Social Protection and Agriculture in Ethiopia 
                        
                                                    Stephen Devereux and Bruce Guenther 
                                           Institute of Development Studies, University of Sussex 
                        
                                 Country case study paper prepared for a review commissioned by the FAO on 
                                          ‘Social Protection and Support to Small Farmer Development’ 
                        
                        
                                                                30 November 2007 
                        
                        
                        
                        
                        
                       1
                             Introduction...........................................................................................................................2 
                       2     Agriculture and social protection in Ethiopia: complementarity or convergence? .................3 
                       3     ‘Land politics’ and social protection in Ethiopia.....................................................................4 
                       4     Agricultural development strategies in Ethiopia ....................................................................5 
                       5     Ethiopia’s Food Security Programme....................................................................................6 
                             5.1     Productive Safety Net Programme (PSNP)...............................................................6 
                                 5.1.1    Direct linkages – PSNP public works.................................................................8 
                                 5.1.2    Indirect linkages – investment of PSNP transfers..............................................9 
                             5.2     Household Extension Packages (HEP)...................................................................10 
                             5.3     Voluntary Resettlement Programme (VRP).............................................................11 
                       6     Other social protection interventions for Ethiopian smallholders.........................................12 
                             6.1     Disaster prevention and risk management..............................................................12 
                             6.2     Weather–indexed drought insurance.......................................................................13 
                             6.4     Ethiopian Commodity Exchange (ECEX)................................................................14 
                       7     Conclusion..........................................................................................................................15 
                       References: Ethiopia case study..................................................................................................17 
                        
                        
                        
                       List of Tables 
                       Table 1      Investment uses of PSNP cash transfers for investment, by income quintile..............9 
                        
                        
                       List of Figures 
                       Figure 1     Objectives of Ethiopia’s Productive Safety Net Programme........................................7 
                       Figure 2     Ethiopia’s proposed integrated drought risk management plan.................................13 
                        
                                                                                                                 Page 1 of 18 
              
              
             1 Introduction 
              
             Agriculture and social protection in Ethiopia are inextricably interconnected. Smallholder farming 
             is the dominant livelihood activity for the majority of Ethiopians, but it is also the major source of 
             vulnerability to poverty, food insecurity and their often fatal consequences – chronic malnutrition, 
             premature mortality, recurrent famines. Ethiopian farmers have been the recipients of enormous 
             volumes of food aid and other humanitarian assistance over recent decades, to such an extent 
             that emergency relief has become institutionalised within government structures and donor 
             agency country programmes. 
              
             The discourse surrounding the complex relationship between agriculture and social protection in 
             Ethiopia can be approached from either perspective. From the agricultural policy perspective, the 
             government’s unshakeable belief in the centrality of farming as the backbone and potential 
             source of growth for the Ethiopian economy is mirrored by its almost ideological view that land is 
             the ultimate ‘safety net’ for rural households, who should therefore be protected against losing 
             their access to land – by being prevented from selling it. From the social protection perspective, 
             awareness that able-bodied farmers are the main ‘beneficiaries’ of safety nets and humanitarian 
             food aid has fuelled the government’s visceral fear of creating ‘dependency’ in rural communities, 
             which in turn explains the predominance of public works projects as their preferred delivery 
             mechanism for food aid, as well as recent shifts in safety net thinking towards cash transfers 
             rather than food aid, with predictable multi-annual transfers expected to lead to ‘graduation’ within 
             a defined time period. 
              
             The discourse on agriculture and social protection (or ‘safety nets’) in Ethiopia might start at a 
             high level of ideological rhetoric and theoretical abstraction, but it plays itself out in real-world 
             policy choices that attempt to find a balance between maximising opportunities and minimising 
             risks. This policy dilemma can be succinctly stated: in a high-risk environment, should you adopt 
             conservative strategies that minimise risk but keep people poor, or push aggressively for growth 
             and ‘grow your way out of poverty’? In the past, the Government has apparently been satisfied 
             with the former approach, but recent policy statements, specifically the ‘Plan for Accelerated and 
             Sustained Development to End Poverty’ (PASDEP), signalled its impatience with the evident 
             failure of this strategy, and embarked on a significant and ambitious shift towards agricultural 
             commercialisation for income generation and wealth creation at household and national levels. 
             Adopted around the same time, the ‘Productive Safety Net Programme’ (PSNP) represents an 
             impatience with decades of food aid that have failed even to sustain rural Ethiopians in their 
             poverty, let alone generated growth, food security and poverty reduction. In a two-pronged attack 
             on rural poverty in Ethiopia, therefore, the PSNP injects cash into a moribund agrarian economy, 
             while PASDEP promotes market chains and export crops that will generate further cash income. 
             This is a major move away from a ‘survivalist’ preoccupation with growing food for subsistence 
             and delivering food aid for survival when food production is inadequate. Whether it will succeed 
             only time will tell. 
              
             This paper explores the linkages between social protection interventions and support to small 
             farmer development in Ethiopia. The paper is divided into six main sections. Following this 
             Introduction, Section 2 argues that agricultural policies and social protection policies in Ethiopia 
             have become increasingly convergent and synergistic in recent years. Section 3 explores the 
             paradoxical relationship of small farmers to land, which is regarded by the government not only 
             as a key productive asset but also as a fundamental ‘safety net’, but could also be interpreted as 
             contributing to the ‘poverty trap’ that millions of Ethiopians find themselves in. Section 4 briefly 
             reviews recent developments in agricultural policy in Ethiopia, focusing on the recent drive for 
             agricultural commercialisation as encapsulated in the shift from ‘ADLI’ to ‘PASDEP’. Section 5 
             analyses the components of the ‘Food Security Programme’ in detail, with empirical evidence of 
             the impact of the Productive Safety Net Programme (PSNP) on agriculture. Section 6 discusses 
             recent innovative interventions that build linkages between social protection and agriculture, such 
             as weather-indexed drought insurance and the design of an ‘integrated drought risk management 
             plan’ and a commodity exchange for Ethiopia. Section 7 concludes. 
                                                               Page 2 of 18 
                     
                     
                    2     Agriculture and social protection in Ethiopia: complementarity or 
                          convergence? 
                     
                    The Government of Ethiopia’s policies for agriculture and social protection follow a trajectory that 
                    can be interpreted as a kind of convergence. In the past, policies of ‘agricultural promotion’ and 
                    interventions that might now be labelled ‘social protection’ were more or less distinct, linked only 
                    by the fact that social protection – or more accurately, safety nets or humanitarian relief – was 
                    triggered as a response to agricultural failure. When harvests failed, safety nets intervened to 
                    protect farming families against the severest consequences. This sequential separation can be 
                    conceptualised as a ‘seasonal policy timeline’, with agricultural policy (such as inputs provision) 
                    delivered during the farming season and safety nets (typically food aid or food-for-work) delivered 
                    during the ‘hungry season’ several months later. In this sense, agricultural policy and social 
                    protection policy are mirror images: the more effectively farming fills household granaries, the 
                    smaller the annual appeal for humanitarian assistance, but several million Ethiopians need 
                    ‘emergency relief’ for several months every year to see them through to the next harvest, and in 
                    years of catastrophic crop failure this figure rises to 12 or 14 million. 
                     
                    Terminology matters: the phrase ‘social protection’ is not yet current in Ethiopia, perhaps because 
                    of its close association with ‘social welfare’ – and by extension, ‘dependency’ – to which the 
                    government is strongly opposed because it believes in self-reliance at household and community 
                    levels, especially in rural areas where most people are either farming or have relatives farming or 
                    otherwise working for them. Instead, Ethiopia has decades of experience with ‘safety nets’, 
                    signifying transitory support mechanisms of last resort (rather than institutionalised permanent 
                    welfare systems), and more recently it has introduced ‘productive safety nets’ (as discussed 
                    below), emphasising the synergies that the government aims to achieve between ‘livelihood 
                    protection’ and ‘livelihood promotion’. 
                     
                    There have been some unintended negative interactions between the two sets of interventions. 
                    For instance, public works projects are often implemented during the months leading up to the 
                    next harvest – at the peak of the hungry season, but also at the peak of the farming season. The 
                    competition for household labour between public works and the household farm often forces 
                    farmers to neglect their fields, compromising the next harvest and perpetuating the cycle of 
                    agricultural under-performance and rural food insecurity. The main policy response has been to 
                    tinker with design and implementation modalities: mainly to try to time public works better, and 
                    there have been suggestions (not yet implemented to our knowledge) that works should be 
                    undertaken before the farming season starts, with payments deferred until the ‘hungry season’. 
                    (This was proposed as long ago as 1993, in the Directives in support of the National Policy on 
                    Disaster Prevention and Management: “coupons, in lieu of wages, could be redeemed 
                    immediately or at a future date”.) A more radical recommendation would be to de-link the delivery 
                    of social protection from any labour requirement altogether. The asset creation benefits of public 
                    works are too inadequate to justify the conscription of cheap labour, while the poorest and most 
                    vulnerable households are typically labour-constrained, rather than underemployed with ‘surplus 
                    labour’ to allocate to public works. 
                     
                    More recently, the positive linkages between agricultural policy and social protection policy have 
                    become increasingly recognised, in general and in Ethiopia, and there are attempts in many 
                    countries to achieve synergies between the two. But recognition that positive linkages might exist 
                    is not new. Thinking on ‘linking relief and development’ in Ethiopia since the early 1990s has 
                    concentrated on efforts to generate agricultural growth through safety nets, by using public works 
                    programmes to simultaneously transfer food rations (a classic ‘consumption smoothing’ safety net 
                    objective) and also (in theory) to construct useful economic infrastructure such as roads (to 
                    integrate markets for farmers and traders), or to subsidise agricultural activities such as vegetable 
                    gardens (promoting production of secondary food crops for both consumption and sale). 
                     
                    Faced with ‘low input, low output’ agriculture, it might seem logical for policy-makers to assume 
                    that farmers face binding input constraints, and that the solution from both an agricultural and a 
                                                                                                   Page 3 of 18 
                     
                    social protection perspective lies in the intensification of smallholder production to maximise 
                    yields from small plots of farmland. This thinking underpins many interventions that can be 
                    described as ‘productivity-enhancing safety nets’ (Devereux 1999), such as Sasakawa Global 
                    2000, which delivers fertiliser and seeds to farmers on a revolving credit basis and has had some 
                    success in raising crop yields in some places – including parts of Ethiopia – at some times. 
                    Unfortunately, revolving credit schemes depend on reliable repayment and do not cope well with 
                    variability in production and repayments. Global 2000 projects are prone to collapse whenever a 
                    bad harvest undermines farmers’ ability to repay their loans. In Ethiopia, a perverse outcome 
                    occurred in the 1990s, when farmers were encouraged to take on inputs packages as loans, and 
                    were actually imprisoned when drought shocks left them unable to repay these loans after a failed 
                    harvest. This experience raises broader questions about the logic of providing social assistance 
                    or income-generating support to poor households in the form of (even zero-interest) loans, which 
                    could simply leave them indebted and worse off than before – another example of a ‘negative 
                    synergy’ between agricultural and social protection objectives. 
                     
                    In fact, efforts at achieving synergies in both directions (promoting agricultural growth in Ethiopia 
                    through ‘productivity-enhancing’ safety nets, or achieving social protection through risk-reducing 
                    agriculture) have been persistently compromised by the instability of the natural environment, 
                    especially fluctuations in rainfall on which smallholder agriculture depends. Neither investments in 
                    agriculture nor investments in social protection appear capable of dealing with this risk. Irrigation 
                    might seem a logical way to proceed, but the scale of the need is so vast and the available 
                    moisture in the highlands is so low that large-scale irrigation has never emerged as a realistic 
                    option, though it is repeatedly recommended. 
                     
                    3     ‘Land politics’ and social protection in Ethiopia 
                     
                    Successive Ethiopian regimes have located the source of Ethiopia’s economic stagnation and 
                    vulnerability in the agriculture sector, yet they have also looked to smallholder agriculture as the 
                    potential source of economic growth, household and national food security and poverty reduction. 
                    In 2000, Prime Minister Meles Zenawi captured this ambivalence succinctly when he said: “The 
                    agricultural sector remains our Achilles heel and source of vulnerability. Nonetheless, we remain 
                    convinced that agricultural based development remains the only source of hope for Ethiopia.” At 
                    the heart of this ambivalence is the politics of land. 
                     
                    The overthrow of Emperor Haile Selassie following the 1974 famine signalled the abrupt ending 
                    of an essentially feudal system in Ethiopian agriculture. Colonel Mengistu’s Derg regime believed 
                    that unequal landholdings and exploitative labour relations based on sharecropping explained 
                    Ethiopia’s persistent vulnerability to famine, as well as constituting a gross social injustice. The 
                    Derg military dictatorship used its unfettered power to force a radical agrarian transformation on 
                    rural Ethiopia. Between 1976 and 1991, all farmland in the highlands was confiscated by the state 
                    and redistributed equally per capita within rural communities. This radical land redistribution was 
                    motivated by both egalitarian and efficiency concerns. The intention was not only to break the 
                    power of the landlords over the peasants, but to give all rural households the means to achieve 
                    sustainable increases in agricultural productivity and rural incomes (Devereux et al., 2005). 
                     
                    Importantly, land was conceptualised as a safety net for rural households by the Derg (a view that 
                    is shared by Meles Zenawi’s EPRDF government). As long as rural families enjoyed guaranteed 
                    access to land, they retained the potential to generate a subsistence livelihood, and in this sense 
                    the Derg’s land redistribution can be seen as a crude form of social protection. Motivated by its 
                    socialist egalitarian ideology, the Derg also implemented agricultural policies such as state farms, 
                    villagisation and forced resettlement, all of which failed to generate the intended growth in 
                    agricultural productivity and were later abandoned. It is no coincidence that the Derg presided 
                    over an even worse famine, in 1984, than the famine of 1974 that had paved its accession to 
                    power. The land redistribution remains as the Derg’s lasting impact on Ethiopian agriculture. 
                     
                    After the Derg was overthrown in 1991, the EPRDF government under Prime Minister Meles 
                    Zenawi reaffirmed the Derg’s commitment to land equality, and Meles has vociferously resisted 
                                                                                                    Page 4 of 18