CURRENT TOPIC
2 BIATEC, Volume X, 4/2002
During 2001, the Statistical Office of the SR (SO SR) carried
out the planned revision of the consumer basket. This means,
that from 2002 inflation will be calculated on the basis of
a new consumer basket. The last revision of the consumer basket
was done 5 years ago. The structure of household expenditure,
which provides a basis for the weight system of the consumer
basket, has changed since then. In principle, we can say
that the fewer consumer basket revisions, the greater the probability
the value of inflation will be distorted. The gradual
change in the purchasing habits of households and the entry of
new products on the market are the reasons behind this, which
may distort the reported increase in the cost of living. Distortions
occur as a result of the fixed structure of items and their
constant weights in the consumer basket. With regard to the
fact that households tend to consume cheaper goods and services,
the value of inflation may be overestimated. According to
foreign studies, marked distortions in the value of inflation
may occur as early as three years after the last revision of the
consumer basket and the difference in inflation is estimated at
around 1 percentage point. The purpose of the revision was to
eliminate distortions in the recoded value of inflation and to
adopt the regulatory directives of EUROSTAT with the aim at
harmonising the calculation of price indices.
Effect of consumer basket revision on inflation
The effect of consumer basket revision on inflation may be
analysed in periods, when the prices of all items are available
from both the old basket and the new one. The year 2001 is
such a period. According to data from the Statistical Office of
the SR, the adoption of the new consumer basket had no significant
effect on the level of overall or core inflation reported.
The rate of inflation, calculated on the basis of the new consumer
basket, would have reached 6.4% in 2001, representing
a difference of 0.1 of a percentage point in comparison with
the actually recorded rate of inflation based on the old basket
(6.5%). The same difference was recorded in the case of core
inflation, which would have reached 3.1% in 2001 according
to the new consumer basket (the actual figure was 3.2%). The
comparison of values at the level of inflation or core inflation
shows no distortions in the Consumer Price Index (CPI), even
five years after the last revision. However, a more detailed
analysis of the new consumer basket or changes in its structure
may indicate significant differences, which could have impact
on the level of inflation in the future. In addition, the
composition of the consumer basket may reveal the current
trend in the standard of living.
Characteristics of the new consumer basket
Until 2001, the Consumer Price Index (CPI) had been based
on a consumer basket of prices from December 1995. The
selection and weight of individual items in the basket were
based on the structure of household expenditure in 1995.
From 2002, the new consumer basket will be used with items
and weights based on the statistics of family accounts from
2000 and data from other sources (statistics on trade, tourism,
national accounts, administrative sources, and the corporate
sector). The December 200 will be the base period for the calculation
of the CPI.
In quantitative terms, the number of items in the new consumer
basket was reduced from 710 in the old basket into
703. Of this figure, 663 items were virtually identical in the
both baskets, representing a share of roughly 97%. The new
items include the services of mobile phone networks (including
mobile phones), new and second-hand motor vehicles,
and new pharmaceuticals. From 1995 to 2001, however, the
Statistical Office made several qualitative changes in the old
basket. Some items were excluded from categories clothing,
restaurants, and hotel services. Since the new items represent
only about 3% of the new consumer basket, qualitative changes
are not expected to affect the Consumer Price Index to
significant extent.
According to COICOP classification (Classification of Individual
Consumption by Purpose), the Statistical Office of the
SR primarily divided the consumer basket into 11 divisions. In
connection with the gradual harmonisation of the method of determining
the CPI with that of EUROSTAT, the new consumer
basket is divided into 12 divisions. Compared with the original
classification (COICOP) the price index is monitored separately
in the division `postal and telecommunications services'.
Change in the structure of household
expenditure
Changes in the weight structure of individual categories are
affected by different developments in prices of individual
items, and changes in household consumption in terms of
both quality and quantity of goods purchased during the years
1995 ­ 2000. In calculating the actual rate of inflation,
the weight of items whose prices increased at a faster rate
than the general price level, will be higher in the current period.
This means that changes in the prices of relatively more
expensive goods (services) will have a higher weight than in
the base period. On the contrary, the weight of a relatively
REVISION OF THE CONSUMER BASKET OF SO SR
Mgr. Miroslav Gavura, National Bank of Slovakia
CURRENT TOPIC
BIATEC, Volume X, 4/2002 3
cheaper commodity (service) will fall. The current weight of
a consumer basket items is determined by the base (constant)
weight of given items and its price increase. Such weight is
referred to as recalculated (normed) weight and represents the
fictitious relative share of expenses for a certain commodity
or service in the current period.
The effect of varying price developments is apparent from
comparison of the old consumer basket with constant 1995
weights and weights recalculated to December 2000. On the
other hand, the change in household consumption outside the
price effect can be assessed from a comparison of the recalculated
weights in the old basket (for December 2000) and
the weights in the new basket based on the structure of household
consumption in 2000.
Comparisons at the level of COICOP categories indicate
that the sharpest fall, from 29.2% in 1995 to 23.6% in 2000,
took place in the weight of foodstuffs (including non-alcoholic
beverages). The fall in the share of foodstuffs in total household
expenditure (more than 5 percentage points) was connected
with the lower dynamics of prices during the period
1995 ­ 2000, and the shift of a part of consumption to other
goods and services. The structure of the new consumer basket
also indicates that households in the SR spent in 2000 less
money on clothing and footwear than in 1995. There was also
fall in expenditure on recreation and culture, and spending
on transport fell slightly as well. Significant increase was recorded,
in line with expectations, in the share of expenses related
to dwelling, due primarily to acceleration in the process
of deregulations in 1999 and 2000. A negative trend in society
is the increase in the weight of alcoholic beverages and tobacco.
In 2000, an average household in Slovakia spent approximately
7% of its net expenditures on alcoholic beverages
and tobacco goods (compared with 4.4% in 1995). The share
of household expenditures on hotels, cafes, restaurants, and
miscellaneous goods and services also increased. In other categories
only moderate changes were recorded.
The comparison between
the old consumer basket
with constant 1995
weights and weights recalculated
to December
2000 indicates that the
largest shifts of weights
within COICOP divisions
were caused by differences
in price development.
On the other hand,
comparison of the recalculated
weights of the
old consumer basket (to
December 2000) and the
weights of the new basket
based on the structure
of household consumption
in 2000 indicates that
the years 1995 ­ 2000
saw no significant change in the structure of household consumption
ascribable to a quantitative or qualitative change in
the purchase of individual consumer-basket items. As a result
of such a change, however, there was a shift in weighting
from the categories `clothing and footwear' and `recreation
and culture' to the categories `alcoholic beverages and tobacco';
`hotels, cafés, and restaurants'; and `miscellaneous goods
and services'.
Comparison of the consumer basket
with those applied in neighbouring countries
The structure of the consumer basket may also indicate
a country's living standard. A typical feature of poorer countries
is the high share of household expenditure on foodstuffs
and dwelling. On the other hand, advanced countries are characterised
by a higher share of expenditure on recreation, culture,
luxury goods and other market services. The share of expenditure
on necessities (food and dwelling) in the net
expenditures of an average household in the SR represented
almost 50% in 2000, which places Slovakia among the poorer
countries. The weight of foodstuffs in the consumer basket
(despite a significant fall in comparison with 1995) was
still much higher than in developed countries (cca 16% in the
EU). As a result of deregulations in 1999 and 2000, the share
of dwelling costs in the SR approached the EU average1.
This means that with continuation of deregulations in 2001
and the following years, may be come the share of dwelling
costs in the SR even higher. Due to lower standard of living
can Slovak citizens afford to spend a smaller share of their inComparison
of consumer baskets by structure according to COICOP classification
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
in %
Foodstuffs
and non-
-alcoholic
beverages
Alcoholic
beverages
and
narcotics
Clothing
and
footwear
Constant weight 1995
Recalculated weight 2000
New weight 2000
Dwelling,
water,
electricity,
gas,
and other
fuels
Furniture,
household
equipment,
and standard
maintenance
of
flats
Health
care
Transport Postal
and tele-
commu-
nications
services
Recreation
and
culture
Education Hotels,
cafes,
and res-
taurants
Miscella-
neous
goods and
services
­­­­­­­­­­­­­­­
1 In contrast with the Harmonised Consumer Price Index used by
EUROSTAT, the consumer basket of the Statistical Office of the SR includes
the item `imputed rent' (hypothetical rent for family houses and flats in private
ownership, representing a weight of 6.3% in the new basket). Excluding
this item out of account, the share of dwelling costs in Slovakia is comparable
with that in the European Union.
comes on recreation and culture than citizens in developed
countries. Significant improvement, however, cannot be expected
in the near future. In connection with the ongoing reform,
the share of expenditure on health care is expected to
increase gradually. The same applies to expenditures on transport,
the lower weight of which is connected with the use of
regulated prices in the area of passenger transport.
Comparison with the neighbouring transition economies
seems to be in favour of the Czech Republic and Hungary.
The structure of household consumption in these countries is
very similar and close to consumption behaviour in developed
countries. The biggest differences in comparison with the
EU can be seen in the composition of household expenditure
in Poland.
Structure
of the consumer
basket by sector
A different aspect of the
change in the structure of
the consumer basket results
from its division into
tradeable and non-tradeable
sectors, i.e.
foodstuffs, tradeable goods,
regulated prices, and
market services. In contrast
with COICOP classification,
price developments
in the individual
sectors according to this
classification follow certain
rules and thus it is
easier to reveal the effects
of individual factors on the price increase. This classification
enables a more detailed view of the possible influence of new
consumer basket on the level of overall and core inflation.
The share of the tradeable sector in the new basket is approximately
61% (67% in the old basket with constant weights).
The decrease in the share of household expenditures on foodstuffs
was in favour of the weights of items with regulated prices
and market services. The tradeable goods (excluding foodstuffs)
maintained in 2000 roughly the same weight as in 1995.
The weight of regulated prices is in the new basket higher than
in 1995, but lower than the weight in the original basket converted
to December 2000. The fall of weigh took place predominantly
in the sector of expenses related to dwelling as expected.
Within this sector, the weight of regulated rent fell in
line with the expectations (due to the sale of flats to private owners),
together with the weight of heat and electrical energy.
CURRENT TOPIC
4 BIATEC, Volume X, 4/2002
Comparison of consumer baskets in selected countries by structure according to COICOP classi-
fication
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
v %
Slovakia
Hungary
European Union
Czech Rep.
Poland
Foodstuffs Other tradeable goods Regulated prices Market services
Constant 1995 weights (old basket) 26.8 40.2 17.8 15.2
Recalculated weights (for Dec. 2000) 23.3 36.6 24.3 15.8
Constant 2000 weights (new basket) 21.4 39.5 21.1 18.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
Structure of the consumer basket in Slovakia (in %)
Foodstuffs
and non-
-alcoholic
beverages
Alcoholic
beverages
and
narcotics
Clothing
and
footwear
Dwelling,
water,
electricity,
gas,
and other
fuels
Furniture,
household
equipment,
and standard
maintenance
of
flats
Health
care
Transport Postal
and tele-
commu-
nications
services
Recreation
and
culture
Education Hotels,
cafes,
and res-
taurants
Miscella-
neous
goods and
services
This means in practice, that the continual deregulation process
should, in the case of the new consumer basket, represent
a smaller contribution to overall inflation than in the case of the
weight system of the basket used prior to 2001.
The increase in the share of household expenditures on other
tradeable goods (the increased weight of this sector in the
new basket compared with the weight recalculated for December
2000) and market services indicate that in the period
from 1995 to 2000 has the living standard of the population
slightly increased. This development is rather surprising, since
the years 1999 and 2000 saw relatively steep increases in
regulated prices, which led to high inflation and a marked fall
in real incomes. In such an environment, it could be expected
that the incomes of households would be used to cover the
costs of necessities (foodstuffs and dwelling costs) and that
the share of expenses on market services and durables would
decrease. However, the changes in the structure of the consumer
basket were probably dominated by the positive development
in real wages in period from 1996 to 1998 (even in
1999 and 2000 were real wages higher than in 1995).
Effect of the revised consumer basket
on the level of inflation
Comparison of the old consumer basket and new one in
terms of structure in a breakdown by COICOP divisions indicates
that from 1995 to 2000 remained the composition of
household expenditure almost unchanged. This indicates that
the adoption of the new consumer basket will not necessarily
represent an unexpected effect on the level of inflation. The
comparison of overall and core inflation in 2001, calculated
on the basis of both baskets, indicates that the revision of the
consumer basket had no marked effect on the level of inflation
in 2001, at least not at the higher level of aggregation. In
the individual months of 2001, however, the difference in inflation
fluctuated up to 0.5 of a percentage point.
Core inflation calculated on the basis of the two consumer
baskets followed roughly the same course of development until
April. From May to September, however, the index of core
inflation based on the new basket exceeded
the original value of the index. In the last months
of the year, the two indices converged again. As
a result of the lower weight of regulated prices in
the new basket (compared with the recalculated
weight of the old basket), the increase in CPI during
the first half of 2001 was slower than in the
case of the original basket. This difference was,
however, reduced by the faster rise in core inflation
in July and August.
Since the majority of items in the two baskets
are identical and there were no significant changes
in weights at the level of COICOP divisions,
the different developments in price indices in some
months were caused by a change in the weight
structure at lower levels of aggregation. It
may be that the weight of a sector in the new basket
remains unchanged in comparison with the recalculated
weight in the old basket, but the price increase in the sector
will differ considerably. Assuming that the items of the given
sector did not undergo a qualitative change, the difference in
price increase would be caused by a weight shift within the
sector in favour of an item with different price dynamics. An
example is the course of prices in the category `miscellaneous
services' within the scope of regulated prices. Based on the
old basket, prices in this category rose by 20% in 2001, and
on the new basket by as much as 35%. The difference is caused
by a fourfold increase in weight within the category in favour
of the item `compulsory insurance of vehicles', the price
of which increased by almost 60% in 2001.
A more detailed analysis of the structure of the new consumer
basket revealed the following facts for the estimation of
inflation in the future:
* in the new consumer basket, the weight of the tradeable
sector (excluding foodstuffs) increased in comparison with the
converted weight of the old basket (for December 2000), to the
detriment of the non-tradeable sector. With slower price dynamics
in the tradeable sector, this could result in lower inflation;
* the inclusion of new items in the consumer basket has
a dampening effect on the development of CPI. This is confirmed,
for example, by a fall in the price of second-hand motor
vehicles or stagnation in charges for the services of mobile
phone networks in 2001;
* unprocessed foodstuffs (meat, fruit, vegetables) retained
a high share in household expenditure in the new consumer
basket. This means that the seasonal volatility of food prices
will continue to cause a certain degree of uncertainty in forecasts
of consumer prices;
* the weight of fuels in the new consumer basket increased
by 1 percentage point. It may be assumed that the irregular
development of fuel prices will according to the new consumer
basket increase volatility to a greater extent than had been
seen prior to 2001;
* the weight of some regulated prices in the new basket fell,
which could result in lower inflation than in the case of the
old basket if the process of deregulation continues. This meCURRENT
TOPIC
BIATEC, Volume X, 4/2002 5
Conversion of consumer prices in 2001 (Dec. 2000 = 100)
107.00
106.00
105.00
104.00
103.00
102.00
101.00
100.00
99.00
XII/00 I/01 II/01 III/01 IV/01 V/01 VI/01 VII/01 VIII/01 IX/01 X/01 XI/01 XII/01
CPI (old basket) CPI (new basket) Core inflation
(old basket)
Core inflation
(new basket)
ans that the 7% increase in the price of heating contributed
0.3 of a percentage point to overall inflation in January 2002,
which would be 0.6 of a percentage point with the old basket.
A similar effect can be expected in the case of electricity prices
and regulated rents. The opposite effect will occur in the
case of a rise in the price of natural gas, whose weight in the
new basket increased somewhat;
* an upward effect on inflation will be exerted in ensuing
years by the harmonisation of excise duties on cigarettes and
tobacco with EU directives. The weight of tobacco goods in
the new basket increased by almost 50%. This means in practice
that while the increase in excise duties in 2002 represents
0.4 of a percentage point in the new consumer basket, the same
increase would have caused a 0.15 percentage point increase
in consumer prices in the old basket.
Although revision of the consumer basket does not involve
an unexpected impact on the level of overall and core inflation
in 2001, the above examples indicate that it has a more
significant effect on the development of price indices at lower
levels of aggregation. Since the effect acts in the opposite direction
in many cases, these effects offset each other to a considerable
extent at the level of overall and/or core inflation.
CURRENT TOPIC
6 BIATEC, Volume X, 4/2002
Structure of the consumer basket Old basket New basket
Recalculated Year-on-year Weight in Year-on-year
weight, to Dec. 20002 inflation Dec. 2000 inflation
Total in % 100.0 6.5 100.0 6.4
Regulated prices in % 24.3 17.2 21.1 18.8
Share of total, in % points 0.0 4.17 3.98
Effect of changes in indirect taxes
on non-regulated prices ­0.03 ­0.07
Core inflation in % 74.0 3.2 78.9 3.1
Share of total, in % points 2.39 2.46
of which: Food prices in % 22.5 3.7 21.4 4.0
Share of total, in % points 0.83 0.86
Foodstuffs ­ processed in % 1 12.6 4.8 12.0 4.6
Share of total, in % points 1 0.60 0.55
Foodstuffs ­ non-processed in % 1 9.9 2.3 9.4 3.2
Share of total, in % points 1 0.23 0.30
Tradeable goods in % 1 35.6 1.0 39.5 0.6
Share of total, in % points 1 0.35 0.23
Tradeable goods excl. fuels in % 1 32.9 2.3 35.3 2.2
Share of total, in % points 1 0.75 0.77
Fuels in % 1 2.6 ­15.6 4.2 ­12.9
Share of total, in % points 1 ­0.41 ­0.54
Market services in % 1 15.8 7.7 18.0 7.6
Share of total, in % points 1 1.22 1.37
Source: Statistical Office of the SR
1 Calculated by the NBS on the basis of data from the Statistical Office of the SR.
2 The sum of the weights is not equal to 100 because of adjustment for changes in indirect taxes.
Processed foodstuffs ­ prices in the categories: 1. bread and cereals; 2. milk, cheese, eggs; 3. oil and grease; 4. sugar, jam, honey, syrup;
5. other.
Unprocessed foodstuffs ­ prices in the categories: 1. meat; 2. fish; 3. fruit; 4. vegetables.
Net inflation ­ includes price increases in the sectors of tradeable goods (excluding foodstuffs) and market services.
Net inflation (excluding fuel prices) ­ net inflation adjusted for the effect of fuel prices.
Core inflation ­ methodology of EUROSTAT ­ core inflation excluding the prices of unprocessed foodstuffs, fuels, and lubricants.
Table 1: Consumer prices in December 2001 based on the old and new baskets
Core inflation (excluding fuels) in % 1 71.3 3.9 74.7 4.0
Share of total, in % points 1 2.80 3.00
Net inflation (excluding the effect of changes in
indirec taxes) in % 1 51.5 3.0 57.5 2.8
Share of total, in % points 1 1.56 1.60
Net inflation (excluding fuels and the effect
of changes in indirect taxes) in % 1 48.8 4.0 53.3 4.0
Share of total, in % points 1 1.97 2.14
Core inflation in % ­ methodology of EUROSTAT 1 61.4 4.2 65.3 4.1
Share of total, in % points 1 2.57 2.69
CURRENT TOPIC
BIATEC, Volume X, 4/2002 7
Const. 1995 Recalculated Const. 2000 Difference Increase in Increase in Difference
weights weights, weights in weight 2001 prices 2001 prices in price
(old basket) Dec. 2000 (new basket) (old basket) (new basket) increase
TOTAL 100,0 100.0 100.0 0.0 106.5 106.4 ­0.16
1 Tradeable goods 66.9 59.9 60.9 1.0 101.9 101.7 ­0.24
11 Foodstuffs 26.8 23.3 21.4 -1.9 103.6 104.0 0.45
Bread and cereals 4.1 3.9 4.0 0.1 109.0 108.4 ­0.58
Meat 8.1 6.8 6.0 ­0.8 100.7 101.2 0.57
Fish 0.5 0.5 0.7 0.2 108.3 110.7 2.37
Milk, cheese, eggs 4.9 4.7 4.1 ­0.7 102.2 102.6 0.38
Oil and grease 1.8 1.4 1.3 ­0.1 100.0 99.7 ­0.35
Fruit 2.1 1.7 1.4 ­0.2 110.4 112.3 1.95
Vegetables incl. potatoes 1.7 1.3 1.3 0.0 97.5 98.7 1.21
Sugar, jam, honey, syrup.
chocolate, sweets 2.5 2.1 1.8 ­0.2 106.4 105.7 ­0.76
Other 1.1 0.9 0.8 ­0.1 101.8 101.5 ­0.25
12 Other tradeables 40.2 36.6 39.5 2.9 100.9 100.4 ­0.47
Non-alcoholic beverages 2.4 1.8 2.2 0.4 99.7 99.4 ­0.30
Alcoholic beverages 2.6 2.1 4.0 2.0 102.1 102.5 0.42
Tobacco 1.9 2.1 2.9 0.8 103.4 106.2 2.81
Clothing 8.1 7.2 5.3 ­1.9 102.2 102.1 ­0.09
Footwear 2.5 2.5 2.0 ­0.5 105.6 106.0 0.38
Dwelling 0.7 0.8 2.0 1.2 100.3 104.5 4.28
Furniture 5.7 5.0 5.0 0.1 100.4 99.6 ­0.80
Transport 6.8 6.4 6.2 ­0.2 94.6 90.9 ­3.66
Recreation and culture 6.3 6.0 5.4 ­0.6 102.5 100.1 ­2.45
Miscellaneous 3.1 2.8 4.3 1.5 102.8 103.2 0.43
2 Non-tradeable goods 33.1 40.1 39.1 ­1.0 113.4 113.7 0.26
21 Regulated prices 17.8 24.3 21.1 ­3.1 117.2 118.8 1.64
Dwelling 9.9 16.0 11.6 ­4.4 120.0 117.7 ­2.27
Health care 1.3 1.4 1.5 0.1 101.8 103.9 2.13
Transport 2.6 2.8 2.3 ­0.5 114.1 121.0 6.94
Recreation and culture 0.4 0.4 0.5 0.1 100.0 100.0 0.00
Education 0.7 0.5 0.3 ­0.1 100.7 99.7 ­0.98
Hotels, cafes, restaurants 1.1 1.1 1.6 0.6 111.1 113.3 2.24
Miscellaneous 1.8 2.2 3.3 1.1 120.1 135.2 15.09
22 Market services 15.2 15.8 18.0 2.2 107.6 107.6 ­0.02
Clothing 0.3 0.3 0.1 ­0.2 106.3 105.7 ­0.57
Footwear 0.1 0.1 0.0 ­0.1 107.0 107.0 0.02
Dwelling 5.9 6.4 7.9 1.5 108.8 108.5 ­0.29
Furniture 0.2 0.2 0.1 ­0.1 104.1 103.8 ­0.21
Transport 0.6 0.8 0.7 0.0 105.1 103.5 ­1.60
Recreation and culture 1.8 2.1 1.3 ­0.8 110.4 110.1 ­0.21
Education 0.6 0.6 0.2 ­0.4 107.5 107.6 0.19
Hotels, cafes, restaurants 3.9 3.6 5.6 2.0 106.3 107.8 1.51
Miscellaneous 1.9 1.7 1.9 0.3 104.6 103.5 ­1.09
Inflation rate. adjusted for
regulated prices (11+12+22) 82.2 75.7 78.9 3.1 103.1 103.0 ­0.09
Regulated prices:
Electricity 2.0 4.1 3.6 ­0.5 116.2 116.2 ­0.04
Heating 3.5 6.2 4.1 ­2.2 120.0 120.0 0.03
Natural gas 1.5 1.9 2.3 0.4 117.9 111.7 ­6.18
Railway traffic 0.3 0.4 0.3 0.0 112.8 109.1 ­3.69
Bus transport 1.2 1.2 1.3 0.1 120.2 130.8 10.56
Water + sewage disposal 0.7 1.1 0.7 ­0.4 120.3 120.3 0.00
Postal services 0.1 0.1 0.2 0.1 108.9 109.0 0.07
Telecommunications 1.2 1.5 2.2 0.7 124.3 136.1 11.85
Rents 1.3 1.4 0.5 ­1.0 138.4 139.4 1.01
Regulated prices in total 11.8 17.9 15.2 ­2.7 120.6 121.9 1.36
Table 2: Structure of the consumer baskets of 1995 and 2002
in % (December 2000 =100)­ not adjusted for changes in indirect taxes