Sources of Growth in Sub-Saharan Africa

WP/04/176


Sources of Growth in Sub-Saharan Africa

Amor Tahari, Dhaneshwar Ghura,
Bernardin Akitoby, and Emmanuel Brou Aka

---

© 2004 International Monetary Fund
WP/04/176

IMF Working Paper

African Department

Sources of Growth in Sub-Saharan Africa

Prepared by Amor Tahari, Dhaneshwar Ghura, Bernardin Akitoby, and Emmanuel Brou Aka1

Authorized for distribution by Anupam Basu

September 2004

Abstract

This Working Paper should not be reported as representing the views of the IMF.
The views expressed in this Working Paper are those of the author(s) and do not necessarily represent
those of the IMF or IMF policy. Working Papers describe research in progress by the author(s) and are
published to elicit comments and to further debate.

Analysis of 1960-2002 data shows that average real GDP growth in sub-Saharan Africa was
low and decelerated continuously before starting to recover in the second part of the 1990s.
Growth was driven primarily by factor accumulation with little role for total factor
productivity (TFP) growth. The recent pickup in economic growth was accompanied by an
increase in TFP growth, namely in the group of countries whose IMF-supported programs
were judged to be on track. Average annual growth in the region, at 3½ percent during 1997-
2002, is less than half of the estimated growth needed to halve the fraction of population
living below $1 per day between 1990 and 2015, one of the Millennium Development Goals.


JEL Classification Numbers: O55, O47

Keywords: Sub-Saharan Africa; Growth Accounting; Growth Prospects

Author(s) E-Mail Address: atahari@imf.org; dghura@imf.org; bakitoby@imf.org




1 Emmanuel Brou Aka was a summer intern in the African Department of the Fund when this
paper was being prepared; he is a doctoral candidate at the Université d’Auvergne, Clermont-
Ferrand, France. The authors are grateful to colleagues in the African Department for useful
comments, and to Stéphanie Denis for invaluable research assistance.

---

- 2 -

Contents Page
I. Introduction ................................................................................................................................3
II. Growth Performance in Sub-Saharan Africa: Some Stylized Facts..........................................4
III. Growth Accounting..................................................................................................................6
IV. Growth Prospects and Challenges ...........................................................................................8
V. Some Concluding Remarks.....................................................................................................10

References....................................................................................................................................22

Tables
1. Sub-Saharan Africa: Subgroups of Countries......................................................................12
2. Sub-Saharan Africa: Real GDP Growth by Country, 1960-2002........................................13
3. Sub-Saharan Africa: Real GDP Growth by Subgroup of Countries, 1960-2002 ................14
4. Sub-Saharan Africa: Sectoral Contribution to Real GDP Growth by Subgroup
of Countries, 1960-2000 ..................................................................................................15
5. Sub-Saharan Africa: Sectoral Contribution to Real GDP Growth, Trade, Investment
and Saving by Subgroup of Countries and by Decades, 1960-2000................................16
6. Sub-Saharan Africa: Sources of Real GDP Growth by Country, 1960-2002......................17
7. Sub-Saharan Africa: Sources of Real GDP Growth by Subgroup of Countries,
1960-2002 ........................................................................................................................18
8. Sub-Saharan Africa: Sources of Real GDP Growth by Subgroup of Countries and
by Decades, 1960-2000....................................................................................................19
9. Sub-Saharan Africa: Sources of Real GDP Growth by Subgroup of Countries,
1990-2002 ........................................................................................................................20
10. Sources of Growth in the World by Region, 1960-2000 .....................................................21

Appendix I: Selected Literature Survey.......................................................................................22

---

- 3 -

I. INTRODUCTION
Poverty reduction is today a central objective of the IMF’s policy design and advice for low-
income countries (LICs), along with the institution’s more traditional focus on correcting
financial imbalances and promoting the development of productive resources and economic
growth. The issue of how to help low-income countries halve poverty by the year 2015
relative to 1990, one of the Millennium Development Goals (MDGs), has increasingly
become a focal point of the dialogue between these countries and the international donor
community. As demonstrated empirically by a number of recent papers,2 economic growth
plays a critical role in lowering poverty. This paper investigates the prospects of lowering
poverty in sub-Saharan Africa, by examining the sources of growth in individual countries
and drawing lessons for growth prospects.
The analysis of the sources of growth, using the growth accounting framework, has received
considerable attention for countries in East Asia.3 The related debate has centered on whether
the East Asian miracle was driven primarily by factor accumulation (capital and labor) or
total factor productivity (TFP). For sub-Saharan Africa, while a number of papers have
looked at the determinants of economic growth in the region,4 more recently, few have
analyzed the sources of growth from a growth accounting perspective. The bulk of the
analyses point to factor accumulation as the main source of growth in sub-Saharan Africa,
with the contribution of TFP growth playing a minor role.5 This paper examines the sources
of growth in sub-Saharan African countries, using the growth accounting framework and
extending the existing analysis both by country and time coverage. In addition, it compares
the sources of growth for subregions and subperiods, and examines the sectoral composition
of output.
Bosworth and Collins (2003) have argued that the growth accounting framework is a useful
tool to understand growth experiences across countries. The same authors have, however,
noted the limitations of this methodology. A key weakness relates to the interpretation that
the measured residual from the growth accounting exercise represents TFP growth. In
practice, in addition to providing a measure of gains in economic efficiency, the residual
may also reflect a number of other factors, including political disturbances and conflicts,
institutional changes, droughts, external shocks, changes in government policies, and

2 See, for example, Dollar and Kraay (2001); and Ghura, Leite, and Tsangarides (2002).
3 See, for example, Bosworth and Collins (2003); Collins and Bosworth (1996); Krugman
(1994); Sarrel (1997); and Young (1995).
4 For recent surveys on this subject, see Fosu (2001); and McPherson and Rakovski (2001).
5 Appendix I provides a brief survey of selected recent papers on economic growth and its
sources in sub-Saharan Africa.

---

- 4 -

measurement errors.6 This limitation is particularly important for sub-Saharan African
countries mired in conflicts and subject to significant drought-related and external shocks.
Also, the results from growth accounting exercise should not be misconstrued as providing
the fundamental causes of growth (rather than the proximate sources of growth).
Notwithstanding the limitations noted above, the paper’s main results are noteworthy. In
particular, average real GDP growth in the region during 1960-2002 was driven primarily by
factor accumulation with little or no role for TFP. The recent increase in growth (during
1997-2002, relative to 1990-96) was accompanied by a pickup in TFP growth, namely in
countries whose IMF-supported programs were adjudged to be on track, were part of the
group of CFA franc zone, and were in the middle-income category. Nonetheless, the average
annual growth performance in the region at 3½ percent during 1997-2002 is less than half of
the estimated growth needed to achieve the MDG of halving the fraction of population living
below $1 per day by 2015. Doubling the average annual growth rate requires a significant
boost in TFP growth, raising the investment-GDP ratio, and the resolution of conflicts.
The remainder of this paper is organized as follows: Section II provides some stylized facts
about growth performance in sub-Saharan Africa; Section III undertakes a growth accounting
analysis; Section IV examines growth prospects and challenges for the region; and Section
VI concludes the paper.
II. GROWTH PERFORMANCE IN SUB-SAHARAN AFRICA: SOME STYLIZED FACTS
This section provides some stylized facts about growth performance in sub-Saharan Africa
during 1960-2002, including for selected subgroups of countries (Table 1). Tables 2 and 3
report the growth performance in individual and subgroups of countries during 1960-2002,
and compares performance over two recent subperiods (1991-96 and 1997-2002); a number
of countries in the region have intensified their efforts to implement policy reforms during
1997-2002.
• The growth performance of the region as a whole was weak during 1960-2002,
with an average growth rate of 3.3 percent. This barely exceeds the average
population growth rate of about 3 percent and is less than half of what is needed
to achieve the MDG goal of halving the fraction of the population living below $1
per day.7

6 Another problem with the growth accounting framework is that it does not decompose
properly growth stemming from the exploitation of natural resources. If a new oil field or
diamond mine comes on stream it will tend to show up more as a boost in TFP than factor
accumulation.
7 The World Bank (2000) estimates that an annual average growth rate of about 7 percent,
and a better distribution of income, would be needed to achieve the MDG of halving the
incidence of severe poverty by 2015.

---

- 5 -

• Only four countries (Botswana, Equatorial Guinea, The Gambia, and Mauritius)
registered an average growth rate of at least 5 percent during 1960-2002. This
number had tripled in the period 1997-2002, with 12 countries registering an
average growth of at least 5 percent.
• Turning to the subgroups of countries, middle-income countries (MICs)
experienced the fastest real GDP growth (an annual average of 4.8 percent), and
the countries mired in conflicts had the weakest growth performance (an annual
average of 2.4 percent), especially during the subperiod 1991-96.
• Most country subgroups, with the exception of countries whose programs were
adjudged to be “off track,” experienced a boost in growth during 1997-2002
(compared with 1991-96).
• The performance of countries whose programs with the IMF were adjudged to be
“on track” improved during 1997-2002 (compared with 1991-96): real GDP
growth in this group of countries increased significantly to an annual average of
4.1 percent during 1997-2002 from 1.5 percent during 1991-96).
Tables 4 and 5 provide the results of an examination of the sectoral contribution to overall
economic growth.
• Following a pattern seen in low-income countries in the world, the share of the
tertiary sector (which includes public and private services) was the largest in total
value added.
• The share of the secondary sector in total value added was the lowest in all
subgroups, with the exception of the middle-income and oil-producing countries.
• Over time, the share of the tertiary sector remained broadly stable, but the share of
the primary sector declined and that of the secondary sector rose, particularly
driven by the group of oil-producing and middle-income countries.
• The tertiary sector contributed the most to real GDP growth (about 50 percent) in
sub-Saharan Africa as a group; this was also the case for the group of all LICs in
the world. The same pattern was observed in the subgroups of countries, with the
exception of the oil-producing countries, where the contribution of the secondary
sector (which includes the oil sector) was the largest.
• On average, growth performance in sub-Saharan Africa deteriorated almost
continuously from the 1960s to the mid-1990s. This was accompanied by the
significant decline in the contribution of agriculture in the 1970s (compared with
the 1960s) and the decline in the contribution of the tertiary sector in the 1980s
and 1990s.
• Real GDP growth was highly correlated with both the secondary sector’s
contribution to growth (correlation of 0.9) and the tertiary sector’s contribution to

---

- 6 -

growth (correlation of 0.8), whereas the correlation with the primary sector’s
contribution was weak.
Tables 4 and 5 also provide some stylized facts on trade, saving, and investment.
• The average share of trade (exports plus imports) in GDP was relatively high in
sub-Saharan Africa during 1960-2002, compared with the group of LICs in the
world. This was driven largely by the performance of middle-income and, over
time, oil-producing countries.
• The average investment-GDP ratio in sub-Saharan Africa during 1960-2002 (at
about 21 percent) was lower than the group of LICs in the world (about
25 percent).
• Moreover, the annual average domestic saving-GDP ratio in sub-Saharan Africa
at about 11½ percent during 1960-2000 was weaker than the annual average of
LICs (about 17 percent). The performance of domestic saving in sub-Saharan
Africa deteriorated significantly over time for virtually all the country groups. In
the group of LICs in the world, the domestic saving-GDP ratio actually increased
over time.
III. GROWTH ACCOUNTING
A standard growth-accounting framework, based on a standard Cobb-Douglas production
function, is used for the analysis:

α
−α
Y =
1
A K L









(1)
t
t
t
t
where Y is output, A is total factor productivity (TFP), K represents physical capital, L is
labor, α is the elasticity of output with respect to the capital input (the value of α is
between 0 and 1), and t is a time index. Given the paucity of the relevant data, the labor
force series is not adjusted for human capital stock; thus in this formulation, educational
attainment is part of TFP. Taking natural logarithm of the production function, and
differentiating with respect to time, gives:
Y&
A&
K&
L&

=
+ α
+ 1
( −α ) ,






(2)
Y
A
K
L
where the dotted variables denote time derivatives.
The data were spliced from the series obtained from the IMF World Economic
Outlook (WEO) and the World Bank World Development Indicators databases. The time
coverage is 1960-2002. Output is measured by the GDP at constant prices (expressed in local
currency units). The labor force is proxied by data on the economically active population.
The capital stock series are constructed using the perpetual inventory methodology. It is
assumed that the initial capital-output ratio in 1960 was 1.5 and the depreciation rate was set
at 6 percent. In line with the literature on production function estimates for developing

---

- 7 -

countries,8 the share of capital (α ) is set at 0.4; the findings are robust to an alternative value
of α of 0.3.
The results from growth accounting exercises for individual countries and for subgroups of
countries over the period 1960-2002 are provided in Tables 6-9:
• The average TFP growth for sub-Saharan Africa as a whole was nil during
1960-2002, contributing to the poor overall growth performance of the region.
This result is consistent with the findings of Bosworth and Collins (2003).
• Real GDP growth was driven largely by factor accumulation.
• Almost half of the countries in the region experienced declines in TFP on average
during 1960-2002. Only five countries (Botswana, Equatorial Guinea, Mauritius,
Swaziland, and Uganda) experienced an average TFP growth of more than
1 percent during 1960-2002.
• During the period 1981-2000, the contribution of both physical capital and TFP
declined significantly, reflecting the poor performance of countries in conflict.
The latter experienced an average TFP decline of 0.8 percent during this period,
and a decline of 2.1 percent on average during 1991-2000.9
• In the 1960s, TFP growth contributed on average to about 30 percent of output
growth in sub-Saharan Africa, but its contribution became negative in the
subsequent decades. Only the subgroups of countries categorized as middle-
income, oil-producing, nonconflict, and CFA franc countries experienced positive
TFP growth on average during 1991-2000.
• An evaluation of economic growth during 1990-2002 shows that average growth
in sub-Saharan Africa rose from 2.1 percent during 1990-96 to 3.6 percent during
1997-2002, driven by a significant increase in TFP growth. To a large extent this
result was propelled by the improved performance in countries whose programs
with the Fund were adjudged to be on track, CFA franc countries, and increased
oil production in Equatorial Guinea.
The improved performance of the CFA franc countries during 1997-2002
most likely reflects the positive impact of the 1994 devaluation on
tradables and complementary structural reforms undertaken by several of

8 Bosworth, Collins, and Chen (1995) provide empirical support for this value.
Senhadji (2000) finds a mean value of 0.43 for group of countries from sub-Saharan Africa.
9 The large decline in TFP could also reflect the “destruction” of human and physical capital
stemming from conflicts.

---

- 8 -

these countries in the postdevaluation period.10 Besides the exchange rate
change, by and large, the governments’ programs in these countries
consisted of (i) internal adjustment measures, including fiscal tightening
and the implementation of structural reforms related to the reorganization
and downsizing of the civil service, (ii) privatization of public enterprises,
(iii) bank restructuring, and (iv) liberalization of domestic prices at the
national level and of interest rates at the regional level.
The increased TFP growth during 1997-2002 in countries whose program
was on track likely reflected the efficiency gains from the implementation
of macroeconomic and structural reforms.
Bosworth and Collins (2003) provide a comparison of the growth performance of various
subregions in the world during 1960-2000 (Table 10). Their results for sub-Saharan Africa,
using a much smaller set of countries and adjusting for educational attainment, are
comparable to those of the current paper. In addition, their results indicate that growth in
the region was the slowest of all other regions in the world, driven largely by the poor record
of TFP growth.
IV. GROWTH PROSPECTS AND CHALLENGES
Although the recent pickup in economic growth has been encouraging, the region has a long
way to go to make up for the ground lost over the past three decades and to catch up with the
growth rates of LICs in other regions. In particular, economic growth rates are still not high
enough to enable sub-Saharan African countries to make a real dent in the pervasive poverty.
There is thus a need to raise substantially real GDP growth rates on a sustained basis, both
through the productivity channel and by raising investment. The rest of this section draws on
the work of the existing empirical studies to draw some lessons on how this can be achieved.
Using the empirical literature on the determinants of TFP growth as a guide,11 it is argued
that the factors that positively influence TFP growth include:
• Good quality institutions. Better institutions include those that deliver better law
and order, better bureaucratic quality, less corruption, lower risks of
expropriation, and lower probability of government repudiation of contracts
(Bosworth and Collins, 2003).12

10 However, over the last two years, performance in a number of these countries has been
adversely affected by the crisis in Côte d’Ivoire.
11 See Berthélemy and Söderling (2001); Bosworth and Collins (2003); and Senhadji (2000).
12 See also Rodrik, Subramanian, and Trebbi (2002) for an analysis of the role of institutions
in economic development.

---

- 9 -

• Human capital development. It will be important to increase the quantity and
quality of basic health care, education, and other high-priority services, with a
view to improving social indicators appreciably over the longer term.
• Favorable macroeconomic policy environment. Following Senhadji (2000), the
elements of such an environment include lower levels of external debt and
government consumption, and higher levels of international reserves.
• Diversification of the economic base. The paper by Berthélemy and Söderling
(2001) shows that the countries in sub-Saharan Africa that have been successful at
diversifying their economic base from agriculture to secondary and tertiary
sectors have experienced higher TFP growth. As noted in Section II above, in the
sample of the countries in this paper, real GDP growth was highly correlated with
both the secondary sector’s contribution to growth (correlation of 0.9) and the
tertiary sector’s contribution to growth (correlation of 0.8), whereas the
correlation with the primary sector’s contribution was weak.
Successful economic diversification will depend critically on decisive steps on structural
reforms including privatization, financial sector reform, and trade liberalization. Calamitsis,
Basu, and Ghura (1999) have noted the need to accelerate the restructuring and privatization
of public enterprises, in order to reduce reliance on budgetary subsidies and transfers, expand
the scope for private sector activity, and promote overall economic efficiency and growth.
In addition, financial sector reform needs to be deepened.13 In this regard, steps need to be
taken to ensure the independence and full accountability of central banks; deepen and
broaden financial markets; establish or strengthen the institutions responsible for the
prudential regulation and supervision of banks; open the banking sectors to healthy
competition and international best practices in bank management; and strengthen the legal
framework for banking activities. Trade liberalization can also contribute to the acceleration
of TFP growth by promoting the competitiveness of domestic producers and speeding up
sub-Saharan Africa’s integration into the global economy.
Nonetheless, boosting TFP alone will be insufficient to raise economic growth rates to levels
required to significantly lower poverty. It will also be essential to boost investment to levels
comparable to other developing countries in the world. Over the period 1991-2000, the
average investment-GDP ratio and the real GDP growth rate was at 18½ percent and
1.8 percent, respectively, in the LICs in sub-Saharan Africa, compared with about 24 percent
and 3 percent in all LICs in the world. The international donor community has a role to
play by raising the level of financial support to boost government investment. The role of
government will need to be focused on the effective delivery of essential public services and
basic infrastructure, as well as the promotion of human resource and social development. As
regards private investment, recent empirical work suggests that it can be boosted by increases
in government investment, a stable macroeconomic environment, financial deepening, and

13 For a survey on the experience with financial sector reforms in sub-Saharan Africa, see
Reinhart and Tokatlidis (2003).

---