Industrial Product Price Indexes (IPPI)
Detailed information for May 2002
The Industrial Product Price Index (IPPI) measures price changes for major commodities sold by manufacturers in Canada.
Data release - June 27, 2002
- Questionnaire(s) and reporting guide(s)
- Data sources and methodology
- Data accuracy
This monthly survey collects transaction prices for products made by domestic Canadian manufacturers. These prices are then used to create various systems of price indexes (macro models such as commodity indexes and industry indexes).
- Industrial product price indexes
- Prices and price indexes
Data sources and methodology
This is a sample survey.
The sampling frame for the IPPI is based on the most recent data available from the ASM. The most common survey unit is the "establishment" which, in Statistics Canada's usage, refers to the smallest unit of an organization that can provide the basic data required for industry statistics, most commonly an individual plant or factory.
Sampling for any directly surveyed elemental index is usually done by a formal probability sampling procedure. Important producers are "must take" respondents. A random selection is taken from the smaller producers. These samples are redrawn on a regular cycle as dictated by changes in the market. A directly surveyed elemental index is usually supported by 12 to 15 price quotes. An index may be derived from as few as 8 quotes or as many as 30 quotes. Each month, about 9,000 price quotes, from about 3,000 producers for goods sold, are used in the calculation of the IPPI.
The entire survey sample is usually reviewed in total once every 5 years. In the interim, re-sampling occurs primarily for commodity groups subject to relatively rapid development in firm structure or product development.
About 700 of the 980 Principal Commodity Group Aggregates (PCGA's) in the manufacturing sector are accommodated by direct survey and these account for about 90% of the value of manufacturing output in 1997. The price movement for the other PCGA's is estimated indirectly from either directly priced PCGA's or through borrowing price movements from other price series.
If a price quotation is not supplied in a given pricing period, an estimate must be made for it. If no action is taken, the last reported quotation will be used as an estimate for the missing one on the assumption that prices have not changed. Though most prices for most commodities do not change in any given month, it is almost certain that some of the missing prices will prove to have changed, so that the more aggregative indexes will need to be revised because of the impact of those individual changed prices.
IPPI series are subject to revision for six months after publication of a given month's data. This is to accommodate late reports, corrections and some price quotations which are supplied quarterly or semi-annually.
Some commodities are not surveyed, but estimated indirectly - some are shown in 62-011. A scale relating to sample error is also published with data for each commodity. Bias in indexes arises most easily in treatment of quality change. All of these, and other related topics are discussed in 62-556 "Industrial Product Price Indexes: 1981=100: Concepts and Methods" (62-556).
The IPPI attempts to express, in a single number, price changes that involve a range of commodities. There is no obvious, let alone absolute, way of perceiving what a composite price movement really is. For the IPPI, the fixed quantities of the commodities are proportional to the sales made by manufacturing establishments in Canada in the basket reference period, that is, 1997. Consequently considerations about the quality of the IPPI relate to these concepts of composite price movements.
The statistical accuracy of this index depends on price and shipment value data. Price data is principally obtained from a sample survey. Shipment value data also mainly rests on other sample surveys, such as the ASM and the input-output tables. Both kinds of input data are subject therefore to their own errors. Users concerned with the accuracy of the underlying input/output tables for 1997 and the ASM are invited to contact the appropriate Statistics Canada Divisions that is Input-Output Division and the Manufacturing, Construction and Energy Division.
The accuracy of the monthly commodity price quotations is maintained through the expertise of the commodity specialists. They develop a thorough industry knowledge, which is supplemented by outside personal contacts for particular industries. Much time and effort is devoted to detecting and following up unusual fluctuations in the monthly pricing patterns of commodities.
In this vein, the IPPI is closer to a typical survey carried out by Statistics Canada than it is to the Consumer Price Index, as the data is almost entirely collected through a survey process. As such, response rates are important to the overall quality of the index, and in this regard the commodity officers strive to ensure that the IPPI is well responded to, with an estimated weighted response rate of 94%.
In addition to sampling errors, producer price indexes are also subject to errors in price collection and editing. This is particularly true when quality change occurs in the sample with regard to the estimation of pure price change. To estimate pure price change over time, the prices to be directly compared through time ought to refer to identical commodities sold by identical establishments under identical conditions of sale. When this is not practical, prices must undergo adjustments for quality differences, which can prove to be a complicated process to implement. Among the thousands of prices that are compared every month to those of the preceding month, some may not fully satisfy these requirements. The total distortion from this source of error is not thought to be as significant at higher levels of IPPI aggregation.
Though the IPPI uses a sample survey methodology to obtain the necessary information, confidence intervals are not currently estimated, due to the longitudinal nature of price index series. Indexes for higher and lower levels of aggregation are considered to be statistically reliable.
- Explanation of methods used and additional sources for industrial price indexes
- Industrial Product Price Indexes - Data Quality Statements