## 2. Restarting Sub‑Saharan Africa’s Growth Engine

## Source details

**Canonical URL:** [2. Restarting Sub‑Saharan Africa’s Growth Engine](https://www.imf.org/-/media/files/publications/reo/afr/2017/may/pdf/sreo0517-chap2.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/afr/2017/may/pdf/sreo0517-chap2.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/afr/2017/may/pdf/sreo0517-chap2.pdf.json)

---

### Key findings and policy implications
- Transitioning from growth turnarounds to periods of sustained growth is difficult and often requires additional efforts beyond what is needed to trigger a growth up‑break.
- Main policy and structural factors that lengthen growth spells:
  - Consistency of macroeconomic policies, proxied by the level of inflation.
  - Sound fiscal policy through reduced debt‑to‑GDP ratios.
  - More outward‑oriented trade policies.
  - Efforts to induce higher investment‑to‑GDP ratios.
  - Improved macro‑structural policies, as captured by lower market distortions.
  - Better‑quality institutions, diversification, and technology adoption.
- Quantified effect: Improvements in each policy variable from its median value by 25 percentile points in the distribution (while keeping other variables constant) increases the length of the spell by one to six years.
- Policy implications (contextualized):
  - For countries in a severe growth slowdown:
    - Preserve macroeconomic stability to help trigger a turnaround and lead to sustained growth through consistent monetary policy to contain inflation, and sound fiscal policy to anchor public debt increases.
    - Foster investment, increased openness to trade, and more stable political environments.
  - For countries continuing to grow rapidly:
    - Rebuild buffers and stem increases in public debt through domestic revenue mobilization to finance a larger part of ambitious infrastructure projects while avoiding overheating.
  - For all countries:
    - Unlock growth potential through economic diversification, deepening technology adoption, financial reforms, and strengthening macro‑structural policies to reduce market distortions and investment risks and to improve the business climate.

### Stylized facts: turning points and growth spells in Sub‑Saharan Africa
- Identification and counts:
  - Turning points identified using a variant of the Bai and Perron (1998) methodology following Berg, Ostry, and Zettelmeyer (2012).
  - About 25 percent of total up‑breaks and down‑breaks identified over 1950–2016 correspond to sub‑Saharan Africa (sample fraction about 28 percent).
  - The region recorded 40 growth spells between 1950 and 2016—about 30 percent of total spells identified over the sample period.
  - Of the 40 growth spells in the region, 25 are ongoing (as of end of sample).
- Temporal patterns:
  - 1950–99: high volatility and frequent shifts; up‑breaks most common during 1985–99; down‑breaks most common during 1970–84.
  - 2000–16: less volatile growth, fewer down‑breaks, similar number of up‑breaks.
  - Growth spells were more concentrated after 1995, reflecting an improved business and macroeconomic environment, increased trade integration, higher average commodity prices, and accommodative global financial conditions.
- Resource intensity and ongoing spells:
  - Resource‑intensive and nonresource‑intensive countries experienced a roughly equal number of growth spells (23 and 17 respectively).
  - Of ongoing growth spells: two are among oil exporters, six among other resource‑intensive countries, and 17 among nonresource‑intensive countries.
  - Country examples with ongoing spells and forecast in 2017 to register growth at 5 percent or higher: Côte d’Ivoire, Ethiopia, Kenya, Rwanda, Tanzania.

### Duration and volatility of spells
- Median lengths and comparisons:
  - Median length of complete growth spells in sub‑Saharan Africa: six years (six years for resource‑intensive countries, and five years for nonresource‑intensive countries).
  - Median length for other regions: 10 years for other developing economies and eight years for emerging market economies.
  - Only one out of three spells in the region lasts at least 10 years compared to more than half in the rest of the developing world and about 80 percent in industrialized economies.
- Large swings around spells:
  - Median annual growth rate before a spell: −4 percent.
  - Median annual growth rate during a spell: 9 percent.
  - Median annual growth rate after the end of the spell: −4 percent.
  - Excluding conflict countries: about −4 percent before spells, about 8 percent during spells, and about −6 percent after spells.
  - Average annual growth after the end of a complete spell in sub‑Saharan Africa falls from close to 10 percent during spells to −6 percent, compared with −1 percent in other developing economies and −2 percent among emerging market economies.

### Exogenous factors around turning points
- Terms of trade and external demand:
  - Growth up‑breaks tend to coincide with improvements in countries’ terms of trade; the opposite holds for down‑breaks (symmetry test confirmed).
  - Stronger growth in trading partners is significantly associated with up‑breaks.
  - Examples of adverse terms‑of‑trade episodes coinciding with down‑breaks: Nigeria in 2010, Equatorial Guinea and the Republic of Congo in 2011.
- Global liquidity:
  - Growth up‑breaks in the region are associated with looser global liquidity conditions, represented by declines in the U.S. federal funds rate of about 2.5 percentage points on average (early 2010s observation).
  - No significant association between changes in the U.S. federal funds rate and down‑breaks.
- Oil prices:
  - Unlike other emerging and developing regions, growth breaks in the region are not significantly associated with changes in oil prices on average (offsetting effects for oil exporters vs. importers).

### Macroeconomic indicators, fiscal and exchange‑rate developments at breaks
- Associations at up‑breaks:
  - Lower inflation.
  - Significant increases in investment.
    - Examples: Zambia in 1999; Equatorial Guinea in 1994.
  - More trade openness.
  - Stronger net inflows of foreign direct investment.
    - Examples: Democratic Republic of Congo in 2009; Zambia in 1999; São Tomé and Príncipe in 1995.
  - Increases in the ratio of credit to the private sector to GDP (not shown in figure).
- Fiscal developments:
  - Up‑breaks accompanied by higher government revenues and better fiscal balances (reflecting better growth and revenue mobilization).
    - Examples: Ethiopia in 2003; Angola in 2000–01; Rwanda and Tanzania in 1994.
  - Down‑breaks coincide with higher debt ratios; large increases in debt‑to‑GDP ratios recorded during downturns.
  - Aid flows rise significantly during down‑breaks, suggesting a strong international response.
- Exchange rate:
  - Growth down‑breaks tend to be associated with more flexible exchange rate regimes, potentially reflecting policy responses that allow more exchange rate flexibility to adjust to a growth deceleration.

### Institutions, social indicators, and structural transformation
- Institutions and conflict:
  - Growth up‑breaks coincide with reductions in political risks, fewer internal conflicts, and improved law and order.
  - Armed conflicts significantly disrupt economic activity and shorten the duration of growth spells in SSA; the number of such conflicts has dramatically decreased over the last two decades.
- Social indicators:
  - Growth up‑breaks tend to be associated with significant declines in inequality (Gini coefficient).
  - Higher levels of education and human capital are associated with longer growth spells; reduction in infant mortality is associated with longer spells.
  - Higher income inequality leads to shorter spells (effect not statistically significant in this sample).
- Structural change and technology:
  - Growth up‑breaks coincide with significant increases in total factor productivity (TFP) growth.
  - Slower adoption of technology (proxied by hampered growth of mobile cell phone lines) is present during down‑breaks.
  - Export diversification, especially higher share of manufacturing exports, supports resilience of growth episodes in SSA (examples: Tanzania, Kenya; Botswana, Cabo Verde, Lesotho, Mauritius increased share of manufactures in exports while sustaining growth).
  - Imports of computers and ICT goods do not show a significant effect for SSA (likely due to still‑low levels and limited data).

### Determinants of growth‑spell duration — one‑by‑one findings
- Exogenous shocks:
  - Tighter global financial conditions (higher U.S. interest rates) significantly increase the probability that growth spells end in SSA and in the rest of the world.
  - Fluctuations in oil prices do not appear to have a significant impact on spells in the region on average.
- Macroeconomic indicators:
  - A better fiscal balance significantly increases the chance that a growth spell will continue.
  - A higher debt burden can accelerate the end of a growth spell (HIPC debt reduction historically provided fiscal space and supported sustainable growth in many countries; example: Tanzania).
  - Reductions in government revenue‑to‑GDP or investment‑to‑GDP increase the probability that a spell will end.
  - Real exchange rate misalignment and extreme misalignment affect the probability that spells end.
  - Trade openness, aid as percent of GDP, and inflation (Log (1+ inflation)) have measurable associations with the probability that a growth spell will end.
  - De facto exchange rate regime flexibility is associated with changes in the probability that a growth spell will end (higher values indicate more flexibility).

### Investment, macroeconomic competitiveness, and institutions in multivariate analysis
- Investment:
  - Investment plays a strong role in warding off growth stops and sustaining growth spells by supporting aggregate demand and expanding productive capacity.
  - Higher investment‑to‑GDP ratio is found to prolong growth spells in SSA in multivariate analysis.
  - Quantified: acceleration in change of investment‑to‑GDP (improvement of 25 percentile points from median) increases expected length by about 1.3 years.
- Trade openness and exchange‑rate policy:
  - Higher trade openness increases duration of growth spells (stronger positive effect in SSA).
    - Quantified: increase in openness (25 percentile points) increases expected length by six years.
  - A more flexible exchange rate regime and less overvalued exchange rate levels seem to extend spells (coefficients not always significant for SSA).
- Debt, inflation, and market distortions:
  - Reduction in total public debt to GDP (25 percentile points improvement) increases expected length by about one year.
  - Reduction in inflation (25 percentile points improvement) increases expected length by 5.4 years.
  - Reduction in the ratio of the relative price of investment to consumption (reduced market distortions) increases expected length by 2.5 years.
- External factors and institutions:
  - Improvement in percent change of terms of trade (25 percentile points) increases expected length by 3.4 years.
  - Increase in oil price (25 percentile points) associated with increase in expected length by 5.7 years.
  - Improvements in Polity 2 rating (25 percentile points) increases expected spell duration by 7.7 years.
- Summary interpretation:
  - Improvements by 25 percentile points in policy variables from their median values increase spell length by one to six years; a similar percentile improvement in institutional environment or exogenous factors raises spell length by four to eight years.
  - Some determinants operate differently in SSA versus other emerging and developing economies: higher investment‑to‑GDP ratio and positive terms‑of‑trade developments prolong growth spells only in SSA in this analysis; trade openness has a stronger positive effect in SSA; macro‑structural distortions do more harm in SSA.

### Selected multivariate hazard‑ratio results (exponentiated coefficients; hazard ratio > 1 shortens spells; < 1 lengthens spells)
- Dataset counts:
  - Observations: 326 (SSA), 1,076 (EMEDEV), 750 (EMEDEV without SSA).
  - Subjects: 288 (SSA), 153 (EMEDEV).
  - Failures: 829 (SSA), 21 (EMEDEV).
  - α: 3.956 (SSA), 1.557 (EMEDEV), 1.342 (EMEDEV without SSA).
  - p‑value: 0.00 (all reported columns).
- Selected hazard ratios and significance:
  - U.S. interest rates: 1.021 (SSA), 1.046 (EMEDEV), 1.037 (EMEDEV without SSA).
  - Oil price, change: 0.935 *** (SSA), 1.005 (EMEDEV), 1.032 (EMEDEV without SSA).
  - Terms of trade growth: 0.959 ** (SSA), 0.996 (EMEDEV), 0.996 (EMEDEV without SSA).
  - Investment to GDP, change: 0.923 * (SSA), 0.962 (EMEDEV), 0.953 (EMEDEV without SSA).
  - Total public debt to GDP: 1.007 ** (SSA), 1.006 ** (EMEDEV), 1.007 (EMEDEV without SSA).
  - Trade openness: 0.98 * (SSA), 0.992 (EMEDEV), 0.994 (EMEDEV without SSA).
  - Capital account openness: 1.359 (SSA), 1.071 (EMEDEV), 1.086 (EMEDEV without SSA).
  - Inflation: 3.242 ** (SSA), 1.472 * (EMEDEV), 1.527 * (EMEDEV without SSA).
  - De facto exchange rate regime: 0.756 (SSA), 0.81 ** (EMEDEV), 0.849 (EMEDEV without SSA).
  - Polity 2, cumulative change: 0.815 ** (SSA), 0.963 (EMEDEV), 1.003 (EMEDEV without SSA).
  - Human capital: 0.159 (SSA), 0.464 (EMEDEV), 0.355 (EMEDEV without SSA).
  - Real GDP per capita growth: 1.115 *** (SSA), 1.049 *** (EMEDEV), 1.064 ** (EMEDEV without SSA).
  - Initial GDP per capita: 4.691 (SSA), 1.531 (EMEDEV), 1.052 (EMEDEV without SSA).
  - Relative price of investment to consumption: 1.015 ** (SSA), 0.994 (EMEDEV), 0.982 *** (EMEDEV without SSA).
  - Decade dummy 2000: 1.380 (SSA), 0.547 (EMEDEV), 0.424 (EMEDEV without SSA).
- Statistical significance legend: *, **, and *** indicate significance at the 10, 5, and 1 percent levels respectively.

### Policy considerations and concluding remarks
- Diagnosis:
  - Resource‑intensive countries in SSA have been severely affected by the slump in commodity prices.
  - Global environment expected to provide little support for growth: growth rebalancing in China, commodity prices expected to remain low for a long period, aid flows expected to become more scarce, and growing risks of inward‑looking policies globally.
  - Implication: impetus to revive and sustain growth must come from domestic macroeconomic and structural policies that support self‑sustaining sources of growth.
- Priorities for countries with marked slowdowns:
  - Maintain macroeconomic stability: improve fiscal balances through targeted cuts and durable revenue measures; ensure monetary policy consistent with controlling inflation.
  - Strengthen macro‑structural policies that foster investment and promote openness and political stability.
- Priorities for countries currently enjoying growth spells:
  - Rebuild buffers in favorable times to stem increases in public debt; pursue gradual fiscal consolidation where warranted.
  - Strengthen monetary policy frameworks with price stability mandates.
- Cross‑cutting measures for all countries:
  - Foster educational development and human capital.
  - Advance economic diversification and increase the share of manufactures in total exports.
  - Deepen technology adoption (ICT goods imports), strengthen financial liberalization, reduce market distortions, and improve the business climate.
  - Promote regional and international trade integration.
- Empirical summary:
  - Sustaining longer spells and avoiding hard landings requires a combination of improved domestic macroeconomic policies, better political institutions, reduced market distortions, higher investment, and favorable external conditions working together.

*Source: sreo0517-chap2 - 2. Restarting Sub‑Saharan Africa’s Growth Engine*

### 2. Restarting Sub-Saharan Africa’s Growth Engine

### 2. Restarting Sub‑Saharan Africa’s Growth Engine

### Key findings and policy implications
- Transitioning from growth turnarounds to periods of sustained growth is difficult and often requires additional efforts beyond what is needed to trigger a growth up‑break.
- Beyond an improved external environment, better macroeconomic policies play a specific and important role in lengthening the duration of a growth spell. Main factors identified:
  - Consistency of macroeconomic policies, proxied by the level of inflation.
  - Sound fiscal policy through reduced debt‑to‑GDP ratios.
  - More outward‑oriented trade policies.
  - Efforts to induce higher investment‑to‑GDP ratios.
  - Improved macro‑structural policies, as captured by lower market distortions.
  - Better‑quality institutions, diversification, and technology adoption.
- Quantified effect: Improvements in each policy variable from its median value by 25 percentile points in the distribution (while keeping other variables constant) increases the length of the spell by one to six years.
- Policy implications (in the context of the current environment):
  - For countries in a severe growth slowdown: preserve macroeconomic stability to help trigger a turnaround and lead to sustained growth through consistent monetary policy to contain inflation, and sound fiscal policy to anchor public debt increases; foster investment, increased openness to trade, and more stable political environments.
  - For countries continuing to grow rapidly: focus on addressing potential vulnerabilities by rebuilding buffers and stemming increases in public debt through domestic revenue mobilization to finance a larger part of ambitious infrastructure projects while avoiding overheating.
  - For all countries: unlock growth potential through economic diversification, deepening technology adoption, financial reforms, and strengthening macro‑structural policies to reduce market distortions and investment risks and to improve the business climate.

### Stylized facts: turning points and growth spells in Sub‑Saharan Africa
- Growth turning points (up‑breaks and down‑breaks) are identified using a variant of the Bai and Perron (1998) methodology following Berg, Ostry, and Zettelmeyer (2012).
- Many turning points in sub‑Saharan Africa:
  - About 25 percent of total up‑breaks and down‑breaks identified over 1950–2016 correspond to sub‑Saharan Africa, in line with the fraction of sub‑Saharan African countries in the sample, about 28 percent.
  - Growth in the region during 1950–99 was characterized by high volatility and frequent shifts between expansion and contraction: up‑breaks most common during 1985–99 and most down‑breaks during 1970–84.
  - During 2000–16 the region experienced less volatile growth and fewer down‑breaks while maintaining about the same number of up‑breaks.
- Large break sizes and volatility:
  - On average, per capita growth in sub‑Saharan African countries was 10 percentage points higher after an up‑break compared to the period before it—the largest break size among all groups examined (higher than the 7 percent break size in other emerging and developing economies).
  - Countries in the region also experience large swings in growth following a down‑break, comparable with DEV MENA and DEV ASIA, but larger than in advanced economies.

### From turning points to growth spells
- Definitions used:
  - Complete growth spell: starts with an up‑break, is followed by an average rate of per capita growth of at least 2 percent, and ends with a down‑break (and a per capita growth rate of less than 2 percent).
  - Incomplete/ongoing growth spell: starts with an up‑break and is still ongoing as of the end of the sample (2016).
- Aggregate counts and patterns:
  - The region recorded 40 growth spells between 1950 and 2016—about 30 percent of total spells identified over the sample period.
  - Of the 40 growth spells in the region, 25 are ongoing.
  - Growth spells were more concentrated after 1995, reflecting an improved business and macroeconomic environment, increased trade integration, higher average commodity prices, and accommodative global financial conditions.
  - More than two‑thirds of countries enjoyed 10 or more years of uninterrupted growth; resource‑intensive and nonresource‑intensive countries experienced a roughly equal number of growth spells (23 for the former group, 17 for the latter).
  - Of the ongoing growth spells: two are among oil exporters, six among other resource‑intensive countries, and 17 among nonresource‑intensive countries.
  - Country examples with ongoing spells and forecast in 2017 to register growth at 5 percent or higher: Côte d’Ivoire, Ethiopia, Kenya, Rwanda, Tanzania.

### Duration and volatility of spells
- Shorter spells than elsewhere:
  - Median length of complete growth spells in sub‑Saharan Africa: six years (six years for resource‑intensive countries, and five years for nonresource‑intensive countries).
  - Median length for other regions: 10 years for other developing economies and eight years for emerging market economies.
  - Only one out of three spells in the region lasts at least 10 years compared to more than half in the rest of the developing world and about 80 percent in industrialized economies.
- Large swings before, during, and after spells:
  - Spells in the region tend to start from worse positions and typically end in “hard landings”:
    - Median annual growth rate before a spell: −4 percent.
    - Median annual growth rate during a spell: 9 percent.
    - Median annual growth rate after the end of the spell: −4 percent.
  - Excluding conflict countries, median annual growth rates remain about −4 percent before spells, about 8 percent during spells, and about −6 percent after spells.
  - When looking at average growth numbers, average annual growth after the end of a complete spell in sub‑Saharan Africa plummets (from close to 10 percent during the growth spells) to −6 percent, as opposed to −1 percent in other developing economies and −2 percent among emerging market economies.

### What happens around growth turning points
- The chapter examines changes around breaks in growth for variables representative of:
  - The external environment.
  - Macroeconomic policies.
  - Structural features.
  - Political and institutional characteristics.
- Robustness checks indicate that natural disasters, epidemics, or HIPC debt relief do not show systematic relationships with breaks in growth or the beginning or end of spells.

*Source: sreo0517-chap2 - 2. Restarting Sub‑Saharan Africa’s Growth Engine*

### 2. RESTARTING SUB-SAHARAN AFRICA’S GROWTH ENGINE

### 2. RESTARTING SUB-SAHARAN AFRICA’S GROWTH ENGINE

### Exogenous Factors
- Growth up-breaks tend to coincide with improvements in the countries’ terms of trade; the opposite holds for down-breaks (statistical test for symmetry between up- and down-breaks confirmed).
  - Worsening terms of trade following the global financial crisis coincided with down-breaks in Nigeria in 2010 and in Equatorial Guinea and the Republic of Congo in 2011.
  - Stronger growth in trading partners is significantly associated with up-breaks.
- Growth up-breaks in the region are associated with looser global liquidity conditions, represented by declines in the U.S. federal funds rate of about 2.5 percentage points on average (as observed in the early 2010s).
  - No significant association between changes in the U.S. federal funds rate and down-breaks.
- Unlike in the rest of the emerging and developing economies, growth breaks in the region are not significantly associated with changes in oil prices (likely because sub-Saharan Africa includes both oil exporters and importers, for which oil-price effects operate in opposite directions).

_Source examples and methods noted: approach follows Jones and Olken (2008); changes in means and symmetry tests are reported (see Annex 2.1)._

### Macroeconomic Indicators (Associations at Growth Breaks)
- Growth up-breaks coincide with:
  - Lower inflation.
  - Significant increases in investment.
    - Examples: Zambia in 1999; Equatorial Guinea in 1994.
  - More trade openness.
  - Stronger net inflows of foreign direct investment.
    - Examples: Democratic Republic of Congo in 2009; Zambia in 1999; São Tomé and Príncipe in 1995.
  - Increases in the ratio of credit to the private sector to GDP (not shown in figure).
- Fiscal developments at growth breaks:
  - Up-breaks accompanied by higher government revenues and better fiscal balances (likely reflecting both better growth and revenue mobilization efforts).
    - Examples: Ethiopia in 2003; Angola in 2000–01; Rwanda and Tanzania in 1994 coincided with improved fiscal balances.
  - Down-breaks coincide with higher debt ratios; large increases in debt-to-GDP ratios recorded during downturns.
  - Aid flows toward the region rise significantly during down-breaks, suggesting a strong international response.
- Exchange rate developments:
  - Growth down-breaks tend to be associated with more flexible exchange rate regimes, potentially reflecting policy responses that allow more exchange rate flexibility to adjust to a growth deceleration.

### Institutions, Social Indicators, and Structural Transformation
- Growth up-breaks coincide with improvements in the institutional environment:
  - Reduction in political risks, fewer internal conflicts, and improved law and order.
  - Growth up-breaks tend to be associated with significant declines in inequality (Gini coefficient).
- Structural factors:
  - Growth up-breaks coincide with significant increases in the rate of growth of total factor productivity (TFP).
  - Down-breaks generally coincide with deterioration in these structural factors.
  - Slower adoption of technology (represented by hampered growth of mobile cell phone lines) is present during down-breaks.
- Note: institutional and social variables are slow-moving; there is a general declining trend in political risks and conflicts which helps explain institutional improvements irrespective of up-breaks or down-breaks.

### Analyzing the Duration of Growth Spells
- Objective: relate the probability that a growth spell will end in the next period to changes in the identified variables, allowing for differences between sub-Saharan Africa and the rest of the world.
- Approach: two-step method due to limited spells and data length:
  1. Test relevance of each factor or policy one by one while controlling for initial conditions at the beginning of each spell (to account for growth convergence).
  2. Select a representative set of factors or policies from all groups while preserving sample size and estimate a multivariate specification to identify effects while controlling for other determinants.
- Methodological note: similar approach to Berg, Ostry, and Zettelmeyer (2012); external shock measures include changes in terms of trade, oil prices, and U.S. interest rates (composite rate combining effective federal funds rate with shadow measures to account for unconventional policies; see Krippner 2016).

### Determinants of Growth Spells: One by One (Key Findings)
- Exogenous shocks:
  - Tighter global financial conditions (higher U.S. interest rates) significantly increase the probability that growth spells end in sub-Saharan Africa as well as in the rest of the world (Figure 2.12).
  - On average, fluctuations in oil prices do not appear to have a significant impact on spells in the region (result may reflect offsetting effects for oil exporters vs. importers).
- Macroeconomic indicators and policies affecting spell duration:
  - A better fiscal balance significantly increases the chance that a growth spell will continue.
  - A higher debt burden can accelerate the end of a growth spell.
    - Historical context: substantial debt reduction after HIPC provided fiscal space to finance productive spending and supported sustainable growth in many countries.
    - Example: Tanzania’s strong growth from the early 1990s accompanied by improved public debt and stepped-up public expenditures.
  - Changes that reduce government revenue-to-GDP or investment-to-GDP increase the probability that a spell will end (Figure 2.13 shows changes in probability associated with a 1 unit change in these variables).
  - Real exchange rate misalignment and extreme real exchange rate misalignment affect the probability that spells end (measured effects shown in Figure 2.13).
  - Trade openness, aid as percent of GDP, and inflation (Log (1+ inflation)) have measurable associations with the probability that a growth spell will end.
  - De facto exchange rate regime flexibility is associated with changes in the probability that a growth spell will end (higher values indicate more flexibility).

_Final methodological note: the size of bars in Figures corresponds to the change in the probability (measured in units) that the growth spell will end associated with a 1 unit change in each variable. EMEDEV = all emerging and developing economies; EMEDEVxSSA = emerging and developing economies excluding sub-Saharan Africa; SSA = sub-Saharan Africa. Statistical significance indicated at the 10, 5, and 1 percent levels in figures._

*Source: Authors’ calculations.*

### 2. Government Revenue, and Total Investment

### 2. Government Revenue, and Total Investment

### Investment and Growth Spell Duration
- Investment plays a strong role in warding off growth stops and sustaining growth spells by supporting aggregate demand and expanding productive capacity.
- Examples: Growth spells in Chad and Lesotho in the 1980s were accompanied by a sharp increase in the investment-to-GDP ratio.
- The momentum includes large infrastructure investment across the region.
- Higher investment-to-GDP ratio is found to prolong growth spells in sub-Saharan Africa (SSA) in multivariate analysis.

### Macroeconomic Factors (competitiveness, trade openness, inflation, exchange rate)
- A more flexible exchange rate regime and less overvalued exchange rate levels seem to extend the length of growth spells (coefficients not significant for SSA in this sample).
- Higher trade openness tends to increase the duration of growth spells (not significant for global sample; stronger positive effect in SSA).
  - Example: Botswana experienced one of the longest growth spells; total trade represented 95 percent of GDP during the period.
- In the case of SSA, aid flows appear to significantly increase the duration of growth spells in the region (a result new to the literature).
- Higher inflation seems to decrease the duration of growth spells in the region (estimated coefficient not significant).

### Institutions, Conflict, and Human Development
- High institutional quality strongly supports growth spells in SSA and more so than in other regions (Polity 2 and related measures).
- Armed conflicts significantly disrupt economic activity and shorten the duration of growth spells in SSA; the number of such conflicts has dramatically decreased over the last two decades in the region.
- Social and human development: higher levels of education and human capital are associated with longer growth spells.
  - Reduction in infant mortality is associated with longer growth spells.
  - Higher income inequality leads to shorter spells (effect not statistically significant in this sample).
- Quantified impacts (from multivariate analysis, improvement of 25 percentile points from median, holding other factors constant):
  - Polity 2, cumulative change: increases expected spell duration by 7.7 years.

### Structural Transformation and Technology Adoption
- Export diversification, especially higher share of manufacturing exports, supports resilience of growth episodes in SSA.
  - Countries with high product diversification (e.g., Tanzania and Kenya) are enjoying ongoing growth spells.
  - Botswana, Cabo Verde, Lesotho, and Mauritius increased share of manufactures in exports while experiencing sustained growth episodes.
- Higher financial liberalization (index reflecting interest rate restrictions, credit controls, competition restrictions, state ownership, banking supervision) is associated with longer growth spells.
- Relative price of investment to consumption goods (proxy for market distortions) is associated with shorter spells (coefficients not statistically significant).
- Imports of computers and ICT goods (adoption of foreign technology) do not show a significant effect for SSA (likely due to still-low levels of such imports and limited data).

### Determinants of Growth Spells: Joint Multivariate Findings
- Full model includes: U.S. interest rate, change in the terms of trade, oil price change; trade openness, capital account openness, inflation, exchange rate regime, debt-to-GDP ratio, change in investment rate; human capital; Polity 2; measure of market distortions; average growth during spell; indicator for 2000–08.
- Results indicate duration of growth spells in SSA is supported by:
  - Improved monetary and fiscal policies
  - Better political institutions
  - Less market distortions
  - Better external environment
- Specific quantified effects (change in expected duration if a variable improves by 25 percentile points from median while other variables constant):
  - Acceleration in change of investment-to-GDP: increases expected length by about 1.3 years.
  - Reduction in total public debt to GDP: increases expected length by about one year.
  - Increase in openness (trade openness): increases expected length by six years.
  - Reduction in inflation: increases expected length by 5.4 years.
  - Reduction in the ratio of the relative price of investment to consumption (reduced market distortions): increases expected length by 2.5 years.
  - Improvement in percent change of terms of trade: increases expected length by 3.4 years.
  - Increase in oil price: associated with increase in expected length by 5.7 years.
  - Improvements in Polity 2 rating: increases duration by 7.7 years.
  - Reduction in GDP per capita growth (sign of overheating): increases expected duration by 2.8 years (interpreted as signs of overheating decrease expected duration; here a reduction in growth increases duration).

- Summary interpretation:
  - Improvements by 25 percentile points in policy variables from their median values increase spell length by one to six years; a similar percentile improvement in institutional environment or exogenous factors raises spell length by four to eight years.
  - Some determinants operate differently in SSA versus other emerging and developing economies: higher investment-to-GDP ratio and positive terms-of-trade developments prolong growth spells only in SSA in this analysis; trade openness has a stronger positive effect in SSA; macro-structural distortions do more harm in SSA.

### Policy Considerations and Concluding Remarks
- Growth turnarounds and sustained growth have become more common in SSA since 2000, but growth spells in the region tend to last shorter, show larger swings, and often end in “hard landings.”
- Sustaining spells for longer and avoiding hard landings requires a combination of factors and policies, not a single driving source.
- Domestic macroeconomic policies are critical, including:
  - Sound fiscal policy to prevent excessive public debt accumulation.
  - Monetary policy geared toward low inflation.
  - Outward-oriented trade policies to exploit global opportunities.
  - Macro-structural policies to reduce market distortions, unleash private-sector potential, and boost investment.
- External environment (terms of trade, global financial conditions) and institutional quality improvements have clearly helped restart and sustain growth but must be complemented by domestic policies.

*Source: Authors’ calculations.*

### 2. RESTARTING SUB-SAHARAN AFRICA’S GROWTH ENGINE

### 2. RESTARTING SUB-SAHARAN AFRICA’S GROWTH ENGINE

### Context and main diagnosis
- Resource-intensive countries in sub-Saharan Africa have been severely affected by the slump in commodity prices.
- Global environment expected to provide little support for growth in the region: growth rebalancing in China toward less resource-intensive sectors, commodity prices expected to remain low for a long period of time, aid flows expected to become more scarce, and growing risks of inward-looking policies across the globe.
- Implication: the impetus to revive growth where it has faltered, and sustain growth where it has remained relatively strong, must come from inside—macroeconomic and structural policies that support self-sustaining sources of growth.

### Policy priorities for countries where growth has slowed markedly
- Maintain macroeconomic stability and set the stage for a growth turnaround through a comprehensive, mutually reinforcing set of policies:
  - Improve fiscal balances through targeted cuts and measures that durably increase revenue to contain the increase in public debt.
  - Ensure monetary policy is consistent with keeping inflation under control.
  - Strengthen macro-structural policies that foster investment and help recover growth.
  - Promote openness and integration with the global economy.
  - Improve human capital and more stable political environments.
- Empirical claim: all these policies are strongly associated with growth turnarounds and sustained growth spells.

### Policy priorities for countries currently enjoying a growth spell
- Focus on prolonging growth and avoiding a hard landing:
  - Rebuild buffers in comparatively favorable times to stem increases in public debt; the analysis shows this is an important factor to sustain the length of growth episodes.
  - Some gradual fiscal consolidation is warranted, especially in the current environment of tighter and more volatile financial markets.
  - Strengthen monetary policy frameworks to ensure objectives are centered around a price stability mandate.

### Cross-cutting measures to unlock growth potential (for all countries)
- Complement macroeconomic policies with efforts to unlock countries’ growth potential:
  - Foster educational development and invest in the development of human capital.
  - Advance economic diversification to increase resilience to shocks; increase diversification of exports in terms of partners and products.
  - Increase the share of manufactures in total exports.
  - Deepen technology adoption through imports of information technology goods (ICT goods imports, imports of computers).
  - Strengthen financial liberalization reforms and reduce risks associated with investment.
  - Promote regional and international trade integration.
  - Invigorate macro-structural policies to reduce market distortions.
  - Strengthen governance frameworks and the business climate to attract investment towards new sectors of growth.
- Empirical evidence: specific measures like increasing manufactured exports, ICT goods imports, and improving investment risk indicators are associated with sustained growth in the analysis.

### Methodology — Structural breaks and growth spells
- Structural breaks identified using a variant of Bai and Perron (1998) as in Antoshin, Berg, and Souto (2008).
  - Parameters used: minimum number of years between potential breaks h = 5, and critical value p = 0.10.
  - Identified 184 up-breaks and 185 down-breaks in economic growth in 146 countries over the period 1950–2016.
  - Because of h = 5, the first and last five years of the estimation sample cannot contain breaks; with 66 years of data, up to 12 possible structural breaks, last possible break in 2011.
  - Breaks grouped by income level and region (advanced, emerging, developing; developing disaggregated into Asia, Middle East and North Africa, sub-Saharan Africa, and Latin America).
- Definition of growth spells (following Berg, Ostry, and Zettelmeyer (2012)):
  - “Complete” growth spells: (1) begin with a growth up-break followed by at least 2 percent average per capita income growth; and (2) end with a growth down-break followed by less than 2 percent average per capita income growth.
  - “Incomplete” growth spells: satisfy condition (1) and are ongoing at the end of the sample.
  - Identified a total of 80 complete and 67 incomplete spells for the full sample.

### Methodology — Changes in variables at turning points
- Use one-sample t-tests (Jones and Olken (2008)) to examine changes in means of variables between periods before and after breaks, for both up-breaks and down-breaks.
- Symmetry tests (two-sample t-tests) evaluate whether mean changes during up-breaks have the same magnitude as those during down-breaks, allowing for different samples and variances (Satterthwaite approximation for degrees of freedom).

### Methodology — Duration analysis of growth spells
- Survival (duration) models used to study how duration of growth episodes varies with covariates.
  - Survival function S(t) = 1 − F(t), with hazard rate defined as the instantaneous probability that a spell ends conditional on survival up to time t.
  - Proportional hazard model assumed, where covariates are multiplicatively related to the hazard at all horizons.
  - Baseline hazard specified as a Weibull distribution.
  - Empirical implementation uses lagged values of covariates to avoid feedback from duration to covariates.

### Duration analysis — Selected multivariate results (hazard ratios and interpretation)
- Dataset counts:
  - Observations: 326 (SSA), 1,076 (EMEDEV), 750 (EMEDEV without SSA)
  - Subjects: 288 (SSA), 153 (EMEDEV), (no separate subject count shown for EMEDEV without SSA)
  - Failures: 829 (SSA), 21 (EMEDEV)
  - α: 3.956 (SSA), 1.557 (EMEDEV), 1.342 (EMEDEV without SSA)
  - p-value: 0.00 (all reported columns)
- Interpretation note: table reports exponentiated regression coefficients (“hazard ratios”): hazard ratio > 1 implies increases in the associated variable shorten spells; hazard ratio < 1 implies a “protective” effect that helps sustain the spell.
- Selected estimated hazard ratios (exponentiated coefficients) and statistical significance:
  - U.S. interest rates: 1.021 (SSA), 1.046 (EMEDEV), 1.037 (EMEDEV without SSA)
  - Oil price, change: 0.935 *** (SSA), 1.005 (EMEDEV), 1.032 (EMEDEV without SSA)
  - Terms of trade growth: 0.959 ** (SSA), 0.996 (EMEDEV), 0.996 (EMEDEV without SSA)
  - Investment to GDP, change: 0.923 * (SSA), 0.962 (EMEDEV), 0.953 (EMEDEV without SSA)
  - Total public debt to GDP: 1.007 ** (SSA), 1.006 ** (EMEDEV), 1.007 (EMEDEV without SSA)
  - Trade openness: 0.98 * (SSA), 0.992 (EMEDEV), 0.994 (EMEDEV without SSA)
  - Capital account openness: 1.359 (SSA), 1.071 (EMEDEV), 1.086 (EMEDEV without SSA)
  - Inflation: 3.242 ** (SSA), 1.472 * (EMEDEV), 1.527 * (EMEDEV without SSA)
  - De facto exchange rate regime: 0.756 (SSA), 0.81 ** (EMEDEV), 0.849 (EMEDEV without SSA)
  - Polity 2, cumulative change: 0.815 ** (SSA), 0.963 (EMEDEV), 1.003 (EMEDEV without SSA)
  - Human capital: 0.159 (SSA), 0.464 (EMEDEV), 0.355 (EMEDEV without SSA)
  - Real GDP per capita growth: 1.115 *** (SSA), 1.049 *** (EMEDEV), 1.064 ** (EMEDEV without SSA)
  - Initial GDP per capita: 4.691 (SSA), 1.531 (EMEDEV), 1.052 (EMEDEV without SSA)
  - Population growth: 1.073 (SSA), 1.004 (EMEDEV), 1.104 (EMEDEV without SSA)
  - Relative price of investment to consumption: 1.015 ** (SSA), 0.994 (EMEDEV), 0.982 *** (EMEDEV without SSA)
  - Decade dummy 2000: 1.380 (SSA), 0.547 (EMEDEV), 0.424 (EMEDEV without SSA)
- Statistical significance legend: *, **, and *** indicate significance at the 10, 5, and 1 percent levels respectively.

*Source: sreo0517-chap2 - 2. RESTARTING SUB-SAHARAN AFRICA’S GROWTH ENGINE (PDF chapter).*

---


_Source: https://www.imf.org/-/media/files/publications/reo/afr/2017/may/pdf/sreo0517-chap2.pdf_