RECHERCHE ET EXPERTISE SUR L’ECONOMIE MONDIALE
Ce que l’on sait, ce que l’on ne sait pas et ce qu’il faut savoir sur le commerce alimentaire informel en Afrique (de l’Ouest) Antoine Bouët (CEPII) SWAC/OECD Experts Meeting: “Intra-regional food trade data in West Africa” 12/13 October in Paris, France
Introduction
• Importance of trade data in agriculture and food – A key input for the balance of payments – A key input for identification of products with comparative advantage – A key input for food safety and the establishment of “food balance sheets” – A key input for the private sector – A key input for academic research
• The African Context – The Malabo declaration – The AfCFTA – Importance of informal trade
Measuring trade in West Africa
• Main databases: • • • • •
COMTRADE BACI FAOSTAT Regional databases (CEDEAO…) Source of data: Customs => National Statistical Institutes
• In (West) Africa: – Informal trade is important • Trade in small quantities • Smuggling • Underinvoicing/misclassification
– Many factors in the African context: costs/benefits of formalization – ECOWAS: weak incentives of customs to record trade - no custom duties
The quality of ‘formal’ databases can be questioned
• Absence of declarations Frequency of declaration of trade flows, COMTRADE, ECOWAS countries from 2010 to 2016 Nigeria Ghana Cote d'Ivoire Mali Senegal Burkina Benin Togo Liberia Guinee Sierra Leone Cape Vert Guinee Bissau Gambie Niger
2010 Y Y Y Y Y Y Y Y
2011 Y Y Y Y Y Y Y Y
2012 Y Y Y Y Y Y Y Y
2013 Y Y Y
Y Y Y Y
2014 Y
2015
Y
Y
Y Y Y Y
Y Y Y Y
Y Y Y Y Y
Y Y Y
Y Y
Y
Y Y
Y
Y
Y
Y
Y Y Y
Y Y
Y Y
Y Y
Y Y
Y Y
Source : Mitaritonna et Traoré, 2017 Note: Y means that data are available
2016 Y Y
How are official databases (e.g. COMTRADE) improved?
• BACI database designed by CEPII from COMTRADE – Procedure that reconciliates mirror flows: declaration of exporting countries and importing countries – Transportation costs are evaluated thanks to a gravity model (proxy of CIF/FOB) – It weighs each declaration by the quality of reporting country according to an econometric procedure
How are official databases (e.g. COMTRADE) improved?
COMTRADE-BACI Comparison Total imports ECOWAS (Millions USD) 160
140 120 100 80 60 40 20 0 2000
2001
2002
2003
2004
2005
2006
2007 Comtrade
2008
2009
2010
2011
Diff BACI
Source : Mitaritonna et Traoré, 2017 Source: BACI et COMTRADE
2012
2013
2014
2015
2016
COMTRADE-BACI Comparison
Intra and extra regional trade – ECOWAS – Imports 2000-2016 Intra_ECOWAS trade (mlns USD) Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Comtrade 2.4 2.0 1.7 3.3 3.2 4.5 4.3 5.9 8.0 3.8 6.2 7.4 8.3 9.2 7.5 4.4 4.3
BACI 2.7 2.7 2.9 4.2 4.6 6.5 6.5 7.4 10.9 6.4 8.6 9.8 11.7 13.9 12.1 7.5 7.0
BACI/Comtrade 1.1 1.3 1.7 1.3 1.4 1.4 1.5 1.2 1.4 1.7 1.4 1.3 1.4 1.5 1.6 1.7 1.6
Source : Mitaritonna et Traoré, 2017 Source: BACI et COMTRADE
Extra_ECOWAS trade (mlns USD) Comtrade 10.9 14.1 15.8 25.1 10.4 16.2 38.9 54.6 55.9 54.9 72.1 94.6 71.1 81.5 73.6 24.8 62.8
BACI 23.1 26.1 28.8 35.6 35.4 43.0 56.7 74.5 92.0 87.7 101.1 131.1 110.5 118.1 121.9 100.2 91.2
BACI/Comtrade 2.1 1.8 1.8 1.4 • 3.4 2.7 • 1.5 1.4 • 1.6 1.6 1.4 1.4 1.6 1.4 1.7 4.0 1.5
- Larger Difference with extraECOWAS trade - Wrong declarations on both sides with intra - Average ratios : 1.90 versus 1.45
Still some limitations…
• Statistical treatment based on mirror flows: limited possibilities • Presence of informal flows • Trade in small quantities • Smuggling • Underinvoicing, misclassification
• Indirect methods of evaluation exist…, but they have their own limitations • National account approach (comparison: production, consumption and recorded trade) • Econometric approach: use of gravity models
What can be done?
• Direct methods of evaluation are needed – More or less complete surveys – Only a few in West Africa: CILSS, ECENE (2010, 2011), Benin and Nigeria – Very interesting survey in Uganda since 2005 (BoU+UBOS)
Studies measuring ICBT in Africa (Source : Bouët, Pace and Glauber, 2018)
Country / region at origin
Name of the initiative
Operated by
Funded by
Borders covered
Years covered
Products covered
Type of ICBT
Survey
Uganda
ICBT Survey
Uganda Bureau of Statistics
Bank of Uganda
Uganda with South Sudan, Congo RD, Rwanda, Tanzania and Kenya
2005-2017
All goods
ICBT Def. A
Survey
Rwanda
ICBT Survey in Rwanda
National Bank of Rwanda, National Institute of Statistics of Rwanda
Government of Rwanda
Rwanda with Burundi, DR Congo, Tanzania and Uganda
2009-2017
All goods
ICBT Def. A
Interviews
DR Congo
Timber Trade in Africa’s Great Lakes
South African Institute of International Affairs
na
DR Congo with Uganda, Burundi, Tanzania and Kenya (through Uganda)
November 2010February 2012
timber
ICBT Def. A
Survey
Benin
ECENE
Benin INSAE
Ministere du Developpement, de l'Analyse Economique et de la Prospective du Benin et Delegation de l'UE au Benin
Benin with Togo, Nigeria, Niger and Burkina
2010 / 2011
all merchandises
ICBT Def. B
Estimation
The Gambia/Senegal
Golub and Mbaye (2009)
World Bank
The World Bank, Agence Francaise du Dévelopement (AFD)
The Gambia with Senegal
2006
Agricultural (unprocessed and processed) goods
ICBT Def. B
Survey
Namibia
Informal CrossBorder Trade Survey (ICBTS)
Namibia Statistics Agency
Bank of Namibia - Namibia Ministry of Finance
Namibia with Angola, South Africa and Zambia
2014-2015
all merchandises
ICBT Def. A
Survey
Cameroon
Cameroon Ministry of Agriculture and Rural Development
Cameroon Ministry of Agriculture and Rural Development
Cameroon with Central African Republic, Chad, Gabon, Congo and Equatorial Guinea
2008
agricultural and horticultural commodities
ICBT Def. A
Type of study
Studies measuring ICBT in Africa (Source : Bouët, Pace and Glauber, 2018)
Type of study
Country / region at origin
Name of the initiative
Meta-Survey
Eastern Africa
East Africa Crossborder Trade
Meta-Survey
Eastern and Southern Africa
Informal Cross-border Trade In Eastern And Southern Africa
Survey
Southern Africa
Cross-border Food Trade Monitoring Initiative
Survey
Kenya
Informal Cross-border Trade Survey
Survey
Tanzania
Survey
Eastern and Southern Africa
Ackello-Ogutu (1996)
Operated by
Funded by
Data provided by the EAGC, USAID, FAO, FEWSNet, Ministère FEWS NET, FAO, NBR and des Affaires Étrangères de la WFP France
Borders covered
ICBT between Tanzania, Burundi, Rwanda, Uganda, Kenya, Somalia, Djibouti, Ethiopia, Sudan, and South Sudan and DRC
Years covered
Products covered
Type of ICBT
2010-2014
Staple food commodities: Maize grain, Rice grain, Maize and wheat flour, Beans and pulses, Cassava, Onions, Tomatoes, Live bovine animals, Milk and cream, Bovine meat, Fish and crustaceans
ICBT Def. A
USAID, ILRI, ReSAKSS
ICBT between Burundi, Democratic Republic of Congo, Djibouti, Ethiopia, Kenya, Malawi, Rwanda, Uganda, Tanzania, Zambia and South Sudan
ACTESA
ACTESA, WFP, FEWSNET
ICBT between Malawi, Mozambique, Zimbabwe and Zambia.
2011-2017
Food products
ICBT Def. A
KNBS
KNBS
Kenya’s ICBT with Uganda, Tanzania, Somali and Ethiopia
2nd quarter of 2011
all merchandise
ICBT Def. A
USAID Africa Bureau
USAID
Tanzania's ICBT with Kenya, Malawi, Zambia, DR Congo, and Uganda
1995-96
all merchandise
ICBT Def. A
USAID
Kenya-Uganda border; Tanzania with Malawi, Zambia, Congo RD and Uganda; Malawi with Mozambique, Tanzania and Zambia; Mozambique with South Africa, Swaziland, Malawi, Zimbabwe, Zambia and Tanzania
1994-95
Some agricultural food products and some manufactured goods
ICBT Def. A
secondary data on ICBT in ESA region collected by UBOS, EAGC, FEWS NET and ACTESA.
Technoserve, Kenya
Studies measuring ICBT in Africa (Source : Bouët, Pace and Glauber, 2018)
Type of study
Country / region at origin
Name of the initiative
Operated by
Econometrics of mirror trade data
Kenya, Mauritius and Nigeria
Bouet and Roy (2010)
IFPRI
Econometrics of mirror trade data
75 countries
Jean and Mitaritonna (2010)
CEPII
Econometrics of mirror trade data
Mozambique
Von Dunem and Arndt (2009)
Econometrics of mirror trade data
Kenya/Tanzania vs. Kenya/UK
Levin and Windell (2014)
Funded by
European Commission
SIDA-U-Forsk
Borders covered
Years covered
Products covered
Type of ICBT
Kenya, Mauritius and Nigeria
2001, 2004
all goods
ICBT Def. C
Burundi, Cameroon, Gabon, Kenya, Madagascar, Malawi, Mauritania, Mauritius, Morocco, Seychelles, Tanzania, Tunisia
2001, 2004
all goods
ICBT Def. C
Mozambique
2003
all goods
ICBT Def. C
Kenya/Tanzania
2000
all goods
ICBT Def. C
The CILSS approach
– Methodology: • Based on APEX organizations • Enumerators collect data on agri. trade along trade corridors and main marketplaces • Key partners: ACTOAH/CILSS – IFPRI - USAID
– CILSS has also collected data on ‘tracasseries administratives’ • road harassment by police, gendarmerie, city officials, sanitary inspectors, custom officers… • Data on bribes and wasted time
– Beyond the collection of data • More information on prices and quantities available for the private sector • Improving national and regional policies
The CILSS approach
• Advantages: • Daily flows • No need for time extrapolation • APEX organizations have an extensive knowledge of what happens • APEX organizations facilitate the collection of data by transporters
• Limitations : • All products are not covered • Limited collection concerning the West and East Basins • Limited collection concerning certain modes of transports (sea, laguna) • Systematic data quality control are being implemented
Comparison CILSS/COMTRADE
Bilateral Flows in 2016 for maize, in thousands of USD UN COMTRADE Burkina Benin
Ghana
CILSS Mali
0.54
Burkina
38.41
Cote d’iv. Ghana Mali Nigeria Togo
58.39
Total
59.68
36.609
81.27
Niger
Togo
2845.58
1.34
848.16
Burkina
3.947 13.77
Cote d’iv. Ghana Mali Nigeria Togo Total
134.32 36.61
Source: COMTRADE and CILSS Exporter in row - Importer in column
Benin
31.99
0.74 0.55 727.92
Burkina
3029.63
1.34
Ghana
Mali
1440.85
Niger
Togo
8.91 900.64
48.03
7761.08
45.88
3254.73 2204.47
24.48
4990.69
65.06 626.38
0.04
679.92
154.74
7579.98
1079.85
5285.00 5038.72
13746.43
45.92
Integrating CILSS data into official data: challenges to be met
• CILSS has already improved many aspects concerning collection techniques – Flows are now recorded according to international norms (HS-NTS nomenclature, choice of units)
– double accounting issue
• Extension of geographic and product coverage • Institutional arrangements between actors (CILSS-WAEMU-ECOWASINS-IFPRI)
Role of IFPRI
• Participation of private sector • Participation of IFPRI – Quality control – Documentation of databases – Communication activities on obstacles to regional trade – Evaluating the impact of road harassment – Identification at a disaggregated level of the location of major food security and nutrition bottlenecks
Conclusion
• Integration of CILSS data into Nat’l Statist’ Institutes databases in progress • CILSS-IFPRI-USAID: successful initiative
• Important for statistical issues but also for food safety • Can it be improved? And implemented at the continental level?