-negligible holdings of land to more moderate holdings. Table 8.3 illustrates this difference.
Latin America shows the same concentration of poverty among the landless or the near landless. In Costa
Rica, wage labor counts heavily among the poor, whereas Peru’s poor are accounted for by small holders and
herders. The poor also participate in rural nonfarm employment, largely cottage and traditional industries, the
products of which are destined for home consumption or local markets.
Table 8.3. Poverty and landholding in rural Bangladesh, 1978–79.
190
Source: World Development Report (World Bank [1990]).
Urban poverty shows the same mix of self-employment and wage labor. Most of the poor reside in the
“informal sector,” which we will study in more detail in Chapter 10. Self-employment is common: as vendors,
petty traders, tea-stall owners, beggars, shoe-shine boys, garbage sifters, load carriers, rickshaw pullers,
roadside hawkers, and so on. Wage employment is often on a casual basis and not subject to minimum wage
laws. Because of the chronic lack of assets, the vulnerability of the poor, quite apart from the low average
levels of living, can be frightening.
Side by side with the scarcity of physical assets are the low levels of human capital. The most important
determinant of the access to human capital is the ability to temporarily remove oneself from the labor force
and use this period to acquire skills. This removal must be covered financially, through either loans or the
support of close family and relatives. This kind of financial cover is the last thing one can associate with the
poor and, consequently, it is far from surprising that the majority of poor have little or no human capital.
Illiteracy rates are very high indeed, and among those who are not illiterate, there is little evidence of schooling
beyond primary levels.
8.3.4. Nutrition
There is an intimate connection between poverty and undernutrition, especially in low-income countries.
With low income, it is difficult for individuals to acquire adequate levels of food and nutrient consumption for
themselves and their families. “Adequacy,” as we shall see, is a loaded word, because the notion depends
fundamentally on the kinds of activities in which an individual is engaged, as well as the nutritional history of
that person. Nevertheless, it is not difficult to see the effects of undernutrition. In children they are particularly
severe: muscle wastage, stunting, and increased susceptibility to illness and infection. Undernutrition can also
affect cognitive skills. In adults, chronic undernutrition diminishes muscular strength, immunity to disease, and
the capacity to do productive work. In the next section, we will see how low nutrition can feed back on a
person’s capacity to do work, thus perpetuating the state of poverty in which they find themselves.
In many countries, poverty and undernutrition are closely related with each other, because the definition of
the poverty line often relies on the expenditure necessary to obtain a certain minimum food or nutrient basket
(plus some margin for nonfood items). Examples include Malaysia and India. Authors such as Lipton [1983]
have argued that using a calorie-based poverty line, or a food adequacy standard, is an appropriate way to
measure moderate or extreme levels of poverty in developing countries.11 In such examples it is not surprising
that poverty and undernutrition are highly correlated. Countries such as Brazil have used measures that are not
obviously nutrition-based, but nevertheless a correlation persists between the subregions or subpopulations of
these countries that display the greatest degree of poverty and the greatest degree of undernutrition. It must be
mentioned, however, that as average income rises, poverty, as measured by household or per capita
consumption (adjusted for the proportion of children in the household), exhibits less of a correlation with
direct anthropometric measures of undernutrition, such as measures of stunting or abnormally low weight in
children.12
Although the incidence of poverty and the incidence of undernutrition may be ordinally related, in the
sense that a poor person is more likely to be undernourished than her richer counterpart, the relationship
between increases in income (or expenditure) and increases in nutrition may or may not be strong. Imagine
that you draw a variety of different graphs to illustrate hypothetical relationships between income earned and
calories consumed. All of these graphs may be increasing in the sense that greater income translates into more
calories consumed. Thus poorer people are more likely to be undernourished, but the flatter curves in that set
of graphs suggest that increases in income may translate (at least over some range) into a small increase in
calorie consumption, whereas the steeper curves suggest a stronger sensitivity of calorie intake to income.
Thus depending on empirical findings, it is perfectly possible for the poor to be undernourished, while at the
same time direct nutrition supplements may have a far greater impact on undernutrition than an increase in
income.
There are two effects that might bear on this phenomenon, and they run in different directions. First,
individuals attach significance to higher nutrition. A state of good nourishment is itself desirable, because it
191
means greater stamina, physical and mental health, and higher resistance to illness. However, nutrition is also
useful in a functional sense, as we shall soon see: it raises work capacity and, therefore, earnings ability. For
both these reasons, an increase in purchasing power tends to raise nutritional status, especially if nutritional
levels are low to begin with.
The second effect has to do with individual preferences for foods that taste good, or, more insidiously,
foods that are well advertised and well packaged, or even worse, foods that are recognized as indicators of
social and economic attainment.13 It is easy enough in economically developed societies to downplay the
strength of this effect, but in societies where food is of extreme importance in the budget, great value is
assigned to the consumption of different food items, and nutrition may not be at the root of all these decisions.
For example, the consumption of meat, or expensive varieties of rice, or even canned food, may be given far
more social importance (as an indicator of status or wealth) than considerations of pure nutritive value
warrant.14 The desire to increase nutrition and the desire to increase food consumption for culinary pleasure
or to signal social standing generally combine to create an intermediate reaction of nutrition to income.
Evidence on this issue is mixed and varies between strong and weak nutrition responses to budget
changes. Overall, an increase in income has a significant effect on nutrition if nutrition is measured by the
consumption of calories. However, the effect is not as strong as we might expect from a pure nutritional
viewpoint.
What might we expect from such a viewpoint? The answer is best stated in terms of elasticities: what is
the percentage change in the consumption of calories15 when household budgets change by one percentage
point? An answer of 1 means that there is an equivalent percentage change in nutrition when budgets change.
Because there are subsistence minima to nutrition levels below which it is difficult to go, this a priori notion is
possibly too high. In other words, if income falls below a certain minimum, individuals may obtain their
nutrition from other sources (support from relatives, for instance). As income increases, individuals
presumably substitute away from these sources. Thus (and simply as a reasonable guess, no more), elasticities
between 0.6 and 0.8 may be good evidence that individuals strongly adjust nutrition to income.
Is this what we observe? The answer seems to be in the negative. Estimates range between elasticities that
are close to zero and those that are in the region of our a priori expectations.16 Table 8.4 summarizes the
estimates obtained in various studies; calorie elasticities are arranged in increasing order of magnitude. Of
course, the idea isn’t to take an average of all these findings, because the methodology and the data sets differ
widely, but we can get a sense of what kind of numbers are available.
Overall, we do obtain some evidence that pure nutritional concerns do not entirely drive household
decision making. However, these overall findings need to be tempered by two observations. First, there is
some evidence that poorer households indeed react more strongly to changes in their budgets by purchasing
more nutrients. Second, the pooling of data across the peak and lean seasons may confound the elasticity
estimates. Because food supply in the peak or harvest season is more abundant, a change in the budget does
not translate into significantly higher nutrient consumption. On the other hand, if food availability is low, as in
the slack season, and credit markets are imperfect so that consumption cannot be fully smoothed over time
(see Chapter 14), an increase in household income in the slack season is more adequately reflected in the
demand for nutrition. Both these points were made by Behrman, Foster, and Rosenzweig [1994] (and by
other authors as well). Behrman, Foster, and Rosenzweig use a data set from rural Pakistan, and found that a
careful distinction between slack and peak seasons pays dividends. Estimated elasticities are high and
significant in the slack. Moreover, they are especially high for people who are landless or near landless.
Later in this chapter, we turn to a converse relationship. What is the relationship between nutrition and the
ability to generate income, or more broadly, on the ability to perform economically productive work?
Nutrition and Income: A Case Study from South India
How do we go about estimating a relationship between nutrition and income? Consider the demand for a basket of food items
consumed by households. The statistician’s choice of the basket depends on the availability of data. Average estimates of the
nutrient content of each food item (its calorie, protein, calcium, and other contents) are available from nutrition data that record such
192
information. Now suppose that household expenditure rises. Then the demand for each of these food items will change, and we can
measure these changes. If we multiply all these changes by the average nutrient content (say, calories per gram or protein per liter)
for each food item and add up, we obtain a measure of the change in nutrient consumption as expenditure changes.
Table 8.4. Elasticities of calorie demand with respect to household budget, arranged in ascending order.
Source: Behrman, Foster, and Rosenzweig [1994, Table 1].
aCalorie elasticity is estimated at the sample means.
bBudget was measured by household income.
cBudget was measured by household expenditure.
dThe first entry pertains to the lean season, the second to the peak season when food is more abundant.
This method does take into account the change in the composition of the food basket as expenditure rises, so that a shift from
more to less nutritious foods can be captured as we move up the expenditure scale. The problem is that the extent to which such
effects can be captured depends on the richness of the data describing food groupings. Often, this is inadequate. For instance, even if
we had data on “rice” rather than “grain,” there are substitutions between short- and long-grained varieties that cannot be picked up.
With the advent of canned, processed, and packaged foods, the possibilities of substitution are endless. Another way of stating the
point is that we cannot assume that nutrient content stays constant within the food item as we move from lower to higher levels of
expenditure. Typically and unfortunately, the nutrient component seems to fall.
A study by Behrman and Deolalikar [1987] dramatically displays this possibility. They used the foregoing method to study six
villages in two states in the semi-arid region of India, known as the ICRISAT villages. 17 For the years 1976–77 and 1977–78,
special nutrition surveys were carried out and nutrient intakes were recorded for households. The nutrition surveys provided
information on nine nutrients: calories, protein, calcium, iron, carotene, thiamine, riboflavin, niacin, and ascorbic acid. This suggests a
direct approach to the problem: simply relate consumption of these nutrients to the expenditure by household.18 Contrast this with
the food basket approach, which the authors discussed as well: they considered consumption changes in six basic foods: sugar,
193
pulses, vegetables, milk, meat, and grains. Table 8.5 summarizes some of their results. Reported in this table are the elasticities of
expenditure on various items with respect to a change in the household budget. This is done first for the commodity groups and then
on the nutrients.19 Thus an entry of 0.57 for sugar means that if household expenditures were altered by 10%, the expenditure on
sugar would increase by 5.7%. An elasticity of 1 means that expenditure on that item grows at the same rate as total expenditure.
We see from Table 8.5 that elasticities for food items are large and significant (the weighted average over food groups is 1.18),
whereas, apart from carotene, there is no strong nutrient effect to speak of (all the estimated coefficients are insignificant at the 5%
level). This raises a puzzle of some significance: Why don’t poor individuals who are generally below the food adequacy standard
(and the individuals in this sample were below the standard, on average) significantly respond to budget increases by increasing their
nutritional intake?
We have discussed this study in some detail because it presents a counterintuitive position in a provocative way. Do not take
this to mean that all subsequent studies find the same low relationship between income (or expenditure) and nutrient intake. There
are significant variations over countries, as well as over studies done at different points in time on the same country, as the main text
illustrates.
Table 8.5. Elasticities of demand for food and nutrient groups.
Source: Behrman and Deolalikar [1987, Table 2].
Notes: An asterisk denotes that the variable was significantly affected by household expenditure (see Appendix 2 for a discussion of
“significance”). The elasticities were evaluated at the sample means.
8.4. The functional impact of poverty
There is very little one can say to captures adequately the degradation, the indignity, and the dehumanization
of utter economic deprivation, so I will not try. We hear often of the joys of a simple, poor life, unencumbered
by materialist ambitions, rich in many ways. There is little doubt that poverty can bring out the best in human
beings, in an environment where the common sharing of transient gains and losses has such immense value. On
the other hand, that is no excuse for poverty, and people singing the praises of the simple, honest, loyal,
trusting poor are well advised to experience a dose of poverty themselves. Economic poverty is the worst
curse there is.
We move on, therefore, to arguments that link the incidence of poverty to mechanisms that drive its
creation. It is also important to understand the informal mechanisms that spontaneously arise to cope with
poverty. These mechanisms tell us something about what causes poverty, as well as the wider effects that
poverty has on the economic system, and they are fundamental to the creation of appropriate policies.
The fundamental feature of poverty is that it affects the access of the poor to markets, and this change in
access has repercussions for the entire economy. Practically all markets are affected: the ability to obtain
credit, to sell labor, to rent land for cultivation. What we are going to discuss next are some of these effects.
Of course, in a natural way, they all tie in with chapters that are devoted to a study of such markets, such as
Chapters 13 and 14, so we will be brief in these matters and refer you to additional material that can be found
194
elsewhere in this book.
8.4.1. Poverty, credit, and insurance
Credit
The market for credit naturally fails for the poor. The poor are unable to obtain loans that can be used to
better their lives by allowing them to invest in a productive activity. The failure occurs for a variety of reasons.
First, the poor lack collateral that can be put up for loan repayment. Collateral is charged for two
reasons. One is that the project to which the loan is being applied may genuinely be unsuccessful, so that the
borrower is unable to repay the loan. Collateral is insurance against this possibility. However, this is not the
principal reason by far. If projects are, on average, successful, then enterprising lenders realize that there are
gains to be made (in an expected sense) and they fill such gaps with loans. Collateral is, more fundamentally, a
means to of prevent intentional default on the part of the borrower. 20 The possibility of lost collateral
reduces the incentive to walk away without repaying the loan. The poor lack the wherewithal to put up
adequate collateral and therefore are denied loans.21. In Chapter 7, we discussed a model in detail that
incorporates this point.
As we will see in more detail in Chapter 14, the inability of the poor to provide appropriate collateral
effectively shuts them out from the formal credit market. Sometimes informal credit sources can step in to fill
this gap, because they can accept collateral in forms that the formal sector cannot. The most important of
these forms is labor. In increasingly mobile societies, this form of collateral is becoming rarer, because
although labor services serve the first function of collateral (which is to provide a backup to the lender in the
even of involuntary default), they are only of limited use in the prevention of intentional default.
Second, it can be argued that the incentives to repay for the poor are limited, independent of (and in
addition to) their inability to put up collateral. To understand this, it suffices to note that each additional unit of
money in hand means far more to a poor individual than to a rich individual: this is just the familiar principle of
diminishing marginal utility. Thus when the time to repay loans comes around, the calculus of whether or
not to default on the loan is naturally twisted in favor of default. Figure 8.1 shows how this works.
I n Figure 8.1, we look at two incomes, Yp (for poor) and YR (for rich). Compare the two cases in a
situation where the same loan L has to be repaid. Because the utility function exhibits diminishing marginal
utility, it is clear that utility loss to the poor from repayment (given by the segment of length A in the diagram)
exceeds the corresponding utility loss to the rich (given by the segment of length B).
Of course, it can be argued, in response to this observation, that the assumption of similar loan size is not
sensible. Typically, the poor receive smaller loans, which destroys the easy comparability of Figure 8.1. In
addition, it can be argued that we are not taking the costs of default into account (as we explicitly did in
Chapter 5). Perhaps the stakes are higher for the poor: they have more to lose from nonrepayment,
particularly in lack of future access to credit.
You could make both these points, and you would be absolutely correct in making them. The poor do get
smaller loans, on average, and for precisely this reason. It is also possible that the poor have much more to
lose from default, but this only reinforces our argument that initial poverty reduces access to the credit market.
Indeed it is always in the interest of the lender to assure that loans do not permanently change the economic
conditions of his borrower, so that the threat of cutting off future credit always has bite.
195
Figure 8.1. Incentives to repay for the poor and rich: a comparison.
We already saw in Chapter 7 that this lack of access implies a loss in national output, because productive
opportunities are not being utilized by a properly functioning credit market. To the extent that lenders are
unable to capture a share of the returns from these activities (because of the fear of default), they will not lend
to allow individuals to exploit these opportunities.
The lack of access to credit market also affects the access of the poor to land tenancy markets. For more
on this, see Chapter 12.
Insurance
On the other hand, mutual opportunities of insurance among the poor are perhaps easier to exploit. To see
why this informal safety net actually works better under conditions of poverty, it is important to take a quick
look at the factors that limit insurance. We take this up in much more detail in Chapter 15, on insurance.
Briefly, then, why do people insure? The reasons are fairly obvious. The future holds risks that we are
unwilling to take. Our house may burn down, we may fall sick or be disabled, we may be laid off, we may run
over someone in our car, and so on. To insure against such contingencies, we typically pay a sum of money,
say every year, to an insurance company. The insurance company collects this money, and typically plays no
role in your life (apart from trying to persuade you to get other things insured) until one of these bad incidents
occurs, say your house burns down. In that case, the insurance company must pay the amount for which you
insured your house.
Now consider what is needed for successful insurance. The first feature of all insurance is that the incident
against which you are insuring must be verifiable, at least to some extent. You will be unable to buy insurance
against the possibility that you might be in a bad mood tomorrow. It’s not that it’s a weird thing to insure
against—people have been known to insure far stranger things—but the point is that such outcomes are not
verifiable, least of all to the insurance company.
The second feature of successful insurance is that whatever you are insuring against is not subject to moral
hazard. Moral hazard is an important economic concept that will be studied in detail in Chapters 12–14, but it
196
is easy to convey the main idea. Suppose that I own a personal computer and insure it against damage. Now
that I am insured, the degree of care that I take to make sure that I do not spill coffee on the keyboard may
decrease, because the cost of the damage to me has been reduced by insurance. The point is that there are
incidents that you might want to insure against, where the probability of occurrence is influenced by
your actions. This creates a dilemma. Perfect insurance is a good idea in principle, but if it blunts the feeling of
responsibility that people have for their own actions, it might make life very costly for the insurance company,
or at any rate for somebody.22
To avoid moral hazard, then, companies typically retreat from the provision of complete insurance. There
are deductibles if you buy a prescription drug or if you spill coffee on the computer keyboard. You typically
incur some of the costs if your house is burgled, and if you buy life insurance, companies will not pay in the
event of suicide, at least in the first few years of the insurance. The list of restrictions is long and varied.
In developing countries, formal insurance schemes are relatively rare. Indeed, on both the abovementioned counts, there are typically severe problems. With the formal legal system at slow and fairly minimal
levels, and with limited powers of verifiability, it is difficult, if not impossible, to obtain formally verifiable
accounts of incidents, such as the exact degree of crop failure on someone’s land holding. The same lack of
information exacerbates issues of moral hazard: it is true that the crop on your land is determined by the
vagaries of the weather (which is why you want insurance in the first place), but it is also the case that it can be
influenced by how hard you work the land, which is very difficult for an insurance company to control.
Moreover, in many cases, what is needed is nonmonetary methods of insurance. Illness in a family might
necessitate the provision of care by another resident in the same village or it might require extra labor at the
time of harvest. Because of these formidable problems, formal insurance is almost always missing.
We will see in Chapter 14 how these formal schemes are typically replaced by informal schemes at the
level of the village community. Village members have access to far better information, and therefore can selfinsure as a group in a way that no formal company can replicate.
Of course, the issues of moral hazard still remain. Perfect insurance of idiosyncratic movements in crop
output may lead to the underprovision of effort by the family farm. The point is, however, that these moral
hazard problems are likely to be smaller for the poor.
It is easy to see why. Almost by definition, the opportunity cost of labor for the poor is lower than that of
richer people. The poor are more likely to be unemployed or underemployed. Even if this were not true, they
are likely to earn lower wages when they are employed and, in general, the cost of their time is lower. This
feature, in turn, permits them to credibly supply more effort to the task at hand (such as farming) without
necessitating a large cutback (or “deductible”) in the degree of insurance that they receive. This low
opportunity cost of effort is helped along by the fact that their marginal utility of consumption is very high (see
the discussion in the previous subsection). Therefore, even if they are participating in schemes that insure them
to a high degree, they will rarely freeload on such schemes. Therefore, when the people involved are poor
rather than rich, it is far easier to have informal schemes that involve a large amount of shared labor and effort,
as well as transfers of money or grain, to tide over bad times.
We will see more of this kind of analysis—and some caveats as well—in Chapter 15.
8.4.2. Poverty, nutrition, and labor markets
Introduction
We already observed that, even by very conservative estimates, over a billion people worldwide were
classified as poor in 1990. We also observed that a large proportion of these individuals are also significantly
below adequate standards of nutrition.
The effects of undernutrition vary widely. We have already mentioned outcomes such as muscle wastage,
retardation of growth, increased illness, vulnerability to infection, and the diminution of work capacity. In
addition, undernourished persons are easily fatigued and exhibit marked psychological changes, manifested in
mental apathy, depression, introversion, lower intellectual capacity, and lack of motivation. Life expectancy
197
among the undernourished is low, but the undernourished do not die immediately.
In this section, we study the relationship that exists between a person’s nutritional status and his capacity
to do sustained work, and we study in Chapter 13 how this relationship creates a vicious cycle in the labor
market: poverty leading to undernutrition, hence the inability to work, which feeds back on the incidence of
poverty. Thus undernutrition plays a functional role apart from being of intrinsic interest. Because
undernutrition affects the capacity to work, it affects the functioning of labor markets in a central way.
Energy balance
To start thinking seriously about this problem, it is useful to examine the simplest story of energy balance
within the human body.23 It has four main components.
1. Energy input. The periodic consumption of food is the main source of energy input to the human body.
It is also the obvious point where nutrition meets economics. Access to food, in most situations, is the same as
access to income. In the case of the poor, income chiefly represents returns to labor supply and (to a lesser
extent) to nonlabor assets such as small quantities of land.
2 . Resting metabolism. This is a significant proportion of the body’s requirements. It represents the
energy required to maintain body temperature, sustain heart and respiratory action, supply the minimum
energy requirements of resting tissues, and support ionic gradients across cell membranes. For the “reference
man” of the Food and Agricultural Organization (FAO), who is a European male and weighs 65 kg, this figure
is around 1,700 kcal per day. Of course, the exact number varies significantly with the characteristics of the
individual and the ambient environment in which he lives. An important determinant, for instance, is body mass:
a higher body mass raises resting metabolism.
3. Energy required for work . The second significant component is energy required to carry out physical
labor. The FAO’s 1973 estimate, applied to their reference man, prescribed 400 kcal per day for “moderate
activity.” Unfortunately, as Clark and Haswell [1970, p. 11] pointed out, the FAO reference man “appears to
be a European weighing 65 kg, and who spends most of his day in a manner rather ambiguously defined, but
apparently not working very hard.” For the poor in less developed countries, who are subject to hard labor of
the most strenuous kind, this may be a somewhat conservative estimate. Although precise estimates are
impossible without knowing the kind of work the individual has to perform, it is probably safe to say that the
figure is significantly higher than 400 kcal per day.
Clark and Haswell’s interesting book contains information on the energy requirements for various types of
physical activity, culled from the work of different authors. Thus, in studies of West African agriculture,
estimates of calorie consumption vary from 213 kcal per hour for carrying a log of 20 kg, to 274 kcal per
hour for hoeing, to 372 kcal per hour for bush clearing, and up to 502 kcal per hour for tree felling. Of
course, these are activities that are not (and cannot) be performed continuously over large stretches of time,
but the European reference man with his allotment of calories for physical activity might be hard pressed to
carry out any of these at minimal levels. The point, then, is clear enough. The labor of the poor is often
physical labor, and physical labor requires significant amounts of energy.
4. Storage and borrowing. It should be quite obvious by now that, over a period of time at least, we can
expects to see some form of balance between item 1, energy input, and the sum of the components in items 2
and 3. In the short or medium run, however, excesses or deficits can be cushioned (to some extent) by the
human body. An energy deficit is met by running down stores from the body. An energy surplus is partly
dissipated, partly stored. Well-fed people in developed countries worry about the second problem (especially
the possibility that energy surpluses may be stored and not dissipated). For the hundreds of millions of people
that suffer undernutrition, the real problem is the first: coping with the threat of an energy deficit. A sustained
deficit leads to undernutrition, and—ultimately—the breakdown of the body via illness, incapacitating debility,
or death.
The point that we need to be aware of—and it is a point that we shall develop in detail in Chapter 13—is
that not only do labor markets generate income and therefore create the principal potential source of nutrition
198
and good health, but good nutrition in turn affects the capacity of the body to perform tasks that generate
income. There is a cycle here, and this cycle alerts us to the possibility that in developing countries, a
significant fraction of the population may be caught in a poverty trap.
To fix our ideas, ignore for the moment the possibility of borrowing or storage. Figure 8.2 shows the
relationship between nutrition and the capacity to perform productive work, which we refer to as the capacity
curve.
Observe closely the labeling of the axes in Figure 8.2. In particular, the x axis, which really should be
“nutrition,” has been labeled “income.” The implicit assumption here is that all income is spent on nutrition.
Nothing of substance is lost by amending this to a more realistic situation where, say, 70% of income is spent
on nutrition, but as you’ll see, the exposition is just easier this way. The y axis is labeled with the vaguesounding phrase “work capacity.” How can we conceptually think about this? The idea is to think of work
capacity as a measure of the total number of tasks an individual can perform during the period under review,
say, the number of bushels of wheat that he can harvest during a day. The capacity curve is found by linking
different nutrition (or income) points to the corresponding levels of work capacity that are generated by the
individual.
Figure 8.2. The capacity curve.
To understand the shape of the capacity curve, ask yourself what happens as we move from left to right
along the x axis; that is, as we increase the amount of income (nutrition) available to the individual. Initially,
most of this nutrition goes into maintaining resting metabolism, and so sustaining the basic frame of the body.
In this stretch very little extra energy is left over for work (remember again that for the moment, we are ruling
out the depletion of body stores of energy). So work capacity in this region is low (close to zero, if you like)
and does not increase too quickly as nutrition levels change. Once resting metabolism is taken care of,
however, there is a marked increase in work capacity, as the lion’s share of additional energy input can now
be funneled to work. This phase is followed by a phase of diminishing returns, as the natural limits imposed by
the body’s frame restrict the conversion of increasing nutrition into ever-increasing work capacity. (The curve
probably even turns downward after a point, reflecting the usual concerns of the developed world, but we
ignore that here.)
199
Nutrition and work capacity
The whole point of developing the biological relationship between nutrition and work capacity is to alert us to
a line of thought that we will pursue in detail in Chapter 13. Although low incomes create low nutrition, low
nutrition is capable of creating low incomes. This is the functional aspect of undernutrition: apart from
being of social and ethical concern in its own right, it has an impact on the ability to earn. Thus it is not difficult
to imagine a vicious circle of poverty in many low-income countries, in which low incomes are responsible for
undernutrition, which in turn perpetuates those low incomes.
In Chapter 13, we take up this theme in some detail, but at this stage, it is worth thinking about how the
argument may run. Several considerations come to mind.
(1) If a low-income-undernutrition-low-income circle is possible in poor countries, why is it not
possible for some groups of people in rich countries?
This question pushes us to think about whether the vicious circle that we’ve described can exist in
isolation, independently of whether the economy is rich or poor. The answer is that, in general, it cannot, and
the reason has to do with the overall supply of labor.
A labor market is tight if the alternatives to working with any particular employer are relatively plentiful
and attractive. Standard supply and demand theory tells us that for a labor market to be tight, there must be
either a low supply relative to demand in that market itself or attractive opportunities in other labor markets.
Now, if the labor market is tight, in the sense that we have just described, the returns to work are high
even though a person may have low work capacity to start with. The circle cannot be completed. These high
returns, in general, permit the individual to consume adequate nutrition and hence raise his work capacity over
time. The limits to which a worker’s income can be pushed depend not on biological considerations, but on
the opportunities available to that worker elsewhere in the labor market. If these latter considerations are
salient, then a vicious circle theory based on undernutrition ceases to be valid.
The tightness of particular labor markets in particular countries is an issue that can be settled only by
detailed and careful empirical work.24
(2) Can’t people simply borrow their way out of the vicious circle?
This is a subtle issue that we cannot address satisfactorily until we study Chapters 13 and 14, but it is
possible to provide some tentative answers. First, the credit market may simply be closed to poor individuals,
for reasons that we have outlined in the preceding sections. This is especially true of consumption credit.
Moneylenders are often interested in funding tangible production projects or providing working capital for
such projects, and consumption loans are difficult to obtain at reasonable terms.
There is a second, more delicate answer. An economy with undernutrition traps of the kind that we are
envisaging here may well be Pareto optimal! That is, there may be no way (in the short run) to make the
undernourished poor better off without some amount of redistribution from the portion of the population with
greater access to income and assets.25 Recall from your introductory economics analysis what Pareto
optimality means. It means that there is no way to rearrange endowments, production, and consumption so
that all economic agents are simultaneously better off. Pareto optimality sounds very nice, and at some level it
is, but it is perfectly compatible with the idea that some people are getting very few of the economic goodies.
The best way to understand this is to think of dividing a cake between two people. As long as you aren’t
throwing any of the cake away, any division is Pareto optimal, including the one in which one person eats all
of it.
Pareto optimality has its implications. If an economy is functioning so that its allocation of goods and
services is Pareto optimal, then introducing a credit market in which people can borrow to stock up on work
capacity cannot have any effect at all! The reason is that for people to lend to such a market, they must
register a gain. The people who borrow presumably gain as well. People who do not participate are
unaffected.26 Then the new allocation achieved by the credit market must make some people better off and
nobody worse off. This contradicts the postulate that the earlier allocation was Pareto optimal.
200
This argument relies on the presumption that the initial outcome is Pareto optimal. We will see more of this
model in Chapter 13.
(3) If work capacity affects future work output, won’t employers wish to offer long-run contracts
that take advantage of this?
It is unclear that such contracts can be enforced unless there is some separate reason why workers want
to stay in such contracts (there may well be, as we will see later in the book). It is unlikely that an employer
will make a long-run contract with his employee just to extract future gains from enhanced work capacity.
There is no guarantee that the employee will be around tomorrow: he may work for a different employer,
perhaps in a different village; he might migrate. Under these circumstances, the employer might be extremely
reluctant to engage in a nutrition-enhancing investment. Second, if a person in good health can be identified by
other employers, the market will bid up the wage rate for such an employee. This means essentially that the
employee will reap the entire benefit of the employer-financed investment in the form of a higher wage. If this
is the case, then why undertake the investment in the first place?
The problem can be overcome if the employee binds himself to a contract that forbids him from working
elsewhere in the future even though the terms elsewhere are better, but this has ethical connotations that make
it unenforceable by a court of law, and rightly so, from a moral point of view.
(4) By the way, if such long-run relationships were somehow in place for other reasons, would this
have an effect on nutritional status?
It might, but in a relationship where nutrition is used positively by the employer to build up work capacity
on the part of her employee, there must be a separate factor, or set of factors, that makes the
relationship inflexible in the sense that the employee is costly to replace. Consider three quick examples.
The slave economy: Slavery is perhaps the most appropriate example. Slaves were bought, and therefore
each act of replacement brought with it a large outlay, apart from the daily costs involved in keeping slaves.
Indeed, in the American South, slave prices rose steeply in the decades before the Civil War (Fogel and
Engerman [1974, pp. 94–102]). Thus an existing slave had great value to the employer/owner. It turns out
that slave diets were plentiful and varied.27 The diet actually exceeded U.S. 1964 levels of recommended
daily allowances for all the chief nutrients. Perhaps more to the point, the calorific value of the average slave
diet exceeded that of all “free men” in 1879 by more than 10% (Fogel and Engerman [1974, p. 113]). In
addition, the maintenance of the health of slaves was repeatedly emphasized in overseer manuals as a central
objective (Fogel and Engerman [1974, p. 117]).
Industry: The effect of adequate nutrition on the productivity of workers has been emphasized repeatedly in
manuals. The monograph by Keyter [1962] on South Africa, for example, contains many such references and
a closing section with fifty-four recipes. This book focuses on industrial feeding, and in so doing squarely
addresses the obvious reasons for feeding in the workplace: by changing the composition of wages in this
manner, it forces the worker to consume a greater proportion of his wage as food.28
Domestic servants: This is another good example of a labor market that is likely to be inflexible. Servants are
associated with characteristics acquired on the job that make them hard to replace. Not only is the loss of a
servant important, but the acquisition of a new servant with minimally acceptable characteristics often results in
an arduous training process. We are interested in seeing studies of this market in the Indian context; casual
empiricism tells me that such studies would prove quite supportive to our thesis.29 We refer the reader,
instead, to an excellent monograph on the subject by McBride [1976], which cites various housekeeping
manuals written for English and French housewives in the nineteenth century. Although McBride found the diet
of servants to be generally parsimonious (relative to that of master and mistress), more than one manual
explicitly suggests means to assure servants a high level of energy. For instance, a popular French manual of
the early nineteenth century recommended that servants be made to abandon the traditional Parisian practice
of café au lait in the morning and substitute a breakfast of soup made from the meat left over from the previous
night, so that the servant would have enough energy to work until 5 p.m. without stopping. Booth’s study of
201
life among London laborers concluded that “the quality of food given to domestic servants . . . is usually very
good, and in all but very rare cases greatly superior to that obtainable by members of the working-class
families from which servants are drawn” (Booth [1903, Vol. 8]).
8.4.3. Poverty and the household
The unequal sharing of poverty
One of the great tragedies of poverty is that the poor may not afford to share their poverty equally. Unequal
sharing arises fundamentally from the fact that certain minimum amounts of nutrition, care, and economic
resources have to be devoted to each person (including each child) in order for that person’s life to be
productive and healthy. In situations of extreme poverty, equal division of household resources might help no
one, because the average amounts are far too small. The potential merit of unequal division is that it helps
some individuals in the household to be minimally productive under extreme circumstances. This takes us right
into the well-known problems of the “lifeboat ethic”: a lifeboat can hold only two people and there are three
individuals to save. One person must die.
The capacity curve gives us a clear idea of how the nutritional problem serves to promote unequal
allocations. Figure 8.3 displays the capacity curve, marked OAEB. The straight line OAB is drawn from the
origin so that the line segment OA equals the line segment AB. The income level corresponding to capacity B is
denoted by Y*. By construction, the income level corresponding to capacity A must be Y*/2.
Now consider a household of just two persons, and suppose that their capacity curves are identical and
given by the curve in Figure 8.3. Suppose that total household income happens to be given by Y*. Think of
two options: the household shares this income equally or one person consumes the entire income.30 Notice
that by the construction of Y*, these two options yield exactly the same total work capacity for the household:
by similar triangles, the height of B must be exactly twice that of A.
Suppose, now, that the household has an income lower than Y*, say Y (see diagram). Equal division
means that each member gets Y/2 and each person therefore has work capacity equal to the height of C.
Total household capacity is therefore given by twice this height, which is just the height of the point D.
Compare this to total household capacity if one person is allocated the entire income for consumption: it is the
height of the point E, which is greater. It follows that at incomes below the critical threshold Y* , unequal
consumption allocations create greater household work capacity than equal allocations . To the extent
that increased household capacity is good for future income-earning potential, we have a dilemma here.
202
Figure 8.3. The capacity curve and unequal allocation.
In contrast, at household incomes above the threshold Y*, equal division does better than unequal division.
The dotted curve ODB was constructed from the capacity curve: it tells us what household capacity is when
household income is divided equally. It lies below the individual capacity curve until the point B and then rises
above it.
This argument suggests why poverty is correlated with unequal allocation. Note well the culprit: it is the
“convex” section of the capacity curve, which captures the fact that certain minimal amounts are required as
nutritional input before productivity gains kick in. Without this section, equal allocations would always be
preferable.31
One reaction to this argument is that it is unrealistic: it is absurd to imagine that in the interests of
maximizing household capacity, one person will be left to starve. This is certainly not the lesson that I want you
to take away: the extremity of the result arises from the simplicity of the model. There are several reasons why
the extreme unequal outcome may not come about, not the least among them being that each family member is
loved and cherished. However, the situation creates a tendency toward unequal treatment to the extent that
the income-earning potential of the household is an issue of some concern.
One common solution to the lifeboat problem is to draw lots: this has at least the virtue of being egalitarian
ex ante. Drawing lots is not an entirely absurd proposition: providers of emergency care in a major disaster
effectively do it all the time. However, we are talking here not about a sudden disaster, but about an ongoing
process of nutritional development (so that drawing and redrawing lots on a daily or weekly basis has the
same effect as equal consumption). Thus the targets of discrimination are established once and for all: certain
individuals may be systematically denied nourishment and medical care, so that scarce resources can be
better focused on some remaining subset of family members.
The receiving end
Who are the individuals who are so denied? They are typically females, both adults and children, and—the
presumed harmonies of the extended family notwithstanding—the old and the infirm. Why the old should be
203
so treated is perhaps relatively easy to understand, especially in the light of the preceding model: nutrition and
medical care serve a functional role apart from being ends in themselves. They provide the foundation for
income-earning capabilities in the future. The old are in less of a position to provide these capabilities. To the
extent that income-earning objectives are internalized in the social dynamics of the family, the elderly will be
discriminated against. That is, no one individual needs to make these hard decisions, but the discrimination will
nevertheless manifest itself in the actions of every family member—perhaps even the elderly themselves.
Consider widows. Rahman, Foster, and Mencken [1992] studied mortality rates for widows in rural
Bangladesh, and Chen and Drèze [1992] carried out a related study for several villages in northern India. The
loss of a husband can be devastating in economic terms unless the widow owns assets such as land, although
here too matters are complicated, because the possibility of land loss can in turn depend on widowhood (Cain
[1981]). As Chen and Drèze [1992] observed, “the basic problem is not only that a widow often depends on
other household members to survive, but also that these other household members typically do not
depend on her for anything essential” (Italics added for emphasis).
Table 8.6. Age-specific death rates for widows in rural Bangladesh.
Source: Rahman, Foster and M encken [1992] and Chen and Dreze [1992].
Table 8.6 shows how age-specific death rates vary with widowhood in rural Bangladesh. The results are
striking. Overall death rates jump by a factor of close to 3 if a woman is a widow, rather than currently
married. In this group, widows who are heads of households do relatively better than the average for all
widows. Widows living in households that are not headed by themselves or by one of their sons do
particularly badly, 32 and the explanation for this cannot rest on the hypothesis that such households are, for
some reason, intrinsically poorer than other households: there is no evidence that households with a widow
have lower per capita expenditures than households without one (Drèze [1990]).
Observations such as these are not restricted to widows alone. In the context of medical care, Kochar’s
[1996] study of extended families in South Asia found that medical expenditures on the elderly vary
systematically (and inversely) with measures of their earnings ability, which implies that the role of the
household as a production unit looms large when nutritional or health expenditure allocations are made. This
bias is reflected not only in smaller allocations of medical expenditures to elders relative to the expected
incidence of illness in higher age groups, but sometimes in absolute terms as well.33
Once we accept the argument that intrahousehold allocation has functional as well as intrinsic motives, the
phenomenon of discrimination against the elderly is easy to understand. It is somewhat more difficult to
appreciate how a similar burden falls on females, both adults and children. Unless we believe that men are
more fit than women for tasks of various sorts, we cannot make the case for discrimination against women on
the basis of the lifeboat argument alone. Intrahousehold discrimination against females reflects the larger
context of gender bias. Suppose, for instance, that women provide household tasks while men earn income. If
household tasks are not properly monetized in the psychology of household resource allocation, then the
204
lifeboat argument applies and we would expect to see discrimination in resource allocation against women.
Likewise, even if women and men are both engaged in monetary employment, but wages to women for
comparable work are lower, this will bias resources away from women.
Table 8.7. Calorie intakes and requirements by sex in rural Bangladesh (1975–76).
Source: Sen [1984, Table 15.3].
The issue is made more complex by the question of how to measure nutritional deprivation. It may not be
enough to simply observe that women receive less nutrition than men: the question is whether they receive less
nutrition relative to their requirements. The evidence on this matter is not as clear-cut as you might expect. For
instance, the Institute of Nutrition and Food Science (INFS) conducted a sample survey of nutritional intake
by household in rural Bangladesh.34 They also used notions of “requirement,” namely, the age- and sexspecific recommendations of the FAO/WHO Expert Committee (1973). Table 8.7 summarizes some of the
INFS observations on calorie intake.
The table is interesting on two counts. First, the second and fourth columns tell us that females receive
systematically lower nutrition in all the age groups surveyed (and the age classification is pretty fine). The
intake shortfall varies from a minimum of 11% (in the youngest age group) and rises to a high of 44% in the
70+ category (in line with the observations on widows made earlier).
Second, and in contrast to the first observation, if the shortfall is measured relative to stated requirements,
this discrepancy goes away. A deficit remains relative to requirements at the two youngest age groups, but
there is a deficit for males as well. This raises the question of just what requirements are and how they are
measured. Apart from considerations of body mass, do they presume different sets of tasks performed by
men and women? In addition, how is it that the energy use of these tasks is accurately estimated without
pinning down a set of tasks completely? As Sen [1984, p. 351] observed, “. . . there are good reasons to
dispute the assumptions about the energy use of activities performed by women, which are not as ‘sedentary’
as calorie calculations tend to assume. Also the extra nutrition requirements of the pregnant women and
lactating mothers require fuller acknowledgement.” Measuring shortfalls relative to some arbitrary notion of
“requirement” can be dangerously misleading.
Thus gender bias may or may not be directly manifested in consumption-requirement ratios, as far as
nutrition is concerned. We may have to probe deeper. Very different sorts of allocation decisions may be at
work, even those that do not have any direct opportunity costs. A female child may not be taken to a clinic
when she is ill even if medical services are free. The cost of taking the child is not the cost of medical care, but
possibly the implied cost of dowry if the child survives to maturity. A female child may not be given education
or her education may be neglected simply because education of female children is not expected to pay off in
larger incomes for that household (and it may not lower the cost of a dowry either). The box on sibling rivalry
in Ghana is an example of research that looks for direct indicators such as these. Finally, sex-based
differences in infant mortality may take care of a large amount of discrimination: the survivors may be treated
205
relatively equally, but in looking for this we fail to count the dead.
These problems are magnified when we lack direct data on intrahousehold allocation and have to make do
with indirect evidence. Deaton [1994] discussed one such method: to look at household consumption of
certain “adult goods” (such as tobacco) and relate this to the proportion of girls in the household (controlling
for total number of children). If there is consumption discrimination against girls, this should be reflected in an
overall increase in adult consumption as the composition of children shifts in favor of females. Deaton [1989],
Subramanian and Deaton [1991], Ahmad and Morduch [1993], and Rudd [1993], among others, took this
interesting methodology to the data. No clear-cut findings were made, even in areas where other indicators of
discrimination (such as sex ratios) were positive. Deaton [1994] observed that “it is certainly something of a
puzzle that the analysis of expenditure patterns so consistently fails to show strong gender effects even when
they are known to exist.”
Sibling Rivalry: Evidence from Ghana
As in many other low-income economies, parents in Ghana often invest less in the human capital of their daughters than their
sons. Primary school enrollments are fairly even, but by secondary school only 28% of females between age 16 and 23 attend
school, whereas 42% of boys are enrolled.
A study by Garg and M orduch [1997] explored how economic constraints exacerbate gender differences in Ghana. The starting
point for this study is that even if parents desire to invest a given amount in their children’s human capital, they may lack the
personal resources to do so, and even if expected returns are high, parents may find it difficult to borrow for such long-term
investments. Children must then compete with their siblings for the resources currently available to parents. Boys have an advantage
in this competition if parents perceive higher returns to this investment. If the total number of their siblings is held constant,
children with fewer brothers also may get more resources than they would otherwise.
The Garg–M orduch study supports this hypothesis in the case of Ghana. For instance, the study shows that children aged 12–
23 with three siblings are over 50% more likely to attend middle or secondary school when all three of their siblings are sisters than
when the three are brothers. The effects are similar for boys and girls and for other sibling groups. Similar results hold for health
outcomes as well. The study is consistent with the idea of “sibling rivalry” caused by parents’ difficulty in borrowing to make
human capital investments in their children. The study illustrates the importance of considering issues of gender within the context
of markets and institutions available to households. The results suggest that improving financial systems can have important indirect
benefits for the health and education of children in Ghana.
What we have learned so far is that there are dimensions along which females are discriminated against,
but the obvious indicator of discrimination—nutrition—does not hold up well unless we have a precise notion
of requirements. There is the additional problem that direct intrahousehold data are hard to obtain. Where
they do exist—as in the Ghana study described in the box—and where data are collected on outcomes other
than nutrition, such as medical care and education, there is clear evidence of discrimination against girls (see
also Subramanian [1994]).
We must, therefore, seek to supplement this sort of research with indicators of differential educational
attainment, direct anthropometric indicators of differential nourishment, or indicators of differential mortality
and morbidity. These indicators are not without problems either, 35 but they serve as another route to
understanding the relationship between poverty and intrahousehold allocation.
Consider educational attainment. The World Development Report (World Bank [1996]) noted that for
low-income countries as a whole, there were almost twice as many female illiterates as there were males in
1995 (the illiteracy rates were 45% for females and 24% for males). This disparity is echoed by enrollment
figures: in low-income countries taken together, male enrollment in primary schools exceeded female
enrollment by over 12%, and the difference exceeded 30% for secondary schools.36 Note well that these are
averages for the countries as a whole. To the extent that the relatively rich in these countries are free of the
resource constraints that lead to discrimination, the corresponding figures for the poor in these countries must
be more dramatic still.37
Consider sex ratios: estimates of female-to-male population in the developing world. In North America
and Europe, the life expectancy of women is somewhat longer than for men. The roots of this difference are
unclear: they may be biological, but there are also possible social and occupational factors at work. The
average ratio of female-to-male population in these countries is around 1.05; that is, there are approximately
206
105 females for every 100 males. Figure 8.4 displays the corresponding sex ratios for many developing
countries. The first panel shows the African data, the second shows the data for Asia, and the last panel
shows the data for Latin America. It is evident that the problem of low female-to-male ratios is predominantly
an Asian problem. The figure for Asia is peppered with data points in the range of the mid-90s, and there are
several instances that are lower still.
These differences imply enormous absolute discrepancies. If the ratio of females to males is 93 (for every
100 males) in India, and India has approximately 440 million males (United Nations [1993]), then about 30
million women are unaccounted for in India alone.38 Thus sex ratios around 95 or so represent prima facie
evidence of substantial discrimination, which might include neglect in infancy or childhood (leading to death) or
practices such as sex-selective abortion.
Figure 8.4. Sex ratios (females per 100 males) in the developing world. Source: United Nations Secretariat [1996].
The relative absence of skewed sex ratios in Africa makes an interesting point. As we’ve noted before,
poverty alone cannot be responsible for the gender biases that we do see in Asia, although poverty serves to
reinforce these biases. The overall social context of discrimination also plays a role. Take, for instance, the
institution of dowry. Families might react to dowry by resorting to sex-selective abortion, female infanticide, or
discriminatory neglect during the infancy of a girl (which amounts to infanticide). Boys are preferred because
they are a source of income and support; girls are not because they impose costs. Nevertheless, once a girl
survives there may be less evidence of discrimination in matters of nutrition and medical care. After all, the
207
costs, say, in terms of the potential for marriage, are only enhanced in the absence of this care. Testing for
gender discrimination is therefore a complicated issue, and it may be unevenly manifested through various
potential channels. There is no reason to expect that all ways and forms of discrimination will be equally in
evidence.
8.5. Summary
Poverty, just like inequality, has intrinsic as well as functional aspects. We are interested in poverty in its own
right, as an outcome that needs to be removed through policy, but poverty also affects other forms of
economic and social functioning. It creates inefficiencies of various kinds and can exacerbate existing forms of
discrimination, such as those against women.
We first studied issues of poverty measurement. The measurement of poverty is based on the notion of a
poverty line, which is constructed from monetary estimates of minimum needs. We noted several problems
with the concept even at this fundamental level: should income or item-by-item expenditure be used to identify
the poor, are notions of the poverty line “absolute” or “relative,” is poverty temporary or chronic, should we
study households or individuals as the basic unit, and so on.
We then turned to well-known poverty measures: among these are the head-count ratio, which simply
measures the fraction of the population below the poverty line. The head-count ratio is a popular measure, but
it fails to adequately account for the intensity of poverty. In particular, a planner who uses the head-count ratio
as a political yardstick for poverty reduction will be tempted to target the segment among the poor who are
very close to the poverty line (and who are arguably not in the greatest need of help). To remedy this
shortcoming, we can use measures such as the poverty gap ratio or the income gap ratio, which look at the
total shortfall of poor incomes from the poverty line and express this shortfall as a fraction of national income
(as in the poverty gap) or as a fraction of the total income required to bring all the poor to the poverty line (as
in the income gap ratio). These measures add to the information contained in the head count, but have their
own drawbacks: in particular, they are indifferent to the relative deprivation of the poor (see the Appendix
to this chapter for more).
We then described some of the characteristics of the poor. Even going by conservative estimates, such as
India’s poverty line applied to the world as a whole, we see that in 1990, over 600 million people were poor.
Poor households tend to be large (though there are some qualifications attached to this statement) and they
are overrepresented by female heads of households. Rural areas tend to display more poverty. Poverty is
highly correlated with the absence of productive asset holdings, such as holdings of land. Poverty is correlated
with lack of education, and there is an intimate connection between nutrition and poverty, although nutrition
levels do not seem to rise as quickly with household income as we might suppose a priori.
The fundamental implication of poverty is that the poor lack access to markets, most notably the markets
for credit, insurance, land, and labor. We discussed how the absence of collateral restricts access to credit
markets and how problems of moral hazard and incomplete information restrict access to insurance. We then
began a study of imperfect access to the labor market (the threads of this story will be taken up again in
Chapter 13). The basic idea is that poverty and undernutrition affect work capacity. The relationship between
nutrition and work capacity can be expressed through the use of a capacity curve. The capacity curve
creates the possibility of a low-income undernutrition trap. Just as low incomes are responsible for low levels
of nutrition, low levels of nutrition work through the capacity curve...
Purchase answer to see full
attachment