Upload
world-bank-publications
View
222
Download
0
Embed Size (px)
Citation preview
7/31/2019 Africa Can Help Feed Africa
1/106
Africa Can Help Feed Africa
Removing barriers to regional trade in food staples
7/31/2019 Africa Can Help Feed Africa
2/106
7/31/2019 Africa Can Help Feed Africa
3/106
Arica Can Help Feed Arica
Removing barriers to regional trade in ood staples
October 2012
Poverty Reduction and Economic Management
Arica Region
7/31/2019 Africa Can Help Feed Africa
4/106
7/31/2019 Africa Can Help Feed Africa
5/106
Foreword
Prices or basic staple ood are back in the headlinesthey are rising again. Tis is bad news
or those among Aricas poor who consume more ood than what they can produce. Tink
especially o poor amilies living in cities: they spend the majority o their income on sim-
ple oodstus. Rising ood prices are also having important macroeconomic impacts on many
Arican countries since more and more ood is being imported rom the global market leading to
worsening balances o trade. Just 5% o Aricas imports o cereals come rom other Arican coun-
tries. Tis issue is not going to go away. Demand or ood will continue to increase, it is project-
ed to double by 2020, and consumers will be increasingly located in Aricas rapidly growing cities.
Fortunately, Arica does have the means and opportunities to deal with and deliver improved
ood security or its citizens. I Arican armers were to achieve the yields that armers are attain-
ing in other developing countries then output o staples would easily double or even triple. On top
o this barely a raction o ertile agricultural land is being cultivatedjust 10% o the 400 million
hectares o agricultural land in the Guinea Savannah zone that covers a large part o Arica. Cul-
tivating this land, while ensuring that existing user rights and the environment are protected, can
play a key role in satisying the rising demand or ood in Aricas cities and ultimately elsewhere
in the world.
But or this to happen current ood trade policies need to change and armers need to be better
linked to both inputs and to consumers. Oen the nearest sources o demand are across a border.
Fragmented regional ood markets and the lack o clear and predictable policies have dissuaded
the private sector rom making the investments that could allow Arica to achieve its potential in
oodrom investments by the poorest armers in raising their productivity to large-scale invest-
ment by nancial institutions to allow role out o insurance-based products to protect armers rom
adverse weather outcomes.
Even though ood staples are very basic products there is still a value-chain that links the armer
to the consumer and at each link in the chain trade barriers and constraints on competition increase
the gap between producer and consumer pricesreducing returns to armers and increasing the
cost o consumers ood baskets. Farmers in Arica ace more barriers in accessing the inputs they
need and in getting their ood to consumers in Arican cities, than do suppliers rom the rest o
the world. Outdated regulations limit access to the best seeds, raise the price o ertilizers and pre-
vent agricultural specialists rom crossing borders to share their knowledge where it is most need-
ed; lack o competition oen ensures high transport costs and poor services; export bans, unnec-essary permits and licenses, costly documentary requirements hit poor small armers and traders
hard, and standards, rather than acilitating are oen a barrier to trade or small producers.
Te concept o ood security has too oen been twisted by narrow nationalistic visions and
requently hijacked by specic interests that benet rom the current situation. It is now an oppor-
tune time to breakout and pursue regionally integrated ood markets that will deliver ood security
more eciently and at the same time allow armers to obtain more o the returns rom their labors.
ii
7/31/2019 Africa Can Help Feed Africa
6/106
What is required? First o all governments in Arica need to engage in a more open and inclusive
dialogue with their stakeholders on policies aecting ood trade and ood security and in particular
to engage with neighbors through the regional communities on pursuing a collective approach to
ood security. oo oen decisions are made without critical analysis and consideration o options
and the interests and views o the broad group o stakeholders in ood staples trade policies are sel-dom represented. Te international community has to be ready to support this dialogue with bet-
ter and more up to date inormation on trends in ood production and regional ood stocks and
analysis o the impact that proposed policy reorms will have on regional ood trade.
Second, there is a need to review and remove barriers to trade along the ood value chain.
Dicult decisions will have to be made and implementedwhile many armers and consumers
will benet a small number o inuential politicians and their riends with interests in the ood
and related sectors will strongly resist. Again the international community can assist with stud-
ies that show the magnitude o the benets rom reorm and in designing packages to oset loss-
es that some may incur.
Obviously, political realities mean that moving to open regional ood markets cannot all bedone at onceso the challenge is to design a processa series o stepsby which the govern-
ment provides clear signals regarding the policy environment or ood trade which gives the pri-
vate sector sucient condence to make additional investments which in turn allows the govern-
ment to take urther steps to reduce intererence in trade and agricultural production and so on.
One example, too oen armers arrive at the border to nd that the government has imposed an
unannounced export ban, or sometimes a ban has been removed but customs ocials have not
been inormed. Ideally governments should commit never to ban exportsexport bans have not
been successul and are typically counter-productivebut at the very least the government should
indicate and stick to clearly dened circumstances under which they will intervene such as when
local ood stocks all below a stated level such as 3 months supply.Food trade has yet to be unlocked within Arica, together with its potential to raise income or
armers and enhance ood security or all. For that, reorms need to happen. And to put reorms
in motion, nothing would help more than evidence brought to the attention o the common citi-
zen. Tis is the ultimate purpose o this report, and o the Arica rade Practice at the World Bank.
Makthar Diop
World Bank Vice President or Arica
Fall 2012
iv
Africa Can Help Feed Africa
7/31/2019 Africa Can Help Feed Africa
7/106
v
Abbreviations and Acronyms
ADMARC Agricultural Development and Marketing Corporation (Malawi)
ACESA Alliance or Commodity rade in Eastern and Southern Arica
AERR ASEAN Emergency Rice Reserve
AFSIS ASEAN Food Security Inormation System
AIFS ASEAN Integrated Food Security
AMIS Agricultural Market Inormation System
ASCE Abuja Securities and Commodity Exchange
ASEAN Association o Southeast Asian Nations
ASFSR ASEAN Food Security Reserve
ASYCUDA Automated System or Customs Data
AU Arican Union
CEN-SAD Communaut des Etats Sahlo-Sahariens (Economic Community o Sahelo-
Sahelian Countries)
CGP COMESA Green Pass
COMESA Common Market or Eastern and Southern Arica
COMRAP COMESA Regional Agro-Inputs Program
DRC Democratic Republic o Congo
EAC East Arican Community
ECCAS Economic Community o Central Arican States
EAERR East Asian Emergency Rice Reserve
ECOWAS Economic Community o West Arican States
ECX Ethiopian Commodity Exchange
FAO Food and Agricultural Organization
FCFA Franc Communaut Financire Aricaine
FRA Food Reserve Agency (Zambia)
GDP Gross Domestic Product
HACCP Hazardanalysis and critical control points
IFPRI International Food Policy Research Institute
IGAD Inter-Governmental Authority on Development
IRG Improved Road ransport Government (USAID)
KACE Kenyan Agricultural Commodity ExchangeMAS Multi-Agency Support eam
NCPB National Cereals and Produce Board (Kenya)
NB Non-tari Barrier
OECD Organization or Economic Cooperation and Development
PHS Plant Health Services (anzania)
RA Regional rade Agreement
7/31/2019 Africa Can Help Feed Africa
8/106
vi
Africa Can Help Feed Africa
SADC Southern Arican Development Community
SAFEX South Arican Futures Exchange
SED Single Entry Document
SOE State Owned Enterprise
SPS Sanitary and Phytosanitary StandardsSQM Standardization, Quality Assurance, Metrology and esting
SSA Sub-Saharan Arica
SR Simplied rade Regime
AEC anzania Atomic Energy Commission
PRI anzania Pesticide Research Institute
FDA anzania Food and Drugs Agency
BS anzania Bureau o Standards
UCE Uganda Commodity Exchange
UNCAD United Nations Conerence on rade and Development
USAID United States Agency or International DevelopmentVA Value-added ax
WAEMU West Arican Economic and Monetary Union
WFP World Food Program
ZABS Zambian Bureau o Standards
ZAMACE Zambia Agricultural Commodity
Acknowledgements
Tis report has been complied by a team led by Paul Brenton (AFPM) and comprising Nora Dihel
(AFP2), Richard Gicho (consultant), Ian Gillson (PRMR), Matthew Harber (consultant), Gozde
Isik (consultant), John Keyser (consultant), Ron Kopicki (consultant), Barbara Rippel (AFPM), and
Andrew Roberts (AFCRI). We are grateul or the comments and suggestions o the peer reviewers:
Bernard Hoekman (PRMR), Will Martin (DECAR) and Stephen Mink (AFSN), and comments
rom Mombert Hoppe (PRMVP). Tis work has been nanced in part by the Multi Donor rust
Fund or rade and Development supported by the governments o Finland, Norway, Sweden, and
the United Kingdom
7/31/2019 Africa Can Help Feed Africa
9/106
vi
Contents
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Introduction and context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Te ood staple value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Regional production and trade patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Agriculture and ood trade policy reorm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Aricas enormous potential or regional trade in staples is not being exploited . . . . . . . . . . . . . . 11
Regulatory barriers to trade undermine Aricas potential in ood trade. . . . . . . . . . . . . . . . . . . . . 17
rade barriers limit access to key inputs: seeds, ertilizers, and extension services . . . . . . . . 19
High transport costs, especially or smallholders, limit access to regional markets . . . . . . . . 27
Opaque and unpredictable trade policies undermine regional ood trade . . . . . . . . . . . . . . . 32
Crossing borders to deliver ood staples to neighbors is costly and dangerous . . . . . . . . . . . 46
Inecient distribution services hamper regional trade in staples and contribute
to ood insecurity in Sub-Saharan Arica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Support institutions that oster ecient and stable regional ood markets. . . . . . . . . . . . . . . . . . . 53
Standards and the capacity to enorce them . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Commodity exchanges and the development o ormal trade. . . . . . . . . . . . . . . . . . . . . . . . . . 57
Market inormation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Institutions that acilitate risk management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Futures and options markets or ood staples 63
Warehouse receipt systems 64
Weather-indexed insurance 66
Political economy issues that constrain open regional trade must be addressed. . . . . . . . . . . . . . 67
Recommendations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Indicative action matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Action Area 1: Improve policies along the value chain to reduce the gap between
producer and consumer prices 75
Action Area 2: Provide a more transparent and predictable ood trade policy regime 77
Reerences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Annex 1: Basic rights and obligations or traders and border ocials . . . . . . . . . . . . . . . . . . . . . . 83
List o BoxesBox 1: Regional trade in rice in West Arica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Box 2: Government imports o maize during the 2001/02 Southern Arica ood crisis . . . . . . . . . 8
Box 3: Maize policy in Kenya. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Box 4: rade Ccsts and rural incomes in Malawi: Improving access to ertilizers . . . . . . . . . . . . . 24
Box 5: Road blocks more than a nuisance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
7/31/2019 Africa Can Help Feed Africa
10/106
viii
Africa Can Help Feed Africa
Box 6: Competing transport services key to lower ood prices or consumers . . . . . . . . . . . . . . 30
Box 7: Mozambique and Uganda open border policy or trade in ood staples . . . . . . . . . . . . . . . 35
Box 8: Te COMESA Green Pass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Box 9: Regional quality standards or ood staples in Arica: Harmonization not always . . . . . . . .
appropriate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40Box 10: Quantiying the eects o non-tari barriers on Arican ood staples . . . . . . . . . . . . . . . 44
Box 11: Te COMESA Simplied rade Regime (SR) or Zambia, Zimbabwe and Malawi . . . 49
Box 12: Te problem o post-harvest loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Box 13: Improving inormation ows: Mobile phones and agricultural markets in Niger. . . . . . 60
List o ables
able 1: Farm gate build-up o 2007/08 ertilizer prices in Malawi (US$/ton). . . . . . . . . . . . . . . . 24
able 2: Value chain indicators or Malawi hybrid maize
(US$/ton tradable grain, 2007/08 prices). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
able 3: otal available prots per hectare rom Malawi hybrid maize, 2007/08 (US$/ha) . . . . . 26able 4: rade policy impacts on maize prices in Zambia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
able 5: Comparison o current and nal dra East Arican standards or maize
with Codex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
able 6: Comparison o dra EAC standards and ZAMACE standards. . . . . . . . . . . . . . . . . . . . . 42
able 7: Weaknesses o SPS standards management in anzania. . . . . . . . . . . . . . . . . . . . . . . . . . . 56
able 8: Major global utures and options exchanges or ood staples. . . . . . . . . . . . . . . . . . . . . . . 63
List o Figures
Figure 1: Key elements in a typical staples value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Figure 2: Te maize ood shed in Southern and Eastern Arica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Figure 3: Te volume o net exports o ood staples by region 19762008 (1,000 tons). . . . . . . . 13
Figure 4: Aggregate ood staple trade by world regions, 1990 and 2010 (millions, US$) . . . . . . . 14
Figure 5: Regional markets can drive trade in staples: Tai exports o cassava 2000 to 2010. . . . 15
Figure 6: Nitrogen-based ertilizer prices in 2010 (US$ per mt o nutrients). . . . . . . . . . . . . . . . . 21
Figure 7: Price raising eect o NMs on rice in Kenya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Figure 8: Bringing staples across borders can be a risky business . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Figure 9: Impact o mobile coverage on the dispersion o agricultural prices in Niger . . . . . . . . 60
7/31/2019 Africa Can Help Feed Africa
11/106
Executive Summary
ix
PhotographbyUN/LucienRajaonin
7/31/2019 Africa Can Help Feed Africa
12/106
x
Africa Can Help Feed Africa
Aricas growing demand or ood has been met increasingly by imports rom the global
market. Tis, coupled with rising global ood prices, brings ever-mounting ood import
bills. In addition, population growth and changing demand patterns will double demands
over the next 10 years. Clearly, business as usual with ood staples is not sustainable.
But there is a solutionand it comes rom within Arica. Trough regional trade, Aricas arm-ers have the potential to meet much o the rising demand, while providing substitutes or more
expensive imports rom the global market.
Tis potential, however, has yet to be exploited because Arican armers ace more trade barri-
ers in accessing the inputs they need, and more trade constraints in getting their ood to consum-
ers in Arican cities, than do suppliers rom the rest o the world. Tus, Arican armers now pro-
vide only ve percento Aricas imports o cereals.
Tis report oers our main messages:
1. Regional trade in staples, which oers the potential to advance ood security and growth,
is not being exploited.Open regional trade is vital, especially as demand or staples becomes more concentrated
in cities, which must rely on ood production rom throughout the continent. And dier-
ent seasons and rainall patterns and variability in production, which will increase as cli-
mate change continues, are not conned to national borders. Tus, an Arica ood secu-
rity model based on nationalsel-suciency cannot work.
Removing barriers to regional trade presents benets to armers, consumers, and gov-
ernments. Farmers gain the incentives to meet the rising regional demand or ood; and
new jobs would be created along the activities o the staples value chainor example,
in producing and distributing seeds and ertilizers, in advisory services, in consolidating
and storing grains, in transport and logistics, in distribution and retailing, and in process-ing. Meanwhile, a regional approach to ood security allows Arican governments to bet-
ter ensure access to ood or their populations.
Te regional value chains that deliver ood rom surplus production areas in one coun-
try to consumers in ood decit areas (such as cities) in another country comprise a num-
ber o stages. Tese include inputs o seeds, ertilizers, labor and knowledge, marketing
and distribution, the requirements related to customs procedures, taxes, and regulatory
requirements in the export market, and then the wholesale and retail stages.
At each o these elements, the extent o competition and the nature o government policies
can inuence the price the consumer pays (and its variability) as well as the price that the arm-
er receives or the product. Tis study urnishes evidence that constraints along the regional value
chain inhibit Aricas cross-border trade in ood staples. Tis leads to the second message:
7/31/2019 Africa Can Help Feed Africa
13/106
x
Executive Summary
2. Regulatory barriers to trade and competition along the whole value chain must be removed
or Arica to reach its potential in regional ood trade.
Te report looks at barriers to trade and competition aecting ve elements o the region-
al value chain:
a. Inputs o seeds and ertilizers and extension services
rade barriers deny Arican armers access to higher yielding seeds and better ertil-
izers available elsewhere in the world. For example, in some Arican countries it can
take two to three years or new seed varieties to be released, even i they are used else-
where on the continent. As new varieties are introduced at a aster rate, Arica alls
arther and arther behind in the use o modern seeds making it dicult to compete
with imports rom the global market. In Ethiopia, or example, use o improved hybrid
maize could help quadruple productivity; and even i just hal the armers achieved
the productivity associated with hybrid seeds, the domestic production would replace
commercial imports (Alemu 2010). But without clarity about regulations, seeds canbe held up at borders or long periods, oen rendering them useless.
Fertilizers oer a similar example, with application rates in Arica substantial-
ly lower than in other parts o the world. Farmers in Arica, especially those in land-
locked countries, ace higher prices or ertilizers than armers in other developing
countries. And markets in many Arican countries are too small to exploit scale econ-
omies linked with ertilizer production and even blending. Part o the reason that
regional ertilizer markets have not emerged is that individual countries more oen
speciy their own ertilizer blend specications and specialty products. Hence, ertil-
izers cannot move reely rom country to country.
As a result, many countries import directly rom the global market but, being smallbuyers, cannot gain the same price as larger buyers. In addition, shipping companies
usually charge more or smaller deliveries. Tereore, regional markets with common
ertilizer specications could generate substantially lower prices when quantities are
ordered or the region. Te lack o an eective system o standards is a major barrier
to cross-border trade and regional ertilizer markets.
wo key issues must be addressed: (a) establishing a consistent and stable policy
environment or regional trade in ertilizers; and (b) investing in institutions that reduce
the transaction costs o coordination ailures. Many countries have enacted new er-
tilizer laws in recent years, but ew have provided the resources to dene and enorce
regulations through standards and testing capacity. Tis report shows that reducing
regulatory burdens on ertilizers and the consequent increase in use o ertilizers would
have substantial impacts on returns to armers, with consequent impacts on poverty.
7/31/2019 Africa Can Help Feed Africa
14/106
xii
Africa Can Help Feed Africa
b. High transport costs in Arica, especially or small armers
High transport costs and the lack o investment in modern trucking and shipping
capacity remain a key actor limiting the movement o surplus staples to areas o
strong demand. Although transport inrastructure needs improvementespecially on
cross-border routes and in linking smallholders to regional networksroads are notthe major constraint. Te critical issue is regulatory reorm that delivers more mod-
ern and competitive transport services.
ransport cartels are still common across Arica, and the incentives to invest in
modern trucks and logistics services are weak. Roadblocks, as well as being a nui-
sance, add considerably to the costs and time to transport, undermining the eciency
o transport operations. Estimates suggest that reorm that delivers more competition
could reduce the cost o transporting staples in West Arica by 50 percent within 10
years. A dierent study nds that a 50 percent reduction in transport costs in Mozam-
bique would increase real agricultural GDP by seven percent and also increase agri-
cultural GDP in Malawi by three percent.c. Opaque and unpredictable trade policies raise trade costs and curb private sector incen-
tives to invest in raising productivity
Among the actors aecting Aricas trade in staples are export and import bans, variable
import taris and quotas, restrictive rules o origin, and price controls. Tese are oen
determined without transparency, and are poorly communicated to traders and ocials
at the border. Tis then creates uncertainty about market conditions, limits cross-bor-
der trade, and raises ood price volatility. Te way that standards are dened and imple-
mented has a critical impact on the propensity to trade. For example, proposed stan-
dards in the EAC on discolored maize could exclude all smallholder-produced maize.
Tis calls or a more open discussion o trade policies beore implementation, witha careul assessment o the costs and benets as well as more inclusive processes or
dening standards to ensure consistency with consumer needs and producer capacities.
d. Crossing borders is costly and dangerous
Hundreds o thousands o Aricans cross borders daily to deliver ood staples rom
surplus areas to higher priced markets. Most o these traders are women, and their
activities provide an essential source o income to many households. But most bor-
der ocials are men, and studies show that cross-border traders regularly suer some
sort o harassment. For example, in the Great Lakes region, poor women cross-border
traders must routinely pay bribes, and a large number o them relate acts o violence,
threats, and sexual harassment, most o which go unreported. Te lack o econom-
ic and physical security undermines the livelihoods o these traders and compounds
their lack o access to nance, inormation, and business knowledge. Reducing the
number o agencies and ocials at the border and increasing the transparency and
predictability o the policy regime is crucial to providing an environment in which
traders ourish and expand their business.
7/31/2019 Africa Can Help Feed Africa
15/106
xii
Executive Summary
e. Inefcient distribution services hamper regional trade in ood
Poor people in the slums o Nairobi pay more pro rata or ood staples than wealthy
people pay at supermarkets. Tis shows the importance o the distribution sector, and
that in many countries the sector is not linking poor armers and poor consumers.
Measures that support both armers and small sellers to become more organizedcan help inormal operators participate in evolving distribution sectors. Reorms are
also needed to provide the regulatory rameworks or modern distribution services.
Te lack o licensing and operation rules or distribution companies, inadequate codes
on investment, commerce, labor, and taxationas well as the lack o bankruptcy pro-
cedurescreate uncertainty and burden rms trying to conduct business operations
in the ormal distribution sectors.
Improvements to regulatory rameworks should eliminate disproportionate entry
requirements, such as lengthy registration procedures, multiple licenses, or inadequate
zoning regulations. In addition, price controls across the region and cartels in sever-
al Arican countries impede competition.
Removing constraints to trade in ood staples along the value chain will reduce transaction
costs but will not ensure good market outcomes. Market ailures, such as those rom asymmet-
ric inormation (or example, a armer knows less about the contents o a bag o ertilizer than the
trader selling it), call or institutions that support market outcomes, including those that dene
and implement standards regimes, commodity market exchanges, and those charged with provid-
ing market inormation. Under previous regimes o heavy state intervention these institutions were
less important. Tis leads to the third message:
3. Build and reorm institutions and invest in their capacity to make staples markets ef-cient and stable
Market-based agricultural production and distribution, especially when based on inte-
grated regional markets, require institutions that aid exchange. Tese market-supporting
institutions can be diverse in character and organizational structure; but their goal should
be to support the inormational and distributional unctions o markets.
But development o these institutions is compromised when the trade policy envi-
ronment or staples is uncertain. Eective standards regimes depend upon private sector
involvement, yet in many countries government agencies dominate the process o den-
ing standards. Commodity exchanges have a great potential to reduce transaction costs
or armers by reducing the number o intermediaries and improving the conditions o
exchange; however, they have not ared well in Arica. One reason is that these institu-
tions cannot thrive without even-handed policies. In addition, operating over larger ter-
ritories allows exchanges to build enough trading volume to exploit scale economies and
be protable, requiring exchanges to be able to operate across borders and requiring pre-
dictable regional trade policies. Investment in acquiring market inormationessential
7/31/2019 Africa Can Help Feed Africa
16/106
xiv
Africa Can Help Feed Africa
or allowing new opportunities or cross-border trade to be exploitedis less likely when
there are risks associated with the policy environment.
Finally, many institutions can help address ood security concerns and thus reduce the
political risk o reorm, allowing countries to pursue the integration o regional markets in
staples and their inputs. For example, utures and options markets or ood staples oer analternative to holding physical stocks o ood staples through ood security reserves and
to trade interventions that limit imports during periods o surplus and exports when pro-
duction is low. Tese contracts guarantee the supply o ood commodities.
Warehouse receipt systems may also negate the need to hold or maintain physical
stocks o ood staples. Tese allow armers to deposit a certain quantity o a commodity
into a private warehouse, where it can be pooled with other commodities o the same qual-
ity. A receipt is issued to the owner as evidence o location and ownership. Te receipt is
a negotiable instrument that can be sold or used as collateral or a loan.
Weather-indexed insurance can lessen the impacts o climatic shocks on armers.
Weather-indexed insurance is a type o nancial derivative written against deviations inaverage rainall or temperature indices constructed rom data measured at weather sta-
tions. For example, i rainall is below a set threshold, leading to low yields, an insured
armer would receive a paid compensation or reduced ood staple production.
Although these institutions can play a key role in supporting greater regional trade, they will
only ourish i there is a change in the way that ood trade policies are dened and implemented.
Tis in turn leads to the ourth message that:
4. Political economy issues that constrain open regional trade must be addressed.
Despite commitments to opening up regional trade in ood, implementation has general-ly been weak; governments continue to restrict trade, maintaining the barriers and con-
straints outlined earlier. A program o regional trade reorm can only be credible i gov-
ernments commit to it, and take ownership o the process.
An indication o this vital commitment and ownership is the extent to which governments
work to build a domestic constituency or reormexplaining the need or, and impacts o pol-
icy change, and generating a political consensus or integrated regional agricultural markets.
Most Arican countries have ailed in these eorts, and there has been little discussion about
the impact o current policy stances and the benets o a regional approach to ood security.
Opening up ood staples to regional trade will create winners and losers. For exam-
ple, where reorm reduces the gap between producer and consumer prices, armers and
poor consumers will gain; intermediaries earning rents, both in public sector agencies and
well-connected private sector interests will lose. Without a political and social consen-
sus on agricultural reorm, it is dicult to implement and sustain policy change. Tis is
especially true i this lack o consensus prevents the creation o new institutional arrange-
ments that moderate the impact o uture shocks and instability in agricultural markets.
7/31/2019 Africa Can Help Feed Africa
17/106
xv
Executive Summary
Te absence o a stable and predictable policy environment at the national and region-
al level in many countries has provoked mistrust between government and the private sec-
tor. Tis, in turn, constrains private sector investment in ood and thus limits the capacity
o the private sector to expand production and trade, while encouraging governments to
hedge against the ailure o the private sector to supply ood when shortages arise. A sit-uation oen exists in which private sector rms are motivated more by ear o loss than
by opportunity or gain. Te key challenge is how to create a competitive environment in
which governments make credible commitments that allow the private sector to invest and
respond in ways that moderate price uctuations and deliver ood security.
wo related actors could help governments build constituencies or reorm and provide a pre-
dictable and stable policy environment:
i. An inclusive dialogue on ood trade reorm inormed by timely and accurate data on
global, regional, and national markets. In many Arican countries decisions about oodtrade policies are made mainly at the highest levels o government, too oen with-
out critical analysis and consideration o options. Food trade policy is rarely subject
to open discussion, and the interests and views o the broad group o stakeholders in
ood staples trade policies are seldom represented. And when there is open discus-
sion about trade reorm, decision makers rely most on the input o those with political
inuencethat is, those in government agencies, whose size and inuence depends
on the current institutional arrangements, and private sector interests, including those
that earn rents as intermediaries.
ii. A reorm strategy that provides a clear transitional path to integrated regional markets
rather than a single but politically uneasible jump to competitive markets. Te rangeo barriers to trade along the value chain, as well as the need to invest in market-sup-
porting institutions, shows that delivering integrated regional ood markets involves
more than a one-o commitment, and that reorms cannot be implemented by the
stoke o a pen. Tus, or many policymakers, attaining open and competitive region-
al markets will not happen during their electoral terms. Te reorm strategy will have
to take place in incremental steps that encourage investment by reducing uncertain-
ties about policies or the private sector and deliver real and visible benets. At the
same time it will allow policymakers to move at a pace consistent with their politi-
cal risk calculations and their capacity to address the concerns o those who will lose
rom the reorm process.
Te rst phase will set clear rules or public sector actions in the staples market to minimize
the uncertainties that discretionary interventions might cause. Te challenge is to provide disci-
pline on short-term policy responses to short-term ood insecurity that have negative long-term
impacts on market development, productivity growth, and ood security.
7/31/2019 Africa Can Help Feed Africa
18/106
xvi
Africa Can Help Feed Africa
o promote transparency and private sector participation, governments could commit to pre-
cise notication procedures, both nationally and regionally, beore restrictions on exports are imple-
mented. Governments would agree to orego the use o export bans and apply an export quota
announced early, which could be increased but not reduced during the marketing year. Countries
would commit to limit public buer stocks to a level that satised three months o emergency needs.Governments would retain the capacity to act in times o short-term ood crises but would assure
the private sector that governments will not arbitrarily buy in or distribute staples rom the stocks,
with subsequent impacts on prices and protability.
Better-inormed and more open discussion o ood trade policies would promote greater trust
and understanding between the private sector and the government over ood security. Te ollow-
ing actions would support this process:
Improving stakeholder access to timely and accurate inormation on ood staples. Lack
o reliable and up-to-date inormation on crop supply, demand, stocks and export avail-
ability has led to hasty, ill-dened, and uncoordinated policy responses to crises that, inturn, have sparked even greater volatility. Although regional communities in Arica have
made some progress in developing inormation systems, the international community
could help build capacity and strengthen these institutions, so that they deliver up-to-
date inormation and sound orecasts o agricultural market developments within Ari-
ca and the global economy.
Knowledge platorms anchored in the Arican Union (AU) that collect, analyze, and diuse
knowledge and best practices on agricultural trade policy reorm across the value chain
to support increased intra-regional trade in Arica. Te platorms would bring together
stakeholders, legislators, and regulators, and unite agriculture and trade specialists. Te
platorms would ensure that countries have access to (a) the best inormation on rulesand regulations that aect trade in staples; (b) the appropriate design o trade and regu-
latory reorms and the capacity or eective implementation; (c) likely outcomes o spe-
cic reorms, including overall benets as well as identication o those who may lose;
and (d) policy options to address any adverse distributional consequences. Te platorms
would bring together all the players to discuss the creation o integrated regional markets
or ood staples. Platorms anchored at the AU would also disseminate experiences and
best practices to and rom regions within Arica. Tis would limit costly duplication o
analysis, advice, and technical assistance across the dierent regional groupings in Ari-
ca. We propose separate platorms, one to support reorm o regulations governing trade
in ood staples and trade in seeds and ertilizers and one to support trade acilitation and
logistics regarding the movement o staples across borders.
7/31/2019 Africa Can Help Feed Africa
19/106
xvi
Executive Summary
Te report highlights the range o barriers to ood trade in Arica along the entire value
chain. Te issues pertain to many ministries and agencies within government: trade, agri-
cultural, health and saety, transport, and nance. Tis in turn requires a whole o gov-
ernment approach to reeing up ood trade, which will require strong and eective lead-
ership to articulate the rationale and sustain the momentum or reorm. Leaders mustalso address the hard choices that will arise in dealing with the political economy con-
straintsconstraints that have until now blocked the capacity o Arica to exploit its enor-
mous potential to eed Aricans.
7/31/2019 Africa Can Help Feed Africa
20/106
7/31/2019 Africa Can Help Feed Africa
21/106
Introduction and context
1
PhotographbyArneHoel,WorldBa
nk
7/31/2019 Africa Can Help Feed Africa
22/106
2
Africa Can Help Feed Africa
he rise in global ood prices and the ever-growing ood import bill have prompted sharp
attention on agricultural policies in Arica. Arican policy makers are grappling with what
unstable ood prices mean or their countries; how these price movements will aect their
ood security situation; how the private sector is likely to respond; and what governments them-
selves can do. In addition, they ear that global warming may signicantly change the location oood production within Arica.
Tis report discusses how opening up cross-border trade will boost the potential or great-
er ood production in Arica and contribute to ood security by improving poor peoples access to
ood and by increasing returns to poor armers or the ood they produce.
Agricultural resources are not allocated equally across countries, or even within them, so bor-
ders oen articially demarcate ood surplus areas rom ood decit ones. For example, Southern
Malawi is not well endowed with agricultural potential and is persistently ood decient. Nearby,
Northern Mozambique is a productive area or growing maize, the main staple o the region, but
it is ar rom the main area o national consumption in the south o the country. Kenya is maize
decit and depends on imports, especially rom anzania. Given dierences in weather patternsbetween countries, regional production is oen less variable than production at the country level.
Regional trade integration can substantially impact ood security by linking armers to con-
sumers across borders, ameliorating the eects o periodic national ood shortages and o escalat-
ing global ood prices. What matters or poor people is local price volatility, and changes in global
prices are only one actor behind domestic price changes. Local crop conditions, supply costs, and
policy measures, including those aecting trade in ood, are oen more important.
However, regional trade in ood staples remains ar rom ree despite the implementation o a
host o regional trade agreements and eorts or policy and regulatory harmonization. Te arbi-
trary and erratic imposition o barriers undermines private sector condence to invest and dis-
torts incentives towards cash crop production and away rom ood staples.Tis report is organized around our main themes or messages:
Te potential or regional trade in ood in Arica is not being exploited;
Barriers to trade along the value chain that prevent this potential rom being realized;
Te need to invest in institutions that support efcient and saer market outcomes; and
Te importance o addressing the political economy realities that have prevented Arican
countries rom realizing the gains rom opening up to regional trade
A nal section oers recommendations to promote dialogue among stakeholders about inte-
grating into regional ood markets, and suggestions on ways the international community can help.
Te report rst provides an overview o the typical value chain or a ood staple as well as a brie
outline o current approaches to regional trade in ood staples in Arica. However, we must rst
dene what the termood staples means.
Te general denition o a staple is a ood that is eaten regularly and constitutes the dominant
part o the diet and supplies a major proportion o energy and nutrient needs. Tus, staple oods
vary by location but typically include grains, tubers, legumes, or seeds. In the Arican context, the
7/31/2019 Africa Can Help Feed Africa
23/106
3
Introduction and context
emphasis has been on cereals, especially maize and rice, and to a lesser extent on starchy root vege-
tables such as potatoes, yams, and cassava as well as pulses (dried legumes) and ruits such as plan-
tain. As will be subsequently discussed, the political economy issues that surround production and
trade o the key cereals has led to problems when implementing reorms. Tis suggests an impor-
tant role or trade in substitutes or the main grains, such as cassava and sorghum, which in thepast have been less vulnerable to political capture.
Te ood staple value chain
When consumers buy a staple ood the price reects that the product has passed through a num-
ber o stages o a value chain. Tis starts rom a seed and continues through a number o inputs,
including ertilizer, labor, and knowledge beore proceeding through a marketing and distribution
system that includes grading, bagging, transportation, and the wholesale and retail stages (Figure
1). When the product or its inputs cross a border there are additional elements linked to customsprocedures, taxes, and regulatory requirements. Te extent o competition and the nature o gov-
ernment policies can, thereore, inuence the price that the consumer pays and the price that the
armer receives or the product at multiple points along the staples value chain at which actors the
extent o competition and the nature o government policies can inuence the price that the con-
sumer pays or the product as well as the price that the armer receives. I the consumer purchas-
es the product in a processed orm (or example, our) the vertical chain linking the armer and
nal consumer increases by an extra level.
Te level o ood staple production in Arica thereore has important implications or employ-
ment beyond that required on the arm. Substantial increases in output that would ollow rom
bridging the gap in yields between Arica and other countries would generate a substantial amounto employment. While much o the employment would be in rural areas, rising agricultural pro-
ductivity and greater regional trade would create jobs in processing and a wide range o services
sectors, including transport, distribution, and retailing.
Te number o stages in the value chain and the value added tend to be less or staples than
or processed products. Tis means that the impact o specic costs along the chain or staples will
likely have a greater proportional impact on price than in more sophisticated chains. Te cost o
transportation comprises a high share o the overall cost o staples given their low-value to weight
ratio. For example, a 2007 value chain study in Cameroon (Keyser, with Nkama and Doya) ound
that domestic handling and transport costs accounted or 2135 percent o total shipment value
or resh cassava over a delivery distance o 130km, compared with less than 12 percent or cotton
lint delivered more than 1,200km rom northern Cameroon to the port o Douala. When the value
chain or staples extends across a border, costs such as those associated with border procedures,
international transportation, and satisying standards in the overseas market will likely have a dis-
proportionate impact relative to higher value products that are less vulnerable to damage.
Economic analysis o the impact o policies aecting agricultural markets typically assumes
competitive outcomes. However, lack o competition at any stage along the value chain can have
7/31/2019 Africa Can Help Feed Africa
24/106
4
Africa Can Help Feed Africa
important implications or the outcome o policy change and external shocks on producers o ood
staples and consumers (McCorriston 2011).1 For example, the decline in consumer prices rom the
removal o a trade barrier will be less i there is market power at some point along the value chain.
Hence, a lack o competition in, say, the transportation and distribution o staples might allow this
sector to capture the largest benets o trade reorm. Clearly, world prices still matter, but the struc-
ture o the value chain and policy interventions along that chain aect the way that changes in glob-
al commodity prices are translated into movements in consumer prices.
Porto et al. (2011) discuss a number o case studies o export crops in Arica to assess the impacto the structure o export value chains on poverty and welare in rural areas. Tey conclude that
greater competition among the processors and traders o cropssuch as cotton in Zambia, Mala-
wi, and Burkina Faso, coee in Uganda, Rwanda, and Cte DIvoire, tobacco in Malawi and Zam-
bia, and cocoa in Cte DIvoire and Ghanais good or smallholders since arm gate prices tend
to be higher. For example, with cotton in Zambia we ound that i the largest processing rm were
split then the income o the average cotton producer would increase by 2.4 percent.
Similarly, market power in the distribution sector can act as an eective import barrier. Fran-
ois and Wooton (2010) show that higher concentration in the distribution sector translates into
signicant trading costs. Tis means that by neglecting the structure o the domestic distribution
sector, the market access benets rom tari reductions may be seriously overestimated. Hence,with staples, a lack o competition in transportation, distribution, and processing will mean lower
arm gate prices or producers and higher consumer prices.
Figure 1: Key elements in a typical staples value chain
Production:Harvest /Collection
Transport /Transit / Storage
Border Wholesale /Retail Markets
Inputs: Seeds,
fuel, fertilizer
Extension
services
Financial
services
Consolidation
of production
Infrastructure
(physical)
Transport /
logistics
services
Storage
facilities
Regulatory
environment /
competition
between
service
providers
Customs
procedures
Border
facilities
Informal
payments
Market access
Food safety
and quality
standards
Competition
Access of
smallholders
to modern
distribution
sector
1 McCorriston highlights that, or the producer o ood staples, the relevant demand is not that o the consumer but rather
the successively related derived demand rom the various actors in the value chain, the nature o which reects the extent
o competition throughout the value chain.
7/31/2019 Africa Can Help Feed Africa
25/106
5
Introduction and context
Regional production and trade patterns
Arica has traditional areas o ood decit and ood surplus. Te ormer include drought prone
areas, such as the Horn o Arica and the Sahel, that oen experience crop ailure. Conict and civil
unrest compounds this problem in the Horn, the Congo Basin, and pockets o West Arica. Te lat-
ter comprise Aricas highly productive agricultural zones, which occur in two types o areas. First
are those with avorable and reliable rainall that do not get too cold in the winter (cultivation o
cassava and banana needs warm conditions) and watersheds, where there is economical irrigation.
An example is Southern Mali, which exports surplus sorghum to Niger and coastal West Arica.
Second are the staple ood basket zones (Jayne et al.2007),where exible climates support
the cultivation o dierent ood crops, particularly cereals, alongside year-round drought-resis-
tant crops or domestic consumption. Tese zones produce a surplus o cereals or export to de-
icit regions in periods o ood stress. Examples include (a) Northern Zambia, where cassava pro-
duction ensures domestic ood security, even in drought years, enabling the region to export maize
to DR Congo, Malawi, and elsewhere in Zambia; (b) Eastern Uganda, where bananas and cassava
ensure ood security, thereby enhancing maize exports to chronically ood decit Kenya; (c) North-
ern Mozambique, where cassava and Irish potato cultivation provide local ood, enabling regular
maize exports north into Kenya and south into Malawi; (d) most o anzania, where a combina-
tion o rice, cassava, bananas, and maize enable regular cereal exports north into Kenya and south
into Malawi; and (e) South Arica, where large-scale commercialization and mechanization com-
bined with modern inputs and irrigation enable high yields or the export o cereals northward to
Zimbabwe, Southern Mozambique, and Malawi (Haggblade 2008).
Maize is the most important ood staple in Eastern and Southern Arica, and the most widely
traded agricultural commodity; in most countries its availability is equated to ood security. Rural
ood surplus production zones supply major decit urban consumption centers as their naturalmarkets. Tese spatially linked clusters o production and consumption zones are reerred to as
ood sheds. Figure 2 shows the maize ood shed in Southern and Eastern Arica and how nation-
al boundaries intersect natural ows o ood. I these borders add signicant costs to moving ood
rom surplus to decit regions then trade ow will be constrained.
In West Arica, there are three major agro-ecological zonesthe Sahelian, Sudanese, and Coast-
al zoneswhere production and consumption o ood staples can be easily classied. In the Sahe-
lian zone, millet is the principal cereal cultivated and consumed. Exceptions include Cape Verde,
where maize and rice are more important; Mauritania, where sorghum and maize are staples; and
Senegal, with rice. Te principle substitutes are sorghum, rice, and cassava our. In the Sudanese
zone,2
maize and sorghum are the principal cereals consumed by most o the population, ollowedby rice, cassava, and yam. In the Coastal zone (with two rainy seasons), yam and maize are the most
important ood products, supplemented by cowpea (USAID 2010).
2 Southern Chad, Central Nigeria, Benin, Ghana, ogo, Cte dIvoire, southern Burkina Faso, Mali, Senegal, Guinea Bis-
sau, Sierra Leone, Liberia).
7/31/2019 Africa Can Help Feed Africa
26/106
6
Africa Can Help Feed Africa
Nigeria is by ar the largest pro-
ducer o ood staples in West Arica
and exports to neighboring countries
(millet, sorghum, and yams). Among
West Arican countries, it is also theone that imports the most ood (in
particular grains) to satisy urban con-
sumption needs (accounting or 36
percent o Economic Community o
West Arican States (ECOWAS) ood
imports). Te most important crops
or Nigeria are roots, tubers, and grains.
Nigeria is the worlds leading produc-
er o cassava, yams and taro root, and
the second largest producer o sweetpotatoes. It accounts or 69 percent
o the regional supply o these prod-
ucts. Nigeria also accounts or hal
o the grains produced in West Ari-
ca and is the worlds largest produc-
er o cowpeas.
Still, while domestic production
provides most o the ood or Nigerias
urban areas, the country has a decit in rice and wheat. Every year Nigeria imports more than one
million tons o rice, making it among the largest rice importing countries in the world, and morethan two million tons o wheat our.
For trade ows, there are three trade basins in West Arica known as the West, Central, and
East basins. In the West basin3 rice is most heavily traded (see Box 1). In the Central basin4 maize
is mostly traded. And in the East basin5 millet is most requently traded.
Agriculture and ood trade policy reorm
Many countries have reormed their traditional state controlled ood marketing systems to dual
systems in which both the private and public sectors are involved in ood staple input and out-
put markets. Following this partial liberalization o ood markets, the private sector has started to
emerge and invest in commodity trading and processing. Most government parastatals no longer
Figure 2: The maize ood shed in Southern andEastern Arica
Source: Haggblade (2008).
3 Mauritania, Senegal, western Mali, Sierra Leone, Guinea, Liberia, the Gambia4 Cte dIvoire, central and eastern Mali, Burkina Faso, Ghana and ogo.5 Niger, Nigeria, Chad and Benin.
7/31/2019 Africa Can Help Feed Africa
27/106
7
Introduction and context
have the nancial or operational clout to meet a countrys entire annual marketing, storage, and
processing requirements, and rely on the private sector or some o these unctions. In many cases,
delivery systems or ertilizer, seeds, and veterinary services have also been passed rom the pub-lic to private service.
Yet government marketing institutions continue to operate in the marketoen alongside the
private sectorboth through their direct procurement and sales operations, including input subsi-
dy programs and through trade policy instruments. Tese institutions oen try to inuence mar-
ket prices through their purchase and sale operations.
Over the last decade, the ood marketing policy environment in SSA has remained sub-optimal
in its support or agricultural productivity growth. Interventions have had an adverse impact on
the regional market or ood staples, especially when prices are articially manipulated and incon-
sistent with world market prices. o benet urban consumers, and to release unwanted stocks, gov-
ernment ood agencies have sold public reserves o ood staples at below-market prices. Tis trig-
gers unpredictability or arms and rms alike, diminishing their productive and trading capacities,
especially during periods o poor harvest,6 and raising ears o being undercut by subsidized gov-
ernment ood.
Box 1: Regional trade in rice in West Arica
With demand outstripping domestic production, West Arica is a major and expanding rice import
market. West Arica accounts or almost 20 percent o world rice imports, amounting to 6.3 mil-
lion tons in 2008. The main importing countries are Senegal, Nigeria, and Cte dIvoire. The larg-
est exporter to the region is Thailand, but China, India, and the US have also become important
sources or some countries. At the regional level, almost all cross-border rice trade is imported
through ormal transit shipments rom the main ports and through inormal trade. The latter is
signicant. For example, Benin ocially imports as much as 800,000 tons o rice in some years;
but the bulk o these imports are parboiled rice, which is trans-shipped via Cotonou to Nigeria
through inormal channels. Almost all trade rom surplus rural production areas to decit urban
markets remains within the country o production due to high transport costs and customs or-
malities. In 2007, ocial regional rice exports were less than 79,000 tons, principally rom Sen-
egal, Ghana, Cte dIvoire, and Togo. Other than Nigerian parboiled rice, inormal exports o
West Arican rice are mostly rom surplus areas close to borders. For example, there are some
exports o domestically cultivated rice rom the Senegal River Valley Mauritania, where rice pric-
es are higher. The largest rice decit countries are Benin, Cameroon, Cte dIvoire, Ghana, Libe-
ria, and Senegal. Most imported rice is consumed in the coastal regions in the larger cities. Dry
rural areas also experience signicant rice decits.
Source: USAID (2009)
6 For example, uncertain output prices in the ace o government intervention in the market coupled with risk-averse
smallholder armers reduces the incentive to use productivity-enhancing inputs such as ertilizer.
7/31/2019 Africa Can Help Feed Africa
28/106
8
Africa Can Help Feed Africa
It can even cause their withdrawal rom the market altogether (schirley et al. 2006). A study
o Zambian maize markets showed that several international grain trading companies le the mar-
ket because o high risks imposed by unpredictable government participation in the maize market
(Nijho et al. 2003a). Consequently, despite intervening in the market, governments have oen been
unable to prevent ood prices rom exceeding import parity levels, generating acute ood shortag-es and disrupting regional trade in these products (Box 2).
Statistical analysis by Minot (2011) nds that the world price o maize has been less volatile
than the price o maize in Arica, suggesting that strategies o national ood sel-suciency have
not been eective in reducing ood price volatility and may even have exacerbated the problem.
Box 2: Government imports o maize during the 2001/02 Southern Aricaood crisis
In 2001/02, the Zambian government publicly announced that it would import 200,000 tons o
maize rom selected South Arican suppliers to cover the national ood decit, and then sell it
below market price to a small number o large ormal sector millers. A subsidy was intended to
limit consumer price increases, paid directly to the South Arican suppliers in addition to the
importers payments. Because o liquidity problems, the subsidy payment was late, thereby delay-
ing the maize imports. When the government instead imported just 130,000 tons very late in the
season, there were maize and maize four shortages, and local market prices exceeded import
parity. Traders and millers not selected to benet rom the scheme, including inormal traders
rom Mozambique, rerained rom commercially importing maize or ear o not being able to
sell once subsidized maize reached the market. Because grain was channeled just to the large
millers, consumers had to pay the higher price or already-rened four instead o being able to
source grain and mill it themselves or though the inormal network o hammer mills.
In the same year, Malawi also aced a modest maize production deciteight percent below the
countrys 10-year average. In September 2001, its grain trading parastatal (ADMARC) announced
a xed price or maize to be sold at its distribution centers and declared its intention to import
maize rom South Arica to maintain this price. The selling price was set considerably lower than
the landed cost o imported maize, leaving private traders with no incentive to import commer-
cially. As with Zambia, the government imports also arrived late and were insucient to meet
demand so prices soared to a peak o US$450 per ton in early 2002. To make matters worse, the
late-to-arrive ADMARC imports arrived during the good 2002 harvest. These were then released
on to the market, resulting in 16 months o continuously alling maize prices to the detriment o
armers. At other times, the sourcing o grain rom South Arica and subsequent release on to the
domestic market through government contracts with South Arican suppliers has also depressed
inormal maize trade with Mozambique. Since Mozambique is the source o inormal trade in
maize to southern Malawi, these government imports also add greater risks and price instability
or Mozambiques smallholder armers.
Sources: Nijho et al. (2002); Jayne et al. (2007); Rubey (2004); Nijho et al. (2003).
7/31/2019 Africa Can Help Feed Africa
29/106
9
Introduction and context
Tere have been ew cases where government attempts to stabilize ood prices have been success-
ul, and interventions to moderate ood price volatility oen impose substantial costs on consum-
ers and armers in neighboring countries (see Box 3).
Cross-border trade in ood staples in Arica remains limited. Prices or staples, especially in
land-locked countries, can vary substantially between years o domestic good harvest and those
o poor harvest. Tis explains why governments still empower agencies to intervene in agricultur-
al markets to supposedly (but ultimately ruitlessly) stabilize prices, and why these interventions
only discourage intra-regional trade. Te challenge is to allow regional trade to improve access to
ood and moderate price volatility while developing institutions that support price stabilization.
Te ollowing sections discuss the key constraints to regional trade in Arica.
Box 3: Maize policy in Kenya
Maize policy in Kenya tries to support and stabilize prices through the National Cereals and Pro-
duce Board (NCPB). In the 1980s the NCPB played a major role in the domestic maize market,
purchasing 600,000800,000 tons annually. Since then, maize markets have been liberalized and
private sector trade plays a much larger role.
However, the NCPB continues to purchase maize to deend a foor price. Since 2000, NCPB pur-
chases have been 30,000190,000 tons per year. NCPB operations are estimated to have increased
domestic maize prices by 20 percent during 19952004 because they account or 2535 percent
o all maize sold by the agricultural sector in Kenya. Most o the maize purchased has been direct-
ly rom large-scale armers in the Rit Valley.
To deend high maize prices, the government has limited maize imports. In mid-2001 a tempo-
rary ban was imposed on cross-border imports o maize because o low prices associated with
a good harvest. Another temporary ban was introduced in 2004 in response to an outbreak o
afatoxin poisoning thought to be caused by imports rom Uganda. And recently Kenya has pre-
vented Tanzanian trucks carrying maize rom entering Kenya, orcing them to o-load and reload
onto Kenyan trucks. The main beneciaries rom high maize prices within the country have been
the largest arms, and the main losers have been the net buyersthat is, urban consumers and
maize-purchasing rural households.
Source: World Bank (2009).
7/31/2019 Africa Can Help Feed Africa
30/106
7/31/2019 Africa Can Help Feed Africa
31/106
Aricas enormouspotential or regionaltrade in staples is notbeing exploited
11
PhotographbyP.Casier,CGIAR
7/31/2019 Africa Can Help Feed Africa
32/106
12
Africa Can Help Feed Africa
Arica has considerable variation in its climatic conditions across countries and even with-
in them. Te distribution o ood crop cultivation coupled with possibilities, where they
exist, or staggered harvesting within the same commodity oer substantial opportunities
or trade. Since production variability is not highly correlated among most countries in the region,
integration through regional trade oers the prospect o canceling the eects o small country sizeon production volatility (Koester 1986). Size matters as larger countries, or groups o countries,
typically have more diverse climatic conditions than smaller ones, which reduces systemic risks at
the country level.
Increased regional trade has the potential to (a) expand the size o Aricas market or ood sta-
ples, (b) boost agricultural growth in surplus zones, and (c) mitigate shortages in decit ones. In
addition, regional trade in ood staples can also help moderate price volatility in Arican ood sta-
ple markets (see Dorosh 2001).
Suppliers rom outside the continent are meeting most o Aricas growing demand or ood.
Te production o ood staples or growing urban markets and ood decit rural areas presents the
largest growth opportunity or Arican armers. Te market value o Aricas ood staple produc-tion is at least US$50 billion per year, equivalent to three-quarters o all agricultural output (World
Bank 2008). Given population growth and increased urbanization, Aricas demand or ood sta-
ples will grow dramatically in the coming decade. Indeed, demand is expected to double by 2020,
primarily in cities. Import bills or ood will rise substantially i Aricas armers do not contribute
more to meeting this demand.
Unortunately, the experience o the past two decades shows that Arican armers have not been
able to do this. And the enormous potential to produce and trade ood staples in Arica remains
unexploited. Tus, despite being rich in agricultural potential, the continent increasingly imports
rom outside o the region to satisy demand. Figure 3 shows that the trend in the volume o net
exports o staples in all our regions o Arica is strongly negative.FAO estimates that Arican cereal imports in 2008 were US$15.2 billion. However, just ve
percent o all grain imported by Arican countries originates rom regional sources. Te volume
o extra-regional ood staple imports shows how regional integration in the ood staple market can
achieve ood security.
Figure 4 compares the deterioration in the net trade position o sub-Saharan Arica as a whole
with that o other regions. While Aricas imports o ood staples have grown at a much aster rate
than exports, other developing regions such as Eastern Europe and Central Asia, South Asia, and East
Asia and the Pacic have seen a turnaround rom a position o net imports in 1990 to net exports 20
years later in 2010. In Latin America the gap between imports and exports has narrowed substantially.
Tese dierential trends in imports between Arica and other developing country regions o
the world suggest that the role o heavily subsidized production and export in OECD countries,
while important, has played a lesser role than domestic actors in explaining the underlying causes
o the increase in net imports o staples in Arica.
Te increasing imports o ood staples are shocking, especially since Arica has the agricul-
tural potential to satisy a large part o this demand. Applying inputs, such as widely available high
yield seeds and ertilizers would substantially increase output in Arica.
7/31/2019 Africa Can Help Feed Africa
33/106
13
Aricas enormous potential or regional trade in staples is not being exploited
A large body o literature has identied the gap in staples yields obtained in Arica relative toyields o staples elsewhere in the world. Te potential to experience a two- to threeold yield increase
among some o the basic ood staples is possible i more armers can access and efciently utilize the
available stock o knowledge and technologies(Nin-Pratt et al. 2009).
Calculations or Central and Western Arica suggest that the ratio o potential to actual out-
put to be around 1.5 or cassava, 1.9 or rice, 2.7 or maize, 3.3 or millet and sorghum, and 5 or
wheat. With regard to catching up to yield potential, Nin-Pratt et al. (2009) estimate that by 2015
imports into the Central and West Arica region would all rom a base case o $12.4 billion to $9
billion, and exports would rise rom $10.6 billion to $16.4 billion, reversing the regions net import
situation. Tis study indicates that i this was accompanied by better trade policies and measures
to improve market conditions, then exports would increase to $22.1 billion, and imports would
increase slightly to $10.1 billion. Tese are optimistic scenariosbut the estimates demonstrate
the potential or increased trade. What willbe important is to remove barriers to higher yielding
seeds, ertilizers, and knowledge as well as link armers to markets and provide the incentives or
investment that delivers higher productivity.
It is also true that large swathes o ertile land in Arica remain idle. For example, the Guinea
Savannah zone covers around 600 million hectares in West Aricathrough Uganda and anzania
Figure 3: The volume of net exports of food staples by sub-region 19762008 (1,000 tons)
1000
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2004
2008
2002
2004
2004
2008
East Africa
Net exports Linear (net exports)
West Africa
Southern Africa Central Africa
5000
50010001500200025003000350040004500
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
0
1000
2000
3000
4000
5000
6000
7000
5000
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2004
2008
2002
2004
2004
2008
4000
3000
2000
10000
1000
2000
3000
4000
5000
0
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
200
400
600
800
1000
1200
Source: FAOStat
7/31/2019 Africa Can Help Feed Africa
34/106
14
Africa Can Help Feed Africa
and encompassing Malawi, Zambia, Angola, and Mozambiquewhich is around one-third o the
total area o sub-Saharan Arica. O this, 400 million hectares can be used or agriculture. How-
ever, less than 10 percent o this area is being cultivated, suggesting huge unutilized agricultural
potential (World Bank 2009b). Te experience o Tailand shows how areas thought to be back-
ward and with little prospect can prosper in the right conditions. Figure 5 shows the recent expan-
sion o Tai cassava exports and how they have been driven primarily by exports to countries in
the East Asia region.
Almost all o the increase in staple ood imports in Arica is caused by the rising demand o
urban populations, which are growing at over our percent annually compared to less than one
percent per year or rural populations. Te shi rom rural to urban liestyles is signicant con-
sidering the high share o staple oods in their total consumption. Among the urban poor, who
comprise the bottom 20 percent in terms o poverty, ood makes up 6070 percent o total expen-
ditures. Across all urban consumers, ood accounts or 4555 percent o total annual household
expenditure (Jayne et al., 2009).
Tis shi in demographics also brings new opportunities as more o the urban demand becomes
commercialized. I Aricas armers can capture a good share o the market created by this growth,
intraregional trade in ood staples could signicantly advance growth and poverty reduction on
the continent.
Global warming may also make ood production and prices in Arica more volatile, increas-
ing the ood security risks and making moving ood rom areas o surplus production to ood de-
icit regions even more urgent. Te IPCC predicts that temperatures in Arica will rise more than
the global median, and extreme weather events will become more prevalent, aecting agriculture
in some regions, while dry areas will become drier, and wet areas will likely receive more rainall.
Figure 4: Aggregate ood staple trade by world regions, 1990 and 2010 (millions, US$)
ECA SAR SSA EAP LAC MENA
Imports
1990 2010
Exports
0
2000
4000
6000
8000
10000
12000
14000
ECA SAR SSA EAP LAC MENA
0
500
1000
1500
2000
2500
3000
3500
4000
Source: COMTRADE.
7/31/2019 Africa Can Help Feed Africa
35/106
15
Aricas enormous potential or regional trade in staples is not being exploited
Schlenker and Lobell (2009) esti-
mate that average maize productivi-
ty in sub-Saharan Arica may decline
22 percent by mid-century. However,
there is considerable heterogeneity inthe impacts o climate change across
countries, and so international agri-
cultural markets may allow or pooling
o the risk posed by local (or nation-
al) climate extremes. Farmers in coun-
tries severely aected by weather may
be able to sell excess supplies to meet
the excess demand rom consumers in
the more severely aected regions. In
the medium to long run, declines inagricultural production rom climate
change in some countries could be o-
set by production increases in other
regions. Again, this puts the emphasis
on linking ood production areas and
major consumption centers through open and predictable regional trade. A regional approach to
ood security will lower the costs o adapting to rising temperatures compared to the costs incurred
i countries isolate production rom regional demand.
anzania is an example o a country where grain production variability may increase because
o climate volatility. Ahmed et al. (2010) analyze the archive o global climate model simulationsand nd that severe dry conditions in anzania will most oen coincide with non-dry conditions
among anzanias key Arican trading partners. Tis suggests that anzania could benet rom
exporting grains to countries within and outside Arica as climate change increases the possibility
o severe precipitation decits in other countries while simultaneously decreasing the likelihood
o severe precipitation decits in anzania. However, i current policy stances to shocks are main-
tained, especially export restrictions, then anzanian armers will not be able to capitalize on his-
torical export expansion opportunities.
Figure 5: Regional markets can drive trade instaples: Thai exports o cassava2000 to 2010
0
200
400
600
800
1000
1200
1400
1600
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
US$Million
Total exportsExports to region
7/31/2019 Africa Can Help Feed Africa
36/106
7/31/2019 Africa Can Help Feed Africa
37/106
17
PhotographbyJosephHadar,WorldBank
Regulatory barriers to tradeundermine Africas potentialin regional food trade
7/31/2019 Africa Can Help Feed Africa
38/106
18
Africa Can Help Feed Africa
Arica must increase outputs to moderate rising ood import bills and satisy growing
demands. Te degree o access to markets or nal products is critical. Linking rural ood
surplus production zones in Arica to major decit urban consumption centers requires
a well-unctioning regional market or these products.
But improved productivity and a decline in the enormous gap in yields between Arica andthe rest o the world requires the using higher-yield seed varieties, more intensive ertilizer use,
and better production techniques. Yet there are major constraints on access to these key inputs,
oen because o trade barriers and limitations on the movement o proessionals across borders.
Te debate about agricultural policy is oen presented as a dilemma between pursuing high-
er producer prices or armers (to encourage investment in higher yield seeds, use o ertilizers,
and better techniques) and delivering lower nal prices or consumers. But this dilemma does not
appear when policy interventions have raised transaction costs between armers and consumers,
and the policy environment has permitted limited competition at key points along the value chain.
Such policies lead to a large gap between producer and consumer prices and entail that the bene-
ts o previous reorms and investments, such as tari removal and better quality roads, may nothave accrued to either armers or consumers, but rather to those providing services along the chain,
such as transport and distribution.
Arican smallholder armers who sell surplus harvest typically receive less than 20 percent o
the consumer price o their products, with the rest eaten away by transaction costs and post har-
vest losses (AGRA 2009). Clearly, this limits the incentive to produce or the market. Policies that
reduce transaction costs and increase competition in the provision o services that aect the pro-
duction and distribution o ood staples could reduce the gap between consumer and producer
prices by reducing consumer prices, increasing producer prices, or both. USAID (2011) estimates
that in West Arica a reduction in transaction costs equivalent to 10 percent o the arm gate price
o agricultural products would lead to a our percent increase in production and a similar increasein the real income o armers, together with an eight percent all in consumer ood prices and a
seven percent increase in real disposable incomes o consumers.
Many o the key barriers to trade in ood staples relate to regulatory and competition issues
at elements along the value chain. As taris have come down, so a tangled web o rules, ees, and
expensive services is strangling Aricas regional trade in ood.
But in many cases an eective market system in ood inputs and nal products requires well-
designed regulationsor example, to ensure that consumers receive reliable inormation about
the quality o the ood or seeds they buy. Tis is important to protect countries rom disease and
pests that could devastate local production. In some cases, the policies restricting trade are deliber-
ately protectionist; but in other instances the lack o regulations limits the development o region-
al markets, such as in the case o ertilizers. In many cases the issue is to dene appropriate regu-
lations and create better institutions to implement them.
Te ollowing section summarizes inormation on the key barriers to trade at dierent points
along the value chain o ood staples. Specically examined are: (a) barriers to trade that limit access
to seeds and ertilizers; (b) actors that lead to high transport costs in Arica; (c) dangers o cross-
ing borders, especially or small inormal traders many o whom are women; (d) costs o opaque
7/31/2019 Africa Can Help Feed Africa
39/106
19
Regulatory barriers to trade undermine Aricas potential in regional ood trade
and unpredictable trade policies that limit trade in staples across borders, mainly non-tari barri-
ers (NBs); (e) inecient distribution services that ail to link poor producers to poor consumers.
rade barriers limit access to key inputs: Seeds, ertilizers,and extension services
Seed security is closely linked with ood security in Arica, but many Arica armers lack access to
modern high-yield seed varieties. In Ethiopia, or example, Alemu (2010) estimates that only about
one-quarter o the demand or improved seed is being satised. Use o modern seed is undamen-
tal to increasing productivity towards its potential; and use o improved hybrid maize in Ethiopia
could quadruple productivity. Even i just hal the armers achieved the productivity associated with
using hybrid seeds, the domestic production would replace commercial imports (Alemu 2010).
rade barriers that create high transaction costs between countries prevent seeds rom mov-
ing quickly across borders when a seed decit occurs due to a disaster, such as ooding, drought,or pest inestation. Inappropriate regulations and burdensome requirements governing the release
o new varieties create small and ragmented seed markets and ensure that varieties available else-
where are not widely disseminated in Arica (Minde and Waithaka 2006). Even within Arica, arm-
ers on one side o a border may not have access to higher yield seeds that have been approved by
the neighboring country under similar agro-climatic conditions.
Seed trade is constrained by regulations and policies established when plant breeding and or-
mal seed production were the responsibility o the public sector. Procedures or the release o new
varieties were designed to meet the needs o public research institutes, and seed certication was
primarily an internal quality control mechanism.
Now, however, a key concern with seed regulation in Arica is the inordinate amount o timerequired or the approval process (ripp 2005), which can take around two to three years or more.
Tis puts Arican armers at a serious disadvantage in competing with imports rom the global
market, which are derived rom seeds that deliver much higher yields, and which may require less
ertilizers and pesticides and be more pest and drought resistant. Arican armers must thereore
be allowed to acquire seeds widely available elsewhere. A rst step would be to ensure that test
data used in the approval process in other countries is included in Arican approval processes to
avoid duplicative testing.
Certication plays a crucial role in a market system in providing condence and bringing buy-
ers and sellers together. In the past, each Arican country developed its own seed regulatory regime;
however, these dierent regulatory arrangements have now created a situation where sourcing seeds
between neighboring countries is complicated, lengthy, and expensive.
While many countries have implemented a degree o liberalization in seed markets, government
and its agencies still heavily control the introduction o new varieties, trade, entry o seed compa-
nies, and in some cases seed prices and production. Central controls on seed production, includ-
ing compulsory certication and licensing or seed arms, can create conicts with pro-market and
pro-reorm processes. As a result, the private sector is oen poorly integrated into the national
7/31/2019 Africa Can Help Feed Africa
40/106
20
Africa Can Help Feed Africa
seed production and distribution system. In Ethiopia, or example, tensions exist between the state
and the emergent private sector, and the argument o the strategic importance o ood and agri-
culture and the perceived weak presence o the private sector is deployed to argue or strong state
control (Alemu 2010 p15).
Restrictions on the cross-border movement o seed are primarily due to NBs, resulting romsanitary, phytosanitary and plant quarantine measures, seed certication, and variety release regu-
lations, which dier across regions and countries. Regional trade in seeds has been hindered by a
lack o agreement among agencies in dierent countries about standards, and by inconsistent and
heavy-handed application o control procedures or imports. Tese costs and delays are exa