THE IMPACT OF IMF STRUCTURAL ADJUSTMENT UNDER KAUNDA'S REGIME

IMPACT OF STRUCTURAL ADJUSTMENT PROGRAMMES ON HOUSEHOLD-LEVEL FOOD SECURITY AND CHILD NUTRITION: THE ZAMBIAN EXPERIENCE

Dr. Katele Kalumba

Senior Fellow,

Dept. of Community Medicine

University of Zambia, 1990

Paper Prepared for the IDRC-supported Workshop on Health and Economic Policy Harare, February, 1990

.INTRODUCTION

Like many other developing countries in Africa today, Zambia has been pursuing some form of structural adjustment programme (SAP)  recommended by  the International Monetary Fund as part of the effort to solve the external account crisis. While these programmes include policy reforms in various sectors, they focus in the initial phase, on outward-oriented trade strategies. But increasingly, concern with internal macroeconomic issues in the 1980s  such as high real interest rates, falling private investment, and rising inflation in countries like Zambia, has pushed fiscal policy debate and actions into reforming the form and pattern of public spending. At the centre of internal fiscal policy stabilization strategies, has been the question of public expenditure for social services and other subsidies directed at lowering the cost of living.

This paper examines the highly problematic question of the social effects of structural adjustment programmes.  Of particular focus is the effect of these programmes on household food security and nutrition. An assessment of their effects at a micro-level removes debate from consideration of statistical aggregates of impact at a macro socio-economic level per se to experiences of real people on the ground.

The paper first examines the methodological problems of linking SAPs to movements in household food security and nutrition at the micro-level. It suggests that part of the problem arises from the difficulty involved in measuring nutrition status specifically and food availability generally. Above all, there is the problem of disaggregating the long-term effects of economic depression, droughts, seasonality  and cultural practices in feeding and production on micro-level nutrition and food security ( Kumar,1988; Reardon et al, 1988; Kennedy and Cogill, 1988; Corbett,1988; Hay, 1988) from specific reform policies under SAPs designed to redress pre-existing economic states.

The paper concedes this methodological problem by suggesting that the impact of SAPs in Zambia during the 1980s on household food security and nutrition can be examined at least at a preliminary level, inferentially. This can be done by a deconstructive analysis which fructures the economic debate by bringing out its presuppositions and showing how these are undone at the level of empirical experience of individual households. At this level the various possibilities and impossibilities offered by the structure of economic reform cannot be fully captured by statistical technology. To ask of statistics to explain the justifiability of poverty even if 'short-term', for people who have known nothing-else but poverty, is to reveal our form of political practice. It must be argued that underlying the question of SAP economic reforms is a political question of 'entitlements' that each group within a public economic space seeks to assert for itself. Such an assertion is a politically subjective construct that takes into account many structural relations including the means of public discourse on SAPs themselves.

This paper critically reviews Zambia's SAP in its socio-political context including its silences on the impact it potentially has on the role of women in production in rural areas. It then presents evidence gathered by various sources documenting increasing trends in malnutrition at household level. It focuses particularly at one rural province-Luapula for illustration of the problem of malnutrition and food insecurity. It concludes that Zambia is currently increasing numbers of food insecure households due to fluctuations in income, inflation and high dependence ratio and urbanisation and decay in basic service infrastructure.

.MALNUTRITION AND ROLE OF SAPS: METHODOLOGICAL PROBLEMS

Zambia's structural adjustment programme in the 1980s  introduced stiff economic policy measures. These  included the decontrol of prices , devaluation of the national currency, removal of food subsidies, and cost recovery in social services including, health. While governments such as Zambia are determined to forge ahead with SAPs, there remains great uncertainty as to the  effects of proposed fiscal and other policy measures on prices and income, money supply and inflation, Zambia's balance of payments, and employment in the long run ( Sheshamani,1989; Olowrski, 1989). Above all, there is general consensus that the short-term effects can be devastating (Mulenga, 1989). In particular, these measures seem to have potentially far‑reaching consequences for the continued provision of basic social amenities upon which the basic needs of the majority of the population depend. These include, education, curative and preventive health services and above all, the nutrition status of children.

Of critical importance to the question of impact of SAPs on the survival of children is the differential levels of nutrition that declines in income might entail. The generation of nutritional problems in a community is a good indicator of the existing form and content of development practice. It also highlights the extent to which development policies downclass women's roles and child survival as priority issues.

It is still a matter of great contention in Zambia by advocates of SAPs whether, the observed decline in nutritional status of children can be attributed to adjustment measures or whether such a decline reflects a trend pre-existing prior to adjustment measures (PIC, 1990).

Research on the direct effects of structural adjustment programmes (SAPs) on nutritional status at the household level is problematic and scanty.  But it must be admitted as an empirical problem that part of the issue is how to disaggregate the effects of long-term trends prior to SAPs on the output indicators. Current proposals on addressing the methodological issues in Zambia have suggested one approach: This requires relating SAPs to nutritional decline through the relationship between their effects on short-term household level incomes, food stocks and food prices (input indicators), changes in food intake and health service provision (process indicators) and nutritional status (output indicators) (PIC, 1990).  These proposals suggest a degree of detail that requires an incomes monitoring system as proposed by Mulenga (1990) that is not at the moment, available. Further, these proposals underestimate the problem of assessing nutritional indicators themselves.

The methodological problems of ascertaining undernutrition in Sub-Saharan Africa are immense (Chaudri and Timmer  1986; Bairagi 1986; Chen and Scrimshaw 1983; Martorell 1985; Srinivasan 1981; Svedberg 1987; Ekstrom 1988). Svedberg's (1987), critical assessment of evidence on undernutrition in Sub-Saharan Africa, exemplifies the thrust of criticisms about methodology. We shall, with great debt, follow his argument at length here.

According to Svedberg there are two main ways employed in nutritional assessments of individuals. One is called the dietary approach. It involves i) estimating an individual's food consumption, ii)  evaluating the nutritional content of the food intake and, iii) estimating what nutritional requirements the individual has in order to comply with the stated objective function. The second method is to measure health and other conditions of the individual that reflect the outcome of an imbalance between her/his nutritional intake and expenditure. This is indicated by anthropometric, clinical, biomedical, mortality and other demographic evidence. Svedberg says these approaches involve many measurement difficulties including, normative evaluation problems. A related, but distinct question, is how rural household members value nutritional adequacy in relation to other "basic needs".

Measurement of food intake, the dietary approach  (eg. in Gatter /Gobezie studies, Gatter 1988) is complicated and costly in practical field situations in Zambia. If household members all eat from the same pot, as happens in most rural communities, there is little possibility of measuring the intra-family distribution. Between meal snacks and away-from -home meals introduce further complications.

Piwoz et al. (1986) and Esterik (1984) have shown that food sharing between households affect children's food intake in times when an individual family may be experiencing food shortage. As social support for food security, food sharing may operate in positively selective ways. Apart from its effect on the households' food accessibility, it also can favor children who walk about and receive tidbits of food at various households (Ekstrom 1988; Richards 1932).  Interviews or stock taking of individual household food reserves may miss this feature of food sharing where it is practiced and misrepresent nutrition status.

Further, estimates of food consumption at aggregate levels, derived from supply-side data (as part of estimating average per capita consumption), are constrained by shaky population and demographic data needed to derive per-capita estimates. Likewise, data on food trade and changes in food stocks, especially, are seldom reliable at the village/household level just as they are at national level.  What a subsistence peasant household grows and consumes can really only be measured through direct observation, which is seldom done (Svedberg, 1987).

The focus of the dietary approach on energy (calorie) content as the major nutritional element raises normative issues. There are several criteria according to which one can relate nutritional "requirements" such as activity, subjective well-being, health, height, weight status, capabilities of various types, etc. Moreover, there is no unambiquous way in which one can establish what level of, for example, physical activity that the "nutritional needs" should be related to, and different activities imply different food requirements.

Anthropometric and Related  Approaches:  Because of the limits of food intake studies, particularly in the assessment of the nutritional status of infants and small children, it is common practice to use anthropometric, clinical and biomedical methods. Also the mortality rate has commonly been used as an indicator of the nutritional standards in a population. Yet, it is important to understand that none of these methods provide full-proof estimates.

The most commonly used anthropometric measures are (i) weight for height (w/h), (ii) height for age (H/a) and (iii) weight for age (w/a). The w/h measure is mainly used as an indicator of temporary, acute undernutrition. The h/a measures permanent, chronic, dietary inadequacy in childhood (which affects also growth in height).  The w/a measure can be used to discriminate between temporary and permanent undernutrition. Other anthropometric measures  include skin-fold thickness and upper arm circumsference. The height and weight- for-age  (h\a) measure is the least affected by temporary food deficiency and, thus, the most reliable indicator of chronic dietary inadequacy in childhood and childhood only. One difficulty, however, with anthropometry in a place like Zambia is that age, needed for the commonly used height- and weight- for- age measures, is usually not recorded adequately. There is thus significant risk for data biases and, random measurement errors of age can produce notable biases in the results (Bariagi 1986).

Infant and child mortality rates, especially, are commonly used as indicators of the nutrition status of a population.  But this approach often ignores the point that mortality is also affected by factors other than inadequate nutrition. Such factors are often difficult to isolate (Ekstrom, 1988). This problem may be understood by conceptualizing the task of reducing mortality when diseases are interdependent (Burnam, 1978).

As suggested earlier, our approach in this paper is inferential in the sense that we leave the social story of SAPs to be intepreted politically from macroeconomic trends. We do this by juxtaposing economic movements against currently existing evidence showing  unabated  trends in malnutrition. The paper argues, along the above stated methodological premise that food security can be enhanced through policies that increase the level of income and reduce the risk posed by recurrent fluctuation in incomes. As proxy to examining the impact of SAPs on food security and nutrition, there is the need, therefore, to examine how these policies affect the price paid by households for food and other basic goods and services. Following Cogill and Zaza (1990), we can set the following analytical premises: Household food security and nutritional status is affected by food availability; ability or entitlement of households to obtain the food and the  desire to claim such entitlement or obtain food through market or home production; intra-household food distribution and food intake. The physiological utilization of food is influenced by policy variables of access to health services, clean water and sanitation.

III.TRENDS IN STRUCTURAL REFORM AND PERFORMANCE IN ZAMBIA'S ECONOMY

The Zambian economy, has been in a state of prolonged crisis since 1975, largely caused by multiple internal and external factors. Zambia's economic crisis is partially illustrated by trends in the per capita GNP, which showed a precipitous decline from just above US$700 in 1981 to US$250 in 1987, at constant prices. An IMF/World Bank supported structural adjustment programme was  initiated  between 1983‑1987 to address these problems. However, its failure to arrest the basic decline in economic growth, together with the social and political consequences of the adjustment measures, culminated in the Government's suspension of the IMF package. From 1988, an Interim Plan, which formed the first phase of the New Economic Recovery Programme (NERP) was designed and implemented by the Government independently of the IMF. In general, the impact of NERP was slow. The good performance of the small scale agricultural sector during 1987‑88 which contributed to a real GDP growth of 2.6 per cent was out‑stripped by overall failure to control money supply and intolerably high rate of inflation. Government deficit financing from the banking system and, inability to regulate the share of domestic consumption in GDP, in addition to increased credit to the non‑government domestic sector, were largely responsible for monetary expansion.

In view of this failure, the Government's new Fourth National Development Plan (1989‑1993) was introduced as the second phase of the NERP. This led to new rapproachment with the IMF and the World bank designed to support the major structural adjustment measures suspended earlier in 1987. The positive effects of these measures are still to be seen. Due to a high and still‑accelerating inflation rate , wage increases have not compensated for the serious erosion in household incomes, particularly for the low income groups.

More importantly, are the social effects of adjustment policies on prices and incomes. Between 1985‑1988, the annual average rate of inflation was around 40% to 55 % for the low income groups and 45‑60% for the high income groups. Consequent to decontrol during 1989 inflation rate increased to 121 % for the low income groups and 119.5 for the high income groups between January and October in 1989.  When put in comparative terms, the total inflation rate from October 1988 to October 1989 was 186.5% for the low and 162.5 for the high income groups (CSO, 1989; UNDP, 1990).

There is a consensus in  Zambia that poverty is now a major national problem, even if the extent is hard to determine. In urban areas, it is estimated that 42% of the urban population are living below the poverty line. The profile of these households suggests that the urban poor are mainly engaged in low income informal sector activities, although it includes lower‑paid formal sector employees. Female‑ headed households are found primarily in the lower‑income sub-sectors of the informal sector. In the rural areas, the most vulnerable rural households include the resource‑poor, in maize and particularly non‑maize growing areas, and female‑headed households (ARPT, 1988, Kalumba, 1990).

At the same time that household incomes have fallen, so has government capacity and commitment to support social policy investments in education, health, public transport and other sectors. In the health sector specifically, economic decline has taken its toll. Real health expenditures have declined by some 41% from 1970 to 1984. More recently, the health sector share of public expenditure has also been eroded. During the five years 1982‑1987, real per capita expenditure in the health sector declined by nearly 50%, and in 1989 stood at K110 (US$2.75).

Exacerbated by the economic crisis resulting in decreasing employment opportunities in the urban areas, the erosion of families' buying power through inflation and the inability of social services and infrastructures to meet the needs of these new urban dwellers, the actual situation in the towns for a significant number of Zambians is one of increasing despair. Conversely, the progressive exodus from the rural areas is reducing possibilities for self‑sustained development as it is contributing to disruptions in the family structure. It is estimated that at least 30% of rural households are female‑headed (CSO, 1980; Kalumba, 1990). Further, the cohesiveness of family networks in a  rapidly changing urban social situation  is in question. Most social policy experts in Zambia believe that traditional assumptions about the social support capacity of the African extended family in urban Zambian no longer seem to hold very neatly. A rise in cross‑ethnic marriages have meant that customary practices supporting family cohesion are not always adhered to when a spouse dies and child care responsibilities are not clearly defined.

IV. EVIDENCE OF NUTRITIONAL DEFICIENCIES AND FOOD INSECURITY AT HOUSEHOLD LEVEL: FOCUS ON LUAPULA PROVINCE

Past trends in the last three years  show that  just  over 40  per cent of child deaths were attributed to the effects of malnutrition. Concern about malnutrition in Zambia has taken on greater emphasis since the mid-80s.

A recent nutrition survey (Cogill and Zaza, 1990) has reported that from a national sample of 2000 children 19.7 per cent were moderatedly undernourished while 5% were severely undernourished. The study also found 10% of children between 12 and 23.9 months of age who were below minus two standard deviations from the median weight for height of the NCHS reference population, suggesting moderate and severe wasting. More dramatically, they found that 59.4 per cent of children were moderatedly or severely stunted. This figure is only surpassed by that reported by Carlson and Wardlaw (1990) for Malawi at 63% of children.

 According to a recent UNICEF report, Zambia's nutritional state of children is comparable to varying degrees with countries in the sub‑region such as Namibia, Zaire, Botswana, Zimbabwe (UNICEF, 1990).

The 1984 PHC evaluation exercise reported that 28% of children below 5 years of age were below the lower line of the growth chart. It also reported that 32% of these children were below in the 0-11 months of age category ( suggesting poor weaning practices). The Snap Survey of Childhood morbidity and mortality in Lusaka (UNICEF/MOH  1987) revealed that paediatric admissions and deaths constituted 21% and 75% of total hospital admissions and deaths at the UTH respectively.  Nearly 60 % of the 156 sick children admitted into UTH who were studied were from the low-income areas of the Lusaka. And, 36.5% of these were suffering from malnutrition (PEM) in the period covered by the study.

At the national level, the National nutrition Surveillance program shows that there are more children  of the age group 0-39 months  under- weight in the rural provinces than in the urban areas. In a 1986  national study (Kauppinen, 1987), 14-48 % of children 0-59 months were below 80%  weight/age of the WHO reference standard. And, a consistently higher percentage of children 12-23 months fell below 80 % of weight/age.  This national picture can be complimented by a look at local level studies in  Luapula, a rural province.

1. MALNUTRITION IN LUAPULA

Nutrition interacts closely with many factors particularly food intake and access to and use of health services in times of illness. Conventionally, the most frequently discussed determinant of food intake is the availability of food. In rural Zambia, food availability often fluctuates with the agricultural circle. Existing studies in Luapula report that pre- and post-harvest periods are seasons of potentially low and high food availability, respectively (ARPT, 1988).

The effect of food availabilIty on nutritional status  is, as noted earlier, difficult to assess because periods of food shortage also coincide with the peak periods of infectious diseases which also have negative effect on nutritional status.  According to available knowledge, food availabilty shows periods of inadequacy that are reflected in lowered food intake among children. Whether it is mediated through a low  meal frequency or not is not addressed in the literature. Evidence for a negative effect of food availability on nutritional status has not been found probably because of confounding factors such as infection. It must be stressed that if infection is adequately measured it can be controlled for in the statistical analysis of the data.

In the 1988 ARPT MABO Nutrition Study in Luapula representing the abundant season, 46% of the 35 children examined with respect to weight for age, were below the lower line of the growth chart. Using the mid-upper arm circumference to measure the effects of long term malnutrition, 38% in the nutrition sample study were below the 13.5cm on the arm-band measuring tape.

In a larger sample of 145 children attending an under-five clinic, 48% were below the lower line of the growth chart in a measure of weight for age. The study found that 46% of the nutrition sample and 51% of the under-five sample fell below the line of the growth chart  for weight/age  in the age range 12-26 months. The 1988 nutrition study thus concluded that "nearly half [the] children in Mabo suffer from undernutrition"   (ARPT, 1988: 5 ).

A similar study on Mabumba close to Mansa, Luapula's provincial capital, by Phillip Gatter (1988) reports that in a 24 hour recall interview, households reported that the problem of malnutrition was not a question of absolute lack of food but inadequate intake of some nutrients. The predominance of cassava porridge as a weaning food contributes considerably to energy and protein deficiency.

In another study of Mabo/Mabumba (Kalumba, 1990) anthropometric measures were taken from 188 children. Analysis of these measures reveal that 30% of children were below the 13.5 cm  WHO Standard in an upper arm circumference measure.  The weight for age measure for which there was more reliable data reveals that 73.5% of children were below the 80% Standard. These findings confirmed what is generally known from various studies. Omissions in the recording of height  rendered the more solid measure of long-term nutritional deficiencies height/age unusable.

2. FOOD SECURITY IN LUAPULA       

The effect of seasonally low food availability reflected in a reduced food intake for populations has been reported in many studies ( Abdullar & Wheeler 1985; Brown et al 1985; Rutishauser 1974 ). Not clear in most studies is the question of children's food consumption across seasons. Studies in this area appear conflicting with some showing seasonal variations and low energy intake (Abdullar & Wheeler 1985; Brown et al 1985) and others finding no seasonal variation but low energy intake (Bidinger et al 1986; Valverde et al 1986).

The major draw back in literature is that there is little effort to measure food availability. The reasons may be that food availability is a complex concept that includes many different components making an estimation of it difficult. Food availability for a household is not only dependent on their agricultural production but other factors such as bartering, food gifts and the household's socioeconomic level have to be taken into account.

When food availability is not measured, the underlying assumption seems to be that food is short at certain periods of the year, and that all households are affected equally. But differences in socioeconomic status of rural households (CSO, 1986) may suggest that this may not be so. Above all, the limits of measuring food availability may be side stepped or measured through a proxy indicator...food accessibility.  The difference between food availability and its accessibilty to the household is that the latter takes into account socioeconomic factors of individual households. An estimation of food accessibility for a household can be achieved  through measuring the types of food consumed. A food diversity score  would show the number of different foods as well as how they are consumed. A high food diversity score is generally correlated with a good accessibilty of food. Focusing on the child, it can be assumed that the food that enters the household comprises the food accessible to the child. This can be estimated by measurements of household food consumption (Ekstrom,1988).

The Kalumba (1990) study of Mabo/Mabumba confirmed that cassava is the main food stay for all households in Mabo/Mabumba. While what the household ate in the past seven days was varied, the major supplementary foods given to children was nshima and vegetables. In more than half of the households in Mabo, food production during last season was inadequate.  94 % of HH in Mabo and 99% of HH in Mabumba reported not having eaten any meat more than three times during the past 7 days, while only 1%  and 6.2 % of households in Mabumba and Mabo respectively had eggs. However, because of the presence of fishing activities in Mabo, 19% reported having eaten fish with nshima a day prior to the interview. The majority however, 58% in Mabumba and 47% in Mabo had only eaten Nshima with vegetables. There are no sources of milk products in the area .

Malnutrition in Mabo/Mabumba reflects malnutrition and the related problems of food production and intake in Luapula province generally. Children in Luapula are known to have one of the poorest nutritional status in Zambia (GRZ/UNICEF,1986).  In January 1984, the National Nutritional Surveillance Programme data show that of children under five years  in Luapula, 33 % were below the lower line on their clinical growth chart, which indicate approximately  80 % of weight-for-age according to WHO standards. Divided into age categories, the proportion of children under the line were: for 0-12 months, 28 percent; for 13-24 months, 35 percent, and for 25 -59 months, 42 percent ( Perez, 1984;  Ekstrom, 1988).  

3. PRODUCTION BURDENS FALL ON WOMEN IN LUAPULA

Official statistics and special studies  on Luapula (Gould & Chongola, 1986; Gobezie, 1985; Perez, 1984; Chilivumbo et al, 1988 ), and on the target Districts (Milimo & Mathai, 1984) and  project areas (ARPT, nd; Gobezie,1985) support the concept of relative deprivation  in  explaining the problems of development for women and the survival of children in Luapula. Expressed in gender terms, such deprivation entails official benign neglect of women in economic policy practice.

Gould and Chongola (1986: 24) report that despite the demographically imposed female predominance in economic, particularly agricultural activities in Luapula, 86% of their female respondents in a provincial study of six micro-regions, reported to have had no access  to any of the five major officially-provided agricultural facilities (including credit).

The vertical concentration of female labour in rural households reveals both the effects of official development practice and the culturally-driven  patriarchy in Zambia generally.  This is particularly revealed in a  study about Luapula women's roles by the Ministry of Agriculture Adaptive Research Planning and Training (ARPT) study on time allocation in Mabumba, one of  our target study site. The  study in Mabumba reported by Allen ( 1988) was an intensive year-long labour/time allocation study  focusing on the differences in the way the time of men, women and children is allocated in households of different types. The study used a carefully stratified sample of 13 households whose activities were recorded on a day to day basis through out the year. The study's findings indicated that in the majority of households, men were doing much less directly productive work than either adult women or female children. Only in households with higher levels of crop production do men begin to work for comparable periods to their women.

Allen  also found out that generally, even women were not working to full capacity. Women in female-headed houselds  tended to be better off than those in male-headed households. Only women in male-headed subsistence level households appeared to be reaching the limits of their physical capacity in some seasons. All others are engaged for considerable periods in social pursuits and unspecified spare time. Thus Allen reports that for both men and women  the time that nobody can account for- "resting time",  is the single most time consuming "activity" by a very wide margin (p12). Women are active for 56 % of the year, whilst men are "resting" for 56 % of the year.

Allen (1988)  ARPT study also revealed that women spend the bulk of their time on food preparation, household maintenance and sickness and contribute 77% of their time to total household productive work.  Allen also reports that women are actually ill 20% more than men and spend five times as much time as men looking after other sick people. They also spend more times on funerals (three times as much).

Allen's time studies have great import in unravelling the problem of women development and child survival. Male labour migration, increased divorce rate and widowing family patterns  increase the ranks of female-headed households. In turn, this often leads to increased workload for women as they have to take over male tasks in the food production chain (Eide & Steady, 1980). Further, introduction of cash crops may also increase women's workload (Mbulo in ZARD, 1985) and tend to reinforce women's subordinate position in society (Tobisson, 1980) making them have less access to various types of resources. Furthermore, seasonality in agricultural activities produces peaks of heavy workload for women (Chambers et al, 1981; Mascarenhas, 1983)

The time and energy consuming tasks in food production may compete with the woman's time available for child care, cooking and feeding. Studies such as Allen's  suggest the need to explore further whether mother's and alternate care takers' time allocation  has a negative effect on children's food intake.

The social reconstruction of women roles and entitlements pertinent to child survival would require a priori, a systematic empirical examination of many variables.  These include among health-specific factors, the interaction between terms of exchange, time allocation in "productive and non-productive"  sectors. Of  particular value here are food and food production systems  and social-cultural functions of child care and social relations. Such analyses cannot be undertaken within the limits of "rapid assessment" research methodologies. Such methodologies or anything resembling them must settle for less demanding objectives.

V. FOOD COUPONS SYSTEM AS GOVERNMENT NATIONAL RESPONSE TO FOOD INSECURITY

In order to respond to increased evidence that Zambia's deteriorating economy and adjustment efforts were affecting income adversely and therefore the capacity of households to purchase basic food staples, the UNIP government introduced a mealie meal coupon system. The primary objective of the coupon system is to provide a safety net for the poor people to get supplementary basic supplies of food during the time that general subsidies are reduced. 

Zambia's coupon system is urban-targeted and in the initial stage, was broadly defined to cover the whole urban population with no discrimination in terms of capacity to afford food at the market price. Later it was restricted to those earning below ZKwacha 20,500 per annum. Further, the total amount of maize meal that a coupon could buy was reduced significantly.

In a study of the operation of the coupon system in Zambia, Pearce (1989) has reached a number of important conclusions. These include: that the value of the coupons is not realistically indexed to an accurate measure of inflation reflecting the consumption basket of target group families. He has suggested that the poorest urban households may be worse off under the coupon system since the system stresses minimizing expenditure rather than improving food security.  Further, by limiting the number of participating retail outlets for subsidized food, constraints have been instituted which limit the capacity of poor households to claim their food entitlements.

According to Pearce, there are targeting problems as people living outside urban areas and those in the informal sector of urban areas have been left out hence limiting the value of the system to those most in need. Financially, the coupon system has met the government's objective of reducing government outlay on consumer subsidies by less than a third of budgetary costs. Without the coupon system, the estimated expenditures on food subsidy would have reached K1.2billion in 1989 as compared to K400m actual cost of the coupon scheme.

VI. CONCLUSION

Indeed there is a growing body of evidence which suggests that households living in conditions which put their entitlement to food security at the recurrent risk imposed by uncertain economic presuppositions, will act to strategically alter the balance of political forces. Zambia's case is no exception.  Food riots and political organization that has ensued since the SAPs measures were introduced suggest a limit in political and social tolerance  for suffering.                              

REFERENCES

Kumar, S.K. (1988) Effect of Seasonal food shortage on Agricultural Production in Zambia World Development 19(9) 1051-1062

Reardon et al, (1988) Coping with Household-level food insecurity in Drought-affected Areas of Burkina Faso World Development 16 (9) 1065-1074

Kennedy E. and B. and Cogill (1988) The Commercialization of Agriculture and Household -Level Food Security: The Case of Southwestern Kenya World Development 16(9) 1075-1080

Corbett, J. (1988) Famine and Household Coping Strategies World Development op cit 1099-1112

Hay, R.W. (1988) Famine Incomes and Employment: Has Botswana Anything to Teach Africa? World Development  op cit 1113-1125

Sheshamani, V. (1989) Towards Structural Transformation with a Human focus: A Study of Zambia's Economic Programmes and Policies in the Eighties. Unpublished paper, Lusaka.

Chaudri, R. and C.P.Timmer (1986) The Impact of changing affluence on Diet and demand patterns of agricultural commodities. World Bank Working Papers,  No. 785

Bairagi, R. (1986) Effects of Bias and Random error in anthropometry and in age estimation of malnutrition Americam Journal of Epidemiology, 123

Chen L.C. and N.S. Scrimshaw eds. (1983) Diarrhoea and malnutrition- Interactions, Mechanisms and Interventions (New York and London: Plenum Press).

Martorell, R. (1985) Genetics, Environment, and Growth: Issues in the Assessment of Nutritional Status, in A. Velazquez and H. Bourges, eds., Genetic Factors in Nutrition (London: Academic Press).

Srinivasan, T.N. (1981) Malnutrition: Measurement and Policy Issues,  Journal of Development Economics, 8.

Svedberg, Peter (1987) Undernutrition in Sub-Saharan Africa: A critical assessment of the evidence, Wider Working Papers, WP15.

Piwoz EG. & F. Viteri (1986)   Studying Health and nutrition behaviour by examining household decision-making, intra-household resource distribution and the role of women in the processes. Food and Nutr Bull 7 (4): 1-31.

Esterik, van P. (1984)  Intra-family food distribution: Its relevance for maternal and child nutrition. Cornell Nutrition Surveillance Programme, Working Papers N.31, Ithaca

Richards, A. ( 1932) Land Labor and Diet in Northern Rhodesia., London

Ekstrom, E.(1988) Time is Food: Women's Time allocatioon and Household food availability as determinants of African children's status. Mimeo IHCAR Karolinska Insitute, Stockholm.

Fourth National Development Plan (1989‑1993)  Zambia, Govt Printers, Lusaka

CSO, 1989 Consumer Price Index

UNDP, (1990) Sectoral Review of the Zambian Economy Vol IV Lusaka.

Kalumba, K (1990) Child Survival and Women Development in Mabo/Mabumba: The need for social mobilization and capacity building at the District level, Research Report to UNICEF/NCDP, Lusaka.

Cogill, B. and M. Zaza (1990) Report of the Nutrition Module as Part of the Crop forecasting survey-Rural Zambia, 1990, CSO/NFNC/MOH/UNICEF, Lusaka Unpublished mimeo.

Carlson, BA and T. Wardlaw (1990) A global regional and country assessment of child malnutrition. UNICEF Staff working Papers No. 7. UNICEF, New York

UNICEF, (1990). State of the World's Children 1990. UNICEF. Oxford U Press. UK.

Kauppinen, M. (1987) NNSP: A Summary Report 1984-1987. Ministry of Health , Lusaka. MOH/SIDA mimeo

Brown et al (1985) Effects of seasonal illness on the dietary intake of weanlings during lomgitudinal studies in Rural Bangladesh, Am. Journ. of Nutrition. 41: 343-355.

Rutishauser, IHE (1974) Factors affecting the intake of energy and protein by Ugandan preschool children,  Ecology of Food and Nutrition 3:213-222

Abdullar, M. & E. Wheeler (1985) Seasonal variations , and the intra-household distribution of food in a Bangladesh village. American J. of Clinical Nutrition. 41: 1305-1313.

Bidinger et al (1986) Nutrition and HEalth consequences of seasonal fluctuations in household food availability, Food and Nutrition Bulletin 8: 36-60.

Valverde et al (1986)  Seasonality in health and nutrition on a Guatemala coffee Planation, Food and Nutrition Bulletin 8: 19-25.

CSO, (1986) Country Profile Zambia, 1985, Lusaka.

GRZ/UNICEF,(1986).   Situation Analysis of Women and Children in ZAmbia. Lusaka.

Perez,L. (1984) An assessment of the nutrition situation in Zambia for Planning Nutrition interventions, Part 1: ZAmbia, Part 2: Luapula Province. Germany Agency for Technical Cooperation.

Gould & Chongola, (1986) The Situation of Women in Luapula, MAWD, Luapula.

Gobezie, A. (1985) ARPT Nutrition Programme: Final Report, ARPT, Lusaka.

Chilivumbo et al, (1988) Small Business Study, Luapula. Zambia. mimeo.

Milimo J. T. & R. Mathai (1984)  Zambia Project Proposal for Integrated Rural Women Education and Child Survival and Development in Mansa and Samfya Districts, NCDP, Lusaka

Pearce, R.(1989) Food Consumption  Subsidies in Zambia, Paper prepared for Unicef, Lusaka, 1989.

ARPT, nd; Mabo Nutrition Survey representing the abundant season, Research Report, Ministry of Agriculture, Mansa, Zambia.

Allen (1988)  Dependent Males, a Research Report, MInistry of Agriculture, Mansa, Zambia.

Eide, W.B. & F.C.  Steady (1980) Individual and social energy flows... In Nutritional Anthropology. Contemporary Approaches to Diet and Culture, eds Jerome, R.F. et al. pp61-84.

Mbulo, MP (1985)  The effects of cash crop production on men's and women's and children's participation in agricultural production: A case study of Magoye resettlements in Mazabuka District, UNZA, Lusaka.

Tobisson, E. ( 1980) Women, Work, Food and Nutrition in Nyamwigura village, Mara region Tanzania,  Tanzania Food and Nutrition Commisiion Report N0. 548, Dar es Salaam.

Chambers et al, (1981) Seasonal Dimensions to Rural Poverty, London.

Mascarenhas, O. (1983) Implications of Constraints in Women's Control and Utilization of Resources for the Food and Nutritional Status of their Families. Unpublished paper presented to the TFNC/UNICEF Workshop on "HUnger and Society", Tanzania, Dec. 1983.