Information about Sub-Saharan Africa África subsahariana
Chapter

2. Fostering Durable and Inclusive Growth

Author(s):
International Monetary Fund
Published Date:
April 2014
Share
  • ShareShare
Information about Sub-Saharan Africa África subsahariana
Show Summary Details

Introduction

Over the last 20 or so years, many countries in sub-Saharan Africa have enjoyed robust and sustained economic growth. Sound economic policies, stronger institutions, and higher levels of public and private investment were behind this success. In most of these countries, sustained economic growth has translated into higher living standards, poverty reduction, and improved social indicators. But in some others, progress in these areas has fallen short of expectations, and the fruits of strong growth have sometimes accrued disproportionately to the better-off. This chapter looks at two areas in which economic policy efforts can help to make growth more inclusive. First, promote job creation, especially in household enterprises that operate in agriculture and services. The ongoing demographic transition in sub-Saharan Africa puts an additional premium on employment creation. Second, complement this effort by fostering financial inclusion and access to finance. Technological advances in mobile banking are opening up new possibilities to boost financial inclusion by reducing intermediation costs.

This chapter documents that per capita income growth in sub-Saharan Africa has been accompanied by better living conditions, as shown by improved human development indicators and lower poverty rates. True, progress has been uneven; but the overall trend is without a doubt favorable, and not limited to natural-resource exporters. In fact, some of the best performers are not resource-rich countries. A large number of people have moved out of poverty, and more individuals than ever before have access to basic education, health care, and clean water. Continued pursuit of sound macroeconomic policies, stronger institutions, debt relief, and, for some countries, the end of armed conflicts, were all instrumental in this transformation.

Despite this progress, there is no scope for complacency. Poverty remains pervasive and living standards need to improve further. Moreover, for a small number of countries, living standards have in fact deteriorated. Thus, the chapter looks at factors that could explain the relative performance of sub-Saharan African countries, both in terms of their increase in per capita income and their achievements in poverty reduction. We find that a stable macroeconomic environment, better infrastructure, and rising agricultural productivity yield the most favorable effects. Although most countries in the region have improved their macroeconomic conditions and infrastructure, lack of economic diversification associated with low productivity gains have held back structural transformation, leaving most of the workforce still underemployed in agriculture and with insufficient employment creation in sectors of the economy with higher value added per worker.

To illustrate this point and its consequences for poverty reduction, we contrast the experiences of Mozambique and Vietnam. These two countries’ growth performance has been similarly strong. But much larger agricultural productivity gains in Vietnam combined with the creation of higher-earning manufacturing jobs, compared with Mozambique, have helped to shift labor out of agriculture toward more productive sectors. In Mozambique, agricultural employment as a share of total employment has remained broadly unchanged. Not surprisingly, Vietnam succeeded in reducing poverty to a larger extent than Mozambique. In fact, most capital accumulation in Mozambique was concentrated in capital-intensive sectors, with certainly favorable effects on growth, but lower impacts on job opportunities and poverty reduction compared with Vietnam.

On the basis of this analysis and in light of the projected increase in the working-age population in sub-Saharan Africa, the chapter argues that, to make growth more inclusive, the overarching policy challenge is to create conditions for taking advantage of a possible demographic dividend. For this purpose, employment and entrepreneurial opportunities will have to be created at a faster pace. Improving infrastructure is also a necessary condition, albeit not sufficient, sustained and inclusive growth. Complementary policies discussed in this chapter are required for growth to achieve its potential and to contribute to poverty reduction. To this purpose, two key areas seem particularly important for policy:

  • First, expanding job opportunities is essential for both further raising per capita GDP growth and reducing poverty. We show that household enterprises are the most likely source of jobs for the majority of the population in sub-Saharan Africa, at least over the short and medium term. Policies should focus on removing obstacles to investment in the service and agricultural sectors—where the bulk of household enterprises perform their activities. The resulting productivity gains would favor the structural transformation process that would shift labor out of agriculture.
  • Second, financial inclusion can play a key role in improving welfare for the poor, and in removing financing obstacles to entrepreneurial activity. Here, the policy emphasis should be on creating an enabling environment to reduce transaction costs by exploiting new technologies, such as mobile and agent banking.

The next section focuses on the key features of the recent growth experience and on the main determinants of sub-Saharan African countries’ relative performance both in terms of growth and standard welfare measures.

Recent Growth and Human Development Performance

Significant and widespread increases in per capita GDP in sub-Saharan African countries have helped improve human development indicators. Between 2000 and 2013, sub-Saharan African countries experienced an increase in the median per capita GDP of 75 percent. Also, most sub-Saharan African countries showed a marked improvement in human development, as measured by the human development indices (HDI) computed by the United Nations, especially those that were worse off at the beginning of the period (Figure 2.1, panels a and b).

Figure 2.1a.Sub-Saharan Africa: GDP per Capita (PPP)

Source: IMF, World Economic Outlook database.

Note: PPP = purchasing power parity.

Figure 2.1b.Sub-Saharan Africa: Human Development Index

Source: United Nations Development Programme, Human Development Report.

Human development indicators have generally evolved in line with changes in GDP per capita. Countries that have experienced the largest increases in income and human development include those rich in mineral resources, such as Angola, Ghana, and Mozambique, as well as countries that are not primarily commodity exporters, such as Ethiopia, Rwanda, and Tanzania (see Chapter 2 of the October 2013 Regional Economic Outlook: Sub-Saharan Africa). Countries with higher GDP per capita tend to have better human development indices, and growth in GDP per capita and improvements in human development have often been larger in some of the countries that have been lagging behind. Take, for example, the case of Mozambique, which has recorded one of the highest improvements, but remains one of the poorest countries on both counts (Figure 2.2, panels a and b).

Figure 2.2a.Sub-Saharan Africa: GDP per Capita (PPP) and Human Development Index, 2012

Sources: IMF, World Economic Outlook database; and United Nations Development Programme, Human Development Report.

Note: Percent rank in sub-Saharan Africa. A higher number indicates higher per capita income and better degree of human development. PPP = purchasing power parity. See page 64 for country name abbreviations.

Figure 2.2b.Sub-Saharan Africa: GDP per Capita (PPP) and Human Development Index Growth, 2000–12

Sources: IMF, World Economic Outlook database; and United Nations Development Programme, Human Development Report.

Note: Percent rank in sub-Saharan Africa. A higher number indicates higher per capita income and better degree of human development. PPP = purchasing power parity. See page 64 for country name abbreviations.

Improvements in human development partly reflect advances in health and education. Primary enrollment and completion rates show remarkable progress, in line with developments in other developing countries. Infant and maternal mortality have declined substantially in the region and have fallen faster than in other developing economies in the last decade (Figure 2.3). The prevalence of undernourishment has also declined significantly. Higher access to clean water and sanitation across the region has helped improve health indicators.

Figure 2.3.Sub-Saharan Africa: Human Development Indicators

Source: World Bank, World Development Indicators.

Despite the overall progress, per capita GDP growth in sub-Saharan Africa only kept pace with the rest of the world after 2000 except in the 10 fastest growing economies, which have started to catch up (Figure 2.4). This partly reflects relatively high population growth (three out of four countries in sub-Saharan Africa made it to the top quartile in the distribution of the world’s population growth rates (Figure 2.5). The region’s growing young population reflects high fertility and declining infant mortality rates.

Figure 2.4.Sub-Saharan Africa: GDP per Capita (PPP)

Source: IMF, World Economic Outlook database.

Note: Each line computes the percent rank in the world distribution of GDP per capita on a constant purchasing power parity (PPP) basis of the lower quartile, median, and top quartile in sub-Saharan Africa.

Figure 2.5.Sub-Saharan Africa: Income and Demographic Indicators, 2000–13

(Percent of Sub-Saharan African Countries in World Quartile distribution)

Source: IMF, World Economic Outlook database.

Note: The figure indicates that 37 percent of the 43 sub-Saharan African countries with available data fell in the top 25 percent of the world distribution of GDP growth rates between 2000 and 2013. PPP = purchasing power parity.

1 Measured as the increase in the share of the population older than 15 years that is employed.

Also, the significant improvement in sub-Saharan Africa’s HDI is lagging relative to the corresponding improvement worldwide (Figure 2.6). In fact, progress in achieving the Millennium Development Goals (MDGs) has been uneven, lagging behind other developing countries and slower than needed to reach the 2015 targets (see the October 2013 Sub-Saharan Africa: Regional Economic Outlook).

Figure 2.6.Sub-Saharan Africa: Human Development Index, 1990–2012

Source: United Nations Development Programme, Human Development Report.

Note: Each line computes the cumulative difference in the percent rank since 1990.

The role of macroeconomic policies

Better macroeconomic policies and political developments played a key role in sub-Saharan Africa’s achievements, in particular in Angola, Burundi, Ethiopia, Ghana, Mozambique, Rwanda, Sierra Leone, Tanzania, Uganda, and Zambia—some of the better performing countries highlighted in Figure 2.2b. Most of these countries benefited from reductions in public debt burdens through the Heavily Indebted Poor Countries (HIPC) Initiative, which gave them fiscal space and facilitated macroeconomic stabilization. Eight of these 10 countries more than quadrupled their exports (out of the 14 sub-Saharan African countries that substantially increased their exports at constant U.S. dollar prices in the period). The region’s exports benefited from expanding global trade, sharp terms-of-trade gains, and associated investments in export-oriented megaprojects. For five of these countries, recovery from macroeconomic disruptions and armed conflicts in the preceding decade was an important factor in explaining their performance, which led to a turnaround in total factor productivity (TFP). Eight of these countries have significantly lowered inflation, six of them into single digits, and only three depend heavily on natural resources.

Policies widely accepted as the most appropriate to encourage growth in low-income countries (LICs) include integration with the global economy, infrastructure development, export diversification, higher school enrollment, access to financial services, formalized property rights, and improved government administration, among others (IMF, 2013d). Using a cross-section regression to assess the relative role of some of these factors in boosting GDP per capita in sub-Saharan Africa, we find that macroeconomic policies, investments in human and physical capital, and structural transformation of a country’s economy that allows the creation of value beyond the agricultural sector all matter for a country’s relative income performance. However, this requires overcoming structural barriers to improving competitiveness. Boosting labor market efficiency and financial market development could play supporting roles (Box 2.1).

Poverty reduction

Sub-Saharan Africa has reduced poverty significantly, but at about 45 percent, poverty remains high. The last two household surveys in each sub-Saharan African country with available data for the last 24 years show that GDP per capita rose in virtually all countries, and poverty fell in about three out of four of those countries (Figure 2.7). In the last decade, the consumption distribution profiles have improved with rising income, as illustrated in Figure 2.8, showing data from six recent household surveys in sub-Saharan African countries.

Figure 2.7.Sub-Saharan Africa: Changes in Poverty and GDP per Capita

Sources: IMF, World Economic Outlook database; and World Bank, World Development Indicators.

Note: The figure shows change in the variables for the last two household surveys within the last 24 years. The sample consists of 26 observations for countries with available data. Poverty is measured as headcount of individuals earning more than $1.25 a day (constant purchasing power parity terms) in percent of total population.

Figure 2.8.Sub-Saharan Africa: Density Estimates of Population-Consumption Distribution

Sources: Country household surveys; and IMF staff calculations.

Note: The chart shows that a growing share of individuals in these countries consumed more than in the previous survey. The area under the distribution within a given range corresponds to the fraction of the population that consumes within that range.

This chapter does not address the topic of inequality and growth, extensively discussed in Ostry, Berg, and Tsangarides (2014). Rather, the focus is on exploring how poverty rates relate to the factors that explain the relative income position of countries around the world to identify the main determinants of poverty in sub-Saharan Africa. The analysis suggests that poverty is positively associated with poor infrastructure, a high share of agriculture in GDP, and high inflation, among others (Box 2.2). The infrastructure gap and the lack of economic diversification appear as the main systematic reasons behind the higher poverty in sub-Saharan Africa, partly offset by the lack of competition, presumably in goods that require low skills.

Structural transformation and inclusive growth

Fostering inclusive growth requires policies that favor both raising income (Box 2.1) and lowering poverty (Box 2.2). The analysis suggests that increasing productivity in sub-Saharan African economies would be an important component of this approach, including by (i) maintaining a stable macroeconomic environment, especially low and stable inflation; (ii) upgrading infrastructure; (iii) diversifying the structure of the economy through reforms that increase competitiveness and gradually reduce the share of agriculture in GDP; and (iv) fostering sound financial systems that safely deliver financial depth and inclusion. To face the significant shortages in infrastructure in sub-Saharan African countries, the challenge for policymakers is to create an appropriate environment to attract domestic and foreign investors.

Box 2.1.Policies that Matter for Increasing a Country’s Per Capita Income Performance

How does a given country’s per capita GDP compare with others around the world? The relative position of a country’s per capita GDP (country’s ranking) is closely associated with a wide array of macroeconomic and structural variables. Cross-section regressions indicate, for example, that a country’s ranking in the distribution of GDP per capita improves when its ranking in the macroeconomic environment and in the provision of health improves, or when the share of the population living in rural areas or the share of agriculture in GDP declines (Table 2.1.1). Among macroeconomic variables, lower inflation, real exchange rate appreciation, and exchange rate volatility result in a lower ranking of GDP per capita. These two-variable relationships also show that a higher degree of financial market development and financial inclusion are good for growth, as is labor market efficiency and the available infrastructure in the economy. These relationships hold for the world and sub-Saharan Africa, although intensity varies.

Table 2.1.1.Cross-Section Regressions on Per Capita GDP
VariableSourceSlope SignSlope on Sub-Saharan Africa DummyTotal Sub-Saharan Africa Slope1Sub-Saharan Africa Dummy LevelAdjusted R-squared
Structural variables
Market size (rank)WEFA+(3)Yes(3)0.41
Rural population, share (percent)WDI+(12)Yes(3)0.69
Agricultural share in GDP (percent)WDI+(14)(−6)+Yes(7)0.77
Health and education (rank)WEFA+(11)0.71
Educational quality (rank)WEFA+(6)(−2)+Yes(5)0.53
Technological readiness (rank)WEFA+(16)0.82
Health (rank)WEFA+(15)0.80
Life expectancy (rank)WEFA(−13)+(6)Yes(−6)0.76
Labor market efficiency (rank)WEFA+(6)Yes(5)0.51
Financial market development (rank)WEFA+(6)Yes(4)0.54
Loan access, ease of (rank)WEFA+(4)Yes(2)0.49
Financial inclusion (accounts per 1,000 adults)CGAP(−4)(−3)Yes(6)0.55
Institutional quality (rank)WEFA+(9)Yes(4)0.65
Gini coefficient (percent)WDI+(2)(−3)Yes(5)0.42
Policies
Macroeconomic environment (rank)WEFA+(4)Yes(3)0.47
Inflation (2000–13) (rank)WEO+(8)(−3)+Yes(7)0.52
Real exchange rate appreciation2 (rank)INS+(3)Yes(6)0.37
Real exchange rate volatility2 (rank)INS+(3)Yes(6)0.37
Other policy-felated factors
Infrastructure (rank)WEFA+(15)0.82
Infrastructure in electricity (rank)WEFA+(17)0.84
Infrastructure in transport (rank)WEFA+(10)Yes(3)0.69
Infrastructure quality (rank)WEFA(−10)0.67
Tariffs (percent)WDI+(4)0.59
Tariffs (percent)WEFA+(7)0.56
Competition (rank)WEFA+(7)Yes(5)0.57
Competition, local (rank)WEFA+(6)Yes(4)0.51
Market dominance (rank)WEFA+(6)(−2)+Yes(5)0.53
Sources: World Economic Forum, 2013–14 (WEFA); IMF, World Economic Outlook database (WEO), IMF, Information Notice System (INS); World Bank, Global Financial Inclusion database (CGAP); and World bank, World Development Indicator (WDI).Note: Cross-section regression by ordinary least squares over latest data available performed on ranks to focus on country’s relative position. Per capita GDP (rank) = constant + (slope+sub-Saharan Africa dummy slope*sub-Saharan Africa dummy)*variable+sub-Saharan Africa dummy level coefficient* sub-Saharan Africa dummy)+ error. Signs shown only for significant variables; t-statistics shown in parentheses.

Computed as adding the point estimates for the world slope adjusted for the sub-Saharan Africa slope premium or discount.

The appreciation is measured between 2000 and 2013, and the volatility as the coefficient of variation of the real exchange rate between 2000 and 2012 with annual data.

Sources: World Economic Forum, 2013–14 (WEFA); IMF, World Economic Outlook database (WEO), IMF, Information Notice System (INS); World Bank, Global Financial Inclusion database (CGAP); and World bank, World Development Indicator (WDI).Note: Cross-section regression by ordinary least squares over latest data available performed on ranks to focus on country’s relative position. Per capita GDP (rank) = constant + (slope+sub-Saharan Africa dummy slope*sub-Saharan Africa dummy)*variable+sub-Saharan Africa dummy level coefficient* sub-Saharan Africa dummy)+ error. Signs shown only for significant variables; t-statistics shown in parentheses.

Computed as adding the point estimates for the world slope adjusted for the sub-Saharan Africa slope premium or discount.

The appreciation is measured between 2000 and 2013, and the volatility as the coefficient of variation of the real exchange rate between 2000 and 2012 with annual data.

Alternatively, a multivariate regression shows that health (ranking), macroeconomic environment (ranking), share of population living in rural areas (percent), and share of agriculture in GDP (percent) improve relative performance. A fifth variable would be infrastructure (ranking), although it appears correlated with labor market efficiency (ranking) or financial market development (ranking) (Table 2.1.2).

Table 2.1.2.Sub-Saharan Africa: Multivariate Linear Regressions on GDP per Capita at Purchasing Power Parity (ranking)
1234
Full sample slopes
Labor market efficiency (ranking)0.0006 (3)
Financial market development (ranking)0.0006 (2)
Infrastructure (ranking)0.0022 (5)
Health (ranking)0.0039 (9)0.0039 (9)0.0029 (6)0.0041 (10)
Share of population in rural areas (percent)0.0024 (4)0.0024 (4)0.0023 (4)0.0024 (4)
Share of agriculture in GDP (percent)0.0062 (5)0.0058 (5)0.0046 (4)0.0062 (5)
Macroeconomic environment0.0005 (2)0.0006 (2)0.0008 (3)
Sub-Saharan Africa slope adjustment
Labor market efficiency (ranking)
Financial market development (ranking)
Infrastructure (ranking)
Health (ranking)−0.0018 (-5)
Share of population in rural areas (percent)
Share of agriculture in GDP (percent)
Macroeconomic environment0.0012 (2)0.0011 (2)0.0015 (4)0.0009 (2)
Sub-Saharan Africa dummy−0.1897 (-3)−0.1876 (-3)−0.1811 (-3)
Memorandum items:
Observations98989898
Adjusted R-squared0.90690.90580.91600.9015
Sources: World Bank, World Development Indicators; and IMF staff calculations.Note: Regressions using ordinary least squares. Constant is not shown. t-statistics shown in parentheses.
Sources: World Bank, World Development Indicators; and IMF staff calculations.Note: Regressions using ordinary least squares. Constant is not shown. t-statistics shown in parentheses.

These regressions do not imply causality, and have many caveats, but are consistent with the statement that macroeconomic policies matter for a country’s relative income performance—that government investments in human and physical capital are worth undertaking, and that the structural transformation of an economy that allows it to create value beyond the agricultural sphere can improve people’s standard of living, which would require overcoming structural barriers to improving competitiveness. Significant gaps with the rest of the world remain in the areas of education, health, the share of agriculture in GDP, financial inclusion, and infrastructure (Table 2.1.3).

Table 2.1.3.Sub-Saharan Africa: Development Gaps
TotalSub-Saharan AfricaNon-Sub-Saharan Africa
Per capita GDP (PPP, 2000)14,7844,25918,485
Poverty (percent)20448
Macroeconomic environment (rank)759568
Primary education (rank)7411262
Health (rank)7512858
Rural (percent)426037
Agricultural share in GDP (percent)13259
Labor market efficiency (rank)747973
Financial market development (rank)749267
Financial inclusion (accounts per 1,000 adults)1,2183161,536
Infrastructure (rank)7411660
Sources: IMF, World Economic Outlook database; World Bank, World Development Indicator; World Economic Forum Global Competitiveness Report 2013–14; and World Bank, Global Financial Inclusion database.Note: PPP = purchasing power parity.
Sources: IMF, World Economic Outlook database; World Bank, World Development Indicator; World Economic Forum Global Competitiveness Report 2013–14; and World Bank, Global Financial Inclusion database.Note: PPP = purchasing power parity.
This box was prepared by Jorge Iván Canales-Kriljenko.

Box 2.2.Structural Factors Affecting Poverty in Sub-Saharan Africa

Cross-section regressions suggest that poverty in sub-Saharan Africa has been higher in countries with a higher share of the population living in rural areas (percent), with less financial inclusion (access), worse infrastructure (ranking), lower tariffs, and lower technological readiness (ranking). Among the macroeconomic variables, higher inflation and real exchange rate volatility have been associated with higher poverty in the region. Given sparse data, the exercise uses the latest data available on poverty across countries with the latest available data on the other variables. The analysis suggests that these relationships only hold statistically for sub-Saharan Africa (Table 2.2.1).

Table 2.2.1.Sub-Saharan Africa: Cross-Section Regressions on Poverty Headcount
VariableSourceSlope SignSlope on Sub-Saharan Africa DummyTotal Sub-Saharan Africa Slope1Sub-Saharan Africa Dummy LevelAdjusted R-squared
Structural variables
Market size (rank)WEFA0.48
Rural population, share (percent)WDI+(2)+0.58
Agricultural share in GDP (percent)WDIYes(4)0.62
Health and education (rank)WEFA0.51
Educational quality (rank)WEFA0.48
Technological readiness (rank)WEFA+(4)+0.61
Health (rank)WEFA+(2)+0.52
Life expectancy (rank)WEFA0.54
Labor market efficiency (rank)WEFAYes(4)0.47
Financial market development (rank)WEFAYes(2)0.48
Ease of loan access (rank)WEFA0.48
Financial inclusion (accounts per 1,000 adults)CGAP(−3)Yes(7)0.59
Institutional quality (rank)WEFA0.50
Gini coefficient (percent)WDIYes(3)0.49
Policies
Macroeconomic environment (rank)WEFAYes(4)0.47
Inflation (2000–13) (rank)WEO+(3)+0.56
Real exchange rate appreciation2 (rank)INSYes(4)0.49
Real exchange rate volatility2 (rank)INS+(3)+Yes(3)0.54
Other policy-related factors
Infrastructure (rank)WEFA+(3)+0.61
Infrastructure in electricity (rank)WEFA+(3)+0.61
Infrastructure in transport (rank)WEFA+(3)+0.57
Infrastructure quality (rank)WEFAYes(3)0.51
Tariffs (percent)WDI(−2)Yes(5)0.55
Tariffs (percent)WEFA(−2)Yes(3)0.48
Competition (rank)WEFAYes(4)0.47
Competition, local (rank)WEFAYes(2)0.47
Market dominance (rank)WEFAYes(3)0.47
Sources: World Economic Forum, 2013–14 (WEFA); IMF, World Economic Outlook database (WEO), IMF, Information Notice System (INS); World Bank, Global Financial Inclusion database (CGAP); and World Bank, World Development Indicators (WDI).Note: Cross-section regression by ordinary least squares over latest data available performed on ranks to focus on country’s relative position. Data for the poverty headcount of those earning below 1.25 dollars per day is the latest available. Poverty (percent) = constant + (slope+sub-Saharan Africa dummy slope*sub-Saharan Africa dummy)*variable+sub-Saharan Africa dummy level coefficient* sub-Saharan Africa dummy)+ error. Signs shown only for significant variables; t-statistics shown in parentheses.

Computed as adding the point estimates for the world slope adjusted for the sub-Saharan Africa slope premium or discount.

The appreciation is measured between 2000 and 2013, and the volatility as the coefficient of variation of the real exchange rate between 2000 and 2012 with annual data.

Sources: World Economic Forum, 2013–14 (WEFA); IMF, World Economic Outlook database (WEO), IMF, Information Notice System (INS); World Bank, Global Financial Inclusion database (CGAP); and World Bank, World Development Indicators (WDI).Note: Cross-section regression by ordinary least squares over latest data available performed on ranks to focus on country’s relative position. Data for the poverty headcount of those earning below 1.25 dollars per day is the latest available. Poverty (percent) = constant + (slope+sub-Saharan Africa dummy slope*sub-Saharan Africa dummy)*variable+sub-Saharan Africa dummy level coefficient* sub-Saharan Africa dummy)+ error. Signs shown only for significant variables; t-statistics shown in parentheses.

Computed as adding the point estimates for the world slope adjusted for the sub-Saharan Africa slope premium or discount.

The appreciation is measured between 2000 and 2013, and the volatility as the coefficient of variation of the real exchange rate between 2000 and 2012 with annual data.

In a multivariate regression that focuses on sub-Saharan Africa, poverty is positively associated with poor infrastructure, high share of agriculture in GDP, lower tariffs, low market dominance, and higher inflation. Similar to the regressions on GDP per capita, there seems to be some close association between financial market development and infrastructure. In contrast, labor market efficiency (as measured by the ranking from the world competitiveness indicators) does not seem to be significant when infrastructure is taken out of the regression (Table 2.2.2).

Table 2.2.2.Sub-Saharan Africa: Multivariate Linear Regressions on Poverty(ranking)
123
Sub-Saharan Africa slope adjustment
Labor market efficiency (ranking)0.035(1)
Financial market development (ranking)0.159(3)
Infrastructure (ranking)0.390(7)
Share of agriculture in GDP (percent)0.594(4)0.497(3)0.347(3)
Tariffs−1.532(−2)−1.966(−3)−1.957(−5)
Market dominance (rank)−0.072(−2)−0.151(−3)−0.174(−5)
Inflation level (rank)13.127(2)20.962(3)14.814(4)
Sub-Saharan Africa dummy17.061(2)15.951(2)−6.039(−1)
Memorandum items:
Observations434343
Adjusted R-squared0.7420.7950.891
Sources: IMF, World Economic Outlook database; and World Bank, World Development Indicators.Note: Regressions using ordinary least squares, with slopes multiplied by the sub-Saharan African dummy. Constant is not shown. t-statistics shown in parentheses.
Sources: IMF, World Economic Outlook database; and World Bank, World Development Indicators.Note: Regressions using ordinary least squares, with slopes multiplied by the sub-Saharan African dummy. Constant is not shown. t-statistics shown in parentheses.

Although regressions do not imply causality, the poverty regressions are at least consistent with the view that policies that improve available infrastructure, diversify the economy, and maintain the purchasing value of domestic currency stable would tend to reduce poverty. Also, because lowering tariffs and increasing competition could cause social dislocations, particularly in the short run, explicit policies to protect the poor and increase their productivity would seem warranted. Interestingly, financial inclusion comes up as significant when inflation is excluded from the regression. This suggests that poor individuals may avoid saving or not afford to save in the financial system when the purchasing power of their savings is at stake.

This box was prepared by Jorge Iván Canales-Kriljenko.

The structural transformation of the economy required to increase total factor productivity significantly from its current low levels should at the same time (i) raise productivity in agriculture, where most of the population is employed (see the section on The Job Creation Challenge), including through amending the country’s legal and institutional frameworks governing agricultural activities; and (ii) boost competitiveness to allow the creation of value beyond basic agriculture, helping to develop a more diversified economic structure. However, there may be a temporary social dislocation associated with a more open and diversifying economy.

Fiscal policy may also play an important role in promoting inclusive growth, in particular through creating fiscal space for investment in human and physical capital. Revenue collection in many sub-Saharan countries remains relatively weak, so the focus should be on broadening tax bases and strengthening administrative tax collection capacity. On the expenditure side, improving the progressivity of public spending through shifting away from general subsidies toward well-targeted expenditure, such as cash transfers, is important (IMF, 2014b).

The Role of Productivity in Reducing Poverty: The Experiences of Mozambique and Vietnam

The experiences of Mozambique and Vietnam illustrate the importance of agricultural productivity gains and economic diversification for poverty reduction. Both countries are star performers that delivered remarkably similar growth in GDP per capita (adjusted for changes in purchasing power), but different results in poverty reduction. In the first five years after their growth takeoffs (1989 in Vietnam and 1992 in Mozambique), GDP per capita rose by 50 percent in both countries, although in Mozambique, the population grew much faster. The two countries endured protracted armed conflicts in the preceding years, transitioning from centrally planned socialist economies to market economies, and undertook important structural reforms. The years when the household surveys were conducted do not fully coincide with the takeoff years, but they show significant differences between the two countries: in Mozambique, headcount poverty fell by 2.4 percent annually between 1996 and 2007, compared with a decline of 4.6 percent annually in Vietnam between 1992 and 2004 (Figures 2.9 and 2.10).

Figure 2.9.Mozambique: Poverty Rates, 1996 and 2007

Source: World Bank, PovcalNet database.

Figure 2.10.Vietnam: Poverty Rates, 1992 and 2004

Source: Wolrd Bank, PovcalNet database.

In the first five years after the takeoff, the share of agriculture declined by a similar magnitude in both countries: 12 percentage points in Vietnam and 11 percentage points in Mozambique, as a result of structural transformation. However, an important difference between the two countries was their relative ability to generate employment outside agriculture. The share of agricultural employment remained almost constant in Mozambique in 2002–08, whereas in Vietnam, it declined by 10 percentage points in 2000–06, in line with improvements in agricultural productivity (Figures 2.11 and 2.12).

Figure 2.11.Mozambique: Employment by Sector, 2002–08

Source: IMF staff estimates based on country houshehold survey data.

Figure 2.12.Vietnam: Employment by Sector, 2000–06

Source: IMF staff estimates based on country houshehold survey data.

Vietnam was able to create more than two million jobs in the industrial sector, whereas the corresponding increase in Mozambique was about one-hundred and sixty thousand jobs, of which only about forty thousand were salaried. Vietnam was able to improve the productivity of a larger fraction of the population by shifting workers into more productive activities, whereas in Mozambique, agricultural productivity did not improve significantly in the period. In fact, unlike total factor productivity, agricultural productivity in Vietnam increased faster and lasted longer than in Mozambique, where it has remained flat in the last few years (Figures 2.13 and 2.14). In Vietnam, initial reforms focused on land use rights, privatization of collective assets, liberalization of prices, and streamlining of agricultural subsidies. These reforms were supported by a large, low-cost skilled labor force and proximity to major economies of the fastest growing region in the world. Foreign direct investment (FDI) from Asian countries accounted for almost 45 percent of cumulative FDI (IMF, 2014e).

Figure 2.13.Mozambique and Vietnam: Real GDP and Total Factory Productivity

Source: IMF staff estimates based on data from the Penn World Tables, version 7.1.

Note: Index, takeoff year = 100. TFP = total factor productivity.

Figure 2.14.Mozambique and Vietnam: Total Factor Productivity in the Agricultural Sector

Source: IMF staff calculations based on estimates by U.S. Department of Agriculture Economic Research Service, International Agricultural Productivity data product.

Note: Index, total factor productivity in agriculture = 100 in the year in which the growth takeoff began, t = 0.

The experience of Mozambique shows that the accumulation of capital concentrated in sectors that generate little employment does not contribute significantly to poverty reduction. Large megaprojects in natural resource enclaves can still generate fast growth, but policies to make growth inclusive are still needed.

The Job Creation Challenge

Sub-Saharan Africa needs to expand the availability and productivity of job opportunities for several reasons. First, most people in sub-Saharan Africa live from their labor income, and the number of poor in the region is still high. Second, the demographic dividend will result in a significant increase in the working age population. By 2020, more than half of sub-Saharan Africa’s population will be below the age of 25, thereby creating massive employment and growth opportunities (Figure 2.15). Absorbing these individuals into productive activities that help reduce poverty rates in due course is crucial to avoiding social and political tensions that could derail much of the progress achieved thus far, particularly in a context of weak institutions.

Figure 2.15.Sub-Saharan Africa: Population by Age Group, 2005–20

Source: United Nations, World Population Prospects: The 2010 Revision.

As the comparison between Mozambique and Vietnam illustrates, a crucial challenge is to increase labor productivity for the largest possible share of the working age population.

Most of the sub-Saharan African population works in low-productivity agriculture and service activities using methods that do not benefit from economies of scale, partly reflecting low labor skills. Except for middle-income countries, labor markets in sub-Saharan Africa are characterized by very low formal sector employment (about 10 percent of the labor force); very low unemployment rates (about 2–3 percent), and a large active population; leading to underemployment. Services are produced to a large extent in household enterprises.1

In line with experience worldwide, including in sub-Saharan Africa, a shift of labor to other more productive sectors will gradually tend to reduce the share of agriculture in GDP. More processing of agricultural products, even if basic, would add value to the tasks of harvesting crops and farming animals, for example, by transforming agricultural products into meals at a restaurant or packaging and preserving food for export.

On the positive side, as employment opportunities are created elsewhere, underemployment in agriculture should be reduced. As less productive workers move out of the sector, value added (and earnings) per worker should rise even if there are no gains in productivity in agriculture. However, the experience of other countries, including Vietnam, suggests that this process of higher productivity in agriculture is enhanced by gains in total factor productivity as the whole economy modernizes and diversifies. However, securing these gains usually implies a faster degree of urbanization to be able to move into higher-value-added activities, including manufacturing and exports of services. The latter would require further reforms that support a more business-friendly environment, such as reducing administrative burdens, simplifying regulations, and promoting competition, which may require investment to catalyze the ability of the country to incorporate the latest technologies. This approach could deliver sizable productivity gains that could generate sufficient employment opportunities to further reduce poverty and income inequality decisively. However, this also implies that country authorities need to invest more resources into urban infrastructure and improve their physical planning frameworks. Fiscal space needs to be generated to support the urbanization process to minimize the risks of social disruption and potential high welfare losses from a disorderly rural to urban transition.

At the same time, there is also significant scope for raising total factor productivity through improvements in agricultural productivity itself. Increasing agricultural productivity is also likely to deliver significant social dividends. Agriculture employs by far the most individuals in sub-Saharan Africa (Figure 2.16). Gains in agriculture are important because many workers will remain in the sector for a long time as agriculture is more labor intensive than other sectors, such as manufacturing and mining. In addition, the skills needed for agricultural activities are less specialized and require less training. Moreover, evidence from the region suggests that growth in agriculture in the last few decades has been among the most important contributors to poverty reduction in the region (Box 2.3). Growth in staple crops reduces poverty more decisively than growth in cash crops (IMF, 2013b).

Figure 2.16.Sub-Saharan Africa: Projected Distribution of Employment by Country Type and Sector, 2020

Sources: Country household surveys; IMF, African Department database; and IMF staff’ calculations.

Agricultural productivity in sub-Saharan Africa lags behind that of other countries, and this gap has widened over time (Figure 2.17). Most growth in sub-Saharan Africa has come from adding factors of production, such as land and labor, into the productive process. The continent combines a large area of uncultivated arable land with relatively low crop yields, and thus has enormous potential. In addition, the area under cultivation has increased in the region, unlike in the rest of the world, where it has declined because of growing urbanization.

Figure 2.17.Total Factor Productivity in Agriculture, 1961–2010

Source: U.S. Department of Agriculture, Economic Research Service.

1 Excludes South Africa.

Box 2.3.Poverty Reduction and the Sectoral Composition of Growth

Progress with poverty reduction has been uneven across sub-Saharan African countries in the last two decades. In some fast-growing countries, the level of poverty has not decreased significantly, whereas the opposite is true in a country such as Ethiopia, which lifted almost one-fourth of its population out of poverty in 2000–10.

Why does growth benefit the poor much more markedly in some countries than in others? One explanation is differences in sectoral composition of growth. Ethiopia’s growth surge has been mainly an agriculture and services story, but in other countries, growth has been mostly driven by extractive activities. Table 2.3.1 explores this idea further in the context of a panel of 35 sub-Saharan African countries and five three-year time periods from 1996 to 2010. Poverty is measured here by the $1.25-a-day headcount index, using data from the World Bank’s PovcalNet database. These data were supplemented with data from the IMF on sectoral real value added for agriculture, manufacturing, services, extractive activities (that is, mining plus manufacturing of petroleum and coal products), construction, and utilities.

Table 2.3.1.Sub-Saharan Africa: Relationship between Sectoral Growth and Poverty Reduction
Dependent variable: Poverty headcount Index12
Agriculture−0.410 **−0.430 ***
(0.047)(0.000)
Services−0.280 ***−0.200 ***
(0.000)(0.000)
Manufacturing−0.130−0.050
(0.160)(0.250)
Extractive activities−0.080 *−0.090 *
(0.092)(0.006)
Construction−0.190−0.080
(0.123)(0.111)
Utilities0.070−0.040
(0.533)(0.761)
Gini0.250 ***
(0.000)
Source: IMF staff calculations.Note: P-values reported in parentheses; and ***, **, and * indicate significance at the 1 percent, 5 percent, and 10 percent levels, respectively. Estimates were obtained using the generalized method of moments.
Source: IMF staff calculations.Note: P-values reported in parentheses; and ***, **, and * indicate significance at the 1 percent, 5 percent, and 10 percent levels, respectively. Estimates were obtained using the generalized method of moments.

Table 2.3.1 presents the results of assessing the poverty-reducing potential of the sectors in sub-Saharan Africa, examining them jointly in columns 1 and 2. Growth in all sectors (except utilities) has a significant and negative impact on poverty. The magnitude of this impact, however, varies widely from one sector to another. Agriculture seems to have by far the strongest bearing on poverty. According to our joint estimation, a 1 percent growth in agriculture pulls 0.41 percent of the population out of poverty. Then follow services: the same 1 percent growth reduces the proportion of poor people in the population by roughly 0.28 percent. Other sectors show a much lower elasticity of poverty to growth, as shown in column 2. These results hold if the Gini coefficient is included as a control variable.

This box was prepared by Isabell Adenauer and Samba Mbaye.

The overall lackluster performance in agricultural productivity reflects significant obstacles related to structural problems that are politically difficult to address decisively. Still, some sub-Saharan African countries, such as Ethiopia and Rwanda, have already increased their yield substantially. In other countries, however, agricultural productivity has stagnated. McKinsey (2010) estimates that shifting to higher-value crops and increasing crop yields, coupled with cultivating more land, could, over in the next two decades, increase growth in the value of the continent’s agricultural production to a rate twice as fast compared to that experienced in the last decade.

Policies for creating job opportunities

Raising agricultural productivity and encouraging economic diversification into labor-intensive activities beyond agriculture requires investments in human and physical capital, both private and public. Given scarce resources, government efforts at reducing poverty need to be well targeted. Because household enterprises absorb most of the labor force in the region, a poverty reduction strategy should focus on nurturing them to grow into more productive firms.

Governments in the region can support household enterprises by providing the following:

  • A business environment that does not overburden them with taxation or regulation. At the moment, household enterprises often pay taxes at a higher rate than large businesses, but receive little in return.
  • Supplementary services such as security, sanitation, electricity, transport, and water supply, among others, that could be thought of as the business infrastructure of the economy.
  • Technical training on job-relevant skills, including business and financial literacy skills to help improve human capital.
  • Raising productivity in the agricultural sector through appropriate land reform, accompanied by improvements in physical infrastructure and increased crop yields. This could assist the structural transformation process and generate widespread economic gains.
  • Policies for financial inclusion that permit their access to financial services (including credit, banking services, and insurance products).

In most sub-Saharan African countries, financial systems are not well developed. On the one hand, private savings are relatively low, and there is limited coverage of social security and social safety nets to rely on during old age or when facing adverse shocks, respectively. On the other hand, as mentioned earlier, labor continues to be the single most important source of income for the vast majority of the region’s population.2 For more enterprises to grow and provide employment opportunities, more people should be willing to take risks. Constrained access to credit would limit the implementation of new projects, especially the most innovative that require a longer period of maturity. Access to savings, insurance, and other financial services is needed to smooth household and business cash flows. Field experiments show that giving individuals access to savings accounts increases savings, productive investment, consumption, investment in preventive health, productivity, and income (Demirgüç-Kunt, Beck, and Honohan, 2008). Finally, broader financial access would help allocate talent across occupations, encouraging entrepreneurs to use their skills to create productive job opportunities.

The Financial Inclusion Challenge

Financial system deepening can help raise investment, spur innovation, facilitate technology transfer, and lead to a more efficient allocation of capital across sectors (Dabla-Norris and others, 2013). However, financial deepening may coexist with exclusion of large segments of the population from financial services. To illustrate the difference between financial deepening and financial inclusion, we plot financial deepening, measured by the ratio of credit to GDP, and financial inclusion, measured by the percent of the adult population holding an account in a formal institution. Figure 2.18 shows that, overall, once a critical level of access is reached, further progress with financial deepening may still help increase saving and investments needed for developing productive activities and generating employment. But for low-income economies, increasing financial deepening without raising financial access significantly will have less impact on the creation of job opportunities and the promotion of business activities that benefit the poor. This requires a proactive approach by policymakers to spur financial access at low levels of intermediation because high transaction costs may discourage financial institutions from granting access to low-income households.

Figure 2.18.Sub-Saharan Africa: Financial Access and Financial Deepening, 2011

Sources: IMF, African Department database; and World Bank, Global Financial Inclusion database.

Note: See page 64 for country name abbreviations.

The impact of financial access on human development and income is not easy to establish empirically because of reversed causality concerns. Using information from the World Bank Global Financial Inclusion database, we explore a possible correlation between these two variables to visualize the impact of policies facilitating higher financial access (Figure 2.19).3 We plot GDP-adjusted financial access performance by country against changes in human development indices between 2000 and 2011 and find a positive correlation for countries with an intermediate level of financial intermediation (credit-to-GDP ratio between 10 percent and 30 percent). This suggests that countries in the transition from a low to a high level of financial intermediation are those enjoying a faster increase in human development indices. In fact, many of the underperformers are countries with low financial intermediation. Likewise, many overperformers are countries that also show higher levels of financial intermediation.

Figure 2.19.Sub-Saharan Africa: Financial Access and Improvements in Human Development

Sources: United Nations Development Programme, Human Development Report; and World Bank, Global Financial Inclusion database.

Note: See page 64 for country name abbreviations.

The problem of financial intermediation is more pressing for the poor. Higher intermediation costs for the poor make financial access less elastic to changes in income for this group. The response of financial access to changes in GDP per capita is stronger for “rich” (top income quintile) than for “poor” (bottom income quintile) populations. The difference between rich and poor is even sharper when using the HDI, suggesting that other components of the HDI, such as life expectancy and educational attainment within a country, may also play an important role in explaining differences in the financial access response to changes in living standards between the poor and the rich (Figures 2.20 and 2.21). For example, primary and secondary education are correlated with the ability to adapt new technologies, an increasingly important element of financial access in sub-Saharan Africa.

Figure 2.20.Sub-Saharan Africa: Financial Access and GDP per Capita, 2011

Sources: IMF, World Economic Outlook database; and World Bank, Global Financial Inclusion database.

Figure 2.21.Sub-Saharan Africa: Financial Access and Human Development Index, 2011

Sources: United Nations Development Programme, Human Development Report; and World Bank, Global Financial Inclusion database.

This evidence suggests that policies that help reduce intermediation costs may have a stronger potential to expand financial inclusion, especially for the poor, than measures that attempt to increase access to credit directly. For many years, financial inclusion policies in LICs, and in sub-Saharan African countries, have focused on providing larger access to credit to the poor, through special credit lines or specialized credit institutions, including state-owned banks and microfinance institutions. However, these policies have had mixed results in terms of improving access to financial services, and have not reduced intermediation costs (World Bank, 2014).

More recent evidence points at technological developments as successful drivers of reductions in intermediation costs, and new studies show that the impact of lower transaction costs on economic activity and welfare can be substantial. From the fast increase in the number and use of ATMs to the impressive growth of mobile money transfers, sub-Saharan African countries are experiencing a substantial transformation in the provision of financial services. In Kenya, mobile banking has become mainstream, and the model is being followed by other sub-Saharan African countries. There is room for exploiting the potential benefits from this innovation, with Angola and Kenya well ahead of other countries in the use of mobile phones for transactions other than transfers (Figure 2.22).

Figure 2.22.Sub-Saharan Africa: Mobile Phones Used to Pay Bills, 2011

Source: World Bank, Global Financial Inclusion database.

Recent research has tried to identify more fundamental issues behind Kenya’s success. Using a regulatory index comprising regulations on e-contracting, consumer protection, interoperability, know-your-customer service, branchless banking, and e-money, Gutierrez and Singh (2013) find that the right balance in regulatory openness and certainty explain Kenya’s success relative to a sample of 34 developing and emerging market economies. In countries moving faster to exploit mobile banking, new business models are contributing to improving financial inclusion and innovation. Other empirical evidence shows that mobile phones have the potential to benefit consumers’ and producers’ welfare and perhaps broader economic development (Aker and Mbiti, 2010).

Policies to foster financial inclusion

Enabling policies can go a long way in supporting financial inclusion, and show a huge potential to incorporate low-income segments of the population into formal financial systems. This includes being flexible in the adoption of innovative products. New products will normally constitute potential competition with established market players, who naturally will be opposed to their introduction and will lobby for additional safeguards before admitting new entrants to financial markets. In addition, the parallel introduction of different platforms can be mutually supportive. The interplay among bank agents, credit bureaus, and mobile banking allows the buildup of a track record for users of financial services from the lower-income segment of the population—which will eventually facilitate access to credit in a more sustainable way than through special programs.

Policies to support financial inclusion should include the following:

  • Promoting competition and removing market imperfections, especially those introducing biases against access by the poor to financial services. Facilitating the use of bank agents—using post offices, supermarkets, grocery stores, and gasoline stations to represent banks outside of their branch network—will make financial services much more accessible to large segments of the population.
  • Focusing improvements in financial and lending infrastructure in areas with higher impact on financial and transaction costs. For example, supporting borrower identification can go a long way in mitigating the reluctance of banks to absorb low-income customers. Also, supporting the information environment will contribute to higher competition and lower costs.
  • Facilitating the operations of credit-reporting institutions allowing effective information that benefits low-scale bank customers. This requires an adequate legal framework that protects the rights of consumers, mitigates information availability risks, and ensures equity and transparency among participants (Brown, Japelli, and Pagano, 2009).
  • Balancing market-friendly actions, appropriate macroprudential oversight, and careful calibration of public policies should be a permanent effort. Policies to broaden financial access require a concomitant widening of the regulatory and supervisory perimeter to minimize regulatory arbitrage and financial system risks.

Conclusion

This chapter focused on identifying key policies that can help to create job opportunities and reduce poverty for the fast-growing population in sub-Saharan African countries, enabling them to continue and possibly enhance their recent favorable performance by making growth more inclusive. Maintaining sound macroeconomic policies and removing structural distortions remain necessary in sustaining economic expansion and supporting job creation. These need to be supplemented by significant improvements in infrastructure to lower the cost of doing business and increase output. This also means that sustainable capital accumulation will be necessary to supplement total factor productivity, which in the past decade has supported economic expansion. Improvements in agricultural productivity alongside structural transformation are essential for continued poverty reduction in sub-Saharan Africa—at least for the next few decades.

In addition, policies focused on fostering financial inclusion are crucial for expanding job opportunities in sub-Saharan African countries and LICs in general, as they increase the participation of the poor in economic activities. Policies that lower financial transaction costs—for example, by facilitating the creation of new business platforms based on technological improvements—and strengthen institutions, including those that promote financial stability, are shown here to have the potential to admit more of the low-income segments of the population into the financial system.

1Household enterprises mostly sell services and internationally and locally produced consumer goods (used clothing, household supplies, and food). They also transform agricultural goods into processed grain and other goods, and natural resources into charcoal, bricks, and iron works. Others pursue artisanal work.
2Not surprisingly, public opinion surveys in sub-Saharan Africa consistently place jobs (or the lack thereof) among the most pressing concerns of the population. See, for example, the series of Afrobarometer opinion polls collected between 2002 and 2012. Jobs and income still outrank the next most pressing problem by almost 8 percentage points.
3We measure financial access performance for sub-Saharan African countries by the difference between the actual share of the adult population holding a bank account (actual financial access) and the level of financial access explained by their GDP per capita based on a simple ordinary least squares (OLS) regression.

    Other Resources Citing This Publication