世界发展银行-March-2021-PovcalNet-Update---What_rsquo_s-New_66页_995kb

Summary of March 2021 PovcalNet Update

Core Content

The March 2021 update to the PovcalNet database includes several significant changes to the data underlying global poverty estimates, enhancing data harmonization, and introducing new survey data and imputation methods for specific countries. This update provides revised poverty estimates for the reference years 2017 and 2018, and adds new estimates for 2019 where possible. The database now includes more than 1,900 surveys, reflecting an expansion in data availability and improved accuracy in poverty measurement.

Main Changes

1. New Surveys and Imputation Methods

• Somalia, South Sudan, and Zimbabwe are included for the first time using imputed consumption data.
• Somalia is now covered with a poverty estimate based on the second wave of the Somali High Frequency Survey (SHFS2).
• South Sudan uses data from Wave 3 of the High Frequency Survey (HFS) 2015-2017.
• Zimbabwe incorporates data from the 2019 Mini-Poverty, Income, Consumption, and Expenditure Survey (Mini-PICES).

2. Improvements in Historical Data

• Over 120 historical surveys were revised, improving the accuracy and coverage of the data.
• These revisions are primarily in the 1990s and early 2000s, leading to more precise poverty estimates.

3. Updated Welfare Aggregates

• Several countries saw changes in their welfare aggregates, reflecting better harmonization and more available information.
• Examples include:
  • Argentina (2003-2018)
  • Bolivia (2000-2018)
  • Brazil (2012-2018)
  • Colombia (2001-2005, 2008-2016)
  • Costa Rica (2001-2004, 2008, 2016-2018)
  • Ecuador (2003, 2005-2018)
  • El Salvador (2000-2006, 2008, 2018)
  • EU-SILC
  • Honduras (2001, 2014-2018)
  • Lao PDR
  • LIS
  • Malaysia (2015)
  • Mexico (1989, 1992, 1994, 1996, 1998)
  • Pakistan (2015)
  • Panama (2000-2004)
  • Paraguay (2006, 2010, 2017-2018)
  • Peru (1997-2014, 2018)
  • Philippines (2015)
  • Sri Lanka (2002-2016)
  • Uruguay

4. Updated PPPs and CPI Data

• PPPs for Somalia are imputed using a cross-country relationship, based on revised 2011 PPPs.
• CPI data has been updated, with the deflation process using data from the Market Price Survey (MPS) and Food Security and Nutrition Analysis Unit (FSNAU) for Somalia.
• National accounts data and population data have also been updated to improve the accuracy of poverty estimates.

5. Survey Year Adjustments

• The update includes changes to survey years, which affect the calculation of poverty estimates.
• Survey years have been adjusted to align with the most recent data available.

6. Comparability Database

• A comparability database has been introduced to ensure consistency across different data sources and surveys.
• This helps in making poverty estimates across time and regions more reliable.

7. Economy-Year Changes

• Some economy-years were removed or added based on the availability of new data.
• The total number of economy-years has increased, providing a more comprehensive dataset.

Key Information

• Global poverty headcount ratio at $1.90 increased slightly from 9.2% to 9.3%.
• Number of poor people increased from 689 million to 696 million.
• Sub-Saharan Africa saw an increase in poverty headcount from 41.0% to 41.2%, mainly due to the inclusion of the first survey-based estimate for Somalia.
• Somalia's poverty headcount at $1.90 is 68.6%, compared to the previously imputed 41.0%.
• South Sudan's poverty headcount at $1.90 is 76.4%, lower than the previously imputed 84.7%.
• Zimbabwe uses the SWIFT method to impute consumption data for households without detailed consumption data.
• The Mini-PICES 2019 survey was conducted between April and May 2019, and is not representative of the entire year, possibly introducing seasonal bias.
• The survey coverage for various regions in 2017 and 2019 is as follows:
  • East Asia and Pacific: 97.5% (2017) and 95.9% (2019)
  • Europe and Central Asia: 89.5% (2017) and 87.4% (2019)
  • Latin America and the Caribbean: 90.2% (2017) and 86.8% (2019)
  • Middle East and North Africa: 58.2% (2017) and 47.9% (2019)
  • Rest of the World: 82.4% (2017) and 64.3% (2019)
  • South Asia: 21.8% (2017) and 21.9% (2019) – no poverty estimates were published for South Asia in 2017 due to insufficient population coverage.

Conclusion

The March 2021 update to PovcalNet represents a significant step forward in global poverty monitoring, incorporating new survey data and imputation methods for countries with limited data availability. The inclusion of imputed data for Somalia, South Sudan, and Zimbabwe has improved the accuracy of poverty estimates in these regions. Additionally, the update reflects ongoing efforts to harmonize data across countries and improve the comparability of poverty estimates over time.