3. Structural Transformation in Sub-Saharan Africa

International Monetary Fund. African Dept.
Published Date:
October 2012
  • ShareShare
Show Summary Details

Introduction and Summary

This chapter analyzes the extent to which structural transformation, defined as the shift of workers from low to high average productivity activities and sectors, occurred across countries in sub-Saharan Africa from 1995 to 2010. This has been a period of high growth for a significant number of countries in sub-Saharan Africa, and this chapter finds that most have experienced some degree of structural transformation, albeit at different speeds and following different paths. But while the Asian countries against which sub-Saharan African countries are benchmarked in this chapter largely transformed through low-wage manufacturing, it is not obvious this will be the path for most sub-Saharan African countries. Depending on resource endowments, labor skills, and other factors, some sub-Saharan African countries may follow the Asian path through low-wage manufacturing, whereas others may transform through services, and still others through the transformation of their agricultural sector.

Many countries in sub-Saharan Africa have experienced unprecedented high, sustained economic growth over the last two decades. This chapter analyzes the extent to which this period of high growth has been accompanied by structural transformation, defined as the shift of workers from activities and sectors with low average labor productivity to those with high average labor productivity, thus contributing to an increase in average labor productivity for the overall economy.1 This chapter’s analysis does, indeed, find that for most countries in the region, growth has been accompanied by some degree of structural transformation, although there has been significant variation in the speed and type of transformation. Most oil exporters, and middle-income and non-fragile low-income countries, have seen sustained increases in average labor productivity, often underpinned by rising productivity in agriculture and resulting in a declining share of GDP from that sector. Fragile countries, in contrast, have generally experienced low and irregular growth, largely as a result of conflict; this poor growth is reflected in the absence of significant structural transformation in most of these fragile countries.

The main findings of the chapter are as follows:

  • While data are limited and not always consistent, it appears agricultural labor productivity growth has been very weak among many low-income countries in sub-Saharan Africa, attributable to poor irrigation, lack of fertilizer, shrinking land use, and infrastructure constraints, although some exceptions exist (Burkina Faso, Malawi).

  • The manufacturing sector is developing in a few countries (Ethiopia, Kenya, Mozambique, and Tanzania), but starting from a low base with slow productivity.

  • The services sector has been more successful in stimulating output, exports, and labor productivity growth, and some countries have moved up the value-added chain (Kenya, Mauritius). Informality in sub-Saharan Africa economies, however, remains pervasive.

The analysis in this chapter takes special note of the agricultural sector, which remains the largest employer in many countries in sub-Saharan Africa.

In sub-Saharan Africa, agriculture accounts, on average, for over half of total employment and one-fifth of gross domestic product (GDP) (Figure 3.1).2

Figure 3.1.Sub-Saharan Africa: Agriculture’s Shares in GDP and Total Employment and Real GDP per capita, 2010


Source: World Bank, World Development Indicators.

Output per worker in the agriculture sector is lower than the nonagriculture in GDP per worker in every country in the global sample of 86 countries considered in Figure 3.2. More important, the difference in average labor productivity between the agricultural sector and the rest of the economy is greater among low-income countries, including the subsample of 23 countries from sub-Saharan Africa (shown in red), than among high-income countries.3

Figure 3.2.Average Labor Productivity

Source: Restuccia, Yang, and Zhu (2008), based on data from the Penn World Table (PWT) and the Food and Agriculture Organization (FAO).

Much of the empirical evidence from other regions suggests rapid productivity growth is often accompanied by declining shares of agriculture in GDP and employment, reflecting the migration of resources toward higher-productivity activities in other sectors. The evidence also shows a concurrent increase in the productivity of labor in agriculture, which releases labor to the rest of the economy. The combination of these shifts could be consistent with increasing agricultural output. Thus, this chapter looks both at the shift of resources out of agriculture and at the evolution of productivity in agriculture among sub-Saharan African countries.

This chapter concludes that increasing agricultural productivity is an essential element of structural transformation in sub-Saharan Africa. Moreover, countries need to continue to emphasize increasing the quality and coverage of education to facilitate the shift of labor into higher value-added industries and services, and sustained improvement in logistics and infrastructure should help the expansion of activities in all sectors.

The rest of this chapter provides a variety of analyses. Section II presents some stylized facts about the extent of structural transformation in sub-Saharan African countries. It also benchmarks that transformation in selected sub-Saharan African countries against a sample of Asian countries. Section III deepens the analysis to include developments in output share and labor productivity in sub-Saharan African countries by sector (agriculture, mining and manufacturing, and services). Section IV briefly discusses some barriers to transformation in sub-Saharan Africa, and Section V concludes. An Appendix provides examples of successful structural transformation in five sub-Saharan African countries.

Stylized Facts

Due to data constraints, as well as the late start of growth and transformation in sub-Saharan Africa, this chapter focuses on a relatively short period (1995 to 2010), combining agricultural data from the Food and Agricultural Organization (FAO) of the United Nations, GDP data by sector based on a survey of the IMF’s African Department country teams, and data on employment by sector from household surveys.4 The following analysis of sub-Saharan Africa’s structural transformation must be interpreted with caution, because the output data are of uneven quality, and few countries have long and consistent time series for employment.

Figure 3.3 underscores the heterogeneity of the growth experience in sub-Saharan Africa.5 Twelve countries in the region have experienced labor productivity declines while, at the other end of the spectrum, 13 have seen productivity grow at rates that would more than double labor productivity in one generation. The rest of sub-Saharan African countries fall between these two extremes, with positive but relatively slower productivity growth. Table 3.1 shows aggregate labor productivity growth, average labor productivity growth in the agricultural sector, and the change in the share of agriculture in GDP (non-resource GDP for resource-abundant countries, shown in italics) over the past decade and a half. Countries are grouped using the Regional Economic Outlook classification (see Statistical Appendix), and within each category they are ranked by their average labor productivity growth. Table 3.2 adds information on the changes in the sectoral composition of output, using the same country groupings. Tables 3.1 and 3.2 reveal several patterns across country groups, which are discussed below. The group of low-income fragile countries shows evidence of limited structural change, with nine out of the 12 experiencing negative overall average labor productivity growth. Given data limitations, the fragile countries are not discussed further in the text.

Figure 3.3.Growth Rate of Real GDP per Person in the Labor Force, 1995–2010

Source: IMF staff estimates.

Table 3.1.Structural Transformation in Sub-Saharan African Countries: Productivity Growth and Change in Agricultural Output Shares, 1995–2010 1(Percent)
Growth rate of real

GDP per capita
Growth rate of real

GDP per person in

the labor force
Annual percent

change in

agricultural output

Average agricultural

productivity growth

(year if not 1995-2010)
Oil-exporting countries
Equatorial Guinea17.1216.86−5.69−0.90
Cameroon1.460.85−0.14−1.60 (1995-2007)
Congo, Rep.1.210.66−0.903.00
Middle-income countries
Cape Verde5.083.36−2.913.90
Mauritius3.462.98−3.341.20 (1995-2009)
Ghana2.782.64−1.061.50 (1998-2005)
Botswana3.492.32−3.562.20 (2001-2010)
Zambia2.082.28−2.664.70 (1998-2010)
Senegal1.350.97−1.404.00 (2001-2005)
South Africa1.580.80−1.167.10 (2000-2009)
Low-income and fragile countries
    Low-income excluding fragile countries
Mozambique4.834.75−2.454.10 (2003-2009)
Uganda3.593.91−3.49−3.50 (2002-2009)
Rwanda3.773.19−0.943.40 (1996-2005)
Tanzania3.203.10−2.072.90 (2000-2009)
Burkina Faso3.152.91−1.411.80
Mali2.372.10−0.053.90 (2001-2009)
Sierra Leone2.051.941.683.00
Gambia, The1.521.150.824.90
    Fragile countries
São Tome and Principe0.801.80−0.232.00
Côte d’Ivoire−0.23−0.510.681.90
Central African Republic−0.47−0.750.652.10
Congo, Dem. Rep. of−0.82−0.98−2.85−2.00
Sources: IMF staff calculations; IMF, African Department database; World Bank, World Development Indicators.

The period for which the change in agriculture’s share of GDP is calculated does not always coincide with 1995–2010 because of lack of data for some countries. Table 3.2 presents the years in parentheses.

Note: Italicized names represent resource-abundant countries, whose sector shares are reported in terms of non-resource GDP, and thus, do not add up to 100 percent.
Sources: IMF staff calculations; IMF, African Department database; World Bank, World Development Indicators.

The period for which the change in agriculture’s share of GDP is calculated does not always coincide with 1995–2010 because of lack of data for some countries. Table 3.2 presents the years in parentheses.

Note: Italicized names represent resource-abundant countries, whose sector shares are reported in terms of non-resource GDP, and thus, do not add up to 100 percent.
Table 3.2.Structural Transformation in Sub-Saharan African Countries: Output Shares of Agriculture, Mining, Manufacturing, Construction, and the Tertiary Sector, 1995–2010
InitialFinalAnnual percent

change in

agricultural GDP

Oil-exporting countries
Equatorial Guinea65.39.6−5.732.355.610.332.43.943.621.011.5
Angola (2000-2010)24.019.0−2.147.841.37.711.88.215.359.953.6
Chad (2005-2010)22.823.10.332.722.
Cameroon (2000-2010)25.124.7−
Congo, Rep. (2002-2008)18.217.2−0.937.828.312.
Middle-income countries
Cape Verde12.87.2−
Mauritius (2000-2010)5.53.7−
Ghana (2000-2010)32.629.2−
Zambia (2000-2010)19.614.4−2.76.810.312.010.65.613.659.458.7
South Africa2.92.4−
Seychelles (2004-2010)3.92.4−
Low-income and fragile countries
Low-income excluding fragile countries
Uganda (2000-2010)24.616.0−
Ethiopia (2000-2010)49.441.0−
Tanzania (1998-2010)31.823.9−
Burkina Faso36.328.6−
Sierra Leone (2001-2010)
Malawi (2002-2010)33.327.2−
Gambia, The (2004-2010)28.730.
Kenya (2000-2010)29.424.4−
Fragile countries
Liberia (2000-2010)66.442.2−
Guinea (2000-2010)21.323.
São Tome and Principe (2001-2010)15.214.9−
Côte d’Ivoire22.825.
Togo (2000-2010)37.644.
Central African Republic (2000-2010)48.351.
Congo, Dem. Rep. (2000-2010)55.839.9−2.910.414.
Guinea-Bissau (2000-2010)41.543.
Eritrea (2000-2010)
Sources: IMF, staff calculations; African department database; World Bank World Development Indicators; and Food and Agriculture of the United Nations (FAOSTAT).Notes: The parentheses next to each country’s name show the years over which the change in agriculture’s share of GDP was calculated. Italicized names represent resource-abundant countries, whose sector shares are reported in terms of non-resource GDP, and thus, do not add up to 100 percent.
Sources: IMF, staff calculations; African department database; World Bank World Development Indicators; and Food and Agriculture of the United Nations (FAOSTAT).Notes: The parentheses next to each country’s name show the years over which the change in agriculture’s share of GDP was calculated. Italicized names represent resource-abundant countries, whose sector shares are reported in terms of non-resource GDP, and thus, do not add up to 100 percent.

Performance by Country Groupings

Oil exporters

Most oil-exporting countries experienced some form of structural transformation, albeit at different rates. Except for Chad, all countries registered a decline in agriculture as a share of non-resource GDP while, aside from Cameroon and Equatorial Guinea, all countries experienced positive labor productivity growth in the agricultural sector.

Within this group, differences in labor productivity growth may be linked partly to variations in the trajectories of oil revenue and exports, through spillovers into the non-oil sector. The three countries that experienced the fastest increase in average labor productivity—Angola, Equatorial Guinea, and Chad—were the ones that experienced the fastest increase in oil production (in the same order). Oil output in Angola increased from around 650 thousand barrels per day (bpd) in 1995 to nearly 2 million bpd in 2010 (an 8 percent annual growth rate), while in Equatorial Guinea it increased from 5 thousand bpd to 320 thousand bpd in the same period (an annual growth rate of 32 percent). In Chad, oil production began only in 2003, growing rapidly to around 125 thousand bpd by 2010. In these three countries, growth has been accompanied by a rise in the output share of construction, underscoring the role of the oil sector as a potential enabler of development in other sectors. By contrast, Cameroon and Gabon, with some of the lowest increases in productivity, saw oil output decline.

Oil production alone does not explain the evolution of average labor productivity. As noted, most countries in this group show significant increases in productivity in agriculture, and some drivers of performance are idiosyncratic and have at most an indirect and remote link to oil. For example, Angola’s emergence in 2002 from protracted civil strife may help explain its high growth rate of average labor productivity in the aggregate and in the agricultural sector, as this growth contains an important “catch up” peace dividend.

Middle-income countries

At a very aggregate level, the middle-income countries (MICs) display the typical pattern of structural transformation. All experienced positive growth in output per worker, and all, except for Seychelles, experienced positive average labor productivity growth in the agricultural sector, as well as a declining share of agriculture in total GDP. In addition, many countries in this group had very high shares of services in output at the start of the period of analysis, increasing further over time, in some cases at the expense of the secondary sector, and not just of the agricultural sector.

Low-income countries (excluding fragile countries)

Almost every country in the non-fragile low-income group experienced some structural transformation, with cross-country differences likely related to variations in education levels. Real output per worker rose in every country but Madagascar and Niger, and every country experienced positive agricultural productivity growth, except Madagascar and Uganda. Low-income countries (LICs) in East and Southern Africa experienced higher average labor productivity growth than countries in West Africa, and this is also true for the change in the agricultural output share. Part of the improved relative performance could be related to the level of education, as the population age 25 years and above among non-fragile LICs in West Africa had 2.2 years of schooling versus 4.3 years in those in East and Southern Africa (2010).6

Education may also have played a part in helping MIC newcomers rise from the ranks of LICs. MIC newcomers have on average higher levels of education than non-fragile LICs (with the exception of Kenya), measured either by average years of schooling or the percentage of the labor force with some secondary or tertiary education. The three newcomers in the MIC group had an average of 6 years of schooling in 2010, up from 5 years in 1995 (Figure 3.4). In contrast, a subsample of 10 non-fragile low-income countries for which data are available (Benin, The Gambia, Kenya, Malawi, Mali, Mozambique, Niger, Rwanda, Uganda, and Tanzania) averaged 3.4 years of schooling in 2010, up from 2.4 years in 1995.

Figure 3.4.Average Year of Total Schooling and Highest Level of Education Attained

Source: Barro and Lee (2011).

Benchmarking Sub-Saharan Africa’s Transformation

To assess sub-Saharan Africa’s performance, Figure 3.5 compares a subset of those countries (Cameroon, Ghana, Mauritius, South Africa, and Tanzania) with the median estimate of a group of low- and middle-income Asian economies that have experienced rapid structural transformation and started their growth takeoffs with similar or lower levels of GDP per capita than the average sub-Saharan African country at the time (Box 3.1). The choice of sub-Saharan countries was determined by the availability of employment data and the desirability of including countries from each grouping. Interesting findings include the following:

Figure 3.5.Sub-Saharan Africa: Sectoral Output and Employment, 1995–2010

Sources: CEIC database; and Haver Analytics database.

  • Ghana and Tanzania have experienced declines in agricultural output and employment shares over time, with Tanzania matching the experience of the comparator Asian economies quite closely.

  • Most middle-income countries have experienced declining manufacturing ratios for the past two decades, consistent with the process in more advanced Asian countries, where services play an ever-increasing role in the economy.

  • Relatively few LICs in sub-Saharan Africa (Mozambique, Tanzania) have been able to raise their manufacturing output and employment shares on a sustained basis.7

Among the sub-Saharan African countries, the upward output and employment trends in the services sector have been stronger than in the Asian economies, suggesting that the path to transformation is likely to take place at least partly through services.

Box 3.1.Structural Transformation in Asia

In all the Asian countries considered, the share of agriculture in output and employment has fallen over time, with most countries experiencing fairly strong labor productivity growth. With workers shifting out of the agricultural sector, labor productivity growth has been quite strong in most cases, ranging from 1 percent in the Philippines to 3–3½ percent in Indonesia and Vietnam. Indeed, one of the main features of the structural transformation process in Asia is its “green revolution.”

Manufacturing output growth has been extremely strong over the past two decades in Bangladesh, Cambodia, and Vietnam, while services output growth in agriculture has been the dominant factor in the more advanced economies (Indonesia, Philippines) and India. In manufacturing, output growth has mainly concentrated in apparel, clothing, and footwear production, mostly for export. The output surge has coincided with modest increases in labor productivity in manufacturing, given the labor intensity of the products. In services, finance and business-process outsourcing have played dominant roles in many countries.

Change in Output and Employment Shares, 1990–2011(Annual, percent)
Output sharesEmployment shares
Cambodia, 1998-2010−−
Indonesia, 1993-2011−0.2−0.10.3−
Source: Haver Analytics; and IMF staff calculations.
Source: Haver Analytics; and IMF staff calculations.

The low wage level relative to productivity is one of the main factors behind this significant structural transformation in Asia. Using data from the World Bank’s Enterprise Surveys, Clarke (2012) has shown that the median labor cost is 28 percent of value-added per worker among East Asia manufacturing firms, compared with 34 percent among low-income and lower-middle-income countries in sub-Saharan Africa.

This box was prepared by Alun Thomas.

An alternative presentation of these data show broad increases in employment in sectors with higher average labor productivity. Figures 3.6 and 3.7 plot annual changes in employment shares against average relative productivity levels for agriculture, manufacturing, mining, and the tertiary sector for sub-Saharan African and Asian countries, respectively.8 Points in the lower left quadrant indicate sectors with below average productivity and declining employment shares, while those in the upper right quadrant indicate sectors with above average productivity and rising employment shares. Interestingly, the picture is quite similar for both regions, with workers moving out of agriculture, where relative productivity levels are low, into the tertiary and manufacturing sectors.

Figure 3.6.Sub-Saharan Africa: Labor Productivity and Change in Employment Shares, 1995–2010

Source: IMF staff calculations.

Figure 3.7.Asia: Labor Productivity and Change in Employment Shares, 1995–2010

Source: IMF staff calculations.

Indeed, the only sub-Saharan African countries with little change, or even an increase, in the agriculture employment share are Cameroon and Zambia.

In all the other sub-Saharan African countries, workers have moved out of the agricultural sector. This suggests that the findings of McMillan and Rodrick (2011), whose analysis of a smaller subset of countries through 2005 found workers moving into low-productivity sectors, may actually be reversed when considering a broader set of countries through 2010.

Sectoral Perspectives

As noted above, structural transformation involves shifting from low value-added activities and sectors to higher value-added activities. This section analyzes the extent to which such structural transformation is taking place in sub-Saharan Africa by evaluating developments at the sectoral level.


As a whole, sub-Saharan Africa has experienced slow growth in average labor productivity in the agricultural sector in the past three decades, indicating that little structural transformation has taken place in this sector. The annual growth rate of average agricultural productivity in sub-Saharan Africa was virtually unchanged in the mid-1990s, at around 0.1 percent to 0.2 percent, and has continued lagging other regions (Figure 3.8). Nevertheless, the regional average masks a wide variation in experiences, including productivity declines in many countries in sub-Saharan Africa and high productivity growth (and presumably structural transformation) in others.9 As Table 3.1 shows, the rate of productivity growth in agriculture is highly correlated with the growth of GDP per capita.

Figure 3.8.Average Labor Productivity in the Agriculture Sector, 1985–2010

Source: World Bank, World Development Indicators.

Fertilizer use and the extension of agriculture to non-arable land are key factors in explaining the lower agriculture labor productivity levels in sub-Saharan Africa compared to Asia. Arable land has been declining by 1 percent a year in both regions. In contrast, non-arable agricultural land has been rising by 1 percent a year in Asia while it has declined by a similar magnitude among sub-Saharan African countries. With respect to fertilizer, while its use has increased by 3.5 percent a year among Asian countries, on average it has not risen in sub-Saharan Africa.

Many governments in sub-Saharan Africa have been concerned with improving agricultural productivity, with some positive results. The Comprehensive Africa Agriculture Development Program (CAADP), created in 2003, focuses attention on a basic strategy to improve agricultural productivity in sub-Saharan Africa involving improved water control, increased use of fertilizer, and development and distribution of better seeds (UNDP, 2012).10 A discussion in this chapter’s appendix of the recent experience with cotton sector production and productivity in Burkina Faso shows how such policies can generate impressive results.

Manufacturing and mining11

Examples exist of strong productivity growth in manufacturing in sub-Saharan Africa, although this is generally restricted to MICs. Among the upper-middle-income sub-Saharan African countries at a more advanced stage of structural transformation, growth rates of productivity in the secondary sector are comparable to those of Asian countries (Figure 3.9). A few low-income sub-Saharan African countries have experienced impressive growth in textiles and leather products—albeit starting from a very low base. Exports of textiles have more than doubled over the past two years in Ethiopia, and leather exports have rebounded, but manufacturing exports account for only 1 percent of output.

Figure 3.9.Manufacturing Productivity Growth vs. Real GDP Per Capita

Source: IMF staff estimates based on household survey data.

Exports of light manufactures have also doubled in Kenya in the past decade, but they still represent only 2 percent of GDP. In general, however, among low-income and lower-middle-income countries (Mozambique, Senegal, Uganda) the productivity growth rates are negative and their magnitude significant. This could be related to data quality, because reported manufacturing employment levels at the beginning of the period in these countries are very low and may reflect poor data, which may have improved over time.

Notwithstanding weak productivity growth, a few low-income sub-Saharan African countries have performed strongly in terms of manufacturing and mining output and exports. In a combined sub-Saharan Africa and Asian sample, a 10 percent increase in manufacturing and mining exports in relation to output is associated with an increase in the corresponding output share of about 2 percent (Figure 3.10).12 The sub-Saharan African countries that have registered an increase in the manufacturing and mining output share in the past 15 years are Madagascar, Mozambique, Namibia, Tanzania, Uganda, and Zambia. This chapter’s appendix provides a discussion of how this was accomplished in two cases: Tanzania, where the result was largely due to renewed macroeconomic stability and a general liberalization of the economy; and Namibia, where a focus on business-oriented infrastructure and support contributed to rising manufacturing.

Figure 3.10.Change in Manufacturing and Mining Output and Export Shares, 1995–2010

Source: IMF staff estimates based on household survey data.

The weaker performance among sub-Saharan African countries in manufacturing may reflect both factor endowments and the timing of growth. At the time of their transformation, most Asian economies had limited land and natural resources and abundant skilled and low-wage workers, and they faced a global marketplace where manufacturing was dominated largely by high-wage industrial countries. So their path to transformation—to higher value-added output—was through low-wage manufacturing.

Sub-Saharan Africa today faces a very different situation. The countries of sub-Saharan Africa generally are land and natural resource abundant, and face a global marketplace where low-wage economies already dominate many markets. So, the market conditions and resource endowments in many sub-Saharan African countries may favor a type of transformation in which the natural resources, agricultural, and services sectors play a more dominant role, with a lesser role for manufacturing.13

While the value of mining output (including petroleum) has increased rapidly in a number of resource-rich countries in the past decade, mining shares have fallen in real terms in some places, such as Angola. The exceptions are Chad, the Republic of Congo, and Zambia. There has also been limited experience with the processing of minerals. In the case of oil, part of the difficulty lies in the small size of the national markets, with a number of countries having developed refineries that could serve the regional market.14


Even though services already play an important role in many sub-Saharan African countries, there are encouraging signs that they could be developed further. Services sector productivity growth among many LICs has been very strong in recent years, higher than industry productivity growth among LICs and far higher than productivity growth in developed countries (Ghani, Grover, and Kharas, 2011). This suggests there may still be room for the sector to grow, especially as it remains heavily concentrated in low value-added informal services such as retail trade. Supporting this view, services sector productivity growth in some MICs (Botswana, Mauritius, South Africa) has been comparable to, if not higher than, middle-income Asian countries in the past decade (Indonesia, Philippines) and generally stronger than in manufacturing. Moreover, productivity increases in a few countries have also been strong (Mali, Mozambique, Senegal), supporting the view that a number of sub-Saharan African countries are possibly following a developmental process that favors services provision over manufacturing.

A few low- and middle-income sub-Saharan African countries are restructuring their economies around services, following the pattern of India. The discussions in this chapter’s appendix of the services sectors in Mauritius and Kenya show that information and communications technology services have grown rapidly in these economies in recent years. In Mauritius, government policies supported this development through strong protection of property rights, flexible labor market regulations, and a simplified tax regime, among other measures. In Kenya, the process has been facilitated by the deepening regional integration of the East African Community (EAC). In addition, one of the major attractive features of these economies is the quality of their labor forces compared to the rest of sub-Saharan Africa. The share of students who have reached the secondary level of education in 2010 is highest in Mauritius at more than 50 percent, while the share of students with tertiary education is comparable between Mauritius and India, and considerably above the sub-Saharan Africa median level in Kenya (Table 3.3).

Table 3.3.Education Characteristics, 2010(Percent of population age 15 and over)
Highest level attained



SSA median32.922.61.95

Among many oil exporters and other resource-based economies, the change in the services sector output share is also considerably larger than the change in the services export share (Figure 3.11). This could reflect the presence of backward linkages with resource extraction, large public sectors, or, most likely a combination of both. Transportation output has grown fastest among the resource exporters: Angola and Nigeria (above 12 percent a year) and Botswana (8–9 percent a year). Public administration also has grown faster among resource-rich countries, at more than 5 percent a year (Botswana, Namibia, Nigeria) compared to 3 percent a year for the rest (median).15

Figure 3.11.Change in Service Output and Export Shares, 1995–2010

Source: IMF staff estimates based on household survey data.

A large proportion of the services jobs in sub-Saharan Africa are in the informal sector, thus making the recent experience of strong services productivity growth in low-income sub-Saharan Africa economies all the more impressive, particularly given the extent of low value-added services such as retail trade (Box 3.2). It does, however, raise the question of whether the current productivity surge can be sustained, although examples such as Kenya and Mauritius show that the switch into higher value-added services can be achieved.

Some Impediments to Structural Transformation

The analysis above indicates that structural transformation in sub-Saharan Africa, while taking place, is doing so at a slow pace. This raises the question of why, despite high growth, transformation has been slow. While a comprehensive answer to that question is beyond the scope of this chapter, this section briefly assesses some key factors retarding progress in the transformation of sub-Saharan African economies.16

The principal factors cited by the private sector as hindering development in sub-Saharan African countries include the cost of power outages, the lack of transportation infrastructure, and the degree of corruption. Table 3.4 presents a comparison of some of these structural impediments for sub-Saharan Africa as a whole and for a few comparator Asian countries, which, while different in some respects, may still provide lessons for sub-Saharan Africa.

Table 3.4.Structural Impediments(2009, unless otherwise noted)
Road Density

(Kilometers of road

per 100 square

kilometers land area,

Cost to Export

(U.S. dollars per

Logistics Performance

Index, Efficiency of

Customs Clearance

(1=low, 5=high)
Value Lost to

Electrical Outages

(Percent of sales)
ICRG Corruption

(0=high risk,

6=low risk)

Accountability, and

Integrity of the Public


(1=low, 6=high)
South East Asia (median)35.07742.
Sub-Saharan Africa114.51,9272.
Sources: World Bank, World Development Indicators; International Country Risk Guide; and IDA Resource Allocation Index.

Road density refers to 2004, and electricity outages to 2007.

Sources: World Bank, World Development Indicators; International Country Risk Guide; and IDA Resource Allocation Index.

Road density refers to 2004, and electricity outages to 2007.

The table shows that the cost of power outages (in terms of sales) is considerably higher among sub-Saharan African countries than in the sample of Asian countries. As for transportation infrastructure, road density and logistical performance are relatively low in sub-Saharan Africa, contributing to the higher cost of exporting goods from sub-Saharan African countries compared to other countries: US$1,974 per container, compared to a median estimate of US$732 for the Asian countries. It also shows that countries in sub-Saharan Africa rank slightly worse than the median country in the Asian sample in the measures of corruption and transparency.

Independently, both an irregular power supply and corruption could hinder production and exports in all three sectors, while high transport costs and low road density could significantly constrain the development of manufacturing and agriculture. But it is the cumulative impact of dealing with all three factors that makes the business environment particularly challenging in parts of sub-Saharan Africa.

Conclusions and Future Prospects

Looking at the 1995–2010 period, rapid economic growth in many countries in sub-Saharan Africa has been accompanied by only limited structural transformation. In particular, the speed of transformation has been slower than that experienced in Asia. Rather than seeing labor shift rapidly from agriculture to manufacturing, many African countries have experienced relatively slow productivity growth in agriculture and a less-pronounced shift from agriculture to services and, to an even lesser extent, manufacturing.

So far, the slower rate of structural transformation would not seem to have held back economic growth in some of sub-Saharan Africa’s most rapidly growing countries. The question is whether it will become an impediment to continued high and sustained growth in the coming decades. Experience both within and outside sub-Saharan Africa suggests that raising agricultural productivity will need to be an important part of the answer. Because the employment share of agriculture in most economies in sub-Saharan Africa is very large and relative productivity levels are so low, there is significant opportunity to raise living standards through agricultural productivity improvements.

Policymakers should and, in some cases, are focusing their efforts on productivity increases in agriculture. While this may be consistent with a steady or rising agricultural output share, if there is to be significant structural transformation in sub-Saharan Africa, the employment share of agriculture will almost certainly decline.

Whether or not rising productivity in agriculture will be accompanied in all countries by an expansion of manufacturing is an open question. The comparative advantage of many countries may lie elsewhere, for example in services. The challenge here is to raise education standards and narrow the infrastructure gap to levels at which the region can compete for higher value-added services. This is already happening in some countries, but a stronger push is needed.

Looking further ahead, the continued rapid growth in real wage levels in Asia, and sub-Saharan Africa’s demographic dividend, with dependency ratios projected to decline, suggest that manufacturing in sub-Saharan Africa could become increasingly competitive. The challenge for sub-Saharan Africa will be to take full advantage of these developments, which will require investments in both human and physical capital. But the aim is clear: to harness Africa’s resources to provide high and sustained growth, raise living standards, and create employment opportunities. Structural transformation will need to be a key part of this process.

Box 3.2:Formal and Informal Employment

Informal workers (including self-employed workers) are the dominant feature of labor markets in developing countries. These workers have no labor contracts and no health or social security benefits. On average, they work longer hours, and their earnings per hour are generally lower than in the formal sector.

Available estimates of formal sector employment in relation to total employment suggest that its share is very low in sub-Saharan Africa. The data are generally based on household surveys conducted infrequently; and definitions of formal employment can vary from study to study, making comparisons difficult. Mindful of these drawbacks, the table (below) provides the latest estimates of formal employment. For low-income sub-Saharan African countries, formal employment covers public sector employment and wage employment in other sectors. For the more advanced countries, formal employment covers workers with an employment relationship that is subject to national labor legislation, income taxation, and social protection.

Formal employment in low-income sub-Saharan African countries is low, ranging from around 7 percent of total employment (Tanzania, Uganda) to about 27 percent (Rwanda). At the other end of the spectrum, formal employment is fairly high in Mauritius and South Africa, ranging from 37 to 51 percent of total employment, with the Indonesia estimate sitting between these figures. Formal employment ratios are positively related to levels of GDP per capita.

Table 1.Employment Indicators(Ratio of total employment, except where stated)
YearSectoral Employment SharesFormal Sector

Advanced stage of transformation
South Africa20075.716.669.036.8
Source: Household surveys.

Latest estimate in percent of working-age population.

Source: Household surveys.

Latest estimate in percent of working-age population.

This box was prepared by Alun Thomas.
Appendix: Selected Country Examples

This appendix highlights some of the successful examples of structural transformation in sub-Saharan economies. The examples were chosen to illustrate a variety of dimensions:

Burkina Faso is a country that has succeeded in raising average agricultural productivity levels in recent years; Tanzania has succeeded in developing its manufacturing sector; Namibia has successfully diversified away from non-renewable natural resources; while Mauritius and Kenya provide examples of countries that have developed a successful service sector.

These experiences may provide useful insights for other economies.

Burkina Faso: Agricultural Productivity Growth1

Cotton is the dominant agricultural product in Burkina Faso: 250,000 households and more than 3 million people (almost 19 percent of the population) depend directly or indirectly on its cultivation for their livelihood. The development of cotton cultivation spurs the development of maize (up by more than 50 percent since 1994), which is planted in rotation with cotton. Until the start of gold mining in 2008, cotton accounted for more than 50 percent of export earnings.

Agricultural production grew at an average rate of almost 4 percent a year between 1995 and 2009, reflecting labor and land productivity gains, and a significant increase in the area of cultivated land, the latter development contrasting with many other sub-Saharan African countries. Until 2006, many of the gains in agricultural productivity were realized in the seed cotton sector, which has grown on average by 7.2 percent a year since 1995. The success of the cotton sector was an important factor in the doubling of real per capita GDP in Burkina Faso between 1995 and 2006.

Following setbacks from 2006 to 2009, ginned cotton production has increased since 2009 by more than 33 percent, underpinned by a series of structural reforms:

  • Setting a realistic and publicly disclosed price for the cotton crop at planting time, negotiated by a grass-roots producers’ association and supported by (i) a fully funded “price smoothing” fund and (ii) increased use of forward sale contracts by ginners.

  • Establishing a revolving-door “inputs fund” that mobilizes seasonal bank financing and fertilizer subsidies.

  • Developing collective liability of producers’ associations for loans connected to agricultural inputs.

  • Introducing new entrants in ginning who provide credible alternatives to the dominant state-owned ginner.

The government has recently prioritized agriculture in policymaking through its development program. It has introduced many interventions that complement the structural reforms described above and that complement the list of measures generally recommended by specialists (UNDP, 2011).

  • Improvements in crop resilience to shocks by using more productive and drought- and pest-resistant seeds.

  • Smart management of soil fertility, given increasing concerns of soil saturation and limits on the availability of arable land.

  • Increased irrigation, including building new dams and rehabilitating existing dams, to hedge against droughts.

  • Promotion of crop diversity to improve farmstead resilience, combined with a focus on larger-scale production of oil seeds and fruit to tap international markets.

  • Promotion of cooperatives and marketing associations to establish economies of scale and scope for small farmers and to promote agribusiness opportunities.

Tanzania: Growth in Manufacturing2

The Tanzanian economy has changed significantly since the early 2000s, with traditional agricultural cash crops (cotton, coffee, tea, sisal, cashew nuts, tobacco) having declined in importance. This reflects the reorientation of agricultural production to the domestic market to meet the demands of a rapidly growing population and the substitution of traditional exports by rising exports of minerals (gold) and manufactures (Figure 3.12).

Figure 3.12.Tanzania: Composition of Exports of Goods and Services

Source: Tanzanian authorities.

Gold exports have grown to almost 30 percent of export receipts, making Tanzania the fourth largest gold exporter in Africa. The expansion largely reflects rapidly rising gold prices because the share of gold production in real GDP has increased only marginally to 2.5 percent in 2010.

In the past decade, the manufacturing sector developed strongly, driven primarily by demand from regional markets, supported by the strengthening of the East African Community (EAC) common market. Manufacturing’s share of Tanzania’s exports grew from 3 percent in 2000 to 17 percent in 2010, with the key products being processed food, beverages, and tobacco products, followed by manufacturing of furniture and other wood products.

Tanzania’s successful diversification away from low value-added agriculture to higher value-added manufacturing is attributable to the policy and structural reforms carried out at the beginning of the decade. These reforms were designed to transform a government-dominated welfare state into a market economy, and covered a broad range of areas: exchange rate, price, and trade liberalization; tax and public finance reforms; introduction of modern monetary policy; liberalization of the financial sector; reforms of the public sector; regulatory and institutional reforms; and the onset of regional integration. In tandem, macroeconomic stabilization advanced rapidly—GDP growth more than doubled from an average of 3 percent in the 1990s to 6.7 percent in the 2000s; inflation declined to single digits; and public debt was reduced to sustainable levels.

At the same time, substantial reductions in tariffs and non-tariff barriers, market deregulation and privatization, and large public investments in infrastructure, including in the energy sector, offered fertile ground for rapid productivity growth in the private sector and the expansion of exports. Indeed, labor and total factor productivity increased by more than 30 percent in the last decade, with manufacturing one of the key sectors benefitting from these reforms.

Namibia: Diversification from Minerals to Manufacturing3

For a decade after independence in 1990, Namibia’s exports relied primarily on the mining sector, with low activity in the manufacturing sector. Mineral exports—mostly gem diamonds, uranium, and base metals—accounted for more than 50 percent of goods and services exports. Within the mining sector, diamond mining played a dominant role, contributing about 74 percent of total mining exports (2000–06 average). With prudent macroeconomic policies and political stability, the government embarked on policy measures to diversify its export base away from minerals to manufacturing. This was achieved through key interventions including the development of export processing zones, industrial parks, common facility centers, and small- and medium-size enterprise (SME) development programs. In addition, the government embarked on active promotion and marketing of Namibia’s investment opportunities to potential foreign investors, and conducted active research on diversification. These policies facilitated an increase of 5 percentage points in the manufacturing sector’s contribution to total exports, reaching 20 percent by 2006. Other exports, including raw hides and skins and arts and crafts, also grew significantly.

Figure 3.13.Namibia: Composition of Manufactured Exports

Sources: Namibian authorities; and IMF staff estimates.

Policy initiatives have also supported diversification in manufactured exports by encouraging textiles and clothing, polished and processed diamonds, and refined copper and zinc. The export share of prepared fish, which traditionally dominated the manufactured exports basket, declined from 49 percent to 36 percent between 1995 and 2006; and the contribution of refined zinc has increased significantly (Figure 3.13). This new area of the manufacturing sector created strong backward and forward linkages within the economy, particularly between the manufacturing and mining sectors.

Mauritius: Shift from Manufacturing to Services Exports4

Since independence, Mauritius has demonstrated a strong economic record based on sound institutions and macroeconomic policies (including prudent fiscal policy and a flexible exchange rate). This performance is impressive given poor natural resource endowments and the high vulnerability to external shocks intrinsic to a small open economy. Economic openness and a political system characterized by coalition governments with a strong track record of protecting property rights (especially for minority ethnic groups) played a key role in the Mauritius success story. In addition, Mauritian policymakers made good use of preferential access to markets in Europe and the United States.

The share of services in total exports grew rapidly in the 2000s (Figure 3.14). Initially driven by growth in tourism (tourist arrivals increased from 240,000 in 1988 to close to 965,000 by end-2011) and financial services, exports of information and communication technologies (ICT), and business process outsourcing (BPO) later followed. Currently, the ICT sector accounts for 2.5 percent of employment and almost 3 percent of exports. New export subsectors are also emerging (health/medical services).

Figure 3.14.Mauritius: Exports and FDI Inflows

Sources: Mauritian authorities; and IMF staff estimates.

In the mid-2000s the country suffered severe external shocks, namely the loss of sugar preferences and the phase-out of textile trade preferences (dismantling of the Multi-Fiber Agreement), which precipitated a shift away from manufacturing exports toward services. Policy measures also influenced this shift, including boosts to foreign direct investment (FDI) in tourism, financial services, and ICT; changes in labor market regulations to increase flexibility; and simplification of the tax regime. The introduction of a 15 percent flat income tax rate for individuals and corporations and the Business Facilitation Act provided an enabling environment for the development of new sources of export growth.

The government has played an important role in fostering new export sectors by formulating sectoral policies and facilitating private sector expansion. Notably, growth in tourism has been supported by government strategic planning complemented by the promotional activities of hotels and by Air Mauritius. Successful structural transformation in Mauritius was supported by a forward-looking approach followed by policymakers, who proactively anticipated responses to underlying changes to the economic environment.

Kenya: Growing Service Exports5

The service sector in Kenya has been the largest contributor to GDP growth and foreign exchange revenue in recent years (Figure 3.15).6 Unlike other LICs, Kenya has several world-class firms exporting high-value off-shore services, such as product development, research and development business ventures, insurance, accounting, and BPO services.

Figure 3.15.Contribution to Real GDP Growth by Sectors, 2005–11

Sources: Kenyan authorities; and IMF staff estimates.

Kenya appears particularly competitive in transportation, communication, and financial services, given the relative abundance of qualified professionals in Kenya employed in the intra-EAC service trade, to which should be added a strong entrepreneurial tradition (World Bank, 2012).

Kenya’s transport services have grown rapidly in recent years, as Kenya’s international airline company has firmly established itself in Africa, consistently realizing growing passenger numbers. Growth in air and shipping freight have been made possible by increasing trade openness in the EAC, partly owing to reforms in the shipping subsector and removal of explicit trade restrictions. Moreover, Nairobi is a key transportation hub for Eastern and Central Africa and the largest city between Cairo and Johannesburg. The port of Mombasa is currently ranked fifth in Africa in terms of handling capacity, having recorded a 145 percent increase between 2006 and 2010, serving more than a dozen countries. Capacity at Mombasa will improve by widening berths to accommodate larger ships and by constructing a second container terminal.

Kenya has increasingly concentrated on high-tech communication and technology services. These exports already account for more than 10 percent of total service exports (Figure 3.16) and close to 20 percent of total foreign direct investment inflows. The success of Safaricom in expanding mobile phone and Internet access and KenCall in providing call center and BPO services to organizations worldwide illustrate the sector’s dynamism. Currently, the Kenyan government is in the process of establishing an ICT Park for BPO, which is expected to further boost service exports.

Figure 3.16.Kenya: ICT Service Exports

Source: Kenyan authorities; and IMF staff estimates.

The cross-border expansion of Kenya’s financial sector has also leveraged regional integration within the EAC and developed a strong regional footprint. Trade in financial services should be further enhanced as barriers, particularly those arising from regulation (no single licensing regime) and capital account mobility, are gradually removed.

This chapter was prepared by Rodrigo Garcia-Verdu, Alun Thomas, and John Wakeman-Linn, with input from Yingying Shi.

This is a narrow sense of structural transformation, because the concept is often defined to include the urbanization and demographic transition that typically accompany economic growth. Other phenomena often associated with transformation are the diversification and increasing sophistication of exports and output and product quality upgrading, all of which are the focus of the recent research project “Economic Diversification for Stability and Growth in LICs,” by the IMF’s Strategy, Policy, and Review Department and Research Department.

The weighted average of value added in agriculture in sub-Saharan Africa was 11.2 percent of GDP in 2010. Given that South Africa accounts for about one-third of the region’s GDP and has a very low share (2.5 percent), this low figure is not surprising. The non-weighted average is close to 20 percent.

Except as otherwise noted, the concept of productivity analyzed throughout this chapter is average labor productivity, as measured by value added per worker, rather than total factor or multifactor productivity.

The FAO growth rates usually coincide with data from the system of national accounts in most countries, although discrepancies exist in a few cases. Part of the difference is explained by the lack of fishing and forestry data in the FAO data on value added, thus explaining why in some countries where these subsectors are important (e.g., fishing in Senegal and Seychelles; forestry in Cameroon) the two sources do not coincide.

For an in-depth discussion of economic growth in sub-Saharan Africa, and the emergence of a group of fast-growing economies, see Ndulu and others (2007, Volumes I and II) and Radelet (2010).

Two counter examples exist: Mozambique reports an average of only 1.2 years of schooling but has experienced rapid average labor productivity growth (perhaps, in part, another peace dividend), while Kenya, with an average of 7 years of schooling, has experienced slow average labor productivity growth.

Although the output share of industry rose briefly in Cameroon following the large devaluation of the CFA franc in 1994, this was only a temporary surge.

The initials appearing in the graphs correspond to the following sub-Saharan African countries: Angola (AGO), Benin (BEN), Botswana (BWA), Burkina Faso (BFA), Burundi (BDI), Cameroon (CMR), Cape Verde (CPV), Central African Republic (CAF), Chad (TCD), Comoros (COM), Côte d’Ivoire (CIV), Dem. Rep. of Congo (COD), Equatorial Guinea (GNQ), Eritrea (ERI), Ethiopia (ETH), Gabon (GAB), The Gambia (GMB), Ghana (GHA), Guinea (GIN), Guinea-Bissau (GNB), Kenya (KEN), Lesotho (LSO), Liberia (LBR), Madagascar (MDG), Malawi (MWI), Mali (MLI), Mauritius (MUS), Mozambique (MOZ), Namibia (NAM), Niger (NER), Nigeria (NGA), Rwanda (RWA), Rep. of Congo (COG), São Tome & Príncipe (STP), Senegal (SEN), Seychelles (SYC), Sierra Leone (SLE), South Africa (ZAF), Swaziland (SWZ), Tanzania (TZA), Togo (TGO), Uganda (UGA), Zambia (ZMB), and Zimbabwe (ZWE). The Asian countries included in the sample are Bangladesh (BNG), Cambodia (KHM), Indonesia (IDN), the Philippines (PHL), and Vietnam (VTM).

Fuglie and Rada (2011) have produced similar results for total factor productivity, with the annual growth rate at 2.1 percent for Southeast Asian countries over the period 1990-2007, 1.6 percent for South Asian countries, and 1.1 percent for sub-Saharan Africa.

This strategy was first adopted in South Africa and led to an almost fourfold increase in cereal yields. Some elements of this strategy are already being followed in other countries. For example, drought-tolerant varieties of millet and sorghum have been developed for farmers in the Sahel and the Horn of Africa, and fertilizer markets in Kenya and Malawi have been liberalized.

While separate discussions of mining and manufacturing would be preferable, data limitations preclude this.

The large absence of value added in manufacturing exports has recently been documented by Johnson and Noguera (2012) who associate it with the emergence of global supply chains in which production stages are sliced up and distributed across countries.

India’s recent experience of rapid growth in its services sector (telecommunications, finance, information technology, etc.) without first having developed a competitive manufacturing sector is a case in point. See Gordon and Gupta (2004), and Kochhar, and others (2006).

Indeed, sub-Saharan Africa could be a net exporter of refined petroleum if the refineries ran at a minimum of 75 percent capacity (Akinkugbe, 2012). However, at the moment refinery use is less than 50 percent in Côte d’Ivoire, Ghana, and Nigeria, and is above 75 percent only in Cameroon (CITAC, 2011).

Burundi appears to be an outlier in terms of its sharp services output growth, which could reflect massive spending by donors following the conflict in the early 2000s channeled through the government.

Important factors not discussed here include land-locked country status, thin financial markets, labor and land market regulations, and limited regional infrastructure.

These case studies were prepared by Antonio David, Floris Fleermuys, Nikoloz Gigineishvili, Ragnar Gudmundsson, Farayhi Gwenhamo, and Jan van Houtte.

For more details on Burkina Faso’s agricultural progress, see World Bank (2010) and Badiane (2011).

For more on Tanzania’s transformation, see Nord and others (2009).

For more on Namibia’s transformation, see Kadhikwa and Ndalikolue (2007) and Zaaruka and Namaklu (2002).

For more details, see Frankel (2010), Subramanian (2009), and Zafar (2011).

For more details, see World Bank (2012) and World Bank (2011).

Dileh and others (2011).

    Other Resources Citing This Publication