TECHNICAL ADVISORY GROUP Inaugural Meeting March 24, 2017 Princeton University, New Jersey
Agenda Introduction of the Technical Advisory Group Members ICP Governance Framework and Terms of Reference of the Technical Advisory Group New ICP Approach and the 2017 Comparison Cycle ICP Research Agenda Setting up ICP Task Forces 2
TECHNICAL ADVISORY GROUP (TAG) MEMBERS 3
TAG Members ALAN HESTON, Professor Emeritus, University of Pennsylvania, Co-founder of the ICP ANA REVENGA, Deputy Chief Economist of the World Bank SIR ANGUS DEATON, 2015 Nobel Prize Laureate in Economics, Princeton University [Chair] D. PRASADA RAO, Professor, University of Queensland PAUL SCHREYER, Deputy Chief Statistician of the OECD [Alternate Chair] 4
TAG Members PRONAB SEN, former Chief Statistician of India ROBERT C. FEENSTRA, Professor, University of California Davis W. ERWIN DIEWERT, Professor, University of British Columbia WALTER RADERMACHER, former Chief Statistician of the EU & Germany XIANCHUN XU, former Deputy Commissioner of NBS China 5
ICP GOVERNANCE FRAMEWORK AND TAG TERMS OF REFERENCE 6
ICP Governance Framework GIA = Global Implementing Agency, RIA = Regional Implementing Agency, NIA = National Implementing Agency
Governing Board Strategic and policy-making body Approves methodology Sets policies and protocols for PPP production Ensures that ICP results are consistent with agreed policies, protocols and methodology Advocates, conducts outreach, promotes openness, raises funding 8
Inter-Agency Coordination Group Coordinating body for program implementation Establishes data collection, validation, calculation and dissemination work plans Develops common standards and protocols to ensure comparability across regions and countries Provides quality assurance standards that national and regional data and metadata must satisfy to be included in the global PPP estimates 9
World Bank Global ICP Unit Coordinates and manages the ICP globally Provides secretariat functions for governance bodies Calculates and disseminates global results Delivers technical support to regions Implements the ICP dissemination and access policy 10
Regional Implementing Agencies* Coordinate and manage regional comparisons/ Eurostat-OECD program Carry out regional capacity-building and technical assistance activities Ensure adequate national and regional data and metadata quality as per the agreed standards Calculate and disseminate regional results * and Eurostat and OECD 11
National Implementing Agencies Conduct country activities Collect and compile the data and metadata required for estimating PPPs Ensure adequate national data and metadata quality per the agreed standards 12
TAG Technical body Assures methodological soundness and overall quality of the PPP estimates Ensures transparency of the PPP estimation process Supports the establishment of a permanent and more frequent ICP The TAG, in collaboration with IACG, sets the ICP research agenda The TAG establishes Task Forces to undertake research into specific technical issues on the research agenda and support the computation of ICP results. 13
TAG Work Modality TAG holds regular annual meetings to DiscussTask Forces proposals on methodological improvements under the research agenda Review the overall quality of PPP estimates Ad-hoc meetings may be organized, if need arises The Global ICP Unit serves as TAG secretariat 14
NEW ICP APPROACH AND THE IMPLEMENTATION OF THE 2017 COMPARISON CYCLE 15
Aspiration: Rolling Survey Approach Spreads the price data collection over 3 years, in order to ease the burden on countries for any given calendar year Allows regions the flexibility to conduct surveys according to their specific circumstances Designed to make comparisons using a mix of actual and extrapolated data that have been collected over 3 years 16
Interim ICP Activities Keep ICP momentum through interim regional comparisons Asia 2015/2016 CIS 2014 Western Asia 2016 (plus results for 2012/2013) Eurostat-OECD regular PPP program Maintain capacity through multiple regional workshops and E-learning course (underway) Increase outreach through seminars, website, video tutorials, papers, leaflets, blogs, and newsletters 17
ICP Timetable 2016-2018 Conduct new ICP surveys Incorporate interim surveys conducted in 2016, where possible End 2019 Link results of interim regional exercises conducted during 2012-2016 and produce global PPP time series Publish ICP 2017 results 2020 and beyond Implement gradually the rolling survey approach to produce results every three years 18
ICP 2017 Cycle ICP 2017 Cycle 2016 2017 2018 2019 Household consumption X => X < = X Housing volume and rentals X < = X Government compensation X < = X Machinery and equipment X < = X Construction and civil engineering X < = X PUBLICATION CPIs, PPIs and national account deflators X X X National account expenditures X X X 19
Future ICP Cycles ICP 2020 Cycle ICP 2023 Cycle 2019 2020 2021 2022 2022 2023 2024 2025 Household consumption X=> X <=X X=> X <=X Housing volume and rentals X X Government compensation X X Machinery and equipment X X Construction and civil engineering X X CPIs, PPIs and national account deflators PUBLICATION OF 2020 RESULTS X X X X X X PUBLICATION OF 2023 RESULTS National account expenditures X X X X X X 20
ICP RESEARCH AGENDA CONTEXT 21
Background The last half-century has seen the ICP grow into a global statistical program Each round of ICP has increased country participation, and improved the methods and governance of the program ICP has reached a stage where it needs to identify its future direction and sustainability 22
Opportunities United Nations mandate A truly permanent global program Stable home Global ICP Unit at the World Bank Opportunity to modernize, innovate and improve Data revolution 23
Challenges Establishing ICP within the work program of the global, regional and national implementing agencies and securing sustainable funding Conducting frequent and regular cycles of ICP through the introduction of a rolling survey approach Ensuring the quality, reliability and consistency of PPPs 24
Consultations IACG Second meeting, Washington DC (September 2016) Experts Meeting in MIT, Boston (May 2016) World Bank management and experts (March 2017) Countries Various regional workshops and events (2016-2017) 25
Research Agenda Focus on methodological aspects related to building PPP time series and fine-tuning established methods and procedures to improve quality Innovations as a cross-cutting theme to ensure that new approaches, technologies, and data sources are utilized in the production of PPPs Suggested Research Agenda includes: Immediate term items Medium term items Ongoing and longer term items 26
RESEARCH AGENDA FOR THE IMMEDIATE TERM 27
I. PPP Time Series & Rolling Survey Approach Issue: Extrapolations of PPPs at the aggregate level over long periods intervening the benchmark comparisons, coupled with major methodological changes between benchmarks, limit the ability to reconcile extrapolations from different benchmarks. Three components: a. Moving Towards Rolling Price Surveys b. Linking Interim Regional Updates into a Global Comparison c. Building PPP Time Series for the Interim Period 28
I.a. Moving Towards Rolling Price Surveys AREAS TO EXAMINE Determine the frequency of data collection for the rolling survey for each aggregate of the GDP Assess availability and quality of national account deflators, CPIs and other price indices to be used in extrapolation to the reference year Determine how to link PPPs from regions with different timetables for data collection Assess methods of extrapolating PPPs for the two years between individual benchmarks, before the full rolling survey approach is implemented 29
I.b. Linking Interim Regional Updates AREAS TO EXAMINE Effect of different timing of regional interim surveys Impact of different degrees of coverage of the GDP Effect of changing regional composition and of countries not participating in regional interim updates Effect of discrepancies in global core list coverage by region on linking Effect of methodological differences between regions on linking 30
I.c. Building Interim PPP Time Series AREAS TO EXAMINE Assess availability and quality of national account deflators, CPIs and other price indices by GDP component Examine consistency of CPI weights and national account weights, and assess the weights effects on PPPs Suggest steps for statistical capacity-building to ensure that CPIs and national account deflators are consistent over time and comparable across countries Ensure consistency between the 2011 and 2017 results and the comparisons for the intervening years 31
II. Dwelling Services Issue: Housing comparisons are difficult due to the varying mix of rented and owner-occupied dwellings, absence of reliable/regular rental surveys, and unreliability of housing expenditures. AREAS TO EXAMINE Assess the quality of housing input data and recommend additional housing quality indicators Propose imputation techniques in cases of missing data Suggest refinements and improvements to existing methods to better account for quality differences Investigate the use of new methods for estimating housing expenditures ISWGNA 32
III. Labor Productivity Adjustment Issue: Adjustment for productivity differentials is designed to improve PPPs for labor; however, the availability and quality of relevant adjustment factors remains challenging. AREAS TO EXAMINE Assess and evaluate the quality of data for labor inputs and capital stocks portion of the productivity adjustment Update productivity adjustment factors for use in the 2017 ICP cycle Examine effects of the adjustment factors not being used in some regions on the global results Examine the scope for productivity adjustment for labor inputs in construction and make recommendations for the 2017 ICP cycle 33
IV. Fine Tuning Global Linking Procedures Issue: Linking via global core list approach and CAR procedure is somewhat straightforward; however, linking in practice had many exceptions. AREAS TO EXAMINE Assess impact of individual countries on global and regional results, and investigate if the reliability of the global comparison can be improved with a two-stage linking Assess the possibility of linking the CIS through the global core approach and CAR procedure Develop a strategy to accommodate special participation cases and specific GDP components 34
V. Quality and Reliability of PPPs Issue: Lack of any indication on the quality of the PPPs and measures of reliability or measures of uncertainty associated with PPPs. Two components: a. Quality of Data b. Reliability Measures for PPPs 35
V.a. Quality of Data AREAS TO EXAMINE Investigate sources of data variability at each level of estimation to assess the overall quality of the results and where data quality can be improved Examine variability in basic heading PPPs and expenditure shares and identify countries and basic headings appearing as outliers Analyze effect of the countries with high variability in prices and expenditure shares on regional and global results ISWGNA 36
V.b. Reliability Measures for PPPs AREAS TO EXAMINE Establish a conceptual framework that underpins reliability measurement Develop a statistical framework for the computation of standard errors Compute standard errors for the 2005, 2011 and 2017 PPPs as well as for PPP changes over time Provide guidance to users as to how these measures of reliability can be used in practice 37
VI. Uses of PPPs for Policy Making Issue: There is an imminent need for outlining the uses of ICP results and demonstrate their relevance and usefulness for policy making. AREAS TO EXAMINE Policy-making uses of PPPs at international level Policy-making uses of PPPs at national level Other uses of PPPs 38
RESEARCH AGENDA FOR THE MEDIUM TERM 39
VII. CPI-ICP Synergies and Sub-National PPPs Issue: There is a need for a closer integration of CPI and ICP activities and better notion of what this integration actually means. AREAS TO EXAMINE Prepare a set of guidelines for countries to facilitate the process of harmonizing CPI and ICP activities Conduct a case study of selected countries to evaluate the use of SPDs in their national CPIs Formulate guidelines on the use of the CPI product lists and prices to compute sub-national PPPs Analyze temporal consistency of CPI and sub-national PPPs Expand work on sub-national PPPs to more countries CPI Expert Group 40
VIII. Exports and Imports Issue: In the ICP, foreign trade sector is incorporated using exchange rates. However, there has been a rich literature on incorporating terms-of-trade gains and losses due to the exportimport price differentials into international comparisons. AREAS TO EXAMINE Investigate data availability and reliability for the terms-of-trade computation, both across space and across time Evaluate various methodologies to incorporate the term-of trade effect into the real GDP and PPP computation, on an experimental basis Incorporate the terms-of-trade effects into the 2005, 2011 and 2017 ICP results, on an experimental basis 41
IX. Construction Issue: 2011 construction results for some countries were not fully plausible; input data quality remains an issue; and the methodology needs to be closely assessed. AREAS TO EXAMINE Investigate the quality of input data for construction Look into methodological improvements (incl. productivity adjustment) Improve estimates of the resource mix weights Review the availability and quality of other construction data and devise methods to incorporate them in PPP estimation ISWGNA 42
X. Health and Education Issue: Compilation of PPPs for health and education is challenging given the complexity of the provision and consumption of related goods and services. AREAS TO EXAMINE Design an output-based approach for estimating health and education PPPs Assess the availability of data in participating countries and other (international) sources Evaluate effects of implementing the output-based approach compared to the regular input-based method 43
ONGOING AND LONGER-TERM RESEARCH AGENDA 44
XI. ICP PPPs and Global Poverty Measurement Issue: PPPs are used in the calibration of the international poverty line (IPL) and in producing estimates of incidence of global poverty. Atkinson Commission recommended that the use of 2011 PPPs be fixed until 2030. AREAS TO EXAMINE Assess impact of 2017 ICP PPPs on global poverty measures Assess possible use of sub-national PPPs for poverty measurement Identify strategies to use data underlying the headline ICP PPPs Investigate whether poverty-specific PPPs are more pertinent for poverty estimates WB Working Groups 45
XII. Innovations in Technology and Data Sources Issue: Availability of scanner data and possibility of collecting prices through web-scraping or crowdsourcing offer a range of alternative sources of price data that need to be explored in future ICP cycles. AREAS TO EXAMINE Assess the quality of prices from alternative sources Compare price levels and dynamics between CPI and ICP price collections with the alternative sources Ottawa Group & GWG on Big Data 46
XIII. Product Quality Differences Issue: Despite the SPD approach, it is still difficult to ensure that there are no differences in qualities of products actually priced in different countries. AREAS TO EXAMINE Use of extended SPD approach to capture quality differences Explore the possibility of incorporating quality indicators in the CPD Conduct a pilot study to measure the extent of bias induced by quality differences 47
SETTING UP ICP TASK FORCES 48
Task Forces (TFs) Each TF will focus on one or more specific topics and it will be provided with TORs covering Background to the problem(s) Clear statement of the technical issues to be considered List of deliverables [report, paper, recommendations, and/or operational guidelines] Timeframe for the completion of the work 49
TF Membership TFs are formed of experts and/or practitioners in the field of index numbers, PPPs, price statistics and national accounts Size of each TF depends on the complexity of the topic Membership of the TF is proposed by the Global ICP Unit for consideration and endorsement by the TAG One TF member is designated to coordinate the work Each TF will include an expert from the Global ICP Unit It is desirable but not necessary to have a TAG member included in the TF 50
TF Work Modality Each TF will have a fixed tenure in line with the timeframe of the assigned topics TFs will meet as needed physically or virtually May join the meetings of the IACG to discuss the operational implications and feasibility of their proposals May join the meetings of the TAG to present their findings and recommendations The Global ICP Unit will provide access to necessary data and materials to facilitate the work of the TFs 51
TFs to be Established in 2017 1. PPP COMPILATION AND COMPUTATION Linking interim regional updates into a global comparison (2017-2019) Building PPP time series for the interim period (2017-2019) Fine tuning global linking procedures (2018-2019) Productivity adjustment for government and construction labor (2018-2019) 2. COUNTRY OPERATIONAL GUIDELINES AND PROCEDURES Moving towards rolling price surveys (2017-2018) CPI-ICP synergies to improve spatial and temporal price consistency (2017-2020) Sub-national PPPs (2018-2020) 52
TFs to be Established in 2017 (Cont'd) 3. HOUSING PPPs and real expenditures for dwelling services (2017-2019) 4. PPP USES Uses of PPPs and ICP data for national and international policy making (2017-2019) 5. EXPORTS AND IMPORTS PPPs for exports and imports (2017-2020) 6. PPPS AND POVERTY MEASUREMENT ICP PPPs and global poverty measurement (2017-2021) 53
TFs to be Established in 2018 & Beyond 1. PPP RELIABILITY Quality of data (2018-2019) Reliability measures for PPPs (2018-2019) Accounting for product quality differences in PPP measurement (2020-2022) 2. NEW DATA SOURCES Exploring innovations in technology and data sources for PPP measurement (2018-2021) 3. NON-MARKET SERVICES PPPs for health and education (2020-2021) 4. CONSTRUCTION PPPs for construction (2020-2021) 54
THANK YOU 55