By: Felix Lung
This research paper considers the case of the recently founded “African Risk Capacity” (ARC), a new Specialized Agency of the African Union that will set up a risk-pooling insurance fund against drought in sub-Saharan Africa. The paper first describes how the envisaged mechanism will function and how it has evolved. Subsequently, it considers the role of ARC in the wider humanitarian response framework for food security in times of drought. Looking at its role in relation to changes in the drought resilience level of African farmers, it shows that ARC can contribute significantly to the overall protection level against drought and possibly revolutionize the way the humanitarian response system functions.
As recent disasters across the Horn of Africa and in the Sahel have shown, drought is a recurring and increasing threat to the population of rural farmers in Africa. At the same time, traditional humanitarian intervention mechanisms to drought are slow and operate on an ad hoc basis. The African Union’s newly founded Specialized Agency, the African Risk Capacity (ARC), may offer a solution to the current system’s deficiencies. This paper argues that the ARC could revolutionize the way drought interventions are conducted in Africa by shortening response time and making contingent funds available more quickly. It first considers the current humanitarian response system to droughts in Africa and how the ARC will function, and then evaluates the greater role of the ARC in the international humanitarian response system.
Drought in sub-Saharan Africa and the Humanitarian Response
Over the past decade, sub-Saharan African countries have faced some of the most catastrophic droughts in history. In 2002, 2009, and 2011 the lack of rain had particularly devastating implications for the Horn of Africa region. With more than 12 million people affected, the Horn of Africa drought of 2011 was the worst in 60 years in some areas, and has had the most severe effects on the region since the 1995 drought.1 The UK’s Department for International Development estimates that between 50,000 to 100,000 people died as a consequence of this crisis.2 Similarly, the recent crisis in the Sahel, where UNICEF estimates 15 million people to be at risk of food insecurity, illustrates the scale of disaster that can be brought on by drought.3 Unfortunately, the situation is expected to get worse. In the light of climate change, droughts have intensified and become more frequent, putting additional stress on future yields of African farmers and their respective livelihoods.4
Traditionally, the international humanitarian community responds to drought disasters on an ad hoc basis. In other words, in case of a drought, the World Food Programme commits a limited amount of immediate contingency funds to finance a first humanitarian response. Subsequently, it conducts a damage assessment in the respective country, launches a funding appeals process, and finally intervenes fully once funds have been obtained. The delay in response caused by these processes is significant; it is estimated that on average seven to eight months pass between the point in time that the rainfall ceases and the height of intervention.5
The length of the period between failing rains and intervention is problematic, because the severity of the consequences felt at the household level increases during this lag. An evaluation of households’ coping strategies with drought shows that in the months after rains fail, farmers will usually look for non-farm work, eat less nutritious food, resort to selling non-productive assets, borrow money, and reduce durable purchases. From one month after the harvest onwards, farmers may reduce input investment, reduce their food intake, and start selling productive assets. From nine months after the harvest onwards, farmers face increased mortality rates.6 Thus, it is of crucial importance to reduce the time it takes for intervention to reach those in need. The following section shows that the ARC can be an effective tool for this.
The African Risk Capacity
The ARC Mechanism
The African Risk Capacity is an innovative project of the African Union designed to improve the current response to drought- related food security emergencies in Africa and to build capacity within AU member states to manage these risks. The ARC is currently in the process of being established as a Specialized Agency of the African Union. The first Meeting of the Conference of Parties took place from the 23rd to the 27th of February, 2013 in Dakar, Senegal, and the first meeting of the ARC Governing Board is scheduled for April 2013. The ARC is based on two key elements: African Risk View (ARV) software and an index-based contingency funding pool.
ARV calculates the number of people who will need food assistance in a given country in Africa during times of drought based on rainfall data. Comparing actual rainfall data with a benchmark value derived from past years, ARV predicts at the time of harvest how many people will need food assistance and how expensive an intervention would be. Thus, it determines beneficiary numbers and intervention costs before the spending actually needs to occur. Until now, these figures were determined by the World Food Programme via lengthy damage assessments, conducted only after farmers had begun experiencing the effects of drought. Replacing the damage assessment with a calculation that estimates the extent of drought effects before they are felt allows for an earlier humanitarian response. This, in turn, prevents the loss of longer-term development gains and yields greater overall benefits compared to how the traditional approach has fared.
Building on this mechanism, a contingency funding pool will be set up, financed through annual premium payments by African member countries. Once the ARV predicts that a certain number of people in a given member country will require food assistance as a result of drought, the required amount of assistance will be automatically released from the pool to the affected country. The received amount will be used to assist the affected households, employing national social safety systems such as cash or in-kind transfers or school feeding, to the greatest extent possible. The advantage of creating a pan-African fund rather than many national ones can be explained through the idea of risk pooling; spreading risk among multiple parties yields a financial advantage. The combined benefit of an early response and risk pooling is currently estimated to be 3 USD for every 1 USD spent, when compared to traditional humanitarian response mechanisms.7
The benefits of the ARC mechanism are not purely financial. By establishing the ARC as a Specialized Agency of the AU, there is ownership created among African states to manage humanitarian drought interventions on the continent. This is strengthened by the fact that the ARC employs national distribution channels for ARC-financed interventions and requires member states to revise their drought contingency plans for increased efficiency. Additionally, while the fund will be capitalized through donor funds, at first, in the long term it will be financed exclusively by African national governments. This creates the possibility of freeing up a substantial portion of the WFP’s budget, since approximately one-third of total WFP expenditures are currently spent on drought interventions in Africa.8
Political evolvement of the ARC
Launched by the Rockefeller Foundation, the WFP’s “Climate Disaster Risk Solutions” (CDRS) unit began working on the design of the Africa Risk View software in 2008. Initially, the work was only intended to expedite the WFP’s humanitarian responses by avoiding lengthy damage assessments in case of drought. However, upon completion of the software in November 2009, the project team both realized the great savings potential of ARV in combination with a pan-African contingency fund for African countries and encountered great interest from the African Union (AU) Commissioner of Rural Agriculture and Economy. Thus, ARC was born. A three- year consultative process between the ARC, WFP, AU Commission (AUC), AU Executive Council, and member states began, culminating in the signing of the Establishment Treaty of the ARC as a Specialized Agency of the AU by eighteen member states on November 23, 2012.
In order for the ARC to be implemented, the support of three key groups was necessary: selected organs of the African Union (the AUC and the Executive Council), AU member states, and donors. There were a number of significant challenges that needed to be overcome in the process of the three year long consultations. They can be grouped as the political environment within the AU and political interests within member states.
Both institutional and cultural qualities of the AU presented themselves as obstacles to a quick adoption of the ARC. As the AU is a relatively new organization, founded in 2002 from the previous Organisation of African Unity, many institutional processes remain to be defined. For example, while member states agree to incorporate Specialized Agencies into the AU system following the UN example, no clear procedures have been established regarding how to do this. This has slowed project progress, as the legal work that was necessary was more cumbersome than expected. There was no existing precedent.
In terms of political culture, the AU is a strongly consensus-driven organization. This is reflected in the fact that decisions by the AU Assembly (made up of the heads of member states) and the Executive Council can only be made with two-thirds majorities.9 However, like in the UN General Assembly, member states tend to try to acheive consensus. This creates a major complication when trying to reach an outcome for two reasons. The general level of trust among member states is relatively low; unlike the European Union, African states do not share a common history of cooperation, but in fact have only been in existence since decolonisation. Secondly, immense cultural and geographic differences play a role as well as dramatic economic differences (five countries cover almost the entire AU budget: South Africa, Nigeria, Egypt, Algeria, and Tunisia). The AU lacks a pre-meeting consultative culture. Other than at the UN, where resolutions are generally negotiated and agreed upon between member state blocks before the meeting where they are to be adopted, all AU negotiations tend to occur at the meeting itself. Only SADC and, less frequently ECOWAS, adopt a country block position. In combination, a consensus-driven environment in which actors have limited trust in each other and discuss relatively little is bound to delay any negotiation process.
Political interests of states play an important role. Support from the member states was key to the ARC implementation process, as consent of the heads of states was required to establish the envisaged AU Specialized Agency. Thus, the ARC team embarked on scoping missions to each of the interested countries and engaged in individual membership negotiations.
The significance of the internal political situation is illustrated in Senegal, where the ARC’s offers to negotiate were declined by the government under President Abdoulaye Wade, but actively pursued after the new president Macky Sall was inaugurated into office in April 2012. In other cases, internal political agendas acted in favor of the ARC membership process. In Kenya, for example, by the time the government was approached by the ARC team, it had already considered contingency funding mechanisms against drought and was in the process of designing the “National Drought Management Authority”, which was eventually established on November 24, 2011.10 Thus, accession talks progressed smoothly. Similarly, in Malawi, the government had already agreed to an index-based weather insurance scheme with the WB and wished to extend its coverage.11 In other countries such as Lesotho, Niger, and Burkina Faso, internal political interests favored the empowerment of the ministry that was to be in charge of drought disaster management.
Finally, concern about the administrative design of the ARC project was often a determining factor in negotiations. Several countries, such as Uganda, questioned whether the ARC secretariat was “African” enough, as two-thirds of the team were not of African descent. In addition, other countries expressed concern about the AU bearing responsibility over a project of immense financial extent and a high level of riskiness. These concerns ultimately played minor roles.
Interpretation and Next Steps
While a wide range of political obstacles needed to be overcome before the ARC could be established, African countries do not seem to have had difficulty in agreeing on the ARC as an Agency. This is somewhat surprising, as the ARC mechanism brings with it significant obligations for the signatory parties: members are required to pay annual premium fees, revise their contingency plans for the effectiveness of distributing payout, and take leadership in the lengthy process of adjusting the ARV software to their domestic environment. Yet these non-negligible commitments have not played as great of a role in the negotiations as one might expect. This might be indicative of the great need for a mechanism of this kind, or reflect the desire to either achieve greater independence from foreign humanitarian aid flows or quicker access to intervention financing in times of crisis. Considering the recent success story of African economic performance, one might interpret this as one indicator of a claim to a new, greater African sovereignty.
The African Risk Capacity in the Greater Humanitarian Response Framework
Not all farming households in sub-Saharan Africa are in need of protection against drought via the ARC mechanism. Some possess enough reserve resources or may have acquired particular coping skills in order to outlast a failing harvest without being affected. Others may have to downscale their eating habits temporarily but do not suffer any further consequences such as having to sell their productive assets. The household’s ability to cope with the effects of drought is described as “resilience”. The household protection achieved by the ARC complements the protection that households have through their existing level of resilience. Thus, in order to be able to determine what role the ARC should play in the overarching humanitarian response framework against drought, an understanding of current resilience levels in sub-Saharan African countries and their expected development is required. This section makes an attempt at providing this analysis.
Household Income as a Resilience Indicator
Droughts occur with differing severity. In order to assess resilience, the first step is to introduce a magnitude to measure the severity of drought. To account for individual country specifics, no absolute measurement is used. Instead, based on historical data, the average least rainfall in five, ten, and fifteen years was determined for each country. In the following section, these droughts will be referred to as 1:5, 1:10, and 1:15 events. Accordingly, a household would be “resilient” against a “1:5 event”if it were able to withstand the effects of a drought whose total rainfall corresponds to the average least amount of rainfall in five years in his country.
The second step is to determine the meaning of “withstand”. As described in the second section, households employ different strategies to cope with drought. The central question is: how “well off” does a household need to be after a drought to be categorized as “resilient”? It is extremely difficult to establish an objective benchmark for this. For this reason, the model does not consider individual households’ living conditions. Instead, using the Food and Agricultural Organization’s Water Requirements Satisfaction Index (WRSI), it assumes that farming households are sufficiently well off when there is an average amount of rainfall.12 Consequently, in order to be categorized as “resilient” against a 1-in- 5, 1-in-10, or 1-in-15 event, a household needs to have an income that exceeds its income obtained at the benchmark rainfall level, adjusted for inflation, by the amount that it loses in such an event. For example, if a household has a monthly income of 80 USD for an average amount of rainfall and faces a 20% income reduction during a 1-in-5 event, it needs to have a monthly income of 100 USD in order to be “resilient” against 1-in-5 events.
The third step is to calculate the required income growth rates for affected households to become resilient in sub-Saharan countries. As before, this is done based on historical data of the FAO’s WRSI index. Consistent with ARC estimates, we assume a “scaling factor” of 1.5. In other words, a 1% downward deviation of rainfall from the annual benchmark level results in a 1.5% livelihood loss. All factors taken together, we can calculate the individual’s loss of livelihood for the respective drought event, which enables estimation of the required income growth rate to become resilient at the farming household level.
For the ARC target countries with available data, the model yields the results found in Table 1.
The results show that the goal of achieving a certain level of resilience for the rural population at risk in sub- Saharan African countries is an ambitious but not impossible task. For example, in order to achieve 1:5 resilience within ten years, income of at-risk farmers would have to grow on average by 2.85% at the household level. Yet it is important to note that there is some significant variability, with Kenya and Mauritania as the upper outliers (8.2% and 7.1% respectively), and Zambia and Mali as the lower ones (each 0.4%).
Achievability of required growth rates
At first glance, agricultural growth in sub-Saharan Africa has performed on par with that of other developing countries over the last twenty years. Since 1990, sub-Saharan agricultural GDP grew on average by 2.9% annually (3.1% in ARC countries), compared to 3.3% in all low-income countries worldwide. Sub- Saharan agricultural growth has recently more momentum recently, with average annual agricultural GDP growth of 4% since 2005 (3.9% in ARC countries).13 Before the global financial crisis, from 2005 to 2009, eight sub-Saharan African countries displayed agricultural GDP growth of more than 6% annually (of which two are ARC countries).
However, there are three significant caveats to this analysis. Due to strong population growth, the gains for sub- Saharan African farmers have been much lower than the numbers would suggest. Sub-Saharan Agricultural GDP per worker has only grown by 0.2% annually since 1990 (1.1% in ARC countries) and 1% since 2005 (1.7% in ARC countries).
Secondly, Sub-Saharan African agriculture is still at a much lower productivity level than in other parts of the developing world. The average cereal yield per hectare in sub-Saharan Africa in 2010 was 1300 kg (1400 kg in ARC countries), compared to the worldwide average of 2100 kg in low income countries, 4600 kg in the East Asia and Pacific region, and 3900 kg in Latin America and the Caribbean.14 Therefore, for agricultural productivity to catch up with levels in other parts of the world, growth in sub-Saharan Africa would have had to be much stronger.
Finally, the variability within sub- Saharan Africa and within the ARC country portfolio is high; for example, while Ethiopia and Mozambique achieved an average agricultural GDP growth of 9.6% and 9.5% per year respectively since 2005, agricultural GDP remained almost unchanged in Zambia and Lesotho and even contracted by 6.7% in Zimbabwe.
Determinants of sub-Saharan African Agricultural Growth
The factors impeding the growth of agricultural GDP in sub-Saharan Africa can be roughly grouped into three categories: geographic and demographic aspects, the policy environment, and the use of technology.
Sub-Saharan Africa faces geographic- demographic challenges in achieving agricultural growth. Two-thirds of the sub-Saharan African population lives in “less-favored areas” for agriculture, where either the agricultural potential or market access is low. In addition, while population density is lower overall than in Asia, and it is estimated that much of the cultivable land remains unused. In 2003, only 180 million hectare of the 420 million hectares of land with high cultivation potential in Africa were used for agricultural purposes. There is a high degree of variability across the continent.15 For example, the land- quality-adjusted population density is higher in Kenya than in Bangladesh. This is a major obstacle to agricultural GDP growth, especially since population growth is high.16
While the overall policy environment has improved significantly over the last few decades, there continues to be a lack of national attention focused on agriculture. During the 1980s and 1990s, African agricultural growth recovered from the agricultural downturn of the 1970s. Major changes in macroeconomic policies were instrumental for this to occur. Key changes included trade liberalisation, often through disempowering national marketing boards, currency devaluation, and lowering the tax burden on producers.17 A 1995 study concluded that almost two-thirds of the improved productivity could be traced back to improvements in macroeconomic policies.18 Today, the worst of the anti- agriculture biases have been removed, but the private sector still needs to evolve as an effective player in crop supply chains.19
Yet in comparison to other regions of the world, attention to agriculture in Africa remains low. In 2003, the Maputo Declaration committed African governments to the Comprehensive Africa Agriculture Development Program (CAAPD), whereby they agreed to raise agricultural expenditure to a minimum of 10% of their national budgets, and to raise annual agricultural GDP growth to 6% by 2008.20 While absolute amounts of spending have increased in many parts of sub-Saharan Africa, relative amounts have declined in most countries since the 1990s. Only six countries had achieved the 10 percent target by 2010: Burkina Faso, Ethiopia, Guinea, Mali, Niger, and Senegal.21 This is particularly low in comparison with the number of Asian countries that spent 15% or more on agriculture during the Asian Green Revolution years.22 Only eight sub-Saharan African countries reached the target of agricultural growth of 6% between 2005 and 2009. Additionally, commitment to funding agricultural research and development is low; despite already low initial levels, national expenditure on R&D fell by nearly half in the 27 sub-Saharan countries with data, and, as a share of GDP, dropped for the whole region in the 1990s.23
Meanwhile, the environment among the international donor community to support agricultural development in Africa has become warmer. Starting with the 2003 CAAPD, initiated by the New Economic Partnership for African Development (NEPAD), several new donor initiatives have been created. These include, for example, the G-8’s “New Alliance for Food Security and Nutrition” (2009), the 2006 “Alliance for a Green Revolution in Africa” (AGRA), and new support for the “Partnership to Cut Hunger and Poverty in Africa” (2001). The newly available funds could offer an opportunity to boost African national budget expenditures for agriculture, after donor support declined sharply after the 1980s. However, their effect in this regard
has been small so far.24
The use of technology in agriculture is extremely low in sub-Saharan Africa. The World Development Report 2008 on agriculture illustrates the lack of sub- Saharan African technology use in three different dimensions25:
- Lack of irrigation: only 4% of the total cultivated area is irrigated compared to 29% in the East Asia & Pacific (EAP) region and 11% in the Latin America and Caribbean (LAC) region.
- Low use of fertilizer: while the use of chemical fertilizer has expanded greatly in the developing world (190 kg nutrients used per hectare of cultivated land EAP and 81 kg in LAC), this is not the case for sub- Saharan Africa (13 kg).
- Low use of improved seeds: only 24% of the cereal area in sub-Saharan Africa is covered by improved seeds, compared to 85% in EAP and 59% in LAC.
There has been a great deal of discussion on why Africa has not had a “Green Revolution” like Asia and Latin America have. In addition to the policy aspects outlined above (low budget commitment, low R&D spending, limited donor commitment), different reasons have impeded its realization:
- While 95% of the food grown in sub-Saharan Africa is rainfed, 45% of the population lives in regions that experience limited rainfall, limiting the impact that improved crops can have.26,27 While this would encourage the construction of irrigation systems, the associated investment costs in doing so are high and sub-Saharan African countries have not been able to tackle this problem sufficiently in the past.28
- There is a certain degree of diversity among cultivated crops. As crops of sub-Saharan Africa differ from those mostly used in Latin America and Asia, and display some variety (they are mainly maize, millet, sorghum), African countries have been unable to benefit significantly from the majority of improved seeds that were designed for wheat and rice in the Asian and Latin American contexts.29
- Infrastructure is insufficient. The lack of roads and integrated markets makes the acquisition of fertilizer, for example, troublesome and expensive. On average, fertilizer is twice as expensive in sub-Saharan African countries than in Asian countries, due to increased transport and marketing costs.30 The World Development Report of 2008 estimates that 34% of the sub-Saharan population lives in regions with limited market access.
Achievability of Required Growth in the Medium Term
Considering that average growth of agricultural GDP per worker in the ARC target countries since 2005 amounts to only 1.7%, targeting to achieve 1:5 resilience in either 10 or 15 years (requiring annual growth of 2.85% and 1.92% respectively) appear to be the two most reasonable among the available options. However, as Table 2 reveals, these averages incorporate a high level of variability within the country sample. Thus, while countries like Ethiopia, Malawi, and Mozambique may outperform what is required of them at the household level and may even achieve resilience levels of 1:10 or 1:15 within 10 years, Kenya, Mauritania, Uganda, Zambia, and Zimbabwe may not even achieve the required growth levels to achieve 1:5 resilience in 15 years. However, one also needs to consider the respective population sizes: If 1:5 resilience was achieved for only Ethiopia, Lesotho, Malawi, Mozambique, Senegal, Swaziland, and Tanzania, this would mean an achievement of resilience for 77% of the target countries’ populations.
Whether or not African agriculture can sustain the required growth level will depend on the ability of African countries to best use the current momentum in agriculture policy reform and address issues that impede growth. While the literature has presented many policy recommendations, most come down to two fundamental objectives: raising productivity and expanding market access for farmers.31 If African states achieve making progress towards these goals, achieving resilience for many parts of the farming population is feasible.
ARC and Protection from Drought
The question of how big a role the ARC can play in the protection of farmers in Africa remains. Initially, a maximum payout of 30 million USD per country per drought emergency is planned. In line with WFP estimates, assuming a humanitarian intervention cost of 100 USD per person, this would insure a maximum of 300,000 people per country. If we further assume that these 30 million USD would be paid out regardless of the severity of the drought, Table 3 shows its protection potential across the ARC target countries.
Although the value of these estimations is limited—after all, the concept of the ARC capitalizes on the fact that drought does not occur in all target countries at the same time, and within the countries only a fraction of the total population will be affected—Table 3 shows that the protective potential of the ARC in its current form is limited. For countries with a small at-risk population, such as Lesotho, Mauritania, and Swaziland, the ARC could protect all people at risk in case of a minor drought. However, in Ethiopia and Tanzania—the countries that have the largest exposed populations—the ARC in its current design only covers between 1.5% and 4% of them. This means that for these populations to be protected, either the ARC has to expand in the future, they have to rely on other protection mechanisms such as micro insurance, or they must continue to rely on ad hoc humanitarian help.
The African Risk Capacity has come a long way. While just an idea in the beginning of 2010, it is expected to be operational as a Specialized Agency of the African Union by mid-2013. It offers innovative ways to finance and deliver humanitarian drought responses in sub-Saharan Africa, and will therefore undoubtedly greatly impact the way that drought assistance is provided in participating countries. Yet the magnitude of risk coverage of the ARC and its overall role in the international humanitarian response framework remains to be determined. In this regard, this paper shows that: i) the extent of coverage depends on the level of drought resilience of farmers’ households; ii) if African agricultural economies manage to stabilize growth at the household level over the next 10 years, most households will be able to become resilient against 1:5 events and, in some countries, 1:10 events; and iii) in order to offer full coverage to African populations at risk, the ARC will have to expand.
Felix Lung is a second-year M.A. student at the Johns Hopkins University School of Advanced International Studies. He concentrates in International Development and spent his first year at the SAIS Bologna Center. Before SAIS, he helped build up and ultimately co-headed PROJECT-E, a German-Austrian NGO focusing on women’s education in Ethiopia.
1. DfID, “Horn of Africa Food Crisis: Latest updates,” 2011, http://www. dfid.gov.uk/news/latest-news/2011/british-aid-battles-starvation-in-africa/#Background.
2. Simon Tisdall, “East Africa’s Drought: The Avoidable Disaster,” The Guardian, January 17, 2012, http://www.guardian. co.uk/world/2012/jan/18/east-africa- drought-disaster-report.
3. UNICEF, “Disaster Stalking Children in Africa’s drought-prone Sahel region, warns UNICEF,” March 16, 2012, http://www.un.org/apps/news/story. asp?NewsID=41555#.UMIQ7OS88aw.
4. World Food Programme, “Hunger and Climate Change,” June 2010.
5. Daniel Clarke and Ruth Vargas Hill, “Cost-Benefit Analysis of the African Risk Capacity Facility,” Oxford University and IFPRI ( June 2012): 46.
6. Ibid, 36.
8. African Risk Capacity, “African Risk Capacity Project Memorandum,” December 2010.
9. AU Constitutive Act, Art. 7, 11, http://www.africa union.org/root/au/aboutau/constitutive_act_en.htm.
10. Government of Kenya, Legal Notice 171, “The National Drought Authority Order,” 2011, http:// www.disasterriskreduction.net/east- central-africa/library/detail/en/?dyna_ fef%5Buid%5D=1905.
11. Joanna Syroka and Antonio Nucifora, “National Drought Insurance for Malawi,” World Bank Research Working Paper No. 5169, January 2010.
12. The WRSI index measures the actual rainfall against a benchmark of what is needed for a certain crop to grow. It is thus an indicator of how well a crop performs given the available water supply.
13. World Bank, World Development Indicators, http://databank.worldbank.org.
15. Xinshen Diao, Derek Headey, and Michael Johnson, “Toward a green revolution in Africa: what would it achieve, and what would it require?,” Agricultural Economics 39 (2008): 539-550.
16. World Bank, “World Development Report 2008: Agriculture for Development” (2007): 55-56.
17. Alexjandro Nin Pratt and Bingxin Yu, “An Updated Look at the Recovery of Agricultural Productivity in Sub- Saharan Africa,” IFPRI Discussion Paper 00787 (August 2008): 29.
18. Steven Block, “The recovery of agricultural productivity in Sub-Saharan Africa,” Food Policy 20 (1995): 385-405.
19. Peter Hazell, “The Asian Green Revolution,” IFPRI Discussion Paper 00911 (November 2009): 23.
20. AU Assembly, “Maputo Declaration on Agriculture and Food Security in Africa,” July 2003.
21. Samuel Benin et al. “Monitoring African agricultural development processes and performance: A comparative analy-sis,” ReSAKSS-Africa Wide Annual Trends and Outlook Report 2010 (February 2011): 49.
22. Hazell, 23.
23. World Bank 2007, 166.
24. Hazell, 23-24.
25. World Bank 2007, 50-65.
26. InterAcademy Council, “Realising the promise and potential of African Agriculture,” (2004): 48.
27. World Bank 2007, 56-57.
28. Liang Zhi You, “Africa Infrastructure Country Diagnostic: Irrigation Investment Needs in Sub-Saharan Africa,” IFPRI and World Bank ( June 2008).
29. World Bank 2007, 55.
30. Derek Byerlee et al. “Fertilizer Use in African Agriculture: Lessons Learned and Good Guidance Practices,” 2007.
31. Diao, Headey, and Johnson, 546.