World Development Vol. 30, No. 11, pp. 1967–1985, 2002
Ó 2002 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
0305-750X/02/$ - see front matter
www.elsevier.com/locate/worlddev
PII: S0305-750X(02)00115-8
False Promise or False Premise? The Experience
of Food and Input Market Reform in Eastern
and Southern Africa
T. S. JAYNE
Michigan State University, USA
J. GOVEREH
Food Security Research Project, Lusaka, Zambia
A. MWANAUMO
Ministry of Agriculture, Lusaka Zambia
J. K. NYORO
Egerton University, Nairobi, Kenya
and
A. CHAPOTO *
Michigan State University, USA
Summary. — The literature on the effects of agricultural market reform in Africa is sharply divided
and inconsistent. This article attempts to reconcile opposing viewpoints on the effects of food and
input market policy reform in eastern and southern Africa. Drawing from studies of Ethiopia,
Kenya, Malawi, Zambia, and Zimbabwe, we argue that a major source of the controversy stems
from assumptions that countries have actually moved to a liberalized market environment. We find
that many of the most fundamental elements of the reform programs either remain unimplemented,
were reversed within several years, or were implemented in such a way as to negate private sector
investment incentives. A framework is developed for explaining why some countries have been able
to liberalize their food and fertilizer markets while others have not. These findings have implications
for how donor assistance and policy-oriented research can more constructively contribute to an
improved policy environment.
Ó 2002 Elsevier Science Ltd. All rights reserved.
Key words — marketing, reform, liberalization, agriculture, institutions, Africa
* Support for this study was provided by the Food Security and Productivity Unit, Office of Sustainable Development, USAID Bureau for Africa; USAID/Zambia;
and USAID/Kenya. Earlier versions of this article have
benefitted from the comments of Steve Buccola, Balu
Bumb, Munhamo Chisvo, Paul Clements, Ralph Cummings, Carl K. Eicher, Steven Haggblade, Stephen
Jones, Isaac Minde, Mulinge Mukumbu, Were Omamo,
Colin Poulton, Allan Schmid, James Shaffer, John Staatz, David Tschirley, Nicholas van de Walle, Michael
Weber, and four anonymous reviewers. The authors
take sole responsibility for any errors or omissions.
Final revision accepted: 30 May 2002.
1967
1968
WORLD DEVELOPMENT
The MMDÕs overzealous pursuance of capitalist policies under the directives of the IMF and World Bank
have registered no success––they have been a disaster
for the nation. We believe that neoliberalism is not
simply an economic doctrine, it is a political project
that seeks to perpetuate the present unfair, exploitative
world economic order. (‘‘An affront to intelligence,’’
Editorial Comments, The Zambia Post, August 10
2000).
The desperate state of our economy bears testimony
to the damage wrought by delayed reform. . . Indeed
there is a price to pay for reform, but there is an even
bigger price to pay for failing to reform. (Frederick
Chiluba, President of Zambia at the time he was
quoted in the Southern Africa Integrated Regional
Information Network News Brief, October 25, 2000).
1. INTRODUCTION
One of the most contentious policy debates in
Africa concerns how agricultural markets
should be organized. Most African governments initiated programs of agricultural market reform in the 1980s as part of economic
structural adjustment programs. Yet many politicians remain unconvinced of the most fundamental elements of the process. In some cases,
politicians openly contend that agricultural
market reform has contributed to the crisis
facing small farm households across the continent, that private sector response has been too
slow and too weak to spur development, and
that the state should get back into direct distribution of strategic inputs and/or commodities.
The academic literature on agricultural
market reform in Africa also ranks among the
most divided and inconsistent within the field
of economic development. While some scholars
find that market reform has generally supported agricultural growth and food security, a
growing literature has explained the poor record of reform in terms of inadequate attention
to the institutional foundations of markets and
weak infrastructure, all of which lead to growth
impeding market failures. 1 This literature has
reinforced the common mainstream perspective
that policy reform has been a false promise. 2
An alternative view examined in this paper
is that market reform has in some cases been
a false premise: it has not actually been implemented, and hence its effects cannot be measured. We argue that a major source of the
controversy stems from assumptions that
countries have actually moved to a liberalized
market environment. Perhaps ironically, market reform has been most severely criticized in
countries where direct government involvement
in marketing has remained entrenched, such
as the former settler maize economies of eastern and southern Africa. In these countries, we
conclude that the basic model of input and
grain market reform has yet to be tested, and
hence conclusions that it has not produced its
anticipated effects are largely unjustified. While
it is difficult to control for all relevant factors,
evidence indicates that countries actually implementing agricultural market reform are
performing no worse, and in some cases clearly
better than those implementing only de jure
reform.
The paper also identifies the phenomenon
of non-reform and policy reversal: Why have
countries such as Uganda, Mali, Ghana, and
Mozambique been able to set a relatively sustained course toward food and input market
reform, while others such as Ethiopia, Kenya,
Malawi, Zambia, and Zimbabwe have not despite the billions of dollars in aid-conditionality
that were the quid pro quo for implementing the
reforms? An understanding of why key aspects
of agricultural policy reform have proven so
difficult may hold valuable lessons for future
development strategies. We conclude that mobilizing the political support for constructive
market reform will require seriously confronting the incentive dilemmas not only within
African governments, but also within donor
organizations and governments in high-income
countries.
2. WHY ARE THE EFFECTS OF REFORM
SO CONTROVERSIAL?
The watershed ‘‘Berg Report’’ (World Bank,
1981) marked international lenders and donorsÕ
attempts to reduce African governmentsÕ control over agricultural markets and increase the
role of the private sector. The basic reform
policy package included two broad features:
privatization (withdrawal of state agencies from
grain pricing and marketing activities) and liberalization (the relaxation of regulatory controls on private marketing). Reform advocates
anticipated that reform would induce agricultural growth. But, agricultural growth in most
eastern and southern African countries in the
1990s has been unimpressive, and there has
been a widespread tendency to attribute this, at
least partially, to the failure of agricultural re-
FALSE PROMISE OR FALSE PREMISE?
1969
Table 1. Per capita agricultural production indices
1989–91 ¼ 100
Year
1980–84
1985–89
1990–94
1995–00
Ethiopia
Kenya
Malawi
Tanzania
Uganda
Zambia
Zimbabwe
116.1
100.0
94.6
98.5
93.1
100.7
92.7
86.6
124.4
107.8
96.3
118.0
105.0
103.7
92.1
83.4
92.4
93.0
99.2
94.6
88.2
101.2
92.7
82.6
99.8
107.5
90.4
96.9
Source: FAO.
Table 2. Total fertilizer nutrient usea
Year
1980–84
1985–89
1990–94
1995–00
Thousands of metric tons
Ethiopia
Kenya
Malawi
Tanzania
Zambia
Zimbabwe
41.2
71.6
92.7
156.0
75.8
115.0
112.7
126.5
37.4
45.2
57.4
52.2
29.2
44.5
44.2
31.0
74.2
83.3
70.2
50.9
159.7
156.4
156.9
165.9
Source: FAO.
Uganda and Mozambique fertilizer use trends are not reported because they have consumed less than 5,000 and
10,000 tons of total nutrient per year since 1980.
a
form (UNCTAD, 2001). Fertilizer use has
often declined and production levels have been
outstripped by population growth in the postreform period in most of the region (Tables 1
and 2). While absolute agricultural production
has increased in the seven countries examined
in these tables, per capita production has been
lower during 1995–2000 than during 1980–84 in
every country except Uganda.
In recent years, with increasing appreciation
of institutional factors in the development of
markets, the early reform programs have been
critiqued for their naivete in assuming that
markets would blossom if only freed of policy
constraints. Even in the design of more recent
reform programs, support for the development
of market institutions and infrastructure to allow smallholder farmers to respond to reform
are arguably still underemphasized. By ‘‘institutions,’’ we mean the rules of the game governing exchange (North, 1987) and agree with
Kydd and DorwardÕs (2001) emphasis on the
need to identify workable ‘‘institutional arrangements’’ to reduce the costs and risks of
entering and participating in markets. It is well
documented that most agricultural marketing
systems in Africa are hampered by market
failure problems such as high transaction costs
and risks, costly means of insuring against risk,
impacted information, weak provision of seasonal finance for purchased farm inputs, and
contract enforcement problems, to name a few
(Barrett, 1997; Barrett & Carter, 1999; Dercon,
2001; Gabre-Madhin, 2001; Kydd & Dorward,
2001; Minten & Kyle, 1999; Shaffer, Weber,
Riley, & Staatz, 1985; Schmid, 1992).
There are, however several problems in generalizing from solid empirical case studies of
market failure to an overarching conclusion
that liberalization has failed. First, market
institutions evolve more gradually than strokeof-the-pen policy changes (Goldsmith, 2001;
Shaffer, 1969). In underdeveloped areas, and
especially where markets have been suppressed, the initial years of reform programs were
by definition constrained to operate in an environment of weak market institutions. As
shown by Bardhan (1989), North (1987),
Shaffer (1969) and Shaffer et al. (1985), the
development of market institutions depend on
the policy environment, and more broadly on
the governance environment. Trade-promoting
market institutions rarely develop in a policy
environment that inhibits the emergence of
market actors (Seidman, 1973; Thomas & Grindle, 1990). It might therefore not be considered
surprising to find studies identifying weak
marketing institutions as impeding the effectiveness of markets. 3 It would be a step forward, conceptually, to ask which institutions
must be in place before reform should be
attempted, and which institutions cannot be
expected to develop until reform has had sufficient time to enable a critical mass of traders
1970
WORLD DEVELOPMENT
to develop new institutions over time, and
lobby for publicly-provided goods in support of
markets (Schmid, 1992).
Second, recent efforts to explain the poor
track record of policy reform––focusing on, for
instance, the naive presumptions that policy
reform was sufficient––while clearly part of the
story, appear to take on face value that the
reforms really occurred. Particularly in many
eastern and southern African countries, a close
examination reveals that many of the most
fundamental elements of the reform process
either remain unimplemented or were reversed
within several years. Consider the following
examples:
(a) Zimbabwe
Price controls on maize meal were reimposed in 1998, five years after the government eliminated them under a 1993 World Bank/IMF
structural adjustment loan program. The Grain
Marketing Board (GMB) has remained the
dominant buyer of grain throughout the reform
process (Durevall & Mabugu, 2000). The GMB
has reverted back to a two-tiered maize pricing
structure; selling maize at a lower price to politically-influential large-scale milling firms than
it does to other buyers. Moreover, throughout
the reform program, the GMB remains the sole
legal exporter and importer of maize, and
continues to offer pan-territorial and pan-seasonal maize prices as it did prior to the reform
program. While this policy environment provided niches for new entry and investment at
certain stages of the maize supply chain, notably in assembly, local milling and retailing, it
impeded private investment at other key stages
(Jayne & Jones, 1997). In 2001, the government
banned all private maize trade (Table 3).
(b) Zambia
The former state maize marketing board,
NAMBOARD, was abolished in 1989 but since
1992, government has designated various parastatal or private companies to distribute fertilizer on behalf of government (Table 4). The
Food Reserve Agency, formed in 1995 and
initially envisioned to play a limited role of
holding maize buffer stocks, became the countryÕs largest distributor of fertilizer in 1997 and
1998. In 1999 with donors increasingly calling
for the government to withdraw from fertilizer
distribution to gain access to external financing,
the government responded by contracting private companies as logistical ‘‘agents’’ to distribute fertilizer to recipients designated by the
Ministry of Agriculture. The designated agents
received a flat fee from government for every
ton distributed. These government programs
have continued to distribute fertilizer at subsidized prices and repayment rates generally
under 40%, undercutting private firmsÕ ability
to distribute fertilizer at commercial prices
(Copestake, 1998; Govereh et al., 2002). While
the government fertilizer distribution program
was ostensibly transformed several times in
response to deficiencies noted in donor-funded
assessments, the central feature throughout the
Table 3. Zimbabwe: chronology of maize market reform, 1991–2000
1991
1993–94
1996–97
1998
2000
2001
––Announcement of Economic Structural Adjustment Program, including a grain market reform
component supported by World Bank, USAID and other donors.
––Control of private maize trade progressively relaxed; registered millers still obliged to procure maize
from GMB.
––Retail maize meal prices decontrolled
––GMB still sole legal importer and exporter of maize
––Formation of Zimbabwe Agricultural Commodity Exchange.
––Maize import/export remains under GMB monopoly.
––Subsidies on GMB trading margin narrows price range within which private traders can operate.
––GMB raises its maize selling price to millers to adjust to prevailing market prices. Millers responded
by raising roller meal price by 21%, causing food riots of January 1998.
––Government reintroduces controls on maize meal prices, May 1998.
––GMB enters the maize milling industry.
––Price controls on maize meal still exist; GMB retains pan-territorial and pan-season producer price
and selling prices
––GMB retains exclusive monopoly over maize import and export.
––Government bans all private maize trade
Sources: Takavarasha (1994); Chapoto (2000); Chisov (2000); IRIN News Service (2001).
FALSE PROMISE OR FALSE PREMISE?
1971
Table 4. Zambia: chronology of fertilizer market reform, 1991–2000
Prior to
1990
1990–93
1994–96
1997–98
1999–
2000
––Importation, distribution, and pricing of fertilizer handled by government marketing agency,
NAMBOARD. Fertilizer subsidy averaged roughly 50% of full retail cost.
––Economic Structural Adjustment Program initiated 1991.
––NAMBOARD abolished in 1990, but fertilizer and credit marketing functions transferred to other
state agencies. State still sets retail fertilizer prices to be paid by smallholders until 1992. Credit recovery
rates under 30%.
––Govt continues to control smallholder fertilizer imports.
––Agricultural Credit Management Programme (ACMP) launched. Government contracts private firms
to distribute fertilizer and seed on credit to farmers. Allocation process determined by government.
––Credit recovery rates near 30%. Private firms asked to absorb some of the risks of government loan
default; they refuse and exit the market. ACMP program abandoned.
––Food Reserve Agency, established in 1995 to manage the national food reserve, takes over fertilizer
distribution on credit to smallholders.
––Donors cease financing of fertilizer imports.
––Pan-territorial pricing re-introduced for FRA-distributed fertilizer; makes private sector fertilizer
uncompetitive in outlying areas.
––Government again contracts private firms to import and distribute fertilizer; in response to
aid-conditionality agreement with World Bank that government not directly distribute fertilizer.
––2000 season loan repayment rate 43%.
––Virtually no fertilizer importers and wholesalers operate in Zambia other than those contracted by
government to distribute fertilizer on their behalf at subsidized prices.
Sources: Govereh et al. (2002); Jayne et al. (1999); Mwanaumo (1999); Pletcher (2000).
reform process has been that government selected the recipients of subsidized fertilizer according to processes that lacked transparency
and allegedly involved interference from state
officials. 4 After almost a decade of aid-conditionality agreements with the World Bank, new
entry of commercial fertilizer firms has been
limited due to the uncertainties associated with
government distribution programs (Govereh
et al., 2002). The party winning the 2001 election ran on a platform of re-introducing agricultural marketing boards to provide fair maize
prices to farmers and consumers and the legislation for the new Crop Marketing Authority
is set for parliamentary deliberation in 2002.
(c) Ethiopia
As part of aid-conditionality agreements, the
Ethiopian government has curtailed the operations of its official state marketing board. But,
in 1995 it permitted the creation of regional
holding companies which are reputedly owned/
managed by influential members of the ruling
party. These holding companies enjoy nearmonopoly rights for the distribution of fertilizer
in their respective regions. Two large private companies have been forced to exit the
fertilizer market because regional governments
have been actively promoting the holding companies while simultaneously raising barriers for
private sector companies (Table 5). There are
widespread allegations that the regional holding
companies are instruments of government policy (Stepanek, 1999).
(d) Kenya
Aid-conditionality agreements pertaining to
maize market reform commenced in the late
1980s with the Agricultural Sector Adjustment
Operation (World Bank) and the Cereal Sector
Reform Programme (EU). Fertilizer markets
have been comprehensively liberalized (Wanzala, Jayne, Staatz, Mugera, & Kirimi, 2002).
Other elements of the reform process have
been marked by increased political interference
in the decisions of key cooperative and jointventure marketing organizations (Kanyinga,
1994). Rent-seeking arrangements that created resistance to reform in the early stages
of the process have been reestablished within
the evolving ‘‘market-oriented’’ institutions
that have developed since reform, a phenomenon that has been observed more widely in
other countries (Bates & Krueger, 1993). The
state-owned marketing board has continued to
support maize prices selectively in areas where
the government derives its main political support (Table 6). Maize import tariffs, marketing
board price supports, and relatively high transport costs have combined to make maize prices
in Kenya among the highest in the world for
countries where maize is a staple crop (Jayne
et al., 1999).
1972
WORLD DEVELOPMENT
Table 5. Ethiopia: fertilizer market reform, 1993–2000
1993
––Ban on private imports of fertilizer lifted; previously, the state Agricultural Inputs Supply
Corporation handles the bulk of fertilizer importation and distribution.
1995
––Ethiopian government agrees to liberalize fertilizer market as part of aid-conditionality agreements
with donors.
––Pan-territorial fertilizer pricing by AISCO impedes private sector fertilizer delivery in outlying areas
––Regional holding companies allegedly owned/managed by influential ruling elites begin engaging in
fertilizer trade.
1996
––Government announces decontrol of import and wholesale prices; retail control prices maintained.
––Government introduces regional ‘‘tenders’’ to award monopoly distribution privileges by district;
regional holding companies win most of the tenders.
1997–2000 ––Price controls on retail fertilizer prices relaxed in 1997.
––Regional tender process generally results in monopoly distribution rights being given to regional
holding companies. State fertilizer marketing firm and regional holding companies have 81% of total
market share in 1998.
––Government New Agricultural Extension Programme accounts for 67% of all fertilizer used by
smallholders in the country; government procures 85% of its fertilizer supply for this program from the
regional holding companies.
Source: Stepanek (1999).
Table 6. Kenya: chronology of maize market reform: 1988–2000
1986–87
1988–91
1992
1994
1996
1998
1999–2000
––Government agrees to implement Agricultural Sector Adjustment Program with World Bank
(1986).
––The state-run National Cereals and Produce Board (NCPB) sets fixed producer and consumer;
private trade across district boundaries prohibited except by license. Commercial millers obligated
to purchase maize supplies from NCPB. Maize meal prices set by government.
––Government agrees to implement Cereal Sector Reform Program with EU (1988) and
Agricultural Sector Adjustment Program II with World Bank (1991).
––Government relaxes movement and price controls on private maize trading
––NCPB continues setting fixed producer and consumer prices; all commercial millers obligated to
purchase a portion of their supplies from NCPB. Maize meal prices fixed by government.
––Government reimposes restrictions on private maize movement and trade
––Government introduces variable import duty following substantial imports by private traders
––The margin between NCPB buying and selling price is too low to cover costs; implicit subsidy on
its trading margin hampers private tradersÕ ability to compete and invest in the marketing system.
––Exports banned after a weak harvest, later replaced by a 25% tariff on maize imports
––Tariff on maize imports increased from 25% to 33% in anticipation of a large harvest.
––Government announced that NCPB would not purchase any maize domestically.
––Maize stabilization policy continued, with the NCPB purchasing 72,000 tonnes of domestically
produced maize.
––Maize import tariff fluctuates from 33% to zero to 75% to 25% over 15-month period (March 1999
to May 2000).
Sources: Nyoro et al. (1999).
Maize import controls remain in Zimbabwe,
Ethiopia, and Kenya. MalawiÕs state-owned
marketing board, ADMARC, is still the dominant maize buyer in Malawi. During 1994–98,
ADMARC and subsidized donor-funded fertilizer accounted for 62% of reported fertilizer
sales in the country, although this proportion
declined steadily over the period (MASIP,
2000). In 1999 and 2000, the government distributed free ‘‘starter packs’’ of maize seed and
fertilizer to most rural households in the
country, but analysis concluded that the pro-
gram may have undercut the market for commercial input suppliers (MASIP, 2000).
These brief examples show that many policy
barriers continue to inhibit the development of
competitive input and commodity markets in
numerous countries. Legitimate objectives may
underlie these policy decisions, but it would be
inappropriate to expect much private sector
response in such an environment. These cases
illustrate how de jure market reform can be
implemented in such a way as to maintain de
facto control over the system. In such cases, the
FALSE PROMISE OR FALSE PREMISE?
market reform process clearly proceeded in a
manner that was unintended by its advocates.
A frequent consequence of evaluating the
effects of reform without carefully determining
how the reforms were actually implemented is
the premature and possibly exaggerated finding
of ‘‘market failures.’’ Researchers have in some
cases correctly noticed a lack of private sector
response but have incompletely identified its
causes. 5 Certainly there are market failure
problems, but explanations for such problems
have emphasized under investments in publiclyprovided goods and market institutions and
have tended to neglect the importance of unresolved policy barriers. In such cases, therefore, frequently-heard conclusions that reform
has failed to produce its intended effects may be
inappropriate, because they are based on a false
premise that reform occurred.
Moreover, the finding of market failures has
often been confused with a failure of the private
sector. Correcting for market failures––arising
from high transaction costs, poor infrastructure, weak systems for contract enforcement,
information asymmetries, and even collusive
trader behavior––is one of the major roles of
government through policy and public goods
investments. The most difficult part of improving agricultural markets is not necessarily the
technical aspect of identifying appropriate institutional and infrastructure investments to
overcome market failure, but rather developing
the incentives for governments to deliver such
investments after being identified.
It is commonly asserted that structural adjustment has involved such serious cutbacks in
governmentsÕ budgets that they have been unable to providing these market-supportive investments at the time that they are most
needed. The empirical record is not so clear.
Several studies show that state revenues as a
proportion of GDP on average declined at
most one percentage point during the course of
World Bank adjustment programs (Jayarajah
& Branson, 1995; Nashashibi, Gupta, Liuksila,
Lorie, & Mahler, 1992). Nashashibi et al. (1992)
provide data suggesting that real government
revenues went up in nine African cases and
down in nine, relative to a base year. According
to published World Bank data across Africa
(excluding South Africa and Nigeria), government revenues have declined from an average
of 16.3% of GDP during 1975–84 to 15.8%
during 1990–96. We argue below that the reasons for selective government under-provision
of market-supportive institutions and invest-
1973
ments, as with policy reform, need to be put in
their political-economy context.
3. EXPLAINING THE PATH OF REFORM
IMPLEMENTATION
A broad assessment of the agricultural reform literature in Africa shows three basic
forms. First, some governments have been
committed to a program of market reform.
Examples in eastern and southern Africa would
arguably include maize and fertilizer marketing
in Mozambique and Uganda. Mali and Ghana
are two other countries commonly cited for
their relatively steady adherence to cereal
market reforms (Dembele & Staatz, 2002;
World Bank, 1994, 2000). This category would
also include countries that may have temporarily reversed course but over the long run
have moved toward a fundamentally marketoriented system (e.g., TanzaniaÕs food markets). A second path includes countries that
have openly resisted reform or reimposed controls after some experimentation with reform.
This category is characterized by transparent
resistance to reform (e.g., maize in Zimbabwe
prior to 1991 and after 1998). The third path
involves de jure reform and de facto state control of marketing, such as the previous example
of fertilizer markets in Zambia. The key feature
of this path is the maintenance of state control
over key aspects of marketing while ostensibly
implementing reforms. Buccola and McCandlish’s (1999) description of coffee marketing in
Malawi and StepanekÕs (1999) analysis of fertilizer in Ethiopia also exemplify this category
of veiled state control.
What accounts for why some African governments have managed to restructure the
stateÕs role in markets, comprehensively remove
policy barriers to private trade, and nurture
key marketing institutions, while others have
achieved very little private sector response
despite at least a decade of attempting such
reforms? Unique country-level conditions warrant caution against overgeneralization. At a
broad level, however, there appear to be two
interrelated determinants: the degree of fiscal
crisis facing the government, and the degree
of perceived political harm posed by reform.
(a) Fiscal crisis
Fiscal crises forced governments to accept
the conditions of international lenders and
1974
WORLD DEVELOPMENT
donors to gain access to external financing
(Devarajan, Dollar, & Holmgren, 2001; Jayne &
Jones, 1997). These conditions were in the form
of agricultural reform programs and economywide structural adjustment programs. While
these programs were official government policy,
many African leaders distanced themselves
from the reforms and later reversed them soon
after gaining access to donor financing (Devarajan et al., 2001; van de Walle, 2001). Some
leaders openly resisted agricultural reform
when state budget deficits were manageable
(Kenya up to the mid-1980s; Zimbabwe
throughout the 1980s) but these same leaders
later accepted reform after their statesÕ deficits
reached crisis proportions.
(b) Political harm
We highlight three components determining
the perceived political risks of implementing
agricultural reform: concerns as to the effects of
reform implementation on constituentsÕ welfare; efforts to articulate the rationale for reform to the public; and the commodityÕs role
in fulfilling patronage and coalition-building
objectives.
(i) Genuine concerns over the effects of the
reforms on constituents’ welfare
Policy change produces both winners and
losers. Especially in countries where subsidies
during the control regime provided widespread
benefits to millions of consumers and producers, politicians have expressed reluctance to
accept the reforms because they signaled a retreat from the ‘‘social contract’’ commitments
made earlier upon African governmentsÕ assumption to power after independence. The rise
of multiparty politics has put additional pressures on politicians to maintain policies with
popular support, despite their potentially adverse long-run consequences. Within the fledgling pluralist societies that are emerging in
some countries, opposition parties, intellectuals, trade unionists, and sometimes nongovernmental organizations (NGOs) assail policy
reforms that incumbent governments have ostensibly accepted. Concern over how policy
reform would fare among the masses sometimes motivates leaders to respond to constituents perceived interests for political gain even
if doing so creates compliance problems with
the donors (van de Walle, 2001).
(ii) Efforts to articulate the rationale for reform
There has been a tendency for policy makers
in the region to view the state and the private
sector as substitutes. To some extent, the ‘‘state
vs. markets’’ view was accentuated by the fact
that few politicians in Southern Africa have
ever experienced a market-oriented economy
before, the history of controls dating back to
the 1930s in Zimbabwe, South Africa, Kenya,
and Zambia. Many government officials explicitly embraced Marxist principles after independence and received their educations in
formerly communist countries. The experiences
of colonialism in many cases reinforced the
Marxist perspective and increased the appeal of
‘‘African socialism’’ (Jenkins, 1997). Many
African policymakers still do not fully accept
the logic or assumptions of a market economy,
despite being compelled to move in this direction under pressure from international lenders
and donors. Furthermore, internal logic of the
‘‘state vs market’’ framework provided very
little rationale for the state to spend public
funds to support private entrepreneurs, who
were seen as competitors. State support for the
private sector was also impeded by the widespread suspicion of ethnic minorities that are
heavily engaged in private food and input
trading in the region (Keyter, 1975; Mosley,
1983). These views partly explain the vast
under-provision of market-supporting public
goods in the region.
Largely as a result of these perceptions, the
private sectorÕs response to the reform process
has been hampered by what has variously been
referred to as ‘‘vicious cycles’’ and ‘‘self-fulfilling prophesies’’ (Jayne et al., 1999). PoliticiansÕ
skepticism over the ability of the private sector
to respond has led to calls for the return to state
intervention, which then scares off private investment, thereby making the rationale for
government intervention more compelling (see,
for example, Mwanaumo, 1999). According to
a worldwide survey of private entrepreneurs,
African business leaders in particular complain
that they are not kept informed about new rules
and regulations, and that they regularly worry
about retroactive changes (Brunetti, Kisunko, &
Weder, 1997). The private sectorÕs response to
reform is likely to be strengthened by more
consistent statements from officials and a perceived commitment to a stable long-term policy
environment. Actions to provide greater policy
stability may become increasingly necessary in
countries such as Zimbabwe, Zambia, and
Kenya, where the private sector has in a num-
FALSE PROMISE OR FALSE PREMISE?
ber of cases been hurt by responding to positive policy incentives, only to see these policies
reversed at a later stage (Copestake, 1998;
Govereh et al., 2002; Jayne et al., 1999).
Conclusions that politicians were not adequately committed to the reforms is often indicative of inadequate design and/or resources for
explaining and educating the public and the
bureaucracy on how the reforms work. Importantly, and in hindsight, there was little effort to
counteract government perceptions of being
spectators to the process––that the reforms required little on their part other than getting out
of the way. One approach to overcoming entrenched ideological convictions is to provide
concrete and easily digestible information on
how the reforms work, and what kind of supportive actions are required by government for
benefits to manifest. The program has to be
expected to build its own constituency. In the
end, the program cannot succeed if a sufficient
mass of local bureaucrats do not understand
and promote it, and hence the entire reform
program strategy can be recast in terms of this
objective (Clements, 2001). Articulating the
concept of reform in a way that more explicitly
rests on the market-supportive role of government may help nurture a greater awareness that
the model of reform posits no deterministic
outcomes––it has a chance to succeed only if
governments undertake important activities to
support the process. Dembele and Staatz (2002)
argue that government ownership of key aspects
of the reform process was critical to the success
of MaliÕs cereal sector reforms. Greater efforts
to engage the public and government in the
logic of reform is likely to be a necessary but
insufficient condition to foster greater local
ownership and commitment.
(iii) Clientelism and patronage
Economic policy analysis has largely ignored
problems identified in the literature on patronage i.e., the political logic of a system in
which the authority of the state is diverted to
enhance private power rather than public interests. As argued by van de Walle (2000), a
relatively autonomous state elite, long used to
exploiting public resources, has generally found
the reform process easy to manipulate to gain
new avenues for patronage. Patronage activities
do not necessarily infer a disregard for public
welfare; they may sometimes be seen as the
means of coalition building.
Because of the tendency for politically influential groups to succeed over time in altering
1975
the design of distribution systems to their interests, they have often been the most formidable critics of reform and of institutional
change that might weaken their position but
provide broad-based benefits (Bardhan, 1989;
Bates, 1981; Seidman, 1973). The ability to
maintain patronage relationships is facilitated
by continued government involvement in the
allocation of commodities or inputs. This has
often been accomplished while ostensibly adhering to the principle of market reform but
taking the stance that the market is unable to
perform certain social functions, therefore requiring some continued government intervention. In this context, reform programs have
been implemented in such a way as to preserve
patronage activities not unlike those that existed before the reforms were implemented
(Tripp, 1997). These dynamics are clearly
present in the government of KenyaÕs attempts
to raise domestic maize prices above market
price levels through marketing board operations and trade barriers (Nyoro et al., 1999)
and the continued role of the GMB in ZimbabweÕs food marketing system. Many state
officials in the region have become large commercial farmers or traders with a stake in
maintaining state involvement in the distribution of inputs on credit, price supports for farm
commodities, and government contracts for
commodity distribution (Toye, 1992). In EthiopiaÕs case, patronage objectives were perpetuated through an opaque fertilizer tender
process that awarded monopoly distribution
rights to selected firms (Stepanek, 1999). In
ZambiaÕs case, patronage activities have been
preserved through the selection of private
‘‘agents’’ and farmer groups to receive fertilizer
on credit, in an environment where, historically, credit repayment rates were very low and
where there has been little effort to sanction
defaulters (Copestake, 1998). The main difficulty with patronage systems is not necessarily
the misappropriated state funds, but rather the
depressing effect of an un-level playing field on
the development of competitive agricultural
marketing systems. Cases such as these, where
some aspects of reform have been implemented,
but where key features of state allocation of
resources remain intact, contribute to the analytical problems of assessing the effects of the
reforms or even reaching consensus on whether
they have occurred.
The most important critique of the patronage
explanation is that it cannot be overgeneralized. Relatively comprehensive market reform
1976
WORLD DEVELOPMENT
has been achieved in Uganda (maize and fertilizer), Kenya (fertilizer), Mozambique (maize
and fertilizer), Tanzania (cereals), Mali (coarse
grains), Ghana (cereals), and Malawi (fertilizer). But these cases arguably represent commodities that did not figure prominently in
statesÕ patronage or coalition-building objectives in the pre-reform period, either because
volumes handled were small, 6 or because patronage activities centered on other commodities of greater strategic importance. Hence
reform could proceed relatively smoothly for
some commodities while taking considerably
longer to achieve for other commodities (e.g.,
tobacco in Malawi; coffee in Tanzania; cocoa in
Ghana; cotton and rice in Mali) 7 or other
sectors of the economy.
(c) Toward a framework of government reform
implementation
The foregoing discussion of factors influencing government behavior toward reform
implementation can be cast in a framework
represented in Figure 1. Where fiscal crises
were serious, but political risks and patronage
activities were relatively low (Quadrant II), reform programs were often implemented relatively smoothly and transparently (e.g., maize
and fertilizer in Uganda and Mozambique).
Quadrant III represents cases where governments openly resist reform because political
costs are perceived to be high and/or the stateÕs
budgetary situation does not require external
financing and the aid-conditionality strings
associated with it. The Zimbabwean and
Kenyan governmentsÕ open resistance to maize
market reform in the 1980s are examples of
this. With deepening fiscal crises in the 1990s,
however these countriesÕ positions in Figure 1
moved upward into Quadrant IV, forcing them
to officially accept reform programs, but implementing them in ways that maintained
de facto control over important elements of
marketing. Other cases of food and input
marketing systems in Quadrant IV as of 2001
include Ethiopia (fertilizer), Malawi (maize),
and Zambia (fertilizer and maize).
The prominence of the former settler economiesÕ maize sectors in Quadrants III and IV is
noteworthy, highlighting the importance of
historical evolution and path dependence in
understanding current political economy situations. In general, the greater the importance of
European agriculture during the colonial period, the greater the degree of state intervention
in food marketing activities, even to the present
(Jayne & Jones, 1997). While the political might
of settler-dominated farm lobbies has declined
in the post-independence period, pressure for
continued state intervention and selectively
disbursed agricultural subsidies have been
solidified as African elites have increasingly
acquired commercial farm land. Second, in
contrast to West Africa where imported rice
and wheat were the most politically sensitive
food commodities, urban food security in most
of eastern and southern Africa depended more
heavily on white maize. Historically, white
maize has not been readily available on world
markets. Post-independence food policy in
these countries was strongly driven by the priority put on white maize self-sufficiency given
the unreliability of alternative sources. A third
Figure 1. Factors influencing government behavior toward agricultural market reform.
FALSE PROMISE OR FALSE PREMISE?
distinction of the former settler states of eastern
and southern Africa was the obligation that the
new leaders of the post-independence states felt
toward the rural population that had provided
the resistance that led to independence. Food
and input markets became the cornerstone of
an implicit and sometimes explicit ‘‘social
contract’’ that the post-independence governments made with the African majority to redress the neglect of smallholder agriculture
during the former colonial period. New patronage activities emerged around the postindependence maize marketing policies (Herbst,
1990). For these reasons, the commonly accepted ‘‘Berg hypothesis’’ that African governments taxed their agriculture, an observation
that was drawn mainly from West African experience, was not appropriate for many countries in eastern and southern Africa (Jayne &
Jones, 1997).
Because the components of political risk are
difficult to quantify in cardinal terms, our
model is not rigorously tested, and is not purported to incorporate all relevant variables influencing the path of reform. One would need
to invoke case-specific factors, for example, to
explain ZimbabweÕs current behavior. Despite
strong patronage activities in the maize sector
and an increasingly severe fiscal crisis since
1998 (which would predict Quadrant IV-type
behavior according to our model), the government has openly reintroduced price controls on
maize and maize meal, and has officially
banned private maize trade, which is indicative
of Quadrant III. In most cases, however and as
concluded by others, the state budgetary situation and strategic political risks appear to play
important roles in influencing the path of government market reform implementation (Devarajan et al., 2001).
One implication of this model for donors and
other reform advocates is to attempt to move
countriesÕ position in Figure 1 leftward along
the x-axis by reducing the political risks of reform. Options for achieving this are examined
in the following section.
4. CHALLENGES FOR THE FUTURE
The foregoing highlights three key challenges
for donors, governments, and researchers to
work together in developing agricultural marketing systems that perform better and contribute to overall economic development. These
are: how to sort out the empirical record on
1977
agricultural market reform; how to build sustained local political support for constructive
agricultural market reform; and how to design
markets so that they can truly contribute to
small farmer productivity and income growth.
(a) Key challenge 1: Seeking convergence on the
empirical record
Policy debates over the effects of agricultural
reform will be difficult to resolve until more
clarity emerges from the academic literature. As
we have argued earlier, the literature is somewhat tainted by the inclusion of studies that
assess the effects of policies that were not implemented. The literature also arguably suffers
from its own political economy problems. Analyses affiliated with the World Bank (e.g.,
Christiaensen, Demery & Paternostro, 2002;
Deininger & Okidi, 2001; Dercon, 2002; Devarajan et al., 2001; World Bank, 1994, 2000)
typically find benefits resulting from reform, as
do analyses funded by other donors supportive
of reform (e.g., Jayne & Jones, 1997; Sahn,
1996). International and civil society organizations critical of reform, for example UNCTAD
and the Catholic Institute of International Relations, have increasingly funded their own
analyses, typically finding negative effects of
reform (e.g., Chisvo, 2000; UNCTAD, 2001). 8
Can certain analytical approaches be used to
establish greater consensus on the empirical
record in the future? We offer these criteria for
discussion. First, assessing the effects of reform
requires more serious attention to how the
process was implemented. Reform cannot be
cast as a ‘‘yes/no’’ issue; there is a wide range of
possible implementation permutations. This
compounds the difficulties of assessing the effects of reform, but it is a step forward analytically to at least specify clearly the policy and
institutional regime shifts that occurred over
the reform process. Second, specify the counterfactual. What policy regime should reform
have been compared to? It has sometimes been
implicitly assumed that the continuation of the
pre-reform policies was a viable option. Such a
view under-appreciates the extent of the crises
faced by many countries that made reform
unavoidable (Sahn et al., 1997). 9 Because
many African governments were not able to
continue incurring the recurrent costs of their
pre-reform agricultural programs, analysis of
reform must give serious consideration to feasible policy alternatives. Third, what factors
cannot be held constant? Constraints on data
1978
WORLD DEVELOPMENT
availability make it difficult to isolate the specific effects of individual policy changes from
other conditions affecting the economy such as
political turmoil, HIV/AIDS, agricultural policies in other countries, and weather. Explicit
attention to these three criteria––implementation specifics, the policy counterfactual, and
uncontrolled variables––may help achieve a
more solid empirical foundation for subsequent
policy discussion of how agricultural markets
should be organized.
(b) Key challenge 2: How to build political
support for constructive agricultural market
reform
In almost all countries of the world, patronage activities influence policy and institutional
choices for strategically important commodities. As we have argued, reform may be especially difficult for countries located in Quadrant
III and IV in Figure 1. One could argue that if
the specifics of a reform program potentially
disrupt the coalition of interests holding the
government together, then nonimplementation
of reform is inevitable. If enough reform programs fail in this way, can reform be defended
as an appropriate strategy?
Shifting statesÕ positions in Figure 1 farther
back along the x-axis requires strategies to
build domestic constituencies for reform. Our
treatment of this issue revolves around evolution in the way that development aid is
disbursed; greater public and stakeholder involvement and education; and perceptions of
reciprocity in policy reform of high-income
countries.
(i) The future of the aid-conditionality model
Most of the food and input reform programs
negotiated between donors and governments
since 1980 have taken place within the context
of aid conditionality. Under this framework,
governments and lending agencies ideally are to
work together to influence economic policy in a
borrowing country through conditions attached to the lending agencyÕs support. Conditionality has been defined more crassly as a
means to buy reforms. Studies evaluating the
effect of aid conditionality on African countriesÕ policy environment mostly conclude that
it has been largely ineffective (Berg, 1997; Collier, 1997; Jones & Wickrema, 1998; Mosley,
Harrigan & Toye, 1991; van de Walle, 2000).
As stated by Collier, ‘‘despite a decade of
conditionality––and an aid flow estimated at
$64 billion under the Special Program of Assistance for Africa 10––African governments
have largely failed to deliver even a minimally
adequate economic environment’’ (p. 58).
Why has aid conditionality generally not
been effective in leveraging policy reform? First,
governments have come to learn that conditionality has no teeth. In hindsight, many
governments apparently agreed to the terms of
aid-conditionality to gain access to the loan resources, and implicitly viewed the quid pro quo
of agreed-upon policy reforms as something to
be negotiated (van de Walle, 2001). African
political leaders have observed that they can
enter into aid conditionality agreements with
international lenders and donors, then later
refuse to implement the reforms after the funds
were disbursed, and still gain access to secondand third-round loans. After having received
the loans, governments frequently cited a
heightened realization of the need to protect
vulnerable groups who would allegedly be hurt
by the reforms, and the need for a gradualist
approach to reform implementation. In many
cases, this has paved the way for new conditionality agreements, often involving the same
policy changes that were part of the terms of
earlier agreements. 11 Organizational and career incentives to ‘‘move money’’ are frequently
cited as reasons why donors, in particular
the World Bank, allow such situations to recur (Berg, 1997; Nelson, 1995; Tendler, 1975;
Werlin, 1998). Kapur, Lewis, and Webb describe how such incentives evolved from AfricaÕs rising debt to the IMF in the early 1980s.
With commercial lenders and many bilateral
donors unwilling to provide African governments with the funds to repay IMF loans,
the Bank helped to ‘‘keep the system whole’’
(Kapur, Lewis, & Webb, 1997, p. 535). It made
available quick-disbursing loans with less
stringent conditions than otherwise would have
been imposed. In this environment, Bank
managers had strong incentives to conclude
successful adjustment agreements with recipient
governments, despite indications that program
loans might not necessarily generate sufficient
growth and revenues to recoup them. This has
put the international payments system on
something of a ‘‘treadmill,’’ and as African
governments have increasingly learned donorsÕ
objective functions, the intent of conditionalitystructured loans has been compromised.
Perhaps the greatest irony of the aid conditionality process in the eastern and southern
Africa region over the past decade is the
FALSE PROMISE OR FALSE PREMISE?
widespread perception that the World Bank
and IMF have dictated the terms of agricultural
policy reform, that weak governments have
actually implemented them under duress, and
that the lack of clear economic turnaround in
the region casts doubt on the technical logic of
the agricultural market reform model. Collier
(1997); Sahn (1996), and Townsend (1999)
argue that in the African countries in which
policy reform has largely been implemented
and sustained, the growth response has been
encouraging. 12 According to Collier, ‘‘the core
failure of conditionality is not, therefore, that
it has been pressing for the wrong policies,
but that it has failed to achieve them’’ (1997,
p. 58).
Aid conditionality has been shown to be
fundamentally incompatible with genuine partnership and local ownership of policy reform
(Collier, 1997). It is now accepted that a reform
program can be credible only to the extent that
governments are truly committed to it and take
ownership of the process. Evidence of this can
be assessed by the degree to which governments
try to explain the logic of the program to the
public and generate broad-based political support for it. So far, such public campaigns in
support of policy reform have been conspicuously absent in most countries in the region
(van de Walle, 2000).
Collier (1997) argues that donors and international lenders might play a more positive role
in supporting agricultural policy reform by
providing aid based on countriesÕ actual policy
environment. This approach makes commitment to the reforms voluntary, which also
transforms the international lendersÕ role from
a ‘‘carrot and stick’’ enforcer to a selective
supporter that makes no demands but provides
positive support for adoption of a particular
policy stance. There are indications that both
World Bank and US government policy is
moving at least somewhat in this direction. 13
Finally, following Clements (1999, 2001),
there is greater need to carefully articulate the
rationale for reform and let the program generate its own local constituency. Support for
local stakeholder groups, carefully chosen to be
representative of societal interests, including
consumers, would nurture this objective. Trading networks in Ghana, Mali, and Senegal have
recently emerged to publically support the economic reforms in these countries (Goldsmith,
2001). The BankÕs Poverty Reduction Strategy
Programs have also been intended to move in
this direction.
1979
(ii) High-income government behavior in
international agreements
Many of the same patronage problems described above are present in high-income countries as well. The Uruguay Round trade
agreements included a 30% reduction in trade
restrictions by the European Union and United
States over five years. By 2000, less than 5% of
this reduction has been implemented. OECD
countries subsidize their own agriculture by
US$300–400 billion per year (McCalla, 2001;
World Bank, 2000), which is greater than the
GDP of sub-Saharan Africa. One study asserts
that these protectionist policies cost lowincome countries US$700 billion in lost trade
per year (Chisvo, 2000). Yet low-income countries are expected to adhere to a robust cutting
of subsidies and protectionist policies mainly
through structural adjustment programs.
The irony of this situation has not gone unnoticed in Africa. As some international agencies push for policy reform in Africa while
another (the World Trade Organization) fails to
achieve much progress in selected high-income
countries, local analysts and politicians openly
wonder whether this is part of a concerted effort
of the capitalist order to gain further advantage
in international markets (see, for example, the
first citation at the top of the article). These
concerns give greater credence to neocolonialist arguments that have stiffened the publicÕs
opposition to market reform in many African
countries. Moreover, most academic analyses in
support of agricultural reform in Africa have
taken world prices as given and left unexplored
the possible strategic objectives that states and
businesses in high-income countries implement
through their agricultural trade policies. Doing
so might alter the long-term welfare implications of reform in low-income areas.
We have already stressed the growing recognition that local ownership in the policy reform process is critical to its overall success.
Ownership is likely to be influenced by perceptions that the reform ‘‘bar’’ is being set
uniformly throughout the world. The past 20
years have seen dramatic enhancements in AfricansÕ access to world news. Arguments that
wealthy countries can afford to pay the costs of
their inefficient agricultural programs while
poor countries cannot contributes to a sense
that the international system is not fair and
undermines the integrity of international economic fora in general. The degree of public
support in Africa for policy reform in particular, and good governance in general, is likely to
1980
WORLD DEVELOPMENT
be influenced by the extent to which agricultural reform is approached in good faith by all
in the international community. 14
(c) Key challenge 3: How to design sustainable
input, credit, and output markets to catalyze
smallholder productivity and income growth
Market liberalization is not an end in itself.
SchultzÕs ‘‘efficient but poor’’ observation of
low-resource farmers also describes the functioning of firms, markets, and entire economies
in many developing areas (Shaffer et al., 1985).
Markets can be efficient in that margins and
price spreads may approximate the costs of
trading, but these costs may be too high and
unstable to encourage rapid private investment
in the marketing system to promote on-farm
productivity growth. The continued prevalence
of private haggling in spot markets over small
volumes serves to keep transaction costs of exchange high. Use of personalized, kin-based
trading networks are still prevalent; these networks have evolved to minimize risks and
transaction costs of exchange but limit the scope
of trading activity. Poor transport and communications restrict the scope of the market and
increase the risk of commercialized production.
The supply of commercial agricultural credit is
extremely scarce due to the weaknesses of systems for legal recourse and contract enforcement (Dorward, Kydd & Poulton, 1998).
Ubiquitous problems of strategic loan default
restrict the supply of agricultural credit for
smallholders and constrain the demand for
productivity-enhancing inputs, which in turn
depress private sector investment in input
supply and contribute to low productivity agriculture. Bottlenecks at one stage of the supply chain depress the incentives for investment
and growth at other stages (Barrett, 1997).
The existence of market failures does not
mean however that reliance on markets is a
poor policy choice. Ameliorating market failures is an important role of donor assistance
but progress also depends on a committed
government. A case can be made that the development of market institutions has been most
difficult in countries where reform has been
resisted. For example, public grain marketing
information systems have been assimilated into
public sector budgets in Mali and Mozambique, while they are currently totally contingent on donor funding in Zambia and Kenya.
Market reform itself is a long-term, continuous process of institutional innovation. The
tendency to view market reform as a short-run
process of handing state activities over to the
private sector under-emphasizes the collaborative supportive role of government in a marketoriented economy. The evolution of more
productive economies over the past 200 years in
high-income countries has featured the evolution from spot markets to more complex marketing and contracting arrangements that deal
with contingent future outcomes. These complex contracting arrangements have successfully reduced risks and transaction costs of
investment in more technically efficient production processes and have hence proven
valuable because they have encouraged productivity growth at other stages of the system
(Bardhan, 1989; North & Thomas, 1973). But
these innovations cannot occur without commensurate advances in the stateÕs legal system.
A complex contracting arrangement between
farmer groups, an out-grower company and an
international marketing firm may involve high
costs in terms of negotiation, legal services,
monitoring, and related public resources to resolve contract disputes if necessary, but such
institutions may provide the stability of returns
to justify major investments in new technology
that lead to productivity gains at other stages of
the food system. In this regard, market reform
policy should be regarded as a continuous
process of searching for alternative institutional
arrangements, adapted to local conditions, capable of promoting new investment and productivity growth throughout the food system.
The kinds of institutional arrangements that
can evolve are of course circumscribed by the
policy environment.
5. CONCLUDING COMMENTS
Substantial controversy remains over the effects of agricultural market reform in Africa.
We have argued that a primary cause of this
confusion stems from differing perceptions as
to whether the reforms were actually implemented. Our review of food and input market
reform in eastern and southern Africa shows
great variation in implementation. Reform implementation appears to have occurred most
comprehensively for commodities that were not
the most strategically important and in countries where fiscal crises made continuation of
government policies unsustainable. In other
countries, particularly the former settler economies where white maize remains the primary
FALSE PROMISE OR FALSE PREMISE?
staple food, fundamental aspects of market
reform remain unimplemented; others were reversed after a short period. But perhaps the
most common path of reform implementation
has been the adoption of some key reforms that
legalize private trade, mixed with continued or
new government activities that erode its profitability. Governments have typically defended
this approach by arguing that markets are unable to perform certain social functions, and
that direct government programs are still necessary. This perspective, while theoretically
compelling, has in practice left the door open
for large-scale state programs channeling key
resources to selected beneficiaries––most often
not the poor (Badiane & Kherallah, 1999;
Jayne & Rubey, 1993; Sahn, 1996). This approach has generally prevailed over strategies
based on public investments to overcome
market failure problems but which do not enable the state to carry out income transfer objectives through the commodity programs. In
this context, reform programs have been implemented in such a way as to preserve patronage linkages not unlike those that existed
before the reforms were implemented. These
activities have clearly affected the risks and incentives for private sector investment in agricultural markets. Cases such as these, where
some aspects of liberalization have been implemented, but where key features of state allocation of resources remain intact, contribute
to the analytical problems of assessing the effects of the reforms or even reaching consensus
on whether they have occurred.
Survey evidence of private traders and potential investors in Africa during the 1990s has
shown that a major impediment to investment
is fear of policy reversal (Brunetti et al., 1997;
Govereh et al., 2002). Such findings indicate
1981
that, in the minds of the people who were
supposed to be the cornerstone of the liberalized marketing systems, a real change in the
policy environment has occurred in only a minority of countries.
Agricultural market reform in Africa over
the past 20 years has been undertaken in the
context of aid conditionality. Perhaps the
greatest irony of the aid conditionality process
in the region is the widespread perception that
the World Bank has forced African governments to implement orthodox agricultural
policy reform, and that the lack of clear economic turnaround in the region casts doubt on
the technical logic of the BankÕs model. The
weight of the empirical evidence indicates that
in the minority of African countries in which
agricultural policy reform has largely been implemented and sustained, the growth response
has been relatively encouraging. This suggests
that the core failure of conditionality is not,
therefore, that it has been based on misguided
policies, but that it has failed to achieve them.
Agricultural policy debate in Africa is still
often framed in broad terms of market liberalization versus government participation in
the economy. While the experiments with government setting prices or acting as the coordinating organization of the economy have not
been successful in promoting a transformation
out of poverty agriculture, neither has the absence of government participation. The most
salient policy questions pertain to how to support governments in making the institutional
and infrastructure investments that are critical
to stimulate a growth response from the agricultural sector, but which are underprovisioned
both by governments and by private enterprise. These challenges are fundamentally governance issues.
NOTES
1. In the academic literature, the diversity of opinion
on the record of agricultural market reforms in Africa
is evident by comparing Barrett and Carter (1999),
Jayne and Jones (1997), Kherallah, Delgado, GabreMadhin, Minot and Johnson (2002), Kydd and Dorward (2001), Mosley (1994), Sahn, Dorosh and
Younger (1997), Sepp€al€a (1998), Stewart (1991) and
World Bank (1994). For a flavor of the market reform
critique, consider the following: ‘‘liberalization too
often exposed the fundamental structural weaknesses
of the rural economy and induced a return to environ-
mentally dangerous extensification,’’ and ‘‘the state
needs to steer a middle course between relying solely on
liberalization, which has not delivered the goods, and
heavy interventionism’’ (Reardon, Barrett, Kelly, &
Savadogo, 1999, p. 390). ‘‘The failure of agrarian
liberalization to generate expected agricultural growth
in some cases can be attributed to the failure of
policies to generate stimulative market-based price
signals to producers or to farm-level constraints uniformly blocking a positive supply response’’ (Carter,
2000, p. 10).
1982
WORLD DEVELOPMENT
2. See for example, Naim (2000).
8. No inferences are intended about the direction of
causality.
3. Even where market institutions have had centuries
to develop, institutional failures are still widespread. For
recent examples, consider the inability of US regulatory
and enforcement organizations to prevent apparently
serious abuse of business accounting practices by several
major US firms, and frequent insider trading scandals
in stock markets in both the US and Europe.
4. Pletcher (2000). PoliticiansÕ financial interest in
government fertilizer distribution surfaced publicly in a
front page article in the countryÕs main newspaper, the
Zambia Times (‘‘Members of Parliament ÔShrinkÕ Over
FRA Debts Debate,’’ November 11, 2000).
5. Take for example ChilowaÕs (1998) analysis of
Malawi concluding that private maize traders response
to market reform has been weak since the late 1980s in
spite of the view that ‘‘there were no significant barriers
to entry in the marketing of smallholder crops’’ (p.
560) and that by 1996, ‘‘there was full liberalization of
crop and input marketing by removing licensing procedures’’ (p. 561). But, his article indicates that the
maize marketing board, ADMARC, has continued
throughout the liberalization program to set the margin
between its maize selling price and buying price too
low to cover marketing costs, without addressing how
this policy might affect private sector trading incentives.
6. e.g., fertilizer imports were under 10,000 tons in
Uganda and Mozambique up till the late 1990s.
7. See Kherallah et al. (2002) for short case studies on
reform implementation for these commodities.
9. For example, heavy subsidies on consumer maize
meal and fertilizer in Zambia in the late 1980s accounted
for over 15% of the government budget during this
period (Howard & Mungoma, 1997). In Kenya, the
maize marketing board ran annual deficits equal to 5%
of GDP.
10. The SPA, launched in 1987, is an association of
donors committed to coordinate support for economic
reforms in Africa.
11. Collier refers to the example of the government of
Kenya, which sold the same agricultural reform to the
World Bank four times over a 15 year period, each time
reversing it after receipt of aid (1997, p. 60).
12. This conclusion is also supported by analysis of
Uganda (Deininger & Okidi, 2001), Mali (Dembele &
Staatz, 2002); Mozambique (Tschirley & Santos, 1999)
and Ghana (Christiaensen et al., 2002), countries considered to had adopted fairly comprehensive agricultural
reform programs. Compared to the largely nonreforming countries for which data are presented in Table 2,
analogous per capita agricultural production indices
during 1995–2000 for Ghana, Mali, Mozambique, and
Uganda were 134, 104.9, 108.5, and 94.6.
13. See for example Devarajan et al. (2001), and recent
statements by US officials on proposed changes in IMF
lending strategy (Financial Times, 2002).
14. The recent European Union decision to eliminate
import duties on exports from selected low-income
countries is a useful step in this direction.
REFERENCES
Badiane, O., & Kherallah, M. (1999). Market liberalization and the poor. Quarterly Journal of International Agriculture, 38, 341–358.
Bardhan, P. (1989). The new institutional economics and
development. World Development, 17(9), 1389–
1395.
Barrett, C. (1997). Food marketing liberalization and
trader entry: evidence from Madagascar. World
Development, 25(5), 763–777.
Barrett, C., & Carter, M. (1999). Microeconomically
coherent agricultural policy reform in Africa. In J.
Paulson (Ed.), The role of the state in key markets.
London: Macmillan.
Bates, R. (1981). Markets and states in tropical Africa:
the political basis of agricultural policies. Berkeley:
University of California Press.
Bates, R., & Krueger, A. (Eds.) (1993). Political and
economic interactions in economic policy reform:
evidence from eight countries. Oxford: Basil Blackwell.
Berg, E. (1997). Dilemmas in donor aid strategies. In:
Gwin, C., & Nelson, J., (Eds.), Perspectives on aid
and development. Policy Essay No. 22, Overseas
Development Council, Washington, DC.
Brunetti, A., Kisunko, G., & Weder, B. (1997). Institutional obstacles to doing business: region-by-region
results from a worldwide survey of the private sector.
Policy Research Working Paper 1759, World Bank
Washington DC.
Buccola, S., & McCandlish, J. (1999). Rent seeking and
rent dissipation in state enterprises. Review of Agricultural Economics, 21(2), 358–373.
FALSE PROMISE OR FALSE PREMISE?
Carter, M. (2000). Land ownership inequality and the
income distribution consequences of economic growth.
Working Paper 201, World Institute for Development Economics Research, Helsinki.
Chapoto, A. (2000). A time series analysis of maize grain
prices in Zimbabwe. M.Phil Thesis, Department of
Agricultural Economics and Extension. University of
Zimbabwe, Harare.
Christiaensen, L., Demery, L., & Paternostro, S. (2002).
Growth, distribution and poverty in Africa: messages
from the 1990s. Policy Research Working Paper,
World Bank Washington DC.
Chilowa, W. (1998). The impact of agricultural liberalization on food security in Malawi. Food Policy, 23,
553–569.
Chisvo, M. (2000). Reaping the whirlwind: economic
liberalisation and food security in Zimbabwe. Report.
London: Catholic Institute for International Relations.
Clements, P. (1999). Informational standards in development agency management. World Development,
27(8), 1359–1381.
Clements, P. (2001). Western Michigan University, Political Science Department, personal e-mail communications, March 30, 2001.
Collier, P. (1997). The failure of conditionality. In Gwin,
C., & Nelson, J., (Eds.), Perspectives on aid and
development. Policy Essay No. 22, Overseas Development Council Washington, DC.
Copestake, J. (1998). Agricultural credit management
in Zambia: business development, social security,
or patronage? Development Policy Review, 16,
5–28.
Deininger, K., & Okidi, J. (2001). Growth and poverty
reduction in Uganda, 1992–2000: panel data evidence.
Washington DC: World Bank and Kampala, Economic Research Council.
Dembele, N., & Staatz, J. (2002). The effects of market
reform on agricultural transformation in Mali. In T.
S. Jayne, I. Minde, & G. Argwings-Kodhek (Eds.),
Perspectives on agricultural transformation: a view
from Africa. Huntington, NY: Nova Science Publishers.
Dercon, S. (2001). Economic reform, growth, and the
poor: evidence from rural Ethiopia. Working Paper,
Center for Study of African Economies Oxford.
Dercon, S. (2002). The impact of economic reforms on
rural households in Ethiopia. Poverty dynamics Africa
series. Washington DC: The World Bank.
Devarajan, S., Dollar, D., & Holmgren, T. (Eds.) (2001).
Aid and reform in Africa: lessons from ten case
studies. Washington DC: The World Bank.
Dorward, A., Kydd, J., & Poulton, C. (Eds.) (1998).
Smallholder cash crop production under market liberalization. New York: CAB International.
Durevall, D., & Mabugu, R. (2000). Maize markets in
Zimbabwe. Country Economic Report 2000:10,
Swedish International Development Agency Stockholm.
Financial Times (2002). OÕNeil urges reform in IMF
lending role. March 1, 6.
Gabre-Madhin, E. (2001). Market institutions, transaction costs, and social capital in the Ethiopian grain
1983
market. Research Report 124. Washington, DC:
International Food Policy Research Institute.
Goldsmith, A. (2001). Institutions and economic growth
in africa. In M. McPherson (Ed.), Restarting and
sustaining growth and development in Africa. Cambridge, MA and Washington DC: Harvard Institute
for International Development, and USAID.
Govereh, J., Jayne, T., Nijhoff, J., Haantuba, H.,
Ngulube, E., Belemu, A., Shawa, J., Banda, A.,
Donovan, C., & Zulu, B. (2002). Fertilizer Market
Reform Strategies in Zambia. Working Paper 4.
Zambia Food Security Research Project Lusaka.
Herbst, J. (1990). State politics in Zimbabwe. Berkeley:
University of California Press.
Howard, J., & Mungoma, C. (1997). ZambiaÕs stop-andgo maize revolution. In D. Byerlee & C. Eicher
(Eds.), AfricaÕs emerging maize revolution. Boulder,
CO: Lynn Reinner Publishers.
IRIN News Service (Integrated Regional Information
Networks) (2001). Southern Africa report. Email
report. United Nations for the Coordination of
Humanitarian Affairs.
Jayarajah, C., & Branson, W. (1995). Structural and
sectoral adjustment: the World Bank experience,
1980–1992. A World Bank operations evaluation
study. Washington: The World Bank.
Jayne, T., & Rubey, L. (1993). Maize milling, market
reform and urban food security: the case of Zimbabwe. World Development, 21(6), 975–988.
Jayne, T., & Jones, S. (1997). Food marketing and
pricing policy in Eastern and Southern Africa: a
Survey. World Development, 25(9), 1505–1527.
Jayne, T., Mukumbu, M., Chisvo, M., Tschirley, D.,
Zulu, B., Weber, M., Johansson, R., Santos, P., &
Soroko, D. (1999). Successes and challenges of food
market reform: experiences from Kenya, Mozambique, Zambia, and Zimbabwe. International Development Working Paper 72, Michigan State
University, East Lansing.
Jenkins, C. (1997). The politics of economic policymaking in Zimbabwe. Journal of Modern African
Economies, 575–602.
Jones, S., & Wickrema, S. (1998). The use of conditionality in reform: food markets in Africa. Policy
Briefing Note No. 1. Oxford Policy Management
Oxford.
Kanyinga, K. (1994). Ethnicity, patronage and class in a
local arena: ‘‘high’’ and ‘‘low’’ Politics in Kiambu,
1982–92. In K. Kanyinga, A. Kiondo, P. Tidemand,
& P. Gibbon (Eds.), The new local level politics in
East Africa, Research Report 95. Nordiska Afrikainstitutet.
Kapur, D., Lewis, J., & Webb, R. (1997). The World
Bank: its first half century. Washington DC: Brookings Institute.
Kherallah, M., Delgado, C., Gabre-Madhin, E., Minot,
N., & Johnson, M. (2002). Reforming agricultural
markets in Africa. Baltimore: Johns Hopkins University Press.
Kydd, J., & Dorward, A. (2001). The Washington
Consensus on poor country agriculture: analysis,
prescription and institutional gaps. Development
Policy Review, 19(4), 467–478.
1984
WORLD DEVELOPMENT
MASIP (Malawi Agricultural Sector Investment Programme) (2000). An action plan for developing
sustainable Agricultural Input Supply Systems in
Malawi, Vols. 1 and 2. Llilongwe: Ministry of
Agriculture and Irrigation.
McCalla, A. (2001). Agriculture in the 21st century: how
were past challenges met? what are the future challenges? Fourth Distinguished Economist Lecture.
Mexico City: CIMMYT.
Minten, B., & Kyle, S. (1999). The impact of distance
and road quality on food collection, marketing margins, and tradersÕ wages: Evidence from the former
Zaire. Journal of Development Economics, 60(2), 467–
495.
Mosley, P. (1983). The settler economies: studies in
the economic history of Kenya and Southern Rhodesia
1900–63. Cambridge: Cambridge University Press.
Mosley, P. (1994). Policy and capital market constraints
to the African green revolution: a study of maize and
sorghum yields in Kenya, Malawi, and Zimbabwe,
1960–91. In G. Andrea Cornia & G. Helleiner (Eds.),
From adjustment to development in Africa: conflict,
controversy, convergence, and consensus? New York:
Macmillan.
Mosley, P., Harrigan, J., & Toye, J. (1991). Aid and
power: The World Bank and policy based lending in
the 1980s. London: Routledge.
Mwanaumo, A. (1999). Agricultural marketing policy
reforms in Zambia. Paper presented at the Fourth
Agricultural Transformation Workshop, Tegemeo
Institute/Egerton University and Eastern and Central Africa Policy for Agricultural Policy Analysis,
Nairobi June 27–30.
Naim, M. (2000). Washington consensus or Washington
confusion? Foreign Policy, 118, 87–103.
Nashashibi, K., Gupta, S., Liuksila, C., Lorie, H., &
Mahler, W. (1992). The fiscal dimensions of adjustment
in low-income countries, IMF Occasional Paper no. 95.
Washington: International Monetary Fund.
Nelson, P. J. (1995). The World Bank and non-governmental organizations: the limits of apolitical development. New York: St. MartinÕs Press.
North, D. (1987). Institutions, transaction costs, and
economic growth. Economic Inquiry, 25(5), 419–428.
North, D., & Thomas, R. (1973). The rise of the Western
world. Cambridge: Cambridge University Press.
Nyoro, J., Kiiru, M., & Jayne, T. (1999). Evolution of
Kenya’s maize marketing systems in the post-liberalization era. Paper presented at the Fourth Agricultural Transformation Workshop, Tegemeo Institute/
Egerton University and Eastern and Central Africa
Policy for Agricultural Policy Analysis, June 27–30,
Nairobi.
Pletcher, J. (2000). The politics of liberalizing ZambiaÕs
maize markets. World Development, 28(1), 129–142.
Reardon, T., Barrett, C., Kelly, V., & Savadogo, K.
(1999). Policy reforms and sustainable agricultural
intensification in Africa. Development Policy Review,
17(4), 375–395.
Sahn, D. (1996). Economic reform and the poor in Africa.
Oxford: Clarendon Press.
Sahn, D., Dorosh, P., & Younger, S. (1997). Structural
adjustment reconsidered: economic policy and pov-
erty in Africa. Cambridge: Cambridge University
Press.
Schmid, A. (1992). Legal foundations of the market:
implications for the formerly socialist countries of
Eastern Europe and Africa. Journal of Economic
Issues, 26(3), 707–732.
Seidman, R. (1973). Contract law, the free market, and
state intervention: a jurisprudential perspective.
Journal of Economic Issues, 7(4), 553–573.
Sepp€al€a, P. (1998). Liberalized and neglected? food
marketing policies in Eastern Africa. World Development Studies No. 12. Helsinki: United Nations
University/World Institute for Development Economics Research.
Shaffer, J. (1969). On institutional obsolescence and
innovation background for professional dialogue on
public policy. American Journal of Agricultural Economics, 51(2), 245–268.
Shaffer, J., Weber, M., Riley, H., & Staatz, J. (1985).
Designing marketing systems to promote Development
in Third World Countries. Proceedings of workshop
on Agricultural Markets in the Semi-Arid Tropics
held at the ICRISAT Center, Pantacheru, India.
October. Pantacheru: International Research Institute for the Semi-Arid Tropics (ICRISAT).
Stepanek, J. (1999). Restructuring fertilizer markets:
lessons from Ethiopia. Ph.D. dissertation, Michigan
State University.
Stewart, F. (1991). Are adjustment policies in Africa
consistent with long-run development needs? Development Policy Review, 9, 413–436.
Takavarasha, T. (1994). Agricultural price policy. In M.
Rukuni & C. Eicher (Eds.), ZimbabweÕs agricultural
revolution. Harare: University of Zimbabwe Press.
Tendler, M. (1975). Inside foreign aid. Baltimore: Johns
Hopkins University Press.
Thomas, J., & Grindle, M. (1990). After the decision:
implementing policy reforms in developing countries.
World Development, 18(8), 1163–1181.
Townsend, R.F. (1999). Agricultural incentives in subSaharan Africa: policy challenges. World Bank Technical Paper No. 444. World Bank Washington, DC.
Toye, J. (1992). Interest group politics and the implementation of adjustment in sub-Saharan Africa.
Journal of International Development, 4(2), 183–198.
Tripp, A. (1997). Changing the rules: the politics of
liberalization and the urban informal economy in
Tanzania. Berkeley: University of California Press.
Tschirley, D., & Santos, A.P. (1999). The effects of maize
trade with Malawi on price levels in Mozambique:
implications for trade and development policy. Food
Security Project Working Paper 34, Ministry of
Agriculture and Rural Development Maputo.
UNCTAD (United Nations Conference on Trade and
development) (2001). Economic Development in
Africa: performance, prospects, and policy issues,
New York: United Nations.
van de Walle, N. (2000). Economic Reform in Africa,
1980–2000: patterns and constraints. Prepared for
The Workshop on Democracy and Development in
Africa. The Gulbenkian Foundation, Lisbon.
van de Walle, N. (2001). African economies and the
politics. Cambridge: Cambridge University Press.
FALSE PROMISE OR FALSE PREMISE?
Wanzala, M., Jayne, T., Staatz, J., Mugera, A., & Kirimi,
J. (2002). Fertilizer markets and Agricultural production incentives: insights from Kenya. Working Paper,
Tegemeo Institute, Egerton University, Nairobi.
Werlin, H. (1998). The mysteries of development: studies
using political elasticity theory. Lanham, MD: University Press of America.
1985
World Bank (1981). Accelerating African agricultural
development. Washington, DC: World Bank.
World Bank (1994). Adjustment in Africa: reforms,
results and the road ahead. New York: Oxford
University Press.
World Bank (2000). Can Africa claim the twenty-first
century? Washington, DC.: World Bank.