A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D I S C U S S I O N PA P E R 0 1 81083 INVESTING IN AGRIBUSINESS: A Retrospective View of a Development Bank’s Investments in Agribusiness in Africa and Southeast Asia and the Pacific GEOFF T YLER AND GRAHAME DIXIE AUGUST 2013 A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D I S C U S S I O N PA P E R 0 1 INVESTING IN AGRIBUSINESS: A Retrospective View of a Development Bank’s Investments in Agribusiness in Africa and Southeast Asia and the Pacific Geoff Tyler and Grahame Dixie © 2013 International Bank for Reconstruction and Development / International Development Association or The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Cover Photo: Shutterstock, LLC. “The more distant we look into the past, the farther we can see into the future.” Winston Churchill CONTENTS V CONTENTS Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Chapter 1: Executive Summary and Lessons Distilled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Chapter 2: Background and Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Chapter 3: The Origins and Evolution of CDC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Chapter 4: The Base Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Chapter 5: Direct Characteristics and Performance of the CDC Agribusiness Portfolio . . . . . . . . . . . . . . . . . . . . . . 17 Chapter 6: Broader Developmental, Environmental ,and Social Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Chapter 7: Explaining Financial Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Chapter 8: Explaining Development Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Chapter 9: Equity Success Stories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Chapter 10: Changing Performance over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Chapter 11: Sub-Saharan Africa versus Southeast Asia and the Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Chapter 12: Estate versus Outgrowers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Chapter 13: Settlers, Smallholders, and Outgrowers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Chapter 14: Size and Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Chapter 15: Start-up versus Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Chapter 16: Pioneering—First mover advantage or paying the price? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Chapter 17: Debt versus Equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Chapter 18: CDC Management—Making a difference? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Chapter 19: Conclusions—Critical Success Factors and Key Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Appendix 1: A Checklist of Some Critical Commercial Success and Failure Factors for Agribusiness Investments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Appendix 2: Data Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Appendix 3: Projects Included in the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R F I G U R E S , TA B L E S , A N D B OX E S VII FIGURES, TABLES, AND BOXES FIGURES Figure 3.1: Investment Strategy over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Figure 5.1: Comparison between the Investment Portfolio Mix between Asia and Africa. . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 5.2: Investment Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 5.3: Market Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 5.4: Investment Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Figure 7.1: Percentage of Projects that Succeeded and Failed with Reasons for Failure . . . . . . . . . . . . . . . . . . . . . . . . . 27 Figure 10.1: Financial Viability over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Figure 12.1: Sustainable Development Impact and Equity Returns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Figure 12.2: Financial and Development Performance of Estate Farming, Nucleus Farms, Outgrower Schemes and Independent Processing Operations: Percentage Classified as Successful or Moderately Successful . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Figure 15.1: Development Impact: Percentage Classified as Success or Moderate Success . . . . . . . . . . . . . . . . . . . . . . 43 Figure 15.2: Percentage of All Projects Classified as Success or Moderate Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 TABLES Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Southeast Asia and the Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Table 5.1: Regional Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 5.2: Country Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 5.3: Enterprise Focus by Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 5.4: Investment Type by Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 5.5: The Sizes, Judged Relative to Industry Norms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Table 5.6: The Market Orientation of Projects—Local Market or Exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Table 5.7: Investment Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Table 5.8: CDC Involvement in the Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Table 5.9: Equity and Loans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R VIII F I G U R E S , TA B L E S , A N D B OX E S Table 5.10: The Cost of CDC's Agribusiness Investments in Africa and Asia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Table 5.11: Investments which Exceeded US$100 Million in 2011 Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Table 5.12: Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Table 7.1: Projects Classified Financial Failures or Moderate Financial Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Table 12.1: Percentage of Projects Classified as Success or Moderate Success. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Table 13.1: Causes of Project Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Table 14.1: Percentage of Projects Classified as Success or Modest Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Table 14.2: Projects Rated as Financial Success or Moderate Financial Success, by Size of CDC Investment . . . . . . . . . . . . 41 Table 15.1: Percentage of Projects Classified as Success or Moderate Success. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Table 16.1: Percentage of Projects Classified as Success or Moderate Success. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Table 17.1: Percentage of Projects Classified as Financial Success or Moderate Financial Success . . . . . . . . . . . . . . . . . . 47 Table 18.1: Percentage of Projects Classified as Financial Success or Moderate Financial Success . . . . . . . . . . . . . . . . . . 49 BOXES Financial Restructuring of the Mpongwe Development Company, Zambia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Sacrificing Development to Avoid Public Controversy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Kilombero Valley Teak Company, Tanzania. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Smallholders and Outgrowers: Spreading the Benefits or Evading Responsibilities? . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 From “Dog” to “Star” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Cutting the Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Surviving the Bad Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 I N V E S T I N G I N A G R I B U S I N E S S : A R E T R O S P E C T I V E V I E W O F A D E V E LO P M E N T B A N K ’ S I N V E S T M E N T S I N A G R I B U S I N E S S ACKNOWLEDGMENTS IX ACKNOWLEDGMENTS The authors are indebted to the Japanese Government, in particular the Ministry of Foreign Affairs, for funding, for being supportive of this work, and for their overarching aim of encouraging increased and responsible investment in agriculture in order to make the world more food secure. The authors would like to gratefully acknowledge the valuable insights provided by peer reviewers Angus Selby, Steven Schonburger, Mike Kubzansky, and Karen Brooks, as well as additional feedback received from comment providers including Aubry Hruby, Bill Vorley, Brian Baldwin, the Lord Cameron of Dillington, John Linton, Jonas Heirman, Jorge A. Muñoz, Keith Clifford Bell, Li Xiande, Gavin Wall, Mike Pfister, Nomathemba Mhlanga, Orin Hasson, Patricia Bonnard, Paul Guenette, Peter White, Ray Goldberg, Ron Appleby, Roble Sabrie, Robynne Anderson, Sila Sahin, and Tim Reardon. Gunnar Larson edited the report. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED 1 Chapter 1: EXECUTIVE SUMMARY AND LESSONS DISTILLED PURPOSE OF THE STUDY mainly private investments in, marketing, mechanization, process- Recent increases in the prices of agricultural commodities have ing, inputs, and other elements that establish the preconditions and spurred a surge of private investment into farming and agribusiness.1 enabling environment for broader agricultural and rural develop- After decades when host developing countries tried with limited ment.3 Given the right types of large-scale investment, this could success to encourage investment in their agriculture sectors, many have a transformative effect in underdeveloped rural areas and have are now faced with difficult decisions about the number, size, and a positive effect on national economic development—including type of investments to accept. A corollary to this increasing interest the provision of domestic food supply to urban areas that could have been rising concerns about whether large-scale investment in reduce dependence on food imports. commercial farming—and more especially large-scale land acquisi- While it is outside the purpose of this study to assume a position tions—do indeed deliver public goods, and about the effects these in that debate, the findings presented in the following document investments and acquisitions have on the rights and livelihoods of suggest that larger-scale agro-investments can help to raise large local communities.2 The issue quickly became highly contentious. numbers of people out of poverty, but that such outcomes are A number of prominent nongovernmental and civil society organi- contingent on a number of factors. These outcomes will be ex- zations take a position that is critical of large-scale land acquisition pressed in terms of the development impacts of a Commonwealth in general, and particularly in countries with weak regulatory capac- Development Corporation (CDC) investment, as distinct from its ity and ill-defined property rights. Some of these organizations ad- technical or financial results. And while there are some very good vocate for a focus on investing in smallholders as the way to enable and very bad agribusiness investments, most lead to a mixture of poor and vulnerable communities to participate in and benefit from positive and negative impacts. The positives are mainly related to agricultural development. Others see efforts to prevent investment economic development in terms of jobs and access to markets, in this area which until so recently was seriously underinvested in, but often also include some investments in social infrastructure, as being inherently misguided, and the overwhelming priority as- improved access to rural infrastructure, the transfer of useful tech- signed to smallholders as being naïve. In their view, the prospects nologies and skills, and in a smaller number of projects, increased for smallholders being able to participate in and benefit from rapid production of staple foods. The negatives are most often associated agricultural development will be largely determined by large-scale, with a lack of consultation with the communities concerned, lim- ited transparency, an absence of mechanisms for resolving disputes, and issues involving land rights—especially informal land rights. 1 The term “agribusiness” is defined for the purpose of this study as com- Negative impacts may also be seen in irresponsible environmental mercial farming (by smallholders, outgrowers, estates, and plantations), fishing, aquaculture, and forestry and directly-related input supply and processing businesses. 3 In fact, the significantly diminished role of the public sector in financing 2 We refer here to projects that are actually implemented. A different and subsidizing these investments compared to its role in the last three “land-grab” issue is where land concessions are obtained on the prom- decades of the 20th century, and the proportionately greater role that ise of agricultural development but once acquired are sold on for specu- private sector investment will play in the present represents a funda- lative gain and/or used for other purposes. mental difference in what the composition of investment will be. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 2 CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED practices and in the social and economic consequences if the in- Twenty-two percent followed the nucleus estate and smallholders vestment fails. In view of these concerns over the risks associated (NES) model in which an investment is made in a processing plant that with increased interest in larger-scale investment in agricultural has an adjoining large-scale farm coupled with outgrowers supplying land, a retrospective review of a large number of private and public the necessary raw material. Twelve percent were outgrower schemes sector agribusiness investments was commissioned to generate with no significant estate element. The remaining 20 percent had no objective empirical knowledge about outcomes; to differentiate farming component. between alternative business models; to provide insights into the Over 60 percent of the projects were start-ups, while one-quarter likely correlates of success and failure over time; and to deliver this involved the expansion of existing, ongoing concerns. Thirteen knowledge into the public domain. percent involved the rehabilitation of moribund enterprises or This study analyzes the experience of the CDC as an investor in assets. commercial smallholder and estate agriculture and agro-processing CDC was the main shareholder in nearly one-third of the projects in Sub-Saharan Africa and Southeast Asia and the Pacific between and managed nearly half of them. A little over one-third of projects 1948 and 2000.4 were promoted by private investors. Twenty percent of CDC’s in- The CDC was established in 1948 as the Colonial Development vestments were in the form of loans to governments or parastatal Corporation. It was created as an agency of the British government enterprises. Using 2011 values, the average investment made by assigned to promote economic development in the remaining CDC was US$33 million, while there were 12 projects in which CDC British Colonies and thereby to improve the availability of food invested over US$100 million. 5 and raw materials within the Sterling Zone. The survey covers 179 projects in 32 countries, representing a large and diverse set of RESULTS agribusiness investment experiences—albeit not necessarily repre- Four components of project performance were assessed for the sentative of either the countries concerned or of their agribusiness whole portfolio: sectors, given that CDC had its own objectives and priorities that changed repeatedly over time.  Technical Results: achievement of production and physical productivity targets, suitability of technologies used SUMMARY OF INVESTMENTS  Direct Development Impact: creation of sustainable liveli- hoods (formal employment, income-earning opportunities), Two-thirds of the projects surveyed were in Sub-Saharan Africa. explicit contribution to national development goals (export Seventy-seven percent were concentrated in 13 countries.6 Nearly earnings, food production) one-third of the projects focused on three crops: oil palm, sugar, and  Financial Viability: creation of financially self-sustaining tea. Nearly 90 percent of the projects involved some form of process- enterprises ing of raw material. Just over 80 percent of the projects were wholly or  Equity Returns: dividends and capital gain to shareholders, partly oriented toward export markets. where equity capital was involved Forty-six percent of the projects were estates or plantations— For each component, project performance was classified as: large-scale farming operations with no smallholder component.  Fail: total or substantial project collapse during implementa- tion or shortly after completion 4 Generally referred to as “Africa” and “Asia” in the report.  Moderate Fail: some positive achievements, but far fewer 5 While agribusiness was a high priority for CDC, it also invested in a wide range of other economic sectors, such as mining, power gen- than planned eration and distribution, manufacturing, housing development, hotels,  Moderate Success: substantial on-going benefits although financial institutions. fewer than planned 6 Côte d'Ivoire, Fiji, Indonesia, Kenya, Malawi, Malaysia, Nigeria, Papua New Guinea, Swaziland, Tanzania, Thailand, Zambia, and Zimbabwe.  Success: Main objectives achieved or exceeded INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED 3 Some projects were technically sound and well-implemented but  Over half achieved reasonable or good overall financial went on to collapse financially owing to low market prices (for exam- performance. ple, tung oil in Malawi) or civil war (for example, rubber in Liberia).  When equity investment was involved, one in six achieved compound equity rates of return of over 12 percent. Some projects contributed substantially to national development Unfortunately it is not possible to calculate the profitability of objectives, but with lower-than-planned profit margins. Financiers CDC’s total or regional agribusiness portfolios with the available consequently had to “write-off ” a substantial portion of their origi- data. CDC itself, after early losses, reported a profit in its accounts nal loans in order to achieve continued financial solvency, while every year from 1955 to 1997. Given that some of CDC’s agribusi- shareholders had to “write-down” the value of their equity stakes ness equity investments yielded very large capital gains (for exam- to reflect a realistic valuation of the net-worth and actual business ple, the BAL plantations in Sabah were sold in 1996 for £100 mil- performance of the company. lion) it is probable that the agribusiness portfolio yielded a positive return overall in monetary terms although not necessarily in real FINANCIAL RESTRUCTURING OF THE MPONGWE terms (after adjusting for inflation). DEVELOPMENT COMPANY, ZAMBIA The Mpongwe arable crops project in Zambia was a case in A simple analysis of the data was undertaken to determine whether point. At the beginning of the 1990s it was insolvent, unable to success and failure could be correlated to any critical factors. service the debt taken on for the development to-date, in spite Seventy-nine (or 49 percent) of the projects were classified as fail- of generating some positive cash flow. ures or moderate failures in financial terms. Both the initial lenders and shareholders (normally one and the same, for example, CDC, International Finance Corporation  In 60 percent of these cases, the major cause of failure was (IFC), and DEG had to take a “haircut.” Technically, CDC convert- that the project concept was fatally flawed, for example ed its debt to equity and then “wrote down” the value of the wrong location, wrong crop, or overoptimistic planning equity in its own balance sheet. The government of Zambia assumptions. About one third of these were unknowable at bought the loans and equity of the other investors for a small the time of appraisal. percentage of the face value, and then also converted these loans to equity.  One in five had the “bad luck” to be adversely affected by government policies (10 percent), or closed down due to This served to establish a joint venture between CDC and the civil unrest (8 percent), or suffered from a collapse in markets government which was almost entirely free of long term debt, (2 percent). providing a fresh, solvent platform on which to base successful  About 20 percent failed due to bad management. expansion in the future. On average, investments in Asia did better than in Africa. For The performance ratings have been based on objective indicators instance, 70 percent of Asian investments were at least mod- where available (actual production, employment, financial solvency, erately successful in financial terms, compared with 44 percent equity returns data, crop production statistics) but there remains a in Africa. subjective element to the classifications, especially in the border- The proportion of projects that suffered from flawed concept line cases. and from bad management was very similar in the two regions. In broad terms the analysis shows that: Overall, just over one quarter of all projects were flawed in their concept and about 8 percent were poorly managed. A signifi-  Fewer than one in five projects surveyed were rated com- plete failures, delivering no significant direct development or cant difference between the two regions was the proportion of financial benefits. projects in which failure was attributable to bad luck. Bad luck  Nearly two-thirds of projects achieved the intended direct caused 13 percent of African projects to fail, but only 2 percent development impact. of Asian projects. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 4 CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED The relatively greater success rate seen in Asia was in part the re- its published Annual Reports. Instead we have reviewed a selection sult of the post World War II boom in palm oil. All 10 investments in of projects which illustrates some of the wider potential and pitfalls Asia which had an equity internal rate of return estimated at more of agribusiness investment without suggesting any overall “success” than 12 percent were oil palm projects. The African counterpart or “failure” ratings. to oil palm has been the success story of sugar and tea. However, In general, CDC strived to be a model promoter or supporter of ag- whereas Asia faced no market limits to the expansion of its palm oil ribusiness ventures, according to contemporary standards—which industries, the growth of the African sugar industry was in practice have however evolved greatly over the past 60 years. Some of CDC’s constrained by domestic demand and EU and U.S. import quotas. earlier activities and priorities would not be acceptable today. World tea prices have been in decline throughout most of the post war period. Perhaps its main weakness was a focus primarily on resolving is- sues within the project boundary (for example, land conservation, NES schemes had a higher probability of success than either stand- pollution control, health and safety standards, employee housing alone estates or stand-alone outgrower schemes, both in terms standards) while paying relatively little attention to broader conse- of development impact and financial performance. CDC rarely in- quences outside (growth of shanty towns for casual labor, project vested in smallholder/outgrower schemes unless their raw material roads providing access to sensitive environments for informal ex- output was closely tied to a related industrial processing facility, as ploitation, impact of promoting smallholder cash crops on house- is the case with green leaf tea, sugarcane, or oil palm fresh fruits hold food production and nutrition). bunches (all of which cannot be stored and must be processed within a reasonable distance from the point of harvest). Typically In most cases CDC avoided controversial land acquisition/resettle- they would only introduce outgrowers into the business model ment issues either by taking over existing moribund estates (for when any technical and production issues had been resolved. example, the BAL plantations in Sabah) or purchasing land that was already in private hands but underutilized (for example, cattle When CDC did venture into supporting smallholder crops that could ranches were acquired for the Swaziland Irrigation Scheme and the be stored and/or sold to third parties, (side-selling), the schemes Kaleya smallholder project). In its earlier years it was not controver- usually ran into credit-recovery difficulties such as in the cases of sial for CDC to convert areas of previously logged, natural forest for oil-seeds in Kenya and tomatoes in the Philippines. agriculture and plantation forestry (Sarawak Oil Palms, Societe de Development des Plantations Forestieres [SODEFOR] teak planta- The results showed significantly higher levels of failure among tions in Ivory Coast) but by the early 1990s it was essential to incor- start-ups and investments in moribund enterprises, compared with porate integrated plans for management of the total concession, investment in expanding existing agribusiness. including areas for preservation, and to consult with, and recognize BROADER DEVELOPMENTAL, ENVIRONMENTAL, the traditional use of forest areas by nearby local communities (for AND SOCIAL IMPACT example, the Kilombero Valley Teak project in Tanzania). It was not until the mid-1990s that CDC began to systematically set When, in 1983/4, CDC did directly venture into leasing land from standards for, and to monitor, the developmental, environmental smallholders (the proposed National Development Corporation and social aspects across its entire investment portfolio, establishing (NDC)/CDC oil palm project at Loreto, Mindanao, Philippines) the a development committee of the main board in 1996, and starting adverse publicity proved to be unsustainable and CDC withdrew. the production of regular development reports to complement its annual financial report and accounts. LESSONS DISTILLED It is therefore not possible to undertake a systematic historical anal- This review of CDC agribusiness investments corroborates the ysis of the broader effects of CDC’s agribusiness portfolio based on view that agribusiness investments are risky, particularly when the INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED 5 investment is in a start-up. While only one fifth of projects were rated 80 percent of equity investments failed. When new management7 complete failures, one third of equity investments generated at least took over at the start of the 1950s, a much more rigorous and moderately attractive internal rates of return, and overall about 55 commercially-orientated approach was taken, including an insist- percent resulted in financially viable projects (that is, financially self- ence on proper agronomic trials before launching into full scale sustaining). The majority of projects in both Asia and Africa ended agribusiness developments. Unviable projects were weeded out. A up being sustainable businesses that delivered broadly the number proper review process was set in place to consider each investment of jobs and level of turnover that had initially been anticipated. This on its merits. Performance significantly increased. The percent- raises the question of why, despite this low level of returns on eq- age of equity investments rated as failures dropped to around 40 uity, these businesses often survive. percent. This improved performance lasted until around the start of the 1970s. The UK government became more involved, insisting The answer appears to be based on sunk cost. Although the initial that the CDC should use its funds to achieve greater development investment often fails to achieve intended levels of profitability, and impacts. More risky projects were taken on. Investments were fre- although project implementation often takes longer than planned, quently made in indigenous businesses. This high toleration of risk if the venture is capable of generating positive cash flows there is resulted in increased levels of equity investment failure, but not to usually nothing to be gained by closing it down. It is therefore either the levels seen during the late 1940s. recapitalized by its owners or sold on, at a discount, to a second or a third investor injecting additional capital. Ultimately, the total amount The findings indicated that nucleus farms have historically been invested is typically more than could be justified on a purely finan- the least risky of the different business models. Although this is cial basis, but the final investor generates a sensible return on their an encouraging result for those advocating the inclusion of small- marginal investment and the business continues as a “going concern.” holder farmers into business models, the nucleus farm model should by no means be seen as a panacea. The reasons for their High ex-ante “hurdle rates” are often set by private investors relatively high success rates are believed to be partly attributable because they need to have potential equity “stars” that can com- to the type of crop enterprises that were invested in. CDC’s focus pensate for some inevitable “dogs.” Actual average rates of return was sensibly on a limited range of industrial crops, including oil realized ex-post are generally much lower. This raises an important palm, tea, sugar, and rubber. The financial institution built up a core question about the potential for “patient” forms of capital. These expertise on business models developed around the production instruments support innovation and long-term development by and marketing of these crops. Most often the production tech- helping the original private investors to bridge the often extended nology was largely resolved, normally initially as an estate type period between their first investment and the eventual realization operation, before smallholder elements were introduced. During of positive financial returns. the periods shortly after countries became independent, CDC Overall, projects were more likely to succeed in the long term when was approached by governments requesting that smallholders the agronomic and economic fundamentals were sound. While bad be included in the agribusiness models generally as outgrowers luck and bad management can destroy a sound enterprise, good producing raw materials to supply the agribusiness.8 These adapta- luck and good management can rarely compensate for a project tions were actually found to work better than many had expected, that is fundamentally flawed. and subsequently became incorporated into the design and plan- ning of more future investments. The importance of incorporating The study demonstrated that outcome is in large measure depend- ent on the attitude and processes of the funding institution. In its 7 Lord Reith, the founder of the BBC, was appointed Executive Chairman first few years the CDC exhibited high levels of overconfidence and on 1 November 1950, with a mandate to sort out the mess. naivety. Projects were approved on the basis of minimal analysis 8 Outgrowers are generally smaller scale farms surrounding the processing plant who grow crops specifically for the agribusiness, and without proper review. The results were very poor. Around generally under some sort of contractual arrangement. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 6 CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED smallholders into proven business models was underscored later Some of these initiatives, in which CDC served as promoter or finan- in the institution’s history when CDC attempted to include small- cier, helped to pioneer innovations that subsequently grew organi- holders in business models before they were thoroughly tested. cally through a series of expansions into very large undertakings. Some of these projects failed, leading to the outgrowers having to These included: shoulder a portion of the downside.  The Federal Land Development Authority in Malaya, which The CDC experience in integrating smallholders into large-scale involved the settlement of landless farmers to become rubber and oil palm outgrowers. Begun in 1957, over the agribusiness investments provides important practical lessons ensuing 40 years, some 120,000 families were settled in over about how the sequencing of the initial investment and engage- 300 new communities; ment with smallholders can affect the level of risk that those  The Kenya Tea Development Authority (KTDA), which smallholders assume in participating. Because the risks entailed involved the promotion of smallholder tea growing to in start-up or venture investment projects are very high, and supply dedicated factories. It began in 1960 with 940 because smallholders’ ability to absorb risk is generally quite low, hectares under smallholder tea cultivation. By 1984, some exposing them to this level of risk is irresponsible. In the not at 145,000 participating smallholders were cultivating tea on 58,000 hectares. Currently Kenya’s small scale tea all unlikely event that the investment goes on to fail, the con- growers produce about two-thirds of the export crop sequences to the outgrower can be catastrophic. Alternatively, which delivers around US$300 million of farm gate income when the outgrower scheme is based on a well-established and to these small holders annually; and successful business model in which problems that arose early on  Zambia Sugar, the first sugar estate in Zambia, began in 1967 in the operation have been effectively resolved, the level of risk is with an initial field and factory capacity for 35,000 tonnes of substantially reduced—and in particular risk to the participating sugar, by 2011 was producing 385,000 tonnes. outgrower. A number of CDC nucleus estate schemes exemplified A number of successful pioneering projects wielded demonstra- this principle. tion effects through which their practical example came to serve as While the finding that start-ups had a lower success rate than models for projects elsewhere: investments into expanded existing concerns is not surprising, it  The KTDA for instance would provide a model for outgrower does suggest useful lessons. First, it explains the focus of the newer tea projects in Uganda, Tanzania, and Malawi. private agriculture funds on existing agribusinesses and farming  The commercial oil palm and cocoa production that CDC enterprises. Second, while being a first mover is risky, a propor- pioneered in Malaya, Sarawak, and Sabah (despite early tion of these investments have a transformative effect. Successful teething problems in the latter) encouraged other investors pioneers can both attract further investment and also stimulate to develop new plantations and to convert existing rubber investment on the part of those who follow. And this must be one estates to these more profitable crops. By 1996, 40 years after CDC introduced the crop, Sabah alone accounting for of the purposes of state-funded development corporations and 7 percent of world palm oil production banks.  CDC was one of the earliest promoters of export-oriented Many of CDC’s agribusiness investments were pioneering. Some horticulture in Kenya on the Osarian and Kuraiha Estates represented the first such operations in a particular country, for and while CDC itself failed to achieve profitability, the farming assets that were established became the nucleus instance the introduction of tilapia in Lake Kariba, Zimbabwe; sug- for the horticultural9 industry which later thrived in arcane in Swaziland and Papua New Guinea; and oil palm in Sabah Kenya. and Sarawak. Others were pioneering in introducing a new produc- tion model, for instance organized smallholder tea production in Kenya and Malawi. 9 Mainly high value vegetables and cut flowers. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 1 — EXECUTIVE SUMMARY AND LESSONS DISTILLED 7 CONCLUSIONS A fourth lesson is to adapt to the sociopolitical realities of the A key lesson of CDC’s experience is the value of patience. Several ul- times. CDC began as an instrument of the British Colonial Office. timately successful investments went through very difficult phases It was able to purchase huge tracts of undeveloped or under- early in their development. Early losses made CDC unsure whether developed land for agricultural development, generating little to cut its losses or persevere. In each case, review teams determined controversy—other than a fear that it would be a waste of British that the fundamentals were good and that CDC should be patient tax-payers’ money, like the “Groundnut Scheme.” CDC could have and commit additional resources. Examples include a loss making faded away along with the Empire. By 1963, when most colonies rice project in Swaziland converted to sugarcane, a struggling arable had achieved independence, CDC reinvented itself as a develop- estate in Zambia that achieved economies of scale via expansion ment agency under the new Ministry of Overseas Development. and merger with an adjacent state farm, an abaca fibre plantation in In the 1970s it supported the nationalization of major agribusi- Sabah successfully converted to oil palm and cocoa, and a rubber ness ventures (for example, Kilombero sugar in Tanzania), the plantation and factory development in Ivory Coast that came on development of state enterprises (for example, Hevecam rubber stream when world markets prices were in a slump—CDC financial- plantations in Cameroon) and the training of local managers to ly supported the venture for several years and realized a substantial take over from its own seconded expatriates. In the late 1980s equity profit during the subsequent commodity price boom. and 1990s it supported privatizations (acquiring the Cavally rubber estate from the government of Ivory Coast), it refocused A second key lesson is the importance of managing the risks of its new investments exclusively in the private sector of poorer investing in agribusiness by establishing a diversified portfolio— countries, and it began to explicitly report on the broader devel- by country and by sector. Commercially, CDC got it wrong when opmental, environmental, and social impact of its activities. By it invested heavily in tung oil plantations in Malawi (market lost to the end of the period under review, CDC had decided that the synthetic substitutes) and got it right when it helped to pioneer political and commercial risk of directly owning and managing oil palm plantations in Southeast Asia. Financially, CDC lost heav- large-scale agricultural plantations was too great and it sold most ily investing in Uganda (considered a jewel in the colonial crown) of its remaining direct agribusiness equity holdings and switched whereas it did very well pioneering commercial agriculture in the to more indirect forms of investment (in private equity funds). colonial backwater that was Swaziland (considered at the time des- tined to be absorbed into the Republic of South Africa). CDC’s experience should help to dispel any pessimistic myths that investing in poor developing countries or in the agribusiness sector A third major lesson is to have a broad outlook, looking beyond the or working with smallholders is doomed to failure. It should also project boundary at the wider developmental, environmental and so- dispel the contrary myth that foreign investors exploiting devel- cial implications of a proposed development. In the colonial era it may oping countries’ agricultural resources always make huge profits. have been sufficient to argue that the expected economic benefits of The analysis of CDC’s agribusiness portfolio demonstrates both a proposed development would outweigh any environmental and historical potential and pitfalls and illustrates the need to continu- social costs. In today’s more democratic, transparent, and contested ously adapt and innovate to achieve both political and commercial world it is necessary to actually mitigate any adverse consequences sustainability. where possible and to adequately compensate where not. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R CHAPTER 2 — BACKGROUND AND SOURCES 9 Chapter 2: BACKGROUND AND SOURCES This study builds on an earlier review of the CDC investment experi- Godfrey Davies (CDC’s Chief Financial Officer) also provided critical ence in Sub-Saharan Africa titled “The Fall and Rise of the Colonial additional information on the outcomes of many of CDC’s more Development Corporation,” which was prepared for the World Bank recent investments.10 as a case study within the All Africa Review of Experiences with A search for similar studies yielded few results. Insofar as we Commercial Agriculture, which in turn formed part of the larger can tell, this type of study is rare in that it deals with the entire study Competitive Commercial Agriculture in Sub-Saharan Africa. investment portfolio from a historical perspective to examine The present paper expands that review to include CDC investments investment trends and their returns. While some similar reports in Southeast Asia and the Pacific, and presents a more detailed have been produced, such as an evaluation of Fundacion Chile account of outcomes and analysis of the reasons for success and and a sector evaluation by Independent Evaluation Group (IEG) failure. CDC annual reports published since 1948 make up most of of IFC’s food and agribusiness operations, they differ with re- the references for this report, which also benefitted from a recent spect to their objectives and scope.11,12 history of CDC. Christopher Brain (the history’s principal author) and 10 Christopher Brain and Michael Cable (2008). Pioneering De- velopment. 11 Fundación Chile: Historia e Impacto (2006) by Jorge Quiroz with Mónica Ríos, Jorge Bravo y Gabriel Piña. 12 Food and Agribusiness: An Evaluation of IFC’s Investments in the Sector. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 3 — T H E O R I G I N S A N D E V O LU T I O N O F C D C 11 Chapter 3: THE ORIGINS AND EVOLUTION OF CDC The Commonwealth Development Corporation, originally the to break even rather than to make a profit, that is it was not required “Colonial Development Corporation” (CDC) was established in 1948 to make a profit beyond that needed to service what it borrowed. As as an agency of the British government. In the immediate aftermath a business model, this was logically flawed because it entailed debt of the Second World War, Britain was short of food and raw materi- on commercial terms while most financing consisted of long-term als. It was also short of US dollars to pay for imports. The Ministry of equity investments. Many of its early investments were moreover Food was therefore determined to promote increased production poorly researched and implemented. CDC was insolvent within from within the Sterling currency zone, that is, mainly the remaining three to four years of start-up. colonies. Within the Colonial Office, the pace of economic develop- CDC initially had no intention of simply being a banker, on-lending ment in the colonies was widely regarded as being too slow, and to public or private ventures at a higher rate of interest and with this was seen as being attributable to the inertia of local adminis- good security. It saw itself directly tackling the type of projects in trations. The solution proposed was to establish a central body to the kind of countries that the private sector would be wary of. The conceive and carry out major projects independently of existing first annual report in 1948 noted that “it is already clear that it is colonial authorities. In the end two separate statutory bodies were in the least developed, rather than the most highly developed ter- created. ritories that the Corporation’s main work will be done . . . The tasks The Overseas Food Corporation came under the Ministry of Food. of development are too large, and the financial return too distant or Its first and last major initiative was the East African Groundnuts the risks too great, to attract sufficient private capital.” scheme in Tanganyika. This was an almost complete agronomic and The Corporation therefore “preferred venture to caution” and deter- commercial failure. The term “groundnut scheme” became a by- mined that the bulk of its investments would be made in the form word in Britain for grandiose, ill-conceived, and poorly implemented of equity. In fact some of its projects were undertaken “directly,” government projects. without incorporating a separate legal entity for the project. This The Colonial Development Corporation came under the Colonial meant that all of the liabilities and risks fell directly onto CDC’s own Office and was assigned a broader purpose. Its mandate was to im- balance sheet. CDC organized itself administratively into production prove “the standard of living of the Colonial peoples by increasing divisions: agriculture, fisheries, forestry, mining, transport, power, their productivity and wealth.” CDC was not envisaged as an “aid” hotels, etc., each with an intended capability to plan, implement, agency. CDC took the form of a statutory corporation, with no share and manage commercial projects in the colonies. capital of its own, but with access to long-term loans on near com- mercial terms from the British Treasury. As a statutory corporation it From the beginning, agriculture—and African agriculture in par- had no equity capital, and was required to break-even each year. A ticular—was always a high priority, and was expected to play a borrowing facility of £100 million was made available—equivalent major part in CDC’s activities. “Africa, the Board believe, is the most to around £2 billion today. CDC’s statutory financial obligation was promising field for large-scale development .  .  . In the sphere of A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 12 C H A P T E R 3 — T H E O R I G I N S A N D E V O LU T I O N O F C D C agriculture much worth-while work can be done immediately by refocused its new investments on development banking, primarily larger production of crops for the local market and by using such providing long-term loans to creditworthy public and private sector schemes to popularize more productive methods of peasant farm- enterprises, often co-financing with the World Bank. This provided ing . . . the Agricultural Division is regarded as potentially the largest CDC with a strong cash flow and ultimately allowed it to build up sphere of the Corporation’s activities.” During its first three years, over a capital base of “reserves” which could be used to finance equity 50 percent of CDC’s investment and financial commitments were for investments, without risking overall solvency. CDC continued to agribusiness ventures. At the end of 1951, 48 percent of CDC actual manage projects where it was the controlling investor. Under this and committed investments were in agribusiness. more risk-averse business model the share of agribusiness in CDC’s investment portfolio shrank to 31 percent by 1962. CDC’s original geographical scope was restricted to British colonies. This mandate was later extended, first to include former-colonies From the mid-1960s, CDC’s geographical remit was gradually ex- that remained within the Commonwealth, and then to include any tended to include all developing countries, and CDC was encour- developing countries as sanctioned by the British government. This aged by the UK government to support projects based upon “re- history explains the early focus on such countries as Swaziland, newable natural resources.” These projects would benefit the Bechuanaland (Botswana), Malaya, North Borneo (Sabah), and the poor more directly, for instance through smallholder agricultural later extension to such countries as Cote d’Ivoire, Mozambique, the schemes. Many were in the public sector, and some with CDC pro- Philippines, and Indonesia. viding management services. CDC also facilitated the nationaliza- tion of some of projects in which it was an investor, for example, CDC has always had the flexibility to promote economic develop- Kilombero sugar in Tanzania, and promoted the training of national ment in a wide variety of ways: initiating its own projects and sup- managers to take over from its seconded expatriates.13 porting the initiatives of others; undertaking projects directly on its own balance sheets (that is, divisions of CDC) and providing equity During the world food crisis of the 1970s, priority shifted to helping and loan finance to separately incorporated companies; investing developing countries to produce more food themselves, with par- in private ventures and in public-private joint ventures; and mak- ticular focus on smallholder production. When several borrowing ing sovereign loans directly to foreign governments. In addition to countries defaulted on their international debts in the 1970s and finance, CDC has provided technical resources and industry special- 1980s, foreign exchange earnings again became a prominent issue ists to plan and manage projects in mining, hotels, electricity gen- and CDC looked to equity investments as a way of soaking up debt eration and distribution, cement manufacture, and other sectors service payments accepted in local currency. CDC shifted its pri- as well as in agriculture. It has also provided technical assistance to mary focus to supporting private sector enterprises through a mix third parties. of debt and equity finance, following the model of the International Finance Corporation (IFC). Yet CDC was unique amongst develop- CDC’s objectives and character have changed substantially over ment finance institutions (DFIs) in continuing to promote and man- time, including the aims of its agribusiness operations which in the age agribusiness ventures in which it had a controlling stake. The beginning focused on helping to feed the British public and supply share of agribusiness in CDC’s portfolio rose to a peak of 53 percent raw materials after the war years. This subsequently changed to a in 1986. focus on providing foreign exchange earnings for colonies about to achieve independence, in line with evolving views of what con- It was not until the mid-1990s that it was given permission to par- stitutes sound economic development. Its original mandate was to ticipate in acquisitions per se, by which time CDC was itself being promote new economic activities and expand existing ones. When targeted for privatization by the British government. Its investment its initial focus on “direct” projects and equity led to near insolvency, CDC switched to more conservative, secure lending, often as sov- 13 CDC established and financed the Managa Agricultural Management ereign debt. After a financial restructuring in the mid 1950s, CDC Centre in Swaziland in 1971. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 3 — T H E O R I G I N S A N D E V O LU T I O N O F C D C 13 FIGURE 3.1: Investment Strategy over Time The changing aims of the CDC’s agricultural investment strategy Food production for the UK Foreign exchange for countries seeking independence Food production for host countries Foreign exchange for debt crisis Profitability via global comparative advantaged 40s 50s 60s 70s 80s 90s strategy once again focused mainly on equity, either in the form CDC announced that its agricultural investments were “for sale,” and of controlling equity stakes in ventures managed by CDC or in CDC was converted into a limited liability company. the form of venture capital style investments, as CDC strived to Ultimately, CDC’s anticipated privatization was cancelled and it was achieve a level of commercial performance that could facilitate its transformed into a “fund of funds” investing in private equity funds own eventual privatization. CDC management believed this strat- rather than directly in underlying projects. egy offered better chances to achieve levels of profitability more typical of private sector firms. These included creating “world class” Throughout the period under review, CDC had the challenge of businesses and focusing on private equity transactions. By the year combining its status as a public body—subject to political pres- 2000, the share of agribusiness in CDC’s portfolio had decreased to sures from governments and civil society in the UK and overseas 20 percent. and needing a clear public purpose to justify its existence—with the need to maintain its own solvency. Its mantra was “doing good with- New management brought in to spearhead the sale of CDC con- out losing money.” In broad terms it reconciled these two objectives cluded that investments in agribusiness were unlikely to achieve by promoting projects that the private sector saw as too risky or an acceptable risk/reward balance. CDC’s willingness to take con- too long-term to develop or by providing concessional finance as trolling stakes in, and manage, large-scale agribusiness enterprises an incentive to private and public sector project developers to re- had allowed it to pioneer many important developments in difficult duce their exposure and risk. However, since most projects had to situations, but it was becoming increasingly untenable politically for compete in competitive markets, CDC expected them to operate CDC to be directly responsible for the livelihoods and working and on a fully commercial basis to give them the best chance of surviv- living conditions of thousands of employees in agribusiness subsidi- ing—and being able to repay CDC’s investment. aries around the world. In its 2000 Annual Report the Chairman of A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 4 — T H E B A S E D ATA 15 Chapter 4: THE BASE DATA Between 1948 and 2000, CDC invested in a heterogeneous mix of Malaya, Sabah, and Sarawak are shown separately above as they were agribusiness ventures with a range of financial instruments and separate colonies while under British rule. However, taking Malaysia technical resources, and with a fluctuating mix of commercial and as one country now, there are 20 countries represented in the Africa developmental motives. This makes for a rich and varied history, survey and 12 in the Asia survey. albeit a challenging one to analyse and interpret, and one that re- Agribusiness is defined for the purposes of this study to include spe- flects CDC’s changing priorities but is by no means representative of cialized agro-inputs (seed companies), farming (estates, plantations, agribusiness in the target countries more generally. smallholders, outgrowers), fishing and aquaculture, forestry (natural The data cover the following countries in Sub-Saharan Africa and and plantation), and primary processing linked to domestic raw ma- Southeast Asia and the Pacific by region and in chronological order ac- terial production. General manufacturing, such as processed-food cording to the year that CDC agribusiness investments began in each. products, rubber goods, and furniture is excluded. The main products and activities that were covered by the CDC sup- SUB-SAHARAN AFRICA 1948 Gambia, Malawi ported projects were: 1949 Swaziland, Nigeria, Tanganyika 1950 Botswana 1951 Seychelles Abattoirs Horticulture: cut flowers, fresh 1955 Kenya Aquaculture: prawns, tilapia vegetables, grapes, tree fruits 1959 Cameroon Livestock: beef, dairy, poultry Arable crops: maize, wheat, soya, rice, 1964 Uganda groundnuts, oil seeds Macadamia 1967 Zambia Bananas Mango 1973 Ethiopia Palm oil Cashew 1974 Cote d’Ivoire Pineapple Citrus 1978 Mauritius, Liberia, Ghana Cocoa Pyrethrum extract (an insecticide) 1984 Zimbabwe Coconut/oleo-chemicals Rattan 1992 Namibia 1996 Mozambique Rubber Coffee 1997 South Africa Seeds Cotton Sisal Fishing Sugar SOUTHEAST ASIA AND THE PACIFIC Flour (wheat) Tea 1948 Sabah Forestry: softwoods, hardwoods, pulp and paper , eucalyptus, gmelina, timber, Tomato paste 1949 Malaya wood chips Wattle extract (for leather tanning) 1961 Fiji 1967 Sarawak 1970 Solomon Islands 1976 Thailand, Indonesia, Papua New Guinea The investments are those made by CDC itself (or by wholly-owned, 1982 Philippines locally incorporated subsidiaries which were sometimes required 1983 Vanuatu 1984 Sri Lanka under local legislation). CDC also established and managed many 1996 Vietnam national and regional development finance companies and venture A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 16 C H A P T E R 4 — T H E B A S E D ATA capital funds, some of which made agribusiness investments, which 9. Pioneer: promoting a technology, crop or system new to are not included in the present study. the country or region 10. Sector: CDC, private, listed company, parastatal, govern- The information provided on the cost of CDC investments is in- ment, or joint venture (JV) dicative only. The information available from CDC annual reports is 11. Investment type: equity, loan or both a mixture of initial approvals, contractual commitments and actual 12. CDC management: whether or not CDC managed the venture for a significant period disbursements. Where different figures were available, the one se- 13. CDC investment (£m)—historic amount in Sterling lected is intended to best reflect the scope of what was planned 14. CDC investment (2011US$m)—amount adjusted (even if not fully implemented). It is important to note that CDC “in- for UK inflation and then converted to US$ at vestment” is not the same as “total capital costs.” Many projects had £1 = US$1.65 co-financiers or coinvestors, and capital costs may be met in part out 15. Technical Performance: physical productivity of self-generated funds, especially in the case of the older projects. 16. Development Impact: sustainable jobs and incomes 17. Project Financial Viability: solvency, eventual financial For each agribusiness investment the following information was self-sufficiency collected, where available. 18. Cause of Financial Failure: flawed concept, ineffective management or exogenous shocks beyond management 1. Main products control 2. Activities: input supply, estate/plantation, outgrower and 19. Equity Returns—profitability for shareholders settler services, processing, marketing, harvesting (fishing/ 20. CDC Investment Performance—whether CDC’s objectives logging) met 3. Scheme type: estate farming (and processor), outgrower 21. Success and Failure Factors: whether the natural resources, supply (and processor), Nucleus Estate and Smallholders chosen technology, market opportunity, management, (NES), independent processor, input supply, financier government actions or civil/military strife had a critical 4. Scale: small, medium, large, mega, as judged in impact on success or failure. relation to the norms for that industry 5. Intensity: intensive (labor, machinery, agricultural inputs) Finally, the data base includes a brief description of each project and or extensive use of land of its current status, where known. A fuller definition of each of the 6. Water supply: irrigation or rainfed above classifications is provided in Appendix 2: Data Classifications. 7. Market: export, local or both A summary listing of all the projects included in the data base is in 8. New or existing: start-up or minimal existing assets, reha- bilitation of a failing business, expansion of an existing, Appendix 3. successful enterprise INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 5 — DIREC T CHARAC TERISTICS AND PERFORMANCE OF THE CDC AGRIBUSINESS PORTFOLIO 17 Chapter 5: DIRECT CHARACTERISTICS AND PERFORMANCE OF THE CDC AGRIBUSINESS PORTFOLIO One hundred seventy-nine investments/projects14 are included in The principal crops and products are shown below: (duplication is the survey, of which 68 percent are in Africa and 32 percent in Asia. included, where a project has more than one principal crop). TABLE 5.1: Regional Focus TABLE 5.3: Enterprise Focus by Region AFRICA ASIA COMBINED AFRICA ASIA No. % No. % No. % Sugar 18 Oil palm 22 122 68 57 32 179 100 Tea 16 Cocoa 14 Cattle/meat 12 Rubber 9 Arable 10 Forestry & wood products 6 The most important countries in terms of number of projects are Rubber 10 Sugar 5 shown below. Forestry & wood products 8 Horticulture 7 TABLE 5.2: Country Focus Fish 7 AFRICA ASIA Tobacco 7 Kenya 15 Malaysia 17 Malawi 13 Indonesia 9 Zambia 13 Thailand 8 CDC’s investments in Asia are clearly dominated by three major Tanzania 12 Fiji 6 plantation crops: oil palm, cocoa, and rubber. In Africa the invest- Swaziland 11 Papua New Guinea 6 ments are more widely spread, although sugar and tea are promi- Nigeria 10 nent among them. Cote d’Ivoire 9 Zimbabwe 8 The split by scheme type is as follows. Total 91 Total 46 TABLE 5.4: Investment Type by Region Together, these 13 countries account for 77 percent of the projects AFRICA ASIA COMBINED in the survey, with the remaining 17 countries accounting for 23 NO. % NO. % NO. % percent. Estates/plantations 55 45 28 49 83 46 Nucleus Estate & Smallholders 21 17 18 31 39 22 Outgrowers 15 12 6 11 21 12 14 There is an arbitrary element in the number; some projects split (BAL Independent processing 23 19 5 9 28 15 and Mostyn Estates), some projects merged (Mpongwe, Munkumpu Input supply 5 4 5 3 and Mpongwe Milling), some integrated projects had components within separate legal entities, for example, AgroLines and Advance Agro, Finance 3 3 3 2 some companies developed diverse activities through a series of proj- Total 122 100 57 100 179 100 ects but all within one legal entity (Tanwat). The guiding principle has been to record projects separately if they appeared as such at any time Projects with some 106 87 50 88 156 87 processing in CDC’s accounts from 1948–2000. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 18 CHAPTER 5 — DIREC T CHARAC TERISTICS AND PERFORMANCEOF THE CDC AGRIBUSINESS PORTFOLIO FIGURE 5.1: Comparison between the Investment Portfolio Mix between Asia and Africa Africa Asia 3% Estates/plantations 9% 4% Nucleus estate & 10% smallholders 19% Outgrowers 45% 49% Independent processing 12% 32% Input supply 17% Finance Nearly half of the projects were based on estate production only, In Africa, 37 percent of relevant projects benefitted from total or (with or without processing) and in 34 percent explicitly involved partial irrigation of farm land, whereas there was only one conven- serviced/contracted outgrowers and/or settlers, with 22 percent tionally irrigated project amongst the Asian investments (and one broadly following the nucleus estate and outgrower/settler model. benefitting from the irrigation of paddy fields to grow eucalyptus on the adjacent bunds). This reflects both the overall drier climate of A large majority of projects explicitly involve some form of industrial much of Africa and CDC’s focus on tropical tree crops in Asia. processing, that is, 156 out of 179. Of the 156, only 28 did not have formal links to outgrower or estate production. FIGURE 5.2: Investment Size Size distribution of agribusiness investments On balance therefore, CDC’s agribusiness investments were orien- in Asia and Africa in percentage tated toward larger ventures in Asia than in Africa. This may reflect the 70 Africa smaller economies of many Sub-Saharan African countries, which lead 60 Asia to relatively small food projects catering to domestic markets. It may 50 46 42 40 37 also reflect the prevalence of large, “industrial” plantations in the Asian 30 23 agribusiness sector. 20 17 14 12 9 Eighty-six percent of relevant projects in Africa are considered ”in- 10 tensive” in the direct or indirect use of land, while for Asia the figure 0 Small Medium Large Mega is a similar 80 percent, giving an average of 84 percent across the FIGURE 5.3: Market Focus portfolio. Extensive land-use is defined as ranching, forestry, and ar- able cropping in low rainfall areas. Market focus of investments in Africa and Asia in percentage 70 Africa 61 TABLE 5.5: The Sizes, Judged Relative to Industry Norms 60 Asia AFRICA ASIA COMBINED 50 44 NO. % NO. % NO. % 40 Small 21 17 5 9 26 15 30 Medium 56 46 21 37 77 43 20 20 20 16 16 Large 28 23 24 42 52 29 11 12 10 Mega 17 14 7 12 24 13 0 Total 122 100 57 100 179 100 Export Export + local Local + export Local only INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 5 — DIREC T CHARAC TERISTICS AND PERFORMANCEOF THE CDC AGRIBUSINESS PORTFOLIO 19 TABLE 5.6: The Market Orientation of Projects—Local Market CDC has invested on its own or with private and public sector part- or Exports ners. It has also made sovereign loans. CDC always managed where AFRICA ASIA COMBINED it had a majority shareholding, and it also provided management NO. % NO. % NO. % services to some joint ventures and government schemes where Export 54 44 35 61 89 50 it did not have a controlling stake. The breakdown is shown below, Export + local 25 20 6 11 31 17 and is based primarily on how projects started: Local + export 19 16 9 16 28 16 Local only 24 20 7 12 31 17 TABLE 5.8: CDC Involvement in the Investment Total 122 100 57 100 179 100 AFRICA ASIA COMBINED NO. % NO. % NO. % CDC’s agribusiness investments have had a clear orientation toward CDC alone 26 21 4 7 30 17 exports. This is not surprising for foreign investors, especially during CDC-led JV 11 9 13 23 24 13 periods of exchange controls. Nevertheless, CDC made a substantial Private 18 15 4 7 22 12 effort to invest in local markets, especially in Africa where one-third Private-led JV 24 20 19 33 43 24 Plc – 2 4 2 1 of investments were wholly or predominantly designed to supply Parastatal or government 27 22 8 14 35 20 local markets. Parastatal-led JV 16 13 7 12 23 13 Total 122 100 57 100 179 100 TABLE 5.7: Investment Timing CDC management 61 50 22 39 83 46 AFRICA ASIA COMBINED NO. % NO. % NO. % CDC was therefore the main shareholder in 30 percent of the projects Greenfield and redevelop 71 58 38 66 109 61 in which it invested. Two-thirds of projects were in the private sector Rehabilitation and turnaround 14 12 6 11 20 11 (defined to include CDC itself ) and one-third in the public sector. Expansion of going concern 37 30 13 23 50 28 There was a slightly higher bias toward public sector investment in Total 122 100 57 100 179 100 Africa (35 percent versus 26 percent in Asia). CDC managed half of There is little difference between the two regions in terms of in- the projects it supported in Africa and 39 percent of those in Asia. vesting in new (green-field or redevelopment) ventures or existing CDC investment usually took the form of equity in, and/or loans ventures, whether rehabilitations of failing businesses or expansions directly to, the project or business entity, but it also made loans to of going concerns. For example, 58 percent of investments in Africa governments, or with government guarantees, for the financing and 66 percent in Asia were in new projects. Both in Africa and Asia of public sector projects or even for the financing of public sector around 36 percent of projects are classified as “pioneering.” participation in private-led joint ventures. The spread of invest- ment types is shown below. FIGURE 5.4: Investment Type TABLE 5.9: Equity and Loans Type of investment project as a percentage AFRICA ASIA COMBINED 70 66 61 NO. % NO. % NO. % 58 60 1 Equity only/direct project 21 17 13 23 34 19 50 Equity + loans 51 42 31 54 82 45 40 30 Loans only (direct to project) 24 20 8 14 32 18 28 30 23 Loans only (to government) 26 21 6 10 32 18 20 12 11 11 Total 122 100 572 100 179 100 10 1 In some cases, CDC did not form a separate legal entity, but financed unincorporated projects 0 directly. Africa Asia Combined 2 Actual total is 58, because one project incorporated both CDC equity and a loan to government to Greenfield & redevelop Rehabilitation & turnaround Expansion of going concern finance their equity stake. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 20 CHAPTER 5 — DIREC T CHARAC TERISTICS AND PERFORMANCEOF THE CDC AGRIBUSINESS PORTFOLIO A little over one-third of CDC’s investments were in the form of loans Technical Performance only, half of which was sovereign lending. CDC took an equity stake in 59 percent of the African projects and in 77 percent of the Asian ones. • Fail—had to be abandoned because resource or tech- nology or management unsuitable In only 19 percent of cases was CDC purely an equity investor, thus • Moderate Fail—productivity achieved just sufficient creating the scope for a divergence between project performance and for survival, but well below target CDC’s financial outcome. • Moderate Success—reasonable productivity achieved, but below planned levels TABLE 5.10: The Cost of CDC’s Agribusiness Investments in • Success—main productivity targets achieved and Africa and Asia broadly a competitive performance AFRICA ASIA COMBINED Direct Development Impact NO. % NO. % NO. % Narrowly defined as the direct impact on jobs and livelihoods Historic cost, £m 535 58 391 42 926 100 and the achievement of any other explicit economic goals 2011 equivalent, £m 2,349 66 1,226 34 3,576 100 such as alleviating foreign exchange shortages or contributing 2011 equivalent, US$m @1.65 3,876 2,024 5,900 to food production average investment, 2011 £m 19.3 21.5 20.0 average investment, 2011 US$m 31.8 35.5 33.0 • Fail—no sustainable incomes/jobs created • Moderate Fail—some worthwhile employment and CDC committed/invested a total of £926 million in agribusiness in income creation continues (either as a business or as viable smallholder production) but far less than planned the two regions15 over a period of 50 years (excluding indirect in- vestments via venture capital funds, and so on), of which 58 percent • Moderate Success—substantial, on-going develop- ment benefits, but less than planned was in Africa. However, UK inflation has averaged 5.5 percent per • Success—substantial commercial activity continues, year since 1948, rendering comparisons based on historical costs either as a business and/or as substantial smallholder production, equalling or exceeding expectations meaningless. Adjusting for inflation, using the British Consumer Price Index, the July 2011 equivalent of CDC’s investments is £3.58 Project Financial Viability billion, (US$5.9 billion) of which 66 percent was in Africa. The mean The establishment of a solvent, “going concern”, that is, finan- investment size in equivalent 2011 values was £20.0 million (US$33.0 cial sustainability. Broadly classified as: million), with investments in Asia being slightly larger on average. • Fail—Business collapsed and ceased trading TABLE 5.11: Investments which Exceeded US$100 Million in • Moderate Fail—A business survived as a going concern, but needed subsidization, for example, via refinancing 2011 Values by shareholders or via negotiated debt write-off or via AFRICA ASIA a sale as a going concern by a liquidator/receiver US$m US$m • Moderate Success—Self-sustaining business established 1949 Usutu: pulp 363 1967 SOP: oil palm 179 in line with expectations, but no significant profits 1950 SIS: sugar/citrus/cattle 246 1948 BAL: oil palm/rubber etc 142 • Success—Positive returns on all capital employed 1957 Mhlume: sugar 210 1970 SIPL: oil palm/cocoa 130 Equity Returns 1957 Camdev: rubber/oil palm etc 168 1976 HOPPL: oil palm/cocoa etc 104 Considered from the perspective of shareholders, where eq- 1987 Sable: arable/coffee etc 127 1976 ORRAF: smallholder rubber 101 uity was involved: 1949 Tanwat: wattle/tea etc 104 1960 KTDA: outgrower tea 100 • Fail—Loss of more than 25 percent of equity value HOPPL = Higaturu Oil Palm Property Ltd.; ORRAF = Office of Rubber • Moderate Fail—loss of equity value, but less than 25 Replanting Aid Fund; SIPL = Solomon Islands Plantation Ltd.; SIS = Sugar percent Industry of Singapore Ltd.; SOP = Sarawak Oil Palms. • Moderate Success—Some return on equity capital, but less than 12 percent IRR • Success—Annualized return of over 12 percent before 15 CDC also had agribusiness investments in Central and South America tax, allowing for dividends and equity sale or valuation and the Caribbean. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 5 — DIREC T CHARAC TERISTICS AND PERFORMANCEOF THE CDC AGRIBUSINESS PORTFOLIO 21 The classifications used for Technical Performance, Development while 29 percent were failures. The development impact success Impact, Financial Viability and Equity Returns are summarized be- rate is higher—71 percent were rated as successful or moderately low, and set out in more detail in Appendix 2. successful. The difference between success rates in terms of invest- ment performance and development impacts implies a number The schedule below analyses the performance of the projects/in- of cases in which worthwhile assets were created even if CDC saw vestments for these four different criteria: little financial benefit. While this may satisfy the criteria used by a development agency, it provides little if any incentive for a private TABLE 5.12: Performance Analysis investor. AFRICA ASIA COMBINED NO. % NO. % NO. % For those projects in which equity investment (by CDC or others) Technical performance was involved, only 16 percent were rated a success—based on es- • Fail 19 16 4 7 23 13 timated or actual compound return on equity of at least 12 percent • Moderate fail 12 10 7 12 19 11 per annum. A further 17 percent saw positive returns, even if mod- • Moderate success 29 24 15 27 44 25 est, so that in one-third of cases shareholders made some return on • Success 61 50 30 54 91 51 their investment while in two-thirds they incurred a partial or total Total 121 100 56 100 177 100 Development Impact loss. • Fail 26 21 6 11 32 18 Investments in Asia fared better on average than those in Africa. • Moderate fail 13 11 7 12 20 11 Positive equity returns (those classified as “successful” and “mod- • Moderate success 12 10 4 7 16 9 erately successful”) were achieved in 44 percent of investments in • Success 70 58 39 70 109 62 Asia compared to 26 percent in Africa. Judging by its projects over Total 121 100 56 100 177 100 Project Financial Viability a 50-year period, CDC found investing in agribusiness to be risky • Fail 37 31 9 16 46 26 everywhere. • Moderate fail 25 21 8 14 33 19 Unfortunately it is not possible, with the available data, to calculate • Moderate success 28 24 6 11 34 20 the profitability of CDC’s total or regional agribusiness portfolios or • Success 29 24 33 59 62 35 to compare the financial performance of the agribusiness portfolio Total 119 100 56 100 175 100 with CDC’s other sectoral investment portfolios. Equity Returns • Fail 53 63 23 49 76 58 CDC itself, after early losses, reported a profit in its accounts every • Moderate fail 9 11 3 7 12 9 year from 1955 to 1997.16 Given that some of CDC’s “star” agribusi- • Moderate success 11 13 11 23 22 17 ness equity investments yielded large capital gains (the SOP plan- • Success 11 13 10 21 21 16 tations in Sarawak were sold for the equivalent of US$115 million Total 84 100 47 100 131 100 in 1995; the BAL plantations in Sabah were sold in 1996 for US$165 CDC Investment Performance million, the Cavally rubber project in Ivory Coast was sold for US$40 • Fail 37 31 15 27 52 29 23 19 5 9 28 16 million in 2007) as well as substantial dividends, it is probable that • Moderate fail 14 12 10 18 24 14 the agribusiness portfolio yielded a positive return overall in mon- • Moderate success 46 38 26 46 72 41 etary terms although not necessarily in real terms (after adjusting • Success Total 120 100 56 100 176 100 for inflation). 16 The economic crisis in Asia in 1998 led CDC to make heavy provisions Forty-five percent of the projects were rated as successful or mod- against its loans to projects in Pakistan and Indonesia and to write down erately successful in terms of CDC’s own investment performance, the value of its equity investments in line with collapsing stock markets, creating an accounting loss of £28 million. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 6 — B R O A D E R D E V E LO P M E N TA L , E N V I R O N M E N TA L , A N D S O C I A L A S P E C T S 23 Chapter 6: BROADER DEVELOPMENTAL, ENVIRONMENTAL, AND SOCIAL ASPECTS While CDC always had a broad goal of supporting economic de- of ever more effective nongovernmental organizations (NGOs) that velopment in host countries by means of commercially sound pro- challenged the status quo. jects, it did not initially seek to demonstrate this in a systematic way. During the 1990s CDC developed manuals and guidelines for both As a consequence it is not possible to analyse CDC’s agribusiness the initial appraisal and subsequent monitoring and reporting of portfolio in terms of its broader developmental, environmental and the ethical, environmental, health and safety and social aspects of social impacts through its Annual Reports, although there are ad projects and sought to avoid any substantial negative impacts— hoc references to the direct and indirect benefits that CDC invest- both for their own sake and to avoid adverse publicity. Formal poli- ments were having on local economies. This section will therefore cies were adopted defining “best practice” standards to be adhered deal with CDC’s broader impact through examples, rather than an to by projects controlled by CDC and to be recommended to other overall portfolio analysis and assessment.17 project sponsors. In 1996 a Development Committee of CDC Board EVOLVING POLICIES AND PRACTICES members was created to review development performance and For many years CDC believed its own profitability and a sound oversee the production of a regular Development Report. balance sheet was sufficient demonstration of its effectiveness in POTENTIAL “NO GO” AREAS supporting worthwhile economic development, given the location In the 1940s and 1950s, consistent with the ethos of the times, CDC of its projects in lower income developing countries and their com- readily promoted and supported the exploitation of natural resourc- mon focus on natural resources. es without too much concern for sustainability or ecological conse- By the early 1980s it had accepted that, in principle, it would be pos- quences, for example, commercial fishing on Lake Malawi, around sible for a project to be financially viable but economically unsound the Seychelles and in the Western Atlantic and logging of the natural if it benefitted from excessive subsidies or tariff protection, and CDC forest in Nigeria. Most of these ventures failed because not only had began to calculate forecast economic rates of return for new invest- CDC not assessed the environmental risks it had not adequately re- ments, as well as financial rates of return. searched the commercial viability of the resource either. In 1986, under pressure from the British government it established an CDC was also ready to support sectors which carried a substantial Evaluation Department to undertake retrospective reviews of the perfor- pollution risk such as pulp and paper in Swaziland, prawn farming in mance of projects—including some notion of their “development value.” Thailand and horticulture in Kenya and Zambia, and to participate in (but not necessarily lead) the gradual tightening of acceptable CDC was also adjusting its operations to the realities of a more effluent and agro-chemical management standards. democratic, transparent world with host governments moving From the beginning, CDC supported tobacco growing projects, from one-party states to multi-party democracies and with the rise especially in Malawi where it developed a large-scale, success- 17 It will not consider the potential adverse economic and social conse- ful smallholder settlement scheme—the Kasungu Flue Cured quences that apply to almost all forms of economic progress, such as Tobacco Authority. Even after the health risks had been clearly rising disposable incomes contributing to the spread of AIDS and al- cohol abuse, which are equally a feature of projects promoted by CDC. established CDC’s position was that poor African farmers should A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 24 C H A P T E R 6 — B R O A D E R D E V E LO P M E N TA L , E N V I R O N M E N TA L , A N D S O C I A L A S P E C T S not be denied support to compete for a place in world markets but moribund estates (BAL plantations) or the purchase of private where they faced subsidized competition from countries such land used for relatively low-value ranching, for conversion to more as the United States or Italy. Nevertheless in the early 1990s CDC intense utilization (the Swaziland Irrigation Scheme, Kaleya small- was directed by the British government, on ethical grounds, not holder sugar in Zambia) or because the land was largely unsettled to make any new investments in the sector and to withdraw from (Mpongwe in Zambia). existing tobacco projects at the earliest practical opportunity. In situations where the continued ownership of large land areas by a ACQUISITION OF LAND foreign entity did become a political issue, CDC negotiated a sale to local interests (for example, listing of SOP on the Kuala Lumpur stock Most of CDC’s large land acquisitions were not controversial at market) or the conversion of free-hold title into a long-lease from the the time, either because they involved the purchase of existing, state (for example, Usutu Pulp in Swaziland). SACRIFICING DEVELOPMENT TO AVOID PUBLIC CONTROVERSY EXPLOITATION OF ECOLOGICALLY SENSITIVE AREAS One of the relatively few occasions when the land required for Several of the projects supported by CDC are in areas that would an estate was already owned by smallholders was the pioneer- today be considered ecologically sensitive—for example, oil ing NDC/Guthrie oil palm plantation in remote Agusan Del Sur Province of Mindanao in the Philippines, which began in palm, rubber, and cocoa in tropical rain forest; prawn farming in 1981. It was to be only the second oil palm plantation in the mangrove swamps. As these habitats have become more scarce Philippines and at 8,000 ha by far the largest, aiming to stem a rising tide of palm oil imports. KILOMBERO VALLEY TEAK COMPANY, TANZANIA CDC’s participation as a lender attracted high profile criticism in the UK—demonstrations outside its London Head Office, a CDC obtained a lease of 28,000 ha in the Kilombero Valley in TV documentary, questions in Parliament. 1992 to develop a teak plantation at an expected cost of US$25 Under the Comprehensive Agrarian Reform Law it was not million in 2011 values. possible for a foreign JV to own large blocks of land. The proj- This was to be the first large-scale private teak plantation in ect therefore negotiated to lease underutilized land from a Africa and a vote of confidence in Tanzania’s stability, as the large number of smallholders, who had been allocated land first significant revenues from the plantation would not be under the reforms but generally lacked the capital and training achieved until 2009, with the commissioning of a sawmill to to use it. There were allegations that a local militia had been process the first commercial thinnings. used by the company to intimidate smallholders into signing In recognition of the growing sensitivities around this kind of the leases, and this cause was taken up by the Roman Catholic development, there was extensive consultation with local villag- Church, and became part of a wider campaign against alleged ers as well as with the government and environmental groups. abuses under the dictatorship of President Marcos. The final plan allowed for just one-third of the site to be As a result of the bad publicity, CDC abandoned a similar oil palm planted to teak in a mosaic pattern amongst the indigenous project in the Province at Loreto that it had been developing forest which would be preserved and protected and with pro- jointly with NDC and for which it had completed feasibility studies vision for wild life corridors (the site is close to the Selous Game and had begun negotiations with the smallholders to be affected, Reserve). It was also agreed to support a parallel outgrower and which would have included a major outgrower scheme. teak planting project. CDC undertook an evaluation of the NDC/Guthrie project in Implementation and operations were formally monitored 1987, including an interview with the priest who had been by the University of Dar es Salaam and the Society for central to the allegations of coercion. He acknowledged that, Environmental Exploration and the Forest Stewardship Council. in spite of his initial concerns, Guthrie had proved to be re- sponsible project managers and the combination of land rent- One significant drawback identified during monitoring was al income and employment opportunities had delivered a ma- that the improvement of road access needed for the project jor improvement in living standards in a remote and poor area. had facilitated increased general access to the area. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 6 — B R O A D E R D E V E LO P M E N TA L , E N V I R O N M E N TA L , A N D S O C I A L A S P E C T S 25 and their economic value better understood and as pro-conser- SMALLHOLDERS AND OUTGROWERS: SPREADING vation organizations have become better organized, so pressure THE BENEFITS OR EVADING RESPONSIBILITIES? mounted on CDC either to avoid sensitive sites completely or to It is conventional to interpret smallholder and outgrower conduct full Environmental Impact Assessments and to develop schemes as a way of spreading the benefits from a core estate integrated utilization, conservation and preservation plans from and/or agro-processing operation to the local people. the outset. When the state-owned Zambia Sugar Company (ZSC) expand- ed its factory capacity in the late 1970s it was politically difficult to acquire more land for estate sugarcane planting. Instead, it EMPLOYMENT CONDITIONS worked with CDC and the government to establish a small- In agribusiness projects promoted by CDC employment conditions holder settlement scheme—the Kaleya Smallholder Company (KSC). (wages, housing, medical facilities, and so on) were generally higher than local norms. There would inevitably be regular, difficult nego- In 1980 the government compulsorily acquired a nearby pri- vate ranch (and in accordance with Zambian law, paid com- tiations with employees’ representatives and periodic strikes, as with pensation only for the improvements to the land, not for most commercial ventures, but there was rarely any political contro- the land itself ) and CDC and the African Development Bank versy, except when the UK press occasionally compared conditions financed the development of 1,800 ha of irrigated cane on with standards of living in the UK. which over 140 families were settled, including, unusually for the times, female-headed households. The weakness in CDC’s approach is that it did not normally look The scheme was a technical success, with average cane yields beyond the project’s boundaries. As a result in some cases, such exceeding those of the ZSC estates, and the smallholders as the Mpongwe arable project in Zambia, “shanty towns” devel- earned relatively high incomes. As a result, some settlers made oped close to projects consisting of those seeking casual work use of informal, hired labor to do much of the work in the fields. Rates of pay for these informal workers were low and they had or waiting for a chance to apply for one of the privileged perma- no access to any of the benefits and protection provided for nent jobs on the project. Unless regulated by the local authori- under national Labour Laws or under ZSC or KSC company ties, such informal settlements could be unhealthy and danger- policies. Some lived in informal settlements on the edge of the ous both for informal settlers and the nearby project employees, scheme. for example, outbreaks of cholera and malaria. During the late Had the extra land been cultivated as an estate, those regu- 1990s the fashion in business for contracting out noncore ac- larly working the land would have been entitled to company housing, medical benefits, social security contributions and tivities, and the adoption internally of demanding housing and the protection of labor officers and trade unions. Indeed some social welfare guidelines for employees, led CDC to reduce per- members of the government were reported to have opposed manent employee numbers where feasible and make greater the smallholder scheme from the beginning, arguing that it use of sub-contracted and casual labor. would be better to provide jobs for state-sector employees than to create a privileged “kulak” class of self-employed settlers. SOCIAL ENGINEERING AND SOCIAL IMPACT Some projects supported by CDC had an explicit “social engineering” In the absence of any base-line surveys and understanding of component. The massive NES schemes in Indonesia, supported by social conditions, including gender relationships, economic ini- both CDC and the World Bank, were part of the country’s transmigrasi tiatives can have unintended social consequences. Traditionally programme of resettling farmers from overpopulated Java to the (but not universally) in much of Africa subsistence food crops are outer islands such as Sumatera and Kalimantan. While a plausible eco- primarily grown by women, as it is considered to be their respon- nomic case could be made for the programme, it was also a means of sibility to feed the family, whereas cash crops—even when other securing increased political control of the outer islands by the central family members contribute to the work—serve to provide the government and has created significant ethnic and cultural tensions. male household head with a cash income. It has been claimed A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 26 C H A P T E R 6 — B R O A D E R D E V E LO P M E N TA L , E N V I R O N M E N TA L , A N D S O C I A L A S P E C T S therefore that the promotion of such outgrower crops as tea Smallholders and Nanga Farms all reduced the flow of water (KTDA) and sugarcane (Mumias) in Kenya by CDC and other through the Kafue Gorge hydro electric station in the country development agencies lead to increasing malnutrition as land is and the Cabora Bassa hydro station in downstream Mozambique, diverted from food crops to cash crops but little of the extra cash reducing the net economic benefit from these important food earned is used to help feed the family. production projects. In the 1960s and 1970s India and Sri Lanka objected to World Bank ECONOMIC EXTERNALITIES and CDC support for expanding tea production in East Africa, such Finally, CDC generally did not always take into consideration as the KTDA, arguing that this undermined their attempts to stabi- major economic externalities of its projects. Thus in Zambia, lize world tea prices through an International Tea Agreement and the promotion of irrigation at Mpongwe, Zambia Sugar, Kaleya export taxes on their own producers. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 7 — E X P L A I N I N G F I N A N C I A L FA I LU R E 27 Chapter 7: EXPLAINING FINANCIAL FAILURE Among those projects classified as financial “failures” or “moderate TABLE 7.1: Projects Classified as Financial Failures or Moderate failures” a summary assessment was used to determine whether Financial Failures these results were primarily the result of bad luck (exogenous fac- AFRICA ASIA COMBINED NO. % NO. % NO. % tors), bad management, or whether the project was “fatally flawed” Fatally flawed concept 35 56 13 76 48 61 in its concept (for instance by ill-conceived financial plan, mistaken Bad luck 16 26 1 6 17 21 assumptions about costs and revenues, or poor location). Projects Bad management 11 18 3 18 14 18 flawed in concept were generally never going to work, however ef- Total 62 100 17 100 79 100 fective the implementation management team. % of all 50% 32% 45% FIGURE 7.1: Percentage of Projects that Succeeded and Failed with Reasons for Failure Percentage of projects that succeeded and failed in Detailed breakdown of the causes of Africa and Asia − with failures categorized by cause failure, expressed as a percentage 100 Success 90 Bad management 100% 80 50 55 Bad luck Bad management 70 68 Fatally flawed 80% Market issues concept Civil unrest 60 Government policies 60% 50 9 8 40 13 10 40% Flawed concept 6 30 2 20 20% 28 25 27 Unknowable flawed 10 concept 0% % 0 Africa Asia Combined A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 28 C H A P T E R 7 — E X P L A I N I N G F I N A N C I A L FA I LU R E While there is inevitably a subjective element in this classification, consistent across both regions. The main reason for the higher it does suggest that in only a small minority of cases can financial failure rate in Africa appears to have been issues characterized failure be attributed to the performance of the management teams as bad luck, principally adverse government policies and civil responsible for implementing and operating projects. This is not unrest. surprising, given that weak managers can be changed and that Of the 60 percent of failed projects that had a flawed concept, operational problems can be sorted out. about two thirds could have been detected at the approval Bad luck was a significant factor in Africa in particular, where civil stage and so could have been controlled by the investor. In strife and nationalization took place in a number of countries. Of the some cases CDC investment analysts and technical specialists 17 cases of “bad luck,” eight were the result of adverse government expressed doubts internally but there was “political” pressure economic policies, seven were the result of civil war, and two were to support projects which were a high priority for either the due to the collapse of export markets. British or the host government or which were seen as strategic for CDC. Host country government decisions about agribusi- By far the most important cause of failure was some aspect of the ness investment projects inevitably take into account more fac- project concept from the start. This means that the planners of the tors than simple internal rate of return calculations. During the projects—be they CDC managers, government agencies or private 1970s for instance, Zambia had a policy of promoting a state sponsors—made major errors of judgement regarding technical, farm and ranch in every province irrespective of agronomic economic, or financial matters. and logistical suitability, on the ground of fairness and national The percentage of projects which failed as the result of a cohesion. flawed concept or because of weak management was broadly INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 8 — E X P L A I N I N G D E V E LO P M E N T FA I LU R E 29 Chapter 8: EXPLAINING DEVELOPMENT FAILURE Among the projects in Asia there were six development failures. In of them. Only one was located on an unsuitable site. Another was three cases the basic natural resources (soils, climate) proved unsuit- affected by low export prices. Two failed as the result of govern- able for commercially viability, and in the other three the technol- ment-related factors. One of these was the result of the revolution in ogy and the management were not adequate. Ethiopia. The other resulted from the withdrawal of export taxes on raw cashew nuts in Mozambique, which led to the collapse of the There were also seven projects in Asia classified as moderate fail- domestic processing industry. ures in terms of development impact. Four of these outcomes were attributable to unsuitable natural resources combined An additional 13 projects in Africa were classified as moderate de- with their inability to compete in highly competitive markets. velopment failures. One of these dates to 1950 and again reflects Two were attributable to inadequate management; and one in CDC’s over-ambition at the time. Five were badly affected by “gov- Papua New Guinea was the result of civil unrest, which led to ernment”—three as the result of the military coups in Uganda and large planted areas being abandoned. Liberia; and two the result of mismanaged state-owned enterprises in Tanzania and Nigeria. Poor natural resources were a major factor By contrast, in Africa there were 26 development impact failures. in three African projects. Weak management also played a major However 10 of these were projects promoted by CDC between 1948 role in three projects, and four were fundamentally not competitive and 1951. These 10 projects reflected CDC’s own naivety at start-up, in the markets they aimed to supply. before it gained experience, and when its decision makers were clearly over-optimistic about the prospects for success in almost any Overall it is clear that the majority of failure must be attributed to setting. In 6 of these 10 early failures the site chosen was unsuitable human error. These may be the result of inadequate or misguided for the intended venture. In the other four, CDC was simply out of its planning, for instance locating projects in sub-optimal agro-climatic depth in terms of management experience and technical know-how. zones, and/or ineffective management during the implementation of the project. Of the 16 development failures that date from 1967 onwards, in- adequate management and technology was a major factor in 13 A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 9 — E Q U I T Y S U C C E S S S TO R I E S 31 Chapter 9: EQUITY SUCCESS STORIES Ten investments in Asia were classified as equity successes, mean- higher prices as an import substitution crop. In addition, former ing that they generated financial returns to shareholders in excess colonies lobbied for and received a share of the UK and then the of about 12 percent per annum. All 10 were oil palm projects. EU preferential import quotas for sugar. Prices were set at levels Southeast Asia and the Pacific enjoyed the good fortune to par- that kept high cost producers in the Caribbean and the EU in busi- ticipate in the post-Second World War palm oil boom. A parallel ness, and were very attractive for the much more efficient, modern can be drawn between the transfer of oil palm from West Africa African producers. to Southeast Asia during the twentieth century, and the earlier The four agro-processing projects benefitted from a good, early transfer of rubber from South America to Southeast Asia during the supply of raw materials and achieved high capacity working quickly. nineteenth century. Both tree crops came to flourish in their new This early profitability helped to yield high compound rates of re- region. Palm oil is cheaper to produce than any other vegetable oil turn on investment. but requires substantial, long-term investment. It takes a long time to develop new production capacity (compared to annual crops). The Usutu wood pulp project in Swaziland is simply a “world class” Palm oil producers have enjoyed good profits as their product has venture: good growing conditions, economies of scale from a large gradually come to account for an ever-increasing share of the global factory, a short rail link to the local seaport, and access to world mar- fats and oils market where, in broad terms, prices were dictated by kets. CDC was able to sell its shareholding as a strategic acquisition the cost of production of more expensive annual oilseed crops. by a global pulp and paper company. Eleven investments in Africa were classified as equity successes. The Cavally rubber project in Cote d’Ivoire combined a nucleus estate and outgrowers with a processing factory. The agro-climatic  4—sugar conditions are excellent and CDC was able to sell its controlling  4—agro-processing stake during the recent rubber commodity price boom.  1—wood pulp  1—rubber The Munkumpu arable crops project in Zambia involved the reha-  1—arable bilitation of an irrigated parastatal wheat and soya scheme. When acquired in a privatization process, CDC was able to achieve a very Like oil palm and rubber in Southeast Asia, sugarcane in Africa is quick turnaround to full capacity and high yielding operations. (CDC a nonnative plant. Sucrose yields respond well to long dry sea- was already managing the adjacent Mpongwe project). CDC’s re- sons which provide many hours of sunshine and cool night tem- turns therefore benefitted from the high sunk-cost of the develop- peratures—characteristic of upland locations in East, Central, and ment incurred by the parastatal which CDC acquired at a substantial Southern Africa—provided there is irrigation. Until the early 1960s, discount to replacement cost. when newly independent countries such as Zambia and Tanzania sought to promote domestic sugar production rather than relying Unsurprisingly, all of the “successful” equity investments benefit- on imports, it was not widely grown on a commercial basis out- ted from good market opportunities and from excellent growing side of the Republic of South Africa. Sugar benefitted initially from conditions. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R CHAPTER 10 — CHANGING PERFORMANCE OVER TIME 33 Chapter 10: CHANGING PERFORMANCE OVER TIME Management attitude and changing strategy had a large impact during this period. From the mid-1960s, the remit of the CDC was on the success rate of investments. At its inception in the late shifted to pursuing a stronger development agenda.18 The num- 1940s the CDC’s management did not have well developed in- ber of investments increased significantly, more than doubling ternal systems for assessing potential investment projects and a between 1970 and 2000 (compared to the period from 1947 “can-do” Executive Chairman inspired a more general tendency to the late 1960s). From the mid-1980s onwards, CDC increas- toward over-confidence. This changed in the 1950s when a ingly focused on private sector projects.19 Investments more fre- new senior management team brought in much more rigorous quently backed indigenous entrepreneurs which contributed to project assessments and took a more conservative attitude to a lower average size of investment. Overall, the financial viability investments that focused more on commercial success. The investment success rate diminished somewhat during the 1970s overall performance of the equity portfolio increased markedly and 1980s. FIGURE 10.1: Financial Viability over Time Financial viability 100% 13 90% 18 18 23 31 80% 13 70% 24 69 29 18 Fails % 60% 26 17 Modest Fail % 50% 18 18 Mod Success % 40% 21 17 Success % 30% 0 48 20% 15 41 38 29 31 10% 15 0% 40s 50s 60s 70s 80s 90s 18 For example, in 1965 the British government offered CDC loans with an interest free grace period of 7 years for investing in agriculture; in 1975 CDC and the British government agreed that CDC should invest pre- dominantly in poor countries and in the Renewable Natural Resources sector. 19 In 1985, the British government explicitly requested CDC to work more with private sector partners and in 1993 set a formal target that at least 80 percent of new investment should be in private sector projects. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 1 1 — S U B - S A H A R A N A F R I C A V E R S U S S O U T H E A S T A S I A A N D T H E PA C I F I C 35 Chapter 11: SUB-SAHARAN AFRICA VERSUS SOUTHEAST ASIA AND THE PACIFIC CDC’s experience investing in agribusiness has been comparatively Although African projects were overall less successful than Asian similar in Sub-Saharan Africa and Southeast Asia and the Pacific ones, the most significant difference was between generating sen- since 1948, which is perhaps surprising given the disparity of de- sible equity returns. Only 26 percent were classified as “success” or velopment progress in the two regions over the same period. This “moderate success” in Africa compared with 44 percent in Asia. Yet 48 similarity becomes even more striking when the CDC’s 10 project percent of African investments ultimately achieved long-term finan- failures in Africa between 1948 and 1951 are omitted. cial viability and 70 percent delivered long term economic benefits. Many of these long term benefits reflect enterprises which CDC de- In part this similarity is because, as an investor rather than an aid veloped and which achieved positive cash flows. They nevertheless agency, CDC only operates where it believes it has a reasonable went on to be sold to new owners at a discount to CDC’s capital cost chance of success, and will suspend operations if the economic and either because earnings were low or the price/earnings ratio was political environment becomes too inhospitable. Liberia, Ghana, low owing to perceived high country and/or sector risks. Examples Nigeria, Ethiopia and Uganda for example would have all figured include eucalyptus plantations in Swaziland (Shiselweni), tea estates more prominently in CDC’s project portfolio had they not gone in Tanzania (Euteco and Tanwat), rubber estates in Nigeria (Illushin), through periods of military coups and economic war on foreign and mixed tobacco/arable/coffee estates in Malawi (Sable/Kawalazi investors. farming group). There were also examples of private sector projects Both regions had their “boom” crops—oil palm and cocoa in in which CDC’s loans were repaid by parent companies to avoid Southeast Asia, sugar and tea in East, Central and Southern Africa. insolvency of a subsidiary. Although these projects had achieved Both regions have also had their share of difficulties. CDC manag- positive cash flows, the cash flows were not sufficient to service the ers were murdered by communist insurgents in Malaya in the 1950s debt, for example, Sugar Corporation of Uganda Limited (SCOUL) and by violent strikers in Papua New Guinea in the 1980s. CDC and rubber in Malawi (Vizara). investments in Sabah and Sarawak were threatened by Indonesia’s To have a chance of success, CDC’s experience in both regions confrontasi policy of the 1960s which led to an outflow of essential demonstrated the advantages of a diversified portfolio (by country, Indonesian migrant workers. Opportunities in Fiji were curtailed by product, and market) and a long-term perspective—holding on military coups. Estates in Papua New Guinea and the Solomon Islands during the bad-times, only quitting in extremis. were over-run by separatist insurgents and had to be abandoned. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 1 2 — E S TAT E V E R S U S O U TG R O W E R S 37 Chapter 12: ESTATE VERSUS OUTGROWERS A crude analysis has been undertaken of success and failure rates TABLE 12.1: Percentage of Projects Classified as Success for different types of scheme, that is, estate, NES, outgrowers and or Moderate Success independent processors. AFRICA ASIA COMBINED DEV FIN DEV FIN DEV FIN Nucleus estates with smallholders/outgrowers (NES) provided the Estate 63 42 69 55 65 47 most successful business model, but only for a limited range of indus- NES 86 66 88 88 87 76 trial crops (oil palm, sugar, tea, rubber), followed by processing. Pure Outgrowers 53 33 83 83 62 48 Processor 65 52 75 75 67 56 outgrower schemes were broadly about as successful as estate farm- DEV = Developmental; FIN = Financial. ing operations. Outgrower schemes worked particularly well in Asia. FIGURE 12.1: Sustainable Development Impact and Equity Returns Percentage of Asian agribusiness Percentage of Asian agribusiness investments that ultimately delivered investments that generated sensible sensible sustainable development impact equity returns for investors 70 70 70 60 60 50 49 50 40 40 30 30 23 21 20 20 11 12 10 7 10 7 0 0 Fail Moderate fail Moderate success Success Fail Moderate fail Moderate success Success Percentage of African agribusiness Percentage of African agribusiness investments that ultimately delivered investments that generated sensible equity delivered sustainable development impact returns for investors 70 70 63 60 58 60 50 50 40 40 30 30 21 20 20 11 11 13 13 10 10 10 0 0 Fail Moderate fail Moderate success Success Fail Moderate fail Moderate success Success (Continued ) A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 38 C H A P T E R 1 2 — E S TAT E V E R S U S O U TG R O W E R S FIGURE 12.1: Sustainable Development Impact and Equity Returns (Continued ) Percentage of African & Asian agribusiness Percentage of African & Asian agribusiness investments that ultimately delivered investments that generated sensible sensible delivered sustainable development impact 70 equity returns for investors 70 62 58 60 60 50 50 40 40 30 30 20 18 20 17 16 11 9 9 10 10 0 0 Fail Moderate fail Moderate success Success Fail Moderate fail Moderate success Success No attempt has been made to identify correlations with other were keen to see CDC invest in inclusive business models which possible explanatory factors. For instance, are NES schemes more incorporated outgrowers. Both because of CDC experience and successful because they are, on average, larger? Or are they suc- the need not to expose smallholders to the high risks of untested cessful because they are more likely to involve crops with good enterprises, these NES were mostly built upon existing successful markets, such as palm oil and sugar? It would probably be wrong estate farming operations. To some extent these projects were to infer that NES are intrinsically less risky than large scale farming therefore self-selecting. Nevertheless CDC managers found that operations. During the 1960s and 1970s, a number of countries NES projects outperformed expectations. FIGURE 12.2: Financial and Development Performance of Estate Farming, Nucleus Farms, Outgrower Schemes, and Independent Processing Operations: Percentage Classified as Successful or Moderately Successful Development impact: 100 Africa 100 Asia Moderate success Success 80 14.3 80 60 8.7 60 14.3 9.1 6.7 40 88.9 83.3 40 71.4 52.7 56.5 53.6 60.0 20 46.7 20 0 0 Estate NES Outgrowers Processors Estate NES Outgrowers Processors Financial viability: 100 Africa 100 Asia Moderate success Success 80 11.1 80 60 23.8 60 4.3 14.3 40 40 77.8 83.3 60.0 27.3 47.6 13.3 20 43.5 20 39.3 20.0 12.7 0 0 Estate NES Outgrowers Processors Estate NES Outgrowers Processors INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 1 3 — S E T T L E R S , S M A L L H O L D E R S , A N D O U TG R O W E R S 39 Chapter 13: SETTLERS, SMALLHOLDERS, AND OUTGROWERS CDC has been involved with many commercially and developmen- Some crops such as rubber, cotton, oil seeds, and coffee have been tally successful projects involving settlers relocated to new land and somewhat less attractive financially for much of the 50 years cov- smallholders farming their own existing land; both of whom may be ered. Schemes involving the production of these crops have not organized as outgrowers supplying a central processing unit (with or done very well, particularly where operations were based in sub- without its own nucleus estate). optimal growing conditions. Where these schemes have been based on financially attractive crops There are well established reasons why some crops are better-suited such as oil palm, cocoa, sugar, and tea, they have generally done well, than others to the NES and processor/contracted outgrower mod- because the outgrowers have a strong incentive to participate and co- els. Those which are better-suited tend to require industrial process- operate. Many such schemes such as Kenya Tea Development Agency ing of a bulky raw material relatively soon after harvesting, including (KTDA), Higaturu Oil Palm Property Ltd. (HOPPL), and the NES schemes sugarcane, green-leaf tea and oil palm fresh fruit bunches. Other in Indonesia were co-financed with the World Bank. In addition the op- crops such as cotton, coffee, cashew, and cereals, the raw material of erators of the “nucleus” normally have high fixed costs and high gross which can be more easily stored, have a variety of low-tech process- margins on factory processing, and therefore have strong incentives to ing options, and afford producers opportunities for “side-selling.” support outgrowers so that the capacity of the factories is highly utilized. These tend to be less well-suited for NES and outgrower models. TABLE 13.1: Causes of Project Failure COUNTRY PROJECT TYPE CROP FAILURE FACTORS Zambia ZCCL factory, estate, smallholders cashew wrong technology; inexperienced management Changanda/Family Farms/Mukonchi estate, settlers tobacco low margins (high labor costs, low prices), high overheads Nigeria Niger Agric Proj settlers mixed arable low yields, low margins Kenya Oil Crop Development outgrowers oil seeds low prices/margins; side-selling Liberia LRDC factory, outgrowers rubber coups and civil war Uganda UTGC factory, outgrowers tea coups and civil war Mozambique Agrimo factory, outgrowers cotton overly-optimistic planning; rehab costs higher than planned Malawi SCA factory, outgrowers coffee small areas suitable for coffee, low production, high overheads Tanzania Kilombero factory, estate, settlers sugar collapse following nationalization and economic mismanagement Philippines Bukidnon factory, outgrowers tomato paste overly-optimistic planning, inexperienced farmers & management; prices paid provided no incentive, side-selling Vanuatu Tana Coffee factory, estate, outgrowers coffee agro-climate unsuitable for good yields, small scale LRDC = Liberia Reconstruction and Development Committee; SCA = Smallholder Coffee Authority; UTGC = Uganda Tea Growers Corporation; ZCCL = Zambia Coffee Company Limited. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 40 C H A P T E R 1 3 — S E T T L E R S , S M A L L H O L D E R S , A N D O U TG R O W E R S Moreover, some crops favor smallholders, such as those which are A full analysis of the advantages and disadvantages of different labor intensive (and can utilize low cost family labor and informally types of outgrower schemes is beyond the scope of this study. hired labor) and which offer few economies of scale. Crops which Table 13.1 on page 39 lists the main failures and moderate failures in involve capital intensive technologies tend to favor large-scale terms of project financial viability in which CDC participated and a operations. summary of why they failed. INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS CHAPTER 14 — SIZE AND SUCCESS 41 Chapter 14: SIZE AND SUCCESS Overall, the analysis suggests that larger projects (large relative to the rate of financial success compared with the size of CDC’s invest- the norms for that sector) supported by CDC were more successful, ment measured in US dollars at July 2011 price levels. especially in financial terms, than smaller ones. There were in par- In fact the rate of financial success in the CDC portfolio appears ticular a large number of financial failures of small projects in Africa, unrelated to the size of investment, except for the very large ones— where 18 of 20 of such investments were rated as financial failures over US$50 million, where success rates are clearly higher, perhaps or as moderate financial failures. in some cases because CDC was rewarding early success with fur- Many sectors exhibit some economies of scale. Relatively ther finance for expansion. smaller projects in these sectors are less likely to be competitive TABLE 14.1: Percentage of Projects Classified as Success unless sheltered from competition. In addition larger projects or Modest Success can generally afford more and better quality management, and AFRICA ASIA COMBINED will attract more head-office attention when things start to go DEV FIN DEV FIN DEV FIN wrong. Small 43 10 60 60 46 20 Medium 77 48 60 40 72 45 It is moreover likely that CDC was more willing to take higher risks Large 61 57 88 88 73 71 with smaller projects because they had less potential to adversely Mega 82 76 100 100 88 83 impact CDC’s balance sheet. An assessment was therefore made of DEV = Developmental; FIN = Financial. TABLE 14.2: Projects Rated as Financial Success or Moderate Financial Success, by Size of CDC Investment1 AFRICA ASIA COMBINED % FIN %FIN %FIN NO. SUCCESS NO. SUCCESS NO. SUCCESS <$5m 30 39 8 43 38 40 >$5m <$15m 28 46 13 69 41 54 >$15m <$30m 25 33 12 75 37 47 >$15m <$50m 17 47 11 55 28 50 >$50m <$100m 15 73 8 88 23 78 >$100m 7 86 5 100 12 92 Total 122 57 179 1 Percentages shown are for the projects for which the financial outcome was known. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 1 5 — S TA R T - U P V E R S U S E X PA N S I O N 43 Chapter 15: START-UP VERSUS EXPANSION As expected, building upon success by expanding existing busi- TABLE 15.1: Percentage of Projects Classified as Success nesses was more likely to deliver higher developmental (92 percent) or Moderate Success and financial (69 percent) success rates than starting from scratch or AFRICA ASIA COMBINED DEV FIN DEV FIN DEV FIN converting moribund existing assets to a different use (61 percent Greenfield and 56 45 70 62 61 51 and 51 percent respectively). redevelop Rehabilitation and 79 29 67 67 75 40 Rehabilitations and turnarounds achieved a reasonable rate of de- turnaround velopment performance but the financial results appear poor. Expansion of going 86 61 100 92 92 69 concern DEV = Developmental; FIN = Financial. FIGURE 15.1: Development Impact: Percentage Classified as Success or Moderate Success Africa Asia 100 100 Mod Success 8 80 80 7 21 11 60 60 100 40 40 73 67 57 58 20 49 20 0 0 New Rehab Existing New Rehab Existing Financial viability: Percentage classified as success or moderate success Africa 100 Asia 100 Mod Success 80 Success 80 60 60 11 19 33 92 40 40 25 50 20 21 41 20 33 18 0 7 0 New Rehab Existing New Rehab Existing A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 44 C H A P T E R 1 5 — S TA R T - U P V E R S U S E X PA N S I O N FIGURE 15.2: Percentage of All Projects Classified as Success The especially low financial success rate of rehabilitation projects in or Moderate Success Africa is perhaps surprising. In broad terms, attempts to turn around 100 Development Financial private sector businesses that were struggling were likely to fail 80 92 because fundamental weaknesses persisted. Attempts to rehabili- 75 tate badly run-down state enterprises through privatization often 60 69 61 proved to be more expensive and to take longer than had been 40 51 budgeted for. This led to poor financial results even though there 40 20 were usually clear development benefits. 0 Greenfield & redevelop Rehabilitation & Expansion of going turnaround concern INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 1 6 — P I O N E E R I N G — F I R S T M O V E R A D VA N TA G E O R PAY I N G T H E P R I C E ? 45 Chapter 16: PIONEERING—FIRST MOVER ADVANTAGE OR PAYING THE PRICE? A simple analysis has been made of the financial and developmen- pioneering work in introducing large-scale production to the re- tal success rates for projects defined as pioneering. Not surprisingly, gion. In terms of public goods, strong arguments can be made for pioneers appear to have had a slightly higher risk of failure, although supporting first movers and pioneers. Conversely, supporting the these findings should not be taken to imply that pioneering should expansion of a new farming sector before the technology, produc- be avoided. Two-thirds were classified as successful or moderately tion systems, and markets have been properly tested risks wasting successful in terms of direct development impact. resources. No account is taken of the likelihood that successful pioneers may TABLE 16.1: Percentage of Projects Classified as Success both attract further investment and encourage others to follow or Moderate Success their example—a role they played in the spectacular success and AFRICA ASIA COMBINED growth of KTDA, which also served as the model for outgrower tea DEV FIN DEV FIN DEV FIN projects in Uganda, Tanzania and Malawi. The emergence of the oil Pioneering 57 42 86 67 66 50 Follower 74 51 71 71 73 58 palm sector in Southeast Asia, to some degree, built upon CDC’s DEV = Developmental; FIN = Financial. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R CHAPTER 17 — DEBT VERSUS EQUIT Y 47 Chapter 17: DEBT VERSUS EQUITY Equity investors are sometimes thought to be more committed to because loan-only investments were more likely to be to existing ensuring the commercial success of projects, whereas lenders are businesses whereas equity would usually be required for start-ups. more inclined to rely on their security (mortgages, guarantees). A comparison has therefore been made of project financial perfor- TABLE 17.1: Percentage of Projects Classified as Financial mance where CDC had an equity stake (usually in addition to loans) Success or Moderate Financial Success and where it was only a lender. The following table suggests little AFRICA ASIA COMBINED difference in the case of CDC. If anything, it is the projects where With CDC equity participation 46 62 52 CDC provided only loans that performed slightly better—perhaps With loan only 50 92 60 A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R CHAPTER 18 — CDC MANAGEMENT—MAKING A DIFFERENCE? 49 Chapter 18: CDC MANAGEMENT—MAKING A DIFFERENCE? CDC managed 46 percent of the projects in which it invested, project directly it was more likely to be a start-up or rehabilita- a technical input that was seen as an important part of its total tion, which is inherently more risky. Of the projects managed by contribution to project performance. A comparison between the CDC, 84 percent were start-ups or rehabilitations, whereas for financial performance of managed and nonmanaged projects nonmanaged projects the figure was 61 percent. reveals that overall, projects were more likely to succeed finan- cially if they were not managed by CDC (see table below). One TABLE 18.1: Percentage of Projects Classified as Financial possible explanation for this is that CDC was prepared to take Success or Moderate Financial Success higher risks (especially in its initial pioneering years) when it was AFRICA ASIA COMBINED providing project management itself, but was more cautious With CDC management 40 64 46 Without CDC management 56 74 62 when backing others. Moreover when CDC was managing a A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R C H A P T E R 1 9 — C O N C LU S I O N S — C R I T I C A L S U C C E S S FA C TO R S A N D K E Y R I S K S 51 Chapter 19: CONCLUSIONS—CRITICAL SUCCESS FACTORS AND KEY RISKS This review of 179 agribusinesses in Sub-Saharan Africa and in Disappointing performance during the early stages of a number Southeast Asia and the Pacific illustrates much of the complexity of projects led to uncertainty on the part of CDC over whether it and many of the risks that were involved in agricultural and agro- should cut its losses and get out or persevere and see the investment processing investment in developing countries over the course of through. In each case a review was undertaken that concluded that a half century. Appendix 1 presents an informal checklist of some the fundamentals such as soils, water, and climate were good and that critical commercial success and failure factors which vary depend- CDC should patiently commit additional resources. In these projects, 20 ing upon the type of agribusiness ventures and/or investments. CDC became a de facto provider of “patient capital,” either as equity or as loans which were flexibly rescheduled or converted to equity. A number of projects that eventually turned out to be successful investments, or at least financially self-sustaining, were problematic CUTTING THE LOSSES and generated losses during the early stages of their development. Gambia Poultry. A huge, integrated, over-ambitious project to supply the UK with eggs. No precedent existed for the proj- FROM “DOG” TO “STAR” ect in Gambia. A large and expensive but inexperienced expa- Inyoni Yami Swaziland Irrigation Scheme (IYSIS): Large- triate workforce of 70, unsuitable soils, low yields of feed crops, scale irrigated agriculture in Swaziland envisaged as a rice poultry diseases, market resistance to imported eggs. Closed scheme, but poor technical performance (water-logging, weeds, down after 3 years. pests). Finally profitable once converted to sugarcane Nigeria Agriculture Scheme. An attempt to replicate the Mpongwe: Irrigated wheat and rainfed soya and maize in Sudanese Gezira arable, irrigated settlement scheme on 15,000 Zambia. Heavy initial financial losses until expansion and merger ha of land in Nigeria. No prior crop research and early yields with adjacent Munkumpu scheme achieved economies of scale were low. Little chance that levies on smallholder profits could combined with government liberalization of food crop markets pay for substantial overheads. Scheme was given to Nigerian government who converted it into a research station. BAL Plantations: Loss making Abaca fibre plantations in Sabah facing declining world market, saved by CDC invest- Cape Rodney Estates. A planned 2,250 ha cocoa estate, in ment to convert to oil palm and cocoa joint venture with PNG government, but early plantings pro- duced poor results and a review team advised that the area Cavally: Rubber plantation in Ivory Coast. CDC invested heav- was unsuitable for commercial cocoa production. CDC with- ily in expansion and new factory, but production came on drew from the joint venture. stream at a time of low world market prices followed by civil war. CDC persevered despite financial losses and risks until suc- Tana Coffee. CDC’s objective was to support economic devel- cessful sale during subsequent commodity price boom. opment in newly independent Vanuatu by promoting a nucleus coffee estate and outgrower scheme on Tana island. The local cli- mate however, which was subject to periodic cyclones, was un- 20 Examples include Primary production (for example, rubber): good grow- suitable. There was no tradition of regular, waged employment ing conditions and low transport costs to market; Primary processing (for example, flour): latest technology, economies of scale, logistics; on the island and labor productivity was low. CDC wrote-off its Consumer goods (for example, prepacked fruit and vegetables): product investment of £1.3 million and “gave” project to the government. quality, reliability of supply chains, managerial flair and innovation. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 52 C H A P T E R 1 9 — C O N C LU S I O N S — C R I T I C A L S U C C E S S FA C TO R S A N D K E Y R I S K S However this role of patient capital provider was limited because CDC management. Good management is a necessary but not sufficient had to protect its own solvency at the total balance sheet level, match- condition for success. It can overcome the inevitable shocks and ing cash flow from its diverse portfolio of investments with its own setbacks that arise during the course of a project’s development debt servicing obligations to the UK Treasury. Weak investments there- and ongoing operations. But even excellent management cannot fore needed to be terminated as early and humanely as possible to compensate for a project that is “fatally flawed” at its planning stage avoid continuing losses when there was no prospect of a turnaround. leading to unsound fundamentals. For instance, while growing con- ditions for a pioneering export-oriented rose production enterprise Country risk, in the form of war, unrest, and nationalization wrecked were judged favorable at the Heleena Farms project in Nigeria, nei- a number of good schemes, including outgrower tea in Uganda ther the sponsor nor the expatriate manager had any experience in (repeated civil wars, economic collapse), sugar in Tanzania (nation- rose production. Poor production and distribution performance led alization), cocoa estates in Papua New Guinea, and oil palm estates the sponsor to abandon the rose venture. in the Solomons (both due to separatist insurrections). CDC management successfully developed the Kasungu tobacco Other projects have survived periods of great economic and/or project in Malawi, but were unable to replicate that success in political risk and stress to produce good financial or at least devel- neighboring Zambia where growing conditions were broadly the opmental benefits. same but the fundamental economic conditions were quite dif- ferent. (The dominant influence of the Zambian copper mines led SURVIVING THE BAD TIMES to high wages and an overvalued exchange rate, which punished Kulai Oil Palm Estate in Malaya: assassination of CDC manag- labor-intensive tobacco production for export, whereas in Malawi ers in 1954 and harassment of workforce by communist insur- gents. Survived to provide seedlings, training and initial crop tobacco was the dominant export crop and the government delib- processing for the FELDA settlement scheme erately maintained wage and foreign exchange rates at levels that Tanwat: survival of Tanzanian export-oriented wattle extract supported the tobacco industry.) venture begun in 1950, despite massive overvaluation of cur- Many of CDC’s agribusiness investments were pioneering. Some rency and shortage of inputs, via diversification into food crops, dairying, forestry plantations, and sawmilling in the 1970s and of these represented the first such enterprises in a particular coun- 1980s. CDC subsequently developed an irrigated tea estate try, for instance tilapia in Lake Kariba, Zimbabwe; sugarcane in and factory and a dendro-thermal power station in the 1990s Swaziland and Papua New Guinea; and oil palm in Sabah, Sarawak as the economy liberalized and the Solomon Islands. Others were pioneering in introducing a Lake Harvest: survival since 1996 of a fish farming venture on new production model, for instance organized smallholder tea pro- Lake Kariba, Zimbabwe, in spite of currency collapse and harass- duction in Kenya and Malawi. ment of foreign investors during the last 15 years; now diversi- fied into crocodile farming and poultry and has 400 employees Some of these initiatives, in which CDC served as promoter or finan- cier, helped to pioneer innovations that subsequently grew organi- The recommended approach is to ensure that the fundamentals are cally through a series of expansions into very large undertakings. sound by assessing the critical success factors for the specific type These included: of agribusiness venture being proposed. Once this is determined, much relies on “good luck” in commodity prices being attractive  The Federal Land Development Authority in Malaya, which when production comes on stream; that the weather patterns are involved the settlement of landless farmers to become normal; and that the venture is supported rather than harassed by rubber and oil palm outgrowers. Begun in 1957, over the ensuing 40 years, some 120,000 families were settled in over local and national authorities. 300 new communities; The role of management is critical, but limited. Bad management can  The Kenya Tea Development Authority (KTDA), which ruin any project and all successful projects have at least adequate involved the promotion of smallholder tea growing to supply INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS C H A P T E R 1 9 — C O N C LU S I O N S — C R I T I C A L S U C C E S S FA C TO R S A N D K E Y R I S K S 53 dedicated factories. It began in 1960 with 940 hectares under objectively high risks of failure when making investment decisions smallholder tea cultivation. By 1984, some 145,000 participat- now in anticipation of future profits: ing smallholders were cultivating tea on 58,000 hectares; and Within any investment portfolio it is normal for a few “star”  Zambia Sugar, the first sugar estate in Zambia, began in 1967 performers to carry many “also-rans” and a few outright “dogs.” with an initial field and factory capacity for 35,000 tonnes of Professional fund managers (who need to regularly attract new sugar, by 2011 was producing 385,000 tonnes. funds to manage) are more likely to publicize their stellar his- torical successes and their ex-ante rate of return expectations A number of successful pioneering projects wielded demonstration from future investments than their actual, historical, realized effects through which their practical examples came to serve as average results. The prominence that is assigned to these few models for projects elsewhere: highly successful outcomes leads to a tendency toward exag- gerated expectations regarding prospective financial returns—  The KTDA for instance would provide a model for outgrower expectations which are in reality based on nonrepresentative tea projects in Uganda, Tanzania and Malawi; outliers.  The commercial oil palm and cocoa production that CDC Moreover, the high rates of return obtained on some venture pioneered in Malaya and Sabah (despite early teething prob- capital and private equity “deals” are achieved by trading as- lems in the latter) encouraged other investors to develop sets over a relatively short time period of time, “buying low” new plantations and to convert existing rubber estates and “selling high.” Virtually all of CDC’s equity investments on to these more profitable crops. Thus by 1996, forty years the other hand consisted of longer term commitments, and the after CDC introduced the crop for the first time, there were rates of return on equity invested depended on the develop- 400,000 ha of smallholder and estate oil palms in Sabah, ment of the underlying businesses.21 accounting for 7 percent of world palm oil production; and  CDC was one of the earliest promoters of export oriented The view of the authors is that commercial investors, development horticulture in Kenya on the Osarian and Kuraiha Estates and agencies and host governments and communities alike should while CDC itself failed to achieve profitability, the farming draw some confidence from the findings that: assets that were established became the nucleus for the floricultural industries which later thrived in Kenya.  Over 80 percent of all agribusiness ventures supported by CDC in Africa and Southeast Asia and the Pacific over a Any overall assessment of whether CDC’s agribusiness investment 50-year period yielded some sustained, direct development portfolio has performed “well” or “badly,” and whether it sends out benefits; positive or negative signals to prospective private investors, devel-  That estate/plantation farming projects and smallholder/ opment agencies and host governments and communities would outgrower projects had similar success rates while combined nucleus estate and smallholder schemes did best of all; depend on the criteria used, would in part be subjective and politi-  That only one quarter of projects failed completely in finan- cal and would need to take into account the broader developmen- cial terms; and tal, environmental and social impacts—which, though difficult to  That one in six equity investments were “stars.” assess systematically, were not universally beneficial. Moreover, a small number of agribusiness investments, such as Given the risks it was designed to take and its developmental objec- FELDA, KTDA, and BAL, ultimately had a transformational effect tives, it would be surprising if CDC consistently achieved levels of fi- greatly magnifying the impact of CDC’s original investment, bring- nancial performance that would be expected from a private investor. ing mostly positive economic and development impact across It would be an illusion however to assume that private investment regions and over decades. is always profitable. Commercial investment in general is inherently risky—a fact that is by no means exclusive to investment in agri- 21 Crudely, equity values are based on earnings multiplied by the price/ business or in developing countries. John Maynard Keynes noted earnings (P/E) ratio. CDC’s focus was on improving the earnings of proj- ects whereas private equity investors often have a focus on improving the significance of what he called “animal spirits” in overcoming the the P/E ratio. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R A P P E N D I X 1 — A C H E C K L I S T O F S O M E C R I T I C A L C O M M E R C I A L S U C C E S S A N D FA I LU R E FA C TO R S 55 Appendix 1: A CHECKLIST OF SOME CRITICAL COMMERCIAL SUCCESS AND FAILURE FACTORS FOR AGRIBUSINESS INVESTMENTS ASPECT OF THE VENTURE POSSIBLE CRITICAL SUCCESS OR FAILURE FACTORS Sector Characteristics: • Political/economic stability All sectors • Rising or declining demand for the product • Market value of existing assets generally at a discount or premium to replacement cost • Dependence on heavy protection from imports • Dependence on export privileges • Profits restricted by government market interference • Competition on level playing field (for example, fiscal advantages, no unfair business practices) • Profits capped by dependence on dominant suppliers or customers (competitive forces) Sector Characteristics: • Technological edge/infrastructure to create barriers to entry Inputs and services • Large market share, to be competitive with rivals Sector Characteristics: • Good growing conditions—high productivity Primary production • Economies of scale—low cost production • Low transport cost (for bulky inputs and to market) • Survivability during low-point in world market price cycles • If labor intensive: scope for smallholder production • If capital intensive: scope for estate/consolidated outgrowers • “Natural” protections (for example, geographical location limiting competition) Sector Characteristics: • Economies of scale in processing and procurement/distribution Primary processing • Any scale or technology barriers to entry • Status of existing capacity utilization within the country • Globally competitive product (esp. if production is of a secondary ingredient for food or other manufacturing) • Reliability of suppliers • Diversified customers Sector Characteristics: • Own brands or reliance on franchising Manufacturing, packaging, marketing • Economies of scale in processing, procurement and distribution • Viable market share within target segment • Consistent product quality • Good customer service via reliable supply chains • Innovative management developing new, higher margin product lines Financial strength of investee company • Strong balance sheet or heavy gearing with third party debt • Fully financed development plan or heavy reliance on forecast self-generated funds to finance development Type of business transaction • Capital intensive start-up or lower risk, modular development • Diversification into unrelated products and markets • Complex rehabilitation/privatization • Highly competitive acquisition (high purchase price incorporating future development potential) • Relatively simple rehabilitation/privatization • Expansion of going concern Scale • Able to afford top quality management • Potential to grow into bigger business (Continued ) A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 56 A P P E N D I X 1 — A C H E C K L I S T O F S O M E C R I T I C A L C O M M E R C I A L S U C C E S S A N D FA I LU R E FA C TO R S ASPECT OF THE VENTURE POSSIBLE CRITICAL SUCCESS OR FAILURE FACTORS Organization and management • Existing management continues • Experienced, new management from within country • Experienced new management from outside country • Inexperienced management • Corporate management • Entrepreneurial manager • Smallholder/outgrower participation • Sponsor objectives Scope for eventual sale of equity stake • Scope for listing on local/regional stock market • Will attract competing buyers, including global strategic players and/or competing local/regional buyers • Management buyout MBO only plausible buyer • JV partner only plausible buyer • Supplier/customer de facto veto INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS A P P E N D I X 2 — D ATA C L A S S I F I C AT I O N 57 Appendix 2: DATA CLASSIFICATION BACKGROUND  Inp—specialist input supply business  Diversified—investment funds or lines of credit specialising CDC’s agribusiness investments were classified according to a range in agribusiness of criteria, which are defined in this note. PROJECT DESCRIPTION INTEGRATED ACTIVITIES Projects have been defined by their principal crops, products or Some components (for example, a nucleus estate or an outgrower activity. scheme) may be financed separately, that is, not part of a single pro- ject. They are noted if they are a central component of an integrated All of the investments that CDC would have considered as scheme, even if not financed by CDC. “Natural Resources” have been included, including fisheries and forestry. SCHEME TYPE Stand-alone processing projects are included only when there is a strong linkage to domestic raw material supply. An overall, summary classification of the projects has been made using the following categories: By the 1990s, CDC was shifting away from financing projects to  Estate: estate/plantation farming and directly related creating and investing in businesses, hence the “dawn raid” on the processing, with zero or minimal supplies from outgrowers/ Thai stock market in 1993 to acquire a controlling stake in United smallholders Palm Oil Industry (Public) ltd, with a view to a merger with CDC’s  Outgrowers: outgrower/smallholder farming and directly existing oil palm interests in the country to build “critical mass” and related processing create an “exit route” rather than for any specific capital develop-  NES: Nucleus estate and outgrowers/smallholders and ment project. directly related processing  Processing; Independent processing operation, obtaining PROJECT ACTIVITIES raw material from the open market, rather than any inte- The activities of projects/businesses are classified as follows: grated estate/outgrower activities  Input: supply of seeds and/or other inputs and services  Est—estate farming  Finance: specialist investment funds/lines of credit  Out—services to outgrowers, farming their own land  Set—services to outgrowers, who are settlers on project land  Proc—substantial processing facilities SCALE  Mkt—independent marketing function (for example, not It is difficult to compare scale for diverse sectors (that is, a large through a separate marketing board) flower project would be 40 hectares under glass, whereas a large  Hvst—harvesting, (for example, fishing, logging of natural sugar estate would be over 10,000 ha). We have therefore attempted forest) A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 58 A P P E N D I X 2 — D ATA C L A S S I F I C AT I O N to categorize into Small, Medium, Large, and Mega relative to the Projects are classified as “rehab” if there is an existing business which norms of the specific sector and region; for example, Sulmac is a large has failed or is underperforming and the objective is to turn it horticulture enterprise by the standards of the horticulture industry. around, improve the capacity utilization and productivity of mainly existing assets and to achieve profitability, for example, Munkumpu, The scale shown is the ambition of the promoters, even where this Rwenzori Highlands Tea. There may also be a component to expand was greater than the actual achievement; for example, Family Farms production capacity. in Zambia had the ambition to settle 6,000 tobacco smallholders and so is classified as large, even though the project failed. Projects are classified as “existing” if there is a viable going concern, and the objective is to build on success via expansion and/or diversification. INTENSITY Intensity is not the same as scale. Irrigated sugar-estates can be huge and are also a very capital and resource intensive use of the PIONEER land. Projects have been considered pioneering if some major aspect was new to the country or region at the time of CDC’s initial Broadly speaking, all plantations and all irrigated and/or highly investment: labor intensive farming have been classified as “intensive,” whereas ranching, forestry, and low-input, rainfed cropping are classified as  Technology—Primo Fina Oleochemicals “extensive.”  Crop—BAL pioneered oil palm in Sabah; Zambia Sugar pioneered sugar growing in Zambia  Organization—KTDA pioneered large-scale organized, inde- WATER pendent, smallholder tea production in Kenya, VIF pioneered Irrigation is identified where known, as perhaps this is an important smallholder sugar production in Swaziland contributor to success or failure. Some businesses were originally pioneering (Triangle sugar in Zimbabwe) but were well established prior to CDC’s investment, MARKET and so the risks of pioneering had already been overcome. Orientation toward exports or local markets may be an important success factor. Where there are significant sales to both markets, the SECTOR main market is shown first. In the 1980s CDC began to focus on private sector investment, but before that it supported many unincorporated government NEW/EXISTING schemes via loans direct to the government as well as investing in parastatal enterprises. This obviously has a bearing on the concept Projects are classified as “new” if they are fully “green-field” develop- of financial success and how to interpret outcomes. We have used ments or if they are substantially new, that is, developing a commer- the following categories: cial project out of a pilot scheme (Mpongwe, Oil Crop Development) or developing a completely moribund asset (Lobatsi Abattoir)  CDC—majority owned/controlled by CDC or converting a low value land-use into an intensive agribusiness  Pvte—majority owned by private investors (other than CDC) (Swaziland Irrigation Scheme, Nanga Farms and Kaleya Smallholders  Para—parastatal enterprise (incorporated) or a govt share- were all used for ranching, prior to acquisition and conversion to holding in a limited company irrigated agribusiness). Projects are also classified as “new” if CDC  Govt—unincorporated government scheme  JV—Joint venture between any of the above, with first funding was committed prior to project commissioning/comple- named partner having control tion, even if work on the project had already started (Advance Agro). INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS A P P E N D I X 2 — D ATA C L A S S I F I C AT I O N 59  Plc—a company already listed on a stock market at the time As many projects were co-financed, the overall project cost is of- of CDC’s investment ten greater than the CDC investment. However, total project cost  TA—technical assistance; that is, management services pro- information is not available on a consistent basis, and as noted vide by CDC even where it was not the controlling investor. above once CDC started to make strategic acquisitions of existing With some projects, control changed over time. They are catego- businesses, the concept of “project cost” becomes less meaningful. rized by the phase we considered most important. CDC INVESTMENT AMOUNT (2011, US$) CDC INVESTMENT TYPE This is the CDC investment adjusted for UK inflation (Consumer CDC invests mainly via equity stakes and/or loans direct to the Price Index) to July 2011 and converted to US$ at an exchange rate project entity. However, for some parastatal and governmental pro- of £1 = US$1.65. jects CDC made a loan to the host government which the govern- The CPI was 31.2 in July 1948 and 935.9 in July 2011, that is, an in- ment then utilized to support the project in its own way (on-lending crease of thirty fold. or even direct government funding). This creates a potential separa- tion of CDC investment performance and project commercial per- For the purpose of this calculation a guesstimate was made of the formance. We have noted therefore when loans were made to the phasing of CDC investments, where it is known that disbursement govt, rather than direct to a project entity. took place over several years (forestry projects). TECHNICAL PERFORMANCE CDC MANAGED Broadly classified as: Whether or not, at any time, CDC had management responsibility, either as majority shareholder, a “corporate” management agree-  Fail—had to be abandoned because resource or technology ment or via secondment of the chief executive. or management unsuitable  Moderate Fail—productivity achieved just sufficient for survival, but well below target CDC INVESTMENT AMOUNT (£M)  Moderate Success—reasonable productivity achieved, but This is normally the investment amount approved by CDC’s board below planned levels or committed via subsequent investment agreements or actually  Success—main productivity targets achieved and broadly a disbursed. It is only indicative as: competitive performance  Some commitments were not fully drawn down by the project. DIRECT DEVELOPMENT IMPACT  Some CDC projects were not incorporated for many years, A narrow definition of development impact has been adopted— and were financed directly from CDC’s bank account. creation of jobs and livelihoods and specific economic objectives  Some developments were financed out of project cash that were an explicit part of the project rationale, for example, flows/retained earnings therefore no specific CDC Board earning or saving foreign exchange, contributing to national food approval exists. production—as less direct impacts cannot normally been inferred  In at least one case (IYSIS) the opposite was true—loans from in CDC’s annual reports. were made to facilitate dividend payments rather than capital expenditure. The broad classification is: The aim is to indicate how significant the project was for CDC.  Fail—no sustainable incomes/jobs created Where the approved/committed amount is totally misleading, then  Moderate Fail—some worthwhile employment and income the amount actually invested is shown instead. creation continues (either as a business or as viable small- holder production) but far less than planned A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 60 A P P E N D I X 2 — D ATA C L A S S I F I C AT I O N  Moderate Success—substantial, on-going development Exogenous: Some unforeseen event or factor seriously harmed the benefits, but less than planned project, for example, civil war (Liberia) or economic war (Nigeria),  Success—substantial commercial activity continues, either nationalization (Tanzania), collapse of market (Tung oil). as a business and/or as substantial smallholder production, equalling or exceeding expectations Management: Owners and/or managers lacked the experience or competence or integrity necessary to achieve project objectives. This classification takes no account of wider economic and develop- Where projects are nationalized and governments impose unsuit- mental impacts, such as: able management this is classified as “exogenous.”  The financial cost to the country (in debt service obligations) of the development impact achieved  The “opportunity” cost (that is, could and would the invest- EQUITY RETURNS ment and resources utilized have had greater development Considered from the perspective of actual or potential private sec- impact if used elsewhere?) tor investors, where equity was involved:  The indirect (backward linkages) and induced (forward linkages) effects on the economy  Fail—loss of more than 25 percent of equity value  External costs (government provision of infrastructure or  Moderate Fail—loss of equity value, but less than 25 percent services, impact on the environment) and benefits (any wel-  Moderate Success—some return on equity capital, but less fare services to employees, their families and communities than 12 percent internal rate of return (IRR) provided by the project)  Success—annualized return of over 12 percent, allowing for dividends and equity sale or valuation PROJECT FINANCIAL VIABILITY Note that no allowance is made for “gearing/leverage” when consid- This relates to the establishment of a solvent, “going concern”, that is, ering target equity returns. financial sustainability. Broadly classified as: Note also that the price at which equity is finally sold has a major  Fail—Business collapsed and ceased trading bearing on equity returns and may not be directly determined by  Moderate Fail—A business survived as a going concern, but project profitability, for example, when shares are listed on a stock needed substantial subsidization, for example, via refinanc- market, such as Ariston in Zimbabwe or Advance Agro in Thailand, ing by shareholders or via negotiated debt write-off or via a the quoted share price will be affected by country issues as well as sale as a going concern by a liquidator/receiver underlying project performance.  Moderate Success—Self-sustaining business established in line with expectations, but no significant profits CDC INVESTMENT PERFORMANCE  Success—Positive returns on all capital employed This is not necessarily the same as project performance since CDC did not always invest directly in the project entity. Sometimes CDC benefited from loan guarantees and sponsor support even when CAUSE OF FINANCIAL FAILURE projects struggled. Conversely some parastatal enterprises per- Projects that were failures or moderate failures in financial terms are formed well (Hevecam), but CDC’s loan was via government which categorized as follows: defaulted on its international obligations. Concept: the project as planned was “fatally flawed” and there was The broad classification is: nothing that operational management could do to retrieve the situ- ation, for example, wrong choice of site; grossly over-optimistic yield  Fail—loss of over 50 percent of capital invested (equity and/ or price assumptions. or loans)  Moderate Fail—loss of 50–100 percent of capital invested INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS A P P E N D I X 2 — D ATA C L A S S I F I C AT I O N 61  Moderate Success—no loss of capital but no significant  Mkt—was the project crippled or boosted by prices on income commodity markets that the project was too small to influ-  Success—compound return on equity investment of over ence? Did the project benefit from preferential markets, for about 5 percent; loans repaid with full interest. example, sugar quotas?  Mgmt—management, to include the role of the sponsor as A further complication is that, in preparation for the planned privati- well as management on the ground. Was the project fatally zation of CDC around 2000 (subsequently abandoned), most sov- flawed, whatever the management did, or was dismal or ereign debts on CDC’s balance sheet were transferred to the British exceptional management performance a decisive factor? government. We have taken a view on whether CDC’s outstanding  Govt—did government economic policies and actions sovereign loans were in the process of being serviced, or were al- (confiscations, punitive taxes, price controls) undo the best efforts of investors and managers? Or on the contrary did ready many years in arrears and unlikely ever to be repaid. government go out of its way to support the project and facilitate success? OUTCOME  War—invasions, civil war, independence movements, terror- This is normally the status at the time CDC’s investment came to ism, aggressive sanctions which severely disrupted econom- ic activities or destroyed them completely. an end. In many cases both important success and failure factors were at However, from 1998 onwards, CDC stopped providing project- work at different times during a projects life; for example, Solomon specific financial data in its Annual Report and Accounts, and so for Islands Plantations (SIPL) operated very successfully with effective some investments there is an element of conjecture. government support for 30 years until Guadalcanal was overrun by secessionist rebels. Similarly, swings in world prices can make “mar- SUCCESS AND FAILURE FACTORS kets” both a failure and a success factor (CDC tried to sell Cavally in This analysis in the Data Base aims to encapsulate any noteworthy 2002 but could not find a buyer because of low rubber prices and factor(s) that made a major contribution to success or failure sum- operating losses; whereas high world market prices in the late 2000s marized as: generated large cash flow surpluses and an attractive exit for CDC,  Res—natural resources: were they of “world class” qual- in spite of the civil strife in Ivory Coast). ity (Mpongwe soils) or totally unsuitable for the venture Subsequently the Data Base provides for each project a brief (Ndolela arable project)? summary of key parameters and events, where known, in the  Tech—the choice of technology/project planning (which can be appropriate and thorough, or misconceived and life of the project, including updates on performance post CDC inadequate)/project scale involvement. A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R A P P E N D I X 3 — P R O J E C T S I N C LU D E D I N T H E S T U DY 63 Appendix 3: PROJECTS INCLUDED IN THE STUDY SOUTHEAST ASIA AND THE PACIFIC 2000 Asiatic Persada Indonesia oil palm 1999 PT Agro Indomas Indonesia oil palm 1997 Keels Plant. Mgt Serv Sri Lanka tea, rubber 1996 PT Harapan Indonesia oil palm 1996 Soucrerie Bourbon Tay Ninh Vietnam sugar 1993 United Palm Oil Industry Thailand oil palm 1993 Advance Agro Ltd Thailand forestry: pulp & paper 1993 Primo Fina Oleochemicals Philippines coconut pdcts 1993 Bukidnon Resources Philippines tomato paste 1992 Kulim Plantations Malaya oil palm, rubber, fruit 1991 Soon Hua Seng Thailand forestry: pulp & paper 1990 Keresa Plantations Sarawak Rattan 1990 Mongkolwat Thailand aquaculture: prawns 1989 Mah Boonkrong Sirichai Thailand cashew 1989 PT Tasik Raya Indonesia oil palm 1988 New Guinea Plantations PNG cocoa,oil palm 1988 Kolombangara Solomon Islands forestry: gmelina 1987 Phansrivivat Thailand oil palm 1987 Desa Tea Malaya tea 1987 South Santo Cattle Vanuatu cattle: ranch 1987 Poliamba PNG oil palm, cocoa 1986 Fiji Forest Industries Fiji forestry: timber 1985 Smallholder Rubber II Indonesia rubber 1985 Milne Bay Estates PNG oil palm, cocoa 1985 Tropik Wood Indus. Fiji forestry: timber, wood chips 1985 Tana Coffee Vanuatu coffee 1984 World Aquaculture Ltd Thailand aquaculture: prawns 1984 Pelwatte Sugar Sri Lanka sugar 1984 Ladang Baturong Sabah oil palm 1983 Metenesel Vanuatu cocoa 1983 NES Project VII Indonesia rubber, oil palm 1982 NDC/Guthrie Philippines oil palm 1982 NES Project VI Indonesia rubber, coconut 1982 Cape Rodney Estates PNG cocoa (Continued ) A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 64 A P P E N D I X 3 — P R O J E C T S I N C LU D E D I N T H E S T U DY SOUTHEAST ASIA AND THE PACIFIC 1980 Ramu Sugar PNG sugar, beef, oil palm 1980 Palong Cocoa Malaya oil palm, cocoa 1979 Fiji Citrus Products Fiji citrus 1978 Coklat Ransiki Indonesia cocoa 1978 Fiji Sugar Corp Fiji sugar 1976 Higaturu (HOPPL) PNG oil palm, cocoa 1976 Gula Pedang Terap Malaya sugar 1976 Tatar Anyar Indonesia tea, rubber 1976 ORRAF Thailand rubber 1975 Fiji Pine Commission Fiji forestry: pines 1974 Darabif Malaya cattle: ranch 1970 Solomon Islands Plantations (SIIPL) Solomon Islands oil palm, cocoa 1970 Chocolate Products Malaysia Cocoa proc 1967 Sarawak Oil Palms (SOP) Sarawak oil palm 1961 Fiji Lumber Co Fiji Timber 1957 Mostyn Estate Ltd Sabah oil palm 1957 Johor Palm Processing Malaya oil palm 1956 Federal Land Dev Auth (FELDA) Malaya oil palm, rubber 1955 United Cocoa Dev Co Malaya cocoa 1950 Kulai Estate Malaysia oil palm 1950 Marudu Rice Farm Sabah rice, groundnuts 1949 Malayan Cocoa Ltd Malaya cocoa 1948 Borneo Abaca Ltd (BAL) Sabah oil palm, rubber, cocoa SUB-SAHARAN AFRICA 1998 Ariston Zimbabwe divers—agribus hold Co 1998 Sulmac Kenya horticulture: carnations, roses, veg 1997 AGRIMO Mozambique cotton 1997 NewFarmers SA divers—invest fund 1996 Cavally Cote d’Ivoire Rubber 1996 Lake Harvest Zimbabwe fish: aquaculture—tilapia 1996 York Farms Zambia horticulture : roses, veg 1996 Cadilu Namibia fish: processing 1996 MOCITA Mozambique Cashew 1995 AgricTrust Zimbabwe divers—line of credit 1995 Mpongwe Milling Zambia flour—wheat 1995 Munkumpu Farm Zambia arable: wheat, soya, maize 1993 Rwenzori Highlands Tea Uganda tea 1993 Heleena Farm Nigeria horticulture—roses 1993 Triangle Ltd Zimbabwe sugar 1993 Hippo Valley Zimbabwe sugar 1993 Cadbury Nigeria Nigeria cocoa: processing 1992 Kilombero Valley Teak Tanzania forestry: teak 1992 Karimjee Agriculture Tanzania tea, sisal INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS A P P E N D I X 3 — P R O J E C T S I N C LU D E D I N T H E S T U DY 65 SUB-SAHARAN AFRICA 1992 FRI Ltd Ghana pineapple, mango 1992 Aussenkehr Farms Namibia horticulture: table grapes 1991 Eglin Plantations Cote d’Ivoire pineapples, bananas 1991 Astek Food Processing Ghana fruit juice 1991 Divine Sea Foods Ghana fish: processing 1991 Sebovia Cote d’Ivoire cattle: ranch, abattoir 1990 Makumbaya Farms Gambia horticulture: chrysanths 1990 Plantations Dam Cote d’Ivoire pineapples 1989 Chrismill Farms Tanzania pineapples 1989 South Nyanza Sugar Co Kenya sugar 1989 Nanga Farms Zambia sugar, wheat, soya, coffee 1989 Masstock Zambia Ltd Zambia Cotton, wheat, marigolds 1988 Swazi Meat Indus Swaziland cattle: ranch, abattoir 1988 Zambia Cashew Co Zambia cashew 1987 Sable/Impala Farming Malawi tobacco, arable, coffee, dairy 1987 Serebou Seeds Cote d’Ivoire seed 1987 Cold Storage Comm. Zimbabwe abattoirs, meat processing 1987 East Usumbara Tea Tanzania tea 1986 Southdown Hldgs Zimbabwe tea 1985 Ndolela Farm Tanzania arable: maize 1985 Oil Crop Dev Co Kenya arable: oilseeds 1984 Kulalu Ranch/AgDevLtd Kenya cattle 1984 Kawalazi/Kavuzi Estates Malawi tea, macadamia 1984 Mpongwe Dev Co Zambia arable—wheat, maize, soya + coffee 1984 Rusitu Valley Dev Co Zimbabwe dairy, coffee, tobacco 1983 Sugar Corp of Uganda Uganda sugar 1983 Rubber Corp of Liberia Liberia rubber 1983 Tamteco Uganda tea 1980 Ngwaketse Pilot Ranch Botswana cattle 1980 Vizara Rubber/Mandala Malawi rubber 1980 Kaleya Smallholders Zambia sugar 1980 Hevecam Cameroun rubber 1980 Decoris Oil Palm Co Liberia oil palm 1979 Southern Paper Mills Tanzania forestry: pulp, paper 1979 Smallholder Coffee Kenya coffee 1979 SODEFOR Cote d’Ivoire forestry: teak 1978 Rubber Outgrowers Cote d’Ivoire rubber 1978 Standard Tobacco Packers Malawi tobacco 1978 Smallholder Sugar Auth Malawi sugar 1978 Smallholder Coffee Auth Malawi coffee 1978 National Seed Co Malawi seeds 1978 TWIFO Oil Palm Ghana oil palm 1977 ZAFFICO Zambia forestry: pines, timber, poles 1978 Liberia Rubber Dev Co Liberia rubber (Continued ) A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R 66 A P P E N D I X 3 — P R O J E C T S I N C LU D E D I N T H E S T U DY SUB-SAHARAN AFRICA 1978 Irrigation Authority Mauritius sugar 1978 Royal Swazi Sugar Swaziland sugar 1977 Dwangwa Sugar Malawi sugar 1977 Changanda Farm Zambia tobacco 1977 Palmindustrie Cote d’Ivoire oil palm 1973 Family Farming Zambia tobacco 1974 Société Africaine de Plantations Cote d’Ivoire rubber d'Hévéas (SAPH) 1973 Gumaro Tea Ethiopia tea 1970 Tanzania Seed Co Tanzania seeds (mostly maize) 1973 Mumias Sugar Kenya sugar 1972 Kuraiha Estate Kenya horticulture, coffee 1971 Mananga Agricultural Management Swaziland na Centre (MAMC) 1971 Oserian Estate Ltd Kenya horticulture 1970 South Chad Irrig Proj Nigeria cotton 1970 Savannah Sugar Nigeria sugar 1970 Oke-Afa Farms Nigeria poultry 1969 Libby’s Swaziland Ltd Swaziland pineapples, citrus 1967 Zambia Sugar Co Zambia sugar 1967 Shiselweni Forestry Swaziland forestry: eucalyptus oil, timber 1967 Smallholder Tea Auth Malawi tea 1967 Mukonchi Tobacco Zambia tobacco 1965 Uganda Tea Grower Corp Uganda tea 1964 AEL: Buganda/Mwenge Uganda tea 1963 Pyrethrum Processing Co Kenya pyrethrum extract 1963 E Nigeria/Cross River Estates Nigeria rubber 1962 S’holder Tea Factories Kenya tea 1962 Vuvulane Irrig Farms Swaziland sugar 1961 Tanganyika Extract Co Tanzania pyrethrum extract 1961 Land Dev & Settle Board Kenya diversified 1960 Kilombero Sugar Tanzania sugar 1960 Spec Crop Dev Auth/KTDA Kenya tea 1959 Camaroun Dev Corp Cameroun rubber, oilpalm, bananas, tea 1959 Maramba Estate Tanzania cocoa, coffee 1959 Nyambeni Tea Co Kenya tea 1958 Bird & Co Tanzania tea, sisal 1957 Ilushin Estates Nigeria rubber 1957 Unga Ltd Kenya flour: wheat 1957 Mhlume Sugar Swaziland sugar 1955 Swaziland Canners Ltd Swaziland pineapples 1955 Meat Commission Kenya abattoir, meat processing 1952 Umbombo Irrig Scheme Swaziland sugar 1951 Seychelles Fisheries Seychelles fisheries 1950 Swazi Irrigation Scheme Swaziland sugar, citrus, cattle (rice) 1950 Kasungu Tobacco Malawi tobacco INVESTING IN AGRIBUSINESS: A RETROSPEC TIVE VIEW OF A DEVELOPMENT BANK’S INVESTMENTS IN AGRIBUSINESS A P P E N D I X 3 — P R O J E C T S I N C LU D E D I N T H E S T U DY 67 SUB-SAHARAN AFRICA 1950 Omo Sawmills Nigeria timber 1950 Bechuanaland Ranch Botswana cattle, arable 1950 Molopo Ranch Botswana cattle 1950 Lobatsi Abattoir/BMC Botswana abattoir 1949 Gambia River Farm Gambia arable 1949 Tanganyika Wattle Co (Tanwat) Tanzania wattle, arable, dairy, timber, tea 1949 Rice Farm Gambia rice 1949 Atlantic Fisheries Gambia fish: shark, tuna 1949 West Africa Fisheries Nigeria fish: trawling, processing 1949 Niger Agric Project Nigeria arable, groundnuts 1949 Usutu Forestry/Pulp Swaziland forestry: pulp 1948 Nyasaland Fisheries Malawi fish: lake fishing 1948 Limpasa Dambo Farm Malawi arable 1948 Tung Oil Plantations Malawi forestry: tung oil 1948 Poultry/Farming Project Gambia poultry, arable A G R I C U LT U R E A N D E N V I R O N M E N TA L S E R V I C E S D E PA R T M E N T D I S C U S S I O N PA P E R Agriculture and Rural Development (ARD) 1818 H Street, NW Washington, D.C. 20433 USA Telephone: 202-477-1000 Internet: www.worldbank.org/ard