From State Dirigisme to Liberalisation in Senegal: Four Decades of Agricultural Policy Shifts and Continuities

Abstract

Structural adjustment in agriculture has been pervasively applied in sub-Saharan Africa since the early 1980s. This article traces the trends and shifts in Senegal's agricultural policy and developments during the post-independence period 1960–2000. The article presents a periodization into policy phases that share commonalities in terms of the macroeconomic context, dominant ideological paradigms, the nature of state intervention and its relation with domestic and external pressures. The article also provides empirical evidence concerning the evolution of major crops and particularly looks at the crisis of the groundnut sector, once the backbone of the Senegalese economy. Agricultural policy in Senegal shifted, somewhat erratically, from a ‘dirigiste’ stance in the first two decades of independence, displaying some continuities with colonial policies, and influenced by a compromise between ‘technocractic developmentalism’ and ‘Rural Socialism’, towards an increasingly incoherent and self-destructive liberalisation programme, triggered by the fiscal and balance of payments crisis of the late 1970s, and driven by competing and shifting demands on the Senegalese government.

L'ajustement structurel dans l'agriculture a été massivement appliqué en Afrique subsaharienne depuis le début des années 80. Cet article met en évidence les tendances et les changements dans la politique agricole du Sénégal au cours de la période post-indépendance 1960–2000. Il présente une périodisation des politiques économiques en fonction du contexte macroéconomique, des paradigmes idéologiques dominants, de la nature de l'intervention de l'Etat et de la prise en compte des pressions internes et externes. Il montre, statistiques à l'appui, l'évolution des principales cultures et accorde une attention particulière au secteur de l'arachide, autrefois pilier de l'économie sénégalaise. La politique agricole au Sénégal a évolué de façon quelque peu erratique. Au cours des deux premières décades suivant l'indépendance, une position dirigiste a été adoptée, montrant une certaine continuité vis-à-vis des politiques coloniales. La politique agricole était alors le fruit d'un compromis entre « développementalisme technocratique » et « socialisme rural ». Elle a évolué ensuite vers un programme de libéralisation de plus en plus incohérent et auto-destructeur, déclenché par la crise fiscale et de la balance des paiements de la fin des années 70 et conduit par des exigences changeantes et concurrentes envers le gouvernement sénégalais.

Notes

 1. For a similar article on a different country (Malawi) see Harrigan (2003).

 2. A summary of the agricultural policies implemented in Senegal between 1960 and 2000 is presented in Table A4 in the form of a matrix.

 3. A focus on these crops is justified not only by the fact that most rural people still depend on them for their survival (according to the latest Agricultural Census, 91% and 73% of agricultural households grow millet/sorghum and groundnuts respectively) or that rice is a major staple for household consumption. It is also the fact that long- and medium-term policy changes can be more clearly traced by looking at major crops with some strategic dimension to the country. Agricultural policies concern other crops, like cotton, maize, niebe, watermelon, horticultural produce, but have not represented the backbone of Senegalese agriculture, even if they have grown in significance recently (Touré and Seck, 2005). In this article, for the same reasons we do not deal with forestry and livestock-related policies. On the latter see some key issues discussed in the proceedings of the regional workshop on Livestock Policies in Dakar 17–18 Nov. 2004, < www.fao.org/ag/againfo/projects/en/pplpi/docarc/mrp_partagees.pdf>.

 4. On the nature and political contradictions surrounding the design of Animation Rurale in Senegal see Gellar et al. (1980).

 5. In practice, policy continuity had been partly facilitated by investments in the training of Senegalese cadres in the second half of the 1950s. This was consistent with the new approach of France towards its colonies, especially Senegal, which envisaged increasing the number of Africans in the colonial administration (50% in 1956 and 67% in 1958) (Marfaing and Sow, 1999).

 6. France maintained support for groundnut prices (prices guaranteed by the Caisse de Stabilisation) until 1967, when European Community regulations came into force, and export prices of some products could only benefit from the preferential treatment given by the EEC through the STABEX programmes.

 7. The notion of ‘developmental state’ has been widely discussed. On the myths and realities of a ‘developmental state’ in Africa see Mkandawire (2001).

 8. Many grands marabouts became very important producers and dominated large-scale farming during this period as a result of these enhanced links with the state (Boone, 1990: 208).

 9. The system of seed distribution to farmers used co-operatives as intermediaries, on the basis of a principle of solidarity. That is, co-operative members had to contribute to compensate for the debt defaults of other members. However, many farmers, especially in the worst seasons, did not respect this principle and defaults were generalised (Kelly et al., 1996: 18; Caswell, 1984: 69).

10. In this period ONCAD became a giant agency with diverse responsibilities, from groundnut marketing, to co-operative management, encompassing input sales, extension, consumer goods and food aid distribution (Caswell, 1984).

11. Seed availability and quality is usually considered a crucial factor in groundnut production. Indeed seed policy failure in the adjustment period underlies the aggravation of the groundnut crisis in the adjustment period (Gaye 1998; Kelly et al., 1996; Freud et al., 1997; CIRAD, 1997).

12. Mbodj (1993: 91) noted that the system of cooperatives, state development agencies and marketing boards ‘was planned to kill two birds, one socio-political, the other economic, with one policy stone’.

13. The ‘Berg report’ of 1981 (World Bank, 1981) had established some clear parameters around a ‘getting prices right’ slogan.

14. According to the World Bank, in the 1990s over $US1 billion was lent by the World Bank to Senegal compared with US$658 million in the 1980s, resulting in a debt that represented 43% of Senegal's total external public debt outstanding in 2001(World Bank, 2004: 2). By 1991, the debt service amounted to 48% of total fiscal revenues and 74% of current government expenditure (Duruflé, 1994: 81).

15. Duruflé (1994) has consistently argued that growth figures in this period are somewhat inflated.

16. Control of the government deficit was rapidly achieved after 1983. However, the burden of fiscal adjustment was borne on the expenditure side. After 1985, capital expenditures on equipment were, in real terms, only 25–35% their levels in 1980 (Oya, 2002: 127). Expenditures on maintenance, intermediate goods and transfers were also drastically reduced in the first years of adjustment, falling by 40% between 1980 and 1986 (Duruflé, 1994: 83). At the same time, however, the number of civil servants continued to increase.

17. For Gerschenkron, a collective action failure in a laissez-faire environment requires the state to mobilise and concentrate capital for social infrastructure and direct credit to sectors with most dynamic potential, reduce the risk of investment, promote exports with performance-based incentives to exporters, and induce the mobilisation of labour to more productive sectors. See Schwartz (2000: 59–62).

18. The pressure to privatise was not exempt of contradictions, as privatisation was accompanied by a reinstatement of protection measures that had been lifted in the previous adjustment loans, just to ‘entice investors and get privatization off the ground’ (Bayliss and Cramer, 2001: 61).

19. The social discontent, particularly in urban areas, was linked to the pernicious effects of adjustment efforts on real wages and the growing de-industrialisation that followed the implementation of the ‘liberal’ New Industrial Policy of 1986. The industrial sector only partially recovered after devaluation as a result of the temporary effect on vegetable oil milling. See Duruflé (1994: 62 and 200), Seck (1997) and Oya (2002).

20. Agricultural data are always problematic in SSA countries (Sender and Smith, 1986: 100–101). For the groundnut sector in Senegal, Freud et al., (1997) had to produce alternative series from 1989 onwards due to systematic over-reporting in official data, especially in years of agricultural recession. Therefore, any unqualified statement concerning these figures (official or unofficial) needs to be taken with caution.

21. Some micro-surveys and official figures themselves suggest that falls in production were mainly accounted for by yield reductions (Badiane and Gaye, 1997; Freud et al., 1997; DISA (Direction des Statistiques Agricoles) data, various years).

22. In this period of impasse, the government strengthened its position as major shareholder of SONACOS, and this enterprise was in fact reinforced and expanded. SONACOS assumed the primary groundnut collection functions of ONCAD and subsequently also took over seed distribution (after SONAR's dissolution), for which a subsidiary, SONAGRAINES, was created (with 100% of capital owned by SONACOS, which was 90% state-owned by 1982).

23. See Martin (1988) for a fuller explanation of the unrealistic aims of the Cereal Plan.

24. See Hirsch (1990) and Dahou (2004) for more on the inconsistencies surrounding the Cereal Plan and rice policies.

25. See GOS (1986), Gaye (1998), Kelly et al. (1996) and Kelly and Delgado (1991).

26. Few existing private traders dealt in fertilisers and farmers had been accustomed to purchasing fertilisers on credit and seldom for cash (Kelly et al., 1996).

27. Arguably, input constraints on groundnuts had negative side-effects on coarse grains since fertilisers and pesticides were often shared across crops (Kelly and Delgado, 1991; Oya, 2002)

28. The number of rural co-operatives was drastically reduced, from about 2,300 to 317. The village co-operatives were reconverted and reduced to sections villageoises (4,472), linked to the new 317 ‘mother’ co-operatives (Gaye, 1998: 54). The continuous reduction in collection points with the acceleration of the liberalisation process between 1992 and 2004 has been considered by the World Bank and the IMF more recently one of the three most important constraints on marketed surplus for groundnut farmers (IMF, 2005: 13; World Bank, 2003).

29. This was particularly true in Touba, the Mouride capital, which soon became the capital of agricultural ‘parallel markets’, hence creating poles of accumulation among large traders and attracting increasing numbers of people from neighbouring areas.

30. Groundnut prices in real terms were in 2003 and 2004 only 65% and 80% respectively of the peak 1975 level and still below the 1960 level.

31. In the latest IMF report on Senegal, it is suggested that in a context of pure competition and a privatised SONACOS, producer prices would be much lower than current prices set by the CNIA. Thus, it is implicitly acknowledged that there may be a trade-off between providing price incentives to farmers and exposing farmers to international competitiveness (IMF, 2005: 11). In a recent interview, the CNIA president AJ Ndiaye declared that the indicative price in 2004 was well above market prices and added: ‘Nous étions cependant conscients du fait que descendre sous la barre de 150 francs le kilo serait donner une potion amère aux paysans qui sortaient d'une campagne agricole particulièrement difficile’. This ‘subsidy’ was only possible with the direct support of the government and the EU (Le Soleil, 21 March 2005).

32. Indeed, similarities with other SSA countries are striking. It is often difficult to distinguish agricultural programmes recommended for different countries. The measures, sequencing and emphasis were repeated invariably from country to country during the 1990s, in spite of the rhetorical call for ‘national ownership of agricultural strategies’.

33. See the results of IFPRI post-devaluation surveys on household consumption in FCFA West Africa published in Diagana et al. (1999) and Diagana and Reardon (1999), the latter specifically for urban Senegal. They present several striking results, which question conventional expectations about the devaluation. For example, total cereal intake fell, especially for the poorest people, but imported rice intake increased; there was a ‘de-diversification’ of the diet and a process of ‘individualisation’ of consumption patterns. These results on the consumption side actually reflected the poor performance of the cereal sector in the supply side and an increase of food prices associated with the inflationary effects of devaluation. Dahou (2004) also notes that costs per hectare of a rice plot increased significantly and became prohibitive for a range of producers.

34. Rice import liberalisation in 1995 clearly offset the expected effects of devaluation on the import bill.

35. By Nov. 2001, some expected but delayed measures had been taken: the dissolution of SONAGRAINES as a body for groundnut marketing and input provision, as well as the creation of ANCAR, a new private-and-demand led rural extension agency, which replaces former public institutions like SODEVA.

36. The privatisation agenda has been stepped up and promoted across the board. SONACOS was finally privatised in 2005, despite the difficulties of finding buyers, which forced the government to accept a bid below expectations. SODEFITEX (cotton-textile) was privatised in 2003. It is worth noting that privatisation has reinforced the domination of external capital in the national industry (Advens group for SONACOS –dominated by the Belgian D. Smet and by Dagris – and group Dagris for cotton).

37. It is not clear how sustainable these measures will be. The same applies to other apparent changes that may be forthcoming with the approval of the Loi d'Orientation Agro-Sylvo-Pastorale in 2005 and the publication of the Lettre de Politique de Développement de la Filière Arachide in 2003 but for the moment this is mainly visible in official documents rather than implemented measures. See different government web declarations in < www.agriculture.gouv.sn> and particularly read < www.agriculture.gouv.sn/pdfs/agrimoteurces.pdf>.

38. However, government efforts to induce a ‘fair’ producer price through the CNIA are more futile now that the market is liberalised and SONACOS buys a declining share of domestic groundnut production. Producer prices paid by private traders tend to be well below the CNIA reference price, which has distressed the farming associations engaged in protest movements in the past few years.

39. On 26 Jan. 2003, more than 20,000 farmers gathered in Dakar football stadium to demonstrate against the government's inaction vis-à-vis agriculture and pronounced a manifesto with their claims.

40. Migration to the cities and overseas, partly pushed by the agricultural crisis has gradually transformed the political economy of Senegal to the extent that international migrants have become a very significant economic and political force in the country (Dahou and Foucher, 2004: 8).

41. The unsystematic and aggregate evidence on industrial real wages from World Bank data points to continuous and unambiguous reductions since stabilisation and adjustment began.

Additional information

Notes on contributors

Carlos Oya

Carlos Oya is Lecturer in the Department of Development Studies at the School of Oriental and African Studies, London, [email protected]