By Gert-Jan Stads. Presented at the ASTI-FARA conference Agricultural R&D: Investing in Africa's Future: Analyzing Trends, Challenges, and Opportunities - Accra, Ghana on December 5-7, 2011. http://www.asti.cgiar.org/2011conf

1. Africa’s Agricultural R&D
Funding Rollercoaster
An Analysis of the Elements of Funding Volatility
Gert-Jan Stads
5–7 December 2011, Accra, Ghana

2. Background: Trends in Agricultural R&D
Investment in Sub-Saharan Africa
Source: Beintema and Stads 2011
 Investments (and human capacity) in agricultural R&D increased by
more than 20% during 2000–08.
 Most of this growth was driven by just a handful of countries (mainly
following boosts in salaries and rehabilitation of infrastructure).
 In many other countries (particularly in francophone West Africa),
investments have declined since 2000.

3. Investment challenge: Underinvestment
Source: Beintema and Stads 2011
 NEPAD target: Allocation of at least 1 % of GDP to R&D
 In 2008, Africa spent $0.61 for every $100 of AgGDP on agricultural R&D.
 Despite an overall increase in recent years, Africa is widely underinvesting
in agricultural R&D.

4. Trends in Agricultural R&D spending
in the “Big Eight” since 2008
30
Change 2008-2010 (%)
20
10
0
-10
-20
-30

5. Investment challenge: Volatility
FASTEN YOUR SEAT
BELT
Keep arms and legs inside vehicle at
all times

7. Severe fluctuations in annual agricultural
R&D investment levels, 1981–2008
Burkina Faso Niger
8 40 8 35
Billion 2005 CFA francs
Billion 2005 CFA francs
Million 2005 PPP dollars
Million 2005 PPP dollars
6 30 6 26
4 20 4 18
2 10 2 9
0 0 0 0
1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008
1.5
South Africa 415 1.0
Gabon
3.9
Billion 2005 CFA francs
Million 2005 PPP dollars
Million 2005 PPP dollars
Billion 2005 rand
1.2 332 0.8 3.1
0.9 249 0.6 2.3
0.6 166 0.4 1.5
0.3 83 0.2 0.8
0.0 0 0.0 0.0
1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 1991 1993 1995 1997 1999 2001 2003 2005 2007

8. Economic Theory on Volatility
 Increased macroeconomic volatility has a negative impact on
economic growth, or is at least closely associated with slower
growth (Aghion et al. 2005; Fatás and Mihov 2006; Hnatkovska
and Loayza 2004; Perry 2009).
 Aid flows in developing countries are more volatile than
government revenues, household consumption, or gross domestic
product (GDP), and aid volatility tends to reinforce
macroeconomic instability and slow down economic growth (Bulíř
and Hamann 2003; Desai and Kharas 2010; Fielding and Mavrotas
2008).
 No literature was found on R&D funding volatility in developing
countries.

9. Why is Stable Agricultural
R&D Funding Important?
 Agricultural R&D investment is positively associated with high
returns, but these returns take time—commonly decades—to
develop.
 Consequently, the inherent lag from the inception of research to
the adoption of a new technology or the introduction of a new
variety calls for sustained and stable R&D funding.
 Severe fluctuations in annual agricultural R&D funding
exacerbate uncertainty at the institute level and renders long-
term R&D budget, staffing, and planning decisions more difficult.
 Therefore, the continuity of research programs is imperiled in
the short run, as is the release of new varieties and technologies
in the long run.

10. Volatility coefficient of
agricultural R&D spending
Growth in agricultural R&D spending (gs) was expressed as follows:
������������
������������ = ln ������������−1
s=1,…, N,
where s is agricultural R&D spending (in constant prices), and t represents the year.
A country’s volatility coefficient (V) of agricultural R&D expenditures
was calculated by taking the standard deviation of growth in annual
agricultural R&D spending:
1 ������ 2, 1 ������
V= ������=1 ������������ − ������ where ������ = ������=1 ������������ .
������ ������

11. Volatility in African
agricultural R&D spending
0.12
(Asia–Pacific 1992–2002)
0.21 >
(SSA 2001–2008)
0.14
(Latin America 2004–2006)
0.09
(SSA agricultural output,
2001–2008)

12. Volatility coefficient
0.0
0.1
0.2
0.3
0.4
0.5
Mauritania
Gabon
Tanzania
Burkina Faso very high
Ethiopia
Namibia
Gambia, The
Mali
Côte d'Ivoire
high
Calculated from Beintema and Stads (2011)
Sierra Leone
Eritrea
Guinea
Sudan
Togo
Nigeria
Burundi
Botswana
Benin
Senegal
Zambia
Uganda
moderate
Kenya
Cross-Country Variation
Ghana
Niger
Volatility Coefficients 2001–08
Mauritius
Madagascar
South Africa
Malawi
low
Congo, Rep.

13. Volatility and Country Groupings
 Agricultural R&D spending in low-income countries
(0.23) is on average more volatile than spending in
middle-income countries (0.16)
 Average volatility was higher in West (0.23) and East
(0.22) Africa than in Southern Africa (0.14)
 Spending at NARS with less than 100 FTEs (0.24) is on
average more volatile than spending at NARS with more
than 100 FTEs (0.19)
 AgR&D expenditures in countries spending less than
0.5% of AgGDP on AgR&D (0.23) are on average more
volatile than those in countries spending more than
1.0% of AgGDP on AgR&D (0.16)

14. Volatility of agricultural R&D
spending across cost categories
Salaries
Operating costs
Capital investments
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Volatility coefficient

15. Funding sources for agricultural R&D
 National government funding: either through direct allocations
or competitive funding schemes
 Donors and development banks: high donor dependency in
low-income countries worldwide
 Production or export levies (mostly on export crops):
e.g. cocoa in Ghana; tea in Tanzania and Kenya; sugarcane in Mauritius, etc.
 Sale of goods and services: e.g. on-demand research for private
companies

16. Benin (INRAB)
Botswana (DAR)
Burkina Faso (INERA, IRSAT, CNSF) Government
Burundi (ISABU) Donors
Côte d'Ivoire (CNRA) Producer organizations
Eritrea (NARI)
Own income
Gambia, The (NARI)
Guinea (IRAG)
Other
Kenya (see footnote)
Madagascar (FOFIFA) Source:
Beintema and Stads (2011)
Mali (IER)
Mauritania (CNERV, CNRADA)
Mauritius (FARC, MSIRI)
Mozambique (IIAM, IIP)
Namibia (DRT)
Niger (INRAN)
Rwanda (ISAR)
Senegal (ISRA, ITA)
Sierra Leone (SLARI)
South Africa (ARC)
Sudan (ARC)
Tanzania (DRD)
Togo (ITRA)
Uganda (NARO)
Zambia (ZARI)
0 20 40 60 80 100 Share of total funding (%)

17. Drivers of Funding Volatility
in African Agricultural R&D
Government
Sale of goods and services
Donors and development banks
Total
Indicates that in many
cases shocks in one 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
funding source are to
Volatility coefficient
some extent absorbed
by reverse shocks in
other funding sources

18. Donor dependency and funding volatility
Average and spreadShare of funding as a % of
of donor Volatility
100
total agriculturaldonorfunding, 2001–08
R&D funding coefficient
<10% 0.19
80
>10% 0.28
>40% 0.31
Share of donor funding in
total annual funding (%)
60
40
20
0

19. Funding sources and cost categories for DRD
(Tanzania) and INERA (Burkina Faso), 2001–08
40 40
DRD – cost categories DRD – funding sources
Million 2005 PPP dollars
Million 2005 PPP dollars
30 30
20 20
10 10
0 0
2001 2002 2003 2004 2005 2006 2007 2008 2001 2002 2003 2004 2005 2006 2007 2008
Salaries Operational Capital Government Donors, development banks, SROs
Producer organizations Sales of goods and services
30 30
INERA – cost categories INERA – funding sources
Million 2005 PPP dollars
Million 2005 PPP dollars
20 20
10 10
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Salaries Operational Capital
Government Donors, development banks, SROs Sales of goods and services

20. Concluding Remarks:
Putting a Halt to Volatility
 Agricultural R&D spending in SSA has been far from stable in recent years.
 Problem is more pronounced in donor-dependent low-income countries.
 Halting excessive volatility in yearly agricultural R&D investment levels
requires a long-term commitment from national governments, donors and
development banks, as well as the private sector.
 Stable and sustainable levels of government funding are key, not just to
secure salaries (which are fundamentally important), but also to enable
necessary nonsalary expenditures.
 Donor and development bank funding needs to be better aligned with
national priorities, and consistency and complementarities among donor
programs need to be assured.
 Mitigating the effects of any single donor’s abrupt change in aid
disbursement is crucial. Need for greater funding diversification (e.g. through
the sale of goods and services or private sector funding).

21. Thank you
Will Africa’s bumpy
rollercoaster ride end here?
2013
2012
2011