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, Senegal12-13 November 2013

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)

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)

Progress at the Africa-wide level, 1995-2010

Share of public agriculture expenditure (PAE) in total expenditures for Africa as a whole declined in 2003-09 (post CAADP) compared to 1995-2003 (pre-CAADP)

Differences across different regions and countries

CA EA NA SA WAAfrica Region

0

2

4

6

8

10 1995-2003

2003-2009

Per

cen

t

Progress in Central Africa, 2003-10

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%

Bur

undi

Cam

eroo

n

Cen

tral

Af.

..

Cha

d

Con

go, D

e...

Con

go, R

ep.

Equ

ator

ial.

..

Sao

Tom

e ...

0

5

10

15 Central AfricaP

erce

nt

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)

Dji

bout

i

Eth

iopi

a

Ken

ya

Mad

agas

car

Mau

riti

us

Rw

anda

Sey

chel

les

Sou

th S

udan

Sud

an

Tan

zani

a

Uga

nda0

510152025

East AfricaP

erce

nt

Shares have stagnated in Mauritania and diverged downwards from the 10% target in the other countries

Alg

eria

Egy

pt

Mau

rita

nia

Mor

occo

Tun

isia

0

5

10

15North AfricaP

erce

nt

Progress in North 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

Ang

ola

Bot

swan

a

Les

otho

Mal

awi

Moz

ambi

que

Nam

ibia

Sou

th A

fric

a

Sw

azil

and

Zam

bia

Zim

babw

e05

1015202530

Southern Africa

Per

cen

t

Progress in Southern 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

Bur

kina

Fas

o

Cap

e V

erde

Cot

e d'

Ivoi

re

Gam

bia

Gha

na

Gui

nea

Gui

nea-

Bi.

..

Lib

eria

Mal

i

Nig

er

Nig

eria

Sen

egal

Sie

rra

Leo

ne

Tog

o0

10

20

West AfricaP

erce

nt

Progress in West Africa, 2003-10

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?

» 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

Summary of progress: and key questions (II)

PAE trends reflect changing role of state (I)PAE as percent of total expenditure

1980 1985 1990 1995 2000Botswana 9.7 9.8 6.5 6.0 4.2Egypt 4.4 4.2 5.4 5.0 6.8Ethiopia 6.9 9.9 6.9 9.1 10.4Ghana 12.2 6.2 6.1 5.1 3.2Kenya 8.4 10.4 6.0 5.5 6.8Malawi 10.2 8.4 11.1 11.1 8.8Morocco 6.5 5.0 5.0 4.2 3.5Tunisia 14.5 8.3 8.5 8.3 9.3Uganda 32.5 3.9 2.2 2.9 2.6Zambia 13.4 10.7 5.6 2.5 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.

PAE trends reflect changing role of state (II)

1980

1985

1990

1995

2000

2005

2010

1980

1985

1990

1995

2000

2005

2010

Ghana Zambia

-1

4

9

14

% o

f to

tal

exp

end

i-tu

re

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

Accounting/Composition of PAE: Case of Ghana

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

2000 2001 2002 2003 2004 2005 2006 2007 2008 20090

2

4

6

8

10

12 Feeder roads

Debt service

PSI

Cocoa

Research

Fisheries

Forestry

Crops & livestock

Perc

ent o

f tot

al e

xpen

ditu

re

How has PAE been allocated?

by sub-sector

by current and investment spending

by function

PAE by subsector, annual average 2003–2007

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.

Cong

o, R

ep.

CAR

Cong

o, D

. R.

S. T

. & P

rinBu

rund

iCh

adD

jibou

tiSe

yche

lles

Uga

nda

Mad

agas

car

Tanz

ania

Nam

ibia

Mal

awi

Zam

bia

Leso

tho

Swaz

iland

Sene

gal

Togo

Cote

d'Iv

oire

Sier

ra L

eone

Mal

i

Central Eastern Southern Western

0%25%50%75%

100%

Fishery Forestry Crops and livestock

Perc

ent o

f PAE

PAE by current and investment spending, annual average 2003–2007

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.

Chad

Cong

o, R

ep.

CAR

Cong

o, D

. R.

Buru

ndi

S. T

. & P

rinSe

yche

lles

Tanz

ania

Djib

outi

Uga

nda

Mad

agas

car

Nam

ibia

Mal

awi

Swaz

iland

Leso

tho

Zam

bia

Sier

ra L

eone

Cote

d'Iv

oire

Togo

Sene

gal

Mal

i

Central Eastern Southern Western

0%

50%

100%

Current Capital

Perc

ent o

f PAE

PAE by function, annual average 2006–2010

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.

0%

20%

40%

60%

80%

100%Other

Inspection

Marketing, storage, and public stockholding

Feeder roads and other infrastructure

Irrigation

Extension, training, technical as-sistance

Research

Subsidies

Perc

ent o

f PAE

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

PAE and agGDP growth: Africa-wide

0 2 4 6 8 10 12 14 16 18 20

-15

-10

-5

0

5

10

15

20

f(x) = 0.152130393953939 x + 1.65002381628054R² = 0.0178841231285043

PAE (% of total expenditures)

agG

DP

gro

wth

rat

e (%

)

-20 -10 0 10 20 30 40

-15

-10

-5

0

5

10

15

20

f(x) = 0.0952498009220817 x + 1.70527465405226R² = 0.0644324268719985

PAE growth rate (%)

agG

DP

gro

wth

rat

e (%

)

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

-20 -15 -10 -5 0 5 10 15 20-5

0

5

10

f(x) = 0.0478456700137788 x + 2.67348731247432R² = 0.0150199687452059

West: Lag 3 years

agR&Dexp growth rate (%)

agG

DP

gro

wth

ra

te (

%)

-30 -20 -10 0 10 20 30

-5

0

5

10

f(x) = 0.0826292689842183 x + 2.76599612010094R² = 0.129578814302894

West: Lag 6 years

agR&Dexp growth rate (%)

agG

DP

gro

wth

ra

te (

%)

-40 -30 -20 -10 0 10 20 30-5

0

5

10

f(x) = 0.148652394727877 x + 0.707175149604037R² = 0.445503708349219

West: Lag 9 years

agR&Dexp growth rate (%)

agG

DP

gro

wth

ra

te (

%)

agR&D exp and agGDP growth: west Africa

Correlation is weak when the data for all countries are pooled in a single estimation for Africa

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

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

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

% of NAIP budget by top 3 objectives/programs Be

nin,

201

0-20

15Bu

rkin

a Fa

so, 2

011-

...Bu

rund

i, 20

12-2

017

Cote

d'Iv

oire

, 201

0-...

Ethi

opia

, 201

0-20

20G

ambi

a, 2

011-

2015

Gha

na, 2

011-

2015

Keny

a, 2

010-

2015

Libe

ria, 2

011-

2015

Mal

awi,

2011

-201

4N

iger

, 201

0-20

12N

iger

ia, 2

011-

2014

Rwan

da, 2

009-

2012

Sene

gal,

2011

-201

5Si

erra

Leo

ne, 2

010-

2014

Tanz

ania

, 201

2-20

16To

go, 2

010-

2015

Uga

nda,

201

1-20

15

0

25

50

75

100Enabling En-vironment

Science&Tech

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

Benin, 2010-15

Burkina Faso,

2011-15

Cote d’Ivoire, 2010-15

Liberia, 2011-15

Mali, 2011-15

Senegal, 2010-15

Togo, 2010-15

0

25

50

75

100

Forestry

Fishery

Livestock

Crops

% of NAIP budget by sub-sector

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

% of NAIP budget by major commodities

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

Country, plan duration Commodities and budget allocation

Benin, 2010-15 Rice=24.9%, Corn=18.7%, Pineapple=4.2%, Vegetables=4.1%

Gambia, 2011-15 Rice=20.1%Malawi, 2011-14 Maize=37.2%

Mali, 2011-15 Rice=30.1%, Corn=12.7%, Millet/Sorghum=7.2%

Nigeria, 2011-14 Cash crops=13%, Rice=2.8%

Senegal, 2010-15 Groundnut=8.9%, Maize=8.6%, Sorghum=4.5%, Cowpea=3.8%, Rice=1.4%

Ben

in, 2

010-

2015

Bur

kina

Fas

o, 2

011.

..

Bur

undi

, 201

2-20

17

Cot

e d'

Ivoi

re, 2

010.

..

Eth

iopi

a, 2

010-

2020

Gam

bia,

201

1-20

15

Gha

na, 2

011-

2015

Ken

ya, 2

010-

2015

Lib

eria

, 201

1-20

15

Mal

awi,

2011

-201

4

Mal

i, 20

11-2

015

Nig

er, 2

010-

2012

Nig

eria

, 201

1-20

14

Rw

anda

, 200

9-20

12

Sene

gal,

2011

-201

5

Sier

ra L

eone

, 201

0...

Tan

zani

a, 2

012-

2016

Tog

o, 2

010-

2015

Uga

nda,

201

1-20

15

0

20

40

60

80

100

Farm Support and Subsidies

NRM

Irrigation

Extension

Research

% of NAIP budget by function

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

Country, plan duration Target population and budget allocation

Liberia, 2011-15 Women and youth=4.8%

Nigeria, 2011-14 Smallholder farmers=35.5%, Commercial farmers=9.6%

Senegal, 2010-15 Youth=48.8%, Men and women=40.3%, Women=0.6%, Men=0.2%

Tanzania, 2012-16 Mainland=92.6%, Zanzibar=7.4%

Uganda, 2011-15 Northern region=2.4%

% of NAIP budget by target population

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

Benin, 2010-15Ethiopia, 2010-20Gambia, 2011-15

Ghana, 2011-15Kenya, 2010-15

Liberia, 2011-15Malawi, 2011-14

Niger, 2010-12Rwanda, 2009-12Senegal, 2011-15

Togo, 2010-15

0% 20% 40% 60% 80% 100%

Government Development partners Others Funding gap

% of NAIP budget by source of funding

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

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

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

Thank You