Transcript Document
ICP Price Collections
Regional Course on Price Statistics and ICP
Male, Maldives
25-29 September 2005
TIMOTHY LO
Statistician, International Comparison Program
Asian Development Bank
1
Operational Stage of the ICP
• The largest coordinated statistical survey undertaken
globally
• Reference period is 2005 covering all six regions for 150
countries
• Involves price survey of the regional product list
• ICP surveys are the primary responsibility of the NSOs
• Countries are encouraged to price all 651 products for
household consumption
2
Scope – defining the target price
Target or ‘ideal’ price
Arithmetic average price of all transactions for the
• whole country
• whole year
“Average annual national price” used to link to official
GDP values
3
Scope – defining the target price
Necessary to sample
• Products what products are purchased -> product
list
• Outlets where the purchases are made -> outlet
sample
• Transactions when the purchases are made -> time
sample
4
Scope of Surveys
• For regional comparability the scope of price surveys
must be the same in every country
• This differs from CPI scope which may be narrower
where CPI basket of goods will vary from country to
country
• ICP price survey is different from the CPI as each
country does not select its own individual basket of
items to price
5
ICP Requirements
Product selection fixed by the regional product
list
• countries do not select products for pricing
A sample of places and times is needed
• for “average annual national price” of items is
required
6
ICP Requirements
Scope of the survey must be the same in every
country
• needed for international comparability
Must use existing infrastructure as far as
possible
• too costly to duplicate CPI infrastructure
7
Sampling Errors
Average annual national price is an estimate.
Therefore subject to:
• sampling error
• non-sampling error
Product & transaction samples are purposive.
Therefore:
• not possible to calculate sampling error
8
Sampling Errors
Outlet sample may be selected based on
probability proportional to size (PPS)
• if it is, then it is possible to calculate sampling
error
Product PPPs are the result of two estimated
averages
• may calculate sampling error of the ratio
9
Probability Sampling
Random or probability sampling is preferred
• unbiased
• efficient because they have minimum variance
Prior information needed to:
• allow stratification
• selection using probability proportional to size
PPS measure (for ICP purposes)
• ideally want product sales volumes
• in practice may use ‘total sales’ or ‘staff numbers’
10
Purposive Sampling
Subjective selection by person or agency
• it is not common to use random sampling
• extensively used in CPI sampling for product &
transaction samples
• used for ICP product and transaction samples
Random or probability sampling is preferred,
however, Purposive sampling used when:
• information not available for random sampling
• cost of random sampling is too high
11
Product Sample
• Regional ICP product list provides the “frame”
• Country lists provide national sample:
includes “representative” & “available” items
excludes “not available” items
try to avoid changes after survey starts (costs/data
quality)
includes items not included in CPI
12
Outlet Sample
Asia regional circumstances vary enormously
Geographically
• City states/Single state countries/Multi-state
countries
Existing CPI infrastructure statistical systems
• single national systems/separate provincial
systems
• highly developed systems/simple limited-scope
systems
13
Outlet Sample
Must include both location & type
• Location: urban/rural provinces
• Type: street stall, small shop, department store
Outlet choice more important for ICP
• Price levels differ by outlet type: cause ICP bias
• CPI more robust: only needs price movements
Avoid price collectors taking easy options
• Pricing in department stores easier
• Risk bias if high priced goods easier to measure
14
Outlet Sample
• Outlet survey designs should reflect individual
country circumstances
• Remember CPI design may not be adequate
15
Informal Markets
Sampling frames may exclude informal outlets
• e.g. Street stalls & itinerant sellers
Weekly markets may be very important & account for
major part of HH purchases
Informal markets may require location sample
• Price collectors then select specific sellers
Bargaining: no fixed prices
• Special care needed by price collectors
• With such markets there is risk of serious bias
16
CPI Outlet Frame
Is a CPI outlet
frame available?
Yes
Is geographic
scope adequate?
Do CPI outlets
cover all ICP
products?
Is the frequency
appropriate?
Yes
No
Use existing CPI
frame
17
Supplement CPI
frame
Develop new
ICP frame
No
Multiple CPI Outlet Frames
• Multi-state countries may have many independent
CPIs (e.g. India, China)
• Design of a single national ICP sample will need input
from sampling experts
A purpose-built sample needed
Outlets should be selected on PPS basis
Stratified sample may be appropriate
Use expenditure weights for PPS, if available
18
Alternative Frames
Alternative frames needed if
CPI frame not available or supplementing
alternative frames are needed
• National level sampling frames
May provide urban/rural, area and other
broad level weights
Country frames may not cover outlets
• Local level outlet sampling frames
Usually are available
19
Alternative Frames
Issues:
• Is it available for statistical purposes?
• Is it available in the required time-frame?
• Is it reliable?
• Is it current?
• Does it have the required coverage?
• Does it include size indicators?
20
Sources of National Frames
• Retail trade census
• Survey of retail sales
• Value Added Tax (VAT) records
• Business registers
• Telephone directories (“Yellow Pages”)
21
Sources of Local Frames
• Local government records
Property tax
Other regulatory systems
• Local chambers of commerce
• Employer organisations
22
Retail Trade Survey
• Frames may exist at national and local levels
• Useful for PPS sampling
• Usually maintained by statistics office
23
Value Added Tax (VAT)
• VAT administrative records provide detailed lists of
businesses
up to date
not be available for statistical purposes
support PPS sampling (provide size details)
exclude some sectors such as food
exclude informal sector businesses
• If available, care needed with exclusions
requires supplementary sources
24
Business Registers
• Maintained for:
Statistical purposes
Administrative purposes
• May contain retail outlet locations/addresses
• May include size measures
sales volume
number of employees
Useful for PPS sampling
25
Telephone/Business Directories
• “Yellow Pages” and similar directories
• Exclude size measures
• May include detailed lists of businesses although
might note be comprehensive
PPS not possible
Simple random sampling can be used
26
Time Sample
• Aim at average prices for the whole year
Need quantity weighted transaction price
Choices: Monthly/Quarterly/Annual
Frequency: determined by variability
• Type of charge: annual, seasonal etc.
27
Price Collections – A Practical
Approach
Procedures:
1. The finalised regional product list of 651 product items
are provided by the RO to the ICP national coordinator
2. Each country then identifies the items from the list on
whether they are representative, available or not
available
3. To ensure comparability, items are reviewed carefully
in terms of the regional product specifications (PS)
28
Price Collections – A Practical
Approach
Regional Product List
•
Contains a comparable and representative regional
basket of goods
•
Minor differences even if the product specifications are
clearly specified
29
Price Collections – A Practical
Approach
Example: Rice
Product Specification: 5 Kg pre-packaged bag of
Jasmine rice with low share of broken rice
Price collected by Country A: 1Kg pre-packaged bag
of short grain rice with high share of broken rice
It is clear that these two product items possess different
characteristics and must not be compared as the same
item
30
ICP Requirements
•
Provision of annual average national prices
corresponding to the product list for the compilation of
PPPs
•
Preparation of a suitable sample frame covering
region, province, city, outlets and urban/rural locations
to compute a national price
•
Identification of representative and available items by
conducting pilot surveys in advance of the reference
year
•
Installation of a fully functional ICP Tool Pack
•
Conduct of Tool Pack Training for participating
countries
31
Frequency of Data Collection
Inflation and Frequency of Data Collection
Inflation affects how often price collections should be
conducted
•
High inflation products such as food prices must be
collected more frequently.
•
Low inflation products such as household durables
price collection can be conducted quarterly or semiannually
32
Frequency of Data Collection
Fluctuation of the prices
Item
Frequency of Data Collection
Monthly
Food items (rice, fresh fruit &
vegetables, fish , meats)
Household durables (washing
machine, and refrigerators)
Utilities (electricity, water,
insurance)
33
Quarterly
SemiAnnually
Frequency of Data Collection
Countries with Reliable CPI and Low Inflation
• Countries with reliable CPI and low inflation semiannual collections would be sufficient
• Average annual national price can be extrapolated
relevant sub-indices from CPI
34
Data Submission
•
Deadline and timetable on submitting ICP price survey
data is set by regional coordinator in agreement with
all countries
•
The national coordinator sets the country timetable for
submission in different collection zones to meet
regional deadline
35
Data Validation
•
Intra-country data validation by RO is conducted each
time a country submits the national averages and
individual price data
•
Inter-country validation are done once submissions
from at least four countries have been received
•
Data review workshops (DRW) are conducted each
quarter by RO to validate data submission from all
participating countries
36
Thank You!
37