Chapter three

DRAFT NOT FOR CITATION

SEASONAL FINANCE FOR STAPLE CROP PRODUCTION: PROBLEMS AND POTENTIAL FOR RURAL LIVELIHOODS IN SUB SAHARAN AFRICA

Andrew Dorward *, Sibusiso Moyo #, Gerhard Coetzee #, Jonathan Kydd * and Colin Poulton * * Imperial College of Science Technology and Medicine, Imperial College at Wye Ashford, Kent TN25 5AH Email: [email protected] # Department of Agricultural Economics, Extension and Rural Development, University of Pretoria, Pretoria 0002 South Africa April 2001

This research was funded by the Department for International Development of the United Kingdom. However, the findings, interpretations and conclusion expressed in this paper are entirely those of the authors and should not be attributed to the Department for International Development, which does not guarantee their accuracy and can accept no responsibility for any consequences of their use.

Seasonal finance for staple crop production in Sub Saharan Africa

i

SEASONAL FINANCE FOR STAPLE CROP PRODUCTION: PROBLEMS AND POTENTIAL FOR RURAL LIVELIHOODS IN SUB SAHARAN AFRICA Andrew Dorward *, Sibusiso Moyo #, Gerhard Coetzee #, Jonathan Kydd * and Colin Poulton * * Imperial College of Science Technology and Medicine, Imperial College at Wye Ashford, Kent TN25 5AH # Department of Agricultural Economics, Extension and Rural Development, University of Pretoria, Pretoria 0002 South Africa April 2001 EXECUTIVE SUMMARY Background: livelihoods, poverty, agriculture and seasonal finance This paper is the first output of PRP project ‘Diverse income sources and seasonal finance for smallholder agriculture: applying a livelihoods approach in South Africa’. The project’s purpose is to develop the sustainable livelihoods (SL) framework and increase its operational utility in a range of situations, particularly in encouraging coherence in understanding and addressing problems of access to seasonal financial assets by subsistence food crop producers and delivery of financial services to subsistence food crop producers. This interim report identifies initial lessons about application of the SL framework to analyse the effects of policies, institutions and processes on access to financial assets and issues to be addressed in the design and implementation of field studies. The paper begins with a review of the broader context and rationale for the project’s focus on seasonal finance for agriculture. It briefly examines poverty in Africa from a livelihoods perspective, stressing the importance of rural poverty in sub Saharan Africa in terms of its large share of overall poverty in the region and its greater depth, incidence and rate of increase as compared with South Asia. It appears that policy reforms have tended to benefit the poor with access to public services and markets, but to have left behind those in remote areas, those growing subsistence crops, and those without work. Livelihoods analysis suggests that poverty reduction needs to involve improved access for the poor to assets, increased productivity of their assets, and reduced vulnerability. Access to assets may be increased by policy or institutional change that redistributes assets or reduces costs of access, or through increased income from existing assets. Section 3 examines the role of agriculture in rural livelihoods in sub Saharan Africa and its potential for increasing the incomes and reducing the vulnerability of the rural poor. A range of recent studies have demonstrated that rural livelihoods are more diversified than was previously recognised, and that the income share directly attributable to agriculture is generally below 50%, and falling. It is important, however, to consider the roles and patterns of diversification both within different household livelihood strategies and within rural economies (between households). While diversification between farm and non-farm activities within livelihoods is extensive and Seasonal finance for staple crop production in Sub Saharan Africa

ii

important, higher income households often have access to higher return non-farm activities, and much of the recent increase in diversification out of agriculture may be due to declining opportunities within agriculture rather than to increasing opportunities in non-farm opportunities (a process of ‘push’ rather than ‘pull’). This leads to a consideration of broader relations between farm and non-farm activities within rural economies. Theoretical and empirical arguments from the ‘linkage’ literature suggest that agriculture is likely to provide the major opportunities for getting growth going in most African rural economies, through its contribution to production of both tradables and non-tradables, but expansion in non-farm activities will be increasingly important in supporting such growth to benefit the poor. Given the importance of agriculture in poverty reducing growth, the recent performance of agriculture in sub Saharan Africa is a matter of much concern, as regards both its slow rate and its reliance on extending cultivated area rather than on intensifying production. An important element in the lack of intensification and growth is low use of inorganic fertilisers, particularly in smallholder production of food grains in more remote areas. A major problem (but not the only problem) is the difficulty that these farmers have in financing seasonal input purchases. Current delivery of rural financial services Problems in sustainable delivery of seasonal finance to smallholder farmers are well understood. A wide range of institutional models and financial products within sub Saharan Africa and beyond are currently serving, or attempting to serve, the poor’s demands for savings and loan services, with varying success. However, very few of these operate in lower density rural areas or in areas where there has not already been some agriculturally based growth in the rural economy, and virtually none are operating in the conditions faced by the majority of poor farmers in sub Saharan Africa. This is partly due to the high costs and risks in the supply of such services, but may also reflect high risks and relatively low returns for borrowers investing in agriculture. However, loan products are often structured in ways that make them particularly unsuited to seasonal lending, unless households have access to alternative cash sources which are not related to agricultural seasonality. There has been a similar lack of interest in and development of micro-insurance services and products, and there appears to be very little in the literature on financial transmission services. Livelihoods in rural South Africa are marked by high poverty incidence and a large share of national poverty, limited importance of agriculture in terms of income share, and high dependence on remittances and state transfers among some poor households. Wage labour is an important component of rural livelihoods, but its high average income share is partly a reflection of high incomes from wage earning among a relatively small but better-off section of the rural community. Despite its relatively low share in income, agriculture is practiced by a significant proportion of the rural population, as part of a ‘survivalist strategy’, to provide for at least some household food needs, or (amongst a minority of ‘commercial’ farmers) as a major source of income. Limited work on agricultural growth multipliers in South Africa also suggests that in some areas at least they are likely to be of similar magnitude as elsewhere in Africa (although building on a lower initial share of the rural economy). There appears, therefore, to be potential for a significant proportion of the rural poor to gain from increased agricultural productivity in terms of both reductions in vulnerability and increases in income. However, this will need increased and more secure access to land and to a wide range of financial services, with the latter being tailored to fit the different roles of agricultural production in the varied livelihood strategies of the rural poor. The high incidence of HIV/AIDS in South Africa has very severe implications for rural livelihoods and access to and viability of financial services, beyond its direct effects on infected individuals. High mortality and morbidity rates among the economically active, and demands on the household for care and medical expenditure, will lead to high dependency rates, labour shortages, and erosion


iii

of assets. These will have adverse effects on household productivity, incomes, ability to invest, and ability to service loans. There are many similarities between South Africa and in the rest of sub Saharan Africa as regards supply and demand in rural finance. Similarities are evident in the range of different types of financial institutions found in rural areas (although South Africa is unusual in the wide range found within one country); in the relative lack of financial services in remoter rural areas and serving poorer households and seasonal agriculture; in the lack of insurance services and common separation of savings, lending, and transmission in different organisations; in the decline of conventional supply led agricultural development organisations; in the absence of current working models to address these gaps; and in the need for wider developments in infrastructure, in property rights and in productive opportunities for the poor to invest in (although the extent and historical basis of inequity in South Africa makes this particularly important). There are, however, also very important differences between rural financial services in South Africa and in the rest of the continent. The ‘special features’ of rural financial markets in South Africa present both opportunities and threats for the development of rural financial services. Opportunities arise from very strong performance of a vibrant micro-finance sector; a wide variety of private, formal and informal financial microfinance institutions; a high degree of sophistication of commercial banks and some other institutions; the strong national infrastructure; the diversification of rural livelihoods and strong urban/rural linkages; and the policy framework provided by the Strauss Commission Threats include likely withdrawal of conventional formal private sector institutions from rural areas; “new” formal private sector financiers’ focus on wage and salary earners; slow progress in increasing access to land, infrastructure and services for rural people; the impact of AIDS; and, with the low shares of agriculture in rural livelihoods, difficulties in identifying productive investments that will lead to significant growth in the rural economy. Lessons for delivery of financial services Taken together, these opportunities and threats suggest an agenda for research and development in rural financial services. Development is needed of financial service products and of the institutions offering them. Greater attention needs to be paid to understanding demand for micro-insurance services and to the development of institutional models and micro-insurance products and technologies. Similarly, access to savings services need to be improved, in view of the current limited services and their likely decline with withdrawal of commercial banks from rural areas. One response to this scenario may be a greater emphasis on institutional models based on group or collective action. Links to high levels of technical expertise and sophistication within the South African financial sector may offer exciting opportunities here, but these need to focus on the use of appropriate technology to overcome the basic problems of maintaining client confidence and loyalty together with effective but low cost transaction and portfolio management systems for small deposits, loans and transactions. Further issues in the development of locally owned and managed institutions are the balance between savings and loan portfolios, relative importance of locally mobilised deposits and outside funds, membership of different household types, management structures and control, and the speed at which these institutions can expand to meet demand. Other institutional models are also likely to be important. There may be potential not only for backward links between group managed institutions and commercial banks, but also for forward links with targeted MFIs, for both savings and loan products. Flexible linking of non-farm income to seasonal loan repayment probably holds the most promise for increasing access to seasonal crop finance for survivalist and subsistence farmers - although there may be problems with the reliability of remittance income, and this approach is unlikely to work well where non-farm income provides a smaller share of household income. It is also unlikely to meet the seasonal needs of small-scale Seasonal finance for staple crop production in Sub Saharan Africa

iv

commercial farmers, although Latin American models may be relevant here, combining interim repayments from non-farm income with non-traditional collateral covering balloon repayments. Finally, the diversity of rural livelihoods leads to demands for a range of different financial services. Further investigation is needed of the aspirations and opportunities for and constraints to agricultural development amongst different farmer types in different areas, with regard to differentiated and targeted development of insurance, savings, transmission and loan products and services for survivalist, subsistence and small-scale commercial farmers, for example. Lessons for the SL approach In the process of the review, important strengths of the SL approach and framework were noted, in its emphasis on putting people first, recognition of the importance of different assets and of livelihood diversification, the concept of ‘livelihood strategies’, stress on vulnerability, and the importance of institutions. However, it was also apparent that the approach gives inadequate attention to diversification between households and the dynamics of market and non-market demand. Minor modifications to the existing SL framework are proposed to allow this to be addressed without any appreciable increase in complexity or loss of current strengths and emphasis. The SL approach and framework also currently lack formal consideration of the various roles of assets within livelihood strategies and of the attributes of assets in fulfilling these roles. This is problematic when considering interventions and services affecting asset investments. An existing conceptual framework is put forward to distinguish between productive, consumption, and convertible assets, and the attributes of assets in fulfilling these roles. Finally, despite the recognition of institutions within the SL framework, there is generally insufficient consideration of the nature of these institutions and of the ways that they relate to each other and to other components of the framework. Institutional arrangements, in particular, are not commonly discussed despite their being the subject of much attention in the microfinance sector. The importance of these relations needs to be given more prominence with a more explicit analytical framework to consider their strengths and weaknesses in different institutional environments, and of the necessary conditions for different forms of institutional arrangement to work for the benefit of the rural poor.


v

TABLE OF CONTENTS 1

Introduction.............................................................................................................................. 1

2

Rural poverty from a livelihoods perspective ....................................................................... 2

3

The role of agriculture in rural livelihoods in sub Saharan Africa .................................... 3 3.1

The extent of rural livelihood diversification in sub Saharan Africa................................ 4

3.2 Relations between farm and non-farm activities in sub Saharan Africa........................... 5 3.2.1 Farm: non-farm activities in diversified livelihoods ..................................................... 5 3.2.2 Farm: non-farm linkages in the rural economy ............................................................. 7 3.3

Patterns of rural livelihood diversification in sub Saharan Africa................................. 13

3.4 The role of agricultural development in rural livelihoods and poverty reduction in sub Saharan Africa............................................................................................................................. 14 3.5 4

Constraints to agricultural development in sub Saharan Africa..................................... 17

Seasonal crop finance in smallholder agriculture: issues and models.............................. 25 4.1 Rural financial markets ................................................................................................... 25 4.1.1 Particular issues in smallholder agricultural finance................................................... 26 4.1.2 Liquidity in poor rural livelihoods............................................................................... 27 4.2 Alternative means of accessing seasonal finance............................................................ 27 4.2.1 Means of ‘Saving up’................................................................................................... 27 4.2.2 Means of borrowing or ‘saving down’ ........................................................................ 28 4.2.3 Means of insuring or ‘saving through’ ........................................................................ 38 4.2.4 Transmission services.................................................................................................. 39 4.3

5

Rural livelihoods in South Africa......................................................................................... 42 5.1

6

7

Access to seasonal crop finance: conclusions................................................................. 40

An overview of rural livelihoods in South Africa ............................................................ 42

The impact of HIV/AIDS on rural livelihoods. ................................................................... 47 6.1

Introduction ..................................................................................................................... 47

6.2

History of HIV/AIDS in South Africa .............................................................................. 47

6.3

Demographic implications of HIV/AIDS ......................................................................... 48

6.4

Social implications of HIV/AIDS ..................................................................................... 48

6.5

Policy implications on strengthening the household’s coping capacity ......................... 48

6.6

Conclusion .......................................................................................................................49

Overview of financial services in South Africa ................................................................... 49 7.1

Introduction ..................................................................................................................... 49

7.2

Aggregate supply ............................................................................................................. 51


vi

7.3 Public sector .................................................................................................................... 52 7.3.1 Land Bank.................................................................................................................... 52 7.3.2 Provincial parastatals ................................................................................................... 52 7.3.3 Post Office Savings Bank ............................................................................................ 53 7.4 Private sector................................................................................................................... 53 7.4.1 NGOs ........................................................................................................................... 53 7.4.2 Village banks ............................................................................................................... 54 7.4.3 Credit Unions............................................................................................................... 55 7.4.4 Co-operatives............................................................................................................... 55 7.4.5 Commercial Banks....................................................................................................... 56 7.4.6 Retail stores ................................................................................................................. 57 7.4.7 TEBA Cash.................................................................................................................. 57 7.4.8 Private sector agricultural firms .................................................................................. 57 7.4.9 Registered small loans industry ................................................................................... 58 7.4.10 Pawn Brokers........................................................................................................... 60 7.5 The Informal sector.......................................................................................................... 60 7.5.1 Mashonisas .................................................................................................................. 60 7.5.2 Stokvels and Burial Societies ...................................................................................... 61 7.6

Financial services in South Africa: discussion................................................................ 62

7.7 Rural and small farmer finance....................................................................................... 62 7.7.1 The recent history of rural and small farmer finance in South Africa......................... 62 7.7.2 The current supply of and access to rural and small farmer finance ........................... 66 7.7.3 Conclusion ................................................................................................................... 71 8

Applying the SL Framework ................................................................................................ 73 8.1

Strengths of the SL approach........................................................................................... 74

8.2

Gaps and Weaknesses in the SL approach ...................................................................... 74

8.3 Extending the Sustainable Livelihoods Framework ........................................................ 75 8.3.1 Extending the SL Framework: Growth and Linkages ................................................. 75 8.3.2 Extending the SL Framework: the Functions of Assets .............................................. 77 8.3.3 Extending the SL Framework: Institutions and Institutional Arrangements............... 81 9

Conclusions and lessons ........................................................................................................ 85 9.1

Current poor provision of seasonal finance and its importance in pro-poor growth ..... 85

9.2

Potential for different mechanisms and ways forward/ necessary conditions ................ 85

9.3

Lessons for further work.................................................................................................. 86

9.4

The SL framework............................................................................................................ 87

References …………………………………………………………………………………………90 1

KwaZulu Natal Case study ................................................................................................... 99

2

Northern Province case study............................................................................................. 100

3

Western Cape case study..................................................................................................... 101


vii

Tables Table 1 Regional Incidence and Distribution of Poverty ............................................................. 2 Table 3.1: Potential of Farm and Non-farm Productivity Growth in Reducing Rural Poverty ......................................................................................................................................................... 16 Table 3.2

Agriculture Sector Performance by Income Level and Region (1965 – 1998)..... 21

Table 3.3 Production and Productivity Changes (% change 1979/81 to 1995/97*)................ 21 Table 4. 1 Performance Indicators of BAAC, Calpiá and CMACs ........................................ 34 Table 4.2 Features of lending technologies.................................................................................. 37 Table 4.3 Options for Seasonal Crop Financing ......................................................................... 41 Table 5.1. Features of rural livelihoods in South Africa........................................................... 43 Table 5.2 Characteristics of ‘commercialisation groups’ in KaNgwane in Mpulumanga...... 44 Table 5.3 Typology of households, central zone of the Eastern Cape....................................... 45 Table 5.4 Access to land, livestock and pension income by poor and non poor in KwaZuluNatal ................................................................................................................................................ 45 Table 5.5 Estimated multipliers, Eastern Cape .......................................................................... 46 Table 7.1 Living standard measurement categories................................................................... 50 Table 7.2. Summary of retail outreach in the micro-finance market in South Africa (1999/2000)...................................................................................................................................... 51 Table 7.3. Summary of type, number and volume of accounts (Aug. 98) ................................ 53 Table 7.4: Profitability of NGOs in South Africa compared with a Latin American Survey54 Table 7.5 Illustrative profile of farming households based on recent surveys in two provinces ......................................................................................................................................................... 69 Table 7.6: Illustrative profile small business households from recent surveys in two provinces ......................................................................................................................................................... 70 Table 8.1 Asset Attributes and Components............................................................................... 80


viii

Figures Figure 3.1 Farm: non-farm linkages in the rural economy ......................................................... 8 Figure 3.2 Farm: non-farm linkages and leakages in the rural economy ................................ 10 Figure 3.3 Linkage and Leakage Effects of Change on the Rural Poor .................................. 12 Figure 8.1. Introducing Markets into the SL Framework ......................................................... 76 Figure 8.2 Asset functions in livelihood strategies...................................................................... 78 Figure 8.3 Policies, Institutions and Processes and Sustainable Livelihoods........................... 84


ix

SEASONAL FINANCE FOR STAPLE CROP PRODUCTION: PROBLEMS AND POTENTIAL FOR RURAL LIVELIHOODS IN SUB SAHARAN AFRICA Andrew Dorward *, Sibusiso Moyo #, Gerhard Coetzee #, Jonathan Kydd * and Colin Poulton * * Imperial College of Science Technology and Medicine, Imperial College at Wye Ashford, Kent TN25 5AH # Department of Agricultural Economics, Extension and Rural Development, University of Pretoria, Pretoria 0002 South Africa April 2001

1 Introduction This paper is the first output of PRP project ‘Diverse income sources and seasonal finance for smallholder agriculture: applying a livelihoods approach in South Africa’. The project’s purpose is to further develop the sustainable rural livelihoods (SRL) framework and increase its operational utility in a range of situations, particularly in encouraging coherence in understanding and addressing problems of access to seasonal financial assets by subsistence food crop producers and delivery of financial services to subsistence food crop producers. This interim report explicitly aims to identify (a) initial lessons about application of the SRL framework to analyse the effects of policies, institutions and processes on access to financial assets and (b) issues to be addressed in the design and implementation of field studies in South Africa. In order to identify these lessons and issues, the paper begins (after this introduction) with a review of the broader context and rationale for the project’s focus on seasonal finance for agriculture. In section 2 it briefly examines poverty in Africa from a livelihoods perspective, before moving on to consider (in section 3) the role of agriculture in rural livelihoods and poverty reduction in sub Saharan Africa. Consideration of the importance of access to seasonal finance for smallholder agriculture leads on, in section 4, to examination of difficulties in its supply, and the record of responses to this in sub Saharan Africa. The paper then moves on from this initial broader review of problems in sub Saharan Africa to consider specific conditions in South Africa. Section 5 describes features of rural livelihoods in South Africa and section 6 discusses the impact of HIV/AIDS on rural livelihoods. Section 7 then provides an overview of financial services in South Africa, leading on to a discussion of rural livelihoods and finance. The final two sections draw together two sets of conclusions from the material covered in the paper. Section 8 discusses the application and contributions of the sustainable rural livelihoods (SRL) framework to the analysis in the paper, identifying strengths and weaknesses and suggesting ways of taking it forward. Section 9 then draws together lessons from the paper as a whole for the conduct of the rest of the research project. These principally concern the design and implementation of the field studies in the next phase of the project.


1

2 Rural poverty from a livelihoods perspective The extent and severity of poverty in Sub Saharan Africa, and the challenges to poverty reduction are well documented (eg World Bank, 2000a) and will not be discussed in detail here, beyond setting out the main stylised facts on the extent of and contributory factors to poverty. The World Development Report 2000 World Bank, 2000c estimates that there were approximately 290 million people living in poverty in sub Saharan Africa in 1998. This represented just under 25% of the global number of poor people. Although there are fewer poor people living in Africa as compared with South Asia, the depth and incidence of poverty in Africa is greater than in South Asia, and poverty is increasing more rapidly in sub Saharan Africa (see table 2.1). Poverty in Africa is largely a rural phenomenon: a little over 70% of the African poor are estimated to live in rural areas, where again the extent and severity of rural poverty are greater than in urban areas. There are also close links between urban and rural poverty. Many poor urban people have strong links with rural areas, and cyclical transfers between urban and rural people are increasingly important (Bryceson, D. F., 1999b). There are no clear patterns of change in poverty incidence across Africa in recent years, with increases and decreases in the incidence and severity of rural and urban poverty observed in different countries. Policy reforms appear to have benefited the poor with access to public services and markets, but to have left behind those in remote areas, those growing subsistence crops, and those without work (World Bank, 2000a p 95). Table 1 Regional Incidence and Distribution of Poverty (people living on less than $1 a day) Poverty incidence

Number of poor (millions)

1987

1998

1987

1998 % increase

26.6%

15.3%

417.5

278.3

-33.3%

excluding China

23.9%

11.3%

114.1

65.1

-42.9%

Europe and Central Asia

0.2%

5.1%

1.1

24

2081.8%

15.3%

15.6%

63.7

78.2

22.8%

4.3%

1.9%

9.3

5.5

-40.9%

South Asia

44.9%

40.0%

474.4

522

10.0%

Sub Saharan Africa

46.6%

46.3%

217.2

290.9

33.9%

East Asia & Pacific

Latin America & Caribbean Middle East and North Africa

Source: World Bank, 2000c

There are a number of contributory factors to the high levels of poverty in Africa and the poor performance of Africa in economic growth and poverty reduction as compared with other countries. These include low savings and investment rates; poor health, education and infrastructure; high dependency ratios; poor management of and access to public services; the spread of HIV/AIDS; poor economic and policy management; worsening terms of trade and continuing dependence on primary exports; poor agricultural performance; low population density; erratic rainfall and limited irrigation potential; ethnic diversity; poor governance; and conflicts. Seasonal finance for staple crop production in Sub Saharan Africa

2

From a livelihoods perspective, poverty may be characterised as low and vulnerable streams of income and of other social and material benefits accruing to individuals as an immediate result of lack of access to assets and of low and uncertain productivity of the assets that they do control. Access to assets and their productivity in turn depend upon the social, economic and institutional environment (determined by local, national and international policies, institutions and processes), the natural environment, and complementarities between asset holdings. These interactions are helpfully illustrated in the sustainable livelihoods framework (Carney, 1998), relating assets to livelihood outcomes, vulnerability and policies institutions and processes, and considering the interactions and complementarities between different types of asset (natural, human, physical, social and financial). Livelihoods analysis suggests that widespread and sustained poverty reduction needs to involve some combination of (a) improved access for the poor to a balanced set of assets, (b) increased productivity of the assets that they hold, and (c) reduced vulnerability to shocks. Key and complementary components of this include improving health and education services to expand human capital, increasing the social capital of disadvantaged and marginalised groups, expanding income opportunities, and reducing vulnerability to seasonal and other variation and shocks. This study focuses on expansion of smallholders’ access to seasonal crop finance as a contribution to improving income opportunities for the rural poor in Sub Saharan Africa (and also to reducing vulnerability), recognising that effective income gains and vulnerability reduction both require and are needed for wider gains in secure access to human and social capital. We begin from the position that expanded income opportunities and reduced income vulnerability for large numbers of people may be achieved by some combination of directly or indirectly increasing (a) secure access to assets that are both constraining large numbers of people and potentially productive for large numbers of people, and/ or (b) the productivity of these assets. Poor people’s access to assets may be increased by changes in policies, institutions or processes that redistribute assets within society, or that reduce the costs of access through subsidy or through infrastructural or institutional change (reducing risks or transaction costs, for example). Access to assets may also be improved through increased income from or productivity of existing assets, as a result of changes in technology, in access to complementary assets, in costs of inputs, or in demand for goods and services supplied by poor people (affecting the volume of demand and/or prices). The next two sections of this paper address two related questions. First we ask how expanded access to seasonal finance for staple crop production can contribute to improved livelihoods for the rural poor in sub Saharan Africa. This requires consideration of patterns of livelihood strategies and diversification, of the role of agriculture (and specifically crop production) in rural livelihoods in sub Saharan Africa, and of the role and nature of seasonal crop finance as a livelihood asset in these rural livelihoods. Second, we ask how access to seasonal crop finance can be expanded. To address this question we review specific problems in the delivery of smallholder agricultural finance before relating this to lessons from historical experience with different institutional innovations in rural financial markets. This provides a platform for subsequent consideration of specific issues in agricultural finance in South Africa. 3 The role of agriculture in rural livelihoods in sub Saharan Africa A major contribution of recent emphasis on ‘livelihoods thinking’ about rural development has been the recognition that poor rural people often engage in a highly diversified portfolio of activities (Reardon, 1998b; Bryceson, Deborah Fahy, 1999a; Barrett et al., 2000; Bryceson, Deborah Fahy, 2000; Ellis, 2000a; Reardon et al., 2000). This literature has shown the extent of


3

non-farm activities1 in rural livelihoods and has begun to examine the way that diversification out of farm activities varies between different types of households, reasons for such variation, and the potential for different types of household to benefit from growth in farm and non-farm activities. Our discussion of different patterns of rural diversification in this section addresses three important and related questions regarding the relative roles of farm and non-farm activities in poverty reduction: ♦ How are rural livelihoods diversified between different farm and non-farm activities, considering diversification both between different households (in the rural economy as a whole) and within household livelihood strategies? ♦ Is it possible to determine if growth in one sector yields greater returns in overall economic growth in rural areas? ♦ How do the poor benefit from growth in the different sectors? We address these questions by first considering evidence regarding the extent of diversification in rural livelihoods. We then examine the possible nature of linkages between farm and non-farm activities. This leads on to consideration of (1) empirical evidence regarding these linkages and (2) different patterns of diversification. We then draw conclusions from this analysis about the role of agricultural development in poverty reduction in sub-Saharan Africa. 3.1 The extent of rural livelihood diversification in sub Saharan Africa A number of recent papers have estimated the balance between farm and non-farm activities in rural livelihoods in sub Saharan Africa (Reardon, 1998b; Bryceson, Deborah Fahy, 1999a; Barrett et al., 2000; Bryceson, Deborah Fahy, 2000; Reardon et al., 2000). A common theme running through these papers is that many rural households in sub Saharan Africa, and particularly poor rural households, obtain a large part of their income, and devote a large part of their resources (especially labour) to non-farm activities. Thus Reardon, 1998b summarising results from a large number of case studies in the 1970s to 90s finds average non farm income shares of 42% in Africa (45% in East and Southern Africa, and 36% in West Africa), although this may mask wide variation in the importance of non-farm income between households with different incomes and livelihood strategies in the same area, and between households in different areas (Barrett et al., 2000). Bryceson, D. F., 1999b finds even higher non-farm income shares of 55 to 80% across a range of case studies in sub Saharan Africa, with evidence that non-farm income shares have increased dramatically in many areas in the late 1980s and in the 1990s2. An immediate conclusion that might be drawn from these data is that if rural households, and especially poor rural households, derive a large part of their incomes from non-farm activities, then actions that are targeted to benefit poor rural people should focus on expanding their opportunities in non-farm activities. Such a conclusion may resonate with policy makers’ disenchantment with the difficulties and disappointments experienced with large investments in agricultural development in Africa over the last 30 years and with enthusiasm for the successes of microfinance initiatives in improving the livelihoods of the poor in Asia and in (largely) non-farm 1

Unfortunately different authors’ definitions of non-farm and off-farm income are not always consistent. Whereas farm income generally refers to income from a household’s own farming activities, and non-farm income refers to income that is not gained from direct engagement in agricultural activities, off-farm income sometimes refers (narrowly) to agricultural wage employment but may sometimes also be used to include nonfarm income as well.

2

South Africa was an exception to the pattern of rising non-farm income shares, but these were already high – several studies indicate non-farm income shares as high as 90% for the rural poor. Seasonal finance for staple crop production in Sub Saharan Africa

4

activities in Africa. However, theory and empirical evidence on the nature of relations between farm and non-farm activities demand caution in arguing for concentration on non-farm activities as the major route for the poor to climb out of poverty. 3.2 Relations between farm and non-farm activities in sub Saharan Africa Relations between farm and non-farm activities need to be examined at two levels or scales of analysis: the relations between farm and non-farm activities within the livelihoods of people or households who engage in diversified livelihood strategies, and relations between these activities within the local economy, irrespective of the degree of integration within individual or household livelihood strategies. 3.2.1 Farm: non-farm activities in diversified livelihoods There are a number of reasons why households or individuals may find it beneficial to integrate both farm and non-farm activities within an overall livelihood strategy. Farm activities (and particularly staple crop production) are often characterised by highly seasonal demands for labour and inputs, delayed and seasonal returns from investment of these resources, uncertainty in yields, dependence on access to land (which may be limited for poorer and more marginalised groups), need for substantial seasonal capital for higher productivity, and poor market opportunities, with high price variations between and within seasons. Engagement in non-farm activities alongside farm activities may allow ♦ complementary use of labour in slack agricultural seasons; ♦ the allocation of labour to different activities according to skill, productivity and earning differentials of different household members; ♦ a better spread of income across the year to match consumption needs; ♦ opportunities for different patterns of income in farm and non-farm activities to cross-finance seasonal expenditure (for example Reardon et al., 1994a, Reardon, 1997) and larger medium to long term investments (see for example Govereh et al., 1999); ♦ diversification of risk by spreading involvement across activities with different production risk characteristics and across markets with different price risk characteristics. On the other hand, retaining some farm activities (rather than leaving agriculture completely) may ♦ allow domestic food production that reduce risks from local price hikes for food items, ♦ provide very high marginal returns to labour at particular times of year, ♦ be important as a means of retaining social and land tenure rights, and ♦ maintain skills, social networks and access to land to provide a ‘safety net’ which although offering relatively low returns represents a ‘survivalist’ strategy offering a ‘fall back’ if nonfarm income opportunities fail (e.g. if a member of the household loses a job or if economic conditions worsen and non-farm earnings decline). The benefits from such diversification are likely to be highest where risk aversion is high, where activities are highly seasonal in terms of production and resource demands, and where markets are thin and poorly developed, detracting from the benefits of specialisation and exacerbating problems of risk and seasonal shortages of labour and capital. However, Reardon et al., 2000 and Barrett et al., 2000, note that paradoxically the poorest households, who have these characteristics and the greatest need to diversify out of agriculture, also have the most difficulty in engaging in higher return non-farm activities as they lack the necessary financial, social and human capital to enter Seasonal finance for staple crop production in Sub Saharan Africa

5

these activities, and thus tend to crowd into low return, seasonal labouring activities. Reardon et al., 2000 summarising evidence from a number of studies in Africa suggest ‘a rough pattern’ (with exceptions) of ‘a positive relationship between non-farm income share (and level) and total household income and/or landholding in much of Africa’ (p272). Barrett et al., 2000 and Toulmin et al., 2000 note a common (but not universal) ‘U shaped’ relationship between the proportion of income earned and total income, with poorer and better off households with a higher proportion of off-farm income, but with very different returns to these activities. Intermediate households, on the other hand, often have lower proportions of earning from off farm activities, as they are able to gain more from farm activities than the poorest households, but are not able to engage in the highest return off farm activities open to the better off households. In this context a number of authors make the helpful if often not precise distinction between push and pull factors promoting diversification out of agriculture into non-farm activities (for example Reardon, 1998b, Ellis, 2000a, Bryceson, Deborah Fahy, 2000). ‘Push’ occurs where households or individuals engage in low return non farm activities because of inadequate returns in agriculture (as a result of chronic low productivity or lack of assets or following shocks such as drought), or where despite higher average returns from agriculture they are forced to diversify out of agriculture to overcome seasonal shortages of capital for consumption or farm investment in seasonal inputs or to reduce the overall risk from their portfolio of activities. The latter situations (where households diversify out of agriculture despite its higher returns) typically occur because of a lack of market or non-market arrangements for savings, credit and insurance, and typically are a reason for diversification by poorer households or by households located in areas with lower or more risky agricultural production. ‘Pull’ occurs where diversification into non-farm activities offers higher returns than agriculture. Bryceson, Deborah Fahy, 2000 argues that much of the diversification into non farm activities in the late 80s and 90s can be characterised as ‘push’, as smallholders have been caught between the scissors of declining profitability of and support for commercial smallholder agriculture on the one hand, and increasing needs for cash to pay for school and health fees and for increasingly expensive consumer goods on the other. As mentioned earlier, the balance between push and pull is affected by the ability or inability of the poor to engage in non-farm activities that yield higher returns to labour: thin and poorly developed markets may not only encourage diversification by the poor but may also result in barriers to entry in some markets (as a result, for example, of demands for social, physical, natural, human or financial capital where markets for these types of capital are weak or missing). These barriers may then prevent the poor from engaging in higher return non-farm activities that are less crowded and yield higher returns than activities without barriers to entry, which more people can engage in, pushing up supply and pushing down prices Reardon et al., 2000. These effects of these barriers to entry and the ways that they may vary between households and areas are illustrated by Barrett et al., 2000 who compare livelihood diversification across different income groups under different agro-socio-economic conditions. They find that in higher potential areas households with little land have a higher proportion of off farm agricultural income, but this declines with increasing farm size until at higher land access, non-farm income suddenly increases dramatically due to access to higher value activities, as ‘with denser financial systems and good market access, commerce becomes extremely competitive and the wealthy often invest in primary agriculture and tertiary sector activities associated with trade or commerce'. Under these conditions there is high correlation of cash crop and livestock income with total income. There is generally greater income diversification in more risky areas, and with greater liquidity constraints and significant transport costs in marketing, high income individuals are more heavily engaged in trade than lower and middle income households, and the correlation of non-farm income with total income increases with poor agro-ecology. Unsurprisingly, the better off earn incomes from cash crop and livestock and non-farm earnings (depending on the area) while the poor are more reliant Seasonal finance for staple crop production in Sub Saharan Africa

6

on wage labour, especially in higher potential areas (although Barrett et al., 2000 find wage labour does not account for more than 10% of income in their study sites in Cote d’Ivoire and Rwanda, the discussion by Whiteside, 2000 of off-farm employment in Malawi suggests that such figures underestimate the importance of off-farm income in the livelihoods of the poor). Four generic livelihood strategies are identified: On Farm Only requires access to land and labour and high potential areas and land/ income correlation leads to an association with wealth and access to output and input markets; the poorer On and Off Farm Only strategy is also found in high potential areas, but includes off farm agricultural wage earnings and describes vulnerable households not able to diversify out of agriculture; the Mixed with Skilled Employment Only strategy is concentrated among the better off; finally the Mixed strategy involving farm, low return and some high return non farm activities is found among poorer households and also better off households with larger families (where some often younger members engage in low return non-farm activities). This discussion of the way that opportunities, returns and ‘barriers to entry’ for on- and off- farm and non-farm activities vary between areas with different agro-ecological and economic characteristics shows that diversification also needs to be examined at a wider scale that considers determinants of local demand for the goods and services produced by low barriers to entry activities. This requires examination of wider demand and supply relations between different activities within the local economy, to which we now turn. 3.2.2 Farm: non-farm linkages in the rural economy A long-standing literature theoretical and empirical literature has examined the linkages between farm and non-farm activities within rural economies (see for a recent example and summary of the literature Delgado et al., 1998). Four types of linkage are commonly identified between agricultural and non-farm activities: direct upstream and downstream production linkages; investment linkages; and indirect consumption (or expenditure) linkages. These linkages are not unique to farm and nonfarm activities, but may arise between any sets of activities, as illustrated in figure 3.1. Starting at the left hand side of the diagram, exogenous change in policies, technologies, markets, infrastructure and capital, for example, may set off changes in prices and productivity in a rural economy. A distinction is made between tradable and non-tradable goods and services, tradable goods and services being those that may be imported or exported to or from the area3. The lower part of the diagram suggests that productivity increases in non-tradable activities will normally lead to a fall in price, as local demand will be constrained by local incomes. A fall in price of tradable or non-tradable goods and services will then lead to an increase in consumers’ real incomes if the good or service commands a high average budget share (for example staple foods in poor communities). A consumption ‘multiplier’ or linkage may then kick in as these increased real incomes lead to increased demand for local (non-tradable) goods and services and this expanded demand generates local employment opportunities. This further raises incomes, contributing to a virtuous circle multiplying the benefits from the original gains in consumer incomes from lower prices.

3

In practice the distinction between tradables and non-tradables is often not distinct, varying with (a) the scale or the boundaries of an area (the larger the area the greater the proportion of non-tradables), (b) its accessibility (the less accessible the greater the proportion of non-tradables) and (c) the comparative production costs inside and outside the area. These factors together determine the relationship between local costs on the one hand and the spread between ‘import’ and ‘export’ parity prices on the other. Although these terms are often associated with international trade, they are equally applicable to intranational trade between different districts or between rural and urban areas. Seasonal finance for staple crop production in Sub Saharan Africa

7

Figure 3.1 Farm: non-farm linkages in the rural economy

Increased productivity of farm & non-farm tradables Policies (inc. taxes, subsidies), Technology, Institutions, Markets (inputs & outputs, national & international prices, price transmission), Infrastructure, Capital

Increased revenues from farm & non-farm tradables Producer benefits

Input & processing demands PRODUCTION LINKAGES

Higher prices of tradables

Increased employment

Local supply response

Producer benefits

CONSUMPTION

Higher volumes

LINKAGES

Lower prices of tradables

Increased productivity& lower prices of non-tradables

Lower prices of goods & services with high budget shares

Consumer benefits

Demand for local nontradable goods and services

Increased earnings / real wages

SAVINGS & INVESTMENT LINKAGES Seasonal finance for staple crop production in Sub Saharan Africa

8

The extent of these gains, however, is limited by ‘leakages’ which are shown in figure 3.2 (as additions to the right hand side of figure 3.1). Thus if local consumers use their extra income to buy tradables (goods and services that are imported from other areas or that are produced locally but would otherwise be exported) then there is reduced stimulus to local demand. Even if there is stimulus to local demand, if local producers cannot respond to the increased demand (due to limited supply of labour or capital, or poor market development and high transaction costs), there will be inflationary pressure on prices, off-setting consumers’ increased incomes. Finally, even if there is a local supply response, if production systems are capital intensive, import intensive or provide returns to a only a limited number of local people, then there will be reduced gains from increased local employment and earnings4. The effects on producers of the initial increase in non-tradable productivity are more mixed. Lower prices may largely off-set producers’ gains from higher productivity5, unless demand is relatively elastic or cost reductions or changes in technology are sufficient to allow expanded labour demands and/or entry of new (poorer) producers into the market. Higher volumes produced and marketed may also provide increased employment opportunities through upstream demand for inputs or downstream processing opportunities, depending upon the extent of such demand associated with the particular non-tradable whose productivity has increased. Production linkage benefits to labour incomes may however also be limited by capital or import intensive production systems, capture of income by elites, or an inelastic supply response. Lower prices for tradables do not benefit existing net producers of these tradables. The negative effects of lower tradable prices on producers are not explicitly shown in figure 3.1, but may be deduced from the positive effects of higher prices. Thus lower prices may depress production and upstream and downstream activities, with associated losses of producer income imposing a drag (in the right hand side of figure 3.1) on the positive consumption linkages that may arise from lower prices and increased consumer incomes. Higher prices for tradables have opposite positive effects on producer incomes, similar to the positive effects of increased productivity in tradable activities. The latter, however, does not have the same negative effects on consumers as do price increases (again, figure 3.1 does not explicitly show consumers’ losses from the negative effects of higher tradables prices). Finally, we consider savings and investment linkages which may arise where increased real incomes allow increased savings and investment in capital, in turn reducing vulnerability and increasing both the productivity of local activities and the potential elasticity of supply responses crucial to consumption linkages (the latter linkage is shown later in figure 3.3 but not in figures 3.1 and 3.2). ‘Leakages’ arise if the returns to local savings and investment are very low, due to lack of secure productive investment opportunities or due to lack of local financial markets linking savers with productive investment opportunities. They may also arise if there are effective financial markets linking the local economy with other economies, so that either local activities are already able to access outside sources of capital or locally generated capital is invested outside the area 6 All these linkages are affected by transaction costs. High transaction costs will generally depress supply responses to increased demand for tradables and non-tradables. Differences in transaction costs for different suppliers may also affect the relative importance of tradable and non-tradable supply responses (Machethe et al., 1997), and hence the extent of leakages.

4

Taxes and savings can be further leakages, not shown in figure 3.2.

5

Where productivity increases result from some form of innovation, early adopters are likely to gain from higher productivity before more widespread adoption lowers prices.

6

Under these latter circumstances finance is ‘tradable’. Seasonal finance for staple crop production in Sub Saharan Africa

9

Figure 3.2 Farm: non-farm linkages and leakages in the rural economy

CAPITAL / IMPORT INTENSIVE PRODUCTION

INELASTIC SUPPLY OF NON-TRADABLES

INEQUITABLE ASSETS & INCOMES Increased productivity of farm & non-farm tradables Policies (inc. taxes, subsidies), Technology, Institutions, Markets (inputs & outputs, national & international prices, price transmission), Infrastructure, Capital

Increased revenues from farm & non-farm tradables

Input & processing demands

Higher prices of tradables Higher volumes

Producer benefits

Increased employment

PRODUCTION LINKAGES

Producer benefits

Local supply response

CONSUMPTION LINKAGES

Lower prices of tradables

Increased productivity& lower prices of non-tradables

Lower prices of goods & services with high budget shares

Consumer benefits

Demand for local non-tradable goods and services

Increased earnings / real wages

SAVINGS &INVESTMENT LINKAGES INTEGRATION WITH WIDER FINANCIAL MARKETS Seasonal finance for staple crop production in Sub Saharan Africa

DEMAND FOR TRADABLES

LOW RETURNS TO LOCAL INVESTMENT 10

Two further types of linkage may arise from growth in production of tradables (Govereh et al., 1999). First, increasing trade flows may lead to improvements in a range of services, particularly in communications (telecommunications and transport services for example) with both investment in improved infrastructure and greater demand for and frequency of services, with greater volumes allowing lower unit costs7. These linkages might be described in terms of economies of scope within the local economy. There may also be economies of scope within particular livelihoods, with, for example, the purchase of farm equipment for production of tradables also being used in the production of non-tradables. An important conclusion from this analysis is that the effects of particular changes on a rural economy and on poor people within it depend crucially upon the nature of the change, on the structure of the local economy, and on different poor peoples’ places within it. This is illustrated by the discussion above of the different effects of price changes and of changes in productivity of nontradables on the poor as producers and labourers on the one hand and as consumers on the other. This point is further emphasised by figure 3.3, which draws out some of the ambiguities in figures 3.1 and 3.2. This figure explores the possible effects of three different types of exogenous change (changes in tradable and non-tradable productivity and in tradable prices, shown on the left of the figure) on poor consumers and poor producers/ suppliers of labour. The nature and scale of possible direct and indirect effects of these changes on the incomes and expenditures of different categories of the poor are then determined by the answers to a set of questions about the characteristics of the goods and services subject to the initial price or productivity change, about their local demand characteristics, about their local production characteristics, about the income elasticities for nontradable goods and services, about the production and supply characteristics of these goods and services and about financial markets. This analysis has not explicitly focussed on an examination of linkages between farm and non-farm activities, but it is highly relevant to the consideration of such linkages. First, it suggests that there can be strong linkages between different activities, and these linkages may work in a number of different ways, and affect different activities and categories of the poor in different ways. Second, it suggests that the nature and strength of these linkages will depend not so much upon general characteristics of farm and non-farm activities but upon the specific characteristics of the farm and non farm activities in question, characteristics which depend upon the specific context of the particular rural economy under consideration, and upon the structure of that economy and of demand and production and the livelihoods of the poor within it. Thus the role that different farm and non-farm activities play in the local economy, the extent to which the different goods and service that they produce are tradable and the importance of different leakages will vary between different areas with different resources, with different communications and outside links, and with different historical patterns of growth and development. Third, it provides a framework for investigation of the linkages and likely contributions of different types of farm and non-farm activities in different rural situations and communities.

7

These greater trade and information flows will also increase the proportion of tradables in the economy. This will increase the consumption leakages and reduce consumption linkages, and may also cause previous producers of non-tradables (for example traditional goods) to lose market share to manufactured items imported into the area. These negative effects should be offset (with time normally to a greater extent) by gains to consumers from these cheaper goods and by new opportunities for expanded tradable production where the area has competitive advantage. Seasonal finance for staple crop production in Sub Saharan Africa

11

Figure 3.3 Linkage and Leakage Effects of Change on the Rural Poor Savings/ investment linkages

Elastic supply

YE

YE

Growth (decline) in non-tradable productivity

High marginal returns to local savings & investment & little wider financial integration?

NO YE

Increase (fall) in demand for goods & services with high marginal budget

Inelastic demand

Strong linkages, benefit (loss) to

Inflationary (deflationary) linkages, damage poor consumers

YE Tradable? NO

YE Reduced (increased)

Weak linkages, by-pass the poor

High average budget share?

Consumer Increase (fall) in real wages / YES incomes

Producers / labour

YES

High local labour

YES

NO Fall (rise) in tradable

Strong linkages, benefit (loss) to poor producers/ labour

Negativ Growth (decline) in tradable Seasonal finance for staple crop production in Sub Saharan Africa

Weak linkages, to poor producers/ labour, increasing inequity

YES Upstream/ downstream

12

These conclusions are borne out by (and were developed from) studies estimating agricultural growth multipliers in different parts of the world, which show that linkages can be strong, and that they do vary in the way described above. These studies provide estimates of agricultural growth multipliers that vary from around 1.5 to over 2.0 (Reardon, 1998b,Delgado et al., 1998) 8. Although earlier studies suggested that agricultural growth multipliers were higher in Asia than in Africa (due to more inelastic supply of non-tradables in Africa and to higher proportions of extra income being spent on agricultural products, which were assumed to be tradables and therefore represented leakages), Delgado et al., 1998 argue that many agricultural products in Africa would be more appropriately considered as non-tradables (or more precisely semi-tradables), and this raises estimates of multipliers in Africa9. Studies across Africa and Asia generally also show that consumption linkages are more important than production linkages. Thus in Burkina Faso, Niger, Senegal and Zambia estimates using data from the mid to late 1980s found consumption linkages accounted for between 50% (in Senegal) and 98% (in Zambia) of overall multipliers calculated (Delgado et al., 1998)10. Bautista and Thomas, 1999 in a study of agricultural growth linkages in Zimbabwe make the important point that different patterns of agricultural growth will have different linkage and multiplier effects, and will affect different categories of poor people differently. They simulate different patterns of growth and find that although productivity growth in smallholder agriculture produces the greatest benefits for the large numbers of poor people working in smallholder agriculture and in communal rural areas and generates the highest multiplier effects (1.9 as compared with 1.5 to 1.6 from growth in large scale commercial farms and 1.5 from labour intensive light manufacturing) poor workers on large scale commercial farms and the urban poor benefit least from this pattern. 3.3 Patterns of rural livelihood diversification in sub Saharan Africa If, as argued above, the importance of the linkages between farm and non-farm activities varies across different rural areas with their agro-ecology, with the nature of the rural economy, and with its place in and links to the broader national economy, then we might expect this to be related to different patterns of rural diversification in different areas. Reardon, 1998b identifies different

8

A multiplier of 1.5 indicates that $1.00 of extra income from production of agricultural tradables stimulates further income growth of $0.5. These estimates are subject to error due to implicit assumptions in the estimation methods that the supply of non-tradables is elastic (leading to an overestimate of the multiplier), and due to failure to allow for the dynamic effects of savings and investment (leading to an underestimate of the multiplier). Allowing for the effects of supply inelasticity in production of nontradables may reduce estimates of multipliers by around 10% in Asia and by 30% in Africa (Haggblade et al., 1991). 9

It also means that agricultural growth has potential for contributing to wider growth both by growth in tradable production and by encouraging a more elastic supply response for non-tradable products. 10

The multiplier estimates cited above do not necessarily assume high elasticity of supply and although they may overestimate consumption linkages, they may also underestimate downstream production linkages and investment linkages. Overall, multipliers are likely to remain significant. Some authors, however, are sceptical of the scale of these multipliers or of the relationships postulated between farm and non-farm activities (see for example comments by Harriss, 1987 and reply by Hazell and Slade, 1987, and a summary by Ellis, 2000b). Wobst, 2000 also argues, on the basis of comparison of partial and full CGE models of Tanzania, that partial equilibrium and multiplier models tend to over-estimate multiplier effects and benefits to rural farmers because these models operate on fixed price assumptions whereas in fact increases in agricultural productivity change the terms of trade for agriculture. However, the standard CGE models find that although net gains for farmers are much lower than in the partial equilibrium model, estimated gains for non-farm households are higher. These findings clearly depend upon the balance between tradables and non-tradables and the operation of non-agricultural labour markets. Seasonal finance for staple crop production in Sub Saharan Africa

13

‘stages of rural non-farm transformation’ and characterises Africa (with South Asia) as being predominantly in stage 1, with most farm: non-farm links being located within rural areas themselves and involving direct (upstream and downstream) production linkages and expenditure linkages. He cites a study in the West African Sahel where more than 80% of local non farm activity involved production linkages with agriculture (Reardon et al., 1994b).11 However, Reardon et al., 2000 also notes that there appear to be differences in the relationship between household incomes on the one hand and the share and level of non-farm income on the other. He postulates that in Africa, where a positive relationship between total incomes and non-farm income share predominates, this arises as a result of a lack of ‘low barrier to entry’ labour intensive earning opportunities, with consequent crowding and low returns as discussed earlier in the context of ‘push’ factors in diversification. Bryceson, D. F., 1999b develops a related categorisation of three broad and often over-lapping diversification ‘complexes’ within SSA. ‘Complex A, local services’ describes diversification in remoter areas, where non-farm activities tend to be the supply of local services and local handicrafts. Remoteness restricts access to a wider market for sale of local goods and services, but also restricts competition from import of goods and services from urban areas. ‘Complex B – trade’ involves significant non-farm income from trading activities and labour migration linking rural and urban areas. In ‘complex C – transfer payments’ these rural-urban links are taken further in areas with historical traditions of labour out-migration and highly mobile populations, to the extent that pensions and remittances become a major source of rural income. Bryceson stresses the importance of investment and consumption linkages. The latter are of particular importance for ‘complex A’ diversification as well as for complex B, where agricultural income is an important determinant of demand for local services and for local purchase of traded goods. Investment linkages are often more important for complex B and, to a lesser extent, complex C, and work the other way, with higher non-farm earnings supporting investment in agriculture where this can offer high or secure returns to larger investments. 3.4

The role of agricultural development in rural livelihoods and poverty reduction in sub Saharan Africa This discussion of the linkages between farm and non-farm activities and of patterns of diversification within households’ livelihood strategies and across rural economies suggests that consideration of the relative roles of farm and non-farm activities in poverty reduction needs to examine the nature of the contribution that different activities can make to growth in the rural economy. The linkage literature reviewed earlier shows that irrespective of debate about the scale of multipliers between farm and non-farm activities and income, two sources of growth within rural

11

Reardon’s ‘first stage of rural non-farm transformation’ can be contrasted with the second and third stages, which predominate in Latin America and South East Asia respectively. These show increasing formal involvement in manufacturing and non-agricultural services and in medium and larger scale agro-industrial activities, with increasing links to urban activities (with commuting to urban areas and sub-contracting of rural businesses by urban or foreign businesses). Wiggins, 2001 suggests that rural areas, and particularly more remote rural areas, suffer from a wide range of economic disadvantages in access to information and input and output markets, and urban areas therefore have a strong comparative advantage for many economic activities. He argues that in the longer run in ‘the deep countryside’ only activities with a strong natural resource base (for example agriculture and in some areas tourism and recreation), local processing of agricultural products, and non-tradable services for the rural population will survive. Exceptions to this may arise where rural labour costs are much lower than urban costs, and there are good communications and road networks in rural areas encouraging labour intensive industries to locate in rural areas (Taiwan is presented as an example here). Seasonal finance for staple crop production in Sub Saharan Africa

14

areas may be identified12: growth in production of tradables (increasing local incomes) and growth in production of non-tradables (increasing local incomes and reducing local prices). Ignoring for the moment consideration of how these two types of growth may arise, however, different conditions are required for them to be effective in reducing poverty. Thus for the first, the tradable must either be widely produced with significant labour demands generating broadly based earnings, so that the poor benefit from the multiplier effects of growth induced in the local economy, or produced by the poor themselves as hired labour or self-employed. Growth and poverty reduction through increased productivity of non-tradables will be effective as a primary source of growth and poverty reduction where the non-tradable good or service is widely consumed (i.e. has a high average budget share), either by the poor themselves or by a large nonpoor population, so that the fall in price (as a result of extra productivity) either directly benefits poor net consumers or benefits a substantial non-poor population with this leading to increased incomes for the poor from the multiplier effects of such growth13. There may also be direct benefits to poor producers if the productivity change is associated with sufficiently large falls in costs or with institutional, market or technical changes that reduce barriers to entry and allow the poor to enter the market. Growth and poverty reduction through increased productivity of non-tradables may also be important as a secondary growth process supporting one of the two primary processes above, responding to increased local demand for non-tradables and benefiting poor producers with expanded income opportunities and poor consumers with stable prices despite increased demand. Here the non-tradable must have a high marginal budget share. The potential for farm or non-farm activities to stimulate and sustain growth in particular areas may then be evaluated against their ability to meet the conditions specified above. Examining first the potential for growth in tradables, in areas characterised by Reardon, 1998b as ‘stage 1’ or by Bryceson, D. F., 1999b as complex A – local services’ there are unlikely to be many tradable nonfarm activities apart from mining that offer broadly based employment opportunities14. Only as links with urban areas develop will opportunities for other non-farm tradable activities develop, as described in Reardon’s stage 2 and stage 3 – it is noticeable that Bryceson’s complex B and complex C describe the importance of trading links and remittances, but not of significant local employment-generating non-farm tradable activities. Even where such activities do develop, Reardon’s analysis suggests that these will often be ‘high barrier to entry’ activities, limiting the direct opportunities for employment and income benefits for the poor, as well as the extent to which wider income gains among the non-poor may lead to employment opportunities for the poor through consumption linkages. Farm activities, on the other hand, do offer opportunities for broadly based expansion in tradable activities (whether cash crops or tradable food crops), with direct and indirect employment and income opportunities for the poor, again depending upon barriers to entry associated with, for example, the nature of the crop, marketing systems, access to land, etc..15

12

Migrant labour and remittances they may be considered a form of tradable, exporting labour to bring extra income into an area. 13

Autarkic subsistence producers will also benefit directly through release of resources to other activities.

14

In some areas tourism and crafts may also offer opportunities for non-farm tradable activities.

15

Even here the poor are unlikely to gain much directly as self-employed producers of tradable agricultural commodities, given their limited access to land and capital and relatively low on-farm incomes. However, there is often considerable potential for them to benefit directly from increased labour demand from significant numbers of less poor farmers producing tradables, and indirectly through increased demand for non-tradables from these farmers. The challenge (addressed later in this paper) is then to improve the access of less poor farmers to the Seasonal finance for staple crop production in Sub Saharan Africa

15

Turning to consider the potential for growth in non-tradables, scope for this as an engine or primary source of growth and poverty reduction may be limited if there are few non-tradable commodities or services with a high average budget share. High average budget shares for food crops in rural areas in Africa (Delgado et al., 1998) suggest that if these food crops are tradables then the nontradable budget share is small. If however they are non-tradables, then the greatest potential for direct growth benefits will arise from growth in farm activities rather than in non-farm activities, provided that there are more winners (subsistence producers and net consumers) than there are losers (net producer employers and employees). There may, however, be greater scope for growth in non-tradable non-farm activities, either in supporting secondary growth processes or where greatly reduced costs or barriers to entry allow an expanded market, with benefits to consumers and to new (poor) producers. In the support of secondary growth processes, a high marginal rather than average budget share is important, but in the second situation (where greatly reduced costs or barriers to entry allow an expanded market) it is a high average budget share that is important. Some local non-tradable services may have relatively high marginal budget shares, but as argued above they will not generally have such high average budget shares (although of course there will be exceptions to this). These characteristics may be shared by livestock and horticultural products in the farm sector, but staple food crops, on the other hand, will tend to have high average budget shares but low marginal budget shares. The relative merits of growth in farm and non-farm tradable activities in supporting poverty reducing growth in a particular situation may therefore depend upon (a) their role in the economic context (supporting secondary growth or driving it); (b) their marginal and average budget shares amongst poor consumers and the local population as a whole; and (c) the respective barriers of entry, with lower barriers of entry both allowing more direct employment benefits for the poor and greater elasticity of supply, and thus lower ‘leakage’. Table 3.1: Potential of Farm and Non-farm Productivity Growth in Reducing Rural Poverty Tradable

Non tradable

Direct gains if high labour content by poor producers or high upstream / downstream linkages have high labour content by poor producers

Direct gains if high average budget share for poor consumers

Indirect gains if high labour content by non-poor and poor benefit from expenditure linkages

Elasticity of supply and low barriers to entry important for goods and services with high marginal budget shares, to support expenditure linkages

Farm activities

Indirect gains if high average budget share for non-poor consumers and poor benefit from expenditure linkages

skills, capital, inputs and output markets to allow them to respond to opportunities in production of farm tradables. Seasonal finance for staple crop production in Sub Saharan Africa

16

Non farm activities

Apart from mining and other NR activities, unlikely without good communications and strong urban or export markets (eg. Reardon stage 2 or 3 of rural non-farm transformation)

Unlikely to have high average budget shares for poor consumers Elasticity of supply and low barriers to entry important for goods and services with high marginal budget shares, to support expenditure linkages

These arguments are summarised in in Table 3.1. A broad conclusion from this, to which there will be significant exceptions, is that in many poorer rural areas in sub-Saharan Africa increasing productivity of farm activities will often have greater potential for stimulating poverty reducing growth, either through direct income benefits in the production of tradables or through consumer benefits in the production of non-tradables. The former will be enhanced if the activities have low barriers to entry and high demand for labour. Increasing productivity of non-tradable farm activities may also have potential for supporting secondary, linkage dependent poverty reducing growth, particularly if the activities have low barriers to entry but increased productivity of non-farm activities is likely to have greater poverty reducing benefits in supporting secondary, linkage dependent poverty reducing growth, again particularly if the activities have low barriers to entry. 3.5 Constraints to agricultural development in sub Saharan Africa The previous section compared in general terms the roles of improved productivity of farm and non-farm activities in stimulating poverty reducing growth in rural areas in Sub Saharan Africa. The importance of agriculture was stressed for poorer households where agriculture is directly responsible for a significant part of their livelihood, where agriculture provides important stability and security even though it is directly responsible for only a small part of their livelihood, and where, although agriculture does not directly contribute to their livelihood, agriculture is important (or potentially important) to the local economy and hence its growth stimulates demand for nonagricultural products and services. We now turn to consider how agricultural productivity might be improved. Despite long standing recognition of the importance of agricultural development, growth in agricultural production has been disappointing. This is illustrated by table 3.2, which presents summary statistics for various agricultural indicators. These clearly show slower rates of productivity growth in sub Saharan Africa as compared with other regions, although growth rates in the different regions appear to have converged somewhat in the 1990s. In Sub Saharan Africa very low rates of growth in the 1970s were followed by increases in the 1980s and 1990s, but per capita growth has been very low or negative over much of the period: thus Sub Saharan Africa is the only region with agriculture growing at a rate below overall population growth from 1965-1998, and at a lower rate than growth in the agricultural labour force from 1980-1998. The data on which these low estimated agricultural growth rates for SSA are calculated, however, contain a number of inconsistencies (for example estimates of agriculture’s shares of GDP in Africa have varied and tend to be (implausibly?) low, as shown in table 2.1) and have been criticised as not reflecting dynamic growth that exists in Sub Saharan African agriculture (Wiggins, 1995) and of being over sensitive to the effects of price changes and currency devaluations (e.g. Block, 1995). Nevertheless, the general picture of low or negative per capita growth in agriculture in much of Sub Saharan Africa over the last 30 years is supported by the high incidence and severity of rural poverty in Sub Saharan Africa as compared with other regions, widespread reports of agricultural stagnation (e.g. Bryceson, Deborah Fahy, 2000), and data on fertiliser use and yields. Seasonal finance for staple crop production in Sub Saharan Africa

17

Table 3.3 presents information on various elements of agricultural productivity that suggest that Sub-Saharan Africa is achieving its agricultural growth largely through a different process from that found in other regions and stands out for increasing its area under cereals dramatically at the expense of other crops16, whereas in other regions the area under cereals has either declined or increased only slightly, and less than the expansion of other crops. Sub Saharan Africa’s increased cereal area is accompanied by a slight fall in overall fertiliser consumption17, a larger fall in rate of fertiliser use, and only a small rise in cereal yields. The area of irrigated land also shows only a small percentage rise18. As a result, whereas other regions have achieved 80% or more of their increased cereal production from yield increases, in SSA more than 70% of increased cereal production is from area increases. This pattern of growth presents a major problem as it is widely recognised that a process of ‘sustainable intensification’ is needed (e.g. World Bank, 1997, Reardon, 1998a,Reardon et al., 1999), to avoid continued ‘soil mining’ and extension of cultivation onto increasingly fragile and vulnerable land. ‘Sustainable intensification’ requires increased use of purchased inputs, especially seeds and inorganic fertilisers (to supplement low external input organic sources plant nutrients). We therefore now turn to consider in more detail information on the use of inorganic fertilisers in Africa, and constraints on their greater use. There are two principle sources of information on rates of fertiliser use in Africa: official national statistics on fertiliser imports and purchases and more piecemeal data from local studies. The former often contain discrepancies that cast doubt on their reliability, while the latter are by nature patchy and difficult to generalise. Nevertheless, national statistics and survey findings paint the same general picture of low rates fertiliser use in SSA as compared with other parts of the world, growth in fertiliser use in the 1970s and 80s stagnating in the 1990s, high variability between and within countries, and discrepancies between rates of application on cash and food crops. Naseem and Kelly, 1999 report that from 1970 to 1995, fertiliser dosage rose from 3.3 kilograms per hectare to 9.9 kilograms per hectare in SSSA, but the average of 8.9 kilograms per hectare for 1991-95 is very low when compared with rates in Latin America (54kg/ha), South Asia (80 kg/ha) and South-East Asia (86.9kg/ha) 19. Furthermore, growth rates in total fertiliser use in SSA have been low (6.9% per annum as compared with 11% per annum for South East Asia, 22% for South Asia and 5% for Latin America) and this is (a) far below the 8 to 10% annual growth rate needed to increase cereal production to a level ensuring food security or the 18% annual growth rate needed to prevent soil mining (Larson and Frisvold, 1996) and (b) has declined in the 1990s, to 2.6% from 1991-9520. World Bank, 2000b paints a more pessimistic picture (see table 3.2), with overall

16

The data in table 3.3 are consistent with earlier discussion of ‘survivalist’ objectives in agriculture and with Bryceson’s analysis of the way that many poorer and more remote African smallholders have been ‘pushed’ out of agriculture. 17

These figures for fertiliser consumption need to be treated with caution: evidence is presented later of gross underestimates for fertiliser use in Ghana, for example. 18

Although this is similar to the percentage increase in irrigated land in the East Asia and Pacific region, SubSaharan Africa’s increase is from a very low base (only 4% of crop land being irrigated, compared with 36% in the East Asia and Pacific region) 19

However, these figures mask very high variability. From 1991 to 1995, for example, four countries (Ethiopia Kenya Nigeria and Zimbabwe) used 60 per cent of all SSA fertiliser (excluding South Africa) (Naseem and Kelly, 1999). Larson and Frisvold, 1996 cite average fertiliser use in Kenya as 48kg/ha, comparable to fertiliser use in South Asia (58kg/ha) and South East Asia (62 kg/ha).


18

fertiliser use in low and middle income countries in SSA declining by 2% over the period 1979/81 to 1995/97, while rates of use per hectare declined by 18%. These figures contrast sharply with those from other regions. More detailed studies of specific countries and of areas within them reach similar conclusions (for example Kherrallah et al., 2000, Jayne, T.S. et al., 1997, Bryceson, D. F., 1999b, Poulton, 1998, Al-Hassan et al., 1999, Reardon et al., 1999, Scoones and Toulmin, 1999, Govereh et al., 1999, Reardon et al., 1997 Brinn et al., 1999, Carr, 1997: that fertiliser use has stagnated in many countries in the 1990s, but that there are also some significant exceptions to this, with fertiliser use increasing for some countries, crops or regions. Many of these studies report that particular difficulties are found with the use of fertilisers on food crops, especially in remote areas (Kherrallah et al., 2000, Scoones and Toulmin, 1999, Govereh et al., 1999, Bryceson, D. F., 1999b, Poulton, 1998, Naseem and Kelly, 1999, Risopoulos et al., 1998), despite the great importance of maize as a food crop in SSA (Byerlee and Heisey, 1996) and its high yield response to fertiliser (Naseem and Kelly, 1999).

20

They note, however, that is the result of declines in consumption in 14 countries (often associated with periods of civil unrest or the early period of structural adjustment programmes), while most countries have increased their consumption levels, (with six countries increasing it more than tenfold). They also note that many of the advances appear to be periods of recovery after earlier declines. Seasonal finance for staple crop production in Sub Saharan Africa

19

Table 3.2

Agriculture Sector Performance by Income Level and Region (1965 – 1998) AGRICULTURAL GROWTH (average annual % value added)

POPULATION GROWTH AGRICULTURA AGRICULTURE SHARE IN L LABOUR GDP (average annual %) (annual % growth) (value added % gdp) Total

Agric.

Agric.

1965-98 1980-90 1990-98 1980-98 1965-98 1980-90 1990-97

1980-98

1970

1980

1998

Low & Middle Income Countries East Asia & Pacific

3.6

4.4

3.5

4.0

1.8

1.2

0.3

..

33

24

15

Latin America and Carribean

2.6

2.1

2.2

2.1

2.1

-0.7

-0.6

..

13

10

8

Middle East and North Africa

4.2

5.5

2.5

4.2

2.8

..

..

2.2

13

10

14

South Asia

2.9

3.2

3.7

3.4

2.2

1.3

1.0

1.5

43

37

28

Sub Saharan Africa

1.9

2.5

2.4

2.5

2.7

2.4

1.7

2.8

21

18

17

Source: World Bank, 2000b, FAO, 2000

Table 3.3 Production and Productivity Changes (% change 1979/81 to 1995/97*) ARABLE & IRRIGATED PERMANENT CROP LAND LAND Changes in (%) Area (ha) Area per ha capita

FERTILISER USE total kg

kg/ha

CEREAL PRODUCTION Area (ha)

OTHER CROPS

Yield (kg/ha) Yield share in prod'n increase

Area (ha)

Low & Middle Income Countries East Asia & Pacific

27%

0%

25%

141%

100%

3%

29%

91%

109%

Latin America and Carribean

16%

-16%

35%

46%

26%

-1%

33%

100%

25%

Middle East and North Africa

23%

-28%

69%

85%

57%

13%

53%

80%

32%

1%

-30%

40%

157%

157%

-1%

39%

100%

5%

20%

-22%

26%

-2%

-18%

71%

28%

28%

-5%

South Asia Sub Saharan Africa * 1996/98 for some variables Source: World Bank, 2000b

21

There are a number of related reasons for the low fertiliser dosages, slow rates of growth in fertiliser usage, high variability between countries and crops, and stagnation and decline in fertiliser use in SSA the 1990s. These may be broadly considered in terms of supply and demand constraints. Supply depends upon the capacity, efficiency and resources of suppliers, their ability to access fertiliser imports, and the incentives for private firms to enter the industry. Similarly demand depends upon farmers’ resources and their capacity to finance input purchases, and the profitability of fertiliser use. This in turn depends upon input and output prices (economic and institutional variables) and yield responses to fertiliser (determined by agro-ecology, technology and complementary inputs).21 In the past, input supply was commonly managed by parastatals and was linked in with pan territorial pricing and monopsonistic crop purchasing. The problems of such parastatals, their increasing inefficiency and ineffectiveness, and the growing fiscal burden they imposed are widely documented. However, liberalisation of the input supply system has not generally lead to an influx of private traders selling inputs to smallholders in marginal areas: such traders are often severely constrained by problems in accessing fertiliser imports and credit for working capital, and face high credit and distribution costs (with poor transport systems and low volumes) with uncertain returns (due to policy uncertainties, variable demand and difficulties in communicating with poor, dispersed farmers) (Bryceson, Deborah Fahy, 1999a, Omamo and Mose, 2001; Reardon et al., 1999; Speirs and Olsen, 1992; Kelly et al., 1999; Larson and Frisvold, 1996; Naseem and Kelly, 1999; Howard et al., 1999; Brinn et al., 1999; Gordon and Goodland, 2000; Karanja et al., 1998; Jones, S., 1998). Low and variable demand is one reason for low profitability and the lack of private sector investment in inorganic fertiliser supplies, but this is itself partly caused by the poor supply system: farmers cannot rely on and wait for uncertain deliveries, often in appropriate package sizes and formulations, and with few (or no) alternative suppliers may be in a weak position to negotiate prices if alternative supplies are not available locally. In addition, uncertain output prices and output marketing opportunities, and relatively higher input prices (as a result of devaluations and subsidy removals) undermine the underlying profitability of fertiliser application (Bryceson, Deborah Fahy, 1999a; Kherrallah et al., 2000; Jayne, T.S. et al., 1997; Reardon et al., 1999; Brinn et al., 1999; Kelly, 1999 #134]; Naseem and Kelly, 1999; Larson and Frisvold, 1996; Poulton, 1998, Carr, 1997, although Scoones and Toulmin, 1999, citing data up to 1993, suggest that output prices have also increased in some countries and in these cases input/output price ratios and hence profitability have not dramatically altered). These effects are most severe for poorer farmers (Dorward, 1999), for more remote areas (where transport and communications costs increase input prices and market uncertainty) and for areas with lower rainfall and poorer soils (with lower or more risky yield responses to fertiliser). In some areas of the Sahel, for example, inorganic fertiliser applications are highly risky investments for farmers, particularly if they are financed on credit. This highlights a widely reported and equally fundamental constraint on input use: farmers’ lack of liquidity to finance input purchases. This particularly affects maize and other food crops following market liberalisation, as output prices have tended to fall while input prices have risen (due to devaluations and subsidy removal), with consequent ‘decapitalisation’ of farmers Al-Hassan et al., 199922, while parastatal interlocking credit systems have faced similar decapitalisation problems 21

Naseem and Kelly, 1999 note that falls in national consumption of fertiliser are associated with periods of civil unrest or the early period of structural adjustment programmes, and these affect both supply and demand. Conversely, many of the increases in national consumption of fertiliser appear to be periods of recovery after such declines. 22

This may occur even if output prices have kept pace with input prices if input and output price rises both occur between seasons (i.e. after output sales but before input purchases). Seasonal finance for staple crop production in Sub Saharan Africa

22

and have been broken up, but have not been replaced by the private sector outgrower systems which have developed with some cash crops (Gordon and Goodland, 2000; Govereh et al., 1999; Al-Hassan et al., 1999; Kherrallah et al., 2000 23; Reardon et al., 1999; Poulton, 1998; Dorward et al., 1998b; Bryceson, Deborah Fahy, 1999a; Jayne, T.S. et al., 1997; Scoones and Toulmin, 1999, Brinn et al., 1999)24. To conclude and summarise this section, agricultural growth in sub Saharan Africa has been slow compared to other regions. While weak output markets, inadequate research and extension, poor communications, lack of education, poor health, and poor governance and civil conflicts all pose important and related difficulties for African smallholders, an important element in the lack of 23

While noting the difficulties facing maize producers in eastern and southern Africa, Kherrallah et al., 2000 argue that there is little evidence of adverse impacts of fertiliser market liberalisation on agricultural production, that production is more sensitive to weather, policy, and exchange rates shifts than to fertiliser policy, and that removal of subsidies has had negligible impacts on welfare (only affecting 15 to 35% of rural households, who are larger farmers in better areas). These conclusions isolate removal of input subsidies from broader structural changes in fertiliser markets, and ignore arguments that the supply response to marketing liberalisation has been poor in Eastern and Southern Africa since food production systems (particularly in remoter areas) were often (on balance) subsidised before liberalisation, with input supply an important part of that (e.g. Jayne, Thomas S. and Jones, 1997). They are also not supported by other studies on this area (e.g. Bryceson, D. F., 1999b). Mosley and Subasat, 1995 argue the volume of credit for agriculture was one of the few significant variables explaining overall economic growth in African economies in early structural adjustment years. The argument also focuses on a before/ after comparison rather than a with/ without analysis, comparing pre- and post- liberalisation situations but overlooking effects of possible lost growth in input use and its wider environmental and employment/growth linkage benefits. 24

An exception to the literature documenting increasing difficulties in crop finance is Kasekende and Atingi-Ego, 1999 who reports that resources allocated to crop finance in Uganda did not decline in real terms after liberalisation. However, no information is presented on smallholder share of this finance. Furthermore the lack of decline following liberalisation may also reflect very low levels of crop finance prior to liberalisation and smallholder difficulties in crop finance for cotton are reported by Gordon and Goodland, 2000.

23

growth in many areas is low use of inorganic fertilisers, particularly on food grains produced by smallholders in more remote areas with moderate or high agro-ecological potential. A major problem here (but not the only problem) is the difficulty that these farmers have in financing seasonal input purchases25. However, our earlier analysis of the role of agriculture in rural development and poverty reduction suggests that it is precisely in remote medium to high potential areas that smallholder production of semi-tradable crops such as food grains that agriculture may have the most critical role to play in promoting poverty reducing growth. We therefore now turn to consider problems constraining small farmers’ access to seasonal finance and means by which their access to seasonal finance may be improved.

25

It is important to stress that seasonal finance constraints are not the only cause of low input use: apart from the wider problems facing smallholder agriculture, input supply and profitability are also important and as Larson and Frisvold, 1996 note, credit is not useful if farmers remain chronically short of cash even after receiving access to seasonal input credit.

24

4 Seasonal crop finance in smallholder agriculture: issues and models Farmers may access seasonal finance to purchase inputs in basically three ways: through their own savings, through borrowing, and through cross subsidisation from non-agricultural activities. The development of financial markets are directly relevant to the first two of these: financial markets may provide more effective and efficient means of savings, and they may provide lower cost and wider access to credit. The two are related, as savings are necessary to provide funds for lending, although a principal purpose of more widely based and developed financial markets is to allow greater sectoral and geographical separation of savers and borrowers. However, the development of financial markets linking savers and borrowers in rural areas and beyond faces severe and well known theoretical and practical difficulties. We therefore begin this section with a brief review of these difficulties and then consider alternative mechanisms for financing seasonal inputs with and without recourse to financial markets. 4.1 Rural financial markets Financial markets are concerned with the transfer of money or liquidity from those with excess liquidity to those who are short of liquidity at a particular time, with the borrower paying for access to liquidity to the saver. Rural financial markets are characterized by frictions, which justify the emergence of financial intermediaries. Some of these frictions are high costs of acquiring information and generally high costs of doing business (transaction costs). Levine, 1997 summarizes the functions of a financial system to five broad categories, which arise to counter the effects of market frictions. They are to facilitate the trading, hedging, diversification, and pooling of risk; to allocate resources; to monitor managers and exert corporate control; mobilize savings, and facilitate the exchange of goods and services. Apart from linking borrowers and savers, financial intermediaries may also provide services in transferring money between two parties. Although transmission services are very important in many rural areas, we shall concentrate our attention initially on saving and borrowing activities. Compared with markets for the exchange of most goods and services, financial markets face a number of difficulties particularly associated with the temporal dimension of saving (or lending) and borrowing, with the transfer of an initial sum from the saver or lender to the borrower followed by payments of interest and repayment of the initial sum (according to some agreed schedule). The difficulties lenders face is in ensuring that borrowers do in fact make these subsequent payments, providing a secure return to the lender. Application of the New Institutional Economics (NIE) to these issues is particularly helpful (Von Pischke, J.D., 1999) with its emphasis on analysis of transaction risks and costs and of the role of institutions (see Poulton et al., 1998 for a summary of the NIE and its application to these issues). Two key and related concepts that emerge from this approach are asymmetric information (whereby lenders do not have as much information as borrowers about the latters’ financial situation and repayment intentions) and adverse selection (whereby those borrowers most likely to default are those most interested to seek out loans and accept high interest rates, with high interest rates increasing adverse selection as not only will they not deter intending defaulters – and may indeed increase incentives to default and hence moral hazard - but they will also reduce lending to low risk/ low return activities and lead to portfolios with a disproportionate number of loans for high risk/ high return activities). As a result lenders need to incur transaction costs (searching for and screening potential borrowers, then monitoring and enforcing contracts) to reduce the risks of transaction failure and loss of both interest and capital. Transaction risks and costs, however, are strongly affected by the nature of the good or service transacted, by the technical and economic environment, and by institutions (defined as ‘the rules of the game’, North, 1990). Institutions may be further considered in terms of the institutional environment and institutional arrangements Seasonal finance for staple crop production in Sub Saharan Africa

25

(Davis and North, 1971), with institutional arrangements being concerned with parties structuring contracts to reduce their transaction costs and risks (Williamson, 1985, Dorward, 2001), in this case to provide incentives for borrowers to repay. 4.1.1 Particular issues in smallholder agricultural finance Lending to smallholder farmers to finance purchases of seasonal inputs poses a number of difficulties to potential lenders: 1. small scale loans lead to very high transaction costs per unit lent (in searching, screening, monitoring and enforcement) 2. these are exacerbated by the dispersion of rural populations and poor communications infrastructure 3. agriculture is risky and insurance markets are usually non-existent (as they face similar problems to financial markets) 4. small holder farmers generally lack collateral (in SSA few smallholders have freehold rights over land) 5. small holder farmers generally lack records and valuations of past income, or future income estimates, or of assets 6. lending to agriculture in an area faces covariant risks from adverse weather or prices affecting large numbers of farmers in similar ways 7. the seasonal characteristics of agriculture lead to lumpy demands for finance at the start of a season, a period of several months without income (from which interest payments might be made) and then a concentrated period after harvest when both interest and principle payments can be made 8. the seasonal characteristics of agriculture leads to simultaneous patterns of lumpy demand and repayment by all farmers 9. financial intermediaries operating in rural areas may face a further difficulty in mobilising savings if agriculture is the dominant economic activity, as savers are likely to withdraw their savings at the time of greatest demand from borrowers. 10. The activities of state and donor sponsored agricultural credit agencies in the past have led to a climate of ‘strategic default’ among farmers in many areas (Poulton et al., 1998) whereby farmers have no incentive to repay loans as a result of past experience of ‘getting away with’ loan default without any penalties of enforced loan recoveries or even reduced access to future credit opportunities 11. There are particular problems in financing inputs for subsistence crop production as by definition the financed inputs will not directly lead to an increased cash flow from which repayments can be made and although cross financing with other household activities may generate cash for repayments, this requires that borrowers can (a) engage in activities generating sufficient cash income, and (b) match the timing of such income streams with the repayment requirements. These various difficulties reinforce each other, for example high risks in agriculture (3, 5) increase the need for searching, screening and monitoring and hence transaction costs, but these are exacerbated by small scale and dispersed borrowers without collateral (1, 2, 4). Large organisations spanning different rural and urban areas may overcome seasonal and covariant risk problems in agricultural lending (5, 6, and 7) but they face particular difficulties dealing with small and dispersed transactions (1 and 2) not backed by collateral (4). For smaller, rural based 26

intermediaries, however, the seasonal and covariant risk problems (5, 6 and 7) tend to be particularly serious, but they are better placed to address problems arising from small and dispersed transactions without collateral (1, 2 and 4). 4.1.2 Liquidity in poor rural livelihoods The discussion of financial markets in the previous sector has focussed on financial services supporting investment in seasonal inputs. However, these need to be considered in the context of the varied activities of poorer rural households and the particular liquidity constraints facing them. Households need liquidity for three main purposes: to make productive investments, for consumption smoothing (where flows of income and consumption do not match), and to meet demands for emergency expenditures. The liquidity required for purchases of seasonal inputs therefore must be seen in the context of the cash flows of other activities, the ways that these and agricultural income relate to consumption flows (for regular consumption as well as seasonal or occasional expenditures for school fees or social commitments, for example), and the need for access to funds for emergencies. Households then save up, borrow (or ‘save down’, to use the helpful terminology of Rutherford S, 2000) and invest in insurance substitutes (or ‘save through’) to allow access to lump sums when needed. For poorer households who spend a large amount of their income on basic commodities, the distinction between productive and consumption expenditure can often be blurred by their need to invest in minimum levels of consumption to maintain the household’s human assets. Such households also have a greater need for insurance substitutes. It is important, then, in considering financial services for subsistence farmers that due attention is paid to a mix of services (saving, borrowing and insurance) and livelihood strategies (farm and non-farm) supporting consumption smoothing and insurance as well as productive investments Matin, Imran et al., 1999. 4.2 Alternative means of accessing seasonal finance We now turn to consider different means whereby smallholder farmers may address the problems of financing purchases of seasonal inputs. We consider first a more general classification of different financial mechanisms whereby households manage their liquidity both to support regular consumption and to access (and pay for) lump sums for larger production, emergency or consumption expenditures. Rutherford et al., 1999 classify means by which the poor ‘save’ to access lump sums in terms of the informal, semi-formal and formal sectors. These categories, however, can overlap and become blurred, as informal mechanisms link or evolve into semi-formal mechanisms, and semi-formal mechanisms link or evolve into formal mechanisms. Within these categories we can also distinguish between mechanisms involving saving up, saving down (borrowing), saving through (insuring), or some combination of these. Discussion focuses on experience from around the world and specifically in SSA. However, discussion of specific experience in South Africa is left until section 6. 4.2.1 Means of ‘Saving up’ Informal ‘saving up’ mechanisms include saving alone, without involving anyone else, either by putting away and accumulating small sums when spending or by allocating and setting aside particular income streams or parts of income streams to finance particular lumpy expenditures. Since regular saving of small amounts of money can be very difficult and risky (in terms of self discipline and protection of savings from theft, accidental loss or calls from family and friends), many people use informal mechanisms to enforce their savings, with savings clubs, ‘money guards’, and fee collecting deposit (or susu) collectors or ‘mobile bankers’. The latter are very common and an important financial institution in West Africa, but are much rarer in East Africa (Rutherford et al., 1999. 27

Pure ‘saving up’ mechanisms are most commonly found in the informal sector (as savings with semi-formal or formal organisations commonly, but not always, also give some access to credit, see below). The volume of such saving by poor households can be very significant, and Rutherford et al., 1999 and Goldstein and Barro, 1999 argue that in both East and West Africa there is an urgent need for the semi-formal sector to develop a greater range of savings ‘products’ for the poor – and particularly in East Africa where neither deposit collectors nor SACCOs are widespread. There are, however, significant difficulties with reliance on such savings or ‘autofinance’ for purchase of seasonal inputs unless households can either (a) rely on some regular income source to support consumption (and precautionary) needs through the ‘hungry gap’ (without jeapordising labour for the crop) or (b) accumulate sufficient savings to support consumption (and precautionary) needs through the ‘hungry gap’ as well as the input purchase. This is not to suggest that cross investment from cash crops, local non-farm income, migrant labour and remittances is not an important means of financing food and cash crop input purchases for some farmers (e.g. Govereh et al., 1999; Reardon, 1997; Reardon et al., 1999, Jayne, T.S., 1994, Savadogo et al., 1998). The review of diversification and relationships between farm and non farm income in section 3.2.1, however, suggests that this is likely to work only for a limited proportion of generally less poor rural households: households able to grow cash crops (and thus in areas where cash crops are grown), households with significant local non -farm incomes (and thus in areas with a reasonable non-farm economy), and households with members able to engage in migrant or more permanent labour markets elsewhere and able and willing to send remit income. Thus Reardon et al., 1999 is concerned that cash crop and non-farm income sources “usually exhibit high Gini coefficients and are distributed regressively, the biggest farmers also earn the most off-farm and from cash crops. The effect is concentration of the capacity to follow sustainable agriculture and intensification among larger operators and commercialised smallholders only, leaving the mass of semisubsistence holders to extensify, intensify unsustainably, or exit from agriculture.” (p 383). These concerns are very challenging but may be addressed to some extent by development of financial markets offering low cost savings opportunities for the poor, by lowering transaction costs and price and other risks in holding non-financial assets (for example small livestock), and by increasing access of the poor to low cost means of borrowing and insurance (saving down and saving through). 4.2.2 Means of borrowing or ‘saving down’ Informal lending institutions in rural areas include moneylenders and traders offering credit to customers. These appear to be much less common Africa as compared with Asia (Rutherford et al., 1999). People may also belong to Rotating Savings and Credit Associations (ROSCAs) and Accumulating Savings and Credit Associations (ASCAs) (both very common in East Africa) although in the terminology of Rutherford et al., 1999 these are ‘saving through’ rather than ‘saving down’ institutions. All these institutions rely on local knowledge to screen and monitor borrowers, with loan repayment enforced through peer pressure, social norms and threats to deny access to future financial services. However, they have a limited capital base and repayment incentive structures mean that they often prefer lending for short term and consumption loans and (with the exception of interlocking transactions by agricultural traders, discussed below) are not keen to lend for agriculture and natural resource based enterprises (Jones, J.H.M. et al., 1999). Informal lending institutions often have problems in accessing wider financial services, due to remoteness from these and due to informational and administrative incompatibilities. The lack of linkage between banks and informal organisations prevents the latter from expanding lending beyond their own limited capital base, and may also increase risks to depositors. However, as compared with formal organisations, they tend to have much greater access to local information 28

and institutional arrangements that lower transaction risks and costs. There may therefore be potential for establishing links with formal organisations that allow them to access more funds and thus increase the volume and spread of their activity, allowing wider access by the poor on more favourable terms (Seibel, H.D., 1999b). However, there will still be a challenge in encouraging greater involvement in seasonal crop finance (Jones, J.H.M. et al., 1999). Interlocking transactions (Poulton et al., 1998) are an institutional arrangement whereby buyers of a crop provide inputs on credit at the beginning of a season and recover the input and credit costs when subsequently purchasing the crop. Conditions under which interlocking transactions will provide a sustainable and equitable means of input financing are quite restrictive (Dorward et al., 1998a, p257-259): strong incentives for traders to increase the volume that they process; competition between traders to protect farmers’ bargaining positions; high returns to seasonal input use (compared to alternative activities, to provide strong incentives for farmers to continue in interlocking relations); some form of restricted competition or information sharing between traders that allows them to penalise defaulting borrowers by denying them access to credit from other possible lenders; and some minimum scale of transactions (suggesting relations with larger scale farmers, with local assembly agents or with farmer groups). In liberalised markets such conditions are normally found only with cash crops with significant economies of scale in marketing or processing, and traders may also be credit constrained themselves (although cash crop traders are often better placed to access formal sources of working capital). However, in the past some parastatals used this method effectively in financing food crop inputs (Kydd et al., 2001), but this required a monopsony in maize purchases and possibly monopolistic and rationed input supply as well. Cash crop traders or processors may also provide inputs for food crop production as a service to attract and support farmers producing cash crops (e.g. Govereh et al., 1999, Poulton, 1998). As with informal lenders, traders offering interlocking contracts have been more common in Asia than in Africa where it is often associated with concerns about the equitability of the relationship between farmers and traders (e.g. Harris White and Janakarajan, 1997, Rajeev and Deb, 1998, 26 and see Dorward et al., 1998b, Govereh et al., 1999, Rusike et al., 2000 and Gordon and Goodland, 2000 for further discussion on this topic). Another institutional arrangement along these lines is the development of supplier credit relationships between input wholesalers and local traders. In Zimbabwe CARE has worked with local traders and input wholesalers to develop systems and bilateral relationships of trust allowing traders to stock fertilisers and other seasonal inputs on credit, and some traders have begun to offer some inputs to farmers on credit Rusike et al., 2000. This works best when traders are involved in both farm input and output marketing. Turning to consider formal sector lending institutions, commercial banks have historically played virtually no part in providing seasonal credit to smallholder farmers. Partly as a consequence of that, prior to the 1970s, concern about smallholder farmers’ access to seasonal finance led to large investments under\ the “old credit policy” (Ellis, 1992). This addressed (naively, with subsidised systems) some of the inherent problems in providing rural credit listed in section 4.1, but exacerbated others and fell out of fashion in the course of the 1980s. The model’s problems are widely recognised, with high costs (for both lenders and borrowers), political patronage, limited outreach and rationing, financial repression, and lack of savings mobilisation making them ineffective, expensive and unsustainable (Von Pischke, J. D. et al., 1983b; Adams, Dale W. et al., 1984; Adams, Dale W. and Vogel, 1986; Braverman and Guasch, 1986; Yaron et al., 1998, Matin, Imran et al., 1999). Recognition of these failings has led to them being largely discredited, and attention of donors and governments has turned elsewhere, although the original critique recognised that they had worked in some situations (eg Braverman and Guasch, 1986) and writers on agricultural development often mention state support in the provision of seasonal finance as an 26

Harriss, 1983 and Stockbridge et al., 1998 provide more positive Asian examples. 29

important and widespread ingredient in government support to successful agricultural modernisation (e.g. Eicher and Kupfuma, 1998; Morris and Byerlee, 1998; Kirsten and van Zyl, 1996) 27. There are also occasional examples of highly successful agricultural development banks serving smallholder farmers, notably (and almost uniquely) BAAC in Thailand and BRI in Indonesia 28. BAAC reaches over 80% of Thailand’s smallholder farmers, using both group and individual loans. It uses funds from both rural depositors and outside sources (whereas BRI’s deposits exceed its loans) and currently charges an average of only 11% on loans (this would need to be raised to 15% to eliminate subsidies) and its operating costs are only 3.5%. BRI makes only individual loans and has higher operating costs, but it charges higher interest rates and earns the highest rate spread, and in 1995 could have reduced its yield on its loan portfolio from 31.6 to 16.3% and remained free of subsidy (Meyer, 2000). Meyer, 2000 identifies the systemic features of success in BAAC and BRI (and the Grameen Bank) as a policy environment that allows them to set their own interest rates and select their own clients29, a legal and regulatory framework that allows them to mobilise deposits, effective organisational information systems, and continuing institutional development of products, services, repayment incentives, management skills, staff incentives, and human capital. However, there are examples of organisations continuing to provide subsidised agricultural credit. An example of this approach continuing in Africa with some modifications and some success comes from a programme supported Sasakawa Global 2000 and the Ethiopian Government. In a recent review of this Howard et al., 1999 reported that although the technical and credit programme had been financially profitable for farmers, economic analysis suggested that its wider viability depended on the overall maize situation in the country: viability of production for export was dubious, but if incremental production substituted for imports then wider economic gains exceeded costs even if extension and credit administration costs were allowed for. However, problems common to such programmes were noted with regard to high credit administration and extension costs and constraints on the supply of finance (with competition for limited funds between alternative uses as the programme has no savings component, relying on revolving donor and government funds, and charging interest rates of between 10 and 15%). Default rates were reported as ‘acceptable’ but this had often required heavy handed recovery methods. These problems are likely to intensify as the programme scales up, expanding to include less favourable zones and poorer farmers, increasing risks to farmers, risks of default, competition for funds, and competition for limited funds and staff. An alternative model for savings and lending institutions is provided by savings and credit cooperatives (SACCOs). Goldstein and Barro, 1999 classifies financial organisations in West Africa as type 1 institutions, large cooperative networks with members’ credit needs almost entirely 27

There are, however, considerable difficulties in empirically estimating the benefits from such support Von Pischke, J. D. et al., 1983a; Adams, Dale W., 1988 28

BRI (Bank Rakyat Indonesia) is widely cited as an example of a successful agricultural bank. However, although it operates very successfully in rural areas in both mobilising deposits and in extending loans, loan products are not designed for seasonal agricultural inputs and only 20% to 25% of lending is directly recorded as agricultural (Seibel, H. D. and Schmid, 1999. Nevertheless, in the context of the massive increase in lending by the Bank, this is a large increase over lending by BIMAS, the earlier subsidised ‘old agricultural credit’ organisation, and Seibel, H. D. and Schmid, 1999 argues that farmers are using BRI loans to finance non-farm activities, and autofinance with profits from these, together with trader credit, allows healthy funding of agricultural activities. The Agricultural Development Bank of Nepal (ADBN) may be another example of an agricultural development bank that is achieving significant success in outreach and sustainability (Seibel, H. D., 1999a). 29

BAAC still administers a small number of ‘government mandated special programmes which grant subsidised agricultural credit to targeted benficiaries… but the negative impact of these programmes on BAAC performance has led to a gradual reduction in their share of the overall loan portfolio’ (Klein et al., 1999 p 38) 30

financed by members’ savings (including membership fees); type 2 institutions, medium-sized cooperative networks with a savings volume insufficient to finance credit needs and therefore using donor-financed credit lines or funds from specialized banks; and type 3 institutions, practicing a credit-first methodology, where saving is not a prerequisite for access to loans and lending is financed through external sources, although compulsory savings may serve as evidence of the ability to put aside money for repayments and to a lesser extent as a loan guarantee. This typology includes savings and credit cooperatives (SACCOs) under types 1 and 2, and micro finance institutions (MFIs) under type 3. Caisse Villageois and Village Banks are forms of client owned financial institution intermediate and evolving between these types (Outtara et al., 1999, see Box 4.1 for a discussion of Caisse Villageois). This categorisation covers a huge range of different types of organisations and approaches, with varying governance and ownership structures, varying savings and loan products, and varying outreach and sustainability. There are examples of SACCOs with wide rural outreach in Francophone Africa, where there is a long history of such mutual organisations (e.g. FCECAM in Benin, RCPB in Burkina Faso, Kafo Jiginew in Mali and FUCEC in Togo). These can be very effective in savings mobilisation, offering competitive and popular savings products. They can also offer a range of loan products, some structured for agricultural lending (with balloon repayment schedules), with a range of mechanisms promoting repayment – collateral, strong social ties, minimum savings balances (often 25 to 30% of the loan amount), and guarantees30. However, their requirement for membership fees, savings and collateral means that they often reach better off villages and individuals, and exclude poorer communities and individuals. Attempts to develop loan products that deepen outreach (for example bringing in poorer women) remove the savings, membership and collateral requirements and use a more conventional ‘Grameen’ type approach (with small short term loans through small solidarity groups and regular small repayments) and hence are not structured to be compatible with lending for seasonal agriculture unless these regular repayments can be funded either from other (informal) borrowing or from other income sources. SACCOs can also provide a route for expansion and ‘graduation’ of village banks and MFIs to become sustainable client owned financial intermediaries. However, rapid expansion of SACCOs as channels of credit from outside funding sources can lead to borrowers’ (short term) interests dominating management decisions and diluting members’ incentives for prudent management and protection of deposits, leading in turn to problems with loan and client screening, loan repayment, control of operating costs, and setting of interest rates paid to depositors and of interest rate spreads (Outtara et al., 1999, Goldstein and Barro, 1999). However, without access to such funds their ability to meet demands for loans to members may be severely constrained, particularly in areas where economic activity is highly dependent on and dominated by agriculture 31. Experience with such organisations in East Africa has been much less positive, and they are much less common (Rutherford et al., 1999).

30

Goldstein and Barro, 1999 suggest that members of these organisations are interested in their savings products primarily as a means of accessing loans, but this is not supported by Outtara et al., 1999.

31

Although if clients include, for example, input and output traders as well as farmers there may be some complementarity between patterns of withdrawals and deposits of these different groups. Goldstein and Barro, 1999 suggest that savings rates are higher among urban members of such organisations and where they have significant presence in both urban and rural areas then urban deposits can be used to finance rural lending (e.g. RCPB in Burkina Faso). This requires a very strong national organisation linking local SACCOs. 31

Box 4.1 Caisse Villageois in the Pays Dogon in Mali Following the development of the CVECA concept by a French NGO (CIDR), and initially successful but not completed pilot projects in Burkina Faso in the early 1980s, the Pays Dogon region in Mali was selected for a more sustained testing of the concept from 1985. The region is characterised by strong social cohesion within Dogon villages (which remained isolated and very traditional), low income levels (with no traditional cash crops) but nevertheless economic vitality shown by the presence of many markets, entrepreneurial spirit, and local savings capacity. The area has a serious lack of arable land but is an important millet growing region, and vegetable growing is important (using rain harvesting techniques). Population density is relatively high (18.4 persons per km2, compared with 6.1 persons per km2 for the whole of Mali). Infrastructure is very poorly developed, with many roads impassable during the rainy season and few telephone lines. By December 1996 there were 52 CVECAs in the Pays Dogon region. These were set up in villages that met project criteria of minimum size (at least 500 inhabitants), social cohesion, market proximity, organisational vibrancy, savings capacity and credit demand. Bank structures are similar to existing traditional groups, and formation of a bank in a village required formal agreement and request from the village to the project. The project then provides training and a safe and book keeping documents, the village nominating a management committee and officers and constructing a building with local materials. Each bank has one to three tellers, with a ‘controller’ who checks the books every two weeks. A management committee is elected by the annual general assembly, which is composed of all village members (irrespective of their membership of the CVECA), and which also approves savings and lending rates and decides on how to allocate profits. The CVECAs offer a range of savings and loan products. Savings products include current, term, deposit and savings plan accounts. Term deposits attract interest at around 21%. By far the majority of loans are short term loans which run from 1 month to a year and from US$5 to US$900 or more, with a single repayment. Loan interest rates in 1996 averaged 43% across the different CVECAs, (most banks had initially set higher rates in order to build up a capital base, and then lowered rates as the CVECAs became more profitable). Collateral is required for loans and although this may be of relatively low market value, loss of use value is an important incentive to repay, although social pressure within the village is probably the most important repayment incentive. Other notable features of CVECAs in Pays Dogon concern membership, structure of members’ savings and borrowings. Membership has expanded dramatically to over 21,000, representing 17% of the adult population of the region and 67% of the adult population in villages with CVECAs. Average bank membership doubled from 1993 to 1996. However, only about half of the membership is active in any one year. Non-members tend to be both the poorer and more wealthy village residents (non-members have a higher mean income but lower median income as compared with members). Members continue to use traditional forms of saving (traditional groups, tontines, or money keepers) to the extent that their cash savings in these and in CVECAs are roughly equal. However CVECAs are strongly preferred for borrowing, accounting for 90% of members loans. However, only about 40% of members take a loan each year. In 1996, 82% of loans were for trade purposes, with 13% for agriculture and stock raising (although these figures probably under report loans taken for consumption purposes). Many loans were very small, with about 60% under US$50. However, loan sizes have been increasing. Although reported recovery rates are very high (at 99% in 1997), this is an underestimate due to unreliable data and methods of calculation: in fact there are some CVECAs in a precarious position with delinquent loans approaching 40% of total outstanding loan volume. Loans have been increasingly financed using outside sources of funds, with donor funds channeled through the BNDA (Banque Nationale de Developpement Agricole) financing 55% of outstanding loans in 1996 (up from 32% in 1993). These funds are charged at 8% p.a. (the same rate as BNDA charges to is best agro-industrial clients). Source: [Fruman, 1998 #347]

32

Many SACCOs in West Africa have programmes within them and links with what Goldstein and Barro, 1999 term ‘type 3 institutions’, identified as the ‘semi-formal sector’ by Rutherford et al., 1999, but more generally known as Micro Finance Institutions (MFIs). Again, there is enormous variation in these regarding the lending and savings products they offer, their governance, and their outreach and sustainability, and some MFIs are now registering as banks. A characteristic shared by most MFIs is a focus on small short term loans associated with compulsory regular (weekly, fortnightly or monthly) savings and repayments. Many of them have a strong emphasis on lending (Rutherford et al., 1999, Hulme et al., 1999, Goldstein and Barro, 1999) with saving seen as a means of developing and demonstrating ability to make regular repayments and of building up individual and group funds to (a) act as loan collateral and (b) provide funds to meet emergencies. Many have used a group lending approach to reduce transaction costs for the MFI and to encourage peer group pressure for loan repayment, although Mosley and Hulme, 1998 report that there are no consistent differences in performance between programmes using individual and group loans. Other common characteristics are graduated access to increasing loan sizes and a wider range of loan products, high interests rates to cover costs, effective management information systems, low operational and fixed costs, and loan officers that are locationally and socially accessible to clients and have clear incentives and delegated authority. Many of these characteristics are similar to those of successful formal sector institutions outlined earlier, but many of them are also not conducive to lending in poor rural areas, or to lending for agriculture. The vast majority of successful MFIs operate either in urban areas or in densely populated rural areas with a strong non-agricultural economy and/or agriculture which has already started to ‘modernise’. Rutherford et al., 1999 and Hulme et al., 1999 report that in East Africa almost all MFIs are located in urban areas or in rural market centres located along major roads, and (with the exception of one or two MFIs) are also largely inactive in and absent from the poorest districts. Similarly, the vast majority of MFI clients tend to be small traders or micro entrepreneurs. (this applies in for example ASA, 2000)32. However, the extent and nature of lending in agriculture is often difficult to determine, as (a) the nature of MFIs means that they often do not collect much information about the precise uses of loans, (b) fungibility makes it difficult to determine what contribution any particular source of liquidity makes to household activities, and (c) where loan uses are recorded, the term agriculture often covers a wide range of activities, and may include small stock rearing, vegetable production and small scale agricultural trading (all short term activities) in addition to seasonal finance for subsistence or cash crops. There are examples of MFIs working in rural areas and offering loans for agricultural purposes. Some of these are well known and documented, for example Financiera Calpia in El Salvador (Klein et al., 1999, Navajas and Gonzalez-Vega, 2000). Financiera Calpia, operating in one of the most densely populated countries in the Western Hemisphere (Navajas and Gonzalez-Vega, 2000), has extended its operations from a largely urban base to target rural areas as well, and has at the same time developed high levels of performance in terms of outreach (with both lending and deposit mobilisation), portfolio performance and profitability. In 1997 92% of its rural loans were 32

These comments apply more widely, for example there are suggestions in Bangladesh that higher default rates may result from selection of ‘sparesely populated’ or ‘unproductive or strictly agricultural areas’ (ASA email from sharenet ASA, 2000). Sharma and Zeller, 1999 find that branch placement across ASA, BRAC and PROSHIKA in Bangladesh tends to favour areas with better transport and communication infrastructure and lower risks of distress but with lower literacy rates. It is not, however, significantly affected by population density or the percentage of the population that is urban. For two of the NGOs, branch placement is also higher in areas with a higher proportion of marginal farms (with area less than 0.5 acres). Once a branch is established, however, rates of outreach and uptake tend to be higher in areas with higher risks of distress and in areas with a higher proportion of farms over 0.5 acres, but is not related to infrastructure. For two of the NGOs studied, rates of outreach and uptake were also positively related to market density. 33

allocated to agricultural lending, with loan write-offs of less than 0.7% (Klein et al., 1999). Navajas and Gonzalez-Vega, 2000, identify key features of Financiera Calpia’s success in a) recognition that the household/ farm is a financial unit integrating a variety of economic activities with expenditures and incomes, with an emphasis in lending decisions and monitoring on repayment capacity rather than funds utilisation; b) management of systemic risk in agriculture by three levels of diversification (household diversification across activities supports both risk control and cash flows for repayment, together with lending for a range of agricultural and non agricultural activities by rural branches, supported by diversification across rural and urban branches); c) promotion of benefits from long term relationships for both lender and borrowers; d) use of collateral, including non-traditional collateral from poorer households, which though it may not perform the function of reducing the lender’s loss from default, signal the lenders’ intentions to recover loans and strongly increase the borrower’s incentive to repay; e) strong emphasis on delegated and decentralised decision making by well trained loan officers who can make prompt decisions about loan authorisation and disbursement; f) regular monitoring of clients’ use of funds, not so much to control their use but to ensure that repayment capacity is not jeopardised, to encourage appropriate changes in use to match changes in opportunities, and to strengthen the borrower-lender relationship and signal the seriousness of the lenders’ intentions; and g) an efficient MIS and a commitment to loan recovery (including seizure of collateral where necessary) to signal to other clients. A key element in Financiera Calpia’s success is the performance of its loan officers. Most of these have a first degree in agricultural science, are experienced and well motivated. Klein et al., 1999 draws similar lessons from success in an examination of BAAC from Thailand, Financiera Calpia, and Credit Municipales de Ahorro y Credito (CMACs) in Peru. A number of questions remain, however, about the broader applicability and sustainability of these models (see table 4.1 below). While BAAC has very high operating efficiency and wide outreach, this is not the case for Financiera Calpia and CMACs, which were still small and incurring high loan administration costs. BAAC’s lower loan administration costs are partly due to the extensive branch network (established with the support of government financed investments over many years), use of group loans, and large loan sizes, but Financiera Calpia, and CMACs also have large loan sizes in comparison with likely loan sizes for food crop producers in sub Saharan Africa. In addition, Financiera Calpia, and CMACs both place emphasis on diversified incomes of borrowers supporting cash flows for repayments (Klein et al., 1999 report that most borrowers from Financiera Calpia possess farm assets with an approximate value of US$7000, have diversified income sources, and live in areas with relatively good infrastructure and transport facilities). It is therefore doubtful that these models would transfer to many areas in sub Saharan Africa (but lessons can nevertheless be learnt from these institutions regarding, for example, risk management and other aspects of good practice). Table 4. 1 Performance Indicators of BAAC, Calpiá and CMACs 33

33

Source: Klein et al., 1999 page 42, data for 1996, except for Financiera Calpia on branches, borrowers and outstanding loans which are approximate estimates from medians per loan officer in 1998 made from Navajas and Gonzalez-Vega, 2000. 34

BAAC

Calpiá

CMACs

650 (850 field offices)

15 out of 68 loan officers, 10 branches

6 out of 12

2 435 836

5850

4 572

Outstanding loan portfolio to individual farmers (US$ million) (1996)

5 589

42.2

7.3

Average Agric. Outstanding loan size per borrower (US$) (1996, 1997 for Calpiá)

2 286

720

1 607

400-500

390 (median)

325-583

Loan administration expenses as % of average outstanding loan portfolio (1996)

3.3%

20.4%

22.2%

Financial expenses as % of average outstanding loan portfolio (1996)

7.1%

9.1%

16.8%

Financial income as % of average outstanding loan portfolio (1996)

10.4%

39.4%

55.0%

No. of branches involved in agricultural lending (1996) No. of rural borrowers (1996)

Outstanding rural loans per loan officer (1996

In addition to the above examples of MFIs providing larger agricultural loans, there are also MFIs in Asia and Africa providing smaller loans to poorer households. Unfortunately it has not yet been possible to obtain detailed information about these (for example BIMAS in India, KREP and WEDCO in Kenya and the Centenary Bank in Uganda (P.Mosley, pers.comm)) or about less well known programmes within established MFIs (for example Matin, M., 1999 reports that prior to 1992 the Grameen Bank in Bangladesh offered small fertiliser and irrigation loans with balloon repayments at harvest but that repayment performance on these loans was generally poor and they were replaced by ‘seasonal loans’ with similar maturity and repayment periods to general loans, although he suggests that these too have suffered from repayment problems as a result of over rapid credit deepening34). Most of these, like Financiera Calpia, and CMACs, harness other, more regular income flows for the gradual repayment of seasonal agricultural loans and may also have lower seasonality of cash flows due to irrigation and/or bimodal rainfall patterns. Repayment of loans may also, of course, be paid for by borrowing (e,g, Matin, M., 1999, Sinha, 1998). Existing MFI lending technologies may therefore finance seasonal input purchases if households can in turn finance the regular repayments either from other income or from other borrowing. Further information about the involvement of MFIs in financing seasonal agriculture will be gathered during the course of this project, but from the evidence presented above it appears that the frequently cited success stories and examples tend: a) to be in higher population density areas, often where agricultural modernisation is already under way, with consequent diversified non-farm economy and non-farm incomes not found in poorer areas;

34

These loans also do not follow the normal Grameen practice of group members taking it in turns to borrow as seasonal loans must all be granted at the same time, but a number of authors report that joint liability for loan repayment is seldom enforced in the Grameen Bank (Matin, M., 1999, Jain, 1996,. 35

b) to provide loans which are often not structured in ways that allow them to be used for financing seasonal crop inputs (for example with regular monthly repayment instalments) unless households’ are able to use some other source of cash to finance loan repayments; c) seldom provide evidence that they are in fact financing seasonal crop inputs35; d) where they are directly engaged in agricultural lending are often in higher population density areas and work with better off farmers with collateral and other sources of income (for guarantee purposes rather than repayment), offering larger scale loans (when compared either with the loan sizes under non-agricultural lending schemes, or with sums required by subsistence farmers to finance seasonal input purchases). There are exceptions to these generalisations, and these require further study, but it appears that there are few if any current working models for extending the frontier for access to seasonal finance for poorer farmers in poorer rural areas. There are, however, a number of lessons that emerge from this review regarding necessary (but not sufficient) conditions for the frontier of access to be extended: a favourable environment, effective organisational management and low cost lending technologies that meet clients’ demand. Many aspects of the environment in which a financial organisation operates will be important to its activities and performance. We highlight the following aspects of particular importance: ♦ a political and regulatory environment that allows deposit mobilisation and independent setting of interest rates, client selection, and loan recovery enforcement; ♦ a reasonably stable economic (and institutional) environment that supports entrepreneurship by yielding returns to investment by borrowers; and ♦ a social environment that allows development of local information networks about borrowers and potential borrowers and, for group lending, some solidarity between group members. High population densities reduce the costs that lenders incur in screening, monitoring and enforcing transactions. Important aspects of effective organisational management found in successful organisations (and missing from many failed organisations) were discussed earlier, and include independent professional managers, governance structures that protect the interests of depositors, well structured staff incentives and training programmes, efficient information systems, and continuing product development.

35

Although problems of fungibility make it difficult to identify what any loan is financing, in the absence of access to equivalent cash sums (from savings, incomes or loans) from other sources, the options for fungibility are limited, demonstrating that either loans are not supporting agriculture directly or indirectly, or their effectiveness relies on a fairly high flows of cash in the household economy).

36

Table 4.2 Features of lending technologies Effectiveness in reducing transaction costs/ or increasing performance in: Screening Lending technology

Self selection

Monitoring

Selection by Lender

Repayment Incentive

Suitability for: Agriculture

Poor

Enforcement

Poor agriculture

?36

Regular repayments Repeat / graduating loans Compulsory Savings

?37

Group lending

?38

?

? ?

39

Collateral

?40

Household budgets

?

Interlocking

36

Regular repayments will pose problems for poorer households without other large and reliable enough regular income sources

37

Savings may be difficult to maintain over the seasonal financing and hungry gap period without other sources of income or borrowing.

38

?

Covariant risk in agriculture may increase the likelihood of the ‘domino effect’ during repayments but benefits in screening remain.: groups may also exclude poorer, higher risk entrants. 39

Collateral requirements should not pose problems for the less poor, especially if non-traditional forms of collateral are used.

40

Drawing up household budgets will be very expensive for lower loan sizes Seasonal finance for staple crop production in Sub Saharan Africa

37

A major feature of the dramatic increase in MFI activity has been the development of lending technologies that both meet demand and are efficient in reducing transaction costs and risk for lending institutions and their clients. Table 4.2 presents some of the different elements in lending technologies reviewed in this section in terms of the way that they reduce costs or increase effectiveness in screening, monitoring and enforcing transactions. The final three columns ask how far these elements may be appropriate in lending products aimed at financing seasonal agricultural inputs, aimed at the rural poor, and aimed at seasonal financing for the poor. It is striking that a number of the key elements of lending technologies offered by MFIs are not suitable for agriculture (e.g. regular repayments and compulsory savings), and the alternative elements introduced to replace these (e.g. collateral, household budgets) are often not suitable for the poor. The table illustrates the difficulties and challenges that exist in designing lending technologies that can be implemented with sufficiently low transaction costs and risks to allow cost recovery from small, dispersed (and thus administratively expensive) loans. Identification of these three conditions necessary for extending the frontier of access to seasonal finance leads to different strategic approaches to the development of financial services (see for example Seibel, H. D., 1999a): regulatory development (improving the policy and regulatory environment, including the framework of financial supervision); institutional development (improving the financial infrastructure and transforming roles and interactions of different players41); organisational development (improving the governance and management of financial organisations); and product development. The focus of this research project is on the last of these, product development, but this can only occur with and must be designed for an effective regulatory environment, working financial institutions, and a well managed organisation. In addition to these conditions necessary for the supply of financial services to seasonal agriculture, there must also, of course, be demand for such services: this requires agricultural technologies, agro-ecological conditions and input and output markets that yield attractive and relatively low risk returns. In some areas where climatic risks are high, average returns in more intensive agriculture may just be too low to justify investments and risks. Elsewhere farmers need some sort of insurance mechanisms to guard against risks of enterprise failure, whether this be due to widespread problems such as drought or low output prices, or due to more specific problems of local flooding or hail, fire, or household problems such as acute or chronic sickness. We therefore turn now to consider farmers’ access to insurance mechanisms. 4.2.3 Means of insuring or ‘saving through’ Zeller and Sharma, 2000 argue that insurance is the ‘missing third of micro- finance’ and that an important part of financial services should be to enhance the poor's ability to bear risks. Provision of insurance services to the poor is not, however, an area that has enjoyed much prominence or success, and this is particularly true in agriculture. Mosley (pers. comm.) argues, however, that whereas many micro finance organisations lead with lending services, followed by saving services 41

An important aspect of institutional development that has not been explicitly discussed elsewhere in this review is the relationship between more local micro-finance institutions and banks. ‘Linkage banking’ is a concept being promoted within Africa (see for example Seibel, H.D., 1999b) and it would appear to offer opportunities for overcoming combining the strengths of both local and national institutions, with the involvement of the former allowing lower information and transaction costs while the latter reduces problems of covariant risk and seasonal cash flows, allows the benefits of financial intermediation across sectors and urban and rural areas, and can provide greater financial regulation and supervision in the microfinance sector. However, with moves by some banks to move into microfinance themselves, and by some microfinance NGOs to become banks, relationships between these institutions can also become competitive Reinke, 2001. Seasonal finance for staple crop production in Sub Saharan Africa

38

(with deposit taking often constrained by banking regulations), and are not involved in insurance at all, the requirements of the poor are first for insurance, then for saving (without minimum deposit or transaction constraints) , and then for borrowing. Insurance in agriculture faces many similar problems as lending, with high transaction costs and risks as a result of asymmetric information, adverse selection, moral hazard, small insured amounts, and high monitoring costs. Informal insurance mechanisms include ROSCAs and membership of social networks (with both horizontal and vertical relations), and households may also engage in precautionary savings, for example investing in jewelry, livestock or savings accounts. These mechanisms may be effective for dealing with idiosyncratic risk (events that strike individual households in a locality) but (with the exception of financial savings42) are much less effective in dealing with covariant risk (for example the effects of drought, adverse price changes, or a disease epidemic). The amount that can be insured by precautionary savings is also constrained by the ability to save and volumes of savings, and although, as we have seen, the poor do save, volumes saved are limited. Insurance is, however, likely to be of particular concern for poor households looking to borrow for a productive investment, as an event that prevents them from realising a return from that investment may place them in the potentially disastrous situation of needing to finance the burden of future loan servicing and repayment without any increased assets or income. This is of especial concern in agriculture where both idiosyncratic and covariant risks tend to be high. Here the needs for and difficulties with insurance are both very high. A number of large scale programmes in the 1960s and 70s failed, and the conventional wisdom has been that such large scale approaches should be approached with great caution (Hazell et al., 1986). There may be scope, however, for agricultural insurance schemes that (a) confine themselves to insuring against covariant rather than idio-syncratic risk, (b) consequently use more regional or area rather than farm based assessments of risk, and (c) operate at a smaller scale with greater use of peer monitoring and other (local) incentives and information to address problems of information assymmetry, adverse selection, moral hazard and hence high transaction risks and costs. Mosley, 2000 reports three schemes adopting different combinations of these. These schemes are in their early stages of design or implementation, and their performance cannot yet be evaluated, but their innovative features hold promise, as does the increasing interest in delivering micro-insurance services to the poor. 4.2.4 Transmission services Transmission services, allowing clients to transfer funds from one area to another are given very little attention in the literature on micro finance services in sub Saharan Africa outside South Africa. This is surprising given the importance of internal and international migration in Francophone West Africa, and the increasing importance of remittances in rural livelihoods throughout the continent (Bryceson, D. F., 1999b). It poses important questions about the demand for such services and the ways that rural people currently transmit funds: for example what is the relative importance of formal and informal transfer mechanisms? what are the relative costs, risks and speed of transfer under these methods? and how does access to them vary for different client groups? This may be an area where there could be important benefits from linkages between commercial banks and microfinance and informal institutions.

42

And the effectiveness of savings is limited by the ability of a household to save up sufficiently, a particular problem for the poor. 39

4.3 Access to seasonal crop finance: conclusions This section has reviewed the features of rural financial markets and seasonal crop finance that commonly lead to market failures and to a particular lack of access for smaller scale subsistence farmers. Two principle mechanisms were identified for accumulating the lumpy funds necessary to purchase seasonal inputs: savings (including cross-investments as a means of autofinance) and borrowing. Although savings are important means whereby some households fund seasonal input purchases for subsistence crops, this is normally only possible if they are growing a cash crop or have a significant non-farm or remittance income, and this excludes many households, particularly poorer households. Conditions necessary for sustainable supply of lending services include a favourable environment, effective organisational management and low cost lending technologies. Smallholder demand for agricultural lending requires profitable investment opportunities in agriculture and access to insurance mechanisms. A variety of informal borrowing mechanisms exist, with low borrower screening and monitoring costs, but they tend to have limited access to wider funds and without such access cannot simultaneously fund input loans for all clients. Individual money lenders are not common in Africa, but SACCOs are common in West Africa and ROSCAs in East Africa. SACCOs involve significant compulsory savings before a member is eligible to borrow, and this may exclude the poor. If linkages are made to external institutions to access funds, there are dangers that governance structures may give insufficient protection to depositors, and thus undermine the sustainability of these organisations: such problems have depressed confidence in SACCOs in East Africa. However, the West African experience suggests that given the right environment they can be viable, but they need strong deposit protecting management, and growth should not be too rapid. Although the West African experience suggests that they have not been very successful in extending agricultural lending, this may be due to low and risky returns to agriculture in the areas where they have grown fastest, and to a lack of effective insurance mechanisms and services. While they may not directly address poorer households’ needs for financing inputs for subsistence crops, they may provide an institutional context for graduation of other approaches to address these needs. ROSCAs require regular contributions from members, and the taking of turns to receive the fund precludes more than a minority of members from using them for seasonal finance. Interlocking seasonal input loans with product purchases can be effective where restrictive conditions are met, but these conditions are not usually satisfied in the marketing of food crops. Formal lending to smallholders has often suffered from an unfavourable investment climate, politicisation, poor management, lack of emphasis on savings mobilisation, and consequently poorly designed products. However, their collapse in sub Saharan Africa has left a gap which has not been filled by the new generation of MFIs which in SSA have tended to focus on urban areas and, where they do work in rural areas, these tend to be the less poor areas with better access and less dependence on agriculture. Examples of MFIs which lend directly for agriculture exist, but for those reported in the literature average loan sizes are relatively large and lending technologies (for example use of collateral and of household budgets) are unlikely to be sustainable for smaller scale and more dispersed loans for subsistence crop production. A small number of MFIs are currently experimenting with seasonal agricultural lending. Understandably these tend to be located in more favourable areas with lower climatic risk, less marked seasonality, and a range of crops, including both cereals and horticultural crops. Further information is needed about these MFIs, which both link repayments of seasonal loans to regular sources of non-farm income, and provide some flexibility in scheduling a greater proportion of repayment after harvest. 40

The current situation is summarised in table 4.3. This suggests possible approaches to increasing access to seasonal finance under different levels of agricultural and non-agricultural activity in an area. In areas with high levels of non-farm activity, cash and subsistence crop farmers may have access to sufficient alternative income streams to use standard (individual loan) MFIs to finance seasonal inputs, and specific agricultural lending (following the BAAC or Financial Calpeo models) may also be viable where households are growing cash crops. In areas without such nonfarm activity, however, these are unlikely to work. Depending upon the nature of the cash crop and market structures, however, interlocking arrangements may be viable in cash crop production (although if farmers have few alternatives income sources and there are few traders, the traders may capture an undue share of the benefits of these transactions). Options for financing inputs for subsistence crops under these circumstances are, however, very limited. In areas with low agricultural potential, there may be insufficient returns or too high a risk to make seasonal borrowing and lending viable. Where, however, the agricultural potential could provide returns to seasonal inputs sufficient to justify investments in them, there is a challenge to develop appropriate institutional models for financial service provision, including saving, borrowing, insurance and transmission services. Table 4.3 Options for Seasonal Crop Financing Level of agricultural activity in area High

Low

Level of non-farm activity in area

Agricultural activity of household

High

Cash crop

MFI (agric, standard)

MFI (standard)

Subsistence crop

MFI (standard)

MFI (standard)

Cash crop

Interlocking?. SACCOs? CVs?

Interlocking? SACCOs? CVs?

Subsistence crop

?

?

Low

41

5 Rural livelihoods in South Africa Following a broader review of rural poverty, rural livelihoods and rural finance in sub Saharan Africa, we now turn to examine these topics in South Africa, in order to (a) identify the similarities and differences between South African rural livelihoods and those further north and (b) develop priorities for field work in the next phase of the project. 5.1 An overview of rural livelihoods in South Africa Rural livelihoods in South Africa must be understood in the context both of the overall structure of the economy and of the history of apartheid. Compared with most other countries in sub Saharan Africa, South Africa has a much more developed and industrialised economy. Thus in 1998 agriculture accounted for 4% of GDP (compared with 17% in SSA as a whole), GNP per capita was just over 7 times that of SSA as a whole (US$3829 compared with US$535, constant 1995 US$), industry and services comprised 32% and 64% of GDP respectively (compared with 29% and 54% in SSA as a whole) and rural people made up 47% of the population (compared with 67% in SSA as a whole) (World Bank, 2000b). The high GNP per capita, however, masks very high income inequality: 50% of the population is defined as poor, with 61% of black people estimated to be poor, compared with 38% of coloured, 5% of Indian and 1% of white people (DFID, 1998). The poorest 20% receive just over 3% of income compared with 47% received by the richest 10%, and per capita incomes in Gauteng (the richest province) are six times greater than per capita incomes in Northern Province (the poorest) (Ibid.). Data showing the low relative importance of agriculture compared to industry and services might suggest that according to Reardon’s analysis of stages of agricultural transformation, South Africa might have progressed beyond stage 1 to stage 2, but with severe inequalities between urban and poorer rural areas, and within rural areas between the former white commercial farming areas and the former black homelands industries have not by and large developed in the areas where most rural people live, and these would be better characterised in terms of Bryceson’s ‘complex C – transfer payments’ with rural-urban, labour out-migration and highly mobile populations, with pensions and remittances become a major source of rural income. Rural areas have been very difficult to clearly define in South Africa due to the prevalence of various settlement regimes (homelands, etc) that were created as a result of previous government’s policies. In some literature the term ‘rural’ has been vaguely defined as non-urban areas. Reference has also been made of rural areas as non-urban parts of the country that are not well served by physical and social infrastructure. According to the Strauss Commission, 1996, available data does not allow a more accurate definition of the rural population in South Africa. However, a broad view of the characteristics of rural livelihoods country wide is provided by Leibbrandt et al., 2000 (see table 5.1 overleaf). This table illustrates a number of important points about rural livelihoods across South Africa: ♦ Poverty incidence is greater in rural areas, with 72% of the poor living in rural areas (DFID, 1998). Roberts and May, 2000 report that in KwaZulu-Natal the incidence and severity of poverty are much higher in rural areas than in urban areas, that both incidence and severity of poverty increased more in rural areas from 1993 to 1998 (for example the headcount ratio for rural areas increased from 44% to 54%, compared with an increase in urban areas from 16% to 21%) , and 87 % of the chronically poor over this period and 77% of the transitorily poor were located in rural areas. ♦ Livelihoods appear to be highly diversified but variation between the percentage of households with access to different income sources and the percentage of income from these sources Seasonal finance for staple crop production in Sub Saharan Africa

42

suggests that there is great diversity in livelihood structures (and in incomes) between households ♦ Agriculture makes up a small proportion of income for a large number of people (almost half rural African households)43 ♦ 60% and 76% of all rural and poor rural African households respectively have no access to wage income, but nevertheless this makes up 53% and 23% of average income, and is highly unequally distributed. ♦ Remittances and state transfers (particularly the former) reach a large number of poorer people, and make very significant contributions to their incomes. The importance of remittances reflects the tendency among rural people to migrate from one locality to another (not necessarily to urban areas), with one or more family members leaving the resident household for varying periods of time, and in so doing making new and different contributions to its well being (although such contributions are not guaranteed by the mere fact of migration, Ellis, 1998). Most evidence points to the desire to increase incomes and diversify sources of income by the family as a major motive for migration. Intra-rural inequality, with economic surplus among the rural rich and labor displacement among the rural poor, is a powerful influence of migration. Much migration remains rural-to-rural. Selective migration by young rural men has caused large, but irregular remittances of savings and serious inequalities among the remaining, often aging rural, rural poor (Strauss Commission, 1996). Table 5.1. Features of rural livelihoods in South Africa All African

Homeland African

Rural Homeland African All households

Poorer households

Share of national hholds

71%

41%

36%

29%

Percentage hholds in poverty

54%

68%

71%

100%

Average hhold income (R/month)

1012

912

856

(1.1)

(1.2)

(1.2)

4.8

5.6

5.6

2.5

2.7

2.6

48%

52%

54%

(Coefficient of variation) Average household size Working age adults Dependency ratio

% hholds access to income from

% income

% access

% income

Remittances

31

44

47

14

54

26

Wages

61

44

40

52

24

23

43

In the country as a whole, Eskom, 1998 report that 2.1 million households (just under 24% of all households) are engaged in small scale agriculture, and of these 7% are emergent commercial farmers, 18% earn some income from commercial agriculture to supplement other (main) sources of income, while 75% are subsistence farmers. Maize contributes 56% of crop incomes, with spinach, potatoes and other vegetables and some fruit making up the balance. 43

Capital (inc. imputed rent)

66

86

88

7

91

8

State transfers

27

36

37

18

36

32

Agriculture

24

39

44

5

47

7

Self-employment

12

13

13

6

11

5

Note: Poorer household are those below household subsistence level (approx. 70%) Source: adapted and rounded from Leibbrandt et al., 2000, p 82 and 93. On the basis of these figures (which are generally supported by other surveys) and of Gini coefficient analysis of the contribution of different income sources to income inequality, Leibbrandt et al., 2000 argue that agriculture is relatively unimportant as an income source for the poor. However, these conclusions need to be considered carefully. First, methodological difficulties in reporting and valuing non-cash agricultural income need to be recognised44, as do the dangers of taking averages across different areas and across households with different livelihood strategies: thus the figures from Leibbrandt et al., 2000 in table 5.1 suggest that although only 47% of the poorest households report agricultural income, contributing an average of 7% of household income, this suggests that for those 47% of households agriculture will contribute a much higher proportion of household income. Coetzee, G. and Vink, 1991(citing Coetzee, G. and van Zyl, 1990 and Lyne, 1998)report that in Kwazulu and KwaNgwane 15% of smallscale farmers produce more than 80% of the marketed maize and groundnuts. Second, and related to this, variation can also be expected between different provinces: Eskom, 1998 estimate that of the 2.1 million households engaged in small scale agriculture, 82% are found in Eastern and Western Cape, in Kwa-Zulu Natal, and in Northern Province. Case studies describing rural livelihoods from areas in each of these provinces are summarised in Appendix 1. The importance and contribution of agriculture in rural livelihoods not only varies between provinces but also between areas within each province. Third, as argued earlier in section 3, the importance of agriculture cannot be related simply to income share. Within household livelihoods, retention of access to land, of agricultural skills and of social relations may be very important parts of a ‘survivalist’ strategy, an important source of security for households vulnerable to loss of employment and/or remittance income. A small number of studies have constructed typologies for rural and farming households in different areas in South Africa (for example Makhura et al., 1998,Laurent et al., 1999). Makhura et al., 1998, for example, using 1987 data from KaNgwane in Mpulumanga identify with cluster analysis seven ‘commercialisation groups’ according to the structure of cash income from non-farm and different farm activities (excluding remittances and state transfers). The relative frequency, description, mean income and land holdings for the seven groups are summarised in table 5.2. Table 5.2 Characteristics of ‘commercialisation groups’ in KaNgwane in Mpulumanga % sample Commericalisation groups

Cash income (Rand) Agric. non-farm

Land (ha)

total

44

Shackleton et al., 2000 review evidence that wild resources and livestock and crop production on communal lands are commonly substantially undervalued due to measurement, market valuation and social valuation problems. Eskom, 1998 report that 88% of households raise crops or livestock only for domestic consumption. 44

Non farm

34%

356

5415

5771

3.7

Very low

23%

190

12

202

4.0

Moderate agriculture

20%

397

809

1206

4.6

High agricultural

14%

3667

434

4101

6.2

Livestock

6%

4195

864

5059

Non farm & agriculture

2%

5348

Highly commercialised agriculture.

2%

9106

4694 10042 411

9517

Source: Makhura et al., 1998 Using data from studies in the mid-late 80s and mid-late 90s in the central zone of the Eastern Cape, Laurent et al., 1999 also construct a typology of seven groups based on cash income from farming. Table 5.3 Typology of households, central zone of the Eastern Cape Moneyless Transfer and remittances Landless

11

111 57% Pensions important, subsistence farming, 10

Not farming Non Farm income

3 14

Minor farming Farming

6% Low income, no regular source, subsistence farming

9

5% No land, mainly remittance and pension income 2% Old/ infirm, mainly remittance and pension income 7% Subsistence farming, income from non-farm activities 5% Subsistence and minor farm sales, other income sources important

36 19% Small scale commercial farmers, high input use, employ labour

Source: Laurent et al., 1999 Although these studies used different methods and are located in widely differing areas, and both studies measured only cash income45, there are striking similarities in the importance of non-farm income and the Eastern Cape study also demonstrates the importance of remittances and pensions (presumably these are important for the ‘very low commercialisation’ group in KwaNgwane). Both studies, however, also show the importance of farming as an activity – in the KwaNgwane study little mention is made of income in kind from subsistence farming, but access and use to land is clearly important, while in the Eastern Cape study subsistence farming is practiced by the vast majority of households. However, little is know about the way that these different income streams and activities complement each other within the overall livelihood strategies of the different groups, nor how they are related to vulnerability and holdings of different assets. Roberts and May, 2000 report that in KwaZulu-Natal a large proportion of the poor own small amounts of land and livestock (see table 5.4). Table 5.4 Access to land, livestock and pension income by poor and non poor in KwaZuluNatal

45

Approximately 90% of the value of agricultural production in the former homelands of Ciskei and Transkei is not marketed (Eastern Cape Province, 1995, cited by Ngqangweni, 1999). Van Zyl and Coetzee, 1990 report that very small proportions of crops or livestock are sold and the majority of the households are deficit producers. 45

Chronically poor

Transitorily poor

Never poor

1993

1998

1993

1998

1993

1998

45

58

39

53

27

33

Mean (median) cultivated land (ha)

0.95 (0.5)

0.65 (0.4)

0.96 (0.5)

0.83 (0.5)

3.85 (0.75)

1.02 (0.5)

Mean cultivated land per capita (ha)

0.12

0.08

0.17

0.13

0.77

0.24

40

44

36

39

20

24

5313 (2934)

6275 (1500)

3626 (1870)

6962 (1800)

4550 (2770)

6416 (3000)

Proportion of households with a pension

35

43

32

39

21

25

Adult literacy (%)

64

73

71

81

86

90

Proportion of households female headed

39

49

32

40

27

35

Proportion of households owning land

Proportion of households owning livestock Mean (median) livestock value

Source: Roberts and May, 2000 Moving on from consideration of within-livelihood diversification, we now consider the relative potential for growth between farm and non-farm activities, and wider linkages between farm and non-farm activities within the rural economy. With regard to growth potential, Ngqangweni et al., 1999b estimate DRC’s for a range of smallholder farm activities in Northern Province, KwaZuluNatal, and Eastern Cape. While warning of difficulties in valuing land and labour inputs, and problems of local marketing of some commodities and of input supplies if production expanded significantly, they find that many activities have DRCs less than 0.6, suggesting real opportunities to exploit comparative advantages of emerging farmers. Very few studies have investigated farm/ non-farm linkages in South Africa. Machethe et al., 1997 recognise that low farm incomes, limited agricultural related non-farm activities, and the important role of large scale firms in agricultural input supply and in supply of consumer goods may undermine linkages (increasing the production linkages for and marginal budget shares of tradables). However, one study that attempted to estimate agricultural growth linkages in a rural area in the Eastern Cape (Ngqangweni, 1999, Ngqangweni et al., 1999a) estimated multipliers very similar to those found by Delgado et al., 1998 (see table 5.5). Table 5.5 Estimated multipliers, Eastern Cape Overall sample

Expenditure Groups

Location

Lower 50%

Upper 50%

Remote area

Small rural town

Tradables

1.00

1.00

1.00

1.00

1.00

Farm non tradable

0.35

-0.35

-0.14

0.06

0.21

Non-farm non-tradable

0.63

0.16

1.22

0.92

0.33

Total

1.98

0.81

2.08

1.98

1.53

Source: Ngqangweni et al., 1999a 46

As compared with our earlier review of rural livelihoods in sub Saharan Africa as a whole, therefore, South African rural livelihoods show both similarities and distinguishing features. Set in a more developed but highly inequitable national economy, national poverty is still dominated by rural poverty, with the rural poor showing a high degree of livelihood diversification. Access to higher earning non-farm income sources tends, as in other parts of Africa, to be differentiated, with limited access by the poor. However, agriculture appears to play a lower role and state transfers a higher role, in rural livelihood diversification in South Africa. The importance of agriculture as a ‘survivalist strategy’ was recognised but not emphasised in our earlier discussion of rural livelihoods in the continent as a whole, but this is clearly of major importance in South Africa. Three broad categories of agricultural activities within wider livelihood strategies are hypothesised as being of value: cash cropping, commercial farmers; subsistence farming as a major activity with non-farm activities earning cash incomes; and survivalist farmers, relying largely on non-farm income (including remittances and welfare) and food purchases, but retaining access to land and agricultural skills for social and insurance purchases. These different groups have different aspirations and constraints with regard to their agricultural activities, and consequently agricultural development strategies will need to address these in different ways, and will look for different benefits. Thus while poverty reducing linkage growth in the rural economy may justify investments in cash cropping and subsistence farming, this (although still an important consideration) may need to take more of a back seat to emphasis on ‘within household’ linkages reducing vulnerability for survivalist households. This raises important research issues: greater attention needs to be given to developing typologies of rural livelihoods taking account not only of the range of different farm and non-farm income combinations, but also of the interaction of these as regards timing in cash flow, and of the asset holdings and flows and vulnerability associated with them. Similarly, different strategies are needed for interventions supporting these different livelihood types, recognising the varied objectives, constraints and opportunities of these households. 6

The impact of HIV/AIDS on rural livelihoods.

6.1 Introduction When reviewing rural livelihoods in South Africa it is critical also to look at the various possible opportunities and threats that the human population is faced with. One such threat that is currently a national menace is the HIV/AIDS pandemic. HIV/AIDS is a big threat as it undermines all the efforts that government and the rest of the population to uplift the standard of living of the South African people. When planning developmental projects it is therefore important for policy makers and all relevant stakeholders to realise the extent of the disease, its possible consequences, and how it can be minimized. This section summarizes the demographic, economic and social implications of AIDS nationally, as there are few studies that are specific to rural areas. 6.2 History of HIV/AIDS in South Africa The first two cases in South Africa were diagnosed in 1982 with the first recorded death occurring in 1985: undoubtedly others before that went unnoticed and it is possible that the disease might have already been in the country, dating back to the 1970s. By the end of 1990 the heterosexual pattern was fast becoming the dominant means of transmission. By February 1993, all but two of the 46 cases diagnosed as AIDS from 1982 to 1986 had died. At the end of 1995, some 9000 cases had been reported of whom some 8000 were still alive. However, estimates from component projection models suggest that barely more than 5% of all cases were reported and only slightly more than 1% of AIDS deaths were recorded. It was estimated that over 1.8million people were 47

infected with HIV at that time. Women were more vulnerable to infection with 73% of all the reported cases in 1996 being female (Department of Social Welfare, 2000). In 1996, the Department of Health estimated that up to 3% of the total population and 7.5% of the sexually active population had been infected by HIV. By the end of 1996, approximately 700 people were becoming infected each day with the rate of new infections doubling every fifteen months. By 1998, these projections had become too modest, the estimated figure now stands at 3.6 million. It is estimated that by 2009, South Africa will reach of figure of six million deaths from AIDS. 6.3 Demographic implications of HIV/AIDS The impact of HIV/AIDS on the population structure is both dynamic and dramatic. There will be decline in the number of people in specific age groups, namely 0-4 year-olds and 25-34 year olds. Since HIV is primarily spread through sexual transmission, the majority of people will be infected during periods of peak sexual activity in their late teens and early twenties and will fall ill and die in their late twenties and early thirties. The concentration of HIV/AIDS in these age groups has important consequences. Over time, the reproductive age groups will move up the population pyramid and so, with increased mortality and deferred births, the structure of the population pyramid will visibly change. Significant gashes will be on the female side of the pyramid and the base will taper inwards due to lowered fertility. Life expectancy for South Africa is definitely being affected by the prevailing AIDS pandemic. Life expectancy is particularly sensitive to AIDS because deaths occurring among young adults, young children and infants result in a large number of years of life lost. Those born with the virus can expect to live for an average of 2.5 years, those born free of the virus but contract it in their youth or early adulthood have a life expectancy of 25 years, and for those who never catch the virus are expected to live into their late sixties. 6.4 Social implications of HIV/AIDS The social impact of HIV infections within households increases certain kinds of long-term expenditure. If infected persons are income earning, their illness and possible death will reduce the household’s income. Special nutrition and medical treatment and the inevitable funeral costs constitute a major financial burden on the household budget. This leads to a further degradation in the household economic status, adversely affecting the living standard and quality of life of surviving members. It is useful to differentiate between the infected and the affected members of the household. The affected are the household members that will care for the infected, and since many studies consider only the loss and cost of care of the infected, the impact of HIV/AIDS is often underestimated. However, when the loss of income (and possible decline in income) of the affected are also considered, together with the impact on markets of the consequent decline in purchasing power, the effects of HIV/AIDs are obviously disastrous. 6.5 Policy implications on strengthening the household’s coping capacity Several policy options can be adopted to strengthen the capacity of rural households to cope with HIV/AIDS. The overall aim of these policies should be to improve both the short and long-term well being of the households in ways which do not create dependency, and at the same time minimize the risk of members being infected with the virus. Mutangandura et al., 1999 propose various policies to help fight the epidemic and to uplift the welfare of rural communities. Part of the recommendations relevant to this work are improving agricultural production and promoting income generation and diversification of income sources. ♦ Improving agricultural production 48

Since a large number of rural households utilise some agricultural production for their livelihoods, strengthening the household’s agricultural production capability is one way in which the impacts of AIDS can be mitigated. Agricultural production ability of the household can be reinforced by improving their access to labor, land, capital, draught power, and management skills, promoting use of existing labor, and capital saving technologies, and by developing technologies that can make use of the available limited resources. Local agricultural may also allow more flexibility for affected carers of AIDs sufferers to combine productive labour with patient care. ♦ Income generation and diversification of income sources Programmes, which help improve and diversify the sources of income of affected households, can help mitigate the impacts of AIDS. Such programmes support household expenditure patterns and thus reduce further losses in household welfare. According to Donahue (1998), quoted by Mutangandura et al., 1999, the first line of response should be to mitigate the impact of AIDS on households by improving their income earning capacities. The aim is to maintain household expenditure patterns and promote savings. This, according to the author can be achieved through micro-credit projects that are typically small and short-term with rapid turnover. Promoting income diversification can strengthen household’s coping capacities. In risky agricultural climates, households with more diversified off-farm income are less vulnerable to food insecurity. Another strategy is to encourage crop diversification and promote a reduction in external input requirements. 6.6 Conclusion This sub-section highlights that great care has to be taken into consideration when planning rural development interventions for South Africa. Assuming that rural areas are affected by the disease at the same rate as the rest of the country, this implies that in the long term the population structure will be such that there are a lot of young orphaned children being kept by their grandparents as their parents would have been swept away by the pandemic. This will result in a serious shortage of active labor in rural areas with a capacity to contribute to the welfare of particular households. This implies, for instance, that any efforts made by developmental agencies to prop up agriculture must be done against background of labor shortages. Other implications from such high incidence of AIDS could be an increased failure by rural households to payback loans, as their productivity would be lowered by AIDS, due primarily to labor shortages. A number of other issues arise due to widespread AIDS related deaths, with major implications for how rural life moves forward. For example some members of the family within which a member has passed away due to the disease may feel the need to migrate to other places: if this were to happen on a wider scale, it could jeopardize a lot of existing developmental activities. There is therefore an urgent need to consider the pandemic and its various effects with great concern. 7

Overview of financial services in South Africa

7.1 Introduction We now turn to review the microfinance market in South Africa, including finance for small farmers. We concentrate on loans below R10000 but will also address larger loans (from R10000 to R100000) in a sub-section on small farmer finance. The demand for banking services by the lower income strata of the population is growing rapidly, due to a variety of factors. These include current low frequency of use of formal banking services 49

by the poor and MSE46 market; increased income of the lower income strata and income redistribution in favour of lower income people in urban areas; urbanisation; rising consumer aspirations; the rapid growth of the informal business sector and, perhaps the most important incentive in the last eight years, the relaxation of the Usury Act on loans below R6000 in 1992 and recently to loans below R10000. In reaction, there has been considerable growth in the microfinance sector, and the biggest source of growth is from the micro-lending sector. The conventional formal banking sector seems inappropriately structured to satisfy this rapid growth in demand and chooses not to engage directly. The “new” commercial banks, those that were formed recently through acquiring shells of banks with licences or by other methods are emphasising the micro-finance market, and are posting quite remarkable results in the process. How big is the micro-finance sector in aggregate economic terms? The micro-finance sector is still quite small in comparison with the formal banking sector, but growing much more rapidly. Its contribution to the national economy is probably not accurately reflected in the national data, because of the informal sector nature of many of the industry members, resulting in an underestimate of size and contribution to GDP. The finance, insurance, real estate and business services sector is a significant contributor to the South African economy, providing approximately 15% of total GDP in real terms in 1998. The total assets of the banking sector at the end of 1998 were R654 billion, with advances totalling R545 billion. In 1999 the size of the micro-finance industry was roughly estimated at R10 to R15 billion with advances at R10 billion (Econometrix, 1999). Recently, a comprehensive analysis of the sector estimated the size of the sector on the basis of current portfolio as R13 billion and the value of loans written per year at R25.8 billion (DTI, 2000). This analysis divided the sector correctly according to loan term and other characteristics, and then compiled an exposure on the basis of annualised figures. However, any estimate of the size of the micro-finance market should also consider the definition of that market. For the purposes of this study the micro-finance market is defined as the market reflecting transactions on loans below R10 000 of poorer people in South Africa. Poorer people are considered as those described in marketing studies as people in and below Living Standard Measurement (LSM) Class Five. In the following table the 1998 results of an LSM classification is presented. Table 7.1 Living standard measurement categories

46

Micro and Small Enterprise (may include small farmers) 50

LSM

Number

%

Female (%)

(16+) (‘000)

Unemployed (%)

Monthly income (R)

Savings (%)

Biggest problem

People in hhold

1

3,738

15

59

80

659

4

Unemployment

3.2

2

3,560

15

52

74

776

7

Unemployment

3.0

3

4,294

18

54

72

929

7

Unemployment

2.9

4

3,314

14

51

65

1,244

8

Unemployment

2.9

5

2,041

8

52

62

1,664

8

Crime

2.9

6

1,888

8

58

61

2,264

6

Crime

3.2

7

2,534

10

50

49

5,675

5

Crime

2.5

8

3,246

13

49

37

9,752

5

Crime

2.4

Source: Eskom, 1998

Micro-finance also refers to the savings transactions of the poorer portion of the population. It is important to consider both savings and credit as this provide a far more comprehensive picture of the financial market within which poor people operate. Of course there are more financial products pertinent in this strata of the financial market. Most important after savings and loans are transmission facilities that are an integral part of the lives of poor people, where support systems and sources quite often do not resort within the same location as households. In addition, insurance products in many guises are also important for the poor. The present usage of banking and financial services by MSEs is very small, with a minute section of the market presently being reached, estimated at less than 1%. This is the most unbanked area of micro-finance, and an area that is still in the introductory phase in the micro-lending industry. No study exists which provides a comprehensive overview and analysis of overall supply to the lower income and MSE market. Existing studies provide an overview of the institutions active in this market, but do not quantify supply in terms of clients reached. This study necessitates an overview of outreach, and therefore an indicative (rather than exhaustive) summary is provided in Table 7.2. In the next section each institutional form noted in the table is then discussed in more detail, focussing on the future role of the institution in the lower income and MSE market. The purpose of this overview is to gauge future formal support to this stratum of the full financial sector. After this section an attempt is made to estimate the rural portion of clients served by each institutional format. 7.2 Aggregate supply In the following table the present supply of micro-finance services in South Africa is described and presented. The assumptions and sources underlying the table are presented in discussion of each institution in the text following the table. Table 7.2. Summary of retail outreach in the micro-finance market in South Africa (1999/2000) Retail institutions

Public sector

Text Para 7.3

Source date

Loans Savings Rm Rm 330

1,646

Outlets Estimated Loan % Rural accounts (‘000) 2,440

36

78

Savings accounts (‘000) 2,840 51

Land Bank

7.3.1 Mar-00

Provincial parastatals

7.3.2

Jun-99

Post Office Outlets 7.3.3

Jun-99

Private sector

7.4

NGOs

7.4.1 Dec-99

Village banks

7.4.2 May-00

Credit Unions

7.4.3

Co-operatives

7.4.4 Dec-99

Apr-00

Commercial Banks 7.4.5 Dec-99

25

80

43

600

50

80

35

1,046

2,365

35

12,599

4,661

17,132

38

7,952

108

5

30

35

66

1

10

100

300

9 12

2,000 4,740 1.1

10 4,000

840

6 1,200

80

4,000

33

1,000

35

2,174

4,000

Retail stores

7.4.6

Apr-00

5,000

TEBA Cash

7.4.7

Apr-00

130

600

172

40

87

700

Private sector

7.4.8 Dec-99

40

45

20

100

25

33

5,600

Registered small 7.4.9 loans industry Pawn Brokers Informal sector

Apr-00

7,000

5,700

35

7.4.10 Apr-00

300

5,000

35

400

1,760 1.15 mill

35

150

25,000

35

1,560 325,000

35

6,500

250

200 800,000

35

8,250

13,329

8,067 1,17mill

35

7.5

Mashonisas

7.5.1

Apr-00

Burial Societies

7.5.2

Feb-99

Stokvels

7.5.2

Apr-00

Total

7.3

30

0

8,030

14,750

22,330

Public sector

7.3.1 Land Bank The Land Bank has largely reformed itself in terms of its political positioning. However, the Bank’s systems and products, as well as clarity on mandate, have not materialised as clearly. In the microfinance arena the Bank managed to launch their Step-up product successfully in April 1998. At the moment they are approaching 43,000 clients and the repayment level is around 80%. The product is being handled by the Start-Up group in Cape Town and the payments and accounts system is handled by First National Bank and the Postbank. The intention is to expand this product. It starts at R250 loans and the ceiling of the last repeat level has recently been increased to R20,000. The mere fact that the Land Bank is entrusting this product to an agent indicates its approach to co-operation and its use of third parties. They lack a service structure, since they have only 25 branches. These branches service their biggest income sources, namely individual farmers and co-operatives. The Land Bank does extend loans to emerging commercial farmers in their bronze range of products (maximum R50 000). 7.3.2 Provincial parastatals A wide array of failures and limited success is a good summary of the state of the provincial parastatal financial institutions (previously homeland institutions). We only concentrate on those 52

providing financial services. The majority of the retail development finance institutions at provincial level are in severe financial problems, based on a combination of inefficiency, bad policy and strategy, and a severe decrease in government transfers. At this stage the only successful institution is the Ithala Development Finance Corporation in KwaZulu-Natal. This institution is the second biggest public sector mobiliser of savings (after the Postbank). Except for Ithala, the rest of the parastatals are in demise. This will leave a tremendous vacuum, and already now we can see a large number of clients with no access to services. In the Eastern Cape a small rural bank has been formed from the ashes of two conventional homeland agricultural banks that were closed recently. The problem with most of these institutions is a lack of physical premises and outreach in that they have no branch network. The reformed parastatal banks are increasingly similar in their approaches to the enterprise lenders, except that they have a far greater existing investment and base to build from. Some have savings as a resource (like Ithala with approximately 800 000 clients) and others have institutional investors, which provides them with cheaper access to capital. Some of the largest lenders include Land Bank (which now has a microlending portfolio of 43,000 clients, but an outstanding book of only about R30 million) and Ithala. 7.3.3 Post Office Savings Bank The reach of Post Bank on a national scale is second only to commercial banks. However, not all branches of the Post Office provides saving services. Table 7.3 provides a summary of the type and volume of savings held with the Post Bank. Table 7.3. Summary of type, number and volume of accounts (Aug. 98) Products

Number accounts

of Share (%)

Balance (Rm)

Share (%)

Distribution Mostly

Savings Bank

1,755,880

85.9

699.6

73.5

Rural

Telebank

224,173

11.0

43.6

4.6

Urban

Savings Certificates

63,507

3.1

208.7

21.9

Urban

Total

2,043,560

100.00

951.9

100.0

The Post Bank has recently completed an investigation into the feasibility of providing loans to existing savings customers. This report has not been implemented yet. Rural people hold the majority of the Savings Bank account of the Post Office. 7.4

Private sector

7.4.1 NGOs As discussed earlier, NGOs an an important part of the financial market fabric in rural areas around the world. The problems of conventional development finance policy have made NGOs appear to be a promising option in the field of financing small and micro enterprises. This view was based on a belief in the ability of the NGO to reach target clients and to capitalise on the vacuum left by the closure of many specialised credit institutions in developing countries. However many NGOs providing financial services have needed continued subsidisation to deliver retail financial services to rural people. This poses serious questions regarding the appropriateness of NGOs as instruments for expanding retail financial services on a substantial scale. This view is echoed in the Strauss 53

Commission’s Final Report (Strauss Commission, 1996). Although the coverage of NGOs by the Commission was not exhaustive, they concluded that NGOs have very limited coverage in rural areas in South Africa. The Strauss Commission also calculated that these institutions run at very high costs, which impacts negatively on their sustainability. When comparing local financial services NGOs with international surveys, as seen in Table 7.447, indications of higher efficiency of the international institutions are visible. Table 7.4: Profitability of NGOs in South Africa compared with a Latin American Survey Small Enterprise Get Ahead Rural Finance Latin American Foundation Foundation Facility Comparison* Loan volume per staff member (R)

26,807

26,138

96,145

54,000

Interest earned/average portfolio

46

42

34

45

Interest paid/average portfolio

14

47

15

29

Gross financial margin

32

-5

19

16

Non-interest exp./average portfolio

134

153

120

72

* Adapted from Schmidt and Zeitinger (1995) and Strauss Commission (1995). An exchange rate of R3.00 = $1 was used for the 1992 figures. The interest earned/average portfolio percentage is based on an assessment of interest in real terms. The use of the Latin American Comparison is indicative, and are averages based on a study of 15 NGOs in the region.

Microenterprise lenders are a special group in the micro credit industry. Around the world, microfinance is associated with enterprise development finance, though in South Africa, microenterprise finance accounts for a very small portion of the microcredit market (consumption finance being considerably more important). Though some “consumption” lending may go towards financing productive activities, microenterprise finance comes largely from NGOs and Trusts. There is very little overt microenterprise finance from the commercial banking sector in South Africa, which has put a floor of R50,000 on enterprise lending. Khula is the major financier of enterprise finance. A recent report indicated R80 million outstanding by about 26 lenders, accounting for loans to approximately 45,000 microenterprises. This is a tiny portion of the entire industry. Average loan sizes among the micro-enterprise lenders are generally in the R3,000 to R6,000 range. 7.4.2 Village banks In 1994 the first Village Bank in South Africa was established in the North West Province. This was the initiative of the International Fund for Agricultural Development (IFAD) and the African Rural and Agricultural Credit Association (AFRACA). A village bank is in essence a savings and credit co-operative. Two more Village Banks were formed until 1996 and thereafter the growth in number was dormant until 1999/2000 when two institutions started a concerted effort to increase the numbers of these village banks. These two institutions, Finasol and the Financial Services Association (FSA) work independently with different sponsors and support groups to increase the number of Village Banks. FSA is the original institution that was involved in the formation of the North West Province village banks 47

The information in these tables are somewhat dated. The Small Enterprise Foundation improved its performance while both Get Ahead Foundation and the Rural Finance Facility closed their doors recently. 54

(albeit in a much less formalised format). FSA is being financed by the Department of Welfare to establish the infrastructure to support Village Banks and to ensure more Village Banks are formed. The United States Aid Agency (USAID) financed Finasol to do in effect the same thing. Finasol had its origin in an endeavour of the South African Sugar Association to reform or restructure its Financial Aid Fund. The USAID financing of Finasol is in question after a recent evaluation of Finasol. It seems as if there is a considerable demand for collective action formats at the grass roots level in South Africa. This is indeed the premise on which most of the activities of FSA and Finasol is based, and their recent success with the formation of more groups are evident of this local demand for financial services. They have assisted in the formation of nearly 30 village banks to date. 7.4.3 Credit Unions Credit unions have a long history in South Africa. After a restructuring of the movement the Savings and Credit Cooperative League of South Africa (Ltd) has now been active in Financial Cooperative Development in SA for the past three years. Currently SACCOL has 21 Savings and Credit Cooperatives (SACCOs) registered with it. Our SACCOs have a combined membership of over 6,000 members with member's savings of just over R10 million mobilised. The loan book has a balance of R9 million. All loans are issued to members who are also saving. The development of the movement has happened with minimal donor or Government support. For the past three years SACCOL has received no donor support, which, while admirable, has constrained their ability and capacity to serve the “poorest of the poor”, due to the high start-up and servicing costs involved in rural development. While SACCOL has not actively sought out rural area development, it continues to happen as requests for SACCOL's assistance in the development of Financial cooperatives keep coming in. 7.4.4 Co-operatives The dominant form of co-operatives in South Africa is agricultural producer co-operatives. Most of these provide financial services to members, mostly commercial farmers who are not the focus clientele of this study. However, some 40 of these producer co-operatives initiated project-based endeavours to assist small farmer development. Some of these services encompass financial services in the form of production loans to small farmers. The extensive infrastructure and existing member base of these co-operatives provide opportunities in the provision of financial services to members. One opportunity exists where commercial banks do not acknowledge the asset value of milk producer’s quotas provided by a sizeable local dairy cooperative. It makes sense that the dairy co-operative attaches value to milk quotas and therefore this presents an opportunity for linking this asset value with other financial services. The interest in financial co-operatives goes wider as farmers in the Eastern Cape/Karoo area have expressed interest in the joining of nearly 1000 farmers in a financial co-operative with the objective to negotiate more favourable priced funds for production than is possible by individual farmers. Due to the over emphasis of producer co-operatives and the retail financial services offered by commercial banks in the past very little attention has been given to financial co-operatives and its potential in commercial agriculture, and extending these services to developing agriculture. The declining importance of agriculture in the portfolios of commercial banks and the costs associated with the provision of retail financial services in rural areas may provide opportunities for collective action in the form of financial co-operatives in rural areas. In effect, the withdrawal of commercial banks in Canada from some farming areas gave rise to the formation of the strong Canadian Credit Union movement. This is something that may hold true for South Africa in the near future. 55

7.4.5 Commercial Banks In the drive to high profits and growth many of the commercial banks entered into mergers and acquisitions with other institutions in order to grow market share. The five largest banks in South Africa (Standard Bank, ABSA, First National Bank, Nedbank, and BOE) presently hold over 80% of the market share. Although the mergers and acquisition approach is in line with international banking trends, closer analysis shows that for South Africa the bigger banks exhibit diseconomies of scale. The banking environment itself has changed significantly during the last few years, with a clear move towards ATM and internet banking, away from the brick and mortar approach of the past. Old delivery channels of service are on the decline. In the market for deposits the banks are losing ground to smaller niche banks, which are more flexible and which can target specific segments more efficiently than the larger banks. Commercial banks only provide a limited range of services in rural areas, although they have the highest incidence of branches in the rural areas, together with the Post Office (Strauss Commission, 1996). South Africa also has a higher ratio of branches per population than elsewhere in Africa. However, this higher incidence of branches is skewed, with rural areas having approximately double the number of people per branch than urban areas. Recently a sharp decline in rural branches is also evident. It is estimated that whereas in 1995 approximately 50 per cent of the South African population had easy access to commercial bank facilities, this number has declined recently to approximately 30 per cent. South African commercial banks are comparable in terms of development and technology to the commercial banks of Europe and the US. Although there may be efficiency differences, South African banks are quite advanced in most aspects. It would therefore be possible to assess international trends in retail banking and expect similar trends from South African banks. Studies show that retail banking would change considerably in the future. Several issues have an impact on commercial banks. Little evidence exists to substantiate scale economies at a “macro” level for consumer banks and financial institutions. This explains the fragmented condition of the sector at present. The shape of retail banking at this stage seems to confirm that those forces pulling banks into smaller, more fragmented units appear to be relatively balanced with those that would naturally lead to consolidation. Consolidation is therefore slow, due to the fact that limited exploitation of scale economies is taking place. The number of products offered by banks is increasing, which drives up the total management cost. It seems that retail banking is inefficient in its current form. Technology is providing a counter to this trend of inefficiency. Technology revolutionises the moving and storing of money and bank product distribution. In addition to decreasing the costs of current bank practices, new technologies are also creating alternative distribution channels for retail banking products. This results in far cheaper ways to reach customers than by retail branches (in the United States telephone banking is more than 50 per cent cheaper than branch banking), and these non-traditional channels are gaining market share. It is especially the younger customers that are abandoning conventional branches for the alternative technology driven outlets. The question of the impact of these trends on rural finance also arises. The process and speed of adjustment and transformation of commercial banks will differ among countries. With increased globalisation, the main constraints on change will be country specific regulations and the ability to take up and apply technology. The banks that change the quickest will be recognisable in that they will sharply diminish their branch network, keeping only those branches with upscale or high net worth demographics. They will segment their client base and cross-sell only to those that clearly offer profit potential. They will increasingly interact with clients through electronic devices (the personal computer, the telephone, the Internet), and will clearly diminish the focus on rural areas, and especially remote rural areas. Areas without high net worth demographics, electricity and communication channels will not be part of the new way of banking. To all intents and purposes no major expansion of commercial bank activities should be expected in rural areas. One would 56

only expect banks to show more interest in rural communities, once technology has been made accessible to rural people. Small farmers and rural people in general will still save with commercial banks, however it will become increasingly costlier to do this as bank branches decrease in number in rural areas. Commercial banks do invest in the micro-finance market, however indirectly through purchasing shares in banks with micro-finance portfolios (the case of ABSA and Unibank) or by striking strategic alliances with commercial banks with micro-finance portfolios (the case of Standard Bank and African Bank). To summarise – ♦ Conventional commercial bank presence in rural areas will decline at an increasing rate, ♦ the farming portion of the commercial bank portfolio will further decline over time, ♦ emphasis will be on the agribusiness sector and larger producers, ♦ financial services to farmers will be provided by the broad retail sections of banks and farm specific portfolios will decline. This will be based on the diverse income sources of future commercial farmers. 7.4.6 Retail stores The furniture and retail store lenders are the latest entrants to the micro-finance market, primarily arriving since the creation of the Micro Finance Regulatory Council (MFRC)48. The furniture industry is already a R15 billion industry per year in South Africa, with about R10 billion of that being sold on credit. Historically, furniture sales have been made under the Credit Agreements Act, which restricted interest rates to the ceiling of the Usury Act, while allowing the seller to retain ownership of the goods sold as collateral. However with the advent of the MFRC and a clearer more transparent regulatory environment for micro-lending, many of the furniture lenders, as well as other retail stores such as Woolworths, have also entered the market. They have registered branches as micro-lenders and are actively promoting micro-loans to their regular, well-known clients. These lenders have a solid credit history on their clients and rely on a credit scoring methodology to assess risk and do not require debit orders or other deductions at the source. However most of their clients are salaried employees. 7.4.7 TEBA Cash Teba Cash is the institution responsible for handling the payments system of mineworkers in South Africa. They have 172 offices over Southern Africa. As they have a large compliment of savings accounts and an immediate market (the mineworkers), they recently applied for a banking licence, which has been granted recently by the Registrar. They provide over the counter loan products (very similar as envisaged for the Postbank) and have a loan book of around R130 million at this stage. 7.4.8 Private sector agricultural firms Started by the Financial Aid Fund of the South African Sugar Association 20 years ago, the trend for private sector processors to offer loans to producers has recently been boosted by the cotton ginners and vegetable processors and agents. Farmers are provided with crop establishment capital 48

The MFRC was launched in June 1999. It has been tasked by the Department of Trade and Industry to regulate the microlending market. Microlenders must register with the MFRC in order to provide loans with no ceiling on interest rates. 57

and in some instances production credit. Some institutions also provide extension services. This interlocking method of finance is quite common in non-farm MSEs and in contract farming. There is also potential for commercial farmers to contract small farmers to ensure throughput and turnover. 7.4.9 Registered small loans industry The quick growth of the micro loans industry is a product of a gap in the market for small loans. It is active in making short to medium term loans available to individual borrowers that normally fall outside the formal banking network, due to the inability of the loan applicants to provide conventional collateral. This gap has been filled by institutions that combined easy access arrangements to homegrown collateral models. These operators provide credit services to clients who can provide them with proof of employment and a bank account. The small loans firm withdraws the payments according to the loan contract after each wage or salary deposit made by the employer into the borrower’s bank account. This is an urban-based product that does not fit the profile of rural people, who often have inconsistent income patterns. It is not seen as a major growth industry in rural areas, except in larger rural towns. There are major drives to organise this industry. Several associations have been formed and registration with the newly established Micro Finance Regulatory Council (MFRC) is compulsory, if operators decide to conduct business within the exemption to the Usury Act49. It is expected that some of the larger more formalised operators (some of which are already listed on the stock exchange) will investigate the formation of banks soon. The available information in South Africa fails to differentiate accurately between rural and urban micro-finance. There are several different segments in the industry: ♦ Formal registered firms, which include commercial banks, financial institutions, section 21 (not for profit) enterprise lenders, developmental lenders, and the larger short term money lenders; ♦ Semi-formal money lenders, which include small unregistered money lenders who are doing it as their main livelihood and the pawnbrokers, who are not formally included in the money lending statistics (yet); and ♦ Purely informal moneylenders such as the township moneylenders (mashonisas) and stokvels, burial societies, and ROSCAs. These different lenders can be regrouped into different categories based on the type of lending that they are involved in. The first four are focused on “consumption” lending and lend only to customers with bank accounts and regular salaries (thus urban orientated). The other lenders, either developmental or enterprise lenders, generally base their repayments on cash flow from the productive activity. The source of repayment is the main differentiating factor between the two groups of lenders. 7.4.9.1 Short Term Cash Lender The short-term cash lender focuses on loans up to 32 days, or the next pay period. On average, these lenders charge an interest rate of 30 percent per month, all fees included. They are the largest number of individual institutions, but each branch tends to be relatively small in size, with a loan book of between R50,000 and R500,000. Their target market is clients with a net income of up to R2,000 per month. The average loan for these firms is about R500, as it takes into very strict 49

Government Gazette of Pretoria, 1 June 1999 (Vol. 408), Department of Trade and Industry, Notice in terms of Section 15A of the Usury Act, 1968 (Act No. 73 of 1968), No. 713. 58

consideration the capacity of the borrower to repay at the end of the month. Capital resources come mainly from their own sources, and occasionally from illegal loans from friends (illegal because this is not allowed under South African law). It is important to note that the rate charged by 30-day cash lenders applies to all loans less than that period or which are repaid on a weekly basis. This raises the effective interest rate of the loan. It is very important to note that even with the bank cards and pin numbers, the default rate on loans was in the neighbourhood of 2.5 - 5 percent. Now that the uses of bank cards and pin numbers has been eliminated, this rate has tended to double among the lenders. 7.4.9.2 Medium Term Cash Lenders There is often a fair amount of overlap between firms that are lending between one and six months and the term lenders. They have a mix of products that are in majority 30-day loans, but also a range of slightly longer-term loans reserved for their better clients. Average loans in the 1-6 month category can increase to a multiple of the person’s actual net take home pay, as they have more time to pay it off. Historically, these lenders have also used the bankcard with pin number as the repayment mechanism. With the restriction of the use of the bankcard, many of these lenders are experimenting with other forms of collection. Interest rates will vary by the term of the loan, but are usually discussed as a flat rate on a declining balance, which effectively increases the effective rate. The nominal interest rate may be between 30 percent (for 30 days) and 12.5 percent (per month for the period), but the effective interest rate is always greater than 20 percent per month. Because these loans are made to better-known clients, the default rate is generally lower, about 2.5 percent, without the bankcard. Since the short-term cash lenders and the medium term cash lenders are often lumped together into the same category, since branches often do both, it is difficult to differentiate between them. Overall, the estimates from the credit bureaux specialising in cash borrowers are that there are now roughly 3,500 – 4,000 storefronts in the country. This figure is down from an estimated 6,000 storefronts two years ago, and is expected to continue decreasing this year to a ceiling of 2,500 storefronts by the end of the year50. 7.4.9.3 Term Lenders The term lenders make loans for periods between six months and 36 months. The industry started through the use of Persal, the government’s central payroll system, using debit orders to get repayment at the source, before the borrower actually had a chance to see the money. Now that these lenders are saturating the market, they are branching out to the larger private companies to establish credit service relationships with them. This is the most rapidly growing segment of the industry, but which has often been restricted by cash to lend. The commercial banks are becoming increasingly involved in this segment of the market, buying up the larger term micro-lenders to develop their access to the market, while reducing the financial constraints on their lending partners. 7.4.9.4 Housing Lenders Housing lenders are closely associated with the term lenders. Most long-term mortgage loans are greater than the ceiling set for micro-loans and fall outside of the exemption. But there are a lot of micro-loans made in the name of housing, as it is the basis of access to a Persal code, which greatly facilitates repayment. Housing loans can also be secured by provident funds, effectively bringing 50

Conversations with Compuscan and MicroLenders Credit Bureau (MLCB) 59

the risk to near zero. Large banks are involved in housing finance, as well as small specialised lending boutiques. The National Housing Finance Corporation (NHFC) has a number of programmes to assist retail lenders to access finance to on-lend to borrowers for housing improvement. The NHFC promotes both urban housing as well as rural housing. Lenders in the housing industry include micro-lenders, small banks, social housing programmes, and non-bank financial institutions (NBFI) such as NGOs. Methods of securitisation include: none (unsecured), provident fund, payroll, and mortgages. The Rural Housing Loan Fund (RHLF) programme works with microlenders doing housing in the rural areas with unsecured loans, for which interest rates of 40 percent (on a declining balance) are common. For provident backed loans, margins above the cost of money are typically between four and five percent. 7.4.10 Pawn Brokers Pawnbrokers comprise one of the oldest industries in South Africa. Pawnbrokers use durable and semi durable goods as collateral against money that they advance to individuals in need of shortterm (generally less than 30 days) funds. These are often used to finance emergencies, or short term cashflow deficiencies in their daily lives and businesses. The advances are made against the pledged item(s) at a rate of 25-30 percent per month, and the borrower has up to three months to reclaim his items by paying off the advance, or else he forfeits the items that he has pledged. Between the time the pawnbroker has made the advance until the time the client comes to reclaim the item, the pawnbroker must store and maintain the item in original condition. If the client forfeits the item, the item then belongs to the pawnbroker and he is free to sell it as a second hand good. Roughly 35 percent of all pawned items are not paid off and collected. While there are many debates about the value that is applied to the items that are pawned compared to their real value, the pawnbroker must incur many costs. The pawnbroker must appraise the item, transport it to his shop, store it for up to three months (with the opportunity cost of capital), and, if the item is not collected, proceed with the sale, which could take several months depending on the demand for the item. There are roughly 5,000 pawnbrokers in South Africa, according to the Association of Pawnbrokers, which officially represents 1,500 of them. Pawnbrokers are registered under the Second Hand Goods Act, so are already regulated and pay VAT on all transactions. The cost structure for pawnbrokers differs radically from the cost structure for micro-lenders, due to their primary operation of storing and selling the second hand goods. Most loans are for one month or less. Estimating that the average portfolio outstanding is about R 60,000 per pawnbroker, there is an outstanding monthly balance of about R300 million. Taking an average term of one month (which may be too long), there is an annual turnover of R3.6 billion. 7.5 7.5.1

The Informal sector Mashonisas

The mashonisas are the informal sector lenders operating completely outside the formal sector. When there was no other alternative for borrowers, the mashonisas were their solution. The mashonisas specialise in short term loans, generally for 30 days. Interest rates run in the range of 50 percent per month, though no additional interest is charged if the borrower is late, effectively reducing the cost of lending. Mashonisas are often women with no other means of support who try to earn a living wage out of this job. They will often have 15-20 clients, borrowing an average of R150 –250 at a time, so individually, they are extremely small players. However, there are many of 60

them, estimated at between 25,000 and 30,000 around the country. Monthly earnings by a mashonisa are often quite small, in absolute terms, in the range of R2-3,000 per month. Interestingly enough, it appears that most mashonisas are very close to their clients and the clients are loyal to them. The borrowers’ survey, referenced below, demonstrates that most clients of mashonisas look to make sure that they can afford the repayment stream before starting their payments. The best estimates on the number of township moneylenders are approximately 30,000. This is the figure that has been developed by Du Plessis and is most commonly referred to by researchers. The research by Jimmy Roth on Township moneylenders in the Grahamstown area can be taken as a proxy for the average lender. This was confirmed by a recent survey in the North51, which found similar statistics. The typical lender has about 15-20 clients with a total outstanding book of about R5,000. Therefore, using this as a proxy, the township lenders account for about 600,000 clients on a monthly basis, equal to the number from the formal moneylenders, but their outstanding book is significantly smaller, about R150 million. On an annual basis this comes out to R1.8 billion. 7.5.2 Stokvels and Burial Societies In this instance, we use Stokvels to refer to the various informal financial institutions that capture member savings and then either save them or on-lend them to some of the members of the groups on a rotating basis. These include the Stokvels, properly stated, burial societies, and rotating savings and credit associations (ROSCA’s). According to the National Association of South Africa of Stokvel Associations (NASASA) there are an estimated 800,000 such institutions comprising about 8.25 million adults accounting for about R200 million a month in savings. They are governed under the Banks Act under the Stokvel exemption. This allows for stokvels, as member based organisations providing services just to their members that are members of the NASASA to be regulated by NASASA. Only 15,000 groups are officially registered with NASASA, but in actuality, as all Stokvels are only providing services to their own members, this is considered to be safe. In terms of calculating costs associated with lending, there are very few. Management of the associations is voluntary, and most of the funds are normally distributed to members at the time of the group meeting and deposit of funds. Since all repayments, with whatever interest, if any, go back into the group fund for redistribution to the members, the interest rates serve more as mechanisms for forced savings for the members. Interest rates may or may not be charged depending on the group and its operating procedures. The amount of savings and loans regarding stokvels reflected in Table 7.2 are merely estimates and represent current portfolio. Stokvels, in the form of ROSCAs rotate pay-outs, and thus mostly have no funds in savings accounts and very little in loans outstanding. The estimates in table 7.2 are likely to be conservative. The ESKOM consumer survey of 1998 (Eskom, 1998) confirms the existence of numerous burial societies. The surveys estimate 6.5 million South Africans are members of burial societies. We have estimated on a conservative basis that if each one of the members contributed R20 per month for a year the accumulated total will be R1.56 billion. This is likely to be a severe under-estimate as many of these societies have monthly contributions per member far in excess of R20. However, we also consider the reality that much of these savings would be deposited with commercial banks, the Post Office Bank or other formal savings facilities. Thus, double counting may be possible and the total savings figure should be treated with caution.

51

By Matome Kgowedi from the University of Pretoria. 61

7.6 Financial services in South Africa: discussion It is clear that the small loans industry (consumer lending) dominates the micro-finance loan market, while the commercial banks dominate the micro-finance savings market. This is not an exhaustive overview but rather an indicative overview. It relates coverage (outreach) with the high cost of intermediation as found by the Strauss Commission, 1996. Table 7.2 indicates a substantial outreach at first glance. However, if these institutions and their clients are studied more carefully, the following observations can be made: ♦ More clients have access to savings and transmission facilities than to credit facilities. ♦ Micro and small entrepreneurs have only limited access to loan facilities. ♦ Consumer lenders provide the majority of loans. ♦ The majority of savings instruments are still informal. ♦ In general, there exists a dichotomy between providers of savings and credit services. ♦ The majority of the services, indicated by the lack of presence of providers in rural areas, are offered in the larger towns. The analysis encapsulates the lack of services in rural areas and specifically those involving people not engaged in wage or salaried labour. However, in the rural context, individual incomes are not as important as household incomes, and thus some households could participate in the market described above. As discussed earlier, it is also quite difficult to divide households into farming and non-farming households. 7.7

Rural and small farmer finance

7.7.1 The recent history of rural and small farmer finance in South Africa Recent history can best be summarised by the work of the Commission of Inquiry into Rural Financial Services, the Strauss Commission, 1996. The Commission was active in 1995 an 1996 and made a range of recommendations of which the most essential are highlighted here. 7.7.1.1 The Strauss CommissionReport The Commission considered the major policy objectives of government and its brief, to contribute towards increasing access to financial services for rural people. These services were identified as transmission services, savings products and loan products for consumption smoothing and productive loans (for farm and off-farm activities). The Commission also identified that state grants would be necessary under certain circumstances. The Commission called for a detailed set of guidelines on the management of subsidies and grants, including their phasing out. A role was identified for the state to facilitate as well as co-ordinate the provision of financial services, with special attention to the needs of women. In this regard the Commission proposed that the Land Bank be tasked to fulfil this role. The way in which the Land Bank, the state and other institutions act must be in support of the market. At the same time the inherent weaknesses in the existing institutional fabric must be acknowledged, as these institutions are not able to contribute to the aims of rural reconstruction on a national basis without appropriate and active support. A review of the policies and activities of different state institutions, such as government departments and parastatals pointed to the dangers of lost development potential unless coordination takes place. The Commission therefore argued that the state must ensure that the improved availability of rural financial services is provided as part of an integrated rural 62

development strategy. This programme must include land reform and housing programmes, the upgrading of infrastructure such as the provision of water, electricity and roads, and capacity building, especially at the level of rural local government. The Commission also argued that the state must acknowledge the current gap in rural finance delivery, and an overlap in the prospective policy briefs of the state funded development finance institutions. International experience points to the success of multi-sector financial institutions at a retail level and a rich discussion exists on apex institutions internationally. However, there is very little experience, and little mention is made, of the advantages and disadvantages of multi-sector versus single sector wholesale institutions. The Commission accepted that all national development finance institutions have mandates to either continue or commence the extension of wholesale lending in rural areas: The National Housing Finance Corporation (NHFC) with earmarked dedicated funds; The Development Bank of Southern Africa (DBSA) funding rural infrastructure projects; Khula Enterprise Finance (KHULA) financing Small, Medium and Micro Enterprises (SMMEs) in rural areas. The Industrial Development Corporation’s portfolio already includes large-scale agro-industrial investments. The Commission supported the need for a financial institution at national level, such as the Land Bank, with both wholesale and retail activities and the responsibility to dedicate special attention to the needs of land reform programme beneficiaries. The Commission’s Final Report (Strauss Commission, 1996) emphasised that the Land Bank should focus on only agriculture as a sector, while the Interim Report (Strauss Commission, 1996) looked at agriculture as the primary focus, but did not exclude a multi-sectoral approach. A focus on a single sector can contribute to increase risk exposure for the institution. Especially in the South African agricultural setting it may imply increased covariant risk. The Commission also recommended that the future role of each of the provincial development corporations should be the result of specific consideration and consultation. This should be a joint approach of the national Ministry of Finance, the national development finance institutions, the provincial departments and the provincial development corporations. The proposed development council should provide guidelines for this purpose, without attempting to design detailed blueprints for implementation at the provincial and retail level. This approach would ensure that within a flexible framework, specific implementation strategies that reflect the reality of a specific setting or province could be worked out at provincial level, rather than being prescribed from the national level. Overall the same broad rules should apply, but implementation should be realistic and pragmatic. At the retail interface with rural clientele, the Commission recognised the important role of the Post Office in satisfying the most basic financial service needs, especially of the poorest, less mobile segments of the rural population. Commercial banks have, however, increased their outreach, and are well-placed to offer savings facilities in the larger rural towns and respond to loan requests from small business entrepreneurs. The NGOs have shown themselves able to reach a microenterprise level that the formal banking institutions do not yet serve. The Commission proposed a multi-pronged strategy at retail level rather than a single institutional strategy. The purpose was to accommodate flexible approaches and reflect the reality that the diverse circumstances in the country require diverse institutional approaches. Furthermore, the analysis done by the Commission showed that none of the current institutional structures proved to be highly successful in reaching rural clients, especially deep rural clients, and therefore it would be short-sighted to choose a single institutional strategy in this regard. In order to foster an integrated approach to both urban and rural development the Commission considered it essential to create an entity to co-ordinate and guide the activities of development finance institutions. This entity, referred to as the Development Council by the Commission, would be supported by a secretariat that could also play a part in structuring the reporting requirements for entities that make use of state support. All development finance institutions and 63

other recipients of state financing would be obliged to conform to appropriate reporting and accounting standards, as the introduction of minimum disclosure requirements would enhance transparency and accountability. This function should be monitored and implemented by the Development Council’s Secretariat. The Commission also recommended that legislation governing agriculture, banking and land, presently in force in different areas of South Africa, should be harmonised as soon as possible. The Commission gave attention to legislation of the hitherto un-legislated sections of the financial market (for example, NGO financial service organisations, village banks type organisations, savings and credit co-operative type institutions) and proposed that these complex issues should be attended to by specialists in the field. The Commission argued that a process actively guided by the state, should be launched as soon as possible to transform the rural financial services sector. This process should build on the strength of existing local level institutions in the private, public and voluntary sectors. The need for a wide range of service providers delivering different products and catering for the diverse rural financial needs was accepted. The Commission favoured fostering a much wider retail financial services network but the Commission was also concerned about longer term institutional sustainability as well as outreach. Under the recommendations, the Land Bank’s wholesale function was to be geared to fostering, nurturing, supporting and coordinating local and provincial level rural financial institutions, be they NGOs, development finance institutions, co-operatives, commercial bank branches, local authorities, or any other kind of institution seeking to render agricultural and agriculturally-related financial services. Essentially the Land Bank’s “wholesale” function should reach those “retailers” who seek to serve the individual and small group agrarian needs of people in the “deep rural” areas, for example small scale individual farmers, groups of female vegetable gardeners and small scale poultry producers who are far away from the relatively limited retail branch network. The responsibility of the Department of Agriculture to promote agricultural development was acknowledged. The Commission, however, recommended that the Department terminate the ACB and suspend current policy initiatives leading it into the role of a direct (central) wholesaler. The Commission identified female farm workers; male farm workers; landless, unemployed rural poor; pensioners; small-holders; contract farmers; rural businesswomen; rural businessmen and small scale employers; and large scale rural employers (including commercial farmers) as requiring special support measures to gain access to financial services. The Commission also drafted guidelines on the application of subsidies and institutional transformation. These recommendations of the Commission were put forward against the background of a set of guidelines, based largely on comparative international experience. The set of guidelines is important for this study and includes: ♦ views on access to financial services; ♦ the application of commercial principles; ♦ diversification in terms of different financial services and sectors; ♦ ensuring that existing capacity is not eroded and refraining from the formation of new structures (rural areas in South Africa are characterised by a plethora of institutions with overlapping competencies and a lack of co-ordination - existing structures should be adapted, rationalised and coordinated to serve the needs of reconstruction); ♦ flexibility where locational differences are acknowledged and incorporated in policy and strategy; 64

♦ incentive based approaches; ♦ improving information flows to ensure better decisions; ♦ the management of risk and the structuring of the rural finance system in order to minimise the impact of covariant risk; and ♦ the detrimental effect of direct intervention by the state in retail financial markets. 7.7.1.2 Discussion of the Strauss Commission reports The Commission Reports emphasised a paradigm shift and a movement away from the supply-led approach, rather than a detailed programme for implementation in order to increase access to financial services in rural areas. This shift was indeed necessary as most of the South African parastatal institutions followed a supply-led approach, as proved by the Commission. Although the Land Bank has been tasked to fulfil certain functions, no point exists where these functions can be coordinated with activities in the provinces. As the provinces have constitutional responsibilities, and the provincial governments are the only shareholders of the respective provincial development finance institutions, they should change their approach and transform these supply-led-type institutions. At this level a number of different activities are underway. It is not clear, though, whether these activities will get adequate political support, neither is it evident that they are coordinated, and no overall actions are taken to solve the core problem – namely, rural access to retail financial services. The Cabinet of the Republic of South Africa accepted the Strauss Commission proposals in August 1997 (Agricultural News, 1997). The Land Bank, which previously reported to Cabinet through the Ministry of Finance, now reports through the Ministry of Agriculture (Agricultural News, 1997). This continues the distinct agricultural emphasis of the Land Bank. The shift in thinking on the role of the Land Bank in the development finance system, but more specifically, as rural financier, also changed between the publication of the Interim Report of the Strauss Commission and the Final Report. In the Interim Report of the Commission it was emphasised that the Land Bank should provide wholesale financial services to retail institutions in rural areas across the whole spectrum of sectors, not only for agriculture. In the Final Report this recommendation was narrowed to include financing only of agricultural activities. This, of course, increases the exposure of the Land Bank to covariant risk, and is a contravention of the Commission’s own guidelines, specifically the guidelines regarding the minimisation of risk. The lack of clarity in the Final Report on coordination of rural financial issues aggravates the overall lack of clarity in the field of rural development co-ordination in the country. Between the Interim and the Final Reports the Commission changed its views on the responsibility to coordinate rural finance issues, which contributed to this lack of clarity. Lastly, the Commission gave a clear proposal on how the overall development finance system should be coordinated. It did not provide any indication of the relevance of the structure of the development finance system in South Africa. Although this was not included in the brief of the Commission, it did define a role for the Land Bank within this system, without questioning the system itself. 7.7.1.3 Conclusion The Strauss Commission considered the reality of the rural areas of South Africa, the demand for financial services and the current institutional structure and, against a set of guidelines, made a number of recommendations directed at different problems. Some proposals attended to the access problem and the expansion of retail financial services in the provinces. Another set of proposals addressed national level responsibility for providing capital and support to provincial level 65

institutions. Legal issues were also addressed, especially the legal status of women. A further set of recommendations aimed to structure a national level support system for rural finance retail institutions in the form of the Land Bank. The Commission also proposed a structure to achieve co-ordination between national level institutions. The most important paradigm shift brought about by the Commission was the rejection of a supplyled system of rural credit. Thus subsidised interest rates and credit programmes in which government decides what is best for the client were rejected. The Commission emphasised that a broad range of services should be accessible, namely saving and transmission facilities and credit products within a demand driven system. It further emphasised the importance of the retail network in rural areas to achieve access to these services. In this regard special mention was made of a role for the Post Office with respect to the provision of savings products, transmission services and other agency services. The Commission also stated that although subsidies are necessary, they should be implemented within strict rules and be finite in nature. The Commission supports an integrated approach, in which it is clearly stated that financial services on their own will be inadequate. The Strauss Commission recommendations pose several challenges. The most important is not the structuring of the national level financier - the real challenge lies at the retail level (Coetzee, G., 1998). Many countries have failed dismally in this regard and few successes can be noted. The challenge is essentially to achieve success within a specific set of guidelines, the most challenging being to achieve sustainable institutions that impact positively on development by ensuring wide access to financial services. The second challenge lies in achieving co-ordination between government departments and a host of institutions to achieve a positive development impact. Coordination between land reform programmes, agricultural development initiatives, provision of infrastructure and basic services and the provision of communication and energy services is an essential requirement of successful rural development. To this should be added the challenge to make local government work. Another prominent challenge is political commitment to meeting these challenges, one of which is to refrain from interfering in the rural financial markets. Past approaches saw the state using financial markets as a vehicle to distribute subsidies. The challenge is to leave financial markets to their real role, that of allocating financial resources to viable investments. 7.7.2 The current supply of and access to rural and small farmer finance Table 7.2 provided some information on which to assess the scope of rural and small farmer finance. Although it is very difficult to clearly distinguish between rural and urban microfinance, an attempt is made in Table 7.2 to estimate the proportion of outlets in microfinance institutions that one would find in rural areas: approximately 37% of 19,500 formal retail outlets and 35% of the 1,150,000 informal “outlets” are estimated to be in rural areas. This is much less than the total proportion of people living in rural areas in South Africa, but it is a fair reflection (even an over statement) of the proportional distribution of economic activity. However, the outlets in rural areas are skewly distributed, as most of the private sector outlets would be focused on larger towns and smaller cities in rural areas where a higher incidence of wage or salary earners prevail. This leaves a limited supply of financial services (judged on presence of outlets) in remoter rural areas. It is even more difficult to identify the extent of financial outlets and services serving small farmer agriculture, as it is expected that many diversions of loan funds occur due to fungibility on the one hand and general unavailability of sources of small farmer agricultural credit. On the other hand, one can also assume that a sizeable proportion of small farmer credit also gets diverted, as many “small farmers” are forced to obtain the credit for small farmer purposes, although they may have more pressing financial needs than for agricultural production. For households engaged in 66

‘survivalist agriculture’, as discussed earlier, much agricultural activity and investment is also cyclical as many of these “small farmers” will leave farming if economic conditions improve and wage income opportunities are more available. This leads back to the difficulties in defining ‘small farmers’ when this term may include small scale commercial farmers, subsistence farming and survivalist farming. The practical difficulties posed by this are illustrated, for example, by the results of surveys amongst small farmers and small businesses, where amongst ‘small businesses’ agricultural income comprised 1% and 9% respectively in Northern Province and KwaZulu-Natal, whereas amongst ‘small farmers’ agricultural income comprised 3% in the two provinces (Ouattara and Graham, 1996). This is not entirely surprising but it demonstrates difficulties with stereotyped notions of small farmers if a “small businessman” has a higher absolute and relative income from farming activities than a “small farmer”. We may also find that the majority of small farmers may be female. Since agriculture plays such a small part in the household income streams of most ‘small farmers’ it is often considered that loans for agriculture are quite scarce and targeted towards those that are more specialised, fulltime and (ironically) higher risk farmers. It is, however, not particularly helpful to refer to ‘small farmer finance’: it is more useful to focus more on ‘rural finance’, which is non-sector specific and emphasises rural areas as the locality of financial transactions, and then within this to consider demand for and access to funds for agricultural investment52. Accepting these definitional difficulties, it is estimated that less than 1% of the portfolio of the Land Bank is for small farmer financing (loans below R10000 to keep with our definition of microfinance). Approximately 30% of the outstanding small business finance book of provincial parastatals will be for small farmer finance. This will decline and should not rise above 25% as most of these banks cap their exposure to agriculture after the restructuring of these institutions. Applying these estimates to the data in table 7.2 gives a total portfolio in small scale agriculture of a little under R100 million, mainly from provincial parastatals, organisations that, with the exception of Ithala in KwaZulu-Natal, are facing serious problems and are unlikely to continue providing services. Even with their involvement, R100 million spread over 2.1 million small scale farming households represents average loan access of less than R50 per household: in fact limited access means that a small number of farm households are accessing larger loans, and the majority are unable to access funds from these sources. In addition to these public sector sources, it is expected that there exists some leakage to farming activities in the portfolios of some institutions without specific farming loan products. These institutions include NGOs, Village Banks, agricultural co-operatives, TEBA Cash, stokvels and even the microlending sector. It is however very difficult to assess the magnitude of this leakage. The Strauss Commission, 1996 analysed data on aggregate urban and rural debt. They found the following: ♦ Rural households have disproportionately few loans. Rural households owed 43% of all debt outstanding, even though 50% of the households were classified as such. 57% of rural households are without debt compared with 45% of urban households. ♦ Rural households have disproportionately small debts. The average loan of a rural household was R4 249, compared to R32 299 for urban households. Although rural households controlled 25% of household expenditure in South Africa and represented 43% of the number of households, they owed only 9% of the outstanding debt.

52

Even the definition of agricultural investment may pose problems, however, for example where equipment is used for both agricultural and non-agricultural activities or where consumption loans may allow a household to supply its own labour to agriculture rather than hire it out to provide for immediate consumption requirements 67

♦ Many rural households use hire purchase agreements and loans from shopkeepers. Hire purchase agreements or debt with shopkeepers accounted for 73% of the number of loans incurred by rural households and for 22% of the amount of debt, compared to 39% and 4% respectively for urban households. The amount of debt owed by rural areas to these sources is roughly half that owed by urban households to the same sources. Thus while large numbers of rural households have hire purchase agreements and loans from shopkeepers, the amount of such debt is disproportionately small. ♦ Informal lenders are important sources of credit in rural areas. About 22% of the rural households with debt have their largest obligations to informal lenders. Information on access to financial services specifically for agricultural uses is very sparse. Surveys of small scale farming and small business households reported by Ouattara and Graham, 1996 support the view that access to credit is very limited (see tables 7.5 and 7.6). The majority of households do not access or are unable to access credit, but where there is access this is from the development corporations (or provincial parastatal financial institutions) and the informal markets, the importance of the latter suggesting an inadequate supply of formal credit either in the form of no access or inappropriate financial products on offer. Coetzee, G. and Vink, 1991 suggest that agricultural credit use differs markedly between emergent commercial farmers and subsistence and survivalist farmers, with subsistence and survivalist farmers making less use of credit, and where they do, making more use of informal sources. It is not clear how far the low use of credit for agricultural investment amongst subsistence and survivalist farmers is demand or supply constrained. Most households are engaged in savings activities, and here a variety of informal financial arrangements was identified by Outtara and Graham, although not at a comparable frequency with urban areas. Most people who save money have commercial bank accounts (Coetzee, G., 1988, cited by Coetzee, G. and Vink, 1991, found that 55% of farmers surveyed in KaNgwane had savings, and 66% of these saved with commercial banks.) This indicates an effort to deposit savings since the majority of rural branches of commercial banks are in medium to larger towns, which are normally far away from these clients. Since commercial banks are not major lenders in rural areas, this also suggests that rural savings are being used to invest in other areas, increasing leakages from rural growth. The PostBank and NGOs did not feature in the survey results. In the case of the Post Bank this may be due to the fact that not all rural Post Office branches offer savings facilities and may also indicate the inefficient services provided based on outdated savings product technology (this has recently been addressed by computerising the manual centralised processing system of the Post Bank). Very few financial NGOs are active in rural areas and the penetration of the rural finance market by NGOs is therefore negligible. Morokolo et al., 1999 in a sample of mainly subsistence farmers in North West Province (of whom 65% had savings accounts with commercial banks) found that saving was positively correlated with both farm and non-farm income (the former being on average approximately 15% of total income) and age, but negatively with dependency ratio, with the main motivation for savings being precautionary (to provide liquid reserves against emergencies) and to provide for grandchildren’s education.

68

Table 7.5 Illustrative profile of farming households based on recent surveys in two provinces

Household characteristics

Northern Province

KwaZulu/ Natal

99

66

-

8.8

Cultivate all land (%)

79

59

Leased land out (%)

8

10

Borrowed farm land (%)

9

14

7

62

95

79

Water source unprotected

83

66

From farming

394

637

From wages

9308

14157

From micro enterprises

149

1294

From remittances

974

4251

From pensions

1489

2667

5

22

5

2

5

6

De facto female household head (%) Household size

Land tenure

Services and energy use (% Use grid electricity hholds) Cook with fire wood Gross annual income (R )

Farm assets - % of hholds Car/Bakkie with: Tractor Plough

Credit used to purchase durables (%hholds) and source (CB = commercial bank, PS = private supplier)

Car/Bakkie

3

CB

10

PS

Generator

1

-

3

-

Refrigerator

4

CB

33

PS

Television

2

-

20

30

61

49

45

0

-

4

21

81

0

30

19

Savings facilities used Banks (%hholds) and distance to Money keeper facility (km) Informal Farm inputs financed on Fertilisers credit (% of households) Seed

7

9

6

5

Farm equipment hired

3

0

Chemicals

1

6

Veterinary medicines

1

0

Livestock feed

0

0

Ploughing services

0

9

Transportation services

8

10

1

1

5

3

10

0

Miller

0

0

Stokvel, Savings club

0

0

Input supplier

0

3

Commercial bank

0

0

Credit source used to Local money lender finance farm inputs and Friend or relative implements (% of h/h) Development corporation

Source: Adapted from Ouattara and Graham, 1996 69

Table 7.6: Illustrative profile small business households from recent surveys in two provinces Household characteristics Business characteristics

Sector of operation (%) Multiple financing start (% )

sources business

of at

Place of work (%)

Selected income, input and output indicators (R )

Use of loans (%)

Savings (%) Electronic banking Handling of shocks or misfortunes (%)

Average age Female (%) Women who can sign contract without husband’s permission (%) Number of years in operation Number of workers at start Average value of business at start (R ) Number of current employees Sole ownership (%) Registered business (%) Construction Manufacturing Services Personal savings Retrenchment package Pension Loan from family or friends Loan from development corporation Income from other business Home Adjacent to home Market Place of delivery of service Average wages received per month Average remittances per month Average pensions per month Average business 1 income per month Average business 2 income per month Average business 3 income per month Average annual agricultural income Average value physical assets on hand Average value inputs per annum Average value output per annum Average labour expenditure p.a. Asset finance: Cash loans Asset finance: Hire Purchase Supplier credit Customer advance Customer loans Shopkeeper loans Formal : Mostly commercial banks Informal: Stokvel Informal: Burial society

Main principle cause: Stolen equipment (% of shocks) Main fallback mechanism: Personal savings (% of shocks) Source: Adapted from Ouattara and Graham, 1996

Northern Province 26 22 33

KwaZuluNatal 46 23 71

8 3.2 5 612 3.5 87 33 38 26 36 93 30 17 10 6 4 37 7 50 5 1326 739 414 3 601 3 796 7 721 1 250 17 913 1 067 92 031 25 360 10 14 7 16 18 56 92 26 64 3 21 52 67

10 4.5 4 509 4.6 94 23 36 27 38 73 12 1 5 6 13 34 5 33 27 1 102 1 360 438 2 057 1 979 2 800 10 886 12 995 12 703 58 044 41 380 10 16 11 29 16 77 76 22 18 52 31 24

70

7.7.3 Conclusion There are many similarities between the state and context of rural finance in South Africa and in the rest of sub Saharan Africa. Similarities are evident in the range of different types of financial institutions found in rural areas (although South Africa is unusual in the wide range found within one country); in the relative lack of financial services in remoter rural areas and serving poorer households and the needs of seasonal agriculture; in the lack of insurance services and common separation of savings, lending, and transmission in different organisations; in the decline of conventional supply led agricultural development organisations; in the absence of current working models to address these gaps; and in the need for wider developments in infrastructure, in property rights and in productive opportunities for the poor to invest in (although the extent of inequity in access to land, infrastructure and services in South Africa makes this particularly important). There are, however, also very important differences between rural financial services in South Africa and in the rest of the continent, which it is useful to discuss in more depth. First, although rural-urban inequality is found throughout the continent, with disproportionate incidence and severity of rural poverty, inequality in South Africa relates to more recent and more extensive disparities in investment in infrastructure and services between urban and former commercial agricultural areas on the one hand and rural areas in the former homelands in the other: previous minority governments’ deliberate policies favoring particular population groups at the expense of others meant that the people who were worst affected by the discriminatory policies were those residing in these rural areas. Past political and economic interference resulted in disempowered rural communities. This had direct effects on access to rural financial services as apartheid policies prevented black people from building up an asset base through land and property ownership, it prevented blacks from owning businesses in white areas and from settling where they chose, and within rural areas they were denied access to significant areas of land and to land titles. Thus millions of South Africans were not able to meet the set of conventional conditions required by formal financial intermediaries such as commercial banks, to provide finance in this market. These political and economic interventions led to increased market failure: in other words government failure led to market failure. The market itself cannot rectify these inefficiencies without an optimum allocation of resources. For the market to intermediate efficiently certain constraints must be must be alleviated (Coetzee, G. K. et al., 1994)m, but if the state is to act to alleviate these constraints without at the same time undermining markets, its actions must be carefully thought through and implemented, the minimum necessary, clearly defined, and transparent. As a consequence of this, one of the more disturbing characteristics of the South African economy is the presence of untapped human resources and underused natural and physical resources in the face of abject poverty. This contradiction is most visible in South Africa’s rural areas. In part they are overcrowded and unproductive; in part capable of producing agricultural surpluses, yet, overall, incapable of sustaining adequate rural livelihoods. The rural areas represent an opportunity for releasing South Africa’s creativity. As the Strauss Commission argued, the provision of financial services in these areas could be one of the more important mechanisms in a rural development strategy for unlocking this potential. The challenge then is to determine how financial services can be extended to rural areas effectively to achieve the greatest possible outreach and impact . Second, South Africa has a vibrant microfinance sector that has grown immensely over the last decade and includes a wide range of institutional forms. The sector’s growth has been remarkable: the current size of the sector on the basis of current portfolio is estimated as R13 billion and the value of loans written per year is estimated at R25.8 billion. It is however focusing on the wage and salary earners and comes mostly targeted as consumer credit and it differs from other parts of Africa in, for example, the extent of the involvement of retail stores in lending in rural areas. Thus 71

despite its vibrancy, diversity and growth, the majority of the sector’s activity take place in urban areas and the present usage of banking and financial services by rural MSEs (including small farmers) is very small, with a minute section of the market presently being reached, estimated at less than 1 %. This is the most unbanked area of microfinance, and an area that is still in the introductory phase in the microlending industry. Third, South Africa has a well developed national communications infrastructure and financial institutions making use of advanced technology to an extent that is rare in other parts of Africa. Thus commercial banks are in many ways equivalent to banks in the North (for example in terms of systems, technology, services and management), and electronic and card systems are widely used by commercial banks, PostBank and other organisations. This offers opportunities for the development of rural financial services and systems not currently viable in other parts of Africa. Fourth, there is a strong network of commercial bank branches in rural areas, and these are used for savings and transmissions services by many poor people and even their informal savings activities eventually land at commercial bank deposit accounts. However, accessing this network is time consuming and expensive for the majority of poor rural people (as banks are not well distributed in the poorer and more densely populated areas), and with changes in technology and relative profitability of different services, the rural branch network is contracting. This increases the transaction costs of rural depositors and contributes to less efficient financial markets. The imbalance between deposit taking and lending by commercial banks in rural areas also leads to a financial drain from rural areas which may result more from differences in commercial banks’ transaction costs in lending in these two markets than from differences in returns to borrowers’ investments. Fifth, rural finance in South Africa received comprehensive attention at the policy level through the activities and reports of the Strauss Commission. Many of the recommendations have been implemented, although implementation is slow. There is, however, strong interest in some areas in improving the supply of financial services to meet the demands of rural people, and willingness to develop and try new models. Sixth, rural livelihoods in South Africa display a greater degree of diversification away from agriculture than is generally found elsewhere in Africa, with strong linkages within households across different rural and urban areas, some households heavily dependent on remittances or state transfers, and more ‘survivalist’ farming. However, as in other parts of Africa there is a high degree of diversity between households, with barriers to entry preventing poorer households from engaging in higher return farm or non-farm activities. Seventh, South Africa is one of the countries in the world with the highest incidence of HIV infection, although some other countries in Southern Africa suffer from similar rates. This has very serious implications for the welfare and productivity of rural people, and for their demands for different financial services and ability to participate in financial markets. These ‘special features’ of rural financial markets in South Africa present a number of opportunities that do not exist to the same extent elsewhere in SSA. Opportunities arise from the strong performance of the micro-finance sector; the variety of private, formal and informal institutions; the sophistication of commercial banks and some other institutions; the strong national infrastructure; the diversification of rural livelihoods and strong urban/rural linkages and the framework provided by the Strauss Commission report. At the same time there are a number of threats to the development of rural financial services: the withdrawal of the conventional formal private sector institutions from rural areas; emphasis on wage and salary earners by the “new” formal private sector financiers; slow progress in increasing access to land, infrastructure and services; the impact of AIDS; and, with the low shares of 72

agriculture in rural livelihoods, difficulties in identifying productive investments that will lead to significant growth in the rural economy. Taken together, these opportunities and threats suggest an agenda for research and development in rural financial services. Development is needed of both financial service products and technologies and of the institutions offering them. In view of the risk aversion of the poor and of their proneness to emergencies demanding cash or curtailing income, greater attention needs to be paid to understanding their demand for microinsurance services and to the development of institutional models and insurance products and technologies that can meet this demand. This may be complicated by (paradoxically) the low share of agriculture in South African rural livelihoods (and thus the greater importance of idiosyncratic risk) and by the spread of AIDS. Micro-insurance and health is, however, receiving increasing attention in the wider literature. Similarly, improved access to savings services needs further attention. Current reliance on commercial banks reflects both high degrees of trust in them and the importance of the transmission services that they offer. However their likely withdrawal from rural areas, and the high transactions costs in depositing and withdrawing funds suggests that more local institutions need to step and fill the gap, but these must also offer transmission services and there must be no loss of security of deposits. One response to this scenario is to recognise that if rural people do not turn to self-help organisations, no real volume of financial services can develop in rural areas. This emphasis on self-help schemes is evident from the richness of informal financial groups in rural areas. These groups or collective action endeavours come in many guises and take place in many different locations, ranging from the joint financial endeavours of the employed to those of the unemployed and destitute. Coetzee, G., 1998 suggests that village banks linking in with commercial banks may provide on model for achieving this. FINASOL is another model seeking to achieve the same goal through franchising of financial services and management systems to village associations. High levels of technical expertise and sophistication within the South African financial sector offer exciting opportunities here. However, these need to focus on the use of appropriate technology to overcome the basic problems of establishing client confidence and loyalty together with effective but low cost transaction and portfolio management systems in handling small deposits, loans and transactions. Further issues in the development of locally owned and managed institutions is the balance between savings and loan portfolios of members, the relative importance of locally mobilised deposits and outside funds in lending, membership of different household types, management structures and control, and the speed at which these institutions can expand to meet demand. These issues affect the incentives for prudent financial management, management skills and effectiveness, relative benefits to different (poor) rural groups, and the ability of the institution to withstand and provide services in the face of covariant shocks. It is also important to recognise the diversity of rural livelihoods and of the demand for different financial services. With regard to support for agricultural investment and productivity, further investigation is needed of the aspirations and opportunities for and constraints to agricultural development amongst different farmer types in different areas, with regard to differentiated and targeted development of insurance, savings, transmission and loan services for commercial, subsistence and survivalist farmers, for example. 8 Applying the SL Framework We now turn to consider initial lessons gained in this review regarding the application of the Sustainable Livelihoods (SL) framework to analysis of the access of subsistence agricultural producers to financial services. The SL approach is taken to refer to a broad approach to 73

development involving six core principles in development, which should be people centred, participatory, sustainable, multi-level, dynamic and involve partnerships. The sustainable livelihoods framework emphasises, inter alia, an emphasis on understanding the livelihoods of the poor in terms of their access to assets; their vulnerability context; livelihood strategies and outcomes; and the policies, institutions and processes affecting access to assets, vulnerability, and returns to different activities (Carney, 1998). 8.1 Strengths of the SL approach Carney’s discussion of the advantages of the SL approach is generally borne out in reports and reviews of the applications of the approach to livelihoods analysis and review. Ashley, 2000, for example, from experience applying the SL approach to analyse how rural livelihoods interact with natural resource management, suggests the main benefits offered by the livelihoods approach are its explicit focus on poor people’s concerns, synthesis across different disciplines, broadening of scope of enquiry, outline of analytical relationships and contribution to practical recommendations. Turton, 2000b makes similar points, adding that in a study identifyng options for support to rural livelihoods in Cambodia it usefully highlighted the importance of macro-micro links. Turton, 2000a used the framework to focus attention on the relations between watershed development and capital assets, and livelihood strategies (including diversification and migration) and outcomes for the poor. Nicol, 2000 finds the approach useful in analysing the important attributes of access to water for poor households, emphasising the multiple roles of water as an asset and good and institutional and sequencing issues in promoting demand led improved access. The formal SL framework was not used explicitly in the structuring of this review, but its principles and elements underpinned much of the initial thinking and subsequent development of the project. Specific elements used include its holistic view of livelihoods and the diversification of different activities within livelihood strategies; the interaction of financial, natural, physical, human and social capital; the importance of vulnerability; the importance of institutions; and the diversity between households with different asset holdings and livelihood strategies. The relationship between livelihood diversification, vulnerability, and households’ different income and expenditure streams was recognised as critical at the inception of the project, and remains a key focus of the project. This relates strongly to the importance of institutions in the supply of financial services and of linkages between the macro environment and the supply and demand of financial services for households with different livelihood strategies. These, together with the people centred approach to development, are aspects of the SL approach which will be applied and further developed as the project begins field work. This review has also, however, identified gaps and weaknesses in the SL approach as we currently understand it, to which we now turn. 8.2 Gaps and Weaknesses in the SL approach We will not attempt to develop a comprehensive critique of the SL approach here: discussion will be limited to issues relevant to the application of the SL framework in this project. It is important, however, to set this in the context of difficulties in application of the approach recognised by other authors. These difficulties may be classified as problems with the basic structure of the framework, gaps in the framework (for example the omission of political capital, empowerment or the role of the private sector and markets), high costs as a result of high resource intensity in data gathering and analysis, and difficulties with quantification (see for example [Ashley, 2000, Nicol, 2000, Hobley, 2001). Some of these are, of course, contradictory in that filling gaps is likely to increase complexity and resource demands, but an important principle of the SL approach is that it should embrace flexibility, and thus in some circumstances simpler cheaper methods will be appropriate, 74

whereas in others it may be necessary to develop the approach (and incur extra costs) to address particular issues in more detail. In the process of this review of the problems and potential for seasonal finance in rural livelihoods in sub Saharan Africa, several difficulties were recognised for the ‘standard’ sustainable livelihoods framework: some of these are specific to our interest in seasonal finance, whereas others are of more general relevance. We briefly discuss these in this section: in section 8.3 suggestions are made as to how the SL framework might be modified or extended to address these issues. First, in the review of the role of agriculture in rural livelihoods, the framework was very helpful in examining the diversity of activities within household livelihood strategies (see section 3.2.1). However, as the discussion in the remainder of section 3 shows, this ignores, and can even obscure, wider consideration of the extent and effects of diversification between households within the rural economy. Thus in general discussion of livelihoods without a clear typology of different livelihood strategies, if averages of income share from different activities are taken to represent diversity within livelihoods, this may over-estimate the importance of within-household linkages between activities and under-estimate the importance of between-household linkages. This draws attention to a more fundamental gap in the livelihoods framework: there is no conceptual recognition or analytical consideration of the dynamic effects the sources of growth in a rural economy and of how different types of growth differ in the ways they may expand (or limit) livelihood opportunities of different livelihood groups. This is related to insufficient attention to market processes (see for example Hobley, 2001). Second, the framework is very strong in recognising the importance of assets to rural people, and in distinguishing between natural, social, physical, human and financial capital. In some circumstances, however, it may be important to determine the roles that particular assets play: are they, for example, more important in reducing or guarding against the effects of vulnerability, or in the role they play in production? What are the trade-offs between these roles, and how do these trade-offs affect specific investment decisions and broader livelihood strategies? When considering rural peoples’ demand for and access to insurance, savings and borrowing services, these are important questions. A third area where the framework, or the thinking around it and its application, could be further developed relates to the role of institutional arrangements. These are discussed in a livelihoods context by Morrison et al., 2000 but otherwise receive very little mention. However, in the context of access to financial services these institutional arrangements are critically important. Their importance is not, of course, confined to access to financial services: institutional arrangements are more widely important in affecting access to other services and assets (for example input and output marketing, land, water, physical capital, etc.). 8.3

Extending the Sustainable Livelihoods Framework

8.3.1 Extending the SL Framework: Growth and Linkages It is not possible to represent the importance of ‘engines of growth’ in the rural economy and the role of linkages or multipliers in determining alternative impacts on the poor with a simple ‘add-on’ to the SL framework. The importance of these linkages is, however, increasingly recognised in a range of different situations, for example the greatest poverty reducing impacts of micro-finance may not be through improving access of the poor to micro-finance but through increased demand for labour by less poor borrowers (Mosley, pers. comm.). However, even if linkages cannot be directly incorporated in the framework, it is important that the concepts of demand are introduced, 75

whether such demand is mediated by market or non-market institutions. This is attempted in Figure 8.1, which presents a simple modification to the basic SL framework. Figure 8.1 contains a small number of modifications to the basic SL framework, but it is not intended to lose anything from that framework (omission of details from the PIPS box and the Livelihood Outcomes box reflects the draft nature of figure 8.1 rather than any need to exclude them for a more developed version). First, in the top right hand corner the influence of ‘demand’ is introduced. This affects the relationship between outputs of livelihood strategies (livelihood strategies are broken into activities and outputs). Demand may be mediated by market or non-market relations (and the distinction between these may often not be clear) and has two components, local and wider (to represent nontradables and tradables respectively). Demand for non-tradables will be influenced by local output whereas net demand for tradables will be influenced by external demand and supply. Livelihood outcomes can also affect local demand (for example through consumption linkages), and demand may also affect asset values and access. The figure distinguishes between demand on the one hand and the structure and operations of market and non-market relations on the other, the latter being properly described by policies, institutions and processes (interactions between these are not shown in the interests of simplicity). A second modification is to link the vulnerability context and policies, institutions and processes together to show how both of them can influence access to assets, allocation of assets to activities, asset productivity and activity outputs, and the contribution of outputs to livelihood outcomes. The standard SL framework appears to place more emphasis on the effects of vulnerability on assets, and less on its effects on livelihood strategies (their activities and outputs), although these are often very important.

LIVELIHOOD ASSETS

DEMAND LOCAL WIDER market/ non-market LIVELIHOOD STRATEGIES

Physical, Social, Natural,Human, Financial

ACTIVITIES

VULNERABILITY CONTEXT SHOCKS, TRENDS SEASONALITY

OUTPUTS

LIVELIHOOD OUTCOMES

POLICIES, INSTITUTIONS & PROCESSES

Figure 8.1. Introducing Markets into the SL Framework It is suggested that these minor amendments introduce into the framework in a simple and accessible way the importance of considering the dynamic relationships between rural livelihoods and markets. 76

8.3.2 Extending the SL Framework: the Functions of Assets53 As suggested earlier, in some situations the analysis of assets in rural livelihoods needs to examine the functions of different assets within the asset portfolios held by the poor. This may need to move beyond categorisation of the types of capital (as emphasised by the SL pentagon) to identify priorities for policy and other interventions supporting expanded access to assets. This prioritisation should relate poor peoples’ access to different types of asset to the functions of those assets within their livelihood strategies, identifying the roles of different assets within those paths. This requires a more systematic analysis of asset functions in the livelihood strategies of the poor. Figure 8.2 presents a conceptual framework bringing together the relationships between the different livelihood functions of assets. The central vertical axis of the figure shows production/ income activities and processes employing productive assets (in the top left corner of the diagram) to generate resources for consumption and social reproduction. The shortcomings of simple income approaches to defining poverty and well-being are highlighted by moving down the diagram to unpack (a) the multiple and often intangible elements that make up people’s wellbeing and (b) the need for flows of resources to match patterns of consumption necessary to maintain wellbeing. For most livelihood activities production or income is discontinuous and to a greater or lesser extent uncertain, not matching resource demands for consumption, which are made up of reasonably continuous elements, time-specific direct consumption elements, time-specific investment in consumption assets, and time neutral investment in consumption assets. Consumption demands are also subject to uncertainty (for example as a result of social obligations or sickness). Households therefore engage in saving and borrowing activities to smooth resource supplies and match them to consumption needs. This is shown by saving and borrowing activities on the right hand side of figure 8.254. The layout of figure 8.2 stresses the dynamic relationship between assets with different functions (in the corners of the figure) and various livelihood activities and processes (in the centre of the diagram) in the pursuit of wellbeing. This does not imply that there is a clear distinction between consumption, productive and convertible assets, as the same asset may fulfil all three functions to some extent, but assets will differ in relative effectiveness with regard to each function. Savings in a highly convertible asset like cash may give no direct production or consumption benefits. Insurance investments may take many forms, and may or may not be associated with production or consumption benefits. Savings in an interest bearing deposit account do yield an income (provided that the real interest rate is positive). The productivity of savings in livestock will vary with markets, management, the type of livestock, etc. This analysis of asset functions in livelihood strategies brings together a number of complex components and attributes of livelihoods. Asset/ process integration: Perhaps the most obvious feature of the framework presented in figure 8.2 is its emphasis on integration between four different types of process (production/income, investment, saving/cashing, and borrowing – insurance needs to be added here) and four associated functions of assets in effective livelihood strategies. The balance between these processes and asset functions, and the balance between the different nature of processes and types of capital asset (human, natural, financial, social and physical) on which they are based will vary between different situations, and one may hypothesise certain broad patterns of change associated with improved livelihoods. Such hypotheses can have important practical implications, and analysis of the ways that different types of process and asset relate to livelihoods may provide a constructive framework 53

This section draws heavily on Dorward et al., 2001

54

Rutherford’s terminology of ‘saving up’, ‘saving down’ and ‘saving through’ (Rutherford S, 2000) might usefully be applied here, but the figure need further adaptation to include insurance or saving through activities. 77

for developing cross-sectoral and inter-disciplinary approaches to research and action in the support of livelihood development, and a basis for developing indicators for livelihood change – an important issue in the operationalisation of the Sustainable Livelihoods approach and development of relatively simple M&E indicators taking account of (complex) changes in assets, income and vulnerability. Figure 8.2 Asset functions in livelihood strategies

78

ASSET FUNCTIONS IN LIVELIHOOD STRATEGIES

Productive assets physical,natural, social, human, financial productivity, security, costs, life

Investment

Production/ income output, timing, robustness, volume

Convertible assets physical, financial, social, natural? costs, security, convertibility, life

Saving & ‘cashing’

Liquid assets / exchange

Borrowing & repayment

Consumption assets physical, natural, social, human, financial utility, security, costs, life

Consumption / Reproduction food, shelter, clothing, leisure, cash, etc. volume, timing, security, control

Debts physical, natural, social, human, financial costs, life, security

WELLBEING Health, Nutrition, Shelter, Clothing, Independence,Status, Dignity, Security, Leisure, Education, Friendship, etc. (requirements demographically socially, & individually determined)

ENVIRONMENT

Natural (biological, physical), social, institutional, political, economic. policies, markets, services shocks, trends, seasonal & other cycles

Key Flows & transformations Overlaps/ blurred distinctions

Activities/ processes

Assets components attributes

79

Asset attributes: Recognition of different asset functions leads onto consideration of the attributes determining the effectiveness of assets in fulfilling these functions. Relevant attributes for each function are listed in figure 8.2, and a number of attributes are, as one would expect, relevant to more than one asset function. This is illustrated in table 8.1, which provides more detail on these asset attributes. Table 8.1 Asset Attributes and Components Main Attribute

Components

Productivity

‘Normal’ productivity, variability, sensitivity, resilience under different conditions, and probability of different conditions

Utility

‘Normal’ utility, variability, sensitivity, resilience under different conditions, and probability of different conditions

Security

Risk of theft, loss of control or access, susceptibility to disease or other ‘natural’ event. For debts: risks to collateral or collateral substitutes

Holding costs

Maintenance and input costs, under normal conditions, variability under different conditions, and probability of different conditions Depreciation in time and in use, under normal conditions, variability under different conditions, and probability of different conditions

Life

Expected period over which asset will be held, under normal conditions, variability under different conditions, and probability of different conditions

Convertibility

Exchange costs, under normal conditions, variability under different conditions, and probability of different conditions Access, under normal conditions, variability under different conditions, and probability of different conditions Lumpiness

Complementarity

Effects on and of other assets and their functions

Ownership/ control

Private (individual, household), communal, public

Some of the distinctions made between attributes or their components are a little arbitrary, but the breakdown in table 8.1 is intended to ensure that the important attributes are allowed for. Under the attribute ‘holding costs’ are included the various costs involved in holding and using an asset for production or consumption: costs involved in acquiring it or disposing of it are allowed for separately under ‘life’ (which describes the frequency of exchange) and ‘convertibility’ (which allows for the various costs involved in converting or exchanging an asset), to distinguish between the production and consumption costs in use from costs of converting or exchanging assets. While it may be expected that high productivity and utility, low holding cost and longer life will be almost universally desirable asset attributes, what is desirable in terms of convertibility is more ambiguous and contextually determined. High convertibility should give more flexibility and lower costs, but people (not just the poor) often impose disciplines and protection on their saving for longer term investment or for future consumption by placing savings ‘out of easy reach’ to prevent 80

them from being cashed and spent by themselves or by others (for example relatives or neighbours) on more immediate consumption needs. Examples of this abound, from pension schemes in more developed economies to the poor investing in ‘lumpy’ assets such as jewelry or livestock or in informal (group or other) savings schemes in economies with less developed financial markets. Shipton, 1990 (p16-17), for example, writing about rural communities in the Gambia states that ‘money is seen as something to get rid of ..... to convert into longer lasting forms. Several features make money an unstable form of wealth in the Gambia: its nearly universal fungibility, its divisibility and its portability. These features make money contestable. Everyone needs it for something...... Rural Gambian saving strategies, then, are largely concerned with removing wealth from the form of readily accessible cash without appearing antisocial’. People thus try to maintain a balance between more and less liquid convertible assets to enable them to maintain their regular consumption requirements while at the same time protecting savings for larger scale investments or to meet large expected consumption needs in the future. Indeed, rather than receiving income from interest on their financial savings, poor people often pay charges to deposit savings. If low convertibility is preferred, then this may be achieved by investment in assets that are relatively lumpy or have high transaction costs or barriers to access55. Lumpiness is associated with reduced diversification, with consequent risk implications (of loss or of forced sale at low price) and with difficulties (for the poor) in building up sufficient other assets to obtain it. Barriers to access may involve some loss of control over a resource, and this too carries risks. High transaction costs may be a deterrent to ‘cashing’ small asset values, but will be important if and when this may be necessary. Similar ambiguities may be encountered as regards asset control. While private (individual or household) control may allow greater capture of benefits from an asset, this may be offset by lower productivity or utility, lower security and greater holding costs. The balance between these attributes is likely to vary with the overall asset holding and the livelihood strategy of different individuals. Consideration of asset functions and attributes in these ways may be helpful in examination of the relationship between holdings of different assets. It may help to pose and answer questions about, for example, the roles of and relationships between access to loans and investment in different forms of social capital, livestock, jewelry, cash, insurance, and savings deposits with different terms and conditions within the portfolios of households with different livelihood strategies. Similarly it may also help to structure consideration of the attributes of different financial services. This conceptual framework for understanding the role of particular assets in livelihoods is currently being developed and applied elsewhere, in the specific context of understanding the role of livestock in livelihoods of the poor56. 8.3.3 Extending the SL Framework: Institutions57 and Institutional Arrangements Although institutions are given a prominent position in the SL framework, there is recognition that the practice of institutional analysis is weak: Hobley, 2001, for example, identifies ‘gaps’ in the PIP toolbox with regard to understanding of existing institutions and of linkages between traditional and informal institutions on the one hand and formal institutions on the other. Morrison et al., 2000 suggest that the distinction between the institutional environment and institutional arrangements 55

This may be relevant to earlier discussion of the importance of commercial banks in rural people’s saving portfolios, despite the high transaction costs involved in depositing and withdrawing funds. 56

Under a DFID financed Livestock Production Programme project commencing April 2001, ‘Understanding small stock as livelihood assets: indicators for facilitating technology development and dissemination’. 57

The term ‘institutions’ is used here in the sense of ‘the rules of the game’, as discussed earlier in section 4.1. 81

can be useful in this and identify three possible entry points to reduce transaction costs and risks: development of the institutional environment, development of more effective institutional arrangements within the existing institutional environment, and development of infrastructure and services to improve communication and reduce its costs. The importance of the last of these is widely recognised, with investments in roads and telecommunications for example, while the first, development of the institutional environment, is the concern of current emphasis on governance and, with regard to financial services, development of financial regulations. Developing institutional arrangements is much less well understood in the SL approach, although it is in practice the focus of development of financial technologies and products in the microfinance sector, and the focus of this project. Morrison et al., 2000 draw a distinction between two related functions of institutional arrangements, to support asset exchange between transacting parties, and to support asset coordination amongst those holding, buying or selling similar assets. They suggest that asset exchange normally involves spot transactions, non-market transactions or some intermediate form of exchange between two parties. Asset coordination, on the other hand, is beneficial where people obtain relatively small individual gains from holding, buying or selling assets (due to low unit value or to small scales of activity). Coordination may be achieved through private, state or collective arrangements, these in turn consolidate asset holding (or buying or selling) at three levels: management, use and ownership. Asset coordination will often involve collective or common property management regimes. To improve its analysis of the influence of institutions, the SL approach needs to develop a more explicit and holistic analysis of policies, institutions and processes and relate them to livelihood assets, vulnerability and activities. In particular, the enabling and disabling effects of the institutional environment on institutional arrangements, and the influence on asset distribution, exchange and productivity need to be given more prominence. Figure 8.3 attempts to present a more ‘institutional view’ of livelihoods. It recognises that asset exchange and coordination are fundamental processes supporting the transformation of peoples’ assets into livelihood outcomes, and these processes may be supported or inhibited by ‘enabling’ or ‘disabling’ institutions and policies. Asset exchange and coordination may occur at macro, meso and micro scales of analysis, and involve linkages between actors within and across the macro/ meso/ micro scales. Actors, all with their own livelihoods and assets to promote and defend, include private, collective and public sector organisations; individuals; and individuals within organisations,. While the focus of sustainable livelihoods development may be in promoting those at the centre of figure 8.3 (poor ‘micro’ livelihoods), the other livelihood interests all influence policies, institutions and processes: this is an important political economy/ New Institutional Economics point that is crucial to livelihoods development. The framework is relevant to a wide range of important issues not currently given enough emphasis in the SL approach, for example regarding entry points and relations between the SL approach and wider issues of governance, micro- and macro- policies, organisational change, decentralisation, and service delivery, and the contribution and relation of SL thinking to other development policy and implementation approaches. Further work is needed in this area: a more explicit analytical framework is needed to consider the strengths and weaknesses of different forms of institutional arrangement in different

82

institutional environments, and of the necessary and sufficient conditions for these to work for the benefit of the rural poor58.

58

Dorward et al., 1998a, for example, have attempted to develop such a framework to identify such conditions for interlocking transactions.

83

Figure 8.3 Policies, Institutions and Processes and Sustainable Livelihoods

Policies, institutions, Incentives, power, vulnerability

processes Incentives, power, vulnerability

enabling, disabling

macro meso

micro

micro

meso

macro

people, assets & activities

Livelihood Outcomes

Policies, institutions, processes


: influence

KEY

: asset exchange & coordination 84

9 Conclusions and lessons This review set out to ♦ test hypotheses underpinning the design of PRP project R7774SL, ‘Diverse income sources and seasonal finance for smallholder agriculture: applying a livelihoods approach in South Africa’; ♦ establish basic information on which to build project field activities in South Africa; and ♦ review the validity and value of the SL approach in encouraging coherence in understanding and addressing problems of access to and delivery of seasonal financial assets in subsistence crop production. 9.1 Current poor provision of seasonal finance and its importance in pro-poor growth The basic hypotheses of project design were confirmed as regards (a) the importance of agriculture as a source of poverty reducing growth in much of rural Africa, (b) the importance of poor access to seasonal finance as a constraint to agricultural productivity and incomes of the poor, and (c) the current lack of viable institutional models for expanding such access. However, the large and often increasing importance of non-farm income in the livelihoods of the poor was recognised, and poverty reducing strategies for rural growth need to take account of this in identifying and supporting the access of the poor to institutional, technical and market changes which may provide income earning opportunities in agricultural and/or non farm activities. Livelihoods in rural South Africa are marked by a high poverty incidence and a large share of national poverty, limited importance of agriculture in terms of income share, and high dependence on remittances and state transfers among some poor households. Wage labour is an important component of rural livelihoods, but its high average share is partly a reflection of high incomes from wage earning among a relatively small but better-off section of the rural community. Despite its relatively low share in income, agriculture is practiced by a very significant proportion of the rural population, as part of a ‘survivalist strategy’, to provide a significant proportion of household food needs, or (amongst a minority of ‘commercial’ farmers) as a major source of income. Very limited work on agricultural growth multipliers in South Africa also suggests that in some areas at least they are likely to be of similar magnitude as elsewhere in Africa (although building on a lower initial share of the rural economy). There appears, therefore, to be significant potential for a significant proportion of the rural poor to gain from increased agricultural productivity in terms of both reductions in vulnerability and increases in income. However, this will need increased and more secure access to land and to a wide range of financial services, with the latter being tailored to fit the different roles of agricultural production in the varied livelihood strategies of the rural poor. The high incidence of HIV/AIDS in South Africa has very severe implications for rural livelihoods and access to and viability of financial services, beyond its direct effects on infected individuals. High mortality and morbidity rates among the economically active, and demands on the household for care and medical expenditure, will lead to high dependency rates, labour shortages, and erosion of assets. These will have adverse effects on household productivity, incomes, ability to invest, and ability to service loans. 9.2 Potential for different mechanisms and ways forward/ necessary conditions The review noted a wide range of institutional models for delivery of financial services to the poor in sub Saharan Africa, each with different advantages and disadvantages. There is a general tendency for low provision of services to the rural poor, and for agriculture to make up a very small proportion of the loan portfolio. This is partly due to the high costs and risks in the supply of such services, but may also reflect high risks and relatively low returns for borrowers investing in Seasonal finance for staple crop production in Sub Saharan Africa

85

agriculture. In addition to the lack of working models for delivery of seasonal agricultural finance (outside of interlocking contracts in production of some cash crops), micro-insurance services have received very little attention among both theoreticians and practitioners in microfinance. This is an important omission. The literature outside South Africa also contains very little discussion of supply and access to transmission services. It is not clear if this is due to lack of interest in this topic among those discussing financial services, or a lack of development of such services. Either way, this is a topic that merits more attention in the development of integrated financial services serving the needs and demands of the poor. Similarly, access to savings services need to be improved, in view of the current limited services and their likely decline with withdrawal of commercial banks from rural areas. One response to this scenario may be a greater emphasis on institutional models based on group or collective action. Links to high levels of technical expertise and sophistication within the South African financial sector may offer exciting opportunities here, but these need to focus on the use of appropriate technology to overcome the basic problems of maintaining client confidence and loyalty together with effective but low cost transaction and portfolio management systems for small deposits, loans and transactions. Further issues in the development of locally owned and managed institutions are the balance between savings and loan portfolios of members, the relative importance of locally mobilised deposits and outside funds in lending, membership of different household types, management structures and control, and the speed at which these institutions can expand to meet demand. These issues affect the incentives for prudent financial management, management skills and effectiveness, relative benefits to different (poor) rural groups, and the ability of the institution to withstand and provide services in the face of covariant shocks. Turning to consider loan products and services, placing them in the context of group managed institutional models is likely to considerably slow their expansion and limit their outreach. It may also impact negatively on their direct outreach to the poor. Other institutional models are therefore also likely to be important, and as seen earlier with SACCOS in West Africa, there may be potential for linking not only group managed institutions with commercial banks, but also for linking targeted MFIs with SACCOs – although the dangers and pitfalls in this must not be underestimated. Attention also needs to be given to development of targeted loan products. Given the importance of non-farm income to many poor households in South Africa, further development of the approach being tried in Kenya and Uganda, of flexible linking of non-farm income to seasonal loan repayment probably holds the most promise for increasing access to seasonal crop finance for survivalist and subsistence farmers - although there may be problems with the reliability of remittance income, and as argued earlier, this approach is not likely to work well in areas of Africa where non-farm income provides a smaller share of household income. This approach is also unlikely to prove sufficient to meet the seasonal finance needs of small-scale commercial farmers, although the El Salvador models (Financiera Calpia) discussed earlier may be relevant here, using a combination of (perhaps small ) repayments from non-farm income together with non-traditional collateral covering balloon repayments. Again, this is likely to be of limited direct relevance to subsistence farmers elsewhere in sub Saharan Africa. Finally, a more general point flowing from this discussion is that the diversity of rural livelihoods leads to demands for a range of different financial services. Further investigation is needed of the aspirations and opportunities for and constraints to agricultural development amongst different farmer types in different areas, with regard to differentiated and targeted development of insurance, savings, transmission and loan products and services for survivalist, subsistence and small-scale commercial farmers, for example. 9.3 Lessons for further work A number of important lessons for work in the remainder of the project emerge from this review. 86

First, the diversity of rural livelihoods and the relationship between farm and non-farm incomes needs further examination. The broad summary of the relative importance of different income sources presented in section 6 needs to be built on with much deeper understanding of the interaction between farm and non-farm activities within different livelihood strategies. More information and understanding is needed, for example, about the way that the cash flows of and labour demands different activities and income sources interact with each other and household demands. Similarly, the role of farm activities on household vulnerability in different livelihood strategies needs further investigation. Intra-household relations extending across different rural and urban locations also need to be recognised in data collection and analysis. Such information needs to be related to the wider structure of the local economy, access to land, and local infrastructure and services. Sites selected for fieldwork, and the issues addressed in them, should be relevant to broader situations and questions that are important elsewhere in South Africa and/or in sub Saharan Africa. Research and data collection methods need to use a range of methods and approaches, including formal comparative studies of both livelihood strategies and structures on the one hand and service providers on the other (using questionnaire survey approaches), focus group interviews, and more participatory, client led methods. Second, the selection of field sites and the sampling system within these sites needs to take account of (a) the structure of local economy (for example the strength of the non-farm economy and the types of agricultural activity engaged in by different households), (b) the balance between different roles of agriculture within livelihood strategies (commercial, subsistence and survivalist), and (c) availability of and access to different institutional forms and products in finance service delivery. Third, explicit attention needs to be paid to the integration of insurance, savings, lending and transmission in both rural livelihood strategies on the one hand and in product design and institutional forms for service delivery on the other. Fourth, households’ demands for financial services need to be examined in the context of their overall asset portfolio and net income stream, and the productive, consumption, savings and precautionary or insurance roles played by these assets. Fifth, the focus of the networking being developed in the project should be expanded to take account of the importance of micro-insurance and transmission services for the rural poor. Particular attention should also be given to involving those practitioners already identified as developing new financial services supporting investment in seasonal agricultural activities. 9.4 The SL framework Important strengths of the SL approach and framework are valuable in analysing and addressing the relationship between poor rural livelihoods and financial service delivery. These include its emphasis on putting people at the centre, recognition of the importance of different assets and of livelihood diversification; use of the concept of ‘livelihood strategies’; stress on vulnerability; and identification of the importance of institutions. However, some of these strengths do not go far enough. First, there is insufficient attention to the importance of diversification between households and the dynamics of market and non-market demand. This is likely to be an important omission in very many situations and applications of the SL framework. Minor modifications to the existing SL framework would allow this to be addressed without any appreciable increase in complexity or loss of current strengths and emphasis. Second, there is a lack of emphasis on the various roles of assets within livelihood strategies, and of the attributes of assets in fulfilling these roles: this is particularly problematic when considering interventions and services affecting asset investments. A conceptual framework that addresses this 87

is put forward. This may be of help in the current project, and is currently being developed and applied elsewhere in the context of understanding the role of livestock in livelihoods of the poor. Finally, despite the recognition of the importance of institutions within the SL framework, the approach has not generally included sufficient consideration of the nature of these institutions and of the ways that they relate to each other and to other components of the SL framework. Institutional arrangements, in particular, are not commonly discussed in SL analysis (nor in analysis associated with other approaches to rural development) although they are the subject of much attention in the microfinance sector. The importance of these relations needs to be given much wider prominence. Where consideration of them is already prominent (as in the microfinance sector) a more explicit analytical framework is needed to consider their strengths and weaknesses in different institutional environments, and of the necessary and sufficient conditions for different forms of institutional arrangement to work for the benefit of the rural poor.

88

References: Adams, D. W. (1988). "The conundrum of successful credit projects in floundering rural financial markets." Economic Development and Cultural Change 36(2): 355-367. Adams, D. W., D. Graham, et al., Eds. (1984). Undermining Rural Development with Cheap Credit. Boulder, Colorado, Westview Press. Adams, D. W. and R. C. Vogel (1986). "Rural financial markets in low-income countries: recent controversies and lessons." World Development 14(4): 477-487. Agricultural News (1997). Agricultural News. Newsletter, National Department of Agriculture, September 8. Pretoria, Department of Agriculture. Al-Hassan, R., A. R. Dorward, et al. (1999). Improving Access to Maize Marketing Opportunities in Remote Areas of Ghana. Workshop on "Improving Smallholder Market Access in Remote Areas of Sub-Saharan Africa", Wye College, July 1999. ASA, B. (2000). Default in microlending. email from sharenet, Tue, 12 Dec 2000. Ashley, C. (2000). Applying livelihood approaches to natural resource management initiatives: experiences in Namibia and Kenya. Working paper No. 134 (Sustainable livelihoods series). London, Overseas Development Institute. Barrett, C. B., M. Besfuneh, et al. (2000). Heterogeneous Constraints, Incentives and Income Diversification Stragies in Rural Africa. mimeo: 40. Bautista, R. and M. Thomas (1999). "Agricultural growth linkages in Zimbabwe: income and equity effects." Agrekon 38(May special issue): 66-75. Block, S. A. (1995). "The recovery of agricultural productivity in sub-Saharan Africa." Food Policy 20(5): 385-405. Braverman, A. and J. L. Guasch (1986). "Rural credit markets and institutions in developing countries: lessons for policy analysis from practice and modern theory." World Development 14(10): 1253 - 1267. Brinn, P., A. Dorward, et al. (1999). Soil fertility management and sustainable rural livelihoods: new approaches to the policy process. Chatham, Natural Resources Institute. Bryceson, D. F. (1999a). "African Rural Labour, Income Diversification and Livelihood Approaches: A Long-Term Development Perspective." Review of African Political Economy 26(80): 171-189. Bryceson, D. F. (1999b). Sub Saharan Africa Betwixt and Between. Working Papers. Leiden, African Studies Centre, University of Leiden, Netherlands. Bryceson, D. F. (2000). "Rural Africa at the Crossroads: Livelihood Practices and Policies." Natural Resources Perspectives(52). Byerlee, D. and P. W. Heisey (1996). "Past and potential impacts of maize research in sub-Saharan Africa: A critical assessment." Food Policy 21(3): 255-277. Carney, D. (1998). Sustainable Livelihoods. Sustainable Livelihoods: What contribution can we make? D. Carney. London, DFID. Carr, S. J. (1997). "A green revolution frustrated: lessons from the Malawi experience." African Crop Science Journal 5(1): 93-98. Coetzee, G. (1988). Die finansiering van landbou-producksie in die kleinboersektor van SuidAfrika. Stellenbosch, University of Stellenbosch. 89

Coetzee, G. (1998). "Retail rural finance in South Africa: from policies to practice." Agrekon 37(4): 517-526. Coetzee, G. and J. van Zyl (1990). An assessment of food security in Southern Africa. Interconference symposium of the International Association of of Economists, Swakopmund, Namibia. Coetzee, G. and N. Vink (1991). "Viewpoint: financing small farmers - quo vadis?" Agrekon 30(3): 153-157. Coetzee, G. K., M. A. Kirsten, et al. (1994). Financing entrepreneurs in South Africa's rural areas: a new approach, Development Bank of Southern Africa. Davis, L. E. and D. C. North (1971). Institutional Change and American Economic Growth. Cambridge., Cambridge University Press. Delgado, L. C., J. Hopkins, et al. (1998). Agricultural Growth Linkages in Sub-Saharan Africa. FPRI Research Report. Washington DC., International Food Policy Research Institute. DFID (1998). South Africa country strategy paper. London, Department for International Development. Dorward, A. (2001). "The effects of transaction costs, power and risk on contractual arrangements: a conceptual framework for quantitative analysis." Journal of Agricultural Economics (in press). Dorward, A., J. Kydd, et al. (1998a). Conclusions: NIE, Policy Debates and the Research Agenda. Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. A. Dorward, J. Kydd and C. Poulton Wallingford, CAB International: 240-265. Dorward, A., J. Kydd, et al., Eds. (1998b). Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. Wallingford., CAB International. Dorward, A. R. (1999). "Farm size and Productivity in Malawian Smallholder Agriculture." Journal of Development Studies 35(5): 141-161. Dorward, A. R., S. Anderson, et al. (2001). Asset functions and livelihood strategies: a framework for pro-poor analysis, policy and practice. Unpublished draft. Wye, Imperial College at Wye, T H Huxley School. DTI (2000). Examination of costs and Interest Rates in the Small Loans Sector. Government Gazette No. 21381, Department of Trade and Industry, South Africa. Econometrix (1999). Report on the micro lending industry in South Africa. Johannesburg., Econometrix. Eicher, C. K. and B. Kupfuma (1998). Zimbabwe's maize revolution: insigths for closing Africa's food gap. International Agricultural Development. C. K. Eicher and J. Staatz. Baltimore and London, John Hopkins University Press: 550-570. Ellis, F. (1992). Agricultural policies in developing countries. Cambridge, Cambridge University Press. Ellis, F. (1998). Livelihood diversification and sustainable rural livelihoods. Sustainable Livelihoods: What contribution can we make? D. Carney. London, Department for International Development (DFID): 53-66. 90

Ellis, F. (2000a). "The determinants of rural livelihood diversification in developing countries." Journal of Agricultural Economics 51(2): 289-302. Ellis, F. (2000b). Rural Livelihoods and Diversity in Developing Countries. Oxford, Oxford University Press. Eskom (1998). Consumers in South Africa. Johannesburg., ESKOM Marketing Division. FAO (2000). FAOSTAT Statistics Database, FAO (Food and Agriculture Organisation). Goldstein, G. and I. Barro (1999). The role and impact of savings mobilization in West Africa: a study of the informal and intermediary financial sectors (Benin, Burkina Faso, Ghana, Guinea, Mali and Togo), MicroSave-Africa/West Africa. Gordon, A. and A. Goodland (2000). "Production credit for African small-holders- conditions for private provision." Savings and development XXIV(1): 55-84. Govereh, J., T. S. Jayne, et al. (1999). Smallholder commercialisation, interlinked markets and food crop productivity: cross country evidence in Eastern and Southern Africa, Michigan State University, Department of Agricultural Economics and Department of Economics: 39. Haggblade, S., J. Hammer, et al. (1991). "Modelling agriculutral growth multipliers." American Journal of Agricultural Economics 73(May): 361-364. Harris White, B. and S. Janakarajan (1997). "From green revolution to rural industrial revolution in South India." Economic and Political Weekly 32(25): 1469-1477. Harriss, B. (1983). Money and commodities: their interaction in a rural Indian setting. Rural Financial Markets in Developing Countries. J. D. Von Pischke, D. Adams and G. Donald. Baltimore, Md., John Hopkins University Press. Harriss, B. (1987). "Regional growth linkages from agriculture." Journal of Development Studies 23(2): 275-289. Hazell, P. B. R., Pomareda, et al. (1986). Agri Insurance CHECK. P. B. R. Hazell and Valdes. Hazell, P. B. R. and R. Slade (1987). "Regional growth linkages from agriculture: a reply." Journal of Development Studies 23(2): 290-294. Hobley, M. (2001). Unpacking the PIP box. Chard, UK, Hobley Shields Associates. Howard, J., V. Kelly, et al. (1999). Green revolution technology takes root in Africa: the promise and challenge of the Ministry of Agriculture /SG2000 experiment with improved cereals technology in Ethiopia. MSU international development working papers No 76. East Lansing, Michigan, Department of Agricultural Economics & Department of Economics, Michigan State University. Hulme, D., L. Mutesasira, et al. (1999). Client Exits (Dropouts) From East African Micro-Finance Institutions. Kampala, MicroSave-Africa. Jain, P. S. (1996). "Managing credit for the rural poor: lessons from the Grameen Bank." World Development 24(1): 79-89. Jayne, T. S. (1994). " Do High Food Marketing Costs Constrain Cash Crop Production? Evidence from Zimbabwe." Economic Development and Cultural Change 42(2): 387-402. Jayne, T. S. and S. Jones (1997). "Food marketing and pricing policy in Eastern and Southern Africa: A survey." World Development 25(9): 1505-1527. Jayne, T. S., J. D. Shaffer, et al. (1997). Improving the impact of market reform on agricultural productivity in Africa: how institutional design makes a difference. MSU International 91

Development Working Paper 66. East Lansing, Michigan, Department of Agricultural Economics, Department of Economics, Michigan State University,. Jones, J. H. M., O. Sakyi-Dawson, et al. (1999). Improving Financial Services for Renewable Natural Resource Development in Ghana: Establishing Policy Guidelines for the Informal Financial Sector., Dept. Agric. Extension and Rural Dev., University of Reading & Dept. Agric. Extension, University of Ghana, Legon. Jones, S. (1998). Liberalised food marketing in developing countries: key policy problems. Oxford Policy Management Papers, No. 46. Oxford, Oxford Policy Management. Karanja, D. D., T. S. Jayne, et al. (1998). Maize producivity and impact of market liberalization in Kenya. Strategies for raising smallholder agricultural productivity and welfare, Egerton University/ Tegemeo Institute of Agricultural Policy and Development, November 1998, Nairobi, Kenya. Kasekende, L. and M. Atingi-Ego (1999). "Impact of Liberalisation on Key Markets in SubSaharan Africa: The Case of Uganda." Journal of International Development 11: 411-436. Kelly, V. A., E. W. Crawford, et al. (1999). Towards a strategy for improving agricultural inputs markets in Africa, USAID: 6. Kherrallah, M., C. Delgado , et al. (2000). The road half-travelled: agricultural market reform in sub-Saharan Africa. FFood Policy Report. Washington D C, International Food Policy Research Institute. Kirsten, J. F. and J. van Zyl (1996). "The Costs and Benefits of Providing Agricultural Support Services to Rural Households in the Developing Areas of South Africa." Development Southern Africa 13(3): 415-28. Klein, B., R. Meyer, et al. (1999). Better practices in agricultural lending. Agricultural Finance Revisited, No 3. Rome, Food and Agricultural Organisation of the UN (FAO), Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ). Kydd, J. G., A. R. Dorward, et al. (2001). Agricultural parastatals revisited: a new institutional economics analysis. D. Byerlee and I. Livingstone, Routledge. Larson, B. A. and G. B. Frisvold (1996). "Fertilizers to support agricultural development in subSaharan Africa: What is needed and why." Food Policy 21(6): 509-525. Laurent, C., C. J. van Rooyen, et al. (1999). "Household typology for relating social diveristy and technical change." Agrekon 38(special issue): 190-208. Leibbrandt, M., C. Woolard, et al. (2000). "The contribution of income components to income inequality in the rural former homelands of South Africa: a decomposable Gini analysis." Journal of African Economies 9(1): 79-99. Levine (1997). "Financial Development and Economic Growth: Views and Agenda." Journal of Economic Literature XXXV: 688-726. Lyne, M. C. (1998). Unpublished survey results. Pietermaritzburg, University of Natal. Machethe, C. L., T. Reardon, et al. (1997). "Promoting farm/non-farm linkages for employment fo the poor in South Africa: a resaerch agenda fociused on small-scale farms and agroindustry." Development Southern Africa 14(3): 377-394. Makhura, M. T., F. M. Goode, et al. (1998). "A cluster analysis of commercialisation of farmers in developing rural areas of South Africa." Development Southern Africa 15(3): 429-448. 92

Matin, I., D. Hulme, et al. (1999). Financial Services for the Poor and Poorest: Deepening Understanding to Improve Provision. Finance and Development Research Programme, Working Paper Series, No 9. Manchester, IDPM, University of Manchester. Matin, M. (1999). Rural Credit Deepening and a Few Concerns: A Study of Grameen Bank in Madhupur. Paper presented at DSA/ESRC conference "The New Poverty Strategies (199099): What have they achieved? What have we learned?" University of Reading, 10-11/4/99. Meyer, R. L. (2000). Asian efforts to develop rural financial markets: some insights for Brazil. Economics and Sociology Occasional Paper No. 2569. Columbus, Ohio, Rural Finance Program, Department of Agricultural, Environmental, and Development Economics, The Ohio State University. Morokolo, B., G. Coetzee, et al. (1999). "Savings behavviour and motivation of resource poor farmers: a case study in the Moretele district of the North West Province." Agrekon 38(2): 852-860. Morris, M. and D. Byerlee (1998). Attaining of productivity gains in post green revolution Asian agriculture. International Agricultural Development. C. K. Eicher and J. Staatz. Baltimore and London, John Hopkins University Press: 458-473. Morrison, J., A. R. Dorward, et al. (2000). Sustainable livelihoods and new institutional economics. Paper prepared for DFID Sustainable Livelihoods Programme, Policies, Institutions and Processes sub Group. Wye, UK, Wye College. Mosley, P. (2000). Financial services diversification and agricultural development: the case of microinsurance. Occasional paper No. 2000.08. Sheffield, Department of Economics, University of Sheffield: 23. Mosley, P. and D. Hulme (1998). "Microenterprise finance: Is there a conflict between growth and poverty alleviation?" World Development 26(5): 783-790. Mosley, P. and T. Subasat (1995). "Assessing adjustment in Africa." World Development 23(9): 1459-1474. Mutangandura, G., D. Mukarazita, et al. (1999). A review of household and community responses to the HIV/AIDS epidemic in the rural areas of sub-Saharan Africa. Best practice collection, key material. Geneva, Switzerland, UNAIDS. Naseem, A. and V. Kelly (1999). Macro trends and determinants of fertiliser use in sub-Saharan Africa. MSU international development working papers No 73. East Lansing Michigan, Department of Agricultural Economics Department of Economics Michigan State University. Navajas, S. and C. Gonzalez-Vega (2000). Innovative approaches to rural lending: Financiera Calpia in El Sa;vador. Economics and Sociology Occasional Paper No. 2569. Columbus, Ohio, Rural Finance Program, Department of Agricultural, Environmental, and Development Economics, The Ohio State University. Ngqangweni, S. (1999). Promoting income and employment growth in the rural economy of the Eastern Cape through smallholder agriculture. Department of Agricultural Economics, Extension and Rural Development. Pretoria, University of Pretoria. Ngqangweni, S., J. F. Kirsten, et al. (1999a). "Exploring growth linkages in a South African smallholder farming area." Agrekon 38(4): 585-593. Ngqangweni, S., J. F. Kirsten, et al. (1999b). "Measuring smallholder comparative advantage in three South African provinces." Agrekon 38(special issue). 93

Nicol, A. (2000). Adopting a sustainable livelihoods approach to water projects: implications for policy and practice. Working paper 133, Sustainable livelihoods series. London, Overseas Development Institute. North, D. C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge, Cambridge University Press. Omamo, S. W. and L. O. Mose (2001). "Fertilizer trade under market liberalization: preliminary evidence from Kenya." Food Policy 26(1): 1-10. Ouattara, K. and D. H. Graham (1996). Rural Finance in the Republic of South Africa: Selected Farm Household Case Studies. Report prepared for the Development Bank of Southern Africa (DBSA). Rural Finance Program, Department of Agricultural Economics, Ohio State University. Outtara, K., C. Gonzalez-Vega, et al. (1999). Village Banks, Caisses Villageoises and Credit Unions: Lessons from Client Owned Micro-finance Organisations in West Africa. Microenterprises best practices. Bethesda MD, Development Alternatives Inc. Poulton, C. (1998). Cotton production and marketing in Northern Ghana: the Dynamics of Competition in a system of Interlocking Transactions. Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. A. R. Dorward, J. Kydd and C. Poulton. Wallingford, CAB International: 56-112. Poulton, C., A. Dorward, et al. (1998). A New Insitutional Perspective on Current Policy Debates. Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. A. Dorward, J. Kydd and C. Poulton. Wallingford, CAB International: 1-55. Rajeev, M. and S. Deb (1998). "Institutional and non-institutional credit in agriculture - Case study of Hugli district of West Bengal." Economic and Political Weekly 33(47-48): 2997-3002. Reardon, T. (1997). "Using evidence of household income diversification to inform study of the rural non-farm labour-market in Africa." World Development 25(5): 735-748. Reardon, T. (1998a). African agriculture: productivity and sustainability issues. International Agricultural Development. C. K. Eicher and J. Staatz. Baltimore and London, John Hopkins University Press: 445-457. Reardon, T. (1998b). Rural non-farm income in developing countries. The State of Food and Agriculture 1998. FAO. Rome, Food and Agriculture Organisation. Reardon, T., C. B. Barrett, et al. (1999). "Policy reforms and sustainable agricultural intensification in Africa." Development Policy Review 17: 375 - 395. Reardon, T., E. Crawford, et al. (1994a). "Links between non-farm income and farm investment in African households: adding the capital market perspective." American Journal of Agricultural Economics 76(5): 1172-1176. Reardon, T., A. Fall, et al. (1994b). Is income diversification agriculture-led in the West African semi-arid tropics? The nature, causes, effects, distribiution and production linkages of offfarm activities. Economic policy experience in Africa: what have we learnt? W. S. Atsain A., and Drabek A.G. Nairobi, African Economic Research Consortium: 207-230. Reardon, T., V. Kelly, et al. (1997). "Promoting sustainable intensification and productivity growth in Sahel agriculture after macroeconomic policy reform." Food Policy 22(4): 317-327.

94

Reardon, T., J. E. Taylor, et al. (2000). "Effects of non-farm employment on rural income inequality in developing countries: an investment perspective." Journal of Agricultural Economics 51(2): 266-288. Reinke, J. (2001). Cooperation or competition: the institutional development of microfinance in Suthern Africa. mimeo. London, CREFSA, London School of Economics. Risopoulos, J., R. Al-Hassan, et al. (1998). Improving smallholder access to maise marketing: opportunities in Southern Africa: a literature review. Wye, Wye College. Roberts, B. and J. May (2000). A dynamic analysis of household livelihoods and asset accumulation in post-apartheid South Africa: evidence from KwaZulu-Natal. CSAE Conference "Opportunities in Africa: micro-evidence on firms and households", St Catherine's College, Oxford, 9th-10th April, 2000. Rusike, J., C. Sukume, et al. (2000). The Economic Potential for Smallholder Soyabean Production in Zimbabwe. Harare, Zimbabwe, Soil Fertility Network for Maize Based Cropping Sytems in Malawi and Zimbabwe, CIMMYT. Rutherford S (2000). "Raising the curtain on the `microfinancial services era'." Small enterprise development 11(1): 13-25. Rutherford, S., L. Mutesasira, et al. (1999). Savings and the Poor: the methods, use and impact of savings by the poor of East Africa. East African Savings Study. Kampala, MicroSaveAfrica. Savadogo, K., T. Reardon, et al. (1998). "Adoption of improved land use technologies to ncrease food security in Burkina Faso: relating animal traction, productivity and non-farm income." Agricultural Systems 58(3): 441-464. Scoones, I. and C. Toulmin (1999). Policies for soils management in Africa. London, DFFID. Seibel, H. D. (1999a). Strategies for developing viable micro finance institutions with sustainable services: the Asian experience. Rural Finance working paper No A3. Rome, IFAD International Fund for Agricultural Development. Seibel, H. D. (1999b). Transforming rural finance in Africa: the role of AFRACA in linkage banking and financial systems development. Rural Finance Working Paper No A 6a. Rome, IFAD. Seibel, H. D. and P. Schmid (1999). How an Agricultural Development Bank Revolutionized Rural Finance:The Case of Bank Rakyat Indonesia. Rural finance working papers, IFAD INTERNATIONAL FUND FOR AGRICULTURAL DEVELOPMENT. Shackleton, S., C. Shackleton, et al. (2000). "Revaluing the communal lands of Southern Africa: new understandings of rural livelihoods." Natural Resources Perspectives 62. Sharma, M. and M. Zeller (1999). "Placement and outreach of group-based credit organizations: The cases of ASA, BRAG, and PROSHIKA in Bangladesh." World Development 27(12): 2123-2136. Shipton, P. (1990). How Gambians save and what their strategies imply for international aid. Policy Research and Extension Affairs Working Papers: Agricultural Policies WPS 395. Washington DC, World Bank. Sinha, S. (1998). "Micro-credit: Impact, targeting and sustainability - Introduction and overview." IDS Bulletin-Institute of Development Studies 29(4): 1-10.

95

Speirs, M. and O. Olsen (1992). Indigenous integrated farming systems in the Sahel. World Bank Technical Paper No 179, Africa Technical Department Series. Washington D.C., World Bank. Stockbridge, M., L. Smith, et al. (1998). Cotton and wheat marketing and the provision of preharvest services in Sindh Province, Pakistan. Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. A. R. Dorward, J. Kydd and C. Poulton. Wallingford, CAB International: 177-239. Strauss Commission (1996). The Interim Report of the Commission of Inquiry into the Provision of Rural Financial Services and the Final Report of the Commission of Inquiry into the Provision of Rural Financial Services. Pretoria, Government Printer. Toulmin, C., R. Leonard, et al. (2000). Diversification of livelihoods: evidence from Ethiopia and Mali. Research Report. Brighton, Institute of Development Studies. 47. Turton, C. (2000a). Enhancing livelihoods through participatory watershed development in India. Working Paper 131, Sustainable livelihoods series. London, Overseas Development Institute. Turton, C. (2000b). The sustainable livelihood approach and programme development in Cambodia. Working paper 130, Sustainable livelihoods series. London, Overseas Development Institute. Van Zyl, J. and G. Coetzee (1990). "Food security and structural adjustment: empirical evidence on the food price dilemma in Southern Africa." Development Southern Africa 7(1): 105116. Von Pischke, J. D. (1999). Overcoming credit access barriers. Microfinance in Africa. S. A. Breth. Mexico City, Sasakawa Africa Association: 1-11. Von Pischke, J. D., D. Adams, et al. (1983a). Fungibility and the design and evaluation of Agricultural Credit Projects. Rural financial markets in developing countries: their use and abuse. J. D. Von Pischke, D. Adams and G. Donald. Baltimore and London, The John Hopkins University Press: 74 - 83. Von Pischke, J. D., D. Adams, et al., Eds. (1983b). Rural Financial Markets in Developing Countries. Baltimore, Md., Johns Hopkins University Press. Whiteside, M. (2000). Ganyu labour in Malawi and its implications for livelihood security interventions - an analysis of recent literature and implications for poverty alleviation. Agricultural Research and Extension Network. London, Overseas Development Institute. Wiggins, S. (1995). "Change in African farming systems between the mid-1970s and the mid1980s." Journal of International Development 7(6): 807-848. Wiggins, S. (2001). "Spatial dimensions of rural development." Department of Agricultural and Food Economics, University of Reading. Williamson, O. E. (1985). The Economic Institutions of Capitalism. New York, The Free Press. Wobst, P. (2000). Why the poor care about partial versus general equilibrium effects: part I methodology and country case. TMD Discussion Paper 60. Washington D.C., International Food and Policy Research Institute. World Bank (1997). Rural development: From vision to action. Washington, D.C, World Bank. World Bank (2000a). Can Africa Claim the 21st Century?, World Bank, African Development Bank, United Nations Economic Commission for Africa. 96

World Bank (2000b). World Development Indicators: tables and CD ROM (some data also gathered from earlier editions). Washington D.C., World Bank. World Bank (2000c). World Development Report, 2000. Washington D.C., World Bank. Yaron, J., M. Benjamin, et al. (1998). "Promoting Efficient Rural Financial Intermediation." World Bank Research Observer 13(2): 147-70. Zeller, M. and M. Sharma (2000). "Many borrow, more save, and all insure: implications for food and micro-finance policy." Food Policy 25(2): 143-167.

97

APPENDIX 1: Rural Livelihoods in South Africa: Case Studies from KwaZulu-Natal, Northern Province and the Eastern Cape 1 KwaZulu Natal Case study May (1998) analyses the patterns of livelihoods that characterize rural households in KwaZulu Natal. He uses survey data collected in the 1990s to examine living standards and incidence of poverty in South Africa. May differentiates households according to their resources and assets (access to physical and human capital), the activities from which they derive incomes, and the outcomes in terms of household and per capita incomes. In his work households are distinguished according to the dominant sources of income within their diversified livelihood packages. These include urban remittances, formal sector employment, and also encompass a wide range of other supplementary sources. The emphasis is that people in rural areas engage in a wide range of activities in order to generate a livelihood. Davies (1996) (in May, 1998), notes that the diversification of income sources is a key factor influencing the well being of households in rural areas. According to him, through diversification households are able to buffer against risk in agrarian environments. By engaging in diversified activities, households may be seen as trying different options, which they hope, will generate an adequate and sustainable livelihood that is resilient to shock. This constitutes what can be called a livelihood strategy, and is an outcome of decisions and conflict over use of the household’s resources. In trying to understand what influences households to choose particular livelihood strategies there is need to have an understanding of the environment in which the households live. KwaZulu-Natal is the province with the largest population in South Africa, estimated at 9.8 million in 1995 living in 1.6 million households. This population grew at about 2.3% per annum then, and was expected to reach 12.9 million by 2010. In 1995, 52% of the population were considered functionally urbanized, one third of which were living in informal settlements. Urbanization is expected to increase to over 60% by 2020. The size of the provincial population, however, will not continue to increase due to the influence of the AIDS pandemic. KwaZuluNatal’s population growth rate declined to nearly 1% due to deaths caused by AIDS. This will have a major influence on the ability of rural people to ensure viable livelihood strategies and also need to be considered in any strategy put forward. As stated earlier, assets are a very important aspect when analyzing the livelihood of a particular group of people. The assets available to rural people in KwaZulu-Natal include land, labor, human capital, agricultural capital, entitlements to pensions, and entitlements to intra-households transfers (May, 1996). Just over half of African rural households in KwaZulu-Natal (53%) have access to land for the cultivation of crops, which is more than twice the incidence of land holding by rural African people in the rest of South Africa. However land holdings vary by region within the province (Muller et al, 1987 in May, 1996). Studies carried out by the Southern African Labor and Development Research Unit (SALDRU) in South Africa in 1993 on incomes and expenditures reveal that average land size for households’ access to land is similar in KwaZulu-Natal to the national average, at two hectares per family. However, average land size per adult equivalent is smaller in KwaZulu-Natal, at half a hectare compared to with 0.8 hectares in the rest of South Africa. This reflects the population densities in rural KwaZulu-Natal, and is consistent with the smaller farm sizes found in the commercial sector (May & Amin, 1995). Livestock also represents a significant resource for the people of KwaZulu-Natal. Livestock ownership is highest in this province than in any other in the country with about 31% of the households owning a herd (May, 1996). The SALDRU survey identified five main activities by which rural households in South Africa generate incomes. These include agriculture, non-farm self-employment, wage labor, claims against Seasonal finance for staple crop production in Sub Saharan Africa

99

the state (old age pensions and disability grants) and claims against household and community members. Ouattara and Graham, 1996 documented conditions and access to financial services among selected homeland farming communities in KwaZulu-Natal & the Northern province in South Africa. Household surveys in the districts of Port Durnford and Hlabisa in KwaZulu-Natal revealed that people in these areas are relatively poor, and most are not significantly engaged in agricultural activities for commercial purposes. For households with a farming tradition, access to formal agricultural extension services was shown to be very limited (Ouattara and Graham, 1996). Other important issues that the study dug out are that: ♦ Access to financial services by households was limited despite the fact that commercial banks were being used extensively in the area. So people were basically depositors, but banks were offering almost non-existent credit in return. ♦ Moneylenders, friends, and relatives formed the bulk of limited source of financing for the household. ♦ Only a very small amount of credit was associated with agricultural production activity with the bulk of it used to purchase consumer durables (television sets, refrigerators, etc) as well as food, clothing, education and medical services. ♦ Major source of income was outside wage employment. ♦ Wealthier people had higher chances of accessing loans. KwaZulu-Natal alone cannot be used as a proxy of what the rural livelihood situation in South Africa is like, because the province seems to have a significant advantage in terms of access to agricultural and productive equipment compared to the rest of rural areas in the country. 2 Northern Province case study Baber (1996), in his paper ‘Current livelihoods in Semi-arid rural areas of South Africa’, compares the patterns of livelihood in two Northern Province communities Mamone and Rantlekane. According to him the communities differ in respect to size, remoteness from employment centers, tenure system, landlessness, the origin of the community, and the length of time in which the households have been settled in the area. One community relies heavily in agriculture whilst the other is typical of many of the Northern province settlements in its meager agricultural resources and greater dependence on the outside economy. The author further stresses that the contrasting pattern of rural livelihoods that emerges in these two settlements helps to illustrate the nature of the rural economy, and can be used in assessing the livelihoods in the semi-arid regions of South Africa. Data analysis results showed that remittances provided two thirds of total household income in Mamone, whilst 93% of households in Rantlekane count remittances as one of their income sources. Most of the remittances are from migrant workers. The greater role of remittances in Rantlekane is congruent with its higher migration ratio and a result of its weak local economy (Baber, 1996). Seventy four percent of the households in Mamone have agriculture as one of their sources of income, compared to 34 % in Rantlekane. The rest of the activities’ contributions to the households are almost similar. The difference between the two settlements ability to provide locally based livelihoods is shown by the greater contribution made in Mamone by informal economic activities, informal employment, local businesses and in particular agriculture. In Mamone just less than half of total resident income at the community level is generated in the local economy, and roughly a quarter of this is in the form of agriculture. In Rantlekane, by contrast, only 30% of total resident income at the community level comes from local economic activity, and agriculture forms 100

only one or two percent of this. The poor state of cropping in Rantlekane relates to landlessness, monoculture (reliance on maize only), the more arid climate and the tenure system (Baber, 1996). According to the author, this hampers backward linkages with other activities, such as ploughing and transport businesses. Also interesting is the fact that in both communities pension income forms a significant proportion of total income, and plays an essential role in lifting some households out of the worst extremes of poverty. There is a tendency by the pensioners to distribute their pension even to those households other than which they live especially female-headed households, thus benefiting those households without remittance income. A study was also done on the former Lebowa homeland, now part of the Northern Province. The revelations were that people were relatively poor, not well educated, and with limited access to public services. The majority of households were headed by a de facto female head, and at least one member of the household was either a full time farmer, or self employed. Just like the studies carried out in some districts of KwaZulu-Natal, household access to financial services and to loans in particular was shown to be limited, although savings were also held in commercial banks (Ouattara and Graham, 1996). Although caution must be exercised when projecting up to regional or national level from the results of handful of micro studies, they undoubtedly are good proxies of the reality of the livelihood situation in rural areas. Of notable significance is the fact that agriculture seems to be playing a limited role in the northern parts of the country. It is probably in this light that households there also have diversified sources of livelihood. It is therefore critical to policymakers that when they make available opportunities that overcome transaction costs faced by such communities like credit markets, they not only fund agriculture but other money generating enterprises as well. 3 Western Cape case study In the Western Cape, Cousins et al (1996) carried out a study on rural livelihoods and small-scale agriculture. Theirs was an evaluation of developmental projects that had been in place in the province since the early 1980s. These projects were aimed at making land and support services available to would-be small scale farmers from the local colored community, who had been denied meaningful opportunities to engage in agricultural production other than as farm workers. The projects were community gardens and a farm whereby part time and full time small holders engaged in various enterprises. The program had two community gardens, consisting of 10m x 15m allotment plots for about 40 members and used mainly for vegetable production, and whilst the farm was 136 ha in extent with six to eight small holders producing a mix of fruit and vegetables under irrigation on holdings ranging from 0.75 to 2 ha each, together with both intensive and extensive small livestock (Cousins, Cousins & Theron, 1996). After enjoying a period of bliss, the project started experiencing problems, some of which are still being faced to date. Studies carried out by Cousins, Cousins & Theron (1996), give a detailed description of the exact nature of the problem as they followed the program from inception through the various phases that it went through. It is in this regard that they came up with lessons that ought to be considered when similar ventures are proposed. Some of the recommendations that the authors came out with for future work on enhancing rural livelihoods are: ♦ They found out that members of the projects demonstrated that aspirant small scale farmers in the region possessed considerable agricultural knowledge and skills as a result of previous involvement in the commercial farming sector, mostly as laborers. This according to them confirmed de Klerk’s (1995) assertion that “commercial farm employees were probably the 101

largest pool of potential small scale farmers in the Western Cape”, but also suggested that exfarm laborers should be included the pool. ♦ Risk management strategies of small farmers must be taken into account when designing programs to promote agriculture as a source of livelihoods, and are likely to influence farmers’ responses to institutional arrangements and financial, technical and other services. This is particularly the case when capital intensive enterprises with long lead times, such as fresh fruit growing, are established. ♦ Diversification came out to be an important risk management strategy. ♦ Adequate levels of financial support are a prerequisite for farming enterprises, and particularly so for capital intensive forms of productions such as fresh fruit. ♦ Support services providing appropriate technical advice, management support and possibly machinery and equipment hire services, are vitally important, particularly in the early stages of small farmer development programs (Cousins, Cousins & Theron, 1996) Considering the fact that the evaluation process revealed that the above project was at crossroads with its future uncertain, the above recommendations ought to be taken seriously by policy makers whenever they are designing probable projects aimed at enhancing rural livelihoods. It is clear that in rural areas of South Africa, as shown by Northern and KwaZulu-Natal provincial studies referred to above, farming on its own rarely provides a sufficient means of survival. What is apparent is that rural households depend on a diverse portfolio of activities and income sources, among which crop and livestock production feature alongside many other contributions to family well-being. The five case studies mentioned here, although regional, reflect to a certain extent the means of livelihoods in rural areas in South Africa. It is clear that some provinces are better endowed than others with respect to resources, thus giving the rural folks more choices from which to generate their livelihoods. In a nutshell, people living in rural areas of South Africa use every means possible to generate incomes and thus diversification of income sources seems to be the prevalent way of making ends meet.

102

Chapter three

Chapter three

Suggest Documents

Chapter Three

Chapter Three

chapter three

chapter three

chapter three

CHAPTER THREE

Chapter Three

chapter three

Chapter three - CiteSeerX

CHAPTER THREE - Resnet

Chapter three

Chapter three Materials & Methods

Chapter Three - Google Groups

chapter three

Chapter Three - unf.edu

Chapter three - AgEcon Search

Chapter Three: Research Methodology

Chapter three

Chapter Three Cases

Chapter Three Materials & Methods

genesis chapter three - SalemBaptistCinci.com

Chapter Three Lecture

chapter thirty-three

Chapter three - CiteSeerX