The issue of malnutrition and income as presented in the literature
The thesis: (A counter-argument)
What to do then?
References
Ecol. of Food and Nutrition, Vol.37(2), 1998.
CLAUDIO SCHUFTAN
schuftan@gmail.com
Abstract:
This article takes issue with mainstream evidence in the literature purporting to show that increasing household income alone is not enough to do something significant towards combating malnutrition in the world. It argues that poverty lines used as a threshold to surpass (in order to show measurable improvements in the nutritional status of vulnerable women and children in the short term) are too low. It further points out that the primary determinant of nutritional status of young children may actually not be total household income, but rather that portion of it earned and controlled by the mother.
Modernization and market expansion have created consumption preferences that compete with food as a need thus leading to decreasing percentages of household income being spent on food. With competing expenditure options, the relative poverty line for a given location (based on current local food prices plus the prices of other needs) is clearly at a higher household income cut-off point than the absolute poverty line that researchers reviewed are using as a threshold. Households that raise their income barely beyond the poverty-line-set-too-low still do not show the expected improvements in the nutritional status of vulnerable household members. The point is made that poverty alleviation of an order of magnitude significantly beyond absolute poverty is needed for lasting nutritional improvement to occur; providing for an adequate minimum nutrition of all household members year-round is simply taking more, because the percentage of total income being spent on food is lower due to competing needs.
It is concluded that poverty-redressing interventions -as much as they may be in the economic realm only- remain central in the ultimate battle against malnutrition worldwide. The paper also explores how further income elasticity of demand assessments are key to redress the misleading perception that improvements in income alone do not lead to measurable improved nutritional status of vulnerable groups.
Ultimately, the paper warns of the dangers of basing mainstream policy to revert malnutrition in the Third World on the conclusions drawn from studies using what the author thinks is misleading data interpretation.
The issue of malnutrition and income as presented in the literature:
Based on recent studies and reviews in well over two dozen countries, strong claims are being made -based on a mass of often contradictory empirical evidence- that improvements in overall household income alone do not lead to improved nutritional status of mothers and under-five children in the short term; it is implied that health, nutrition and sanitation interventions have an edge over more directly combating poverty as policy choices to revert trends of malnutrition in the world. (IFPRI, 1985; Alderman, 1986; Kennedy and Cogill, 1987; Behrman and Kennan, 1988; von Braun et al., 1989; ACC/SCN, 1990; Bouis and Haddad, 1990; von Braun et al., 1991; TFNC, 1992; and especially Marek, 1992)
“Programs with the sole goal of enhancing income”, it is, for instance, claimed, “have not been as effective in alleviating malnutrition as expected… (Only) in the longer run, or in ‘some’ cases is increasing income likely to improve… nutrition.” (Alderman, 1986; Pinstrup-Andersen, 1991) A more compromising position admits that “income level has a positive effect on household food intake, but a less robust effect on nutritional status.” (Marek, 1992)
Despite this marginal overall effectiveness attributed to income improvements alone, the data do show that an improvement in nutritional status can be expected for the lowest income deciles households when overall household income indicators improve. (von Braun et al., 1989; Saenz de Tejada, 1989)
Computer-run models based on some of the International Food Policy Research Institute’s studies imply that even doubling household income would not significantly improve the nutritional status of vulnerable women and children across the board. (Kennedy, 1989; von Braun and Pandya-Lorch, 1991)
There are actually arguments against and in favor of the leading role of income in reverting malnutrition in the literature; the most salient ones are analyzed here before being discussed later on.
There are at least four further arguments that are used to downplay the importance of increasing household income alone (in-kind and disposable) as a central intervention to combat malnutrition: (See Table 1)
TABLE 1: Arguments casting a doubt about the leading role of income in reverting malnutrition:
ARGUMENTS + RESPECTIVE REFERENCE
1.- The inflection seen in the weight-for-age growth curve of some children in low income house-holds after 24 months of age shows that their growth curve is flat or descending between 12 and 24 months (when the infant still depends on others to be fed) and catches-up (resumes an ascending slope) thereafter when the child is assumed to be better endowed to fetch food for him/herself; this is interpreted as an indication that income-related household food security alone is not enough to prevent malnutrition in infants after weaning and that infant and child care/feeding practices are more important. (UNICEF, 1990)
2.- In longitudinal studies, children who are doing well at the baseline, have a high probability of doing well at follow-up, the opposite being true for those doing poorly; this is shown to be independent of the level of household income. (Kennedy, 1989)
3.- Parents of children with poor nutrition, them-selves have satisfactory weight or are overweight. (*) (Ferro-Luzzi, 1992)
4.- Total household calorie availability is sufficient in many households that still have malnourished under-fives. (*) (Pinstrup-Andersen + Garcia, 1984)
(*): Arguments 3 + 4 are equally used to downplay the importance of income.
On a technical note, one of my major criticisms of the literature reviewed is that not enough is understood about the timing or level of responsiveness of nutritional outcomes to varying economic measures such as changes in disposable income. Is there a lag period in the response and, if so, how long a lag time?
On the other hand, we know that, in response to abrupt income declines, the very factor households will protect through a variety of coping mechanisms (e.g., deterioration of the mother’s food consumption) is the short-term nutritional status of children; thus the question is: is the reverse true following an abrupt income improvement? (Csete, 1991) Although in a review of fifteen countries, a hike in income had the greatest effect on caloric intake during lean times, a 10% increase in income in families who consumed 1750-2000 calories per capita per day, only led to about a 4% increase in caloric intake. (von Braun and Kennedy, 1986; Bouis and Haddad, 1990)
As said, there are also arguments in the literature supporting the role of income in one way or another critically determining nutritional status; these are summarized in Table 2:
TABLE 2: Arguments directly or indirectly supporting the role of income in reverting malnutrition:
ARGUMENTS + RESPECTIVE REFERENCE
1.- Increasing rural income has been shown to improve food security in terms of access to food, significantly reducing malnutrition in countries with very low per-capita income. Doubling per capita income (from $300-$600) resulted in a reduction of about 40% in the prevalence of below standard weight-for-age of children. Looked at from the opposite vantage point, there is evidence that malnourished rural poor households tend to have much lower incomes than non-malnourished rural poor households. (von Braun and Pandya-Lorch, 1991, p.p. 11+29)
2.- Better nutritional status has been associated with higher wages. (Haddad + Bouis, 1991)
3.- Increases in income have been associated with increases in caloric intake of staple and non-staple foods, especially for the poorer households. (Alderman, 1986)
4.- The caloric intake of preschoolers has been shown to rise with increasing income. (Kennedy + Payonga-yong, )
5.- It has been pointed out that the effect of in-come on nutritional status may not always be through increased food consumption, but indirectly through achieving better access to clean water and sanitation. (von Braun et al., 1992)
6.- Time spent on generating income has been identified as an important (positive) determinant of household nutritional status. But the nutrition effect of increased income may be less than expected, because women may allocate more time to earning money and less to child care. Although results are situation-specific, the opposite may also be true: as more households pass the threshold of a realistic poverty line for a given location, women will be able to decrease their workload and free up some more time to devote to infant and child care and feeding. (Pinstrup-Andersen, 1991; von Braun + Pandya-Lorch, 1991,p.28 Leslie + Paolisso, 1990)
7.- A direct relationship has long been established between socioeconomic status and birthweights. (Kehrer + Wolin, 1979, TFNC, 1992)
8.- Policies that increase the prices of essential items, such as food, fuel or housing have shown to have profound repercussions on levels of malnutrition, especially if the consumption of these items is price-elastic. (Alderman, 1986; Cooper et al., 1990)
9.- Higher food prices have been shown to have a significant detrimental impact on the nutritional status of both adults and children, particularly those above 3 years of age. (Lavy et al., 1992)
10.- The combined effect of three factors -changes in distribution of income, level of income and staple food prices- has been shown to have a substantial impact on nutritional status. (Ravaillon 1992)
11.- In households with severely malnourished members, insufficient income has been shown to be the most limiting factor and education alone, aimed at mothers reallocating a given amount of real income to food, was not shown to be effective; i.e., nutrition education was shown not to change competing household priorities. (Garcia + Pinstrup-Andersen, 1987)
12.- Access to credit and labor-saving technologies, water supply, health and literacy have been identified as the explicit priorities identified by women themselves, with credit having been the top priority chosen. A program of collateral-free loans to women as the only intervention (used to start up village-based micro-enterprises) has proven to have significant positive health and nutrition impact on participating households. (Leslie + Paolisso, 1990; Ward, 1991)
13.- Finally, and most importantly, it has been reported that the primary determinant of nutritional status of young children (or at least of their food energy consumption) may actually not be (total) household income as such, but rather that portion of continual, small-sum (non-lump-sum) income earned and controlled by the mother. The issue of intra-household resource control and decision-making has been proposed as being central since women are more likely to spend more of their income on family welfare and on food. (Kennedy, 1989; IFPRI, 1990; von Braun and P.-Lorch, 1991, p. 28 Levinson 1991; Hoddinott + Haddad, 1991)
In all the studies reviewed, the common denominator actually is that it is for moderate malnutrition that it seems to be more difficult to show significant improvements when income alone increases.
Considering all facts found -and for the same reasons that I had insisted upon much earlier (Schuftan, 1979)- I have extreme difficulty in taking the evidence from the literature as ‘proof’ of the fact that “increasing-income-is-not-enough-to-effectively-combat-malnutrition in the world”. (Marek, 1992)
I am particularly weary about the normative implications such ‘proofs’ may have for policies to combat malnutrition in what is left of the century. I am talking about the dangers of basing mainstream policy to revert malnutrition in the Third World on the conclusions drawn from studies using what I contend is misleading data interpretation.
Finally, as pertains income distribution -a key indicator very few of the studies marginally touch upon- one has to keep in mind that too many “development” projects and most Structural Adjustment Programs have resulted in reductions in the disposable income of the poor, particularly the malnourished rural poor, thus increasing their household poverty and malnutrition (at least for sometimes long initial periods). At the extremes of poverty, most household income is spent on food and water (if the latter is not available free), plus a smaller amount on shelter, leaving little surplus. For the poor people concerned, reduced income leads to malnutrition and ill-health. Reduced income for the very poor can thus result from policies that cause either a reduction in income across the population or, more often, from development projects’ caused increased maldistribution of income.
The thesis: (A counter-argument)
Poverty is the major cause of protein-energy malnutrition and should not be (but is being) obscured and bypassed as the main determinant to tackle when choosing solutions to combat this form of malnutrition in Third World countries.
The extreme poor can be acceptably defined as those whose food-energy consumption falls below levels at which a healthy life is ensured. (von Braun and Pandya-Lorch, 1991) Since the most urgent problem we face in the world too often still is severe malnutrition (more frequently found among the poorest of the poor), anything we do to first eradicate these cases of malnutrition is already a laudable (and monumental) task. [It is estimated that 455 million people in developing countries already in 1988 were too poor to obtain sufficient energy for minimal activity among adults and healthy growth among children. (Millman, 1990)]. Implementing (and not only designing) measures that rapidly improve total income of the poorest rural populations thus seems to me to be urgent, equity-oriented and well worth the while.
As we saw, the literature says that malnutrition is effectively addressed by household income growth; but it is further said that increasing household income is not enough and health and sanitation actions are indispensable. (von Braun and Pandya-Lorch, 1991, p.p.xviii + 45; Marek, 1992) I have problems with the latter argument taken at face value, if nothing else, for the reason that nutrition, health and sanitation interventions before income growth have simply not convincingly proven to decrease the prevalence of malnutrition in any significant or sustainable way. (von Braun and Pandya-Lorch, 1991, p.29; von Braun et al., 1992)
In India, target-oriented health and nutrition interventions were recently assessed as having been able to only partially alleviate malnutrition among the vulnerable groups studied; it was judged that, to eradicate the worst forms of malnutrition, increasing the purchasing power of the poor and strengthening the public distribution system were indispensable. (Rao et al., 1991)
I, therefore, feel inclined to disagree with those who claim that income growth in poor areas has an indirect rather than a direct effect on nutritional improvement. (von Braun and Pandya-Lorch, 1991, p.29)
Additionally, one should not underestimate the so many other desirable effects that an improvement in especially non-farm/non-agricultural household income has, even if these effects are totally unrelated to nutrition. (Kennedy, 1989) (This non-farm/non-agricultural income represents 45-59% of rural income in the Third World; rural households with lower incomes, on average, have a greater degree of non-agricultural income, most probably out of sheer need). (von Braun and Pandya-Lorch, 1991, p.p.8, 34 + 35)
Expenditures on food and non-food items or services: Non-food expenditures of poor households often rise more rapidly with increased income than do food expenditures, the fact notwithstanding that income elasticities for such necessities are location- context- and time-specific. [Remember at this point that food expenditure per capita is a better and more dependable indicator of wellbeing than income. (Anand and Harris, 1990)].
Moreover, the availability of desired local goods and services is considered to play a crucial incentive role for rural growth.
This availability of affordable goods and services explains why patterns of household spending and consumption -while rational to the household head- may not necessarily be optimal from a nutritional perspective; it also explains why, if development policies are geared to increase the availability of these goods and services at low prices, nutrition is likely to suffer (due to a substitution effect) if overall household income does not increase substantially at the same time. (In a regression model using data from 21 countries, the income elasticity of nutritional improvement ranged only from .33 to .35). (von Braun and Pandya-Lorch, 1991, pp.xviii, xix + 11; von Braun et al., 1992)
It is important to keep in mind that poverty is essentially -but not only- a matter of low income; poverty is also related to the cost of locally acquiring certain commodities and services; being able to afford these determines an income- or expenditure-based poverty line (heavily based on minimum-cost-diet considerations). If poverty is measured directly as under-consumption of calories, then one is measuring absolute poverty with a (food) poverty line cut-off point at 80% of recommended caloric consumption. (von Braun and Pandya-Lorch, 1991)
The relationship between absolute poverty (measured as prevalence of malnutrition) and average rural per capita income is strong. Even if the cut-off point chosen to define poverty is nutritional, malnourished rural poor households still need mostly income (in cash or in-kind) to surpass that poverty line, no matter how the latter is defined. The cut-off point thus almost unavoidably gets monetized!
Interestingly, low income/low expenditure in severely malnourished rural poor households correlate considerably stronger with an anthropometric indicator (weight for age in under-fives) than with overall caloric under-consumption; this, actually comes as no surprise since the weight deficit is a poverty outcome indicator that is more directly linked to the percentage of income spent on non-foods, thus being more linked to relative than to absolute poverty. (von Braun and Pandya-Lorch, 1991, p.p.xiv, 10, 30, 31, 33 + 68)
Poverty lines and income elasticities of demand: I remain convinced that the main flaw in the reasoning in all the literature criticized above is that: cut-off points for absolute poverty lines are being used to measure the significance of improvements in household income and these cut-off points are set at levels I believe too low to show measurable improvements in the nutritional status of household members in the short term when surpassed.
Ergo, households that raise their income barely beyond that poverty-line-set-too-low still are not showing the expected improvements in the nutritional status of their more vulnerable household members.
What I contend has happened, both in urban and in rural areas in the Third World, is that the relentless twin processes of modernization and market expansion -particularly as seen in the last twenty years- have caused income elasticities of demand of the poor for non-food items to catch up or even surpass those for staple foods. In China, this has been shown for income elasticities for more expensive and/or processed food items, as well as for an array of diverse services: they are simply often higher than those for staple foods. (Yang et al., 1990)
Modernization and market expansion have basically created alternative consumption preferences/needs that compete with food as a need thus leading to decreasing percentages of disposable household income being spent on food. In short, the trade-offs the mother has to make divert her effective demand away from traditional foods, towards food variety and towards non-foods.
In the rural areas, elasticity of demand of peasant households crucially depend on the availability of affordable (low-priced) manufactured consumption goods such as textiles, footwear, processed foods and beverages, building materials, transistor radios and means of transportation. Modernization and market expansion are making this increasingly possible. (von Braun and Kennedy, 1986; Behrman and Deolalikar, 1988; de Janvry et al., 1990; von Braun and Pandya-Lorch, 1991, p.19)
With competing expenditure options/preferences/needs for income disposal (even greater in urban areas), the relative, but more realistic, poverty line for a given location (linked to current local food prices plus the prices of other felt needs) is clearly at a higher household income cut-off point than the absolute poverty line that researchers are using as a threshold that needs to be surpassed to improve nutrition.
Poverty alleviation significantly beyond absolute poverty is needed for lasting nutritional improvements: Providing for an adequate minimum nutrition of all household members year-round is simply taking more, because the percentage of total income being spent on food is lower due to competing needs.
Moreover, in urban areas the amount spent per-calorie increases with rising income; therefore, the minimum level of total income required for adequate household nutrition increases further with modernization. (von Braun et al., 1989; Bouis and Haddadd, 1990; NCH, 1991) This probably explains why, in Asia, other social indicators of welfare have gone up at a higher rate than nutrition indicators in the last 15 years. (Jonsson, 1995)
Therefore, in the presence of competing needs (and choices) for household income expenditure, the relative (higher) poverty line -that would still allow sufficient allocation of cash for enough food for the household- is probably quite significantly higher. Distributions below and above an income-based or expenditure-based poverty line are thus particularly important to determine these households’ respective percentages of total income spent on food. (von Braun and Pandya-Lorch, 1991, p.1)
[We are reminded here that measuring poverty entails two distinct problems, namely, defining a poverty line (allowing a ranking), and defining the characteristics and the intensity of the poverty of those falling below the line (and household spending behavior clearly contributes to defining the latter characteristics). (Sen, 1982)]
With others (Wilkinson, 1992), I thus caution readers to reassess whether relative poverty is not more relevant for us to consider for effective policy-making than the narrower spectrum of determinants that define absolute, extreme (food) poverty (which a World Bank staffer euphemistically calls “hard-core poverty”, I guess to emphasize that poverty is as much of a scandal as pornography…). (World Bank, 1992) In a way, people -through their economic behavior- determine whether they are poor… (Schuftan, 1979)
For better or for worse, the people are the ultimate judges of where their priorities lie, and trying to improve their nutrition ignoring how they are voting with their pockets will only keep pushing us in the wrong policy directions. Therefore, under current circumstances, trying -for example through education- to significantly limit the diversion of household and female income from traditional foods to other items is a futile enterprise. (Garcia and Pinstrup-Andersen, 1987)
These are the reason, then, why I posit that household income, as a major socioeconomic status indicator, may not be correlating positively with the nutritional status of women and children in the short term in the studies criticized above.
It is really regrettable that the empirical experience thus far reported tends to indicate that even well conceived studies demonstrate a direct link between nutrition and macro-level policy issues only with great statistical difficulty, with extraordinary effort and expenditure and too often unconvincingly. (Csete, 1990)
I think that orders-of-magnitude-in-the-income-increase have been too low for the interpretation of such studies to shed any positive light on the above link. The fact that diversity in income levels of households has not been showing as differences in nutritional status in those studies has most probably been due to the fact that the range of this diversity has been too narrow to make a difference.
[From empirical experience, as an expatriate living in the Third World, I feel one can be quite categorical about this issue of orders of magnitude: the local urban minimum wages would indeed require a several fold increase to allow a family of five or six to purchase sufficient food].
Note, finally, that I am not here blaming modernization and the market; I am here voicing a strong disagreement with and raising a counter-argument against those researchers that are concluding -without more strictly qualifying their conclusions- that income increases (evidently too low to have a nutritional impact) are having negligible effect in reverting malnutrition. This, because, among other, they are not considering this fast growing elasticity of demand for non-foods. They surely empirically accept that poverty is at the base of malnutrition, yet they are shy to the need to guide the recommendations they make toward income increases/redistribution of the order of magnitude that could have a chance of making a difference.
If in the regression model quoted earlier (in Table 2, No. 1) -that showed that rural preschool malnutrition would be reduced by 14 percentage points (a 40% reduction) if per capita rural income increased from $300 to $600/yr.- one actually examines the full regression curve, it turns out that, for the same 27 countries with GNP per capita of less than $1,000/yr., a 300% increase in income, from $200 to $800, would actually reduce the prevalence of malnutrition by roughly 26 percentage points (a 60% reduction), certainly a laudable improvement. (von Braun and Pandya-Lorch, 1991, p.p. 10 + 11)
I thus conclude that mainstream thought and man-made models in this area are somehow not accurately interpreting the facts. I find logical inconsistencies that I think call for alternative interpretations of the data. (Schuftan, 1995) For now, the danger I see is that mainstream policy is currently based on these studies’ conclusions.
I acknowledge that, for now, this paper mainly critiques the mainstream paradigm portrayed in the literature reviewed; it does not elaborate on an alternative, one beyond some generalizations that show, at most, a direction. What follows is an attempt at defining such direction.
What to do then?:
If as researchers we want to contribute to making a meaningful difference for the people we research on, I think it is now called for us to launch a series of worldwide income elasticity of demand studies to reassess the role of modernization and market expansion on household expenditure behavior of at least the lower quartile income group in specific rural and urban settings.
I suggest we also need to commission corresponding, complementary income distribution studies to periodically reassess and document the (more often than not) relentless process that is further skewing income distribution to the detriment of the poor. (WHO, 1995)
Only then will we know who the lower quartile is, how this group’s income relates to the poverty line, how nutritionally at risk the people under this cut-off point are and how they behave economically. (Schuftan, 1985) It is not that this type of studies is not carried out at all, but they need to be updated and published periodically (and not only sporadically) and the data used for setting up corrective measures.
Once we have reassessed the percentage of household income the poorest decile households realistically spend on food year-round both in rural and urban settings, then we have to carry things one step further and estimate more objectively how significant the increase in household income will have to be to better assure covering minimum caloric needs for all household members. I suspect this will most probably be more than the 100% increase in the poorest households’ income the computer models of IFPRI looked at. (Kennedy, 1989; von Braun and Pandya-Lorch, 1991)
It is even more. I contend it is conceivable that such threshold income studies may find locus-specific cut-off points for percentage of total net household income spent on food, below which malnutrition rates of under-fives drop sharply. (I predict that this percentage of total income spent on food that has a preemptive effect on malnutrition will most often be below 55%; and to reach this point, rather sizeable increases of overall household income will have to be realized). Although such a rough index risks being limited to particular settings, it can increasingly become a very useful indicator in the battle against malnutrition and could lead to a more systematic characterization of households at risk. (Schuftan, 1992) As an indicator, this percentage-of-income-spent-on-food index would complement the one already suggested earlier that estimates the number of hours of work necessary to buy a basic ‘food basket’ which incorporates both wage and price effects. (Alderman, 1986)
All interventions geared at increasing disposable household income to ultimately affect nutritional status of all household members need to be rooted in an understanding of existing local household income strategies and their respective underlying comparative advantages.
A list of some of such potential interventions is presented in Table 3:
TABLE 3: Potential interventions to increase disposable household income: (*)
– Investments in human capital formation including literacy and primary/technical education.
– Time/labor-saving technological changes.
– Targeted price subsidies (food subsidies, for example, increase the real income of households with access to the subsidies). (**)
– Rural income diversification to strengthen the households’ year-round income generation capacity. (***)
– Wage employment generation (preferably with wages paid on a weekly basis).
– Access to credit/saving schemes (to alleviate production instability and uneven seasonal income streams). (***)
– Enforcing minimum wages, social security and equitable access to/ownership of the basic means of production.
(*): All these to be seriously considered with a skew towards women.
(**): von Braun et al., 1992
(***): von Braun and Pandya-Lorch, 1991.
With the more direct causes of poverty differing according to location and the degree of purchasing power for different goods and services varying from diverse income sources, there is little basis for making generalizations and for deriving blanket conclusions pertaining income targeting. But a ‘package’ with combinations of the above-mentioned interventions -having the additional potential of generating sizeable income and employment multipliers elsewhere in the economy- would, if significant, exercise a strong enough influence over the income and nutritional status of the malnourished poor; the latter respond rather quickly to adjust their income strategies to take advantage of favorable new circumstances. (Pinstrup-Andersen, 1991; von Braun and Pandya-Lorch, 1991, p.p. 39, 42 + 43; Marek, 1992)
All that would be left, then, would be to fine-tune how to best improve the food security effect and nutritional impact of income-generating policies themselves. (Kennedy and Haddad, 1991)
Only after I see such a ‘package’ being used in an effort to determine the role of income improvements alone on the nutritional status of the more socially vulnerable, will I let myself be convinced that such improvements do or do not lead to improved nutritional status in the relatively short term.
China can be taken as an example: During the period 1978-88, nominal per-capita incomes quadrupled in rural areas and more than tripled in urban areas (real per-capita incomes more than doubled in rural areas while the increase in urban areas was about 87%); at the same time, an improvement in the rural-urban income equality was also attained. (Pinstrup-Andersen et al., 1990)
Income transfers to the rural and urban poor did become a cost-effective means of significantly expanding their food consumption. (Yang et al., 1990; Pinstrup-Andersen et al., 1990)
Higher incomes resulted in higher consumption of calories, protein and fat. Thus, increasing incomes that much did result in an increase in the demand for almost all major foods except low prestige coarse grains. In the process, the diet became more diversified. Households made large adjustments in their consumption of calories and protein in response to income increases, much more so than in response to changes in relative food prices. (Zhu, 1990) (One can here also empirically consider the fact that Latinamerica, for instance -with much better nutrition indicators than Africa- has on average, three times more income per capita than Africa).
All that is said in this section does not amount to a policy prescription. In fact, alternatives listed are only very lightly, if at all, discussed; a full other paper would be needed to justify each of them.
Finally, let me emphatically state that the last thing I wanted to do here was to grossly downplay the role of non-income related variables (e.g., morbidity, household size, education), and non-income related interventions (e.g., health, water, sanitation, nutrition and child care interventions) in determining and affecting the nutritional status of vulnerable groups – especially if properly targeted. (Kennedy, 1991)
But somehow, current interpretation of available data is leading to a double message: income generation has a role to play in reverting malnutrition in-the-long-run, and non-income related interventions in-the-short-run. As a result, policies derived from this message notoriously first and foremost target the short term interventions -considered to be more ‘doable’- and leave the long term ones for ‘others’ to do something about.
I repeat, large absolute increases in poor households’ incomes are necessary along with synergistic public expenditures in health, water, sanitation, nutrition and women’s and children’s care in order to root out poverty-linked malnutrition in the world.
In the name of what I consider fairness, I have only here attempted to resurrect the (endangered) concept that poverty still is the major determinant of protein-energy malnutrition in the world as conventional wisdom always has indicated. I am thus putting all those other variables and interventions in the proper context and perspective, and calling on all of us for a more determined and concerted action on the income determinants.
I still see poverty-redressing interventions, as much as they may be in the economic realm only, as central in the ultimate battle against malnutrition worldwide, particularly against malnutrition of the more severe type, the one we decisively want to rid this planet of.
We need to advance realistic policies and interventions that will more effectively redress poverty and its accompanying ill-health and malnutrition. It is policy failure, more than anything else, that prevents us from overcoming constraints to equity.
References
ACC/SCN. (late 1990). SCN News, 6, p.46.
Alderman, H.(1986). The effects of food price and income changes on the acquisition of food by low-income households. IFPRI Research Report, May.
Anand, S. and C. Harris.(1990). Food and standard of living. In J. Dreze and A. Sen (Eds.), The Political Economy of Hunger. Clarendon Press, Oxford.
Behrman, J.R., and A.B. Deolalikar. (1988). Is variety the spice of life?: Implications for nutrient responses to income. Discussion Paper No.1371, Harvard Inst. of Econ. Res., March.
Behrman, J.R., and W.R. Kennan. (1988). Nutrition and incomes: Tightly wedded or losely meshed?. Pew/Cornell Lecture Series on Food and Nutrition Policy, Ithaca, New York.
Bouis, H., and L. Haddad. (1990). Effects of agricultural commercialization and land tenure, household resource allocation and nutrition in the Philippines. IFPRI Research Report No. 79.
Cooper, D.E, et al. (1990). The Impact of Development Policies on Health: A Review of the Literature. WHO, Geneva, pp. 29 and 161.
Csete, J. (1991). The use of nutritional measures in consideration of effects of adjustment policy in Africa. In Deng, L., M. Kostner, and C. Young. (Eds.), Democratization and Structural Adjustment in Africa in the 1990s. University of Wisconsin-Madison, pp. 135-141. de Janvry, A., M. Fafchamps, and E. Sadoulet. (1990). Peasant household behavior with missing markets: Some paradoxes explained. As cited in Democratization and Structural Adjustment in Africa in the 1990s, p. 21. (See reference above).
Ferro-Luzzi, A., et al. (1992). As cited in A. Berg, New and Noteworthy in Nutrition, 18, p.5, September 11, World Bank.
Garcia, M., and P. Pinstrup-Andersen. (1987). The pilot food price subsidy scheme in the Philippines: Its impact on income, food consumption and nutritional status. IFPRI Research Report No. 61.
Haddad, L.J., and H.E. Bouis. (1991). The impact of nutritional status on agricultural productivity: Wage evidence from the Philippines. Oxford Bull.of Economics and Statistics, 53(1), 45-68.
Hoddinott, J., and L.J. Haddad. (1991). Household expenditures, child anthropometric status and the intra-household division of income: Evidence from the Cote d’Ivoire. Mimeo, IFPRI.
IFPRI. (1985). Nutrition effects of technological change in agriculture and related public policies and projects: selected case studies. IFPRI Report to IFAD.
IFPRI. (1990). 1990 Report.
Jonsson, U. (1995). Success factors in community-based nutrition-oriented programs and projects. Draft report from a rapid appraisal of programs and projects in South Asia, p.4, November.
Kehrer, B.H., and C.M. Wolin. (1979). Impact of income maintenance on low birthweight. J. Human Resources, 14, 434-462.
Kennedy, E., and B. Cogill. (1987). Income and nutritional effects of the commercialization of agriculture in SW Kenya. IFPRI Research Report No. 63.
Kennedy, E. (1989). The effects of sugarcane production on food security and nutrition in Kenya: A longitudinal analysis. IFPRI Research Report No. 78.
Kennedy, E. (1991). Successful nutrition programs in Africa – What makes them work?. Policy, Research and External Affairs Working Papers, WPS 706, The World Bank, June.
Kennedy, E., and E. Payongayong. (1991). Inventory of food and nutrition monitoring systems. Mimeo, IFPRI.
Kennedy E., and L. Haddad. (1991). Testimony in front of the US Congress, October.
Lavy, V., D. Thomas, and J. Strauss. (1992). Public policy and anthropometric outcomes in Cote d’Ivoire. LSMS Working Paper No. 89, World Bank.
Leslie, J., and M. Paolisso (Eds.) (1990). Women’s Work and Child Welfare in the Third World. AS Selected Symposia Series 110, Westview Press, Boulder, CO.
Levinson, J. (1991). Addressing malnutrition in Africa – Low cost program possibilities for government agencies and donors. Social Dimension of Adjustment in Sub-Saharan Africa Working Paper No.13, World Bank.
Marek, T. (1992). Ending malnutrition: Why increasing income is not enough. Technical Working Paper No. 5, The Worl Bank, Afr. Tech. Dpt., PHN Div., October.
Millman, S. (1990). Hunger in the 1980s: Backdrop for policy in the 1990s. Food Policy, 15(4), 277-285.
NCH. (1991). NCH Poverty and Nutrition Survey of Britain. National Children’s Home, London, June.
Pinstrup-Andersen, P., and M. Garcia. (1984). Household vs. individual food consumption as indicators of the nutritional impact of food policy. Paper presented at Tufts University, October 23.
Pinstrup-Andersen, P., D. Yang, Z. Xian, and Y. Yang. (1990). Changes in incomes, expenditures and food consumption among rural and urban households in China during the period 1978-88. Proceedings of the Intl. Symposium on Food, Nutrition and Social-economic Development, 447-458, Beijing, June.
Pinstrup-Andersen, P. (1991). Targeted nutrition interventions. Food and Nutr. Bull., 13:3, p.168.
Rao, N.P., J.G. Sastry, A.N. Naidu, and V. Reddy. (1991). Trends in Nutrition – NNMB Survey. Nutrition News, 12:3, National Institute of Nutrition, Hyderabad, May.
Ravaillon, M. (1992). Does undernutrition respond to incomes and prices? Dominance tests for Indonesia. The World Bank Economic Review, 6(1), January.
Saenz de Tejada, S. (1989). Food consumption and its relation to production: A survey in Lesotho. Mimeo. Univ. of Arizona, Phoenix, AZ.
Schuftan, C. (1979). Household purchasing power deficit -a more operational indicator to express malnutrition. Ecology of Food Nutr., 8:1, 29-35, June.
Schuftan, C. (1985). Hunger and malnutrition: Outlook for changes in the Third World. J. Trop. Pediatrics, 31:6, 299-300, December.
Schuftan, C. (1992). The household entitlements revolution, or an oikonocentric (household-centered) approach to family security, unpublished open letter to the Director General of UNICEF, August 9.
Schuftan, C. (1997). Development nemesis. Africa Development, 2.
Sen, A. (1982). Choice, Welfare and Measurement. Basil Blackwell, Oxford. (As cited in Democratization and Structural Adjustment in Africa in the 1990s, p.2; see ref. by Csete above).
TFNC. (1992). The food and nutrition situation in Tanzania. TFNC Report No.1445, p.20. Country Paper for the ICN in Rome, December.
UNICEF. (1990). Nutrition Strategy. New York.
von Braun, J., and E. Kennedy. (1986). Commercialization of subsistence agriculture: Income and nutritional effects in developing countries. Working Papers on Commercialization of Agriculture and Nutrition, IFPRI..
von Braun, J., D. Hotchkiss, and M. Immink. (1989). Non-traditional export crops in Guatemala: Effects on production, income and nutrition. IFPRI Research Report No. 73.
von Braun, J., D. Puetz, and P. Webb. (1989). Irrigation technology and commercialization of rice in The Gambia: Effects on income and nutrition. IFPRI Research Report No. 75.
von Braun, J., H. de Haen, and J. Blanken.(1991). Commercialization of agriculture under population pressure: Effects on production, consumption and nutrition in Rwanda. IFPRI Research Report No. 85.
von Braun, J. and R. Pandya-Lorch. (Eds.) (1991). Income sources of malnourished people in rural areas: Micro-level information and policy implications. IFPRI Working Paper on Commercialization of Agriculture and Nutrition No. 5.
von Braun, J., H. Bouis, S. Kumar, and R. Pandya-Lorch. (1992). Improving food security of the poor: Concept, policy and programs. Mimeo. IFPRI.
Ward, V.E. (1991). The impact of women’s income on community health in Ghana. Women’s International Public Health Network News, 10, p.7, Winter.
WHO. (1995). Bridging the Gaps: The World Health Report 1995. Geneva.
Wilkinson, R.G. (1992). National mortality rates: the impact of inequality?. American J. of Public Health, 82:8, 1082-1084.
World Bank. (1992). World Bank structural adjustment operations. Mimeo. Operations Evaluation Department, June.
Yang, D., Y. Yang, and P. Pinstrup-Andersen. (1990). Food consumption and food acquisition behavior among urban households in China 1987 and 1988. Proceedings of the Intl. Symposium on Food, Nutrition and Social-economic Development, p.31, Beijing, June.
Zhu, X., Z. Xian, and P. Pinstrup-Andersen. (1990). Food consumption and food acquisition behavior among rural households in China 1988. Proceedings of the Intl. Symposium on Food, Nutrition and Social-economic Development, Beijing, June.
Claudio Schuftan
Ho Chi Minh City, Vietnam.