II. Distributional effects of reforming the groundnut sector1
6. As part of its poverty reduction and growth strategy, Senegal is about to complete two major reforms in the groundnut sector: the sale of the state-owned groundnut processing company and the elimination of tax and tariff preferences protecting that company. The first section of this chapter reviews the history of the reforms, followed, in section 2, by an analysis of their distributional effects on groundnut farmers. The third section discusses the need for further reforms. The fourth section addresses policy responses to environmental risks. The concluding section presents the key findings and policy recommendations.
A. Background: Structure of the Sector and Attempts at Reform
7. The groundnut sector accounts for a dwindling size of the Senegalese economy, but plays a critical role in the livelihood of about 30 percent of the population, especially in the rural area where the poverty rate is the highest (65 percent). In 2003, it represented about 1.4 percent of GDP and contributed to 4.6 percent of exports (against, respectively, 7 percent of GDP and 80 percent of exports in 1960). Climatic changes, a declining fertility of the soil, environmental degradation, but also failed interventionist policies, contributed to a secular decline of the sector.
8. The groundnut sector is dominated by the state-owned Société Nationale de Commercialisation des Oléagineux du Sénégal (SONACOS), the fifth largest company in Senegal. It employs about 950 people and has five industrial plants, with excess capacity and outmoded technology. During the crop year 2003/042, about 320,000 households produced 445,000 tons of groundnuts, and SONACOS processed about 100,000 tons in groundnut oil and animal feed for export. The balance was either purchased by NOVASEN, a much smaller (and private) oil milling company, used for domestic consumption, sold in neighboring countries or stored as seeds for future crops. SONACOS is an important player on the world market for groundnut oil, holding a market share of about 20-25 percent. But the world market has been declining by one percent on average during the last twenty years, as cheaper and healthier vegetable oils (such as soy and palm oil) have emerged.
9. The price at which groundnut is purchased from farmers is set at the start of the harvest season by the Centre National Interprofessionnel de l’Arachide (CNIA), a quasi-public regulatory agency. SONACOS, NOVASEN, groundnut producers, private marketing agents, transporters, seed distributors, the state-owned agricultural bank, Caisse Nationale de Crédit Agricole du Sénégal (CNCAS), and the Ministry of Agriculture and Livestock are represented in the CNIA. The price is set taking into account developments in world market price for groundnut oil (Figure 1).3
Figure 1.Senegal: Groundnut Prices and Production
10. SONACOS also has a dominant—and protected—position in processing imported unrefined vegetable oil for the local market. Its market share of almost 75 percent in that market was, until recently, shielded by a combination of tariff and domestic taxes which resulted in a protection equivalent to about 30 percent of the price of refined vegetable oil. Indeed until end-May 2004, refined vegetable oil imports were subject to a total tax rate of about 65 percent, while imports of unrefined vegetable oil were subject to a tax rate of about 35 percent (Table 1). In June 2004, however, the 10 percent contingent import tax on refined vegetable oil was suspended. The excise duty on refined (15 percent) and unrefined (5 percent) vegetable oil is scheduled for repeal by the end of 2005. At that time, SONACOS’ tariff and tax protection against refined vegetable oil imports will have been reduced to 10 percent, the difference between the Common external tariffs (CET) on final consumption goods (20 percent) and intermediate goods (10 percent) imports.
|Import duties||(in percent)|
|Common external tariff (CET)||20.0||10.0||10.0|
|Contingent import tax||10.0||10.0|
11. In spite of these dominant positions, SONACOS made large losses during three of the last six years. It accumulated losses of CFAF 60 billion (US$ 120 million) in 2000-01, which necessitated budget transfers of CFAF 65 billion (1.9 percent of GDP). It has been barely profitable since then.
12. SONACOS’ privatization and the elimination of specific protective taxes are the last major steps needed to fully liberalize the groundnut sector. After failed interventionist policies that left the government with a debt burden equivalent to 16 percent of GDP in 1980, the government started to gradually disengage from the agricultural sector. In the aftermath of the CFA franc devaluation in 1994, the prices of inputs were liberalized, SONACOS lost its monopoly in the collection of groundnuts for oil processing, and CNIA was put in charge of setting producer prices as a function of world market prices for groundnut oil. During the 2000/01 agricultural campaign, the government returned to a more interventionist policy characterized by setting of producer prices above the level determined by CNIA and the pricing of inputs below market prices, which caused losses for SONACOS and its subsidiary SONAGRAINES, active in the distribution of seeds and fertilizers and the collection of groundnuts. In 2001, the government liquidated SONAGRAINES, financially rehabilitated SONACOS with the assistance of donors, notably the European Union, and committed to privatizing SONACOS and fully liberalizing the groundnut sector. This commitment was confirmed in the Groundnut Sector Development Policy Letter adopted by the government in May 2003 and supported by the World Bank.
13. The authorities have decided to privatize SONACOS and eliminate the specific protective taxes on vegetable oil. After two failed privatization attempts in 1995 and 1999, and following several months of negotiation, the authorities have decided to sell SONACOS to the only bidder that made an offer in June 2004. The transfer of ownership should be completed by end-June 2005. As mentioned above (paragraph 10), the authorities have also started to eliminate the specific tariff and tax preferences protecting SONACOS to provide a more level playing field for the operators in this sector.
B. Distributional Impact of Ongoing Reforms
14. An important question is whether either the elimination of the specific tariff and tax or the privatization of SONACOS will have significant distributional consequences, especially for poor farmers. This section will discuss each of these reforms in turn.
Elimination of tariff and tax preferences
15. It is generally assumed that the tax and tariff preferences that SONACOS has enjoyed have increased the price of imported vegetable oil for consumers. In fact, the rationale for these preferences was to allow SONACOS to extract monopoly rents with which it could subsidize the price paid for groundnuts. The preferences give SONACOS market power that should, in fact, allow it to charge a higher price than would prevail under pure competition. Two questions arise in this connection: (a) what is the distributional effect of a lower price for vegetable oil; and (b) are the higher profits for SONACOS passed on to groundnut farmers?
16. The analysis below shows that the impact of the elimination of these preferences on consumption should be small, but poor consumers may benefit relatively more. The consumption of imported vegetable oil is small relative to the consumption of groundnut oil, especially for lower income groups (Figure 2). 4 The direct effect of a price reduction would therefore be small—less than one half of one percent of consumption, assuming that the entire tax reduction is passed through to consumers. Moreover, since the share of vegetable oil in total consumption is—with the exception of the highest income group—relatively flat across quintiles, the direct benefits of a reduction in price would be roughly proportional to total consumption. If imported oil is a close substitute for groundnut oil, however, the drop in vegetable oil price could lead consumers to substitute vegetable oil for groundnut oil. This would amplify the benefits of the price reduction. To the extent that poorer households consume relatively more groundnut oil, the substitution effect can benefit them more. 5
Figure 2.Senegal: Vegetable Oil Consumption
Source: Enquête Statistique auprès des Ménages (ESAM‐II) and World Bank staff estimates.
17. There is no evidence that the repeal of SONACOS’ tax and tariff preferences will have an adverse impact on groundnut prices. The impact would depend on the extent to which the current groundnut prices are cross-subsidized. CNIA is supposed to set prices on the basis of the world market price, not the preferences enjoyed by SONACOS. There is scant evidence that producer prices set by CNIA have been consistently or heavily subsidized (Figure 1). Therefore, the elimination of the tax and tariff preference should not result in a reduction of groundnut prices. The strongest argument that producer prices will be reduced when tax and tariff preferences are eliminated is that a reduction is necessary for the survival of SONACOS. However, it is unlikely that SONACOS still has the market power to fully pass the required price reduction to groundnut producers as alternative markets (NOVASEN, small groundnut oil refineries, neighboring countries, …) have developed into which producers can sell. 6
18. One reason why the government has been hesitant about selling SONACOS is the concern that a private SONACOS would increase uncertainty about the income of groundnut farmers. For instance, a private SONACOS may be unwilling to absorb the supply of groundnuts, especially in years when harvests are bountiful or the CNIA price is higher than the world market price. Alternatively, it could fight for a lower producer price or question the manner in which CNIA currently sets prices. The discussion above (paragraph 17) suggests that SONACOS would have a difficult time obtaining a lower producer price from CNIA under the current price-setting regime. However, CNIA’s charter has expired 7, and private owners might insist on a different regulatory environment or curtail their purchases when the CNIA price is considered too high. Key issues in the privatization are whether the new owners will be required to: (a) accept something close to the current structure of CNIA, or (b) commit to a minimum level of groundnut purchases. These issues create a two-edged sword. Without a commitment to provide a market for groundnuts at a price determined by a broad committee of sector participants, privatization will indeed increase uncertainty or at least change the nature of the uncertainty in the system. Such a commitment, on the other hand, would subject SONACOS to an overly rigid regulatory environment, in which the full benefits of privatization would not be realized.
19. The potential for increased risk engendered by privatization will depend on the degree of distortion in the current market for groundnuts. If the current CNIA price is too high relative to the world market price—which presently does not seem to be the case (Figure 1)—a private SONACOS facing the same groundnut price will be hard pressed to survive financially, especially without the excess profits generated by the current tax and tariff preferences. Alternatively, if the current price reflects market forces, SONACOS would have no power to reduce it. Moreover, the privatization of SONACOS and the elimination of tax and tariff preferences will open the field for potential competitors, so it will be even more difficult for SONACOS to exert market power to reduce prices, and SONACOS will be under pressure to become more efficient. 8
20. Even if SONACOS could reduce in the price it pays private marketing agents for groundnuts, the ultimate effect on farmers would be mitigated. As noted above, farmers have access to alternative markets that would indirectly limit the ability of SONACOS to cut prices. These outlets would also mitigate the effect of a reduction in the SONACOS price on the average price for a farmer’s entire crop. Moreover, the effect of a lower SONACOS price would also depend on the degree to which private marketing agents can pass this reduction through to farmers. It would likely be a mistake to expect much mitigation from this source, however, without other reforms (see paragraphs 23-26), since the need for quick cash and the lack of access to credit apparently force farmers to supply at least some of the crop at low prices.
21. In sum, although the sale of groundnuts is an important source of income for rural farmers, the privatization of SONACOS need not have a large effect on the well being of poor groundnut farmers. Uncertainty about the institutional arrangements that determine the price SONACOS must pay for groundnuts and the quantity it buys present some risk, especially to the extent that market forces are not adequately reflected in the current environment. However, this uncertainty does not present a greater risk than farmers have experienced in recent years for other reasons. Indeed, the staff estimates that the average cash flow to groundnut farmers has varied by a factor of four during the last decade (Figure 3), mostly as a consequence of the sharp slump in production in 2002/03, following the drought in 2002 (Figure 1).
Figure 3.Senegal: Average Cash Flow to Groundut Farmers
Source: Ministry of Agriculture and Livestock and staff estimates.
C. The Need for Additional Reforms in the Groundnut Sector
22. Privatization of SONACOS has received the bulk of the attention with respect to the reform of the groundnut sector, but some of the other problems currently faced by groundnut farmers are likely to present greater challenges. These include market imperfections in the distribution chain, and the need to adjust to market developments.
Possible reforms to the distribution chain
23. The current distribution chain suffers from a number of market imperfections. First, the CNIA price is not set—and SONACOS does not begin its purchasing campaign—until after the groundnut harvest has begun. This delay puts farmers who face serious liquidity constraints at a disadvantage, as they must sell at least part of their crop without access to SONACOS as a buyer and without the knowledge of the price it will eventually pay. The new institutional framework after the privatization of SONACOS remains unclear. However, if some of the features of the current system are kept, it would be beneficial to farmers to receive more timely information and begin the sale of groundnuts to SONACOS earlier. This would improve the functioning of the market and reduce pressure on farmers to sell their crops in alternative markets at low prices.
24. Another serious problem for groundnut farmers is lack of access to credit. This shortcoming can force farmers to sell early and at a low price. The absence of land titles, unreliable financial statements, high level of non-performing loans, and weak enforcement of creditor rights are the main reasons for the low access to bank credit by most small and medium sized enterprises, and a fortiori groundnut farmers9. Access to microfinance credit is also hampered by the below average loan repayment track record of groundnut farmers (ADE, 2002, p. 65). An elaborate government program to provide or guarantee credit would likely be costly and run the risk of weakening market discipline. Addressing information asymmetries and weaknesses in the legal and judicial framework, notably the absence of land titles, are essential if credit availability is to improve, obviating the need for farmers to sell their crop under duress.
25. Direct access to SONACOS collection points is available for both individual farmers and farmer cooperatives, but it is not often used because of its inconvenience. One vision for restructuring the collection of groundnuts with the liquidation of SONAGRAINES in 2001 was to have farmers cooperatives deal directly with SONACOS, obviating the middlemen. However, the limited availability of SONACOS collection points10 and lags in grading the crops and payment by SONACOS have been sufficient that farmers typically have preferred to deal with the private agents, even if it has meant a lower price. Consequently, the role of the private marketing agents was expanded, and the collection and distribution of groundnuts became less transparent. If the privatization leads to easier access to SONACOS as well as faster quality grading and payment, it will provide a viable option to dealing with private marketing agents. It could not only increase the price received by the farmer, but also provide indirect regulation for the private agents.
26. Finally, the current system of dealing through private marketing agents poses two problems. First, the monopsony of private marketing agents in remote areas could lower the price paid to farmers. Second, farmers have no information about prices and quantities purchased in different parts of the country. A system for monitoring prices actually paid by marketing agents to producers would improve the transparency of the groundnut market, allow farmers to make more informed marketing choices, and alert the government about potential abuses of power arising from lack of competition in some market segments.
Need to adjust to market developments
27. Competitive factors and consumers’ changing preferences for different vegetable oils pose a challenge for groundnut farmers, whose primary—if not the sole—export market is that of processed oil. The competitive pressures arise from the inefficiency of the SONACOS oil refining operations (owing to old plants and excess capacity). At the same time, the worldwide export market for groundnut oil is declining, as groundnut oil is replaced by healthier or less expensive alternatives. The percentage of groundnut output worldwide processed into oil declined from 60 percent in 1980 to 48 percent in 2000 (Diop, et al., 2004). SONACOS and possible competitors will have to shape their activities to reflect these market forces.
28. A more rigorous quality assurance capacity than is currently available may be necessary to diversify into confectionary nuts. The percentage of groundnuts worldwide used for food has risen, and roasted nuts for snacks and confectionary nuts are also potentially lucrative exports. To enter this market, however, requires different seeds and production and processing methods that ensure high enough quality for export to developed countries. Population growth in Senegal and the establishment of a regional common market, present opportunities for marketing groundnuts in domestic and regional markets, which could be better exploited.
D. Response to Environmental and Other Structural Risks
29. Environmental risks are the most important challenge faced by groundnut farmers. First, rainfall in most of Senegal and in the groundnut basin in particular is both highly variable and subject to a significant downward trend (Figure 4). The variability creates substantial short-term risk, such as was evidenced in very low 2002-03 crop. The risk of drought, in particular, has been—and is likely to continue to be—one of the main sources of risk for groundnut farmers, with significant implications for short-term income variability. Second, the downward trend in rainfall presents medium and long-term risks. Since the level of rainfall increases from north to south, this trend has caused a gradual shift of groundnut farming toward the south. This shift has been accompanied by the clear cutting of new land reducing forest coverage and subjecting the landscape to additional environmental risk. Finally, soil quality in the groundnut basin has been deteriorating for several reasons, since farmers do not have the option of practicing proper soil maintenance. They cannot afford to rotate crops or to allow fields to rest. In addition, fertilizers are expensive, so they are often applied sparingly, adding to the strain on the soil.
Figure 4.Senegal: Rainfall and Groundnut Production
30. In the medium to long term, the authorities will have to decide what steps they want to take to protect the natural resources exploited in the groundnut sector. These steps could involve land management practices that would reduce the current intensity of use in exchange for longer-run sustainability. Such measures could, in turn, have a negative effect on groundnut farmers that would require some compensatory measure.
31. Government policy with respect to the groundnut sector—no matter how enlightened—cannot eliminate the risks groundnut farmers face. The average scale of groundnut farming is likely to have to increase in order to raise farmers’ living standards, and it will not be able to support the share of the rural population that it currently does. As the rural population grows, economic pressures in the groundnut farming areas will intensify. It is important that residents in these areas be able to develop the skills necessary to allow them to engage in alternative activities, either in rural or urban areas. This could entail not only promoting education and health, but access to more specific training to broaden job opportunities.
32. A safety net is needed to provide assistance to farmers in the event of a serious shock. The evidence presented above is that modest price adjustments to promote market efficiency are not as serious a problem for farmers as a large shock, such as a drought or a collapse of world prices. The regulation of SONACOS or price setting mechanism cannot adequately compensate for such a shock without seriously distorting the sector and creating nontransparent cross-subsidies. It is better to allow the market to make the necessary adjustments to a shock and then compensate those adversely affected in a transparent and equitable manner.
33. A safety net based on price subsidization, although easy to administer, would have drawbacks. A price subsidy would benefit large groundnut producers more than the smaller producers, but that does not mean that such a subsidy necessarily benefits the richer farmer more than the poorer ones because the volume of groundnut output is not a good proxy for total income. A better proxy would be the level of total consumption. Groundnut production is relatively uniformly distributed among the four lowest and five highest deciles of the farmers’ population grouped by total consumption levels (Figure 5). 11 Consequently, a price subsidy would benefit relatively richer farmers (including those who are above the poverty line) as much as the poorer ones (i.e., those with small consumption levels). 12 Moreover, farmers who do not grow groundnuts—but who may be equally poor—would receive none of the benefits. In addition, a groundnut price subsidy would divert the allocation of resources towards the groundnut sector and away from more productive uses.
Figure 5.Senegal: Share of Groundut Production
Source: World Bank Staff Estimates.
34. A more efficient social safety net program would explicitly target poor farmers for compensation in response to a severe shock. In the short run, the authorities should use budgetary transfers directed to the poorest farmers to dampen the impact of a price or weather-related shock. Senegal should develop the capacity to target assistance based on household characteristics that are good proxies of overall wealth. In the long-run, conditioning transfers to poor households on acquiring human capital to get access to alternative sources of income would reduce poor farmers’ dependence on groundnut farming.
E. Summary of the Conclusions
35. The sale of SONACOS and the elimination of the tax and tariff preferences should not have a negative impact on the income of groundnut farmers. Privatization does present potential risks for groundnut farmers, as a new system of price setting will be negotiated and the privatized SONACOS may want to reduce producer prices to restore its profitability. But SONACOS is unlikely to have the market power to impose a lower than market price on groundnut producers. In view of the small weight of vegetable oil in consumption and the uniform distribution of vegetable oil consumption across the population, the tax and tariff repeal should benefit consumers across the board. However, poor farmers may benefit more to the extent they can shift a larger proportion of their consumption from the more expensive groundnut oil to the cheaper imported vegetable oil.
36. More reforms are needed to address market imperfections in the distribution chain and encourage adjustment to likely market developments. Market imperfections arise from a lack of timely information about the groundnut price for oil production and the monopsony position of marketing agents in some areas. A policy that improves the information flow, supports the development of alternative marketing options for groundnut farmers, and enhances the availability of credit to farmers will help them face the challenge of evolving groundnut market conditions.
37. Environmental risks are the most important challenge faced by groundnut farmers, which may necessitate a safety net. Rainfall is volatile and decreasing and groundnut soil is being depleted. A long-term plan for protecting the natural resources exploited in the groundnut sector and, by implication, the livelihood of groundnut farmers needs to be developed. Through training opportunities and other human capital development policies, groundnut farmers should also be encouraged to diversify their activities to reduce their exposure to exogenous shocks. The safety net against severe exogenous shocks should take the form of direct transfers to the poorest farmers, rather than subsidized prices. A price subsidy would benefit poor and rich farmers equally and distort resource allocation.
Aide à la Décision Economique (ADE)2002Sénégal: Evaluation à mi-parcours du programme de relance de la filière arachide Rapport provisoire Louvain-la-Neuve.
BadianeCoura2001 “Senegal’s Trade in Groundnuts: Economic, Social, and Environmental Implications,” TED Case Studies Number 646.
BoccanfusoDorothee and SavardLuc2004 “Impacts Analysis of the Liberalization of Groundnut Production in Senegal: A Multiple Household Computable General Equilibrium Model,” Background paper for the forthcoming “World Bank 2005 Sénégal: Réformes du secteur de l’arachide et pauvretéWashington, DC.
International Monetary Fund2001 “Senegal: Selected Issues,” IMF Country report no. 01/188.
Ministère de l’Agriculture2001Recensement National De L’Agriculture 1998-1999 Volume 4: Rapport Général du Recensement de l’agriculture Pluviale.
NdiameDiopBeghinJohn and SewadehMirvat2004 “Groundnut Policies, Global Trade Dynamics, and the Impact of Trade Liberalization,” draft report for the World Bank.
Prepared by Robert Gillingham, David Newhouse, and Christian Josz
The world price of groundnut in Figure 1 is estimated by assuming that it is a fixed fraction of the world price of groundnut oil (Boccunfuso and Savard, 2004). The fixed fraction is obtained by dividing the world price of groundnut in 2001 reported in Badiane (2001) by the world price of groundnut oil in 2001. The error in this estimated price will depend on the accuracy of the observation in 2001, as well as the assumption of a fixed ratio. Even as an approximation, however, it should provide some evidence on whether CNIA has allowed the producer price in Senegal to stray from the world price.
Groundnut oil is supplied by smaller refineries producing solely for the domestic market.
The tax repeal could also have distributional effects through the fiscal policy reaction to the revenue loss, if any. However the revenue loss should be small as these taxes raised less than CFAF 3.5 billion (0.1 percent of GDP) in 2003, and additional vegetable oil imports triggered by the elimination of these taxes should generate more custom duty and VAT receipts.
Although SONACOS is still by far the largest buyer of groundnuts in Senegal, the share of the groundnut output it purchases, 40 percent on average over the last 20 years, has been on a decreasing trend since the 1960s.
Since SONACOS was scheduled to be privatized in 2003, the charter for CNIA was allowed to expire at the end of 2003 to allow SONACOS’ new owners to help structure a new charter. CNIA is continuing to operate under the expired rules pending the completion of the privatization. It is likely that a new charter will be negotiated in the near future.
SONACOS currently employs substantially more workers than it needs to conduct its core businesses and has two idle groundnut refineries. Consequently, it is likely that a new owner could improve efficiency, as long as it was given sufficient latitude to restructure its operations.
See “Senegal-Financial System Stability Assessment-Update”, IMF Country Report No. 05/126.
Groundnuts can be harvested as early as October, but SONACOS does not accept groundnuts for processing until the beginning of the groundnut campaign, which typically takes place in December. According to a recent survey by the World Bank, the shortage of collection points counts among the main problems faced by groundnut producers in marketing their crops, together with low prices and the lack of access to credit.
Figure 5 shows the distribution of groundnut output for a sample of farmers in groundnut producing regions divided into 10 groups in ascending order of their consumption levels (deciles on the horizontal axis). The bar charts represent the share of groundnut output produced by each decile of the farmers’ population. Excluding the 5th decile, the distribution is relatively uniform.
The three top deciles of Figure 5 (representing approximately 35 percent of the rural population), though largely poor, are above the poverty line. About 65 percent of the rural population falls below the poverty line.