This is the old United Nations University website. Visit the new site at http://unu.edu

Food policy

Mitigating the effects of drought through food security and nutrition monitoring: Lessons from Malawi

Mitigating the effects of drought through food security and nutrition monitoring: Lessons from Malawi

Abstract
Introduction
Historical context of drought management in Southern Africa
The FSNM system
Use of FSNM information in emergency planning: The case of the 1991-92 drought
Lessons from the use of FSNM for drought management in Malawi
Concluding remarks
Acknowledgements
References
Appendix 1. Chronology of events in the management of the 1991-92 drought in Malawi

Suresh Chandra Babu and Evance Chapasuka

Suresh Chandra Babu is with the International Food Policy Research Institute in Washington, DC. Evance Chapasuka is with the Ministry of Agriculture and Livestock Development in Lilongwe, Malawi.

Abstract

This article uses a case study from Malawi to demonstrate the use of a Food Security and Nutrition Monitoring (FSNM) system for managing and mitigating the effects of drought. The implementation of FSNM during the drought emergency is presented, along with a description of the process of information generation and the use of results in emergency interventions. Presenting the lessons learned from the Malawi experience for the benefit of future drought-management and disaster-prevention efforts in other sub-Saharan African countries, the article concludes that a decentralized system of data collection, processing, and analysis is more likely to be successful in planning interventions for food security and nutritional improvements during periods of drought.

Introduction

The food security and nutritional status of households in the Southern Africa region continue to be threatened by recurring droughts. The drought of 1991-92 tested the region’s ability to face, manage, and mitigate the effects of food shortage on the livelihood of the households. In spite of the fairly well-established early warning systems for food shortage at the national and regional levels, the response systems were not well developed and were not fully operative to address the food insecurity and malnutrition problems at the household level [1]. This is partly because the early warning systems provided very little information on the causes and conditions of household food security, health, sanitation, and availability of drinking water during periods of drought. Countries with established systems of monitoring food availability and nutritional status at the household level managed drought emergencies better than other countries in the region.

For successful planning and management of drought and its effects on food availability, periodic information on the factors that affect the food security and nutritional status of the population is essential. This requires a system of information generation, compilation, processing, and analysis at different levels of emergency intervention and planning. Food Security and Nutrition Monitoring (FSNM) is one way to provide periodic essential and adequate information on impending food and nutrition crises for planning emergency interventions. Based on the principles of nutritional surveillance [2] and the experiences gained from implementing it in several developing countries, FSNM has also been widely used for programme management and evaluation and for policy planning and implementation [3]. In general, recent developments in FSNM methodology reflect the importance of nutritional surveillance systems to adopt the technical requirements of surveillance to the decision-making environment that prevails in a given country [4, 5].

Very few countries have designed and implemented successful FSNM systems in sub-Saharan Africa to mitigate the impact of drought on household welfare. Drought continues to be a major constraint on development in a large number of countries in Eastern and Southern Africa. A major concern for the governments and donors in these countries is the establishment of early warning systems to predict and prevent drought-induced famine at the household level. In response to the food crisis of the 1970s and as a follow-up to the 1974 World Food Conference, a Global Information and Early Warning System (GIEWS) was established by the Food and Agriculture Organization (FAO), which has assisted sub-Saharan African countries to set up National Early Warning Systems (NEWS) [6]. A recent review of country experiences of FSNM systems in Africa shows that timely warning of national food shortages is by far the major purpose of the NEWS [7]. For example, in the Southern Africa region, all 10 Southern African Development Community (SADC) countries have NEWS, which are primarily used for the prevention and management of drought. The Regional Early Warning System operated by SADC Food Security United, based in Harare, Zimbabwe, in collaboration with the NEWS, reviews and interprets nationally and provincially aggregated data relating to crop production, trade, and storage in order to assess food availability in a growing season. Although they are helpful in organizing food imports during periods of food emergencies, the NEWS fall short of providing information on household food insecurity and on the vulnerable population groups.

Malawi was the first country in the SADC region to establish a household FSNM system [8]. Recently, based on the lessons learned from the Malawi experience [5], a Regional Food Security and Nutrition Information System has been established by SADC with technical assistance from FAO to provide selected member states (Mozambique, Swaziland, and Zimbabwe) with information on households suffering from food stress and probable causes of food insecurity [9]. Countries in the Southern Africa region were the most affected by the drought of 1991-92. Among these countries, Botswana had a well-established and relatively successful drought relief management programme. The programme had been reasonably effective in safeguarding the welfare of vulnerable population groups during the long droughts of 1982-87 and 1991-92. The district-level capacity to monitor food emergencies through a timely and fairly comprehensive FSNM system has been cited as one of the contributing factors for Botswana’s success in mitigating the effects of drought [10]. Countries such as Zimbabwe, Zambia, and Mozambique, on the other hand, had challenges in constructing a profile of the food-insecure regions. Although these countries have various data-generating mechanisms in place for managing food emergencies, if these systems of monitoring are judged by the extent to which the data are used, most of them have not been fully effective, partly because they have focused on food supply as the major source of information [11].

This article, using a case study from Malawi, demonstrates the use of an FSNM system for managing and mitigating the effects of drought and derives lessons from the Malawi experience for similar efforts elsewhere. It begins with a brief historical background to drought in Malawi. The implementation of FSNM during the 1991-92 drought emergency is described next. The next section presents the process of information generation and the use of results in emergency interventions. Lessons learned from the Malawi experience are presented in the last section for the benefit of future drought-management and disaster-prevention efforts in other sub-Saharan African countries.

Historical context of drought management in Southern Africa

Several countries in Eastern and Southern Africa have recently faced chronic drought-induced food insecurity, as witnessed by the Sahelian drought in the early 1970s and the African food crisis of the 1980s. The lessons learned from the Sahelian drought, however, were overshadowed by more pressing problems of debt and balance of payments, which left the governments in the region unprepared for the African food crisis of the 1980s [12]. Although massive free food distribution through commercial imports and food aid was helpful in mitigating the effects of drought, the early warning systems that had been in place since the early 1970s were of only limited help in warning of food needs [13]. Also missed were the opportunities to identify the vulnerable regions of the affected countries and the vulnerable households affected by the food crisis.

The famine of 1949-50 in Malawi was a major disaster that had striking similarities to the famine of the early 1980s in other parts of Africa [14]. By the mid-1980s, about 27 sub-Saharan African countries had been identified by the FAO and the World Food Programme (WFP) as critically affected by droughts and as facing severe food shortages [12]. All other countries in Southern Africa except Malawi and Swaziland were listed as “at risk” countries. In fact, in 1984-85, when the media were filled with stories of mass starvation in Ethiopia, Mozambique, and various parts of West Africa, Malawi stood out as a country that was able not only to feed itself but also to export substantial quantities of maize to its neighbours who were short of food [15]. For most of the 1960s, 1970s, and 1980s, Malawi was one of about seven African nations that was relatively consistent in meeting its goals of self-sufficiency in food production [15]. Earlier famines during this century in Malawi include the famine of 1922 and the famine of 1903, which drove people into Malawi from Mozambique. It is said that nothing comparable to the 1949 famine had happened in Malawi until the 1991-92 drought [16].

Since 1985, however, Malawi has witnessed a number of disasters that have brought home the need for emergency planning. These include the influx of Mozambican refugees, which began in 1986, the mealy bug attack on cassava in 1987, the drought of 1990, and the flooding in Phalombe in 1991. Despite its lack of experience of major disasters and therefore of disaster management, the government managed to handle these situations success fully [17, 18]. The 1991-92 drought in Malawi and other countries of Southern Africa was the worst the region had witnessed for 40 years [19]. Malawi was the most seriously affected country, with an estimated affected population of 5.7 million, compared with 100,000 in Botswana [4,20]. The production of maize, the staple food of 90% of the population, fell from 1,638,000 tonnes to 662,000 tonnes - an unprecedented drop of 60%. To meet this production deficit, about 582,000 tonnes of maize was pledged by donors immediately after the government’s declaration of emergency. The country subsequently received about 356,000 tonnes of maize as food aid. The government also imported 53,000 tonnes of maize during the latter part of 1992, and an additional 100,000 tonnes of yellow maize was bought to meet the production shortfall.

There was, however, no mass famine, starvation, or death. Relief camps were not set up, there were no mass migrations from the affected areas, and there were no catastrophic disease outbreaks such as occurred in recent famines in Ethiopia and Sudan [21]. Malawi demonstrated the capacity to respond to drought by mobilizing resources, importing food (both as food aid and on a commercial basis), and organizing a large-scale distribution effort. This successful response was in part due to a well-established infrastructure for collecting, analysing, and disseminating information, and a district-level administrative system that could respond quickly to information and logistical needs. (A chronology of events in the management of the 1991-92 drought in Malawi is provided at the end of this article in the form of an appendix with an emphasis on the use of information from FSNM for decision-making).

Strong coping mechanisms at the household level, the carry-over effect of grain reserves from earlier years, solid inter-household redistribution and exchange mechanisms, and the effectiveness of food aid are among the reasons that have been suggested for Malawi’s success in preventing famine [21, 22]. A factor that is frequently overlooked, however, is the ability of Malawi’s institutions to react to early indicators of drought and to use existing information to create meaningful policies and programmes. We argue here that the well-established institutional structure for information generation and use was one of the leading factors that contributed to the drought management and subsequent disaster prevention in Malawi, and we illustrate the argument with a case study of the FSNM system.

The FSNM system

Periodic droughts and deficits in food self-sufficiency continue to be recurrent phenomena in most sub-Saharan African countries. Depending on the nature and extent of the drought, planning efforts range from emergency food aid to short-, medium-, and long-term preventive interventions. The lack of adequate information about the food security and nutritional status of the population is the major constraint on such planning and policy-making activities [23]. Periodic monitoring of household food security and the nutritional status of individuals is essential to generate adequate information about household food security and the factors that influence it [24]. It is also important to monitor and evaluate the impact of policy interventions and of the structural adjustment and stabilization policies that are being implemented by various sub-Saharan African countries [25]. The Malawi approach to FSNM has focused primarily on developing the capacity for converting information into planning and policy interventions. It is also flexible enough to meet information needs during periods of emergency planning.

The FSNM surveys are conducted twice a year to capture the effects of different seasons on the food security and nutritional status of households: in June-July, immediately after the harvest, and in November-December, at the beginning of planting. Information is collected under four different modules: household food security, household income and expenditure, markets and prices, and nutrition monitoring. The food security and markets and price modules are used in each of the rounds. This generates information on food production, storage, and purchases; food availability; pattern and types of food intake; availability of food markets; food prices; farm and non-farm employment; wage rates and remittances; and the coping strategies of households. The household expenditure and nutrition-monitoring modules and the food security modules are used alternately. The expenditure module generates broad information on the quantity of food and non-food commodities consumed, their value in terms of cash or other commodities, and the quantity of food consumed from home production. Under the nutrition-monitoring module, information is collected on the nutritional status of preschool children, their morbidity and mortality, and the availability of health, sanitation, and drinking water. The sample households of the FSNM system are a subsample of the Annual Survey of Agriculture, in which crop production and farm management information is collected.

Before each survey, a three-day training session is conducted for the enumerators and field supervisors to help them understand the need for collecting data on each of the variables and to train them in the use of an enumerator manual. A data-entry programme is written to compile the data in dBASE, and a manual on guidelines for data analysis in the SPSS programme is also prepared and distributed to the Agricultural Development Divisions (ADDs). A week long training course in statistical data analysis is conducted after every survey for the evaluation staff of the ADDs, which enables them to undertake advanced analysis of the data collected. Using the results of the analysis, they prepare a Food Security Working Paper, which is discussed within each ADD to arrive at recommendations for interventions and to identify persons responsible for action. These proposals are then presented to the senior officers of the Ministry of Agriculture to prepare a plan of action [18], which is sent to the Economic Planning and Development Department of the Office of the President and the Cabinet. A national report is then prepared for the use of senior officials in the Ministry of Agriculture, as well as other ministries and donor agencies, as a quick reference guide to indicators and variables affecting food security [26].

Efforts have been made to ensure that the food security and nutrition information is regularly used in emergency planning and project implementation. To this end, a Food and Nutrition Information Committee (FNIC) has been established in the Ministry of Agriculture to review the existing information, identify additional needs, channel information to other ministries and donor agencies, and recommend interventions to the Ministry of Agriculture. The FNIC is made up of members of the Food and Nutrition, Evaluation, Marketing and Pricing, Early Warning, and Data Processing Units in the Ministry of Agriculture, and it is represented on the National Committee on Drought Management, which also includes staff from the Economic Planning, Transport, Health, Local Government, Community Services, and Statistics Departments and from academic institutions.

Data on the food security and nutrition situation collected from different parts of the country are thus analysed and reported in three stages. In the first stage, the ADDs analyse the data for intervention planning and policy-making. The results, in the form of working papers, are discussed with the ADD management and subject specialists. In the second stage, the reports from the ADDs are complied at the Ministry of Agriculture into a national report on the food security and nutrition situation, primarily in order to present decision makers with data in an easily understandable form. The national report also includes a list of all recommendations for interventions generated by the ADDs. The report is circulated to the members of the FNIC for comments and further suggestions, and the results are sent to different users, including the Drought Relief and Rehabilitation Unit in the Office of the President and the Disaster Preparedness Committee (DPC), for immediate action. In the third stage, the data are used to prepare medium- and long-term plans and to train government officials in conducting food security and nutrition policy research. A series of policy research papers is prepared, based on the data collected by the FSNM surveys, along with other available data sets in the Ministry of Agriculture [27]. Although not designed specifically for drought management, the infrastructure, institutions, and human capacity developed for FSNM were fully deployed for drought monitoring and management during the 1991-92 drought. Additional details of the FSNM system in Malawi, including costs and the number of staff involved, are given elsewhere [8].

Use of FSNM information in emergency planning: The case of the 1991-92 drought

The success of any monitoring system depends not so much on the effectiveness of information generation but on the efficiency with which the information is used for decision-making on interventions. Malawi’s FSNM system stands out in enabling the better use of information from the monitoring systems in emergency planning. This section describes the process of drought management in Malawi as it relates to the actual use of information from the FSNM system, along with comparisons of such information with that in a normal year.

The approach followed in FSNM in Malawi is multidisciplinary, in which nutritionists and social scientists work together with agricultural subject matter specialists at both the regional and headquarters levels. In strengthening the national capacity to generate FSNM information and in designing emergency interventions, the Malawi approach has also emphasized decentralization of the activities so that each ADD can undertake the FSNM surveys without much help from the headquarters [28]. Such a process of decentralization has been useful in generating food and nutrition interventions from the lowest levels that are specific to their area, such as Extension Planning Areas (EPAs). This provides an opportunity for the officials involved, in implementing interventions at the lowest levels, to participate in the design of emergency intervention projects and to be fully involved in the implementation of these projects with enhanced motivation. The discussions of the FSNM survey results at the ADD and Rural Development Project (RDP) levels and the policy recommendations made from them have demonstrated that such a bottom-up approach in designing and implementing interventions is possible in Malawi.

The rest of this section illustrates, in chronological order, the use of information from the FSNM system in the design and implementation of interventions during the 1991-92 drought. In response to a special alert prepared by FAO’s GIEWS on “Southern Africa’s Drought and Its Threat to the Region’s Food Supply” in February 1992, the Malawi Ministry of Agriculture requested the NEWS and the FSNM system to assess food production and food availability for the 1992 harvest season. Information collected through the FSNM survey conducted just before the planting season (December 1991) provided an indication of the areas chronically affected by food shortages at the household level. For example, as shown in table 1, even during normal years, the majority of rural households in Malawi ran out of food stocks before the next planting season. This gave an indication of the vulnerable regions during the periods of drought. Southern ADDs such as Blantyre and Ngabu were more exposed to food insecurity during drought than their northern counterparts. Combining this information with production estimates for the 1991-92 crop season provided a better picture of the impending food shortages. Table 1 also shows the extent of food insecurity during drought, as indicated by the number of households running out of food during July 1992.

TABLE 1. Cumulative percentage of households running out of food stocks in normal (1991) and drought (1992) years in Malawi, by Agricultural Development Division (ADD)

ADD	August		September		October		November		December
	1991	1992	1991	1992	1991	1992	1991	1992	1991	1992
Karonga	22.6	28.1	29.6	31.4	37.1	43.2	46.6	54.8	67.2	73.8
Mzuzu	32.5	34.6	40.3	44.3	50.4	58.5	57.9	62.5	71.1	75.0
Kasungu	23.1	28.4	28.1	32.5	42.5	45.8	53.4	58.3	63.8	68.5
Salima	12.5	18.9	20.8	25.8	31.9	33.7	37.9	40.2	39.5	48.3
Lilongwe	23.0	26.5	31.0	35.0	44.0	48.0	56.0	59.0	68.0	75.0
Liwonde	35.8	70.8	39.5	75.6	47.3	85.3	63.3	91.2	75.8	98.0
Blantyre	36.7	76.4	41.4	83.4	51.2	88.6	66.3	90.3	73.5	96.8
Ngabu	40.5	80.2	52.4	85.6	58.5	90.1	77.3	95.6	80.1	100

The production estimates were generated by the ADD offices (table 2) and compiled at the national level. Collected and updated throughout the cropping season, this provided an early indication of the food requirements at the ADD level. The crop estimates for the ADDs and RDPs showed that southern ADDs, such as Blantyre, Liwonde, and Ngabu, were the most affected, with food deficits ranging from 63% to 100%. Salima district in the central region was also affected. However, district-level information based on estimates of production forecasts did not serve the purpose of targeting the distribution of food. Thus, FSNM surveys conducted immediately after harvesting generated information on the food security situation in specific EPAs within each district and the level of food stock held by the households after the harvest of the crop during the drought season.

For example, during the 1991-92 drought, Liwonde, Blantyre, and Ngabu were the most severely affected ADDs. The FSNM survey conducted in July 1992 indicated that all the EPAs in southern Malawi would have no food left at the household level by December 1992 (table 3). Similar statistics for the same districts during a normal year, shown in table 3, show the severity of shortfall in food production due to drought. This information was provided to the DPC, which enabled focusing of the disaster relief to the identified EPAs. In order to target the households within the EPAs on the basis of the information generated from the FSNM surveys, the ADDs generated information on the number of farm families affected, the number of affected children under five years of age, and the food required per month to avoid starvation. Table 4 provides such information for one ADD. This was further used to refine the food requirement estimates and to reallocate the food aid distribution among the EPAs by providing additional information to the DPC (see appendix 1).

The FSNM surveys also collected information on the nutritional status of children under five years of age. This information was essential to monitor the nutritional welfare of the population during the drought period. It was also used to target the health facilities and to improve the availability of protected water to the rural households. Table 5 presents information submitted to the DPC on stunting, wasting, and underweight at the ADD level. Whereas stunting and underweight did not show a marked difference between normal and drought years, there was a remarkable increase in the percentage of wasted children due to the incidence of drought. As can be seen, wasting, an indication of acute malnutrition due to a sudden drop in food availability, was very high in the southern ADDs (Blantyre and Ngabu). Following this information from Blantyre, Ngabu, and Lilongwe districts where the wasting was high, the DPC ordered intensified food distribution in the affected EPAs. It also recommended that the nutritional status of children in the affected areas be monitored, along with related information such as incidence of illness and availability of protected drinking water. On the basis of this recommendation, brief reports were prepared from the FSNM data, with graphic representation of food insecurity and malnutrition, and submitted periodically to the DPC.

TABLE 2. Food production estimates and deficits during 1991-92 drought in Malawi, by Agricultural
Development Division (ADD) and Rural Development Protect (RDP)

ADD/RDP		No. farm families	Total population	Food required (tonnes)	Production 1991-92 (tonnes)	Deficit (tonnes)	% surplus or deficit
Lilongwe
	Ntcheu	58,771	264,470	47,605	21,189	-26,415	-55
	Dedza Hills	44,629	200,831	36,149	37,601	1,452	4
	Thiwi Lifidzi	45,096	202,932	36,528	50,478	13,950	38
	L. East	84,643	380,894	68,561	80,796	12,235	18
	L. West	11,401	514,805	92,665	105,331	12,866	14
Blantyre
	Mwanza	26,308	118,386	21,309	2,560	-18,749	-88
	Bt/Shire	217,410	978,345	176,102	22,198	-153,904	-87
	Mulanju	94,437	424,967	76,494	8,413	-68,081	-89
	Phalombe	49,301	221,855	39,934	14,639	-25,294	-63
Liwonde
	Mangochi	75,746	340,857	61,354	894	-60,460	-99
	Banwera	60,201	270,905	48,763	15,380	-33,383	-68
	Balaka	57,081	256,865	46,236	185	-46,051	-100
	Kawinga	66,069	297,311	53,516	7,436	-46,080	-86
	Zomba	123,512	555,804	100,045	16,791	-83,074	-83
Mzuzu
	Rumphi	33,647	168,235	30,282	18,714	-11,568	-3
	C. Mzimba	37,802	189,010	34,022	37,365	3,343	10
	S. Mzimba	25,368	126,840	22,831	23,850	1,018	4
	Nkhata Bay	30,715	153,575	27,644	2,260	-25,284	-92
Kasungu
	Kasungu	62,723	313,615	56,451	64,051	7,600	13
	Mchinju	69,641	348,205	62,677	32,096	-30,581	-49
	Dowa West	45,431	227,155	40,808	47,343	6,455	16
	Dowa East	33,274	166,370	29,947	21,137	-8,810	-29
	Ntchisi	24,425	122,125	21,983	46,161	24,179	110
Ngabu
	Nsanje	60,290	301,450	54,261	206	-54,055	-100
	Chikwawa	39,635	198,175	35,672	902	-34,769	-97
Karonga
	Karonga	22,404	112,020	20,164	5,657	-14,507	-72
	Chitipa	13,771	68,855	12,394	13,096	702	6
Salima
	Bwanje	41,301	198,245	35,684	5,657	-30,027	-84
	Salima	42,701	209,235	37,662	6,271	-31,391	-83
	Nkhotakota	31,706	15,830	28,535	5,195	-23,340	-82
Malawi		1,732,439	8,090,862	1,456,355	714,232	742,123	-51

Interventions to reduce food insecurity during the periods of drought will have limited nutritional benefits unless the process of converting food intake into improved nutrition is also considered. This process requires access to protected drinking water, sanitation, and primary health care. The FSNM surveys obtained information on these variables during the drought period. For example, in Blantyre ADD only 34% of the households interviewed used water from protected sources. However, among the households that had access to safe drinking water, 70% of them travelled less than 2 km to fetch water. After the presentation of this information to the DPC in July 1992, it was recommended that the Water Department work with UNICEF to install 3,000 bore wells and hand pumps in the affected areas. This was accomplished in the next two-month period. Information gathered on the household sanitation measures indicated that about 65% of the households interviewed in Blantyre ADD owned and used a pit latrine. UNICEF was given the responsibility of installing 20,000 pit latrines over a period of eight months. Investigation of the availability of primary health care during drought in the same ADD showed that about 45% of the households had to travel more than 5 km to their closest primary health-care facility. About 35% of the households reported that a primary health-care committee existed in their area, and about 16% of the households were participating in the activities of the committee.

TABLE 3. Availability of food stocks in the affected areas at the household level in 1991 (normal) and 1992 (drought) years in Malawi, by Agricultural Development Division (ADD) and Extension Planning Area (EPA)

ADD	No. of EPAs	Percentage of EPAs in which food stocks ran out
		May		June		August		December
		1991	1992	1991	1992	1991	1992	1991	1992
Salima	14	7	28	7	28	7	28	7	28
Lilongwe	26	4	11	8	26	8	26	9	35
Liwonde	26	11	57	15	84	20	88	23	100
Blantyre	33	9	60	15	85	18	96	30	100
Ngabu	11	9	100	18	100	27	100	36	100

TABLE 4. Food requirements at the decentralized level in Ngabu Agricultural Development Division (ADD) during the 1991-92 drought in Malawi, according to Rural Development Project (RDP) and Extension Planning Area (EPA)

RDP/EPA		No. of farm families affected	No. of children under 5 yr affected	% of children under 5 affected	Monthly food requirement (tonnes)
Chikwawa RDP
	Kalambo	9,588	8,198	100	621
	Mitole	12,185	10,418	100	783
	Mikalango	12,291	10,509	100	801
	Livunzu	13,528	11,566	100	882
	Mikalango	12,291	10,509	100	801
	Dolo	6,918	5,915	100	450
	Mbewe	8,145	6,925	100	531
Nsanje RDP
	Magoti	7,135	6,100	100	459
	Mankianga	7,600	6,575	100	504
	Mpatzoo	4,663	3,988	100	306
	Nsanje	6,906	5,905	100	450
	Nyachilenda	4,344	3,714	100	288

In addition to the food security and nutrition indicators, information collected on the coping strategies of the households during drought indicated that households resorted to casual farm labour and sale of liquid assets such as livestock to reduce the severity of drought-induced food insecurity. Households also changed their food preferences to cope with reduced food availability. A majority of the households indicated that they shifted from eating ufa (fine maize flour with 40% extraction) to eating mgaiwa (coarse maize flour with 96% extraction). Households also adjusted their eating patterns by reducing the number of meals eaten in a day. For example, in Lilongwe ADD between June and December 1992, the percentage of households eating only one meal per day increased from 22% to 45%, while the percentage of households eating three meals per day dropped from 27% to 16%. Monitoring the changes in the food intake patterns, particularly the deviation of the number of meals eaten from the normal pattern during the period of drought, was found to be extremely useful in planning distribution of emergency food aid.

TABLE 5. Percentage of child malnutrition in 1991 (normal) and 1992 (drought) years in Malawi, by Agricultural Development Division (ADD)

ADD	Stunting (ZHA < -2)		Wasting (ZWH < -2)		Underweight (ZWA < -2)
	June 1991	July 1992	June 1991	July 1992	June 1991	July 1992
Karonga	42.17	43.77	2.1	5.86	20.10	23.49
Mzuzu	48.21	49.78	1.7	5.38	23.12	28.06
Kasungu	56.17	58.64	3.2	6.49	26.07	32.68
Salima	41.24	44.75	1.8	3.79	32.17	30.77
Lilongwe	45.16	49.21	4.1	8.45	30.41	32.74
Liwonde	40.06	43.24	2.0	4.79	26.52	27.56
Blantyre	48.15	46.23	3.1	12.1	33.48	36.89
Ngabu	40.15	48.07	2.8	13.60	33.11	34.46
National average	44.53	48.44	2.6	7.54	28.12	31.28

Abbreviations: ZHA, Height-for-age Z score; ZWH, weight-for-height Z score; ZWA, weight-for-age Z score.

The analysis of the data from the FSNM system was useful beyond emergency planning for drought. For example, the data were analysed more vigorously to make medium and long-term food-security interventions. Table 6 shows the incidence of food poverty based on the expenditure level to meet the daily nutritional requirement of 2,200 kcal. There was a marked increase in the level of food poverty due to drought in several ADDs. Monitoring the level of food poverty on a regular basis and understanding its causal factors is fundamental to the design of food and nutrition policies and programmes [29]. This enabled the Malawian response to drought to go beyond the distribution of free food. At the sectoral level, information on variables such as resource availability, technology adoption, food production, and patterns of expenditure on various food and non-food items has been used to design long-term intervention policies and programmes. These programmes concern agricultural research, extension, and input supply and are funded by bilateral donor agencies, such as the International Fund for Agricultural Development and the World Bank [30].

TABLE 6. Percentage of households with food poverty based on insufficient food expenditure to meet the daily nutritional requirement of 2,200 kcal in November 1991 (normal) and May 1992 (drought) in Malawi, according to Agricultural Development Division (ADD)

ADD	FSNM round 4 Nov 1991	FSNM round 5 May 1992
Karonga	46.15	52.17
Mzuzu	38.53	54.49
Kasungu	51.22	58.36
Salima	67.81	57.72
Lilongwe	53.63	56.84
Liwonde	66.50	56.08
Blantyre	54.90	66.02
Ngabu	62.91	63.22
National average	54.85	58.24

On the basis of the intervention policies generated from the survey results, appropriate projects have been identified by the ADDs for implementation with an objective to improve food security and nutrition. In general, these projects are implemented in small areas as pilot schemes to learn from their actual impact on the rural committees. These projects are evaluated for their impact on food security and nutrition status at the household level with the FSNM system. Lessons learned from the evaluation of these projects are helpful in replicating the specific projects in other ADDs with any modifications that may be required. A detailed account of how the FSNM system is used to design long-term development policies that go beyond free food distribution is given elsewhere [8].

Lessons from the use of FSNM for drought management in Malawi

Although FSNM systems have been implemented for the past 15 years in several sub-Saharan African countries, only a handful of these countries could provide concrete examples of the use of information for effective interventions to alleviate food insecurity and malnutrition. Even among those countries which have established an operational system, several issues relating to the principles of FSNM need to be addressed. Some of the emerging lessons from the implementation of the FSNM system in Malawi for drought management are presented in this section.

Information obtained through FSNM could be used for emergency planning, programme evaluation, and policy-making objectives in a wide range of sectors, including agriculture, health, economic planning, social welfare, labour, and famine relief. To meet these diverse information needs and the various objectives of food and nutrition interventions, the FSNM system should be flexible. This flexibility in Malawi’s FSNM system enabled a quick transformation of an information system originally designed for policy planning into a system supporting the information needs for drought management. Such flexibility does not exist in the currently implemented monitoring systems in Botswana [10], Zambia [11], or Tanzania [3].

One of the constraints on developing a comprehensive FSNM system in sub-Saharan Africa has been the limited infrastructure to collect and process the data from the field. In countries with a well-developed network of data-collection systems, it is generally suggested that the existing infrastructure should be effectively used. However, in countries where there are no organized data-collection systems for food-security monitoring, such as Eritrea and Namibia, it may be necessary to create new infrastructure. Elsewhere, it is possible to combine the existing but poorly organized structures and reorient them towards specific goals of FSNM. In Malawi, as in Botswana and Kenya, the use of data-collection institutions that were in place for the collection of information in farm management and agricultural surveys made possible a cost-effective FSNM system [28].

One of the major criticisms of currently operating monitoring systems is that too many data are collected, few data are analysed, and little is reported of what has been analysed. The time gaps between collection of data and analysis, and between analysis and reporting, have been so wide in the past that when the information has been given to the decision makers, it has been too late to make effective use of it. To minimize this time lag, it is essential to decentralize the data-processing and analysis systems. To meet this task, attention should be paid to developing appropriate but flexible computer data-processing systems at these levels. The FSNM system established in Malawi is capable of generating situation analysis reports within four to six weeks from the time of training the enumerators for data collection. Botswana also provides an excellent example of a drought-monitoring system that reduces the time involved in reporting after collection of information from the field [10]. This is possible because of the trained human capacity and resources for data processing, analysis, and preparation of food security working papers at the ADD level [28].

At all levels of decision-making, namely, the national, regional, and local levels, the data analysis should closely match the needs of the decision makers, and various alternative intervention scenarios should be presented to them. Lack of attention to this important issue has rendered several well-intended FSNM systems data-driven rather than user-driven. Experience from Namibia and Zambia clearly indicates that reducing the number of variables on which the data are collected will increase the use of information in decision-making [11]. One way to prevent the efforts spent on data collection and analysis from becoming futile is to involve the decision makers in the generation of information. This is best done by developing the monitoring system in continuous collaboration with the decision makers through the initial stages and frequently modifying the content of the information generated to improve the relevance of the information for their decision-making needs. During the 1991-92 drought in Malawi, the information needs for drought-management planning were constantly evaluated by the designers of the monitoring system by working closely with the DPC. This increased user participation in the monitoring system and reduced the probability of its becoming a process-driven system.

Once the data collected from the field have been converted into information that leads to potential interventions, implementing these interventions is usually left to the decision makers. The political, cultural, and bureaucratic environments in which these decision makers operate tend to influence the speed with which the decisions are converted into action. In addition, the availability of timely information for making appropriate decisions and the quality of information presented to the decision makers play an important role in strengthening the information-action linkages. The FSNM reports generated by ADD-level officers in Malawi not only contained information on the nature and extent of food insecurity and malnutrition but also included potential solutions and possible interventions to mitigate the effects of drought [31]. Similarly, in Tanzania the decentralized, community-based nutritional surveillance programme in Iringa proved very effective for programme management and evaluation of nutritional interventions by increasing the use of information in a regular and timely fashion [3].

To increase the likelihood of the use of information in decision-making and the timely implementation of intervention policies, several workshops and other advocacy activities were organized during the period of drought to sensitize the decision makers towards understanding food security and nutrition issues and the importance of information-based decisions that reduce food insecurity and malnutrition.

Although the FSNM systems could be implemented by various sectoral ministries to meet their planning and policy-making objectives, it is important to have a focal office at the national level to provide an overview of various systems operating in a country. This helps in maintaining and providing an inventory of activities and in avoiding duplication of efforts among the sectoral ministries. More importantly, the FSNM focal office should play the role of a clearing house for information related to the food security and nutritional situation of the population in a country. Unfortunately, in several countries, including Kenya, Zambia, Zimbabwe, and Swaziland, the efforts to organize information generation and use have been hamstrung by bureaucratic barriers against coming together through multisectoral nutritional committees [7]. During the periods of drought management in Malawi, the DPC, coordinated by the Office of the President and Cabinet, acted as the decision-making group and was responsible for dissemination of information to donors and other international organizations. This substantially reduced the duplication of information-gathering efforts by several independent and non-governmental organizations (NGOs).

For successful implementation of the FSNM systems, having a critical mass of people trained in food and nutrition from various sectors, including the university and other academic institutions, is a prerequisite. Lack of such capacity to translate the data collected by monitoring systems into policy decisions and to design interventions poses a formidable challenge in all sub-Saharan African countries. Despite the efforts of several donors and institutions in developing capacity for the past 10 years, only a little more than 200 participants have been trained in food and nutritional surveillance and policy analysis [8]. Associated with this is the need for strengthening policy analysis capacity at the decentralized levels. This would require reorientation of the approaches to capacity-building in food and nutrition policy analysis to meet the specific requirements for such an approach. The human capacity that was put in place in Malawi for information generation and analysis was developed over a period of three years through constant monitoring and training of nutritionists and social scientists at various levels. This proved very helpful in designing drought monitoring and management efforts.

Concluding remarks

This article presents Malawi’s approach to FSNM for emergency planning. By using the case study of the 1991-92 drought in Southern Africa, it demonstrates that decentralizing the activities of monitoring could be an effective method of converting data into useful information that could be used to design intervention programmes at the local levels. Although it is difficult to predict the outcome of drought in the absence of an information system, comparison of Malawi’s experience with that of other countries in the region indicates that food insecurity and child malnutrition would have worsened because of delays in getting food to vulnerable people in the drought-affected parts of the country. Despite the overall success in preventing a drought-related famine through food aid and food trade during 1991-92, the region still lacks a long-run strategy to address the effects of food shortages induced by recurring droughts. Mitigating the impacts of recurring droughts on household food security and child nutrition, and linking these relief efforts to long-term development, provides an appropriate context to discuss some of the critical issues that must be addressed in this region to pursue appropriate action for increasing food security and nutritional wellbeing at the regional, national, and household levels. In meeting this objective, the need to strengthen institutional capacity to implement and sustain policies and actions for improving household food security and child nutrition, along with improving the data-generation and analytical capacity of institutions charged with food security and nutrition programmes, can hardly be overemphasized.

Acknowledgements

This article was prepared with financial support from UNICEF-Malawi through the UNICEF/International Food Policy Research Institute (IFPRI) Joint Food Security Programme, Bunda College of Agriculture, Malawi. The authors would like to thank Per Pinstrup-Andersen, Ken Williams, Victoria Quinn, Ruth Ayoade, Jin Qian, and Natalie Hahn for their guidance and involvement in the Malawi approach to FSNM and for suggestions at various stages of the preparation of this article. Thanks are also due to two anonymous reviewers whose comments and suggestions helped to sharpen the focus of the article. The authors alone are responsible for any remaining deficiencies.

References

1. Thompson C. Drought management in Southern Africa. Report prepared for UNICEF, Harare, 1993.

2. Mason JB, Haaga JG, Tabatabai H. Valverde V. Nutritional surveillance. Geneva: World Health Organization, 1984.

3. Pelletier D, Jonsson U. The use of information in the Iringa nutrition program: some global lessons for nutritional surveillance. Food Policy 1994; 19: 301-13.

4. United Nations Children’s Fund. Emergency in Malawi. UNICEF-Malawi, 1992.

5. Babu SC. Improved policies through food security and nutrition monitoring. Food Policy 1992; 17: 384-6.

6. Maxwell S. Frankenberger T. Household food security - concepts, indicators, measurements: a technical review. New York and Rome: UNICEF/IFAD, 1992.

7. Quinn VJ, Kennedy E. Food security and nutrition monitoring systems in Africa. Food Policy 1994; 19: 234-54.

8. Babu SC, Mthindi GB. Household food security and nutrition monitoring - the Malawi approach to development planning and policy interventions. Food Policy 1994; 19: 272-84.

9. Southern Africa Development Community. Development of a regional food security and nutrition information system. Harare, Zimbabwe: SADC, 1995.

10. Belbase K, Morgan R. Food security and nutrition monitoring for drought relief management. Food Policy 1994; 19: 285-300.

11. Eele D. Indicators for food security and nutrition monitoring. Food Policy 1994; 19: 314-28.

12. Curtis D, Hubbard M, Shepherd A. Preventing famine - policies and prospects. London: Routledge, 1988.

13. Borton J. Clay E. The African food crisis of 1982-1986. Disasters 1986; 10(4): 258-576.

14. Vaughan M. The story of an African famine. Cambridge, UK: Cambridge University Press, 1987.

15. Liebenow JG. Food self-sufficiency in Malawi: Are successes transferable? In: Glantz MH, ed. Drought and hunger in Africa. London: Cambridge University Press, 1987: 367-90.

16. Green RH. Sound the tocsin: the third horseman mounts to ride: drought in Southern and South Africa 1991-1993 Discussion Paper. Sussex, UK: Institute for Development Studies, 1993.

17. Lampell S. Disaster preparedness and relief in Malawi. Lilongwe, Malawi: Office of the President and Cabinet, 1988.

18. Ministry of Agriculture. Food Security and Nutrition Monitoring Report No. 2. Lilongwe, Malawi: Ministry of Agriculture, 1992.

19. United States Agency for International Development. Southern Africa Drought Assessment-Malawi. Washington DC: Office of US Foreign Disaster Assistance, 1992: 77-89.

20. Ministry of Agriculture. Nutrition Monitoring for the Drought, February 1993. Lilongwe, Malawi: Food and Nutrition Unit, Ministry of Agriculture, 1993.

21. Ejigu M. Towards improved response to drought in Malawi and Mozambique. New York: UNICEF, 1993.

22. Save the Children Fund. Report on the Baseline Nutrition Assessment Conducted in Chikwawa District, Lilongwe. Lilongwe, Malawi: Save the Children Fund, 1992.

23. Pinstrup-Andersen P. Government policies, food security and nutrition interventions. Pew/Cornell Lecture Series on Food and Nutrition Policy. Ithaca, NY, USA: Cornell University Food and Nutrition Policy Program, 1989.

24. Babu SC, Quinn V. Food security and nutrition monitoring in Africa - introduction and historical background. Food Policy 1994; 19: 211-7.

25. Babu SC. Agricultural policies and food security in Eastern and Southern Africa. Food Policy 1994; 19: 7983.

26. Ministry of Agriculture Food Security and Nutrition Monitoring. Report No. 1. Lilongwe, Malawi: Ministry of Agriculture, 1991.

27. Jere P. Babu SC. An analysis of indicators of poverty in Malawi. Agric Econ Analysis Rural Dev 1994; 4(2): 716.

28. Babu SC, Mthindi GB. Developing decentralized capacity for disaster prevention: lessons from food security and nutrition monitoring in Malawi. Disasters 1995; 19(2): 127-39.

29. Pinstrup-Andersen P, Pandya-Lorch R, Babu SC. Prospects for a 2020 vision for food, agriculture and the environment in Southern Africa. Working Paper. Washington, DC: International Food Policy Research Institute, 1996.

30. Simler K. Coping strategies of smallholder households and the scope for targeted interventions to improve food security. Malawi Agricultural Sector Memorandum Working Paper. Washington, DC: World Bank, 1993.

31. Chapasuka E. Monitoring of food security and nutrition in Malawi. In: Mataya C, Babu S, eds. Manual on food and nutrition policy analysis. Lilongwe, Malawi: Agricultural Policy Analysis Training Programme, Bunda College of Agriculture, 1995: 132-60.

Appendix 1. Chronology of events in the management of the 1991-92 drought in Malawi

Date	Event	Policy change or implementation
21 Feb 92	Special Confidential Alert from GIEWS of FAO on “Southern Africa’s Drought and Its Threat to Region’s Food Supply.”	Ministry of Agriculture requested NEWS and FSNM system to assess food production and food availability for 1992 harvest season.
26 Feb 92 a	Report on rainfall pattern and expected crop production during 1991-92 cropping season prepared for discussion by the Ministry of Agriculture with the help of NEWS and FSNM systems.	Report concluded that due to occurrence of dry spell from the middle of Dec 1991 to the middle of Jan 1992, more than 50% of the crops in the Lower Shire area in the south were wilting due to severe moisture stress, and in Blantyre and Liwonde districts about 20% loss was estimated.
6 Mar 92 a	First emergency meeting on food production estimates organized by the Ministry of Agriculture with use of FSNM reports from Nov 1991 household survey.	Concluded an earlier food loss estimate of 0.5 million tonnes of food was an overestimate. Realistic estimates of production and food aid needs based on food stocks at the household levels needed for presentation at the forthcoming donors’ meeting on drought.
16 Mar 92	First emergency meeting of the executive members of the DPC, consisting of government, NGO, and donor representatives.	DPC officially organized. Recommended that free food distribution should start immediately in Nsanje and Chikwawa districts in the south with 20,000 tonnes of food from USAID. Food Assessment and Monitoring (FAM) subcommittee formed with Minister of Agriculture as chair. Minister of Health and Minister of Works given responsibility of chairing Nutrition and Water Supply subcommittees, respectively.
10 Apr 92 a	Emergency assessment of effects of drought (United Nations report released) called for major food distribution effort to be jointly organized by donors under WFP coordination.	Estimated 354,000 people affected in Nsanje district alone. Free distribution of food started in the southern part of the country.
25 Apr 92 a	Nutrition subcommittee released its first report on the health and nutrition situation in the areas affected by drought.	Ministry of Health and Ministry of Agriculture jointly used the FSNM system that was already in place.
8 May 92 a	Food assessment and monitoring reports from the affected areas released by the DPC.	District commissioners given the mandate for distribution of free food based on the estimates prepared by the Ministry of Agriculture through FSNM reports.
30 Jun 92 a	Government of Malawi released confidential report on food security for donors.	Final crop estimates showed 59% production loss. Commercial and food aid imports begun, with 16,700 tonnes arriving in the country.
20 July 92 a	Nutrition Monitoring During Drought initiated under FSNM system by the Ministry of Agriculture.	About 49,000 tonnes of donor-pledged food imported so far. Strategic Grain Reserve continued to be depleted. Internal food aid distribution already under way in several districts for some months. Also seed and fertilizer distribution for the next cropping season initiated by NGOs in severely affected areas.
31 Aug 92 a	Food Security Update released by the Government of Malawi.	Estimated 58,000 tonnes of maize delivered to rural distribution centres since May 92, most of which was distributed to the affected households.
31 Oct 92 a	Food Security Update released by the Government of Malawi.	Government imported an additional 28,502 tonnes of maize for the month of October, while the target for imports was 35,000 tonnes per month. Government also borrowed 44,433 tonnes of maize imported as food aid for refugees for free distribution. Food aid pledged from donors for the marketing year 1992-93 continued to be 400,000 tonnes.
15 Dec 92 a	DPC meeting to review food distribution and the aftermath of the drought.	Reports from various sources indicated that the food distribution programmes were implemented well, although logistical problems in some districts slowed the delivery of food. Good rains reported in all parts of the country. Earlier estimates of crop area indicated a good crop for 1992-93.
28 Feb 93 a	Ministry of Agriculture released Nutrition Monitoring During Drought report.	Drop of 46.9% in the rate of acute malnutrition in Nsanje, the worst-affected district, compared with the previous month (Jan) indicated that food distribution had a positive impact in reducing acute malnutrition. Similar reduction in malnutrition figures reported from other districts.
11 May 93 a	Ministry of Agriculture released crop estimates for 1992-93.	Record harvest of maize reported, with an estimated production of 2.0 million tonnes for the 1992-93 crop year. Given the food requirement of 1.68 million tonnes, this left a surplus of 320,000 tonnes.
20 May 93 a	DPC met to review the effects of drought.	Government announced the need to phase out free food distribution by the end of May. Reconciliation exercise of food distribution begun. Government also announced plan to fill the strategic grain reserve by procuring 180,000 tonnes of maize. This would leave 140,000 tonnes as surplus from the current harvest.
2 Jun 93	National Workshop on Information Systems for Disaster Preparedness-	Assessment of drought information systems of government, NGOs, and donors was made, and a plan of action was prepared for information sharing for early warning of drought and for monitoring and evaluation of disaster-relief programmes.
13-17 Dec 93	National Drought Management Workshop to evaluate 1992-93 drought-relief programmes used reports from drought-monitoring surveys of FSNM.	Government officially announced that drought and its impacts had shown signs of disappearance. A long-term disaster preparedness plan was organized by a Disaster Assistance and Rehabilitation Unit established under the Office of the President and Cabinet in the event of future droughts.

a. Use of FSNM information in decision-making.