Speaker:
Andrew Dorward, Research Fellow, Centre for Development, Environment and Policy, SOAS, University of London

Discussant:
Paul Harvey, Research Fellow, Humanitarian Policy Group, ODI

Chair:
Martin Prowse, Research Officer, Protected Livelihoods and Agricultural Growth (PLAG) Programme, ODI

Meeting Report

1. Martin Prowse, in the chair, introduced the fourth meeting in the 'From broad themes to practical policy' series being co-hosted by ODI and FAC. The meeting would be considering the limits to state action in the agriculture sector, using the recent Malawian fertilizer subsidy as a case study.

‘Testing the limits of state action: The Malawi fertiliser subsidy’
Andrew Dorward, Professor of Development Economics, School of Oriental & African Studies, University of London

2. Andrew Dorward presented an evaluation of the 2006/7 Malawian agricultural input supply programme and its maize fertilizer subsidies, focusing on the background to the programme, its development and implementation, and its impacts.

3. Malawi suffers from high rates and wide geographical dispersion of poverty. Most farming households are poor due to low levels of food production, particularly maize, which require them to buy in additional food. Maize prices within Malawi are also highly volatile, forcing poor farmers to grow maize despite low yields, and resulting in a rural economy trapped in a low productivity maize production. There is a history of providing financial assistance to farmers in the country, with fertilizers subsidised in the 1980s and 90s. Both maize yields and fertilizer subsidies remain contentious political issues.

4. Poor maize harvests and associated economic problems in the early 2000s led to both main parties in the 2004 elections advocating fertilizer subsidies, though there was no subsidy in place in 2004/5. In 2005/6 the Malawian government implemented a fertilizer subsidy despite mixed donor views and no direct donor support (though general budget support was received). Coupons were used to allocate fertilizer by hectarage and the increase in fertilizer application, in conjunction with good rains, led to record maize production that year. The subsidy was therefore hailed as a success by the government.

5. The policy objectives of subsidies have changed in recent years, due in part to the failure of the green revolution in Africa. The conventional view of subsidies used to promote technical change and productivity, which in the past have resulted in issues of cost control, inequality in the distribution of benefits and the crowding out of the private sector, is being replaced with new objectives in which the emphasis is on subsidies which lower staple food prices and raise wages, provide social protection and develop the private sector.

6. A DFID/USAID funded evaluation of the 2006/7 Malawian agricultural input supply programme, a programme designed to provide subsidised  maize and tobacco fertilizer and maize seeds to farming households, took place in 2007. The evaluation considered the input subsidy implementation; the effects on rural households; and the effects on the input supply system, but did not consider the effects on the macro-economy or the programme in the context of other policies. It should be noted that the evaluation dealt with complex issues and involved multiple and diverse stakeholders as well as data difficulties.

7. The 2006/7 programme experienced constant change including alterations to donor participation and implementation procedures, which led to uncertainty and delay for all stakeholders. In particular, the original supply of coupons, used to allocate fertilizer and maize seeds, were unable to meet demand and supplementary (and unbudgeted for) coupons were distributed. This supplementary allocation was regionally uneven, and the distribution process was not transparent. This was felt to be due to a conflict regarding the objectives of the programme, as the Ministry of Agriculture directed additional subsidies to those areas of Malawi where they could be most effectively used, rather than to those areas which required the additional social protection the subsidies could provide. The systems of fertilizer and maize seed distribution also varied between districts, due in part to the inconsistent involvement of local leaders and a lack of clarity in the targeting criteria.

8. Sales of fertilizer and maize seeds were through two channels – parastatals and large private sector distributors, while small agro-dealers were not involved directly. Issues with sales included reports of ‘tips’ being paid to buy fertilizer and the exchange of coupons for other items. Access issues to the subsidised fertiliser and maize seeds included high transport costs for those households located far from the supplies, an issue which may be solvable by the inclusion of local agro-dealers in more remote areas, and the difficulties some households faced in obtaining the cash required to buy supplies, even with the subsidy in place.

9. The total cost of the programme to the government has been calculated at US$64million, US$13million over budget and a substantial portion of the Ministry of Agriculture’s total budget of approximately US$120million. The cost to donors has been calculated at US$9.5million. The overspend was due to the distribution of the supplementary coupons.

10. The DFID/USAID evaluation looked in particular at whether the benefits of the 2006/7 subsidy outweighed the costs, and how the benefit-cost ratio could be improved. The complexity involved in answering these questions was again emphasised.

11. Direct impacts of the programme included:

• Record national fertiliser sales, though displacement of private sales by subsidised sales led to less of an increase in total sales than desired;
• Record crop production, though this was due in part to good climatic conditions;
• A reduced number of people seeking casual labour, and higher wage rates; and
• Lower maize prices for consumers, but lower returns for surplus maize growers.

12. Further economic and social impacts of the programme proved complex to measure and varied according to household characteristics, access to and use of coupons, and the area in question. Any evaluation of these wider impacts is also dependent in part on the objectives against which they are judged, and of the wider policy and economic climate.

13. Suggested changes to the agricultural input supply programme to improve the benefit-cost ratio included:

• Agreeing objectives for the programme which all parties abide by;
• A more effective system for distributing subsidy coupons; and
• Efforts to encourage the private sector, in particular fertiliser distributors, to move into more remote rural areas.

Discussant's comments
Paul Harvey, Research Fellow, Humanitarian Policy Group, ODI

14. Paul agreed from experience that fertilizer was highly sought-after amongst poor Malawian farmers in 2000/1, but could not be afforded. He was pleased that the case study described subsidies as a potentially effective public policy with a positive impact on food security, and that while it was important to be careful about the results and the complexities of the case, it was equally important to share positive outcomes. He also felt that the case study showed democracy in action, with a government being elected on a policy platform, the policy being implemented, and the government receiving praise from the electorate for it.

Discussion

15. It was asked if Andrew felt the results of the evaluation implied other countries would benefit from similar programmes. Andrew agreed that the evaluation has provided lessons for other countries, but that the process should be considered in context before implementation and not simply copied.

16. Alternatives to fertilizer subsidies were discussed. A participant suggested that in many areas of Malawi, the increasing population and consequential reduction in the amount of land available per family has made it harder to feed the population without the use of fertilizers, and that subsidies are a better solution in the short- and medium-term than food aid. Andrew agreed that there has been little success shown by alternative schemes to date, and suggested that this was a question for the future, to be considered once production has been increased and maize prices have stabilised.

17. The role of donors in the process was questioned. Andrew explained that many donors are supportive of the programme and many observers have praised the way in which the government has led the process and donors have helped where appropriate. There are also still some questions regarding the objectives and the benefits of the programme which are currently being discussed by donors and the government.

18. Additional issues that arose included:

• The extent to which maize prices are protected from externalities;
• The ease of phasing out the programme; and
• The likely success or otherwise of the programme in years with bad climatic conditions.