## 1. Countries Including E-Commerce in CPI (43 economies)

## Source details

**Canonical URL:** [1. Countries Including E-Commerce in CPI (43 economies)](https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019163-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2019/wpiea2019163-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2019/wpiea2019163-print-pdf.pdf.json)

---

### Overview
- Content unit: a listing of countries that include e-commerce in their Consumer Price Index (CPI).
- Number of economies covered: 43 economies.
- Location in source: page 22.

### Survey findings on e-commerce coverage in CPI (Joint IMF-OECD, August 2017; sample size 43 economies)
- 77 percent (33 countries) include e-commerce in the CPI.
- Domestic e-commerce included more widely than cross-border e-commerce.
- Advanced economies in sample: 100 percent include e-commerce in CPI.
- Emerging and developing economies in sample: 37.5 percent (6 countries) include e-commerce in prices or weights.
- 97 percent of advanced economies sampled (26 countries) include domestic e-commerce in CPI; 37 percent (10 countries) include cross-border e-commerce.
- Divisions most commonly priced online:
  - Recreation and culture (09)
  - Furnishing, household equipment and routine household maintenance (05)
  - Transport (07)
- Only 3 of 43 economies include sharing economy services in CPI.

### Quality adjustment methods reported for CPI (survey of 43 economies)
- 6 economies do not use any quality adjustment method.
- 27 economies use several methods depending on product group.
- 8 economies use one quality adjustment method for all products.
- Main methods reported:
  - Overlap: 14 economies
  - Hedonic regression: 13 economies
  - Judgmental methods: 13 economies

### Implications and recommended priorities (as reflected in source)
- Need to improve treatment of digitalized goods and services in CPI quality adjustment practices.
- Further research required on use of scanner and web-scraped data, classification and processing of big data, and improved quality adjustment methods for digitalized goods and services.
- Repeat study in five years recommended to monitor compilation improvements and potential extension to export and import price indexes.

*Source: wpiea2019163-print-pdf — "1. Countries Including E-Commerce in CPI (43 economies)" (page 22) and related survey results cited in the chapter.*

### 1. Countries Including E-Commerce in CPI (43 economies) ..................................................22

### 1. Countries Including E-Commerce in CPI (43 economies) ..................................................22

### Summary
- Content unit: a listing of countries that include e-commerce in their Consumer Price Index (CPI).
- Number of economies covered: 43 economies.
- Location in source: page 22.

### Related table-of-contents context (from the same PDF)
- Adjacent sections:
  - "2. Countries Including E-Commerce in PPI (36 economies) ...................................................24"
  - "3. Summary Conclusions .........................................................................................................25"
- Related figures referenced elsewhere in the PDF:
  - "12. Quality Adjustment Methods in the CPI ............................................................................23"

*Source: wpiea2019163-print-pdf - 1. Countries Including E-Commerce in CPI (43 economies) ..................................................22*

### 13. Quality Adjustment Methods in PPI ..................................................................................

### 13. Quality Adjustment Methods in PPI

### I. Introduction
- Price indexes covered: consumer price index (CPI), producer price index (PPI), export price index (XPI), import price index (MPI), residential property price index (RPPI).
- Sample and groupings:
  - 196 economies in sample.
  - 37 advanced economies and 159 emerging and developing economies.
  - Emerging and developing economies subgrouped into 50 inflation-targeting and 109 non-inflation-targeting economies.
- Analytical focus: (1) frequency of updating weights; (2) coverage of CPI and PPI; (3) timeliness of CPI and PPI; (4) classification systems used in CPI and PPI compilation.
- Data sources and inputs:
  - IMF technical assistance experience: about 60 TA missions and training activities per year since 2000; over 1,200 technical assistance and training missions on price statistics compilation over past two decades.
  - Joint IMF-OECD questionnaire (2017): 43 countries responded on CPIs, 36 on PPIs.
  - IMF Dissemination Standards Bulletin Board (DSBB) and RCDC inputs.

### II. Status of CPI Compilation
- Universality and uses:
  - CPI compiled by all 196 economies.
  - Common uses: indexation of wages, rents, contracts, social benefits; deflation in national accounts; monetary and fiscal policy tool.
- Frequency of updating CPI weights:
  - Recommendation: CPI weights no more than five years old (CPI Manual 2004 and update).
  - Approximately 45 percent of all economies have CPI weights updated within five years.
  - Advanced and inflation-targeting economies update more frequently:
    - 89 percent (33 economies) of advanced economies revised weights in last five years.
    - 52 percent (26 economies) of inflation-targeting emerging and developing economies revised weights in last five years.
  - Weight age examples: remaining four advanced economies and 17 inflation-targeting emerging/developing economies have weight reference period between 6 and 10 years old.
- Geographical coverage of CPI (expenditure weights):
  - 149 economies (76 percent) — weights based on expenditure of all households (urban and rural).
  - 34 of 37 advanced economies (92 percent) have national coverage.
  - 115 of 159 emerging and developing economies (72 percent) have national coverage.
  - 28 economies — weights reflect urban areas only.
  - 19 economies — weights reflect only the capital city.
  - In emerging and developing economies: 80 percent of inflation-targeting and 69 percent of non-inflation-targeting economies have national coverage.
- Timeliness and frequency:
  - IMF data dissemination standards: SDDS prescribes monthly dissemination; e-GDDS suggests dissemination within eight weeks.
  - 75 economies (41 percent) are SDDS Plus or SDDS subscribers; 108 economies (59 percent) report data according to e-GDDS.
  - Approximately 86 percent of countries compile and disseminate a monthly CPI and publish within four weeks after the reference period.
  - 13 economies compile monthly but publish more than one month after the reference period (lags span 6 to 12 weeks).
  - 15 economies (7 percent) compile quarterly CPI published within 1 to 3 months after reference quarter.
- Classification system:
  - COICOP recommended and widely used.
  - Approximately 76 percent of economies (148 economies) use COICOP.
  - 95 percent of advanced economies and 81 percent of inflation-targeting economies use COICOP.
  - COICOP 2018 structure: 13 divisions (two-digit), 63 groups, 186 classes, 338 subclasses.

### III. Status of PPI Compilation
- Availability and uses:
  - PPIs compiled by 126 economies (64 percent of the sample).
  - Of PPI compilers: 35 economies (28 percent) are advanced economies and 91 are emerging and developing economies.
  - 81 PPI compilers are inflation-targeting economies; 45 are non-inflation-targeting economies.
  - PPI uses: deflation of national accounts, short-term indicator of inflationary trends, indexation in contracts, leading indicator for consumer inflation.
- Frequency of updating PPI weights:
  - Best practice: update at least every five years.
  - Approximately 36 percent of all PPI-compiling countries update weights within five years.
  - Advanced economies: 57 percent (20 economies) updated weights within five years; remaining 43 percent have weights between 6 and 10 years old.
  - Emerging and developing economies: 28 percent (25 economies) updated within five years; 47 percent (43 economies) have weights between 6 and 10 years old; 25 percent (23 economies) have weights older than 10 years.
  - Inflation-targeting emerging/developing: 35 percent maintain weights 5 years old or less; non-inflation-targeting: 20 percent within 5 years; 33 percent of non-inflation-targeting have weights older than 10 years vs. 17 percent of inflation-targeting.
- PPI coverage (industrial scope):
  - Recommended start: mining, manufacturing, utilities (MMU) and expand to services, construction, agriculture.
  - 33 percent of economies compile PPIs with more comprehensive coverage (commonly including agriculture and/or construction).
  - 40 percent (51 economies) cover only MMU.
  - 27 percent (34 economies) cover less than MMU (often manufacturing only).
  - Note: MMU can contribute as little as 10–15 percent of total GDP in some economies; services accounted for more than 65 percent of global GDP in 2017 and services account for more than 40 percent of total GDP in approximately 80 percent of economies.
- Timeliness and frequency:
  - SDDS requires monthly PPI with one-month lag; e-GDDS encourages dissemination within eight weeks.
  - Of 126 PPI-compiling economies:
    - 96 economies (77 percent) disseminate monthly.
    - 86 publish within one month after reference month.
    - 11 publish within 3 months after reference month.
    - 22 economies (17 percent) compile quarterly PPI: 4 publish within one month after reference quarter; 18 publish usually within 3 months after reference quarter.
    - 8 economies compile semiannual or annual PPI.
- Classification:
  - Activity classification recommended: ISIC Rev.4; EU uses NACE Rev.2 (based on ISIC Rev.4); NAICS comparable at upper levels.
  - Product classification recommended: Central Product Classification Version 2.1 (CPC); EU uses CPA.
  - Among PPI compilers:
    - ISIC Rev.4 or comparable used by 68 economies (54 percent).
    - 27 economies (21 percent) use older versions of ISIC.
    - 31 economies (25 percent) use other classifications.

### IV. Frontier Issues in Price Statistics
- Overview:
  - Emerging topics: scanner data, web scraping, digital economy coverage, quality adjustment challenges from new data sources and new items.
- Big data and price indexes:
  - Scanner data: high-volume POS transaction data capturing individual transactions, dates, quantities, descriptions, values.
    - Benefits: complete coverage of items sold by a retailer across locations; quantities and revenue enabling more accurate weights; potential to increase item samples and frequency.
    - Challenges: data access (purchase vs. free provision), IT infrastructure for processing, classification to CPI structure, data quality checks, index formula development, treatment of quality changes.
    - Limitations: scanner data often do not cover universe of CPI expenditure (may miss small independent stores and non-retail outlets).
  - Web-scraped data (internet):
    - Benefits: expand product coverage, more frequent pricing, automate large-scale data collection.
    - Challenges: limited item coverage in some countries, websites blocking scraping tools, legal/practical barriers.
    - Use cases: CPI price collection and residential property price indexes (RPPIs).
- Digitalization and quality adjustments in CPI:
  - Core principle: measure price change holding quality constant; quality adjustment required when items evolve or are replaced.
  - Types of new goods/services:
    - Evolutionary new goods: advanced versions/extensions of existing goods (example: mobile phones evolving into smartphones).
    - Revolutionary new goods: totally new products (example: tablets).
    - Services present additional measurement challenges (e.g., sharing economy platforms like Uber, Airbnb).
  - Survey results (Joint IMF-OECD, August 2017; sample size 43 economies):
    - 77 percent (33 countries) include e-commerce in the CPI.
    - Domestic e-commerce included more widely than cross-border e-commerce.
    - Advanced economies in sample: 100 percent include e-commerce in CPI.
    - Emerging and developing economies in sample: 37.5 percent (6 countries) include e-commerce in prices or weights.
    - 97 percent of advanced economies sampled (26 countries) include domestic e-commerce in CPI; 37 percent (10 countries) include cross-border e-commerce.
    - Divisions most commonly priced online: Recreation and culture (09), Furnishing, household equipment and routine household maintenance (05), Transport (07).
    - Only 3 of 43 economies include sharing economy services in CPI.
  - Quality adjustment methods in CPI (survey of 43 economies):
    - 6 economies do not use any quality adjustment method.
    - 27 economies use several methods depending on product group.
    - 8 economies use one quality adjustment method for all products.
    - Main methods: overlap (14 economies), hedonic regression (13 economies), judgmental methods (13 economies).
- Digitalization and quality adjustments in PPI:
  - Survey results (Joint IMF-OECD; sample size 36 economies):
    - 53 percent (19 countries) include e-commerce activities in PPIs; 89 percent of those are advanced economies.
    - Domestic e-commerce coverage patterns in PPI respondents:
      - Prices and weights included: 12 countries.
      - Weights only: 19 countries.
      - Prices only: 1 country.
      - 5 economies include domestic e-commerce but not cross-border for export.
    - Only two emerging and developing economies include e-commerce in PPIs.
    - Quality adjustment in PPI (36 respondents):
      - 9 economies (25 percent) do not quality adjust when making replacements.
      - 15 economies use various methods depending on product.
      - 12 economies use one quality adjustment method in all cases.
      - Main methods: overlap, hedonic regression, judgmental quality adjustment.
    - Only 2 of 36 countries include sharing economy services in the PPI.
  - Observations: PPI coverage of e-commerce and exports needs improvement.

### V. Summary and Conclusions
- Key overall findings:
  - CPI:
    - Availability: 196 economies.
    - Frequency of compilation: Monthly in 181 economies; Quarterly in 15 economies.
    - Timeliness of publication:
      - Less than 1 month: 168 economies.
      - Greater than 1 month: 13 economies.
      - Less than 1 quarter: 8 economies.
      - Greater than 1 quarter: 7 economies.
    - Classification: COICOP used by 148 economies; Other used by 48 economies.
    - Geographical coverage (expenditure weights): National 149 economies; Urban 28 economies; Capital City 19 economies.
    - Frequency of updating weights:
      - Within 5 years: 88 economies.
      - 6 - 10 years: 75 economies.
      - 11 - 15 years: 27 economies.
      - Greater than 15 years: 6 economies.
  - PPI:
    - Availability: 126 economies.
    - Frequency of compilation: Monthly 96 economies; Quarterly 22 economies; Other 2 economies.
    - Timeliness of publication (publication lags reported in Table 3):
      - Less than 1 month: 85 economies.
      - Greater than 1 month: 11 economies.
      - Less than 1 quarter: 4 economies.
      - Greater than 1 quarter: 18 economies.
      - Other: 8 economies.
    - Classification:
      - ISIC Rev.4: 68 economies.
      - ISIC Rev.3: 17 economies.
      - ISIC Other: 10 economies.
      - Other classification systems: 31 economies.
    - Frequency of updating weights:
      - Within 5 years: 45 economies.
      - 6 - 10 years: 58 economies.
      - 11 - 15 years: 12 economies.
      - Greater than 15 years: 11 economies.
    - Industrial coverage:
      - More than MMU: 34 economies.
      - MMU: 51 economies.
      - Less than MMU: 41 economies.
- Performance assessment:
  - CPIs: universally compiled, timely, and fairly well-aligned with international classification standards; but weights are outdated in 55 percent of economies; national coverage in 76 percent; 65 economies (33 percent) comply with international standards on all four evaluated criteria (weights, coverage, timeliness, classification).
  - PPIs: timely in many cases but major issues are undercoverage and age of weights; approximately 34 economies (27 percent) cover less than MMU; only 36 percent of economies have updated PPI weights within last five years; 30 economies (24 percent) comply with international standards on all four criteria.
- Policy and methodological implications and priorities:
  - Need to expand PPI coverage to include services and exports where relevant.
  - Need to update CPI and PPI weights more frequently, especially in economies with outdated weights (weights older than 5 years; many PPI weights older than 10 years in some groups).
  - Low-cost short-term improvements: data validation processes, treatment of temporarily missing prices, quality adjustment, use of proper index formulas.
  - Further research required on use of scanner and web-scraped data, classification and processing of big data, and improved quality adjustment methods for digitalized goods and services.
  - Repeat study in five years recommended to monitor compilation improvements; potential extension to export and import price indexes.
- Selected surveillance and capacity-development context:
  - IMF technical assistance and RCDCs support development in more than 108 economies.
  - Resource constraints and data availability drive some countries to postpone weight updates; donor support and targeted TA have supported price index projects.

*Source: IMF working paper chapter "13. Quality Adjustment Methods in PPI" (staff estimates, joint IMF-OECD survey results, and IMF DSBB inputs as cited in the chapter).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019163-print-pdf.pdf_