Bulgaria
Official title
Consumer Price Index
Scope
The index is computed and published monthly and covers primarily
low- and middle-income households in Bulgaria.
Official base
Base period for computation: December of previous year = 100.
Base period for publication: December of previous year = 100,
previous month = 100.
Sources of weights
The weights of the index are primarily derived from the Household
Budget Survey data aggregated to the national level. 1998 CPI is
computed with weights derived from the 1997 HBS. During the
period May 1990-December 1992, CPI was calculated with the
weights for elementary aggregates derived from the retail trade
and for the main group, from HBS. 1993-95 CPI was calculated
with weights fully derived from 1992 HBS. Index reference period
was December 1992. Because of the considerable changes in the
consumption pattern, since 1996, the weights has been updated
annually. At the beginning of 1999 NSI shall introduce CPI
weights which will be updated for price change to align the
weights and base price period.
HBS covers approximately 3,000 primarily low-and middle-income
households distributed throughout the country, which stay in a
sample for one month. Due to the reduction of the sample size
from 6,000 to 3,000, survey data are no more reliable for deriving
the city level weights.
Weights and composition
| Major groups
| Number of items
| Weights
| Approx. No.
of price quotation
|
| Food
| 306
| 55.777
| 42,840
|
| Alcoholic beverages
| 17
| 1.258
| 1,904
|
| Tobacco products
| 4
| 2.416
| 224
|
| Housing
| 38
| 2.869
| 2,128
|
| Power for home use
| 10
| 10.194
| 560
|
| House furnishings
| 109
| 3.678
| 6,104
|
| Clothing, footwear, personal belongings
| 172
| 8.655
| 9,632
|
| Personal hygiene and health care
| 99
| 4.690
| 5,544
|
| Education and recreation
| 72
| 2.339
| 4,032
|
| Transport and communications
| 66
| 7.883
| 3,696
|
| Others
| 6
| 0.242
| 336
|
| Total
| 899
| 100.000
| 77,000
|
Classification is adapted to the SNA Classification of Household
Final Consumption expenditure.
Household consumption expenditure
Household consumption expenditure refers to all financial
spending for the purchase of consumption goods and services used
for direct satisfaction of individual needs by the index
population. The criterion for determining the individual
consumption expenditures is in the methodological compliance with
the principals of SNA.
Sample selection
Price data are collected in 28 towns selected according to their
geographical location, number of inhabitants and retail trade
turnover. These towns represent 44% of total population and 65%
of total retail turnover. In 1996, an experimental observation
was made in a sample of small towns in the country. The results
obtained showed similar price movement as in the observed big
towns. That means that the current sample which includes only
big towns is sufficiently representative.
The number and the structure of outlets, where the prices are
observed, are determined by purposive sampling. The sample of
outlets includes the biggest and the most popular shops, markets,
service establishments, etc. from which the index population
makes its purchases, with special attention to different forms of
trade.
The items are selected according to their significance in
personal consumption and their representativness for items not
included in the list.
Method of data collection
Prices are collected monthly, between 5th and 25th day of the
month, by personal visits to approximately 5,500 outlets in 28
urban settlements. Prices of goods and services are collected
regularly in fixed periods. The number of price quotations for
each item depends on the commodity group to which that item
belongs. At least 6 price quotations are obtained for the main
food items in each of the 28 cities. 4 prices are obtained for
other food items and 2 prices for the non-food items.
Discounts, sale prices, free-market prices for items which also
have official prices, black market prices, hire-purchases and
second-hand prices, as well import prices are not taken into
consideration in price collection and calculation of national
CPI.
Housing
The rent index includes non-market rents, fixed by the
government, and rents on a free contract basis. Price collection
covers rents for 4 different types of dwellings. Price
collectors select 2 dwellings for each type of dwelling and visit
them two times a month. Owner occupied housing is not included
in the index.
Specification of varieties
Two kinds of specifications are used: Tight specifications are
used for those items whose prices are set centrally by the
government. In such a case there is no difference between
representative item and particular variety. The specifications
are very detailed.
In the case of loose specifications the price collector has to
choose the particular variety to be priced. As the supply of
goods or services is not stable at the moment these
specifications are not too strict. There is intention to proceed
to more detailed description of the representative goods and
services in the future.
Substitution, quality changes
Quality adjustment is not applied.
The list of items is updated annually, taking into account
appearance of new products. National Statistical Institute (NSI)
analyses the structure of household expenditure for newly
significant goods (accounting for more than one part in a thousand
of the total household expenditure) and takes decision about
their inclusion in the index. Another source is price collectors
who annually inform the NSI about the appearance of new goods and
services on the market.
Up to 1997, when a given type or quality disappeared from the
market, the last observed price was carried forward. Since then
no imputation has been made. One or two additional outlets have
been identified in each city and, in the case of a missing item,
the price collector obtains the price from this additional
outlet. If this fails, the city average is computed using
available price observations. If the price is missing in
subsequent months, a replacement is selected. The price
collector is instructed to select the variety that is most
comparable with the previous one.
Seasonal items
Since 1997 price indices of seasonal fresh fruit and vegetable
during the off-season are computed by the method of imputed
prices. When the supply of a seasonal item is limited or missing
an imputed price is computed by using the price movements of the
available items. This method is applied keeping in mind the
following:
- comparison is done with December of the previous year
- seasonal items for which this method is applied are
determined in advance for each month
- particular item or consumption group whose index will be used
for imputation is carefully chosen.
This method is also applied to some non-food and service items.
In 1998 NSI will start to compute the index for seasonal
items on the basis of prices that are available for both
current and base period.
Computation
The index is computed according to the modified Laspeyres
formula, as a weighted arithmetic average with a fixed base,
using weights representing structure of household's money
expenditure for acquisition of respective groups of goods and
services. National average monthly prices for each item are
computed as a weighted arithmetic mean of prices for 28 towns,
with weights representing structure of the population in these
towns. Elementary aggregate indices are calculated as geometric
mean of price relatives of each item included in the elementary
aggregate.
The annual indexes are computed as a simple arithmetic mean of
monthly indexes for two subsequent years, based on one and the
same base.
Other information
None.
Organisation and publication
National Statistical Institute of Bulgaria
Statistical Yearbook of the Republic of Bulgaria
,
Prices, Consumer Price Indexes and Inflation
,
Statistical News
.
On Internet website: http://www.nsi.bg