Previous Page Table of Contents Next Page


WHY NUTRITION IMPROVES


The Role of Poverty Alleviation Schemes in Improving Nutrition
Why Nutrition Improved in Andhra Pradesh
Why Nutrition Improved in Tamil Nadu

The Role of Poverty Alleviation Schemes in Improving Nutrition

The issue of how to alleviate poverty (and thus poverty-related nutritional deprivation) in India has generated several, long sustained controversies, the bases of which are outlined here (see also Clay et al. 1988):

i) Growth or welfare? Planning orthodoxy has it that investing in productive activity has to precede investment in welfare, that the latter is most dispensable and that generalized welfare can be replaced by targeted welfare. This conflicts with evidence from Sri Lanka, Kerala, Costa Rica where striking welfare gains derived from long-term investment in education, health and social welfare, in the absence of prior growth and in the presence of considerable poverty and low levels of food consumption. Is there necessarily a trade-off between growth and equity (poverty alleviation) in Indian development? This has been called simplistic (Basu 1991) because i) there is sufficient slack in the Indian economy for both growth and equity (it need not be a zero-sum game between the two), ii) there may be no relation whatsoever between the two anyway, and iii) the fact that policies tend to come in bundles (which can be disaggregated) may exaggerate the notion of a necessary trade-off. Basu argues against the belief that it is necessary to prune poverty alleviation programmes if growth is desired, asserting that it is a question of keeping the fiscal burden manageable (e.g. through targeting), providing the political climate remains stable and conducive.

ii) Poverty alleviation or social security? Investment in assets and livelihood creation are said by some to have greater consumption multipliers than social security. On the other hand, unemployment, sickness and old age benefits are precisely targeted to those of the poor in households in crisis who are least able to access poverty alleviation programmes as presently administered. Another factor is relative cost - administrative costs of poverty alleviation programmes generally exceed those of social security provision, although the types and targeting of social security are controversial issues.

iii) Asset endowment or wage employment? Regarding feasibility, some argue that an insignificant fraction of the poor can be given assets without land and property reform, which is the only guarantee for a mass base to growth. Asset endowment programmes such as IRDP strain administrative and financial capacity, and are more open to corruption e.g. asset capture by untargeted elite groups. Employment provision can be self-targeting and is asserted to increase consumption linkages more than assets creation. It may also preferentially benefit the incomes of women who often form a majority of the workforce. Wage differentials between sexes can be eliminated, with possible knock-on effects on local labour markets. The assets built may include roads, drainage and irrigation channels, which could themselves provide a long-term source of employment, through repair and maintenance as well as stimulating rural non-farm activities, especially trades and services. Reforestation, erosion control and the rehabilitation of agricultural land are other possibilities. On the other hand, some would argue that the relegation of poor to wage labour does not promote long-term independence.

iv) Cash or kind wage payment? Cash is much more feasible administratively, less vulnerable to corruption and leakages, although cash payment requires that such newly created demand can be satisfied through marketing channels that function reasonably well. If this is not the case (as in many famine situations), then kind wages, as food, may be a better option. Kind payment recognizes both that buffer stocks need turning over and that markets may not function efficiently.

v) "Top-down" or "bottom-up"? A top-down approach would argue for the need to control funding, avoid low level administrative corruption, and ensure regional equity. A "bottom-up" community-based approach would stress the need for flexibility. Administrative capacity may be improved by an increase in responsibility at lower echelons. The direct involvement of the "targets" themselves is truly democratic.

vi) Targeting versus universalism. This is relevant to many interventions, not just poverty alleviation. While the fate and final shape of the draft nutrition policy is yet to be decided, among the key factors likely to influence and justify government commitment to nutrition is the notion of targeting actions to those most needy - particularly relevant in the face of the severe economic situation which is not expected to ease in the short or medium term. The issue of targeting is not clear-cut. The argument in favour hinges on the efficient selection of the most needy reducing overall programme costs. However, targeting is not in itself cost-free, for the following reasons:

- the selection of beneficiaries based on application of certain criteria is costly to administrate;

- the targeted (usually the poorest groups) do not have sufficient political power to ensure the services they receive are of sound quality;

- the notion of targeting implies passive acceptance of a benefit, and can be demeaning for the recipient;

- the system of selection may be actively distorted by some non-needy households (e.g. income under-reporting) to qualify for eligibility;

- there may be an disincentive element whereby borderline eligible cases reduce their output/status to qualify;

- targeting is invasive involving constant probing and policing.

This is not at all to suggest that targeting is not a good idea, just that the question of how far the process of selection should be pushed, needs to be considered carefully. The costs above need to be weighed against the overall costs of universal coverage of a programme. One option with great potential is self-selection of the beneficiary. This can be done, for example, through subsidising less-preferred coarse grains - normally the staples of the poorest groups (i.e. commodity targeting in the PDS), or providing guaranteed work at wage rates that, while providing for a decent livelihood, are not sufficiently high to attract less-poor groups (e.g NREP).

While the ‘nutrition-connection’ between poverty alleviation programmes and nutritional well-being is implicit in the definition of a poverty line based on calorie criteria, neither the NREP or IRDP have been explicitly evaluated for their effects on nutrition. Poverty alleviation schemes are potentially highly important for future nutritional improvement. Associations between poverty reduction and improved nutritional status are well documented in the literature, and it is also evident from prior research (e.g. Leslie and Paolisso 1989), that an improvement in women’s income has a much greater impact on nutritional status of vulnerable groups as compared with a comparable improvement in total family income.

In the face of the limited success of the IRDP based on poverty criteria especially in reaching female-headed households as well as the poorest groups, the plausible impact on nutritional well being is likely to be small. In Andhra Pradesh, nonetheless, area targeting of IRDP is high, thus improving the likelihood of some positive impact. Investment and self-employment schemes such as the IRDP and TRYSEM may not be suited to the needs and level of infrastructure development of the poorest states. A tendency towards such inappropriateness is seen in a state-wise analysis wherein Bihar and West Bengal (among the poorest states) have received much lower IRDP inputs than would be justified by their poverty ratios. On the other hand, the poor in the better-off states, such as Andhra Pradesh, may be able to derive greater benefit from such schemes (to that extent, ironically, a mismatch between IRDP inputs and state-wise poverty ratios may in fact facilitate the programme’s impact on such groups).

In the context of the NREP and RLEGP at an all-India level, the fact that the two programmes together provide only about 1% of total employment, and women’s participation (as well as those of scheduled castes/tribes) is as low as 20%, nutritional impact is likely to be limited. But again, in both Tamil Nadu and Andhra Pradesh, where women’s participation is much higher (42% and 44% respectively) than other states, it is likely that NREP had a relatively greater positive impact on nutrition. No evaluations, except one small study (Radhaiah et al 1985) touch upon the potential or accrued nutritional benefits. Nutritional objectives should be explicitly defined for poverty alleviation schemes, reflected in programme design, and thereafter used to evaluate nutritional benefits against in all future efforts. The disproportionately higher share of both Andhra Pradesh and Tamil Nadu in NREP, as also the greater participation of women in the NREP may have contributed to the observed nutritional improvements in these states. In the absence of hard data, however, these conclusions must remain tentative.

Given the history of anti-poverty programmes in India, self-targeting rural works schemes may be seen as a better alternative for the poorest, provided these are preceded by adequate planning to link community infrastructure needs with works, and better area-based and seasonal targeting. Such a shift in policy will necessitate further decentralization of planning to the district/block level - where needs and opportunities can be better assessed by beneficiary communities and households. While some self-targeting is intrinsically built in to rural works programmes (and this may be NREP’s greatest advantage), the scope for focusing in further on the poorest needs to be capitalized upon, especially in the light of the changing economic scenario in the country. Thus it may be nutritionally a better option to provide more person-days to fewer (poorer) people. This built-in self-selection of the poorest, however, does not include the old and disabled, so programmes such as the NREP (now rechristened Jawahar Rojgar Yojana or JRY) should be complemented with social security for the old and disabled. Finally, there is a need for a better monitoring and information system to improve accountability and reduce corruption, and an explicit recognition of nutritional objectives from planning to evaluation.

Earlier sections of this report have shown that the improvement in nutritional status of young children in Andhra Pradesh and Tamil Nadu has been modest over the last fifteen years, with the improvement being manifested in a decline in severely underweight prevalences (through a rightwards shift in the anthropometric distribution) with little change in other categories. In the following two sections, we draw upon data presented earlier to analyze and prioritize the factors that may have contributed to this decline in the two states, and those that potentially limited it. Tables 9 - 11 summarize salient aspects of the nutrition-relevant actions in Andhra Pradesh and Tamil Nadu examined in the sections above. Table 9 considers whether interventions are underway in the right places (coverage with respect to need, area and beneficiary targeting), whether the objectives and services are appropriate to the problem, as well as the quality of implementation and whether any impact has been documented. Table 10 assesses interventions from a Triple A perspective, and addresses the following questions: was ‘need’ identified?, was the intervention targeted with respect to this need?, was the community empowered by the intervention?, and was institutional support built into its design? Finally, Table 11 provides a matrix which aims to score different interventions with respect to certain factors (identified by Shrimpton 1989) relating to the type and extent of community involvement. In each of Tables 9 to 11 the interventions are scored according to a scale of 0 to 5 (worst to best). This has been done by the authors, based on the analysis of the different actions presented here - hence subjective based on objective data.

TABLE 7

THE PUBLIC DISTRIBUTION SYSTEM IN TAMIL NADU
TARGETING OF PDS COMMODITIES
(As on 31st March, 1988)

SI. No.

Name of the commodity

Scale of supply

Retail issue price (Rs./kg)

1.

Rice

For Card holders whose monthly income is below Rs. 1,000 raw rice or boiled rice at 4 kg per adult and 2 kg per child subject to the maximum of 12 kg. per family in addition to that of 3 kg. raw rice per family card.




Common rice Rs 1.75




Fine rice Rs 2.25


For Card holders whose monthly income is above Rs. 1,000 raw rice only is issued at 4 kg per adult and 2 kg per child subject to the maximum of 12 kg. per family card.




Superfine rice Rs 2.50

2.

Sugar

500 gms. per capita per month

5.10

3.

Wheat

20 kg. per card per month in Plain areas and 30 kg per card per month in Hill areas.

2.20

4.

Palm oil

Scales fixed from time to time

13.65



subject to availability


5.

Kerosene

Madras city and headquarters of Municipalities (10 litres per family card)




Other municipalities (6 litres per family card)

Price ranging from Rs. 2.19 to Rs. 2.53



Township and town panchayats (5 litres per family card)




Rural areas (3 litres per family card)


SOURCE: Statistical Hand Book of Tamil Nadu, 1988

TABLE 9

ASSESSMENT OF INTERVENTIONS VIS A VIS NEED

TABLE 9 ASSESSMENT OF INTERVENTIONS VIS A VIS NEED

INTERVENTION

TARGET GROUP

COVERAGE

SERVICES/OBJECTIVES

QUALITY

PROGRAM IMPACT








DOES IT TARGET THE RIGHT GROUP?

WHAT % OF VULNERABLE GROUP DID IT COVER?*

DO THE SERVICES OFFERED MATCH THE NEEDS?

IS IT WELL IMPLEMENTED

HAS CONVINCING IMPACT? BEEN DEMONSTRATED?


AREA

BENEFICIARY

Al

AP

TN


Al

AP

TN

Al

AP

TN


TARG.

TARGETING
























ICDS

1

3

51%

34%

18%

3

2

2

3

1

1

1

TINP

3

5

-

-

53%

4

-

-

4

-

-

4

MDM

0

1

-

-

NA

1

-

-

3

-

-

4

VIT A
PROPH.

3

4

?

?

?

3

?

?

?

?

?

?

ANEMIA PROPH.

3

4

?

?

?

3

?

?

?

?

?

?

PDS

2.5

0 2 4 #

?

>100%

>100%

5

1

3

3

1

3

3

IRDP

1

2 3 3 #

7.7%

?

?

3

1

2

2

2

?

?

NREP/RLEGP

1?

2 3 3 #

NA

NA

NA

4

1

2

2

7

?

?

NOTES:

1. Al = ALL INDIA NATIONAL AVERAGE; AP = ANDHRA PRADESH; TN = TAMIL NADU.
2. * Percent children covered currently to those below the poverty line, assuming perfect targeting.
3. # Numbers refer to Al, AP & TN respectively.
4. $ 3.747 Million families covered (Planning Commission, 1988). Using family size estimate of 5, this is equal to 18.735 Million Pop = 7.7% of 242 Million Pop below poverty line.

TABLE 10

ASSESSMENT OF NUTRITION-RELEVANT ACTIONS IN THE CONTEXT OF THE TRIPLE A STRATEGY

INTERVENTION

WAS "NEED" IDENTIFIED?
(ASSESSMENT)

WAS PROGRAM TARGETED TO NEED?
(ANALYSIS OF RESOURCES AND NEEDS)

WAS COMMUNITY "EMPOWERED"?
(ACTION & COMM. INVOLVEMENT)

WAS INSTITUTIONAL SUPPORT BUILT IN TO PROGRAM DESIGN






ICDS

3 (Poverty criteria)

3

1

3

TINP

4 (Poverty & Nutritional crit)

4

2

5

MDM

0 (No)

0

0

1

PDS

3 (Poverty criteria)

3

0

3

IRDP

3 (Poverty criteria)

2

3

2

NREP/RLEGP

3 (Poverty criteria)

3

3

2

NOTES:

1. Rankings are made on a scale of 0 to 5 (Worst to best).
2. All rankings are subjective and are based on program operations in Andhra Pradesh and Tamil Nadu.

TABLE 11

COMMUNITY INVOLVEMENT IN DIRECT & INDIRECT NUTRITION INTERVENTIONS*

NEEDS ASSESSMENT/ ACTION CHOICE

ORGANIZATION

LEADERSHIP

TRAINING

RESOURCE MOBILIZATION

MANAGEMENT

ORIENTATION OF ACTIONS

MONITORING EVALUATION/INFORMATION EXCHANGE









INTERVENTION

















ICDS

2

2

2

3

1.5

2

3

1.5

TINP

2

3

2.5

5

1.5

2

5

3

MDM

1

1

1

1

1

1

1

1

PDS

2

1

1

1

1

1

3

2

IRDP

2.5?

2

2.5

?

1

2

3

2

NREP/RLEGP

2.5?

2

2.5

?

1

2

3

2

NOTES:

1. Ratings are based on the matrix for ranking Nutrition-Relevant Actions with respect to levels of community involvement suggested by Shrimpton (1989).

Why Nutrition Improved in Andhra Pradesh

Overall declines in poverty levels in Andhra Pradesh have been close to the average decline noted in the rest of the country as a whole, and may have contributed to some of the observed improvement in the nutrition profile. Direct expenditure on nutrition in Andhra Pradesh has been very small compared with other states, with only one direct nutrition intervention, the ICDS, of limited coverage.

Given the poverty ratios, Andhra Pradesh has received more than its fair share of IRDP, NREP and PDS inputs. A disaggregated analysis at the district level shows that the two districts that exhibited the maximal decline in underweight prevalences in the last fifteen years were Guntur and Chittoor (Figure 14). However, these two districts experienced only moderate declines in poverty (11.0 and 13.4% respectively as compared with 22.8% decline for the state overall) over the period 1977-87/88, so that changes in poverty can not readily explain the large nutritional improvement (Figure 15), unless the ultra-poor in these districts benefited disproportionately. ICDS coverage was very low in Guntur, and less than that seen in other districts in Chittoor. IRDP coverage was also low in both these districts (Figure 12). Instead, a more likely explanation for the improvement lies in the disproportionately high PDS inputs in the two districts. The percentage of preferential green PDS cards far exceeded underweight prevalence levels or poverty ratios (Figure 11) and hence probably provided an adequate household food security buffer.

These data should not be interpreted to advocate inequity in PDS allocations, but to show that when PDS supplies are far in excess of need, benefits do trickle through to the poor. However, when PDS supplies are limited, as was the case overall in Andhra Pradesh, a better management of these should entail tighter targeting to ensure preferential supply to the needy. These conclusions are supported by the observed negative correlation (Pearsons r = -0.42, Spearmans r = -0.50) between underweight prevalences and adequacy of energy intakes (observed for the eight districts for which data were available) implying a strong household food security problem in 1977. This correlation was reduced (to +0.15 and +0.15) in the 1988-90 data set, implying that food security may not have been a major limiting factor in 1988-90 after the PDS was in place in the eight districts surveyed by NNMB (which received relatively high PDS inputs).

Figure 14. DECLINE IN UNDERWEIGHT PREVALENCE - Preschool children in A.P.

SOURCE: NNMB (1991)

Figure 15. TRENDS IN POVERTY - By District in A.P.

Direct nutrition interventions such as the ICDS in Andhra Pradesh had a relatively low coverage, and are unlikely to have had a substantial impact. Their impact could, perhaps have been enhanced by better area and beneficiary targeting, and by improved implementation.

We therefore conclude that the PDS which aims at improving household food security, and possibly the NREP with high female participation (42% of the workers) and self-targeting to the poorest households, are both likely to have had major roles in facilitating the observed nutritional improvements seen in the eight NNMB-surveyed districts of Andhra Pradesh over the last fifteen years. Where food entitlements under PDS were targeted in excess of ‘nutritional need’ (as in the case of both Andhra Pradesh and Tamil Nadu states), the impact was much more pronounced. In the light of resource constraints though, better area and beneficiary targeting in the future is warranted. The lack of adequate direct-nutrition interventions focusing on improvement of care factors and access to health care may have however limited the nutritional improvement, especially in areas with a substantial reduction in poverty ratios. Perhaps, a better support structure for improving access and utilization of health facilities and direct nutritional interventions with a strong focus on improving care factors are needed to shift the nutrition profile of the less severely malnourished in Andhra Pradesh.

Why Nutrition Improved in Tamil Nadu

The overall changes in the nutritional profile in Tamil Nadu are very similar to those in Andhra Pradesh i.e. a reduction in the proportion of severely underweight children, with little change in prevalence of the less severe forms, except in TINP-I districts where an improvement is also observed in the less severe forms. This, despite Tamil Nadu’s unprecendentedly large investment in direct nutrition expenditures.

As discussed earlier, a major chunk of the nutrition expenditure in Tamil Nadu is directed to the totally untargeted Noon Meals school feeding scheme with questionable nutritional benefits. Similarly, benefits from the ICDS in Tamil Nadu are likely to have been small. In contrast, NNMB surveys have registered a greater improvement in the nutritional profile in districts where the TINP-I has been operational (see Table 6). Improvement has been observed in both a reduction in the prevalence of the severely underweight and an increase in the percentages of children in the ‘normal’ and less severe weight-for-age categories. The targeted and tailored inputs of the TINP-I, with a strong emphasis on the maternal and child care (and to a limited extent health) inputs, in complement with the strong and relatively well-targeted PDS inputs which provided the much-needed boost to household food security among the poorest, were all important factors. The fact that incidence of vitamin A deficiency in Tamil Nadu reduced much more dramatically than that in Andhra Pradesh, despite a near constant intake of vitamin A, also suggests that programmes such as the TINP-I (which included prophylaxis against vitamin A deficiency, and improved maternal care factors such as breastfeeding and use of colostrum etc) may have contributed to the improved nutritional profile. In addition, Tamil Nadu had strong NREP inputs with a relatively high participation of women (44%, as against 20% national-level), which is likely to have contributed to some of the observed improvement in both the TINP and non-TINP areas.


Previous Page Top of Page Next Page