Module H5 Session 3
Session 3. Consumer price indices
Learning objectives
At the end of this session the students will be familiar with

consumer price indices and their uses



key issues relating to coverage
the sources of data and methods of data collection
the internationally recommended classification of expenditure by purpose
The use of consumer price indices
Recapping from Session 1, write down at least three uses of a CPI
Issues of CPI coverage
There are some important and possibly controversial questions concerning the scope or
coverage of the CPI (or CPIs if there are several versions covering differing reference
populations), including:


should the CPI cover the rural as well as the urban population?
should the CPI cover the consumption of own-produce (non-monetary) as well as
purchased (monetary) expenditure?
SADC Course in Statistics
Module H5 Session 3 – Page 1
Module H5 Session 3
Geographical coverage
The international guidelines are quoted below.
Urban and rural
Geographical coverage may refer either to the geographical coverage of expenditures or the coverage of price
collection.
Ideally these two should coincide, whether the CPI is intended to be a national or a regional index. In most
countries, prices are collected in urban areas only since their movements [not necessarily their levels] are
considered to be representative of the price movements in rural areas. In these cases national weights are
applied and the resulting index can be considered a national CPI.
If price movements in urban and rural areas are felt to be sufficiently different – although price collection is
restricted to urban areas because of resource constraints – then urban weights should be applied and the
resulting index must be considered as purely an urban and not a national CPI.
For example, the following countries cover urban households only (expenditure weights and prices):
Australia, Mexico, Republic of Korea, Turkey, United States.
Most other developed countries tend to use weights covering urban and rural households, although in nearly
every case price collection takes place in urban areas only.
Of course, the borderline between urban and rural is inevitably arbitrary and may vary from country to
country. For example, in France urban price collection is interpreted to include villages with as few as 2,000
residents.
Decisions about geographical coverage in terms of urban versus rural coverage will depend on population
distribution and the extent to which expenditure patterns and the movements of prices tend to differ between
urban and rural areas.
Source: Consumer price index manual: Theory and practice (Chapter 3) ILO, Geneva, 2004
In most SADC countries, a majority of the population lives in rural areas, often with very
low levels of income and expenditure. The expenditure patterns of the rural population is
likely to be very different from those of the urban population. A very large proportion of
consumption in rural areas is food.
Do prices move differently in rural areas? In general, while the level of prices may differ,
available evidence suggests that their movements do not differ significantly. However
food prices in local areas may vary in the short term according to fluctuations in supply
(caused by drought for example). In some countries, prices in rural markets are monitored
for food security purposes, so these data (suitably weighted) could be used in the CPI.
SADC Course in Statistics
Module H5 Session 3 – Page 2
Module H5 Session 3
Consumption of own produce
Extract from he international guidelines on subsistence agriculture and on owner-occupied
housing are given below.
Subsistence agriculture
In principle, the prices of the outputs from own-account agricultural production may be included in CPIs,
even though they are imputed. On the other hand, for households relying on subsistence agriculture, the
prices of inputs of agricultural materials purchased on the market may be their main exposure to inflation.
Two points may be noted. First, the imputed market value of the output should usually be greater than the
costs of the purchased inputs, if only because it should cover the costs of the labour inputs provided by the
household. Thus, pricing the purchased inputs rather than the outputs may mean that the consumption of
own agricultural production in CPIs does not receive sufficient weight. Second, double counting should be
avoided. If the imputed prices of the outputs are included, the actual prices of the inputs consumed should
not be included as well.
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
Owner-occupied housing
In the case of owner-occupied housing, the situation is complicated by the fact that the production requires
the use of the capital services provided by a large fixed asset in the form of the dwelling itself. Even if the
inputs into the production of housing services are priced for CPI purposes, it is still necessary to impute
prices for the inputs of capital services (mainly depreciation plus interest) provided by the dwelling. Some
countries therefore prefer to impute the prices of the outputs of housing services actually consumed on the
basis of the rents payable for the same kind of dwellings rented on the market. The treatment of owneroccupied housing is complex, and somewhat controversial.
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
Discussion
For the purpose of monetary policy, what would be the best solution


To include or exclude rural expenditure patterns in the CPI weights?
To include or exclude the consumption of own produce?
SADC Course in Statistics
Module H5 Session 3 – Page 3
Module H5 Session 3
Two stages in CPI calculations
There are two main stages in the calculation of a CPI. The first is the collection of the price
data and the calculation of the elementary price indices. The second is the averaging of the
elementary price indices to arrive at price indices at higher levels of aggregation up to the
overall CPI itself. Expenditure data are needed for the elementary aggregates that can be
used as weights in the second stage.
Stage 1: Elementary aggregation
Elementary aggregate
An elementary aggregate is the smallest aggregate for which expenditure data are available and used for CPI
purposes. The values of the elementary aggregates are used to weight the price indices for elementary
aggregates to obtain higher-level indices. The range of goods and services covered by an elementary
aggregate should be relatively narrow, and may be further narrowed by confining the goods and services to
those sold in particular types of outlet or in particular locations.
Elementary price index
An elementary index is a price index for an elementary aggregate. Expenditure weights cannot usually be
assigned to the price relatives for the sampled products within an elementary aggregate, although other kinds
of weighting may be explicitly or implicitly introduced into the calculation of elementary indices. Three
examples of elementary index number formulae are the Carli, the Dutot and the Jevons.
Carli price index
An elementary price index defined as a simple, or unweighted, arithmetic average of the sample
price relatives.
Dutot price index
An elementary price index defined as the ratio of the unweighted arithmetic averages of the prices in the two
periods compared.
Jevons price index
An elementary price index defined as the unweighted geometric average of the sample price relatives.
Source: Consumer price index manual: Theory and practice (Glossary) ILO, Geneva, 2004,
The method recommended internationally is the Jevons method. Reasons for this include:



Unlike the Carli, the Jevons index is independent of the chosen price base period
Unlike the Dutot, the Jevons index is independent of the levels of price quotations
Prices are always positive and tend to follow a log-normal distribution
Question
Which method is used in your country?
SADC Course in Statistics
Module H5 Session 3 – Page 4
Module H5 Session 3
Stage 2: Higher levels of aggregation
Methods of aggregation used to combine the elementary price indices were discussed in
session 2.
Questions
1. Which method is used in your country?
2. What is the index reference period?
3. What is the weighting reference period?
Classification
The classification of household expenditures used in a CPI provides the necessary framework for the various
stages of CPI compilation. It provides a structure for purposes of weighting and aggregation. The goods and
services covered by a CPI may be classified in several ways: not simply on the basis of their physical
characteristics but also by the purposes they serve and the degree of similarity of their price behaviour.
Product-based and purpose-based classifications differ but can usually be successfully mapped onto each
other. In practice, most countries use a hybrid classification system in which the breakdown at the highest
level is by purpose while the lower-level breakdowns are by product type.
[The] internationally agreed Classification of Individual Consumption according to Purpose (COICOP)
provides a suitable classification for CPI purposes… It is desirable for the price movements of the individual
products within the elementary aggregates to be as homogeneous as possible.
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
A summary of the COICOP classification is given in the Annex to these notes
Discussion point
How suitable is COICOP for countries in the SADC region?
SADC Course in Statistics
Module H5 Session 3 – Page 5
Module H5 Session 3
Sources of data for CPI expenditure weights
Household expenditure surveys and national accounts
The principal data source for household consumption expenditures in most countries is a household
expenditure survey (HES). An HES is a sample survey of thousands of households that are asked [for details]
of their expenditures on different kinds of consumer goods and services [during] a specified period of time,
such as a week or longer. The size of the sample obviously depends on the resources available, but also on
the extent to which it is desired to break down the survey results by region or type of household. HESs are
costly operations.
This [session] is not concerned with the conduct of HESs or with general sampling survey techniques or
procedures. There are several standard texts on survey methods to which reference may be made. Household
expenditure surveys may be taken at specified intervals of time, such as every five years, or they may be taken
each year on a continuing basis.
HESs can impose heavy burdens on the respondents, [and they] tend to have some systematic biases. For
example, many households either deliberately, or unconsciously, understate the amounts of their expenditures
on certain ‘‘undesirable’’ products, such as gambling, alcoholic drink, tobacco or drugs. Corrections can be
made for such biases. Moreover, the data collected in HESs may also need to be adjusted to bring them into
line with the concept of expenditure required by the CPI. For example, the imputed expenditures on the
housing services produced and consumed by owner-occupiers may not be collected in HESs.
The use of the commodity flow method within the supply and use tables (SUTs) of the System of National
Accounts enables data drawn from different primary sources to be reconciled and balanced against each
other. The commodity flow method may be used to improve estimates of household consumption
expenditures derived from expenditure surveys by adjusting them to take account of the additional
information provided by statistics on the sales, production, imports and exports of consumer goods and
services. By drawing on various sources, the household expenditure data in the national accounts may provide
the best estimates of aggregate household expenditures, although the classifications used may not be fine
enough for CPI purposes… It is important to note, however, that national accounts should not be viewed as
if they were an alternative, independent data source to HESs. On the contrary, HESs [are a major] source for
the expenditure data on household consumption used to compile [benchmark estimates of GDP].
[Benchmarking GDP estimates will be covered in a later session.]
Household expenditure surveys in many countries may not be conducted as frequently as might be desired
for CPI, or national accounts, purposes…[As a result,] CPIs in many countries are Lowe indices that use the
quantities of some base period b that precedes the price reference period 0 [by a year or more] and period t
by many years.
Countries that conduct continuous expenditure surveys are able to revise and update their expenditure
weights each year so that the CPI becomes a chain index with annual linking. Even with continuous
expenditure surveys, however, there is a lag between the time at which the data are collected and the time at
which the results are processed and ready for use, so that it is never possible to have survey results that are
contemporaneous with the price changes. Thus, even when the weights are updated annually, they still refer
to some period that precedes the time reference period. For example, when the price reference period is
January 2007, the expenditure weights may refer to 2004 or 2005, or perhaps both years. (Some countries
prefer to use expenditure weights that are the average rates of expenditure over periods of two or three years
in order to reduce ‘‘noise’’ caused by errors of estimation (the expenditure surveys are only samples) or erratic
consumer behaviour over short periods of time.)
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
SADC Course in Statistics
Module H5 Session 3 – Page 6
Module H5 Session 3
Issues in price data collection
Methods of price collection
The following is an extract from the international guidelines:
This section is concerned with some of the operational aspects of price collection.
Local price collection
When prices are collected from local outlets, the individual products selected for pricing can be determined in
two ways. One way is for a specific list of individual products to be determined in advance by the central
office responsible for the CPI. Alternatively, the price collector can be given the discretion to choose from a
specified range of products. The collector may use some kind of random selection procedure, or select the
products that sell the most or are recommended by the shop owner or manager. An individual product
selected for pricing in an individual outlet may be described as a sampled product. It may be a good or a
service.
When the list of products is determined in advance by the central office, the objective is usually to select
products that are considered to be representative of the larger group of products within an elementary
aggregate. The central office also has to decide how loosely or tightly to describe, or specify, the
representative products selected for pricing. In theory, the number of different products that might be
identified is to some extent arbitrary, depending on the number of economic characteristics that are deemed
to be relevant or important. For example, ‘‘beef ’’ is a generic term for a group of similar but nevertheless
distinct products. There are many different cuts of beef, such as minced beef, stewing steak or rump steak,
each of which can be considered a different product and which can sell at very different prices…
Tightening the specifications ensures that the central office has more control over the items actually priced in
the outlets, but it also increases the chance that some products may not actually be available in some outlets.
Loosening the specifications means that more items may be priced but leaves the individual price collectors
with more discretion with regard to the items actually priced. [However, this may add significant and
undesirable variation into the index.]
Central price collection
Many important prices can be collected directly from the head office of the organization responsible for .
fixing the prices. When prices are the same throughout the country, collection from individual outlets is
unnecessary:
 Some tariffs or service charges are fixed nationally and apply throughout the country. This may be
the case for public utilities such as water and electricity, postal services and telephone charges, or
public transport. The prices or charges can be obtained from the relevant head offices
 Many of these prices determined centrally may change very infrequently, perhaps only once or twice
or year, so they do not have to be collected monthly. Moreover, many of these prices can be
collected by telephone, fax or email and may not require visits to the head offices concerned.
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
SADC Course in Statistics
Module H5 Session 3 – Page 7
Module H5 Session 3
Continuity of price collection
A CPI is intended to measure pure price changes. The products whose prices are collected and compared in
successive time periods should ideally be perfectly matched; that is, they should be identical in respect of their
physical and economic characteristics. When the products are perfectly matched, the observed price changes
are pure price changes. When selecting representative products, it is therefore necessary to ensure that
enough of them can be expected to remain on the market over a reasonably long period of time in exactly the
same form or condition as when first selected. Without continuity, there would not be enough price changes
to measure.
Having identified the items whose prices are to be collected, the normal strategy is to continue pricing exactly
those same items for as long as possible. Price collectors can do this if they are provided with very precise, or
tight, specifications of the items to be priced. Alternatively, they must keep very detailed records themselves
of the items that they have selected to price.
Source: Consumer price index manual: Theory and practice (Chapter 1) ILO, Geneva, 2004
Dealing with changes in the available goods or services will be discussed in a later session.
Discussion points
1. Which of the two methods of local data collection would you use?
2. Why is it so important to price identical products in each time period?
SADC Course in Statistics
Module H5 Session 3 – Page 8
Module H5 Session 3
Annex – COICOP
version 1.1
01 - Food and non-alcoholic beverages
01.1 - Food
01.2 - Non-alcoholic beverages
02 - Alcoholic beverages, tobacco and narcotics
02.1 - Alcoholic beverages
02.2 - Tobacco
02.3 - Narcotics
03 - Clothing and footwear
03.1 - Clothing
03.2 - Footwear
04 - Housing, water, electricity, gas and other fuels
04.1 - Actual rentals for housing
04.2 - Imputed rentals for housing
04.3 - Maintenance and repair of the dwelling
04.4 - Water supply and miscellaneous services relating to the dwelling
04.5 - Electricity, gas and other fuels
05 - Furnishings, household equipment and routine household maintenance
05.1 - Furniture and furnishings, carpets and other floor coverings
05.2 - Household textiles
05.3 - Household appliances
05.4 - Glassware, tableware and household utensils
05.5 - Tools and equipment for house and garden
05.6 - Goods and services for routine household maintenance
06 - Health
06.1 - Medical products, appliances and equipment
06.2 - Outpatient services
06.3 - Hospital services
07 - Transport
07.1 - Purchase of vehicles
07.2 - Operation of personal transport equipment
07.3 - Transport services
08 - Communication
08.1 - Postal services
08.2 - Telephone and telefax equipment
08.3 - Telephone and telefax services
09 - Recreation and culture
09.1 - Audio-visual, photographic and information processing equipment
09.2 - Other major durables for recreation and culture
09.3 - Other recreational items and equipment, gardens and pets
09.4 - Recreational and cultural services
09.5 - Newspapers, books and stationery
09.6 - Package holidays
10 - Education
10.1 - Pre-primary and primary education
10.2 - Secondary education
10.3 - Post-secondary non-tertiary education
10.4 - Tertiary education
10.5 - Education not definable by level
11 - Restaurants and hotels
11.1 - Catering services
11.2 - Accommodation services
12 - Miscellaneous goods and services
12.1 - Personal care
12.2 - Prostitution
12.3 - Personal effects n.e.c.
12.4 - Social protection
12.5 - Insurance
12.6 - Financial services n.e.c.
12.7 - Other services n.e.c.
SADC Course in Statistics
Module H5 Session 3 – Page 9