Global Donor Platform for Rural Development

Platform Policy Brief

No. 5 // October 2011

T Aid to agriculture, rural development and food security

Platform Policy Briefs are designed to inform and guide members in the delivery of assistance in agriculture and rural development. Global Donor Platform for Rural Development Tackling rural poverty, together

Unpacking aid flows for enhanced transparency, accountability and aid effectiveness
SUMMARY

onors have long recognised the importance of agriculture for the economic development of low-income countries, for reducing rural poverty and for long term conservation of natural resources. Yet, for the last couple of decades, agricultural and food aid suffered a steady decline, both in absolute and relative terms, according to OECD statistics. Largely due to the 2007-2008 food price crisis, this trend is now being reversed. In 2009, G8 donors and international organisations pledged additional resources to support agriculture and food security, renewing emphasis on the sectors importance to reducing world poverty and hunger, in what became known as the LAquila Food Security Initiative. The scale and nature of the earlier decline needs to be better understood, however, as approaches to agricultural development have changed considerably since the days it received record-high aid flows in the early-1980s. At the same time, the forms of delivering development assistance have also changed, with programme-based approaches now increasingly used at the country level. This policy brief reports the findings of a study on the measurement of aid to agriculture and rural development (ARD) and food security. The study concludes that important policy changes are not easily traceable by means of standard statistical measures. One example concerns recent shifts towards a more explicit consideration of the factors that have limited supply responses to agricultural incentive policies, market and enterprise development, financial market deepening and trade facilitation. Failing to account for policy shifts in an accurate manner may, however, compromise the effectiveness of aid allocation and country-level planning, and undermine efforts to strengthen global aid transparency and accountability. The study also noted that institutional structures and political and administrative incentives, at the level of both donor and recipient governments, present challenges to building a common framework for measuring aid allocation and use in a way that is consistent for international comparison and useful in terms of establishing indicators for a results approach to aid management. These are not insurmountable obstacles. The global call for transparency, accountability and results may expose the challenges but it also encourages the search for practical solutions to address them. Three policy recommendations stand out from the study: Greater cross-donor coherence in defining the scope of the sector, so that cross donor trends can be meaningfully compared and global commitments can be accurately tracked and accounted for Unpacking the objectives of aid programmes targeting agriculture, rural development and food security, so as to generate more manageable results frameworks Breaking sectoral boundaries in aid measurement for improved consistency between policy and resource allocation.

Author: Ldia Cabral of Overseas Development Institute, London

Platform Policy Brief I No. 5

T ANALYSIS
The study summarised in this policy brief analyses development assistance to ARD and food security by focusing on the quality of aid measurement. It investigates the extent to which aid data provides an accurate indication of policy priorities and the changing policy context for ARD and aid more generally.

The response of the agricultural sector to such policy-related assistance has been mixed, leading to a more explicit consideration of the factors that have limited a strong supply response to policy incentives. Hence, in the 2000s more aid is being directed at the business environment, with private enterprise promotion, value-chain development, financial market deepening and trade facilitation strongly favoured by some donors. Since the late 1990s in particular, expenditures on relief and social protection have risen, reflecting growing concerns over fragility and vulnerability of many rural populations.

// The changing policy context

The ARD and food security policy domain has changed considerably over the last decades.
7] Partly compiled from Investing in Women as Drivers of Agricultural Growth (J. Ashby, M. Hartl, Y. Lambrou, G. Larson, A. Lubbock, E. Pehu, C. Ragasa) and IFAD (2010) Special Session Report The Farmers Forum in Conjunction with the ThirtyThird Session of IFADs Governing Council. Special Session: Promoting womens leadership in farmers and rural producers organizations. Held at IFAD Headquarters, Rome, on 13 February 2010. 8] These are the Gender in Agriculture Sourcebook http://worldbank.org/genderinag, The World Development Report: Agriculture for DevelopmentAgriculture for Development. The 2008 World Development Report. The World Bank. Washington, DC. 2007. and the International Assessment of Agricultural Science and Technology for Development [www.agassessment.org ]. 9] Farnworth, C.R. (2010) Gender aware approaches in agricultural programmes: a study of Sida-supported agricultural programmes. Sida Evaluation 2010: 3 10] Lastarria-Cornhiel, S. (2008) The Feminization of agriculture: trends and driving forces. Background paper for the World Development Report. http://siteresources.worldbank.org/I NTWDR2008/Resources/27950871191427986785/LastarriaCornhiel_F eminizationOfAgri.pdf 11] IAASTD (2008) Executive Summary of the Synthesis Report. 12] G. Rebosio, S. Gammage, and C. Manfre, A Pro-Poor Analysis of the Artichoke Value Chain in Peru,www.microlinks.org/file_down load.php/Artichoke_Peru_Research_ Brief.pdf?URL_ID=18386&filename=1 1861594421Artichoke_Peru_Researc h_Brief.pdf&filetype=application%2F pdf&filesize=299504&name=Artichok e_Peru_ Research_Brief.pdf&location=user-S. 13] Barrientos, S., C. Dolan, A. Tallontire. "A Gendered Value Chain Approach to Codes of Conduct in AFrican Horticlture." World Development 31 (9) (2003): 1511-1526. 14] Action Aid Securing womens right to land and livelihoods: a key to ending hunger and fighting AIDS http://www.actionaid.org/micrositeAs sets/eu/assets/women's%20right%2 0to%20land%20hiv%20and%20hunge donorplatform.org r%20jun08final.pdf

Up to the mid-1970s, agricultural aid was principally focussed on increasing production and productivity through input subsidisation and irrigation. Concerns about addressing producers outside the higher potential areas, led to the emergence of integrated rural development projects in the 1980s. The shift also stimulated additional expenditure on the underlying impediments to increased agricultural incomes. The disappointing outcomes of agriculture and rural development projects, and the perilous state of many economies, led to a policy shift away from projects towards programmes and balance of payments support, conditional upon economic policy reforms in the 1990s. For agricultural development, the often explicit policy objective was to increase profitability through a more favourable exchange rate for agricultural producers, especially in export markets. A few years later, into the late-1990s and early 2000s, the policy shift also incorporated new forms of budgetary aid and sector-specific programme aid.

// Aid flows to ARD and food security

The studys analysis of aid flows drew on data from the Creditor Reporting System (CRS) of the OECDs Development Assistance Committee (DAC). Agricultural official development assistance (ODA) is conventionally measured as the sum of sector allocable aid to agriculture, forestry and fishing, here referred to as AFF. A broader measure, AFF+, adds to the calculation aid flows coded by the DAC as: rural development, development food aid and emergency food aid. According to the DAC, while these three additional socalled purpose codes do not support agricultural development directly, they contribute to improving rural livelihoods and food security. The study considered the period 1975-2009 and revealed, inter alia, the trends and patterns documented below. Overall, the downward trend in aid to ARD and food security is now being reversed and a significant share of funds are being channelled to agricultural policy and administration and emergency food aid. Sub-Saharan Africa is where most of the recovery is taking place.

Aid to agriculture, forestry and fishing and proportion of total ODA, 1975-2009
12,000 10,000 8,000 USD-Million 6,000
X X X X X X X X X X X X X X X X X X X X X X

25% 20% 20% 15% 10% 5% 0 Percentage ODA

4,000
X

X X X X X X

X X

2,000 0

Data source: OECD/DAC Creditor Reporting System

1975 1076 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

share of ODA

DAC AFF constant prices (2008)

DAC AFF current prices

Platform Policy Brief I No. 5

TSelected trends and patterns in aid to ARD and food security, 1975-2009
Steady decline in aid to agriculture, forestry and fishing during the last two decades of the twentieth century, both in absolute nominal and real terms and as a share of total ODA Reversal in the downward trend in the 21st century, particularly since 2005/06 and especially in Sub-Saharan Africa Earlier reversal if the broader measure (AFF+) is considered, with emergency food aid accounting for much of the growth and a reaching record value in 2008 Sharp drop in aid to development food security programmes from the record levels of the 1970s and early 1980s, recent surge but have been surpassed by emergency food aid Irregular pattern in aid to rural development, albeit with a descending trend line Significant increase in support to agricultural policy and administration increasing, particularly in Sub-Saharan Africa, which becomes the largest component of agricultural aid Steady decline in aid to food crop production throughout the 1980s, but sharp increase in recent years, particularly in Sub-Saharan Africa Declining support to agricultural inputs, agricultural financial services, marketing storage and transportation services Sub-Saharan Africa and South and Central Asia accounting for much of aid flows to the sector, and where most of the decline in the 1980s and 1990s happened Disbursements rate lower for either AFF or AFF+ than for overall ODA, suggesting the challenges of fulfilling commitments in this sector

// Aid flows versus policy trends

While global trends and patterns may seem broadly consistent with widespread perceptions on aid to the sector, important elements of the history of assistance to ARD and food security that are not revealed by the data, suggesting a gap between aid policy trends and financial flows. An example of this is the emphasis on the factors that had limited a supply response to policy incentives through support directed towards improving the business environment, developing value chains, deepening financial markets and facilitating trade. A case-study on the International Fund for Agricultural Development (IFAD) published as a working paper provides an illustration of the failure of global aid statistics to capture some important trends in the financing of agriculture. Conventional ARD and food security measurement excludes an important part of IFADs assistanceThis is because a large part of IFADs projects is coded using categories of aid (or purpose codes, in DAC terminology) which are not part of the DACs calculation of AFF and AFF+. They include, among others: financial services, and small and medium-sized enterprise development. The importance of support to financial services is particularly striking: it represents 24 per cent of IFADs assistance over the period 2003-09, the bulk of which corresponds to projects supporting rural informal and semi-formal financial intermediaries. Yet, since such proj-

ects do not restrict credit to farmers, the agricultural financial services DAC purpose code is not used to classify the corresponding aid flows. Instead the bulk of IFADs assistance to rural finance is coded as banking and financial services, a category lying outside AFF and AFF+ calculations. Support to the development of rural value-chains, coded as business support services or as enterprise development is similarly left out of the calculations.

T IFADs assistance to ARD and food

security: comparing IFAD and DAC measures
IFADs assistance to ARD and food security, i.e. the total volume of IFADs portfolio, has remained broadly stable (in constant 2009 prices) over the past 15 years. On average, the Fund has provided a total of $492 million in grants and loans to developing countries over the period. Taking DACs AFF definition, however, an abrupt decline is noticeable in IFADs assistance between 1997 and 2002. This decline continues into 2006 if the AFF+ definition is considered. These measures are misleading in the IFAD case. For example, AFF+ fails to account for 58 per cent of IFADs assistance in 2005 and 32 per cent in 2009.

donorplatform.org

Platform Policy Brief I No. 5

IFADs assistance to ARD and food security: comparing IFAD and DAC measures
700 600 500 400 USD-Million 300 200 100 0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

IFAD AFF

IFAD AFF +

IFAD AFF Total

Data source: OECD/DAC Creditor Reporting System

// Insights from donor head offices and recipient countries

Donors have different ways of accounting for ARD and food aid, largely due to differences in institutional structures and in the constituencies to which they report. Case-studies on donor practices revealed the sheer variety of institutional arrangements for aid management. While the United States has at least three channels for ARD and food security support, the United Kingdom, has a single agency responsible for ARD and food security spending. Despite the structural differences, all the agencies reviewed have adopted a similar approach to tracking expenditure against what are seen as principal policy areas or themes. For example, Germany has developed a marker system; agricultural expenditure comes under the Rural Development marker, which is also the measure used for tracking on LAquila commitments. Such a thematic approach to aid measurement holds much more importance to individual donors than their reporting obligations to the CRS. It is seen as making aid intelligible and hence defensible, to the wider constituency. Yet, for ARD and food security aid, the main consequence of this thematic approach is that what may seem to some individual officials a coherent policy domain may not be considered by others as a theme in its own right. Therefore, ARD and food security becomes a fairly widely dispersed set of activities that is not reported upon in a consolidated way, bringing into question the value of current coding for international standards such as the CRS. One of the main reservations is that the CRS mixes sectors (e.g. agriculture), with modali-

ties (e.g. budget support) and themes (e.g. rural development). Furthermore, there are differences in data handling that lead to differences in the figures produced by agencies themselves and those reported by the DAC. For example, the UK reports on budgetary support aid according to estimated percentage distribution across sectors, whereas the DAC reports budgetary support as a single multisectoral spending category. In addition, the UK disaggregates multi-sectoral projects into its individual sectoral components, whereas the DAC uses one single code per project choosing either the dominant sectoral component or coding into a single multi-sectoral spending category. The World Bank, on the other hand, categorises expenditure by both sector and theme and allows up to five sectors or themes to be allocated to a multi-sectoral loan. Adding to the gaps between CRS and donor own statistics, the study also revealed discrepancies, at the level of individual donors, between aid policy trends and assistance flows. Decentralisation in aid management and greater autonomy given to country offices to design country programmes, implies that headquarters policy guidance is less influential than claimed. As for recipient countries, their main concern is not with linking policies with aid flows but with monitoring aid flows at the project level to ensure commitments are met and disbursement constraints are addressed. Strategic reviews on the nature of assistance and cross-sectoral consistency are not yet primary concerns and would prove difficult due to the considerable fragmentation of aid systems, and, by consequence, the significance of off-budgetary aid.

18] IFPRI, 2009 19] IFPRI, 2009 20] See compilation of studies in Farnworth, C.R. (2008) Module 5: Gender and Agricultural Markets. In Gender in Agriculture Sourcebook. World Bank. http://worldbank.org/genderinag 21] Kitinoja, Lisa. 2002. Identifying Scale-Appropriate Postharvest Technology. In Postharvest Technology of Horticultural Crops, 3rd ed., ed. Adel A. Kader, 48190. Oakland,CA: Division of Agriculture and Natural Resources and University of California. 22] Gurung, C., 2006. The role of women in the fruit and vegetable supply chain in Maharashtra and Tamil Nadu India: the new and expanded social and economic opportunities for vulnerable groups task order under the Women in Development IQC. Washington, DC: USAID. 23] USAID, Gender and Economic Value Chains: Two Case Studies from the GATE Project, www.usaid.gov/our_work/cross-cutting_programs/wid/eg/gate_valuechai n.html.24] Food and Agriculture Organization, 2009. The state of food insecurity in the World 2009. Rome. http://www.fao.org/docrep/012/i0876e /i0876e00.htm 25] DFID/DEFRA (March 2010) DFID/Defra Policy Narrative on Global Food Security and Sustainable Agriculture. 26] Farnworth, C.R. (2010) Zambia Country Report: Sida UTV Working donorplatform.org Paper 2010:8

Platform Policy Brief I No. 5

// Towards a broader measure of ARD and food security aid for global transparency
In an attempt to address the coverage shortfall in the CRS, the study proposes a broader measurement of ARD and food security aid, which incorporates CRS purpose codes considered relevant and adds these to the DACs AFF+ measure. The aim is to capture, in aid flow terms, the importance of the policy trends identified above. The first set of trends considered relates to the balance in priorities between aid measures concerned with: (i) agriculture as an economic sector; (ii) promoting rural development more broadly, including physical infrastructure and social and governance structures; and (iii) immediate relief of poverty and suffering. Most donors support all of these measures and there is not necessarily a trade-off between any one or the other. Yet, there is evidence of a relative decline in support to agriculture as an economic sector, at least until very recently, and a relative increase in immediate relief. The second set of trends related to the policy choices in supporting agriculture. There appears to be a trend away from direct support to producers and selected commodities towards more indirect measures, notably support for the design of incentive policies, more effective government performance and market development. This recalculation of ARD and food security aid considers, for example, trade facilitation and attributes 20 per cent of it, corresponding to the average weight of agricultural commodities in developing countries trade flows, to agriculture. Likewise, 10 per cent, corresponding to the share of agricultural value added in developing countries GDP, is applied to financial services aid and added in. Thereby, volume is added to the DAC AFF and AFF+ measures. For example, for the period 1995-2009, the proposed measure adds 33 per cent of volume to AFF+ and more than doubles AFF. Furthermore, from the late 1990s, the proposed measure starts growing more rapidly than AFF+, due to the steep growth of some of the aid categories added in, such as democratic participation and civil society, business support services and SME development. This is significant as these areas represent for some donors at least a new approach for supporting ARD and food security objectives. While such a recalculation improves historical analyses of global trends, there remains the need, particularly at country level, to systematically and accurately track commitments and assess the effectiveness of aid by clarifying the link between spending activities and results.

27] DFID/DEFRA (March 2010) DFID/Defra Policy Narrative on Global Food Security and Sustainable Agriculture. 28] Farnworth, C.R. (2010) Gender aware approaches in agricultural programmes: a study of Sida-supported agridonorplatform.org cultural programmes. Sida Evaluation 2010: 3

Platform Policy Brief I No. 5

TPOLICY RECOMMENDATIONS
// Clarifying the scope of ARD and food security aid for improved transparency and accountability
Greater cross-donor coherence is needed in defining the scope of the sector, so that crossdonor trends can be meaningfully compared and global commitments can be accurately tracked and accounted for. Global aid systems should draw on the strengths of individual donor practices in aid measurement and accountability. Individually, donors are generating policy-relevant aid data, mainly for accountability purposes within their own domestic constituencies. The strengths and limitations of different approaches need to be considered collectively. Yet, the incentives for standardising aid data globally are relatively weak. New attention on how low-income countries can cope with food price increases and volatility to which the AFSI is a response may change this, alongside mounting pressure to improve transparency and accountability. Thus: Agricultural production and productivity would be a component of results in the theme of economic growth Rural socioeconomic development would be part of results for, say, a theme of sustainable increases in living standards Under the theme of reducing vulnerability, a component on rural livelihoods recovery would be included.

// Breaking sectoral boundaries in aid measurement for improved consistency between policy and resource allocation
Finally, strengthening consistency between policy and resource allocation requires looking at expenditure across sectoral boundaries. The structural changes in the ARD and food domain and the growing connectedness of its individual policy components with other policy domains should be borne in mind and lead to advocacy and management strategies which break beyond artificial sectoral boundaries. Supporting agricultural production and productivity is no longer simply about aiding production systems, through input subsidisation or irrigation systems development, but also about creating a more conducive environment for viable agricultural production. A step forward would be to design aid measurement mechanisms that incorporate the breadth of interventions necessary to promote the distinct policy objectives underlying ARD and food security irrespective of the sector labels these intervention may carry with them.

// Unpacking the objectives of ARD and food security aid for manageable results frameworks
It is results-based aid that is likely to be more challenging to establishing a distinct and authoritative profile for ARD and food security. Two major challenges need attention. One is attribution, as the link between spending activities and results is complicated by the wide range of factors that impact upon production, incomes, employment, living standards and natural resource use in rural areas. The other challenge is the existence of at least three relatively distinct policy domains within the term agriculture, rural development and food security one is focused on production and productivity, another is about providing opportunities for improving the living standards and economic development of rural areas, and the third is about addressing high levels of risk and vulnerability. This does not, however, mean that the three domains are to be tacked separately, but rather that their intrinsically different objectives need to be understood and reconciled. A step forward would be to develop results frameworks for the different policy objectives within the broader themes now favoured by most donors.

// Recommended reading
Coppard, C. (2009) Agricultural development assistance: a summary review of trends and the challenges of monitoring progress, a Development Initiatives report commissioned by ONE. Islam, N. (2011) Foreign aid to agriculture: review of facts and analysis, IFPRI Discussion Paper 01053, Director Generals Office, International Food Policy Research Institute. Komorowska, J. (2010) A proposed methodology for measuring government expenditures in support of food and agriculture sector development and application in the case of Uganda, paper presented at the Global Forum on Agriculture, OECD, November 29-30, 2010.

donorplatform.org