Fiji
Official title
Consumer Price Index.
Scope
The index is computed monthly and covers urban households
in six major urban centres.
Official base
1985 = 100.
Source of weights
The weights and selected items were derived from a household expenditure
survey conducted from February 1983 to January 1984 among a sample of
4,402 households throughout Fiji. The index relates to 1,756 households
in urban areas. The weights were not adjusted to take into account
price changes between the survey period and the base period.
Weights and composition
| Major groups
| Number of items
| Weights
| Approximate number of price quotations
|
| Food
| 103 | 339.3 | ...
|
| Alcoholic drinks and tobacco
| 7 | 63.8 | ...
|
| Housing (rent, rates, maintenance)
| 9 | 186.1 | ...
|
| Fuel and light
| 3 | 48.7 | ...
|
| Durable household goods
| 36 | 75.6 | ...
|
| Clothing and footwear
| 32 | 63.1 | ...
|
| Transport
| 15 | 112.9 | ...
|
| Services
| 27 | 67.5 | ...
|
| Miscellaneous
| 35 | 43.0 | ...
|
| Total
| 267 | 1000.0 | ...
|
Household consumption expenditure
Excluded are income tax, payments to savings and pension funds,
life insurance premiums, house purchases and mortgage interest,
subscriptions to unions and betting payments.
Method of data collection
Prices are collected during the middle of two weeks each month
from centres in the Suva-Nausori area and the middle week
from centres in Ba, Lautoka, Nadi and Labasa. As an index based
on the consumption pattern of all urban dwellers in Fiji, the
Suva-Nausori area was considered as representative of the central
division, Labasa the northern division and centres in Ba, Lautoka
and Nadi were considered as representative of in the
western division. Together, the price changes in these areas are
considered to be representative of price movements in all urban
areas of Fiji.
The prices collected for the construction of the index are
those charged in cash transactions. Price collectors visit each
centre and personally note down the prices and weights of items
displayed in shops. If the prices are not marked or
are illegible, the investigators are instructed to obtain the
information directly from supermarket managers or supervisors. Prices
for items on sale
or specials
are recorded when they apply
to items in the index, except when the price has been reduced due to
damage, shop-soiled
or any form of deterioration of the goods.
The latter conditions reflect changes in the quality of the merchandise
and price variations caused by quality differences are excluded from the
index.
Officers of the Bureau of Statistics visit each pricing
establishment personally to record prices. Information
relating to services levied by the Government and statutory
bodies is obtained directly from them. Consistency in
terms of visiting the establishment at the same time and day of the
week is maintained each month, especially at regional municipal
markets where there can be considerable variation in prices
between the morning and afternoon.
Food items for pricing purposes are divided into municipal
market items and supermarket items. Items in supermarkets
are well defined in terms of size and brand, and prices are read
from the tags. With the market items, however, officers carry
their own scales to weigh and price items. Generally, three
samples of an item are taken from various vendors in the market
and the average price per unit weight is later computed in the
office. Markets are visited on Fridays and Saturdays at
specified times in the morning. This is to capture normal
prices for perishable goods which often deteriorate towards the
end of the day when they are generally sold at reduced
prices.
Cooking gas, white benzine and kerosene are priced from the usual
pricing establishments while electricity tariffs obtained directly from
the Fiji Electricity Authority are regularly checked for possible
changes.
Housing
The major component of this section is rent. The sample was chosen
from among the rented households identified during the expenditure
survey, and represents a random selection of all
rented households of Fiji citizens during 1983. Data on
rental charges are collected each month along with
prices for other items. On each visit, households are asked for
any improvements to the houses and any increase in rent due
to such improvements is adjusted for a change in quality. Other
items, such as those used for maintenance, are priced in the usual
way during visits to hardware shops each month.
Specification of varieties
An important consideration after the items were selected for the
index was to decide which particular brand and what size of
packet or how many brands of a particular item were to be priced.
If a particular item was overwhelmingly popular, it was easy to choose
this from the expenditure and brand details recorded by
households in their weekly diaries issued during the survey.
However, in many cases, additional details had to be obtained
from distributors and supermarket supervisors before final
selections could be made. The criteria for an item to be
included in the index were that it should be representative, a
good-seller
and generally available in major centres
throughout Fiji.
Substitution, quality change, etc.
If an item is not available, the price is generally repeated for
three months and then a substitute is selected. Care is taken to
ensure that the substitute item is as similar to the original
as possible in terms of quality, size and end-use.
Seasonal items
For seasonal items such as fresh fruit and vegetables, the
closing prices for the previous season are maintained during the
off season.
Computation
The national urban index (for the main island of Fiji) is
computed as the weighted average of three separate indices
relating to the central, western and northern regions. Each
regional index is computed according to the Laspeyres formula as
a weighted arithmetic average with fixed base, using weights
corresponding to February 1983-January 1984. The
three regional indices are then combined into a national urban
index according to the proportionate expenditure for each region,
the proportions being 0.571 for the central region,
0.361 for the western region and 0.068 for the
northern region.
To compute the item indices, the weighted average of price
relatives between the current and base periods is used.
Other information
Major sub-group indices are published.
Organisation and publication
Bureau of Statistics, Suva: Statistical News
.
Idem: Methodological Report on the Consumer Price Index
(Base average 12 months 1985 = 100)
, Parliamentary Paper No. 22
of 1988.