systemic change case study on Naatal Mbay.pdf

CASE STUDIES ON FACILITATING

SYSTEMIC CHANGE

A SYNTHESIS OF CASES FROM GHANA, SENEGAL, ZAMBIA, AND RWANDA

LEO REPORT NO. 49

OCTOBER 2016

CASE STUDIES ON FACILITATING

SYSTEMIC CHANGE

A SYNTHESIS OF CASES FROM GHANA, SENEGAL,

ZAMBIA, AND RWANDA

CONTRACT NUMBER: AID-OAA-C-13-00130

COR USAID: KRISTIN O’PLANICK

CHIEF OF PARTY: ANNA GARLOCH

LEO REPORT NO. 49

OCTOBER 2016

DISCLAIMER

This publication was produced for review by the United States Agency for International Development. It was prepared by

Olaf Kula of ACDI/VOCA (synthesis), Charles Addo, independent consultant (Ghana case), MarketShare Associates (Senegal and Rwanda cases), and Dan White of ACDI/VOCA (Zambia case) with funding from USAID’s Leveraging Economic Op-portunities (LEO) project.

i

CONTENTS

EXECUTIVE SUMMARY

I. BACKGROUND

A. Objectives of the Study B. Methodology

II. SPOTLIGHT ON INNOVATION: A THEORY OF CHANGE FOR SYSTEMIC CHANGE

III. COUNTRY CASE STUDIES

A. SENEGAL: SYSTEMIC CHANGE CASE STUDY ON NAATAL MBAY

Summary Context Project Strategy Systemic Change Area 1: Contract Farming Evidence of Systemic Change systemic change area 2: equipment leasing Evidence of Systemic Change Conclusions

B. ZAMBIA: SYSTEMIC CHANGE CASE STUDY ON PROFIT+

Summary Context Project Strategy Evidence of Systemic Change Conclusions

C. RWANDA: SYSTEMIC CHANGE CASE STUDY ON RDCP II

Summary Context Project Strategy Evidence of Systemic Change Conclusions

D. GHANA: SYSTEMIC CHANGE STUDY ON ADVANCE II

Summary Context Project Strategy Evidence of Systemic Change Conclusions

IV. BENEFICIARY SPOTLIGHTS

ANNEX A. BIBLIOGRAPHY

ii

ACRONYMS & ABBREVIATIONS

ABS Agro Business Services (See under “Ghana”)

ABS African Breeders Services (See under “Rwanda”)

ADVANCE Agricultural Development and Value Chain Enhancement project

AGDP Agricultural Gross Domestic Product

ANCAR National Agency for Rural and Agricultural Advisory Services

(Agence National Conseil Agricole et Rural)

BRB Bonzali Rural Bank

CAD Community agrodealers

CIRIZ Interprofessional Rice Committee in Senegal

(Comité Interprofessionnel du Riz au Sénégal)

CNAAS National Agricultural Insurance Company for Senegal

(Compagnie nationale d'assurance agricole du Sénégal)

CNCAS National Fund for Agricultural Development of Senegal

(Caisse Nationale de Crédit Agricole du Sénégal)

COMESA Common Market for Eastern and Southern Africa

DCA Development Credit Authority

DFID United Kingdom Department for International Development

DRDR Regional Directorate of Agriculture (Direction Régionales de Développement Rural)

ECOWAS/CEDAO Economic Community of West African States/ Communauté économique des États de l'Afrique de l'Ouest

EDPRS Economic Development and Poverty Reduction Strategy

FI Financial institutions

FPA Fédération des Perimetres Autogeres

GAP Good Agricultural Practices

GGC Ghana Grains Council

GH₵ Ghanaian Cedi

GNI Gross National Income

GoR Government of Rwanda

GoS Government of Senegal

Ha Hectare

HACCP Hazard Analysis Critical Control Point

ISRA National Institute for Agricultural Research of Senegal

(Institut Sénégalais de Recherches Agricoles)

JTI Japanese Tobacco International

Kcal Kilo calories

LOL Land O’Lakes

MDG Millennium Development Goal

MSA MarketShare Associates

MT Metric tons

NGO Non-governmental organization

OB Outgrower business

OG Outgrower

PC Producer company

PCE Economic Growth Project (Projet Croissance Economique)

PFL Premium Foods Ltd.

PPP Public-Private Partnerships iii

PRACAS Accelerated Program for Agriculture in Senegal

(Programme de relance et d’accélération de la cadence de l’agriculture au Sénégal)

PROFIT+ Production, Finance, and Improved Technology Plus

RALIS Rwanda Agriculture and Livestock Inspection Services

RDCP Rwanda Dairy Competitiveness Program

RND Rwanda National Diary

SAED National Society of Land Management of the Senegal Delta and River Valley

(Société nationale d'Aménagement des Terres du Delta et de la Vallée du Fleuve Sénégal)

SAT Sinapi Aba Trust

SH Smallholder farmers

SWT Strength of weak ties

USAID United States Agency for International Development iv

ACKNOWLEDGEMENTS

The authors wish to thank the LEO COR at USAID, Kristin O’Planick, for her support of, and patience for, this study. Key feedback was also received from a variety of USAID staff in the Bureau for Food Security, with the valuable assistance of Su-san Pologruto and championing of Meredith Soule. This study is part of a series of studies on systemic change, a topic which provides a significant leap forward in our understanding of how market systems anticipate, respond, and adapt to change. We further thank Anna Garloch, the Program Manager of the LEO project for her support and multiple edits.

We acknowledge with gratitude the cooperation of the staff and management of the four projects included in these case stud-ies. Despite multiple iterations of questions and clarifications, all four Feed the Future implementation teams remained patient and helpful. Even greater appreciation goes to the multiple value chain actors and service providers, for their cooperation was entirely voluntary and they are not compensated for the many hours they spend speaking to consultants. Through multiple clarifications, there is always a possibility of errors. These are entirely the responsibility of the author.

Finally, we wish to acknowledge the contributions of Sally Oh, and William Vu, of the LEO project. They ensured that the researchers got where they needed to go when they needed to be there and made a rather rough document appear as clean as the one we hope you will enjoy reading.

EXECUTIVE SUMMARY

Feed the Future (FTF) is facilitating changes in core agricultural systems that contributes to more sustainable and scalable de-velopment objectives. This report summarizes the findings from research into four FtF projects, selected as illustrations of observable systemic change. The four projects are:

FTF Senegal Naatal Mbay, which has introduced various alterations to the prevailing model for contract farming of paddy rice, including a price discovery process that reduced uncertainty that in turn unleashed widespread investment by financial institutions and processors into the more beneficial contract farming system, as well as an increase and improvement in the services to value chain actors, particularly agricultural machinery leasing.

FTF Zambia Production, Finance, and Improved Technology (PROFIT) Plus, which is in the early stages of introducing changes in the structure of the rural input supply system through new aggregation models and agents, improving smallholder access to input and extension services. Interestingly, this has taken place in the context of two years of heavy drought, shifting behaviors from those that are revenue maximizing to those that are risk mitigating and resilience maximizing.

FTF Rwanda Dairy Competitiveness Program (RDCP) II, which has introduced quality grades and standards into the dairy industry both through support for more formal policy-level changes as well as through firm-led behav-iors and models that incentivize and reward for quality. Like Zambia, these changes are early in the systemic change process, but there are strong indications of imitation by other lead processors, independent replication, and that these behaviors and practices are beginning to become institutionalized and a ‘new normal’.

FTF Ghana Agricultural Development and Value Chain Enhancement (ADVANCE) II, which has supported the emergence of a relatively new actor in commodity value chains, the outgrower business; this is changing the net-work structure of input and output systems in the target areas, increasing smallholder access to quality inputs, financ-ing, and output markets.

The cases describe projects in four unique enabling business environments, with each systemic change in various stages of ma-turity. These cases also span a number of value chains: Zambia Profit + and Ghana ADVANCE II focus on maize; Senegal

Naatal Mbay focuses on paddy rice, and Rwanda RDCP II focuses on dairy, with a particular focus on the urban market.

What is Systemic Change?

Conceptualizing and defining systemic change is an evolving, fertile space in development. As discussed in Section I.B, there is no agreement within the market systems development field about how to neatly define systemic change. There is, however, a common emphasis on changes in the underlying structural elements of a system. These may include institutions, policies, be-havioral norms, networks, and perceptions. In commissioning these case studies, USAID targeted projects they believed had facilitated changes at the structural level, and the researchers then utilized qualitative methods (e.g. focus groups, key inform-ant interviews, document review) to assess those changes - supplemented with available project monitoring data when possible

- organized around two categories of indicators: buy-in and imitation. These categories, presented below, draw from the 2014 literature review on Evaluating Systems and Systemic Change for Inclusive Market Development, produced by LEO:

1. Buy-in1 indicators, which measure the degree to which market actors have taken ownership over the new business models, technologies, practices and behavior changes that were introduced and/or supported by the intervention.

Some examples of buy-in indicators include the following:

Adaptation or innovation to the original, program-sponsored model(s)

Continued, independent investment after program sponsorship ends

1 “Buy-in” refers to much more than a mental assent or philosophical agreement with project-promoted models, technologies or behaviors. It signifies evidence of ownership through significant investment of financial capital, other resources, time, and reputation.

Repeat behavior

Satisfaction with program-facilitated changes

2. Imitation indicators, which measure the scale or breadth of program-supported behavior change within a system.

There are two prominent examples of imitation indicators:

Crowding-in by other businesses that imitate program-sponsored business models originally adopted and demonstrated by businesses that collaborate with the implementer

Copying, mentioned less often than crowding-in, refers to imitation at the target beneficiary level by market actors (firms, farms, households or individuals) that imitate the new practices originally adopted and demon-strated by the target beneficiaries of the intervention.

Each case study explores the identified systemic change through these two lenses – buy-in and imitation. These two key do-mains of indicators have recently been expanded upon as LEO and others have further articulated elements of systemic change. This includes a greater focus on network and institutional structure, emergent patterns, and sensing changes at both the individual agent (e.g. single farmer, single firm, etc) and collective levels. Importantly, no project operated entirely outside the realm of subsidies – with some relying more or less on them, to support different elements, and at different stages of the process. This is area that deserves more attention in future studies, and as noted elsewhere in the report, supports ex-post as-sessments to better validate the sustainability of changes after projects end.

Why Focus on Systemic Change?

The concept of systemic change is gaining increasing attention in donor-funded market development projects - and under-standably so. Market development projects often involve investments of scarce donor resources in actors who are not part of the intended beneficiary group in order to make systems in which large numbers of the poor participate work better, and work better in ways which allow the poor to benefit. It is a legitimate and compelling question to ask: how can we be sure that in-vesting in getting actors within a system to do things, or organize themselves differently than they have until now, in order to achieve inclusive growth actually achieves these expectations? How can we be sure that inclusive growth will continue after the program is over? Simply put, donors are interested in systemic change because of their interest in enhancing the scalability and sustainability of development outcomes. In recent years, across USAID, systemic change has gained increasing attention: in

2014 the Agency released Local Systems: A Framework for Supporting Sustained Development, which explored the logic behind linking systemic change to enhanced development outcomes and outlined ten principles for engaging in local systems2.

The urgency of the scalability and sustainability goals in some ways has put the determination of whether systemic change has occurred, ahead of the question, what is systemic change and how does one make sure that it happens. This report is an initial effort to respond to both questions by looking forward at four cases and backwards at a body of literature to support a theory of change.

In the growing literature on innovation and systemic change there are two distinct paths. The first begins with the introduc-tion of a new way of doing or organizing things - an innovation - and asks how and with whom should this innovation be in-troduced and how and when can we determine that the introduced innovation becomes systemic, i.e. when forces adopting change outweigh forces opposing it. All the case studies fall into this category.

2 Access the framework at: https://www.usaid.gov/policy/local-systems-framework. For more on the relationship between local systems and market sys-tems, see ACDI/VOCA, 2016. Local Systems and Market Systems.: www.microlinks.org/library/local-systems-and-market-systems.

https://www.usaid.gov/policy/local-systems-framework https://www.microlinks.org/library/local-systems-and-market-systems

The second, called systems thinking, looks at how the process of change becomes systemic, i.e. how are new ideas, norms, and processes, drawn in by members of a network to disrupt a status quo in order to achieve greater growth and more inclu-sive growth. In systemic thought, issues including feedback loops, customer and SME churn (i.e. attrition) rates, flow rates

(information and finance), alignment of these factors across levels (micro, meso, macro), etc. all come in3.

Systems thinking, however important is not a focus of this report. This report focuses on the first set of questions - how an innovation in rules, norms, and or processes is introduced, what factors affect the rate of adoption of the innovation, and how can one determine when the innovation has acquired its own momentum or the point where forces within a system in favor of an innovation have become stronger than the forces supporting the status quo.

Each project4 case study (see Sections III A-D) focuses on a particular intervention that project management and the respec-tive USAID mission believed represented the best illustration of systemic change among multiple interventions and multiple projects. Together they illustrate elements of the theory of change elaborated later in this report.

Each project introduced an innovation that disrupted a status quo in a set of processes, norms and or relationships – such as in

Senegal, where the innovation involved revamping the prevailing model of contract farming in paddy rice. Each worked through actors within a system or network. These included actors within their respective value chains but also included a range of firms that provide services to, or markets for, those value chain actors. Each involved the introduction of an ‘innovation’ across rather than within a group, bridging groups across different functions in the value chain. Each described factors that either accelerated or slowed down the transfer of the innovation from one group to another – such as in Zambia, where the presence of a drought influenced the take-up of localized agro-dealer agents and community agro-dealer-run companies. In RDCP II in Rwanda, the

‘innovation’ was a policy, and reminds us of how quickly the status quo can change when policy changes or policy constraints are lifted. In Ghana, ADVANCE II had such an active level of copying and crowding in by actors who were not directly supported by the project that it suggests that the particular innovation - a change in the nature of the relationship between smallholders and middlemen - had likely reached the point where the forces supporting change had outweighed the advocates of the status quo.

The presence of all of the elements that enable an innovation to become systemic (such as disrupting the status quo in a set of processes, norms, relationships; working through local actors; introducing an innovation that bridges groups, etc) does not ensure that the change has become systemic. The authors of the four case studies, therefore, also looked for evidence that each of the four interventions demonstrated one or more steps in the process of change becoming systemic, as evidenced by the level of ‘buy-in’ by project stakeholders. These steps, while difficult to quantify, represent the sequence from introduction of an innovation to evidence of broad use and adaptation of the innovation - evidence that it had become systemic.

Table 1.

Evidence of Buy-in Naatal Mbay ADVANCE II PROFIT + RCDP

Satisfaction

Continued use

Adaptation of the model

Further Investments

Replication

As elaborated more fully in Section III, all four cases illustrate evidence of customer satisfaction and continued use by its in-tended clients (which is context specific, but includes farmers, processors, SMEs, etc). Naatal Mbay in Senegal demonstrated all the steps except for evidence of adaptation of the model. This is not surprising because in this case, the innovation was

3 See MarketShare Associates, 2016. Disrupting System Dynamics: A Framework for Understanding Systemic Change. www.microlinks.org/library/disrupting-sys-tem-dynamics-framework-understanding-systemic-changes.

4 Throughout this document, “project” is used in the generic sense to refer to donor-funded activities, rather than the USAID-specific defi-nition of this word.

http://www.microlinks.org/library/disrupting-system-dynamics-framework-understanding-systemic-changes http://www.microlinks.org/library/disrupting-system-dynamics-framework-understanding-systemic-changes buy-in by multiple actors into a common price discovery process; adaptation would mean non-acceptance of the innovation.

ADVANCE II showed strong evidence of all steps in the process including broad replication of at least parts of the upgraded relationship between aggregators as service providers and smallholder farmers. RCDP II did not evidence adaptation either and for the same reason. The innovation - the introduction of dairy standards - could only be adopted. Adaptation to the standards would have meant rejection of them; however, businesses up and down the value chain did have to adapt their busi-ness models and practices in order to respond to these emerging quality norms.

In Zambia, PROFIT + is a unique case that illustrates several of the key stages towards systemic change, with wide adaptation of the model. It also illustrates the role of the external environment in the rate of adoption of an innovation. In the

PROFIT+ case, two seasons of severe drought led participating stakeholders to make use of the innovation, in this case the placement of community level agro-dealers (CADs), but not as the project had initially intended. Instead of sourcing high cost, high yielding seed, fertilizer and crop protection inputs for their maize plots consistent with the vision of the project, smallholders used the CADs to source vegetable seeds and inputs for small livestock rearing. Initial analysis suggested that farmers able to access a range of inputs from retailers in their own village were demonstrating greater resilience to the drought conditions by diversifying their activities and avoiding the financial risk associated with high yielding maize seed, favoring recy-cling of their old seed instead.

I. BACKGROUND

Feed the Future (FTF) has made significant progress in providing technologies, market opportunities, and nutritional approaches to large numbers of rural people in FTF focus countries. However, FTF has a greater ambition than developing good service delivery models. USAID’s interventions are often designed to facilitate the creation of new market opportunities, farmer-market linkages, or channels for seed and fertilizer delivery, that—if successful—are “self-replicating” with no additional implied financial burden on either donor or host government. These “self-replicating” changes occur largely by identifying and facilitating new opportunities in which for-profit actors—whether traditional traders, or seed suppliers, or nucleus farm owners—are facilitated in taking advantage of new market opportunities that increase their own profits by opening up new opportunities for poor rural people. The motivation for the case studies, therefore, is to dig deeper than simple FTF results reporting, and identify, describe, and analyze strong case studies of FTF value chain programming significantly contributing to systemic change.

A. Objectives of the Study

The objective of this study was to support USAID’s Bureau for Food Security (BFS) to (i) identify countries or regions where

USAID has been instrumental in promoting systemic change; (ii) document in case studies the importance and if possible the impact of that systemic change; (iii) document in the same case studies factors and processes that led to systemic change; (iv) based on these case studies, suggest implications for future programming, including possible metrics for measuring systemic change; and (v) prepare short (one page or less) “success story” versions of each case study that both identifies the systemic change and its impact and—by telling the story of one or more poor rural people who have benefitted—puts a human face on articulating the “systemic change” approach of FTF.

The metrics element of the fourth objective was subsequently addressed more thoroughly in two companion LEO publica-tions: Guidelines for Monitoring, Evaluation, and Learning in Market Systems Development and Disrupting System Dynamics: A Framework for Understanding Systemic Change5. As such, it was not an explicit focus of these case studies.

B. Methodology

The four projects profiled in this report were identified by USAID and recommended to LEO to include in this study. Follow-ing this, the research team then collected and reviewed relevant project documents (e.g. annual performance reports, work plans, results reports, etc), had discussions with project management, technical staff, and associated USAID contacts, and then prepared for field research, which generally took place between May and August 2016, lasting 1-2 weeks in country.

Field research for each of the four cases involved interviews with the key actors to identify indications of buy-in and imitation, as well as focus group discussions with key beneficiary groups to ascertain change in resilience (approximated through diversifica-tion of crops) and welfare from adopting innovations and forming new relationships. Post-field work, these case studies and the overarching synthesis were then drafted. A webinar6 was hosted on September 8, 2016 to preview key findings and solicit feed-back from the general practitioner community.

Identifying and Measuring Systemic Change

As discussed in “Evaluating Systems and Systemic Change for Inclusive Market Development” (Dunn and Fowler, 2014) published by USAID through the LEO project, there is no agreement about how to define systemic change. Definitions include:

5 Both reports are available at www.microlinks.org/leo.

6 For a recording of the webinar and slides, visit https://www.microlinks.org/facilitating-systemic-change-insights-feed-future-programs-rwanda-senegal-ghana-and-zambia.

http://www.microlinks.org/leo https://www.microlinks.org/facilitating-systemic-change-insights-feed-future-programs-rwanda-senegal-ghana-and-zambia https://www.microlinks.org/facilitating-systemic-change-insights-feed-future-programs-rwanda-senegal-ghana-and-zambia

“[S]hifts in patterns (similarities and differences) of system relationships, boundaries, focus, timing, events and behaviors over time and space.” (Parsons and Hargreaves, 2009)

“Transformations in the structure or dynamics of a system that leads to impacts on large numbers of people, either in their material conditions or in their behavior.” (Osorio-Cortes and Jenal, 2013)

“Change in the underlying causes of market system performance – typically in the rules and supporting functions – that can bring about more effective, sustainable and inclusive functioning of the market system.” (DFID and SDC, 2008)

“Systems are groups of agents that interact with each other, producing emergent patterns of collective behavior. They are dynamic – constantly changing – as agents are constantly acting, producing emergent patterns that in turn influence indi-vidual behaviors in a never-ending feedback loop. Because systems are constantly changing, “systemic change” refers to the diversion of a system down a new evolutionary path, not the introduction of movement where there was none previ-ously (there is always movement). We can observe indications that systems are changing at two levels: behavior changes and characteristics of individual agents (e.g. people, businesses, other market actors); and collective shifts in interactions between individual agents. Systems are constantly changing in both positive and negative ways. For the purposes of mar-ket systems development, positive systemic changes result in more sustainable, inclusive benefits to agents in the system.”

(MarketShare Associates, 2016).

These and other definitions of systemic change emphasize the need to change the underlying structural elements of a system.

These may include institutions, policies, behavioral norms, and perceptions. The Donor Committee for Enterprise Develop-ment (DCED, 2014) further identifies three characteristics of systemic change: scale (“Systemic changes influence and benefit a large number of people who were not directly involved in the original intervention”), sustainability (“Systemic changes continue past the end of the programme, without further external assistance”) and resilience (“Market players can adapt models and insti-tutions to continue delivering pro-poor growth as the market and external environment changes”).

In addition, this report posits that a social or economic system in which systemic change has taken place should be fundamentally different as a result of the change, i.e. transformative. While systemic change can be positive, neutral, or negative, at least for some of the actors in the system, development practitioners aim for these observed changes to contribute to positive development out-comes, manifesting in the increased resilience and or welfare of individuals and communities. Finally, systemic change is inher-ently disruptive. In order for change to become systemic, it must ‘disrupt’ a status quo whether in the relationships, rules, pro-cesses, technologies, network, norms and or behaviors of actors within a system to the point where the forces favoring a change exceed those seeking to maintain it. These two particular characteristics are explored further in the next section.

In conducting these case studies and profiling examples of systemic change, these various unifying elements of systemic change were incorporated. In capturing indications of change, as presented above, this research focused on those presented in

Fowler and Dunn, 2014: buy-in, and imitation.

II. SPOTLIGHT ON INNOVATION: A

THEORY OF CHANGE FOR SYSTEMIC

CHANGE

Building on the findings from the four case studies, as well of the broader body of literature on the topic of systemic change, this paper posits the following theory as it relates to how systemic changes occur:

Innovations introduced within a system become self-replicating and capable of disrupting a status quo without further external force, ergo systemic, through the transfer of an innovation between groups characterized by weak ties between them. Innovations are spread across ‘bridges’ from actors with prior knowledge of a new process, practice, technology or behavior, however recently acquired, to actors who would benefit from adopting it. The rate of adoption of innovations is determined by characteristics of the individuals forming the bridge, social norms and customs, and environmental

(physical, climate, and business enabling) factors. Members of a group or network may replicate and adapt innovations without external support once said innovation is adopted by 16-20%7 of their members.

In reviewing the literature on systemic change, innovation diffusion and the review of the four case studies included in this paper, the authors formulated a theory of change that should guide the design and implementation of any projects leveraging the power of private sector actors to disseminate innovations on a sustainable and systemic basis.

A social or economic system in which systemic change has taken place should be fundamentally different as a result of the change. While change can be positive or negative, innovations in rules, norms networks, or processes in order to achieve do-nor identified objectives, should result the greater resilience and or welfare of individuals and communities affected by this change. Thus as stated above, in order for change to become systemic, it often will ‘disrupt’ a status quo. In many cases, change becomes systemic when the factors supporting a new evolutionary path overcome the factors supporting the status quo. While never neutral, because the end game must be significantly different from the status quo ante, it can be good or bad for some participants in a system. Importantly, systemic change can occur with or without an external intervention.

Much of the literature around systemic change in market development begins its analysis phase after the introduction of an innovation. Yet, as a complement, a comprehensive theory of change must begin sooner in the change process and identify where the end is of any need for continued subsidy or support to accelerate and render systemic a desired change.

While systemic change can be positive or negative, depending on the power within a system to protect or overturn an ineffi-cient status quo, innovation, at least in this context, refers to change that results in higher economic and social benefits and is therefore a positive force. Part of the systemic change challenge is that innovation begins with an agent from within or out-side a system, with the motivation to disrupt a status quo in norms/rules, net-works, product, information or service flows, and ends with a large enough mass of actors adopting an innovation to ensure that forces favoring the new condi-tion have surpassed forces invested in maintaining the status quo. A compelling theory of change must begin with the conditions that cause an agent or agents to disrupt the status quo, a method for the identification of those individuals and their motivation, the mechanism by which innovations are spread, and ulti-mately, a determination of the tipping point beyond which the forces of change have surpassed those favoring the status quo. It is at this last point that subsidy or support is no longer required, at least for a particular innovation.

This TOC therefore, needs to answer several questions. It must be able to explain most if not all cases of introduced innova-tion becoming systemic for objectives from resilience to economic growth and across multiple environments. It must address how in the process of becoming systemic, it acquires momentum and becomes self-replicating. At the very least the TOC8 must answer the following questions:

1. How do we identify system actors motivated to disrupt the status quo by innovating within a system?

2. How is the innovation spread beyond the innovator to adopters?

3. What are the factors that affect the rate at which innovation occurs and the rate at which said innovation is taken up by other actors?

7 This hypothesis is based on assumptions from innovation diffusion theory (Rogers, 1962), which posits that once the innovators, the early adopters, and the early majority of a cohort, change their behavior, the remainder of the set of the early and late majority will observe and adopt the new behavior of their own volition.

8 A distinction should be made between a theory for how change occurs, which the synthesis of these studies is attempting to do, and a theory of how a change process is integrated into networks. The latter is much more dynamic but outside the scope of this report.

Innovation: the introduction of new products, rules, norms, organizational models or networks of actors in order to generate inclusive economic growth and/or social benefits. Economic growth is the optimization of the utilization of factors and the measure of success is how well the factor utiliza-tion is optimized.

https://en.wikipedia.org/wiki/Optimization https://en.wikipedia.org/wiki/Utilization https://en.wikipedia.org/wiki/Factor_analysis

4. How do we measure progress from the introduction of an innovation to the point at which the introduced change becomes systemic? What are the observable steps in the adoption and replication process, and;

5. How long will it take and what percentage of a population do we have to reach before an innovation has become sys-temic and we can move on to the next thing?

Elements of this process have been identified and integrated in previous LEO materials, notably the concepts of leverage and momentum, comfort and learning zones and the diffusion theory bell curve9. Our TOC will draw from a literature review on both systemic change and on the introduction and diffusion of innovation. The remainder of this section provides a theoreti-cal foundation for each of the above questions.

1. How do we identify system actors motivated to disrupt the status quo by innovating within a system?

Market facilitation approaches often emphasize tactics that enable buyers and sellers to learn to cooperate more effectively10.

But who among the potential large set of vertically linked firms is likely to introduce or be receptive to introduced innovation?

Mark Grannovetter in his seminal work the Strength of Weak Ties11 and the Strength of Weak Ties: A Network Theory Revisited12 de-scribes the innovator in a system as "the innovativeness of [actors] is shackled by vested intellectual interests (or perspectives) then new ideas must emanate from the margins of the network."

Otherwise stated, membership in groups (such as an association of traders, processors, input companies or farmers) tends to stifle innovation, maintaining an internal status quo. This is one reason that donor-funded entities so often introduce innova-tion, as they are external to any networks within the market system. But external introduction of innovation is a one-off activ-ity. The implication for a development partner is that innovation must be introduced through an actor who is somehow mar-ginal to the group in which they participate. An actor, whether a miller, an input supplier, or a wholesaler, becomes marginal to a group or network when she/he faces incentives to overcome a status quo situation. She/he might be trying to capture mar-ket share from other members of her/his group, be a member of a different ethnic group than others, have a different educa-tional level or other factors that predispose her/him to overcome a status quo condition. With some exceptions, the actor in a group with the greatest market share will be less inclined to innovate; their investment in the status quo has worked well for them so far.

2. How are innovations spread beyond the innovator to adopters?

Once the challenge of selecting a firm or firms with the incentives to introduce an innovation, the systemic change program must identify to whom a particular technology should be introduced. Many programs continue to try and train as close to

100% of a targeted population as possible for social equity reasons. This approach tends to be costly, and less effective than those that target assistance to individuals most likely to adopt a particular technology and disseminate that technology within her or his group. Here the notion of bridges as described by Grannovetter is important13. Bridges are weak ties between members of two unlike groups, one who has access to information and or technologies of value to the other and the incentives to disseminate;

the second, who is more disposed than other members of her or his group to test, and if successful, adopt the new technology

(see figure 1). In the context of the four cases in this paper the holder of innovation could be a processor, an input or veterinary services provider, a lead firm, or nucleus farmer or outgrower business. In some instances, an individual can serve as a bridge be-tween two groups. The outgrower businesses (OB) in the ADVANCE II project are an illustration of this.

9 Rogers, Everett M. 2003. Diffusion of innovations. New York: Free Press.

10 For more on the facilitation approach, see www.microlinks.org/good-practice-center/value-chain-wiki/facilitation. LEO has also produced a number of program-focused learning tools that build capacity in staff to apply the facilitation approach and specific intervention tactics. See www.micro-links.org/library/market-systems-development-cartoon-based-learning-tools.

11 Granovetter, Mark, The Strength of Weak Ties (1973). American Journal of Sociology, Vol. 78, Issue 6, p. 1360-13 1973. Available at

SSRN: http://ssrn.com/abstract=1504479 12 Granovetter, Mark, The Strength of Weak Ties: A Network Theory Revisited, (1983). Sociological Theory, Vol. 1 (1983), pp. 201-23

13 Idem http://www.microlinks.org/good-practice-center/value-chain-wiki/facilitation http://www.microlinks.org/library/market-systems-development-cartoon-based-learning-tools http://www.microlinks.org/library/market-systems-development-cartoon-based-learning-tools http://ssrn.com/abstract=1504479

To effectively use limited resources, the firm introducing change must identify whom in a receiving group is most inclined to adopt and if successful, disseminate within her/his own group.

Here diffusion theory is helpful. Diffusion theory posits that within any group or network, members of the group respond to, and adopt innova-tions differently. Members can be classified as innovators, early adopters, early and late majorities and laggards. Surprisingly the distri-bution of a population around these types is remarkably consistent across types of networks, culture, age, etc. Innovation diffusion the-ory14 states that approximately 2.5% of any population are predisposed to innovate (e.g. try new technologies, before they have observed the re-sults). Innovation diffusion theory suggests that a program can be more effective when it identifies the innovators with from within a network, introduces a new, or bolsters an existing, innovation, and supports inno-vators in disseminating (e.g. ‘diffusing’) this innovation to other members of her/his network. The community agro-dealers (CAD) in the Zambia

PROFIT+ project illustrate this, but so does any bridge between a private actor who has successfully identified the innovators with in any group to which she/he wishes to introduce a technology. Innovators within a group are not difficult to identify because most of the members of a group or network already know who they are. Market system development programs aiming to facilitate systemic change should take care to ensure that members of the group identify the innovators and that they function as one side of the technology transfer bridge; this can both save resources and accelerate the diffusion to the larger group.

3. What are the factors that affect the rate at which innovation occurs and the rate at which said innovation is taken up by other actors?

Each of the four cases in this study illustrate key elements of systemic change. Each however varies by the level and rate of copying and crowding in by actors not directly supported by the particular project. We assume that the rate of adoption of innovations, and importantly for systemic change, the rate of imitation and crowding in by other actors, will vary from case to case. But why? And are there ways to change, i.e. accelerate change within a system?

Rogers (1995) developed a model for the adoption of innovation within a system that illustrates the variables in this process

(see figure 2). These include individuals, social systems and perceptions about a particular innovation. All four projects in this study intervene at the individual level using strategic subsidies, cost sharing grants, as well as heightened recognition and activi-ties to elevate her/his status from adoption, to positively affect the individual’s (innovator) uptake of innovation. Likewise, programs use demonstrations to increase an individual’s perception of an innovation; demonstration plots and field days are a common mechanism to achieve this. An additional factor, not included in Roger’s framework but important in a developing economic context and illustrated in the four cases in this study are environmental factors. Included in environmental factors are the physical, political, economic, and climate.

14 Rogers, Everett M. 2003. Diffusion of innovations. New York: Free Press.

Figure 1: Diffusion Theory Bell Curve

Figure 2: Rogers’ Model for Adoption of Innovation in a System

Each of the four cases provides strong evidence of systemic change each was affected by, or importantly, had an effect upon, the environment. The Rwanda dairy case is an illustration of how the introduction of standards resulted in an industry policy change that in turn affected the adoption of standards across the whole industry. The Zambia case illustrates how in the face of persistent drought the rate of innovation adoption is slowed, even while strengthening the resilience of current adopters; in the absence of drought conditions, Zambia might be expected to demonstrate a higher level of adoption, imitation and crowding in. Ghana and Senegal experience relatively normal rainfall patterns; both countries suffer from an absence of a robust private sector seed market and low levels of high quality seed use15. Yet the rate of imitation and crowding in by multiple private sector actors in Ghana seems significantly higher than in Senegal. Since imitation concerns actors not directly targeted by the project, this difference cannot be attributed to differences in management or implementation. Imitation behavior by actors not directly supported by the project are exogenous to differences in management.

The causes of this difference are beyond the scope of these studies to determine; one hypothesis is that the difference is due to the business enabling environment. A quick comparison of the World Bank’s Doing Business indicators (see table 2) shows much lower ranks for the Rwanda and Zambia, but also shows a 20-point difference in Ghana’s favor over Senegal; lower scores reflect more ease in doing business. Table 2. Ease of Doing Business

Innovation diffusion theory is useful at identifying where a project can intervene to accelerate the adoption of innovation as well as understanding factors outside of the project’s control that will affect the rate of adoption and dissemination of a particular innovation. Weaknesses of the model, however, are in its failure to describe what happens when innovations are not taken up and how to mitigate this. Figure 3 below lays out empirically observable steps in the innovation acceptance and diffusion process.

15 The Senegal Naatal Mbay project is implementing a seed multiplication activity but it is too early to assess the effectiveness of this intervention at lifting the seed availability constraint for large numbers of farmers.

Country Ease of Doing Business Rank

Rwanda 62

Zambia 97

Ghana 114

Senegal 153

Source: www.doingbusiness.org/

4. How do we measure progress from the introduction of an innovation to the point at which the introduced change becomes systemic? What are the observable steps in the adoption and replication process?

Figure 3 lays out a five step process to assess progress towards buy-in, which is one key indication of systemic change. These steps can be built into a program’s internal M&E system, permitting users to monitor progress towards buy-in as it is taking place. Progress towards and beyond each of the five steps can be monitored using qualitative data collection techniques. If weak ties bridges identified above reject an innovation or there is no evidence that the network or group is investing in the innovation, project staff can dig deeper to understand why and to modify implementation strategy based on what they find when they dig.

5. How long will it take and what percentage of a population do we have to reach before an innovation has become systemic and we can move on to the next thing?

Figure 4 below and basic calculus would suggest that the ‘tipping point’ occurs where the slope of the blue curve changes from concave to convex to its origin - or when the early majority be-gins to replicate the behaviors of the early adopters. If this is in-deed the case, the targeted results of many market systems devel-opment programs in terms of outreach are excessive – in other words, projects only need to aspire to outreach targets that ap-proximate the ‘tipping point’, not canvas the entire population.

There are also numerous factors that break ties between one subpopulation and another; this phenomena, if established would warrant interventions in which the tipping point was achieved within two or more subpopulations, with the expecta-tion that the status quo was overcome and the desired change would continue to spread. Empirical, post-project research is needed before this theoretical tipping point can be confirmed in the field.

Figure 3. Scale of Evidence of Buy -In

Rogers16 diffusion of innovations graph illustrates Malcolm Gladwell's tipping point notion. Once innovators and all the early adopters have tried, adapted and replicated a new innovation for their own benefit, other members of their group or net-work will follow suit, so long as the resources required to adopt a given innovation are still available. This is what Mal-colm Gladwell refers to as the ‘tipping point.’ Beyond this point, if members of the early majority have the same access to the bundle of services that the innovators and early adopters had, they should, in principle, go out and invest in the same innovation. This of course assumes that actors who are part of the early majority can access that bundle of goods and or services under the same terms and conditions. This point of inflection in the adjoining figure occurs after when approximately all early adopters (i.e., 16-18% of a population) have adopted and are replicating a particular innovation.

From a development resource management perspective, be-yond this point, funds invested to ensure wide adoption of a particular change within a particular environment may not be necessary. This needs to be tested empirically of course. An ex-ception to this hypothesis is if the ties between subgroups of a population lack bridges (see above). This can occur among groups in conflict or severe isolation from one another.

This appears to be supported by the up-take of improved seed by farmers. Fig-ure 5 illustrates uptake of improved seed by farmers in a number of Sub-Sa-haran countries. In this graph, use of improved seed by country shows a sig-nificant cluster at or below 20% and use of private proprietary seed far lower.

Three countries, however, Senegal, Kenya, and South Africa, have uptake of improved seed at or above 50%.

There are many variables in play in this graph but in West African markets, use of improved seed barely reaches 5%, and seed markets flounder; in most of the COMESA countries, hybrid seed use has surpassed 20% and these mar-kets are growing, mostly without exter-nal subsidy.

From this theoretical foundation and an evaluation of the four cases we can hypothesize a Theory of Change for Systemic

Change as:

Innovations introduced within a system become self-replicating and capable of disrupting a status quo without further external force, ergo systemic, through the transfer of an innovation between groups characterized by weak ties between them. Innovations are spread across ‘bridges’ from actors

16 Ibid

Figure 4. Innovation Diffusion and Tipping

Points

Figure 5: Improved Seed Uptake in Sub Saharan Africa with prior knowledge of a new process, practice, technology or behavior, however recently acquired, to actors who would benefit from adopting it. The rate of adoption of innovations is determined by characteristics of the individuals forming the bridge, social norms and customs, and environmental

(physical, climate, and business enabling) factors. Members of a group or network may replicate and adapt innovations without external support once said innovation is adopted by 16-20%17 of their members.

17 This 16%-20% figure requires additional empirical testing.

III. COUNTRY CASE STUDIES

A. SENEGAL: SYSTEMIC CHANGE CASE STUDY ON NAATAL

MBAY

SUMMARY

Naatal Mbay is a large-scale, market systems development project targeting the rice, maize, and millet value chains in the Sene-gal River Valley and the South Forest Zone. This case study highlights two key areas in which the project is facilitating changes at a systemic level: revamping the prevailing contract farming system in paddy rice by introducing a more inclusive, competi-tive model, and introducing agricultural equipment leasing into a new geographic area, tailored for a new category of clients

(millers and processors).

About the project: Naatal Mbay is the successor to the Project Croissance

Economique (PCE), which was launched in April 2009 and ended in May 2015.

PCE focused on the rice, maize and millet value chains, targeting the Senegal

River Valley for…