KEYNOTE ADDRESS: Achieving the Maputo Declaration Target and Prioritizing Public Agricultural Expenditures, presented by Godfrey Bahiigwa, ReSAKSS Africawide Coordinator, IFPRI at the 2013 ReSAKSS annual ReSAKSS conference

1. Complying with the
Maputo Declaration target:
Trends in public agricultural expenditures and
implications for pursuit of optimal allocation of
public agricultural spending
Godfrey Bahiigwa and Sam Benin, IFPRI
ReSAKSS Conference
Dakar, Senegal
12-13 November 2013

2. Background and Motivation

In 2003, African governments adopted the Comprehensive Africa
Agriculture Development Programme (CAADP) with two key targets
» Achieve 6% ag GDP growth rate per year
» Spend 10% of national expenditure on agriculture – Maputo
Declaration

Public spending (fiscal policy in general) in agriculture is a key
instrument for most developing country governments to achieve
national development objectives:
» Most of the poor work in the agriculture sector and in rural areas
» Sector employs 65% of the labor force and accounts for 32% of
GDP
» Evidence that public agriculture investment (particularly in R&D)
has large poverty-reducing effects
» Experience of Green Revolution (especially India and China)

3. Objectives and outline of presentation

Present patterns and trends in public agricultural
expenditure (PAE) in Africa

Assess progress in achieving the Maputo Declaration
target of spending 10% of national expenditures in
the agriculture sector

Draw implications of the Declaration on spending
behavior and optimal PAE allocation

Assess PAE data requirements for the joint agriculture
sector reviews (JSRs)

4. Progress at the Africa-wide level, 1995-2010
1995-2003
8
Percent
10
2003-2009
6
4
2
0
CA


EA
NA
SA
WA
Share of public agriculture expenditure (PAE) in total
expenditures for Africa as a whole declined in 200309 (post CAADP) compared to 1995-2003 (pre-CAADP)
Differences across different regions and countries

5. Progress in Central Africa, 2003-10
Central Africa
10
Sao Tome
& Principe
Equatorial
Guinea
Congo, Rep
.
Congo, De
m. Rep.
Cameroon
Burundi
0
Chad
5
Central
African…
Percent
15
Shares increased in Burundi, Rep of Congo, and São
Tomé and Principe
 Shares declined or stagnated in other countries, which
already spend less than 5%


6. 25
20
15
10
5
0
Uganda
Tanzania
Sudan
South…
Seychelles
Rwanda
Mauritius
Madagascar
Kenya
Ethiopia
East Africa
Djibouti
Percent
Progress in East Africa, 2003-10
Many countries in East Africa spend 5-10% percent of
total expenditure on agriculture
 Shares have increased over time in several countries
(especially Ethiopia, Rwanda, Sudan)


7. Progress in North Africa, 2003-10
Percent
15
North Africa
10
5

Tunisia
Morocco
Mauritania
Egypt
Algeria
0
Shares have stagnated in Mauritania and diverged
downwards from the 10% target in the other
countries

8. 30
25
20
15
10
5
0
Zimbabwe
Zambia
Swaziland
South
Africa
Namibia
Mozambiqu
e
Malawi
Lesotho
Botswana
Southern
Africa
Angola
Percent
Progress in Southern Africa, 2003-10
Malawi is an outstanding performer, with nearly
three times the target share in recent times
 Apart from Zambia, shares have stagnated or declined
in the other countries


9. 


Togo
Sierra…
Senegal
Nigeria
Niger
Mali
Liberia
Guinea-…
Guinea
Ghana
Gambia
Cote…
Cape Verde
25
20
15
10
5
0
West Africa
Burkina…
Percent
Progress in West Africa, 2003-10
Region where many countries have achieved target
Shares have increased in many countries
Burkina Faso and Mali (and Niger in recent times) have
consistently cut back on the shares to the target level

10. Summary of progress: and key questions (I)

Since 2003 when the declaration was made
» only 13 countries have surpassed the target in any
year—Burundi, Burkina Faso, Republic of
Congo, Ethiopia, Ghana, Guinea, Madagascar, Malawi,
Mali, Niger, Senegal, Zambia, and Zimbabwe
» only 7 have surpassed it in many years

Where the shares have been increasing or are high:
» Especially among countries in east and west Africa, is it
because they have observed positive returns or
because they think the 10% is optimal?

Where the shares have been declining:
» Especially in Burkina Faso, Mali and Niger where the
shares were higher than 10%, is it because they are not
getting the expected returns?

11. Summary of progress: and key questions (II)
» For middle income countries with other sources of
growth and development (esp. in north and southern
Africa), is it because the return from additional
spending in agriculture is lower than in the other
sectors?

Where the shares have stagnated:
» Is it because they have reached equilibrium, where
returns from additional spending in agriculture and
non-agriculture are equal?

These questions reflect the issue of the composition
of public agriculture expenditure (PAE):
» Role of government: large variation over time reflects
changing involvement of government in the sector
» Accounting problem: PAE depends of how PAE is
accounted for and reported in different countries

12. PAE trends reflect changing role of state (I)
Botswana
Egypt
Ethiopia
Ghana
Kenya
Malawi
Morocco
Tunisia
Uganda
Zambia
1980
9.7
4.4
6.9
12.2
8.4
10.2
6.5
14.5
32.5
13.4
PAE as percent of total expenditure
1985
1990
1995
9.8
6.5
6.0
4.2
5.4
5.0
9.9
6.9
9.1
6.2
6.1
5.1
10.4
6.0
5.5
8.4
11.1
11.1
5.0
5.0
4.2
8.3
8.5
8.3
3.9
2.2
2.9
10.7
5.6
2.5
2000
4.2
6.8
10.4
3.2
6.8
8.8
3.5
9.3
2.6
2.1
Compared to pre-structural adjustment periods, share of PAE has
declined substantially. Governments were directly involved in
agriculture production, cooperatives, marketing boards, etc.

13. Ghana
2010
2005
2000
1995
1990
1985
1980
2010
2005
2000
1995
1990
1985
14
12
10
8
6
4
2
0
1980
% of total expenditure
PAE trends reflect changing role of state (II)
Zambia

New form of direct governmental involvement in the sector in
recent times in form of heavy farm support subsidies

Issue is extent to which these programs have been refurbished to
take account of their negative experiences in the past

14. Percent of total expenditure
Accounting/Composition of PAE: Case of Ghana
12
Feeder roads
10
Debt service
PSI
8
Cocoa
6
Research
4
Fisheries
2
Forestry
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009


Crops &
livestock
Accounting changed over time. Is it merely to show compliance
with CAADP target? Or is effectiveness of portfolio considered?
Also reflects changing role of MDAs in the sector

15. How has PAE been allocated?

by sub-sector

by current and investment spending

by function

16. 100%
75%
50%
25%
0%
Congo, Rep.
CAR
Congo, D. R.
S. T. & Prin
Burundi
Chad
Djibouti
Seychelles
Uganda
Madagascar
Tanzania
Namibia
Malawi
Zambia
Lesotho
Swaziland
Senegal
Togo
Cote d'Ivoire
Sierra Leone
Mali
Percent of PAE
PAE by subsector, annual average 2003–2007
Central
Fishery
Eastern
Forestry
Southern
Western
Crops and livestock
Expenditures on crops and livestock dominate PAE.
 Share of PAE on forestry is higher in the central and eastern African
countries; not surprising, given dominance of forests there.
 Share on fisheries is higher in island countries and countries with large
coastlines.


17. 100%
75%
50%
25%
0%
Chad
Congo, Rep.
CAR
Congo, D. R.
Burundi
S. T. & Prin
Seychelles
Tanzania
Djibouti
Uganda
Madagascar
Namibia
Malawi
Swaziland
Lesotho
Zambia
Sierra Leone
Cote d'Ivoire
Togo
Senegal
Mali
Percent of PAE
PAE by current and investment spending, annual
average 2003–2007
Central
Eastern
Current
Southern
Capital
Western
Wide variation in share of PAE for investments—6% in Seychelles to
88% percent in Madagascar.
 Artifact of how different countries classify current expenditures and
investments. In many countries, all of the expenditures financed by
donors are classified as investment irrespective of what they are
actually spent on.


18. PAE by function, annual average 2006–2010
100%
Other
Inspection
Percent of PAE
80%
Marketing, storage, and public
stockholding
Feeder roads and other infrastructure
60%
40%
Irrigation
20%
Extension, training, technical
assistance
Research
0%
Subsidies
Burkina FasoKenya
Mali
Uganda Tanzania
Bulk of annual PAE was spent on subsidies.
 Share of PAE on research was moderate, at about 10–15
percent, although it was relatively low in Mali, at about 5 percent.
 Overall, the functional distribution of PAE seems to be more balanced in
Mali compared to the other four countries.


19. How has PAE trends contributed to growth?

Used simple correlations to assess co-trends
between PAE and agricultural GDP growth
rate

Results do not imply cause-effect
relationships, which require detailed PAE
and other data and advanced quantitative
methods beyond the scope of this report

20. 

Positive correlation
between PAE and
agGDP growth;
Different results for
different regions, with
largest correlations in
east Africa, which is a
top performer in both
indicators of PAE
Low significance using
aggregate PAE points
to the importance of
composition of PAE
agGDP growth rate (%)

20
15
10
5
0
-5 0
5
10
-10
-15
agGDP growth rate
(%)
PAE and agGDP
growth: Africa-wide
15
20
y = 0.1521x + 1.65
R² = 0.0179
PAE (% of total expenditures)
20
15
10
5
0
-20
-10 -5 0
10
-10
20
30
40
y = 0.0952x + 1.7053
R² = 0.0644
-15
PAE growth rate (%)

21. 

Results uphold common
knowledge that agR&D
investments take time to
manifest
Results (magnitude of
correlation, lags, and
statistical significance) are
different for different
regions
agGDP growth
rate (%)
Correlation is weak when
the data for all countries
are pooled in a single
estimation for Africa
West: Lag 3 years
-20
10
5
0
-15
-10
-5
-30
5
10
15
20
agR&Dexp growth rate (%)
West: Lag 6 years
10
8
6
4
2
-20
-10
y = 0.082x + 2.766
R² = 0.129
-40
0
-5
y = 0.047x + 2.673
R² = 0.015
agGDP growth
rate (%)

agGDP growth
rate (%)
agR&D exp and agGDP
growth: west Africa
0
-2 0
-4
10
20
30
agR&Dexp growth rate (%)
West: Lag 9 years
-30
y = 0.148x + 0.707
R² = 0.445
-20
-10
8
6
4
2
0
-2 0
-4
-6
10
20
30
agR&Dexp growth rate (%)

22. PAE data challenges and requirements for JSRs

To answer the questions posed earlier in a
comprehensive manner is very challenging; it is virtually
impossible with existing data for many countries

Some analysis on the efficiency and effectiveness of PAE
exists in a handful of countries only

We are faced with PAE measurement problems
» Most of data are at higher aggregate level
» Data systems reflect outlays associated with
organizational structures of governments rather than
objectives sought and functions performed
» Several PAE undertaken outside traditional ag MDAs

We need to do better data collection and management
for successful JSRs

23. NAIPs and implications for PAE data and analysis

NAIP budget allocated:
» Objectives and
programs
» Sub-sector
» Commodity and
commodity groups
» Economic use and
functions
» Target population
» Sources of financing

Need PAE data
accordingly: for
review, learning, and
further planning

24. % of NAIP budget by top 3 objectives/programs
100
Enabling
Environment
75
Science&Tech
50
25


Benin, 2010-2015
Burkina…
Burundi, 2012-2017
Cote…
Ethiopia, 2010-2020
Gambia, 2011-2015
Ghana, 2011-2015
Kenya, 2010-2015
Liberia, 2011-2015
Malawi, 2011-2014
Niger, 2010-2012
Nigeria, 2011-2014
Rwanda, 2009-2012
Senegal, 2011-2015
Sierra…
Tanzania, 2012-…
Togo, 2010-2015
Uganda, 2011-2015
0
NRM
Markets
Productivity
Food Security
Food and nutrition security and increasing productivity dominate
planned expenditures in many countries
Improving markets and sustainable NRM also take a large share

25. % of NAIP budget by sub-sector
100
Forestry
75
50
25
0
Fishery
Livestock
Crops
Benin,Burkina Faso, 2011-15Liberia, 2011-15 2011-15 2010-152010-15
2010-15 Cote d’Ivoire, 2010-15 Mali, Senegal, Togo,


Very few NAIPs had a breakdown by subsector, which is surprising
given that PAE is typically reported by subsector
In general, crops subsector dominates; share of other subsectors
depends on country context

26. % of NAIP budget by major commodities
Country, plan duration
Benin, 2010-15
Gambia, 2011-15
Malawi, 2011-14
Mali, 2011-15
Nigeria, 2011-14
Senegal, 2010-15

Commodities and budget allocation
Rice=24.9%, Corn=18.7%, Pineapple=4.2%,
Vegetables=4.1%
Rice=20.1%
Maize=37.2%
Rice=30.1%, Corn=12.7%, Millet/Sorghum=
7.2%
Cash crops=13%, Rice=2.8%
Groundnut=8.9%, Maize=8.6%,
Sorghum=4.5%, Cowpea=3.8%, Rice=1.4%
All the NAIPs identified specific commodities to lead overall
agricultural growth and development. Only few had specific
budget allocations. Maize and rice received the largest shares

27. % of NAIP budget by function
100
Farm Support
and Subsidies
80
60
NRM
40
0
Benin, 2010-2015
Burkina…
Burundi, 2012-…
Cote…
Ethiopia, 2010-…
Gambia, 2011-…
Ghana, 2011-2015
Kenya, 2010-2015
Liberia, 2011-2015
Malawi, 2011-…
Mali, 2011-2015
Niger, 2010-2012
Nigeria, 2011-2014
Rwanda, 2009-…
Senegal, 2011-…
Sierra…
Tanzania, 2012-…
Togo, 2010-2015
Uganda, 2011-…
20
Irrigation
Extension
Research
NRM and farm support and subsidies dominate planned
expenditures, followed by irrigation
 Research and extension are stated priorities with specific budget
allocations in a few countries only


28. % of NAIP budget by target population
Country, plan duration
Target population and budget allocation
Liberia, 2011-15
Women and youth=4.8%
Smallholder farmers=35.5%, Commercial
farmers=9.6%
Nigeria, 2011-14
Tanzania, 2012-16
Youth=48.8%, Men and
women=40.3%, Women=0.6%, Men=0.2%
Mainland=92.6%, Zanzibar=7.4%
Uganda, 2011-15
Northern region=2.4%
Senegal, 2010-15

Very few NAIPs had a breakdown of the budget by target
population, even though targeting and different target groups
were discussed in all of them

29. % of NAIP budget by source of funding
0%
20%
40%
60%
80%
100%
Others
Funding gap
Benin, 2010-15
Ethiopia, 2010-20
Gambia, 2011-15
Ghana, 2011-15
Kenya, 2010-15
Liberia, 2011-15
Malawi, 2011-14
Niger, 2010-12
Rwanda, 2009-12
Senegal, 2011-15
Togo, 2010-15
Government
Development partners
Dependence on external sources for financing the NAIPs
 Only in a couple of countries is government financing at least 50%
 More than 50% financing gaps in Benin, Gambia, Ghana, Senegal, Togo


30. Conclusions and Implications (I)

The amount of PAE in Africa as a whole increased
rapidly, but at a slower pace than the growth in total
expenditures resulting in a decline in the share of PAE
in total expenditures for Africa as a whole

Some governments’ reports on compliance with the
Maputo Declaration have generated controversy on
what to count as PAE
» resulting in a debate that may be polarizing behavior
around the fundamental issue of the investments
needed to achieve development results
» i.e. what types of investment, how much of each type
of investment, where should they be invested, and
when should they be invested

31. Conclusions and Implications (II)

Prioritization of investments has to be based on
analysis of the efficiency and effectiveness of
different types of public spending.
Therefore, disaggregation of public expenditure data
by type, across space, and over time is critical.
» Need public expenditure accounting and reporting
systems with unique codes or identifiers that also
reflect the objectives and functions that the outlays are
undertaken for (Kenya’s Open data on public
expenditure is a very good example).
» This is important for review of the NAIPs (as in JSRs)
» Will enhance the political accountability of government
to its citizens

32. Thank You