Research Paper Research

Ministries of Agriculture: Structures, Capacity and Coordination at District Level in Malawi

Blessings Chinsinga February 2008

Research Paper 013 | October 2009 www.future-agricultures.org Figure 1. Map of Malawi Source: World Relief Institute

Research Paper 013 | October 2009 ii www.future-agricultures.org Table of Contents Table of Contents...... iii, iv 1. Introduction...... 1 1.1 Background to the Study...... 1 1.2 Methodology...... 1 1.3 Organisation of the Report...... 2 2. Organisational Setup of the MoA at National and District Level...... 3 2.1 Introduction...... 3 2.2 MoA at the National Level...... 3 2.3 MoA at District Level ...... 5 2.4 Decentralization and Ongoing Changes at District Level...... 5 2.4 Conclusion...... 7 3. Overview of District Agricultures...... 8 3.1 Introduction...... 8 3.2 Key Agricultural Activities...... 8 3.3 Key challenges and opportunities in the district agricultures...... 8 3.5 Conclusion...... 8 4. MoA Functions and Performance...... 8 4.1 Introduction...... 8 4.2 Perceptions of the Functions MoA...... 8 4.2.1. Farmers’ Perceptions of MoA Functions...... 8 4.2.2. Other Agricultural Service Providers’ Perception of MoA Functions...... 11 4.3 MoA’s Failure to Perform Key Functions...... 13 4.4 Farmers’ Perception of MoA Performance...... 13 4.5 Conclusion...... 13 5. Structures and Decision Making Processes for MoA at District Level...... 15 5.1 Introduction...... 15 5.2 Reporting Structures of MoA...... 16 5.3 Decision-making processes...... 18 5.4 Opportunities and challenges arising from decentralisation policy...... 20 5.5 Conclusion...... 21 6. MoA Operative Capacity of MoA...... 22 6.1 Introduction...... 22 6.2 Technical Capacity...... 22 6.2.1. Staffi ng Situation...... 22 6.2.2. Causes and Remedies to the Staffi ng Crisis...... 23 6.2.3. Career Prospects and Stability of Tenure...... 25 6.2.4. Strategies for Improving Technical Capacity...... 25 6.3 Financial Capacity...... 26 6.3.1. Budgeting Process...... 26 6.3.2. Budget Implementation...... 27 6.3.3. Other Sources of Funds...... 29 6.3.4. Strategies for Improving MoA’s Financial Capacity...... 30 6.4 Conclusion...... 31 7. Interface between MoA and other stakeholders...... 31 7.1 Introduction...... 31 7.2 Forms and levels of interface...... 31 7.3 Mechanisms for coordination and collaboration...... 32 7.4 Status of interface between MoA and relevant stakeholders...... 33

Research Paper 013 | October 2009 iii www.future-agricultures.org 7.4.1. Intra MoA interface...... 33 7.4.2. Interface between MoA and other public sector agencies...... 33 7.4.3. Interface between MoA with other agriculture sector stakeholders...... 34 7.5 Strategies for Improving Interface between MoA and Stakeholders...... 35 7.6 Conclusion...... 36 8. Concluding Reflections...... 36

Appendix I: Checklists for Field Work...... 39 References...... 44

List of Tables Table 1: Stakeholder Perceptions of Opportunities and Challenges at District Level...... 9 Table 2: Phases of Agricultural Development and Performance of MoA in Dedza...... 14 Table 3: Ranking of the Phases of Agricultural Development and MoA Performance in Dedza...... 16 Table 4: Phases of Agricultural Development and Performance of MoA in Thyolo...... 17 Table 5: Ranking of the Phases of Agricultural Development and MoA Performance in Thyolo.....18 Table 6: Established Posts for Graduate SMS Filled in Thyolo District...... 22 Table 7: Incentive Differentials between the NGO Sector and MoA...... 23

List of Figures Figure 1: Map of Malawi...... ii Figure 2: The Structure of Ministry of Agriculture at the National Level...... 4 Figure 3: The Structure of MoA at the District Level...... 6 Figure 4: Expenditures by Programmes at the National Level between 1990 & 2007...... 27 Figure 5: Budget Estimates for ADDs for the 2006/2007 and 2007/2008 Fiscal Years...... 28 Figure 6: Budget Estimates for RDPs for the 2006/2007 and 2007/2008 Fiscal Years...... 29 Figure 7: Total Budget Allocations to MoA HQs, ADDs and Districts...... 30

Research Paper 013 | October 2009 iv www.future-agricultures.org 1. Introduction on their capacity to play a coordination role and 1.1. Background to the Study be demand driven. The justifi cation for studying This study was carried out under the auspices the MoA at the district level was that it is the of the Future Agricultures Consortium (FAC) best place to get insights into how the ministry politics and policy processes sub-theme. interacts and cooperates with other stake- Building on the earlier work of the sub-theme holders at fi eld level which is as close as possible on the debates about the Ministries of Agriculture to the point of service delivery. This should in (MoAs) in developing countries, the study was turn generate good insights to feedback up to intended as an entry point for grasping the func- central policy makers and serve as the basis for tions, structures, rules, financial and human informing the subsequent phases of FAC’s capacities of MoAs in Africa. endeavours. Specifi cally, the study intended to This study was inspired by the apparent provide insights to the following four funda- return of agriculture onto the international mental questions: agenda as a primary engine of growth and • To what extent do MoA perceptions on its poverty reduction in predominantly agrarian key functions match other stakeholders’ societies (cf. Timer, 2004; DFID, 2005; OPM, 2007). (particularly farmers) perceptions on what But this is taking place against the backdrop of the MoA is doing as well as their expecta- three distinct policy narratives about the role tions on what it should be doing? of MoAs. These are: 1) MoAs having the capacity • Is there a gap in the functions, activities, and policy clout to address the major constraints services being performed and delivered by of agriculture in which case they ought to be MoA? If so, is this gap being fi lled by other strong and well funded; 2) MoAs taking on a players in the sector (NGOs? Private minimal role focused on oversight and regula- sector?) tion whereas the private sector (non-state • What impacts do internal constraints have actors) assume a dominant role in the agricul- on the ability of the MoA to perform its tural sector in a free market environment; and current functions? 3) MoAs in partnership with other state agencies • How has the performance of the MoA should play a key role in addressing the coordi- changed over time and why? nation and intermediation of getting markets to work eff ectively while ensuring at the same 1.2. Methodology time concerted the public eff orts targeted to The study was conducted in two districts, poverty reduction (cf. Cabral and Scoones, namely: Dedza and Thyolo. Malawi is divided 2006). into three administrative regions, namely: north, The major focus of the study was therefore centre and south. These are in turn divided into exploring how the MoA interact with other key 28 districts. There are 6 districts in the north; 9 players at the district level which, inter alia, in the centre; and 13 districts in the south. Dedza include other government agencies, farmers, district is in the centre whereas Thyolo district the private sector and NGOs. The underlying is in the south. The decision to choose these two idea was to generate evidence on patterns and districts was based on the following three trends in the scope and leverage of the MoA in considerations. devising and delivery of agricultural services and These districts are currently politically promi- policies with the view to drawing implications nent. The president comes from Thyolo whereas

Research Paper 013 | October 2009 1 www.future-agricultures.org the leader of opposition comes from Dedza. The the agricultural sector as well as the changing latter has been one of the leading advocates for role of the MoA regarding its functions (see the fertiliser subsidy programme and used his section 3 Appendix I). A total of 4 FGDs were political clout in parliament to get the govern- conducted, two in each district. The same check- ment’s subsidy initiative extended to benefit list that was used for MoA personnel at the tobacco production in addition to maize. district level was also used for staff at the local Thyolo faces serious land constraints among level with appropriate modifications. The list of smallholder farmers compared to Dedza. Thyolo stakeholders consulted is provided in Appendix is one of the densely populated districts and II. The major limitation of the study was in terms much of the cultivable arable land has been of coverage as it was only carried in two districts taken up by tea and coffee estates. Dedza is one in addition to the very limited coverage of the of the leading maize and tobacco growing perceptions of private sector actors and donors. districts with relatively manageable land The main reason for this is that few private sector constraints. The average landholding size in the district ranges between 0.7-1ha compared to actors operate at district level whereas donors the national average estimated at 0.5ha. do not have representation at district level. These districts fall into different agro-ecolog- ical zones as classified by the MoA. Malawi is 1.3. Organisation of the Report divided into four agro-ecological zones based This report is divided into eight sections struc- on altitude. These are Lower Shire Valley (<200m), tured as follows: Section 1 introduces the scope Low altitude (200-760m), Middle altitude of the study and Section 2 provides an overview (760-1300m) and High altitude (>1300m). Dedza of the structural and organisational set up of and Thyolo fall into Middle and High altitude the MoA both at national and district levels for respectively1. purposes of setting the context for the rest of The data collection for the study was done the report. Section 3 examines the status of for a period of three weeks: one and half weeks agriculture in the districts of Dedza and Thyolo in each district. Field work was divided into two districts highlighting key agricultural activities, main phases, namely data collection at the constraints and opportunities. Section 4 anal- district and at the community level. Data collec- yses the perceptions of the roles and functions tion at the district level involved the use of semi- of MoA by farmers and other stakeholders structured interviews with relevant stakeholders, comparing them to the impressions of MoA including MoA officials, other relevant govern- officials and what is actually prevailing on the ment agencies, NGOs and private sector agents ground. In Section 5, decision making processes even though these were not readily available of MoA district level offices are examined (very few private sector organizations operate at the district level). Separate but related check- including how work plans are formulated, modi- lists were prepared for the MoA and the other fied and implemented. Section 6 evaluates the stakeholders respectively (see sections 1& 2 operative capacity of MoA offices at the district Appendix I). Data collection at the community level from both technical and financial perspec- level involved semi-structured interviews with tives and Section 7 examines the forms of inter- extension workers and focus group discussions face within MoA and between MoA and other (FGDs) with mixed groups of farmers (men and stakeholders paying attention to the inherent women) focusing mainly on their perceptions challenges. Finally section 8 offers some about the changes that have taken place within concluding reflections.

Research Paper 013 | October 2009 2 www.future-agricultures.org 2. Organisational Setup of the MoA • The Department of Agricultural Extension at National and District Levels Services (DAES) 2.1. Introduction • The Department of Land Resources and This section briefl y outlines the organisational Conservation (DLRC) setup of the MoA at national and district levels. The MoA is structured just like any other The main purpose of this section is to set the government ministry. The top management context for the rest of the report. It is important structure is headed by the Secretary for to note that the MoA has experienced institu- Agriculture below which there are the six depart- tional restructuring regularly and this has been ments. Below the departments there are eight motivated by challenges being faced at partic- Agricultural Development Divisions (ADDs) ular points in time. It used to be known as the which replicate the activities of the six depart- Ministry of Agriculture; it became Ministry of ments at the national level. The ADDs are Agriculture, Irrigation and Food Security; Karonga, Mzuzu, Kasungu, Lilongwe, Salima, Ministry of Agriculture and Irrigation; and Machinga, Blantyre and Ngabu. Each ADD currently it is known as the Ministry of Agriculture covers several districts but this does not however and Food Security2. The Department of Irrigation coincide neatly with regional boundaries. The has been moved to the Ministry of Water ADDs used to be split into 30 Rural Development Development. The MoA has also experienced Projects and these have now been restructured substantial structural changes as a result of the into the 28 District Agriculture Development ongoing decentralisation policy reforms since Offices. There are further divided into 154 the turn of the 1990s. Extension Planning Areas (EPAs). The EPAs are subdivided into Sections which is the point of 2.2. MoA at the National Level service delivery to farmers. In addition, DARTS The structure of the MoA is largely infl uenced operates a network of 16 experimental sites and shaped by the frequent reorganisations. The strategically located throughout the country. number of departments depends on the nomen- clature of the MoA at a particular point in time. The MoA structure is illustrated in Figure 2. MoA’s departments have now been reduced to The overall mandate of the MoA as expressed six following the shift of the Department of in its 2003-2008 strategic plan is to promote and Irrigation to the Ministry of Water Development. accelerate broad based, sustainable agricultural Current departments are as follows: [and irrigation policies] so as to promote • Administration and General management. economic growth and contribute to poverty This comprises of Administration, Human reduction. The specifi c functions of the MoA Resource Management and Development, include the following: Finance, Internal Audit, Procurement, • To attain and sustain household food suffi - Transport and Planning Departments ciency and to improve the nutritional status • The Department of Agricultural Research of the population; and Technical Services (DARTS) • To expand and diversity agricultural produc- • The Department of Animal Health and tion and exports; Livestock Production (DAHLP) • To increase farm incomes; • The Department of Crop Production (DCP) • To conserve the natural resources base;

Research Paper 013 | October 2009 3 www.future-agricultures.org •• To promote agricultural policies, legislation According to the strategic plan, MoA head- and regulations with stakeholder quarters concentrates on policy formulation and participation; regulation, coordination of training and collabo- •• To generate and disseminate agricultural ration with other stakeholders in the sector. The information and technologies; ADDs interpret policies from MoA headquarters •• To regulate and ensure quality control of for implementation at the points of service agricultural produce and services; and delivery, coordinate subject matter specialists •• To monitor and manage the food security (SMSs), supervises programmes, develop tech- situation. nical messages and train SMSs. The District

Figure 2. The Structure of Ministry of Agriculture and Food Security in Malawi Source: GoM (2003)

Research Paper 013 | October 2009 4 www.future-agricultures.org Agriculture Development Offices (DADO) Ministry of Natural Resources and disseminate messages, train EPA staff and Environmental Aff airs. farmers, provides technical advice and supervi- The administrative section of the MoA at the sion of EPA staff . The MoA also has technical district level comprises human resource responsibilities over several parastatal organisa- management, fi nance, registry and planning tions primarily to ensure that thir activities are sections. As the overall in charge of MoA at in tune with government policy priorities at all district level, the DADO is the head of adminis- times. These include Agricultural Development tration. He/She is assisted by the ADADO who and Marketing Corporation (ADMARC), National is designated as the head of technical services. Food Reserve Agency (NAFRA), Smallholder The ADADO works very closely with subject Farmer Fertiliser Revolving Fund of Malawi matter specialists (SMSs) in the various sub- (SFFRM) and the Tobacco Control Commission programmes in coordinating the delivery of (TCC). services to farmers working with and through Agriculture Extension Development 2.3. MoA at District Level Coordinators (AEDECs) at EPA levels and The MoA structure at the district level more or Agriculture Extension Development Officers less replicate structures at the national level. The (AEDOs) at section levels. The DADO reports head of the MoA at district level is the District both to the Programme Manager at the ADD Agricultural Development Offi cer (DADO) who level on technical issues and to the District on is assisted by an Assistant District Agricultural administrative matters. Development Officer (ADADO). The District Agricultural Development Offi ce has two a tech- 2.4. Decentralization and Ongoing Changes nical and an administrative arm. There are fi ve at District Level technical departments at the district level. These This dual reporting structure has been driven include: by the ongoing decentralisation policy reforms. • Extension Services with the following The reforms advocate for the devolution of subsections: Extension Methodology, Food and Nutrition, Agricultural Communication, service provision from the centre to the district Agricultural Gender Roles Extension Support level. The main objective of the reforms is to Services (AGRESS) and Agribusiness. ultimately transform districts into focal points • Crops with subsections focusing on horti- for planning and service delivery with the view culture, cereals, legumes and tobacco. This to improving efficiency, effectiveness and section is also responsible for plant protec- responsiveness of service provision. MoA is one tion and farm mechanisation. of the ten ministries whose powers, functions • Livestock section focusing on animal health and resources have been fully devolved to the and production. district level since 2005. The decentralisation • Land resources and conservation section policy reforms have entailed the development promoting soil fertility restoring technolo- and the reinvigoration of participatory struc- gies and controlling rampant soil degrada- tures below the district level. The overall coor- tion by promoting proper husbandry dinating body in the district is known as the practices. District Assembly. Immediately below are the • Fisheries section, which has just moved to Area Development Committees (ADCs) and at MoA. It used to be part of the Department the bottom of the hierarchy are the Village of Fisheries and Environmental Aff airs in the Development Committees (VDCs). The structure

Research Paper 013 | October 2009 5 www.future-agricultures.org Figure 3.The Structure of MoA at the District Level Source: GoM (2003)

Research Paper 013 | October 2009 6 www.future-agricultures.org of the MoA at the district level is as depicted in include MoA offi cials, service providers (NGOs Figure 3: and FBOs), agro-dealers and representatives of The structural reorganisation that MoA has farmers. The panels shall work closely with the experienced at the district level has been greatly District Agricultural Committee of the DAs. The influenced by the decentralisation policy Area Stakeholder Panels are envisaged to reforms. Prior to the creation of the District comprise at least 50% of smallholder farmers, Assemblies (DAs), the offi ce of the DADO did farmer based organisations, senior traditional not exist. There were, as indicated above, 30 leaders, MoA offi cials and other service providers Rural Development Projects (RDPs) headed by within the agricultural sector. Agricultural a Projects Offi cer. These RDPs did not conform committees shall be established at the village to district boundaries as is the case of DADO. level comprising extension workers, village Project Officers reported directly to Project heads and farmers. The extension policy is being Managers at the ADDs. The 30 RDPs have been implemented using the model village approach reorganised into 28 District Agricultural even though on a very much ad hoc basis. The Development Offi ces conforming to the bound- idea is that farmers should be assisted to under- aries of the 28 districts in the country designated take situation analyses of their agricultural port- as a unit of operation within the decentralisation folio which will form the basis for them to framework. articulate their demands for extension services Decentralisation has also infl uenced signifi - to the existing gamut of service providers. cantly the adoption of a new policy on the provi- sion of agricultural extension services. The 2.4. Conclusion justifi cation is that decentralisation off ers an The organisational setup of the MoA both at the opportunity to bring the control of extension national and district levels is essentially unstable. to the people and off er services that fi t better It is shaped to a great extent by the frequent with the local situation. Thus a district must institutional changes to the MoA. Reorganisation interpret and implement agricultural extension has meant either adding new departments from policy in a manner that suits its own special other ministries or shifting departments to other circumstances. The policy aims to develop ministries. These changes have largely been pluralistic and demand-driven extension infl uenced by the recurrent episodes of food services by changing resource management insecurity the country has been regularly facing structures to involve local stakeholders and since the turn of the 1990s. These have essen- promote participatory planning and implemen- tially politically driven initiatives in an attempt tation of agricultural programmes. The policy fi nd lasting solutions to the enduring problem recognises that extension services are no longer of food insecurity. Decentralisation policy a public sector monopoly. There are several reforms meant to enhance effi ciency, eff ective- other actors playing an important role, such as ness, effi ciency and responsiveness of service churches, NGOs and farmer associations. delivery have pushed further structural reor- The idea in the new extension policy is to ganisation at the district level including the institutionalise the provision of extension adoption of the new policy on extension advo- service through an institutional framework cating for the development of pluralistic and patterned along the lines of the decentralised demand-driven extension services. sub-district participatory structures. It is envis- Decentralisation has not aff ected the centre aged that District Stakeholder Panels will be much in terms of structural set-up but rather in established. Membership to these panels shall terms of shedding off responsibilities to the

Research Paper 013 | October 2009 7 www.future-agricultures.org lower levels of the ministry’s hierarchy particu- 3.5. Conclusion larly with regard to financial and spending There are several agricultural activities taking decisions. place in Dedza and Thyolo districts determined largely by their agro-ecological conditions. 3. Overview of District Agricultures While these districts generally face similar 3.1. Introduction constraints in their agricultural activities, there This section provides a quick overview of the are some constraints unique to Thyolo. These agriculture sector in Dedza and Thyolo. This constraints arise from the enormous land pres- entails, inter alia, an analysis of the key agricul- sure and mountainous terrain of the district. The tural activities and main opportunities and chal- opportunities identified in the agricultural lenges facing the sector in these districts. sector are the same for both districts. It is, however, striking to note that there were signifi- 3.2. Key Agricultural Activities cant differences between stakeholders and Agricultural activities in Dedza and Thyolo are farmers in terms of constraints and opportuni- determined largely by their agro-ecological ties emphasised much as there were in some characteristics, the availability of land and their cases overlaps topography. Unlike Dedza, Thyolo has for the most part a hilly terrain and experiences severe 4. MoA Functions and Performance land constraints. Much of the cultivable land is 4.1Introduction taken up by tea and coffee estates. In this section, the functions of MoA are exam- The major crops grown in Thyolo include tea, ined. In particular, the focus is on the divergence coffee, bananas, maize, sweet potatoes, cotton, between stakeholders’ perceptions of the func- macadamia and vegetables - it is a major source tions and roles of MoA that are stipulated in its of vegetables for the city of Blantyre. The major mandates against the backdrop of what exactly crops cultivated in Dedza include maize, tobacco, the MoA is doing at the district level. An attempt beans, Irish potatoes, soya beans, vegetables is also made to chart out the evolution of the and a wide range of horticultural crops. Although roles and functions as well as the performance generally considered in decline, livestock of MoA from a historical perspective over the production featured as one of the major agri- last two decades. This analysis is based on the cultural activity in both districts. Chickens, pigs, views of farmers captured in the focus group goats and cattle are raised on a significant discussions as the major beneficiaries of the scale. MoA services. It was difficult to do the same with the other stakeholders because the majority of 3.3. Key challenges and opportunities in the them have been in these districts for a period district agricultures of less than three years. Several challenges and opportunities facing the agricultural sector were identified in these two 4.2. Perceptions of the Functions MoA districts. These challenges were strikingly similar 4.2.1. Farmers’ Perceptions of MoA although there were slight differences service Functions providers and farmers. Unique constraints were The perceptions of the farmers were sought identified for Thyolo district particularly in rela- through focus group discussions carried out in tion to the land availability per capita. These the two districts. Besides seeking their percep- challenges and opportunities are summed up tions of the roles and functions of MoA, farmers in Table 1: were also asked to characterize the performance

Research Paper 013 | October 2009 8 www.future-agricultures.org Table 1. Stakeholder Perceptions of Opportunities and Challenges at District Level Dedza Thyolo Challenges Farmers •• High prices of farm inputs (seeds •• High prices of farm inputs (seeds and and fertilizer) fertilizer) •• Lack of markets to dispose of their •• Lack of markets to dispose of their produce at profitable prices produce at profitable prices •• Lack of access to improved tech- •• Lack of access to improved technol- nology particularly with regard to ogy particularly with regard to irrigation irrigation •• Erratic rainfall patterns in recent •• Erratic rainfall patterns in recent years years •• Exploitation by private traders •• Exploitation by private traders who who offer farmers very low offer farmers very low prices prices •• The devastating impact of the HIV/ •• The devastating impact of the HIV/ AIDS pandemic AIDS pandemic •• Hilly terrain coupled with very small landholding sizes per capita

MoA •• Erratic rainfall patterns in recent •• Excessive land pressure since much officials years of the cultivable land is taken up tea •• Limited funding leading to and coffee estates substantial scale down of exten- •• Land shortages also impacting nega- sion activities such as agricultural tively on livestock production shows, demonstration farms etc •• Poor road infrastructure to support •• Serious staffing shortages espe- market functions cially at service delivery levels •• Limited funding leading to substan- •• HIV/AIDS worsening the acute tial scale down of extension activities shortage of personnel and such as agricultural shows, demon- constraining active farmer stration farms etc involvement •• Serious staffing shortages especially •• Poor road infrastructure to support at service delivery levels market functions •• HIV/AIDS worsening the acute short- •• Low uptake of technology among age of personnel and constraining farmers active farmer involvement •• Non-functioning laboratories which make it extremely difficult for the livestock section to make appropriate diagnosis Other •• Erratic rainfall patterns in recent •• Erratic rainfall patterns in recent stake- years years holders •• Serious shortage of qualified •• Serious shortage of qualified exten- extension staff at service delivery sion staff at service delivery levels levels •• Failure of the MoA to articulate and •• Failure of the MoA to articulate enforce policy directions and and enforce policy directions •• guidelines resulting in lack of coor- and dination and monitoring

Research Paper 013 | October 2009 9 www.future-agricultures.org •• guidelines resulting in lack of •• Poor road infrastructure to support coordination and monitoring market functions •• Poor road infrastructure to support •• Excessive land shortages since much market functions of the cultivable land is taken up tea and coffee estates Opportunities Farmers •• The introduction of the fertilizer •• The introduction of the fertilizer subsidy programme (agriculture subsidy programme (agriculture is is alive again) alive again) •• Introduction of village extension •• Introduction of village extension multipliers to meet the shortage multipliers to meet the shortage of of extension staff extension staff •• The introduction of the new exten- •• Proximity to Blantyre offering pros- sion system with structures pects for markets for well organized extending down to the farmers grassroots •• Complementary efforts by numer- ous stakeholders (NGOs) in the agricultural sector MoA •• Existence of perennial rivers to •• The adoption of the farmer technician officials ensure irrigation on a sustainable concept to alleviate the current seri- basis ous shortage of extension •• The adoption of the farmer techni- personnel cian concept to alleviate the •• An increasing number of projects current serious shortage of exten- targeting the agricultural sector in sion personnel areas f marketing, production, insti- •• An increasing number of projects tutional and capacity development, targeting the agricultural sector irrigation and farmer organizations in areas f marketing, production, •• Newly introduced extension policy institutional and capacity devel- offering prospects for institutiona- opment, irrigation and farmer lised collaboration and coordination organizations among stakeholders in the agricul- •• Newly introduced extension tural sector policy offering prospects for insti- tutionalised collaboration and coordination among stakeholders in the agricultural sector Other •• The availability of funding for MoA •• The availability of funding for MoA stake- in specialized areas through proj- in specialized areas through projects holders ects likely to improve capacity likely to improve capacity •• Initiatives to make up for the •• Initiatives to make up for the current current serious shortage of exten- serious shortage of extension person- sion personnel on the ground nel on the ground •• An increasing number of non- •• Proximity to Blantyre offers prospects state actors within the agricultural for viable markets for organised sector complementing the servies farmers of MoA •• An increasing number of non-state actors within the agricultural sector complementing the servies of MoA

Research Paper 013 | October 2009 10 www.future-agricultures.org of the MoA in service delivery for a period of •• The MoA should vigorously promote live- over two decades. For the farmers just like MoA stock for two reasons. This was particularly officials emphasis was on functions bordering emphasised by farmers from in Thyolo. First, on direct service delivery even though elements livestock is in great decline which has impli- of the regulatory role of the MoA was empha- cations for nutrition of the people and sised with regard to marketing. The roles and income earning opportunities. Second, the functions of MoA highlighted by farmers promotion of livestock would ease the diffi- included the following: culties encountered in accessing fertiliser •• Provision of basic farm inputs particularly due to exorbitant prices by making the improved maize seed varieties and fertiliser manure alternative within easy reach. including regulating the prices of fertiliser. In Dedza in particular, the issue of facilitating 4.2.2. Other Agricultural Service Providers’ farmers’ access to farm inputs through the Perception of MoA Functions club system was particularly emphasised. There were striking differences regarding the Farmers argued for the resurrection of this perception of the functions of MoA at district alternative due to the fact that most of the level between MoA officials and other stake- existing credit institutions are exploitative. holders. While MoA officials emphasised roles bordering on direct service delivery, for instance, They further argued that if going back to provision of inputs, extension messages and the club system is not possible then the MoA training particularly in agribusiness, most of the should play a leading role in identifying and other stakeholders felt that MoA should concern linking up farmers with credit institutions itself with issues of policy, regulation and moni- with fair lending terms. toring. Even in terms of service provision, MoA •• Facilitating access for farmers to markets officials demonstrated bias toward the provision including regulating the prices offered to of services that are geared toward the achieve- them in these markets. In both districts, the ment of food security and income enhancement farmers argued that they have been victims at household level. This should not be surprising of market liberalization. They argued that since these functions are prioritized by MoA as private traders (vendors) are not only stipulated in its vision and mission statement dictating prices but are also manipulating (cf. MoA, 2005) (see also section 2.2). MoA func- the measurement scales to farmers’ disad- tions pertaining to coordination, policy gover- vantage. Consequently farmers are not nance and monitoring are intimated but come getting fair returns for their produce. toward the end of a long list of functions which •• Checking the quality of agricultural inputs may well suggest that they are not priorities. sold by the small scale private traders For most of the stakeholders, the MoA at the (vendors). This was particularly raised with district level has the following key functions to reference to pesticides as both districts are play which they feel the MoA is currently not major producers of vegetables. The farmers performing: observed that oftentimes the vendors sell them substandard pesticides (mixed with Coordination3 maize flour, sand and even ash). The pesti- They contended that MoA has a key role to facili- cides therefore do not achieve the desired tate coordination among various stakeholders impact on the vegetable fields urging the in the agricultural sector broadly understood MoA to intervene. as mechanisms for bringing

Research Paper 013 | October 2009 11 www.future-agricultures.org different stakeholders together to brief them the role of the MoA staff at the district level must on government policy regarding the agricultural be reoriented accordingly. sector. Coordination was emphasised as extremely vital for purposes of ensuring that all Monitoring stakeholders in the sector are operating at least Even some of the MoA officials emphasised the on the same wavelength. This observation was critical role of the monitoring function in view constantly made with reference to the competi- of the new policy on extension that advocate tive tendencies of NGOs, which in the interest for the development of pluralistic and demand of achieving impact at a stroke, tend to cut driven extension services. The gist of this policy corners or use conflicting strategies on the same is that the provision of extension services is no issue. For instance, some NGOs operating in the longer a monopoly of the public sector. Other same area and on similar projects pay benefi- key players include churches, NGOs, farmer ciaries for taking part while others do not. This organisations etc. It was therefore pointed out has tended to create confusion on the part of that it is imperative for the MoA to closely and communities as primary beneficiaries of the consistently monitor how these new actors are services. Most stakeholders pointed to the intro- providing their services to farmers. The idea duction of the DAESS as an opportunity for the would be to ensure that the services provided are in tune with the levels of quality and stan- MoA to institutionalise coordination as one of dards as prescribed in the relevant policies and their key responsibilities. This is particularly guidelines. The inability of the MoA to perform important in view of the pluralism in service the monitoring role was expressed as a serious provision advocated for by the new policy on concern because most of the service providers agricultural extension services. are increasingly employing their own cadres of extension staff who are often under qualified. Policy Formulation According to MoA officials, most of these service Stakeholders emphasised that the MoA should providers are recruiting their staff from numerous assume a leading role in sensitising actors in unaccredited training institutions that have the sector about government policies. The MoA emerged following the liberalization of the should not only sensitise the stakeholders about education sector. The nationally recognized these policies but should also interpret them to training institution for extension workers did ensure that stakeholders’ activities are guided not enroll any students from 1994 to about three by the same operative framework. Many stake- years ago. This has greatly compromised the holders interviewed argued that this should be quality of extension services provided. a primary role of the MoA especially since the In short, in performing these functions, the new policy on agricultural extension services MoA would have been fulfilling the following emphasise that districts, now operating in a roles: decentralised environment, must interpret and •• Coordinating and providing leadership for implement agricultural extension policy in a agricultural activities at district level. manner that suits their own special circum- •• Serving as a source of technical agricultural stances. Thus according to the decentralisation information for stakeholders such as policy reforms, backed up by the Local markets, rainfall data, agro-ecological zones, Government Act, districts are “key planning and farm families etc. operational level of extension and need to be •• Providing technical support to service recognized as such” (MoA, 2000: 32). This means providers in the agricultural sector.

Research Paper 013 | October 2009 12 www.future-agricultures.org •• Creating a link between stakeholders in the four different phases. These included: the 1980- agricultural sector and the government. 1990 phase; the 1990-1995 phase; the 1995- 2000 phase; and the 2000-2007 phase. In Thyolo, 4.3. MoA’s Failure to Perform Key farmers distinguished three phases as follows: Functions the 1980-1993 phase; the 1994-2005 phase; and Several reasons were cited for the failure of the the 2005-2007 phase. While there are notable MoA to perform the key functions as perceived differences in the temporal characterization of by other stakeholders. the phases between farmers in Dedza and On their part, MoA officials pointed out that Thyolo, the overall characterization of MoA their failure to perform these functions is performance is strikingly similar. Farmers were primarily due to financial constraints. As further also asked to rank the MoA performance demonstrated below, the MoA in both districts according to the phases identified. The charac- has been operating under serious financial terization of the phases and the outcome of the constraints which make it impossible for them ranking exercise are summed up in Tables 2, 3 to carryout coordination and monitoring func- 4 & 5. tions. They argued that these two functions The perceptions of the farmers about the would require a significant amount of resources functions of MoA are clearly nostalgic about the for allowances, fuel and even functioning vehi- historical role of ADMARC, which for a very long cles. Besides they argued that that most of the period of time played an important role in non-state service providers are reluctant to be providing marketing services and supplying monitored. They are very reluctant to provide inputs to farmers. It is thus not surprising that reports of their activities to MoA even when they farmers both in Dedza and Thyolo ranked the are specifically requested to do so. first phase as the best period in the agricultural The other stakeholders emphasised different development history of the country. At its peak constraints altogether. The main reason given ADMARC maintained a network of markets was that perhaps the MoA officials do not know extending to every possible corner of the that they are supposed to carryout these func- country which provided farmers not only with tions. They therefore argued that these key tasks ready markets for their produce but also for their do not feature as an integral part of their work basic farm inputs. ADMARC has substantially plans. However, granted that the MoA staff is streamlined its market network over the years aware of these nominal functions, the failure to as a result of structural adjustment programmes deliver is attributed to a bias in practice to direct instigated by the IMF and World Bank. The interventions in the sector and the limited of majority of the markets targeted under the technical capacity in the district - DADO senior auspices of these reforms were located in the officials are often young and recent graduates remote rural areas. Private traders have not filled without the required professional experience. up this vacuum leaving farmers in these areas without any viable market outlets. 4.4. Farmers’ Perception of MoA Performance 4.5. Conclusion In assessing the performance of MoA over the There are significant variations between the last 20 years, the farmers distinguished various perceptions of stakeholders about what the phases of MoA performance segmented on a MoA should be doing and what it actually does temporal basis. In this exercise 1980 was used at the district level. While stakeholders see coor- as a benchmark. Farmers in Dedza distinguished dination, policy governance and monitoring as

Research Paper 013 | October 2009 13 www.future-agricultures.org Table 2. Phases of Agricultural Development and Performance of MoA in Dedza Trends and Time Period Performance 1980-1990 1991-1994 1995-1999 2000-2007 High points of •• Flourishing of •• Introduction •• Promotion of •• MoA intensifies performance vibrant farmer of the hybrid sasakawa land management clubs maize technology campaigns •• Very good technologies •• Promotion of •• MoA intensifies performance of •• Growing use land conserva- hybrid technology the tobacco among tion technolo- promotion industry and smallholder gies •• High adoption rates food security at farmers of •• Promotion of for hybrid maize household level hybrid maize irrigation varieties which was •• Availability and as cash crop technology no longer consid- easy access to ered cash crop but agricultural virtually a substi- inputs made tute for local possible by MoA varieties as the input loan main subsistence schemes to food crop smallholder •• More awareness schemes. created on the need •• Very good for land manage- contact with the ment practices. extension •• Excellent as MoA workers as they intensifies farmer to were closer to the farmer extension farmers and methodologies staying within especially through the villages the introduction of •• Excellent farmer “volunteer” lead mobilisation for farmers or exten- various interven- sion multipliers tions through the •• Monthly meetings Block Extension and trainings for System stakeholder panels •• Formation of •• Adjusting to vibrant village democratization by based farmer empowering the groups farmers to diagnose •• Practice of low own needs. MoA input agriculture more responsive to owing to good grassroots farmers fertility levels problems – less concerned about soil degradation

Research Paper 013 | October 2009 14 www.future-agricultures.org Low points •• Heavy •• High input prices •• Serious land •• Food Insecurity of handed •• Disbandment of farmer degradation reaches disas- performance tactics in clubs high default rates. levels trous levels. enforcing Negative influence of •• High input Significant loan multi partyism with prices number of payment proponents encourag- •• Food insecurity people died of ing for loan non became an malnutrition and -repayment claiming endemic feature men were greatly loans were democratic of households affected. privileges •• Absence of •• Abolishment of Block strong linkages Extension System between MoA •• Extension worker image and the severely ‘denigrated’ as grassroots misguided multi party structure e.g. advocates portrayed village volun- him/her as an agent of teers as used in oppressive loan policies Ministry of •• Extension functions Health adversely affected by poor relations as farmers no longer trusted extension workers •• Extension messages lost value amongst farmers because inputs were highly unaffordable.

the primary functions for MoA, most of the MoA both districts that MoA’s performance is at least officials emphasise on tasks bordering on direct improving since the turn of the millennium after service provision. This is clearly the case because a period of near total collapse throughout the of the emphasis on these functions in the official 1990s. Stakeholders contend that tasks of coor- mandates and responsibilities of the MoA stipu- dination, policy governance and monitoring lated in its vision and mission statement. Much should be the core tasks of MoA at district level as the perceptions of the farmers are in tune mainly due to the adoption of the new policy with the popular perception of MoA officials on extension advocating for the development about their roles, it is worth pointing out that of pluralistic and demand driven agricultural the farmers are also keen to see the MoA extension services. performing some regulatory functions particu- larly with regard to marketing. The perceptions 5. Structures and Decision Making of the farmers about the role of the MoA are Processes for MoA at District Level very much a historical legacy of ADMARC which 5.1. Introduction played a dominant role in providing farmers This section examines the decision making with agricultural inputs, extension and processes of the MoA at district level vis-à-vis marketing services. There is a strong feeling in its organisational setup. Particular attention is

Research Paper 013 | October 2009 15 www.future-agricultures.org Table 3. Ranking of the Phases of Agricultural Development and MoA Performance in Dedza Criteria Rank Strength Weakness 1 1980-1990 •• Extension services were very close to the farmers. Extension workers belonged to the community 2 2000-2007 •• Introduction of extension •• Mobility constraints and staff shortages multipliers and coverage of wide area •• Good linkage through stakeholder panels 3 1995-1999 •• Introduction of “useful” •• Retrogressive political agenda as there was no hybrid technologies will to advance agricultural development through Sasakawa •• Security breakdown •• Hostility to extension workers 4 1991-1994 •• Extension role severely compromised by credit recovery role •• Hostility to extension workers •• No political will as extension workers were largely lazing about and no effort made to motivate extension workers and fill existing vacancies paid to the influence and impact of the decen- has taken the initiative to orient us on our terms tralisation policy reforms on the decision- of reference”. Second, DADOs in both districts making processes of the MoA at this level. expressed concern with the current setup of their offices. The Assistant DADO is designated 5.2. Reporting Structures of MoA as head of technical services whereas the DADO Two concerns were raised with regard to the is the head of administration yet the task of MoA reporting structures at the district level. reporting on both technical and administrative While the SMSs at the district level are under matters is exclusively for DADOs. This was seen the DADOs, they report directly to the divisional as a major handicap because by virtue of being heads at the ADD level. Likewise, divisional the head of administration the DADO rarely goes heads communicates directly to the SMSs with out to the field on technical missions. They there- copies to the DADOs in both cases. This clearly fore argued that since they are responsible for indicates that divisional heads wield much more the overall functioning of their offices, both power than DADOs. Most SMSs interviewed technical and administrative functions should indicated that there are no standard reporting be primarily be vested in the DADOs. The main formats and procedures. One of them in Dedza problem with this arrangement is that DADOs confessed that they actually don’t know what feel that they take responsibility for activities the appropriate procedures are “since nobody over which they have no direct control.

Research Paper 013 | October 2009 16 www.future-agricultures.org Table 4. Phases of Agricultural Development and Performance of MoA in Thyolo Trends and Time Periods performance 1980-1993 1994-2005 2006-07 of MoA High points • Quality extension services • MoA intervenes in of that were utilizing vibrant the input market performance farmer clubs through the • Use of several extension coupons system approaches e.g. agricultural • Extension Services shows, residential trainings, slowly being food and nutrition and home rejuvenated economics displays • More awareness • Easy coverage of the wide created on need for area by extension services land management since extension workers were practices resident in communities • Intensive campaigns • Easy marketing processes on soil and water • Good soil fertility warranting conservation no land conservation eff orts practices • Food security at household level and farm produce stored in granaries outside the homesteads • Availability and easy access to fertiliser through a vibrant club system Low points • Government literally • Offi cial corruption of abandoned the farmer. by government, performance “Disaster is the ADMARC, private catchword for the traders, police in period characterised by handling coupons high theft rates” for fertiliser subsidy • High input prices programme • Vendors controlling • Targeted input ADMARC markets approach benefi ted • Output markets for very few farmers crops became almost • Few selling points to non-existent cater for all farmers • Extension service in good time for key system almost agricultural collapsed as there were practices no trainings and • Late availability of agricultural shows inputs aff ecting the • Role and infl uence of utility of the inputs the extension worker toward ensuring diminished bumper harvests

Research Paper 013 | October 2009 17 www.future-agricultures.org Table 5. Ranking of the Phases of Agricultural Development and MoA Performance in Thyolo Criteria Rank Period Strength Weakness 1 1980 -1993 •• Vibrant farmer clubs •• Easy access to fertiliser and related inputs •• Good efforts at promotion of good land husbandry practices •• Prominent and effective extension approaches like shows, residential trainings and the block extension system 2 2006- 2007 •• MoA responsive to farmer demands •• Official corruption in operating •• Introduction of the “subsidy programme” the coupon system •• Availability of good extension services •• Extension messages not appropriate-emphasis on manure against fertiliser which is more effective •• High input prices and limited markets for farm produce 3 1994 -2005 •• Introduction of extension multipliers •• MoA not responsive to farmer and coverage of wide area demands •• Good linkage through stakeholder •• Vendors left to exploit farmers panels at will

5.3. Decision-making processes these needs into EPA-wide priorities. These Procedurally work plans for the MoA at the EPA-wide priorities are then sent to the DADO district level are supposed to be prepared in a where the SMSs scrutinize these submissions bottom-up fashion. The introduction of the to come up with district wide priorities taking pluralistic and demand-driven provision of into account their own plans of action. The extension services policy further entrenches the district-wide work plans are then submitted to centrality of the bottom-up approach to the the DC and to the ADD. The latter is mainly inter- development work plans. Work plans are critical ested with the technical aspects of the work since they form the basis for budgeting and plans while the former focuses predominantly implementation of MoA activities. on the finances needed to execute the work In the spirit of the bottom-up strategy as plans. advocated by the new policy on extension, the MoA officials interviewed both in Dedza and starting point for the work plans the district Thyolo indicated that these work plans are Sections. The AEDOs are expected to consult hardly implemented as submitted. They are with farmers in their respective Sections to iden- always subject to extensive revisions because tify priority needs for extension. The AEDOs then of funding constraints4 and because these work forward these priorities to the EPA where, plans have to take into account centrally deter- working closely with the AEDECs, they integrate mined priorities.

Research Paper 013 | October 2009 18 www.future-agricultures.org The central MoA priorities are often worked The decision-making processes are further out on the basis of priorities stipulated in the aff ected by the frequent reorganisation of the national development strategies often cham- MoA. There is a tendency to either bring in new pioned by the political leadership. For instance, departments or shift some departments out of currently the MoA has to be seen translating MoA. For instance, the Department of Fisheries some of the priorities laid out in the Malawi has recently been shifted to MoA from the Growth and Development Strategy (MGDS) into Ministry of Natural Resources and Environmental actionable programmes. The other source of Aff airs whereas the Department of Irrigation priorities is the endless donor funded projects formerly in MoA has been shifted to the Ministry spearheaded by MoA. For most MoA offi cials of Water Development. Almost all MoA offi cials interviewed, the challenge becomes how to argued that these frequent shifts, which are meaningfully interface the local and national easily done on paper but not in practice, are a priorities particularly in the context of very huge challenge to local decision-making. For limited fi nancial resources. the incoming departments, the challenge is The concern expressed by the MoA offi cials often to create a favourable working rhythm was that when faced with a choice between with the rest of the departments and for the prioritising national and local level priorities, outgoing departments the challenge is to create preference is always given to the former. The viable coordination mechanisms with the desti- consequence is that the work plans have to be nation ministries without jeopardizing service revised accordingly. It was observed that the delivery. This particular change was specially preponderance of national over local priorities emphasised with reference to the shift of the means MoA at the district levels are operating Department of Irrigation to the Ministry of Water using work plans with packages of intervention Development. AEDOs are jacks of all trades prescribed from the centre when in fact the work including irrigation at the point of service plans were supposed to be developed on the delivery to the farmers. The shift of the basis of problems discerned in the fi eld. Department has therefore been largely in terms The major challenge arising from this modus of resources and not personnel. This is creating operandi is that programmes implemented do problems with regard to irrigation since for the not address the actual problems in the fi eld. This AEDOs to off er irrigation services they have now is the case because according to MoA offi cials to liaise with offi cials from the Ministry of Water most of the problems diagnosed during their Development (MoWD) as MoA no longer needs assessment exercises tend to be area controls the budget line for irrigation. The offi - specific and as such the universal outputs cials from MoWD cannot meaningfully function prescribed by MoA national headquarters do on their own because they are not well versed not address the real problems. It was argued with the agronomy of irrigation. This requires that local level priorities tend to be inferior when some kind of coordination between MoA and faced with a choice of what to prioritise since MoWD which is not always easy to achieve. the centrally determined MoA universal outputs The interaction between MoA offi cials at the tend to be tied to the budgetary ceilings. The district level and the ADDs with regards to deci- view of nearly every MoA offi cial interviewed sion making was considered to work relatively was that this defeats the logic of the new exten- well in both districts. This interaction is entirely sion policy which advocates for pluralistic and on technical issues; fi nancial matters are taken demand-driven extension services. MoA decen- to the District Commissioner’s Offi ce. The MoA tralisation is therefore largely theoretical. offi cials indicated that there are a number of

Research Paper 013 | October 2009 19 www.future-agricultures.org issues that they have to refer to the ADDs for started bearing fruits in Lobi EPA, Dedza guidance. They felt that the ADDs are generally where the area stakeholder panel has responsive but the degree of responsiveness already been constituted and is quite varies depending on the issues at stake. Some vibrant. issues are responded to instantly but most of •• MoA has a chance to learn best practices them take on average two weeks to a month from other players who are well funded and especially if they are matters of policy. This is sometimes have access to new the case because the ADD officials have to liaise innovations. with MoA national headquarters. The decentralisation policy reforms have, as hinted above, brought up considerable chal- 5.4. Opportunities and challenges arising lenges that are invariably creating perceptions from decentralisation policy that provision of agricultural services has been According to this study, decentralisation policy negatively. These challenges relate mainly to reforms have created much more confusion lack of clarity of the dynamics of the decentrali- than facilitating a propitious atmosphere for sation policy reforms to the key stakeholders. improved service delivery in the agricultural According to MoA officials, this is the case sector. Nonetheless the major positive contribu- because the operational implications of the tion of the reforms, as further elaborated in the decentralisation policy reforms have not been next chapter, is that it has transformed districts fully explained to the stakeholders at the district or DADOs offices into cost centres. Prior to the level. Some of the challenges are as follows: designation of DCs as controlling officers in 2005, budgets for MoA district offices were held DADO’s dual reporting structure at ADDs. Decentralisation has allowed budgetary DADOs have a dual reporting obligation. They allocations to be provided for close to the points have to report to the DC at the district level and of service delivery. Each EPA now has a monthly to the PM at the ADD. It was argued that this budgetary allocation since 2005 as the districts arrangement has created a silent but destructive are now budget holders. Potential opportunities tug of war regarding who is who between the of decentralisation policy reforms for the agri- DC and the Programme Manager even though cultural sector cited included the following: DADOs report to them on different issues. The •• Bringing more ownership among farmers challenge is that both the DC and the PM are at which will in turn provide effective solutions the same rank in the civil service hierarchy and to the challenges and the problems that the central government has done very little, if they face within the agricultural sector. any, to clarify the situation. This has created •• Through decentralised structures like Area divided loyalties among MoA staff at the district Stakeholder Panels farmers will effectively level especially since the role of ADDs in district monitor the performance of their respective agriculture is not clearly spelt out making ADD’s extension workers. Non-performing and more or less redundant. Most MoA officials actu- lazy extension workers will be exposed and ally questioned the presence of SMSs at the ADD this will act as a motivation for extension level as well as the rationale for the continued workers to work hard. existence of ADDs within the framework of •• The panels will be as an effective tool for decentralised provision of agricultural services. effecting desired change among farmers as The following sentiments are illustrative of the one of their terms of reference is to provide divided loyalties of the MoA officials at the farmer to farmer extension. This has already district level:

Research Paper 013 | October 2009 20 www.future-agricultures.org “To be honest although we are decentralised, Recentralization tendencies we owe our allegiance to, and take decisions as Concerns about tendencies toward recentraliza- prescribed by the ADD. To the Assembly we just tion were expressed in both Dedza and Thyolo present highlights of our activities. Technical district with regard, for example, to the Farm supervision is superior to administrative one.” Mechanisation Programme that was launched (MoA offi cial, Thyolo). in 2004. It was observed that both the ADDs “Though the ADD no longer controls other and MoA district level offi ces do not have control recurrent expenditures and doing very little to over certain key aspects of the programme that assist districts fi nancially, they still retain the would have been best dealt with at the district supervisory mandate. We therefore owe alle- level. The programme procured animals and giance to the ADD more than we do the District equipment such as tractors and ploughs to Assembly since they are the ones who hold the facilitate mechanisation on smallholder farms. fate of our careers in their hands.” (MoA offi cial, The MoA offi cials contended that the programme Dedza). risks being run down completely as a result of bureaucratic or inflexible decision-making Lack of clarity on key issues tendencies despite the adoption of the decen- There are a number of grey areas regarding tralised mode of service delivery. Some animals several policy issues, for instance, transfers, disci- have overgrown while others have died. The pline and fi nances. This is an issue because all MoA officials interviewed indicated that the regulations governing fi nance and personnel programme has not made any progress at all are centrally determined. The huge discrepancy because “a circular from above strongly prohibits is that while the ADD is responsible for staffi ng offi cers both at ADD and district levels from and salaries for MoA personnel at the district replacing or exchanging the oxen even where level, operational funds for the same come one paired ox is sick or has died”. This is despite through the DC’s offi ce. The paradox is that the continued advice from the officers on the DCs control operational resources yet they ground. It was strongly felt that the directive cannot either discipline or promote MoA staff . from MoA headquarters regarding the handling The ADD reportedly plays a vital role in lobbying of the animals on Farm Mechanization eff ec- for increases in funding for MoA district offi ces tively overrules the ADD and district implemen- but they do not have the mandate to audit the tation mandate. MoA fi nances at this level. The ADDs are unable to keep track of expenditures at district level let 5.5. Conclusion alone institute audits because they do not have The decision-making processes are in theory the mandate since 2005 when DCs were desig- designed to follow a bottom-up strategy rein- nated as controlling offi cers. The MoA funds at forced by the adoption of a new policy on agri- district level can only be audited by offi cials from cultural extension advocating for the pluralistic the Ministry of Local Government and Rural and demand driven delivery of extension Development. The ADDs are simply given services. These decision making processes are expenditure returns. These problems were to be mediated through the MoA structures that generally attributed to the apparent incom- do go all the way down to the grassroots level. pleteness of the decentralisation policy reforms The major challenge is that the practice tends characterised by lack of clarity of operating to be entirely diff erent from the policy rhetoric. principles, standards and procedures. Local level priorities are often superseded by central level priorities in the development of

Research Paper 013 | October 2009 21 www.future-agricultures.org work plans. There is thus evidence of district The MoA staff establishment is clearly biased offices trying to be demand driven but they are toward the technical personnel which is being constrained in their actions by decisions commendable as often establishment tend to taken at the centre often influenced by the be skewed in favour of the support functions. grand national development strategies and There are very few positions in the administra- donor funded vertical projects. While decentrali- tion section (see Figure 2). In both Dedza and sation policy reforms have created opportuni- Thyolo districts, the MoA offices are grappling ties for improved service delivery in the with serious staffing problems. Most key posi- agricultural sector, they have also brought about tions are either vacant or filled simply in an considerable challenges. Most of the challenges acting capacity by people who are highly under are arising from lack of clarity on the operating qualified. principles, standards and procedures governing The SMSs at the district level are supposed service delivery in a decentralised environment to be graduates but in both districts most of largely due to the fact that the decentralisation these officers are certificate holders. process itself is incomplete in a number of Comparatively the staffing problem is less respects. pronounced in Dedza than Thyolo even though it cannot be described as being satisfactory. 6. MoA Operative Capacity of MoA There is no sub programme that is completely 6.1. Introduction without personnel in place even though a good This section assesses the operative capacity for number of them are holding these positions in MoA at district level from technical and financial an acting capacity. For example, only 3 of the 6 perspectives. The justification for this assess- crops officer posts are currently filled and the ment is that the operative capacity of organisa- same happens for extension officer (3 of 5) and tions is key to ensuring success in fulfilling their land resources officer posts (2 in 3). In Thyolo, mandates. It is not enough for an organisation most of the positions are vacant and nearly all to have a technically competent staff compli- the people that are holding the remaining posi- ment without the necessary financial capacity tions are doing in an acting capacity. The staffing to discharge their functions. situation in Thyolo is captured in Table 6. The staffing situation is not any better below 6.2. Technical Capacity the district level. Dedza and Thyolo are divided 6.2.1. Staffing Situation into 10 and 6 EPAs and 169 and 142 Sections,

Table 6. Established Posts for Graduate SMS Filled in Thyolo District

Field Established Posts Filled (Graduates) Filled (Certificate Holders) Extension 5 1 2 Crops 5 0 2 Livestock 1 0 3 Land Resources 2 0 2 Planning 1 0 0 Source: Clerical Officer in the Planning Office, July 2007

Research Paper 013 | October 2009 22 www.future-agricultures.org respectively. It was found that out of the 10 EPAs, wide range of actors which include NGOs, FBOs, only 5 are fi lled by qualifi ed AEDECs in Dedza farmer associations and to some extent the and none of the 6 EPAs in Thyolo is manned by private sector. These actors are providing a wide a qualifi ed AEDEC. Out of the 169 sections in range of extension services to farmers including Dedza there are only 82 AEDOs to oversee in some cases facilitating access to markets. service delivery. For the 142 sections, Thyolo has There is no doubt that these actors are comple- only 56 AEDOs. Out of the 82 AEDOs in Dedza, menting government’s effort but there are 10 of them are on a month to month contract nevertheless concerns that have been raised. having been called out of retirement. On average These actors rely very much on government therefore the AEDOs take care of 3-4 sections. extension workers by inducing them with attrac- The serious shortage of staff is refl ected in the tive allowance packages. The extension workers high extension worker-farmer ratios estimated are demoralized because of low salaries and as at 1: 1000 for Dedza and 1: 3000 for Thyolo. It such they either abandon their positions and needs not be overemphasised that the current become employees of these service providers staffi ng crisis has had signifi cant implications formally or simply accumulate both jobs. on the reach let alone the quality of service delivery to the farmers. There is no doubt that 6.2.2 Causes and Remedies to the Staffi ng technical capacity of MoA is substantially Crisis compromised by the fact that most of the key The worsening staff situation in the MoA cadres positions are yet to be filled by competent was characterized as inevitable. The MoA staff people as confessed by an MoA official in has been progressively declining tending toward Dedza: crisis proportions due to a number of reasons. “I am only a technical offi cer with a certifi cate MoA has lost and continues to lose an increasing in agriculture. I am currently holding three posi- number of personnel to the HIV/AIDS pandemic tions of which two I am acting. I am substantively while at the same time an increasing number the Principal for the Residential Training Centre of the staff are retiring from active service. Death but I am an acting Food and Nutrition Education and retirement have had a huge negative impact Officer and an Acting Agricultural on MoA’s staff complement because these could Communications Offi cer.” not be replaced. The only nationally recognized The shortage of personnel coupled with training institution for extension workers perennial funding constraints as further illus- remained closed for a period of more than a trated below means that MoA is not able to decade. It reopened its doors three years ago deliver all its mandated services to farmers. The now as a semi-public institution. Meanwhile the gap in service delivery has been taken up by a MoA district level offi ces have been drawing

Table 7. Incentive Diff erentials between the NGO Sector and MoA

Position Salary in MoA Salary in NGO DADO MK 39,000 MK 250,000 AEDEC MK 13,000 MK 100,000 Driver MK 5, 000 MK 78,000

Source: Fieldwork, July 2007

Research Paper 013 | October 2009 23 www.future-agricultures.org from EPAs for the existing army of acting SMSs. on the accomplishment of their responsibilities. The combined effects of these dynamics have Most of the MoA officials interviewed confessed seen the MoA at all levels grappling with serious that the search for allowances has led to the staffing problems. prioritization of planning meetings at the The staffing crisis at MoA was further attrib- expense implementation. Given perpetual uted to the failure to retain graduate SMSs. This serious funding constraints, a disproportionate is the case due to poor incentive packages. Not share of any funding opportunity is devoted to only are salaries low but the general conditions planning. In short, MoA officials are preoccupied of service are not attractive at all. It was argued with the struggle for survival. The main motiva- that most of the graduate SMSs simply use MoA tion for MoA officials to attend workshops is the as a stepping stone for careers elsewhere. The allowance without which most of them would majority end up pursing career opportunities not turn up. In the words of the DADO in Thyolo in the NGO sector. If they do not move on, they “people are motivated to work when there are are quickly posted to the ADDs or MoA national allowances and the majority of the old employees headquarters. The NGO sector is an attractive are hanging on simply to complete their service destination for MoA technical officers due to in order not to forfeit their pension the attractive incentive packages. In Dedza, it entitlements”. was, for instance, observed that “almost the •• The MoA has responded to the worsening entire technical team for Concern Universal, staffing crisis in a number of ways without from managers to field staff, used to be employed which perhaps it would have collapsed alto- by the MoA though not specifically from Dedza gether. The strategies have included the district”. Table 6 sums up the incentive differen- following: tials between the NGO sector and MoA equiva- •• Recalling retired extension workers into lent positions. active service on a month to month basis. MoA officials further feel that their remunera- •• The introduction of the farmer technician tion packages are low even when compared to concept (lead farmers) popularly known as some public sector professions especially those extension multipliers 5. in health and education. The concern was that •• Recruiting school leavers as assistant exten- their colleagues on the same grades from these sion workers subjecting them to the on the ministries receive professional and risk allow- job training6. ances while they do not. This is apparently These strategies have not been very effective, demoralizing for MoA employees to the extent however. The MoA’s technical capacity is quite that the majority of those interviewed wondered weak and according to stakeholders outside why agriculture sector professionals are not MoA this has been further weakened by insuf- given better treatment given that the sector is ficient (if they exist at all) renewal strategies to the backbone of the country’s economy. cope with emerging issues granted its dimin- The low salaries have invariably created and ishing staff complement. These stakeholders entrenched what is described as an allowance argued that the MoA has technically failed to culture or syndrome. This is not however unique position its staff to adequately respond to the to the MoA employees. It is a deeply entrenched forces affecting modern agriculture. It was, for practice in the Malawian public service. It essen- instance, argued that the existing cadre of tially implies that employees are always looking AEDOs has been in their positions for the last for opportunities that can spin off allowances 10 years yet they were trained to deal with issues even though they do not have a direct bearing based on research relevant 20 years ago. These

Research Paper 013 | October 2009 24 www.future-agricultures.org cannot be expected to meaningfully adjust to There is generally stability of tenure for MoA challenges in a highly dynamic agricultural offi cials below the district level in both districts. sector. Besides, MoA staff has to endure lack of There is, however, frequent turnover over of staff operational resources to showcase their skills; at district level due to either posting or resigna- lack of capacity building programmes; and a tion. It was observed in both districts that it is poor incentive system that does not recognize very rare for graduate SMSs to serve for over and reward good performance. Such a reward two years before either quitting for greener system could, inter alia, include performance- pasture or being posted away usually to the ADD based salary acceleration and recognition like or the MoA national headquarters. The insta- upgrading courses for best performers. bility of the top positions was perfectly exempli- fi ed in the case of Thyolo. In the last three years, 6.2.3. Career Prospects and Stability of Thyolo has had four DADOs and the incumbent Tenure indicated that in the last three years he has been Prospects of career progression for MoA transferred four times. It was argued that that personnel at least exist. This can be achieved the frequent change of staff at the top has had either through promotion or training. Career negative impact on service delivery mainly advancement on the basis of promotion is through lack of continuity as a result of a heavily limited by the kind of qualification one has. fragmented institutional memory. The regular Those holding diplomas and below cannot ordi- transfers are a huge disturbance to service narily be promoted beyond the position of Chief delivery since diff erent offi cers bring in new Technical Offi cer (CTO) which is an equivalent management styles and an individual needs a entry point for degree holders. Those holding minimum of two to three years to begin making diplomas are now being encouraged to enrol impact. for degree programmes and degree holders to MoA offi cials below the district level can stay enrol for masters’ programmes. Upgrading one’s in the same area for 15-20 years. They are rarely qualifi cation is therefore key to career advance- transferred unless there are serious problems ment not only in MoA but in the public sector with the community in which they work or they as a whole. have been promoted to take a higher post. The There were, however, concerns that career transfers, if they do happen, are often intra progression opportunities vary between sub- district. One of the reasons the transfers are programmes within MoA. This was particularly infrequent is that it is costly since posting for highlighted by those in the Department of this cadre of staff has been decentralised. The Animal Health and Livestock Production. They stability of tenure is quite important because it argued that it is possible for an AEDO belonging enables them to get used to their places of work to the Department of Agricultural Extension even though sometimes overstaying risks the Services at the same grade (Technical Offi ecer) extension workers becoming fully absorbed into as Assistant Veterinary Officer (AVO) to be village life to the extent of neglecting their considered for promotion to Professional Offi cer work. effectively jumping two posts which is not possible for the latter. It is not possible for an 6.2.4 Strategies for Improving Technical AVO to jump posts and one has to serve on a Capacity position a minimum of 4 years in order to be A number of suggestions were made by the considered for promotion. interviewees with the view to improving MoA’s

Research Paper 013 | October 2009 25 www.future-agricultures.org technical capacity. These included the workers at the Section level presenting their following: targets and budgets to the AEDECs who then •• Need for regularized and institutionalised present the budget estimates to subject matter capacity building programmes for field staff. specialists for consolidation and further refine- They need to be appraised about recent ment into a district-wide budget for the agri- technological developments within the cultural sector which is presented to the DADO sector. for submission to the DC and the PM at the ADD. •• Improvement of staff incentives in order to In principle, the budgeting process bottom-up motivate them. The conditions of service starting from EPAs but then the budget priorities have to be revised by, among other things, are reordered by subject matter specialists to renovating institutional houses and improve- align them with national level priorities. ments in transport. Currently extension The budget formulation process is guided by workers get a monthly push bicycle allow- budgetary ceilings given out by Ministry of ance of MK 180 which is not enough to fix Finance headquarters. The budgetary ceilings a puncture. are basically incremental in nature, being •• Establishing more posts at Professional adjusted by approximately 10-15% annually. The Officer grade and progressively sending the budget is structured according to a standard existing cadre of staff for refresher courses classification system which breaks down expen- and graduate training programmes. diture by type (eg. Salaries, goods and services). •• Investing in technical research on various Following the designations of DCs as controlling issues, for instance, new production tech- officers in 2005, MoA offices at the district level niques as well as methodologies for under- have been turned into cost centres. They hold taking production estimates. their own budget for recurrent expenditure and •• Urgently reconsider the remuneration of no longer have to seek authorization for expen- frontline workers who are always left out diture from ADDs. They are now self-accounting when the revisions of allowances are made. institutions. The DADO authorizes expenditure They do not basically have activities for subject to the approval of the DC. The transfor- which they would be entitled to claim allow- mation of MoA district offices into cost centres ances. This leads to AEDOs working for NGOs has ensured that resources should now at least almost abandoning their positions as filter down to the EPAs. government employees. The funds are released from the Treasury to •• Investing in capital resources particularly the MoA district offices through the office of laboratory equipment as most of the labo- the DC on a monthly basis. The procedures for ratories are currently dysfunctional or in disbursing funds to EPA were found to be state of disrepair. different for Dedza and Thyolo. In Dedza, the disbursement of the funds starts by first giving 6.3. Financial Capacity each EPA an equal amount for basic items such 6.3.1. Budgeting Process as fuel, paper, pens and allowances. The The budgeting exercise is closely related to the remaining funds are shared proportionally to development of work plans at the development EPAs. The rule of thumb is that EPAs imple- plan but almost entirely divorced at implemen- menting specific projects get a slightly higher tation. Thus the MoA budgets are linked to the allocation than those without. Sometimes the work plans. The budgeting process is conducted differential disbursement of funds takes into in a bottom-up manner. It starts with extension account the number of sections in each EPA.

Research Paper 013 | October 2009 26 www.future-agricultures.org The funds are kept at MoA district offi ces. EPAs resources. Extension services are treated as a simply make claims against their monthly alloca- separate budget item separate from administra- tion. Each EPA in Thyolo gets 1% of the total tion as illustrated in Figure 4. budget. Monthly allocations to EPAs are decided The budgetary cuts were described as a on the basis of the 1% staggered over the entire normal feature not only for MoA but for the fi scal year. These arrangements have ensured public sector generally. These cuts are some- that ‘more’ resources are allocated to EPAs than times communicated but it was emphasised that to the district headquarters. The annual budgets the silent rule is that it is of no use to seek feed- for Thyolo and Dedza were pegged at MK 1.4 back since the funds are channelled through a and MK 1.7 million respectively. diff erent sector arguing that Local Government offi cials cannot have the required answers since 6.3.2. Budget Implementation they merely serve as a conduit for the resources. The district MoA offices always experiences Usually no reasons are given for budgetary cuts. budgetary cuts even though the budgeting These announcements do not necessary result exercise is guided by ceilings provided by the in corresponding readjustments of the work MoA headquarters. The ceilings were therefore plans as refl ected in the following sentiments: described as mere guidelines. The fl uctuations “The ‘approved budget’ is presented some- in the funding patterns are perhaps aptly illus- times towards the end of the fi nancial year when trated at the national level for which data was we have already overspent on some items. This available in Figure III 3. The funding levels for results in chaos because we have to readjust the various sub-programmes have been highly our priorities in line with the magnitude of the in consistent fluctuating year by year with budget. In the worst case scenario we completely administration taking up a larger share of the cancel other important activities but sometimes

Figure 4. Expenditures by Programmes at the National Level between 1990 & 2007 Source: Various Budget Documents between 1990 and 2007

Research Paper 013 | October 2009 27 www.future-agricultures.org we have to borrow from other sub-programmes.” There are, however, three major challenges (MoA official, Dedza). with regard to budget implementation. The “We are just communicated as to how much quality of the overall budgeting process is very we have been funded. Sometimes the cuts are low. In both Dedza and Thyolo, the planning explained by the accounts section as emanating sections in the DADOs’ offices are manned by from errors at the Treasury. For example, we extremely under qualified staff. The planner in might be told that funding allocation to the the DADO’s office is supposed to be a graduate agricultural sector has been increased but when in a relevant field but currently in both districts we query the DC’s office we are told the increase the planning sections are overseen by officers has only been effected for the votes controlled at a clerical officer grade. The ADDs and the MoA by the MoA headquarters where the ADDs get district offices are funded differently. The ADDs their funding” (MoA official, Thyolo) are funded directly through the MoA headquar- There are no budgetary freezes but the ters whereas the district offices are funded budget ceiling for some expenditure items- through the Ministry of Local Government and particularly fuel, stationary and allowances - is Rural Development. This creates a management reached before the end of the financial year The dilemma in the sense that the ADDs cannot most affected sub-items are fuel, stationery and audit MoA accounts at district level yet they allowances. They argued that this is mostly the reportedly play a key role in lobbying for funding case due to ad hoc programmes often directed increases for the district offices. The paradox is from the centre. that the ADDs retain a supervisory mandate over

Figure 5. Budget Estimates for ADDs for the 2006/2007 and 2007/2008 Fiscal Years Source: Ministry of Finance Budget Documents 2006/2007 and 2007/2008

Research Paper 013 | October 2009 28 www.future-agricultures.org Figure 6. Budget Estimates for RDPs for the 2006/2007 and 2007/2008 Fiscal Years Source: Ministry of Finance Budget Documents 2006/2007 and 2007/2008

technical matters in the agricultural sector at offi ces are less than a quarter of the magnitude the district levels. The budgetary cuts are of resources earmarked for expenditure at the extremely ad hoc. They do not take into account MoA headquarters. While it is said that the of sub-programme outputs, targets and resource fl ows to district levels have improve activities. following the adoption of decentralization The ad hoc nature of the budgeting process policy reforms in 2005, the amounts are clearly appears to be a normal phenomenon at al levels. insignifi cant to register the desired impact. The There are, for instance, considerable variations disproportionate share of budgetary resources in the budget estimates for the 2006-2007 and in favour of MoA headquarters means that much 2007-2008 fiscal years at both Agricultural is spent n administrative overheads compared Development Division (ADD) and Rural to the actual technical work that would posi- Development Project (RDP)7 levels as depicted tively impact on the agricultural sector. in fi gures 4 and 5 respectively. It is worth noting as shown in fi gure 6 above 6.3.3. Other Sources of Funds that there are huge imbalances in terms of The MoA district offi ces do get funds from other resource allocation at the three levels of MoA’s sources besides the regular budgetary entitle- operation, namely: headquarters, ADD and ments from the central government which are district levels. The budgetary estimates for the not generally refl ected in the national budget. 2006-2007 and 2007-2008 clearly show that They get funds from special projects within MoA resources are concentrated disproportionately and from NGOs. at the headquarters barely trickling down to the Some of the major projects include the ADD and district levels. The combined volume following: Irrigation and Rural Livelihoods of resources allocated to ADDs and district Development Programme (IRLAD); Smallholder

Research Paper 013 | October 2009 29 www.future-agricultures.org Figure 7. Total Budget Allocations to MoA HQs, ADDs and Districts Source: Ministry of Finance Budget Documents 2006/2007 and 2007/2008

Irrigation Project (SHIP); Farm Income the Rabies Week Vaccination Campaign in both Diversification Project (FIDP); Institutional districts. Development Fund for Agriculture (IDAF); and Rural Income Enhancement Project (RIEP). MoA 6.3.4. Strategies for Improving MoA’s officials at the district level do not however Financial Capacity exercise any discretionary powers of expendi- Several ways in which the financial capacity of ture over project financial resources. Expenditure MoA at the district level can be enhanced were is strictly on activities prioritized by the projects’ suggested by various stakeholders. These coordination units at MoA headquarters. included the following: Similarly NGOs provide funding only for those •• Budgetary cuts should not only be consulta- areas and activities that are of direct relevance tive but should also be communicated in to their respective programme portfolios. It was good time with concrete explanations for in fact emphasised that very rarely do NGOs respond to requests for activities that do not the district level MoA offices to make the fall within their areas of interest but are of partic- necessary adjustments. ular importance to MoA. Some stakeholders do •• Approved budgets should be communi- support MoA’s activities on a one off basis. cated to MoA officials timely. Delayed Companies such as BAT Malawi, Toyota Malawi communication greatly affects implementa- and RAB Processors have supported agricultural tion because it is not easy to readjust the shows by making available gifts to farmers. The work plans to conform to the budget which Ministry of Health has also supported MoA on has also been greatly trimmed.

Research Paper 013 | October 2009 30 www.future-agricultures.org • Careful planning should be undertaken from other sources such as projects and NGOs when making policy changes that affect but the constraint is that they are not in a posi- budgetary allocations. An example given tion to exercise any discretionary powers over was the recent increase of allowances from these resources. The usage of the funds is prede- MK 1060 to MK 6000. This was deemed as a termined. The combined eff ects of weak tech- substantial increase which has had tremen- nical competence and fi nancial capability have dous knock down eff ects on several activi- manifested in the less satisfactory service ties as there have been no corresponding delivery by MoA at district levels. increases in the funding levels. • Monthly allocations should consider that 7. Interface between MoA and other agriculture is a seasonal enterprise and as stakeholders such allocations should be seasonally sensi- 7.1. Introduction tive. Funding allocations should refl ect this This section explores the extent of interface seasonality between MoA and other stakeholders at district level not only in service delivery but also in terms 6.4. Conclusion of policy design, budget preparation and, mobil- The MoA district offi ces have very weak opera- isation of resources. The main purpose is to tive capacity both in terms of technical compe- establish areas of interface and whether there tence and financial capability to effectively are constraints faced and how these constraints deliver services to the farmers. The MoA district can possibly be alleviated. The issue of interface level offi ces are grappling with serious personnel borders on coordination and collaboration shortages at all levels. Extension staff at EPA and among stakeholders in the delivery of a service section levels is in short supply and qualifi ed to benefi ciaries in whatever sector. SMSs at the district level are almost non-existent. The problem of shortage of personnel was 7.2. Forms and levels of interface particularly exacerbated by the closure of the Several forms and levels of possible interface only training institution for more than a decade between MoA and other stakeholders were besides the devastating impact of the HIV/AIDS discerned during fi eldwork. There is need for pandemic on the extension staff cadre. It has interaction between different parts of MoA; also been difficult to retain graduate SMSs interface between MoA and other closely related because of unattractive conditions of service in public sector agencies; and interface between MoA. Most of them tend to move on to the NGO MoA and other stakeholders particularly NGOs, sector. FBOs and the private sector actors. All these The funding situation for district level MoA forms of interface can take place at district and offi ces is hugely unpredictable due to irregu- community levels. The possible areas of inter- larity in funding patterns even though since the face between MoA and other stakeholders designation of DCs as controlling offi cers, the include policy design, budget preparation and amount of resources fi nding their way close to service delivery which can be achieved through points of service delivery has greatly improved. consultation, briefi ng, exchange of information, The major challenge for MoA at district level is mobilisation or provision of resources, service that the budgetary cuts are ad hoc and are not delivery etc. communicated to them. This makes it extremely The new policy on extension makes the inter- diffi cult for them to readjust their work plans face between MoA and other stakeholders accordingly. MoA district offi ces get funding imperative. The gist of the policy, as stated

Research Paper 013 | October 2009 31 www.future-agricultures.org elsewhere in the report, is that farmers must be overall agricultural agenda for a district but also served through more pluralistic and demand- ensuring that stakeholders work toward a driven extension services after a process of common goal. The institutionalisation of the stakeholder consultation towards defining DAESS is waiting for funding to be provided by better policy and practice. This is aptly captured IDAF. in the following observations made in the policy In the absence of DAESS, the District Executive itself. (DEC) deputizes as a coordinating framework “Coordination at a decentralised level is of for agricultural activities in the district. DEC is a particular importance, especially with greater technical advisory body to the District pluralism, in that, it is possible to achieve stake- Assemblies that bring together heads of sectors holder interaction, linkages and harmony and NGOs operating in a district. It has several towards fulfilling common goals. Decentralised sub-committees including the District sectoral coordination and linkages will, there- Agricultural Coordinating Committee. Most of fore, enhance complementarity of efforts, these committees, according to the consulta- improve provision of quality services and reduce tions carried out, are non-functional. DECs are wastage of scarce resources.” (MoA, 2000: 26). scheduled to meet on a monthly basis but this This is further underscored by MoA’s new is rarely the case due to funding constraints. For model village operational strategy. The model this reason, DECs are not seen as effective vehi- village approach advocates for a multisectoral cles for information exchange, discussion and approach to development with the view to strategizing on technically orientated issues as benefiting from synergies arising from collabo- reflected in the following sentiments: ration with other stakeholders (government “At district level, there is no sharing of infor- ministries and departments, NGOs, FBOs and mation and DEC meetings are only held at the private sector) in a wide range of rural livelihood convenience of NGOs and the only way to raise activities. The first stage in developing a model an issue is through AOB.” (MoA official, village is therefore joint assessment of village Thyolo). needs, planning possible interventions and “We have several burning issues to raise with implementation. MoA but for some reason we do not have a communication channel. We assume the SMSs 7.3. Mechanisms for coordination and know how we should operate but the problem collaboration is that the information does not trickle down A systematic coordinating structure for agricul- through to EPA staff.” (NGO official, Thyolo). tural related activities does not exist at the “There are no clear mechanisms for coordi- district level. The potential coordinating frame- nating sectoral issues at district level. Sectors work for agricultural activities at this level is the with burning issues do call for DEC meetings. District Agricultural Extension Support System These are often impromptu and usually are (DAESS) which is still very much in the pipeline, called by outsiders. If a sector has an issue it is under the auspices of the new policy on agri- tabled as AOB and at this time participants are cultural extension services. It is, as intimated tired and little is gained in discussing technical elsewhere in this report, envisaged that the issues.” (NGO Official, Dedza) institutionalisation of the DAESS will see the “As things stand, there are no regular meet- constitution of District Stakeholder Panels ings amongst stakeholders not even to talk bringing together key actors in the agricultural about technical meetings. The only time interac- sector for purposes of not only charting out the tion takes place is during DEC meetings and

Research Paper 013 | October 2009 32 www.future-agricultures.org these are usually called by a department or to MoA but unlike the rest of its departments, agency that has a specifi c issue to communicate the Department of Fisheries continues getting to DEC. And DEC meetings are hardly a condu- funding directly from the headquarters. These cive forum for addressing technical issues since problems are also encountered between sub- the only time MoA or any agency is given an programmes that have always been an integral opportunity to share its activities, programmes part of MoA. MoA offi cials in both districts, for and issues is through AOB and practically there instance, argued that there is practically no is little to be gained through this approach.” interaction between crops and agribusiness (MoA offi cial, Dedza). sub-programmes in as far as deriving synergistic “Stakeholders interact through DEC meetings effects from their converging niches is but these are irregular and the agenda and concerned. In providing services to crop-based membership changes almost all the time so associations, there is no evidence of joint plan- issues cannot be properly followed up. In all ning or even sharing of information by these fairness, this is not the right forum because in sub-programmes right from the planning stages. most cases sector heads tend to delegate to very On paper, there is supposed to be joint technical junior staff and (sometimes administrative staff ) backstopping on production and organisational who do not have the required capacity to discuss from the crops sub-programme, and on technically orientated issues at this level.” (MoA marketing from the agribusiness sub-pro- offi cial, Dedza). gramme. Instead their eff orts are isolated and The conclusion is therefore that there is very not complementary. The result is that most crop- little interface between MoA and other stake- based associations are struggling with marketing holders on policy issues (see also section 4.2). as a result of lack of proper interface between The new policy on extension mandates districts the sub-programmes. to develop district specifi c agricultural policies Special projects within MoA also jeopardise taking into account the uniqueness of their interface mainly due to infi ghting amongst sub- circumstances. In both Dedza and Thyolo, the programmes. These projects are often inte- policies are yet to be produced and this was grated or multifaceted in nature. Fights arise attributed to the excessive delays in institution- over which activities constitute the core yet it alising DAESS. is obvious that MoA by its very nature is a sector that thrives on strong complementary intra- 7.4. Status of interface between MoA and sector linkages. Several MoA offi cials confessed relevant stakeholders that cases abound of failure of MoA integrated 7.4.1. Intra MoA interface projects due to coordinators’ inability to account Internal MoA problems of interaction and coor- for the inputs to be provided by other sub- dination exist. These can be attributed to programmes. In the process, other players are frequent reorganisation of MoA and to the effectively blocked from fully implementing endless introduction of special projects within their designated functions in the project. MoA. Departments that are brought into the MoA fold feel that they are very much outsiders. 7.4.2. Interface between MoA and other It was argued that this is the case because joint public sector agencies sub-programmes meetings in MoA are irregular The adoption of the model village approach and most of the activities where there is close portended a platform for deepening interface collaboration are ad hoc in nature. For instance, between MoA and several other relevant public the Department of Fisheries has been shifted sector agencies in the areas of planning,

Research Paper 013 | October 2009 33 www.future-agricultures.org implementation, service delivery, resource predominantly forestry activities like bee mobilisation, monitoring and evaluation. In keeping. Oftentimes, MoA have facilitated both districts, the experience with the model production of beehives with wrong specifica- village approach with regard to deepening tions and only invites DoF officials when they interface between MoA and the other public encounter serious technical problems.” (DoF). sector agencies was regarded as less satisfac- In a broader scheme of analysis, problems of tory. Stakeholders interviewed indicated that proper interface exist because these public the interface did not go beyond diagnostic exer- sector agencies are at different stages of decen- cises. This was attributed to limited funding, lack tralisation which makes coordinating decision of commitment from field staff and bad experi- processes fairly challenging. While for instance ence with operationalising multisectoral MoA is decentralised, DoF is still centralised. This approaches. means that officers at the same level have Beyond the model village approach, the differing mandates to make decisions at the problematic interface between MoA and other point of service delivery. It is was therefore relevant public sector agencies was attributed argued that until all sectors and departments to the lack of a culture of intersectoral collabora- are fully decentralised, it would be expecting tion even where it is obvious that the success too much to realize meaningful interface among of the activities concerned is dependent on the public sector agencies in the areas of planning, sectors working closely. In both districts, this mobilisation, service delivery, monitoring and challenge was exemplified by the problematic evaluation. interface between MoA and the Department of Forestry. DoF officials accused MoA of not only 7.4.3. Interface between MoA with other disregarding collaboration but also as intruding agriculture sector stakeholders in its areas of jurisdiction using sometimes MoA’s interface with other stakeholders particu- distorted and divergent approaches. The larly NGOs is probably the most challenging. following sentiments underlie the lack of proper Interface between MoA and the private sector interface between MoA and DoF: is virtually non-existent. The interface between “There is conflict on land use as regards river MoA and NGOs is largely on a bilateral basis and bank cultivation. DoF’s recommendation allows it is generally characterized by accusations and cultivation about 30 m from the river bank but counter accusations. The major accusation by with the introduction of the treadle pump tech- MoA officials against NGOs is that they prefer nology this is not considered.” (DoF official, to go it alone consulting only after they Dedza). encounter problems in the field. NGOs accuse “There is duplication of efforts for example MoA officials of scaring them away because of in the Rural Livelihoods Support Programme the unnecessary demands for allowances. This where in the same village MoA and DoF officials is perhaps aptly captured in the following are setting different nurseries and conducting sentiments: different trainings on the same aspects of “Some activities for which allowances are nursery management and agro-forestry respec- demanded do not meet the criteria for payment tively.” (DoF official, Thyolo). because they are a core function for MoA like “Collaboration efforts are partly being supervision.” (NGO official, Dedza official, hampered by the greed of the officers involved. Dedza). Occasions abound when MoA officials have “MoA officials demand night allowances for secured donor funding to implement jobs done while at their workstations. NGOs are

Research Paper 013 | October 2009 34 www.future-agricultures.org therefore fi nding it hard to pay high allowances to interface with MoA structures at this level for government offi cials. It was aff ordable when because they do not have the requisite technical their allowances were pegged at MK 1060. Now, know how to execute their programmes. MK 6000 is too high.” (NGO offi cial, Dedza). The interface amongst NGOs themselves is Despite these constraints two NGOs were not any better. Incidences of rivalry between particularly singled out as working very closely NGOs were frequently reported in both districts. with MoA at all levels: planning, budget prepara- There is a widespread tendency among NGOs tion, service delivery, monitoring and evalua- to oversubscribe to a particular area when other tion. These are Concern Universal in Dedza and areas in the same district are not served at all Oxfam in Thyolo as highlighted in the following implementing almost the same interventions. observations: They thus largely view themselves not as part- “With Oxfam, there is a systematic mecha- ners but rather as competitors. This is the case nism. It provides funding for sector specific because NGOs are very much concerned with planning and review sessions; multisectoral achieving signifi cant impact at a stroke in order planning and review sessions; and joint fi eld to retain or court further donor support. supervisory visits as well as community review Attempts in both districts have been made to sessions where community leaders and exten- establish an NGO consortium as a forum for sion workers from all sectors discuss progress entrenching modalities for partnership, coordi- of various development interventions.” (MoA nation and collaboration but the initiative has offi cial, Thyolo) not been a success. This was attributed to the “Of all stakeholders, Concern Universal is the lack of powers of censure in the offi ce of the DC. one that has a systematic approach because we The point is that the recommendations of the plan together at district level and we also imple- NGO consortia would be meaningless if not ment hand in hand at EPA level. Others despite backed up by any sanctioning power, which that they do not have the technical capacity do should ideally reside in the DC’s offi ce in their not involve MoA structures but only consult capacity as controlling offi cers. when something goes wrong.” (MoA offi cial, Dedza) 7.5. Strategies for Improving Interface Strikingly the level of interface between MoA between MoA and Stakeholders and NGOs is relatively deeper at the EPA level Several suggestions were put forward regarding than is the case at the district level. MoA offi cials how the interface between MoA and its stake- at the EPA level even work closely with the holders can be improved. These include the emerging network of agro-dealers. Agro-dealers following: work with MoA staff at this level in carrying out • Strictly enforce the draft MoU with the demonstration plots for new varieties of crops Council of Non Governmental Organisations and brands of pesticides. One of the agro- (CONGOMA) on the protocol of NGOs opera- dealers interviewed observed that “there are no tions in the agricultural sector. formalities involved in our interaction with EPA • Accelerate the institutionalisation of the staff . They are close by and we just work with DAESS as a mechanism for interaction of the them as it is within their job description”. NGOs stakeholders in the agricultural sector at the pointed out that it is easier for them to interface disctrit level so that stakeholders learn from much more with offi cials at EPA than at district each other. This would, inter alia, help curb level because they do not demand huge allow- the competitive tendencies among stake- ances. For most MoA offi cials, NGOs are forced holders especially NGOs.

Research Paper 013 | October 2009 35 www.future-agricultures.org •• Redefining the roles and functions of MoA Development Projects (RDPs) whose areas of district staff. They should not regard them- jurisdiction did not necessarily coincide with selves as implementers but rather concern district boundaries. The decentralisation policy themselves with much more strategic issues reforms have also led to dramatic changes in such as policy governance, coordination and the funding modalities for MoA district level monitoring. offices. Instead of getting funded directly from •• Providing for the sanctioning powers in the MoA headquarters, MoA district offices are office of the DC for deviant stakeholders in funded for their operational costs through the the course of service delivery in their respec- Ministry of Local Government and Rural tive areas of jurisdiction. Development. The funds are disbursed to MoA district offices through the DCs’ offices. The share 7.6. Conclusion of budgetary resources is, however greatly There is no doubt that the interface between skewed in favour of the MoA headquarters, with MoA and other stakeholders is far from ideal. relatively insignificant amounts finding their There are problems of interface with all the three way to the ADD and district levels even though forms of MoA interface. This means that there it is generally said that there have been signifi- is very little interaction among stakeholders cant improvements in resource flows from MoA within the agricultural sector bordering on plan- to the district level since the reorganization of ning, implementation, resource mobilisation, RDPs into offices under the auspices of monitoring and evaluation. It is striking that decentralization. MoA interface with other stakeholders is much Furthermore, decentralisation reforms have deeper at EPA level than is the case at the district led to the adoption of a new policy on extension level. For NGOs this is the case because of the advocating for the development of pluralistic excessive demands for allowances by the district and demand-driven agricultural extension level staff whereas from the perspective of MoA services through the involvement of stake- officials, NGOs are forced to interface with MoA holders and promotion of participatory plan- structures because they simply do not have the necessary capacity to undertake their ning and implementation of agricultural programmes effectively. The interface among programmes. The policy envisages the establish- NGOs themselves is not without problems ment of a participatory institutional framework either. They regard each other not as partners patterned along the lines of the sub-district but rather as competitors. The bottom line is participatory structures currently in place under that the interface between agriculture sector the auspices of the decentralisation policy stakeholders has to be improved. reforms. These changes have also been taking place against the backdrop of frequent tenden- 8. Concluding Reflections cies to reorganize the MoA depending on the There is little doubt that the MoA is undergoing prevailing challenges at particular points in time. through a period of significant transformation The reorganisation has involved either shifting at the district level spearheaded by the ongoing departments out of MoA or bringing new decentralisation policy reforms. The reforms departments into MoA fold. The major issues have driven significant structural changes to arising out this study can be summed up as MoA at district level. Prior to the reforms there follows: were no MoA district offices. Service delivery at •• Service providers and beneficiaries empha- district level was coordinated through the Rural sised on different constraints and challenges

Research Paper 013 | October 2009 36 www.future-agricultures.org facing district agricultures even though they devastating impact of the HIV/AIDS were some overlaps. pandemic; and an increasing number of staff • There are signifi cant variations between the proceeding on retirement without replace- perceptions of stakeholders about what the ment. The funding situation is hugely unpre- MoA should be doing and what it actually dictable due to irregularity in funding does at the district level. While most MoA patterns coupled with the fact that offi cials and smallholder farmers empha- budgetary cuts are ad hoc and rarely sised on tasks bordering on service delivery, communicated to MoA officials in good other stakeholders see coordination, policy time. The combined effects of these governance and monitoring as the primary constraints have tremendously negatively functions for MoA. impacted on service delivery to the • While decision-making processes are farmers. designed to follow a bottom-up strategy • The interface between agriculture sector strongly reinforced by the adoption of a new stakeholders and internal coordination policy on agricultural extension advocating within the MoA are problematic. This means for the pluralistic and demand-driven that there is very little interaction among delivery of extension services, the practice stakeholders within the agricultural sector tends to be entirely diff erent from this policy regarding planning, implementation, rhetoric. There is evidence of the MoA district resources mobilisation, monitoring and offi ces trying to be demand driven but they evaluation. It is, however, striking to note are being constrained in their actions by that MoA interface and NGOs is relatively decisions taken at the centre usually driven deeper at EPA level than is the case at district by the grand national development strate- level. The interface among NGOs themselves gies and donor-funded vertical projects. is not without problems either. They regard • Much as the decentralisation policy reforms each other not as partners but rather as have increased opportunities for improved competitors. service delivery in the agricultural sector, it Given the current operative capacity and has also brought about considerable chal- environment for MoA district level offices, it lenges. Most of the challenges are arising would be expecting too much for them to mean- from lack of clarity on operating principles, ingfully play a coordination role and be demand- standards and procedures governing service driven. While a policy framework for the MoA delivery in a decentralised environment to assume a coordinating role in a demand- largely due to the fact that the decentralisa- driven fashion exists, the MoA district level tion process itself is incomplete in a number offi ces are clearly constrained by limited tech- of respects. nical and fi nancial capacities. The MoA district • The MoA district level offi ces have a very offi ces are grappling with acute staff shortages weak operative capacity both in terms of at all levels coupled with teething fi nancial prob- technical competence and fi nancial capa- lems to provide even the most basic services. bility to effectively deliver services to The majority of the existing staff complement farmers. The MoA district offi ces are grap- is orientated toward roles and functions pling with serious personnel shortages at bordering on direct service provision and it all levels due to several factors: the closure would require reorienting them toward of the only nationally accredited training executing strategic functions such as coordina- institution for nearly a decade; the tion, policy governance and monitoring. It is,

Research Paper 013 | October 2009 37 www.future-agricultures.org however, doubtful whether this would be a success given that the majority of the people holding key posts at the district level are highly under qualified. The capacity of MoA to play a coordinating role at the district level is further constrained by the delays in the institutionalisation of the DAESS under the auspices of the new policy on agricultural extension services. The potential of the DAESS may, given the current orientation and capacity of the existing cadre of MoA staff, be overestimated. Moreover, it is relatively easy to put structures in place but very hard to make them function effectively. This is related to the apparent contradictions between rhetoric at the policy level and the actual practice. While advo- cating for the provision of extension services on a demand-driven basis, the MoA district offices are constrained by the continued domi- nance of the central over the local priorities in framing work plans for implementation within the agricultural sector. This is further compounded by challenges brought about the implementation of the decentralisation policy reforms. A number of issues remain unclear and unresolved that if clarified and resolved would go along way in propping up the capacity of MoA to assume a coordination role and be demand-driven. In the final analysis, it is impera- tive that the core functions of MoA district offices be redefined to prioritize tasks of coor- dination, policy governance and monitoring especially following the adoption of the new policy on extension advocating for the develop- ment of pluralistic and demand-driven agricul- tural extension services.

Research Paper 013 | October 2009 38 www.future-agricultures.org Appendix I: Checklists for Field MoA offi cials, one to be used in interviews with Work non-MoA informants, and one for focus group Three sets of questions are proposed to guide discussions. the fi eld work: one to be used in interviews with

Section I: Questions for MoA offi cials: These should include DAEO, Crops Offi cer and 2+ extension offi cers

Issues Guiding questions Context – the • What are the key agricultural activities in the district? sector • What are the key challenges associated with those activities (factors constrain- ing growth, profi tability etc)? • What new opportunities for agriculture in the district? MoA functions • What are the key functions of the MoA at district level? How do these functions link with the policy objectives for the sector? • What services/activities does the MoA provide/carry out in the district? Technical • How many staff do you have in the district? What is the proportion of adminis- capacity trative/technical staff ? Try to get numbers by rank/type (e.g. extension offi cers, DAEO) • How does this number compare with that 5 years ago, 10 years ago? (get trend if not actual fi gures, but actual preferred) • What is the qualifi cations profi le of technical staff in the district? • What are the career progression opportunities for technical staff ? • What are pay levels for fi eld level technical staff , and for District level offi cers? How do these compare to (a) work with NGOs, (b) work in private sector (e.g. a fertiliser company)? • How long have are fi eld staff usually in post before transfer? How long are District offi cers in post before transfer? • How adequate are this structure and staffi ng levels to perform the MoA functions and pursue its policy objectives? • [For extension offi cers]: how many farmers within your area are you in regular contact with? MoA structures • To whom are district-based agricultural offi cers accountable? How are objectives and decision- and workplans set at district level? making • Ask: tell me about the last time you had to refer to a higher level to get a decision � what was the issue, how long before you got an answer, and was it useful? Then tell me about the time before that. Financial • How are you involved in annual budgeting? Do you submit a budget? To whom? capacity How do you do this? What factors do you take into account? Are you given clear (resource guidelines and ceilings? Is budgeting just done by District MoA, or do other mobilisation actors participate? and use) • How do your submissions turn out? Do you normally get all you request, or just a part? Are you consulted about cuts? • How long does it take for AIEs (authority to incur expenditure) to arrive when the budget has been drawn up at HQ?

Research Paper 013 | October 2009 39 www.future-agricultures.org Issues Guiding questions •• Have you suffered further cuts to your budget during the FY, or had freezes on spending imposed? •• Do budgeted non-salary funds (e.g. fuel allowance) last the whole of the financial year? If not, how many months are you without such funds? •• Do you access funds from any other source than your own Ministry? •• Resources are always scarce: if it were up to you, would you allocate staff and resources to their current tasks? What changes would you make if you had your way? •• What additional changes would you make if you had a bigger budget? MoA interface •• How does the MoA interact with other government agencies in the district? Of with other particular interest is interaction with other NR-based ministries (e.g. livestock, players in the forestry, water, cooperatives). sector •• How does MoA interact with non-governmental stakeholders in the district? •• For each of the above groups, explore: -- Are there clear and structured mechanisms for interaction? -- What type of interaction – exchange of information, provision of resources or services…? •• With whom do you regularly exchange information? •• Do you plan jointly for some activities with other agencies? •• Are there projects where implementation is shared? •• Do you allocate funds to other agencies, or receive funds from other agencies? •• How well does co-ordination work, in your opinion? How might it be improved?

Section I: Questions for MoA officials: These should include staff of other ministries, NGO representatives, farmers (or farmer organisa- tion representatives) and private sector stakeholders (e.g. input stockists, crop buyers) Issues Guiding questions Context – the •• What are the key agricultural activities in the district? ag sector •• What are the key challenges associated with those activities (factors constraining growth, profitability etc)? •• What new opportunities for agriculture in the district? MoA •• What do you think should be the key functions of the MoA? functions, •• What does the MoA actually do? structures •• If there is a discrepancy between these two answers, why do you think this is the and decision- case? And are the functions being performed by other players (NGOs? Private making sector?)? MoA interface •• Do you have any needs that you would wish the MoA to assist you with or issues with other that you would wish to raise with MoA? If yes, how would you present these needs/ players in the issues to the MoA? sector •• In your experience, how responsive is MoA to such needs/issues? •• Do you have any regular interaction with the MoA in the district? -- Are there clear and structured mechanisms for interaction? -- What type of interaction/exchange – information/dialogue, resources, services? -- With whom to you interact in the MoA?

Research Paper 013 | October 2009 40 www.future-agricultures.org Issues Guiding questions - How frequently does interaction occur? - What are the strengths and weaknesses of the interaction? - What could be done better? What might be the obstacles for improving interaction? • Does the MoA bring stakeholders in the district (or sector) together? If so, for what purposes? • How well does co-ordination between stakeholders within the district (or sector) work? How might it be improved? Performance • What is your opinion on performance of the MoA – highlight strengths and and evolution weaknesses? of the MoA • Has there been any signifi cant change in the MoA during the time that you have worked in the district – in terms of functions, structures, decision-making, interface with other players?

Section II: Questions for focus group discussions: Each focus group should assemble 4-7 older farmers with a long-term perspective on agriculture within the district. The groups might be organized on the basis of agro-ecological zones within the district Issues Guiding questions Performance • The discussion should explore the following issues: and evolution - distinguish time periods (since 1980? Since Independence?) according to how of the MoA well agriculture in the district was performing. Note what was doing well or badly in each period; - assess what the MoA did within the district during each period. Where the same activities are listed for each period, get respondents to assess performance of these activities in each period and give their reasons for diff erences in perfor- mance over time - ask respondents to rank periods by strength of MoA performance

Research Paper 013 | October 2009 41 www.future-agricultures.org Appendix II: A List of Stakeholders Consulted Dedza District Name Organisation Position Kamlomo T. MoA Acting DADO Mangani G. MoA Acting AGRESS Officer Dzonzi W. MoA Extension Methodologies Officer Hara O. Self Helf International Project Manager Chizimba L. Concern Universal Project Manager Kumwenda B. Concern Universal Project Manager Chizonga J. District Assembly Acting Director of Planning and Development Dzimbiri K. MoA Principal Residential Training Centre Gausi H. MoA District Animal Health and Livestock Development Officer Gondwe M. MoA Assistant Veterinary Officer Kabuli H. MoA Agribusiness Officer Nyoni M. MoA Grain Legumes Officer Njobvu T. MoA District Fisheries Officer

Thyolo District Name Organisation Position Limbani L. MoA DADO Kaphuka K. District Assembly Director of Planning and Development Nkhata G. World Vision International Capacity Development Officer Fatchi H. MoA Acting ADADO/Extension Methodologies Officer Magombo P. MoA Acting Land Resources Conservation Officer Muthali G. MoA Fisheries Assistant Joshua C. MoA Fisheries Assistant Chikanda A. MoA Assistant AEDEC Nangwale B. District Forestry Office Assistant District Forestry Officer Malobe P. Malacha Agro-dealers Partner Kamgwira Y. Oxfam Programme Coordinator

Research Paper 013 | October 2009 42 www.future-agricultures.org End notes 1 Middle altitude zone (760-1300m) covers 75% of the country’s agriculture. It is the main agricultural area with maize, tobacco, tea, groundnuts grown as the main crops where as High altitude (> 1300) includes highlands and mountains with mean temperature ranges between 10-15 degrees and with annual rainfall above 1200m. 2 These changes have mainly been as a result of political initiatives independent of the ongoing public sector reforms at the national. The shape of the MoA at the district level has been greatly been infl uenced by the decentralization policy reforms. The main infl uence in the constant reconfi guration of MoA at the national level has been the regular episodes of the hunger crisis. 3 It is important not that from the fi eldwork, the notion of coordination is multifaceted. It means diff erent things to diff erent stakeholders. This is to say that stakeholders tend to emphasise on diff erent aspects of the notion of coordination. Aspects of coordination intimated included the following: sharing of information about what diff erent actors are doing, establishing partnerships between diff erent players, organising events to make the parties meet to carry out reviews and monitoring activities etc. 4 The exercise of developing work plans usually starts before indications of funding ceilings are given out but then even when the work plans are developed on the basis of indicative budgetary ceilings, these ceilings are rarely honoured. The work plans are therefore revised in accordance with the funding fi nally made available. 5 These can be simply be described as village based AEDOs. The design is to train 10 lead farmers per village: 2 in land resources; 2 in food and nutrition; 2 in AGRESS; 2 in Agribusiness; and 2 in agricultural communications. The concept was introduced two years ago and the idea is that the lead farmers should deputize for the absence of extension workers. They interact with qualifi ed extension workers for capacity building at monthly intervals. 6 These are attached to qualifi ed extension workers for purposes of mentorship. They are then off ered scholarships to study for a diploma at the Natural Resources College (NRC) signing a bond that after completing their studies they would work for the government for a period of at least fi ve years. The experience to date is not promising. The turn over of the fi rst cohort of NRC graduates is quite high regardless of having been bonded to MoA for a period of fi ve years. Most of them are taking up lucrative positions in the NGO sector. 7 It is quite surprising that the budgeting is still done at the RDP level when the district is now the operational level following the reorganization of RDPs into district agricultural level offi ces.

Research Paper 013 | October 2009 43 www.future-agricultures.org References

Cabral, L., and Scoones I., (2006) What role for Ministries of Agriculture? Narratives and Policy Space: Draft based on Working Paper presented at the Future Agricultures Workshop: Brighton 21-22 March 2006

Department for International Development (DFID) (2005) “Growth and Poverty Reduction: The Role of Agriculture”, A DIFID Policy Paper, DFID, London

Ministry of Agriculture (MoA) (2000) “Agricultural Extension in the New Millennium: Towards Pluralistic and Demand Driven Services in Malawi”, Policy Document, Ministry of Agriculture, Lilongwe, Malawi

Ministry of Agriculture (MoA) (2005) A New Agricultural Policy, A Strategic Agenda for Addressing Economic Development and Food Security in Malawi, Ministry of Agriculture, Lilongwe, Malawi

Ministry of Agriculture (MoA) Strategic Plan 2003-2008, Ministry of Agriculture, Lilongwe, Malawi

Oxford Briefi ng Management (2007) “The Decline in Public Spending to Agriculture: Does it Matter?” OPM Briefi ng Notes, Oxford

Timer, P., (2004) Food Security and Economic Growth: An Asian Perspective. H.W Arndt Memorial Lecture. Australian National University, November 22, 2004.

This Research Paper was written by Blessings Chinsinga of the Future Agricultures Consortium. The series editor is David Hughes. Further information about this series of Working Papers at: www. future-agricultures.org

The Future Agricultures Consortium aims to encourage critical debate and policy dialogue on the future of agriculture in Africa. The Consortium is a partnership between research-based organisations across Africa and in the UK. Future Agricultures Consortium Secretariat at the University of Sussex, Brighton BN1 9RE UK T +44 (0) 1273 915670 E [email protected]

Readers are encouraged to quote or reproduce material from Future Agricultures Briefi ngs in their own publications. In return, the Future Agricultures Consortium requests due acknowledgement and a copy of the publication.

FAC appreciates the support of the UK Department for International Development (DfID)

Research Paper 013 | October 2009 www.future-agricultures.org