## wpiea2022088-print-pdf

## Source details

**Canonical URL:** [wpiea2022088-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022088-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022088-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022088-print-pdf.pdf.json)

---

### Shapley decompositions and predictors (general)
- The Shapley decomposition of projected year-on-year real GDP growth in the quarter of interest is provided for country cases.
- Nigeria Shapley decomposition note:
  - Sample Av. = Sample average.
  - KENPMI = Kenya’s PMI.
  - COVIDDUM = COVID-19 dummy.
  - MATP = Materials price index.
  - GHAPMI = Ghana’s PMI.
  - xZAFSTOCK = Stock market index in selected sub-Saharan African countries excluding South Africa.
  - INDXNGA = India’s exports to Nigeria.
  - NGAPMI = Nigeria’s PMI.
  - CHNPMI = China’s PMI.
  - ZAFNVS = South Africa’s new vehicles sold.
  - ZAFSTOCK = South Africa’s stock market index.
- Botswana Shapley decomposition note:
  - Sample Av. = Sample average.
  - USGDP = United States’ GDP.
  - IndiaGDP = India’s GDP.
  - DiamondsExports = Botswana’s Diamonds Exports.
  - Inflation = Botswana’s Inflation.
  - Imports = Botswana’s Imports.
  - RealEffectiveExchangeRate = Botswana’s Real Effective Exchange Rate.
  - GoogleSVITravelBotswana = Botswana’s Google SVI (Travel).
  - ChinaGDP = China’s GDP.

### Botswana out-of-sample nowcast (second quarter of 2020)
- Context and predictors:
  - Target: project GDP growth for 2020Q2 (first full quarter under COVID-19) when 2020Q2 GDP data were not yet released.
  - Predictors included:
    - Google search volume index (SVI) in the “Travel” category for the word “Botswana”.
    - Botswana’s real effective exchange rate.
    - Stock exchange index.
    - Inflation.
    - Imports.
    - Bank loans to households.
    - Diamond exports.
    - Electricity generation.
    - Tax revenue.
    - GDPs for India and the United States.
  - Data start from 2008Q2.
- Model performance (out-of-sample RMSE, percent):
  - OLS: 9.9
  - Best Model: 2.4
  - Random Walk: 13.4
  - AR(1): 12.9
- Model selection and nowcast:
  - “Support Vector Machine” has the lowest out-of-sample RMSE among the models considered and is the selected model.
  - Selected model nowcasts about −19 percent year-on-year growth for 2020Q2 (model run with GDP data only up to 2020Q1).
  - Actual growth for 2020Q2 was −24 percent.
  - Interpretation: the framework relatively well identified the “turning point” of GDP growth, giving policymakers more space for early action.
  - Shapley decomposition indicates US GDP, India GDP, and diamond exports explain the nowcast (−19 percent), which is a lot lower than the sample average of GDP growth.

### Botswana all-period projections
- The nowcasting framework’s projections move closely with data (Figure 7: Botswana (All Period): Year-on-Year Rolling Quarterly Real GDP Growth, Data and Projections).

### Model comparison remark
- The selected model has the out-of-sample RMSE lower than traditional methods such as OLS, random walk, and AR(1).
- Random walk and AR(1) projections are based on GDP data during the in-sample period only.

### Sub-Saharan Africa, COVID-19 impact (summary)
- Sub-Saharan Africa faced an unprecedented health and economic crisis.
- The second quarter of 2020 had a year-on-year growth rate of −8.5 percent.
- Since 2020Q2, sub-Saharan Africa experienced a partial recovery throughout the year.

### Nowcasting framework and coverage
- Tracks year-on-year real GDP growth using a nowcasting framework applied to four economies:
  - (i) South Africa
  - (ii) Nigeria
  - (iii) Angola
  - (iv) an aggregate of Botswana, Cameroon, Côte d’Ivoire, Ghana, Kenya, Lesotho, Namibia, and Tanzania (weighted by purchasing-power-parity [PPP] GDPs)
- These 11 countries are used as a proxy for sub-Saharan Africa and account for about ¾ of the region’s PPP GDP.
- Quarterly GDP data are available at least from the first quarter of 2010 for the included countries.
- Ethiopia is excluded due to the unavailability of quarterly GDP data.
- Growth in the first quarter of 2021 is assumed to be the same as in the last quarter available for countries whose data are missing.

### Findings and projections
- The pandemic led to an unprecedented contraction in sub-Saharan Africa:
  - Initial estimate: 3.6 percent in 2020 (October 2020 Regional Economic Outlook: Sub-Saharan Africa)
  - Revised to a contraction of −1.7 percent (October 2021 Regional Economic Outlook: Sub-Saharan Africa)
- The projections closely move together with available GDP growth observations up to the first quarter of 2021.
- As a second quarter 2021 nowcast, the framework predicts high year-on-year growth (about 10 percent), largely reflecting the base effect from the bottom in the second quarter of 2020.
- Growth for the third quarter of 2021 is projected to decline to 3.6 percent, partly reflecting the impact of the third COVID-19 wave in sub-Saharan Africa.
- The nowcasts reproduce observed trends up to 2021: recovery after the second quarter of 2020 contraction, with the third wave having large impact.

### Drivers of the 2020 slowdown and subsequent dynamics
- Key factors contributing to the second quarter of 2020 slowdown:
  - Lockdown measures and voluntary social distancing (quarantine requirements)
  - Lower commodity prices and lower external demand
  - West Texas Intermediate (WTI) crude oil prices declined from about $60 before the pandemic to below $20 in April 2020
  - Eurobond spreads significantly widened at the height of the crisis
  - Substantial drop in flight arrivals reflecting worldwide travel restrictions
- COVID-19 dynamics:
  - Sub-Saharan Africa faced three waves of the COVID-19 pandemic, the third one having the largest impact.
  - New COVID-19 cases increased during the second quarter of 2020.
  - Containment measures (Oxford Government Response Stringency Index) strengthened in March and April 2020, gradually loosened until late 2020, and then remained at a similar level.
  - Mobility to workplaces remained lower than pre-COVID-19 until mid-2021, with temporary declines around the second and third waves.

### Policy implications and recommended uses
- The nowcasting framework provides a valuable addition to the policymaker’s toolkit by:
  - Enhancing regional surveillance through monthly releases of high-frequency data to inform IMF staff and policymakers about the latest status of regional economic development.
  - Producing nowcasts of GDP growth for individual countries (where quarterly GDP data are available) as a regular reference for surveillance.
  - Improving nowcasts further using country-specific information provided by IMF country teams.
- Potential operational uses:
  - Incorporation of a nowcasting dashboard based on the framework for regular tracking of real-time economic activity.
  - Training sessions at the IMF on the model, and presentations of nowcasting results for several individual countries to staff.
  - Workshops to member-country authorities to provide technical expertise and to help expand their internal forecasting capacities.

### Limitations and avenues for future research
- Key limitation: lack of quarterly GDP data for many sub-Saharan African countries poses statistical challenges for supervised machine-learning algorithms.
- Suggested future directions:
  - Improve coverage of countries that do not publish quarterly GDP data, potentially via a panel approach with a big data angle.
  - Use non-traditional data sources—satellite, mobility, Google trends, flight, and textual data—to address timely-data gaps in some countries.

### Key statistics and exact figures
- 3.6 percent in 2020 (October 2020 Regional Economic Outlook: Sub-Saharan Africa)
- −1.7 percent contraction (October 2021 Regional Economic Outlook: Sub-Saharan Africa)
- The 11-country proxy accounts for about ¾ of the region’s PPP GDP
- Quarterly GDP data available at least from the first quarter of 2010 for included countries
- Second quarter 2021 nowcast: about 10 percent year-on-year growth
- Third quarter 2021 projection: 3.6 percent year-on-year growth
- WTI crude oil prices declined from about $60 before the pandemic to below $20 in April 2020
- Botswana out-of-sample RMSEs (percent): OLS 9.9; Best Model 2.4; Random Walk 13.4; AR(1) 12.9
- Botswana 2020Q2 selected-model nowcast: −19 percent year-on-year growth; actual 2020Q2 growth: −24 percent
- Sub-Saharan Africa 2020Q2 year-on-year growth: −8.5 percent

*International Monetary Fund — content from wpiea2022088-print-pdf*

### 4.3 percent.

### 4.3 percent.

### Shapley decompositions and predictors (general)
- The Shapley decomposition of projected year-on-year real GDP growth in the quarter of interest is provided for country cases.
- Nigeria Shapley decomposition note:
  - Sample Av. = Sample average.
  - KENPMI = Kenya’s PMI.
  - COVIDDUM = COVID-19 dummy.
  - MATP = Materials price index.
  - GHAPMI = Ghana’s PMI.
  - xZAFSTOCK = Stock market index in selected sub-Saharan African countries excluding South Africa.
  - INDXNGA = India’s exports to Nigeria.
  - NGAPMI = Nigeria’s PMI.
  - CHNPMI = China’s PMI.
  - ZAFNVS = South Africa’s new vehicles sold.
  - ZAFSTOCK = South Africa’s stock market index.
- Botswana Shapley decomposition note:
  - Sample Av. = Sample average.
  - USGDP = United States’ GDP.
  - IndiaGDP = India’s GDP.
  - DiamondsExports = Botswana’s Diamonds Exports.
  - Inflation = Botswana’s Inflation.
  - Imports = Botswana’s Imports.
  - RealEffectiveExchangeRate = Botswana’s Real Effective Exchange Rate.
  - GoogleSVITravelBotswana = Botswana’s Google SVI (Travel).
  - ChinaGDP = China’s GDP.

### Botswana out-of-sample nowcast (second quarter of 2020)
- Context and predictors:
  - Target: project GDP growth for 2020Q2 (first full quarter under COVID-19) when 2020Q2 GDP data were not yet released.
  - Predictors included:
    - Google search volume index (SVI) in the “Travel” category for the word “Botswana”.
    - Botswana’s real effective exchange rate.
    - Stock exchange index.
    - Inflation.
    - Imports.
    - Bank loans to households.
    - Diamond exports.
    - Electricity generation.
    - Tax revenue.
    - GDPs for India and the United States.
  - Data start from 2008Q2.
- Model performance (out-of-sample RMSE, percent):
  - OLS: 9.9
  - Best Model: 2.4
  - Random Walk: 13.4
  - AR(1): 12.9
- Model selection and nowcast:
  - “Support Vector Machine” has the lowest out-of-sample RMSE among the models considered and is the selected model.
  - Selected model nowcasts about −19 percent year-on-year growth for 2020Q2 (model run with GDP data only up to 2020Q1).
  - Actual growth for 2020Q2 was −24 percent.
  - Interpretation: the framework relatively well identified the “turning point” of GDP growth, giving policymakers more space for early action.
  - Shapley decomposition indicates US GDP, India GDP, and diamond exports explain the nowcast (−19 percent), which is a lot lower than the sample average of GDP growth.

### Botswana all-period projections
- The nowcasting framework’s projections move closely with data (Figure 7: Botswana (All Period): Year-on-Year Rolling Quarterly Real GDP Growth, Data and Projections).

### Model comparison remark (Figure 6 caption)
- The selected model has the out-of-sample RMSE lower than traditional methods such as OLS, random walk, and AR(1).
- Random walk and AR(1) projections are based on GDP data during the in-sample period only.

### Sub-Saharan Africa, COVID-19 impact (summary)
- Sub-Saharan Africa faced an unprecedented health and economic crisis.
- The second quarter of 2020 had a year-on-year growth rate of −8.5 percent.
- Since 2020Q2, sub-Saharan Africa experienced a partial recovery throughout the year.

*Source: IMF staff calculations and country authorities as presented in the provided content.*

### 3.6 percent in 2020 (October 2020

### wpiea2022088-print-pdf - 3.6 percent in 2020 (October 2020

### Nowcasting framework and coverage
- Tracks year-on-year real GDP growth using a nowcasting framework applied to four economies:
  - (i) South Africa
  - (ii) Nigeria
  - (iii) Angola
  - (iv) an aggregate of Botswana, Cameroon, Côte d’Ivoire, Ghana, Kenya, Lesotho, Namibia, and Tanzania (weighted by purchasing-power-parity [PPP] GDPs)
- These 11 countries are used as a proxy for sub-Saharan Africa and account for about ¾ of the region’s PPP GDP.
- Quarterly GDP data are available at least from the first quarter of 2010 for the included countries.
- Ethiopia is excluded due to the unavailability of quarterly GDP data.
- Growth in the first quarter of 2021 is assumed to be the same as in the last quarter available for countries whose data are missing.

### Findings and projections
- The pandemic led to an unprecedented contraction in sub-Saharan Africa:
  - Initial estimate: 3.6 percent in 2020 (October 2020 Regional Economic Outlook: Sub-Saharan Africa)
  - Revised to a contraction of −1.7 percent (October 2021 Regional Economic Outlook: Sub-Saharan Africa)
- The projections closely move together with available GDP growth observations up to the first quarter of 2021.
- As a second quarter 2021 nowcast, the framework predicts high year-on-year growth (about 10 percent), largely reflecting the base effect from the bottom in the second quarter of 2020.
- Growth for the third quarter of 2021 is projected to decline to 3.6 percent, partly reflecting the impact of the third COVID-19 wave in sub-Saharan Africa.
- The nowcasts reproduce observed trends up to 2021: recovery after the second quarter of 2020 contraction, with the third wave having large impact.

### Drivers of the 2020 slowdown and subsequent dynamics
- Key factors contributing to the second quarter of 2020 slowdown:
  - Lockdown measures and voluntary social distancing (quarantine requirements)
  - Lower commodity prices and lower external demand
  - West Texas Intermediate (WTI) crude oil prices declined from about $60 before the pandemic to below $20 in April 2020
  - Eurobond spreads significantly widened at the height of the crisis
  - Substantial drop in flight arrivals reflecting worldwide travel restrictions
- COVID-19 dynamics:
  - Sub-Saharan Africa faced three waves of the COVID-19 pandemic, the third one having the largest impact.
  - New COVID-19 cases increased during the second quarter of 2020.
  - Containment measures (Oxford Government Response Stringency Index) strengthened in March and April 2020, gradually loosened until late 2020, and then remained at a similar level.
  - Mobility to workplaces remained lower than pre-COVID-19 until mid-2021, with temporary declines around the second and third waves.

### Policy implications and recommended uses
- The nowcasting framework provides a valuable addition to the policymaker’s toolkit by:
  - Enhancing regional surveillance through monthly releases of high-frequency data to inform IMF staff and policymakers about the latest status of regional economic development.
  - Producing nowcasts of GDP growth for individual countries (where quarterly GDP data are available) as a regular reference for surveillance.
  - Improving nowcasts further using country-specific information provided by IMF country teams.
- Potential operational uses:
  - Incorporation of a nowcasting dashboard based on the framework for regular tracking of real-time economic activity.
  - Training sessions at the IMF on the model, and presentations of nowcasting results for several individual countries to staff.
  - Workshops to member-country authorities to provide technical expertise and to help expand their internal forecasting capacities.

### Limitations and avenues for future research
- Key limitation: lack of quarterly GDP data for many sub-Saharan African countries poses statistical challenges for supervised machine-learning algorithms.
- Suggested future directions:
  - Improve coverage of countries that do not publish quarterly GDP data, potentially via a panel approach with a big data angle.
  - Use non-traditional data sources—satellite, mobility, Google trends, flight, and textual data—to address timely-data gaps in some countries.

### Key statistics and exact figures
- 3.6 percent in 2020 (October 2020 Regional Economic Outlook: Sub-Saharan Africa)
- −1.7 percent contraction (October 2021 Regional Economic Outlook: Sub-Saharan Africa)
- The 11-country proxy accounts for about ¾ of the region’s PPP GDP
- Quarterly GDP data available at least from the first quarter of 2010 for included countries
- Second quarter 2021 nowcast: about 10 percent year-on-year growth
- Third quarter 2021 projection: 3.6 percent year-on-year growth
- WTI crude oil prices declined from about $60 before the pandemic to below $20 in April 2020

*International Monetary Fund — content from wpiea2022088-print-pdf - 3.6 percent in 2020 (October 2020)*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022088-print-pdf.pdf_