5 th Meeting of the International Comparison Program (ICP) - PowerPoint PPT Presentation

5 th Meeting of the International Comparison Program (ICP) Governing Board December 13, 2019 Washington, DC

ICP Revision Policy ICP Global Office, World Bank

Revision Policy Rationale ICP benchmark and non-benchmark year results should be frequently revised, to ensure that the program’s outputs: � Reflect the latest available information and methodologies; � Are of the highest quality; and � Remain relevant to users. 3

ICP Indicators � Purchasing Power Parities (PPPs) � Price Level Index (PLIs) � Real expenditures � Per capita (p.c.) real expenditures � Nominal expenditures � Per capita (p.c.) nominal expenditures for all headings under the ICP Classification of Final Expenditure on GDP 4

Triggers for Revisions: Input Data [1] 1. REVISIONS IN THE AGGREGATE GDP ESTIMATES � Revise real and p.c. real expenditures and nominal and p.c. nominal expenditures. 2. REVISIONS IN THE COMPONENTS OF GDP � Revise real and p.c. real expenditures and nominal and p.c. nominal expenditures. � May also trigger a revision of PPPs and PLIs. 3. REVISIONS IN POPULATION FIGURES � Revise p.c. real expenditures and p.c. nominal expenditures. 5

Triggers for Revisions: Input Data [2] 4. CHANGES IN COUNTRIES’ CURRENCY UNITS � Revise real and p.c. real expenditures, PPPs, PLIs, and nominal and p.c. nominal expenditures. 5. CORRECTION OF ERRORS IN SOURCE DATA OR RESULTS � May trigger a revision of real and p.c. real expenditures, PPPs and PLIs, and nominal and p.c. nominal expenditures. 6

Triggers for Revisions: New Methodology [1] 1. SIGNIFICANTLY IMPROVED PPP COMPUTATION AND AGGREGATION METHODS � Revise real and p.c. real expenditures, PPPs and PLIs. 2. SIGNIFICANTLY IMPROVED GLOBAL LINKING APPROACH � Revise real and p.c. real expenditures, PPPs and PLIs. 7

Triggers for Revision: New Methodology [2] 3. SIGNIFICANTLY IMPROVED RETROPOLATION/ INTERPOLATION/ EXTRAPOLATION METHOD � Revise timeseries of real and p.c. real expenditures, PPPs and PLIs. 4. A CHANGE TO THE ICP CLASSIFICATION OF FINAL EXPENDITURE ON GDP � Triggered by the System of National Accounts (SNA) or Classification of Individual Consumption According to Purpose (COICOP) revisions. � Revise real and p.c. real expenditures, PPPs, PLIs, and nominal and p.c. nominal expenditures. 8

Revisions to ICP Indicators Purchasing Triggers Per capita Per capita Power Parity Real Nominal Real Nominal and Price Expenditures Expenditures Expenditures Expenditures Level Index ✓ ✓ ✓ ✓ Revision to Aggregate GDP ✓ ✓ ✓ ✓ ? Revision to GDP components ✓ ✓ Changes to Revision to Population input data ✓ ✓ ✓ ✓ ✓ Change in Currency Unit Correction of errors in source ? ? ? ? ? data or results Significantly improved PPP ✓ ✓ ✓ computation and aggregation methods Significantly improved global ✓ ✓ ✓ New linking approach methodology Significantly improved ✓ ✓ ✓ retropolation/interpolation/ extrapolation method Change to the ICP ✓ ✓ ✓ ✓ ✓ classification of final 9 expenditure on GDP

Guidelines for Revisions [1] A. HISTORICAL REVISIONS � Changes in historical input data from countries may trigger a revision to ICP results. � Historical revisions based on new methodologies are subject to the feasibility of introducing the new methodology retrospectively. B. GEOGRAPHICAL SCOPE � Regions are responsible for revising benchmark indicators and associated timeseries at the regional level. � Revisions to regional data should precede those applied to global data. � Revised global results should maintain regional fixity. 10

Guidelines for Revisions [2] C. CATEGORIES OF INDICATORS � Real and p.c. real expenditures will be revised when GDP expenditure data and/or population data are revised. � PPPs and PLIs may be revised depending on the level of detail of the GDP expenditure revisions, and/or a change to the ICP classification. � Nominal and p.c. nominal expenditures will be revised when GDP expenditure data and/or population data are revised. D. QUALITY REVIEW � Revised ICP data will be subject to an expert review before they are published. 11

Guidelines for Revision [3] E. PUBLICATION OF REVISED RESULTS � Revised ICP indicators will be published once they have been compiled and undergone quality review. � Revised ICP indicators will be released at the publication level of the benchmark results. � The ICP Global Office will archive the various vintages of the data. F. UNPUBLISHED DATA � When revising published ICP indicators, unpublished data may need to be revised to maintain the consistency between the two datasets. 12

Timing and Communication of Revisions [1] TIMING OF REVISIONS � Revisions to real and p.c. real expenditures and nominal and p.c. nominal expenditures [triggered by GDP expenditure and population data revisions] are introduced on an annual basis. � Revisions to PPPs and PLIs are introduced in conjunction with the release of new benchmark results. 13

Timing and Communication of Revisions [2] COMMUNICATION OF REVISIONS � The schedule of revisions will be announced to stakeholders and users well in advance. � When methodology is improved, the new methods will be communicated to users well in advance. � Results will be made publicly available on the ICP Website, the World Bank’s Data Portal, the World Development Indicators, and other relevant sites. 14

THANK YOU 15