World Development Vol. 40, No. 1, pp.

96–109, 2012
Ó 2011 Elsevier Ltd. All rights reserved
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev
doi:10.1016/j.worlddev.2011.05.017

Can Public–Private Partnerships Leverage Private Investment

in Agricultural Value Chains in Africa? A Preliminary Review
COLIN POULTON
School of Oriental and African Studies (SOAS), UK

and

JON MACARTNEY *
Independent Consultant
Summary. — Public–private partnerships (PPPs) may be one way of increasing the level of private sector investment into poorly per-
forming agricultural value chains. This paper considers a range of PPP mechanisms that respond to different market failures affecting
such chains and draws on principal–agent theory to illustrate the challenges. It reviews emerging experience with a number of these
mechanisms along with experience from other sectors that may shed light on “generic” problems of implementing PPPs in Africa. While
finding some positive impacts on investment, it notes that state failures can also undermine PPP effectiveness. As the evidence base is still
limited, it calls on organizations promoting innovative PPPs to disclose available information for critical examination.
Ó 2011 Elsevier Ltd. All rights reserved.

Key words — market failure, state failure, public private partnerships (PPP), agriculture, Africa

1. INTRODUCTION sibility for design and execution”. All these definitions

Many parts of African agriculture may be thought of as
struggling between market and state failure. There has been
some disappointment with the private sector response to
agricultural market liberalization, which is perhaps best
described as “uneven”. Thus after a period when the major
policy emphasis was on getting the state out of agricultural
marketing activity in Africa, over the past decade or so there
has been a search for a more active role for the state that goes
beyond the creation of a basic enabling environment toward
encouraging market development. Simultaneously, there is
ongoing disappointment with the performance of public sector
service delivery, including in—but not restricted to—the
agricultural sector. This has prompted a search for ways to
bring the private sector into provision of services that were
traditionally considered to be the preserve of state agencies.
This paper reviews a variety of so-called public–private
partnerships designed to leverage private investment both
where existing participation in value chains is considered too
limited and into the provision of services traditionally
provided by the public sector. The scope of the term
“public–private partnership” (PPP) is subject to considerable
debate (Hodge & Greve, 2007). Hodge and Greve (2007) cite
a definition from Van Ham and Koppenjan (2001, p. 598) of
“cooperation of some sort of durability between public and
private actors in which they jointly develop products and ser-
vices and share risks, costs, and resources which are connected
with these products”. Narrod et al. (2009, p. 10) define a PPP
as “a cooperative venture between the public and private
sectors built on the expertise of each partner that best meets
clearly defined goals through the appropriate allocation of re-
sources, risks and rewards”. According to the World Eco-
nomic Forum definition, cited by Warner, Kahan, and Lehel
(2008, p. 9), PPPs entail “reciprocal obligations and mutual
accountability, voluntary or contractual relationships, the
sharing of investment and reputational risks, and joint respon-
96
emphasize new investment facilitated by a cooperative ap-
proach and risk sharing.
Through PPPs the state (or other “public” actor, such as a
donor or nonprofit foundation) seeks to align the incentives
facing private sector actors with public policy goals, for exam-
ple the provision of public goods or of services to poor groups.
In this paper we focus on cases where the mechanism formally
used to seek this alignment is some type of contract.
In recent years there has been growing interest in possible
PPP mechanisms to improve the performance of poorly func-
tioning agricultural value chains in Africa and elsewhere. Thus,
World Bank (2007) notes that PPPs “offer much potential” for
stimulating innovation, highlighting biotechnology research
and value chains for high value agricultural products as two
areas of particular promise. Langyintuo et al. (2010) argue that
seed systems in southern and eastern Africa would benefit from
support from PPP interventions. However, in practice there has
been rather less action: many interventions are still at proof-of-
concept stage. Perhaps understandably, there have been even
fewer formal evaluations. Nevertheless, it is noteworthy how
little information on innovative PPPs for agricultural value
chains is readily available. This paper does not provide an
exhaustive review of available evidence, but it does set out what
the authors believe to be important general lessons for Africa
from experience so far. One objective is to stimulate deeper
* A longer version of this paper was commissioned by the Trade and
Markets Division of FAO under the EU-funded All ACP Agricultural
Commodities Programme. The authors are grateful for comments on
the paper from participants at a workshop at FAO, Rome in June
2010. Sections of the paper also benefited from research assistance from
Elena Baglioni and from discussions with Mark Thomas (manager of
DFID’s Food Retail Industry Challenge Fund), Brinda Ganguly
(Rockefeller Foundation, New York) and Akin Adesina (AGRA,
Nairobi). Final revision accepted: April 12, 2011.
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
debate on the potential of PPPs in the hope of bringing infor-
mation currently buried in gray literature out into the full pub-
lic arena.
In this paper some reference is made to PPPs in agricul-
tural value chains in other continents, but experience with
PPPs in agricultural value chains in Africa is also compared
with experience within nonagricultural sectors in Africa.
Much of the literature here is quite critical of PPPs in prac-
tice and sometimes also as a concept (foisted by donors on
unconvinced governments for ideological reasons where priv-
atization of state activities was not considered a viable op-
tion). Nonagricultural examples are considered relevant
here because the low capacity of many African states and of-
ten ambiguous attitudes toward private sector-led develop-
ment among political elites can be expected to affect public
engagement with the private sector irrespective of sectoral
boundaries.
The paper proceeds as follows: In Section 2 reasons for the
poor functioning of agricultural value chains are briefly dis-
cussed, leading to an understanding of the types of market
(and associated state) failures that PPP mechanisms may be
designed to address. Section 3 presents a typology of PPP
mechanisms that is used to structure later sections. Section 4
draws on principal–agent theory to highlight the major design
challenges facing PPP mechanisms. Sections 5–8 review expe-
rience with different mechanisms both in nonagricultural and
agricultural contexts to address the following questions:
 Can PPPs remove or alleviate binding constraints on pri-
vate investment?
 Can contractual arrangements between state and private
sector actors incentivize the latter to provide supply chain
services in ways that meet public policy objectives?
Finally, Section 9 concludes.
2. REASONS FOR POOR FUNCTIONING OF AGRI-
CULTURAL VALUE CHAINS
Following agricultural market liberalization in Africa pri-
vate traders have taken up many opportunities to purchase
output from producers, although levels of entry and compe-
tition vary across geographic space. In general, private sector
participation in the provision of pre-harvest services has been
more limited than participation in output marketing. How-
ever, there are significant differences across crop groups, with
the incentives for investment in service provision for tradi-
tional export cash crops generally much stronger than for
food crops (Poulton, Dorward, & Kydd, 2010). Private
investment in crop storage has been another “disappointing”
area, contributing to increased price volatility post-liberaliza-
tion.
The reasons for these outcomes remain contested. Jayne,
Govereh, Mwanaumo, Nyoro, and Chapoto (2002) argue
that states have not fully withdrawn from many input and
output markets; their continued presence—and/or the contin-
ued threat of state intervention—discourages private invest-
ment. Commission for Africa (2005) and Jayne, Govereh,
Wanzala, and Demeke (2003) emphasize the impact of low
public investment in basic infrastructure on private invest-
ment opportunities in agricultural marketing. Kherallah,
Delgado, Gabre-Madhin, Minot, and Johnson (2000) and
Fafchamps (2004) argue that important institutions required
to support efficient private markets have not been established.
Finally, according to Poulton, Kydd, and Dorward (2006),
important coordination issues have to be tackled to overcome
“low level equilibrium traps” constraining agricultural
97
production and marketing activity in many parts of rural
Africa. Meanwhile, in important export markets, consumers,
retailers, and legislators are demanding greater assurances
about produce quality and safety. This makes it difficult for
African supply chains, especially those relying on large num-
bers of smallholder producers, to stay competitive (Henson &
Reardon, 2005; Otsuki, Wilson, & Sewadeh, 2001), thus
discouraging investment.
As greater emphasis has been placed on the role of transac-
tion costs and the associated risks in constraining private sec-
tor activity in African agricultural markets, so there has been a
search for ways in which public agencies might share some of
these costs and risks. Many of the interventions discussed in
this paper have this objective. However, high transaction costs
are not the only constraint to greater private sector investment
in African agricultural markets, nor even necessarily the bind-
ing constraint. PPPs will only encourage greater private invest-
ment where they tackle the binding constraint to such
investment. While a sharing of transactions costs and risks
could partly compensate for high costs due to an unstable
macro-economic environment or inadequate physical infra-
structure, it is unlikely to stimulate greater private investment
where unpredictable state policies are what currently discour-
age such investment.
Table 1 summarizes a range of market failures affecting
African agricultural markets and indicates where some form
of PPP might stimulate greater private involvement. Note that
there may also be other public interventions that can stimulate
greater private sector participation. For example, publicly-
funded marketing information or business development ser-
vices may reduce barriers to entry in agricultural markets.
However, it is only interventions where a state agency enters
into some form of contractual relationship with a private sec-
tor partner(s) that are defined as PPPs and reviewed in this pa-
per.
Meanwhile, some PPPs attempt to stimulate private involve-
ment in the provision of services traditionally provided by the
public sector (row 2). Often, the original rationale for public
sector provision was market failure, emanating from the pub-
lic or merit good attributes of the services in question. How-
ever, where state provision is also perceived to have failed,
innovative ways are being sought to bring the private sector
in after all.
3. PUBLIC–PRIVATE PARTNERSHIPS: A TYPOLOGY
Table 2 categorizes PPPs according to the type of investment
that they seek to stimulate or other role that they might play.
While in practice the boundaries between categories can be
blurred, the rows in Table 2 are intended to correspond to
those in Table 1. Thus, for example, PPPs to develop new prod-
ucts and services principally tackle problems of barriers to en-
try (row 3 in both tables). Table 2 identifies sectors outside
agriculture where the relevant mechanisms have been used, as
well as applications related to agricultural supply chains.
Two details of Table 2 may require more explanation at this
point. Firstly, row 2 distinguishes (a) contracts for delivery of
services where public sector financing is expected to be re-
quired for the foreseeable future, from (b) good or services
where, after a period of improved access (hence greater famil-
iarity), consumers’ willingness to pay might be sufficient to
sustain a commercial service. In the latter case, public funds
may be used to stimulate demand in the short-medium term.
This, then, represents an intermediate case between contract-
ing out of a public service and support for the development
98 WORLD DEVELOPMENT

Table 1. Sources of failure in agricultural markets

Source of market failure: What is the constraint to private sector involvement? Agricultural examples Possible PPP solutions
1. Lack of enabling Unstable macro-economic environment; – –
environment
Inadequate physical infrastructure; Rural roads, irrigation Provision and/or
infrastructure maintenance of infrastructure
Weak property rights and/or contract enforcement; – –
State as problem (crowding out, unpredictable policies) – –
2. Public or Merit Market under-provides due to:
goods
-Nonexcludability, nonsubtractability OPV seed research Contracting out service
delivery
-Social benefits exceed private benefits Fertilizer (e.g., if food price falls) Voucher schemes
-Lack of effective demand Extension
3. Barriers to entry Lack of access to:
Capital Stockists, Seed companies Loan guarantees
-Technical knowledge – –
-Market information – –
High fixed costs/risks Biotechnology research, credit, Risk sharing schemes
smallholder engagement in high
value commodity chains
4. Coordination failures Asymmetric information, no mechanism to Complementary investments in Deliberative fora
enforce commitments; lack of trust service provision to smallholders
or along value chain

Table 2. A typology of PPPs for leveraging private sector investment

Category
1. Capital Investment
(and Maintenance)
2. Service Delivery
(ongoing)
3. New Products and
Services (“bottom of the
pyramid”)
4. Coordination
(Deliberative fora)
Sub-type Nonagriculture Agriculture
Roads Rural roads
Prisons Irrigation
Schools Market Places
Hospitals Rural ICT
a) Service Contracts Water Extension contracting
Solid waste
Primary health care
b) Demand stimulation Bednet distribution Fertilizer vouchers
Extension vouchers
a) Investment Funds Neglected health research Smallholder inclusion in high value
chains
b) IPR deals Seed varietal research
c) Loan Guarantee Funds – Guarantees to banks for loans to
stockists, seed companies, farmers
Health investment funds Stakeholders’ fora (whole value chain)

of new products and services by the private sector (row 3).
Distinguishing service delivery and new products development
in this way highlights the importance of considering exit strat-
egies for the public sector from PPPs.
Secondly, deliberative fora (the final row of Table 2) are a
bit different in nature from the other mechanisms discussed
in this paper. Such fora bring together stakeholders from pub-
lic and private sectors for dialog that, it is hoped, will generate
mutual understanding and trust and thereby lead to joint ac-
tion to strengthen value chains. This joint action could include
the types of contractual arrangements discussed elsewhere in
the paper. However, it could also include coordinated actions
by private players, where the role of the state is limited to that
of discussion facilitator, or private investments that are
encouraged by changes made within the policy or institutional
environment as a result of the forum deliberations.
4. THEORETICAL CONSIDERATIONS
Principal–agent (P–A) theory can shed light on the chal-
lenges inherent in devising and managing PPPs. Assume that
the state (or other “public” actor) is the principal, while the
agent is the private company, partnership, or consortium
that the state contracts with. Formally, the state wishes to
harness the capacity (human and investment), entrepreneur-
ship and innovation of the private sector “agent” to achieve
public policy goals. However, it has to recognize that private
sector “agents” (1) have their own objectives and (2) will
only enter into PPP agreements that will further these objec-
tives. Specifically, their expected “utility” from PPP agree-
ments should exceed the opportunity cost of the resources
that they will have to contribute. (This is the agent’s reserva-
tion position).
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
To illustrate the divergence in objectives, assume that the
private sector ultimately seeks to maximize profits. This can
encourage firms to cut costs (raise efficiency), innovate, or ex-
pand service coverage (in such a way that the total size of the
market increases). PPP agreements aim to harness such entre-
preneurial energy in pursuit of public goals. However, firms
only act like this if external conditions (market competi-
tion—or, failing that, the incentives provided by a contract)
demand it. Alternatively, they might seek profit through
increasing margins, reducing quality, or displacing competi-
tors. A contract with a public agency could assist them to
do any of these things.
As a result some observers are wary of PPP agreements,
which could simply squander public resources on the enlarge-
ment of private profit, with no improvement in service deliv-
ery. From the perspective of P–A theory, however, the
challenge facing the state is to:
Identify appropriate private sector players to contract with.
Through the terms of the contract, structure incentives to
encourage the private “agent” to act in line with the “public
interest”.
Achieve all this at reasonable cost.
The problem is asymmetric information in the presence of
risk. If the state had perfect information on the capability
and motives of potential private sector partners prior to
signing a contract and on their actions and motivations dur-
ing its implementation, contracting would be straightfor-
ward. Unfortunately, these things are at least partly
hidden from everybody except the senior management of
the firm itself.
Hart and Holmstrom (1987) distinguish three types of P–A
models. Under adverse selection the main challenge facing the
principal is to choose an agent with the capability and motiva-
tion to do what the principal wants. Under moral hazard (hid-
den action), the principal cannot observe the actions of the
agent, although it can observe the outcomes. Under moral
hazard (hidden information), the principal observes the ac-
tions of the agent, but cannot observe the contingencies under
which they were taken.
In all cases risk factors beyond the agent’s control affect the
relationship between the agent’s action and the observed out-
come. Hence, the principal cannot be sure whether an undesir-
able outcome results from opportunistic actions or decisions
by the agent or from adverse circumstances. The combination
of the asymmetric information constraint and the risk factors
restricts the principal’s choice of indicators for inclusion in the
contract: what is observable and, therefore, measurable? How-
ever, use of “second-best” indicators can distort the incentives
faced by the agent, so considerable care has to be taken in con-
tract design. Implementation could produce poor outcomes
(i.e., not contributing to public policy goals) simply because
the contract was poorly specified.
Four additional issues are worthy of note. Firstly, P–A
models generally assume that principals have a reasonable
understanding of the “utility functions” and “reservation posi-
tions” of the population of agents with whom they might do
business. 1 Indeed, the principal maximizes his/her expected
profit by providing a set of incentives that will allow the agent
to just achieve his/her reservation position. While some public
agencies may employ personnel with this level of understand-
ing of large private sector operations, the analytical capacity
of many African state agencies with respect to private sector
business models is arguably rather more limited.
Secondly, it is assumed that the principal can push the agent
down to his/her reservation position because the agent is a
price (or contractual term) “taker” within a competitive
99
market. However, where PPP agreements are designed to
leverage greater private sector involvement in poorly function-
ing value chains, the number of potential partners will often be
small. This gives the agent more active bargaining power. Fur-
thermore, one way of avoiding adverse selection is to work
with “established” players. However, narrowing the choice
in this way increases the chances that the agent will exercise
power in contractual negotiations. Indeed, given the disparity
in resources between African state agencies (or their person-
nel) and some large private companies, private sector agents
may not just exert power in contract negotiations, but may
even “capture” the agency or personnel that are supposedly
acting as principal.
This leads to the fourth and perhaps most fundamental point.
While a state might have an agreed set of public policy objec-
tives (summarized, for example, in a national development plan
or Poverty Reduction Strategy Paper), individual agents within
the state—including the most senior ones—might pursue quite
different private objectives (van de Walle, 2001). Thus, the insti-
tutional set-up for devising and implementing PPP agreements
should include provision for holding the “principals” account-
able for their actions. Asymmetric information again comes
into play. In some contexts public officials—or consultants
hired by such officials—who truly understand private sector
business models and operations are scarce. If they defend PPP
agreements as the best deal that the public sector could obtain,
it may be costly for those holding them accountable to discern
the truth or otherwise of such claims.
5. PPPS FOR CAPITAL INVESTMENT
To date these are more common in OECD countries and La-
tin America than in Africa.
(a) Nonagricultural infrastructure
OECD experience with infrastructure PPPs is reviewed by
Hodge and Greve(2007, 2009). They note positive assessments
from both individual studies and reviews, but argue that
equally authoritative studies and reviews reach skeptical or
negative conclusions. Two main arguments are advanced for
PPPs in infrastructure investment: (i) that they can relieve
pressure on the public budget, because private firms assume
responsibility for raising finance necessary for infrastructure
investment; (ii) that by involving the private sector in this
way, the state can achieve better value-for-money in infra-
structure provision. On the first claim, Hodge and Greve
(2007) observe that there is little evidence that money “saved”
through resort to infrastructure PPPs is diverted to other
pressing uses. Moreover, it is only the need for lump sum pay-
ments that is saved by PPPs: states still have to remunerate
the private partners eventually, unless some of the investment
can be recouped through user fees. Evidence on the second
claim is mixed. A methodological challenge is to establish
the appropriate public-sector-only counterfactual against
which costs of a PPP project can be compared. Assessment
is further complicated by the complex remuneration formulae
embodied in some long-term infrastructure maintenance con-
tracts.
Warner et al. (2008, p. 45–46), commenting on PPPs for
water, health, and education infrastructure in developing
countries, observe that private operators frequently complain
of insufficient government commitment to agreements, espe-
cially after political change-overs. Politicians may view PPPs
as ways of getting more for less or as opportunities for
100 WORLD DEVELOPMENT
rent-seeking. Thus, if the real constraint to investment is lack
of political commitment to public service delivery, PPPs may
not solve the problem. Warner et al. (2008) recommend:
More transparent procurement and bidding processes.
Mechanisms, such as multi-stakeholder fora, to give infra-
structure beneficiaries a voice regarding service delivery
options.
That private operators build strong relationships with cus-
tomers, who “may then be a source of support in disputes
with political leaders” (p. 46).
(b) PPPs for construction and maintenance of agricultural
infrastructure
Improved infrastructure is vital to increase competitiveness
in many African agricultural value chains (World Bank,
2009), but government and donor investments in rural
infrastructure projects fell during the 1980s and 1990s.
Volumes of business in much of rural Africa remain too low
to directly recover investment costs in infrastructure, while
revenue streams that depend on rainfed smallholder agricul-
tural production are considered too risky by many private
investors.
Nevertheless, there may be advantages to private sector par-
ticipation in infrastructure investment, including the private
sector’s ability to source investment capital (easing immediate
constraints on infrastructure investment imposed by limited
state development budgets, albeit at the cost of higher recur-
rent budgets in future) and its project management skills. Un-
der a conventional construction contract, a public agency
might pay a private contractor to construct the asset in ques-
tion, with subsequent management and maintenance of that
asset handled (often by the public agency) through separate
arrangements. A PPP agreement can engage the same private
firm both in construction and subsequent management, with
the firm contributing investment capital that is recouped
through revenue linked to asset management, utilization, or
performance. This can reduce the incentives to cut construc-
tion costs in ways that undermine future performance of the
asset, while also sharing the risks from cost overruns or lower
than expected utilization of the asset.
Key questions concern the revenue stream for the private
company. Potential for user fees varies by type of rural infra-
structure (compare irrigation with feeder roads). Shadow tar-
iffs, whereby the sum paid by the public partner to the private
operator depends on utilization rates or other dimensions of
performance, are a way of encouraging risk sharing. Warner
et al. (2008) argue that infrastructure investment can be made
more attractive to the private sector through “bundling” con-
tracts for complementary investments—for example, feeder
roads + agro-processing or irrigation + power generation—
which have different revenue and risk profiles.
In practice, Warner et al. (2008, p. 10–11) cite World Bank
figures showing that only 1% of private investment in
infrastructure during 2003–05 was in investments directly or
indirectly related to agriculture (i.e., including rural electrifica-
tion and telecommunications). Based on a review of 18 PPPs
for rural infrastructure (seven of which are in Africa), they con-
clude that PPPs can have a number of advantages and that con-
ditions are in many ways right for their use to be expanded.
However, they are “by no means a panacea” (p. 6) for problems
of rural infrastructural investment. Potential pitfalls include:
Lack of real political will—politicians can have various
motives for seeking to leverage private investment and
may renege on agreements that turn out to be politically
unpopular, for example due to high user fees charged.
Lack of capacity (on both sides, but particularly within
public agencies) to structure and negotiate deals that deli-
ver value for money for the state and its citizens, while still
providing a sufficient return on investment to be attractive
to private investors.
Competitive bidding processes to ensure transparency in
contract allocation have high fixed costs. Relatively large
projects are required to make these costs worthwhile.
Equally, the set-up costs of a PPP, which also include the
costs of negotiating a final contract with the selected
bidder, have to be offset against any efficiency gains from
private sector involvement.
In the absence of a suitable regulatory framework, private
investors may pursue pricing policies that exclude poor
groups, thereby undermining the “public” nature of bene-
fits from the infrastructure investment.
6. CONTRACTING THE PRIVATE SECTOR FOR
SERVICE DELIVERY
As with PPPs for infrastructural investment, experience with
contracting out of service delivery 2 in water, waste manage-
ment, and health sectors in developing countries shows that
this rarely avoids the problems of state failure that previously
afflicted public service delivery, while the fundamental
challenges of service delivery remain.
PPP water contracts in Africa have struggled in the face of
high costs of constructing new distribution (pipe) networks
on the one hand and low willingness to pay for public water
supplies (due to low incomes and popular beliefs that access
to water should be a right) on the other (Kayaga, 2008). How-
ever, both Kayaga (2008) and Tati (2005) also highlight limi-
tations in contract design and the contractor selection
processes. Where services have traditionally been provided
by the public sector, one should not expect strong domestic
private sector capacity to exist.
Awortwi (2004) assesses the contracting out of solid waste
collection services by municipal governments in Ghana. Many
of the fundamental rules of P–A theory were ignored in prac-
tice. As a result, the volume of solid waste that was collected
increased (this was the basis on which the companies were
paid), but so too did the cost per ton of solid waste collected.
Two of the major contracts were not competitively tendered,
which discouraged efficiency savings. Instead, “politics and
patronage became a factor in awarding contracts” (Awortwi,
2004, p. 218). Meanwhile, inflexible civil service regulations
meant that few jobs were saved by contracting services out,
but few resources were made available for monitoring contrac-
tual performance. However, 10 years of contracting out had
stimulated capacity for service delivery among community-
based organizations, which could stimulate future competition
and performance.
Palmer and Mills (2005) examine three contracts for pri-
mary health care services in South Africa and Lesotho. Pri-
mary health care is complex, involving a broad range of
services. This makes it difficult to specify complete contracts,
documenting all services to be rendered and under what
terms. This is particularly the case in rural locations where
it is rare for more than one service provider to operate in
a given area. Similarly, the known pool of potential provid-
ers is often low, so rural contracts may be managed relation-
ally. Palmer and Mills (2005) note that, if problems were
identified, contractor and provider generally talked them
over, as contract termination was considered problematic.
Monitoring was also problematic: the resources within the
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
contracting agency for visiting rural care centers were lim-
ited.
Within the agricultural sector, Hubbard (2003, p. 148–157)
reviews four cases where state agencies (two African and
two South Asian) contracted private providers for diverse ser-
vices (crop storage or export, rural road maintenance). He
concludes that the agencies could generally handle basic con-
tracting tasks, but that unwillingness of state agencies to
reconfigure themselves as regulators, rather than service pro-
viders, constitutes a significant obstacle to the expansion of
contracting.
(a) Agricultural extension services
Anderson and Feder (2007) quote estimates that “80% of the
world’s extension services are publicly funded and delivered by
civil servants” (p. 2345), while only 5% are private. Except
where supply chains are characterized by strong vertical coor-
dination between farmers and buyers, two over-riding con-
straints discourage private investment in extension service
delivery: much extension information has public or common
pool good attributes, while variable weather and/or market
prices render the valuation of advice, and thus ability of ser-
vice providers to appropriate a suitable user charge, difficult
(Hanson & Just, 2001). The attributes of the extension process
also pose formidable challenges for management of effective
delivery, irrespective of the agent (public, private, NGO).
Attributing positive agricultural outcomes to particular exten-
sion input or disappointing agricultural performance to poor
extension services is often not possible (Israel, 1987). The effort
of front-line extension staff is also difficult to monitor, espe-
cially in quality terms, given the dispersed locations that they
visit. Only farmers can really say what quality of support they
have received.
However, public extension services are often centralized,
hierarchical, and unresponsive to the diverse needs of farmers.
Contracting private extension service providers could increase
responsiveness, if there is a mechanism through which farmers
can express their preferences across competing providers.
Given that willingness to pay for extension is likely to be
low, this could happen through distribution of vouchers or
alternatively through farmers’ representation on the bodies
that allocate contracts in a given area.
Chile and Uganda provide useful lessons on involving pri-
vate service providers in national extension systems. In Chile,
a subsidized voucher system was introduced in 1978 in the
hope of quickly transitioning to a commercial extension mod-
el. As the value of the subsidy was progressively reduced, how-
ever, farmers proved reluctant to pay increasing fees, resulting
in declines in service quality. In addition, monitoring was
inadequate and Bebbington and Sotomayor (1998) found evi-
dence of collusion between farmers and providers to share
voucher values. From 1983, therefore, vouchers were aban-
doned. Instead, providers bid for service contracts in a given
geographic area, based on specified work plans. This model
has remained, with farmer representatives playing an increas-
ing role in contract allocation and monitoring, though their
control over these processes remains inadequate (Cox &
Ortega, 2004).
In Uganda the extension service was reformed starting in
2001 and renamed National Agricultural Advisory Service
(NAADS). Under the new approach, farmer groups con-
tracted service providers, including both commercial private
sector and NGOs, to provide advice through short-term
contracts. Benin et al. (2007) found that the quantity and
(perhaps) quality of advice services in Uganda were greater
101
in NAADS sub-counties than in their matched control group
of nonNAADS sub-counties. However, adoption rates of new
technologies were insufficient to generate higher yield growth
in NAADS sub-counties than others. Benin et al. (2007) also
found that NAADS service providers were more effective at
promoting the adoption of new crop varieties and yield
enhancing technologies than at promoting improved soil fertil-
ity management, while Bahiigwa, Rigby, and Woodhouse
(2005) found that NAADS focused primarily on commercial
producers and crops. Advice on new crop varieties and yield
enhancing technologies is fairly readily time-bound, with
adoption relatively easily linked to the advice received (hence
well suited to a contracting approach), whereas soil fertility
management is more complex and the impacts longer term.
In this sense, these findings are consistent with the predictions
of theory.
Wider lessons from these two cases include the following.
Firstly, given the pre-reform history of public extension provi-
sion in both Chile and Uganda, most of the “private” provid-
ers post-reform are agents who have come out of the public
system. In extension there are no international specialist ser-
vice providers ready to enter national markets to compete
for contracts. Thus, reform programs should build in an ele-
ment of capacity building for commercial extension providers.
Secondly, effective farmer voice in contract allocation and
monitoring requires functioning decentralized administrative
systems and strong farmer organizations. These two elements
are missing in much of eastern and southern Africa. While
NAADS shows that the latter can be developed as part of
an extension reform program, assuming sufficient (donor?)
funding, the former is a broader political issue. This leads to
arguably the most fundamental point: extension reform in
both Chile and Uganda has benefited from clear local “owner-
ship” and political support (see Bebbington & Sotomayor,
1998 for the Chile case 3). Where public sector “principals”
are incentivized to deliver on stated public policy objectives,
interactions with private sector agents can be reviewed and re-
fined over time to ensure that those objectives are met.
Finally, although it does not directly contract commercial
private sector operators, the Maendeleo Agricultural Technol-
ogy Fund (MATF) is worthy of note. It aims to promote dis-
semination of “proven” technologies to farming communities
in East Africa by facilitating multi-stakeholder partnerships
for technology transfer and its main tool for identifying prom-
ising technologies and partnerships is a system of competitive
grants (i.e., a challenge fund). MATF’s expectation is that pro-
vision of the new products or services becomes self-sustaining
by the end of the grant period, which in practice has proven
extremely challenging (Poulton, Mangheni, Okwadi, & Kilewe,
2006). If smallholder farmers are to adopt new technologies,
they require access to a range of support services, including
credit, efficient supply of the technology itself, access to remu-
nerative output markets—and perhaps some ongoing techni-
cal advice as well. Finessing demand and supply for these
services within a 2–3 year project, such that by the end of
the project volumes of business are attractive to commercial
service providers, is a daunting task. Organizing provision of
credit has proven particularly problematic.
Nevertheless, MATF can provide valuable lessons. The
“advisory panel”, which selected projects for approval,
worked hard and in an objective manner. Being part of a
donor-funded, NGO-managed project perhaps helped to insu-
late them from rent-seeking or other external interference. The
panel played an active role in monitoring, including making a
pre-contract visit to check out the capabilities of the partners
against the claims made in the bid. The time that they and
102 WORLD DEVELOPMENT
project staff spent in the field generated impressive organiza-
tional learning, with regard to both the requirements for suc-
cessful technology adoption and the capabilities necessary
within a partnership for this to occur. However, MATF’s
pro-active management approach, when combined with the
small size of individual projects, came at a cost: total adminis-
trative “overheads” came to 40% of funds disbursed (Poulton,
Mangheni et al., 2006).
(b) Demand stimulation
Table 2 distinguishes ongoing service delivery contracts
from public efforts to stimulate demand for private services
in the hope that a self-sustaining private market could develop
once clients become familiarized with the product or service in
question. Jones, Quigley, and Foster (2006) discuss an exam-
ple of the latter—the Tanzania National Voucher System
(TNVS) that distributed vouchers to women in remote rural
areas to subsidize the purchase of insecticide-treated bednets
for malaria control. The vouchers could be exchanged at par-
ticipating local traders upon payment of a small transaction
fee. According to Jones et al. (2006), use of vouchers and nets
were both high. In addition, the extra demand created by
TNVS encouraged the expansion of distributor networks in
the study area, where sales of unsubsidized nets rose 30%.
Are there agricultural services for which public and private
sectors could share the costs and risks of stimulating demand
for commercial service delivery? Returning to extension, Cox
and Ortega (2004) report that, after 25 years of promoting pri-
vate involvement in extension provision, the state in Chile still
pays 80% of the cost of extension delivery. While vouchers
were abandoned in 1983 and full privatization has apparently
fallen off the policy agenda, it is perhaps surprising that a fully
private “top end” service has not emerged for better-off pro-
ducers. This suggests that, if other states embark on service
contracts for extension provision, they should do so in the
expectation that they will be funding these contracts for some
time to come.
Donors are also keen to ask similar questions of fertilizer
voucher (subsidy) schemes, such as that in operation in Mala-
wi since 2005/06. However, there are important differences be-
tween bednets and fertilizer. Firstly, bednets were a genuinely
new technology to many rural Tanzanian households, whereas
Malawian farm households have long had (intermittent) ac-
cess to fertilizer. The issue for Malawian farm households is
affordability, while some of the most important benefits from
the subsidy scheme are indirect, that is, increased demand for
casual labor combined with lower food prices (Dorward &
Chirwa, 2011). Secondly, the challenges of operating at scale
make it difficult for the fertilizer subsidy program to facilitate
experimentation with fertilizer use, 4 such that farmers might
improve fertilizer use efficiency and thereby be willing to pay
more for fertilizer. Thirdly, fertilizer vouchers are a potent tool
of political patronage, whereas distribution of vouchers for
bednets is not so easily transformed into political gain, espe-
cially when linked to attendance at ante-natal clinics. Perhaps
partly for this reason, private stockists have been excluded
from distribution of subsidized fertilizer (though not seed).
Continuing political interference in the fertilizer market can
be expected, thereby increasing the risks of private sector
investment in fertilizer distribution.
In Sections 5 and 6 we have reviewed experience with con-
tracting the private sector to construct infrastructure and to
deliver a range of services. The capacity of state agencies
and the incentives facing state agents to administer contracts
in a way that achieves stated public objectives have been
recurring themes. The nature of rural services also poses chal-
lenges for contract design, as illustrated by the complexity of
rural health care and the difficulties of specifying or monitor-
ing extension quality. The PPP approaches that we review in
the remainder of the paper have been pioneered by interna-
tional organizations, so the capacity and incentive issues are
more readily dealt with, albeit at the same time raising ques-
tions of replicability. Meanwhile, challenges for contract de-
sign remain.
7. PPPS FOR NEW PRODUCTS AND SERVICES
Margins on new products and services for so-called “bottom
of the pyramid” consumers are typically low, so firms have to
achieve scale if their investment is to be worthwhile. This
makes the investment very risky. PPPs, therefore, aim to share
costs and/or risks at critical stages of the product development
process. Once the viability of the product or service is estab-
lished, however, the private sector partner is expected to make
the necessary investment to scale up delivery.
(a) Investment funds
Classic challenge funds disburse grants on the basis of a
periodic, competitive bidding process. Other funds are man-
aged with greater flexibility or discretion, even if the manager
ultimately has to seek approval from an oversight board to
fund a particular proposal. Thus proposals might be funded
on a first-come-first-served basis, providing that certain basic
criteria are fulfilled, or managers may pro-actively seek pro-
posals from particular groups, seeing capacity building of po-
tential bidders as part of their role.
Advocates of the competitive bidding approach highlight
several advantages. Only the best proposals received in any gi-
ven period are backed—not all that cross a minimum thresh-
old. Transparency of decision making is maximized as
proposals are explicitly assessed against each other. If African
states establish investment funds to support their local private
sectors, the governance advantage of a competitive (challenge)
fund should be a compelling argument. Finally, it is believed
that open competition allows the best ideas to reveal them-
selves. By contrast, a more pro-active approach by fund man-
agers can look like an attempt to “pick winners”—something
that can be done well, but which is difficult to achieve, with the
danger of large, wasted subsidies when done badly. On the
other hand, competitive bidding processes can be time-con-
suming and may not fit the investment timetable of key private
sector actors. If private sector capacity is limited, so too will be
the numbers bidding in open competition and hence the ben-
efits of competitive bidding.
(i) Health investment funds
Moran, Ropars, Guzman, Diaz, and Garrison (2005) ex-
plain how a series of investment funds has helped to transform
the landscape of research into so-called “neglected” diseases
(including HIV/AIDS, malaria and tuberculosis). 5 Where dis-
eases are prevalent in low income countries, the ability of most
potential consumers to pay for treatment is low, yet drug
development costs are large, with large risks associated with
the development of any particular treatment. Hence, while
drug development was driven by the value of patent rights
alone, diseases of the poor were neglected by major pharma-
ceutical companies. In recent years increasing political and
civil society pressure has led to companies to seek new models
of drug development that combine social responsibility and
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
commercial viability. The willingness of PPP investment
funds to share costs and risks at later (higher cost) stages of
the drug development process—for example, large-scale
clinical trials—was critical for the emergence of what has
been called a “no profit—no loss” model (Moran et al.,
2005, p. 12).
The investment funds in question are pro-actively managed
funds that also seek to network players working on particular
problems. Nwaka (2005) concludes that the Medicines for
Malaria Venture (MMV), established in 1999, “radically chan-
ged the antimalarial R&D landscape” (pS20). Given the large
sums of money involved in medical research projects, high
quality fund management, including the ability to manage
relationships, structure contracts, and negotiate IP agree-
ments, was critical. Some of the MMV-funded projects did
fail, but overall the portfolio performed well. Chataway and
Smith (2006) note the achievements of the International AIDS
Vaccine Initiative (IAVI) but also note the importance of insti-
tutional learning. In its early years IAVI was accused of being
a rather arrogant organization, insensitive to local cultures
and needs, but, according to both Chataway and Smith
(2006) and Rosiello and Smith (2008), it has since invested
in more genuine partnerships in many of the countries in
which it works. A similar learning process is documented for
the African Comprehensive HIV/AIDS Partnership, involving
Merck, Government of Botswana and Bill and Melinda Gates
Foundation (BMGF). Initial criticisms focused on Merck’s
apparent desire to achieve quick public relations gains from
the venture, rather than to form genuine partnerships, but
these tensions have now been at least partly addressed. The
experience highlights the useful role that a “third party”,
BMGF, could play in the process of building trust between
Merck and GoB (Clark & O’Brien, 2003; Ramiah & Reich,
2006).
Buse and Harmer (2007) take a critical look at 23 health
PPPs, arguing that these funds have distorted national health
priorities in countries in which they work, while imposing
uncoordinated reporting requirements on national partners
(as a result of their proliferation and silo-based approaches).
Buse and Harmer (2007) also criticize the internal governance
of (many) health PPPs for over-representation of the private
sector in decision making compared with its financial contri-
bution, a lack of measurable objectives, and poor transpar-
ency. Nevertheless, they also recognize their “impressive
contributions to efforts to tackle neglected diseases” (p. 261).
(ii) DFID challenge funds
DFID has so far established at least eight challenge funds
(some in collaboration with other donors), with most of these
supporting some projects directly relevant to the agricultural
sector. 6 Thus, for example, the Financial Deepening Chal-
lenge Fund (FDCF) supported roll-out costs for the now-fa-
mous M-PESA mobile phone-based money transfer system
in Kenya; the Africa Enterprise Challenge Fund runs a win-
dow to encourage businesses commercialize technologies
developed within public agricultural research systems, and
the Food Retail Industry Challenge Fund assists agribusiness
to link smallholders into agricultural export supply chains. As
is now widely recognized, there are significant fixed costs asso-
ciated with sourcing from smallholders—for example, group
mobilization and training—that can be sufficient to prevent
smallholder inclusion (Boselie, Henson, & Weatherspoon,
2003; Poulton et al., 2010; Raynolds, 2004), but which could
be partially met through such a fund.
All the DFID funds have sought (or seek) to attract “expe-
rienced” private sector investors who are required to contrib-
103
ute at least 50% of total project costs. Given the assumption
that private sector capacity, entrepreneurship and spirit of
innovation that are relevant to the sector(s) concerned already
exist, one can ask why challenge funds are needed to stimulate
private sector investment. Irwin and Porteous (2005) explain
their rationale in terms of the risk-return relationship of so-
cially beneficial projects and hence the chances of these pro-
jects being prioritized in the internal competition for
investment funding within firms that may otherwise target
higher income or less risky markets.
This, however, raises the so-called “additionality” question
(an adverse selection issue): is/was the public contribution
from the challenge fund really necessary for the innovation
to proceed? Outsiders can glean at best partial information
about the internal processes of competition for investment
funds within firms, while project proposals have to argue that
the public contribution is essential if they are to stand a chance
of being funded.
Moving away from the “classical” challenge fund model can
help here. Initial concept notes typically provide little insight
into the additionality question. However, many private sector
players have limited experience in putting funding proposals
together, while fund managers have an interest in receiving
the best possible proposals, so there is then a case for personal
interaction between bidders and managers. This can provide
managers with useful insights into how the project is seen
within the firm, which in turn they can feed into the decision
making process if requested by the panel or board.
For external evaluators, however, the problem is still not
solved. Irwin and Porteous (2005) classify projects funded by
FDCF in terms of whether the fund was a “catalyst” for the
project (i.e., essential to it going ahead), an “accelerator”
(i.e., it probably would eventually have gone ahead anyway,
but later and/or more slowly) or where the public funding
probably made little difference to the process and speed of
innovation. The difficulty of assessing additionality is high-
lighted by a mobile banking scheme in Kenya, which Irwin
and Porteous (2005, p. 24) place in the “would have gone
ahead anyway” category, but Enterplan (2004, footnote 10),
citing a CGAP report, hailed as a clear case of additionality.
Experience with the DFID challenge funds highlights the
following practical lessons if a fund is to leverage private sec-
tor investment in a cost-effective manner:
The “challenge” and associated outcome indicators against
which projects will be assessed need to be clearly defined,
but should neither be so narrow and specific that innovative
new ideas are excluded, nor so broad that the projects col-
lectively fail to trigger any systemic change in their (sub-)
sector. 7
There should be a clear understanding of why a challenge
fund might leverage private investment that would other-
wise not be forthcoming, given that successful investments
are expected to achieve both sustainability and replicability
within short periods of time.
If a fund is too large, large sums of public money may be
squandered if the management is poor. If it is too small,
the overhead rate (administrative costs as a share of fund
size) will be high, given the fixed costs associated with man-
agement both of individual projects and of a challenge fund
as a whole. 8 This overhead rate has to be set against the
social benefits derived from PPP implementation.
Enterplan (2004) and Irwin and Porteous (2005), both high-
light that, without active and ongoing efforts to market a
fund, it runs the danger of servicing a small number of pri-
vate actors, whose motives for engagement could readily
slip from help for innovation to rent-seeking.
104 WORLD DEVELOPMENT
Active monitoring—and swift action on the information it
provides—are also important. Monitoring and evaluation
may not resolve the additionality problem, but are necessary
for demonstrating the eventual impact of a fund’s investment,
can shed light on the types of projects and organizations
likely to make good use of funds and also help to minimize
shirking (a form of moral hazard) by successful bidders.
Where a project is not performing, the grant should be can-
celed quickly, so as to maintain incentives for future bidders
and protect the overall performance of the fund portfolio.
However, to do this, managers and their boards must be
shielded from political pressures by the institutional arrange-
ments of the fund. Being donor-funded may be a considerable
benefit here. 9
(b) The CGIAR and PPPs
In the field of agricultural research there has been interest
in PPPs for at least a decade. Publicly funded national and
international agricultural research centers engage with
private sector partners in a number of ways. Spielman, Hart-
wich, and von Grebmer (2007) review 75 PPPs (loosely de-
fined) that involved centers within the CGIAR system. 10
They noted two distinct patterns of interaction, depending
on the nature of the private sector partners. Center relation-
ships with local private sector enterprises were typically in-
tended to encourage commercialization of technologies
developed within the centers by the private sector partners
(e.g., local seed companies). By contrast, through relation-
ships with multi-national crop science firms, centers often
aimed to access proprietary technologies or other re-
sources—recognition of the huge investment made by such
firms in crop biotechnology and the resultant ownership of
intellectual property rights (Pingali & Traxler, 2002). On
their part, crop science firms may be willing to grant access
to proprietary technology as part of their corporate social
responsibility, recognizing that markets in countries where
CGIAR centers are based currently offer them few commer-
cial opportunities.
Nevertheless, Spielman et al. (2007) conclude that “the po-
tential benefits of PPPs have yet to be fully realized in the
CGIAR” (p. 3), a view echoed by World Bank (2007). They
observe that it has proven difficult to bridge the cultural gulf
between public and private sector organizations, and that
negotiations over intellectual property rights are difficult
(partly because public organizations have much less informa-
tion on the technologies available from private partners than
vice versa)—even though private companies conduct such
negotiations with each other all the time. Systematic compar-
ison of this experience with that of the health investment
funds discussed above could prove insightful. Here we note
that there are few examples of what Spielman et al. (2007)
call “frontier” agricultural research, that is, collaborative ef-
forts to develop new technologies, between public and private
sectors, unlike the efforts catalyzed by the health investment
funds. The lack of investment funds to catalyze this within
the field of agricultural research may be a contributory fac-
tor, but CGIAR centers do have their own funding streams
to draw upon. We hypothesize that another factor differenti-
ating the two sectors is the much stronger political and civil
society pressure upon life science companies to demonstrate
positive impact upon poor populations than exists for crop
science firms, whose proprietary technologies are viewed by
many civil society groups as undesirable for smallholder
producers.
(c) Loan guarantee funds
Loan guarantee funds encourage commercial lending to
players within agricultural value chains who otherwise suffer
from sub-optimal access to capital. Two scenarios may be
considered. In the simplest case, the risk of lending to the
new borrowers is perceived differently by the lender and the
third party willing to offer the guarantee. For example, the
new borrowers may be unknown to the lender, but known
and trusted by the third party. The guarantee provides the len-
der with sufficient assurance to lend to the new groups for an
exploratory period. If the borrowers repay all their loans, the
lender should then be willing to advance loans without guar-
antee, based on his/her own acquired knowledge of the bor-
rowers. The best case scenario is that new lender–borrower
relationships are established without any recourse to the loan
guarantee fund.
Where the third party has no direct knowledge of the poten-
tial new borrowers, the rationale may be as follows: The third
party believes that increased lending is in the public interest,
but recognizes that successful lending to the new groups will
require a period of learning about the new borrower popula-
tions. Given that perceived lending risks during the learning
phase are sufficient to discourage “uninsured” experimenta-
tion, the guarantee provides the lender with a degree of insur-
ance against expected losses during this phase. The best case
scenario is that, by the end of the guarantee period, the lender
will develop suitable new loan products and modes of delivery
and will know enough about the new groups to be able to
identify creditworthy individuals. Nevertheless, some recourse
to the loan guarantee fund is expected.
(i) Guarantees for supplier credit to agro-dealers
This model was pioneered by CARE in Zimbabwe and
Kenya in the 1990s (Rusike & Gandanhamo, 1999) and has
subsequently been promoted by projects funded by Rockefel-
ler Foundation and AGRA. In these projects, access to credit
is one component of a capacity building program for agro-
dealers, alongside training in business management skills
and/or technical knowledge regarding agro-inputs. 11 Through
the training programs the implementers get to know which
agro-dealers are competent and diligent. They then offer guar-
antees to input wholesalers to stimulate supplier’s credit for
these individuals.
World Bank (2006) reports briefly on experience in Mala-
wi. Over 300 trained agro-dealers were enabled to acquire
stock on 30–60 day credit from input supply companies.
The volumes of sold inputs rose impressively as a result. In
3 years the default rate on this suppliers’ credit was below
1%. The authors understand that default rates have been sim-
ilarly low within the parallel program in Kenya, where over
200 agro-dealers were trained in the west of the country by
2007.
However, as noted earlier, the agro-dealer network in Mala-
wi has also had to contend with the introduction of the major
fertilizer subsidy program. If they were allowed to distribute
the subsidized fertilizer and reclaim the value of the vouchers
from the state, the scheme would benefit private agro-dealers.
Instead, the Malawi government has channeled all sales of
subsidized fertilizer through two parastatal organizations,
thereby crowding out the emerging agro-dealer network (Fu-
ture Agricultures Consortium, 2009). While access to capital
and training were undoubtedly constraints on agro-dealer
expansion, the nature of ongoing state intervention could be-
come the more binding constraint.
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
(ii) Guarantees to commercial banks to support agricultural
lending
In 2005 Rockefeller Foundation provided a loan guarantee
fund of US$500,000 to Centenary Rural Development Bank
(CERUDEB) in Uganda, on the basis of which CERUDEB
lent US$1 million to farmers over the period 2005–08. The
agreement specified that the two parties would share the costs
of any loan defaults equally. Over this period the drawdown
on the guarantee fund to cover loan default was only
US$10,400—less than the guarantee fund accumulated in
interest (AGRA, 2009). Subsequently, the agreement was
renegotiated for a further 2 years with the same US$0.5 mil-
lion guarantee expected to leverage US$4 million of lending
during this extension period. 12
The total number of farmers benefitting from the first phase
of the arrangement (some of them through membership of
farmer groups that received collective loans) was in excess of
2000. To ensure a “small farmer” focus Rockefeller stipulated
that borrowers should not be cultivating more than 10 ha. The
average loan size was in the range US$500–1000, so these were
certainly “better off” smallholders. It is assumed that most
were new borrowers, as most loans were directed to previously
under-served rural areas and CERUDEB had little history of
lending to smallholders prior to this.
Drawing on this experience, AGRA has negotiated further
loan guarantee deals, including a US$5 million guarantee to
Equity Bank in Kenya, a guarantee fund of US$10 million
to Standard Bank expected to leverage US$100 million of
new lending to the agricultural sectors in four African coun-
tries 13 and a guarantee fund with National Microfinance Bank
in Tanzania, expected to leverage US$5 million in loans to
agro-dealers (AGRA, 2009; AGRA & Millennium Challenge
Account Mozambique, 2009). These are innovative deals de-
signed to tackle one of the major constraints facing farmers
and other players in agricultural value chains in Sub-Saharan
Africa. However, important questions remain regarding the
detailed design of the agreements.
AGRA (2009) emphasize that these deals reduce the interest
rates that farmers and other borrowers have to pay for credit.
AGRA’s view is that commercial bank lending to smallholder
producers in Africa is impeded by an unjustifiably pessimistic
view of the risks involved. The guarantees reduce the risk in
two ways: by providing partial insurance against losses and
indirectly by strengthening the relevant value chains through
simultaneously financing players at multiple stages. From this
perspective, the reduced interest rates secured are close to what
“true” market rates should be. Moreover, AGRA hopes that
there will be dynamic benefits as the commercial banks realize
the potential of the sector and begin to develop new products
to respond to additional opportunities that they see.
An alternative view is that, at existing interest rates, lenders
are unable to cover the high transaction costs and associated
default risks of lending to poor, dispersed clients engaging in
rainfed agricultural production in areas where roads are poor.
Loan guarantees reduce the cost of lending and this “divi-
dend” can then be used either to reduce the interest rate or
to expand service provision to new client groups while retain-
ing interest rates at the pre-existing level. The danger with the
former route is that few genuinely new or poor borrowers are
reached. Existing clients receive loans on improved terms or
clients may be attracted from other banks without expanding
the size of the market (the public policy objective). Further-
more, if guarantees are used to reduce interest rates, rather
than to support the development of new loan products or cost
effective modes of delivery, most “gains” may be reversed once
the loan guarantee is removed. In terms of Table 2, the effect
105
of the guarantee fund could be more to support ongoing ser-
vice provision than to stimulate the development of new prod-
ucts or services. In principal–agent terms, the outcome may
advance the objectives of the agent, but the principal has
not incentivized the agent to act in accordance with long-term
public policy objectives for the development of rural financial
markets.
Of course, African smallholders are not a homogeneous
group, so both analyses may have some validity for different
smallholder types. Moreover, the authors understand that
deals may also include a concessional finance package—an
additional cost “dividend” that offers the same opportunities
and choices as the loan guarantee. As loan guarantee deals
are still in their infancy, the discussion flags issues to be
tracked during monitoring of the guarantee agreements. Col-
lecting data on geographical spread of borrowers and on loan
sizes is a useful first step. However, a further step should be to
require that monitoring data distinguish between existing bor-
rowers from the bank in question, existing borrowers from
other financial institutions and genuinely new clients who have
been able to access a loan for the first time as a result of the
loan guarantee agreement.
8. DISCUSSION
Can contractual arrangements between public and private
sector actors effectively incentivize the latter to provide ser-
vices in ways that meet public policy objectives? The answer
to this question is a cautious “yes”, but several important
caveats are in order.
Firstly, as principal–agent theory tells us, detailed mecha-
nism design is important. Table 3 considers three of the PPP
mechanisms reviewed in this paper, which between them illus-
trate all three main asymmetric information problems distin-
guished by Hart and Holmstrom (1987).
Table 3 also indicates mechanisms that can be (and in some
cases have been) employed to tackle these asymmetric infor-
mation problems. However, designing effective mechanisms re-
quires expertize. Similarly, we have argued in this paper that
designing a contract to align the incentives of private sector
agents with stated public objectives in a cost-effective manner
requires familiarity with the basic business models of those
private agents. What type and level of incentive is required
to influence their behavior? As CGIAR experience with PPPs
illustrates, such information can be hard to obtain. However,
it may be particularly challenging for African state agencies
that have traditionally been suspicious of the private sector.
In the case of DFID challenge funds, this problem of unfa-
miliarity is typically overcome by contracting fund manage-
ment to people with requisite private sector experience,
something that can also enhance the credibility of the fund
with the private sector agents that it is seeking to engage.
However, international agencies are more likely than African
state agencies to be able to pay salaries sufficient to attract
people with the necessary experience. Furthermore, this cre-
ates a “nested” principal–agent problem, where the contracted
“principal” enjoys an asymmetric information advantage over
the ultimate principal in regard to the private sector agent—an
advantage that an opportunistic manager could exploit, for
example, through collusion with favored private sector part-
ners. The factor that minimizes this danger in the context of
a typical international agency is the overall framework of per-
formance incentives. Thus, challenge fund managers are ex-
pected to present evidence of the effectiveness and impact of
the investments their funds have made. By contrast, the overall
106 WORLD DEVELOPMENT

Table 3. PPP mechanisms as principal–agent models

Classification
Desired outcome
Agent’s action
Major danger
Are there risk factors beyond the
agent’s control that affect the
relationship between the agent’s
action and the observed
outcome?
To what extent are agent’s
actions or effort observable?
Incentives to avoid moral hazard
Mechanisms to avoid adverse
selection
Note: in the final two rows we identify means of protecting against both moral hazard and adverse selection for all three cases. However, the classification
in the top row shows which of these problems is considered the dominant one in each case.
Extension contracts
Moral hazard (Hidden action)
Technology adoption by farmers
Delivery of extension advice
Shirking on extension quality
Yes—weather, farmer
preferences (poorly understood?),
market prices, availability of
complementary services
Only farmers can observe
extension quality; principal may
try to track quantity of visits
Contract renewal; involvement of
farmers in selection
Competitive bidding
Challenge fund
Adverse selection
New products adopted by
farmers
New product roll-out
Subsidy to poor project or
project not requiring public
support
Yes—consumer acceptance: high
risk in rolling out new product,
especially with low margins,
hence scale needed
Fund manager can track through
regular reporting requirements
and follow-up
Threat of grant cancellation
Competitive bidding,
independent panel; work with
established players
Guarantee fund
Moral hazard (Hidden
information)
Greater access to services for
poor/rural clients; no recourse to
guarantee fund
Services provided to new clients
Guarantee used to obtain
competitive advantage in serving
existing market
On recourse to fund, yes—major
covariant risks affect both
fertilizer sales and agricultural
finance
Loan disbursements can be
tracked with strong M&E system
Partial loss guarantee
Due diligence on commercial
partner; work with established
players

framework of performance incentives in many African state
agencies is weak (van de Walle, 2001). On a practical note this
paper has highlighted the importance of thinking hard about
monitoring requirements when designing contracts. This mon-
itoring is important for structuring incentives not just for the
private sector agents, but also for those responsible for con-
tracting them.
Assuming there is strong commitment to achieve improve-
ments in service delivery—as with reforms to the national exten-
sion services in Chile and Uganda?—it seems inevitable from
the foregoing discussions that there will also be a learning curve
as successive contracts are drawn up. (This is one of the reasons
why more cases are required before firm conclusions can be
reached on the efficacy of PPP arrangements). The notion of a
learning curve suggests that the “cost effectiveness” of PPP
arrangements (from a public perspective) should also improve
over time. However, this is an area on which very little informa-
tion is yet available. Exactly what cost effectiveness means will
depend on the type of mechanism. In PPP contracts for infra-
structure investment or service delivery a cost comparison
may be done with “conventional” options. There is an inconclu-
sive literature on cost-effectiveness in the case of PPP contracts
for infrastructure investment (Hodge & Greve, 2007). For chal-
lenge funds the “without fund” scenario is no new products;
hence ultimately the cost–benefit analysis should consider the
cost of the public investment (including fund administration)
vs. the social and economic impacts of the new products and ser-
vices. This is very difficult to do. Existing assessments focus en-
tirely on the administrative costs of a challenge fund (as a share
of funds invested) and how this compares with figures reported
for venture capital funds. The discussion in Irwin and Porteous
(2005) illustrates that even this is not a precise science!
Learning within the state sector is one reason why contracts
may be refined and terms improve over time. Another is that it
might take time for private agents to gain trust in public sector
partners. Where trust is low, the public partner is likely to
have to offer generous terms to achieve collaboration from
private sector. As observed by Chitundu, Droppelmann, and
Haggblade (2006), lack of trust between public and private ac-
tors is often mutual, suggesting a role for institutions that can
bring different parties together.
(a) Deliberative fora
Hall and Soskice (2001, p. 10–11) explain the importance of
institutions that facilitate “collaboration and strategic interac-
tion” within OECD economies (and sectors) by assisting ex-
change of information, monitoring behavior, sanctioning
defection and facilitating “deliberation”. Deliberation, that
is, the chance to get together and discuss, can build mutual
understanding of the roles of different actors in tackling a par-
ticular problem or of the long-run development vision of a sec-
tor. The trust that comes from this mutual understanding
makes collaboration easier.
However, where coordinated actions by both state and pri-
vate actors are required, Hall and Soskice (2001) caution that,
if the state is too strong, “firms are reluctant to share ... infor-
mation with governments whose position as powerful actors
under a range of unpredictable influences raises the risks that
they will defect from any agreement and use the information
they have acquired against the firm” (p. 47). The “coordinated
market economies” of northern Europe overcome this prob-
lem because powerful employer associations and trade unions
exert influence within political parties, so they can “sanction”
governments that abuse their power.
Poulton, Kydd et al. (2006) highlight the need for comple-
mentary investments (by both private and public actors) to
support service delivery to staple food producers in Africa,
as incentives rarely exist for a single private investor to provide
multiple services to smallholders through contract farming
arrangements. A similar argument may hold for investments
at different stages of a nascent supply chain. Where the return
on investment by one player is dependent on the existence of a
complementary investment(s) by another player(s), the situa-
tion is analogous to that of asset specific investment as ana-
lyzed by Williamson (1985), with the same basic problem of
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
hold-up. Thus, mechanisms for coordinating investments are
required if significant private investment is to be forthcoming.
Spielman et al. (2007) note the efforts of some CGIAR cen-
ters to facilitate coordination across value chain actors for
crops that they are working on. Chitundu et al. (2006) report
on the activities of a task-force to bring together stakeholders
within the cassava value chain in Zambia. They note that the
“level of public–private interaction [that the task force repre-
sented] does not come naturally in an economy such as
Zambia ... Mutual mistrust between the public and private
sector are [sic] common” (p. 19). The opportunity to bring
“interested players together around a table in search of very
concrete solutions to identified opportunities” was positive.
However, we also note that there are few mechanisms by
which most private sector actors (and especially indigenous
small or medium sized enterprises) can sanction the state if
it acts in a way that is damaging to their interests. Thus, delib-
erative fora alone may be insufficient to encourage greater pri-
vate sector investment, especially in politically high profile
value chains where politicians and public agencies are more
inclined to intervene.
Deliberative multi-stakeholder processes, facilitated by
donor-funded projects, have led to the development of ware-
house receipt systems in several African countries. These are
designed to increase traders’ access to capital for commodity
purchase and to promote private storage activity. The pres-
ence of inspection companies at coffee and cotton processing
plants in Tanzania has increased banks’ confidence to lend
to these processors. However, warehouse receipt systems have
had a less durable impact on maize trading, where lack of pri-
vate investment in storage has been one of the factors contrib-
uting to increased price volatility post-liberalization. Coulter
and Onumah (2002) observe that ad hoc government interven-
tions in maize markets can depress prices, causing losses to
traders (and farmers) who have invested in storage. This
undermines the incentive to invest, even where inventory credit
is available to assist this. Such difficulties have befallen the
Zambian scheme that Coulter and Onumah (2002) describe
[Gideon Onumah, pers.comm.]. Following Hall and Soskice
(2001), deliberation alone is insufficient to ensure sustained
strategic interaction where private parties are powerless to
influence (damaging) state actions. Following Jayne et al.
(2002), this is a case where the unpredictable and uncontrolla-
NOTES
1. As pointed out by one reviewer, the central challenge in hidden
information models is that the principal does not know the utility function
of individual agents. The literature on mechanism design (see Mas-Colell,
Whinston, & Green, 1995, chapters 14 and 23) thus explores ways in which
principals can induce agents to truthfully reveal their preferences. Even
these, however, generally assume that the preferences of individual agents
are drawn from a commonly known preference set. In the context of PPPs
for agricultural value chains, this would still require that principals have a
fair degree of understanding of private agents’ business models, strategies,
and objectives.
2. Hodge and Greve (2007) argue that short-term service contracts
should not be considered as PPPs. Rather, PPP contracts should be of
sufficient duration to provide incentives for significant investment by the
private partner in improved service delivery. Whether or not all the
contractual arrangements reviewed here meet this criterion, we argue that
common lessons emerge regarding state capacity and incentives for
effective contracting.
107
ble nature of state intervention—and fear of its implications
for future business development—is more of a binding con-
straint on private investment than failures in financial or other
markets.
9. CONCLUSIONS
The evidence base regarding the effectiveness of PPPs in
leveraging private investment in poorly functioning agricul-
tural value chains and the impact of such interventions on
smallholder producers is still limited. Many of the mechanisms
reviewed in this paper are “pioneering” examples: the number
of cases to observe is still low. Moreover, many are still suffi-
ciently recent that one cannot expect impact evaluations to
have been conducted. On the other hand, initial evidence on
their effectiveness in leveraging private investment should be
available, but can be hard to come by.
Insofar as evidence does exist, some mechanisms do seem to
alleviate binding constraints to private investment. For exam-
ple, reviews of challenge funds have tended to conclude that
they encourage (or hasten) investment with high social returns,
although the additionality question remains unresolved. Loan
guarantees to agro-dealers have enabled them to expand the
volume of their operations.
However, it should be noted that these cases were pioneered
by international organizations. This paper has not reviewed
any cases of successful deployment of PPPs for strengthening
agricultural value chains by African state agencies. 14 By con-
trast, it has noted cases where the main constraint on greater
private investment in such chains was unpredictable public
policy. Even where market failures are present, current state
capacity to design and administer effective PPPs may remain
a constraint to private investment. This paper has argued that
the general framework of performance incentives within many
African state agencies is at the root of this capacity problem,
which does mean that political commitment can make a differ-
ence. If so, there may be an advocacy role for AGRA, among
others, in persuading African states to experiment with PPP-
type arrangements for poorly functioning agricultural value
chains. However, this advocacy role should be accompanied
by full disclosure of available information on “pioneering”
examples for critical examination.
3. In Uganda Joughin and Kjaer (2010) suggest that the nature of
NAADS may now be changing as increasing political competition causes a
search for new sources of patronage.
4. A limited range of fertilizers is offered and coupons are for 50 kg bags,
rather than small packs. The administrative effort of distributing subsi-
dized fertilizer reduces time that extension staff have to provide advice on
fertilizer use (Chinsinga & Cabral, 2010).
5. A more recent survey by Moran et al. (2009) found that in 2007
around US$ 2.5 billion was invested in R&D for new products for
neglected diseases, with 90% of this coming from public (including
philanthropic) sources. About a quarter of donor funding was routed to
PPPs, such as the International AIDS Vaccine Initiative and the Medicines
for Malaria Venture.
6. Information on several of these funds can be accessed through
www.frich.co.uk/html/challengefunds.html. Other funds include the
108 WORLD DEVELOPMENT
Financial Deepening Challenge Fund (1999–2007) and the Business
Linkages Challenge Fund (2000–08).
7. Some form of systemic change is the “big win” that all challenge funds
would like to trigger [Mark Thomas, pers.comm.]
8. Irwin and Porteous (2005) judged the overhead rate of around 20% for
Financial Deepening Challenge Fund to be reasonable. This compares
with the 40% overhead rate for MATF, which funds much smaller projects
as well as having to engage in more capacity building with (potential)
bidders.
9. Khan (2000) argues that the ability of the South Korean state to
withdraw rents from companies that were not delivering on national
development priorities, so as to reallocate them to others, was central to its
success in stimulating economic growth. However, few other states have
been able to do this so effectively.
10. The Consultative Group on International Agricultural Research
(CGIAR) comprises 15 international agricultural research centers located
in Africa, Asia and Latin America.
11. Adesina et al. (2011) report that around 10,000 agro-dealers across 13
countries have now been trained, enhancing farmers’ access to inputs
through reduced distances from farmer to stockist and expanded ranges of
products sold.
REFERENCES
Adesina, A., Langyintuo, A., Bugo, N., Makinde, K., Bigirwa, G. and
Wakiumu, J. (2011) Improving farmers’ access to agricultural inputs and
finance: Approaches and lessons from Sub-Saharan Africa. Paper
presented at the IFAD Conference on “New Directions for Small-
holder Agriculture”, January 24–25, 2011, Rome.
AGRA (2009). Financing growth for Africa’s smallholder farmers, Alliance
for a Green Revolution in Africa, Nairobi.
AGRA, Standard Bank and Millennium Challenge Account Mozam-
bique, (2009) Fact sheet: New loan programme for African agricultural
development, Nairobi.
Anderson, J., & Feder, G. (2007). Agricultural extension. In R. Evenson,
& P. Pingali (Eds.). Handbook of agricultural economics (Vol. 3,
pp. 2343–2378). Elsevier.
Awortwi, N. (2004). Getting the fundamentals wrong: Woes of public–
private partnerships in solid waste collection in three Ghanaian cities.
Public Administration and Development, 24(3), 213–224.
Bahiigwa, G., Rigby, D., & Woodhouse, P. (2005). Right target, wrong
mechanism? Agricultural modernization and poverty reduction in
Uganda. World Development, 33(2), 481–496.
Bebbington, A., & Sotomayor, O. (1998). Demand-led and poverty-
oriented or just sub-contracted and efficient? Learning from (semi-)
privatised technology transfer programmes in Chile. Journal of
International Development, 10(1), 17–34.
Benin, S., Nkonya, E., Okecho, G., Pender, J., Nahdi, S., Mugarura, S.,
Kato, E. and Kayobyo, G. (2007) Assessing the impact of the National
Agricultural Advisory Service (NAADS) in the Uganda rural liveli-
hoods. IFPRI Discussion Paper 00724, International Food Policy
Research Institute, Washington DC.
Booth, D. and Golooba-Mutebi, F. (2011) Developmental patrimonialism?
The case of Rwanda. Africa Power and Politics Programme Working
Paper 16, Overseas Development Institute, London.
Boselie, D., Henson, S., & Weatherspoon, D. (2003). Supermarket
procurement practices in developing countries: Redefining the roles
of the public and private sectors. American Journal of Agricultural
Economics, 85(5), 1155–1161.
Buse, K., & Harmer, A. (2007). Seven habits of highly effective global
public–private health partnerships: Practice and potential. Social
Science and Medicine, 64(2), 259–271.
Chataway, J., & Smith, J. (2006). The International AIDS Vaccine Initia-
tive (IAVI): Is it getting new science and technology to the world’s
neglected majority?. World Development, 34(1), 16–30.
12. Rockefeller Foundation conducted an evaluation of the initial 2005–
08 loan guarantee prior to negotiating the extension. However, the
resulting report remains an internal document [Brinda Ganguly,
pers.comm. 15/07/2009].
13. Unlike the CERUDEB agreement described above, this is a “first
loss” agreement, whereby AGRA will cover all losses up to 20% of the
portfolio size in the first year, 15% in the second year and 10% in the third
year. This structure clearly assumes that Standard Bank will undergo a
learning process over the period of the guarantee.
14. Booth and Golooba-Mutebi (2011) suggest that the Rwanda horti-
culture sector may in due course provide positive examples. If so, this may
be “the exception that proves the rule” for the arguments advanced in this
paper. Booth and Golooba-Mutebi (2011) note the unusually strong
discipline and performance incentives within the post-2000 Rwandan
state. The model for PPPs in horticulture is likely to be the activities of
Tri-Star Investments S.A.R.L. (now registered as Crystal Ventures Ltd),
owned by the ruling Rwandan Patriotic Front party. Tri-Star, inter alia,
partnered the mobile company MTN, subsidizing its learning costs so as
to encourage it to establish operations in Rwanda. Linking to our
discussion in Section 8, Booth and Golooba-Mutebi (2011, p. 13) note that
Tri-Star “has a robustly internationalist approach to filling skill gaps in its
firms, with business efficiency and the meeting of strategic social objectives
taking precedence over commitments to local hiring and capacity
development”.
Chinsinga, B. and Cabral, L. (2010) The limits of decentralised governance:
the case of agriculture in Malawi. Policy Brief 033, Future Agricultures
Consortium, Brighton.
Chitundu, M., Droppelmann, K. and Haggblade, S. (2006) A value chain task
force approach for managing public–private partnerships: Zambia’s task-
force on acceleration of cassava utilization. Food Security Research
Project Working Paper #21, Michigan State University, Lusaka, Zambia.
Clark, P., & O’Brien, K. (2003). Fighting AIDS in Sub-Saharan Africa: Is
a public–private partnership a viable paradigm?. Medical Science
Monitor, 9(9), ET28–ET39.
Commission for Africa (2005) Our common interest: Report of the
Commission for Africa, London.
Coulter, J., & Onumah, G. (2002). The role of warehouse receipt systems
in enhanced commodity marketing and rural livelihoods in Africa.
Food Policy, 27(4), 319–337.
Cox, M. and Ortega, H. (2004) Chile: Origin and Evolution of a Privatized
Extension System. In: W. Rivera and G. Alex (Eds.). Extension reform
for rural development Volume 2: Privatization of extension systems—
Case studies of international initiatives. Washington DC, World Bank
ARD Discussion Paper No.9: 9–15.
Dorward, A., & Chirwa, E. (2011). The Malawi agricultural input subsidy
programme: 2005–6 to 2008–9. International Journal of Agricultural
Sustainability, 9(1), 232–247.
Enterplan (2004). Discussion document—FDCF: Assessing its achievements
and possible future directions. Report for DFID under contract CNTR
99 7888A.
Fafchamps, M. (2004). Market institutions in Sub-Saharan Africa: Theory
and evidence. Mass, MIT Press: Cambridge.
Future Agricultures Consortium (2009). Fertiliser subsidies: Lessons from
Malawi for Kenya. Future Agricultures Briefing Paper, Institute of
Development Studies, Brighton.
Hall, P., & Soskice, D. (2001). An Introduction to Varieties of Capitalism.
In P. Hall, & D. Soskice (Eds.), Varieties of capitalism: The institutional
foundations of comparative advantage (pp. 1–68). Oxford: Oxford
University Press.
Hanson, J., & Just, R. (2001). The potential for transition to paid
extension: Some guiding economic principles. American Journal of
Agricultural Economics, 8(3), 777–784.
Hart, O., & Holmstrom, B. (1987). The theory of contracts. In T. Bewley
(Ed.), Advances in economic theory: Fifth World Congress (pp. 71–155).
Cambridge: Cambridge University Press.
 PUBLIC–PRIVATE PARTNERSHIPS LEVERAGE PRIVATE INVESTMENTIN AGRICULTURAL VALUE CHAINS IN AFRICA
Henson, S., & Reardon, T. (2005). Private agri-food standards: Implications
for food policy and the agri-food system. Food policy, 30(3), 241–253.
Hodge, G., & Greve, C. (2007). Public–private partnerships: An international
performance review. Public Administration Review, 67(3), 545–558.
Hodge, G., & Greve, C. (2009). PPPs: The passage of time permits a sober
reflection. Economic Affairs, 29(1), 33–39.
Hubbard, M. (2003). Developing agricultural trade: New roles for
government in poor countries. Basingstoke: Palgrave Macmillan.
Irwin, D. and Porteous, D. (2005) Financial Deepening Challenge Fund
Strategic Project Review. Report funded by DFID and BMGF, Irwin
Grayson Associates, Northumberland.
Israel, A. (1987). Institutional development: Incentives to performance.
Baltimore: Johns Hopkins University Press.
Jayne, T., Govereh, J., Mwanaumo, A., Nyoro, J., & Chapoto, A. (2002).
False promise or false premise? The Experience of food and input
market reform in Eastern and Southern Africa. World Development,
30(11), 1967–1985.
Jayne, T., Govereh, J., Wanzala, M., & Demeke, M. (2003). Fertilizer
market development: A comparative analysis of Ethiopia, Kenya, and
Zambia. Food Policy, 28(4), 293–316.
Jones, L., Quigley, J., & Foster, G. (2006). Vouchers revisited—Can small
enterprises save government programmes?. Small Enterprise Develop-
ment, 17(4), 43–51.
Joughin, J., & Kjaer, A. (2010). The politics of agricultural policy reform:
The case of Uganda. Forum for Development Studies, 37(1), 61–78.
Kayaga, S. (2008). Public–private delivery of urban water services in
Africa. Proceedings of Institution of Civil Engineers: Management,
Procurement and Law, 161(4), 147–155.
Khan, M. (2000). Rent-seeking as process. In M. Khan, & K. Jomo (Eds.),
Rents, rent-seeking and economic development: Theory and evidence in
Asia (pp. 70–144). Cambridge: Cambridge University Press.
Kherallah, M., Delgado, C., Gabre-Madhin, E., Minot, N., & Johnson,
M. (2000). The Road Half Travelled: Agricultural Market Reform in
Sub-Saharan Africa, Food Policy Report. Washington DC: Interna-
tional Food Policy Research Institute.
Langyintuo, A., Mwangi, W., Diallo, A., MacRobert, J., Dixon, J., &
Bänziger, M. (2010). Challenges of the maize seed industry in eastern
and southern Africa: A compelling case for private–public intervention
to promote growth. Food Policy, 35(4), 323–331.
Mas-Colell, A., Whinston, M., & Green, J. (1995). Microeconomic Theory.
New York: Oxford University Press.
Moran, M., Ropars, A., Guzman, J., Diaz, J. and Garrison, C. (2005). The
new landscape of neglected disease drug development, Pharmaceutical
R&D Policy Project: George Institute for International Health/
London School of Economics/Wellcome Trust, London.
Moran, M., Guzman, J., Ropars, A.-L., McDonald, A., Jameson, N.,
Omune, B., et al. (2009). Neglected disease research and development:
How much are we really spending?. PLoS Medicine, 6(2), e1000030.
doi:10.1371/journal.pmed.1000030.
Narrod, C., Roy, D., Okello, J., Avendano, B., Rich, K., & Thorat, A.
(2009). Public–private partnerships and collective action in high value
fruit and vegetable supply chains. Food Policy, 34(1), 8–15.
Nwaka, S. (2005). Drug discovery and beyond: The role of public–private
partnerships in improving access to new malaria medicines. Transac-
tions of the Royal Society of Tropical Medicine and Hygiene, 99(Suppl.
1), S20–S29.
109
Otsuki, T., Wilson, J., & Sewadeh, M. (2001). Saving two in a billion:
Quantifying the trade effect of european food safety standards on
African exports. Food Policy, 26, 495–514.
Palmer, N., & Mills, A. (2005). Contracts in the real world: Case studies
from Southern Africa. Social Science and Medicine, 60(11), 2505–2514.
Pingali, P., & Traxler, G. (2002). Changing locus of agricultural research:
Will the poor benefit from biotechnology and privatization trends?.
Food Policy, 27(3), 223–238.
Poulton, C., Dorward, A., & Kydd, J. (2010). The future of small farms:
New directions for services, institutions, and intermediation. World
Development, 38(10), 1413–1428.
Poulton, C., Mangheni, M., Okwadi, J., Kilewe, A. (2006) External
evaluation report for the maendeleo agricultural technology fund. Report
to FARM-Africa, Nairobi.
Poulton, C., Kydd, J., & Dorward, A. (2006). Overcoming market
constraints on pro-poor agricultural growth in Sub-Saharan Africa.
Development Policy Review, 24(3), 243–277.
Ramiah, I., & Reich, M. (2006). Building effective public-private partner-
ships: Experiences and lessons from the African Comprehensive HIV/
AIDS Partnerships (ACHAP). Social Science and Medicine, 63(2),
397–408.
Raynolds, L. (2004). The globalization of organic agro-food networks.
World Development, 32(5), 725–743.
Rosiello, A., & Smith, J. (2008). A sociological economy of HIV/AIDS
vaccine partnerships: Case studies from Africa and India. International
Review of Sociology, 18(2), 283–299.
Rusike, J. and Gandanhamo, T. (1999) The impact of the CARE
Zimbabwe Agribusiness Entrepreneur Network and Training (AGENT)
programme: A survey of farm input dealers in Makoni North and South
Districts. Report produced for DFID project R6720(H) “Linking Seed
Producers and Consumers: Diagnosing Constraints to Institutional
Performance”.
Spielman, D., Hartwich, F. and von Grebmer, K. (2007) Sharing Science,
Building Bridges, and Enhancing Impact: Public–Private Partnerships in
the CGIAR. Discussion Paper 00708, International Food Policy
Research Institute, ISNAR Division, Washington DC.
Tati, G. (2005). Public—Private partnership (PPP) and water-supply
provision in urban Africa: The experience of Congo-Brazzaville.
Development in Practice, 15(3–4), 316–324.
van de Walle, N. (2001). African Economies and the Politics of Permanent
Crisis, 1979–1999. New York: Cambridge University Press.
Van Ham, H., & Koppenjan, J. (2001). Building public–private partner-
ships: Assessing and managing risks in port development. Public
Management Review, 4(1), 593–616.
Warner, M., Kahan, D. and Lehel, S. (2008) Market-oriented agricultural
infrastructure: Appraisal of public–private partnerships, FAO/ODI,
Rome.
Williamson, O. E. (1985). The economic institutions of capitalism. New
York: The Free Press.
World Bank (2007). World Development Report 2008: Agriculture for
Development, Washington DC.
World Bank (2006). Developing rural agricultural input supply systems for
farmers in Africa. Case study prepared for Fertilizer Toolkit.
World Bank (2009). Awakening Africa’s sleeping giant: Prospects for
commercial agriculture in the Guinea Savannah Zone and beyond
directions in development. Washington DC: World Bank.