Adding Value to Root and Tuber Crops

ISBN 958-9439-14-4

Adding Value to Rootand Tuber Crops

,,4 1Heueuat tUe- 'P~'D~~

Christopher Wheatley, Processing Specialist for East andSoutheast Asia and the Pacific Region, CentroInternacional de la Papa (CIP), Bogor, Indonesia

Gregory J. Scott, Leader, Postharvest Management andMarketing Program, CIP, Lima, Peru

Rupert Best, Leader, Cassava Program, CIAT, Cali, Colombia

Siert Wiersema, International Consultant, University ofWageningen, Department of Crop Science, Acacialaan,The Netherlands

r~_onooroInternational Center for Tropical Agriculture

ISBN 958-9439-14-4

Adding Value to Rootand Tuber Crops

,,4 1Heueuat tUe- 'P~'D~~

Christopher Wheatley, Processing Specialist for East andSoutheast Asia and the Pacific Region, CentroInternacional de la Papa (CIP), Bogor, Indonesia

Gregory J. Scott, Leader, Postharvest Management andMarketing Program, CIP, Lima, Peru

Rupert Best, Leader, Cassava Program, CIAT, Cali, Colombia

Siert Wiersema, International Consultant, University ofWageningen, Department of Crop Science, Acacialaan,The Netherlands

r~_onooroInternational Center for Tropical Agriculture

Centro Intemacional de Agricultura TropicalInternational Center for '[}-opical AgricultureApartado Aereo 6713Cali, Colombia

CIAT Publication No. 247ISBN 958-9439-14-4Press run: 1,500Printed in ColombiaSeptember 1995

Adding value to root and tuber crops: A manual on product development I ChristopherWheatley, Gregory J. Scott, Rupert Best, and Siert Wiersema. -- Cali, Colombia: CentroIntemacional de Agricultura Tropical, 1995. 166 p. -- (CIAT publication; no. 247)

Centro Intemacional de Agricultura TropicalInternational Center for '[}-opical AgricultureApartado Aereo 6713Cali, Colombia

CIAT Publication No. 247ISBN 958-9439-14-4Press run: 1,500Printed in ColombiaSeptember 1995

Adding value to root and tuber crops: A manual on product development I ChristopherWheatley, Gregory J. Scott, Rupert Best, and Siert Wiersema. -- Cali, Colombia: CentroIntemacional de Agricultura Tropical, 1995. 166 p. -- (CIAT publication; no. 247)

Contents

Preface

PART I: OPPORTUNITIES, LESSONS, AND GUIDELINES

Unit 1: Needs and Opportunities in Product DevelopmentCharacteristics of Roots and TubersProduction Trends

Trends in Consumption and UtilizationFuture Prospects

Unit 2: An Integrated Approach to Product Development

Product Development and Integrated ProjectsThe Elements of Project DesignWho Benefits and How?Integration: The Key to Successful Product DevelopmentA Model for Product DevelopmentFinal Tips on Project Design

Unit 3: Identifying Product IdeasHow to Generate Product Ideas

Selecting the Project RegionHow to Screen Product IdeasIdeas Define Action

Unit 4: Research for Product and Process DevelopmentDesigning an Ideal SystemMarket, Consumer, and Farm-Oriented Research

Technical Research on the Product and ProcessA Preliminary Status Report

Unit 5: The Pilot Phase

Setting Up the Pilot PlantRefining Plant Operations

Finding a Niche in the MarketTraining ProcessorsTest Markets for Consumer ProductsThe Feasibility StudyTiming the Transition

iii

Page

vii

1

336

1217

23232526283134

37

37

4142

47

515153

6168

73

73

7883

8586

8790

Contents

Preface

PART I: OPPORTUNITIES, LESSONS, AND GUIDELINES

Unit 1: Needs and Opportunities in Product DevelopmentCharacteristics of Roots and TubersProduction Trends

Trends in Consumption and UtilizationFuture Prospects


Product Development and Integrated ProjectsThe Elements of Project DesignWho Benefits and How?Integration: The Key to Successful Product DevelopmentA Model for Product DevelopmentFinal Tips on Project Design

Unit 3: Identifying Product IdeasHow to Generate Product Ideas

Selecting the Project RegionHow to Screen Product IdeasIdeas Define Action

Unit 4: Research for Product and Process DevelopmentDesigning an Ideal SystemMarket, Consumer, and Farm-Oriented Research

Technical Research on the Product and ProcessA Preliminary Status Report


Setting Up the Pilot PlantRefining Plant Operations

Finding a Niche in the MarketTraining ProcessorsTest Markets for Consumer ProductsThe Feasibility StudyTiming the Transition

iii

Page

vii

1

336

1217

23232526283134

37

37

4142

47

515153

6168

73

73

7883

8586

8790

Unit 6: The Commercial PhaseGetting Organized

The Product and Process Revisited

Investment and Impact

'Ibward a Self-Reliant Agroindustry

A Final Status Report

Unit 7: A Review ofMajor Issues

Key Aspects of Product Development

Future Prospects

PART II: SUMMARIES OF PRODUCT DEVELOPMENT CASE STUDIES

Adding \blue to Root and Tube,. Crops

Page

939397

105

106

110

113

113

116

117

Case 1:

Case 2:Case 3:

Case 4:

Case 5:

Case 6:

Case 7:

Case 8:

Case 9:

Case 10:

Dried Cassava for Animal Feed in Colombia

Conserved Fresh Cassava for Human Consumption in Colombia

Cassava Flour for Human Consumption in Colombia

Diversifying Cassava Markets in Ecuador

An Integrated Cassava Project in Brazil

Drying Potato in India's Villages

Processing Dried Potato in Peru

Simple Potato Processing in Colombia

Development of a Sweet Potato Beverage in the Philippines

Development of Root Soy Sauce in the Philippines

119

122126130

135139143

147

150153

Appendix: Contributors to this Manual

Acronyms

iv

157

164

Unit 6: The Commercial PhaseGetting Organized

The Product and Process Revisited


'Ibward a Self-Reliant Agroindustry



Key Aspects of Product Development

Future Prospects

PART II: SUMMARIES OF PRODUCT DEVELOPMENT CASE STUDIES

Adding \blue to Root and Tube,. Crops

Page

939397

105

106

110

113

113

116

117

Case 1:

Case 2:Case 3:

Case 4:

Case 5:

Case 6:

Case 7:

Case 8:

Case 9:

Case 10:


Conserved Fresh Cassava for Human Consumption in Colombia







Development of a Sweet Potato Beverage in the Philippines


119

122126130

135139143

147

150153

Appendix: Contributors to this Manual

Acronyms

iv

157

164

Contents

Boxes

Page

Unit 21:

2:3:4:

Unit 3

5:

Unit 4

6:7:

8:

9:

Defining the Beneficiaries of a Potato Project in Peru (Case 7)

Multidisciplinary Teamwork in the Philippines (Case 9)

Farmer-to-Farmer Training in Ecuador (Case 4)

The Four Stages of Product Development-Dried Cassava in Colombia (Case 1)

Initial Screening of Cassava Product Ideas in Colombia (Case 1)

Consumption of Processed Potato Products in Peru (Case 7)

Assessing the Potential for Potato Processing in Colombia (Case 8)

Selection of Equipment for a Cassava Flour Project in Colombia (Case 3)

Selecting a Name for Fresh, Storable Cassava in Colombia (Case 2)

27

293032

43

57606669

Unit 5

10: Organizing Farmers in Colombia (Case 1)

11: Selecting the Site for a Pilot Plant in Colombia (Case 3)

12: Obtaining Raw Material for Cassava Processing in Nigeria

13: Dealing with Raw Material Quality in Colombia (Case 3)

14: A Study of the Market for Cassava Flour in Colombia (Case 3)

15: Applying a Financial Model to Cassava Flour Production in Colombia (Case 3)

16: Costs and Returns in Simple Potato Processing in India (Case 6)

Unit 617: Second-Order Organizations in Colombia (Case 1)

18: Second-Order Organizations in Ecuador (Case 4)

19: Interinstitutional Committees in Brazil (Case 5)

20: Interinstitutional Committees in the Philippines (Case 10)

21: A Three-Tiered System for Distributing Potato Products in India (Case 6)

22: Dividing and Conquering with Potato Products in Peru (Case 7)

23: Distribution of Fresh Cassava in Colombia (Case 2)

24: A Comprehensive Strategy for Project Evaluation in Brazil (Case 5)

25: Versatile Processing Enterprises in Ecuador (Case 4)

Unit 726: Problem Solving at the Pilot Phase in Colombia (Case 3)

v

76

78

8081

8491

91

95969899

101102103108109

115

Contents

Boxes

Page

Unit 21:

2:3:4:

Unit 3

5:

Unit 4

6:7:

8:

9:

Defining the Beneficiaries of a Potato Project in Peru (Case 7)



The Four Stages of Product Development-Dried Cassava in Colombia (Case 1)

Initial Screening of Cassava Product Ideas in Colombia (Case 1)

Consumption of Processed Potato Products in Peru (Case 7)


Selection of Equipment for a Cassava Flour Project in Colombia (Case 3)


27

293032

43

57606669

Unit 5

10: Organizing Farmers in Colombia (Case 1)

11: Selecting the Site for a Pilot Plant in Colombia (Case 3)

12: Obtaining Raw Material for Cassava Processing in Nigeria

13: Dealing with Raw Material Quality in Colombia (Case 3)

14: A Study of the Market for Cassava Flour in Colombia (Case 3)

15: Applying a Financial Model to Cassava Flour Production in Colombia (Case 3)

16: Costs and Returns in Simple Potato Processing in India (Case 6)

Unit 617: Second-Order Organizations in Colombia (Case 1)

18: Second-Order Organizations in Ecuador (Case 4)

19: Interinstitutional Committees in Brazil (Case 5)

20: Interinstitutional Committees in the Philippines (Case 10)

21: A Three-Tiered System for Distributing Potato Products in India (Case 6)

22: Dividing and Conquering with Potato Products in Peru (Case 7)

23: Distribution of Fresh Cassava in Colombia (Case 2)

24: A Comprehensive Strategy for Project Evaluation in Brazil (Case 5)

25: Versatile Processing Enterprises in Ecuador (Case 4)

Unit 726: Problem Solving at the Pilot Phase in Colombia (Case 3)

v

76

78

8081

8491

91

95969899

101102103108109

115

Unit 3

1:2:3:

4:

Unit 45:

6:7:8:9:10:

11:

Unit 512:

13:

14:

Checklists

Markets for Root and Tuber Crops

Root and Tuber Products

Processes for Transforming Roots and Tubers

Final Screening of Products

Components of an Ideal System for Producing and Commercializing aRoot and Tuber Product

Information Gathered Through Market Research

Questions for Consumer Research

Essential Information on Raw Material Supplies

Factors Determining Raw Material Quality

Main Activities of Technical Research for Product Development

Elements of a Prefeasibility Study

Site Selection Criteria

Elements in the Design of a Pilot Plant

Requirements for Test Marketing

Adding \fllue to Root and Tuber Crops

Page

39404144

53555658596370

77

79

86

Unit 6

15: Information Gathered by a Project Monitoring and Evaluation System

vi

107

Unit 3

1:2:3:

4:

Unit 45:

6:7:8:9:10:

11:

Unit 512:

13:

14:

Checklists

Markets for Root and Tuber Crops




Components of an Ideal System for Producing and Commercializing aRoot and Tuber Product





Main Activities of Technical Research for Product Development





Adding \fllue to Root and Tuber Crops

Page

39404144

53555658596370

77

79

86

Unit 6

15: Information Gathered by a Project Monitoring and Evaluation System

vi

107

Preface

Root and tuber crops, also referred to simply as"root crops," contribute importantly to income andfood security in developing countries. Thesecommodities are grown mainly by small-scalefarmers, and most yield more (in terms of caloriesper hectare per day) than other crops.

There are many opportunities to improvetraditional uses of root crops and introduce theminto a wide range of new food and feed markets,particularly in the rapidly urbanizing societies ofthe developing world. A concerted effort to realizethe promise of these crops could give them a morecentral role in development.

The key to fulfilling the potential of rootcrops is to establish strong links between smallscale producers and new markets. Numerousbarriers now separate them. Root crops areperishable and bulky; capital resources are scarcein rural areas; and organizing market channels iscomplex. Th overcome these obstacles requiresappropriate strategies and technology forpostharvest processing and utilization.

Over the past decade, several nationalagricultural research and development systemsand three international agricultural researchcenters-CIAT, CIP, and the InternationalInstitute of Tropical Agriculture (lITA)-havededicated appreciable resources to lessening thedifficulty of root crop development. This manualdraws lessons from their experience in improvingroot crop utilization and marketing. Based onthose lessons, we present guidelines for thedevelopment of self-supporting agroindustries.

The manual should be useful to anyoneinterested in developing root crops to generateincome and increase supplies of human andanimal food. That includes researchers, extensionofficers, rural entrepreneurs, policy makers,planners, and other staff of government andnongovernment organizations.

The manual is divided into two parts. Part I,consisting of seven units, presents an approach toproduct development and explains how to applyit. In Unit 1 we describe the characteristics of root

vii

crops, analyze trends in their production andutilization, and discuss their potential forcontributing to the socioeconomic advancement ofdeveloping countries. Unit 2 gives an overview ofthe method put forward in this manual,discussing key principles of project design andoutlining the various stages in developing ruralagroindustries: from the generation of productideas, to experimentation and testing of productsand processes on a pilot scale, to production at thecommercial level.

In subsequent units we explain each stage infurther detail. Unit 3 tells how to identify andscreen product ideas. Unit 4 deals with issues inresearch on products and processes, whichprovides information needed for prefeasibilitystudies. Unit 5 explains how to test proposedprocessing technology on a pilot scale. This taskincludes market testing of the products and fullassessment of their commercial feasibility. Unit 6considers key factors determining whetherproducts and processes tested at the pilot levelcan succeed commercially. Units 3-6 contain aseries of checklists designed to help you gatherkey information relevant to particular tasks anddecisions.

In Unit 7 we highlight important issues thatcut across all the stages described in Part I. Thesuccess of any effort to develop root crop products,processes, and markets will depend to a largedegree on how well it addresses those issues.

Part II presents summaries of 10 case studies(selected from a total of 16) on productdevelopment projects in various countries. Inaddition, we draw on the experience of theseprojects throughout the book (mostly in a series ofboxes) to illustrate particular points about thedevelopment of root crop products and processes.We are grateful to Trudy Brekelbaum forpreparing first drafts of the case study summariesand for editing an early draft of the entiremanual.

Many colleagues in national researchsystems and in regional and international

Preface

Root and tuber crops, also referred to simply as"root crops," contribute importantly to income andfood security in developing countries. Thesecommodities are grown mainly by small-scalefarmers, and most yield more (in terms of caloriesper hectare per day) than other crops.

There are many opportunities to improvetraditional uses of root crops and introduce theminto a wide range of new food and feed markets,particularly in the rapidly urbanizing societies ofthe developing world. A concerted effort to realizethe promise of these crops could give them a morecentral role in development.

The key to fulfilling the potential of rootcrops is to establish strong links between smallscale producers and new markets. Numerousbarriers now separate them. Root crops areperishable and bulky; capital resources are scarcein rural areas; and organizing market channels iscomplex. Th overcome these obstacles requiresappropriate strategies and technology forpostharvest processing and utilization.

Over the past decade, several nationalagricultural research and development systemsand three international agricultural researchcenters-CIAT, CIP, and the InternationalInstitute of Tropical Agriculture (lITA)-havededicated appreciable resources to lessening thedifficulty of root crop development. This manualdraws lessons from their experience in improvingroot crop utilization and marketing. Based onthose lessons, we present guidelines for thedevelopment of self-supporting agroindustries.

The manual should be useful to anyoneinterested in developing root crops to generateincome and increase supplies of human andanimal food. That includes researchers, extensionofficers, rural entrepreneurs, policy makers,planners, and other staff of government andnongovernment organizations.

The manual is divided into two parts. Part I,consisting of seven units, presents an approach toproduct development and explains how to applyit. In Unit 1 we describe the characteristics of root

vii

crops, analyze trends in their production andutilization, and discuss their potential forcontributing to the socioeconomic advancement ofdeveloping countries. Unit 2 gives an overview ofthe method put forward in this manual,discussing key principles of project design andoutlining the various stages in developing ruralagroindustries: from the generation of productideas, to experimentation and testing of productsand processes on a pilot scale, to production at thecommercial level.

In subsequent units we explain each stage infurther detail. Unit 3 tells how to identify andscreen product ideas. Unit 4 deals with issues inresearch on products and processes, whichprovides information needed for prefeasibilitystudies. Unit 5 explains how to test proposedprocessing technology on a pilot scale. This taskincludes market testing of the products and fullassessment of their commercial feasibility. Unit 6considers key factors determining whetherproducts and processes tested at the pilot levelcan succeed commercially. Units 3-6 contain aseries of checklists designed to help you gatherkey information relevant to particular tasks anddecisions.

In Unit 7 we highlight important issues thatcut across all the stages described in Part I. Thesuccess of any effort to develop root crop products,processes, and markets will depend to a largedegree on how well it addresses those issues.

Part II presents summaries of 10 case studies(selected from a total of 16) on productdevelopment projects in various countries. Inaddition, we draw on the experience of theseprojects throughout the book (mostly in a series ofboxes) to illustrate particular points about thedevelopment of root crop products and processes.We are grateful to Trudy Brekelbaum forpreparing first drafts of the case study summariesand for editing an early draft of the entiremanual.

Many colleagues in national researchsystems and in regional and international

organizations have contributed to this publicationby preparing case studies, consulting with us, andreviewing the first draft at regional workshopsheld in Latin America, Asia, and Africa (theproceedings of those meetings are valuablecompanions to the manual). In the Appendix welist everyone who helped prepare the manualthrough their participation in the regionalworkshops and other events. We are greatlyindebted to them for sharing so generously theirknowledge and experience.

viii

Adding \blue to Root and Tuber Crops

Much additional information is available inthe literature on product development. We'vetried to include all the major references inbibliographies at the end of particular units.

Preparation of this manual was madepossible by the United Nations DevelopmentProgramme (UNDP), which provided fundingthrough a project entitled Human ResourceDevelopment for Generation and Transfer of Rootand Tuber Crops Technology (GLO/87/001).

organizations have contributed to this publicationby preparing case studies, consulting with us, andreviewing the first draft at regional workshopsheld in Latin America, Asia, and Africa (theproceedings of those meetings are valuablecompanions to the manual). In the Appendix welist everyone who helped prepare the manualthrough their participation in the regionalworkshops and other events. We are greatlyindebted to them for sharing so generously theirknowledge and experience.

viii

Adding \blue to Root and Tuber Crops

Much additional information is available inthe literature on product development. We'vetried to include all the major references inbibliographies at the end of particular units.

Preparation of this manual was madepossible by the United Nations DevelopmentProgramme (UNDP), which provided fundingthrough a project entitled Human ResourceDevelopment for Generation and Transfer of Rootand Tuber Crops Technology (GLO/87/001).

\\

Unit 1

Needs and Opportunities in Product Development

Over the last three decades, many developingcountries have achieved remarkable increases infood production. But decision makers andscientists believe the possibilities for furtherimprovement in productivity, income, andconsumption are far from exhausted. Specialistsin various disciplines are convinced that to reachthis goal will require major efforts to expandutilization of agricultural commodities through aprocess we refer to here as product development.

Though by no means novel, this concept hasaroused new interest in developing countryagriculture on the eve of the 21st century. Productdevelopment is frequently associated withmanufactured goods, with the use of hightechnology, and with developed country economies(Kotler, 1986). As used here, though, the termapplies to food crops-produced with labor- orcapital-intensive techniques-and to markets indeveloping countries (Austin, 1992). Moreover,this manual focuses specifically on cassava.potato, and sweet potato-crops frequentlycharacterized as "traditional" or "subsistence" andnot commonly transformed for sale in a modern,commercial context.

There is little doubt that one can developnew products for root and tubers. But do theresults justify the investment? In this unit weaddress that question by describing variouscircumstances that make product development anattractive and even necessary approach for rootcrops in developing countries.

First, we examine the agronomic andbiochemical characteristics of roots and tubers,highlighting the features that permit, if notrequire, transformation to increase utilization ofthese crops. Then, we analyze production trendsto pinpoint tendencies in output, area, and yieldthat represent opportunities (signaled by growingproduction) and needs (where area planted isdeclining) for alternative uses of these crops.

Subsequently, we discuss recent changes inthe prevailing patterns of utilization, emphasizing

3

how experiences at specific locations might be putto wider use. We briefly mention developments inmarkets beyond those for root and tuber cropsthat, nonetheless, influence trends in these crops.Finally, we comment on trends affecting thefuture prospects of roots and tubers.

Characteristics of Roots andTubers

This manual focuses mainly on the three majorroot and tuber crops----<:assava (Manihotesculenta), sweet potato (Ipomoea batatas), andpotato (Solanum tuberosum). We also give someattention to yams (Dioscorea spp.), cocoyams(taro, yautia = Colocasia esculenta), and tannia(Xanthosoma spp.).

Collectively, these crops occupy about50 million hectares worldwide (Horton et aI.,1984). Annual production exceeds 550 milliontons, about two-thirds of which is harvested in thedeveloping world. Cassava alone is an importantfood crop for some 500 million people indeveloping countries (De Bruijn and Fresco.1989). Cassava, sweet potato, and potato aregrown in roughly 100 of those countries under awide range of growing conditions.

Agronomic traits

Root crops can adapt to diverse environments,partly as a result of their agronomiccharacteristics.

Growing cycle: Potato and sweet potato fitparticularly well into complex cropping systems,because they have a shorter vegetative cycle thanother root crops, such as cassava, whose growingperiod ranges from 9 to 24 months, depending onsoil fertility and ambient temperature (Table 1).

Thmperature: Although potato yields bestunder cool conditions. the crop is grownextensively in areas (e.g., in Tunisia andBangladesh) with high daytime temperatures

Unit 1

Needs and Opportunities in Product Development

Over the last three decades, many developingcountries have achieved remarkable increases infood production. But decision makers andscientists believe the possibilities for furtherimprovement in productivity, income, andconsumption are far from exhausted. Specialistsin various disciplines are convinced that to reachthis goal will require major efforts to expandutilization of agricultural commodities through aprocess we refer to here as product development.

Though by no means novel, this concept hasaroused new interest in developing countryagriculture on the eve of the 21st century. Productdevelopment is frequently associated withmanufactured goods, with the use of hightechnology, and with developed country economies(Kotler, 1986). As used here, though, the termapplies to food crops-produced with labor- orcapital-intensive techniques-and to markets indeveloping countries (Austin, 1992). Moreover,this manual focuses specifically on cassava.potato, and sweet potato-crops frequentlycharacterized as "traditional" or "subsistence" andnot commonly transformed for sale in a modern,commercial context.

There is little doubt that one can developnew products for root and tubers. But do theresults justify the investment? In this unit weaddress that question by describing variouscircumstances that make product development anattractive and even necessary approach for rootcrops in developing countries.

First, we examine the agronomic andbiochemical characteristics of roots and tubers,highlighting the features that permit, if notrequire, transformation to increase utilization ofthese crops. Then, we analyze production trendsto pinpoint tendencies in output, area, and yieldthat represent opportunities (signaled by growingproduction) and needs (where area planted isdeclining) for alternative uses of these crops.

Subsequently, we discuss recent changes inthe prevailing patterns of utilization, emphasizing

3

how experiences at specific locations might be putto wider use. We briefly mention developments inmarkets beyond those for root and tuber cropsthat, nonetheless, influence trends in these crops.Finally, we comment on trends affecting thefuture prospects of roots and tubers.

Characteristics of Roots andTubers

This manual focuses mainly on the three majorroot and tuber crops----<:assava (Manihotesculenta), sweet potato (Ipomoea batatas), andpotato (Solanum tuberosum). We also give someattention to yams (Dioscorea spp.), cocoyams(taro, yautia = Colocasia esculenta), and tannia(Xanthosoma spp.).

Collectively, these crops occupy about50 million hectares worldwide (Horton et aI.,1984). Annual production exceeds 550 milliontons, about two-thirds of which is harvested in thedeveloping world. Cassava alone is an importantfood crop for some 500 million people indeveloping countries (De Bruijn and Fresco.1989). Cassava, sweet potato, and potato aregrown in roughly 100 of those countries under awide range of growing conditions.

Agronomic traits

Root crops can adapt to diverse environments,partly as a result of their agronomiccharacteristics.

Growing cycle: Potato and sweet potato fitparticularly well into complex cropping systems,because they have a shorter vegetative cycle thanother root crops, such as cassava, whose growingperiod ranges from 9 to 24 months, depending onsoil fertility and ambient temperature (Table 1).

Thmperature: Although potato yields bestunder cool conditions. the crop is grownextensively in areas (e.g., in Tunisia andBangladesh) with high daytime temperatures

Adding '\.blue to Root ami Tuber Crops

Table 1. Main characteristics ofroot crops.

Characteristics Cassava Sweet Potato Tannia Taro Vampotato

Growth period (mo.) 9-24 3-7 3-8 9-12 6-18 8-11Annual or perennial plant Per. Ann. Per. Per. Per. Ann.Optimal rainfall (em) 100-150 50-75 75-100 140-200 250 115Optimal temperature (OC) 25-29 15-18 >24 13-29 21-27 30

Drought resistant Ves No Ves No No Ve.Optimal pH 5-6 5.5-6.0 5.6-6.6 5.5-6.5 5.5-6.5 n.a.Fertility requirement Low High Low High High HighOrganic matter requirement Low High Low High High High

Growable on swampy, No No No No Ves Nowaterlogged soilPlanting material Stem Tubers, Vine, Conns! Corms! Tubers

cuttings cuttings connels cormelsStorage time in ground Long Short Long Long Moderate LongPostharvest storage life Short Long ShOl·t Long Variable Long

n.a. =Data not available.

SOURCES: Derived from Kay (1973), as presented in Horton (198H).

(Horton and Monares, 1986; Scott, 1988). Butwhen nighttime temperatures exceed 20 "C,potato does not tuberize well (Midmore andRhoades, 1987). In contrast, sweet potato,cassava, and other root crops are best cultivatedunder higher temperatures.

Rainfall: Potato requires less total rainfallthan cassava, but rain must be fairly continuous,particularly at the beginning and during thetuberization phase of the vegetative cycle.Consequently, drought can devastate potato cropsbut is much less damaging to cassava and sweetpotato.

Crop management: To produce high yields,potato needs ample fertilizer and organic matter.These inputs are less important for cassava andsweet potato, which can give good returns even inpoor soils. In general, cultural practices for thosecrops (e.g., seed handling, weeding, and pestcontrol) are less demanding than for potato.Moreover, potato crops can normally be left in theground for only a short time after their primeharvesting date, whereas cassava can be "stored"in the ground, unharvested, for many months.

Biochemical traits

The fact that root and tuber production isgenerally confined to particular seasons is one

4

major reason why these crops require some sort oftransformation to permit continuous usethroughout the year. Another is their physical andbiochemical characteristics, the most singular ofwhich are their bulkiness and perishability.

Fresh cassava has a dry matter content ofabout 40%, sweet potato 30%, and potato 20%.Physiological deterioration of fresh cassava rootsbegins 1-3 days after harvest (Ospina andWheatley, 1992). Sweet potato and potato have amuch longer shelf life. Even so, their harvestedroots and tubers are living organisms that requireadequate respiration and proper handling toprevent sprouting, spoilage, or pest damage.

If some biochemical traits of roots and tuberspresent serious obstacles, others-much lessfrequently mentioned-offer compelling reasonsto expand utilization of these crops.

Horton (1988) points out that "root crops areoften thought of as 'starchy staples' that providelow-cost energy but little else to the human diet."Contrary to this misleading generalization, thecontent of protein, essential vitamins, andminerals varies considerably among roots andtubers. On average, cooked potato and yam haveabout 2% protein, twice that of cassava (Table 2).Cassava, sweet potato, and potato all provideascorbic acid, whereas cereal-based foods have

Adding '\.blue to Root ami Tuber Crops

Table 1. Main characteristics ofroot crops.

Characteristics Cassava Sweet Potato Tannia Taro Vampotato

Growth period (mo.) 9-24 3-7 3-8 9-12 6-18 8-11Annual or perennial plant Per. Ann. Per. Per. Per. Ann.Optimal rainfall (em) 100-150 50-75 75-100 140-200 250 115Optimal temperature (OC) 25-29 15-18 >24 13-29 21-27 30

Drought resistant Ves No Ves No No Ve.Optimal pH 5-6 5.5-6.0 5.6-6.6 5.5-6.5 5.5-6.5 n.a.Fertility requirement Low High Low High High HighOrganic matter requirement Low High Low High High High

Growable on swampy, No No No No Ves Nowaterlogged soilPlanting material Stem Tubers, Vine, Conns! Corms! Tubers

cuttings cuttings connels cormelsStorage time in ground Long Short Long Long Moderate LongPostharvest storage life Short Long ShOl·t Long Variable Long


SOURCES: Derived from Kay (1973), as presented in Horton (198H).

(Horton and Monares, 1986; Scott, 1988). Butwhen nighttime temperatures exceed 20 "C,potato does not tuberize well (Midmore andRhoades, 1987). In contrast, sweet potato,cassava, and other root crops are best cultivatedunder higher temperatures.

Rainfall: Potato requires less total rainfallthan cassava, but rain must be fairly continuous,particularly at the beginning and during thetuberization phase of the vegetative cycle.Consequently, drought can devastate potato cropsbut is much less damaging to cassava and sweetpotato.

Crop management: To produce high yields,potato needs ample fertilizer and organic matter.These inputs are less important for cassava andsweet potato, which can give good returns even inpoor soils. In general, cultural practices for thosecrops (e.g., seed handling, weeding, and pestcontrol) are less demanding than for potato.Moreover, potato crops can normally be left in theground for only a short time after their primeharvesting date, whereas cassava can be "stored"in the ground, unharvested, for many months.

Biochemical traits

The fact that root and tuber production isgenerally confined to particular seasons is one

4

major reason why these crops require some sort oftransformation to permit continuous usethroughout the year. Another is their physical andbiochemical characteristics, the most singular ofwhich are their bulkiness and perishability.

Fresh cassava has a dry matter content ofabout 40%, sweet potato 30%, and potato 20%.Physiological deterioration of fresh cassava rootsbegins 1-3 days after harvest (Ospina andWheatley, 1992). Sweet potato and potato have amuch longer shelf life. Even so, their harvestedroots and tubers are living organisms that requireadequate respiration and proper handling toprevent sprouting, spoilage, or pest damage.

If some biochemical traits of roots and tuberspresent serious obstacles, others-much lessfrequently mentioned-offer compelling reasonsto expand utilization of these crops.

Horton (1988) points out that "root crops areoften thought of as 'starchy staples' that providelow-cost energy but little else to the human diet."Contrary to this misleading generalization, thecontent of protein, essential vitamins, andminerals varies considerably among roots andtubers. On average, cooked potato and yam haveabout 2% protein, twice that of cassava (Table 2).Cassava, sweet potato, and potato all provideascorbic acid, whereas cereal-based foods have

Table 2. Nutritional composition ofa lOO...g edibteportion ofvarious foods.

Fooda Water Protein Food Protein! Fats Ash Ca P Fe Na K Thiamine Riboflavin Niacin Ascorbic(%) (g) energy calorie (g) (mg) (mg) (mg) (mg) (mg) (mg) (mg) (mg) (mg) acid

(kcal) ratio (mg)(glOOOkcall

Maize (grits) 87 1.2 51 24 0.1 0.6 1 10 0.1 205 11 0.02 0.01 0.2 0

Potato 80 2.1 76 27 0.1 0.9 7 53 0.6 3 407 0.09 0.04 1.5 16

Plantain 80 1.3 77 17 0.1 0.7 - 0

Taro (raw) 73 1.9 98 19 0.2 1.2 28 61 1.0 7 514 0.13 0.04 1.1 4

'" Yam (raw) 74 2.1 101 21 0.2 1.0 20 69 0.6 - 600 0.10 0.04 0.5 9

Rice 73 2.0 109 18 0.1 1.1 10 28 0.2 374 28 0.02 0.01 0.4 0

Spaghetti 72 3.4 111 31 0.4 1.2 8 50 0.4 1 61 0.01 0.01 0.3 0

Sweet potato 71 1.7 114 15 0.4 1.0 32 47 0.7 10 243 0.09 0.06 0.6 17

Common bean 69 7.8 118 66 0.6 1.4 50 148 2.7 7 416 0.14 0.07 0.7 0

I Cassava 68 0.9 124 7 0.1 0.6 - - - - - - - - 26

I\. Fresh white bread 36 8.7 269 32 3.2 1.9 70 87 0.7 507 85 0.09 0.08 1.2 -Trace I)

a. Boiled unless otherwise indicated; edible portions of potatoes and other root crops and plantains do not include peels.b. Dashes denote lack of reliable data.

SOURCES: USDA(1975) and Wu-Leung et al. (1968), as presented in Horton (1988).

Table 2. Nutritional composition ofa lOO...g edibteportion ofvarious foods.

Fooda Water Protein Food Protein! Fats Ash Ca P Fe Na K Thiamine Riboflavin Niacin Ascorbic(%) (g) energy calorie (g) (mg) (mg) (mg) (mg) (mg) (mg) (mg) (mg) (mg) acid

(kcal) ratio (mg)(glOOOkcall

Maize (grits) 87 1.2 51 24 0.1 0.6 1 10 0.1 205 11 0.02 0.01 0.2 0

Potato 80 2.1 76 27 0.1 0.9 7 53 0.6 3 407 0.09 0.04 1.5 16

Plantain 80 1.3 77 17 0.1 0.7 - 0

Taro (raw) 73 1.9 98 19 0.2 1.2 28 61 1.0 7 514 0.13 0.04 1.1 4

Yam (raw) 74 2.1 101 21 0.2 1.0 20 69 0.6 600 0.10 0.04 0.5 9

Rice 73 2.0 109 18 0.1 1.1 10 28 0.2 374 28 0.02 0.01 0.4 0

Spaghetti 72 3.4 111 31 0.4 1.2 8 50 0.4 1 61 0.01 0.01 0.3 0

Sweet potato 71 1.7 114 15 0.4 1.0 32 47 0.7 10 243 0.09 0.06 0.6 17

Common bean 69 7.8 118 66 0.6 1.4 50 148 2.7 7 416 0.14 0.07 0.7 0

I Cassava 68 0.9 124 7 0.1 0.6 26

I\. Fresh white bread 36 8.7 269 32 3.2 1.9 70 87 0.7 507 85 0.09 0.08 1.2 -Trace I)

a. Boiled unless otherwise indicated; edible portions of potatoes and other root crops and plantains do not include peels.b. Dashes denote lack of reliable data.

SOURCES: USDA(1975) and Wu-Leung et al. (1968), as presented in Horton (1988).

none. Sweet potato and potato also contain theimportant amino acid lysine, in whichcommodities such as rice are deficient (Woolfe,1987; 1992).

Furthermore, cassava, sweet potato, andpotato significantly outyield the cereals in drymatter (Le., calorie) production per unit area.Potato is particularly productive in terms ofcarbohydrate per hectare per day (Table 3).

Production Trends

Developing countries produced 149 million tons ofcassava, 122 million tons of sweet potato, and79 million tons of potato in 1988-1990 (Table 4).Between 1961 and 1988, production of the threecrops increased 100%,31%, and 173%,respectively, although output of sweet potatoactually declined by 14% in the latter half of thisperiod. The area planted to sweet potato has alsofallen sharply in recent years, but this was morethan offset by a near doubling of yields. Areaplanted in potato grew steadily over the last threedecades-up 79% since 1961 (more, in fact, thanany other major food crop except soybean andtomatol-even though it started from a muchsmaller base than coarse grains or beans. Area incassava production increased by about 50% andyields by 32%.

Within these overall patterns, which broadlyinfluence the prospects for product development

r------- ---.-

Adding value to Root and Thber Crops

of roots and tubers, production trends varygreatly, not only among commodities, but also forthe same commodity across and even withinregions. Statistics on the evolution of output, areaplanted, and yield therefore merit closer scrutiny.

Regional distribution

Although cassava, sweet potato, and potato alloriginated in Latin America, the bulk ofproduction has shifted away from this region. Asiacurrently accounts for two-thirds of the output ofthese crops in developing countries. The highconcentration of root crops on that continentmainly reflects its large share of sweet potato(93%) and potato production (76%). China aloneproduces over 85% of the developing world's sweetpotatoes and nearly 40% of its potatoes (Tables 5and 6). Africa produces nearly half (44%) thedeveloping world's cassava (Table 7). Since 1961output of cassava and potato has grown muchfaster in Africa and Asia than in Latin America,where production of sweet potato has actuallydeclined by nearly 22%.

Africa has the most producers of cassava(38 countries) and sweet potato (36). Asia has thelargest number of potato-producing countries,with 33, and the heaviest concentration of bigproducers (Table 8). In Latin America there arenumerous minor producers «10,000 t per year) ofcassava (12 countries) and sweet potato (also 12).

Table 3. 1bp-ranking food crops in developing market economies in terms ofdry matter, edible energy, and protein production.

Dry matter production (kg/ha) Energy production (MJlha Protein production (kg/haper day) peT day)

Cassava 3.0 Potatoes 216 Cabbages 2.0Yams 2.4 Yams 182 Dry broad beans 1.6Potatoes 2.2 Carrots 162 Potatoes 1.4

Sweet potatoes 2.1 Maize 159 Dry peas 1.4Rice 1.9 Cabbages 156 Eggplants 1.4

Carrots 1.7 Sweet potatoes 152 Wheat 1.3

ICabbages 1.6 Rice 151 Lentils 1.3Bananas 1.5 Wheat 135 'Ibmatoes 1.2

IWheat 1.3 Cassava 121 Chickpeas 1.1 IMaize 1.3 Eggplants 120 Carrots 1.0

...../SOURCE: HOrWn and Fano (1985).

6

none. Sweet potato and potato also contain theimportant amino acid lysine, in whichcommodities such as rice are deficient (Woolfe,1987; 1992).

Furthermore, cassava, sweet potato, andpotato significantly outyield the cereals in drymatter (Le., calorie) production per unit area.Potato is particularly productive in terms ofcarbohydrate per hectare per day (Table 3).

Production Trends

Developing countries produced 149 million tons ofcassava, 122 million tons of sweet potato, and79 million tons of potato in 1988-1990 (Table 4).Between 1961 and 1988, production of the threecrops increased 100%,31%, and 173%,respectively, although output of sweet potatoactually declined by 14% in the latter half of thisperiod. The area planted to sweet potato has alsofallen sharply in recent years, but this was morethan offset by a near doubling of yields. Areaplanted in potato grew steadily over the last threedecades-up 79% since 1961 (more, in fact, thanany other major food crop except soybean andtomatol-even though it started from a muchsmaller base than coarse grains or beans. Area incassava production increased by about 50% andyields by 32%.

Within these overall patterns, which broadlyinfluence the prospects for product development

r------- ---.-

Adding value to Root and Thber Crops

of roots and tubers, production trends varygreatly, not only among commodities, but also forthe same commodity across and even withinregions. Statistics on the evolution of output, areaplanted, and yield therefore merit closer scrutiny.

Regional distribution

Although cassava, sweet potato, and potato alloriginated in Latin America, the bulk ofproduction has shifted away from this region. Asiacurrently accounts for two-thirds of the output ofthese crops in developing countries. The highconcentration of root crops on that continentmainly reflects its large share of sweet potato(93%) and potato production (76%). China aloneproduces over 85% of the developing world's sweetpotatoes and nearly 40% of its potatoes (Tables 5and 6). Africa produces nearly half (44%) thedeveloping world's cassava (Table 7). Since 1961output of cassava and potato has grown muchfaster in Africa and Asia than in Latin America,where production of sweet potato has actuallydeclined by nearly 22%.

Africa has the most producers of cassava(38 countries) and sweet potato (36). Asia has thelargest number of potato-producing countries,with 33, and the heaviest concentration of bigproducers (Table 8). In Latin America there arenumerous minor producers «10,000 t per year) ofcassava (12 countries) and sweet potato (also 12).

Table 3. 1bp-ranking food crops in developing market economies in terms ofdry matter, edible energy, and protein production.

Dry matter production (kg/ha) Energy production (MJlha Protein production (kg/haper day) peT day)

Cassava 3.0 Potatoes 216 Cabbages 2.0Yams 2.4 Yams 182 Dry broad beans 1.6Potatoes 2.2 Carrots 162 Potatoes 1.4

Sweet potatoes 2.1 Maize 159 Dry peas 1.4Rice 1.9 Cabbages 156 Eggplants 1.4

Carrots 1.7 Sweet potatoes 152 Wheat 1.3

ICabbages 1.6 Rice 151 Lentils 1.3Bananas 1.5 Wheat 135 'Ibmatoes 1.2

IWheat 1.3 Cassava 121 Chickpeas 1.1 IMaize 1.3 Eggplants 120 Carrots 1.0

...../SOURCE: HOrWn and Fano (1985).

6

Table 4. Food crop production in developing countries, 1961-1990.

Crop 1988-1990 Percent change in:-..~--,,- -,--- _.._._~--- ~--- ---

Production Area Yield Production Area Yield(000 tl (000 hal (tJhal

1 2 3 1 2 3 1 2 3~--~

Rice (paddy) 484,379 143,000 3.4 52.3 52.6 132.4 17.7 6.6 25.5 29.5 43.1 85.2Wheat 224,612 100,450 2.2 79.5 94.0 246.3 19.8 12.9 35.3 49.8 71.8 157.4Maize 196,916 63,213 2.4 73.1 65.1 185.9 22.0 14.7 40.0 41.9 43.9 104.3Cassava 149,193 15,074 9.9 41.2 41.3 99.5 26.4 17.8 51.3 9.9 19.9 31.8

I Sweet potato 122,057 9,063 13.5 52.2 -14.1 30.6 -4.1 -24.9 -26.0 58.7 14.3 81.4Potato 79,066 6,363 12.4 73.8 56.8 172.5 31.9 35.7 79.0 31.7 15.6 52.3...,Soybean 48,857 30,255 1.6 130.7 169.2 521.0 27.6 104.7 161.3 80.8 31.5 137.3Banana 44,766 4,004 11.2 43.7 45.4 108.9 41.0 25.7 77.2 1.9 15.6 17.9Sorghum 42,288 38,600 1.1 38.7 3.4 43.4 -0.8 -5.7 .jl.5 39.9 9.7 53.4

lbmato 33,881 1,848 18.3 78.0 118.8 269.4 48.0 59.0 135.3 20.3 37.6 65.5Millets 26,894 34,994 0.8 17.9 2.0 20.3 2.3 -12.5 -10.5 15.3 16.6 34.4Barley 25,156 18,401 1.4 -7.0 35.3 25.9 -15.6 6.8 -9.9 10.2 26.8 39.7Groundnut

in shell 21,907 19,687 1.1 19.0 34.5 60.0 14.8 4.5 19.9 3.6 28.7 33.4Yam 18,083 2,087 8.7 53.6 36.2 109.2 34.3 31.0 76.0 14.4 3.9 18.9

Cabbage 15,774 816 19.3 65.9 71.4 184.3 27.8 18.7 51.8 29.8 44.3 87.3Bean, dry 13,324 23,889 0.6 17.9 24.4 46.7 15.4 15.1 32.9 2.1 8.1 10.4Chickpe~ 6,294 9,204 0.7 -18.3 9.4 -10.6 -16.3 -4.2 -19.8 -2.4 14.2 11.5Broad bean, dry 3,657 2,822 1.3 -22.4 2.6 -20.4 -23.7 -19.9 -38.9 1.6 26.1 30.2Lentil 2,093 2,900 0.7 24.5 108.9 160.0 22.2 61.1 96.8 1.9 29.7 32.1

------~.

a. 1 = 1973-1975 VB. 1961·1963,2 =1988-1990 VB. 1973-1975, and 3 = 1988-1990 V8. 1961-1963,

SOURCE: FAO Basic Data Unit, unpublished statistics.

Table 5. Sweet potato production, area, and yield in selected developing countries, 1961-1990.

CountlY 1988-1990 Percent change in:a

Production Area Yield Production Area Yield

~(000 t) (000 hal (t/ha)

1 2 3 1 2 3 1 2 3

Africab 6,492 1,315 4.9 44.5 -29.7 87.4 73.0 18.3 104.6 -16.4 9.6 -8.4,

Madagascar 479 91 5.3 14.2 81.3 55.6 0 57.8 57.8 -14.2 14.9 -1.4

Sub-SaharanAfricac 6,401 1,311 4.9 46.1 29.5 89.3 73.6 18.3 105.4 -15.8 9.5 -7.9

Uganda 1,718 412 4.2 201.2 3.7 212.4 209.1 -15.1 162.4 -2.6 22.2 19.0Rwanda 817 147 5.5 12.0 43.1 60.2 9.0 81.9 98.2 2.7 -21.3 -19.2Burundi 655 91 7.2 -11.3 57.3 39.4 -0.5 36.5 35.8 -10.9 15.2 2.7Kenya 561 57 9.8 80.0 107.7 273.8 23.4 81.1 123.4 45.9 14.7 67.3Tanzania 526 245 2.1 92.3 29.1 148.3 367.4 70.9 698.9 -58.9 -24.5 -68.9

Asiad 113,380 7.453 15.2 53.7 -15.4 30.1 -8.9 -29.5 -35.7 68.6 20.0 102.4China 104,824 6,304 16.6 58.0 -15.7 33.2 -10.5 -31.8 -39.0 76.5 23.7 118.3

et:; Indonesia 2,118 232 9.1 -20.9 -12.8 -31.0 -26.6 -31.9 -50.0 7.8 27.9 37.9Vietnam 1,920 328 5.9 -4.6 74.6 66.6 -5.1 50.1 42.5 0.5 16.4 16.9India 1,262 159 8.0 56.2 -18.4 27.6 49.5 -26.7 9.7 4.5 11.3 16.3Philippines 675 140 4.8 11.8 -21.2 53.7 64.1 -19.3 32.5 18.8 -2.3 16.0Bangladesh 538 52 10.4 95.0 -17.1 -7.3 17.6 -19.6 -5.4 -4.9 3.1 -2.0South Korea 528 23 22.6 45.2 -68.8 -54.7 18.8 -74.3 -69.4 22.2 21.4 48.4North Korea 501 35 14.4 38.4 57.8 118.3 42.9 48.6 112.2 -3.1 6.2 2.9Papua New

Guinea 460 101 4.6 37.9 12.8 55.6 21.6 12.2 36.5 13.4 0.5 14.0

Latin America" 2,185 295 7.4 14.5 -316 -21.6 12.5 -23.2 -13.6 1.8 -11.0 -9.3Brazil 683 68 10.1 15.5 -59.2 -52.9 6.0 -55.9 -53.2 8.9 -7.6 0.7

Thtal 122,057 9,603 13.5 55.2 -14.1 30.6 -4.1 -24.9 -28.0 58.7 14.3 81.4 ,-_/

a. 1 = 1973-1975 VB. 1961-1963,2 = 1988-1990 vs. 1973-1975. and 3 = 1988-1990vs. 1961-1963.b. Not including South Africa.c. Not including Algeria, Egypt. Libya, Morocco, Tunisia, and SouthAfrica.d. Does not include Israel and Japan but does include Oceania (except Australia and New Zealand).e. Including Central, North (except Canada and the USA), and South America.

SOURCE: FAD Basic Data Unit, unpublished statistics.

Table 6. Potato production, area, and yield in selected developing countries, 1961-1990.

Country 1966-1990 Percent change in:-

Production Area Yield Production Aree Yield(ooot) (000 hal (tJha)

1 2 3 1 2 3 1 2 3------ ~~~------_.. _--- - -- _.~--~----- _._~---_.__._-- --------- --_._-----

Mricah 6,166 650 9.4 75.0 91.1 234.3 66.7 48.0 176.4 -6.3 29.1 21.0

Bubo-SaharanAfricae 2,292 381 6.0 70.3 40.8 139.9 83.3 31.4 140.9 ·7.1 7.2 -0.4

Egypt 1,719 80 21.4 90.7 131.7 341.9 75.0 91.3 234.7 9.0 21.1 32.0Algeria 903 106 8.5 81.3 113.5 297.0 161.4 73.8 354.3 -30.7 22.8 -14.8

Asiad 60,073 4,695 12.8 89.7 58.3 200.4 39.9 48.0 104.3 35.6 8.4 47.0China 31,597 2,801 11.3 99.9 22.5 144.8 41.9 37.6 95.3 40.8 -11.0 25.4India 14,558 925 15.7 82.1 181.1 411.9 41.8 69.7 140.7 29.4 65.6 112.7South Korea 2,042 155 13.2 55.2 82.7 183.6 36.4 54.7 110.9 13.8 18.1 34.4Bangladesh 1,144 117 9.8 127.4 44.8 229.3 49.4 38.2 106.5 52.2 4.8 59.5

'" LatinAmericae 12,877 1,018 12.7 29.2 38.8 79.3 1.4 -1.7 -0.3 27.4 41.2 79.9Argentina 2,658 112 23.6 7.4 57.5 69.2 -30.7 -5.1 -34.2 54.9 66.0 157.0Colombia 2,580 168 15.2 68.5 128.4 284.8 58.0 67.4 161.1 8.0 36.4 47.4Brazil 2,222 183 13.6 37.9 42.9 97.1 -2.6 -14.5 -16.7 41.5 67.2 136.6Peru 1,651 192 8.6 17.1 -2.4 14.2 2.9 ·26.9 -24.8 13.8 33.6 52.0Mexico 1,044 72 14.4 76.5 61.8 185.6 16.9 30.7 52.8 51.0 23.8 86.9Chile 880 60 14.7 -3.3 11.2 7.5 -15.9 -22.4 -34.8 15.0 43.3 64.8Bolivia 695 131 5.3 45.2 -9.9 30.9 10.0 8.3 19.1 32.0 -16.8 9.9

Other countriesThrkey 4,237 191 22.2 55.0 82.5 182.8 29.1 5.5 35.1 21.0 72.9 109.3Iran 1,964 124 16.0 64.6 290.5 526.4 83.9 271.0 508.2 0.4 2.6 3.0

Total 79,066 6,383 12.4 73.8 56.8 172.5 31.9 35.7 79.0 31.7 15.6 52.3-------

8. 1 = 1973·1975 VB. 1961-1963,2= 1988-1990 vs. 1973·1975, and 3 = 1988-1990 va. 1961-1963.b. Not including South Africa.c. Not including Algeria, Egypt, Libya, Morocco, Tunisia, and South Africa.d. Does not include Israel and Japan hut does include Oceania (except Australia and New Zealand).e. Includes Central, North (except Canada and the USA), and South America.

SOURCE: FAO Basic Data Unit. unpublished statistics.

Table 7. Cassava production, area, and yield in se1ected developing countries, 1961-1990.

Country 1988-1990 Percent change in:-

Production Area Yield Production Area Yield(000 t) (000 hal (tlha)

1 2 3 1 2 3 1 2 3---,-,. -------'---- .. _---

Sub-SaharanAfrica" 65,344 8,440 7.7 35.2 49.6 102.3 25.3 18.4 48.3 7.9 26.4 36.4

Zaire 17,333 2,256 7.7 29.1 51.6 95.6 22.4 36.9 67.5 5.5 10.7 16.8Nigeria 16,363 1,504 ILl 30.1 68.6 119.4 23.5 49.4 84.5 5.4 12.9 18.9Tanzania 7,230 671 10.8 54.4 61.5 149.3 35.1 -14.3 15.7 14.3 88.5 115.4Mozambique 4,019 941 4.3 23.1 25.6 54.6 6.7 17.7 25.5 15.4 6.7 23.1Uganda 3,393 375 9.1 125.5 36.2 207.0 82.9 -29.1 29.6 23.2 92.2 136.8Ghana 3,115 364 8.6 50.3 83.2 175.3 67.2 48.8 148.7 -10.1 23.2 10.7Madagascar 2,252 339 6.6 30.5 82.0 135.3 19.5 74.6 lOB.6 9.3 3.2 12.8

.... Angola 1,817 500 3.6 28.6 10.8 42.5 25.2 4.2 30.4 2.7 6.3 9.20

Asia' 52,836 4,013 13.2 61.4 74.7 181.8 27.0 36.5 73.4 27.1 27.9 62.6Thailand 22,424 1,543 14.5 221.3 254.0 1,037.5 283.3 235.4 1,185.6 -16.2 5.6 -11.5Indonesia 16,139 1,341 12.0 6.0 33.3 41.3 -3.9 -7.5 -11.1 10.3 44.1 59.0India 5,308 269 19.8 240.3 -16.7 183.4 42.4 -28.0 2.5 139.0 15.6 176.4China 3,271 230 14.2 99.1 38.5 175.9 93.4 18.2 128.5 3.0 17.3 27.0Vietnam 2,650 293 9.1 -0.7 134.4 132.8 0.4 90.9 91.7 -Ll 22.8 21.5

LatinAmericad 31,013 2,621 11.8 33.4 -2.1 30.6 39.1 ·3.7 33.9 -4.1 1.7 -2.5Brazil 23,209 1,856 12.5 28.7 -lD.1 15.8 37.5 -9.5 24.4 -6.3 -0.7 -7.0Paraguay 3,806 235 16.2 31.4 190.5 281.8 25.9 184.0 232.5 4.4 10.0 14.8

Thtal 149,193 15,074 9.9 41.2 41.3 99.5 28.4 17.8 51.3 9.9 19.9 31.8-----",

a. 1 =1973-1975 vs. 1961-1963,2 =1988-1990 vs. 1973·1975, and 3 = 1988-1990 VB. 1961-1963.b. Not including Algeria, Egypt, Libya. Morocco, Tunisia, and SouthAfrica.c. Does not include Israel and Japan but does include Oceania (except Australia and New Zealand).d. Including Central, North (except Canada and the USA), and South America.

SOURCE: FAO Basic Data Unit, unpublished. statistics.

Unit 1: Needs and Opportunities in Product Development

~-~----~ -----,Table 8. Distribution ofdeveloping countries by volume ofcassava (C), potato (P), and sweet potato (SP) production, 1988-1990.

Production Number ofcountries----------- -~-------

Africa Latin America Asia Thtal-------

C p SP C P SP C P SP C P SP- -- ---------

oor no info. 16 24 18 18 25 25 44 32 40 78 81 83< 10,0001 4 6 10 12 4 12 6 6 8 22 16 30< 50,0001 6 11 11 5 6 13 3 6 4 14 23 16< 250,000 I 8 7 7 4 4 5 4 9 4 16 20 16> 250,000 I 20 6 8 8 8 4 8 12 9 36 26 21

Thtal 54 54 54 47 47 47 65 65 65 166 166 166~--- ~---------_.

SOURCE: FAO Basic Data Unit, unpublished statistics.

Cassava

Five countries (Brazil, Thailand, Zaire, Nigeria,and Indonesia) produce 64% of the developingworld's cassava (Table 7). In Latin America thecrop ranks second in annual production (on afresh-weight basis) among the 19 major foodcrops, though 74% of the regional total isharvested by just one country-Brazil. Zaire andNigeria account for about 50% of Africa'sproduction and Thailand and Indonesia for over70% of Asia's. Trends in cassava production, likethose for potato and sweet potato, have beenhighly uneven across countries.

Among the eight largest cassava producers insub-Saharan Africa, six increased their output by90% or more from 1961 to 1990 (Table 7).According to Dorosh (1989), this growth waslargely a consequence of the crop's "low laborinput requirements, ability to produce a crop ondegraded soils, and drought tolerance." UnlikeAsia, Africa mostly has a low ratio of population toagricultural land. In addition, African farmerstend to engage in a more diverse set of farm andnonfarm occupations. Cassava is an attractiveoption for these growers, because its cultivationrequires relatively little labor and its culturalpractices are fairly flexible. Other factors thataccount for the popularity of this crop are growingpopulation pressure, shorter fallows, and scarcityof fertilizers (norosh, 1989).

Cassava is popular in Thailand because of itsdrought tolerance, stable yield, and flexibleplanting and harvesting dates (Konjing, 1989).Largely to satisfY strong demand for dried

11

cassava in the European Union (EU), thiscountry has expanded production significantly(Phillips, 1979; Sarma and Kunchai, 1989). Inthe Philippines the favorable agronomic traits ofcassava have also contributed to growth inoutput (Cabanilla, 1989). But in contrast withdevelopments in Thailand, growth in thePhilippines has been driven by strong domesticdemand for cassava (the country exports onlynegligible quantities), which is used as food inrural areas and in animal feed andmanufacturing. In Indonesia farmers cater bothto export and domestic markets. Cassava exportsare considerable in terms of volume and value.'But roughly 35% of production goes to localindustries producing starch for humanconsumption (Kasryno, 1989).

In Brazil growers have responded to weakdemand for cassava by switching to other, moreprofitable crops. As a result, production declined10% and area planted by the same percentagebetween 1973 and 1990. More recently, goodprospects for expanding utilization of cassava inNortheast Brazil appear to have slowed, if notreversed, this trend (Ospina and Wheatley,1992). Brazilian experience with cassava andsimilar developments in Colombia and Ecuador(Best and Wheatley, 1990), along with trends inThailand and Indonesia, have stimulatedrenewed interest in utilization of roots andtubers generally.

1. Exports of gaplek (based on dried cassava chips) have fluctuatedbetween 149,000 and 710,000 t since 1970. In 1986,97% of theseexports were to Germany (Kasryno, 1989).

Sweet potato

Although cassava, sweet potato, and potato aregrown in many countries, production isconcentrated in a relatively small number ofthem. For example, the 16 largest sweet potatoproducers account for nearly 97% of totalproduction (Table 5) and also for 98% of thechange in output since 1961.

Recent trends in area planted and productionhave been highly uneven. In some countries(notably Burundi, Kenya, Madagascar, NorthKorea, and Rwanda), sweet potato outputexpanded rapidly over the last decade. Often, thiswas primarily the result of rapid populationgrowth, which increased the pressure onfarmland, as in Rwanda (Von Braun et aI., 1991).Another contributor to expansion was theminimal production costs of the crop and itsability to do well even on marginal soils (Ewelland Kirkby, 1991). In a number of countries,

. however (e.g., Brazil, China, Indonesia, and thePhilippines), sweet potato output and area havefallen since the mid-1970s.

Reasons cited for this include the expansionof infrastructure (mainly irrigation) forproduction of other crops and a switch to highervalue vegetables in response to growth inurbanization, income, and the associated demandfor a more diverse diet (Calkins, 1979; Chin,1989).' In 7 of the 16 largest producers, output ofsweet potato has dropped since 1973-1975. In theabsence of detailed information, most observersattribute this to weak demand or to the lack ofalternative markets (CIP, 1988a; 1988b; 1989b). Ina recent survey, national program scientists citedthese two circumstances as, by far, the mostimportant production constraints (CIP, 1989a).

Potato

Eighteen of the largest producers grow nearly90% of the developing world's potatoes (Table 6)and account for 90% of the increase in productionsince 1961. Most of this growth has taken place in

2. Little research has been dune on the relationship between changesin income and the consumption of fresh sweet potato roots indeveloping countries. Studies in the Philippines (Alkuinu. 1983;Bouis, 1991) and Peru (Collins, 1989) show that this rdation is muchmore complex than previously believed. Consumers' rtJlolpunsedepends on various factors, including their income, place ofresidence. and the sweet potato variety (Watson, 1989).

12

Adding \blu~ to Root and Thber Crop,

China, South Asia (Bangladesh, India, andPakistan), North Africa (Algeria, Egypt, andMorocco), and the Middle East (Iran andTurkey).

In all these regions, the introduction ofimproved, short-duration varieties of wheat andrice has created a new niche for potato in theagricultural calendar. The availability ofirrigation, abundant supplies of cheap labor(particularly in South Asia), and the introductionof improved varieties and chemical fertilizershave permitted rapid increases in yields and areaplanted (Chowdhury and Sen, 1981; Kokab andSmith, 1989; Scott, 1988). The huge increase inColombia's potato output is partly attributable torapid expansion in the processing subsector(Rodriguez and Rodriguez, 1992).

Other contributors to growth in potatoproduction are strong domestic demand for food,tbe desire of most low-income consumers todiversify their diets, modest per capitaconsumption of potato, the lucrative Europeanmarket (which has helped the North Africancountries in particular), and expansion of coldstorage facilities in South Asia. In several LatinAmerican countries, on the other hand, potatoproduction has been hurt by drought, cheapimported cereals, and in Chile a shift to highervalue fruit and vegetable crops for export (Fu,1979; Scott, 1985).

Other less important producers (e.g., Syria)have substantially increased potato output overthe last three decades. In these countries, as inthose mentioned above, policy makers and potatoscientists are increasingly interested indeveloping alternative uses for this commodity toprevent a collapse in prices resulting from abruptsaturation of the domestic market. In a recentsurvey, researchers in national programs citedunstable prices and supplies as the mostimportant constraints to increased potatoproduction (Scott, 1991).

Trends in Consumption andUtilization

Roots and tubers are generally considered foodcrops, first and last. More than half theproduction in fresh form is for human

Unit 1: Needa and Opportunitks in Product J)eVf!lopment

Table 9. Pereent ch'Jl8O' ib utiJi..tion or.......va (C), potatoao (P), and .weet potatoao (SP) in devoloping countrie., 1961-1990.

Utilization 1961·1963' 1973-1975' 1988·1990'--- - --- - -----

C P SP C P SP C P SP---------

Food 69.5 63.4 74.4 68.7 60.7 69.2 71.0 61.4 54.8Feed 13.0 13.0 10.0 14.0 17.0 16.2 12.0 14.8 29.4Processingb 0 2.2 3.8 0.1 3.9 4.0 0.7 5.4 4.4Seed 0 13.3 3.7 0 10.3 2.4 0 9.4 2.9W..te 13.8 7.9 7.0 14.0 7.9 8.2 13.8 8.9 8.6

----------

a. Ttrt.als may not add up to 100 due to roundmg.b. Includes other uses.

SOURCE: FAO Food Balance Sheets, unpublished statistics.

consumption. Nonetheless, other uses of thesecrops are also important, although the share ofoutput devoted to them varies considerablyacross crops, regions, and countries. Patternsin the utilization of sweet potato and, to alesser extent, cassava have changedsignificantly over the last 30 years. Processingfor animal feed, for example, has assumedmajor importance in a number of countries. Asa result of this trend and good prospects forsimilar developments in the future, patterns inthe mean consumption and utilization of rootsand tubers are receiving closer scrutiny.

Estimates of "processing" or "waste" as apercentage of root and tuber production aredifficult to interpret, however. In manycountries waste in the form of damaged rootsand vines is processed or fed to livestock.Production is also lost to physical or autolyticprocesses, microbiological attack, and pestdamage (NAS, 1978). Furthermore, because ofthe higher water content and bulkiness of rootsand tubers, postharvest losses appear to behigher than for cereal crops (Coursey, 1982).Even so, reliable data on production losses arescarce. The information we have is based oninferences (e.g., since sweet potatoes areperishable, a certain percentage of the harvestis lost) or "guesstimates." The statisticsavailable must therefore be interpreted withcaution.

Potato

An estimated 60% of potato production indeveloping countries is for human

13

consumption, 15% for feed, 10% for seed, and 5%for processing; about 10% is lost to waste (Table 9).'

In the tropics potato performs one of fourdifferent functions in the human diet, serving as a1) basic staple, 2) complementary vegetable,3) seasonal vegetable, or 4) delicacy to be consumedon special occasions (Poats, 1983). In mostcountries it performs the third or fourth function;consumers like the taste of potato but cannot affordto eat it daily. Potato is the centerpiece of the dietonly in the highlands of South America and parts ofCentral Africa. The crop is rapidly gainingimportance as a complementary vegetable in SouthAsia (e.g., in Bangladesh, India, and Pakistan),North Africa (e.g., in Egypt and Tunisia), andeastern and southern Africa (e.g., in Kenya,Madagascar, and Rwanda).

Traditional or simple rustic processing ofpotato for human consumption is common in Peru,Bolivia, and to a lesser extent, Ecuador. Typically,potato is solar dried (Yamamoto, 1987). Amodernized version of this practice has been triedin Colombia (Mantilla, 1988) and Guatemala(Esquite and Perez, 1991). Simple processing ofpotato is also practiced to a limited extent inMadagascar (Rasolo et aI., 1987) and is being testedexperimentally in Cameroon (Nave, 1989) andZaire (Ravuna, 1990). Potato is processed at thevillage level in Bangladesh, India, and Pakistan(CTI, n.d.; Sikka, 1988).

The potato is a relatively costly commodity inmost developing countries. One exception is China,

3. According to the FAD Food Balance Sheets. these percentages haveremained constant over time.

where a growing proportion of production goes tonoodle-making, flour, and snack foods(Gitomer, 1987). When value is added to potatothrough transformation, the resulting productis either prohibitively expensive for all but thehighest income households (as are processedfoods) or simply cannot compete with cheapersubstitutes in industrial uses (e.g., as starch)and livestock feed (G6mez and Wong, 1989). Ingeneral, producers have received quitefavorable prices for potato in relation to unitproduction costs, as evidenced by the sharpincrease in area planted in Asia, sub-SaharanAfrica, and parts of Latin America. Only in thelast decade or so, have bumper crops of potatoin a number of countries stirred seriousinterest in developing alternative markets.

In an important exception to the generalpattern, the fast food industry has experiencedremarkable growth in Southeast Asia, CentralAmerica, Mexico, and parts of South America(especially Colombia and Brazil). Strongdemand for potato in this particular marketniche apparently overrides cost considerationsin the short run and in the medium termoffers growers an attractive financial incentiveto expand the local supply for industrial use(Scott et aI., 1992a).

The substantial use of potato for pig feedin China-a purpose served mainly by small,decayed tubers and vines (Gitomer,1987)-accounts in great measure for totalfeed use of the crop. Potato is used only to alimited extent for feed in Africa and LatinAmerica, and this is largely confined to onfarm utilization of unmarketable tubers.

Sweet potato

Patterns in the utilization of sweet potato havechanged markedly over the last three decades,particularly in Asia (Scott et aI., 1992b).Although more than half the output in freshform still goes for human consumption in allregions of the developing world, nearly 40%now serves as animal feed (Table 9). There isalso growing interest in processing swcetpotato for human consumption and industrialuse.

14

Adding \-blue to Root and 7Uber Crops

Much more is known about patterns in theconsumption of cassava and potato than aboutsweet potato (Horton et aI., 1984). The latter isconsidered a "poor man's food" or a survivalcrop in many parts of Latin America, Africa,and Asia (Collins, 1989; Watson, 1989). But it isalso eaten as a seasonal vegetable. Undercertain market conditions, sweet potatocommands an even higher price than potato(Maggi, 1990).

Sweet potato is most commonly boiled indeveloping countries. In China, for example, itis peeled and cooked with rice to make abreakfast porridge; it is also served fried,roasted, or mashed. Sweet potato leaves or"tips" are a delicacy in the Philippines and atcertain times of the year provide an importantsupplementary source of essential vitamins andminerals. 4

Sweet potato processing for humanconsumption is remarkably diverse andwidespread (Woolfe, 1992). Some 5%-10% ofChina's annual production is processed intonoodles, starch, chips, and candy (Tang et aI.,1990). In the Philippines sweet potato is used tomake ketchup, a soft drink, cakes, and candies(Van Den, 1989). Dulce de batata, a cheese-likesweet, is among the most popular dessert dishesin Argentina (Boy et aI., 1989). Substitution ofgrated, fresh sweet potato for imported wheatflour has gained a foothold in the Peruvianbread market (Cavero et aI., 1991).

Wherever sweet potato is produced indeveloping countries, it is almost always used insome form as animal feed (Table 10). Accordingto estimates made by the Food and AgricultureOrganization (FAO), 40% of total output isdevoted to this purpose in China, 35% in Brazil,30% in Madagascar, 17% in Korea, and 5% orless in the remaining 11 of the 15 largest sweetpotato producers. These estimated percentageshave remained stable during the last threedecades in all countries except China (12% in1961-1963) and Korea (2% in 1961-1963). Evenso, recent estimates from China indicate that as

4. Personal communicatiun, Dr. Howarth Bouis, International FoodPolicy Research Institute OFPRIl, Washington, D.C., USA.

Unit 1: Needs and 0pporlunities in Product Development

1B.ble 10. ;U.ofsweetproiato asanjmal feed in Asia. Africa, and Latin America.

Plant partCountryf---------- -----

Asia

Form Animals

BangladeshChina

'ThiwanIndiaIndonesia:

JavaIrian,Jaya

South Korea

Papua New Guinea

Philippines

Vietnam

AfrIca

EgyptKenyaMozambiqueRwandaUganda

LatinAmerlca

ArgentinaBrazil

Dominican Republic

Ecuador

Haiti

Jamaica

I Peru

lVenezuela

n.a. = not available.

VinesRoots

VinesWaste from process·starch, noodlesRootsRoots

Roots, culls, vinesRoots

Roots, cuBs, stored roots

Vines, foliageRootsLeaves, vinesRoots

VinesRootsVines

VinesVinesVinesDamaged roots, vinesSurplus roots, vinesLeaves

Roots, vinesRoots, vinesVinesRootsVinesRootsVinesCulls, roots left in fieldafter harvestRootsVinesRootsVines

Roots, vines

GreenSliced, dried ground,cookedGreen, from silageWaste water

Sliced, driedSun-dried chips

FreshFresh

Fresh, stored,limitedquantity for highcarbohydrate feedSilageFresh storedGreenCooked, dried chips,composite feedn.a.Fresh, sliced. driedn.a.

Green fodderGreen fodderGreen fodderFreshFreshFresh

FreshFreshn.a.FreshGreen, groundFreshGreen fodderFresh

FreshFreshFreshFresh

Fresh

CattlePrincipally pigs but also forcattle, poultryIbidPigs

PigsPigs

CattlePigs

Pigs, composite feedsfor pigs, poultry, otherdomestic animalsLivestockPigsPigsPigs mainly but alsopoultryPigs, water buffaloPigsPigs

CattleCattle, pigsSmall animalsLivestockLivestock, pigsFish

Pigs, cattleDairy and beefcattle

PigsCattlePigs, goats, beefcattleBeef cattle, goatsPigs

PigsPigs, cattle, other farm animalsCattle, pigs, rabbitsFodder for dairy cattle,small ruminantsLivestock

SOURCES: Boy et 81. (1989); CIP 0988a; 1988b; 1989b); Mackay et al. (1989).

15

much as 35% of the country's sweet potato outputnow goes to animal feed'

Explanations for this sharp increase sincethe early 1960s include growth in cerealproduction (which lowers the amount of sweetpotato needed to supplement cereals for humanconsumption); rising demand for meat products(principally pork), for which sweet potato servesas a feed component (Ge, 1992); and changes ingovernment policy (e.g., the introduction of the"responsibility system," which permits the sale ofagricultural surpluses for profit). Anothercontributor was China's bilateral agreement withthe EU, which allowed the country to export up to600,000 t of dried sweet potato chips duty free tomember countries during the 1980s (Calpe, 1992).

When using sweet potato as animal feed,farmers in Africa, Asia, and Latin America mostcommonly give the roots to pigs and the vines tocattle, as indicated in Table 10 (Scott, 1992;Woolfe, 1992). In northern China many farmersslice and then dry the roots before using them aspig feed (Lu et aI., 1989). This type of simpleprocessing is often done in the field. Slicing andthen sun-drying the roots is a well-knownprocedure for producing pig feed in Taiwan(Calkins, 1979.; Tsou et aI., 1989). It has also beenpracticed, though on a more limited scale, in thePhilippines (Palomar et aI., 1989) and Vietnam(Hoang et aI., 1989)6 Virtually all production offeed from sweet potato takes place at the farm orvillage level. Only limited quantities of compositefeeds are produced industrially.

Cassava

About 70% of the cassava produced in Africa andAsia is for human consumption; less than 50% iseaten fresh in Latin America. According to FAOestimates, these percentages have remainedstable over the last three decades (Table 9). Freshcassava is a basic foodstuff for rural households inCentral and West Africa, parts of South Asia,Latin America, and the Far East. But it's also a

5. Personal communication, Pruf. Z. Tunc, Crup Research Inf<t.itutc,Sichuan Academy ofAgricultural Sciences, China.

6. In the PhilippineH, drying and slicing the J"(Juls has apparently nutbeen profitable (Palomar ct al., 19R9).

16

Adeling \blue to Root and 7Uber Crops

high-priced vegetable in the urban markets ofmany of these same regions (Horton et aI., 1984).The form in which people consume cassava variesconsiderably.

In Latin America, Lynam (1989a) hasobserved that the roots are traditionally eaten inone of three principal forms: 1) fresh (boiled orfried); 2) as a roasted flour called farinha demandioca, particularly in North and NortheastBrazil and neighboring territories; and 3) as anunleavened bread, called casabe, in the Caribbeanbasin.7

In sub-Saharan Africa, the roots are a majorstaple, consumed in processed form in many areasand as a vegetable in others (Dorosh, 1989).Cassava leaves are also eaten as a vegetable,particularly in Central Africa. In West Africa thecrop is most commonly consumed as gari, a drygranular meal made from fermented cassava.Dorosh (1989) estimates that "gari may accountfor more than 70% of cassava consumption inNigeria, 40%-50% in Cameroon, 40% ofconsumption in Ghana, 30% in Cote d'Ivoire."Cassava is also consumed in the form of a sundried flour (called lafu.n in southwest Nigeria) anda sticky puree or heavy soup made fromfermented cassava (Nigerian fu.fu). In East Africacassava is commonly made into a flour from driedroots or root chunks.

According to George (1989), cassava isconsumed principally in the form of baked roots inIndia. He also observes that cassava is used insmall amounts "to make chips, flour and sago, atype of wet starch that is roasted, dried, andfinished." In Indonesia cassava roots are eatenboiled, fried, or steamed (Kasryno, 1989); they arealso processed into gaplek (dried cassava chips)and starch.

Cassava has three other important uses indeveloping countries: 1) animal feed for thedomestic market, 2) industrial purposes (such asstarch and glue production), and 3) processinginto dried chips for export.

7. For a dct.ailcd review of traditicmal cassava pmccssing [ur humancunsumption, soc Lancaster ct al. (1982).

Unit 1: Needs and OpportunuU!8 in Product Development

In sub-Saharan Mrica only negligibleamounts of cassava are used for these purposes.8

In Latin America they account for almost half ofthe output, with 37% going to feed and 7.6% toother nonfood uses. About 10% of cassavaproduction in Asia is for local feed and processing.The region also exports 20 million tons (freshweight) annually in the form of driedcassava-nearly the equivalent of regionalproduction-primarily to Europe. Although thedomestic use of cassava for animal feed hasattracted much attention in Asia (Phillips, 1979;Calpe, 1992), it is not nearly as common as inLatin America. There six and a half times morecassava, on a fresh weight basis (11.2 million tonsvs. 1.7 in Asia), is used for feed.

The highly lucrative EU market for driedcassava has given rise to huge productionincreaaes most notably in Thailand. There and inother Asian countries, it has also helped create astrong profit orientation in the cassava processingindustry. The EU market has been the mostdynamic, with annual demand increasing from anaverage of 1.7 million tons offresh roots in 19611963 to 20.3 million tons in 1981-1983 (Sarma andKunchai, 1989).9 Impending changes in tradeagreement&-under negotiation in the UruguayRound of the General Agreement on Tariffs andTrade (GATT}-have diminished the prospects forincreased (if not continued) cassava exports toEurope and other developed countries.

In Latin America the continuation of pricesupports and subsidies on locally produced feedinputs, along with cheap imports (also largely theresult of government policy), have dampened theeffect of mushrooming demand for meat productsas a stimulant to cassava production andprocessing (Lynam, 1989b). The lesson from theseexperiences is that one must be cautious aboutrelying on special trade arrangements tostimulate the emergence and growth of cassavautilization. Partly for that reason, manydeveloping countries are now focusing more onthe internal market for cassava-based products.

8. Ac:cording to the FAO Food Balance Sheets, less than 2% ofcassavaproduction goes to feed use or processing.

9. Thailand alone accounted for 17.6 million tons, China for 1.5 million.and Indonesia 1.1 million in 1981-1983.

17

Over the last 10 years, rural developmentprojects in parts of Colombia, Ecuador, andNortheast Brazil have shown that processing ofdry cassava for concentrated feed rations is aneffective means of stimulating growth (Best andWheatley, 1990; Ospina and Wheatley, 1992). Theextent to which this success can be repeated inother parts of the developing world dependspartly on future developments in a number of keyareas.

Future Prospects

Many factors that lie beyond the realm of cassava,potato, and sweet potato production andutilization will influence the prospects for rootand tuber product development. These factorsinclude demographic patterns, growth in incomes,availability of substitute food and feed sources,evolution of the market for derivative products(e.g., meat and processed foods), governmentpolicies on agriculture and trade, and improvedtechnology for production and processing.

Demographic pattern.

With few exceptions (China, Brazil, India,Indonesia, and the Philippines), populationgrowth rates in the developing countries areexpected to remain well over 2.0% during 19892000 (World Bank, 1990). Since the majority ofhouseholds in these countries are located in ruralareas, this growth will greatly increase pressureon farmland. That in turn will have importantimplications for both the production andutilization of food crops. Farmers will becompelled to place greater emphasis on higheryielding commodities, to bring more marginalland into regular cultivation, and to seek ways ofconverting raw materials into higher valueproducts.

1b meet the rising demand for food-both onand off the farm-rural families will have toexploit the production potential of their crops tothe utmost, partly by reducing postharvest lossescaused by dehydration, spoilage, and pestdamage. Farmers will also have a strong incentiveto convert what cannot be readily sold orconsumed at harvest into marketable products. Inthe search for new products, notes Coursey

(1982), there is great potential for enhancingtraditional practices with knowledge derived frommodern science.

Urbanization will have a strong influence onproduct development. Nearly one in threeconsumers in the developing world now resides inurban areas, and rates of urbanization are two tothree times population growth rates. If roots andtubers are to compete with alternativecommodities (e.g., through reduced transportationcosts), they will have to be processed in largerquantities. Urban consumers will increasinglydemand food items that are easier to prepare andpreserve. They will also shift to a more diversediet that depends less on plants and more onlivestock products, particularly meat. As a result,roots and tubers (which already occupy a niche inthe market for animal feed) could become an evenmore substantial source of ingredients for feedconcentrates.

Increased incomes

This has a more complex effect on the outlook forproduct development. Where incomes increaserapidly, fresh roots and tubers become lessattractive to certain types of consumers, whileprocessed food products, such as French-friedpotatoes, become more affordable. In addition,rising incomes typically increase the demand forlivestock products. Unable to supply enough feedby expanding cereal production, many developingcountries have satisfied this demand withprocessed roots and tubers. China is a primeexample (Gitomer, 1987).

Government policies

In many countries governments have drasticallyaltered their policies on food and feed importsover the last few years. This is partly the result ofchanges in world markets but is also a way tocope with debt burdens and to create moreopportunities for domestic agriculturalproduction. In the years ahead, changes in policywill strongly influence the potential of root andtuber crops as processed products.

It will take time for some of theagrobiological and socioeconomic factors discussed

18

Adding Value tu Root and Thber Crops

here to exert their full impact. In the meantimethe countries with the best prospects forexpanding the use of roots and tubers asprocessed products appear to be the ones thatalready have a substantial supply of thecommodity; are experiencing shortages of food,feed, or both; and cannot (for economic or politicalreasons) continue or expand imports of food orfeed. A number of countries in Asia and some inLatin America appear to meet these criteria, atleast for increased use of roots and tubers asanimal feed. In China, for example, there are goodprospects for morc intensive utilization of sweetpotato (Gitomer, 1987) and in Brazil for cassava(Ospina and Wheatley, 1992). In addition,markets for processed potato appear to beemerging in Central America and Southeast Asia.

Technical change

Improvements in the yield, dry matter content,and digestibility of cassava, potato, and sweetpotato should make them more attractive asprimary materials for processed products.

In developing countries average potato andcassava yields are well below their potential.Potato yields are about half those in mostdeveloped countries, and cassava yields in 8ubSaharan Africa are half those in Asia (Table 7).Average yields of sweet potato have doubled indeveloping countries over the last 25 years,mainly because of increases in China (Table 1).The increase has apparently resulted more fromimproved cultural practices (mainly higher plantdensity) than from improved varieties or chemicalfertilizers and pesticides (Mackay, 1989). That isto be expected, since until recently only limitedresources for research and development havebeen committed to sweet potato."

Most varieties of this crop cultivated indeveloping countries have a dry matter content ofaround 30%. Results of research at the AsianVegetable Research and Development Center(AVRDC) show that "the mean dry matter contentufbreeding lines improved from 25.9% to 35.1% infive years. Theoretically this program increased

IO. RdatiVl~ tH thu value of prnduetion, funding fur o;wcet potatoresearch worldwide has been lower than for any other major foodcommudity (Gregory et al. 19R9).

Unit 1: Needs and Opportunities in Product Development

chip yield for animal feed by 40%" (Tsou et aI.,

1989). The international collection of sweet

germplasm includes varieties whose dry matter

content is as high as 45%." Materials with similar

advantages are available in the cassava and

potato collections. Such varieties have much

potential for making root and tuber crops better

suited to processing.

References

Alkuino, J. 1983. An econometric analysis of the

demand for sweet potatoes in the Philippines.

Unpublished Ph.D. diss., Department of

Agricultural and Resource Economics, University

of Hawaii, Manoa, USA

Austin, J. 1992. Agro-industrial project analysis. 2d ed.

Johns Hopkins University Press, Baltimore, MD,

USA

Best, R. and Wheatley, C. 1990. Dried cassava for

animal feed: A case study of the Colombian

experience. CIAT, Cali, Colombia.

Bows, H. 1991. Potato and sweet potato demand

elasticities for Bangladesh, Pakistan, and the

Philippines: Impacts of price and income changes

on consumption. International Food Policy

Research Institute (lFPRl), Washington, DC,

USA.

Boy, A.; Cantos, F.; Fano, H.; and Fernandez, F. (eds.).

1989. La batata en la Argentina. Resumenes y

conclusiones del Taller sobre la Producci6n y Uso

de la Batata. Instituto Nacional de Tecnologia

Agropecuaria (INTA), Buenos Aires.

Cabanilla, L. 1989. Trends and prospects for cassava in

the Philippines. In: Sarma (1989).

Calkins, PH. 1979. Production, distribution and final

uses of sweet potato in Taiwan. In: Plucknett

(1979).

Calpe, C.A. 1992. Roots, tubers and plantains: Recent

trends in production, trade, and use. In: Machin,

D. and Nyvold, S. (eds.). Roots, tubers, plantains

and bananas in animal feeding. Proceedings of

the FAD expert consultation. Food and

Agriculture Organization, Rome, Italy.

11. Personal communication with Dr. Zosimo Huaman, Department (,f

Genetic Resources, elP.

19

Cavero, W.; Chumbe, v.; and Peralta, P 1991. Estudio

sabre producci6n y consumo de pan de camote.

Instituto Nacional de Investigaci6n Agraria y

Agroindustrial (lNIAA), Lima, Peru.

Chin, M.S. 1989. The outlook for sweet potato in

Korea. In: CIP (1989b).

Chowdhury, S.K and Sen, A. 1981. Economics of potato

production and marketing in W. Bengal.

Agroeconomic Research Center, Visua-Bharati,

West Bengal, India.

CIP (Centro Internacional de la Papa). 1988a.

Improvement of sweet potato (Ipomoea balalas)

in East Africa. Report of the Workshop on Sweet

Potato Improvement in Africa. Lima, Peru.

___. 1988b. Mejoramiento de la batata (Ipomoea

batatas) en Latinoamerica. Memorias del

Seminario sobre Mejoramiento de la Batata

(Ipomoea batatas) en Latinoamerica. Lima, Peru.

___. 1989a. Annual report CIP 1989. Lima, Peru.

___. 1989b. Improvement of sweet potato

(Ipomoea balalas) in Asia. Report of the

Workshop on Sweet Potato Improvement in Asia.

Lima, Peru.

Col1ins, M. 1989. Economic analysis of wholesale

demand for sweet potatoes in Lima, Peru.

Unpublished M.Sc. thesis, Department of

Agricultural and Resource Economics, University

of Florida, Gainesville, USA.

CTI (Compatible Technology, Inc.). n.d. Solar potato

drying for small or cottage entrepreneurships.

Society for Development of Appropriate

Technology, Bareilly, Uttar Pradesh, India.

Coursey, D.G. 1982. Traditional root crop technology:

Some interactions with modern science. In:

Feeding the hungry: A role for postharvest

technology? Institute for Development Studies

Bulletin, vol. 13, no. 3. Sussex, UK.

De Bruijn, G. and Fresco, L. 1989. The importance of

cassava in world food production. Netherlands

,Journal of Agricultural Science 37:21-34.

Dorosh, P. 1989. Economics of cassava in Africa. In:

Sarma (1989).

Esquite, A. and Perez, G. 1991. Estudio exploratorio de

la papa deshidratada en Guatemala. In: Scoot, G.

and Herrera, J. (eds.). Mercadeo agricola:

Metodologias de investigacion. Selected papers

presented at the Latin American Workshop on

Methods for Agricultural Marketing Research.

CIP and the Instituto Interamericano de

Cooperaci6n para la Agricultura, Lima, Peru.

Ewell, P. and Kirkby, R. 1991. Roots, tubers and beansin the food systems of Eastern and SoutheasternAfrica. In: Veelbehr, E. (ed.). Dialogue andtraining for the promotion of roots, tubers andlegumes in Africa. German Foundation forInternational Development, Feldafing, Germany.

Fu, G. 1979. Producci6n y utilizacion de la papa enChile. eIP, Lima, Peru.

Ge, L.W. 1992. Sweet potato in China. In: Scott et al.(1992b).

George, ~S. 1989. Trends and prospects for cassava inIndia. In: Sarma (1989).

Gitomer, C. 1987. Sweet potato and white potatodevelopment in China. A compendium of basicdata. IFPRI, Washington, DC, USA.

Gomez, R. and Wong, D. 1989. Procesados de papa: Unmercado potencial. Cuadernos de Investigacion,no. 11. Universidad del Pacifico, Lima, Peru.

Gregory, :P.; Iwanaga, M.; and Horton, D. 1989. Sweetpotato research at the International PotatoCenter. In: CIP (1989b).

Hoang, \':1'.; Hoanh, M.T.; Tien, T.N.; and Truong, \':H.1989. The sweet potato in Vietnam. In: CIP(1989b).

Horton, D. 1988. Underground crops. Long-term trendsin production of roots and tubers. WinrockInternational, Morrilton, AK, USA.

Horton, D. and Fano, H. 1985. Potato atlas. CIP, Lima,Peru.

Horton, D. and Monares, A. 1986. A small effectiveseed multiplication program: Thnisia. SocialScience Department Working Paper 1984-2. CIP,Lima, Peru.

Horton, D.; Lynam, J.; and Knipscher, H. 1984. Rootcrops in developing countries. An economicappraisal. In: Proceedings Sixth Symposium forthe International Society for Tropical Root Crops.CIp, Lima, Peru.

Kasryno, F. 1989. Trends and prospects for cassava inIndonesia. In: Sarma (1989).

Kay, D.E. 1973. Root crops. Tropical Products Institute,London, UK

Kokab, A. and Smith, A. 1989. Marketing potatoes inPakistan. Pakistan-Swiss Potato DevelopmentProject, Pakistan Agricultural Research Council,Islamabad.

Konjing, C. 1989. Trends and prospects for cassava inThailand. In: Sarma (1989).

20

Adding \blue to R()(Jt and Tuber Crops

Kotler, P. 1986. Principles of marketing (3rd ed.).Prentice-Hall, Englewood Cliffs, NJ, USA.

Lancaster, PA.; Ingram, J.S.; Lim, M.Y.; and Coursey,D.G. 1982. Traditional cassava-based foods:Survey of processing techniques. EconomicBotany 36: I 2-45.

Lu, S.Y.; Xue, Q.H.; Zhang, D.P.; and Song, B.F. 1989.Sweet potato production and research in China.In: CIP (1989b).

Lynam, .J. 1989a. The evaluation of cassavaconsumption in Latin America. In: Sarma (1989).

1989b. The meat of the matter: Cassava'spotential as a feed source in tropical LatinAmerica. In: Sarma (1989).

Mackay, K.T. 1989. Sweet potato, small farmers andneed for co-operative research. In: Mackay et al.(1989).

Mackay, KT.; Palomar, M.K; and Sanico, R.T. (eds.).1989. Sweet potato research and development forsmall farmers. Proceedings of the InternationalSweet Potato Symposium. Southeast AsianRegional Center for Agriculture, Leyte,Philippines.

Maggi, C. 1990. La comercializacion de batata enArgentina: Un estudio basado en informacion delMercado Central de Buenos Aires. INTA, Institutode Economia y Sociologia Rural (lSER), BuenosAires.

Mantilla, J. 1988. Informe tecnico: Comercializaci6n deproductos procesados de papa en la zona dePamplona (Colombia). Memorias de la ReunionAnual. Programa Andino de Investigacion enPapa (PRAClPA), Lima, Peru.

Midmore, D. and Rhoades, R. 1987. Application ofagrometeorological principles to potato productionin warm climates. Acta Horticulturae 214:103-136.

NAS (National Academy of Sciences). 1978.Postharvest food losses in developing countries.Washington, DC, USA.

Nave, R. 1989. Pilot project for processing root andtuber crops in the Northwest Province ofCameroon. Report on the mission. FAO, Rome,Italy.

Ospina, B. and Wheatley, C. 1992. Processing ofcassava tuber meals and chips. In: Machin, D.and Nyvold, S. (eds.). Roots, tubers, plantainsand bananas in animal feeding. Proceedings ofthe FAO expert consultation. FAO, Rome, Italy.

Unit 1: Needs and Opporlunuies in Product Development

Palomar, M.K; Bulayog, E.F.; and Van Den, T. 1989.Sweet potato research and development in thePhilippines. In: CIP (1989b).

Phillips, T. 1979. The implications of cassavaprocessing and marketing for other root crops. In:P1ucknett (1979).

Plucknett, D. (ed.). 1979. Small-scale processing andstorage of tropical root crops. Westview Press,Boulder, CO, USA.

Poats, S. 1983. Beyond the farmer: Potato consumptionin the tropics. In: Hooker, W.J. (ed.). Research forthe potato in the year 2000. CIF, Lima, Peru.

Rasolo, F.; Randrianaivo, D.; Ratovo, H.; Andrianovosa,D.; Adriambahoaka, D.; Razafindraibe, R.;Rakotondramanana; and Scott, G. 1987. Lapomme de terre pour l'autosuffisance alimentairea Madagascar. FOFIFA-FIFAMANOR-CIP.

Ravuna, M. 1990. Project de transformation de pommede terre en vue d'augmenter la production et lesrevenus des fermiers de Lubero, Nord-Kivu.Memoire. Institut Superior de DeveloppementRural, Bukavu, Kivu, ZaIre.

Rodriguez, P. and Rodriguez, A. 1992. Algunos aspectosde la industrializaci6n de la papa en Colombia.Revista Papa 5:4-7.

Sarma, J.S. (ed.) 1989. Summary proceedings of aWorkshop on Trends and Prospects of Cassava inthe Third World. IFPRI, Washington. DC, USA.

Sarma, J.S. and Kunchai, D. 1989. Trends andprospects for cassava in the Third World. In:Sarma (1989).

Scott, G. 1985. Markets, myths and middlemen: A casestudy of potato marketing in central Peru. CIP,Lima.

___. 1988. Marketing Bangladesh's potatoes:Present patterns and future prospects. CIP andthe Australian Development Assistance Bureau,Dhaka, Bangladesh.

___. 1991. CIP's mission and moleculartechniques for germplasm enhancement: Somestrategic considerations for future impact. In:Report of the planning conference Application ofMolecular Thchniques to GermplasmEnhancement. Lima, Peru.

___. 1992. Sweet potato for animal feed indeveloping countries: Present patterns and futureprospects. In: Machin, D. and Nyvold, S. (eds.).Roots, tubers, plantains and bananas in animalfeeding. Proceedings of the FAO expertconsultation. FAO, Rome, Italy.

21

Scott, G.; Herrera, J.E.; Espinola, N.; Daza, M.;Fonseca, C.; Fano, H.; and Benavides, M. (eds.).1992a. Desarrollo de productos de raices ytuberculos, vol. II-America Latina. Memorias delTaller sobre Procesamiento, Comercializaci6n, yUtilizaci6n de Raices y Tuberculos en AmericaLatina. CIP, Lima, Peru.

Scott, G.; Wiersema, S.; and Ferguson, P.I. (eds.).1992b. Product development for root and tubercrops, voL I-Asia. Proceedings of the InternationalWorkshop. CIP, Lima, Peru.

Sikka, B. 1988. Marketing of processed potato productsin Delhi. Agro-Economic Research Center,Himachal Pradesh University, Shimla, India.

Tang, Z.; Li, R.; Lin, L.; Wan, Y; Fu, M.; Song, B.; andWiersema, S. 1990. Sweet potato processing andutilization in China. Annual project progressreport. CIP, Lima, Peru.

Thou, S.C.S.; Kan, KK; and Wang, S.J. 1989.Biochemical studies on sweet potato for betterutilization at AVRDC. In: Mackay et al. (1989).

USDA (US Department of Agriculture). 1975.Composition of foods. Washington, DC.

Van Den, T. 1989. New developments in processingsweet potato for food. In: Mackay et al. (1989).

Von Braun, J.; de Harn, H.; and Blanken, J. 1991.Commercialization of agriculture underconditions of population pressure: A study inRwanda on production, consumption, andnutritional effects, and their policy implications.IFPRI, Washington, DC, USA.

Watson, G.A. 1989. Sweet potato production andconsumption surveys: Variability and varieties.In: CIP (1989b).

Woolfe, J. 1987. The potato in the human diet.Cambridge University Press, UK.

___. 1992. Sweet potato: An untapped foodresource. Cambridge University Press, UK.

World Bank. 1990. World Development Report, 1990.Oxford University Press, New York, NY, USA.

Wu-Leung, W-T.; Busson, F.; and Jardin, C. 1968. Foodcomposition table for use in Africa. USDepartment of Health, Education, and Welfare,Public Health Service, Bethesda, MD, USA.

Yamamoto, N. 1987. Potato processing: Learning froma traditional Andean system. In: Report of theThird Social Science Planning Conference. CIP,Lima, Peru.

Unit 2

An Integrated Approach to Product Development

This unit gives an overview of productdevelopment, based on the authors' experiencein root and tuber projects over the last decade.Here we focus on general concepts underlyingthe whole process. In Units 3-6, we providedetails on each of its four phases.

In this unit we first define productdevelopment and explain its relationship to theconcept of integrated projects for roots andtubers. Then, we explain the importance ofclearly spelling out objectives, constraints, andother key considerations in project design.Another factor is the choice of beneficiaries. Butbecause of its special importance, we treat thisissue in a separate section of the unit. Next, wetalk about integration of project activities, whichis key to the success of product development.Then, we briefly describe four distinct phases inthat process, setting the scene for Units 3-6. Weconclude with some final tips on project design.

Product Development andIntegrated Projects

'Ib realize the sizable potential of root and tubercrops for contributing to socioeconomicdevelopment in rural areas requires acombination of efficient, sustainable cropproduction with new or improved products andmarkets. Projects that work toward this goalshould be organized on a scale that isappropriate for developing country farmers.Most of these growers have small or, in somecases, medium-sized operations, limited accessto processing technology, and weak links tonontraditional markets.

Product and process development (oftenabbreviated to product development) may bedefined as generating ideas for new or improvedproducts, selecting the best ideas, anddeveloping these into commercially successfulproducts. The entire procedure consists of fourdistinct phases:

23

• Identification of product ideas

• Research

• Pilot phase

• Commercial phase

Product development for root and tuber cropsrequires a high level of commitment, broad socialparticipation, and effective administration. Youneed to take particular care in designing theorganizational strategy. Its ultimate goal is totransfer the administration of a self-sustainingagroindustrial enterprise to competent farmerorganizations or agroentrepreneurs. If the projecthas been well conceived and executed, theenterprise should contribute to ruraldevelopment. 'Ib reach that goal requires activecooperation among the various public and privateinstitutions that provide support in technologygeneration, product design, extension, training,farmer organization, credit, marketing, andproduct promotion.

As shown in Figure I, product developmentis a key part of the broader concept of integratedprojects for production, processing, and marketingof root and tuber crops (abbreviated here tointegrated projects). Such projects combineresearch and development in those three areaswith other activities to achieve community-basedrural development in a specific region. Integratedprojects take place in four phases:

• Macroplanning

• Microplanning

• Pilot phase

• Commercial phase

The purpose of macroplanning at thenational level is to ensure that the mostappropriate region of a country and the mostpromising markets are selected. Throughmicroplanning the project then gathers enoughinformation to define market characteristics,production practices, constraints, and so forth inthe target region.

Adding \blue to Root and TUber Crops

Productdevelopment

Integratedprojects

Research anddevelopment forcrop production

/

Diagnosis

Componentresearch

/

'Thchnologyadaptationand validation

_.

--~

'Thchnologydiffiuion

________-,'1'--- __~1/-

al

on

-----_.- ------ --------

on

on

ngand

anning

alon

technicaland

anning

Identification of Macroplproduct idea

Reaearch Micropl

I~ 1'\"

- Farmer

Pilot phase organizati

- Institutioncoordinati

---~-- Farmer

organizati

Commercial - Institution

phase coordinati

- Credit,assistanceinputs

- Monitorievaluation

_..._--

Figure 1. The relationship between product development and integrated projects for root and tuber crops.

24


Integrated projects have proven highlysuccessful in facilitating the agroindustrialtransformation of crops. They accomplish this bylinking improved production technologies withproduct development and social organizationthrough interinstitutional cooperation. You cannotachieve product development throughtechnological innovation alone. It also requiresfarmer organization and training (technical,administrative, and organizational), provision ofcredit, establishment of efficient distributionchannels, and interinstitutional coordination.These activities are especially important whenproduct development reaches the commerciallevel. At that point they can best be carried outthrough links with integrated projects.

some overlap between their aims, the project isdoomed to failure. In finding a common groundamong participants, you need to determinecarefully the relative weight the project will giveto different objectives, since this issue can be amajor source of conflict. The goal of generatingforeign exchange, for example (which issometimes the overriding national priority), maybe difficult to reconcile with the objective ofimproving the well-being of small-scale farmers.

Th avoid difficulties at this early stage,project participants must state their ownobjectives explicitly and then agree upon the mainobjective of the project. Below we list three maingroups of actors and describe the objectives towhich they generally assign high priority:

Integrated projects of the sort described hereought not be confused with the well-knownintegrated rural development projects sponsoredby the World Bank. The former operate in smallerareas and have more limited objectives. Moreimportant, they are distinguished by theiremphasis on technology-as opposed to theinfusion of capital for infrastructure oroperations-as the key means of triggeringdevelopment. Technological innovations allowrural people to benefit from previously underusedresources, especially local ones. No doubt, productdevelopment and integrated projects on roots andtubers should be part of some integrated ruraldevelopment projects.

The Elements of Project Design

Good project design starts with a thoroughanalysis of the local situation: the actors and theiragendas, constraints, and opportunities. Havingidentified constraints, you should then determinewhich ones lend themselves to change. Based on apreliminary assessment of possible solutions, youcan start to determine the project's institutionalrequirements and technological and marketoptions.

•

•

•

National research and developmentinstitutions seek policies and solutions thatcontribute to the country's socioeconomicdevelopment by increasing foreign exchangeearnings, permitting import substitution, orboth. These institutions may seek to slow orreverse migration to urban centers, increasethe incomes and well-being of small-scalefarmers, lower the cost of basic foods, andimprove the nutritional status of thepopulation.

Local development and extension agenciesand farmer groups work at the regional levelto provide solutions to problems affectingparticular segments of the rural population.They may seek to provide employment forlandless laborers, increase the incomes ofsmall-scale farmers by introducing new cropsor diversifying markets for traditional ones,reduce postharvest losses, protect theenvironment, and improve the position ofwomen and youth in agriculture.

Commercial enterprises are primarily profitoriented. They may consider it in their longterm interest to promote concern for theenvironment and help improve the generalwell-being oflocal communities.

The actors and their objectives

Product development involves a variety ofactors-from farmers to scientists-who oftenhave quite different objectives. Unless there is

25

Anticipating constraints

Every project is faced with an array ofconstraints. If not dealt with specifically, these

may make it difficult, if not impossible, for theproject to achieve its objectives. Here we divideconstraints into five broad categories:

• The institutional mandates of participatingorganizations may impose limits with respectto location and target group, farm size, typeof processing organizations involved (co-op,small business, or agrarian reform group),and crops or raw materials used. Thisproblem is especially serious when themandate does not permit an institution toengage in technology transfer at the locallevel.

Adding Value to Root and Tuber Crops

Who Benefits and How?

Product development projects are undertaken tobenefit target groups of people. The technologyemployed is a means toward that end, not an endin itself. Benefits may be:

• Economic (increased income)

• Nutritional

• Time-saving (e.g., beneficiaries can spendtime saved in processing on other incomegenerating or family-related activities).

•

•

•

Financial constraints affect both institutionsand farmers or processors. Institutions maylack the necessary research facilities,technical expertise, and operational funds.Small-scale farmers and processors rarelyhave much capital, generally do not meet theconditions for credit, and may be unwilling torisk their land in a new venture.

Legal constraints include restrictions on theuse of public funds for certain target groups,licencing requirements, and the need tocomply with government regulations onproduct nutritional and sanitary standards,which may be designed more for urban thanrural conditions.

Infrastructure may not be adequate fortransporting raw material to the processingplant, or there may be no appropriate marketstructure for the processed product.

•

•

•

•

•

•

•

•

These benefits accrue to various groups:

Small-scale farmers who are directlyinvolved in the project

Small-scale farmers who supply raw materialfor processing

All farmers, regardless of farm size or capitalholdings

Rural intermediaries and assembly agents

Urban wholesalers

Rural and/or urban retailers

Rural and/or urban consumers (divided byincome strata)

The rural community in general (throughimproved infrastructure, spillover effects inthe rural economy, and stronger communityspirit)

• Common environmental constraints areunreliable rainfall, low soil fertility, and highambient relative humidity.

Seizing opportunities

In their initial assessment, project planners tendto focus on the negativc--Le., on problems andlimitations in a given country or region.Obviously, you should also be on the lookout foropportunities in the form of undevelopedresources and institutions that are not directlyinvolved in the project but might contributeknowledge, expertise, and other resources.

26

Among farmers and consumers, you canfurther characterize project beneficiaries bygender and age (e.g., dependant children may bethe primary beneficiaries of a project focusing onnutritional improvement).

The following developments can generatemonetary benefits:

• Improved prices for raw material (freshroots) benefit all farmers producing the crop.

• The labor required for processing, marketing,etc., provides a source of income for peopledirectly involved in the project.


•

•

Profits generated by the enterprise benefitits stakeholders.

Reduced prices for the final product benefitconsumers and can lead to increased demandfor the product, benefiting farmers andprocessors.

presented in this manual is its emphasis onweeding out bad product ideas at an early stage.Despite this precaution, though, the product andprocess may still fail, even in the pilot phase. Butsince the venture is operating on a small scale atthat point, the resulting losses should be minimal.

• Improving the quality of the final productbenefits consumers and, if demand rises,farmers and processors as well.

Institutions participating in the project mayalso benefit, as their staff gain more experience,which they can apply in other projects.

In addition to analyzing potential payoffs,you have to confront the possibility of failure.Should this happen, who will suffer and howmuch? One of the strengths of the method

Choosing target beneficiaries

It may not be possible to benefit all needy groupsat the same time. For example, unless marketingmargins are reduced, you cannot help farmersthrough higher prices for fresh roots andsimultaneously aid consumers through lowerprices for the end product. Target beneficiariesmust therefore be selected with care. A project inPeru dealt with this issue (as described in Box 1)

by shifting its market focus from low-income

Box 1

Defining the Benefi.ciaries ofaPotato Project in.P·eru (Case 7)

The original objective of Centro IDEAS inPeru was to benefit small-scale potatofarmers and low-income consumers,primarily those participatingin •st~w foodprograms. A pilot plant was set up toproduce a multiple-use food product basedon a mixture of precooked p.otatoanduncooked cereal (rice,barley, and maizE!)and legume (broad bean) flours. The:rnostimportant property of this natural :produet(intended as a substitute for rice, noodles,and flours) was its nutritional quality.Project planners carried out little marketresearch, since they aimed to introduce theproduct through state institutions.

A number of problems arose.. Theprocessed product had a higher price th~nthe substitutes, and certain characteristics(e.g., bitterness, nonuniform pres.entation,and long preparation time) made it lessacceptable. Low-income consumerspreferred a less expensive product (gr~insor simple flours), even ifit wasalsolessnutritious. Given these circumstaMes,farmers had no incentive to participate inthe project.

27

Consequently, project planners had toabandon their initial goal of producing alow-cost, high-quality nutritional productand at the same time benefiting potatofarmers. Instead, they decided to develop amore sophisticated product, which could besold to higher income groups throughsupermarket chains. But in attempting todo so, the project ran into cash flowproblems and was unable to utilize itsprocessing capacity properly. To remedy thelatter difficulty, the project had to modifythe pilot plant and purchase differentequipment. Meanwhile, cottage industriesadopted the project's idea and purchasedabout 50 hammer mills from themanufacturing firm that had built theproject's original equipment.

If project planners had conducted aprefeasibility study and analyzed themarket, they would have realized that inthis case the goal of benefiting smallfarmers was incompatible with that ofhelping low-income consumers. They mightalso have built the pilot plant on a moreappropriate scale.

consumers, who were very price conscious, tohigh-income consumers, who could pay apremium for a quality product.

Size and distribution ofbenefits

The size and distribution of benefits depends inpart on the type of organization d~ing theprocessing. Cooperatives tend to emphasize socialbenefits for their members but may also try toimprove the welfare of other groups (such asnonmember farmers). The central goals of smallbusiness, on the other hand, are to capitalize theenterprise, improve the efficiency of itsoperations, and maximize profits for individuals.In the long run, cooperatives need to strike abalance between the potentially conflicting goalsof redistributing profits to members andnonmembers (by maintaining high prices for rawmaterials) and ensuring the viability of theenterprise (by capitalizing it, making it lessdependent on credit, and so forth).

Integration: The Key toSuccessful ProductDevelopment

Based on a wide range of experience in processand product development for root and tuber crops,we have found that, if any single characteristic isvital for effectiveness, it is thorough integration ofproject activities and participants.

Actors and activities

•

•

•

•

A"dlTlg Value to Root and Thber Crops

It is essential that farmers expandproduction fast enough to keep pace withprojected demand for the processed product.Given that roots and tubers are long-cyclecrops, farmers will respond only if there is asure market and good price for their output.Another issue is the quality of the rawmaterial. Are farmers' traditional varietiesand cultural practices suitable in thisrespect? Plant breeders can contributeimportantly to better productivity andquality, agronomists to improvement incultural practices and cropping systems, andagroecologists to the proper analysis ofresource management issues.

To obtain appropriate equipment, the projectmust seek the services of local designers andmanufacturers, who can develop prototypesfor testing at an early stage in the project.

Project staff must ensure that policy makersare informed about product development andintegrated projects and grasp theirimportance for national development. Thiscould reduce the possibility that governmentofficials will create policies with adverseeffects on project activities.

Consumers should participate in integratedprojects from the very beginning of productdevelopment. Unless this process is closelylinked to consumer requirements andpreferences, the project will fail when itattempts to place production on a commercialfooting.

Any commodity system has three maincomponents: production, processing, andmarketing. Integrating them is the key tosuccessful product development. To accomplishthis the project must build strong ties with a widerange of institutions-both public and privateengaged in research, extension, and socialdevelopment. The exact character of theselinkages will vary according to the stage of theproject in technology generation and transfer.

Below we comment on specific groups withwhich product development must be closelylinked:

Other important acturs are governmentinstitutions involved in n~search anddevelopment, NGOs working in rural areas,producer cooperatives, and small-scaleagrobusinesses. Their participation is especiallycritical during the pilot phase. That is whenresearchers are still testing and adapting thetechnology; extension workers can still modifytheir organizational schemes (in collaborationwith farmer groups) to facilitate technologytransfer; input from the private sector can helpthe project adapt technology generated byresearch institutions to the demands ofcommercial production; and producers can be

Unit 2: An Inugrated Approach to Product Development

connected with consumers of the end product fortheir mutual benefit.

Multidisciplinary cooperation

Th analyze the potential role of root and tubercrops, you need to determine production potential,cost structures, and competitive prices andpotential demand for the crop and its products.You should also identify local farmer groups andother cooperating institutions and examineadditional factors that will affect project activities.

Rarely does a single institution have enoughstaff to supply the wide range of expertise neededto carry out these tasks. For that reason projectparticipants, from the public and private sectors,must resort to creative institutionalarrangements. As illustrated in Box 2, these canbe built around working groups, designed toaccomplish specific tasks by drawing on all thedisciplines available. Th be effective these groupsneed appropriate leadership and plenty ofteamwork, based on a clear understanding of thedistinct perspective of each discipline.

Planning and participation

involved, each with different resources, loyalties,objectives, and strategies.

All participating institutions, includingfarmers groups, should be involved in planningthe program. But the number of participants inplanning meetings should be kept as small aspossible, and decisions should be made by thepeople actually working in the project, not bythose at a higher, more political level. Althoughthe project needs political support, decisions aboutits activities should be taken strictly on atechnical basis.

Genuine integration comes from meaningfulparticipation in project activities. Producers,administrators, and technicians must interact inspecific ways, sharing perspectives, needs, andorganizational capacities. As mentioned above,consumers need to be involved early on, so thatthe project can take into account their concernsfrom the start. You should also determine whatmotivates farmers to participate and how muchlabor and money they are willing to sacrifice inworking toward project objectives.

For farmers and processors, participation is akind of apprenticeship, in which they learn to:

Th integrate multiple actors is easier said thandone. The first step toward this goal is toestablish common objectives and plans of work. Inthe absence of a good general strategy and clearguidelines for project management, the assistanceoffered by others is unlikely to be effective,particularly where various organizations are

•

•

•

Operate and maintain the processing plant.

Discuss issues constructively, speak inpublic, negotiate, and identify and solveproblems.

Understand the technology and the market.

Box 2


When the Visayas State College ofAgriculture MSCA) developed a sweetpotato beverage, it entered into a contractwith a large food and beverage company.The two institutions created amultidisciplinary team with all theexpertise needed for successful productdevelopment. The team consisted of a food

29

product development specialist, a processdevelopment engineer, a marketingspecialist, a fruit processing plant manager,an agricultural extension specialist, and theViSCA researcher/inventor. Leadership wasrotated among team members, dependingon the task at hand.

Addmg Value to Root and Thber Crops

They acquire these skills gradually throughdaily involvement in solving problems, working ingroups, and training new members.

•

•

•

Administer resources and keep accounts.

Make sound investments.

Plan future activities.

which experienced processors train new ones.First used to transfer dried cassava technologyfrom Colombia to Ecuador, this model is now animportant feature of a product developmentproject in the latter country.

Culture and gender

TrainingThe outcomes of product development can beaffected in important ways by these two factors.

A product development project cannot operatewithout qualified human resources. Project staffmust learn how to work on multidisciplinaryteams, communicate with groups of farmers andprocessors, apply project methodologies, monitorand evaluate activities, and so forth. The farmers,many of whom may be functionally illiterate, needtraining to improve both their technical andadministrative skills. In developing an effectivetraining program, the project may benefit fromlinks with institutions that specialize in workingwith groups of small-scale farmers.

Another good approach is the farmer-tofarmer training model (described in Box 3), in

Culture gives people a distinct identity withrespect to communication, dress, food habits, timeconsciousness, values, norms, relationships,reward systems, learning processes, beliefs, andattitudes. These factors, particularly attitudesand habits related to food, can have importantimplications for project design. In some regions ofAfrica, for example, male and female members ofa household may operate totally separateeconomic units. Consequently, to increase theincome of the male may not improve the quality oflife of other family members, who depend on theIncome of the female. At the outset of projectdesign, planners should explore cultural factors inconsultation with farmers, consumers, and others.

Box 3


Promoters must be members of a localAsociaci6n de Productores y Procesadoresde Yuca (APPY) and have ample experiencein cassava production and processing(including construction of processingplants). They also need to be able to read,write, and use a calculator. Theirappointment must be approved by the headof promotion, by the administrator ofUAPPY, and by the APPY to which theybelong. In addition to training, promotersexamine any problems that arise, proposesolutions to the APPY, and communicatethis information to UAPPY. They dedicateup to 3 days a week to their duties aspromoters; their per diem expenses are paidby the UAPPY and the APPY that is beingtrained.

farmers with much experiencei drying traveled to Ecuador'sManabi Province, where they explained anddemonstrated the process to farmers who

st get started. They also helpedct a g plant. The training and

sistance becausethe

counterpartsnicated with them in terms theystand.

1b train ditional groups of farmers,Ecua de Asociaciones dePr y Procesadores de Yuca(U prepared several outstanding localfarmers to promote the new technology.They also developed a special promoters'manual (Romanoff, 1989).


Socially determined attributes of men andwomen define to a large extent what is or is notan "acceptable" role or responsibility for them inagriculture. Failure to take gender issues intoaccount has often resulted in the development ofinappropriate solutions to local problems.Mechanization of processes that havetraditionally been done manually is a classicexample of how product development candisplace female labor by introducing maleoperated equipment.

The last United Nations developmentdecade was dedicated to gender issues and thecurrent one to cultural issues and theirimplications for development (see the list ofpublications on such issues at the end of thisunit). In recent years agricultural researchersand project designers have become moresensitive to those concerns and more adept atincorporating them into integrated projects. Thisnew awareness is partly the result of a growingrecognition that the principal actors and endusers must be full participants in projectactivities and that to achieve this requires alarger role for social scientists on research anddevelopment teams.

A Model for ProductDevelopment

The kind of integration required for successfulproduct development does not normally occurspontaneously. Most often, it is the result ofdeliberate efforts within the framework of a welldesigned project, in which the objectives,activities, and responsibilities of eachparticipating institution are clearly defined.

The whole process of product developmentcan be divided into four distinct phases, as shownin Figure 1 (page 24) and described below. Theseare not theoretical or abstract categories butspecific events, which we've watched unfold inthe various projects with which we've beenassociated.

A project focusing on dry cassava inColombia (see Box 4) provided much of theexperience on which the methodology we presenthere was originally based. Since that project

31

started before the methodology emerged, not allthe activities were carried out in the order wenow recommend.

Phase 1: Identifying product ideas

The approach we outline here is market orconsumer oriented. The first step is to identifyunsatisfied consumer needs (reflected inproblems with existing products) oropportunities for developing new products(evident from gaps in existing markets).

To identify an opportunity for productdevelopment, you must first generate a largenumber of ideas and then select the best optionsfor further investigation. In coming up withideas, you should consider both market andtechnological factors. But the choice of bestproduct alternatives ought to be based largelyon market conditions.

Phase 2: Research

Although research may be conducted in allphases of product development, it is the mainactivity of phase 2. Market studies andconsumer research at this point determine thedemand for and required characteristics of theselected product. Technical research focuses onthe product and the process for manufacturing11. By the end of this phase, you should havedeveloped a prototype product and process andf(lrmed some idea of demand for the product andthe costs of manufacturing it. With thislI1formation, you can prepare a prefeasibility'itudy.

In phase 2 the project must use itslI1stitutional contacts to harness technologicaldevelopments in production and processing aswell as expertise in the social sCiences. Usually,natlOnal agricultural research programs carry(lut production research, while universities andfood and feed research institutes JI1vestigateprocessing and utilization. Universities andspecialized institutes may be engaged in marketand consumer research as well. NGOs can alsocontribute importantly to research. It may evenbe necessary to obtain specialized assistancefrom institutions in other countries.

Adding Value to Root arul Thber Crops

Box 4

The Four Stages ofProduct Development-Dried Cassavain Colombia (Case 1)

coasts most

). Basedstudy,

in Sucretent w'ith small ers who

dy organized and were receivingadequate support from national researchand development institutions.

2. Research (1976-1980): CIAT had alreadydeveloped appropriate cassava chipping anddrying technology, don 'ence inThailand. Center s refin rentaspects of it during the pilot stage.

3. The pilot phase (1981·1983): During thisperiod the project expanded from a singlepilot drying plant to seven. The feed hopperof the chipping machine was redesigned to

Phase 3: The pilot phase

Once the processing technology has beenthoroughly tested and if market and consumerstudies show a demand for the product, it is timeto proceed with the pilot phase. The objective is tointroduce a new product or process on a limitedscale in a specific region where roots and tubersare grown. This step may be associated with theintroduction of improved technology aimed atincreasing crop productivity and reducingproduction costs. If so, an integrated project canprovide the framework in which production,processing, and marketing activities are linked inthe same region.

During the pilot phase, you should evaluatethe technical, economic, and operational

achieve a more even flow of roots; thismodification oubled the machine'scapacity. Th ss was economicallyviable with cassava yields of 8 tIh (theaverage in the project area) and aconversion rate of 2.5.

4. The commercial phase (1984·1989): On thestrength of success in the pilot phase, thenumber of processing plants rose quicklyfrom 7 to 36. Further adjustments weremade in the equipment (e.g., the capacity ofthe motor was inadequate for continualuse). Farmers received intensive training inadministration, so that they couldeventually assume full responsibility. Sincethe end of a formal interinstitutional projectin 1989, cassava drying has expandedsignificantly in the region, particularly asprivate individuals have gotten involved. By1992 over 150 drying plants were operatingin Colombia. This model is also beingadopted in other parts of the country wherethere is potential for producing driedcassava chips.

feasibility of new technology under realisticconditions. Much experience has shown (seeCases 4 and 5 in Part II) that many problemsarise at this stage, which are not apparent in themore controlled environment of a research center.For that reason the pilot plant should beoperated, not by researchers, but by the project'sintended beneficiaries. Any difficulties inmarketing, organization, etc., will becomeevident. Once these problems have been resolved,the project can start to market the productcommercially on a small scale.

By the end of the pilot phase, you ought tohave enough firsthand information to confirm ordeny most of the assumptions made in theprefeasibility study. With solid data derived fromthe pilot experience, you can determine with


confidence whether the project is justified inproceeding to the commercial phase.

The pilot phase involves significant risks.The product and process are still unproven, yetfarmers and others are investing time andresources in the project. It is only fair that theproject cover most of the costs incurred at thispoint, especially that of any untested equipment.Donor agencies recognize the importance of pilotoperations and are generally willing to fundthem, at least judging from the experience ofmany of the projects described in this manual.Assuming that the pilot phase gives positiveresults, further project activities should entailonly minimal risk. The project should then beable to finance additional plants on credit.

Phase 4: The commercial phase

On the foundation of its experience in the pilotstage, the project can replicate or expand the useof new technology and its products. Once youknow what the technology costs commercially,you can calculate the resources needed to meetthe following requirements for widespreadtechnology adoption:

• Credit for crop production, establishment ofa processing capacity, and working capital

It takes time to develop and consolidaterural enterprises and also to legalize them-anecessary step for obtaining commercial credit. Afrequent condition for legal registration is thatall members of the cooperative have titles totheir land, something that few possess. Since themanagers of these enterprises generally havelittle formal education, they are poorly preparedto deal with these matters on their own. Th getaround such obstacles, the project must rely onits links with other institutions involved in ruraldevelopment.

As the project starts to expand, it shouldestablish a system to monitor commercialsuccess and to determine whether the size anddistribution of benefits are as planned. Thissystem can employ mechanisms created duringthe pilot phase to gather information for thefeasibility study.

The outcome of this phase should be a selfsupporting rural agroindustry. As the enterprisegrows stronger, it should gradually take over thefunctions of project institutions. This is a slowand difficult task-one that few projects haveaccomplished fully. Nevertheless, projectpersonnel need to realize that they are not apermanent fixture, that the project framework istemporary.

•

•

Training and technical assistance forfarmers and processors

A plan for product distribution andpromotion

Decision points and checklists

Each of the four phases in product developmenthas a concrete outcome:

• Formation of a second-order organization(Le., a co-op or federation that supports anumber of first-order farmers groups or coops) to coordinate marketing, farmertraining, and provision of credit

•

•

•

Phase 1: one or more products selected forone or more regions

Phase 2: a prefeasibility study

Phase 3: a feasibility study

Primary processing of root and tuber cropsshould be done in rural areas either bycooperatives, associations, or small businesses.Th expand operations from a pilot to acommercial scale, the project will need toestablish several plants rather than one or twolarge ones. Special enterprises located near eachlarge urban market or second-orderorganizations can handle distribution of theproducts or carry out any secondary processing.

33

• Phase 4: a self-supporting ruralagroindustry

The product selected at the end of phase 1determines how the project will develop insubsequent phases. The products of phases 2 and3 will, in the first instance, decide whether theproject should even proceed; if so, they will thenhelp shape the activities of phase 4. The productof this last phase is a measure of the project'ssuccess.

In this manual we provide checklists ofimportant elements you need to consider at eachcritical stage in the four phases of productdevelopment. We don't mean for you to adhere tothese strictly, since not all items are relevant toall situations. We've made the lists as complete aspossible to ensure that nothing is overlooked ortaken for granted. Particularly in the earlier unitsof the manual, we hope the checklists (such as theone for final idea selection in Unit 3) do notrequire that you generate a lot of new primaryinformation. As the project reaches the pilot andcommercial phases, though, your decisions shouldbe based on precise primary information, with aminimum of assumptions.

Final Tips on Project Design

The project's ultimate success hinges oncommercial acceptance of its final product. That iswhy product development must start with theconsumer and a clear definition of productcharacteristics that make consumers want toacquire it. Once the project gets underway,however, other components of the wholeendeavor-production, utilization, andcommercialization-will each become temporarilypreeminent, as the project advances from onephase to another in product development.

Since project design is complex, werecommend that you start even before the projectbegins. It is helpful to prepare a documentoutlining the project's aims and plans,particularly if you intend to seek externalfunding. If not, the document still helps defineproject activities and ensure that they are relatedto its objectives. (See the list of publications onproject design.)

Given that product development can bedivided into four distinct phases, each endingwith a concrete product, the whole undertakingcould be similarly organized into four projects.The first would identify the product and targetregion, the second would conduct research, and soforth. Unfortunately, though, few donorsrecognize the first phase as a concrete activityrequiring their support. So, often, projectplanners will have to identify opportunities beforesubmitting a research proposal.

Add,,,/? Value to Root and Thber Crops

One purpose of this manual is to highlightthe importance of this first phase and the need toassign it sufficient time and resources. Unlessphase 1 is taken seriously, the project may wastevaluable resources on products that have nocommercial prospects and should never have beenselected.

References

Romanoff, S. 1989. Manual de referencia para lapromoci6n de asociaciones de productores yprocesadores de yuca. Fundaci6n Ecuatoriana deInvestigaciones Agropecuarias (FUNDAGRO) andCIAT, Quito, Ecuador.

Further Reading

Gender

Feldstein, H.S. and Poats, S.Y" 1990. Working together:Gender analysis in agriculture. 2 vols. KumarianPress, West Hartford, CT, USA.

Poats, S.V 1991. The role of gender in agriculturaldevelopment. Issues in Agriculture no. 3.Consultative Group on International Agriculture,Washington, DC, USA.

Culture

Bammate, N. 1977. Cultural aspects of development.In' Nossin, J.J. (ed.). Surveys for development.Elsevier Scientific Publishing, Amsterdam,Netherlands.

Brislm, RW. 1981. Cross-cultural encounters: Face-toface interaction. Pergamon Press, New York,USA

Hall, E.T. 1979. Beyond culture. Anchor Press/Doubleday, Garden City, NY, USA.

Harris, PR and Moran, RT i979. Managing culturaldifferences. Gulf Publishing, Houston, TX, USA.

Hofstede, G. 1980. Culture's consequence:International differencei' in work-related values.Sage, Beverly Hills, CA. USA.

Moran, RT. and Harris, PR 1981. Managing culturalsynergy. Gulf Publishing, Houston, TX, USA.

Project design

Austin, V. 1984. Rural project management: Ahandbook for students and practitioners. BatsfordAcademic and Educational Ltd., London, UK.

Unit 2: An Integrated Approach tv Product Development

Bamberger, M. and Cheema, S. 1990. Case studies of

project sustainability. World Bank, Washington,

DC,USA.

Brekelbaum, T. and Rizo, N. (comp.). Proyectos

agricolas (lista de referencias bibliograficas).

CIAT, Cali, Colombia. (Mimeo.)

CIAT (Centro Internacional de Agricultura Tropical).

1990. Manual para disenar proyectos integrales

de yuca (version preliminar). Cali, Colombia.

Conrad, D.L. 1980. The quick proposal workbook

(includes computer program and handbook).

Public Management Institute, San Francisco, CA,

USA.

Delp, E; Thesen, A.; Motivalla, J.; and Seshadri, N.

1977. Systems tools for project planning. Program

Advanced Studies in Institution Building and

Technical Assistance Methodology, International

Development Institute, Indiana State University,

Bloomington, IN, USA.

Gittinger, J.E 1982. Economic analysis of agricultural

projects (2nd ed.). Johns Hopkins University

Press for Economic Development Institute and

the World Bank, Baltimore, MD, USA.

International Women's Tribune Center. 1989. Proposal

writing and financial/technical assistance. New

York, NY, USA.

Mayo-Smith, I. 1980. Planning a performance

improvement project: A practical guide.

Kumarian Press, West Hartford, CT, USA.

Motta, F. 1990. Marco logico para el diseno de un

proyecto. CIAT, Cali, Colombia. (Mimeo.)

Schubert, B; Nagel, v'J.; Denning, G.L.; and Pingali,

PL. 1991. A logical framework for planning an

agricultural research program. International Rice

Research Institute (IRRl), Los Banos, Philippines.

Vanderschmidt, L. and Lent, M. 1981. Systematic

project design: A handbook for volunteers. Center

for Educational Development. Boston University,

Boston, MA, USA.

Zivetz, L. 1990. Project identification, design and

appraisal: A manual for NGOs. Australian

Council for Overseas Aid (ACFOA).

Unit 3

Identifying Product Ideas

This unit explains how to generate and selectideas for products based on root and tuber crops.As shown in Figure 2, the outcome of this phase isselected product ideas. Project staff then studythese in the research phase, as described in Unit 4.

In that and all subsequent units, we assumeresearch is being carried out under a project withspecific objectives and outputs. Even where this isnot the case, you should organize the research asif it were a project to increase efficiency and thechances of success.

At the start of the project, only its objective isdefined. As indicated in Table 11, you may havedecided on related matters as well, including thebeneficiaries, the region where the project will belocated, or others, such as the raw material to beused (Le., which root crop) and the type ofenterprise that will produce and market theproduct. At the end of this and subsequent units,we present revised versions of Table 11, notingprogress made in resolving major project issues.

How to Generate Product Ideas

Ideas may come from a variety of sources,including:

• Consumers

• Traders

• Market researchers

• Technical researchers

• Managers

• Policy makers

• Development agencies

• Extension workers

You can also generate ideas through varioustechniques, such as:

• Brainstorming sessions withmultidisciplinary groups

• Interaction of researchers with a smallnumber of consumers, traders, and others

37

selected according to certain demographic orsocial characteristics

• Visits to markets and stores

• Analysis of secondary information (e.g., onfood consumption, purchasing habits,agricultural trends, and prices)

To identify a successful product, you need totake into account the objectives, constraints,assumptions, and beneficiaries of the project. Youshould also make sure that there is a market forthe product, that the crop provides suitable rawmaterial for it, and that this material can beprocessed cost-effectively.

Finding a market niche

Root and tuber products are sold in a wide rangeof markets (see Checklist 1), each with distinctcharacteristics and requirements. In foodmarkets, low- and high-income consumers oftendiffer in their consumption habits andexpectations of food products. High-incomeconsumers tend to value convenience, quality, andlong product shelf life, whereas low-incomeconsumers are likely to be more concerned withprice. Roots and tubers have enormous potentialas low-cost, locally available raw materials for awide range of processed products, provided thesecrops can compete with cereals in price, quality,and availability.

In developing countries consumption ofanimal products (meat, milk, and eggs) and theirderivatives is increasing rapidly. So is the use ofroot and tuber crops as low-cost carbohydratesources for livestock production.

In the many industries that use starches andflours, there is potential for replacing importedwheat and other cereals with locally producedroot and tuber products. Over the last twodecades, government policy in many countries hasfavored imported cereals at the expense of thosecrops. The withdrawal of subsidies in manycountries should enable roots and tubers to

rGovernment

policy

Addfll;: Value to Root and 1\lber Crops

Raw material i

characteristics I

iiL .....-i~_R.~~~~J

~I

" iInformation on market, l .environment, i ~socioeconomic conditions, Iand crop production !t--_·· .

No

~.~~..

( Product 1"-------------

... _.1 ..

Generate product ideas

List of products

,Initial idea screening

Final idea screening

OK

Product 2

Define projectobjectives,constraints,assumptions,and beneficiaries

..

Product 3

Figure 2. Stages in generating and selecting Ideas for products based on root and tuber crops (phase 1

Unit 3: Identifying Product Ideas

Thble 11. Status of a product development project at theoutset ofidea identification (phase 1).

products-from traditional foods to sophisticatedfood and industrial products.

compete against imported cereals on a moreequitable basis.

Why (objective)?Where (region)?What (product)?How (process)?How much (market)?By whom (type of enterprise)?For whom (beneficiaries)?

a. Possibly implied by project objective.

Defined Many industries that could use root andtuber crops maintain rigid standards for specificraw material qualities (such as purity andhygiene, physicochemical composition, andfunctional properties). Some options for productdevelopment will simply not be feasible unlessfarmers and processors can meet those standards.In general, improving quality of the raw materialor end product increases its market potential.

Options for processing

A cornucopia ofroot and tuberproducts

Because root and tuber crops are so versatile,they offer many opportunities for productdevelopment (see Checklist 2). Although some ofthese crops are relatively rich in protein orcertain vitamins, they are mainly used as acarbohydrate source in a wide range of

Products of root and tuber crops aremanufactured through a wide range of processes,as indicated in Checklist 3. Some products, suchas beverages made from sweet potatoes, mayacquire additional processing.

For each process there are generally variousoptions that differ in technological sophistication.Starch, for example, can be extracted either

Checklist 1

Markets for Root and Thber Crops

Food

• Export

• Urban consumers (upper, middle, andlow income)

• Rural consumers (many of whomproduce these crops and consume themon-farm)

• Food industries, small (e.g., localbakers) or large scale (e.g., producers ofstarch for packet soups)

Animal feed

• Export or domestic use

• Source of carbohydrate (roots andtubers) or protein (leaves and vines)

• Functions related to specific properties(e.g., use of cassava as an agglutinantfor shrimp feed)

• Animal feed companies or livestockproducers

• Integrated livestock systems based onroots and tubers

Industrial

• Functions related to properties ofstarch, product purity, particle size, andprice

• Production of starch, flour, and modifiedstarches used as raw materials

• Major industries (e.g., paper, textile,glue, plywood, cardboard, oil, andpharmaceutical)

Adding Value to Root and Thber Crops

Checklist 2


Primary Secondary

• Fresh roots and tubers: • Starch and flour:

At harvest Bakery goods, soup and sauce bases,

Cleaned and graded processed meats, pastas, noodles,

With improved shelf life beverage bases, and snack foods

Prepared Textiles, paper, glue, plywood, oil,and pharmaceuticals

Noncommercial quality (for animalAnimal feed rationsfeed)

Boiled (tubers) to remove Alcohol

antinutritional factors Glucose, mannitol, sorbitol, etc.

Ensiled (roots) for storage as Dextrin

animal feed Monosodium glutamate

• Fresh leaves and vines • Fresh roots and tubers:

• Sun-dried pieces, chips, and slices of Frozen, canned, and vacuum-packedroots and tubers

Flakes, granules, and cubesProtein enriched via fermentation•

• FlourUsed in beverage, jam, and sauceproduction

• Native starch Snack foods (chips, crisps, etc.)• Fermented starch

• Modified starches • By-products:

• Leaf meal Animal feed

• Fresh leaves and vines (ensiled) Products of industrial processing

manually with rustic equipment or by means ofhigh technology in a fully automated factory. Thetype and complexity of the technology you usedepends on several factors:

• Scale of operation• Capital investment• Operators' level of education

• Value of product• Conversion or extraction rate

••••

•

Product quality or purityRaw material characteristicsRaw material costAvailability of services (water, fuel, andelectricity)

Cost of labor

41

This is not the appropriate place todiscuss the details of specific types ofequipment. To select product ideas, youneed only a general idea of the scale ofyour operation and the complexity of thetechnology. (See the list of publications onproducts, processes, and markets at the end ofthis unit.)

Raw material characteristics

In identifying opportunities for productdevelopment, keep in mind the characteristics ofroots and tubers as raw materials, includingtheir chemical composition, functional propertiesin specific products (especially those containing


Checklist 3


• Selection/grading and cleaning mesh or centrifugal action, or by

Peeling-manual, mechanical, chemicalsedimentation in tanks or settling

• channels(lye), or steam

• Fermentation, usually in a solid state• Sulphating to prevent enzymic with natural inoculum

browning• Milling (in hammer, pin, or roller mills)

• Reduction of size by slicing, chipping, or and grading by mesh size according tograting/rasping the end use

• Blanching to prevent enzyme action and • Boiling and cooking of fresh roots forpartially cook or sterilize prepared foods, purees, or drying

• Drying-solar or artificial, using batch • Frying for fresh roots, snack foods, etc.or continuous processes

• Extrusion of starch or flour for snack• Starch extraction by separation of and other foods

starch in solution from other rootcomponents and sedimentation of • Baking for flours and starchessoluble starch, by separation using· fine

starch), and some of the storage and other traitsof fresh roots. Some of these features are listed inTable 2 in Unit 1; Table 12 gives others. (For moreinformation on this subject, see the list ofpublications at the end of this unit.)

The chemical composition of root crops cansuggest potential products. In the Philippines, forexample, scientists at ViSCA decided to develop afruit-juice-like beverage from sweet potatobecause of its high content of sugar and vitamin Cand because some varieties are yellow or orange.Likewise, the high starch content of cassavamakes it a good candidate for starch-basedproducts.

Raw material characteristics also affect youroptions in processing. The high dry mattercontent of cassava, for example, allows it to besun-dried easily. For potato, in contrast, solarassisted or artificial drying is more suitable(except where ambient humidity is very low andsolar radiation is high, as on an altiplano),because a greater quantity of water must beremoved from this crop to obtain a stable product

41

Some raw material characteristics of rootand tuber crops vary considerably and may alsobe affected by environmental conditions. The datain Table 12 are averages. If possible, obtain morespecific information about the chemicalcomposition, etc., of locally available varieties inthe project area.

Selecting the Project Region

At this early stage in product development, manyproject activities can go forward even before youhave decided where the final product will bemarketed. Even so, it is helpful to have a roughIdea of the target region.

This is defined automatically where productdevelopment is associated with a ruraldevelopment project in a specific region. But if itIS part of a national or international project, youwi II need to select one or a few specific targetn~b'1ons. This will help you identify potentialpfllducts by narrowing the geographical focus ofyour study of existing products and their raw


Table 12. Raw material characteristics of root and tuber crops that are relevant to opportunity identification.

Cassava Sweet·potato

Dry matter (%) 30-40 19-35

Starch (% FW) 27-36 18-28

'Ibtal sugars (% FW) 0.5-2.5 1.5-5.0

Protein (% FW) 0.5-2.0 1.0-2.5

Fiber(%FW) 1.0 1.0

Lipids (% FW) 0.5 0.5-6.5

VitaminA(~g/100 g FW) 17 900

Vitamin C (mg/100 g FW) 50 35

Ash (%FW) 0.5-1.5 1.0

Energy (kJ/I00 g) 607 490

Antinutritional factors Cyanogens Trypsininhibitors

Starch extraction rate (%) 22-25 10-15

Starch grain size (micron) 5-50 2-42

Amylose(%) 15-29 8-32

Max. viscosity (BU) 700-1,100 n.a.

Gelatinization temp. (OC) 49-73 58-85------.-_..~-


materials. It will also help screen product ideas,as discussed below. Here are some criteria forselecting target regions:

•

Potato Yam Aroids

20 28 22-27

13-16 18-25 19-21

0-2.0 0.5-1.0 2.0

2.0 2.5 1.5-3.0

0.5 0.6 0.5-3.0

0.1 0.2 0-1.5

Trace 117 0-42

31 24 9

1.0-1.5 0.5-1.0 0.5-1.5

318 439 390

Solanine Alkaloids, Oxalatestannins

8-12 n.a. n.a.

15-100 1-70 1-12

22-25 10-30 3-45

n.a. 100-200 n.a.

63-66 69-88 68-75

It is advantageous if the rural population hasexperience in root crop production andprocessing.

•

•

There is enough root and tuber cropproduction to establish an agroindustry.

Fresh root prices are low enough to permitprocessing if the project aims to manufacturea low-value product.

•

•

It is helpful if institutions in the region havesuccessfully supported the formation ofcooperatives or small enterprises in the past.

Credit or other sources of finance areavailable for replicating project activities.

•

•

•

•

•

Harvest is spread over several or manymonths.

One or more institutions in the region isengaged in crop production research andextension.

For food products, there is a major urbanmarket within or close to the region selected.

The distances over which both raw materialand final or intermediate products wouldhave to be transported are acceptable.

For products that require other rawmaterials or inputs (e.g., for packaging orfue!), these are available at reasonableprices.

4 ')L.

• Policy makers in the region clearly supportthe improvement of rural welfare throughproduct development for root and tubercrops.

How to Screen Product Ideas

Having developed product ideas, you should nowscreen them to pinpoint those with the bestchance of success. This is a two-stage process,consisting of an initial and final screening.

Initial screening

First, screen your product ideas for theircompatibility with project objectives, assumptions,and constraints (as described in Box 5). Suppose,

Unit 3: Ickntifying Product Ideas

Box 5

InitiaISereenhtg of Cassava Product Ideas in Colombia (Case 1)

I

l'~-

In the country's Atlantic coast region. freshcassava is a major staple for both rural andurban populations. although consumptionin urban areas is declining. Cassava isgrown exclusively by small-scale farmers inthe region. which provides 40% ofColombia's total supply of the crop. Becauseofsharp fl uations in price, cassavaprodu me risky in the 19708.Output declined. and farmers had littleincentive to adopt new productiontechnology. There were few alternativecrops for this seasonally dry area withinfertile soils.

In the 1980s opportunities wereidentified for developing alternative uses ofcassava (starch extraction and drying foranimal feed) and for improving storage offresh cassava in traditional markets. Theseoptions were screened in light of theproject's:

ObJectives

• Improve cassava markets and linksmall-scale farmers to them.

• Increase the incomes of small-scalefarmers and landless laborers.

COnBtrainl.

• The Programa de Desarrollo RuralIntegrado (DIU) was relying exclusivelyon fanners groups and cooperatives forsmall-enterprise development.

• Little capital was available forinvestment.

• The options were limited to small-scaleprocessing. using locally manufacturedequipment.

A.BumptionB

• External fmancing was available forbuilding a pilot plant.

• Local fmancing was available forinstitutional support.

Sun-drying of cassava chips was foundto be an economically viable option. andseveral other processes and productsshowed potential.

for example, that the project aims to improve thewelfare of small-scale farmers by adding value toroot and tuber crops through rural processing.You would eliminate processes such as freezing,canning, and vacuum packing, since these involverelatively complex technology and are capitalintensive, making it difficult for small-scale ruralindustries to adopt them.

Similarly, if the project's objective is toimprove the nutritional status of low-incomeurban consumers, you would reject this samegroup of technologies, because their high cost willresult in a product that only higher incomeconsumers can afford. Based on the project's socialobjectives and financial limitations, you can thusreduce the product options to a manageablenumber for more detailed study.

43

The remaining options will normally beproducts that require relatively simple, low-costprocesses and have a large potential market.These will tend to be fresh, prepared, or storedproducts as well as flour, starch, and productsderived from them. Since flour includes itemssuitable for animal feed, human food, andindustrial use, you can subdivide this category ofproduct. For each potential product, indicate adistinct quality or specification. This in turndetermines the process required and the cost ofthe final product.

Final screening

You can screen the remaining products or marketcategories (using Checklist 4), on the basis ofpotential demand, raw material supply, and

AddIng Value to Root and Thber Crops

Checklist 4


Potential demand

• What is the target market (city, region,etc.)?

• How large is the market (as apercentage of the total number ofconsumers)?

• How much product will each clientconsume per year?

• Is the potential market expanding?

• Are consumption or food purchasinghabits changing?

• Does the product fit these changes?

• If the product is novel, will consumersaccept it?

• If the product is competing againstothers, will its price and quality bebetter?

• For industrial markets, what are thevolumes and prices of competing rawmaterials?

Raw material supply

• What is the volume of production in thetarget region?

• How much demand is there for thisproduction in other markets?

• Is production and demand in othermarkets seasonal?

• Are there price fluctuations or cycles?

• What are the main characteristics ofavailable varieties (growth cycles,quality, and yield)?

• Are other (improved) varietiesavailable?

• What is the potential for increasingproduction (by increasing area or yield)?

• What are the constraints to increasedproduction (diseases, erosion, drought,etc.)?

• What are the main handling andstorage factors?

• If the crop is to replace another sourceof raw material, what will be the effectsof reduced demand for this other crop?

• What environmental effects could resultfrom expanding or intensifying cropproduction?

Physical factors

(These determine harvest times and thefeasibility of natural drying.)

• What is the rainfall pattern (dry andwet seasons)?

• Does the dry season coincide withharvest time?

• What is the temperature pattern?

• What is the pattern of relativehumidity?

Utilities

• Does the target region have access to:

Electricity, gas, coal, etc?

Clean water (especially for starchextraction)?

Sewerage or other waste disposalsystems?

• Are there adequate roads and access totransport?

Organizational aspects

• Are farmers linked to markets?

• Are farmers willing to experiment?

• Does the target region have a history ofpositive experience with co-ops?

• Is institutional support available for co-ops or small businesses?

• Is credit available?

• Are capital requirements manageable?

• Is a separate distribution entity needed,and are technical and financial supportavailable for this?

• Will many institutions be involved? Isinterinstitutional coordinationsatisfactory?

(Continued)

Existence ofsimilar activities •• Would processing by small-scale farmers

be new or merely a continuation ofcurrent practices? •

• What are the strengths and weaknessesof existing operations? •

• What is the scale of existing operations?

• How can weaknesses be corrected? •• Could you use existing facilities to

reduce capital outlays? •


Checklist 4. Continued.

Capital requirements

• How much money would you have toborrow and under what conditions(interest rate, grace period, etc.)?

• What are the collateral requirements(landholdings, etc.)?

• Would you have to obtain the loanunder standard banking arrangements,or is there a special loan scheme forsmall businesses or co-ops?

Labor

• Costs• Availability (seasonality)

• Educational level (including literacy)

• Gender issues

Technology

• Has it already been developed, or isfurther research required? If so, howmuch, and what are the chances ofsuccess?

various other factors. This brief, but systematic,analysis is preferable to a long-term, costlyresearch project. How long it takes to workthrough Checklist 4 will depend on the amount ofdata available. You should be able to fill any gapsin the information by consulting secondary datasources and by making a brief visit to the projectregion. There is no need to conduct an elaborateformal survey.

45

Is the technology imported or local? Ifimported, are spare parts and so forthavailable?

Can farmer groups manage thetechnology?

Is training required (process operation,business, marketing, etc.)? If so, whocan provide and finance the training?

Can local labor and materials be used inconstruction?

How will the process and its wastesaffect the environment?

Consumer acceptance

• Is the crop (fresh or processed) alreadypart of the local diet?

• Is its image good or not?

• Is the product already being made? Ifnot, why not?

• If the product is already available, howcan it be improved?

• If the product is novel, is the crop usedin other foods or with other ingredients?

Benefits

• How well will the product meet projectobjectives?

• Who will benefit and how much?

• What is the risk of failure? Who willsuffer if the project fails?

• How great will the benefits be (largeenough to make the project economicallyfeasible)?

Be sure to review all your assumptions withcare, since common knowledge is frequentlymistaken. For example, in areas originallyconsidered to have excessive rainfall, cassavadrying is now commercially successful.

The next step is to decide whIch of severalinitial product ideas to study in more detail, basedon your responses to the items in Checklist 4. Be

flexible in using the checklist. For some products,a single negative response to a crucial questionmay be enough to eliminate them. For others,negative responses to several questions may notmatter, if they relate to conditions that can bechanged, given sufficient funds, training, or time.

One advantage of using the checklist is thatit can help you generate additional product ideas.As you consider each item in the list, suggestionsor observations may occur to you for modifyingthe original product idea (e.g., selling a powderlike potato product to industrial clients ratherthan a dried wafer or chip directly to consumers)


or for replacing it altogether (e.g., with aproduct for animal rather than humanconsumption).

Tables 13 and 14 give the results of usingChecklist 4 in particular cases. A project innorthern India screened four ideas for potatoproducts, as shown in Table 13. The mostfavorable option was to store potatoes for sale inthe off-season, when prices are higher. But sincethe project's objective was to maximize ruralemployment, not farmer incomes, it opted todevelop several processed products for humanconsumption as a means of creating jobs.

Table 13. Final screening of ideas for potato processing in northern India (Case 6),f--------------- ..---------..

Checklist

Fresh storage Dry processed

Product ideas

Chips(animal feed)

Starch

Demand

Raw material supply

Physical factors

Organizationalaspects

Existence ofsimilaractivities

Consumer acceptance

Capital requirements

Labor

Technology

Good; fills gap inmarket (out of season);traditional product,good acceptance acrossincomes

Excellent; buy at lowprice and sell whenhigh

Need to store duringhot season, technologydevelopment needed?

Family level or crop

Storage of other cropscarried out

Good if quality ok

Moderate

Little used

Needs work but simple

Novel product for'high-income consumers(10% of population )

Limited to June-March,at main harvest, whenprice is low

DI'y season comcideswith harvest; good flJrnatural drying

Small business

Processing well knownand accepted

Good at upper mcomelevels

Moderate

Much needed; someskilled

Needs work; rathercomplex

Depends on price,unlikely to becompetitive

Family level or crop

Industrial market,depends on price

Low

Unskilled only

8nnple

Depends on price ofraw material andstarch yield; uncertain

Water availability poor

Small business

Not in this area

Moderate to high

Much needed; someskilled

Known; relativelysimple

Benefits:FarmersProcessorsConsumersEconomy

+++ + ++None ++t ++

++ None+++ ++ +

++++

++

Note: Size ofbencfits: +++ =many, ++ =SlIIIlC, and + = few.

46


Table 14. Final screening of ideas for cassava product development in the Atlantic coast region of Colombia.

Dried cassava Cassava flour Fresh Starch(animal feed) (human consump.) storage

Demand +++ +++ ++ ?

Raw material supply +++ +++ +++ +++

Climate, water, electricity +++ ++ ++ No

Farmer organization Some Some Some Some

Existing activity No No No No

Consumer acceptance +++? ++? ? ++

Capital needs Low Medium Low Medium

Labor availability +++ +++ +++ +++

Technology existing Yes No No Yes?

Benefits farmers +++ +++ +++ ?

Benefits landless labor +++ ++ +

Benefits urbandistributors No + ++ No

Benefits consumers No No +++ No

Conclusion Proceed to Research processes Research processes Developpilot stage and markets and markets elsewhere

Note: +++ =excellent potential, ++ =gaud potential, and + = some potential.No =constraint identified.? = uncertain, more research required.

In the Atlantic coast region of Colombia,where there is a sizable market for fresh cassava,project planners perceived a need for new marketoptions (Table 14). Because of drastic fluctuationsin the price of this commodity, its production hadbecome risky for small-scale farmers. The projectelected to produce dried cassava for animal feed,because it requires only simple sun-dryingtechnology and offers promise as a substitute forimported feedstuffs. Another option-productionof flour for human consumption-required moreapplied research. So did the development of amore storable fresh root to supply urban markets.Project planners ruled out starch extraction,because it would have been difficult to obtain areliable supply of good water in this seasonallydry environment.

Ideas Define ActionWe cannot overstate the importance of correctlyidentifying product opportunities at the outset. Alltoo often, projects are started on the basis of acursory evaluation of the processing technologiesavailable. The danger in this approach is thatconsiderable resources may be committed todeveloping a product with only a minimal chanceof success.

47

By taking time to generate a large number ofproduct ideas and then evaluating them, you canidentify quite efficiently the one or more productsthat are most likely to succeed. The screeningprocedure need not be time-consuming orexpensive. In most cases you can obtain all thenecessary information from secondary sourcesrather than generate a lot of new informationthrough research surveys and other means.

In addition to generating one or a fewproduct ideas, this stage in product developmentleads to the identification of a target region,where the product will be tested and eventuallycommercialized (Table 15). As the project movesthrough successive phases, more of its activitieswill take place in the target region rather than aresearch center.

In selecting the product idea, you will alsodefine the general processes to be used in itsmanufacture (issues such as equipment needs andscale of the operation are dealt with in theresearch phase). Even though you may not havequantified demand for the product, you will atleast have determined that it exists. You shouldalso have a clear idea of the project's potentialbeneficiaries.

Adding Value to Root arul Thber Crops

Why (objective)?

Where (region)?

What (product)?

How (process)?

How much (market)?

By whom (type of enterprise)?

For whom (beneficiaries)?

a. Possibly implied by project objective.

Phase I-Start

Defined

-"

Phase I-Finish

Better defined as project progresses

Identified (general area)

Idea selected

General process determined by type of product

Not quantified but good potential seen

Defined

Further Reading

Root and tuber markets, processes, andproducts

Bacigalupo, A. (ed.). 1985. Technical manual on basicfood processing. Food and AgricultureOrganization (FAO), Regional Office for LatinAmerica and the Caribbean, Santiago, Chile.

Balagopalan, C.; Padmaja, G.; Nanda, S.K.; andMoorthy, S.N. 1988. Cassava in food, feed andindustry. CRC Press, Boca Raton, FL, USA

Buitrago A., J.A. 1990. La yuca en la alimentacionanimal. CIAT, Cali, Colombia.

Chan, H.T. and Chan, H.T., Jr. 1983. Handbook oftropical foods. Marcel Dekker Inc., New York, NY,USA.

CIAT Cassava Program. Study guides and constructionmanuals for cassava storage, drying and starchextraction technologies. CIAT, Cali, Colombia.

FAO (Food and Agriculture Organization). 1988. Rootand tuber crops, plantains and bananas indeveloping countries: Challenges andopportunities. FAO Plant Protection andProduction Paper no. 87. Rome, Italy.

Fellows, P.J. 1988. Food processing technology:Principles and practice. Ellis Horwood, New York,NY,USA.

Grace, M.R. 1977. Cassava processing. FAO PlantProduction and Protection Series no. 3. FAG,Rome, Italy.

Hollyer, J.R. and Sato, D.M. (eds.). 1990. Proceedings,Taking Taro into the 1990's: A Taro Conference.Research Extension Series no. 114. HawaiiInstitute of Tropical Agriculture and HumanResources, Hilo, HI, USA

I1TA (International Institute of Tropical Agriculture).1990. Cassava in tropical Africa: A referencemanual. Ibadan, Nigeria.

[]TA Post-harvest Technology Unit. Manuals and plansfor gari and other cassava process equipment.lITA, Ibadan, Nigeria.

Kawabata, A; Garcia, VV; and Rosario, R.R. 1981.Processing and utilization of root crops in thetropics. In: Uritani, 1. and Reyes, E.D. (eds.).Rioresources investigation on production, storage,processing and vegetation of root crops in thetropics. 1981 interim report, Nagoya UniversityCooperation. Nagoya, ,Japan.

Lancaster, PA and Coursey, D.G. 1984. Traditionalpost harvest technology of perishable tropicalstaples. FAO Agricultural Services Bulletinno. 59. FAO, Rome, Italy.

Muchnik, J. and Vinck, D. 1984. La transformation dumanioc: Technologies autochtones. Agence deCooperation Culturelle et Technique ConseilInternational de la Langue Fran~aise. PressesUniversitaires de France

Pacey, A and Payne, C. 1985. Agricultural developmentand nutrition. FAOfUnited Nations InternationalChildren's Fund (UNICEF)/Hutchinson, London,UK

Plucknett, D.L. (ed.). 1979. Small scale processing andstorage of tropical root crops. Westview Press,Boulder, CO, USA

Unit 3: Identifying ProdUA:t Ideas

Scott, G. 1987. Potato marketing and demand indeveloping countries. In: Report of the ThirdSocial Science Planning Conference. CIP, Lima,Peru.

Scott, G.; Herrera, J.E.; Espinola, N.; Daza, M.;Fonseca, C.; Fano, H.; and Benavides, M. 1992.Desarrollo de produetos de raices y tuberculos.vol. II - America Latina. Memorias de un taller.CIP, Lima, Peru.

Scott, G.; Wiersema, S.; and Ferguson, P.I. 1992.Product development for root and tuber crops,vol. I - Asia. Proceedings of a workshop. CIP,Lima, Peru.

Weber, E.J.; Cock, J.H.; and Chouinard, A (eds.J. 1978.Cassava harvesting and processing: Proceedingsof a workshop held at CIAT. InternationalDevelopment Research Centre (lDRCJ, Ottawa,Canada.

Wolfe, J. 1987. The potato in the human diet.Cambridge University Press, Cambridge, UK.

___. 1992. Sweet potato: An untapped foodresource. Cambridge University Press,Cambridge, UK.

49

Composition and characteristics of roots andtubers

Bradbury, J.H. and Holloway, W.D. 1988. Chemistry oftropical root crops: Significance for nutrition andagriculture in the Pacific. Australian Centre forInternational Agricultural Research (ACIAR)Monograph no. 6. Canberra.

Burton, w.G. 1989. The potato. Longman Group,Essex, UK.

Cooke, R.D.; Rickard, J.E.; and Thompson, AK. 1988.The storage of tropical root and tuber crops:Cassava, yam and edible aroids. ExperimentalAgriculture 24:457-470.

Kay, D.E. 1987. Crop and product digest no. 2: Rootcrops. 2nd ed., rev. Tropical DevelopmentResearch Institute, London, UK

Wheatley, C.C. and Chuzel, G. 1993. Cassava: Thenature of the tuberous root. In: Encyclopedia offood science, food technology and nutrition.Academic Press, Orlando, FL, USA

Unit 4

Research for Product and Process Development

By now you should have formed a general ideafor a product and decided in which region it willbe developed. The next steps are to define thecharacteristics of the product as well as theprocess for manufacturing it and to quantify moreprecisely its market demand and potentialprofitability. Th complete these tasks requiresmore information, which you can generatethrough two types of research:

• Market, consumer, and farm-orientedresearch to characterize existing patterns inproduction, marketing, and consumption ofthe selected commodity and of products thatmay compete with your own. Informationgathered through these studies will highlightsocioeconomic, cultural, and technicalconstraints that must be overcome to makethe product a success.

• Technical research to develop a productthat satisfies consumer demands andpreferences and to design appropriateprocessing technology.

In this unit we first discuss the notion of anideal system for producing and commercializing aroot and tuber product (see Figure 3). Then, wedescribe how to examine the system's variouscomponents and links to determine whether anyof these are missing. The missing elementsprovide the basis for an agenda of market,consumer, and farm-oriented research.

The next step is to develop a product brief,which gives the characteristics of the product.The technical research needed to develop theproduct and process consists of two stages: Thefirst takes place in the laboratory, while thesecond focuses on prototypes of new equipment(assuming that existing equipment is unsuitable).With the information generated through research,you can prepare a prefeasibility study, whichindicates whether the project is justified inproceeding to the pilot phase.

f)l

Designing an Ideal System

The first step in research planning is to design anideal system for converting the raw material intothe desired product. Using Checklist 5, you canbreak the system down into its component parts,which must be linked if the project is to succeedcommercially.

It is important at this stage to analyzegovernment policies that may affect the project.The existence of state-run marketing boards andmonopolies, for example, may determine themarketing channel you use.

Some components of the ideal system mayalready exist. For example, farmers in the targetregion may already be producing the targetcrop. If the project's objective is simply to improvetraditional products, all the components and linksmay be in place. Th diversify the market for theseproducts by improving their quality, you will needto establish new links to different markets andconsumers. If the purpose is to develop a novelproduct, you may have to establish all thecomponents and links of the system-fromprocessing to consumption.

Having designed an ideal system, you needto Investigate various socioeconomic, marketing,and technical issues to determine whether thesystem is viable and to identify specificconstraints.

Take the case of a project in Colombia, whoseobjective was to link small-scale cassava farmersto the urban market for bakery products throughrural processing of cassava into flour. Th facilitateproduction of a composite wheat-cassava flour, itwas proposed that wheat mills do the mixing forsubsequent sale to bakeries through the wheatmills' distribution channels. To determine thefeasibility of this system, the project researched awide range of issues (as indicated m Figure 4),usmg a holistic approach.

Adding \Ulue to Root and Tuber Crops

Product 1

t

Create idealized system

,Characterize system components and links

L- . _. ~ _

Research: market, consumer, on-farm, and raw material quality'------,--~------

.----.1--

Concept testing'-------,-------

._____----- ____t __C Product brief-----------,.--.--_._-------~-

Identify missingcomponents and/or links

Research agenda

Product development-laboratory scale

Product development-prototype production and evaluation

Prefeasibility study

OK?

Yes

Pilot phase

Figure 3. Stages in research for product development (phase 2)

52

NoI

-.j New ideas or Ii change pro~~:.~bri':~_l

Unit 4: Research for Product and Process Development

Checklist 5

Components of an Ideal System for Producing andCommercializing a Root and 'fuber Product

• Production to supply the raw material.

• Processing to transform the rawmaterial into an end product. Often,processing has more than onecomponent. For example, the rawmaterial may first be transformed into aprimary product, such as flour or starch,from which the final product ismanufactured through furtherprocessing.

Market, Consumer, and FarmOriented Research

At this stage the technical problems associatedwith developing a new product and process mayseem predominant. Yet, the commercial success ofsimple processing depends, not just on itstechnical feasibility, but on its attractiveness andutility for consumers. It is therefore essential thatresearch on technical questions be balanced by athorough investigation of socioeconomic issues.

What's in the market?

The aim of this research is to supply informationon the components and links that are missingfrom your ideal system. It does so by defining theexisting market structure, identifying productsalready on the market, and determining theircosts. Checklist 6 can help guide your search forthis information.

An easy way to prepare a list of processedproducts of a particular type or from a given foodcrop is to visit representative sales points in thetarget city or region. As indicated in Checklist 6,the list should include information on productcharacteristics. Intermediate users (e.g., bakeriesand restaurants) and industries (such as flourmills and confectionery companies) can add to the

53

• Marketing (which includes distributionand promotion) to make the productavailable to consumers or clients. Thisconsists of two components if wholesaleand retail marketing are distinct.

• Consumption of the product byconsumers, industrial clients, or both.

list and pinpoint problems in the quality, cost, andavailability of products or their raw materials.This information may suggest marketopportunities.

To determine whether different products aremarketable, it is helpful to talk with retailers,wholesalers, and industrial users. Could traderssell more of the product if they were supplied withmore? Is price the key factor influencing sales?

Specialized libraries and documentationcenters are a good source of useful secondaryinformation. In addition, try to obtain feasibilitystudies on projects similar to yours fromagroindustrial institutes, departments of foodtechnology or business studies at localuniversities, regional offices of the agricultureministry, rural development agencies, or creditinstitutions. Statistics on urban or rural foodconsumption habits may also be available.

Ask the consumer: Concept testing

Consumers' opinions should be taken into accountas early as possible. Their tastes and preferenceswill indicate the type and quality of product thatis most likely to succeed commercially. Thequantity of product you can sell will depend onconsumers' purchasing power.

---t>..!On-farm use---... Present system

-----------... Proposed system

market

...

RuraIJurban0..'

Cassavaroots

Wheat(imported!domestic)

Cassavafarms

Cassava roots...

Flourprocess

Cassava flour-------------. Wheat

flour mills(mixing)

Wheat flour ..----------_._---- ...

Wheaticassavaflour mixes

Bakeries

Bakery products......

Bakery products

• !

Consumer I

~::';1,;....

II Socioeconomic

• Cost of production withand without new technology.

• Output prices in traditionalcassava markets.

- Potential for productionincrease with new markets.

- Changes required insystem to meet requirementsofprocessing plant.

• Potential increase in farmerincomes due to developmentofflour market.

Technical

- Investment costs ofalternative processingmethods.Alternative organizationand managementschemes.

- Per unit costs of flourprocessing.

- Model of pricedetermination for wheatand bread demand.

- Cost increase for cassavaflour production.

• Price differential requiredbetween cassava and wheatflour to motivate cassava flourproduction.

Price differential requiredto motivate use ofcassavaflour.

- Consumer preference fordifferent bread types.

• Need for a price differentialto motivate consumeracceptance.

- Flour quality ofvarietiesat different planting andharvesting dates.

• Improved productiontechnology.

Washing, peeling, drying,milling, and storagemethods for producingcassava flour.Quality control of product.

Additional equipmentneeded to produce cassavaflour.

- Storage characteristics ofcassava flour.

- Local bread types, and quality parameters.• Effect ofdifferent mixture percentages on bread quality.

Effect ofvariety, harvesting age, and processing on flourcharacteristics and bread quality.Adjustment ofbread·making techniques for cassava flour.

Figure 4. An ideal cassava flour system and major issues for research.


Checklist 6


• Market structure and costs from farmgate to consumer. This includes thenumber of intermediaries, volumeshandled, margins obtained, frequency ofpurchase, quality problems, and productwaste.

• Proposed or similar products in themarketplace.

• Characteristics of relevant productsalready on sale: color, size, form, unitsize/weight, taste, package type, andstorage properties.

• Sales volumes of relevant products,including growth trends.

• Locations where relevant products aresold (e.g., small stores, markets, androadsides).

• Product characteristics (such as priceand quality) that are critical forimproving sales and consumeracceptance (from the viewpoint ofretailers and wholesalers).

• Technology used by industry to producethe desired product or similar productsand problems with the raw materialsused currently.

Central objectives of consumer research areto determine the characteristics (such as color,shape, and size) that consumers desire in aproduct-as well as those they dislike-and todetermine which products are the most and leastaffordable. You can obtain this information byasking consumers about their food preferencesand habits in purchasing and preparing food (seeChecklist 7).

•

•

•

4-kg units for sale to middle and upperincome groups who shop weekly insupermarkets

I5-kg units for sale to restaurants andinstitutions

I5-kg units for sale to neighborhoodshopkeepers, who sell the product to lowincome consumers in smaller subunits

Consumer research separates the productsthat have high potential from those with limitedpossibilities. See Box 6 for an example of howmarket surveys and consumer interviews areused to evaluate the commercial potential ofprocessed products.

Based on the information you have gatheredthrough market and consumer research, form aseries of concepts about how different consumergroups might use root and tuber products.Suppose, for example, that you want to marketfresh cassava, packed in polyethylene bags, withgood eating quality and a 2-week storage life (asdescribed in Case 2). The product concepts mightbe as follows:

55

For a more complex product-high-qualitycassava flour for human consumption (as inCase 3)-some likely options are:

• 25- or 50-kg units for sale to industrialconcerns, which would incorporate the flourinto a range of food products

• 2.5-kg units for sale to flour wholesalers, whowould mix it with wheat flour and sell thecomposite flour to local bakers

• 250- or 500-g units for sale through localretailers to consumers for household foodpreparation

Add,ng Value to Root and Thber Crops

Checklist 7


Food preferences

• What product size, color, weight, use,storage life, preparation time, andprice is acceptable to consumers?

• Do they consider the product to benutritious, tasty, filling?

• How does the product compare withsubstitutes in terms of price, quality,and availability?

• Is the product of good hygienicquality?

• Is there a need for new or differentproducts?

The next step is to test these concepts bypresenting them to consumers through acombination of text and photographs. Try to getanswers to the following questions:

• Do consumers comprehend the productconcept?

• Are its benefits relevant to them?

• How would they use the product?

• Given the suggested unit price, would theypurchase the product?

Concept testing is a rapid, cost-effectivemethod of defining important productcharacteristics, based on consumer needs. It helpseliminate uses or ways of presenting a productthat are unattractive to consumers.

A stable supply ofraw materials

Whereas agricultural supply is naturally cyclical,the demands of processing are mostly stable. Thatis one reason why satisfying the raw materialrequirements of a processing plant can be a majorchallenge.

Habits in purchasing food

• Where do consumers purchase foodproducts, how often, and in whatvolumes?

• How aware are consumers of processedproducts in the market?

Background information

• What is the level of consumption of rootcrops and processed foods (by incomegroup and season)?

• What are the principal types ofprocessed foods available in the market(identify gaps)?

In some cases, depending on operating costsand returns per production unit, it may beprofitable to carry out processing during only afew months of the year. But in others processingmay not be economically viable unless it can bekept going year-round. To further complicatematters for processors, growers may already havea ready outlet for their produce, even in the peakharvesting period, or they may have onlyoccasional difficulty in selling their crops.

Given these uncertainties, processors mustseek answers to a number of questions. Can Iobtain reliable supplies of raw materials? Can Iget enough to justify establishing a processingfacility? Are the raw materials of adequatequality? Is the price acceptable? To answer theseand related questions requires productionn~search.

The results should define current patterns ofsupply and determine whether price movementsare such that growers will be interested in sellingto a processor or engaging III processingthemselves. Important issues for this research arethe quality and uniformity of supply and the costs

Unit 4: Research {or Product and Process Development

Box 6

Consumption ofProcessed Potato Products in Peru (Case 7)

The problem

By 1987 technical and market research on potatoin Peru had yielded valuable information onsimple processing technology (Keane et a!., 1986),traditional processing in the highlands (Werge,1979), prospects for increased consumption oftraditional processed products in Lima (Benavidesand Horton, 1979), and the feasibility ofintroducing certain types of processed productsinto the diets of low-income people in the capital(Benavides and Rhoades, 1987).

Nevertheless, not much was known aboutconsumption of various processed potato productsby different income groups in Lima. 1b gather thisbasic information, a team of social scientists wasformed at the Universidad del Pacffico in 1987(G6mez and Wong, 1988). The approaches theyemployed and the outcomes of their work areoutlined below.

Method.

• Literature review• Participant observation• Informal interviews with potato processors

and traders• Pilot consumption survey, using a structured

questionnaire

The survey covered consumers (n =199) withhigh (n =19), medium (n =81), and low incomes(n =99). They were interviewed in supermarkets,shops, stores, and markets. The preliminaryfindings were presented to a group of processorsand traders.

ReBUlt.

• The following processed potato products wereavailable:

Potato starchTradiiional dehydrated potato (papa seca)Impo~instant potatoesPotato breadPotatocnsp.sBleached,dehydrated traditional potato(papacku7io)Peeled, precut potatoes for restaurants

57

• The total annual requirement of freshpotato for the Lima market alone wasestimated at 36,000 t.

• Consumers were not aware of the range ofproducts available in Lima and wereinterested in knowing more about themand their uses.

• Consumers' perceptions of productattributes and defects were as follows:

Potato starch: tastes good but tends togo lumpy.Potato crisps: practical and readilyavailable but are greasy and salty andspoil quickly.Instant potatoes: easy to prepare buthave an acidic taste.Dehydrated potatoes: filling but havea bitter taste and are of unevenquality.Bleached/dehydrated potatoes:nutritious, but the pieces are toosmall.

• Demand for snack foods and convenienceproducts (such as crisps and instantpotatoes) was more responsive to simulateddeclines in price than that for productswith specific uses (such as potato starchand traditional dehydrated potato).

• People of different income groups variedmarkedly in their knowledge of particularprocessed products. High-incomeconsumers knew much less abouttraditional bleached and dehydratedpotatoes than did middle- and low-incomegroups.

Impact

Based on the results of this survey, productquality was improved with respect to color(specifically, yellow and gold were found to bethe most desirable colors for dehydratedpotato) as well as purity and cleanliness. Later,these improvements led to successful contractsfor test marketing of an improved product bytwo major supermarket chains in Lima.

of assembly and transportation. The topicsindicated in Checklist 8 can serve as a guide toplanning production research.

It is particularly important to determine atthis stage whether the quality of the raw materialis adequate for manufacturing the desiredproduct. 'Ib enhance quality requires a significantinvestment in research (on genetic improvementor agronomic practices), which tends to have along lead time. For that reason problems withquality can greatly diminish the feasibility of aproject.

In assessing raw material quality, considerthe factors indicated in Checklist 9. Obviously, therelevance of each depends on the end product.

'Ib obtain information of the sort described inChecklist 8, you can synthesize secondary data,review the available literature, and collectprimary data. In accumulating primaryinformation, it is particularly important that youinterview different types of representativegrowers. Try to measure the interest, not just oflarge-scale growers who seek to maximize profits,but also of subsistence farmers who may haveminimal surpluses to sell and hence a limitedinterest in commercial processing.

Data on the annual volume of cropproduction in a particular region can generally be


obtained from the Ministry of Agriculture orNational Bureau of Statistics. Most commodityresearch programs have information on plantingand harvesting dates. You can estimate the outputof different categories of farmers and determineparticular groups' share of total production fromagricultural census data or by synthesizing theresults of formal farm surveys. Producer pricesfor specific crops (by calendar year or month tomonth) are often available from the Ministry ofAgriculture.

Analyzing information on production may besufficient by itself for gauging the potential ofcertain crops or regions for processing. Forexample, if an area produces only a small amountof sweet potatoes for the fresh market and pricesare always high, farmers are unlikely to beinterested in sweet potato processing.

On the other hand, if your analysis suggeststhat there are good possibilities for producing aparticular product at a given location, check withresearch centers (such as university libraries andagroindustrial institutes) for information on past,farm-level attempts to introduce or improveprocessing. Examine closely the objectives, trackrecord, and major constraints of previous orongoing experiments or projects that have provenunsuccessful.

Checklist 8


• Varieties planted, planting and harvesttimes j and seasonality of supply

• Experience of farmers in producing thecrop

• Postharvest grading by size and qualityand use by grade of produce

• Reasons·Carmers prefer particularvarieties

• Rural marketing channels, seasonalstability, typical traders, volumes traded,and forms of payment

• Potential for changing the existing systemwith regard to variety used, harvest times,marketing arrangements, etc.

• Farmers' potential or limitations forproducing andlor marketing processedproducts


Checklist 9


• Physical: root size, shape, anduniformity; peel thickness and color;parenchyma color and hardness

• Chemical: dry matter, fiber, starch, etc.;presence of antinutritional factors ortoxins

Formal or informal methods?

In general, you can make sound decisions aboutproduct development on the basis of indicative, asopposed to definitive, information. For that reasonunstructured survey techniques are usuallyadequate for market, consumer, and productionresearch. They are also faster and cheaper thanformal surveys.

Only in certain cases (e.g., a consumersurvey in which it is important to differentiateamong income groups) is the extra precision of aformal survey warranted. A sample of 50-200consumers should be adequate. Pretest formalquestionnaires with a small group of targetrespondents.

Formal methods may also be necessary forsampling farmers' opinions and getting at theunderlying reasons for them. The larger thenumber of alternative products and processes, themore precise the quantitative informationrequired and the more important it is that youconduct a formal survey. If you can reduce thepossibilities to one or two (e.g., through analysisof secondary data), informal interviews may besufficient.

If you need information from consumers of aparticular product, interview purchasers of theseproducts at the point of sale rather than conduct ageneral consumer survey through householdvisits. The type of consumer you interviewdepends on the product. For example, if you arestudying the prospects of intermediate products

• Organoleptic: aroma, taste, and texture

• Functional properties (of starch andflour): viscosity, etc.

(flours, starch, etc.), industrial clients,restaurants, and bakeries should be the principalsources of information.

Interviews should take place at theconvenience of the respondents, not theinterviewers (busy traders and tired housewiveshave little time to spare and under pressure maygive inaccurate replies). Conduct interviews in alow-key, nonaggressive fashion to gain theconfidence of respondents (this is especiallyimportant for market agents and industrialists);they will then be more likely to answersupplementary questions aimed at resolvingcontradictions.

Whether you are preparing structured orunstructured surveys, phrase each questioncarefully to avoid ambiguity. Collect only theessential information to reduce costs and improvethe quality of the results. To make the best use ofsurvey results, the project must be able to processand analyze the data rapidly.

If you have little experience ill this kind ofresearch, it may be helpful to memorize a list ofthe topics you wish to cover during informalsurveys. Compare the results with the opinions oftechnical specialists.

Where possible the same team of researchersthat compiles the secondary information shouldalso carry out the market and consumer surveys.The team should consist of technical as well associal scientists. We cannot overemphasize thelmportance of getting technical researchers out of

the lab or workshop and into the market. Thisinterdisciplinary approach can open newavenues for technical research and help sharpenits focus on the most relevant problems andopportunities.

Addinl{ Value to Root and Tuber Crops

Box 7 illustrates the use of a formalquestionnaire to evaluate the potential of aparticular process. (For more information onsurvey techniques, see the list of publications atthe end of this unit.)

Box 7


The problem

Potato researchers in northeasternColombia decided to explore the potential ofsimple potato processing in an effort todevelop alternative uses and markets forthe crop and thus stabilize prices andimprove growers' incomes. They set up asmall pilot plant in the Pamplona region todemonstrate that processed potato productscould be produced, using simple technologyand local varieties, infrastructure, andtechnical personnel. Afterwards, a researchproject assessed the socioeconomicfeasibility of the technology, emphasizingthe needs and interests of local producers.The outcomes of this study are outlinedbelow.

Methods

• Review of available studies andsecondary data on potato productionand marketing in the region

• Demonstrations for local producers ofproducts and dishes prepared with them(e.g., cakes and soups)

• Formal survey of 81 growers from fivedistricts in the region. Information ontechnical aspects of potato productionwas collected, and farmers' interest inprocessing was assessed.

Results

• Sixty-three percent of the farmersinterviewed had produced potato forover 20 years. In 1986 average potatoproduction per farmer was 52.7 t, ofwhich only 9% was sold.

• Most potatoes were sold from Septemberto January, when supplies areabundant, and less than 40% in AprilMay, when supplies are lower and priceshigher. Farmers rarely stored the cropfor sale at a later date.

• Growers reported that, although priceshad fluctuated somewhat, they had notbeen low for several years.

• Seventy-three percent of farmers reliedon family labor; shortages at harvesttime were common.

• Most farmers (72%) were familiar withpotato crisps, but 28% knew of noprocessed potato product.

• Almost all growers were interested inprocessing part of their production toimprove prices, diversify the family diet,and provide employment for householdmembers. They mentioned suchproducts as potato chips, potato flour,and French fries.

Impact

Though interested in processing, farmerslacked a compelling incentive to pursue it,since they generally had little difficulty inselling their harvest at reasonable prices.Moreover, the shortage of labor at harvestwould have made it practically impossiblefor them to adopt a labor-intensivetechnology. If this survey had beenconducted earlier, researchers could havelocated the pilot plant in a more suitableregion.


Developing the research agenda andproduct brief

The research agenda comprises the topics youmust investigate to supply missing componentsand links in the processing system and to provideinformation for a prefeasibility study. Theresearch needed for the first purpose should beevident from your analysis of components andlinks (discussed above). The elements of aprefeasibility study are listed in Checklist 11 onpage 70.

Based on the results of market, consumer,and product research, you should be able toprepare a product brief (see the examples inTable 16). This document gives the specificationsof the final product as well as its raw materialand processing requirements. Some productsrequire more than one process or a combination ofraw materials. Processing more than one rootcrop (assuming that their harvest times differ) orother crops, such as plaintain, with similarprocessing requirements may help you increasethe number of months per year the plant canfunction.

Table 16. Product briefs for two cassava products.

Dry cassava for animal feed (Case 1)

Technical Research on theProduct and Process

In this section we describe technical research,whose purpose is to ensure that the productmeets the expectations-in terms of price andquality-of consumers or clients. This researchfocuses both on the product itself and on theprocess by which it will be made.

There are two types of product: first andsecond generation. The former results from whatwe describe as primary transformation. Examplesof first-generation products of roots and tubersare flour and starch as well as fresh roots thathave been selected, treated, and packaged toimprove their presentation and prolong their shelflife.

Products such as flour and starch are oftenused as raw materials in secondarytransformation, which gives rise to secondgeneration products. In some processes of thistype, other ingredients are added to the rawmaterials (as in the production of balanced animalfeed rations and composite flour) without altering

Fresh stored cassava for human consumption (Case 2)

Product

Raw material

Processing

Packaging

Marketing

Consumers

A carbohydrate source in the form of dried cassavachips for incorporation into balanced feed rations tocompete with traditional carbohydrate feeds suchas sorghum and maize.

Cassava roots with high dry matter and low/intennediate cyanide content.

Roots washed, chipped, and dried by either naturalor artificial means; processing to be carried out byfarmer co-ops or small- to medium-scaleagroentrepreneurs.

The dry root chips to be packed in 50-kg sisal orpolypropylene sacks; closure of sacks to be donemanually or by machine.

The dried chips to be sold directly to animal feedconcentrate companies or to livestock producers;promotion to emphasize high starch digestibility,availability and relative cost

See marketing above.

61

A high-quality fresh cassava root product with astorage life of up to 2 weeks.

Cassava roots, selected by size and eating quality.

Roots treated to suppress physiologIcal and microbialdeterioration; treatment and packing to be done byfarmer co-ops or assembly agents.

Roots to be packed (in polyethylene bags, plasticcrates, or wooden boxes) according to market outlet.

The stored cassava to be sold through supermarketsand local neighborhood shops; promotion toemphasize freshness and storability

Families from all socioeconomic groups, favoringthose with poor market access

their physical characteristics. Other processesmodify the raw materials, biochemically orphysically, through cooking, extrusion,fermentation, and so forth.

Investors in processing

One vital question that may still be unansweredat this point is who will invest in manufacturingthe product. Though in any situation you willhave various options, it may not yet be possible toselect one. This is not an issue if your purpose issimply to improve the marketability of an existingproduct by reducing production costs orimproving quality. But if you intend to establish anew product, look seriously at the followingcandidates:

• Existing agro- or food industries: Thisalternative, which you can explore whileconducting market research, is mostappropriate for making second-generationproducts, since a certain level of skill andexperience is required to achieve the desiredquality.

• Entrepreneurs: You may be able to identifyinterested parties through groups ofentrepreneurs, such as food manufacturersassociations or growers federations. Theentrepreneur need not have previousexperience in agroindustry but should beinvolved in the project from an early stage.

• Farmers organizations: Producerassociations or co-ops may be the idealcandidates, since they have direct controlover raw material supplies and may alsohave the necessary managerial andadministrative capability. In some cases,though, it may be necessary to form such anorganization to get processing established. Ifso, it should normally focus on firstgeneration products, at least initially.

Your choice among these alternatives willgreatly influence the technology selected formanufacturing the product.

1Wo stages ofresearch

Technical research takes place in two stages. Inthe first your principal aim is to develop, on a lab

Addwg Value to Root and Tuber Crops

or bench scale, a prototype product withacceptable physicochemical and, in the case offoods, organoleptic characteristics. A taste panel isessential for checking the acceptability of a foodproduct. In the second stage, you select anddevelop prototype equipment for manufacturingthe product under experimental conditions. Thisequipment can then be incorporated into the pilotplant. Consult Checklist 10 for ideas in planningtechnical research.

Raw material quality

In Unit 1 we discussed the physical or mechanicalcharacteristics and chemical constituents of rootand tuber crops. These traits determine to a largeextent the type of processing technology used.

The end product results from the interactionbetween raw materials and processes. To take asimple example, a root crop with a high drymatter content (> 40%) can be dried at low costthrough natural drying, whereas a root with only15% dry matter will take longer to dry andrequire more sophisticated drying procedures andequipment to achieve comparable product qualityafter drying. In general, process economicsimprove with increasing dry matter or starchcontent; that is, fewer tons of raw material arerequired to make a ton of end product.

If the raw material has toxic orantinutritional factors, extra processing ormcreased processing time may be required toeliminate or reduce them to acceptable levels.Table 17 shows how particular standards in endproduct quality can be met through processing orimproving raw material quality.

The nonuniform size of some roots andtubers reduces the efficiency of processing,particularly during initial operations, such aspeeling. One solution is to grade roots by sizebefore peeling, but this adds to labor costs.Another option, hand-peeling, is also laborintensive and in some areas possiblyuneconomicaL It may still be Justified, though, asa source of employment.

The right raw material specifications arethose that, in combination WI th the appropriateprocesses, give a product of the requiredquality. If only a small percentage of the


Checklist 10

MainActivities of Technical Research for Product Development

Laboratory research

• Make the product in small quantities.

• Have taste panels and consumers testthe product.

• Test product quality. using chemical andfunctional methods.

• Define raw material standards.

• Define. select. and test process optionsona smallscale.

Prototype development

• Determine the scale on whichprocessing will take place.

.------------------------- - - ---------

• If necessary, develop and test equipment.

• Determine the layout of process operations.

• Test product quality, process efficiency, etc.

• Revise raw material and product qualityspecifications.

• Determine process operating parameters.such as conversion rates and amounts ofinputs (fuel, labor, water. and raw material)required per ton of product.

• Determine packaging materials andstorage conditions and times.

• Establish the costs of equipment andinfrastructure and amount of workingcapital needed.

Table 17. The relationship between end-proouctquality, the process, and raw material.

End-product quality

Dry matter < 14%

Low microbial count

Fiber content < 3%

Protein content> 5%

White color

Process

Natural or artificial drying

Hygienic conditions,water treatment, rapid drying

Peeling, sifting

Fermentation

Sulphating, rapid drying,removal of impurities,water treatment, rapid processing

Raw material

High dry matter content needed

Absence of preharvest rots, etc

Age at harvest, variety, environmentalconditions

Variety

Variety

harvested roots and tubers meet thosespecifications, farmers must have a good marketfor rejected roots or receive a price thatencourages them to cater for such a demandingmarket.

If locally available varieties fail to meetquality standards, you can consider introducingnew ones. Where promising varieties areavailable, they will need to be evaluated on-farm.For some crops, such as cassava, multiplyingenough planting material can take considerable

63

time. If no suitable varieties can be found, it maybe necessary to initiate a breeding program.

If so, should new varieties be developedspecifically for processing, or must they satisfythe requirements both of processing and thetraditional market? In varieties aimed at thefresh market, organoleptic and other factorsrelated to consumer preference are important. Thbreed these traits into new varieties that haveimproved processing attributes may be complexand time-consuming.

On the other hand, farmers may resist theintroduction of an "industrial" variety that isunacceptable to the fresh market, unless they areprovided with price incentives, a secure market,or both. Farmers prefer a variety that can be soldin several markets. Being overly dependent on asingle market is risky for them, because itexposes them to the danger of price fluctuations.If you plan to introduce an industrial variety, youmust also find a way to compensate farmers forincreased risk.

Product quality

The results of consumer and market studiesshould give you a clear idea of the end product'scharacteristics and quality. The product may alsohave to meet certain legal standards (such asthose established by governments for foods).Since both the raw material and processdetermine end-product quality, you need toidentify the aspects of both that are crucial formeeting quality requirements economically (seeTable 17).

At the outset of technical research, monitorthe quality of the end product as well as theefficiency of the process. The most importantquality traits are chemical composition,functional properties, and use characteristics(e.g., storage time), which affect the product'sappearance, organoleptic properties, hygiene,and performance.

Selecting technology and equipment

Your choices should be based on various factors,including:

• Scale of the operation: If the value addedby processing roots and tubers is to remainwithin rural communities, small-scaleprocessing is preferable. Many processes aredesigned to operate on a larger scale than isfeasible for rural enterprises. Even so,small-scale processing is carried out in manyparts of the world, and much research hasbeen conducted in support of this activity. Ifyou are introducing a novel process orscaling down a large operation, research willbe needed to develop suitable equipment.

64

•

•

•

Adding Yhlue to Root and Thber Crops

Capital investment: The more capital theenterprise requires, the more likely it is tofail. This is especially true for small-scalefarmers organizations that depend on creditfor investment in infrastructure andequipment. As a general rule, newenterprises should start with simpleprocesses requiring little capital investment.Only after gaining some business experience,should they expand into more complexprocesses requiring greater investment. In acassava project in Colombia, for example, acassava flour process was introduced to agroup that was already managingsuccessfully a simpler, natural drying processfor producing animal feed.

Wherever possible, use equipment that is orcan be manufactured locally. This not onlycreates jobs locally and reduces costs butsimplifies maintenance and repairs. If at allpossible, avoid using imported equipment.

Conversion or extraction rate: This isprobably the single most important variableaffecting the financial feasibility of theprocess. It depends both on the raw materialand process. Choose the process that offersthe best conversion or extraction rate andincurs reasonable energy and labor costs. Ifnone of the available equipment meets thesecriteria, you may have to develop new ormodified equipment.

In the cassava flour project mentioned above,this was necessary in two cases. First, toachieve a good fresh-to-flour conversion rate,root peeling had to be dropped in favor ofpeel removal during milling and separationof dried chips. Second, a small-scale mill hadto be developed that could convert chips intoflour at a rate of 80%-85%.

Amount of value added: If the cost of theraw material is high relative to the value ofthe final product (i.e., if the amount of valueadded is low), costly processing is not viable,unless the volume of production is high. Inconstrast, an end product aimed at highincome or export markets requires a certaintype of processing and equipment to meethigh standards in quality and packaging.


•

For example, in producing dried cassava foranimal feed, natural drying of the cassavachips is essential for keeping the price of thisproduct low. Only when chips are to bemilled for human consumption-and thussold at a higher price in a differentmarket-is artificial drying a viable option.

Availability ofservices and utilities:Many processes require potable water,electricity or natural gas, or goodcommunications infrastructure (such asroads, telephone, and radio). But roots andtubers are frequently produced in the poorestareas with the least access to basic services.Under these circumstances, the project'soptions are to:

Develop a process that can be carriedout only in those limited areas whereutilities or services are available-andthus restrict the benefits to communitiesthat are already better off than most.

Develop a process that is suitable for allareas (e.g., by using biogas energy,water power, natural drying, etc.).

Divide the process into two stages. Thefirst would require few, if any, externalinputs. The material would then betransported to a centralized secondaryprocessing plant that has the utilitiesand services required to manufacturethe final product.

•

•

•

Here are a few other points to consider indeciding what equipment to use:

The different items of equipment should becompatible in terms of capacity. Look forpotential bottlenecks and decide how youmight expand capacity: by duplicating theoriginal process or by increasing the capacityof one or more items of equipment.

Some processes (e.g., root peeling and slicingand starch separation) can be carried outeither manually or mechanically. Animportant advantage of mechanizedprocesses is that they give uniformthroughput and product quality. Manuallabor, on the other hand, generatesemployment, especially among women andthe elderly. But if labor is expensive it mayalso greatly increase costs. Where genderissues are particularly important, it may bevaluable to use manual labor in at least onestage of a process. Keep in mind thatmechanization often means replacing manywomen with one man.

If the project requires a new type ofequipment, try to pick up useful ideas (e.g.,about design, power sources, andconstruction materials) from otheragricultural processes in the area or fromprocessing equipment employed for roots andtubers in other parts of the world beforeembarking on prototype development.

•

Bear in mind that some services, such astelephones, may be essential for marketingthe product.

Managers and operators' level ofeducation: The process should not be socomplex or require such precise control overcertain variables (such as dryingtemperature and the chlorine content of thewater) that rural people will have difficultycarrying out or monitoring it. Literacy isessential for certain processes and formaintaining business records. Even so,illiterate farmers are quite capable ofperforming many tasks (e.g., judging themoisture content of dried cassava chips basedon empirical measures, such as chip texture).

65

Designing equipment is a specialized taskrequiring the services of qualified mechanicalengineers. Few national agricultural ordevelopment institutions have such specialists.But they may be found in universities, industrialresearch centers, and technical colleges. It mayalso be useful to link up with local metalworkshops that have experience in manufacturingsmall equipment. Designing and testingprototypes of small equipment has proved to be anexcellent subject for student thesis projects. SeeBox 8 for an example of equipment selection.

Laboratory taste panels

Laboratory analysis of an end product canindicate whether it meets certain quality

(~-------

!

Addmg Value to Root arul Thber Crops

Box 8

Selection ofEquipment for a Cassava FlourProject in Colombia (Case 3)

The production of high-quality cassava flourfor use in food products is a simple process,involving washing and peeling roots;chipping or cutting them into uniformpieces; and drying, milling, and sifting thechips or pieces. Much research has beenconducted on the design and development ofsmall-scale equipment for processing rootcrops. Sometimes locally availableequipment is suitable for milling and otherpurposes.

The first steps in selecting equipmentfor this project were a literature search andvisits to local agricultural engineeringcompanies and other agroindustrialente that operate on a fairly smallseal ch as coffee drying and maizeprocessing plants in Colombia).

er/peeler: Equipment based onm the literature were built and

teste . Although it removed the peeleffectively, this eq caused a largeamount of the root p yma to be lost.Project staff selected instead a local washerused in small-scale plants to extract cassavastarch. This equipment removed much ofthe outer bark but not the peel itself.

Root chipper: Prototypes based on designsfrom Brazil, Malaysia, and Thailand were

specifications. But only tests with human subjectscan tell whether it also possesses more subjectivetraits, including an appealing taste, aroma, feel inthe mouth, texture, etc. Since these traits areessential for gaining consumer acceptance, it iscritical that you evaluate the product for themduring the research phase.

For each product attribute (e.g., sweet tasteand hard texture), taste panels can provide twotypes of information: an intensity rating and anacceptability rating. Only a trained panel canprovide the first type, because this is more or less

6f1

built and tested. The best features of theThai and Malaysian designs wereincorporated into a composite design.

Root chip drying: Artificial drying wasfound to be more suitable than naturaldrying, both for reasons of product qualityand because it would permit year-roundprocessing. Through-circulation bin driersfor cassava, developed at the University ofVicosa in Brazil, were adapted toColombian conditions.

Coal-fired bu.rner and fan: Initialeconomic studies. suggested that coal,which is abundant and cheap in Colombia,would be the best fuel. A customizedversion of a local proprietary design wasbuilt through a student project. The fanwas purchased locally.

Dry chip milling and flour sifting:Local hammer and other mills were testedbut could not provide high enoughextraction rates for cassava flour. A localmachine shop developed a small-scalesifter, based on a larger Swiss design. Localwheat flour mills also milled chips in thestandard roller mills, with an extractionrate of more than 85%, from washed,unpeeled roots (the peel was removed inthe by-product).

objective information; it has no relation toconsumer likes and dislikes, which can vary byregion, income group, etc. The second type ofinformation, in contrast, is directly related to thepreferences of taste panel members. You cannotassume that other sectors of the population willgive the same responses. For this type of testing,relatively untrained panelists and less controlledconditions are suitable.

Laboratory taste panels can test productsfairly objectively under controlled conditions. Forthis purpose it is not enough just to gather the lab


Some general evaluation criteria andsample rating scales are listed below:

• Intensity of quality characteristics relativeto a standard (e.g., much sweeter than X)

• Intensity of particular qualitycharacteristics (e.g., for sweetness: low tovery intense)

workers for half an hour every week. You needto select panelists who are representative of thetarget population and have a good sense oftaste, smell, etc. (You can screen candidates forthis latter ability by asking them to identify andrank sweet, salty, bitter, and acid solutions ofdiffering concentrations).

ingredients, cooking times, etc.) vary greatly, asdo their perceptions and food preferences. As aconsequence, the results of this evaluationwill reflect various factors, giving an overallmeasure of consumer acceptability or intent topurchase.

To organize and conduct tests in theconsumers' homes can be very time-consuming.You may have to visit some participantsrepeatedly; drop-out rates will tend to be high;and some consumers may not follow theinstructions properly (e.g., on food preparationdates). If the objective is to gauge the response ofhigh-income consumers, you may have difficultygetting enough people to participate. Because ofsuch difficulties, only a limited number ofdifferent products can be tested in this way.

Packaging and shelflife

Because consumers' opinions are so variable,your sample should be relatively large. It shouldalso represent the target population accurately,with respect to income level, proportion living inrural and urban areas, family size, etc. In urbanareas census data can help you determine theproper composition of the sample.

Acceptability (very acceptable to notacceptable)

•

Though you need not construct a specialfacility for testing products, it is important thatthe panel meet in a quiet place with controlledlighting and that samples be presented inrandom order. This will keep human and othererrors to a minimum. Early on, the groupshould hold discussions to reach a consensus onthe quality characteristics it will evaluate. Therating scales used should be easy to understand.

• Preference relative to standard (prefer X toY)

• Hedonic evaluation (like or dislike)

The literature contains many examples ofproduct testing (see Watts et al., 1990, in the listof publications at the end of this unit). To ensureproper interpretation of the results, subjectthem to statistical analysis; the method you usedepends on the evaluation scale andexperimental design.

Consumer testing ofproducts

Evaluating food products in the lab is notenough. At some point in the latter stages ofresearch (certainly no later than the pilotphase), they must also be tested by ordinaryconsumers in their own homes. Consumers'methods of preparation (including use of other

Near the end of the research phase, you will needto decide on packaging and determine theappropriate shelf life. You can obtain much of theinformation needed for this purpose fromconsumer and market research. Inquire about thelength of time that consumer goods normallyremain in shops and homes and industrial goodsin storage before further processing. Throughconsumer surveys, you can gather other usefulinformation, such as appropriate package sizesand storage conditions at home and in stores.

Next, you should evaluate a range ofpackaging materials under typical storageconditions for changes in chemical composition,functional properties, appearance' (especiallycolor). and organoleptic characteristics (usingtaste panels) as well as for signs ()f insects orcontamination by microbes.

Based on the results, you can then design thehest package, taking into account the product's

67

stability over time and the cost of materials. Inpreparing the prefeasibility study, be sure toindicate packaging costs, including that ofprinting logos, instructions, and so forth.

Product name

Even in the research stage, it is not too soon toselect a name for the product and have it legallyregistered. Often, researchers give little thoughtto this task, using names that have littleconsumer appeal. The product name should:

• Highlight the benefits of the product forconsumers.

• Suggest product attributes (such as color,use, and taste).

• Be easy to pronounce and recognize.

• Be distinct from names of similar products.

A well-chosen name will make the finalproduct more marketable by associating it withcharacteristics that consumers like. A fast andeasy way to test product names is to interviewconsumers who purchase similar products (see theresults of one such survey in Box 9). Repeat thisexercise in each of the product's potentialmarkets.

As is evident from the survey described inBox 9, legal registration of a brand name isimportant. The first step is to search the lists ofregistered names for the one you have selected(many names are registered but not usedcommercially). If the name is not listed, a lawyerwill be needed to register it. This will preventothers from using the same name once theproduct is established.

The product name must have an owner.Some options are the project's executing agency, afarmer group or cooperative, or a small business.An even better candidate is some second-orderorganization that supports the project, since it canrepresent all the organizations involved.

The prefeasibility study

The purpose of a prefeasibility study is todetermine whether a particular product andprocess should go on to the pilot stage. This

Ac1dtng Value to Root arui Thber Crops

decision should be based on a careful analysis oftechnical, economic, financial, and commercialinformation (see Checklist 11). Although much ofthis information may still be tentative, it shouldbe adequate for making reasonable assumptionsabout such matters as the production capacity ofthe plant, the processing technologies to use, theraw material inputs and total investment costs,production costs, sales revenues, and returns oninvestment.

Based on these assumptions, you candetermine rates of return and other financialinformation by means of a SImple financialmodel (Ostertag and Wheatley, n.d.). At thisstage or even earlier, the model can help youassess the merit of different options inprocessing. For example, if several dryingsystems show potential, you can feed theinvestment and operating costs of each into themodel to determine which is best. Another use ofthe model is to determine the minimum size atwhich the pilot plant can operate profitably. Ifthe proposed operation appears unprofitable, themodel can help identify ways to improve theprocess, reduce costs, and so forth.

The types of research described in this unitshould generate enough information to preparea prefeasibility study and decide whether to setup a pilot plant in the target region. (For moreinformation on feasibility analysis and relatedtopics, see the list of publications at the end ofthis unit).

A Preliminary Status Report

By the end of the research phase (as indicated inTable 19), the product has been produced on asmall scale, using prototype equipment. It hasalso undergone a preliminary consumerevaluation. You have determined theapproximate costs both of capital and operationsand selected the target region. Market researchhas helped define the product's potential. If theresults of the prefeasibility study are positive,you can proceed to the pilot phase, in which theproduct and process will be tested on a smallscale in a commercial environment. At this pointyou need to identify the enterprise or group thatwill manage and operate the pilot plant.


Box 9


In preparation for marketing storablecassava roots, packaged in polyethylenebags and treated with a thiabendazolebased chemical, project staff interviewedconsumers in supermarkets as well asneighborhood shops and markets in threeColombian cities. Each participant wasasked to select one product name from a listof five. The results are given in Table 18.

Consumers showed clear preferences forsome names over others. Interestingly, twonames that are virtually identical, Yucaricaand Ricayuca, received very different scores.The two highest scoring names, Yucafrescaand Superyuca, both suggest the specialcharacter of storable cassava. Superyuca wasselected, since Yucafresca had already beenregistered by another enterprise.

---------------- -----

Table 18. Consumer responses to product names.-_._----,.""._-----

Percentage of consumers interviewed

Bogota Cali Barranquilla Average

20 24 20 21

18 22 36 29

10 12 10 11

41 26 26 31

8 14 4 9Delicious cassava

English translation

Tasty cassava

Super cassava

Tasty cassava

Fresh cassava

Yucarica

Name

Superyucai

I

I

, i Ricayuca

i I Yucafrescai I

! i Deliyuc.a

L:=------

Table 19. Status of a product development project at the end of the research phase (phase 2).

Phase 1- Start Phase I-Finish Phase 2-Fmlsh

Why (objective)?

Where (region)?

What (product)?

How (process)?

Defined

-'

Defined

General area identified

Idea selected

General process identified

Well defined

General area identified

Idea and concept tested

Options considered, tested andbest selected

How much (market)?

By whom (type of enterprise)?

For whom (beneficiaries)?

Loosely identified

Defined

a. May be defined in project objective.

69

AddIng Value to Root and Tuber Crops

Checklist 11


• 7echnical feasibility: Whether theproduct can meet quality standards andbe made efficiently depends on variousaspects of the:

Raw material-time betweenharvest and processing, variety,time to maturity, levels of damagecaused by root rots and otherdiseases or pests, dry mattercontent, and other root qualityattributes

Process-conversion rate for freshroot to final product, raw materialrejection rate, time and laborrequirements for all stages of theprocess, amount of other inputs(e.g., water, electricity, fuel, andother raw materials) required perton of final product, andparameters affecting productquality (e.g., drying andfermentation temperatures andpH)

End product-moisture content,chemical composition, purity,microbial standards, color, particleor unit size, and storage life

• Commercial feasibility: Whether aproduct of the required quality can besold at the price offered depends onmarketing variables:

Product-quality and availabilitythroughout the year

Price-ability to compete withsimilar products and potentialmarket response to increases ordecreases in price

Distribution-potential channelsand size of margins

ill

Promotion-approach used,expected effect on demand, andcost-effectiveness

• Financial feasibility: Whether theproduct can yield the required rate ofreturn on capital invested depends on:

Size of capital investmentestimated costs of infrastructureand equipment

Working capital needed-for rawmaterial and other inputs

Scale of operation-plant capacityand utilization rate

Processing costs-number of unitsand cost per unit of each input(e.g., raw materials, water, fuel,labor, etc.) per ton of final product

Margin expected-sale price offinal product and any by-products(weighted average price)

• Business feasibility: Whether theenterprise can be run profitably withoutoutside support in the medium termdepends on:

Type of business proposedcooperative or commercialenterprise

Managers and workers' level ofeducation

Training needs and capacity tosupply training

Complexity of the process

Other aspects of the business-rawmaterial supply, marketdevelopment, accounting, etc.

Unit 4: Research fur Product and Process Develupment

References

Anderson, AM. and Earle, M.D. 1985. Systematicproduct design. In: Product and processdevelopment in the food industry. HardwoodAcademic Publishers, Amsterdam, Netherlands.

Benavides, M. and Horton, D. 1979. La perspectiva delconsumo de la papa seca en Lima, Peru. CIP,Lima. (Mimeo.)

Benavides, M. and Rhoades, R 1987. Socio-economicconditions, food habits and formulated foodprograms in the Pueblos J6venes of Lima, Peru.Archivos Latinoamericanos de Nutrici6n37(2):259-281.

G6mez, R and Wong, D. 1988. Procesados de papa: Unmercado potencial. Centro de Investigaci6n de laUniversidad del Pacifico (CIUP), Lima, Peru.

Keane, P; Booth, R; and Beltran, N. 1986. Appropriatetechniques for development and manufacture oflow cost, potato-based, food products indeveloping countries. CIP, Lima, Peru.

Ostertag, C. and Wheatley, C. n.d. A financial modelfor the implementation and evaluation of smallscale agroenterprise. In: Scott, G. (ed.). Methodsfor agricultural marketing research in developingcountries. CIP, Lima, Peru. (In press.)

Watts, B.M.; Ylimaki, G.L.; Jeffery, L.E.; and Elias,L.G. 1990. Basic sensory methods for foodevaluation. IDRC, Ottawa, Canada.

Werge, R 1979. Potato processing in CentralHighlands of Peru. Ecology of Food and Nutrition7:229-234.

Further Reading

Survey techniques

Ashby, J.A 1990. Evaluating technology with farmers:A handbook. CIAT, Cali, Colombia.

Cervinskas, J. and Young, RH. 1990. Communitynutrition research: Making it rapid, responsiveand relevant. IDRC, Ottawa, Canada.

71

Holtzman, J. 1986. Rapid reconnaissance guidelines foragricultural marketing and food systems researchin developing countries. Michigan StateUniversity International Development Papers,Working Paper no. 30. Department ofAgricultural Economics, Michigan StateUniversity, East Lansing, MI, USA

Horton, D. 1982. Tips for planning formal surveys.Social Sciences Department, Training Document1982-6. CIP, Lima, Peru.

Khon Kaen University. 1987. Proceedings of the 1985International Conference on Rapid RuralAppraisal. Rural systems research and farmingsystems research projects. Khon Kaen, Thailand.

Rhoades, R. 1982. The art of the informal agriculturalsurvey. Social Sciences Department, TrainingDocument 1982-2. CIP, Lima, Peru.

Hhoades, RE.; Sandoval, VN.; and Ragalanon, C.P(cds.). 1991. Asian training of trainers on farmhousehold diagnostic skills. CIPIUPWARD, LosBanos, Philippines.

Product testing

Watts, B.M.; Ylimaki, G.L.; Jeffery, L.E.; and Elias,L.G 1989. Basic sensory methods for foodevaluation. IDRC, Ottawa, Canada.

Prefeasibility studies

Anderson, AM. and Earle, M.D. 1985. Systematicproduct design. In: Product and processdevelopment in the food industry. HardwoodAcademic Publishers, Amsterdam, Netherlands.

Austin, ,J. 1981. Agro-industrial project analysis. JohnHopkins University Press, Baltimore, MD, USA

Ostertag, C.l". 1992. Comparaci6n de la rentabilidad dediferentes procesos de yuca en Colombia. In:Desarrollo de productos de rakes y tuberculos.vo!. II - America Latina. Memorias de un taller.CIP, Lima, Peru.

Wilkinson, B.H.P 1985. Feasibility studies for thesmall business. In: Product and processdevelopment in the food industry. HardwoodAcademic Publishers, Amsterdam, Netherlands.

Unit 5

The Pilot Phase

I

In this unit we describe the pilot stage ofresearch and development, focusing on anexperimental product and process. The purpose ofthis stage is to generate enough information todetermine whether the project is justified inexpanding to a commercial scale.

The pilot stage cannot be undertaken withina research institution. It must be located in theenvironment where the project's beneficiariesthe people who will use the technology-live andwork. The idea is to test the product and processunder the conditions in which it will eventuallyhave to survive commercially, though on a smallerscale. If the aim of the project is to increasefarmers' incomes by getting them involved inprocessing roots and tubers, the pilot stage is thetime for them to start. Farmers can gain valuableexperience by operating and, if possible,managing the pilot plant.

Pilot testing of a new product and processmust not take place in isolation. Rather, it shouldbe closely linked to research and extension aimedat improving production of the target crop as wellas to systems that provide institutional support(supplying credit, helping processors acquirebusiness skills, and so forth). These activities arekey components of integrated projects, which wedescribed in Unit 2. The pilot phase is whenproduct development should become part of anintegrated project in the target region.

This is also the time for the project to makecontact with local manufacturers of equipment. Ifyou have developed new or modified processingequipment during the research phase, it isimportant to verify that large- or small-scalemetalwork shops can replicate this equipment ata reasonable cost.

In this section we first discuss key issues andtasks involved in setting up the pilot plant. One ofthe early steps, as indicated in Figure 5, is toselect a suitable site. Then, we cover key aspectsof plant operations, which should first take place

on an experimental basis and then on a largerscale under semicommercial conditions. If theproduct is aimed at the consumer market, youmay want to evaluate it in a test market. Finally,you need to prepare a feasibility study, whichconcludes the pilot phase of product development.

Setting Up the Pilot Plant

In the following sections, we describe the stepsleading up to the start of processing. Thoughsome of these, such as construction, are quitecomplex, they mostly involve knowledge and skillsthat are commonly available. Instead ofdescribing those steps in detail, we comment hereon important decisions and measures that haveparticularly to do with setting up a processingplant.

Organizational models

Processing roots and tubers usually demands agood deal of teamwork. To market the finalproduct calls for additional effort and skills. Abasic accounting system is needed to control thepurchase of inputs and sale of outputs. To ensurethat both the work and income of the enterpriseare distributed efficiently requires carefulorganization.

Here are several alternative models fororganizing workers:

• Individual worker or familylhouseholdenterprise

• Cooperatives or associations

• Small businesses, consisting of one or moreowners with employees

• State-owned enterprises

Co-ops and associations are socially oriented.By providing services and generatingemployment, they can distribute significant

Integrated projects:MacroplanningMicroplannig

Crop productionresearch

\

"Site selection for

pilot project

Pilot project designbased on site

Plant construction

'"


Project formulation

Equipmentmakers

.1 Plant operation

l,- Experimental- Semicommercial

!

Feasibility study

Test market

--r. .---i.

OK?

Yes

No

.'

•

1· Rede~ign Pilo~p~je~~-J

J Ch~;e location.,-l.L ~__~

J.Feedba~k~~~~search~-l ~ .__ ._ ... ~__.. _

~ Abort1. .._--

Commercial expansion

Figure 5. Flow diagram showing stages involved in pilot phase.

74


benefits to target groups. In many countries,though, these organizations have a history ofabsolute or relative failure. Even promising co-opshave had difficulty consolidating and expandingtheir gains. One reason is that in some placescultural factors favor individual rather thancooperative action.

Businesses, in contrast, operate for profit,much of which is divided among the owners orshareholders according to their stake in theenterprise. Provided that the rate of return oninvestment is attractive, business may offer thefastest means of developing a rural agroindustry.There is a risk, though, that a large share of thereturns may not reach the project's intendedbeneficiaries.

State-owned businesses, marketing boards,etc., have a sad history of inefficiency andcumbersome bureaucracy. Frequently, politicalpressures adversely affect their decisions oncommercial matters.

One alternative is an intermediate model, inwhich a small business functions within theorganizational framework of a cooperative.Managers are given sufficient freedom andincentives to operate the business efficiently at itsfull market potential. The profits accrue to the coop, which in turn transfers social benefits to itsmembers.

A further possibility is to create differenttypes of organizations for successive stages inprocessing and marketing. For example, one ormore co-ops might produce an intermediateproduct for further processing by a second-orderfederation of cooperatives. A small distributionenterprise could then market the final product ina specific urban area.

It may seem premature to discuss theorganization of second-order federations and soforth at the pilot stage. Yet, much experience hasshown (see Box 10) that mistakes made at thispoint are difficult to correct later on.

For that reason it is risky to initiate a pilotproject with a group or enterprise that has beenformed only recently. Since its members havelittle experience in working together, they are ill

75

prepared to handle the complex task ofintroducing new processing technology andmaking it a commercial success. The pilot projectcannot achieve this goal, even with a potentiallyprofitable technology, unless it builds on a strongfarmer organization or small enterprise andprovides it with appropriate technical,organizational, and financial assistance.

The scale ofpilot operations

Choosing the size of the pilot plant is not easy. Onthe one hand, it must be large enough to providereliable data on the efficiency of the process andits labor requirements, costs, etc. The plant mustalso make enough product to allow a meaningfulassessment of its marketability, based onsignificant levels of sales.

But since there is a considerable risk offailure, you need to keep the amount of capitalinvested in the pilot plant to a minimum. For thesame reason and because the pilot plant isunlikely to make a sufficient profit to permitrepayment of a loan until it expands to acommercial scale, this investment should consistof a donation or soft loan. If weighed down withheavy loan repayments from the start, anotherwise successful pilot project may be doomedto failure.

Data derived from the pilot plant provide thebasis for a financial feasibility study. Its resultsindicate the plant size that is most attractivefinancially. With that information you can seekcommercial credit for plant construction (asdescribed in Unit 6). One exception to this generalpattern may be an existing enterprise that wishesto take on a new process or product. Since theplant already has facilities that can be assigned,at least temporarily, to pilot testing, the newproduct should require little additionalinvestment, which can be financed withcommercial credit.

There is no point in establishing more thanone plant in the pilot phase. To do so will merelyincrease the financial risk without increasing thechances of success. It is better to concentrate onsolving the inevitable problems of one pilot plantrather than dividing your efforts among many.


Box 10

Organizing Farmers in Colombia (Case 1)

Experience in Colombia has shown thatconflicts can arise when cooperatives andprivate enterprises engage in the sameprocess in the same area. In this case bothgroups belonged to a second-ordermarketing organization, which wasresponsible for recommending raw materialprices. The co-ops wanted a high price tobenefit the largest possible number offarmers (members and nonmembers),whereas the private sector plants wanted alow price to maximize their profits. As aresult of this and other conflicts, thesecond-order organization split in two andlost its power to negotiate with the animalfeed companies that purchase the endproduct.

The first farmer co-op for drying cassavawas formed in 1981. Even though theproject was limited to forming andfinancing co-op plants, many additionalorganization models evolved spontaneouslyin the region over the next decade:

• Small-scale co-op or association withabout 20 members, consisting of smallscale farmers or landless laborers

"---------

Site selection

The location of the pilot plant should be based onvarious criteria, as indicated in Checklist 12 andillustrated in Box 11. A major requirement is thatan adequate supply of raw material be availableat a reasonable distance from the plant(transportation of bulky fresh roots iscostly). Good roads are also important fortransporting the raw material as well as shippingthe final product to its principal market, whichideally should also lie within a reasonabledistance. If the plant operates only during the dryseason, road conditions in the rainy season areobviously not important.

76

• Large co-op with 100-400 members

• Association with 2-4 members, mainlylarge-scale farmers with their owncapital for plant construction

• Privately owned plant that purchasescassava from local farmers

• Private business that rents drying floorspace to farmers

• Private business that carries a mobilechipper to farmers and dries cassava onplastic sheets

• Small, on-farm drying plant located on acattle ranch

• Starch producers, who switch to dryingcassava when the price is low

• Individual farmers, who chip cassavamanually and dry it on any surfacewhen the price of fresh cassava is low orwhere access to the fresh market is poor

If at all possible, locate the pilot plant near agroup of farmers who have some experienceworking together. This will reduce the risk oforganizational problems, which could make itimpossible to give new technology a valid test. Thfurther ensure that the group functions well, it isessential that the pilot plant have ready access toinstitutional support in both technical andorganizational matters.

Obviously, you will not be able to find onesite that offers all these advantages. Nor is itpossible, or even necessary, to assign all theselection criteria equal weight. For example,having electricity at the site may be absolutely


Checklist 12


/'-----

r

,

•

Raw material

• Availability (harvest months per year)

• Current and future surpluses

• Existence of competing markets (e.g.,for fresh roots)

• Quality factors (e.g., dry matter andstarch contents)

• Concentration and distance from plant(which affects transportation costs)

• Price

Processing

• Infrastructure (e.g., water, electricity,and roads)

• Local capacity to build the processingplant

• Local capacity to build and maintainequipment

• Suitability of climate (e.g., for naturaldrying)

Farmer organization

• Presence of farmers organizations orsmall rural enterprises

• Farmers' level of education

• Interest in the project

• Resource availability (e.g., labor, capital,

and land)

• Organizational capacity (e.g.,leadership, level and quality ofparticipation in decision making, and

management of conflict)

Institutional support

• Institutions present and their relative

strength

• Interest in the project

• Delineation of responsibilities

• Availability of funds

• Technical knowledge of technicians and

field staff

• Availability of credit

• Existence of mechanisms forinterinstitutional coordination

Markets

• Prices in different markets

• Cost and accessibility of transport

• Ease of commercial contact

• Stability (constant versus seasonal)

• Size and future growth potential

Environmental factors

• Prospects in the area around the site for

increased crop production

• Waste disposal or treatment facilities

near the site

essential but a paved road less so. The weight

assigned to each criterion will depend on the

product and process.

Designing the pilot project andconstructing the plant

The next step is to draw up a plan for adapting

the design and other aspects of the processing

plant to the site where it will be constructed. For

this purpose refer to the points in Checklist 13.

'7~, '

Building a pilot plant is much the same as

any type of construction for agroindustry. It starts

with preliminary land preparation and ends with

the testing of equipment. Other important steps

are to obtain water and electricity, build access

roads. and manufacture and install equipment.

Once the equipment has been installed,

make several trial runs to improve its

functioning, efficiency, etc. Some minor

modifications may be needed.

Add,ng Wllue to Root and Thber Crops

Box 11

Selecting the Site for a Pilot Plant in Colombia (Case 3)

Production of cassava flour involvesrelatively complex technology. For thatreason planners of a project in the Atlanticcoast region of Colombia decided tointegrate the pilot plant into the work of aco-op already producing dried cassava foranimal feed. In preliminary screening, fourco-ops were identified for furtherconsideration. Each was rated according tothe criteria considered most important. Theresults are given in Table 20.

Since the project planned to dry cassavaartificially, it needed a year-round supply ofroots. Access·to electricity and water wasessential. Stro.llg institutional support wasalso consider.ed.vital to the project's success.Visits to all four co-ope confirmed thatChinu was theheet option, especiallybecause of its continuous supply of rawmaterial.

Table 20. Rating sites according to their suitability for cassava processing.·

~------------~~-_._~-------

Chinti Betulia Palmar

Availability ofland for cassava production 3 3 3Potential for yield improvement 3 3 2Availability offresh roots for processing 3 3 IContinuity ofsupply during the year 3 1 2Infrastructure (electricity. water, and roads) 3 3 3Distance to major markets 3 3 .3Institutions in the region 3 :3 3Socioeconomic importance ofcassava 3 3 3Current institutional support to co-op 3 3 2

'!bta] 27 25 21

a. MaJdmum score of 3 per criterion.~

21

322223232

Pivijay

-~

- I

1

I

I

I

J

Cooperative

----------------Selection criteria

At the same time, you can start selectingpersonnel to work in the pilot plant. If it will berun by a co-op or association, ensure that themembers identify operators with sufficient skill tohandle the processing equipment and keep theirnumber to a minimum so as not to inflateoperating costs. It is not a good idea to include anadministrator or secretary at this point. Thevolume of production simply does not justify theexpense. Once the personnel have been selected,you need to draw up a training plan.

forth under experimental conditions. Once theplant is working efficiently and turning out a finalproduct of consistent quality, it can be run on asemicommercial scale. In this stage the productshould be sold regularly in the target market toobtain information about its acceptability toconsumers or clients. In the following sections, wediscuss important issues that you need to dealwith in the experimental and semicommercialstages of pilot processing.

Refining Plant Operations

Your central task in the pilot phase is to get theprocessing plant going. The first step is to tryoutthe equipment, raw material, operators, and so

Raw material

In Unit 4 we described the challenge of securing a.,-;table supply of raw material and raised theIssues of quality and price. During the pilotphase, you need to move beyond assumptions

71",


Checklist 13


• Adjust the capacity of the plant to theproduction of roots and tubers and themarket for them in the area around theplant as well as to the funds available.

• Prepare the building site (this includesstudies of the topography and soils andmay involve land levelling), and obtainpermits for electricity, water, or both.

• Design the pilot plant, infrastructure,and equipment according to the capacityand characteristics of the site. Wherethe construction. requires the services ofarchitects or civil engineers, initiate abidding process.

about these matters and decide how exactly theywill be dealt with.

Supply. Th speed capital flows and avoid lossof quality, roots and tubers should be kept in theplant for as short a time as possible beforeprocessing. This is especially important forcassava, which can be stored safely for only1-2 days. If the plant requires a daily supply offresh roots, you should try to organize dailydeliveries. This may result in some unused rawmaterial, especially at the beginning of the pilotphase, when the volume of raw material neededtends to be small. Although other root crops areless perishable than cassava, it is still importantto minimize their storage times before processing.

Since fresh roots contain 65% or more water(which is eliminated in many processes), they arealways expensive to transport. That is why theprocessing plant should be built in an area wheresufficient raw material is produced within a shortradius. If only limited amounts of raw materialare available nearby (perhaps because the plantwas built on the basis of future productionpotentiaD, you will have to cover the cost oftransporting raw material over longer distancesin the early stages of processing.

79

• Develop a plan for training plantoperators, with emphasis on conceptsand procedures in quality control,hygiene (especially for foodstuffs), andbookkeeping.

• Plan for experimental andsemicommercial operation of the pilotplant.

• Design a marketing plan that identifiesthe most attractive markets in terms ofmargins, proximity to plant, etc.

• Analyze the institutional supportavailable, and take measures to ensureadequate support in all areas.

There are several ways to organize rawmaterial supplies. You may want to integrate cropproduction with processing, at least to someextent. This can help make up for shortfalls insupplies from other sources. If the processing is acooperative venture, the entire co-op or individualmembers can also produce the raw material.Another option is to have intermediaries identifyfarmers with harvestable roots and tubers. Theintermediaries may also transport raw material tothe processing plant, although this could addconsiderably to the cost of their services. Anotherarrangement is to contract farmers to produceroots and tubers. As illustrated in Box 12, it mightbe worthwhile to provide them with inputs (suchas planting material and fertilizer) to ensure goodyields and quality.

All these options entail risk. Small-scalefarmers in particular are liable to break theiragreement with the processing plant if offered ahigher price for their production in anothermarket.

Harvest periods. Th organize raw materialsupplies, you need to determine local harvesttimes. These can vary greatly from one area toanother, depending on the months in which


Even so, for lack of raw material, thefactory was able to work at only 25%capacity. Problems arose from:

• The perishability of fresh cassava

• High and fluctuating raw material prices

'-------------------------- -------

rainfall is adequate for planting and initial cropdevelopment. In some places two harvests a yearare possible, while in others root crops can beharvested continuously, either because rainfall isevenly distributed or irrigation is available.Under some circumstances (e.g., where theproduct must be dried naturally), the processingcan be done for an even shorter time than theharvest period.

Price. The price of fresh roots may varygreatly during a normal year, being lowest atharvest. A preliminary study of patterns in theprice will give you an idea of how much the plantcan expect to pay for raw material. Unless thisprice is comparable to that of the freshmarket, you may have difficulty obtaining rawmaterial, except at harvest time, when suppliespeak.

Competition for raw material suppliesbetween fresh and processing markets has causedseveral large-scale cassava processing plants tofail in Latin America. One way of avoiding thisproblem is by locating the plant in an area withlimited access to markets. In general, a processingplant whose viability depends on low-cost rawmaterial should be located away from majorconsumption centers, where the fresh market isimportant.

• Small-scale farmers' lack of control overplanting harvest time as well asland and r shortages

• High transport costs

• Pest problems

The solution recommended was toinitiate an outgrowers scheme, in which anassociation of small-scale farmers wouldproduce cassava, while the company wouldprovide inputs and carry out the harvest.

___ JQuality. In general, processing plants

are less demanding with respect to rawmaterial quality than the fresh market,where intermediaries and consumersnormally express strong quality preferences.In fact, it may be possible to supply a pilotplant with roots and tubers that are notacceptable for the fresh market. Butindiscriminate use of noncommercial or poorquality raw material is not an option forprocesses and products that require high drymatter or starch content.

Determine as early as possible in thepilot phase whether the raw materialrequired for processing is different from thatdestined for other uses. If the differences aresignificant, it may be necessary to include aselection stage in the process, during whichsuitable roots and tubers can be identified. Ifa large percentage have to be rejected, youwill need to determine an alternate use forthem. Box 13 gives an example.

Early in the pilot phase. It should beobvious whether the raw material factorsdiscussed here (quality, prIce, etc.) arefavorable for securing a continuous supply ofraw material of adequate quality.


Box 13

DealiBIJwith &awMaterial Quality in Colombia (Case 3)

make dried cassava chips for animal feed, aproduct for which roots of lower quality areacceptable. Later, the cooperative shifted itsselection process to the. farm level. Thismeasure saved labor at the processing plantand reduced transport cosU!.

Production of cassava flour for humanconsumption requires a high-quality endproduct. 'lb meet quality specifications, ap Colombia had to reject a

ntage of roots during theselection stage. The farmer cooperative thatoperated the plant used the rejected roots to

I

l---------

Processing

During the experimental stage of pilot plantoperations, you need to adjust the equipment tomaximize efficiency. In some cases this may takeonly a few days. In others it may take more timeand money to modify defects in construction(above all, in prototype machinery). Moreover, youmay have to test several options for a process todetermine which is most efficient, easiest tomanage, and produces the best quality product.

One of the most important factorsdetermining the profitability of many processes isthe conversion factor (or tons of raw materialrequired to produce a ton of final product). Howgood, or low, the conversion factor is depends onthe quality of the raw material (e.g., dry mattercontent and percent peel) as well as the efficiencyof the process. For each step in the process,establish realistic specifications for all importantvariables (such as use of labor and fuel, processtimes, and product characteristics).

When the supply of raw material permitscontinuous processing without furthermodifications or adjustments, the plant is ready tooperate on a semicommercial scale. At this stageyou need to evaluate the following aspects of plantoperations:

• Performance of equipment under continuoususe, including energy consumption andefficiency

• Need for further research on equipment orits handling

• Bottlenecks in the process

• Actual operating costs

• Raw material supply, price, and qualitythroughout the period processing is expectedto take place

By the end of this stage, you will havecompleted all necessary adjustments in theprocess, including management and labor use.Often, the plant workers themselves can findpractical solutions to problems and offer valuablesuggestions. Try to establish mechanisms forincorporating their contributions into theevaluation of pilot plant operations.

The product

In the experimental phase of plant operations, itis not enough just to improve the efficiency of themachinery. The process is not ready for furthertesting until the quality and quantity of the finalproduct are right as well.

Product quality-a result of the interactionbetween the raw material and the processconSIsts of three main groups of traits:

•

•

•

Ease with which operators handle theprocess

Labor efficiency

Effectiveness of operator training• Chemical, physical, and microbial

composition of the product

81

Adding \.blue to Root and Thber Crops

• extremely important (see the discussion ofconsumer testing of these traits on page 67).

• Establish specifications for raw materials,other supplies, processing operations, andthe final product, together with itspackaging.

• Develop procedures to measure each qualityfactor (often, official methods are suitable).

Though normally associated with the foodindustry, quality control is just as applicable toother industrial uses of crops. Th maximizeproduction without sacrificing the quality of thefinal product, plant management rather than theproduction section should be directly responsiblefor quality control. This consists of the followingtasks:

Develop sampling procedures that givereliable results at minimum cost.

Design recording and reporting forms for useby production operatives.

Train production operatives in the use ofquality control tests.

•

•

•

Traits perceived by consumers (organolepticfactors, etc.)

• Features, such as shelf life, that are relatedto uses of the product

The most important chemical parameters forprimary processed products (flours, starches, etc.)are usually dry matter and starch content,followed by other nutrients, such as protein andvitamins. All food and feed products should befree of mycotoxins. For many foods microbialcounts are another important measure of hygiene.In some root crops, it is also important to monitortoxic or antinutritional components (e.g., cyanide,alkaloids, and trypsin inhibitors) throughoutprocessing and in the final product. In cassavaproducts, for example, maximum acceptable levelsof cyanogens have been determined.

Food products must comply with any legallimits on microbial counts, etc. It may also benecessary to obtain a sanitary license from theMinistry of Health. Some clients have evenstricter quality standards than those of thenational standards institute or those required bylaw.

Identify a laboratory that can analyze criticalquality factors routinely. This will ensure that youhave selected the correct process during theexperimental stage of plant operations and allowyou to monitor product quality during thesemicommercial stage. The project should coverthe costs of these analyses as long as the plant isoperating on an experimental basis.

If the product does not meet qualitystandards consistently in the experimental stage,you need to find out why. The problem must lieeither in the raw material or in the conditionsunder which processing takes place, which mayvary from those of the research station orresearch phase of the project. It is especiallyimportant that rigorous standards be maintainedin personal hygiene and in the cleanliness of theprocessing equipment. In contrast, manytraditional root crop products are made in thecomplete absence of sanitary controls and, as aresult, have high levels of contamination.

In products destined for humanconsumption, organoleptic characteristics are

Quality control should take place in threestages:

• Raw material control

• Process control

• Inspection of the final product

Theoretically, if you do a thorough job in thefirst two stages, the third should besuperfluous. In practice, though, it is still a goodidea to inspect the final product, especially sincemany root and tuber processing plants operateunder less than ideal environmental conditions. Ifthe raw material and process are carefullycontrolled, the proportion of final product rejectedshould be relatively small.

Shelflife and packaging

No matter how short a product's journey down themarket chain, its quality must hold up for somelength of time. It also has to resist significantchanges during storage. The following factors canaffect the product's useful life:

82


•

•

•

•

•

Moisture content

Relative humidity and temperature duringstorage

Contamination by fungi, bacteria, andinsects during processing

Type of packaging

Size of packaging (product weight per unit)

dealing with these markets is that the monthlyunit of purchase can be very large, equaling orexceeding the plant's production capacity.

In consumer markets a primary or secondaryproduct is sold through a wholesale distributionchannel. This is quite complex. Th be successfulthe product must be purchased by thousands ormillions of consumers, who are daily bombardedwith publicity about competing products.

Packaging serves as a barrier between theproduct and its environment. Even so, beingporous, it permits the interchange of moistureand gases (C02and 02) and can be penetrated byinsects and rodents, which in turn facilitatecontamination by microorganisms. For thatreason the packaging must be, not onlyreasonably priced and easy to obtain, but alsoadequate to protect the product after purchaseuntil it is used or consumed.

Fungal growth and mycotoxins are not aproblem in products that are dried to moisturelevels suitable for long-term storage. For someprimary processed products, packaging is not amajor consideration, since the industrial e;:;··r willmill or otherwise process the product furthershortly after purchasing it.

In contrast, processed products for humanconsumption, such as flours, noodles, and cookiesrequire packaging that resists insect attack andchanges in moisture content, particularly if theturnaround time is slow. Some products require aspecific type of packaging to preserve theirdistinctive characteristics (e.g., fresh cassava inpolyethylene bags).

Finding a Niche in the Market

There are two types of markets for the pilotplant's products: industry and individualconsumers.

In industrial markets enterprises buy aprimary product, which they transform orincorporate into another product. This market isrelatively simple. Since it generally consists of fewclients, you can establish direct contact with theirheads of purchasing, whose decisions are based onlogic, price comparisons, availability andsuppliers' performance. A further advantage of

Whether consumers buy a given productdepends on many illogical factors, such as itsimage and status. Th reach large numbers ofconsumers requires an efficient distributionsystem at terminal markets as well as resourcesand capacity for mass promotion.

Small-scale farmers organizations canseldom meet these requirements on their own. So,you may have to contract or create otherorganizations to handle product distribution andpromotion. This, of course, involves additionalcosts.

Target markets

Having selected a site for the pilot plant, you needto identify target markets and then carry out amarketing study. Here are the main groups ofoptions:

• Local: rural areas around the pilot plant

• Regional: the nearest large urban center

• National: the capital and other major cities

• National: all locations of the target industrialmarket

• Export: neighboring countnes, the USA,Europe, Japan, etc.

You should normally tackle local and regionalmarkets first and only later consider exportmarkets. Unless the product can compete locally,it is unlikely to succeed internationally.

The relevance of national markets depends,not so much on their size, as on the type ofproduct. If the plant is producing for industrialmarkets, it should adopt a national strategy fromthe Htart, since clients will probably be located farfrom the zone where root and tuber crops are

produced. Usually, there are only a few suchplants, so transportation ought not pose too greata problem. On the other hand, if the pilot plantcaters to consumers, it should avoid breaking intonational markets until it has first passed the testof a well-defined local or regional market.

Adding Wllue 10 Root and TUber Crops

product of acceptable quality. Until then theproject covered the resulting losses.

After making the first sales to an industrialmarket, inquire in detail about the following:

• Satisfaction with the product

The tlesign of the market study willnaturally vary according to the target market. Iffocused on consumer markets, the study shouldcover intermediaries, final outlets, andconsumers. Studies of industrial markets includeonly potential clients, as illustrated in Box 14.

Easing into the market

•

•

•

•

•

{)se of the product

Evaluation of its quality

Comparison with other raw materials

Attractiveness of the price

Estimate of demand and potential forincreased purchases

Based on the results of the marketing study, youcan draw up a plan for selling the product. Thisshould be implemented after the experimentalstage of pilot plant operations. By then you willhave fine-tuned the process and should be turningout an acceptable product. It is crucial that thefirst samples clients purchase be of optimalquality.

Production that does not meet qualityrequirements can be disposed of in other marketswith less exacting standards. For example, thecassava flour project described in Box 13 initiallysold to the animal feed market until it couldconsistently provide the food industry with a

• Consumers' reaction to the final product

• Desired unit of purchase and frequency ofdelivery

For sales to consumer markets, it ispreferable to begin in a limited geographical area,distributing to several nearby shops or to a chainof supermarkets. This will enable you to gather aconsiderable amount of information with aminimum of resources.

Once the pilot plant is operating on asemicommercial basis, product quality andoperating costs should be stable. That is the timeto proceed with your plan for developing a

Box 14

UaSStllVa Flour in Colombia (Case 3)

a special marketingu<:tr a wellThere

ongackers

andr with the

mam <: aract ssava flour andwith its differences from wheat flour.

A study of the prospects for overcoming~~heseobstacle~was conducted in two

stages. The first was a preliminary surveyon the use of flours in generaL In thesecond, industries were given samples withwhich to conduct partial substitution trialsin various products. Then, a survey wasmade of the trial results and interest inpurchasing the neW product. The studyyielded concrete information on thepotential use of cassava flour in a widerange of products. And it was a good dealless trouble than extensive laboratory trials.


market. Explore several options for presentingthe product (as described on page 68). To selectthe best, it is helpful to test the reaction of themarket to different alternatives.

Product pricing and terms ofpayment

There are two ways to set the price of a newproduct: by calculating production costs plus aprofit margin or by determining the price themarket will bear, based on the prices of competingproducts.

A new product should be better, more readilyavailable, or cheaper than those already in themarket. Be cautious about promoting a productwhose sole advantage is lower price, since this isoften equated with low quality. If the product is ofhigher quality than others, it should cost at leastas much as inferior competitors.

Although it is important to cover costs, theyare not the only basis for determining a product'sfinal price. When introducing a product in themarket, for example, it may be helpful to offerdiscount prices for a limited period to encouragefirst-time sales.

The terms of trade offered to purchasers ofthe product should be at least as favorable asthose of competing products. But they also dependon the financial situation of the enterprise. Forexample, if cash flow is a problem, you mightconsider offering discounts for cash purchases,bulk orders, or regular contracted deliveries.Payment after 30 days is the norm in someindustries or where state marketing organizationsare involved. If you allow sales on credit, expect asmall percentage of nonpayers. If such sales arecommon, the associated costs should be built intothe enterprise's financial projections.

activity in product development must be to trainworkers, not just to operate the processingequipment, but also to maintain high productquality-a task involving hygiene, raw materialquality, etc.

During the pilot phase, the project is thelogical candidate to offer training on technicalaspects of processing, based on its experience inthe research phase. It is unrealistic to expect theextension service, for example, to mount a formaltraining program on a product and process thatmayor may not succeed. If extension or otherservices do help with training at this point, theproject must be prepared to finance theirparticipation. If the pilot project is successful, it isthen reasonable to expect that extension andother training agencies will incorporate the newproducts and processes into their portfolio oftechnologies for the region.

The processing operation must be managedlike any business but without neglecting socialaims that led to its formation. To do this requiresmany skills farmers normally do not possess,especially if the product is aimed at urbanmarkets. Thus, even in the pilot phase, you mayneed to provide a relatively small number ofpeople with intensive training in business,marketing, and accounting.

If project staff are not qualified to conducttraining on these subjects, you need to find anagency that is. In addition to training, the agencycan perhaps give ongoing support in thecommercial aspects of the processing operation.Some likely partners for training are:

• University economics or businessdepartments

• NGOs supporting small businesses (often inurban areas)

Training Processors • Private sector businesses

In most cases the pilot plant will be located in arural or periurban area. The people whomanufacture the product will either be farmers,landless laborers, or traditional root cropprocessors who want to expand their productrange or improve existing processes or products.Most will have little formal education or previoustraining and experience. For that reason a major

85

It may seem surprising that the privatesector would assist possible competitors. Yet, inAsia and Latin America, large companies activelysupport the establishment of small businesses,particularly those having commercial interestsoutside, but related to, their own. Business skillsare, after all, independent of the productmanufactured and sold.

Here are some useful approaches to training:

• Practical training at the prototype plantconstructed in the research phase of theproject

• Formal instruction on basic principles of foodmanufacturing (e.g., hygiene, producthandling, etc.)

• In-service training provided by the privatesector on topics such as business skills,marketing, and quality control

• Formal courses in accounting and so forth

• Visits to other projects

Given that a wide variety of training must beprovided for a fairly small number of people, it isprobably necessary to involve various institutions.Since not all will be project participants, you willhave to contract some of them for specificfunctions.

By the end of the pilot phase, the project willhave invested heavily in training for farmers andothers involved in plant operations andmanagement. For that reason you need to selectboth the site for the pilot plant and theparticipants very carefully. If you have peoplewho are strongly committed to the project, theycan help train other farmers when the operationstarts to expand.


Test Markets for ConsumerProducts

To guarantee the success of a product in consumermarkets, you must make a considerable effort(much more than is required for a product used byindustry) to organize its distribution andpromotion. Since these tasks are quitecomplicated and costly, it is a good idea tointroduce the product in a test market beforelaunching a full-scale project to commercialize it.

A test market is typical of the total potentialmarket but smaller. It enables you to test thedistribution system and promotional activities atlow cost within a reduced area. From the resultsit should be clear whether the product cansucceed on a wider scale.

To test market a product, the pilot projectmust satisfy the requirements indicated inChecklist 14.

To identify a suitable urban test market,consider the following options:

• Shops (small, local concerns or larger oneseither in low- or high-income areas)

• Institutions (schools, hospitals, army bases,etc.)

• Supermarkets

Checklist 14


• A continuous supply of the product orenough inventory to meet expectedmarket demand

• A product of the specified quality

• An attractive promotional price

• Promotional materials, together with anadvertising campaign based on them

• A distribution system that is adequate,not only for the initial volume, but forexpanded sales within a short time

Kf;

• A system for obtaining feedback to judgethe product's chances of success(including weekly data on volumesdelivered and sold, by shop, and onpurchases by a sample of consumers,information that indicates the rate ofrepeated purchases, compared withinitial purchases)


• Shopkeepers' co-ops • Radio and television commercials

• Restaurants

• Stalls in wholesale or retail markets

It is easier to deal with a market consistingof only a few high-volume clients than one withnumerous clients purchasing small amounts. Butsince the latter may be the largest markets, it isworthwhile to examine these options for wideningproduct distribution:

An advertising campaign that uses all thosemedia is beyond the budget of most projects. Evenso, you can still achieve significant impact withlimited resources by:

• Clearly identifying the target market, suchas low-income housewives from the poorestresidential districts, and using only themedia that reach them (e.g., a commercial onthe favorite radio station of these consumers)

Promoting the product on the news andother programs of public interest, above all toconsumers

•

Contract a private distributor on anonexclusive basis.

Set up an enterprise that specializes indistributing products in urban centers.

•

•

• Distribute the product to a centralwarehouse operated by a shopkeepersassociation (possibly one organized expresslyto facilitate distribution).

• Organizing special campaigns forshopkeepers (e.g., using leaflets that explainthe benefits of the product), aimed at gettingthem to promote the product among theirclients

• Obtain warehouse space in a wholesalemarket where shopkeepers purchase otherfood products.

Each of these systems has advantages anddisadvantages. Select the one that keepsdistribution costs to a minimum and thus allowsyou to sell the product at a competitive price.

Th design and execute a promotionalcampaign-even with simple, low-cost mediayou need the help of individuals or firms thathave experience in advertising and know how togain maximum publicity with scarce resources.

The key to promoting a product in consumermarkets is to have an attractive, legallyregistered brand name that reflects the product'sadvantages, together with a logo and slogan.Advertising media include:

As shown in Figure 6, the product is asuccess if numerous buyers make initialpurchases and a large percentage buy it again. Ifinitial purchases are high but only a few peoplebuy a second time, the product is based on a goodconcept but does not live up to its promises. Inother words, consumer are disillusioned withit. When initial purchases are low but repeat salesare high, you have a good product but poordistribution or promotion. If both initial andrepeat sales are low, the product is a failure.

Test market trials are difficult but quiteeffective for detecting problems early, beforeanyone has made a major investment. Forproducts aimed at the consumer market, thesetrials are an essential part of the pilot phase inproduct development.

•

•

•

••

The package (with logo, slogan, recipes,instructions for use, packing or expirationdates, name, and sanitary license)

Promotional materials (such as posters andleaflets) at the point of sale

Supermarket promoters and salesrepresentatives

Newspaper advertisements

Billboards

The Feasibility Study

The semicommercial stage ends when you havecollected enough information to complete afeasibility study-the main output of the pilotphase. The key items of information are the sameas those needed for a prefeasibility study (seeChecklist 11, Unit 4). The results indicatewhether the project should proceed to thecommercial expansion phase or be abandoned.

87

Ad(hng Value to Root and Thber Crops

Repeat purchases

Low

Initialpurchases

High

Low

Successfulproduct

Distributionor promotionproblem

Good productconcept butpoor productquality

Failedproduct

Figure 6. Illustration of Kotler's hypotheses regal'ding product success (Koller, 1986).

Normally, it takes a full year to gather

information on raw material supply and product

demand across seasons. But you may need more

time if problems such as poor rainfall or pest

attacks are so severe as to affect raw material

supply, price, or quality. If the experimental stage

in plant operations takes longer than expected,

avoid shortening the semicommercial stage

correspondingly, since this may leave too little

time to collect information.

If the pilot phase has gone well, the supply,

price, and quality of the raw material will be

adequate; the process will be functioning

efficiently; operators will be trained and working

efficiently; the product will meet specifications

and be of uniform quality; and the market for the

product will be expanding.

The feasibility study essentially documents

all this experience. Based on the outcomes of

semicommercial plant operations and test

marketing (if this was necessary), the study

examines the financial feasibility of the new

product, given the investment required to produce

and market it. For help in deciding whether the

product is an attractive investment, you can apply

financial models, which, in addition to

determining profitability, indicate how it can be

increased through improvements in processing

and other aspects of the enterprise.

Models for estimating financial rates

ofreturn

Using a basic microcomputer spreadsheet

program (e.g., Lotus 1-2-:3), you can construct a

model that describes the process and other

operations. You can easily adjust it to changes in

costs and process parameters and carry out

sensitivity analyses quickly to determine how

rates of return are affected by changing costs,

lmprovements in processmg efficiency, and so

forth

There are numerous measures of financial

feasibility. One is the financial rate of return

(FRR), which is defined as "that rate which

discounts annual cash flow to the project start

(time 0) in such a way that its present value is

equal to the initial investment" (Gittinger, 1982).

The FRR takes the point of view of the business

(i.e., the processing unit), not that of society in

Unit 5: TM Pilot Phase

• Economic rate of return

general, which is measured by the economic rateof return (ERR). Differences between these twomeasures are summarized as follows:

•

Takes society's point of view.

Does not consider taxes as costs; viewssubsidies as costs to society.

Does not take into account financial costs.

Does not always use market costs and prices.

Financial rate of return

Takes the business's point of view.

Considers taxes as costs and subsidies asincome to the business.

•

model does not take this factor into account,then the minimum acceptable FRR value isthat of the opportunity cost of capital minusthe inflation rate. If inflation is taken intoaccount, you need to decide whether or not touse a constant rate, and the same rate forcosts and income, throughout the life of theproject.

Salvage value of the capital investment.At the end of the project, the plant willpresumably be sold and all working capitalreturned. The model must assume a valuefor the plant, as some percentage of theinvestment cost. If the model includesinflation, it must take this into account aswell.

May take into account financial costs.

Always uses market costs and prices.

The model has various components, eachdealing with a key aspect of the finances of theprocessing operation:

Before developing the model for a givenprocess, you need to make basic decisions about:

• Investment

•

•

•

•

Project life. The longer the life of theproject, the higher the FRR; the higher theprofitability, the less impact project life hason FRR.

Production capacity. Because of theseasonal availability of raw material and thehigh cost of storing fresh roots, processingplants often operate for fewer than12 months a year. The maximum productioncapacity of a plant will therefore be less thanthe theoretical capacity of its equipmentduring a full year. On the other hand, theplant may operate at more than 100%capacity if it processes during a greater partof each year than originally planned.

Capacity utilization. For a number ofreasons (problems in the supply or quality ofraw material, power failures, breakdown ofequipment, etc.), processing plants do notnormally operate at full capacity. Based onexperience in the pilot stage, you canestimate more or less accurately whatpercentage of its capacity the plant willutilize.

Inflation. In many countncs inflation is aserious problem for small enterprises If the

H9

•

•

Preliminary estimates of costs of equipmentand infrastructure. Initial working capital isalso included. The amount required dependson.

Production costs

Volumes produced

Time taken to distribute, store, and sellthe product

Payment schedules, especially forindustrial clients

Maintenance

The cost of keeping equipment andtnfrastructure in good workmg condition isdivided by the production volume andrepresented as a fixed cost per ton ofproduct

Basic information

Plant capacity

Capaclty utihzatlOll

Processing parameters i e.g., conversionfactor for fresh roots to finished product,labor requirements in worker-hours perlon, vnergy and water use per ton as

•

•

well as the unit prices of each of theseelements in the cost structure)

Variable costs

Those that vary according to the scale ofproduction, such as cost per ton of:

Raw material (including transport costs)LaborPackagingFuel and other energyWaterTransportCommissions and contingencies

Fixed costs

Those that do not vary with the volume ofproduction:

•

•

•

•

•

•

•

•

Addmg Value to Root and Thber Crops

Sensitivity analysis to determine whichcomponent most affects the FRR

Product cost and price structure

The model can help you:

Optimize the process.

Set maximum acceptable prices forpurchased inputs.

Assess financial feasibility.

Select the best option for financing thebusiness.

Provide solid information for donors, banks,project advisors, and the owners andoperators of the enterprise.

Determine the sales pnce.

•

•

•

•

•

•

Investment in infrastructure andequipment, reflected in finance costsAdministrative costsPlant maintenance

(Be sure to include all costs so as not tooverestimate financial feasibility.)

Sale price ofproducts

The model uses a weighted average of thecosts of the principal product and any byproducts. 'Ib calculate this you need to knowthe:

Sale price of each product and byproductPercentage of each producedProcessing losses (e.g., in milling andtransport)

The model's outputs are the:

Financial rate of return

Gross margin (sales price less variable costs)

Net margin (sales price less variable andfixed costs)

Cash flows (income from sales minusvariable and fixed costs plus increases inworking capital due to inflation)

Net present value

90

• Identify improvements in the plant andprocess that could increase FRR.

Box 15 shows how a cassava flour project inColombia applied the financial model. Evenwithout a computer-assisted model, you can learna lot simply by calculating all costs and returnscarefully and by looking at net returns per unit ofoutput, as illustrated in Box 16.

Timing the Transition

In this unit we have described how to move aproduct from small-scale, experimentalproduction through the pilot stage and markettesting to a feasibility study. If the outcome isfavorable, you can proceed to full-scalecommercial production, with the aims of meetinga significant proportion of demand and making anattractive return on investment. To reach thesegoals, you can expand the existing plant,construct more units, or both

How do you determine when the pilot stageends and the commercial phase begins? This is avery important decision. If the operation expandstoo soon, before enough information is available,it may fail because of unforeseen problems. Onthe other hand, if the pilot phase is prolongedunnecessarily, you run the risk of losingcommercial opportunities or prolonging the life ofa bad product and multiplying financial losses.

For help in timing the transition correctly,consult Checklist 11 (Unit 4), which indicates theinformation required to assess feasibility. As soonas you can replace all important assumptionswith solid data obtained under commercialconditions, it is time to assess feasibility.

Even if the results are encouraging, youcannot launch the commercial phaseimmediately. It takes time to d~velop plans,obtain financing for expansion, and so forth. Inthe meantime the pilot plant should continueoperating to keep the market alreadyobtained. The plant may still not be operating ona large enough scale to earn a profit. In that caseyou may need to obtain a bridging loan or makea small investment to boost capacity to acommercial scale.

If the outcome of the feasibility study isnegative, you have several options:

• Redesign the pilot plant. By replacingequipment or redesigning theinfrastructure, you may be able to overcomeproblems identified in the pilot phase. Thismeans backtracking, repeating certain pilotactivities, and conducting a new feasibilitystudy. If funds are not available to cover thecosts or provide extra working capital, thisoption may not be viable.

• Move the pilot plant to anotherlocation. This may help if there areproblems with raw material supply orquality or with the organization operatingthe plant. This option may not be feasible,though, if you have made a significantinvestment in one site, which cannot betransferred easily to another.

Addmg Value to Root and Tuber Crops

• Do more research. If a problem arose in thepilot phase that was not investigated in thepreceding stage, you may need to conductmore research. This could lead tomodifications in the process, improvements inproduct quality, etc. In this case the pilotstage has fulfilled its function, but it is stilltoo early to determine the feasibility of theenterprise.

• Abort the project. The pilot stage gives youthe option of discarding an unsuitable productbefore making a significant investment in itsmanufacture. Since the project-not theprocessors-should absorb most, if not all, therisk involved, aborting the project shouldhave few negative consequences. Keep inmind that in the food industry only a verysmall percentage of product ideas ever reachthe market and succeed. You should nothesitate to kill a product at this stage if youare convinced that it has no commercialfuture. That is the best way to ensure that allprojects reaching the commercial phase standan excellent chance of success.

By the end of the pilot phase, after you havecompleted the feasibility study, the status of theproject should be as indicated in Table 21.

References

Gittinger, J.P. 1982. Economic analysis of agriculturalprojects. 2nd ed. The Johns Hopkins UniversityPress, Baltimore, MD, USA.

Kotler, P. 1986. Principles of marketing. 3rd ed. PrenticeHall, Englewood Cliffs, N,J, USA.

Table 21. Status of a product development project at the end of phase 3 (pilot phase).

II

f·Why (objective)?Where (region)?What (product)?How (process)?

How much (market)?By whom (type of enterprise)?For whom (beneficiaries)?

Phase I-Finish

DefinedGeneral area identifiedIdea selectedGeneral process identified

Phase 2-Finish

Well definedGeneral area identifiedIdea and concept testedOpt ions evaluated andbest selectedIU"ntified

J)pfined

Phase 3-Finish

Well definedPilot site selectedProduct trial marketedProcess feasibility evaluated

Market potential evaluatedEnterpnse evaluatedDefined

'.'--~-------

a. May be defined in project objective.


Appl

Box 15

a FinaJaeial Model to Cassava Flourin Colombia (Case 3)

After operating for a year on a pilot basis, aproject set up to produce high-qualitycassava flour for human consumption inColombia ncial model to estimatethe financi ility of the enterprise.The fol information was available:

• Amount of investment in plant andequipment

• Data on plant operations (e.g., capacityand costs of labor, raw material, andenergy)

• Variable costs of labor, energy, andtransport per ton of product

• Fixed costs per ton of product• Sale price of final product

It was assumed that the project wouldhave a life of 8 years, that inflation wouldbe 25% per year, and that the plant wouldoperate at 90% of its capacity.

Box 16

Costs and Returns in Simple Potato Processing in India (Cas.e 6)

Dehydrated potato chips and potato flourwere identified as potential. products forhuman consumption. Initial trialsdemonstrated/that •both .proeesses weretechnically/feasible, even if conducted on avery small. scale, producing 200 kglday.

Project planners needed to determinethe minimum scale of operation that wouldcover .the costs of equipment. First, theyinventoried all infrastructure andequipment and then monitored each processcarefully to determine the amounts of labor,material, and fuel used as well as the costsof all inputs and outputs. With thisinformation, they calculated operatingbUdgets for processing on different scales.Some of the key results were as follows:

• Units costs are 25% lower at1,000 kglday than at 200 kg/day.

91

• Variable costs are 80% of total costs.• Although the purchase of equipment

requires large cash outlays, the amountis still only 5% of annual operating costs.

• As the scale of the operation and lengthof the processing season increase, so doprofits.

• Increasing conversion rates by 1%improves profitability by 6% whenproduction is 200 kg/day for 90 days/year.

Based on the results of this analysis, theproject focused on improving conversionrates, rationalizing the use of labor, andlowering transport and marketing costsrather than on reducing drying costs. Plantswere encouraged to produce at least600 kg/day. And efforts were made to prolongthe processing season.

Unit 6

The Commercial Phase

Assuming that the pilot phase demonstrates thefeasibility of a new processing enterprise, you cantake one of two approaches in expanding it to acommercial scale. The first is simply to add extraplants in a more or less ad hoc fashion as demandfor the product increases. The other is to preparea plan of action for the commercial phase beforeexpanding the operation.

In this unit we strongly recommend thesecond course, as indicated in Figure 7. The planof action describes a series of activities aimed atreaching specific commercial targets. Its goal is toensure that supply and demand grow at the samepace. The plan can pay high dividends if based oncareful examination of all aspects of commercialprocessing. Since any plan is fallible, though, itshould be executed with flexibility.

This unit deals mostly with the planningprocess. (For further details on planning for thecommercial phase, see the list of publications atthe end of this unit.) First, we describe options fororganizing farmers and the various institutionsinvolved in the commercial phase. Next, wediscuss various aspects of plant operations,including:

• Raw material supply

commercial phase of product development hastaken place within the framework of anintegrated project. As explained in Unit 2, suchprojects deal with a wide range of activities,including crop production, institutional support,and credit. In this unit we refer to those activitiesbut deal with none in detail. Some are complexenough to merit a manual of their own. Toimprove crop production, for example, requires amajor effort to disseminate better varieties andcrop management practices. Any productdevelopment project should take such activitiesinto account, even if it is not part of a largerintegrated project.

Getting Organized

If small-scale farmers are to benefit fromeconomies of scale, learn to carry out specializedfunctions, and obtain credit for root and tuberprocessing, it is essential that they be organizedinto groups (either cooperatives or smallbusinesses). In deciding how to do this, take intoaccount farmers' traditional forms of organizationas well as the laws of the country. (For furthermformation on farmer organizations, see the listof publications at the end of this unit.)

• Processing Patterns offarmer organization• Product packaging, distribution, and

promotion

• Farmer training

Then comes a discussion on sources offunding and approaches to evaluating progress.

The end product of this phase should be agrowing rural agroindustry that is commerciallyviable and meets project objectives. Once theenterprise is reasonably well developed and hasthe capacity to develop new products,institutional support should be withdrawn.

This manual draws heavily on cassavaprojects in Latin America. In all of them, the

93

In Latm America many integrated cassavaprojects have worked with village-level farmerassociations or cooperatives. Most of these have15-:30 members, who live near one another andmayor may not produce roots for processing.Groups are typically structured so as to investmaximum authority in the general assembly ofmembers. Their central purpose is to generatesocial benefits for processors as well as producers.Toward that end they tend to set the price of theraw material at higher than commercial rates.

In some countries larger cooperatives withmany other activities have also taken up cassavaprocessing. Both this and the village-level

Adding Value to Root and TIther Crops

Pilot Project

G;~~~te project with

government policy 14

makers, industry, etc.-_.__.__.. _-_.".,~ ..._--~,_._ .. _._,- -

Projectformulation

i~ufr~

Plan commercial expansion

-- !

Execute plan LCrop~roduction

I-B~~iness ---11management I

No OK?

Yes.!

I, Start project phaseout.-----_ ..,,--

•Self-supporting rural

~, agroindustry----~-~-_.. ~

Governmentpolicy

BanksCropproduction.

Other farmactivities

Figure 7. Steps in the commercial phase of product development

organizational model are applicable to processingof cassava and other root and tuber crops.

As noted in Unit 2, private sectororganizations distribute benefits quite differentlyfrom cooperatives. Rather than spread benefitswidely, the joint owners of an enterprise try tomaximize profits by seeking raw material at thelowest possible price, among other means. As aconsequence, privately owned enterprises have aneasier time building up capital and reducing theirdependence on credit.

94

One type of private sector enterprise is thefamily or household processing operation. It issomewhat like a co-op, in that members shareresponsibility for decisions and work. But thedistribution of benefits is relatively limited, as inany private business. If such an enterpriseexpands (e.g., by employing labor from outside thefamily), its character may change radically. Forthat reason, some family or household enterprisesmay not be interested in processing on a largerscale.

Unit 6: The Commercial Phase

Regardless of how a farmer association isorganized, it needs to obtain legal status. Often,this is a prerequisite for obtaining inexpensivecredit or technical assistance from the publicsector. In some countries there are legalrestrictions on the sale of produce for profit.These force loosely organized associations to formcooperatives, a step that involves considerablebureaucratic red tape.

promotion and face the difficulty of negotiatingwith high-powered, private sector enterprises. Tostrengthen their position in the commercial world,it is a good idea for these groups to form secondorder organizations of the type described inBoxes 17 and 18.

The functions of a second-order organizationare to:

When local groups or enterprises attempt tomarket their products outside the region, theygenerally run into problems with pricing, volume,and transportation. They also incur high costs for

•

••

Conduct further processing to convert anintermediate into an end product.

Package and store the product.

Distribute, market, and promote the product.

Box 17

Se~ond-Order.Org.nizations in Colombia (Case 1)

The Asociaci6n Nacional de Productores yProcesadores de Yuca (ANPPY), founded in1986, is open to individuals and legallyconstituted groups engaged in cassavaproduction and processing. By 1988, 53 ofthe country's 59 drying plants wereaffiliated.

ANPPY negotiates the sale of driedcassava, based on recommendations fromthe managers of CQ.-Ops and from regionalcassa.vatechnieal committees, who meetwith associa.tion management to discussproduction costs, etc., and reach aconsensus on the sale price. Members arealso free to engage in commercial contractsoutside this structure.

An important objective ofANPPY is torepresent and protect the interests of itsmembers before the government and otherpublic and private entities. It is concernedparticularly with policies that affectproduction, agroindustry, prices, markets,imports, and exports. In addition, theassociation has established marketingchannels for dried cassava and developedan information system dealing with pricesand markets. It can also provide trainingand technical assistance in production,

--------~ _.------ ------- ---------

95

marketing, bookkeeping, and financial andlegal matters.

Since ANPPY is not a co-op (members donot contribute any capitan, it has only alimited capacity to carry out many of thoseactivities. The association is supposed toreceive 1% of sales to cover its expenses.But since ANPPY is not actually engaged inselling, it depends entirely on members'willingness to pay the commission.

Another problem is management of thegeneral assembly. Because its social base iswidely dispersed, the association has noorganizational background, and membershave little sense of belonging. Recently,tensions between farmer associations andthe owners and operators of private sectordrying plants-groups with quite differentobjectives and social outlooks-have furtherweakened the organization.

In 1991 a new second-order associationof cooperatives-ASOCOSTA-was founded.As nonvoting members, individuals benefitfrom services but have no influence onpolicy. Increasingly, ASOCOSTA is seen asrepresenting farmers. groups and ANPPYthe private sector.

Addlllg Value to Root and Thber Crops

Box 18

Second-Order Organizations in Ecuador (Case 4)

The Uni6n de Asociaciones de Productoresand Procesadores de Yuca (UAPPY), foundedin 1986, coordinates cassava production,processing, and marketing for 18 associations(360 farmers). Its functions are to:

•

•

•

•

•

•

Seek funding (which enables members toobtain loans for three purposes-capitalinvestment in land and processing plants,working capital, and cassavaproduction-at interest rates set yearlyby the general assembly).

Market members' products, with the aimof increasing their profits.

Further process the associations'products to increase their worth andprofit margins.

Train members and ofTer them highquality, low-cost technical assistance.

Participate in research for thedevelopment of new technology.

Try to diversify the associations'products.

• Stimulate the socioeconomic progress ofmembers.

UAPPY set up a demonstration centerto test new technology, mill dried cassavachips and produce refined flour, carry outquality control, maintain machinery, andtrain members. The union handles fiveproducts (whole cassava flour, cassava flour,industrial starch, starch for humanconsumption, and starch bagasse), of whicha total of 1,750 t were manufactured in1990. The gene.ral assembly ofrepresentatives meets monthly to decidesuch matters as product prices.

DAPPY.has established strong linkswith other. institutions to execute anambitious plan of work in four areas:1) research dealing with production(includingseed),socioeconomic issues,andprocessing and utilization; 2) extension;3) education; and 4) institutional support.

• Provide first-order organizations withtechnical assistance and training in cropproduction, processing, accounting, etc.

• Identify new opportunities for products.

• Coordinate research and development,focusing on new products.

• Manage interinstitutional relations.

• Obtain funds through donations and creditand operate rotating funds, which providecredit to first-order organizations.

• Monitor and evaluate first-orderorganizations.

• Represent the processing sector beforegovernment policy makers.

Third-order organizations have not emergedin Latin America, although Colombia's ANPPYand the Brazilian National Cassava Congress

have attempted to shape policies at the nationallevel. ANPPY was instrumental in price settinguntil it split into regional groups, as a result ofdiverging opinions between farmer groups andprivate processors.

Organization ofcooperatinginstitutions

:-luccessful product development results from the,'olll'ctive work of various mstitutions. The idealarrangement for such efforts has been called anIllterorganizational collectivity," in which two or

more institutions make dl'cisions and act onuehalf of others, Their purpose is to promote andprotect common interests and obtain and allocatemuch larger resources than could any singleparticipant.


Finding organizational mechanisms toaccomplish these goals has proved difficult. Herewe examine two models. One is theinterinstitutional committee, an approachdiscussed in Boxes 19 and 20.

The second model is the technical team,whose purpose is to coordinate the work of fieldstaff in different disciplines and institutions. In aproject for cassava drying on the Atlantic coast ofColombia, state-level technical assistance teamswere established for this purpose. First, theyconducted a study to identify productionsurpluses and then suggested alternate markets.

Other functions of the teams were to carryout feasibility studies (a requirement forobtaining group credit), define productionstrategies (covering seed, inputs, machinery,labor, and credit), plan for expansion ofprocessing, and coordinate training. In this workthe teams received support and in-servicetraining from local and national institutions.

You might expect the private sector to showstrong interest in root crop processing once a pilotproject has demonstrated its economic feasibility.Yet, this is often not the case.

In Colombia, for example, individuals did notinvest in cassava drying plants until 7 years afterthe initial success, by which time more than 50cooperative plants were operating. One reason forthe private sector's reluctance may have been thelow status of cassava as a poor man's crop.Investors tend to focus on other options, such ascotton, even if these entail greater risk.

In other countries, such as Indonesia andThailand, where the private sector has a betterrecord of agroindustrial innovation, it has notonly taken the lead in large-scale processing butalso encouraged small-scale intermediateprocessing, sometimes involving farmercooperatives.

Even though the private sector may havedifferent objectives from those of a project fordeveloping root crop products, the two have manycommon interests. On that basis they can explorethe possibilities for a private sector role in theproject. In doing so it is important that theyidentify potential areas of conflict at an earlystage.

97

The Product and ProcessRevisited

Resolving basic issues of farmer and institutionalorganization is the first step in planning forcommercial expansion of the processingenterprise. The next step is to review carefullythe whole range of topics-from site selection totraining-that were dealt with in the pilot phase.

Site selection

In the pilot phase, you made a list of criteria forselecting the pilot plant site. If necessary, revisethe checklist in light of subsequent experiencewith the processing operation. Give particularattention to the following criteria:

• Rural versus urban locations

• Infrastructure needs

• Environmental impact (the effects of wastes,etc.)

• Farmer groups that have a sufficient priceincentive to plant more than they require foron-farm consumption

It is useful to separate essential criteria,such as availability of electricity or water, frommerely desirable ones (e.g., existence of a farmergroup or other enterprise at the site). Since somenew plants built in the commercial phase will belocated at less than ideal sites, it is also helpful toidentify the minimum set of conditions that mustbe met to justify building a plant.

Raw material supply

To estimate accurately the availability of rawmaterials within a reasonable distance of theprocessing plant, you need to:

• Identify actual and potential productionregions.

• Identify farmer groups or other enterprisesin each region that can ensure an adequateraw material supply of the appropriatequality and who wish to participate inprimary processing.

• Study competing markets to ensure that theprice the plant can pay for raw materialgives farmers a sufficient incentive to sell.

,-\(j'rU:If.{ ~hLue to Root and 1)Lber Crops

Box 19

Interinstitutional Committees in Brazil (Case 5)

A primary objective of a pilot project forcassava developIXlent in the .state of Ceara,Brazil, is. to strengthen communityorganizations, with heavy emphasis onparticipatory management.

Brazil has national, state, regional, andlocal institutions concerned with productionand processing of this crop. In the originalproject design, it was proposed that anadvisory council be established at the nationallevel to provide general guidance. Butplanners later decided that this wasunnecessary.

'Ib coordinate work on cassava at the statelevel, several agricultural institutions formedthe Ceara State Cassava Committee, as shownin Figure 8. It is composed of technical andadministrative representatives of four agenciesand at one time included the CIAT projectleader. The committee is chaired by the statesecretary of agriculture and meets monthly. Toensure effective coordination, it appointed anexecutive leader and established channels ofcommunication with the offices of research andextension services. The Committee also helpedidentify institutions to participate in workinggroups that coordinate local action.

The Ceara Cassava Committee is nowgenerally recognized as the coordinating body

for all activities related to cassavadevelopment in the state. It received increasedsupport as a result of a tour in project areas,which was organized by CIAT in mid-1989 forpolicy makers from Northeast Brazil(including the Ceara secretary of agriculture).

At the regional level, the organizationalplan calls for Regional Cassava Committees ineach project zone. Composed ofrepresentatives from technical supportagencies and farmer organizations, thesecommittees are intended to decentralizeproject administration and facilitate localparticipation in decision making.

At the local level, the Regional Committeescoordinate the work of technical teams,composed of extension workers and subjectmatter specialists from various research andextension agencies. The purpose of thesegroups is to stimulate the formation ofcommunity-based farmer groups forintegrated cassava production and processing.Initially, progress was hampered by the slowdevelopment of the State and RegionalCommittees. In the first year, 12 existing butdormant farmer groups were reorganized orreactivated and another 12 were organized. Bythe end of 1990,59 groups were engaged incassava drying.

Farmerorganizations

Federation of Cassava Farmersof the State of Ceara

(future goal)

Second-ordercassava farmer organizations

(cooperatives and associations)

Level Institutionalsupport

Ceara Cassava StateCommittee

Regional CassavaCommittee

Cassava farmerorganizations

MUnIcipality

Rural communities

+---+ Technical teams

+---+ Technicians

Figure 8. Farmer and institutional organization in an integrated cassava project, Ceara, Brazil.


Box 20

Interin$titutional Committees in the Philippines (Case 10)

The mission of the Philippine Root CropResearehandi Training Center (PRCRTC)and the Postharvest 'Iechnology Section ofthe Visayas State College ofAgriculture(ViSCA) is.to adopt, modify, or develop newpostharvest teehnologies, products, or usesto increase root crop production andutilization in the country.

In 1984., PRCRTC and ViSCAimplement~ed a project to produce a soy~

flavored sauce based on root crops; developand adopt suitable processing equipment;and devise utilization and marketingschemes. Recognizing that the project calledfor a multidisciplinary approach with closeintegration of its various components, thetwo organizations formed a team composedof a postharvest technologist, an economist,and an agricultural engineer.

Where there is a competing market for freshroots and tubers, try to obtain data on historicaltrends in prices, so you can identify seasonalpatterns. Creating a new market for a root cropshould help stabilize prices.

It is also critical that you take into accountseasonal variation in production. Since roots andtubers are highly perishable, it is rarelyeconomically feasible to store fresh produce forlater processing. As a consequence, the plant willprobably be able to operate only during harvestperiods. If the process involves sun drying of rawmaterial, you need to check whether the dryseason coincides with harvest time. Experience inLatin America suggests that planting and harvesttimes as well as dry season months varyconsiderably, even within one region of a country.Thus, you cannot safely extrapolate results fromthe pilot plant site to the whole region.

It should be easier to answer questions aboutraw material supply once various international

99

Each member was assigned specific tasksand given responsibility for making a programof activities for each project component. Theentire team then coUated and discussed theseprograms to define strategies and planactivities. Project leadership was rotatedaccording to the activity under consideration.

The team held a series of meetings withlocal support agencies to organize a projectmanagement board. A memorandum ofagreement was drawn up, defining the tasksand responsibilities of each participatingagency and specifying the terms andconditions under which farmer groups couldbecome owners of the venture. A core team ofeight farmer leaders was organized to build uptheir management capacity; this team alsoincluded two representatives of a localfoundation.

agricultural research centers have completed aproject underway now to map root and tuberproduction areas by edaphoclimatic zone. Withmore detailed information about target productionareas, you will be able to identify and prioritizethem more accurately.

Processing plants

The size of the processing plant affects thecomplexity of the operation as well as its ability toobtain an adequate raw material supply and dealwith equity issues. In general, many small plantsotTer more advantages than a few large ones.

Members of co-ops and other ruralenterprises generally have little education andexperience in operating and maintainingmachinery. These people will need substantialtechnical assistance and training III primaryprocessing, even if the procedures and equipmenttested in the pilot phase are appropriate for asmall-scale plant in an urban area

When a processing enterprise shifts from thepilot to commercial phase, the process itselfshould require few changes. This is particularlytrue if the commercialization strategy is toreplicate the pilot plant at other sites rather thanexpand its operations at the original location.Nonetheless, you may have to increase the scaleof the operation somewhat. If so you need toidentify stages in the process where bottleneckscould form as the plant's capacity grows. It is alsoimportant to make sure that processing capacitydoes not exceed the managerial ability of thesmall association or co-op.

Timing is crucial in the construction of newplants, especially where processing seasons arewell-defined. Even so, the timetable must beflexible. In deciding on a completion date, allowplenty of time to get the operation started and tofinish training before the main harvest season.Also take into account the inevitable delays inconstruction, resulting from such problems asshortage of labor at the harvest times of othercrops, shortage of building materials, and delaysin obtaining credit.

If the new plant is to be operated by a co-op,its members should help with construction asmuch as possible (under the supervision of anexperienced journeyman or master workman).This will not only help reduce costs but alsoconsolidate the group by focusing its energies on ajoint task.

The existence of many small processingplants can complicate quality control. To ensurethat product quality remains consistent as theoperation expands, you need to standardize asmany aspects of processing as possible across coops. The way raw materials are received andselected and storage time before processing haveespecially pronounced effects on product quality.Both managers and operators of small plants needto understand the importance of establishing andmaintaining strict quality standards. If a secondlevel organization is responsible forcommercializing the product, it too should beactively involved in quality control.

With products for human consumption and,to a lesser extent, those used as animal feed,sanitary and hygiene regulations are especially

j I II I

Addmg Value to Root and TUber Crops

critical. It may be difficult to comply with some ofthese either for technical reasons (e.g.,inadequate water quality) or lack of knowledge.Before the plant starts to operate, you need todeal with any shortcomings through technicaladjustments in the process or training. Before theplant is even built, it is a good idea to check local,regional, and national health regulations andlicensing requirements.

Th ensure that new plants operate efficientlyand economically, managers need extensivetraining and support in administration andaccounting. The agency providing credit forexpansion should help provide these services,since it has a strong interest in seeing that thefunds are well used. State entities and NGOs canalso contribute importantly. Some trainingfunctions could be performed by a second-orderorganization of co-ops, created specifically for thispurpose.

The pilot plant can be quite useful fortraining and demonstration. Operators of newenterprises can spend some time working there togain practical experience. In addition, people whohave worked in the pilot plant can spend timehelping those who are starting new ones. It's amistake, though, to overburden experiencedprocessors with training responsibilities.

Draw up a training plan to ensure thatenough operators and managers are available forthe number of plants required to meet expecteddemand for the products.

Product distribution

As noted in Unit 5, processors themselves candistribute a product aimed at the industrialmarket. But to reach consumers they need aspecialized distribution network. In marketing animproved version of a traditional product, youmay be able to use the existing network. Chancesare, though, it will have several levels ofIl1termediaries, each requiring a marketingmargin to cover expenses and provide income. Asthe processing enterprises evolve, they willalmost certainly want to increase theirparticipation in the marketing chain, taking oversume of the intermediaries' functions andmargins


You need to decide at this point how far totake product distribution. Do you rely onindependent wholesalers, create a second-orderenterprise to take charge of wholesale activities,or expand the scope of distribution to includesales at the retail level? The answer depends onthe characteristics of the product and market aswell as consumers' purchasing habits. SeeBoxes 21, 22, and 23 for examples of differentapproaches to product distribution.

If you opt for a distribution enterprise, itmust operate on a purely commercial basis. If it isunder the umbrella of a second-level co-op, itshould be administratively and financiallyseparate. Only then can the enterprise be trulycompetitive and maximize profits for members.

The main responsibilities of a distributionenterprise are to:

• Coordinate the supply of products fromdifferent enterprises according to marketdemand.

• Oversee quality control of finished productsand, if necessary, divide them according todifferent quality standards.

• Manage warehousing, stocks, andinventories.

• Sell and distribute the product to wholesalersand retailers.

• Coordinate promotional campaigns withdistribution.

Box 21

A Three-TIered System for Distributing Potato Productsin India (Case 6)

Ii

l

In Uttar Pradesh, India's largest potatogrowing state, nearly all of the 6 milliontons produced are harvested in just onemonth. Since farmers need cash and cannotafford to the tubers in cold stor theyare force 1 at very low prices. segrowers crease comes bycreati ith alte marketsthrough -level processing.

In 1985 the Society for Development ofAppropriate Technology (SO"I'EC) wasestab d received a three-year grantfrom work on problems with dryingpotatoes and using them in Indian foods.Project staff realized that a processingplant, by itself, would be incapable ofhandling all the activities needed to makevillage-level processing a success. For thatreason SOTEC established a three-tierstructure with:

1. Village-level drying plants.

2. A unit that sorts chips from 8-12 nearbydrying units (quality controD, grindsthem into powder, packages the powderin bulk or for retail, stores the product,and fills orders.

10]

3. A unit that hand.les sales and marketing,ad.vertising, package design, distributionofprQd.uctstoretailers, billing, andcollection. It also establishes productiontargets for tiers 1. and 2 and may obtainbank loans for purchasing products.

Initially, SOTEe operated at all threelevels. In 1988it turned over most of thesales activities to ind.ependent companiesand by 1990 was ablato withdraw from salesaltogether. Currently, SOTEe is establishinga federation, which will eventually assumeall the responsibilities of tier 2 as well asequipment supply. It will also help procureraw materials and financing.

One problem with this system is in thecollection of dried chips. Villagers lackadequate storage facUities. And transportcosts are high, because the product is of lowdensity and is shipped in lots ofless than atruckload. There are also difficulties withstorage of the final product. Currently, it istaken to SOTEC headquarters for sortingand storage and then reshipped to themarket as required.

Addlllg Value to Root arul Thber Crops

Box 22

Dividing and Conquering with Potato Products in Peru (Case 7)

The agroindustrial program of the Centrode Investigaci6n, Documentacion,Asesoramiento y Servicios (IDEAS)established a pilot plant at Concepcion,Junfn, to promote small-scale manufactureof processed potato products in ruralareas. The products were to be sold in theprincipal supermarkets of Lima under thebrand name Abril. The project encounteredseveral problems:

1. Target group. The products wereoriginally intended to improve thenutrition of low-income urbanconsumers, who would be reachedthrough official food programs orinstitutional markets. Afterwards,other products were to be developed formiddle- and high-income consumers.Although no detailed marketingsurveys were conducted, projectplanners did learn that low-incomeconsumers preferred cheaper products(such as grains and simpleunprocessed flours) to processedproducts. As a result, the project had toalter its strategy.

2. Pricing. The first product developed,a flour mixture, cost more per kilogramthan simple unprocessed flours andshowed no apparent advantage overother processed products. Its onlybenefit was low cost per nutrient andration. And since consumers cannotreadily perceive this, the productwould have been difficult to market.Changes in government policy onsubsidies for imported foods put theproduct at an even greaterdisadvantage. Consequently, projectstaff decided to modify it.

3. Marketing strategy. Initially, theproject had no strategy, because theoriginal idea was that its productswould be marketed through

government and private institutions.Thus, even though recommendationswere made about the name, packaging,etc., no action was taken on theseissues.

Because of its difficulties in identifyinga suitable flour mixture, the project decidedto market three other products: precookedflours ("creams"); precooked, peeled grains;and specialized products (such as a milksubstitute named Chicolac) based onpotatoes, maize, quinoa, etc. The marketwas divided into three segments: 1) middleand high-income consumers, who shop insupermarkets in Lima; 2) low- and middleincome consumers who would be reachedthrough shops, markets, and specialpromotions in the area around the plant;and 3) the institutional market, principallygovernment programs offering free mealsfor the poor.

The project eventually sold productsthrough five of the six supermarket chains,distributing directly to 57 points ofpurchase. Advertising campaigns wereconsidered unnecessary, because thepackage was attractive, the quality of theproducts was good, and prices were fairlylow. Chicolac was introduced successfully inthe municipal Glass of Milk Program inHuancayo and Concepcion, which included70,000 children and mothers. Project staffpromoted this product by demonstrating itsuse for mothers and by letting childrentaste it. The project promoted otherproducts by offering introductory discountsand distributing recipes.

Although the Abril products cost morethan competing items, consumers werewilling to pay the difference. The volume ofsales was limited by lack of working capital.As a result, the project used only 50% of itsproduction capacity.


Box 23

Distribution ofFresh Cassava in Colombia (Case 2)

•

•

In spite of its high consumer appeal, acoI1SElrvedcassa"Va. pl'oductwas•trading atlo\V volume--only 1.2%of the market(10tlweekcompared with a total marketvolume ofBOOtiweek). Apparently, theproblem was insufficient promotion. Anadvertising. agency was hired to develop abrand narne(Yucafreska)and slogan anddesign a promotional campaign. But thePrograma .de Desarrollo Rurallntegrado(DRI) was unable to obtain funding to carryout the campaign.

Initially, each cooperativemanufacturing the product also handledmarketing. Then, as shown in Figure 9, theproject proposed to establish a centralmarketing organization, which wouldcoordinate regional supply and qualitycontrol. During the pilot phase, theCooperativa de Producci6n y Mercadeo deRepe16n (COOPROMERCAR), Atlfmtico,was selected for this purpose.

In the commercial phase, responsibilityfor marketing was given to the Federaci6nde Organi~acionesAgropecuarias deColombia (FAGROCOL), a second-orderfederation of co-ops of cassava producers

Monitor the volume of sales, including thoseof competitors, as well as the profit marginsof wholesalers and retailers.

Give feedback both to processing groups andproduct organizations.

and shopkeepers formed in Barranquilla in1989. Since ANPPY did not authorize thisgroup to use the brand name Yucafreska, itadopted the name Superyuca in mid-1990but has not yet registered it.

FAGROCOL has the capacity to sell 50 tof cassava weekly, of which 15 t are freshcassava, conserved and sold in bags tosupermarkets. From November 1989 toJune 1990, the federation sold 102 t ofbagged cassava. It has also improved theprocessing technology and product quality.Return of deteriorated cassava has fallenfrom 20% to 5%. Nonetheless, the operationis not profitable, since it must market 26.6 ta month to break even.

Currently, the project is trying to reducecosts. But to increase the volumes ofproduction and sales, it will need additionalsupport to establish a processing plant andcarry out a large-scale publicity campaign.Some wholesalers have opted to store andsell untreated cassava roots packed inpolypropylene bags. The roots can be storedfor 1 or 2 days, which was previouslyimpossible.

knowledge of product storage. State agencies arenot the place to look for these skills, since they donot operate on a commercial basis. Instead, youmust seek input from organizations that supportsmall businesses and can provide relevanttraining and technical assistance.

The location of the enterprise's central officesand warehouse is very important. If the productwill be sold in small shops, the warehouse shouldbe near the central wholesale market whereshopkeepers purchase most of their goods.

Obviously, to manage distribution demandsquite different skills from those needed inprocessing. The former requires a high degree ofcompetence in marketing and business and a good

103

Product promotion

To meet the objectives outlined in the marketplan, you need to make consumers aware of theproduct and its advantages and inform themwhere it can be purchased. Many publicationsdescribe how to plan and execute a promotionalcampaign (see the list at the end of this unit).Here we summarize a few basic principles.

Production

Wholesale


Cassava cooperatives

2,4,12 kg

tWarehouse: second-order organizationor small businessPI

- --- ~---,------ -------_._-_.

2,4 kg 2,4,12 kg, 12 kg 12 kg

Retail SupermarketsP 1,2

Corner shops ABP 1,2,3,4

2,4 kg

,

2,4 kg Loose rootsor loose roots

, ,

ConsumptionHousewife ABP 6,4,2,3,8

Housewife ABP6,4,7

Housewife CDP6,5,7,3

RestaurantsP2,3,1

Promotional materials:

P1- free sampleP2 - visit by salesmanP3 - pamphletP4 - newspaper advertisementP5 - radio advertisementP6 - regional TV advertisementP7 - posterP8 - material distributed at point of purchase

Bag size (kg of roots per bag):

2,4,12

Socioeconomic standing of retailersand consumers:

AB - middle to higher incomeCD - lower income

Figure 9. A scheme for distributing and promoting storable cassava in bags in Colombia.

104


When deciding what media to use, you needto take into account the characteristics of targetconsumers (e.g., newspaper advertisements willnot reach barely literate people in the low-incomestrata). In developing promotional materials,avoid using regional expressions, dialect, and thelike, since they may limit the geographical area inwhich the materials can be used. In all materialsthe brand name, logo, and slogan should figureprominently.

A small co-op cannot afford to launch a newproduct with the same degree of fanfare as largefood companies. Even if it did have the resourcesto bombard consumers with publicity, the co-opwould be hard pressed to satisfy the resultingsudden increase in demand. A safer alternative isto increase promotional activities gradually, as thedistribution network expands and product supplyincreases. To create a market costs money. But inthe long run it should pay for itself.

Training

To train farmers and other rural people inprocessing and product distribution requires astrong commitment from the institutionscollaborating in the integrated project. In additionto dominating the technical aspects of plantoperations, these people need to acquire newskills in business administration, personnelmanagement, financial analysis, andbookkeeping. They must also learn how tonegotiate, motivate employees, and work ingroups. A big part of the challenge is to findappropriate training methods for the manyfarmers who may be quite astute but functionallyilliterate. They may need extensive on-the-jobtraining, until they acquire enough skill andconfidence to manage the processing plantsthemselves.

Personnel from both the private and publicsectors should receive training in the technicalaspects of root and tuber production andprocessing as well as in methods for working withfarmer groups and small-scale agroindustries.One problem to watch for is high turnover ofpersonnel, which can greatly increase theinvestment required in training.

10f)


In addition to covering all aspects of processingand marketing new products, your plan of actionfor the commercial phase must answer twoquestions that have a direct bearing on the longterm success of the processing enterprise. First,how will the activities of this phase be financed?And second, how will investors or creditors beassured that the project is working steadilytoward its objectives and achieving the expectedimpact?

Sources offinancing

The pilot phase is normally funded throughspecial projects, since no small business orcooperative can be expected to finance thedevelopment of a product that has not proven tobe technically and economically feasible. But oncethe plant reaches the commercial phase, it shouldbe able to obtain credit for replicating orexpanding its manufacturing capability andproviding sufficient working capital to coverinitial expenses. Even at this stage, though, thereis an element of risk, since the success of theproject depends on its ability to develop aprocessing enterprise with small-scale farmers.

For that reason you will probably have toseek credit on relatively easy terms from a publicsector organization. It is also important thatcredit be accompanied by a well-designed programof training and technical and administrativesupport, provided either by the public or privatesector. Such activities can generally be financedby state organizations, assuming that theirdevelopment priorities are in accord with those ofthe project. In countries where root and tubercrops receive low priority, a more likely source offunding may be NGOs that are active in ruraldevelopment and are working with farmergroups.

Small-scale producers and processors rarelyhave enough capital to finance the promotion of aconsumer product on a commercial scale. This isalso the activity for which it is hardest to findoutside financial support. By and large productpromotion is not considered part of a rural

development project. That is why it is importantto find a second-order organization willing toundertake this task. Possible candidates areinstitutions interested in improving theavailability of staple foods in urban areas.

'Ib reduce their dependence on externalcredit, processors need to build up capital. For coops this means finding a balance betweendistributing profits and reinvesting them in thebusiness to finance growth and reducedependence on credit. Poorer groups in particularare under a lot of pressure to distribute profits.

Monitoring and evaluation

If the project goes ahead with the commercialphase, it will need a system to monitor theprogress of the product in penetrating targetmarkets. Whether the project achieves itsobjectives depends to a great extent on the qualityof the market plan, which must be both detailedand flexible. You may have to modify the plan inresponse to developments in production,processing, and markets. For example, if thegrowing season is so poor that the supply of rootsand tubers falls dramatically, you will need to cutback processing or find other sources of supply.

If the enterprise depends on funds from thepublic sector, it is also important to monitorchanges in government policy. In addition, themarketing plan should take into account probabledelays in forming and consolidating co-ops,obtaining credit, and constructing processingplants. 'Ib ignore these possibilities may causeother project activities to get out of phase.

Make an effort to document experiencegained at the outset of the commercial phase. Thisinformation may help you improve plans forbreaking into other markets.

The monitoring system should focus inparticular on documenting:

Adding Value 10 Rool and TUber Crops

• Consumer satisfaction with product quality,usefulness, and price

Checklist 15 indicates other types ofinformation you should be able to obtain from themonitoring and evaluation system. It has to be inplace when the project begins and must bemaintained by the participating organizations. Alarge part of the monitoring can be based on theproject's administrative records. It might also beuseful to make an annual survey to obtainanswers to specific questions.

'Ib learn first-hand whether they have madea sound investment, the organizations that havesupported product development ought to conductan evaluation of its commercial impact. Asillustrated in Box 24, they should judge this, notJust by the financial viability of the processingenterprise, but also by its distribution of benefits.These can accrue to farmers, landless laborers,rural transporters, urban distributors, retailshopkeepers, and consumers. It is also importantto determine whether the enterprise has had anyadverse effects.

Data generated by mOl1ltoring andevaluating the project are a potential source ofideas for new markets or products. Thisinformation should be communicated to thetechnical research components of the project forfurther evaluation. (For more information onproject monitoring and evaluation, see the list ofpublications at the end of thiS unit.)

Toward a Self-ReliantAgroindustry

Good planning must be followed by decisiveaction. In this section we outline the series ofsteps that you need to take after developing anaction plan for the commercIal phase.

Executing the plan

•

•

Sales volumes (including totals as well asbreakdowns by market outlet and byconsumers' socioeconomic status),inventories, and stocks

Consumer behavior with respect to first andrepeat sales

Executing a plan to expand the agroindustry islargely a matter of integrating its differentcomponents. For example, efforts to increase cropproduction must be geared to the construction ofnew processing plants. Proper integrationdepends III turn on close coordination of theactivities of participating organizations (research


Checklist 15

Information Gathered by a Project Monitoringand Evaluation System

Raw material supply

• Origin

Distance to plant

Percentage brought by middlemen

Percentage supplied by village

Supply from co-op members inrelation to break-even point

• Quality

Dry matter content

Variety

Planting period

• Regularity (quantity processed by lot)

• Value

Purchase price

Percentage of remainders

Processing

• Utilization of capacity (a measure ofoverall efficiency)

extension agencies, NGOs, farmer groups, theprivate sector, etc.). This is a highly complex task,for which a single entity ought to be given overallresponsibility.

As the commercial phase moves forward, youneed to make good use of feedback from themonitoring system to keep the project on trackand modify the plan of action in response tounforeseen developments. This is also the time toprepare for the eventual withdrawal of projectsupport. Th avoid creating dependence, the projectmust ensure that farmer groups get enoughtraining and experience to handle all aspects ofthe enterprise. Even if these groups have tocontract out certain commercial or financial tasks,they should always keep these activities undertheir overall control.

107

• Product quality

Moisture content

Conversion factor

Color

• Production

Cost structure

Costs in relation to selling price

Commercialization

• Packaging

Types of materials

Bulk

Sizes

• Utilization of warehouse (measured bythe ratio of processed product to storagecapacity)

• Rotation of capital (days of delay inpayment)

• Ratio of inventory value to workingcapital

Phasing out the project

The end product of the commerclcd phase is aneconomically viable agroindustry that requires nooutside support to survive, fulfills the socialobjectives of the project, and is able to expand andadapt to a changing commercial environment (seeBox 25 for an example). Whether the projectachIeves this objective depends on numeroustechnIcal, economic, and human factors.

I f the agroindustry has expanded by buildinga i;lrgc numbc;r of small processing units, ratherthan c;xpand the original pilot plant, itsproductll1n capacity will be fragmented. ToadlIcve commercial success, the agroindustry willreqUIre close coordination in product marketingand distnbution. This will depend on the strength


Box 24

A Comprehensive Strategy for Project Evaluationin Brazil (Case 5)

A pilot project for cassava development inCeara, Brazil, is tracking progres.s towardkey objectives through a comprehensiveevaluation strategy. This involvesmonitoring the following activities:

• Daily progress in cassava processing

• Impact on cassava production

• Distribution of benefits

The project also assesses thesocioeconomic status of participatingfarmers to determine impact on incomegeneration and distribution, educationlevels, farmer organizations, access to thepolitical support system, and social benefits.This monitoring takes place at three levels,with a different method at each:

• Specific project objectives: collection ofbaseline data on participatingorganizations.

of the second-order organization formed to handlethose functions.

Phasing out a project smoothly requires goodmanagement. This underscores the importance oftraining key people in the enterprise. One optionis to hire managers working in other businesses.But there is a downside to bringing outsiders intoa community enterprise. To avoid problems theoutside manager would need to be under thesupervision of well-trained personnelrepresenting the target beneficiaries.

Training managers is complex, since bydefinition they must coordinate a wide range offunctions (technical, financial, personnel, sales,etc.). Projects in Colombia and Ecuador have beendeveloping managerial capacity for more than adecade. To consolidate this achievement takes ageneration. The children of original members offarmer co-ops are just now completing their

ilK

• Target population: a survey of 150participating cassava farmers.

• Target areas: small-group surveys andintensive follow-up with a limitednumber of farmers (sample populationsfor this level are drawn from the secondlevel).

Information from the project's databasesis reported in periodic bulletins. On thisbasis the technical team of the CassavaCommittee gauges project performance, andthe farmer groups evaluate their ownperformance relative to one other. Projectstaff have found that getting feedback toindividual groups rapidly is critical fordetecting and correcting problems. Thedatabases have proved useful for preparingdetailed project reports and demonstratingthe size and distribution of benefits to theKellogg Foundation, which is funding thisproject.

formal education. With the technical or businessskills they have acquired in school, plus theexperience they gain by working in processingenterprises, these young people will be wellprepared to provide solid leadership in the future.

Institutional support should not bewithdrawn before the enterprise is financiallysolvent and has reasonable prospects forcontinued success. Above all, it must have accessto a sustainable source of credit. Some options arecommercial bank loans, a special credit lineestablished for small-scale agroindustries, or arevolving credit fund, which the enterprisemanages, using funds left over from the productdevelopment project.

Finally, the enterprise needs to establishchannels of communication with governmentpolicy makers. It can do this through localgovernment or sectoral organizations operating at


Box 25

Versatile Processing Enterprises in Ecuador (Case 4)

For several years Ecuador's UAPPY; whichsupplied dry cassava for the shrimp feedin.dustI1; enjoyed steady growth inmembership iand volume of production.Then, in 1989 it suffered a major setback.The country's shrimp industry collapsed asa result offoreign competition and otherproblems. The demand for dried cassava fellsharply, and large stocks of product wereleft.unsoldat the end of the processingseason.

The organization responded in twoways: first, by seeking other markets forcassava. chip and flour products in Ecuadorand, second, by starting to produce starchfrom cassava. UAPPY found demand inseveral industrial markets, specificallyamong producers of plywood, glue, andcardboard boxes. But to take advantage ofthese opportunities, processors had to raisethe quality of their products. Someenterprises .started producing flour frompeeled roots and thus created many newjobs for women as manual laborers.

In the ManaM area of Ecuador, wherethe project was located, there is a tradition

the national level (e.g., associations of ruralagroindustries and crop producers and unions ofsmall-scale farmers). It is increasingly commonfor producers and importers of nationallysignificant raw materials and products to havepolitical lobbies. These groups can achievesignificant shifts in government policy, especiallywith respect to support prices, import tariffs, etc.The agroindustrial enterprise must have arepresentative at that level to ensure that itsinterests are taken into account.

Occasionally, policies formulated for onesector of the economy may unintentionally harmthe interests of the processing enterprise. Or itscompetitive position may be weakened by theefforts of one interest group to gain advantagesover another. In Colombia, for example, the

109

of small-scale cassava starch production.The UAPPY saw an opportunity to expandthe market for this product by improving itsquality. A women's group was formed forstarch extraction and successfullyestablished a profitable operation.

The union next conducted a marketsurvey of food, feed, and other industries toestimate the potential demand for itsvarious products. By that time cassavaprocessors were turning out flour, starch,and by-products of different qualitystandards. Based on the results of thesurvey, UAPPY was able to allocateresources to the different processesaccording to their market potential.

Through this experience UAPPYmembers learned the importance of productand market diversification for ensuring thelong-term viability of the processingenterprise. The union set up a special unitto develop new and improved processes aswell as a demonstration plant for training.The organization's research capacity shouldenable it adapt to changing circumstancesin the future.

microbial standard was set higher for cassavaflour than for wheat flour, because the standardscommittee included representatives of wheat flourmillers (importers) but not of cassava producers.

The sign ofself-reliance

While some basic products have a long life span,consumer items tend to be relatively short-lived.No matter what market an enterprise caters to, itcannot expect to survive indefinitely by producingJust one product.

This is especially true under current globaleconomic conditions. Increasingly, the marketdetermines raw material and product prices, andnational economies are being opened up tointernational competition. World market prices

Ai/i/IfIA' Value to Root arul Thber Crops

for many agricultural goods fluctuate widely.The resulting uncertainty makes thedevelopment of root and tuber products tocompete with animal feed grains or wheat flour,for example, a fairly risky business.

plant itself. An in-house capacity to develop newproducts is a sure sign of the agroindustry's selfreliance and vitality.


Table 22. Status of a product development project at itsconclusion.

By the end of the commercial phase, the status ofa product development project should be asdescribed in Table 22. This final phase has twomain outputs: a commercially successful productand a financially sound enterprise. If the projecthas dom~ a good job of monitoring progress, you

Another reason not to rely exclusively onone product is that consumer tastes and habitschange, though more gradually than economicconditions.

If the project's commercial phase has beensuccessful, it must ensure (before institutionalsupport is withdrawn) that the enterpriseacquires a capacity for continued development ofnew or improved products. Here are severalstrategies for developing a more versatileagroindustry: I'roJect component Status at end of Phase 4

• Improve the quality of existing products, sothey can enter new markets.

Defined, with good potential ofbemg reached

• Develop totally new products.H,egiol-j Industry spreading from initial

to other regions

This last strategy involves a series ofincreasingly complex processing procedures.Each must be built on a solid foundationestablished in previous steps. In other words,don't try to master all the procedures at once.

During the commercial phase, the projectcomes into close contact with the market. Thiscan be an important source of ideas for new orimproved products. To act on these ](}eas, th.,enterprise must repea t the process of prod uctdevelopment, starting with re~search, a~

described in Unit 4.

At this point the project may be prolongedto support the exploration of new opportunI tie~,

But a better option is for the enterprise to buildits own product development capacity. It mayhave to contract out research for which it lacksthe necessary expertise (e.g., in equipmentdesign or analysis of product quality). Butassuming that the new product reaches the pilotphase, the enterprise should operate the pilot

Commercially successful; moreIdeas now generated

True potential being realized

Spjpcted model successful, moreappparing

.\1 on Itoring to ensure impactmaches target

En1.erpnse type

llen,of;cl"ries

l'rod'l('l

Marke!

AllOther output oft1'1' :lroduct is a set ofl!l"ils fl!r new products, ThE' enterprise should be<lui" tu devc.)op these itself "ven it has to'1Illtral't uut some tasks Th' enterprise may

;:c'l'd projC'ct funding for thi,; purpose, If so it"hould be' actively involvl:d lTl formulating and,·xl'cutini.! the projecto In Latin America two

~l(just.rlPs have recentlv taken this step. Ino,J!unbia :\80('081'A ohtamcd governmentJ'l'di1 I Ii a pl',I.Wet tli produce balanced animalI'l'll ratIOn:,: using cassava. tn Ecuador UAPPY

has cCl,'~ved credit and donations for various

shuuld be able to trace its effects on targetlwneficlaries. Since it takes time for quantifiablelll'lwfits to accrue, impact IJllght not be assessedI; I"~ 'm J ture~ly

Further process existing products (e.g.,from chips to flour to feed rations for animalproduction).

•

j \1

Unit 6: The ComrMrcial Phase

Further Reading

Planning for the commercial phase

Brace Research Institute. 1992 Guia de planificaci6nde pequefias empresas agroalimentarias.Translated from French and published by theInstituto Interamericano de Cooperaci6n para laAgricultura (IICA), San Jose, Costa Rica.Originally published by the Brace ResearchInstitute, Faculty of Engineering, McGillUniversity, Quebec, Canada.

Fernandez, R.A. (ed.). 1992. Commercialization ofresearch results. Southeast Asian RegionalCenter for Graduate Study and Research inAgriculture (SEARCA), Los Bafios, Philippines.

Howell, J. 1979. Assessing management andorganisations for agricultural developmentprojects. In: Institutions, management andagricultural development. Occasional Paper no. 3.Overseas Development Institute, AgriculturalAdministration Unit, London, UK.

United Nations Development Programme (UNDP),Government of The Netherlands, InternationalLabor Organization (ILO), and United NationsIndustrial Development Organization (UNIDO).1988. Development of rural small-scaleenterprises: Lessons from experience. UNDP,New York, USA.

Small-scale farmer organizations

Brekelbaum, T. 1990. Organizaciones campesinas:Sinopsis de una revisi6n de literatura. In: Manualpara disefiar proyectos integrales de yuca. CIAT,Cali, Colombia.

Romanoff, S.A. 1989. Manual de referencia para lapromocion de asociaciones de productores yprocesadores de yuca. Serie Tecnica Manual no. 1.Fundacion Ecuatoriana de InvestigacionesAgropecuarias (FUNDAGRO), Quito, Ecuador.

____. 1991. Farmer organizations in integratedcassava projects. In: Perez-Crespo, C.A. (ed.).Integrated cassava projects. Working Documentno. 78. CIAT, Cali, Colombia.

Product promotion

Abbott, J.C. 1987. Agricultural marketing enterprisesfor the developing world. Cambridge UniversityPress, Cambridge, UK.

111

Hiebing, Jr., R.G. and Cooper, S.W. 1990. Thesuccessful marketing plan. NTC PublishingGroup., Lincolnwood, IL, USA.

Kotler, P. and Armstrong, G. 1992. Marketing: Anintroduction. 3rd ed. Prentice Hall, New York,NY, USA.

Stanton, w.J.; Ezel, M.J.; and Walker, B.J. 1993.Fundamentals of marketing. 10th ed. McGrawHill, New York, NY, USA.

Project monitoring and evaluation

Ahmed, V. and Bamberger, B. 1989. Monitoring andevaluating development projects World Bank,Washington, DC, USA.

Bamberger, M. and Cheema, S. 1990 Case studies ofproject sustainability. World Bank, Washington,DC, USA.

Henitt~z, R. 1987. Manual de asistencl3 tecnica delProyecto DRIIPMA para el desarrollo agroindustrial del cultivo de la yuca en la CostaAtlantica. Corporaci6n Fondo de Apoyo aEmpresas Asociativas, Bogota, Colombia.

Bode, P. 1991. Monitoring and evaluatIOn systems forcassava drying projects. In: Perez-Crespo, C.A.(ed.) Integrated cassava projects. WorkingDocument no. 78. CIAT, Cali, Colombia.

Casley, D..]. 1987. Project monitoring and evaluation inagriculture. Johns Hopkins University Press forthe World Bank, Baltimore, MD, USA.

Casley, D.J. and Kumar, K. 1988. Collection, analysisand use of monitoring and evaluation data. JohnsHopkins University Press for the World Bank,Baltimore, MD, USA.

Clayton, E. and Petry, F. (eds.). Monitoring systems foragricultural and rural development projects. FAOEconomic and Social Development Paper no. 12.Food and Agriculture Organization (FAO), Rome,Italy

Dlaz, KO. 1990. Marco metodol6gico para laplaneaci6n de un proyecto de produccion de yucay cultivos asociados; aplicaci6n del monitoreo. In:Manual para disefiar proyectos integrales deyuca. CIAT, Cali, Colombia.

Dorward, A. 1988. The classification and organizationof monitoring and evaluation activities inagncultural development projects. AgriculturalAdministration and Extension 30(3):107-127.

McLean, D. 1988. The logical framework in researchplanning and evaluation. Working Paper no. 12.International Service for National AgriculturalResearch (ISNAR), The Hague, Netherlands.

Maddock, N. 1987. On the monitoring and evaluationof agricultural development projects. AgriculturalAdministration and Extension 25(3):177-188.

Oven, R. von. 1979. Common problems in monitoringand evaluation of agricultural developmentprojects. Zeitschrift fur AuslandischLandwirtschaft 18(3):226-238.

Petry, F. 1981. A typology of indicators for monitoringagricultural projects and programmes. FAOEconomic and Social Development Paper12:253-261.

l~


Scott, C 1985. Sampling for monitoring andevaluation. World Bank, Washington, DC, USA.

Smith, PJ. 1985. Monitoring and evaluation ofagricultural development projects: Definitionsand methodology. Agricultural Administrationand Extension 18(2):107-120.

Unit 7

A Review of Major Issues

In this final unit, we review important issuesraised in Units 3-6 and conclude with a look to thefuture of small-scale agroindustries for root andtuber product development.

Key Aspects of ProductDevelopment

Here we discuss key aspects of productdevelopment, drawing lessons from the casestudies to underscore important points.

Project objectives

Product development projects are complex,requiring careful integration of many diversegroups and tasks. A clear statement of objectives,with which all parties agree, can reduce thepotential for confusion and conflict. It is especiallyimportant to be clear from the start about projectbeneficiaries, since you will probably have tochoose between rural producers and urbanconsumers.

As the project proceeds, new experience mayrequire that you modify the objectives. If so, dothis explicitly and with the consent of all parties.If the private sector is involved in the project, asan investor in processing or as a source of funds,it is particularly necessary that the project worksteadily toward common objectives.

Identifying opportunities

Public resources for research are scarce. Toensure that they are wisely invested in productdevelopment, you need to do a thorough job ofidentifying the best possible product. Resist thetemptation to rush into the investigation oftechnical aspects of processing after consideringproduct options only superficially. Extra timespent on this latter task will pay dividends later.

1] :3

There are many examples of research thatled nowhere, because it focused on a product thatwas unprofitable or technically difficult or had nomarket. In most instances a more comprehensiveinvestigation of opportunities at the outset wouldhave given the research a different focus.

Take the case of a project in the Chapareregion of Bolivia, where five plants were built fornatural drying of cassava to produce chips foranimal feed. Once the plants were in operation, itbecame apparent that, because of high rainfall inthe region, cassava required artificial drying. Italso came to light that this product could notcompete with low-priced maize in the local targetmarket.

The project turned instead to producinghigh-quality flour (which requires artificialdrying) as a substitute for expensive, importedwheat. If the project had made a more thoroughevaluation of product options at the outset, itwould have saved time and money in the long runand benefited small-scale farmers much sooner.

Putting the market first

In this manual we emphasize the importance ofputting market factors first in key decisions aboutproduct development. We also take into accounttechnology options but always in view of theirprospects in the market.

Many research institutes do Just theopposite, letting technology serve as the drivingforce behind product development. The mainattraction of this approach is that it givesresearchers maximum freedom to exercise theircreativity. But they also run a high risk ofdedicating scarce resources to processes andproducts with little chance of commercial success.We're convinced that research is more efficientand more likely to payoff if guided by marketmformation from the start.

Market-driven research must beinterdisciplinary. It requires that technicalresearchers learn to deal with the results ofmarket and social studies and to interact with awide array of project cooperators andbeneficiaries. The value of interdisciplinaryresearch is evident from the experience of theVisayas State College of Agriculture (ViSCA) inthe Philippines. By participating in market andconsumer surveys, technical researchers gainednew insights into the needs and problems ofconsumers, while market researchers learnednew ways to interpret their data.

If the institute responsible for technicalresearch in your project has few economists orother social scientists, contact local universities.With them you can explore the possibilities ofconducting market research through studentthesis projects and other forms of collaboration.Universities should be able to provide low-cost,rapid methods of gathering market informationthat do not slow down technical research.

Pilot projects

The pilot phase is a central part of productdevelopment. Its purpose is to test the processand product on a small scale under "real life"commercial conditions.

In many research institutes, what passes fora pilot plant is more correctly termed an"experimental plant." The difference is that inthe latter researchers control the process andfrequently the raw material supply as well.Moreover, the product is marketed underartificial conditions; production costs areunrealistic, and the product is sold largely toemployees.

An experimental plant may be essential forresearch and useful for training anddemonstration. But it has limited value forjudging the commercial feasibility of a processingplant. To examine the key factors that determinethis (such as logistics, ra w material quality, andenergy and labor supply), you must establish apilot plant under realistic circumstances. Asillustrated in Box 26, the technical problems thatmay arise under these conditions provide a focusfor adaptive research in the pilot phase.


Cooperatives or small businesses?

At an early stage in product development, youneed to weigh the merits of differentorganizational models and decide which is themost appropriate vehicle for commercializing aparticular product. As discussed in Unit 2, thereare two main options: to organize the enterpriseas a cooperative or association with primarilysocial ends or as a small business with chieflycommercial ends. Whether you choose one or theother or some combination of the two shoulddepend on project objectives, on the social andpolitical environment of the target region, and onthe technology involved.

The advantage of cooperatives is that theydistribute benefits over a wide social base. Theyare the obvious choice if the project's overridingconcern is to benefit the rural poor. Smallbusinesses are generally more effective and growfaster than co-ops but generate fewer socialbenefits. This shortcoming may not matter if theproject's objective is to benefit urban consumers.

Whether participating IOstitutions choose towork through cooperatives also depends on thetrack record of such groups in the target region.In some parts of the world, social factors orgovernment policies work against cooperatives.

Another important comnderation is thesophistication of the technology. If it is suitable forlarge-scale production, you might consider joiningforces with a big food company. You can thenshare development costs and achieve greaterimpact more rapidly. A further advantage is thatprivate enterprise can promote the product on alarge scale, a task that is beyond the resources ofpublic sector projects. In order for such anarrangement to work, however, it must becompatible with the project's objectives. Smallscale farmers will benefit only if the market forraw material greatly expands; they will gainnothing from processing or added value.

In the Philippines, ViSCA has taken severalproducts (a sweet potato beverage, ketchup, andDelicious SP, a dried snack.1 as far as the researchphase and then passed them on to privatecompanies for manufacturing on a pilot andeventually commercial scale In the first such


Box 26

Problem Solving at the Pilot Phase in Colombia (Case 3)

A pilot plant was set up in Colombia todetermine if it would be economicallyfeasible for a small-scale farmer cooperativeto produce high-quality flour for the foodindustry. The plant was based on a designtested under experimental conditions atCIAT. The process was found to be efficient,especially in artificial drying of cassavachips, and the end product met Colombianquality standards for total and coliformbacteria levels.

Nonetheless, initial results from thepilot plant showed that the artificial dryingsystem was not performing efficientlyenough and that product quality wasinadequate, especially with respect tomicrobial standards. Research conducted atthe plant identified two major problems.First, there were delays of up to 2 daysbetween harvest and processing. Andsecond, drying time was long when theplant was run at full capacity (3 t of freshchips per batch).

case, the college made an exclusive contract witha company, which never developed the productany further. Since then ViSCA has usednonexclusive contracts. One of the terms is thatthe college helps companies conduct pilot testingof products, and in return they donate equipmentto the college.

Phasing out the project

Most of the recommendations in this manual arebased on a wide range of experience in variousparts of the world. The one step about which wecan't speak with much certainty is the commercialphase of product development, during which theproject framework is withdrawn. The problem isthat there are still few cases in which this hashappened. The only projects we can describe inthe past tense are those established in Colombiafor producing dried and fresh cassava. And eventhese are hardly perfect examples.

l1S

The solutions were tighter control overthe supply of fresh roots to ensure that nomore than 24 hours elapsed betweenharvest and processing and the purchase ofan additional coal-fired burner to raisedrying temperature, reduce drying time,and improve the quality of the dried chips.

With these improvements the plant wasable to produce chips and flour for humanconsumption that satisfied qualitystandards. It obtained a license for foodproduct manufacture and began sellingflour to the local food industry.

This experience underscores theimportance of the pilot phase. Without thisstep the project would not have identifiedand solved at an early stage problems(especially those relating to fresh rootsupply) that did not appear in theexperimental plant at CIAT.

The cassava drying project, for instance, wasnot phased out through a conscious decision. Thisjust happened as the agroindustry created by theproject spread far beyond the original targetregion, making project activities essentiallyirrelevant. The private sector became increasinglyinvolved in cassava drying, and governmentreduced its support of rural cooperativedevelopment.

The fresh cassava project suffered a similarfate. It lost support from the government's ruraldevelopment agency because of changes inpersonnel and institutional mandate. At the sametime, CIAT decided to terminate its work on thisproduct as an outcome of budget cuts andstrategic planning. The project thus came to ahalt at the end of the pilot stage. Two years laterbusinesses set up on a pilot basis for storing freshcassava are still operating at a profit. By notproceeding to the commercial phase, we may havemissed a valuable opportunity.

Institutional support to the Union deAsociaciones de Productores y Procesadores deYuca (UAPPY) in Ecuador, is currently beingphased out in an orderly fashion over a two-yearperiod. In preparation the union is building itsown capacity to develop new projects in directcontact with international donors rather thanthrough institutional intermediaries.

Root and tuber products under developmentin the Philippines are all in the pilot phase,except sweet potato ketchup, which is beingmanufactured successfully by private enterprises.Production on a commercial scale could spread tonew locations with further support fromresearchers at ViSCA.

In India the potato processing enterprisesnow underway require further institutionalsupport to guarantee their long-time profitabilityand diversify their range of products.

In Peru processed potato products are nowavailable. A project aimed at expanding themarket would be helpful.

Based on the limited experience we have sofar, it seems clear that to phase out a projectsuccessfully two conditions must be met. First,the enterprise must have a product with a goodprofit margin. And second, it must be under soundmanagement.

A good indicator of the strength of anagroindustry is its ability to develop newproducts. Cooperatives in Colombia, for example,are now commercializing maize, producing animalfeed rations, and so forth. If the enterprise needssupport (for example, in applied research on cropproduction or process development), it shouldcontract other organizations to prOVide thisservice.

Future Prospects

In the 1990s two general trends have a directbearing on the prospects of small-scale, ruralagroindustries for developing products based onroot and tuber crops. Markets are graduallybecoming less distorted, and countries areopening up their economies to external trade andstrongly encouraging exports.

Add,,,,, Value to Root and Thber Crops

Agriculture is probably the sector with themost distortions. This is largely the result ofheavy subsidies maintained by developedeconomies in Europe and North America and ofthe negative effects of politically motivated foodaid on agricultural production in the developingworld. It seems likely that developed countrieswill make some progress over the next decadetoward reducing subsidies and allowing freemarket conditions to prevail in agriculture.

Production of root and tuber crops indeveloping countries can succeed if it provideslow-cost, high-quality, locally available rawmaterials for rapidly developing food, feed, andother industries. These crops will be usedprimarily in starch- and flour-based products!serving mainly as ingredients of animal feed), inprocessed foods, and for a wide range of otherindustrial purposes.

The central premises of this manual are thatroots and tubers can perform this role and thatsmall-scale rural processing (at least formanufacturing intermediate products, such asstarch and flour) is a viable way for small-scalefarmers to get a share of the benefits from valueadded products. If these are sound assumptions,root and tuber crops could contribute importantlyto sustained economic development.

In order for these crops tll contribute';lgllificantly, though, it is vital that governmentsreduce or eliminate subsidies on imported,ndustrial raw materials, that rural people receivetraIning in both the technical and businessaspects of operating processing plants, and thateffective managers be selecte(l fiJT productdevelopment projects.

The key advantage of TOC:" and tubers is thattlwy an' low-cost sources of cdrbohydrate. To.:apnaJize on this advantage, rural people must;,dd value to these crops through small-scale;J!'\ ,cessing of products whose q uaJity and price are. JlTij.Jarable to those of cereals ff this happens·'juJtable rural development cliuld become a'ahty in the disadvantaged areas where roots

!I,d tubers are usually grown

Part II

Case 1


Objective: 'Ib raise the incomes of small-scalecassava farmers and landless laborers byintroducing a new cassava product-dried, wholeroot chips-for sale to producers of animal feedconcentrate.

Project area: The departments of Sucre andCordoba on the Atlantic coast; the idea laterspread to other departments on the coast and inother regions of the country.

Time frame: Pilot phase, 1981-1984; commercialphase, 1985-1989; phase-out started in 1990. Theproject drew on research conducted by CIAT inthe mid-1970s.

Background

In the 1970s the country's Programa deDesarrollo Rural Integrado (DRO tried to improvethe welfare of small-scale farmers in this regionby providing credit for cassava production.Although production increased, demand in thefresh market remained constant, leading to aprice decline and widespread default on loans.

In an effort to diversify cassava markets,DRI joined CIAT and various national institutionsto mount a pilot project in which cooperativeswould produce dried cassava chips on a smallscale, using natural-drying technology from

119

Thailand. The idea was that this domestic productwould compete with imported sorghum, which theColombian government was purchasing insignificant amounts and selling at support pricesabove those in the world market. Economicstudies suggested the project's strategy wasfeasible, assuming that farmers could reduceproduction costs and increase yields.

Project EvolutionDRI organized small-scale farmers intocommunity co-ops or associations, andinterinstitutional teams coordinated by DRI gavethe farmer groups integrated support in technicalmatters and credit. CIAT focused particularly oncassava drying.

The first pilot plant was set up with fundsfrom the Canadian International DevelopmentAgency (CIDA). Other plants soon followed(mostly In Sucre and Cordoba); by the mid-1980sthere were nearly 40. The number of plantsincreased again in the late 1980s and early 1990sbeeause of continued support to farmer groupsand Increased interest among entrepreneursoutside the project. By 1992 more than 150 plantswere producing over 25,000 t of dried cassavachips

Technical and other challenges

The plants sold their product to the country'sJllaJor animal feed companies, which incorporatedthe chips to varying degrees into balanced feedsfOI cattle, SWllle, poultry, and other livestock.

Although generally successful, thismurketing approach did present problems for bothprocpssors and their customers. Because thefanner groups catered mainly to industriesaccustomed to paying for goods 30-90 days afterdelJvery, they had difficulty maintaining workingcupltal The feed companies, on the other hand,often complained about inadequate chip quality, aresult of poor drying, but detected no aflatoxins.

The farmer groups also faced challenges inpJant operations. For example, scarcity of cassavafresh roots sometimes forced them to look outsidethe group for new supplies. And this reduced theirprocessing efficiency, particularly in thecommercial phase.

Since the cost of raw material accounted for74% of processing costs, including transportation,the project paid close attention to its links withcassava production. It also improved the chipperdeveloped in Thailand, increasing its capacityfrom 3 to 12 tlh. The project proved economicallyviable with a cassava yield of 8 tlha and aconversion rate (for fresh to dried cassava) of2.5:1. Delays between harvesting and processingand extended drying time had a negative effect onquality of the final product. Inadequately driedcassava chips (i.e.., exceeding a maximummoisture content of 12% on a wet basis) wererejected by the feed industry.

In addition to technical support, the farmergroups received extensive training in all aspectsof production, processing, commercialization, andadministration. Some members needed literacyclasses. Training in plant administration,especially financial management, was notsuccessful. This and marketing required skillsand communication channels that the farmergroups lacked.

Institutional support proved costly. In itsfirst year, the project received 220 days of supportfrom various groups; the requirement dropped toabout half that amount after a few years. Farmerscommonly complained that technical support waspoor and inopportune, largely because they hadno control over the technical teams.

Spreading the benefits

Despite its difficulties, the project generatedsignificant benefits. To spread these as widely aspossible, planners opted to replicate the farmergroups and processing plants rather than expandexisting ones.

When enough groups had been formed, theproject encouraged them to establish a secondorder organization, which would handle productmarketing (particularly negotiation of prices with

l~O


animal feed companies), channel technicalassistance and other services to farmer groups,and protect farmers' interests in the politicalarena. The Asociaci6n Nacional de Productores yProcesadores de Yuca (ANPPY) was establishedfor these purposes in 1986.

By 1992, however, the organization had splitinto regional factions. Between 1990 and 1992,private owners of drying plants took control ofANPPY, prompting the co-ops to form rivalsecond-order organizations, the Asociaci6n deCooperativas de la Costa (ASOCOSTA) andFederaci6n de Cooperativas de la Sabana(FEDECOSABANA).

Producers earned income from processing byvarious means: 1) selling raw material directly tothe plant; 2) working in processing or itsadministration; 3) profits from the plant or otherbenefits of group membership, such as access tocredit and training; and 4) income paid to familyor contracted labor for producing raw material.Early results showed that small-scale farmersreceived a large share of the income fromprocessing. Some also went to landless laborers.

The rate of return from processing wasinitially very favorable (more than 70% in 1990)but has declined in recent years because ofincreased competition from cheap imported maizeand sorghum. Even so, in 1993 the price of driedcassava still exceeded production costs.

As a consequence of the project, cassavademand increased, farmers adopted improvedproduction technology, and yields rose. TheInstituto Colombiano Agropecuario (lCA) releasedtwo new varieties. But farmers resisted the ideaof differential prices for roots based on dry mattercontent. And this has made it difficult to improvethe processing plants' efficiency.

The loss of institutional support

CIAT terminated its involvement in the project in19H9. By 1992, DRI had also reduced its role inthe work, as the government switched to a morefn~e-rnarketapproach in agriculture, abandonedsupport prices, and restructured and reduced theSize of public institutions. The interinstitutionaltechnical teams were disbanded. Lines of credit

Sources

Best, R; Sarria, H.; and Ospina, B. 1991. Establishingthe dry-cassava industry on the Atlantic coast ofColombia. In: Perez-Crespo, C.A. (ed.). Integratedcassava projects, Chapter 8. Working Documentno. 78. CIAT, Cali, Colombia. p. 112-127.

Brekelbaum, T. (comp.). 1991. Secado natural de yucaen la Costa Norte de Colombia. Cuaderno deAgroindustria Rural Doc-Esp-6. CentroLatinoamericano de Tecnologia y Educaci6nRural, Cali, Colombia, and Institutolnteramericano de Cooperaci6n para laAgricultura, San Jose, Costa Rica. 57 p.

Case 1: Dried Cassava for Animal Feed in Colombw

for small-scale farmers were reduced, and accessto them made more difficult.

Once the project lost its institutionalframework, technical support collapsed. Perhapsthis would not have happened if the service hadbeen established within a second-orderorganization from the beginning, as was done inEcuador (see Case 4). Nonetheless, the groupscontinued to function, though often at reducedproduction levels and narrower margins. Somegroups still receive institutional support throughthe new second-order organizations.

Future directions

Under the free-market economic policies thegovernment has pursued since 1990, farmersmust rapidly improve cassava productivity butwithout degrading the land. Farmer groups alsoneed to diversify the market for this crop byfinding products with higher value added, such ascassava flour for food and industrial uses (seeCase 3). These two steps are essential fortransforming cassava into an agroindustrial cropthat benefits small-scale farmers and processors.

The success of this project prompted DRI andother institutions to launch additional projects.Their aim was to extend drying technology tofarmer groups in other regions of the country andtest other products, such as fresh, storablecassava for the fresh market and cassava flour forhuman consumption (see Cases 2 and 3).

Lessons Learned

• When production is dispersed among manyfarmer groups or cooperatives, an effectivesecond-order organization is essential for

12l

•

•

increasing their bargaining power andconcentrating product supplies.

Farmer cooperatives seek to maximizebenefits to their members by maintaining ahigh margin for crop production. The privatesector, in contrast, aims to maximize theprofits from processing. Where a secondorder organization includes both co-ops andprivate sector processors, it must reconcilethe different interests of these groups toavoid conflict. If the organization maintainsa balance between groups, the private sectorcan spur the co-ops to achieve greaterefficiency.

Distance to the market is critical because ofthe high cost of transporting such a bulkyproduct. Increasing the ratio of weight tovolume by reducing the size of the driedproduct is important. When other cash cropsare being harvested and trucking lines areoperating at full capacity, cassava chips maynot be collected on time, giving rise to cashnow problems for the cooperatives.

AddlTlM 'value to Root and Thber Crops

Case 2

Conserved Fresh Cassava for HumanConsumption in Colombia

Objective: To demonstrate the technical andeconomic feasibility of new cassava storagetechnology on a pilot basis and study its impact.

Project area: Primarily the city of Barranquilla(population 1.2 million) and surrounding areason the Atlantic coast.

Time frame: Pilot phase, 1987-1988; commercialphase, 1989-1992.

Background

Consumption of fresh cassava is declining inmany urban areas of Latin America. Among thebiggest losers are small-scale farmers, whoseldom have alternative markets for this highlyperishable crop. With no way to process andstore the fresh product, they are forced to sellimmediately at low prices.

In search of alternatives, this projectexamined ways to slow postharvest deteriorationof fresh cassava, improve its quality, and reduceits price to consumers. The most promisingoption for preserving fresh roots was to treatthem with an antimicrobial agent and storethem in polyethylene bags.

In 1985 a pilot project \part of a DRI projectin Bucaramanga, Santander del Sur)demonstrated the viability of the new storagetechnology. Consumers liked the bagged cassavaand bought it from retailers m the main marketand shopkeepers in several neighborhoods.Farmer groups earned net profits of just overUS$10/t

Project Evolution

A new project was set up in Barranquilla for threereasons: 1) small-scale farmers on the Atlanticcoast were already developing associations fordrying cassava; 2) this is the country's mostimportant cassava producing region; and 3) it hasthe highest rate of cassava consumption percapita (54.3 kg/year vs. 25.5 kg nationally).

Institutional partners

Several co-ops took part in the project. One wasthe 82-member Cooperativa de Producci6n yMercadeo de Repel6n (COOPROMERCAR), some90 km from Barranquilla, which was growing andmarketing tomatoes, plantain, and cassava.Having gotten low tomato Yields and prices forseveral years, the co-op was interested in thetechnolo6'Y for conserving fresh cassava. Anotherparticipant, Cooperativa Agroindustrial del NorOnente del Atlantico (COOAGRONOR), wasalready drying cassava chips and began sellingbagged cassava to a supermarket chain. Severalothl'r co-ops got involved in the project as well.

DRI coordinated the activities of institutions:-iupporting the farmer groups. ClAT providedtl'chnical assistance and train mg. The ServicioNacional de Aprendizaje (SENA) gave training inco-op organization, administration, andbookkeeping. lCA developed Improved varieties

Case 2: Conserved Fresh Cassava for Hilman Consampl",n '" Colombia

and technology packages. Statistics on priceswere obtained from the Central de Cooperativasde la Reforma Agraria (CECORA). TheCorporaci6n Fondo de Apoyo a EmpresasAsociativas (CORFAS) provided the co-ops withtechnical assistance, along with credit forproduction, commercialization, and capitalinvestment. Other organizations, including theCaja Agraria, Instituto de Financiamiento yDesarrollo Cooperativo de Colombia(FINANCIACOOP), and Instituto Colombiano dela Reforma Agraria (lNCORA), also madeavailable credit for production and processing.

Striving for a quality product

The success of the cassava preservationtechnology depends to a large extent on rootquality and shape. Farmers must take care notto include damaged roots, since these can bestored safely for only 4-5 days. Initially,operators removed the stem from the roots witha machete. But since even experienced workersoften damaged the roots, they switched to agarden pruner for this purpose.

Early on, large sacks of roots wereimmersed in a barrel containing a solution ofMertect (a fungicide), followed by a 15- to3D-minute wait to remove excess moisture. Thiswas replaced by a faster method, in which thesolution was sprayed with a backpack sprayerdirectly into the polyethylene bags of cassava. Asmall hole was cut in the corner of the bags toallow excess solution to drain out. This methodused less solution and required that the roots bepacked only once. Costs fell from US$12 to $4/t.

Storage conditions proved critical.Operators learned the importance of aircirculation around the bags to prevent lossesfrom microbial growth and internal rotting.

Despite these measures, product qualityvaries significantly, even among roots harvestedfrom the same field. In addition, rain and poorlydrained soils reduce the starch content ofcassava roots, an issue that requires muchfurther research.

Another difficulty is the limited correlationbetween chemical composition of cassava roots

and easily measurable traits, except for bitternessand total root cyanogen content. Roots from tall,unstressed plants tended to have better chemicaland organoleptic qualities than those fromshorter, stressed plants. Since root quality canvary with soil and climate, roots must be tastetested to ensure consistent quality. In some casesup to 20% of roots had to be discarded; normallythe rejection rate should not exceed 5%. The ratewas particularly high during the rainy season.

The prohibitive cost ofaggressivemarketing

Market and consumer tests documented the maincharacteristics of the fresh cassava market inBarranquilla. The city consumes about 32,000 t ofcassava yearly. In the central marketplace,cassava is purchased both for wholesale andretail. Consumers buy the crop mostly from smallshops (numbering 5,000 and accounting for 65%of the volume), supermarkets, and street vendors.Retail prices are 50% higher in urban than ruralareas. Although marketing margins are extremelyhigh (more than twice the farm-gate price), theprofits go mainly to retailers, who buy and sellsmall quantities, rather than to intermediaries.

In consumer testing of potential brandnames, the preferred name was one suggestingthe product's freshness (Yucafreska). But sincethis name was already registered, the projectsettled for Superyuca. This and the marketingcampaibJTl slogan-calidad par laargo rata(suggesting "laasting quality")-were bothregistered in the name of the ANPPY A bag wasdesigned and posters printed for point-of-salepromotion in supermarkets.

Lack of funds prohibited a more ambitiousmarketing campaign aimed at creating generalawareness of the new product and its advantagesamong consumers and traders. Even more modestmeasures (such as replacing the plainpolyethylene bag with more sophisticatedpackaging printed in two colors) would haveadded 40(fr-80(fr) to the current production cost of$O.Ol/kg. These steps would have been moreaffordable If the project had marketed largervolumes of product. But it was unable to do soIJl'cause of limited financing.

A fragile distribution system

The Federaci6n de Organizaciones

Agropecuarias de Colombia (FAGROCOL) set up

a central collection point and undertook urban

distribution of cassava. In 4 months it sold 288 t

of fresh, untreated cassava in polypropylene

bags; 125 t of treated cassava (Superyuca) in 3-,

15-, and 50-kg bags; and 15 t of chopped waste

roots for animal feed.

'Ib facilitate treatment of roots at the

collection center, they were packed in

polypropylene sacks and shipped the next day

for treatment with the antimicrobial solution

and repacking in polyethylene bags for retail

sale. In addition, cut roots were protected with

sodium disulphate and calcium carbonate

powder, which reduced the quantity of rejected

roots.

Noting that FAGROCOL was able to move

72 t of cassava each month, other wholesalers

adopted its approach to commercializing

untreated cassava in polypropylene bags. This

enabled wholesalers to store the crop in their

warehouses for 1 or 2 days without losses.

Granabastos, the central collection center for

public sector wholesalers in Barranquilla,

became interested in the storage technology as

well.

The co-ops ran into serious economic

trouble when FAGROCOL was liquidated as a

result of circumstances unrelated to the project.

The lack of a good distribution network and a

centrally located warehouse that caters to

shopkeepers has made it difficult to expand sales

beyond 10 tJweek, even though the total market

volume is about 800 tJweek.

Benefits for all concerned

Economic analyses of this technology in the pilot

phase showed that it benefited all concerned.

Producers received slightly more for their roots,

and participating co-ops earned net profits

ranging from US$13.50 to $72.75/t. The lower

figure is for COOPROMERCAR, a commercial

enterprise that pays administrative costs of

1:24


nearly $0.02/kg of treated roots, and the higher

is for a co-op that catered to the export market.

Consumers paid slightly more ($0.03/kg) for

conserved cassava but received a more

convenient, higher quality product. As the

volumes of sales increase, supermarkets will be

able to reduce their margins, particularly since

product losses from deterioration are minimal.

Margins should be further reduced, as continued

spread of the conservation technology gives rise

to competition. Narrower margins will benefit

consumers.

The technology has already been adopted in

other regions of the country. For example, at

Socorro in the department of Santander, SENA

has trained various co-ops, and they are treating

roots to provide a product called Yucarica for the

Bogota market. They are also drying cassava for

feed concentrate plants in Bucaramanga and

Bogota. The success of the new technology has

prompted farmers to seek improved production

technology and varieties with high starch

content and desirable eating qualities.

Future directions

Since the co-ops are currently satisfying only

18% of potential demand, they have much room

for expansion. DRI should take the lead in

supporting and financing the efforts of co-ops

and small processing businesses to meet the

demand of current and new markets for bagged

cassava. These processors urgently require

financing for capital investment, working

capital, and technical assistance.

A simple, low-cost system for monitoring

and evaluating processing operations in the

coastal region and eventually nationwide is

needed so that decisions can be based on precise,

current information. CIAT designed and

installed a computerized system (Sysyuca)

that covers production, processing,

commercialization, and socioeconomic indicators

of benefits. Implemented originally for a cassava

drying project, this system has been operated by

CORFAS and SENA since the Center's

withdrawal from that project.

Case 2: Conserved Fresh Cassava for Human Consumptwn in Colombia

Lessons Learned

• In handling and marketing fresh producewhere presentation and quality are critical,the co-ops must find alternative uses (such asdrying or starch extraction) for rejectedroots, which can account for up to 50% of thetotal weight of the harvest.

• Package design is very important forconsumer products.

• It is considerably more difficult to penetratea consumer market with a new product thanan industrial market like that described inCase 1. Since the costs of product promotionand distribution tend to be high, the projectshould first target specific market niches.Experience suggests that only large-scalesecond-order organizations with some degreeof economic power and know-how shouldattempt to enter consumer markets.

• As the co-ops expand their processingactivities, they need to identify alternativesources of roots during the rainy season andmaintain records of dry matter and starchcontent, as a means of ensuring acceptableproduct quality.

125

Source

Wheatley, C. and Izquierdo, D. 199] Case study: Freshcassava storage. In: Perez-Crespo, C.A. (ed.).Integrated cassava projects, Chapter 5. WorkingDocument no. 78. CIAT, Cali, Colombia. p. 62-92.

Add",!! Value to Root and Tuber Crops

Case 3


Objectives: In research, to determine thetechnical and economic requirements ofdeveloping a rural cassava flour industry; in thepilot phase, to integrate cassava production,processing, and marketing under typicalsocioeconomic conditions; and in the commercialphase, to prepare for expansion by improving theprofitability of the pilot plant.

Project area: Chinu, department of C6rdoba, onColombia's Atlantic coast.

Time frame: Research, 1985-1988; pilot phase,1989-1991; commercial phase, 1992-1994.

Background

In the last 40 years, rapid urbanization in LatinAmerica has changed people's dietary habits.Starchy staples, such as maize, plantain, androot crops, have given way to more convenientfoods, such as rice and processed products ofwheat (e.g., bread and pasta). Researchers areseeking alternative markets for the traditionalstaples, so that rural people can continue to .derive a livelihood from these crops.

One promising option is to use cassava flouras a partial substitute for wheat in bread, pasta,and other foods. Three factors favor the use ofthis flour: 1) it is relatively cheap, costing 15%20% less than wheat; 2) in some food products, itcan be substituted at medium to high levels forother types of flour; and 3) it has functionaladvantages over wheat flour in some foods,absorbing more water and giving bettercrispness. Cassava flour is especiallyadvantageous in processed meats, biscuits,cones, spices, pastry (for empanadas), andmixtures for breading and frying. The potentialdemand for cassava flour among industrial andsmall-scale food processors in Colombia isestimated at 40,000 tJyear.

lLh

Project Evolution

In the past, large-scale cassava flour plants havefailed, because they were poorly linked withproduction and often dependent on expensivefossil fuels. An alternative strategy is to establishsmall plants in important cassava-growing ruralareas. 'Ibward this end, the project used fundingfrom the International Development ResearchCentre (lDRC) to design a system for producingand marketing cassava flour


CIAT conducted research on cassava production,carried out prefeasibility and feasibility studies,and designed and developed processingequipment. The Instituto de InvestigacionesTecnol6gicas (lIT) in Bogota developed bakeryproducts and performed economic studies. TheUniversidad del Valle at Cali designed anddeveloped the processing plant and devisedformulas for products based on cassava flour. Incooperation with ICA, on-farm trials wereconducted to identify varieties for intercroppingand second-semester planting (which wouldensure a continuous supply of cassava forprocessing) and to find ways of increasing theproductivity of current cropping systems.

During the pilot phase, DRI coordinated theintegration of cassava production and flourprocessing and marketing into its program ofcassava development on the Atlantic coast. Sincethe end of the pilot phase, the Natural ResourcesInstitute (NRI) in the UK has supported researchaimed at Improving flour quality, focusingparticularly on the use of varieties with highcyanogen content that are commonly grown inBrazil and Africa.

Case 3: Cassava Flour for Human Consumption In Columbia

Developing prototype equipment andtesting the product

The processing plant and equipment weredeveloped according to a batch or semibatchmodular design, with a capacity of 1 t of cassavaflour per day. Flour yields from dry cassava chipsranged from 83% in a laboratory version to 98% inan industrial mill. Research showed that duringmilling fiber and lignified bark in the chips areseparated from the flour by sifting. Resultssuggested that milling the chips in a roller millwould eliminate the need for peeling roots. Theproject developed a simple roller mechanism thatcould handle chips small enough to allowcontinuous feeding.

The prototype plant was designed for bothnatural and artificial drying of cassava chips. Thechips are dried naturally for one day on inclinedmesh-bottomed trays positioned on racks. Thenthey are loaded into a bin dryer, which has anindirect coal-fired burner, is coupled to a 5-hpcentrifugal fan, and heats the air to 60 8C.

In an evaluation of composite flour producedwith this equipment, bakers complained about thehandling properties of the dough. In breadmaking the main drawback was loss of volume inthe final product. To avoid this problem requiressubstantial adjustments in the bread recipe. Suchchanges are unnecessary with products, such asbiscuits, cakes, and crackers, whose quality is lessdependent on the use of wheat flour.

The project also conducted a blind test ofbread made from wheat flour, compared withbread made from a composite flour, with 200families representing different social strata inBogota. The panelists detected differences inappearance, aroma, and freshness. More than80% liked the bread made from composite flourand 15% preferred it.

The pilot phase

After an economic evaluation of research resultsindicated that production of cassava flour isfeasible, a pilot plant was established at Chinu,Cordoba. This site was chosen because land andraw material are available throughout the year,there is potential for increasing cassava yields,

127

infrastructure is adequate, markets are nearby,and institutional support is available. Based onsubsequent experience, the project added severalitems to its criteria for site selection, includingthe educational level of co-op members, theexecuting organization's entrepreneurial ability,proximity to fuel sources and machine repairshops, availability of transportation, and rawmaterial quality.

The project began working with a farmer coop, Cooperativa de Productores de Algarrobos(COOPROALGA), which operates a plant thatproduces chips for animal feed. The co-op's 41members are relatively well educated. Afterseveral trials, adjustments were made in theprocess, and personnel selected by the co-op'sadministrative council received training.

The plant soon ran into problems with itswater supply. The project had planned to obtainwater from a well it constructed fiJr two nearbyvillages. But the mayor refused to go along,maintaining that the well's capacity wasinadequate. Consequently, the project had to buildnew wells 2 km away. And since it had notbudgeted for extra pipe and a pump, the start ofprocessing was delayed. The project's financialinvestment in the pilot plant, excluding the cost ofobtaining a water supply, amounted to US$48,179($44,389 for infrastructure and processingequipment and the rest for working capital).

The pilot plant had the capacity to produce200 t of cassava flour per year. In 1991 it operatedfor 9 months, but output was low 143 t of chips,with a conversion factor of 2.92) because ofequipment breakdowns, energy cuts, andinsufficient working capital. The chips weremilled at a wheat mill in Medellin, with a flourextraction rate of 87%.

The financial rate of return or FRR (seeexplanation on page 88) to the pilot plant wascalculated to determine its profitability. Giventhat the opportunity cost of capital in Colombia iscurrently 22(k" the FRR would have to be a11ll111nlUm 01'30% to cover annuallOflation of27%.Based on data for 1992, the FRR was only 19%.Further analysis showed that it is very sensitiveto utilization of plant capacity, price of rawmaterial, the conversion factor, coal consumption,

and sale price. On this basis the project developeda strategy to raise the FRR to 31%.

During the pilot phase, problems withproduct quality emerged that had not beenevident in testing of prototype equipment. Onefactor that lowers quality is the high moisturecontent of the roots during the rainy season.Others are the high incidence of bacteriosis,fungal infection, and termite attack. Some rootsare contaminated with fecal coliforms under thepoor sanitary conditions of small farms. As aresult, the quality of the cassava flour did notmeet industry's microbiological standards for foodproducts and had to be sold to manufacturers ofanimal feed.

At the end of the pilot phase, the project hadtwo options: to sell the chips to millers orsubcontract the milling and sell cassava flour. Itchose the latter and in promoting this product,the project concentrated on Medellin, the largestand most promising market. The wheat flour millthat milled cassava chips expressed interest inpromoting cassava flour among its clients.

The commercial phase

'Ib increase COOPROALGA's participation il1:plant operations, the project put the cooperative'smanager in charge of the two processing plants,and plant operators' positions were stabilized.

In 1992 the project concentrated onimproving the microbiological quality of itsproduct. Sanitary conditions were raised bytreating washing water with sodium hypochlorite.disinfecting roots, cleaning the equipment morethoroughly, enforcing high standards of hygieneamong workers, and establishing specificationsfor raw material quality. 'Ib control the physicaland biochemical quality of raw material, operatorswere encouraged to wash the roots adequately,adjust the distance between premilling rollers,and periodically measure the moisture content ofthe chips. In addition, chips were dried at highertemperature and for a longer time. Variousmodifications were made to reduce drying time.Another burner was added and the drying areaexpanded to improve air flow. As a result of t,hesemeasures, the cassava flour now conforms tostandards for wheat flour.

Adding Value to Root arul Tuber Crops

Given the high cost of milling in Medellin, asmall-scale mill and flour classifying system wasdeveloped at the Universidad del Valle andincorporated into the process. The cooperativenow sells flour directly to buyers. The projectestablished a flexible price scheme for cassavaflour, in which the price varied from 15% to 20%below that of wheat flour, according to thelocation and type of client (e.g., a bakery versus aprocessing firm). A price list was prepared withdiscounts for the initial purchase (5%), paymentin cash (1.5%), and volume purchases.

Future directions

During much of the project's lifetime, cassavaflour could be produced at a competitive price72% that of wheat flour, assuming a 25% profitmargin for both the cassava grower and processor.The future of this product depends very much ongovernment policy, attitudes in the private sector,and trends in production.

The government's current emphasis on tradeliberalization policy has severely damaged theability of cassava flour to compete with importedwheat flour, whose price has declined in realterms over the last 3 years. Unfavorable policiesare reinforced by vested interests in some sectorsof the economy. As the volume and price ofimported wheat decline, cassava growers willneed to adopt improved production technology tokeep cassava flour competItive.

In an effort to deal with these issues, theproject established a consultative body withrepresentatives from public and privateorganizations having direct or indirect interestsin wheat, cassava, and bread. In general, publicagencies responsible for rural developmentsupported the project, while the private sectorwas noticeably cautious. Industrialists continue tolargely ignore the nonfood uses of cassava flour,making it difficult to expand the market for thisproduct

Although the future of cassava flour isuncertain, there are some positive signs. Forexample, the Colombian government has recentlydeveloped a comprehensive plan to promotecassava. Moreover, COOPROALGA is takingvarious steps to improve its services to the local

Case 3: Cassava Flour for Human Consumption in Colombia

community. Specifically, it seeks to gain greatercommunity acceptance, stimulate cassavaproduction, and help meet local needs for water,housing, and education.

'Ib manage the processing operation on theirown, staff of COOPROALGA need furthertraining. For this purpose the project developedtwo reference manuals for the managers ofcassava flour plants. Another challenge is for theplants to obtain the technical capacity to performfinal milling of cassava flour.

Projects like this one have been launched inBolivia, Brazil, Ecuador, and Peru. In addition,the NRI is studying the market potential forcassava flour in selected Mrican countries. A newresearch project will identify flour propertiesrelated to end-product quality and novel productscontaining cassava flour.

Lessons Learned

• Varieties differ significantly in dry mattercontent. Venezolana, for example, has highdry matter content, which makes it wellsuited for processing into flour. ManihoicaP-12, in contrast, has a lower dry mattercontent. Although farmers tend to plantseveral varieties in the same plot, they favorP-12. The project paid a premium price forhigh quality roots to give growers anincentive to plant varieties with high drymatter content.

The functional properties of cassava flourconfer desirable characteristics on these endproducts.

• During the pilot phase, a n umber oftechnical problems arose that were notevident from testing of the prototypeprocessing equipment. This highlights theimportance of having a pilot phase beforeinvesting heavily in commercial developmentof the product.

• Competitiveness of cassava flour is stronglyinfluenced by raw material price. Researchon crop production, by increasingproductivity and lowering costs, caneventually complement projects onpostharvest processing and marketing.

• To penetrate a market that IS accustomed toconventional raw materials with a newproduct, like cassava flour, requires athorough understanding of thephysicochemical and functional properties ofthe new product. This information isfundamental in promoting the productamong clients.

Sources

CIAT; Instituto de Investigaciones Tecnol6gicas; andUniversidad del Valle. 1988. The production anduse of cassava flour for human consumption.Final report, research phase. Working Documentno 66. CIA'l~ Cali, Colombia. 85 p.

• Correctly identifying the most promisingmarket for a product is critical to its success.During the research phase, this project.considered bakers of bread its main market.When they expressed concern about cassavaflour's effect on bread quality, the projectshifted its focus in the pilot phase to otherfoods, such as processed meats and biscuits.

129

ClAT; Universidad del Valle; and Programa deDesarrollo Rural Integrado. 1992. Producci6n ycomercializaci6n de harina de yuca para consumohumano. In: Ostertag, C. and Wheatley, C. (ed.)lnforme final, fase de proyecto piloto. Volumen 1,Aspectos generales y resumenes CIAT, Cali,Colombia. 32 p.

Adding villae to Root and Tuber Crops

Case 4


Project Evolution

The project in Ecuador was based to a largeextent on experience in Colombia but with greateremphasis on building a strong second-orderorganization to provide key services to cassavaprocessors. The challenge of this project was torepeat the technical success of the Colombianprojects but at a lower cost 111 terms of1I1stitutional support.


PartiCIpating institutions contributed as follows:

Objective: To provide farmers with alternativemarkets by researching and promotingprocessing of cassava flour from chips, cassavastarch for human consumption and industrialuse, and preserved fresh cassava and byimproving interinstitutional cooperation, withthe aim of expanding integrated cassavaprojects.

Project area: Manabf province in coastalEcuador.

Time frame: Pilot phase, 1985-1987;commercial phase, 1988 to present.

Background

During the mid-1980s, cassava production in theproject area declined, mainly because farmershad difficulty disposing of extra production.Major barriers were the perishability of cassavaroots, limited markets, and low prices. In theproject area, there was considerable potential forinvolving small-scale cassava farmers in productdevelopment, because community developmentprojects in Bijahual-Alajuela and Bellavista hadalready organized cassava growers intoassociations.

i .j()

•

•

•

•

•

Ministry of Agriculture and Livestock: In thepilot phase provided three staff, one tocoordinate the project, another to serve on itsadvisory committee, and the third to overseetraining and technology transfer and supportthe Asociaciones de Productores yProcesadores de Yuca (APPYs) inorganization and extension.

Fundaci6n Ecuatonana de InvestigacionesAgropecuarias (FUNDAGRO): Providesadministrative and technical assistance to asecond-order organizatIon, the Union deAPPYs (UAPPY),

lnstituto Nacional de InvestigacionesAgropecuarias (lNIAP J: Supports on-farmresearch. As a result of the project, INIAPrevived its research on tropical roots andtubers.

Community development projects: Provideadvisors to farmer associations at Joboncilloand Bijahual.

ClAT: Provided an anthropologist to helpfarmers and rural institutions establish asmall-scale cassava industry as well asagronomists and economists to give technicalsupport and training. The Center also

Case 4: Diversifying Cassava Markets in ECllacJ"r

arranged for a Colombian farmer to trainEcuadorian farmers in plant constructionand operation and donated two chippers.

• British and Canadian embassies: Initiallyfunded the APPYs' demonstration projects.

• US Agency for International Development(USAID): During 1985-1988 provided theAPPYs' main source of grant funding andloans and paid the salary of the CIATanthropologist through FUNDAGRO.

• IDRC: Funded maize and cassava researchand monitoring of drying plants.

Farmer-to-farmer training

The project has put as much as possible of theresponsibility for supporting project activities inthe hands of the farmer organizations engaged incassava processing. The objectives are to reducethe cost and increase the effectiveness of supportprovided by the above-mentioned institutions andfoster independence among project beneficiariesfrom the start.

One effective tactic is farmer-to-farmertraining and technology transfer, which CIAT hasguided and fostered. Early on, Colombian farmersvisited counterparts in Manabi to teach them howto dry cassava. Then, growers from Manabivisited Colombia to see the drying technology inaction and learn how to organize and manage aprocessing plant. A Colombian farmer leaderhelped design and build plants in Manabi. AtBijahual and Jaboncillo, members of farmergroups that predate the project experimentedwith the new technology and formed additionalgroups to apply it and market the product.

Building a strong second-orderorganization

The number of cassava processing operations orAPPYs grew quickly from 2 to 16. The APPYs buyfresh cassava from members and other producers,chip the fresh roots, dry the chips, put them insacks, and sell the product to UAPPY. Thisorganization evolved from a marketing unionestablished by the first APPYs. It was assumedthat UAPPY could support as many as 20 local

chipping associations (each with 15-20 members)within a radius of about 70 km. Roughly 30% ofUAPPY members are women, concentrated infour APPYs consisting entirely of women.

The local APPYs and the regional UAPPYshare market risk and responsibilities. The latterprovides portable milling equipment andmarketing services to the APPYs in exchange for30(k of the markup between fresh cassava andflour (about 10% of the gross price of the flour).The remaining 70% covers the APPYs' operationalexpenses, capitalization, and distribution ofprofits to members. In addition to milling andmarketing, the UAPPY manages loans anddonations, provides credit to the APPY's forprocessing, exercises product quality control, andhandles accounting and transportation.

Until 1993, UAPPY's administrator was afarmer and founding member of one of the firstAPPYs. He was assisted by representatives fromCIAT and FUNDAGRO, who together formed amanagement and planning committee, whichorganized training for the UAPPY's assembly andboard and involved them in planning.

By making good use of talented peoplewithin the organization, UAPPY has remainedsomewhat independent of supporting institutions.This has enabled it to negotiate with them on amore equal footing and obtain the servicesmembers require.

Even so, as the organizations that helpedstart the processing operation withdraw support,UAPPY faces major challenges. For example, itmay prove difficult to find and keep goodmanagers. It is therefore critical that the board ofdlrectors of the processing operation learn now toappreciate a good manager. Otherwise, they willtend to strictly limit his or her authority andsalary. Turnover is generally high amongmanagers who are also farmers. Another problemonce institutional support comes to an end is thatnew APPYs are less likely to obtain grants forbuilding processing plants and have to obtainluans at interest rates close to market rates.

The UAPPY assembly, which consists ofAPPY presidents, needs to provide training for itsthree commissions (responsible for membereducation, marketing, and auditing). In addition,

association members need training in financialmanagement.

In addition, UAPPY should contract anaccounting firm to improve its inventory system,depreciation schedule, and system of records formembers. The Union also needs advice onminimizing taxes and maximizing memberbenefits.

An ambitious sales campaign

In its pilot phase, the project tested naturaldrying of cassava; evaluated cassava starch as abinder for shrimp feed, cattle feed consisting ofdried cassava and chicken manure, and use ofcassava silage for swine feed; and carried outtrials to identify varieties with high dry mattercontent and drought tolerance. Based on theresults, the project adopted a strategy in itscommercial phase of promoting the use of cassavaflour through an ambitious sales campaign aimedat the animal feed industry.

The first attempt focused on manufacturersof chicken feed in the Portoviejo area Thisinitiative was not as successful as comparableefforts in Colombia, because the industry wasreluctant to try a new ingredient, even at a lowprice.

The second attempt, aimed at shrimp feedmanufacturers in Guayaquil, was moresuccessful. Ecuador's shrimp industry wasseeking an alternative to formaldehyde to holdfeed pellets together under water. US healthauthorities had banned the importation of shrimpraised on feed containing that ingredient. Cassavaflour was an ideal substitute, because its highstarch content makes it an excellent binder. Eventhough cassava comprises only 2%-14% of eachpellet, the scale of shrimp production and feeddemand was such that the demand for cassavaflour exceeded 8,000 tlyear. The shrimp feedindustry paid a higher price for this flour, becauseof its agglutinant characteristics, than didproducers of poultry and livestock feed.

Toward product diversification

From an early stage, UAPPY strongly encouragedproduct diversification. In 1987, for example, two

1


APPYs experimented with preparing, packaging,and selling fresh cassava for export (see Case 2).At about the same time, two new APPYs, made upentirely of women, were established to producecassava starch for human consumption. Theirmain clients were stores and food processors. In1989, UAPPY sold industrial-quality starch to alarge cardboard factory in Guayaquil, whichsubstituted it for maize starch in glue forcorrugated boxes. A 1989 study estimated highpotential demand for industry-quality cassavastarch, which is easier to produce than starch forhuman consumption.

Meanwhile, UAPPY had developed a millingcapacity to produce cassava flour from chips forthe shrimp feed industry. In 1988 several factoriescomplained that cassava flour made from wholeunpeeled roots contained too much ash. Inresponse UAPPY began selling flour made fromwhole or peeled roots according to the clients'[weds.

At first APPY members resisted peeling,hL'cause they were not accustomed to this practiceand lt involved much additional labor. Even so,peeling soon became an important source ofadditional income for the famities of members andnonmembers. They could earn as much as amonth's minimum wage in 2 weeks by peelingcassava in their spare time. In the 1990-1991processing season, UAPPY pmd more thanUS$16,000 in wages for cassava peeling. Most ofthe money went to poor women, children, and theL-ldl~r1y. For that reason, the UAPPY is nowreluctant to introduce mechanized peeling.

A valuable lesson in the need fordiversification

Despite its early moves toward diversification,UAPPY still catered primarily to a single market,the shrimp feed industry. In 1989 this began tochangl'. Because of competition from Asiansh rimp producers and a shortage of larvae tostock ponds, there was a slump in local feedpruduction, which halted the purchase of cassava!lour. To make matters worse, many APPYs hadburrowed heavily to expand their drying capacity.

GAPPY responded by slashing expenses andlaunching an all-out campaign to sell cassava

Case 4: Diversifying Cassava Markets in Ecuador

flour to other industries. As a result of thesemeasures and a recovery in the shrimp industry,the union was able to sell all production instorage by 1990. Although its economic balancefor the year was poor, DAPPY had learned avaluable lesson. From then on it assigned highpriority to diversifying the market for currentproducts and building the capacity to producenew ones.

In 1990, for example, DAPPY began refiningwhole-root cassava flour by passing it through amechanical vibrating sifter. The product can serveas a partial substitute for wheat flour. It was soldto factories that use it as a filler for resins inmanufacturing plywood. Bran, the byproduct ofsifting, is sold as a source of fiber to livestock feedindustries in the highlands. DAPPY also uses themechanical sifter to produce a white flour formaking noodles. Although the union sells only asmall amount of flour for human consumption, itis testing various methods, with a view toexpanding production. As a result of suchmeasures, DAPPY reduced its dependence on theshrimp industry by about 30% (Table 23).

Spreading and increasing the benefits

The demand for cassava flour and starch grewrapidly. DAPPY's sales quadrupled in 1988 andagain the next year, with annual profits of 27(k.

Average payments to individual farmers increasedfrom DS$100 in 1985 to more than $300 in 1988.

Since some members provide more cassavathan others, distributing the association'sproceeds equally among members may not beacceptable. Larger producers are in effect

subsidizing the smaller ones and may insist onproportional payment. By-laws must be specificwith respect to management of members' capital,payment to former members and heirs, conditionsfor membership, etc.

Families engaged in cassava processing havea strong incentive to increase cassava areas. Ifthey increase the area from 0.5 to 2.0 ha (as manyAPPY members have done since 1985), theirincome may grow by 10% to 100%, depending ontheir land holdings, other crops, managementability, and the weather. The opportunity costs ofthis increase are low, since pastures are the mostcommon alternative use of cassava land andfamily labor is a key component of increasingcosts

Future directions

UAPPY has dearly demonstrated an ability toplace cassava processing on a firm commercialfooting. It is therefore not surprising thatindustrial interests are now asking how they can"create a DAPPY" to supply them with cassavaproducts

To continue generating benefits for membersand nonmembers, DAPPY must expand itsmarkets, diversify its array of products, andimprove product quality. Processing research cansupport this efTort, particularly by developingmethods for quality control and monitoringmarket demand. Researchers must also addressthe problem of contamination caused by wastewater from starch processing. Research oncassava production should concentrate onvarieties that are suitable for dry zones and

Table 23. Market share ofannual processed cassava production for 1989-1990 and 1990-1991.

Final product UAPPY products Tht.al t.ons sold Percentage of totalamount sold

Shrimp feeds

Cardboard boxes

Plywood

Traditional cassava starch bread

Pastas and noodles

I Cattle and swine feeds

\ Thtal"'------------

White and whole industrial flour

Industrial starch

Refined whole industrial flour

Food starch

Refined whit.e food flour

St.arch bagasse and flour bran

133

191'l9-90

974.4

20a

1,015.0

1990-91

1,207.2

209.1

226.5

5.2

69

871

1,740.0

191'l9-90

96.0

2.0

100.0

19~~11'69.3

12.0

13.0 I0.3 i

I0.4 I5.0 !

mixed cropping on hillsides and have high drymatter content, which makes them more suitablefor processing.

•

Addln/i Value to Root and Thber Crops

A strong second-order organization canreduce the cost of institutional support forcassava processing and foster self-relianceamong project beneficiaries.

Lessons Learned

• Farmer-to-farmer training is an effectiveapproach to technical assistance. Colombianfarmers were able to explain anddemonstrate processing technology toEcuadorian farmers in language they couldunderstand. The project trained a team ofoutstanding Ecuadorian farmers anddeveloped a manual to promote thetechnology among new groups.

• Diverse markets and products are criticalrequirements for long-term commercialsuccess in cassava processing.

Source

CIAT (Centro Internacional de Agricultura Tropical).1992. Ecuador integrated cassava project. In:Cassava Program, 1987-1991, Chapter 5. WorkingDocument no. 116. CIAT, Cali, Colombia.p. 273-295.

Case 5


Objective: Th improve the welfare of ruralcommunities in major cassava production areasby applying a model for integrated production,processing, and commercialization of driedcassava chips for use as animal feed and bystrengthening farmer organizations throughparticipatory management.

Project area: Ceara state in Northeast Brazil.

Time frame: Pilot phase, 1989-1992.

Background

Brazil is the world's second largest producer ofcassava, contributing 16% of the world total,and accounts for 75% of production in LatmAmerica.

The crop is particularly important 111

Northeast Brazil, where a yearly average of113,000 ha were planted to cassava during1985-1987. Most of the region's production IS

used for farinha de mandioca (a toasted flour).Some is consumed fresh, and minor amountsare fed to animals. The crop is a major source ofcalories in the human diet, providing 27% oftotal caloric intake.

In 1990 the total population of NortheastBrazil was estimated at 43 million, 42(!f) ofwhom live in rural areas. This region has thecountry's highest levels of poverty andunemployment. Land distribution is skewedtoward large-scale growers. Even thoughholdings of less than 10 ha make up nearly halfthe total number of farms, they occupy onlyabout 4% of the total cultivated area.

There is considerable potential forexpanding the use of cassava as animal feed.Since Brazil is an important beef producer andthe world's third biggest producer of poultry,the market for feed rations is large. Once amaize exporter, Brazil must now import3.5 million tons of this commodity annually to

135

nled demand. The Northeast in particular has alarge deficit of maize and animal feed.

Project Evolution

The pnlject was funded by the KelloggFuundation and executed jointly by CIAT and theCeara Secretaria de Agricultura e ReformaAgraria (SEARAJ, which acted through twoagl~nCtes, the Empresa de Assistencia Tecnica eExttmsao Rural do Ceara (EMATERCE) andEllipresa de Pesquisa Agropecuaria do Ceara(EPAeEI. Their work built on the experience of 11farml!l' groups organized in the 1980s for cassavapn1lessing. The project also relied heavily on theCeara State Cassava Committee, which was setu~: In 1988 to coordinate work on cassava stateWldl' and consists mainly of representatives fromEMATEHCE and EPACE.

Levels ofinstitutional support

As a result of good initial results from this project,thl! State Cassava Committee gained considerablerecognition and credibility. This helped it identifysources of financial support for project activitiesand establish useful contacts with state andnatlonal development agencies. The Committeewas made a permanent member of the nationwideCamara Sectorial de Mandioca (Cassava SectorChamber), which represents producers,agromdustrialists, consumers, and governmentorganizations and whose main purpose is torecommend policies to the Ministry of Agriculture.

Coordination of project activities was placedIT1 t he hands of Regional Cassava Committees, ofwhich five were established by 1992, with twomore planned. They are composed ofrepresentatives from the main agencies executingthe project as well as farmer groups. The RegionalCommittees enabled the project to rapidlydecentralize its activities and helped improvecOIlllllunication among extension officers.

Farmers get organized

The project concentrated on organizing farmergroups around agroindustries based on cassavadrying. It reactivated the 11 groups establishedpreviously and set up another 135, with the totalnumber of farmers involved exceeding 3,000.Nearly 60% of group members are smallholders(of whom 28% live in land reform settlements),29% lease the land, and 13% are sharecroppers.More than half the farmers plant less than 1 ha tocassava; only 15% plant more than 2 ha.

Each group obtained a grant from localagencies to finance the installation of dryingfacilities, provide credit for production andprocessing, distribute planting material, and soforth. Few members availed themselves of credit,though, because the high inflation rate made itrisky for them.

Farmers were trained in production,processing, and commercialization as well ascommunity organization through more than 100training events (such as field trips and courses),involving nearly 850 technicians and well over2,000 farmers. Since more than half of groupmembers are illiterate, the Kellogg Foundationfunded a separate project aimed at increasingliteracy.

In addition, the project stimulated thecreation of three regional second-orderassociations to improve the farmer groups'bargaining power and participation in projectplanning.

Processing and its benefits

The farmer groups readily adapted to the newtechnology for processing dry cassava because ofits simplicity and their ample experience inproducing farinha de mandioca. From 1989 to1992, the groups processed 7,094 t of fresh rootsinto 2,677 t of dry chips. The proportion of rootsgoing to farinha production fell from 65(}b in 1989to 38% in 1992. Even so, group members'consumption of this product increased from 4.6 to6.2 kglhousehold per week, probably because ofthe extra income farmers generated by sellingroots for dry cassava production.

Adding Value 10 Rool and TUber Crops

The project was successful in opening andconsolidating an alternative market for cassavaproduction. In 3 years a total of 975 clientspurchased dry cassava; 93% of them bought lowvolumes (less than 5 tJyear), accounting for 32% oftotal output. Contrary to experience in othercountries, less than 5% of the consumerspurchased large volumes of product, whichrepresented 59% of the total.

The financial success of new cassavaagroindustries will depend largely on the dynamicrelationship (with respect to prices, costs, etc.)between the two main outlets for cassavafarinha and dry chips. The project helped farmergroups build the capacity to choose the best option10 a given year.

Farmer groups benefited by gaining a newmarket for cassava roots, additional employmentopportunities, and a share of the profits fromcassava processing. Of the total income earned,59(!c went to smallholders, 32% to renters, and 9%to sharecroppers. More than 70% of the incomewent to farmers who planted only 1-2 ha.

From 1989 to 1992, earnings generated forfarmers through processing totaled US$163,689.Of this amount 37% came from cassava sales, 10%from wages for processing, and 53% from farmers'share of annual profits. Smallholders received59% of the earnings, renters 32%, andsharecroppers only 9%; 77% went to growersplanting no more than 2 ha.

The project tried to attract credit sources inthe public and private sectors. But it was unabletu do so, largely because of the country's difficulteconomic conditions. Even so, the project grew asa result of strong support from local agencies,which provided farmers with grants to establishagroindustries. Growth would perhaps have been"tronger, except that local agencies responsible forImplementing the project are undergoing radicalreorganization.

Limited adoption ofimprovedproduction technology

To improve the efficiency of cassava production,the project sought solutions to two major

Case 5: An Integrated Cassava Project in Brazil

constraints, low adoption of improvedtechnology and a lack of good planting material.

It addressed the first problem byestablishing "preproduction trials" under farmermanagement. The improved technology showeda decided yield advantage over traditionalpractice. But it remains to be seen whetherfarmers will invest scarce resources in fertilizerand weed control.

Since small-scale farmers apply littlefertilizer on cassava, they can maintain soilfertility only by extending the fallow period.Although organic manure is available in someareas, farmers cannot afford it. They urgentlyneed new options for maintaining andincreasing soil fertility (e.g., mulches and greenmanures) to raise production and preserve theagricultural resource base.

To deal with the scarcity of good plantingmaterial, the project established communal plotsfor cassava propagation, but this approach wasnot successful.

Monitoring and evaluation

A major activity of the project was to establish asystem for monitoring and evaluating itsactivities. This involved collection of baselinedata, annual surveys, and intensive monitoringof a small subsample of participating farmers.These data were collected by managers of thecassava processing groups and extension agentsunder the coordination of the Regional CassavaCommittees. The information was then analyzedby the State Committee and the resultsdistributed through monthly and annualreports.

The baseline data (on cassava productionpotential, farmer characteristics, etc.) helpedidentify areas for project expansion. Monthlyreports on the performance of farmer groups inprocessing proved useful for planning projectactivities and calculating benefits and theirdistribution. Annual reports kept cooperators.donors, and decision makers abreast of projectactivities.

137

Future directions

Since the demand for dry cassava as a partialsubstitute for cereals in the production of animalfeed concentrates far exceeds the supply, there ismuch potential for expanding project activities toother regions of the country. It is encouragingthat the organizational structure of the project(particularly the State and Regional Committees)was widely accepted and worked effectively.Another positive sign is the success of the farmergroups in creating employment opportunities,opening an alternative market, stimulating localindustry, and raising incomes,

Even so, to consolidate these benefits,important constraints must be addressed. It isessential, for example, that the organizationalstructure adopted under the project be made alegal part of the local institutional framework. ItIS also critical that researchers find ways to speedthe adoption of improved technology, partly as ameans of reducing costs. Whether farmer groupscan secure their hold on the new market for drycassava will depend largely on their ability toprovide sufficient quantities of product atcom peti tive prices.

Lessons Learned

• The formation of State and Regional CassavaCommittees proved to be an effective meansof coordinating project activities. The StateCommittee helped the project establish closetips with the country's agricultural policymakers. This was an essential step forensuring that small-scale cassava farmersare represented.

• Although local markets can absorb relativelysmall volumes, they offer good opportunitiesfor commercializing a new product. Pricesare often higher, and quality requirementsJess strict than in large-scale industrialmarkets

• It takes time to turn cassava farmers intosmall-scale agroindustrial entrepreneurs.Improvement of literacy and participatorytl'chniques are essential for success.

• Interaction of project personnel withsimilar projects in other countries (seeCases 1 and 4) contributed importantly tohuman resource development andmotivation.

13~


Source

CIAT and Ceara State Cassava Committee. 1992.Cassava development: Pilot project Ceara,Northeast Brazil. Integrated production,processing and commercialization of dry cassavachIps for animal consumption. Final Report.CIAT, Cali, Colombia. 55 p.

Case 6


Objective: 'Ib give small-scale farmers analternative for disposing of surplus potatoproduction by introducing a simple dryingprocess and by developing the infrastructureneeded to ensure regular supplies of rawmaterial as well as technical assistance,financing, and help with marketing.

Project area: Potato-growing areas in theBareilly district of western Uttar Pradesh.

'lime frame: Development of prototypeequipment and product, late 1970s;establishment of pilot processing unit, 1985;expansion in number of processing units, 1986to present.

Background

Uttar Pradesh produces an average of6.4 million tons of potato annually. About everysecond year, the market is glutted at harvest,and prices are low. For that reason and becausecold storage space is scarce and expensive,farmers leave some fields unharvested. Even inyears when there is no glut, only 70% of theharvest can be stored until prices are morefavorable.

139

Project Evolution

Various organizations joined forces to providegrowers with the option of drying potato forurban markets.


Compatible Technology, Inc. (CTD, a USvoluntary agency dedicated to helping the poorestablish viable rural enterprises, offered avehicle and funds to set up processing units andcover operating costs. The Society forDevelopment of Appropriate Technology(SOTEC) established a demonstration processingunit, along with simple storage facilities, andpurchased 1 ha of land to develop the ResearchTraining and Village Development (RTVD)Centre. CIP funded development of the storagefacilities, equipment, and recipes for preparingfoods from the dried potato products.Appropriate Technology International (AT!) inthe USA provided funds to develop, test, andcommercialize procedures for potato storage andprocessing. The Nonconventional EnergyDevelopment Agency of Uttar Pradesh made agrant to develop a solar drier.

Project organization

A noteworthy feature of the project is its threetier organizational structure.

The first level consists of village dryingunits. After a prototype unit was established in1985, another 21 were set up during 1986-1990.Training workshops were organized for staff ofvoluntary and government agencies toencourage them to establish the additionalprocessing units in rural areas. About half ofthose units were still operating at the end of1989.

It took time to determine the optimum sizeand output for drying units. At first the project'sgoal was for each unit to have 10 workers,processing 60 t of fresh potato in 60 working days.But this proved impossible except underunusually favorable conditions. Normally, 15people can process 40 t in 60-70 working days. By1990 village units were carrying out all of theprocessing and selling their product for 25%-30%less than some automated drying plants.

Since the village processing units are toosmall to handle their own marketing and lackstorage capacity for large quantities of finishedproduct, the project placed responsibility for thesefunctions at two additional organizational levels.The second takes care of quality control (throughsorting), grinding, packaging, storing, anddespatching orders; while the third establishesproduction targets and handles sales, marketing,invoicing, and collection of payments for thesecond tier. In addition, the third tier mayeventually seek loans to purchase the product, sothat the lower tiers can receive payments morequickly. Initially, SOTEC was involved at all threelevels. But by 1988 it had turned over 95% of thesales to two independent companies.

Since the drying units have chronic cashflow problems, they often cannot deliver a fulltruckload of dried product. This, together with thelow product density, contributes to hightransportation costs. To deal with this problem,SOTEC collected the product from severalneighboring units and took it to the RTVD Centrefor inspection, repacking, storage, and dispatch.

SOTEC must comply with a variety ofgovernment regulations, including food andstorage licensing requirements, sales tax andpackaging laws, and labor laws controllingminimum wage, permanency of employment, andnumber of employees. Wages are high,particularly during wheat harvest.

To keep labor costs and taxes low, SOTEChas tried to keep the units small. This also helpsthem avoid competition with large-scalecommercial operations. If village-level processingwere to become conspicuously successful, groupswith large amounts of capital might try to takeover. The project has therefore emphasized the

140

Adding \-hlue to Root and 'LUber Crops

use of family labor and simple machinery, withlow overhead.

Most low-income rural people in the projectarea have limited formal education and considerthemselves capable only of menial tasks. To helpovercome this limitation, SOTEC trains workersof the village drying units and visits themregularly to ensure that processing is goingsmoothly.

Searching for a marketable product

The project's original idea was to promote driedpotato slices as a rehydrated vegetable. But itsoon realized that consumers traditionally usechunks and do not like the flat slices. Driedchunks more than 7 mm thick give poor results interms of rehydration and texture.

As part of an effort to create interest in driedpotato slices, the project organized a contest toencourage the development of recipes using thisproduct. In addition, staff of the Nave TechnicalInstitute prepared 50 dishes from the dried slices.In all but one case, though, the slices had to berehydrated, cooked, and mashed, making themjust as time-consuming to prepare as freshpotatoes.

Subsequently, the project tried grindingdried slices into a coarse powder. This provedeasier to use and improved the recipes. But sinceconsumers were unfamiliar with potato powder,the project was unable to introduce this productfor household use. Where SOTEC conductedproduct demonstrations, initial sales were good,but consumers continued buying the productonly where SOTEC had personal contacts.Retailers wanted attractive, high-qualitypackaging. But there was little money for this,and it would have increased the price of theproduct.

The project was more successful in sellingpotato powder for reprocessing into extrudedsnacks. The introduction of extrusion cookinggenerated interest in potato-based snacks. Thepowder (finer than 60 mesh) and granules (40mesh) produced by the village processing unitsproved satisfactory for this purpose and lessexpensive than factory-produced powder.

Case 6: Drying Potato in India's Villages

In conjunction with the search for a viableproduct, research was conducted on appropriateequipment for the drying units. The project firstexamined implements available in the kitchens ofordinary homes. But to increase productivity andmake processing economical, it was necessary todevelop sturdier equipment in consultation withlocal blacksmiths and carpenters.

Local industries, in contrast, proveduncooperative in sharing technical know-how. Asa result, it took 2 years to develop a technique forapplying abrasive grit to the surface of the potatopeeler. Two experienced technicians made finaladjustments in the equipment. Even so, theproject continued improving the design of newimplements to reduce their cost, which rose as aresult of increased taxes on metals. Local artisansmanufactured the equipment, sometimes withgrants from SOTEC.

Several challenges rem ain to be met. Forexample, small rural industries often do not havethe capital to purchase equipment, and spareparts are hard to come by. Moreover, sincecustomers require products of differing particlesizes, processors need a proper screen mill. Amotorized sieving machine would reduce laborcosts and losses through dust.

Drying and storage

The project learned several useful lessons aboutdrying and storage.

It found, for example, that white-skinnedpotatoes dry better than red-skinned ones andthat certain varieties are more suitable for drying.It is also important to select appropriate tubers(large, uniform in shape, with shallow eyes, andfree of damage and disease) and then cure themby removing the aerial part and leaving thetubers in the ground for 1-2 weeks.

Sun-drying can be done during only4 months of the year. If farmers do not haveenough tubers for processing, they have to buythem just after the main harvest, when prices arelow, and store them. Potatoes can be stored for up10-12 weeks and still give good dried products.The rate of recovery of dried chips or strips mayreach 18% but depends on the variety and tuber

quality. Some starch is recovered from thewashing tubs (5-10 kg from the average 800 kg ofpotato processed daily), but it cannot be producedeconomically in this manner.

Since dried slices take up almost as muchspace as fresh potatoes, those that are to be soldin powder form should be ground at once.Discolored chips should be removed and thosewith visible insect damage or fungal growth usedfor animal feed. 'Ib prevent chips from absorbingmoisture, they must be packed in sisal-type sackslined with plastic.

The economics ofpotato drying

The project closely studied the economics ofvillage potato drying.

To construct a cement drying floor, simplestores, and a water tank and obtain equipment(including a washer/peeler, slicer, and dryingracks) requires an initial investment of US$3,850.The units can be economically viable if theyoperate for a minimum of 60 days/year. Assuminga product recovery rate of 18%, the dried productcan be produced at a cost of $0.47/kg. If a unitprocesses 60 t in 60 days, it can repay a capitalloan (at 12% interest) in 4 years and a loan foroperating costs in 4 months, assuming theproduct is sold at $O.71/kg. But since most unitsprocess 800 kg/day, they would have to operate75 days and sell the chips at $0.88/kg. Recoveredstarch earns another $88.25/year.

The sales record of the processing units isencouraging. By 1990 demand for the processedproducts was so great that 90% of the productionwas sold as soon as it was available. Approximatesales figures are as follows:

F:d~C'Amount sold (t)

1987 1988 1989"

Chips 0.2 2.0 3.5

Strips 0.2 1.5 3.3

Powder 2.0 6.0 6.0

Starch 0.4 1.2

Thtal 2.4 9.9 13.7

a. March-August unly.

141

A more disturbing element in the economicpicture of the drying units is credit. As a result ofproduction problems, most units have had toreschedule their payments on loans forequipment. SQTEC has not been able to chargeinterest on these loans. Payments are made whenthe product is delivered to the RTVD Centre.

There are clearly major obstacles toextending credit for rural processing. The peopleinvolved have no business experience and lack theresources to survive without immediate paymentfor the goods and services they provide. Yet,customers who purchase in bulk commonly insiston 30-45 days' credit and then further delaypayment. Under these circumstances, credit forestablishing and running a processing unit ismore likely to enslave than liberate villagers.

Future directions

For the period 1989-1991, SOTEC received fundsfrom ATI to develop, field test, and establishcommercial operations for sorting, processing,drying, and grinding potato. During the first twoyears, a demonstration project built several potatostores, established three processing units, and setup a milling unit.

Depending on the outcome of thedemonstration project, an expansion phase wasplanned to set up six more processing units withsupporting facilities. This phase was also toinclude an economic evaluation of the processingand marketing of potato products. The resultsshould indicate alternatives to the original projectdesign that facilitate diffusion of the processingtechnology and adoption of particularly attractivecomponents (e.g., storage).

142

Adding \blue to Root and 'fuber Crops

Lessons Learned

• The development, as opposed to transfer, ofvillage-level technology takes time. Eventhough the basic principles of potatoprocessing are well known, it still takesseveral years of applied research to applythis knowledge at specific locations.

• Problems in management and motivation arejust as important, if not more so, as purelytechnical constraints. Whereas the latter canbe solved with temporary outside support,the former require on-the-job training andcoordination with public and privateorganizations over an extended period oftime.

• Small-scale processing can compete withlarge-scale operations, because the formerpay lower overhead and taxes and have thenecessary flexibility to extend or reduce theperiod of plant operations.

• Quality control is critical in marketing toindustrial clients. In this case qualitystandards are met by undertaking grindingand sieving at a centralized location andchecking products from the village units formoisture and foreign matter.

Source

Nave, R. and Scott, G.J. 1991. Village-level potatoprocessing in developing countries: A case studyof the SOTEC project in India. CIP, Lima, Peru.40 p.

Case 7


Objective: '1b produce nutritious processedproducts for low-income consumers and raise theincomes of small-scale potato farmers throughincreased sales and reduced production costs.

Project area: The Mantaro Valley aroundHuancayo, Peru.

Time frame: Research on processing andmarketing, 1977-1984; semicommercial andcommercial production of dried potato products,1984-1990.

Background

Potato production covers 10% of the cultivatedarea in Peru; only maize, covering 15%, is moreextensive. Most of the country's potato productionis concentrated in the central sierra. Because ofthe wide range of agroclimatic zones in this area,the crop can be harvested throughout the year.Potato production became more market orientedas the country improved its highway network andexpanded intra- and interregional trade. There isgrowing interest in processing potato to reduce

143

transportation costs and thus give farmers largerprofit margins, to supply urban markets moreregularly, and to store the product when marketsare saturated, until prices increase. Peru's coastalcities offer a potential market for inexpensive butgood quality potato products.

For centuries small-scale farmers in thesouthern sierra have used solar energy to producethree products: 1) chuno, which is used to preparesoups and mazamorra, a maize-based pap,2) dried potatoes, and 3) starch. Traditionalprocessing could be improved to give productsthat better meet the requirements of urbanmarkets in terms of color, consistency, and purity.

Since the 1980s potato area and productionhave fallen dramatically. Average yield has risenfrom 6.5 to 8.3 tlha, but this seems insignificantgiven that potential yield in the highlands is30-50 tlha. About 90% of farmers plant less than1 ha to potato, mainly for on-farm consumption;9% plant up to 3 ha, mostly for the market; and1% are commercial growers who plant as much as100 ha for seed production and urban markets.

Project Evolution

This project drew upon considerable previousresearch on potato processing. One group, forexample, field tested an improved solardehydrator, which consisted of a black wooden boxwith a removable plastic top. They hoped it wouldreduce the time required to dehydrate potatoeswhile preserving their nutritional and culinaryquality, but this method did not give better resultsthan traditional processing. One disadvantage ofthe black box was that it lacked ventilation,trapping moisture inside; removing the plastic didnot eliminate this problem.

The Universidad Nacional Agraria supporteda commercial-scale drying plant in Muquiyauyo,but it never functioned as planned because of thehigh price of tubers and limited market for driedpotato. Moreover, administration and technical

assistance proved more demanding than expected,so finally the plant was abandoned. Scientistsrealized the importance of developing newprocessed products that would be less subject tofluctuations in the price of potato.

A nutritious product from local crops

Researchers at CIP focused on identifyingcombinations of local crops that could beprocessed into low-cost products of highnutritional value. Since low-income consumersmake up a large share of the market, it waspossible to lower costs by processing a largervolume of product. To explore the possibilities,CIP staff experimented with different cropcombinations, constructed a pilot processingplant, and carried out taste trials, marketingstudies, and an economic feasibility study.

Various crops grown by small-scale farmers(including quinoa, Andean lupine, faba beans,oats, and barley) were evaluated for theiracceptability and compatibility with potato.Prototype mixtures were produced and evaluatedaccording to their organoleptic qualities. The mostacceptable were then analyzed for theirnutritional value. The best product contained 30%dried potato mixed with rice, faba bean, oat,barley, and maize flours. Adding 1 L of water to80 g of the mixture and then cooking for25 minutes gives a thick porridge; the product canalso be used as a thickener. It has 10.6% proteinand 333 kcal of energy per 100 g of mixture. Itsprotein efficiency is 86%, compared to 82% forpotato, 70% for rice, and 41% for faba bean alone.

A pilot plant, built near Huancayo, could dry1-2.5 t of mixture weekly. After being cooked andmashed, moist potatoes were mixed with flour ofthe cereals and legumes.

The mixture, with a moisture content of 52%,was spread on wooden trays on shelves in thedrying chamber (5 kg/m2

). A hundred trayscontaining 500 kg produced 250 kg of driedmixture in 48 hours. A ton of product gave 50,000portions (250 g each), which when rehydrated,provide enough food for 396 families of six to havethree meals a day for one week.

The drying chamber included a solarcollector made of eucalyptus wood and adobe

144

Adding 'l-nlue to Root and Thber Crops

brick. The walls and floor of the drying floor werepainted black to absorb more sunlight. The otherwalls had six windows 50 cm from the floor topermit air circulation; a fan extracted saturatedair.

Too expensive for the poor

Demonstrations were organized to promote theproduct. A thousand potential consumers sampledit and were given half a kilogram to take home.The product was also served for a year atcommunity kitchens in low-incomeneighborhoods. In both cases the product was wellreceived. In addition, taste trials were conductedin 12 pueblos j6venes (squatters' settlements onthe outskirts of Lima), at another communitykitchen, and in 2,000 schools across the centralsierra. The product's flavor and consistency werefound acceptable.

In calculating the economic feasibility of theproduct, it was assumed that the processingoperation would require an investment ofUS$15,000 (not including land), with an annualinterest rate of 12.5%. The product would be soldfor $1 and cost about $0.60 to process, dependingon the scale of production. A study conducted inthe pueblos j6venes determined that processedproducts like this one are too expensive for thepoor. But scientists working on the processdismissed these findings as too pessimistic.

Trouble from the start

The technology's developers decided to help a localNGO, the Centro de Investigaci6n,Documentaci6n, Asesoramiento y Servicios(IDEAS), prepare a project proposal and identify apotential donor. The project was to consist of fivephases: assimilation, development, investment,operation, and replication/impact. The originalplan called for a pilot plant at Huancayo andthree more in Puno, Cajamarca, and Piura,requiring a total investment of $295,000 in fixedassets and $10,000 in working capital. Thisestimate included the cost of land as well asconstruction, a hammer mill, and industrialpotato press. Analyses suggested that theoperation would be profitable under highlyfavorable conditions but show a net loss undermore realistic circumstances.

Case 7: Processing Dried Potato in Peru

The project was implemented by Industria deDerivados del Agro S.A. (IDEAGRO) with a loanfrom Centro IDEAS. Payments on the loan wereto be deposited in a rotating fund that wouldenable other firms to repeat the processelsewhere. Even though the financial viability ofthe project was questionable, it was viewed as anexperiment from which valuable lessons could bedrawn.

An additional drying chamber increased theprocessing plant's capacity to 9.6 t of product perweek. Its staff consisted of an industrial engineeror food technician, an administrator, head,secretary/sales assistant, and eight laborers. Theoperation was weak in administration andaccounting, had no sales force, and made noarrangements to ensure that it met legalrequirements in such matters as plantconstruction, municipal and health licensing, andregistration of a trademark.

Marketing studies were conducted in twouniversity cafeterias and three communitykitchens. Although the product was generally wellreceived, some people complained that it "smellsbad, like oil," has a "rancid or bitter taste,""doesn't thicken," and "yields less than it'ssupposed to."

A change in strategy

After 6 months the project was clearly in trouble.The product didn't satisfy consumer tastes andwas more expensive than some wheat products.'Ib remedy these problems, IDEAGRO changedthe project's strategy. It launched a search fornew, better quality products that would appeal tourban consumers and, for financial reasons,permit a production increase from 9.6 to16 t/week.

Oats were eliminated from the mix becausethey were not readily available, and peas weresubstituted for faba beans to eliminate the bittertaste. Cooking time was reduced from 25 to15 minutes.

In searching for new products, IDEAGROmade profitability its main criterion. Consumersfelt that two aspects of the original productneeded improvement: its texture and taste. To

145

accomplish this, project staff used precookedpotato flour and left out the rice, which was toocoarse. They also developed several new products,including Chicolac (4% potato, 70% maize, milk,and cocoa), a potato cream and semolina, andlegume and cereal lines. The project purchasednew equipment as well.

The price of Chicolac was higher than that ofcomparable products in Lima's wholesale andmunicipal markets but considered intermediatecompared to prices in supermarkets. The projectregistered its products under the brand nameAbril and promoted them by means of flyersdistributed in supermarkets. Chicolac won a firstprize at the Huancayo fair, where it waspresented to committees responsible fordistributing milk to children.

Sales of the products were low, apparentlybecause housewives lacked information onpreparing dishes with them. 'Ib remedy thisproblem, the project distributed recipes insupermarkets. The potential monthly market inLima was estimated at 113 t of dried potato and7 t of potato flakes. Nonetheless, average monthlysales in 1988 amounted to only 8.2 t, and only50% of the pilot plant's processing capacity wasused.

Working under difficultcircumstances

The country's deteriorating economic situation inthe late 1980s made it difficult for IDEAGRO towork effectively. Gross national product dropped9% between 1986 and 1989, plunging 26% in 1988alone. High inflation distorted prices, andgovernment policies (such as reduced barriers tofood imports, soft credit, fertilizer subsidies) alongwith drought destabilized the prices of key projectinputs.

Problems arose in managing the projectbecause of fundamental differences betweenIDEAS and IDEAGRO about project strategy. Itproved difficult for staff to make decisions thatsatisfied both organizations.

Three factors eroded the potentialprofitability of the operation. First, money waslost through unpaid credit. Second, the project

Source

Wong, D.; Alvares, M.; Scott, G.J.; and Yupanqui, A.T.1991. Papas, mezclas y cremas: Un estudio decaso del procesamiento de papa en el Peru. CIPand Centro de Investigaci6n, Documentaci6n,Asesoramiento y Servicios, Lima, Peru. 30 p.

had to pay workers even when processing washalted for adjustments in the product. And third,the prices ofAbril products did not increase at thesame rate as those for similar products. The firm'sability to pay its loan in the short term wasrestricted by two factors. First, capitalinvestments absorbed much of the firm'sresources (75% versus an ideal 25%), leaving toolittle working capital. Second, working capitalwas immobilized because clients delayed paymentfor purchased product by up to 2 months.

Future directions

Several agroindustrial enterprises have replicatedthe product of this project as well as some of themachinery and marketing strategies, includingproduct presentation and points of purchase.Upon seeing the effects of competition from abetter quality product, wholesalers began toimplement quality control and segment themarket according to sales expectations. There ispotential for small-scale farmers to replicate theproduct and process, provided they have access toprocessing equipment and technical assistance.

Lessons Learned

• The experience of this project illustrates thevital importance of information about marketcharacteristics, such as raw material supplyand price, quality requirements for processed

146

•

•

•


products, and potential demand. Trying toenter three markets at once was overlyambitious, and assigning highest priority tothe low-income market was a mistake.Projects of this sort should concentrate firston the product with the most potential forprofit to get the firm well established.

This project underscores the need forcontinual modification of processes andproducts in response to changing marketconditions. Small-scale commercialenterprises require technical support frominstitutions with the appropriate expertise.

Availability and cost of raw material are keyconsiderations. Purchasing raw materialfrom wholesalers rather than producersproved less costly and more predictable.

Estimates of cost and cash flow must berealistic. The high cost of credit to clients,which limited growth, should have beentaken into account in estimating the amountof working capital the project would need.

Case 8


Objective: 'Ib create new marketing alternativesfor potato producers and generate ruralemployment by developing an industry thatproduces solar dried chips, cubes, and flour.

Project area: Areas of North Santander andNariiio departments where potato is grown on asmall scale.

Time frame: 1987-1991.

Background

From 1970 to 1988, annual potato production inColombia expanded from just under 1 million toabout 2.5 million tons. Researchers at ICA wereconcerned about the effect of such a large increaseon potato prices and producers' incomes. A sharpdrop in the price during 1984-1985 prompted adecision to explore alternative potato productsand markets more actively. Other good reasons forthis initiative are the seasonal abundance,bulkiness, and perishability of potato, whichgreatly complicate marketing.

A Colombian technician was sent to Peru fortraining in simple potato processing. By 1986 asmall pilot plant had been set up to determine thefeasibility of producing solar dried chips, cubes,and flour. Colombian potato farmers, unlike theircounterparts in Bolivia and Peru, where simplepotato processing is centuries old, are unfamiliarwith these techniques.

After successful trials in a pilot plant,marketing studies were conducted. One objectivewas to gauge the interest of farmers in producing,selling, and consuming potato products. Anotherwas to get feedback from urban consumers onproduct attributes as a basis for improvingprocessing methods.

Colombian policy makers and potatoresearchers had long been interested in thedevelopment of new processed products, such aspotato chips, French fries, flour, and starch. Somesaw these products as a way to expand potato

147

exports, while others were more concerned abouttheir prospects in domestic markets. The expertsagreed that to diversify potato products andmarkets successfully would require a betterunderstanding of market demands and ruralurban links.

Project Evolution

From 1987 to 1991, ICA undertook a project withtechnical support from CIP, aimed at developingmarkets for processed potato products. A largeshare of the operating costs were paid by IDRC aspart of its support for a regional potato researchnetwork in the Andean zone. The goals of thenetwork were to address high priority problems inmember countries, train scientists to conductmarketing research, and facilitate the exchange ofexperiences and methods.

The project's approach was to explore theprospects for commercial potato processing, usingdifferent levels of technology (rustic, semiindustrial, and industrial). This involved regularinteraction between researchers concerned withthe technical aspects of processing and socialscientists engaged in market research. Growerassociations, women's groups, and the privatesector also participated in the project. Thestrategy was flexible enough that the projectcould shift its focus as new opportunitiesemerged. Researchers were free to respond to theneed for market information and also explorethoroughly the feasibility of technical optionsbased on socioeconomic studies.

Obstacles to simple processing

The project concentrated initially on assessing thecommercial potential of products intended forhuman consumption and involving simpleprocessing methods. This work took place in thearea around Pamplona in northeastern Colombia.In Pamplona simple processing was a response tothe crash in potato prices during 1984-1985.

Case 8: Simple Potato Proc'essing in Colombia

ICA now receives requests for germplasmwith particular skin color and dry matter content;its plant breeders used to reject some suchmaterials. These staff have formed a newperspective on increased potato production.Rather than fear a drastic fall in prices, they seehigher production as necessary to meet risingdemand for fresh and processed potato products.

Lessons Learned

•

people at the lower end of the income scale.This approach proved overly ambitious. Suchprojects should probably concentrate first onone product for one segment of the market.

Experience in Colombia also underscores theimportance of consumers' perceptions inproduct development. In Pamplona, forexample, the processed products had anacceptable taste, but consumers wereconcerned about their color and texture.

•

•

Although technology is central to a project ofthis sort, it must not be emphasized at theexpense of other factors. The threeexperiences described here underscore theimportance of analyzing the market first.Characteristics of the product and targetmarket, along with sources of inputs andmarketing channels for processed products,must be clearly identified at the outset.

A related challenge is segmentation ofmarkets. In Colombia three products wereaimed at three different groups of users,although initially the emphasis was on

149

• Product development may involve newprocesses and products, such as simpleprocedures for solar drying of potato to makeflour or cubes. It may be more effective,however, to improve existing processes first(e.g., semi-industrial processing withimproved raw material).

Source

Scott, G.J. 1993. Processed potato products inColombia. CIP, Lima, Peru. 8 p.

Adding WJlue to Root and TUber Crops

Case 9

Development of a Sweet Potato Beveragein the Philippines

Objectives: 'Ib identify food products that can bemade from sweet potato, develop technology formanufacturing these products, and transfer thetechnology to processors.

Project area: Baybay, Leyte, the Philippines.

Time frame: Identification of product ideas,1985-1986; product development, 1986-1989;commercial phase, 1989 to present.

Background

Sweet potato is an important food crop in thiscountry, ranking third after rice and maize. Morethan 800,000 t are produced on about 165,000 ha.But per capita consumption of sweet potato isdeclining, and few alternative markets for thecrop have been developed.

Project Evolution

The Visayas State College of Agriculture (ViSCA)was seeking ways to increase sweet potatoutilization and raise its value for producers.College staff found that a number of processedsweet potato products (e.g., fried chips, candies,flour, and local delicacies) were already beingproduced commercially and in households.

150

The project considered three possibleapproaches: 1) to improve existing processes,2) adopt technologies from other countries, and3) develop appealing, nontraditional productswith good market potential. ViSCA chose thethird, based on the hypothesis that roots could beprocessed into products traditionally made fromfruit.

Sweet potato's appeal

Sweet potato has considerable appeal as acompetitive substitute for fruits in some products.The crop is nonseasonal and inexpensive andoffers excellent nutritional value. In addition totheir high starch content, the roots show highlevels of vitamin C. Moreover, varieties withorange root flesh contain as much B-carotene(provitamin A) as carrots and more than othervegetables and fruits.

In search of product ideas, project staff madean inventory of commercial fruit products,including dried fruits, jams, canned fruits, juices,and drinks packaged in various forms. Since freshfruits are seasonal, these products tended to beexpensive and thus accessible only to high-incomeconsumers and export markets.

Product research

ViSCA developed three products on a trial basis:dried sweet potato with a sweet and sour taste,jam, and catsup. Consumer tests gaveencouraging results, with 80% of the respondentssaying they liked the products moderately to verymuch.

Based on these results and the interest oflocal food processors in such products, the collegeproceeded with research on a fourth product, anonalcoholic sweet potato beverage (SPB). Inevaluations of varieties with different root fleshcolors, orange varieties were found most

Case 9: Development ofa Sweet Potato Beverage in the Philippines

appealing. Adding artificial orange flavoring orfruit juice or pulp from guava, pineapple, orPhilippine lemon significantly improved theproduct's aroma.

SPB had a higher vitamin A content thancommercial fruit drinks in cans and tetrapacks.An 8-oz or 237-ml serving satisfied the averageFilipino's daily requirement of this vitamin. Inaddition, the drink was fortified with vitamin C,as is commonly done in fruit juice processing. As aresult, it provided 40% of the recommended dailyallowance of this vitamin. The phosphorus andcalcium levels of SPB were significantly higherthan those of Hi-C orange, mango, and pineapplejuices, while its content of magnesium andpotassium levels was comparable to that of fruitdrinks.

A laboratory taste panel found no significantdifferences in sensory attributes between the newbeverage and commercial products. In fact, itreceived higher scores for aroma and generalacceptability than papaya nectar and pineappleorange drink. SPB with guava was rated higherthan guava-flavored fruit nectar in cans. Thenatural orange color of the beverage, an indicatorof high vitamin A content, gave it a markedadvantage over commercial fruit drinks, to whichartificial coloring and flavoring are added.

Samples of SPB, without flavoring and withripe guava, were tested on four groups ofconsumers, each representing a different agebracket. They were asked to rate the products ona seven-point scale. The results generallyconfirmed those of the lab test. Even so, someconsumers who were aware that the beverage isbased on sweet potato, were apprehensive that itmight cause flatulence, as a consequence of alarge intake of undigested starch. But the starchcontent of SPB is only 0.8-1.0 g/100 g of productlower than that of passion fruit. Ten volunteersconfirmed that consuming one 8-oz bottle of thebeverage daily did not cause flatulence.

Toward commercial processing

In 1989, ViSCA obtained a patent to protect itsrights to the processing technology andstrengthen its bargaining position with industry.The college also prepared a strategy for

151

technology transfer, involving dissemination ofinformation to the public through scientificreports, newspaper announcements, aDepartment of Trade and Industry investmentforum, and science and technology fairs.

In the end, ViSCA offered the technology, onan exclusive basis for 5 years, to a group ofbusinessmen who were establishing a foodprocessing company. The college was criticized forthis decision, because it seemed inconsistent withViSCA's mandate to improve the welfare of smallscale farmers. Project staff hoped, however, thatcommercial processing of sweet potato wouldincrease the demand for raw material, resultingin higher prices for the crop and boosting farmers'incomes. Unfortunately, the processors had littleexperience in marketing food products, and theirfacilities proved inadequate. When fire destroyedthe processing operation, this arrangement wasvoided by mutual consent.

ViSCA next entered into an agreement witha large food and beverage company, which wasexploring the potential of indigenous rawmaterials in the production of nutritious,inexpensive food products for poor consumers.The company intended to contract farmers toproduce raw material, thus integratingproduction with processing and marketing.

ViSCA's alliance with local industry

Under ViSCA's agreement with the company, theprocessing technology was made available on anonexclusive basis, with the understanding thatany improvements the company made would beits property for 5 years from the time the productwent into commercial production. Moreover, alltrademarks developed for the product wouldbelong to the company.

In return, it donated food processingequipment to ViSCA for research anddevelopment. The company also provided fundsand the use of its facilities for production of SPB,first on a pilot basis and then semicommercially.

This work was done by a team consisting ofthe ViSCA scientist who invented the product,together with company specialists in product andprocess development, plant management, andagricultural extension. Their main tasks were to

raise product quality to company standards, carryout trials with local varieties of sweet potato andthe improved cultivar VSP-1, identify the factorsaffecting quality and processing, evaluate theproduct for sensory attributes and shelf life, andanalyze costs (taking into account fluctuations inthe price of raw material).

Adding \blue to Root and Thber Crops

can market this new product without muchadditional investment in advertising anddistribution. That in turn increases the project'spotential for impact over a wider area.

Lessons Learned

Source

Truong, YD. 1991. Development of a sweet potatobeverage in the Philippines. ViSCA, Baybay,Leyte, Philippines. 14 p.

• Strong ties between production andprocessing are critical to successful productdevelopment and should be establishedbefore the product reaches commercialproduction. This step is just as important ina project like this one, which links smallscale crop production to large-scaleprocessing, as in the other nine casesdescribed here, whose main aim is toestablish small- to medium-scaleagroindustries in rural areas.

1b ensure a steady supply of high-qualityraw material, ViSCA offered assistance inimproving cultural practices, postharvesthandling, and grading of roots. It also trainedstudents at a local agricultural school to producesweet potato planting materials.

Future directions

The project seems to have resolved the majority ofits challenges in postharvest processing andmarketing. Nonetheless, it will take a substantialeffort to provide the company with a reliablesupply of raw material. This could prove difficult,since the plan is to produce SPB only whencertain fruits are not in season.

It remains to be seen whether, by placing theprocessing technology in the hands of a largeprivate company, the project can achieve itsoriginal objective of benefiting small-scale sweetpotato farmers. They may not be able to modifycrop production practices to meet the processor'sdemands.

Even so, the arrangement with this companydoes offer a decided advantage. Since it is alreadyproducing beverages similar to SPB, the company

152

•

•

Close analysis of the physicochemical andfunctional properties of the raw materialoften leads to a new product idea.

Teams consisting of specialists fromuniversities and industry can be effective incarrying a product from laboratory testing topilot-scale production.

Case 10


Objectives: In research, to determine thetechnical and economic feasibility of producing asoy-flavored sauce from root crop flours; in thepilot phase, to refine the processing technology,transfer it to farmer organizations, and develop amarketing strategy.

Project area: Maasin, southern Leyte, thePhilippines.

Time frame: Research, 1980-1987; pilot phase,1988-1990.

Background

The Philippine Root Crop Research and TrainingCenter (PRCRTC) and the PostharvestTechnology Section of ViSCA develop and improvepostharvest technology for root crops, with a viewto increasing supplies of better quality food, feed,and industrial products for low-income people. Aprimary objective of both institutions is toincrease commercial use of root crops.

One option PRCRTC has explored is todevelop a soy sauce in which flour made from rootcrops (sweet potato, cassava, and taro) substitutesfor wheat flour, a premium raw material that isoften in short supply. Soy sauce is a popularcondiment throughout the country. Ifmanufactured from roots crops, the product wouldbe cheaper, could be produced in larger quantities,and would provide farmers with a new outlet forroot crop production.

Project Evolution

The project consisted of two phases: 1) researchand 2) a pilot study conducted in a localcommunity. The research was carried out by aPRCRTCMSCA team, composed of a postharvesttechnologist, food microbiologists, economist,nutritionist, and agricultural engineer. The pilotproject was implemented by a PRCRTCpostharvest specialist, agricultural engineer, andengineer. The pilot phase was funded by the

153

Department of Science and Technology-RegionVIII. The Department of Agriculture providedtechnicians, and the municipal government ofMaasin supplied equipment and local staff.

Technical and market research

Soy sauce is normally prepared from a mixture ofsoybean and wheat, which is fermented throughthe action of the fungus Aspergillus oryzae. 'Ibmanufacture this product from root crops involvesthree main steps:

1. Root crop flour is prepared by washing,peeling, chipping, and sun-drying, andgrinding (the chips may also be cooked beforedrying). The dry ground flour is thenroasted.

2. Starter is made by first soaking, draining,and sterilizing rice. Mter the resulting mashhas cooled to room temperature, it is seededwith spores of A. oryzae or A. sojae and left toincubate for 4-6 days or until greenish sporesform.

3. Soy sauce is derived by first mixing steamedsoybean with roasted root crop flour andstarter. The mixture is incubated, initially for4-5 days to allow growth of themicroorganism, and then for 3 months, withthe addition of a brine solution, until the pHreaches 5.05-5.50. The resulting mash ispressed and strained to obtain a liquor,which is left to settle overnight. The soysauce is then decanted. This process isrepeated twice, after which the three liquorsobtained are combined. Molasses is added togive the sauce a darker color and greaterviscosity. The mixture is pasteurized at 800

for 3 minutes.

The research team examined varioustechnical issues involved in this process. Forexample, it compared three methods of extractingthe sauce: manual (in muslin) and mechanical (in

nylon cloth), using a screw- or lever-type press.The teams also conducted three laboratory trialsto determine the most effective medium forfungal development.

Mold grew most abundantly on cookedsweet potato medium. In the first trial, cookedsweet potato and wheat flour gave the highestyields and cassava flour the lowest, especiallyuncooked cassava. Cooking gelatinizes thestarch, making it more readily available for themicroorganisms to act upon, and sweet potatoeshave more digestible matter than cassava (88%versus 76%). In the second and third trials,production did not vary significantly, perhapsbecause of the manner in which the sauce wasextracted.

Production costs were highest for wheatflour (US$0.37), followed by sweet potato ($0.18),and cassava ($0.14).

The properties of soy sauce made fromcooked sweet potato flour are comparable tothose of sauce made with wheat flour. Oneexception is the low salt concentration of theformer, which can easily be adjusted.

The soy sauce was compared on the basis oforganoleptic qualities with two locally availablecommercial brands (A, which is dark-colored andinexpensive, and B, which is lighter and morecostly). PRCRTC researchers and laborers wereasked to rate the sauces (on a nine-point scale) inpure form, as a dip for broiled fish, and as amarinade for beef. The panelists gave the sweetpotato soy sauce the same rating as bothcommercial brands. They considered the cassavasoy sauce as good as brand B but felt that bothwere inferior to brand A in terms of color, aroma,and consistency. Taro soy sauce received thelowest rating, although it was found comparablein aroma to brand B.

ViSCA identified 12 processing devicessuitable for village-level processing andevaluated them in the laboratory. These includeda root crop washer, pedal-operated root chipper/grater, a modified copra dryer, a portable Almedaattrition mill (designed for rice and maize), and ascrew-press sauce extractor. Other devices weredesigned specifically for the project, including a

154

Adding \-blue to Root and Thber Crops

flour roaster, mixer, crown cap sealer, andcharcoal stove.

The name Root Soy Sauce was selected forthe product through a contest. Consumersurveys were conducted at ViSCA and in fournearby communities. Samples (50 ml) weredistributed among households, and a week laterpeople were asked to comment on the product.Survey results indicated that the aroma was toostrong, salt content needed to be increased, andcolor should be darker.

Marketing channels were identifiedthrough a study of retail stores and cooperativesin the area around Baybay and a survey of 300consumers chosen at random. The preferredpoint of purchase was retail stores and thepreferred size 12 oz or 320 ml (about the size of abeer bottle). Consumer preference wasdetermined largely by availability, price, taste,and promotional activities.

A feasibility study was conducted in Leyte,because it has great potential as a source of rawmaterial and is readily accessible from ViSCA.Data were collected through interviews, analysisof processing costs in the lab and on a pilotscale, and market testing. Potential monthlydemand for root soy sauce was estimated at1,814 L. The capital investment required to meet50% of this demand was estimated at aboutUS$3,220 and just over $4,140 for the wholeproject.

The pilot phase

A project management board was established tosupervise and monitor the pilot phase. Theboard consisted of the ViSCA-PRCRTC team, theplant manager, and representatives from theDepartment of Science and Technology, afarmers' federation, and the municipality.Maasin was selected as the site for the pilotplant, because it has good crop productionpotential, farmers there showed strong interestin the project, institutional support is available,the infrastructure is good, and the municipalityis under progressive leadership. The pilot plantwas inaugurated in May 1989.

Case 10: Development ofRoot Soy Sauce in the Philippines

The project strongly emphasized institutionbuilding as a means of making the processingoperation viable. A farmers federation (consistingof 40 farmer associations) was organized andmade responsible for plant management. Theplant's technical staff, all recruited from the localcommunity, received training both in the lab andat the pilot site. Training was combined withtesting of processing equipment and trials todetermine optimum operational schemes.

After evaluating the operational viability andorganizational and management structure of thepilot plant, the project moved into a transitionphase in preparation for commercial production.The process and plant infrastructure weremodified; a full feasibility study was conducted;and ViSCA staff began gradually to withdrawsupport. The farmers federation continues toimprove produce quality, stability, andpresentation. These changes will be followed byfurther consumer testing and marketing trials.Farmers are being encouraged to adopt improvedsweet potato cultivars with high dry mattercontent.

Future directions

Assuming the transition phase described abovegives satisfactory results, the project will embarkon commercial production. To finance thisventure, the project has submitted a proposal forexternal funding.

155

Lessons Learned

• Farmer groups that undertakeagroindustrial processing need strongsupport in organization and conflictmanagement.

• Process and product development can take along time. In this case it took 10 years toassess the commercial viability of theproduct.

• A pilot phase is essential for laying thegroundwork of commercial production.During this phase, it is critical thatmultidisciplinary teams conduct planning,that institutional linkages be established,and that local groups participate inmanagement.

• Information gathered through consumeracceptance trials and marketing surveys isvital for orienting research aimed atimproving product quality and presentation.

Source

Data, E.S. and Roa, J.R. 1991. Development of soysauce from root crop residues: PRCRTC'sexperience. Philippine Root Crop Research andTraining Center, Visayas State College ofAgriculture, Baybay, Leyte, Philippines. 36 p.

7b obtain source materials

Source materials for the 10 case studies includedin this manual, as well as for six other studies notcovered here, are available from the CassavaProgram, CIAT, at US$0.10 per photocopied page(including postage). The six additional studiesare:

Agroindustrial program for cassava in Mexico byAsunci6n Mendez. (Spanish, 6 p.)

Cassava root and leaf meal in balanced feed rations byLuis Fernando Gerhard. (Portuguese, 222 p.)

Development and utilization of nationally producedtuber and cereal flours to substitute wheat byAhmed El-Dash. (English, 43 p.)

Nontechnical problems associated with improvedvillage-scale gari processing technology byAurea Almazan. (English, 15 p.)

Pilot testing of commercial formulations of root cropbased feeds by Guindolino Gerona. (English,14 p.)

Thailand's experience in the development of the driedcassava industry by Alistair Hicks. (English,19 p.)

Much valuable information is also containedin the set of workshop proceedings listed below.They present information on cassava production,processing, and marketing in the countries

156

Adding \.htue to Root and 'fuber Crops

represented at the workshops and also describeresearch; processing methodologies; trade in rootsand tubers; the work and experience of differententities, including commercial firms; and theregional potential for root and tuber processing.

Scott, G.; Wiersema, S.; and Ferguson, P.I. (eds.). 1992.Product development for root and tuber crops,vol. I-Asia. Proceedings of the InternationalWorkshop. CIP, Lima, Peru. 384 p.

Scott, G.; Herrera, J.E.; Espinola, N.; Daza, M.;Fonseca, C.; Fano, H.; and Benavides, M. (eds.).1992. Desarrollo de productos de raices ytuberculos, vol. II-America Latina. Memorias delTaller sobre Procesamiento, Comercializaci6n yUtilizaci6n de Rakes y Tuberculos en AmericaLatina. CIP, Lima, Peru. 375 p.

Scott, G.; Ferguson, P.I.; and Herrera, J.E. (eds.). 1992.Product development for root and tuber crops,vol. III-Africa. Proceedings of the Workshop onProcessing, Marketing, and Utilization of Rootand Tuber Crops in Africa. CIP, Lima, Peru.506 p.

All three volumes are available from CIP.The price per volume is US$15 for developingcountries and $30 for developed countries, plus$10 per copy for handling and shipping. A limitednumber of copies are available free to researchersin developing countries.

Appendix

Contributors to this Manual

Participants in a meeting held on 28-29 April 1988at the Overseas Development Natural ResourcesInstitute (ODNRI, now NRI), London, UK:

Bob Baulch and Nigel PoulterODNRI, London, UK

Rupert Best and Christopher WheatleyCIAT, Cali, Colombia

Y.W. Jeon and Aurea AlmazanIntemational Institute of Tropical Agriculture(UTA), Ibadan, Nigeria

Gregory Scott and Siert WiersemaCentro Intemacional de la Papa (CIP), Lima,Peru

Participants in a meeting held on 7-9 November1988 at the Ambassador Hotel, Bangkok,Thailand:

Rupert Best and Christopher WheatleyCIAT, Cali, Colombia

Robert BoothPrevention of Food Losses Programme, Food andAgriculture Organization (FAO), Rome, Italy

Emma DataPost-harvest Technology Section, Philippine RootCrop Research and Training Center, Leyte,Philippines

Ahmed EI-DashFaculty of Food Engineering, Universidad deCampinas, Campinas, Sao Paulo, Brazil

Alistair HicksAgricultural Engineer/Agro-industries,FAO Regional Office for Asia and the Pacific,Bangkok, Thailand

Y.W. Jeon and Aurea AlmazanUTA, Ibadan, Nigeria

Leticia LaniyunoCommittee on Women in Development, Oyo State,Nigeria

Robert NaveSociety for the Development of AppropriateTechnology, Bareilly, India

157

Truong Van DenDepartment of Agricultural Chemistry and FoodScience, Visayas State College of Agriculture,Leyte, Philippines

Siert WiersemaCIP, Lima, Peru

Participants in the Taller Colaborativo de CIP,CIAT e UTA con el Instituto de Ciencia yTecnologia Agricolas (lCTA) sobre Procesamiento,Mercadeo y Utilizaci6n de Raices y Tuberculos enAmerica Latina, held at leTA, Villa Nueva,Guatemala, on 8-12 April 1991:

Colombia

Andres Alvaro Alvarez SotoMejorador, Tuberosas TropicalesInstituto Colombiano AgropecuarioEl Carmen de Bolivar

Alejandro Femandez Q.Profesor, Facultad de IngenieriaSecci6n de AlimentosUniversidad del ValleCali

Pedro Rodriguez QuijanoIngeniero Agr6nomo, Secci6n PapaInstituto Colombiano AgropecuarioSantafe de Bogota

Christopher C. WheatleyJefe, Utilizaci6n de YucaCentro Intemacional de Agricultura TropicalCali

Costa Rica

Fran~ois BoucherEspecialista, AgroindustriaInstituto Interamericano de Cooperaci6n para la

AgriculturaCoronado

David Ricardo Carmona BrownCoordinador EjecutivoConsejo Nacional de Producci6nMinisterio de Agricultura y GanaderiaSan Jose

Alfonso Delgado SalasJefe, Departamento TecnicoPrograma Agrario NacionalConfederaci6n Nacional de TrabajadoresSan Jose

Carlos Roberto Ramirez AguilarEspecialista, Cultivo de PapaMinisterio de Agricultura y GanaderiaCartago

CubaPedro Luis Dominguez GuardeInvestigador TitularJefe, Laboratorio Nutrici6nInstituto de Investigaciones PorcinasPunta Brava LisaHabana

Sergio Rodriguez MoralesSubdirector de InvestigacionesInstituto de Investigaci6n en Viandas TropicalesSanto Domingo

Dominican RepublicOscar S. MalamudCentro Internacional de la PapaDirecci6n RegionalSanto Domingo

EcuadorGloria Cobena RuizInvestigador AgropecuarioInstituto Nacional de Investigaciones AgropecuariasPortoviejo, Santa AnaManabf

Vicente Ruiz ChavezJefe de ProyectosUni6n de Asociaciones de Productores y

Procesadores de YucaPortoviejo

GuatemalaRicardo BressaniInstituto de Nutrici6n de Centro America y PanamaGuatemala

Axel Esquite CastilloEncargado de AgroindustriaInstituto de Ciencia y Tecnologfa AgricolasBarcena, Villa Nueva

Carlos Alfredo Egtiez AlavaCoordinador, Programa YucaFundaci6n para el Desarrollo AgropecuarioQuito

Adding \blue to Root and Thber Crops: A Manual on Product Development

Alvaro Roberto EI Cid HerreraCoordinador, Programa de HortalizasInstituto de Ciencia y Tecnologfa AgricolasBarcena, Villa Nueva

Marcio Aristides Ibarra MenendezTecnico del Equipo de Prueba, Validaci6n y

Transferencia de TecnologfaInstituto de Ciencia y Tecnologfa AgrfcolasBarcena, Villa Nueva

Luis Maria Soto GuevaraTecnico del Programa de HortalizasInstituto de Ciencia y Tecnologfa AgricolasBarcena, Villa Nueva

Rosemary Vargas LundiasOficial de ProgramasPrograma de las Naciones Unidas para el DesarrolloGuatemala

MexicoJose Joaquin Bonilla BadaInvestigador de la Red de Divulgaci6nInstituto Nacional de Investigaciones Forestales y

AgropecuariasPuebla

PanamaJose Antonio Aguilar L6pezInvestigador AgricolaPrograma de Rakes y TuberculosInstituto de Investigaci6n Agropecuaria de PanamaPanama

Bernardo Enrique Puga SantosJefe, Programa de Rakes y TuberculosDirecci6n de AgriculturaMinisterio de Desarrollo AgropecuarioSantiago

Javier E. RibaGerente GeneralRiba Smith, SAPanama

PeruMaria AlvarezGerente, Industria de Derivados del Agro S.A.Centro de Investigaci6n, Documentaci6n,

Asesoramiento y ServiciosLima

Pelayo PeraltaDir. Of. AgroeconomiaInstituto Nacional de Investigaci6n Agraria y

AgroindustrialLima

158

Appendir Contributo,.. w this Manual

Gregory J. ScottLfder, Programa de Poscosecha y Utilizaci6nCentro Internacional de la PapaLima

Participants in the International Workshop onRoot and 'fuber Crop Processing, Marketing, andUtilization in Asia, organized by CIP, CIAT, andUTA with ViSCAand held 22 April to 1 May 1991at ViSCA, Baybay, Leyte, Philippines:

AustraliaUthai CenpukdeeResearcherUniversity of Queensland

ColombiaRupert BestLeader, Cassava ProgramCentro Intemacional de Agricultura TropicalCali

Christopher WheatleyHead, Utilization SectionCassava ProgramCentro Intemacional de Agricultura TropicalCali

ChinaShengwu WangFood TechnologistXuzhou Sweet Potato Institute

Li Wei GeTemporary DirectorFeed InstituteChinese Academy of Agricultural SciencesBeijing

IndiaCherukat BalagopalanHead, Division of Postharvest TechnologyCentral Tuber Crops Research InstituteThriruvananthapuram GevanelvumKerala

Robert W. NaveProject DirectorSociety for Development of Appropriate TechnologyBareilly, U.P.

159

Indonesia

Djoko DamarcJjatiResearch Coordinator, Postharvest Technology!

Food ScientistCentral Research Institute for Food CropsBogor

Abdul RachimAssociate Agricultural EconomistCentral Research Institute for Food CropsBogor

Agus SetyonoPostharvest Technologist/Food ScientistCentral Research Institute for Food CropsBogor

J.W. Taco BottemaAgricultural EconomistCoarse Grains, Pulses, Roots and Tuber Crops CenterBogor

Italy

Ester BonitatibusAssociate Professional OfficerPrevention of Food LossesAgricultural Services DivisionFood and Agriculture Organization of the

United NationsRome

Korea

Byeong Choon JeongMokpo Branch StationCrop Experiment StationRural Development AdministrationMuan Chunnam Province

Netherlands

Siert WiersemaIntemational ConsultantDepartment of Crop ScienceUniversity of WageningenAcacialaan

Peru

Dr. Gregory J. ScottLeader, Postharvest Management,

Marketing ProgramIntemational Potato CenterLima

PhilippinesAna AbejuelaHead, Research and Utilization ProjectSoutheast Asian Regional Center for

Graduate Study and Research in AgricultureCollege, Laguna

Jose M. Alkuino, Jr.ProfessorlHead of Agricultural Chemistry and

Food ScienceVisayas State College of AgricultureBaybay, Leyte

Felix J. Amesto80Assistant Professor and Research AssistantFood 'Thchnology SectionDepartment of Agricultural Chemistry and

Food ScienceVisayas State College of AgricultureBaybay, Leyte

Jose L. BacusmoAssociate ProfessorPhilippine Root Crops Research and Training CenterVisayas State College of AgricultureBaybay,Leyte

Liborio S. CabanillaChairman, College of Economics and ManagementUniversity of the PhilippinesLos Banos

Vu Manh CuongInternational Potato CenterInternational Rice Research InstituteManila

Guindolino R. GeronaProfessor, Department of Animal Science and

Veterinary MedicineVisayas State College of AgricultureBaybay, Leyte

Phung Huu HaoInternational Potato CenterInternational Rice Research InstituteManila

Alan B. LoretoResearcherPhilippine Root Crop Research and Training CenterVisayas State College of AgricultureBaybay, Leyte

Lutgarda S. PalomarProfessor, Department of Agricultural Chemistry

and Food ScienceVisayas State College of AgricultureBaybay, Leyte

Adding \blue to Root and Thber Crops: A Manual on Product DelJelopment

Hilda QuindaraInternational Potato CenterInternational Rice Research InstituteManila

Julieta R. RoaHead, Socio-Economics SectionPhilippine Root Crop Research and Training CenterVisayas State College of AgricultureBaybay, Leyte

Henry Samar, Sr.President, BIADS Foundation, Inc.Legazpi City

Daniel Leslie TanAgricultural EngineerPhilippine Root Crop Research and Training CenterVisayas State College of AgricultureBaybay, Leyte

Zenaida ThqueroPostharvest Economics Advisor (Asia)International Development Research CentreVisiting ScientistSoutheast Asian Regional Center for

Graduate Study and Research in AgricultureCollege, Goruro

Truong Van DenFood Technologist, Department of Agricultural

Chemistry and Food ScienceVisayas State College of AgricultureBaybay, Leyte

ThailandSarath IlangantilekeChairman, Division of Agricultural and

Food EngineeringAsian Institute of TechnologyBangkok

Saipin ManeepunDirectorInstitute of Food Research and Product DevelopmentKasetsart UniversityBangkok

Sumalee SoontornnarurngsiHome Economist, Department of Agricultural

ExtensionMinistry of Agriculture and CooperativeBangkok

UKWinston TimminsNatural Resources InstituteChatham Maritime

160

Appendi:&; Contributors to this Manual

Vietnam

Quach NghiemHead, Department of Biochemistry and

Food TechnologyNational Institute of Agricultural SciencesHanoi

Western Samoa

F. BjornaAssociate Professional OfficerFood and Agriculture Organization of the

United NationsApia

Participants in the International Workshop onRoot and Tuber Crop Processing, Marketing, andUtilization in Mrica, organized by CIP, UTA, andCIAT and held 26 October to 2 November 1991 atUTA, Ibadan, Nigeria:

Burundi

M. BeavoguiChief Technical AdviserFood and Agriculture Organization of the

United NationsBujumbura

D. BerriosPhysiologistInternational Potato CenterBujumbura

Cameroon

J.T.AmbeInstitut de la Recherche AgronomiqueNgaoundere

A. FoaguegueInstitut de la Recherche AgronomiqueDjombe Station

Rose WanzieProvincial Subject Matter Specialist Post-harvestThe Provincial Delegation of AgricultureBamenda, North West Province

Colombia

Guy HenryEconomistCassava ProgramCentro Internacional de Agricultura TropicalCali

161

Christopher WheatleyHead, Utilization SectionCassava ProgramCentro Internacional de Agricultura TropicalCali

Congo

G. BoukambouVice PresidentAGRICONGOBrazzaville

O. LegrosD.E.E.AGRICONGOBrazzaville

J. MassambaNutritionist, Departement BPAFaculte des SciencesBrazzaville

S. TrecheResponsable, Lab. d'Etudes sur la Nutrition et

l'AIimentationInstitut Francais de RechercheScientifique pour Ie Developpement en CooperationBrazzaville

Cote d'Ivoire

N'dri CoulibalyUniversite d'Abidjan

Kouadio TanoIDESSABouake

France

Brook A. GreeneInstitut de Gestion Internationale Agro-AIimentaireEtablissement d'Enseignement Superieur PriveCergy-Pontoise Cedex

Gabon

J. BrochierGeneral ManagerAGRICONGOLibreville

Ghana

Ramatu M. AI-HassanDepartment of Agricultural EconomicsFaculty of AgricultureUniversity of GhanaLegon

Kenya

Peter EwellRegional RepresentativeInternational Potato CenterNairobi

Jasper K ImungiDept. of Food Technology and NutritionFaculty of AgricultureUniversity of NairobiNairobi

Jackson N. KabiraCoordinator for Root and 'fuber CropsKenya Agricultural Research InstituteLimuru

E.G. KaruriDept. of Food Technology and NutritionUniversity of NairobiNairobi

N.KO.OjijoCoffee Research FoundationRuiru

Nigeria

IA AdeyemiResearcher, Department of Food Science and

EngineeringLadoke Akintola University of TechnologyOgbomoso

O. AjoboDepartment of Agricultural EconomicsObafemi Awolowo UniversityHe-Ife

O.B. AreneAssistant Director, PlanninglProject DevelopmentNational Root Crops Research InstituteUmidike

P.AySociologistInternational Institute of Tropical AgricultureIbadan

J.S.T. BogunjokoTechnical DirectorCadbury Nigeria PLCIkeja

L.S.O. EneDirectorNational Root Crops Research InstituteUmudike

162

Adding Value to Root and Thber Crops: A Manool on Product Development

N.OA EzehIndustrial EconomistNational Root Crops Research InstituteUmudike

L.S. HalosResearch Specialist, Postharvest Technology UnitInternational Institute of Tropical AgricultureIbadan

Yw. JeonHead, Postharvest Technology UnitInternational Institute of Tropical AgricultureIbadan

Felix I. NwekeAgricultural Economist and Project LeaderCollaborative Study of Cassava in Africa ProjectInternational Institute of Tropical AgricultureIbadan

S.O. OdurukweAssistant DirectorDepartment of Farming Systems Research and

ExtensionNational Root Crops Research InstituteUmudike

O.L. OkeResearcher, Department of Pure and

Applied ChemistryLadoke Akintola University of TechnologyOgbomoso

P. OmoakaResearch AssistantInternational Institute of Tropical AgricultureIbadan

G.OrkworNational Root Crops Research InstituteUmudike

B.O. UgwuNational Research CoordinatorNigeria COSCA projectNational Root Crops Research InstituteUmudike

01umide O. TeweProfessor, Dept. of Animal ScienceUniversity of Ibadan

Peru

Gregory J. ScottLeader, Postharvest Management,

Marketing ProgramInternational Potato CenterLima

AppendU: Contributors to this Manual

Rwanda

David G. Tardif-DouglinAgricultural EconomistDevelopment Alternatives Inc.Division des Statistiques Agricoles, MINAGRIKigali

Tanzania

M.A.M. MsabahaMwanza Research Station

Gabriel T. NdunguruFood Science and Technology DepartmentTanzania Food and Nutrition CentreDar es Salaam

B.W. RwenyagiraMinistry of Agriculture

Uganda

R.O.M. MwangaResearcherDepartment of AgricultureNamulonge Research StationKampala

J.U.A. Opio-OdongoProfessorMakerere University

G.w. Otim-NapeNamulonge Research StationKampala

UKM.H. FowlerAgricultural EconomistNatural Resources InstituteChatham Maritime

163

A.M. StabrawaAgricultural EconomistNatural Resources InstituteChatham Maritime

USABernard BashaashaPh.D. Student, Department of Agricultural EconomicsOhio State UniversityColumbus, OH

Gillian EgglestonFood TechnologistlBiochemistColumbia, MO

Njeri GakonjoPh.D. StudentDepartment of Agricultural EconomicsCornell UniversityIthaca, NY

Adhiambo OdagaThe World BankAFTHR DivisionWashington, D.C.

Truong Van DenPostdoctoral ScientistDepartment of Food ScienceNorth Carolina State UniversityRaleigh, NC

Vietnam

Dang Thi LanHead, Postharvest SectionPotato and Vegetable Research CenterInstitute for Agricultural ResearchHanoi

Adding \blue to Root and Tuber Crops: A Manual on Product Dewlopment

Acronyms

ACFOA

ACIAR

ANPPy

APPy

ASOCOSTA

ATI

AVRDC

CECORA

CIAT

CIDA

CIP

CIUP

COOAGRONOR

COOPROALGA

Australian Council for OverseasAid

Australian Centre forInternational AgriculturalResearch

Asociaci6n Nacional deProductores y Procesadores deYuca (National Association ofCassava Producers andProcessors), Colombia

Asociaci6n de Productores yProcesadores de Yuca(Association of CassavaProducers and Processors),Ecuador

Asociaci6n de Cooperativas de laCosta (Association of CoastalCooperatives), Colombia

Appropriate TechnologyInternational, USA

Asian Vegetable Research andDevelopment Center, Taiwan

Central de Cooperativas de laReforma Agraria (Central Officeof Cooperatives for AgrarianReform), Colombia

Centro Internacional deAgricultura Tropical(International Center forTropical Agriculture), Colombia

Canadian InternationalDevelopment Agency

Centro Internacional de la Papa(International Potato Center),Peru

Centro de Investigaci6n de laUniversidad del Pacifico, Peru

Cooperativa Agroindustrial delNor-Oriente del Athintico(Agroindustrial Cooperative forthe Northeast of AtlanticoDepartment), Colombia

Cooperativa de Productores deAlgarrobos (Cooperative ofCarob Producers), Colombia

164

COOPROMERCAR Cooperativa de Producci6n yMercadeo de Repel6n (Repel6nCooperative for Production andMarketing), Colombia

CORFAS Corporaci6n Fondo de Apoyo aEmpresas Asociativas(Corporation Fund for Supportto Associations), Colombia

CTI Compatible Technology, Inc.,a US agency operating in India

DRI Programa de Desarrollo RuralIntegrado (Program forIntegrated Rural Development),Colombia

EMATERCE Empresa de Assistencia Tecnicae Extens8.o Rural do Ceara(State Technical Assistance andRural Extension Agency ofCeara), Brazil

EPACE Empresa de PesquisaAgropecuaria do Ceara (StateAgricultural Research Agency ofCeara), Brazil

ERR Economic rate of return

ED European Union

FAGROCOL Federaci6n de OrganizacionesAgropecuarias de Colombia(Colombian Federation ofAgricultural Organizations)

FAO Food and AgricultureOrganization of the UnitedNations

FEDECOSABANA Federaci6n de Cooperativas dela Sabana (Federation ofCooperatives in the Savanna),Colombia

FIFAMANOR Fiompiana-Fambolena-Malagasy-Norveziana(Malagasy-Norwegiancooperative project on potato,wheat, and milk production),Madagascar

FINANCIACOOP Instituto de Financiamiento yDesarrollo Cooperativo deColombia (Colombian Institutefor Cooperative Finance andDevelopment)

Acronyma

FOFIFA Foibem-pirenena momba ny INCORA Instituto Colombiano de lafikarohana ampiharina amin'ny Reforma Agraria (Colombianfampandrosoana ny Institute for Agrarian Reform)ambanivohitra (National Centerfor Applied Research on Rural INIAA Instituto Nacional deDevelopment, Department of Investigaci6n Agraria yResearch and Development), Agroindustrial (NationalMadagascar Institute for Agrarian and

Agroindustrial Research), PeruFRR Financial rate of return

INIAP Instituto Nacional de

FUNDAGRO Fundaci6n Ecuatoriana de Investigaciones Agropecuarias

Investigaciones Agropecuarias (National Institute for

(Ecuadoran Fund for Agricultural Research), Ecuador

Agricultural Research)INTA Instituto Nacional de Thcnologia

GAIT General Agreement on TariffsAgropecuaria (NationalInstitute for Agricultural

and TradeTechnology), Argentina

ICA Instituto Colombiano IRRI International Rice ResearchAgropecuario (Colombian Institute, PhilippinesInstitute for Agriculture andLivestock) ISER Instituto de Economfa y

ICTA Instituto de Ciencia y TecnologiaSociologia Rural (Institute ofEconomics and Rural Sociology),

Agricolas (Institute of INTA, ArgentinaAgricultural Science andTechnology), Guatemala ISNAR International Service for

National Agricultural Research,IDEAGRO Industria de Derivados del Agro The Netherlands

S.A. (Agricultural By-ProductsIndustry), Peru NAS National Academy of Sciences,

USAIDEAS Centro de Investigaci6n,

Documentaci6n, Asesoramiento NGO Nongovernment organizationy Servicios (Center for Research,

NRI Natural Resources Institute, UKDocumentation, Assistance, andServices), Peru

ODNRI Overseas Development Natural

IDRC International DevelopmentResources Institute, UK (now

Research Centre, CanadaNRI)

IFPRI International Food PolicyPRACIPA Programa Andino de

Investigaci6n en Papa (AndeanResearch Institute, USA Program for Potato Research),

PeruIICA Instituto Interamericano de

Cooperaci6n para la Agricultura PRCRTC Philippine Root Crop Research(Inter-American Institute for and Training CenterCooperation in Agriculture),Costa Rica RTVD Research Training and Vi,llage

Development Centre, IndialIT Instituto de Investigaciones

Tecnol6gicas (Institute for SEARA Secretaria de Agricultura eTechnical Research), Colombia Reforma Agraria (Secretariat of

Agriculture and AgrarianIITA International Institute of Reform), Brazil

Tropical Agriculture, NigeriaSENA Servicio Nacional de

ILO International Labor Aprendizaje (National TrainingOrganization Service), Colombia

165

Adding \i}lue to Root and Thber Crops: A Manual on Product Development

SOTEC

UAPPy

UNDP

UNICEF

Society for Development ofAppropriate Technology, India

Uni6n de Asociaciones deProductores y Procesadores deYuca (Union of Associations ofCassava Producers andProcessors), Ecuador

United Nations DevelopmentProgramme

United Nations InternationalChildren's Fund

166

UNIDO

USAID

USDA

ViSCA

United Nations IndustrialDevelopment Organization

US Agency for InternationalDevelopment

US Department of Agriculture

Visayas State College ofAgriculture, Philippines

Adding Value to Root and Tuber Crops

Documents