## _wp14143

## Source details

**Canonical URL:** [_wp14143](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14143.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14143.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14143.pdf.json)

---

### Introduction and context
- Africa will account for 3.2 billion of the projected 4 billion increase in the global population by 2100 (Table 1).
- Africa's working age population will increase by 2.1 billion between 2010 and 2100, compared to a net global increase of 2 billion.
- With declining mortality and fertility, Africa’s share of the working age population will increase from about 54 percent in 2010 to peak at about 64 percent in 2090.
- Africa’s share of the global working age population is projected to increase from 12.6 percent in 2010 to over 41 percent by 2100.
- Definitions:
  - Working age population is defined as those aged between 15-64.
  - The share of working age population is the working age population divided by the total population.
- Source of population estimates: IMF staff estimates based on UN World Population Database.

### Key descriptive statistics and table 1 highlights
- Total population (bn):
  - World: 2010 = 6.9; 2100 = 10.9; Change = 4.0
  - Africa: 2010 = 1.0; 2100 = 4.2; Change = 3.2
- Working age population (bn):
  - World: 2010 = 4.5; 2100 = 6.5; Change = 2.0
  - Africa: 2010 = 0.6; 2100 = 2.7; Change = 2.1
- Africa/World (percent):
  - Total population: 2010 = 14.9; 2100 = 38.5
  - Working age population: 2010 = 12.6; 2100 = 41.2

### Research objective, model, and empirical strategy
- Objective: Provide estimates of the potential magnitude of the demographic dividend for Sub-Saharan Africa (SSA).
- Theoretical approach:
  - Simple general equilibrium overlapping generations (OLG) model (two-period Diamond-type) with production Y = A K^α L^{1-α} and log utility.
  - Steady-state capital-labor ratio: k* = [β(1–α)/(1+β)(1+n)]^{1/(1–α)}.
  - Calibration choices:
    - target interest rate: around 5 percent per annum;
    - one model period = 30 years;
    - initial population growth rate: 2.5 percent per annum;
    - quarterly discount rate: 0.99;
    - share of capital α = 1/3.
- Empirical approach:
  - Follows Bloom et al. (2010) with methodological improvements (country and year fixed effects based on Hausman test; wider panel estimation techniques).
  - Emphasizes role of human capital and tests pace/magnitude of transition and threshold effects (income and education).
  - Panel: 172 countries, 5-year intervals, 1960–2010, sourced from World Development Indicators (2012).
  - Main specification includes: initial working age share (WAS), change in WAS (ΔWAS), initial GDP per capita, trade openness, investment, sectoral change (share of agriculture), and education dummies.
  - Estimators: Fixed Effects (FE) baseline; two-step system GMM for robustness (ΔWAS and sectoral change treated as endogenous).

### Main empirical findings and quantitative estimates
- Confirmed general result:
  - "We confirm the unambiguous impact of demographic transitions on growth. A 1 percentage point change in the working age population increases real per capita GDP growth by" 0.5 percentage point, with the range varying between 0 and 1.1 percentage points.
- Baseline FE results (Table 2, column (1); 5-year growth):
  - Log Initial GDP per capita: -5.22*** (t-stat -6.33)
  - Trade Openness: 0.02*** (2.90)
  - Sectoral Change (Agr): -0.16*** (-3.30)
  - Log(WAS): 12.57*** (3.37)
  - Δ WAS: 0.53*** (3.75)
  - Observations: 1,100; Number of countries: 172; R-squared: 0.29
- System GMM (Table 2, column (2)):
  - Log(WAS): 14.05*** (6.97)
  - Δ WAS: 0.76*** (3.37)
  - AR(1) pval: 0.00; AR(2) pval: 0.14; Hansen J pval: 0.35; Nr. Instruments: 101
- Education interaction (Table 2, column (3)):
  - Δ WAS * High education: 0.31* (1.65)
  - Interpretation: Countries with higher education benefit roughly twice as much from the demographic dividend.

### Regional differences and human capital role
- Asia and Europe:
  - Show unambiguous demographic dividends; Asia’s rapid growth tied to leveraging demographic transition via human and physical capital investment and labor-intensive export-led growth.
- Latin America:
  - Did not fully harness the dividend; weaker policy environment and less investment cited.
- Sub-Saharan Africa:
  - Transition is later and longer; working age share peak is smaller and occurs later (peak around 2090).
  - Heterogeneity: five groups by transition start decade; majority (29 of 46 countries) started transition between 1981 and 2000 and are only about 25 percent through transition.
  - Low-education African countries experienced declines in growth following increases in working age population (Malthusian phase); investment in education reverses this effect.
  - Moving from first to second education quartile yields a net positive dividend for African countries.

### Threshold effects by income level
- Interaction coefficients (demonstrating diminishing demographic effects with higher income):
  - Log(WAS)*Log Initial GDP per capita: coefficient -4.59** (t-stat -2.22)
  - Δ WAS*Log Initial GDP per capita: coefficient -0.20** (t-stat -2.22)
- Marginal effect analysis:
  - Demographic variables’ contribution to growth becomes non-significant when GDP per capita reaches around US$5,162 (text presents US$5,162).
  - Interpretation: beyond roughly US$5,100 per capita, changing economic structure and reduced reliance on labor-intensive strategies lower the marginal contribution of demographic factors.

### Projections for Sub-Saharan Africa (2010–2100) and median-country scenario
- Definition: demographic dividend = GDP per capita with demographic transition minus GDP per capita without demographic transition.
- Scenario assumptions from 2010:
  - (i) Share of working age population remains constant at 50 percent.
  - (ii) Share of working age population rises as per UN median scenario.
  - Method: use coefficient estimates from FE baseline (column (1) of Table 1) to compute average 5-year growth rates; initial GDP per capita and trade openness taken from 2010; sectoral change from agriculture assumed zero.
- Median-country projection (initial per capita income around US$560 in 2010):
  - Share of working age population projected to peak at 65 percent in 2090.
  - Baseline GDP per capita peak (without demographic effect) around US$2,475.
  - Factoring in demographic developments increases GDP per capita to US$3,865 by 2100.
  - Implied demographic dividend: about US$1,350 for the median country.
  - US$3,865 is about 56 percent higher compared with the unchanged working age share scenario.
- Timing and uncertainty:
  - The demographic dividend initially accrues at a low pace and only takes off in the second half of the century.
  - Appendix 2 provides country specific estimates; uncertainty is much greater for these.

### Country-level peak dividends (selected per capita peak values, Constant 2005 USD)
- Representative peak per capita dividend values appearing in Appendix Figure 1 include: 1665, 1042, 1254, 1064, 1859, 1492, 1358, 1223, 1621, 1285, 1674, 1429, 1556, 1635, 1478, 1572, 1581, 1817, 1687, 1608, 1573, 1491, 1537, 1474, 1605, 1864, 1777, 1454, 2462, 1697, 1698, 1897, 1743, 1829, 2582, 2567, 3001, 2141, 2967, 4808, 517, 612, 615, 638, 653, 673, 717, 725, 758, 759, 760, 781, 785, 811, 822, 827, 854, 868, 877, 885, 893, 944, 969, 994, 998, 1007, 1025, 1037, 1051, 1059, 1072, 1100, 1119, 1128, 1200, 1249, 1314, 1391, 1496, 1913, 2100, 2056, 2044, 2067, 2100, 2084, 2097, 2065, 2093, 2070, 2100, 2084, 2085, 2088, 2089, 2074, 2098, 2095, 2079, 2090, 2070, 2085, 2090, 2085, 2084, 2047, 2095, 2085, 2085, 2095, 2069, 2095, 2090, 2090, 2094, 2052, 2070, 2095, 2050, 2070.
- Note: projections assume a baseline WAS of 50 percent in 2010.

### Policy implications and recommendations
- Prioritize investment in human capital (education and health) to magnify dividend effects.
- Emphasize policies that support labor-intensive export-led growth initially to create employment and facilitate transition into higher TFP sectors.
- Raise labor force participation, including for women, to maximize benefits from increases in labor supply.
- Strengthen policies that attract physical capital and promote structural transformation away from low-productivity agriculture.
- Prepare for longer-term shifts in income thresholds where demographic contributions wane by fostering productivity-enhancing investments and institutions.
- Address social risks by ensuring job creation keeps pace with labor market entrants to avoid rising unemployment and social tensions.

*Source: IMF working paper text provided in the content unit.*

### References .............................................................................................................

### _wp14143 - References

### Introduction and context
- Africa will account for 3.2 billion of the projected 4 billion increase in the global population by 2100 (Table 1).
- Africa's working age population will increase by 2.1 billion between 2010 and 2100, compared to a net global increase of 2 billion.
- With declining mortality and fertility, Africa’s share of the working age population will increase from about 54 percent in 2010 to peak at about 64 percent in 2090.
- Africa’s share of the global working age population is projected to increase from 12.6 percent in 2010 to over 41 percent by 2100.
- Definitions:
  - Working age population is defined as those aged between 15-64.
  - The share of working age population is the working age population divided by the total population.
- Source of population estimates: IMF staff estimates based on UN World Population Database.

### Key table: Table 1. Africa Rising (as reported)
- Total population (bn)
  - World: 2010 = 6.9; 2100 = 10.9; Change = 4.0
  - Africa: 2010 = 1.0; 2100 = 4.2; Change = 3.2
- Working age population (bn)
  - World: 2010 = 4.5; 2100 = 6.5; Change = 2.0
  - Africa: 2010 = 0.6; 2100 = 2.7; Change = 2.1
- Africa/World (percent)
  - Total population: 2010 = 14.9; 2100 = 38.5
  - Working age population: 2010 = 12.6; 2100 = 41.2

### Research focus and methodology
- Objective: Provide estimates of the potential magnitude of the demographic dividend for Sub-Saharan Africa (SSA).
- Theoretical approach: Develop a simple general equilibrium overlapping generations model to show potential growth effects predicted by theory from the demographic transition.
- Empirical approach:
  - Follows Bloom et al. (2010) but introduces methodological improvements.
  - Introduces country and year fixed effects (FE) on the basis of the Hausmann test.
  - Uses a wider range of panel estimation techniques to show robustness under various settings.
  - Emphasizes the role of human capital in explaining regional differences in the magnitude of the demographic dividend.
  - Investigates how the pace of transition and its magnitude matter for growth.
  - Considers whether thresholds exist with regard to income and education levels and how these matter for the dividend.
- Outputs: Initial estimates of the dividend that African countries can expect as a result of the transition.

### Scenarios and implications
- Three possible scenarios for countries undergoing demographic transition:
  - Positive: The window of opportunity is properly tapped, generating a “demographic dividend” from higher growth (as occurred in several countries in Asia).
  - Less benign: The window of opportunity passes without meaningful progress due to failure to implement enabling policies.
  - Worst case: The transition results in an army of unemployed youth, significantly increasing social risks and tensions.

### Figures and tables listed in the unit
- Figures referenced:
  - 1. Global trends in working age population
  - 2. Evolution of shares of working age population in Sub-Saharan Africa
  - 3. Diverging experiences in Asia and Latin America: 1965-2010
  - 4. Population growth and output
  - 5. Demographic dividend and development level
  - 6 (a). GDP growth in median SSA country
  - 6 (b). Demographic dividend and working age population
- Tables referenced:
  - 1. Africa rising
  - 2. Baseline results for 5-year growth
  - 3. Demographic dividends by regions
  - 4. Demographic dividends and threshold effects
- Appendix items:
  - Table 1: Summary statistics
  - Table 2: Demographic dividend by regions
  - Figure 1: Country specific demographic dividends in peak year

### Reported empirical emphasis and partial result
- The study reports five main findings.
- Reported (partial) finding:
  - "We confirm the unambiguous impact of demographic transitions on growth. A 1 percentage point change in the working age population increases real per capita GDP growth by" (text ends here in the provided content; no further numeric value is supplied).

*Source: Excerpt from _wp14143 - References (IMF staff content provided).*

### 0.5 percentage point, with the range varying between 0 and 1.1 percentage points

### _wp14143 - 0.5 percentage point, with the range varying between 0 and 1.1 percentage points

### Key Findings
- A 1 percentage point change in the working age population increases real per capita GDP growth by 0.5 percentage point, with the range varying between 0 and 1.1 percentage points depending on the region.
- Both the magnitude and speed of the demographic dividend depend on:
  - the country’s initial level of the working age share and its rate of change;
  - the income level.
- Investment in human capital is critical: countries with higher education levels benefit the most; education helps explain the Asian miracle.
- Demographic factors contribute significantly to the dividend only up to a certain income level (around US$5,100 per capita).
- Low income countries in Africa have the potential to benefit most from the nascent demographic transition.
- The median African country with an initial per capita income level of around US$550 in 2010 can expect a demographic dividend (beyond growth with unchanged demographic structure) of about US$1,350 by 2100, producing GDP per capita of US$3,865 — about 56 percent higher compared with a scenario of unchanged share of working age population.

### Literature and Theory
- Demographic transition: changes in age structure (rise in working age share) increase labor supply and growth potential, via higher output, savings, and investment.
- Demographic transitions often considered key driver of the Asian miracle; absence/delay in Africa cited as a growth constraint.
- Transition stages: before transition population growth can be negative for growth; during transition higher life expectancy accelerates growth by favoring human capital and TFP, raising savings and female labor participation.
- Window of opportunity: demographic dividend is realizable if good policies are in place; otherwise risks include rising unemployment and social pressures.
- International spillovers: aging countries with excess capital can transfer resources to countries undergoing transition.

### Economic Model (OLG framework) and Calibration
- Model: two-period overlapping generations (Diamond-type) with production Y = A K^α L^{1-α} and log utility; steady-state capital-labor ratio k* = [β(1–α)/(1+β)(1+n)]^{1/(1–α)}.
- Calibration choices:
  - target interest rate: around 5 percent per annum;
  - one model period = 30 years;
  - initial population growth rate: 2.5 percent per annum;
  - quarterly discount rate: 0.99;
  - share of capital α = 1/3.
- Implication: an increase in population growth rate increases labor force, lowers capital per worker and output per worker, but raises aggregate output.

### Empirical Strategy
- Decompose output per capita growth into growth of output per worker and growth of working age share (assuming constant participation).
- Growth of income per worker modeled with convergence speed and initial deviations from steady state.
- Panel: 172 countries, 5-year intervals, 1960–2010, sourced from World Development Indicators (2012).
- Main empirical specification includes: initial working age share (WAS), change in WAS (ΔWAS), initial GDP per capita, trade openness, investment, sectoral change (share of agriculture), and education dummies.
- Estimators used: Fixed Effects (FE) baseline (Hausman test), two-step system GMM for robustness (treating ΔWAS and sectoral change as endogenous).

### Baseline Empirical Results
- FE baseline (column (1) of Table 2):
  - Log Initial GDP per capita: -5.22*** (t-stat -6.33)
  - Trade Openness: 0.02*** (2.90)
  - Sectoral Change (Agr): -0.16*** (-3.30)
  - Log(WAS): 12.57*** (3.37)
  - Δ WAS: 0.53*** (3.75)
  - Observations: 1,100; Number of countries: 172; R-squared: 0.29
- System GMM (column (2) of Table 2) yields similar or slightly higher estimates:
  - Log(WAS): 14.05*** (6.97)
  - Δ WAS: 0.76*** (3.37)
  - AR(1) pval: 0.00; AR(2) pval: 0.14; Hansen J pval: 0.35; Nr. Instruments: 101
- Interaction with education (column (3) of Table 2):
  - Δ WAS * High education: 0.31* (1.65)
  - Countries with higher education benefit roughly twice as much from demographic dividend.

### Regional Differences and Role of Human Capital
- Asia and Europe show unambiguous demographic dividends; Asia’s rapid growth tied to making the most of demographic transition, investment in human and physical capital, higher labor participation, and labor-intensive export-led growth.
- Latin America did not fully harness the dividend; weaker policy environment and less investment cited.
- Sub-Saharan Africa:
  - Transition is later and longer; Africa’s working age share peak is smaller and occurs later (peak around 2090).
  - Heterogeneity across African countries: five groups by transition start decade; majority (29 of 46 countries) started transition between 1981 and 2000 and are only about 25 percent through transition.
  - For Africa, low-education countries experienced declines in growth following increases in working age population (Malthusian phase). Countries that invested in education reversed this effect and obtained positive dividends.
  - Moving from first to second education quartile yields net positive dividend for African countries.

### Threshold Effects (Demographic dividend vs. income level)
- Inclusion of interactions between demographic variables and log initial GDP per capita shows negative interaction terms:
  - Log(WAS)*Log Initial GDP per capita: coefficient -4.59** (t-stat -2.22)
  - Δ WAS*Log Initial GDP per capita: coefficient -0.20** (t-stat -2.22)
- Marginal effect analysis indicates demographic variables’ contribution to growth becomes non-significant when GDP per capita reaches around US$5,162 (presentation in text uses US$ 5,162).
- Interpretation: beyond roughly US$5,100 per capita, changing economic structure and less reliance on labor-intensive strategies reduce demographic factors’ marginal contribution.

### Estimates for Sub-Saharan Africa (2010–2100 scenarios)
- Definition: demographic dividend = GDP per capita with demographic transition minus GDP per capita without demographic transition.
- Scenarios modeled from 2010 start:
  - (i) Share of working age population remains constant at 50 percent.
  - (ii) Share of working age population rises as per UN median scenario.
- Methodology specifics:
  - Use coefficient estimates from column (1) of Table 1 (FE baseline) to compute average 5-year growth rates.
  - Initial GDP per capita level and trade openness taken from 2010.
  - Sectoral change from agriculture assumed zero (no further shift from agriculture to non-agriculture).
- Projection highlights (median country with initial per capita income ~US$560):
  - Share of working age population projected to peak at 65 percent in 2090.
  - Baseline GDP per capita peak (without demographic effect) around US$2,475.
  - Factoring in demographic developments increases GDP per capita to US$3,865 by 2100 (implied demographic dividend of about US$1,350 for the median country).
  - The US$3,865 level is about 56 percent higher compared with the unchanged working age share scenario.

### Policy Implications and Recommendations
- Translating Africa’s demographic potential into realized dividends requires appropriate policies:
  - Prioritize investment in human capital (education and health) to magnify dividend effects.
  - Emphasize policies that support labor-intensive export-led growth initially to create employment and facilitate transition into higher TFP sectors.
  - Raise labor force participation, including for women, to maximize benefits from increases in labor supply.
  - Strengthen policies that attract physical capital and promote structural transformation away from low-productivity agriculture.
  - Prepare for longer-term shifts in income thresholds where demographic contributions wane, by fostering productivity-enhancing investments and institutions.

*Italic source: IMF working paper text provided in the content unit.*

### 2100. The demographic dividend of about US$1,350 implies that per capita GDP is higher by

### _wp14143 - 2100. The demographic dividend of about US$1,350 implies that per capita GDP is higher by

### Main finding on the demographic dividend
- The demographic dividend is about US$1,350 and implies that per capita GDP is higher by nearly 56 percent.
- The demographic dividend initially accrues at a low pace and only takes off in the second half of the century.
- Appendix 2 provides country specific estimates, although uncertainty is much greater for these.

### Projections and graphical evidence (as reported)
- Figure 6(a) and Figure 6(b) present:
  - GDP per capita (Constant 2005 USD) with demographic transition and baseline trajectories for 2010 through 2100.
  - Working age population (% of total, RHS) trajectories for 2010 through 2100.
  - Time series span shown: 2010, 2015, 2020, 2025, 2030, 2035, 2040, 2045, 2050, 2055, 2060, 2065, 2070, 2075, 2080, 2085, 2090, 2095, 2100.
- Figures show demographic dividends measured in Constant 2005 USD alongside working age share (% of total, RHS) across the century.

### Conclusions and policy implications
- The projected increase in Africa’s working age population this century represents a window of opportunity that, if properly tapped, can be transformational.
- Growth can be substantially higher, with low income countries standing to benefit the most since the magnitude of the dividend declines as economies mature.
- Risks: the transition can create potential social risks if the economy does not create jobs on a sufficient scale to absorb those joining the labor market.
- Human capital interaction:
  - Significant interaction between human capital and the magnitude of the demographic dividend indicates that improving and increasing access to education is critical to improve worker productivity and support a transition to higher value added sectors.
- Structural transformation:
  - A shift away from agriculture is conducive to harnessing the demographic dividend.
  - Policies that remove impediments to private sector development and enable labor intensive manufacturing could help position Africa to make the most of its resources.

### Key summary statistics (Appendix Table 1)
- Average 5-year growth rate of real GDP per capita: Obs 1489; Mean 1.43; S.D. 3.26; Min -22.56; Max 26.90. Source: WDI (2012)
- Log Initial GDP per capita: Obs 1493; Mean 7.89; S.D. 1.61; Min 3.91; Max 11.75. Source: WDI (2012)
- Ratio of Investment to GDP: Obs 1438; Mean 0.11; S.D. 0.11; Min -0.23; Max 1.16. Source: PWT 8.0
- Trade Openness: Obs 1395; Mean 76.03; S.D. 48.86; Min 1.08; Max 429.95. Source: WDI (2012)
- Years of Schooling: Obs 1420; Mean 5.61; S.D. 3.03; Min 0.04; Max 12.91. Source: Barro and Lee (2013)
- Bureaucratic quality: Obs 659; Mean 2.13; S.D. 1.21; Min 0.00; Max 4.00. Source: ICRG (2012)
- Sectoral Change (Agriculture): Obs 1211; Mean -1.31; S.D. 4.45; Min -43.13; Max 26.22. Source: WDI (2012)
- Life Expectancy: Obs 1904; Mean 61.88; S.D. 11.70; Min 30.31; Max 81.98. Source: WDI (2012)
- Working age share: Obs 1908; Mean 57.91; S.D. 6.54; Min 45.27; Max 79.14. Source: WDI (2012) and UN
- Change in working age share: Obs 1908; Mean 0.60; S.D. 1.44; Min -6.27; Max 7.81. Source: WDI (2012) and UN

### Regional regression results (Appendix Table 2 — selected coefficients)
- Regressions reported by region with respective sample periods and sizes. Coefficients shown with significance indicators (***, **, *) and t-statistics in parentheses:
  - Asia (Period 1965-2010; Observations 287; Number of countries 53; R-squared 0.41)
    - Log Initial GDP per capita: -6.14***  (-3.02)
    - Sectoral Change (Agr): -0.27***  (-2.84)
    - Log(WAS): 19.19**  (2.56)
    - Δ WAS: 0.81***  (3.11)
    - Constant: -31.66  (-1.11)
  - Europe (Period 1965-2005; Observations 226; Number of countries 39; R-squared 0.75)
    - Log Initial GDP per capita: -13.06***  (-9.87)
    - Sectoral Change (Agr): 0.33***  (4.64)
    - Log(WAS): 41.39***  (4.50)
    - Δ WAS: 1.04***  (3.81)
    - Constant: -46.80  (-1.13)
  - Latin America (Period 1965-2010; Observations 204; Number of countries 31; R-squared 0.35)
    - Log Initial GDP per capita: -7.15***  (-5.38)
    - Trade Openness: 0.02  (1.36)
    - Sectoral Change (Agr): 0.02  (0.21)
    - Log(WAS): 1.52  (0.21)
    - Δ WAS: -0.25  (-1.19)
    - Constant: 52.08**  (2.07)
  - SSA (Period 1985-2010; Observations 197; Number of countries 44; R-squared 0.46)
    - Log Initial GDP per capita: -6.02**  (-2.49)
    - Trade Openness: 0.03  (1.40)
    - Sectoral Change (Agr): -0.18**  (-2.33)
    - Log(WAS): -34.27**  (-2.32)
    - Δ WAS: 0.02  (0.08)
    - Constant: 176.34***  (3.31)
- Note: Robust t-statistics in parentheses. ***, ** and * denote significance at the 1%, 5%, and 10% levels, respectively. All regressions include time fixed effects.

### Country-level peak dividends (Appendix Figure 1 notes)
- The projected dividend for countries at their peak is based on a baseline WAS of 50 percent in 2010. Countries are ranked by the size of the demographic dividend per capita in their peak year.
- Representative peak per capita dividend values (Constant 2005 USD) appearing in the figure include: 1665, 1042, 1254, 1064, 1859, 1492, 1358, 1223, 1621, 1285, 1674, 1429, 1556, 1635, 1478, 1572, 1581, 1817, 1687, 1608, 1573, 1491, 1537, 1474, 1605, 1864, 1777, 1454, 2462, 1697, 1698, 1897, 1743, 1829, 2582, 2567, 3001, 2141, 2967, 4808, 517, 612, 615, 638, 653, 673, 717, 725, 758, 759, 760, 781, 785, 811, 822, 827, 854, 868, 877, 885, 893, 944, 969, 994, 998, 1007, 1025, 1037, 1051, 1059, 1072, 1100, 1119, 1128, 1200, 1249, 1314, 1391, 1496, 1913, 2100, 2056, 2044, 2067, 2100, 2084, 2097, 2065, 2093, 2070, 2100, 2084, 2085, 2088, 2089, 2074, 2098, 2095, 2079, 2090, 2070, 2085, 2090, 2085, 2084, 2047, 2095, 2085, 2085, 2095, 2069, 2095, 2090, 2090, 2094, 2052, 2070, 2095, 2050, 2070.
- Note: Countries are ordered and labeled in the figure; the baseline WAS used for the projection is 50 percent in 2010.

*Source: Authors’ calculations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14143.pdf_