National Statistical Institute of the Czech Republic

Price Statistics / CPI section

Na padesátém 81 /100 82 Praha 10 — Strašnice /CZECH REPUBLIC

The harmonised index of consumer prices (HICP) is a consumer price index (CPI) that is calculated according to a harmonised approach. It measures the change over time of the prices of consumer goods and services acquired by households (inflation).

Due to the common methodology, the HICPs of the countries and European aggregates can be directly compared.

European classification of individual consumption according to purpose (ECOICOP)

The HICP covers the final monetary consumption expenditure of the household sector.

The main statistical variables are price indices.

The basic unit of statistical observation are prices for consumer products.

3.6.1. Statistical target population

The target statistical universe is the 'household final monetary consumption expenditure' (HFMCE) on the economic territory of the country by both resident and non-resident households. The household sector to which the definition refers, includes all individuals or groups of individuals irrespective of, in particular, the type of area in which they live, their position in the income distribution and their nationality or residence status. These definitions follow the national accounts concepts in the European System of Accounts.

3.6.2. Coverage error population

No national deviation from the definition of the statistical target population.

3.7.1. Geographical coverage

The HICP refers to the economic territory of a country as referred to in paragraph 2.05 of Annex A to ESA 2010, with the exception that the extraterritorial enclaves situated within the boundaries of a Member State or a country are included and the territorial enclaves situated in the rest of the world are excluded.

3.7.2. Coverage error regions

Generally, no geographical parts are excluded from coverage in weights; the price collection is not done in small towns and villages and in some districts, but the price level and development is well represented by the data collection elsewhere.

3.8.1. Start of time series

The HICP series started in January 1997.

3.8.2. Start of time series - national specifics

See the HICP database.

Czech HICP index series starts from January 1996 (for all-items HICP and 12 main headings).

Indices for classes and groups (ECOICOP3 and ECOICOP4) are available from December 1999.

Indices for sub-groups (ECOICOP5) are available from December 2014.

2015=100

The following units are used:

• Index point
• Percentage change on the same period of the previous year (rates);
• Percentage change on the previous period (rates);
• Percentage share of the total (weights).

HICP is a monthly statistics.

Harmonised Indices of Consumer Prices (HICPs) are harmonised inflation figures required under the Treaty on the Functioning of the European Union. Regulation (EU) 2016/792 of the European Parliament and the Council of 11 May 2016 (OJ L 135) sets the legal basis for establishing a harmonised methodology for the compilation of the HICP and the HICP-CT.

This regulation is implemented by Commission Regulation (EU) 2020/1148 of 31 July 2020.

Further documentation, can be found in Eurostat’s website - HICP dedicated section, namely recommendations on specific topics, under the methodology page, and guidelines, under the quality page.

None.

 Regulation (EC) No 223/2009 of the European Parliament and of the Council, of 11 March 2009, on the transmission of data subject to statistical confidentiality to the Statistical Office of the European Communities.

Aggregation of disclosive information, aggregation rules on aggregated confidential data, primary confidentiality with regard to single data values.

In line with the Community legal framework and the European Statistics Code of Practice Eurostat disseminates European statistics on Eurostat's website (see point 10 - 'Accessibility and clarity') respecting professional independence and in an objective, professional and transparent manner in which all users are treated equitably. The detailed arrangements are governed by the Protocol on impartial access to Eurostat data for users.

The HICP is released according to Eurostat’s Release calendar.

The calendar is publically available and published at the end of the year for the full following year.

HICP see   Eurostat's website https://ec.europa.eu/eurostat

CPI:

Aggregate consumer price index (for households total) broken by 12 divisions of ECOICOP classification is published as 'Press Release' on the 10^th calendar day following the end of reporting months at http://www.czso.cz/. The following day the publication 'Consumer Price Indices – Basic Information' is published.

More detailed data on the consumer price indices (down to ECOICOP4 = class level are then published monthly on web site in the document 'Consumer Price Indices (cost of living) – Detailed Information' which is released on the 25^th day following the end of reporting period. The publication 'Consumer Price Indices (cost of living) – Detailed Information' providing data for the whole year is published on yearly basis on the 25^th January. 

This means, that HICP details (down to ECOICOP5 = sub-class level) are published earlier in the Eurostat online database.

Publications are released according to the Catalogue of Products (https://www.czso.cz/csu/czso/catalogue-of-products) published on web sites of the CZSO.

Publications with the consumer price indices:

Monthly: Consumer Price Indices – Basic information; Consumer Price Index – Detailed information

Quarterly: Indicators of Social and Economic Development of the Czech Republic

List of news releases with consumer price indices:

Monthly: Consumer Price Indices – Inflation

Users are informed that the data is being released by a publication calendar.

There is no pre-access for any user.

Press release for CPI includes also HICP annual and monthly rate for all-items level. Published on the internet for all users. No privileged users exist. The link to Eurostat website is included. No more HICP data published by the NSI.

Monthly

Detailed HICP is not published by the Czech Statistical Office.

CPI is published up to COICOP3, for food up to COICOP4 levels

Flash Estimate is not calculated.

CPI All published indices are rounded to 1 decimal place including the basic index (2015=100).

All indices derived from basic indices are rounded to 1 decimal place before their publishing.

For HICP only monthly and annual rate is published, rounded to 1 decimal (same derivation procedure as Eurostat uses). HICP is marked as preliminary in CPI Press release. It contains a link to the Eurostat Press Release and the database for definitive HICP.

Press release and most of the publications are both in Czech and English.

HICP total index is published together with CPI Press Release (https://www.czso.cz/csu/czso/consumer-price-indices-inflation).

Press release for CPI includes also HICP annual and monthly rate for all-items level (Czech, English).

Press release is published on the internet for all users. No privileged users exist. The link to Eurostat website is included. No more HICP data published by the NSI.

Detailed HICP is not published by the Czech Statistical Office (CSO).

CSO publishes only CPI. Publications refer to Eurostat HICP database for HICP detailed indices.

CPI is published as follows:

Monthly data:

• Basic indices 2015=100
• Annual rates of change
• Monthly rates of change
• 12-month average rate of change

Regularly are calculated also breakdowns of m-o-m and y-o-y changes in the consumer price indices expressing the contribution (in % or percentage points) of individual components of consumption basket to the aggregate index change.

Link to the online database where the HICP is disseminated: HICP database.

Micro data are provided on demand only at NSI.

No privileged users exist, but there is closer cooperation with Czech National Bank and Ministry of Finance.

Paper dissemination was abandoned.

The HICP Methodological Manual provides the reference methodology for the production of HICP. (https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/KS-GQ-17-015)

10.6.1. Documentation on methodology - national specifics

Methodology description in English on website page: Inflation, Inflation Rate - Methodology | CZSO.

It includes:

• User's methodological manual
• Composition of the consumer basket (including history; down to ECOICOP5).

See Eurostat's Compliance Monitoring Report of January 2020 and follow-up report of February 2023. Both can be found on the page Quality - Harmonised Indices of Consumer Prices (HICP) - Eurostat (europa.eu).

System of regular mapping of user requirements, expert advisory boards, external and internal methodical audits, system of continuous data checks, detailed monitoring of data collection.

11.1.1. Quality management - Compliance Monitoring

Compliance Monitoring

11.1.2. Quality assurance - national specifics

Production/methodology

CPI production is subject to audits of experts from both the NSI and external users. Audits are targeted to methodology and the organisation of the CPI production process. Last national audits on CPI production were performed in 2005 and 2012.

Every year CSO holds regular meeting of advisory Committee on CPI based of experts from main users to ensure relevance of CPI production.

Details of the CPI/HICP are often solved in Advisory Committee

Continuous control of the data flow

Continuous controls are targeted on the quality of the data.

Stage 1 – Price information gathered by the price collectors is checked by the special central office staff. Price collection takes the first 3 weeks of the reference month. Extreme price deviations are sorted out with the help of software. There is no automatic rejection; all suspicious data are re-checked by collectors. Centrally collected prices are entered into the system by central office staff. A 'preliminary index' is then processed by central office staff. This stage ends towards the end of the reference month. Price collection independently supervised.

Price collection is randomly checked by a supervisor. (Price can be checked the same day in the same outlet by independent person)

Stage 2 – Central staff checks the preliminary results. It is possible to see all price details and to perform checking of both individual prices and indices. Then quality adjustment and re-sampling/replacement procedures are carried out. The CPI is then compiled again. This can be repeated several times. Final results are approved and prepared for publication.

Stage 3 – HICP is derived from CPI sub-indices. HICP-CT is derived from HICP sub-indices. HICP thus inherits data checks of CPI.

11.2.1. Compliance monitoring - last report and main results

The last available compliance or follow-up report can be found in the dedicated HICP section 'Compliance Monitoring' of Eurostat’s website, under 'Methodology'.

11.2.2. Quality assessment - national specifics

 National statistical audits targeted to the methodology and the organisation of the CPI production process.

CZSO has the policy of regular audit on important statistical indicators, including CPI. The audit is done by the senior statisticians from the Office, with participation of the High School of Economy in Prague and possible participation from important users.

Link to the page where the reports are published: Quality - Harmonised Indices of Consumer Prices (HICP) - Eurostat (europa.eu)

Eurostat recommends that / Follow-up to recommendations

As regards newly significant outlets and internet shopping, further analysis should be carried out in order to provide evidence that they are correctly represented in the HICP.

Survey of newly significant outlets and internet shopping: data of GfK were regularly used to map the market; representativity of the sample improved after analysis of first SD obtained; internet collection extended – see also 'Sampling design: outlets – detailed information'. In real market internet purchases are spread over many ECOICOP sub-classes and it is difficult to define representative items (EAs) to be collected by the regular price collectors. CZSO has rather plan to reduce slightly the number of regular price collectors and introduce 'internet price collectors' instead, who would survey prices for a specific 'internet consumer basket' reflecting the typical household’s consumption via internet.

The HICP sample be reviewed and updated annually, in particular for critical expenditure groups, in order to be able to adapt promptly to changes in market structures and include newly significant expenditures and outlets into the HICP as they become evident. The CPI/HICP sample is reviewed and updated annually. See also 'Comparability over time'; representativity of the sample improved after analysis of SD obtained; data of GfK are also used to monitor changes. Substantial methodical changes and changes of the sample are discussed also with the Advisory Committee for CPI. See also 'User satisfaction'.

Further analysis be carried out for distinguishing foreign business expenditures from foreign tourist expenditures.

Distinction of foreign business expenditures from foreign tourist expenditures is done by NA. Data are derived from Tourism Satellite Account (TAS).  See also 'Weights'.

Foreign expenditure on medical services in the economic territory be closely monitored and its significance for the HICP properly assessed. Eurostat welcomes the plans of CzSO to introduce dentist services paid by foreigners from 2009. Foreign expenditure on medical services are concentrated mainly to dentist services. The prices (for all medical services, if not payed from insurance) are the same as for the domestic patients. The prices are surveyed by price collectors.

Eurostat welcomes the plans of CzSO to introduce exchange office services into the Czech HICP. Exchange office services were not introduced into the Czech HICP because (1) Payments are concentrated to credit/debit cards (2) Methodology is not known/clear/harmonised.

Concerning income-dependent prices, price changes resulting from income changes of households, in particular services by elderly homes be reflected in the Czech HICP, as required by Article 4 of Council Regulation (EC) No 2166/1999 (repealed and replaced by Commission Implementing Regulation (EU) 1148/2020). Income-dependent prices – the area is monitored. See 'Social protection'.

The share of government payments for elderly houses remains negligible and constant over time: The elementary aggregate - E12.402.01 'Accommodation in old people´s home' has a relative HICP weight 4.24‰, of which estimated max 2 % is paid by the government.

For passenger transport by air, PCs as well as for fruits and vegetables, the CzSO review and monitor the representativity of the sample and its elementary aggregates, also taking into account the forthcoming standard on the seasonal items.

For passenger transport by air - introduction of internet prices collection was added to monitor properly the market (consumers buy often via internet and well in advance).

Total weight for passenger transport by air is not important. See 'Flights'.

Some items if fruit products were added like strawberry, watermelon and kiwis, as they are these days available the whole year. There is practically no important fresh fruits and vegetables, which are not included in the basket. See 'Seasonal items – detailed information'.

Regional expenditure weights should be used instead of population weights when aggregating regional data. Regional expenditure weights are now used instead of population weights. Regional weights are estimated as a proportion of relative share of HFMCE within country. The number of population and average salary is used for the estimation. See also 'List of elementary aggregates'.

CzSO undertakes the necessary steps to verify that, in case of missing observations, price changes are reproduced correctly and the representativity of the sample is maintained from month to month.

Supervision of price collectors improved. See also 'Devices for price collection'.

The CzSO review their procedure for case specific quality adjustment methods. In particular, Eurostat encourages CzSO to speed up the implementation of recommended quality adjustment (QA) procedures, with special respect to the implementation of consumption segments in case of personal computers. Number of representative items in PC equipment increased – there are 3 now: Notebook, Tablet, Notebook - sale via Internet. Expert judgement is used for QA.

In addition to being a general measure of inflation, the HICP is also used in the areas of:

• wages, social benefit and contract indexation;
• economic forecasting and analysis;
• measuring specific price trends;
• accounting purposes and deflating other series;
• inflation targeting by central banks;
• cross-country economic comparisons.

The euro area (evolving composition) index is used by the European Central Bank (ECB) as the main indicator for monetary policy management. The ECB and the European Commission's Directorate-General for Economic and Financial Affairs (DG ECFIN) use the HICP for assessing price stability and price convergence required for entry into European Monetary Union.

Other users include: National Central Banks, financial institutions, economic analysts, the media and the public at large.

12.1.1. User Needs - national specifics

In national legislation CPI serves mainly as compensative factor (e.g. national pension scheme).

Besides as a general measure of inflation, CPI have a variety of potential other uses, for example (level of importance descending):

• inflation targeting by central banks.
• wage, social benefit and contract indexation;
• accounting purposes and deflating other series;
• input to economic forecasting and analysis;
• measuring specific price trends;

At national level CPI is preferred to HICP.  Generally, HICPs are in particular considered suitable for cross-country economic comparisons.

To assist creation of CPI concepts, Advisory Committee is summoned by the Czech Statistical Office.

The Advisory Committee consists of representatives of ministries, CNB, research institutes, Economic University, trade-unions, organizations of the retired and selected branches of the CZSO. Meetings are held regularly.

The Committee discusses the updating of the consumer basket and methodology of the consumer price processing in the next periods. But the board has no direct influence on the HICP. However the active involvement of users at a national level may influence discussion at the European level.

Opinions of the users not in Committee are not systematically collected.

All COICOP indices at 5-digit level (ECOICOP) accounting for more than one part in a thousand of the total expenditure are produced.

Exceptions (no NA or HBS information is available concerning the value of consumption):

02.2.0.3                Other tobacco products

04.1.2                   Other actual rentals

04.1.2.1                Actual rentals paid by tenants for secondary residences

04.3.2.5                Services of carpenters

05.2.0.9                Other household textiles

09.2.1                   Major durables for outdoor recreation

The same is valid for HICP-CT.

The accuracy of HICP is generally considered to be high. The accuracy of source data is monitored by assessing the methodological soundness of price and weight sources and the adherence to the methodological recommendations. There is a variety of data sources both for weights (National Account data, Household Budget Survey data, etc.) and prices (visits to local retailers and service providers and central collection via mail, telephone, e-mail and the internet are used). The type of survey and the price collection methods ensure sufficient coverage and timeliness. The outlets, from which prices are collected, are chosen to represent the existing trade and services network and they are based usually on three main criteria:

• Popularity with consumers,
• Significant turnover from consumer sales and
• Availability of goods and services included in the HICP basket.

All the private households in the economic territory of the country are covered, whether resident or not and irrespective of their income.

The HICPs are statistical estimates that are subject to sampling errors because they are based on a sample of consumer prices and household expenditures, which are not the complete universe of all prices/expenditures.

The exact sampling error is not known.

The NSI tries to reduce the sampling errors by using a sample of consumer prices that is as large as possible, given resource constraints. The NSIs use models that optimise the allocation of resources by indicating the number of prices that should be observed in each geographic area and each item category, in order to minimize the variance of the all-items index.

Non-probability sampling is used.

Possible influence to the non-sampling error and the bias risks associated with HICP:

• Under coverage - consumption of upper class is not well known.
• Survey instrument, respondent and interviewer where relevant - minimised by independent controls of price collectors.
• Unit (non)response including causes for non-response, item non-response for key variables - no serious problems so far registered.
• Data editing, coding and imputation where relevant - minimised by consistency checks, comparison of similar prices, previous prices; imputations is limited only to seasonal items

The full set of HICPs is published each month according to Eurostat’s Release calendar, usually between 15 and 18 days after the end of the reference month.

Each year, the January release is published at the end of February to allow for the annual update of the weights, both of individual product groups and the relative country weights of Members States in the country group aggregates.

The euro area flash estimate is published on the last working day of the reference month or shortly after that.

Since the March 1997, launch of the HICP release, the HICP for the country groups aggregates has always been published on the dates announced in Eurostat’s Release calendar.

HICPs across Member States aim to be comparable. Any differences at all levels of detail should only reflect differences in price changes or expenditure patterns.

To this end, concepts and methods have been harmonised by means of legislation. HICPs that deviate from these concepts and methods are deemed comparable if they result in an index that is estimated to differ systematically by less than or equal to 0.1 percentage points on average over one year against the previous year (Article 4 of Council and Parliament Regulation (EU) 2016/792).

HICP data are fully comparable over time. There have been several improvements in methodology since the HICP was introduced with the aim of improving reliability and comparability of the HICP. 

Information on the main re-referencings and revisions

HICP is derived from CPI, CPI is biannually chained index (minor updates of sample continuously, minor methodological amendments annually).

No revisions of published CPI or HICP indices ever occurred (since 1995). 

Overview of HICP re-referencing:  1996 = 100, 2005 = 100, 2015=100.

No statistical breaks or interruptions in the HICP series due to changes in the methods have occurred.

Differences between the HICP and national CPI

The following expenditure is included in the HICP but excluded from the national CPI: consumption expenditure of non-residents in the economic territory.

The following consumption expenditure is included in national CPI but excluded from HICP: Monetary expenditure of investments in the owner-occupied house is measured (purchase of the dwelling itself, self-builders, renovations and major repairs). Insurance of dwellings of owner-occupiers is included.

Games of chance, drugs, prostitution etc. are excluded both from CPI and HICP.

Methodology, territorial coverage and product coverage is the same. Small exception HICP index for COICOP 11.2 Accommodation services differs from CPI (luxury tourist hotels, namely in Prague are not in CPI, but they are in HICP).

Sharing information between statistical units

There is methodological cooperation and sharing of experience namely between: CPI and PPI, CPI and Business Services Price Index.

For revisions of CPI/HICP cooperation with NA and HBS is necessary.

The HICPs are internally coherent. Higher level aggregations are derived from detailed indices according to well-defined procedures.

Not available

The HICP series, including back data, is revisable under the terms set in Articles 17-20 of Commission Implementing Regulation (EU) 2020/1148.

17.1.1. Data revision - policy - national specifics

There is no specific national policy for correcting mistakes and errors in CPI/HICP production. Discovery of an error would have to be considered by a special body, consisting of the President, statistical directors and relevant experts. The solution should follow CZSO general principles on dissemination policy. Regarding the specific solution found, a possible revision of previous results might occur. For HICP the relevant Commission Implementing Regulation (EU) 2020/1148 articles concerning revisions of HICP would be respected.

No mistakes have been discovered yet that would have required treatment according to this regulation on the compilation of the HICP.

Transparency and control of disseminated data is supported by the fact that CPI detailed indices (to the level of EA) can be provided to users on demand. Typically Czech National Bank and Ministry of Finance require regularly the information (data and explanation). 

Revisions of the results (index numbers)

Index series were not revised in the period of 2010-2022.

Revisions of the methodology and weighting schemes

Major (planned) revisions of detailed CPI basket weights based on specifically targeted Household Budget Survey are performed every 5-6 years. The last one was done in 2021. For HICP based on CPI sub-indices, the revision (planned) of weights based mainly on National Accounts data is performed every year, with the possible inclusion of new sub-indices. Methodical changes induced by the harmonisation process are usually introduced both in CPI and HICP. Where the HICP methodology might be contradictory to the national concept of CPI, a specific sub-index might be calculated as a duplicate – one for CPI use, the other for HICP.

18.1.1. Weights

See details in the following sections.

18.1.1.1. Compilation at elementary aggregate level

There are 463 explicit weights at Elementary Aggregate level plus 22 explicit weights of complex sub-indices (mostly tariff items, like electricity, gas, transport services, financial services, insurance, etc.).

Each tariff sub-index has internal weights of its items (but variable number through time).

The weighting reference period used in 2023 is the year 2022; the weighting reference period used in 2022 was the year 2021.

Generally, the expenditure shares at sub-index level are derived from data sources for year t-2 and (partly) year t-1; these shares are updated to make them representative for t-1.  

Weights of elementary aggregates are updated every 5-6 years (more often for some main headings, like food and beverages). Weights of complex sub-indices and their internal weights (weights of its internal structures) are updated annually.

The lowest aggregated level is ECOICOP5 where weights are sum of subordinated EAs and (possible) sub-indices.

The main source of EA weights is HBS; also NA, scanner data (namely for food and beverages), market research (GfK).

Other important information sources to improve details (individual products): Insurance Association (insurance), electricity and gas distributors (electricity, gas), Czech Railways (passenger railway transport), banks (financial services), external experts (new and second-hand cars), municipalities (local public transport in big cities).

Judgmental adjustments or equal weights are used only if there is no other source of data available (exceptionally).

For the purpose of the consumer price revision the Social Surveys Unit of the CZSO carried out an extraordinary annual survey with very detailed breakdown of household financial resources in 2016.

Source for weights for regions is regional NA.

HBS data are often corrected by NA. Weights are corrected for net weight approach for insurances. Moreover the data are compared with previous sets of weights to identify possible errors and to trace possible/impossible changes within consumption structure of households. Results are discussed with NA in order to improve the quality of weights. 

No explicit weights for outlets are generally used. Where scanner data is used (food and non-alcoholic beverages, alcoholic beverages and tobacco) the weights for outlets (retail chains) are derived from the scanner data. For tariff price index - explicit weights are derived from the turnover of the tariff providers (eg. electricity - from the electricity companies).

18.1.1.2. Compilation of sub-index weights

General procedure

The weights at sub-index level (ECOICOP 5) are generally derived from annual National Accounts based on t-2 and usually considered to be an appropriate estimate of t-1 expenditures. Weights are always price updated to December t-1. Weights for insurance services  are calculated as a moving average of the last 3 years. Weights of EA are not explicitly price updated.

Specific procedure (weights for 2023)

The expenditure shares at sub-index level: The estimates for the 2022 expenditures (t-1) were based on the available quarterly NA data of the year 2022 at the level of 3-digit ECOICOP. This information was used as a 'base frame' for the weighting scheme to be used for HICP in 2023. Then, the most recent version of 2021 NA expenditures (t-2) were used as a source to make finer breakdown of the weighting scheme at ECOICOP 4-digit and 5-digit levels. The estimates based on NA data were adjusted for the differences between HFMCE and the total final household consumption. The final household consumption expenditures used followed the domestic concept.

Main data sources for the 2023 HICP relative weights

• Quarterly NA 2022 household expenditures as a frame for the compilation of the weights (Q1, Q2 and Q3; Q4 was estimated),
• Yearly NA 2021 household expenditures for most of the detailed weights,
• 2022 Short term statistics (STS) for some individual detailed weights (where necessary),

Integration of the 4th quarter of t-1 (2022) into the calculations of the sub-index weights

The national accounts ECOICOP 3-digit level data were available for the first three quarters of 2022 and for all quarters of 2021 and 2020, while there were no data for the fourth quarter of 2022. Data for the fourth quarter of 2022 were estimated by applying the average quarterly change in consumption between Q1-Q3 of 2022 and Q1-Q3 of 2021 applied to Q4 of 2021. The calculations were done at the level of each ECOICOP group (3-digit). Some further adjustments of the Q4 were done.  

18.1.1.3. Compilation of sub-index weights

The weighting reference period for ECOICOP 5 levels and higher used in 2023 is the year 2022.

Weights of higher aggregates are sum of weights of subordinate levels (consistency).

18.1.1.4. Weights – plausibility checking

Weights for the annual updating obtained from NA are checked against long term trends in the time series or relative weights.

Similar process is done for the weights at lower levels, if possible/applicable.

18.1.1.5. Price updating

General procedure

Price-updating (from t-2) to the previous year's (t-1) average price level is not done (usually).
Price-updating (from t-1) to the previous year's December price level (Dec of t-1) is performed every year.
Price-updating practice is applied only at ECOICOP 5 level.
Relative weights for elementary aggregates within sub-groups (ECOICOP 5 level) are updated only as a part of resampling or ad-hoc changes (similar to resampling).

Specific procedure (weights for 2023)

Since the HICP of year t is defined as a Laspeyres-type index where the weights refer to t-1 and the price reference period corresponds to December t-1, the expenditure shares for 2022, obtained for each sub-index, were multiplied by the difference (ratio) between the corresponding HICP sub-index for the average of 2022 and December 2022, thus following the normal procedure. The price-updating between the year t-1 (2022) and the December t-1 (Dec 2022) was done at the level of sub-classes (5-digit ECOICOP). 

18.1.1.6. Compilation of total household final monetary consumption expenditure

Data sources used to estimate the total Household Final Monetary Consumption Expenditure of 2023

The main sources used were the household final consumption expenditures (HFCE) for the first three quarters of 2022 and for all quarters of 2021 (from the quarterly national accounts).
An auxiliary source used was both the total and the detailed yearly HFCE for 2021 (from the preliminary NA data), where the detailed expenditures were used as a secondary data source for some expenditure categories excluded from HICP country weights.

Data for the fourth quarter of 2022 were estimated by applying the average quarterly change in consumption between Q1-Q3 of 2021 and Q1-Q3 of 2022 applied to Q4 of 2021. 

The sources used for the compilation of sub-index weights and for the compilation of country weights were the same and therefore compatible. 

18.1.2. Prices

The price data is mostly based on survey; administrative data sources are used only in specific cases.

18.1.2.1. Data Source - overview  

The price data is mostly based on survey; administrative data sources are used only in specific cases. Scanner data are collected, but not used as price information yet.

Main source of weights on the level of COICOP5 is NA (because of the domestic concept HBS data have to be corrected by NA).

The main sources for the detailed weights are National Accounts and the Household Budget Survey combined with other auxiliary sources.

There are 430 HICP Elementary Aggregate weights at national level of items plus 17 weights of various (mostly tariff-like) sub-indices. Each sub-index has internal weights of its items (but variable number through time of items within each of the 17 tariff sub-indices).

Considerable part of the consumer basket (cca 31% in weight terms) is based on scanner data (SD).  This concerns ECOICOP main headings 01 Food and Beverages, 02 Alcoholic Bevarages and Tobacco and some other smaller parts of the basket. The source of detailed weights under the level of ECOICOP5 is based soleley on SD. There is approximately 150 000 individual SD items with its own relative weight. 

HBS data are corrected by NA. Weights are corrected for net weight approach for insurances. Moreover the data are compared with previous sets of weights to identify possible errors and to trace possible/impossible changes within consumption structure of households. Results are discussed with NA in order to improve the quality of weights used in final stage (e.g. for the calculation of the new chain in the index).

18.1.2.2. Scanner data - general information

 Scanner data are collected and processed for eleven big hypermarket chains and three small supermarket chains. Data have been included into CPI/HICP calculations.

18.1.2.3. Web scraping - general information

Bulk web scraped data not used.

18.1.3. Sampling

18.1.3.1. Sampling design: locations for survey

HICP covers the entire area of the country. All of the 14 regions (NUTS3) are covered. Within each region several districts are selected (that way, about half of all existing districts are selected).

Prices are collected in all the regional capitals and in of all of the selected district centres. Rental is also collected for some smaller towns.

Villages are excluded from price collection.

18.1.3.2. Sampling design: outlets

In total, about 8 000 outlets in the country are chosen for the survey. Outlets are selected on the regional level (but with central recommendations as to the proportion of basic outlet types) according to local knowledge about retail sales distribution for each of the consumer basket items (EA-elementary aggregate). In practice, outlets are selected/replaced by price collectors mostly after identifying the non-availability/non-representativity of the outlet previously selected.

Outlet selection is centrally adjusted based on information about shares of sales for individual groups of products between basic outlet types and between individual hypermarket chains. A major outlet sample update occurs every 5-6 years and coincides with the general index revision (based on detailed Household Budget Survey).

Internet shopping is included in the index; market stalls are included for fresh fruit and vegetables.

A kind of general outlet survey is done every year in cooperation with GfK for particular goods and services.

Reporting network is constituted by shops (including hypermarkets, supermarkets, chain stores), service shops and other institutions providing services (approx. 8000) – hereinafter only the 'respondents' (outlets).

Consumer prices are surveyed in 35 selected districts in the Czech Republic and in the Capital of Prague.

The Capital of Prague splits into 8 territories (strata), Brno into 3 strata, Plzeň and Liberec split into 2 strata. Maps of strata are deposited in the Regional Office of the CZSO in Hradec Králové (hereafter RO in Hradec Králové) and also with one of the respective employees of field survey department.

Majority of respondents are selected by the field survey department employees of the CZSO upon an agreement with the owner (authorised person) of a selected reporting unit of the reporting network. For centrally surveyed prices the selection of reporting units is done by the Consumer Prices Statistics Section of the CZSO Prague.

Respondents are important centres of sale or provision of services to the population in the district, i.e. the reporting network includes hypermarkets and supermarkets. Included are also Asian sellers in stone shops, however, not those in the open-air market places.

18.1.3.3. Sampling design: newly significant goods and services

Identification of new products is done both on the level of price collectors and the regional (2)/central (1) staff. Decision about necessity of their inclusion into survey is often subjective.

There is no systematic comprehensive source of information but a collection of small market researches on particular goods and service classes, possibly with the help of producer organisations. 

CZSO regularly uses results of GfK market survey. Next sources are scanner data, detailed HBS, detailed data from providers of (tariff) services.

The inclusion of a new variety (change of parameters within the description of EA) is done every month. Quite a ('radically') new EA is normally introduced only in December, but only exceptionally in the current month, if necessary

Price data is collected every month.

18.3.1. Price collection surveys

Definition of units (for details see Organisation Structure of the Czech Statistical Office | CZSO

https://www.czso.cz/csu/czso/organisation_structure_of_the_czech_statistical_office )

Central Price Statistics Department - Consumer Prices Statistics Unit (C).

Regional Office Hradec Králové - Section for Processing of Consumer and Agricultural Prices (R).

Regional price data collectors (DC).

Price collection is performed in Prague and in 35 districts (NUTS4) out of 75 altogether. The districts are evenly spread over the country. The regional collection is done in the principal town of the districts and close surroundings (namely hypermarkets and supermarkets around). Rental is surveyed also in other small towns outside the seat town of the district.

Procedure

Stage 1 – Data collection, data validation, pre-processing

Price information gathered by the collectors (DC) is checked by the (R). In case of doubt they will contact by phone the field survey collectors (DC) and verify the prices. In exceptional cases they verify the surveyed price directly in the reporting unit. In addition, they perform scheduled control of the field survey department employees directly in the field. On these inspections records on defects, if any, are taken stating the date of their removal.

Price collection takes first 3 weeks of the reference month. Extreme price deviations are sorted out with the help of software. No automatic rejection is done, all suspicious data are checked again by (DC). Centrally collected prices are entered into the system by (C). (The database can be simultaneously shared between (C) and (R)). A 'preliminary index' is then processed by (R). This stage ends towards the end of the reference month.

Stage 2 - QA-procedures, resampling, result validation, dissemination

Staff (C) checks the preliminary results. It is possible for (C) to see all price details and to perform checking of both individual prices and indices. Quality adjustment and resampling/replacement procedures are done by (C). The 'Index' is then compiled again by (R). This process may take several rounds. Final results are approved by (C) and prepared for publication.

There exists two different manners of data collection. Part of consumer basket is constituted by centrally surveyed prices  (e.g. same prices for the whole Czech Republic or average prices detectable from various reports, prices surveyed on the Internet). These prices are observed and surveyed by employees of the Consumer Prices Statistics Section of the CZSO Prague (C), the RO employees in Hradec Králové (R) and employees of the Field Surveys Department (DC). Prices are surveyed on monthly basis.

The share of centrally collected prices is about 10 % (5,500 to 6,000).

Price collectors have permanent job. They are partly used also for Household Budget Survey, which enables to share the experience with a related statistical field.  There is, altogether, 47 price collectors, of which 8 in Prague.

Price collectors (DC) have written methodological instructions 'Instructions for survey, verification and presentation of data on consumer prices of goods and services included in the consumer basket' (about 30 pages including appendices) concerning namely: outlet selection, priced item selection (EA-list), treatment of price reduction, data recording, treatment of seasonal items, replacements, etc.) Amended one time every year. Prepared by (C) together with (D).

Every year there is a meeting/training for data collectors (DC) and regional processing staff (R) organized by central staff (1).  The meeting is held on level (C). It includes also multilateral exchange of information between (C), (R) and (DC), like news about changes in the market and how to treat them; the changes of the concept, practical experience/solutions etc.)

18.3.2. Timing of price collection

The prices are collected between the 1st and the 20th day of each month.

Generally, all EAs are priced every month, but some of them do not have price or do not occur continually in the market like:

• seasonal goods (clothing, footwear),
• yearly charges and fees (education fees, dancing courses, almanac/diary…)

Prices collected more frequently than monthly apply to fresh fruit and vegetables – pricing is spread evenly over first 3 weeks of the current month. This is arranged by the appropriate division of the set of surveyed regions.

Petrol and diesel fuels for personal transport equipment are priced every week – four times per month.

Prices of goods are entered for the month in which they are observed, prices of services are entered for the month in which the consumption of the service at the observed prices can begin. Package holidays are treated as services.

Price collectors: Electronic devices used for primary data collection/editing – first validation for price extremes and correct editing.

Data processing: Second validation – checking for extraordinary low/high prices within Elementary Aggregate, checking for correct codes, checking for extra price movements within time.

Final calculations, publications: Validation of individual results after quality adjustments procedures. Assessment of overall results.

18.4.1. Data validation - price data

Price collectors: Electronic devices used for primary data collection/editing – first validation for price extremes and correct editing.

Price information gathered by the price collectors is checked by the special central office staff. Price collection takes the first 3 weeks of the reference month. Extreme price deviations are sorted out with the help of software. There is no automatic rejection; all suspicious data are re-checked by collectors. Centrally collected prices are entered into the system by central office staff.

Price collection is randomly checked by a supervisor. (Price can be checked the same day in the same outlet by independent person)

Data processing: Second validation – checking for extraordinary low/high prices within Elementary Aggregate, checking for correct codes, checking for extra price movements within time. A 'preliminary index' is then processed by central office staff. This stage ends towards the end of the reference month.

Central staff checks the preliminary results. It is possible to see all price details and to perform checking of both individual prices and indices. Then quality adjustment and re-sampling/replacement procedures are carried out. The CPI is then compiled again. This can be repeated several times. 

Checks for data consistency.

Final calculations, publications: Validation of individual results after quality adjustments procedures. Assessment of overall results. Final results are approved and prepared for publication.

HICP is derived from CPI sub-indices. HICP-CT is derived from HICP sub-indices.

CPIs and HICPs are approved by the highest management of the CZSO in the context of the current economic situation in the country.

Price Statistics Department has to explain all the price movements in CPI in relation with the changes of the supply/demand on the consumer market, administrative changes and external influences on the economy.

18.5.1. Elementary price index formulae

Czech CPI/HICP is a Laspeyres-type index covering the ECOICOP international classification.

The ratio of the arithmetic means is universally used for computing the elementary price indices. No alternative formula is used.

The average price of the elementary aggregate for the CR is calculated as weighted arithmetic mean of prices for regions calculated by simple arithmetic mean. The weight corresponds to the estimated relative consumption in the region in 2020.

An exception is the average price calculation for the Czech Republic in respect of the following representative items: rentals and services related to the use of dwelling in rental houses, payment for the use of co-operative dwellings and services related to the use of co-operative dwelling, spa treatment fully covered by patient, second-hand cars, fuel, air transport, tow tickets, holiday abroad (stay by the sea-side), school fees at higher technical school, university admissions fee, school fee at private university, lunch in the canteen, accommodation in student hostel, insurance of a house, dwelling and accident insurance, insurance of cars including compulsory cover. Average price calculation for these items is specific (e.g. number of patients in spa).

The formula is compliant with HICP legislation. Using of arithmetic averages (rather than geometric mean) is preferred for the parallel use of mean prices as public 'Average prices' (better understanding, interpretation). 

The number of decimals

Full price for observation (= one decimal: price xx.x CZK), no rounding.

For consecutive calculations 2 decimals are used (= usually 4 to 6 valid digits).

For weights, 6 decimals are used.

For internal calculations (aggregations, chaining) 4 decimals are used for basic indices.

Basic indices are then rounded to 1 decimal place.

Rates of change are derived from rounded basic indices.

All published indices are rounded to 1 decimal place.

All indices derived from basic indices are rounded to 1 decimal place.

All published indices are rounded to 1 decimal place.

Truncation is not used.

18.5.2. Aggregation of different data sources

HICP is derived from CPI; CPI is chained index with bi-annual chaining - see 18.5.1.3. Chaining and linking method.

CPI index is Laspeyres index. Price indices for EAs are aggregated bottom-up.

Some parts of consumer basket are not covered by EAs, but by a more complex ready index (called 'specific sub-index'). These are cases with more complex price changes, which cannot be simply captured by a few EAs - typically tariff prices.

The use of specific sub-indices was newly introduced after the 2005 CPI revision. These apply to base indices of some groups on the ECOICOP5 level which are calculated outside the consumer basket. They belong to the group of centrally observed prices (sub-indices are compiled by the Consumer Prices Statistics Section of the CZSO Prague). This applies especially to the groups of goods and services with tariff prices (electricity, gas, railway transport, telecommunications, financial services) where there exists a wide range of frequently updated supply of products. For these items the method of selecting representatives applied for the consumer basket formation could result to certain index deterioration. To this end, the price development of these items is calculated independently either on the basis of entirely exhausted supply or from strongly prevailing supply of these products. Sub-indices are used in the formula for the overall consumer price index calculation and by their weight in the consumer basket (determined like for other items) they are aggregated to higher aggregates and the total. Besides the above groups the method of using of the sub-indices is applied for market rentals, imputed rentals (CPI only), prescription medicines, over-the-counter medicines and other pharmaceuticals, passenger transport by air and insurance.

Data are generally processed in ORACLE, some small, but complicated parts (e.g. indices for tariff prices) in Excel. All outputs, including intermediary, are in Excel. Excel is used for outputs, including publications. Excel is also used for intermediary outputs and for specific calculation of some sub-indices.

18.5.3. Chaining, linking and splicing methods

CPI calculation

Starting from 2022, the price base period was changed from December 2019 to December 2021. The calculated indices are chained at all levels of the consumer basket to the basic index (with index base 2015 = 100). Therefore, a multiplicative constant is used to transform the index with the base period December 2021 = 100 into an index with the base 2015 = 100. Thereby, a continuation of the existing long index time series 2015 = 100 is ensured.

CPI is a chained index with bi-annual chaining.

Then, the remaining indices are derived (previous month = 100, corresponding period of last year = 100 and annual rolling average, i.e. the average of index numbers over the last 12 months to the average for the previous 12 months).

HICP calculation

HICP is derived from CPI the following way:

• HICP link index (base December 2022 = 100) is equal to the corresponding CPI index (base December 2020 = 100) for sub-class (ECOICOP5) (some exceptions exists).
• HICP link index (base December 2022 = 100) for higher aggregate of COICOP classification (class, group, division, total) is calculated as arithmetic weighted average of all subordinated indices on sub-class level (HICP link index with base December 2022 = 100).
• HICP is chained index with annual chaining.

HICP and CPI weights differ (mainly due to domestic vs. national concept and non-inclusion vs. inclusion of OOH)

18.5.4. Quality adjustment – Detailed information

The generally used QA method is bridged overlap within EA. This means first: if a product-offer (variety) can no longer be found in an outlet, another product-offer (the same or a different variety within the same EA) is selected to replace the old one. Only if these product-offers are not considered essentially equivalent (i.e. the product-offers do not belong to the same variety or the outlet changes), this observation is effectively excluded from the index and the price change is estimated using the remaining observations. (A new base price is imputed to obtain this effect.) After this QA procedure, the number of price quotations within EAs remains constant through time.

A (straight) overlap method is used for the replacement of EA (EA replaced by EA '1:1'). This overlap itself does not affect the price index (at least in formal terms) either. Typically (but not necessarily) this happens in the month of December.

On a more general level, re-sampling and replenishment procedures are possible higher up, within the COICOP 5-digit (national extension of COICOP) level, too, in that a number of EAs are replaced by a number of new EAs (EAs replaced by EAs 'm:n'). This involves changing the weighting scheme within the relevant COICOP group.

Explicit methods (mostly expert judgement) are used only in exceptional cases and are mainly used for the replacement of EA by new EA, where the overlap or bridged overlap method is not considered to be appropriate/sufficient.

Note that resampling itself, on whatsoever level, does not change the price index (generalised overlap).

Missing EA in current month is exceptional phenomena and is treated individually. Last year 11 EAs of 430 in the basket were replaced (mostly in December as planned replacements). SD items are replaced whenever necessary.

Quality adjustment of the price index is performed when the product which was subject to price survey is replaced by another one. The change is due to the absence of the original product in the market or decreased demand. Then the representative in the consumer basket is replaced by a similar product or product in great demand. The other, much more frequent case of substitution, takes place when the product selected by the field survey department employee for the purpose of price survey (variety corresponding with the description) ceases to be sold in the selected shop or when the shop is liquidated. The filed survey department employee shall select (being advised by the shop staff that the absence is permanent) another similar product (complying with the description) or survey prices in other shop.

 A) Direct adjustment

 A1) Direct comparison of prices (dealing with two comparable products)

New item is considered directly comparable, if: 

• It is produced by the same manufacturer
• It is produced from the same materials and has the same or similar technical parameters which are important for customers (use value)
• Has the same measurement unit
• Has the same type of package
• The difference between the original and new items are not significant and may refer to the customer taste and their personal preferences, such as colour, design, shape, decorative elements, etc. 

If the products are considered comparable in terms of quality, then the difference in their price shall fully reflect price increase or decrease of the price index. Such method is used by the CZSO for the substitution of varieties within price index.

A2)   Expert estimate (dealing with 2 incomparable products)

However, in most cases the original and new products are of different quality and their prices cannot be compared directly. In such case it is necessary to adjust the price difference by quality change of a new and original product which means to express in figures which part of price difference of both products is caused by the quality change and which part represents price increase or decrease.

Information can be obtained as follows:

• by an expert estimate of the field survey employee in collaboration with the shop staff
• by an expert estimate of external experts
• by the use of hedonic regressive patterns

The CZSO sometimes uses especially services of external experts during the substitution of a representative in the consumer basket. The CZSO does not so far use the last alternative because hedonic pattern requires a large-size database with a wide scope of product characteristics which demands significant expenses related to the creation and maintenance of such database.

B) Imputation methods (dealing with 2 incomparable products)

B1) OVERLAP method 

The overlap method is used where it is possible to observe the price of both the original and new product in the same time period (month) which is determined as an overlap. In the overlap month the prices of both products are surveyed. Price development in this month is measured on the original product, next month on a new product using its price surveyed in the previous overlap month. The price difference of the original and new products will not feed into the price index. This method is used by the CZSO for the substitution of representatives in the consumer basket (especially in the price index revision when new components are imputed into the consumer basket).

B2) BRIDGED OVERLAP method 

This method is used where it is impossible to survey the price of new and original component in one month. Price development of similar products which appeared in the reference as well as previous month will be assigned to the product newly included into the consumer basket. This method is used also in cases when it was impossible due to various reasons to survey the price of the component in both months (missing prices or, in contrast, newly appearing ones). The CZSO uses this method each month for adjusting of variants surveyed by the field survey employees and for adjusting of prices surveyed only in one month. 

In substitutions of representatives, the development of a similar representative or group of representatives can be used as a bridge. This method is used by the CZSO in substitution of representatives in the consumer basket (especially in the revision of price indices when new components are imputed into the consumer basket).

Note. The use of bridged overlap is limited in several cases listed in HICP recommendation on bridged overlap. The following cases are considered separately:

• The last price of the replaced (old) product-offer is a reduced price.
• The first price of the new product-offer is a reduced price.
• The first price of the new product-offer is unusually high.
• The matched sample of product-offers includes reduced or atypical prices, or shows a downward price trend during the product life cycle.

In the abovementioned cases, the preferred method instead of bridged overlap is direct comparison, if considered appropriate. Otherwise, an expert view on QA factor might be considered (explicit QA method).

Practical use of QA

In price index quality adjustment the most serious problem is to decide to which extent the newly include product is comparable with the original one, i.e. when to use direct price comparison and when and by which method to make the quality adjustment of prices. To this end the CZSO uses the following procedures:

 a) Substitution of a variety by the field survey employee during the data collection:

The main arbiter for the assessment of comparability during the variety change is the field survey employee who knows best both types of product is able to obtain information from the shop staff. If he/she considers the product comparable the product price is marked by 'v' code (similar product). If he/she considers the difference in quality important the price is marked by '3' code (another variety). In both cases the arbiter shall specify the main characteristics of a new product in the note. This information is sent to RO in Hradec Králové where price verifiers on the basis of data in the note concerning current and previous months (or telephone questions) will verify the proposal of the field survey employee. The last supervision of substitutions and the 'BRIDGED OVERLAP' method of adjustment is performed by employees of the Consumer Price Statistics Department.

 Example: For 'A' representative the following price groups were surveyed in month _t-1 and month _t:

The non-adjusted index _t / _t-1 = 120/100 = 1,2

In quality adjustment using the BRIDGED OVERLAP method for the m-o-m measurement of 'A' representative development only those prices were used which were comparable in both reported months. 

Index  _t / _t-1 =  110 / 102 = 1,08

The quality adjusted price index is imputed to the 'A' representative by means of the base period price change: 

Price in BP _t-1 – price in base period of 'A” representative in month _t-1         80,0

IBAZ _{t-1  -} base index of “A” representative in month _t-1                              125,0

 IP _t  unadjusted – m-o-m index unadjusted                                             120,0

 IP _t  adjusted – m-o-m index – adjusted                                                  108,0

 IBAZ _t – base index of 'A' representative in month _t                                     125,0 * 108,0/100= 135,0

 Price _t – average price of 'A' representative in month _t                                                 120,0

 Price in BP _t – base period price for 'A' representative in month _t                                      120,0/1,35 = 88,89

 or

 Price in BP ZO _t – price in base period for 'A' representative in month _t            80,0 * 1,2/1,08 = 88,89

 b) substitution of a representative in consumer basket     

Substitution of representatives (EAs) in consumer basket takes place in the consumer price index revision continuously, as needed in the period between revisions. The substitution owes to a constant update of assortment in the market and to changing consumption patterns of the population. For products newly included into the consumer basket it is difficult (practically impossible) to back trace the price in base period since the product has not existed then. In such cases the new price representative follows the previous development of the original representative using the OVERLAP or BRIDGED OVERLAP method.

 Example:

'A' REPRESENTANTIVE will be replaced by 'B' REPRESENTATIVE.

 - by the OVERLAP method 

- by the BRIDGED OVERLAP method

c) quantity adjustment

Quantity adjustment refers to items whose price in the period t was surveyed for other unit of measurement than in the period t-1. Quality adjustment is performed by a linear conversion to the same unit of measurement. If (upon conversion to the same unit of measurement) the price is changed due to a different unit of measurement, the following procedure depends on whether or not:

• original package disappeared from the market (e.g. weight reduction of a chocolate bar from 50 to 40 g)
• original package is not available in the shop store only in month t  (e.g. washing powder of the weight of 1 kg is replaced by 3 kg package)

In the first case the price change is reflected in the index as a price increase or decrease. In the second case the products are considered incomparable and the quality adjustment procedure is the same as in the variety substitution. This means that the field survey employee will assign code 3 to the surveyed price (change of variety or weight in grams) and this price difference is not included into the index.

18.5.5. Seasonal items

For fruit and vegetables the prices are collected evenly from the 1st to 20th day of a month, always in 1/3 of selected districts in a week. The field survey employee collects 3 prices for these items. There are no EAs in the sample, for which the price would be unavailable. (Some strongly seasonal fruit and vegetables with very limited season are excluded from CPI/HICP basket - like cherries).

Fish is not considered seasonal. Very small weight.

Clothing, footwear are seasonal - see 18.5.5.3.

Package holiday sub-index as a whole is not considered to be seasonal (some EAs are basically available all the year-round; however some EAs are considered seasonal).

Not applicable.

None.