PRODUCER PRICE INDEX ON DOMESTIC MARKET
METHODOLOGY
1. Purpose, nature and use
Producer Price Index (PPI) is one of the main short-term business indicators; it is regarded as one of the important measures of the economic situation in the Country. Indexes for particular economic activity measure the average change in the prices of industrial products, which are produced and sold by Bulgarian enterprises. This is done on the bases of constant sample of groups of products, produced by the activity and sold on the domestic market and that sample is representative for total industrial production.
The purpose of the price survey is price indexes to be calculated at different levels of production, price indexes are based on the monthly survey on prices applied from Bulgarian producers on domestic market. In this way a system of price indexes will be build, that will serve the needs of aggregated information as well as the needs of more detailed information (example price changes of particular activities or price changes of main industrial groupings).
Producer price index on domestic market is considered as one of the main producer price index, that has several main uses:
- As a main short-term business indicator that predicts the tendencies in inflation rate; PPI can be considered as an early measure of the current inflation process in the economy;
- As a deflator for calculation the values at constant prices. Example for deflation in National accounts, or for deflation of industrial sales on domestic market and then for compilation the industrial production index;
- Some ministries and branch associations use PPI for monitoring price pressures on sub-sectors of Bulgarian industry;
- Trade associations and companies use PPI as an impartial measure of prices for contractual purposes and for comparing with their own patterns of purchases and sales.
- As a short-term indicator required from different international organizations like Eurostat, International Monetary Fond, European Central Bank, that use PPI for international comparisons and economic supervision.
2. Definitions and sources of information
For the purposes of the producer price indexes survey the surveyed price is defined as net revenue received from particular enterprise, for particular product, sold in particular day of the month.
PPI measure price changes between two different periods of time. The average price change cannot be measured directly; it must be estimated through survey of the actual prices of the products in particular point of time. PPI do not attempt to measure the actual price levels, they are limited in measuring of the average price changes between two points of time.
Information for calculation of PPI is received on the bases of monthly survey on producer prices of industrial products, sold on domestic market. The aim of Producer price survey is: (1) to be compared prices of specific products with equal quality in the two surveyed time periods; (2) In the case of quality change of the surveyed specific products this quality change to be taken into account.
3. Coverage and sampling technique
The PPI survey covers mining, manufacturing industries and production and distribution of electricity, steam, and natural gas and water supply.
Indexes are calculated at the group level (third digit), division level (second digit), at section and subsections of National Classifications of Economic Activities (NACE.BG). Indexes are calculated for Main Industrial Groupings (MIGS: Intermediate products, Investment products, Consumer durable products, Consumer non-durable products, Energy products), following EU Regulation 586/2001 concerning Short-term statistics as regards the Main industrial groupings.
The sampling method used in the PPI survey involves a three stage sampling process: first of PRODCOM groups, second of reporting units and then of specific products (transactions).
3.1. Sample of PRODCOM product groups and Sample of reporting units
First on the base of the annual survey on production and sales of industrial products - PRODCOM survey the sample of products groups is made. After that for each of the selected PRODCOM groups the sample of reporting units is made. These units (enterprises) will participate in the price survey. A probabilistic method - proportional to size is used to ensure sufficient coverage. Probability proportional to size techniques involves identifying a stratum that will be selected with certainty (i.e. probability=1) and the remaining strata selected with probabilities based on their relative contribution to sales on domestic market. The number of the units selected in the sample depends on the degree of industrial concentration of the enterprises from each sampled PRODCOM group. In production with high degree of industrial concentration where largest 5 enterprises have 75% of sales only these 5 enterprises are selected. For the other groups with smaller industrial concentration where the largest 5 enterprises have less than 75% sales share more enterprises are selected and sampling technique ‘probability proportional to size’ is applied. In this way companies that are small sized in terms of the turnover may be selected but that companies have important share of the market of a particular product group.
3.2. Sample of specific products (transactions)
For each enterprise with the corresponding PRODCOM groups that were sampled for price survey, the Price index estimation are based on the sample of specific product (specifications). For each enterprise the selected PRODCOM product groups are desegregated into group of product that show equal price changes and from these groups representative specification (specific products) are selected. The selected specifications reflect in the best way the price changes of the whole group.
At that stage of sampling so called ‘purposive sampling’ or ‘expert choice’ is applied. The experts from the selected enterprises choose the price quotations, in other words products with the highest sales volume are selected and long presence on the market. During the initiation process each enterprise sampled for price survey specifies the specifics products (specifications), which prices it will report every month. These specifications must have the following attributes:
-
Are representative of the company’s production in the relevant product groups selected in the sample
-
Are manufactured and sold within the country. The definition of Bulgarian manufacture is that the nature of the good must be changed in the country. As examples, assembly of imported parts within the country would be included, but not a simple repackaging of an imported goods.
After the specific products for monthly price survey are selected, they have to fulfill the following conditions:
-
The products must be quoted on the same basis each month – i.e. the same product and the same terms of sale.
-
The product and the terms of sale must be described in enough detail to ensure consistency each month.
-
The price must be recorded at the time at which the product is ordered.
-
Prices are ex-factory selling price, i.e. excluding any delivery charge outside the factory.
Companies that report prices of the selected products are not allowed to report average prices of all transactions of the selected product that happened during the month. Such average prices tend to be very violate and can be affected by different discounts that enterprises provide for different customers for large orders. Such fluctuations reflect changes in customer mix rather than genuine changes in prices. For this reason, companies are asked to quote prices for similar transaction each month. Though there are industries where average prices are accepted as a preferable alternative to unrealistic list prices.
4. Calculation of Producer Price Indexes
Producer price indexes are base weighted (Laspeyres) indexes - that is to say they are weighted according to the sales in the base year, currently 2000. After specification’s price collection from the enterprises, elementary indexes (price relatives) are calculated for each specification (price relative – specification price in the current month divided by average price of that specification in the base year). After that price relatives are weighted with sales structure in the base year. The weight reference period according to the EU Regulation 1165/98 is updated at every five years, which must be ending at ‘0’ or ‘5’.
Producer price indexes at different levels of aggregation are calculated from successive aggregations in which each level of aggregation uses the arithmetic mean of indexes at the level below, duly weighted with sales structure in the base year following the structure of the National Classification of Economic Activities, which is consistent with NACE Rev 1.1.
In the weighting system there are two categories of weights, which depend on the level of aggregation:
-
At group level – 3-th digit of NACE.BG indexes are calculated as price relatives are weighted with sales structure on domestic market by PRODCOM groups. These weights are received by annual full coverage PRODCOM survey in the base year.
-
Higher-level indexes are weighted with the structure of sales on the domestic market, derived from Structural Business Statistics in the base year 2000.
4.1. Calculation of monthly base year price index (2000 =100)
4.1.1. Calculation of elementary indexes (price relatives) at representative product (specification) level.
Elementary indexes are calculated as the price of the selected specific product in the current month is divided by its average price in the base year.
where:
 |
Elementary index of price change of the representative product ‘i’ in month ‘t’ |
 |
Current price of the representative product ‘i’ in month ‘t’ |
 |
Average monthly price of the representative product ‘i’ in the base year |
4.1.2 Index calculation at PRODCOM group level
where:
 |
PRODCOM product position, that consist of ‘Ê’ representative products ( i = 1. . . . . . k ) |
 |
Price index for PRODCOM product position ‘Í’ |
 |
Price indexes of the representative products from PRODCOM product position ‘Í’ |
4.1.3. Producer price indexes at 3-th digit level of NACE Rev.1.1.
where:
 |
Price indexes at 3-th digit level of NACE Rev.1.1. |
 |
Price index for PRODCOM product position ‘Í’ |
 |
Value of sales on domestic market for PRODCOM product position ‘Í’ in the base year |
4.1.4. Producer price index calculation at higher levels of aggregation - 2-nd digit level of NACE.BG, section, sub section, Main industrial groupings and total Industry level.
Producer price index calculation at higher levels of aggregation are base weighted Laspeyres indexes that is to say they are weighted according to the sales on the domestic market in the base year – 2000.
where:
 |
Sales structure on domestic market in 2000 at correspondent level of aggregation. |
 |
Value of sales on domestic market at correspondent level of aggregation in 2000. |
Calculation of Producer price indexes on domestic market at different level of aggregation is done consecutively as follows:
-
Indexes at 2-nd digit level of NACE Rev.1.1 are calculated as monthly indexes at 3-th digit level are weighted with sales structure on domestic market at 3-th digit level of NACE.BG
-
Indexes by sectors, sub sectors and by Main industrial groupings are calculated as indexes at 2-nd digit level are weighted with sales structure on domestic market of 2 digit levels in the correspondent nomenclature.
-
Indexes at total ‘Industry’ level are calculated as indexes at 2-nd digit level are weighted with sales structure on domestic market of 2-nd digit levels in the tolal industry sales.
In general Producer price indexes at higher levels of aggregation (2-nd digit level and higher) are calculated as sum of product of indexes at lower levels, multiplied by sales structure.
where:
 |
Producer price index in month ‘t’ |
 |
Price index for group (division, section, sub section, industry-total) |
 |
Sales structure on domestic market in 2000 at relevant level of aggregation |
4.1.5. Calculation of base year price indexes for different periods of time (2000 =100).
Indexes for different periods of time at 2000 as a base can be calculated as a simple average of the monthly base year price indexes. For example the price index for the period from the beginning of the year can be calculated as a sum of the monthly base year indexes from the beginning of the year divided by the number of the months within that period.
where:
 |
Index for the period from the beginning of the year at 2000 as a base’ |
 |
Price index for month ‘t’ |
| n
|
Number of the months within the period |
4.2. Calculation of the monthly indexes and indexes for periods at different index base.
From calculated in such a way base year indexes might be calculated monthly indexes (example – index for the current month according to the previous month or to the corresponding month of the previous year) and indexes for periods of time (example – index for the period from the beginning of the year to the corresponding period of the previous year) at different index base.
where:
 |
Index according to the previous month |
 |
Index according to the same month of the previous year |
5. Replacement of representative products (Quality corrections):
In the producer price survey it is very important to take account the changes in quality by means of price changes i.e. to take account price changes that are due to inflation but not these that are due to changes in the quality of the priced products. It is very important product with constant quantity to be priced every month, because if quality changes in products are not identified and isolated/removed from reported prices before being used as indicators of pure price change in the index, then the accuracy of the PPI will suffer and calculated PPIs will be significantly biased. In the case of quality changes of the surveyed products care is taken to ensure that only 'pure' price changes are recorded for PPI purposes and the old product must be replaced from the new product from the same product group. Currently there are four types of changes and methods for quality corrections in the new PPI system.
5.1. Method for direct replacement. If the main characteristics of the new product can be directly compared with the main characteristics of the old product (the product that was priced to the moment), in the case that there are not fundamental change, then the base price of the new product is not recalculated and in the following months the price of old product is directly replaced with the price of the new product.
5.2. Method for direct quality adjustment. This method is applied when the value of the quality change is known. In that case the producer is asked to report what part of the price increase is due to increase of the quality of the new product. Then for recalculation of the base price of the new product this part is extracted from the price of the new product for the current month.
where:
 |
Recalculated base price of the new product |
 |
The price of the new product in current month |
 |
The price of the old product in the base period |
 |
The value of quality changes of the new product |
5.4. Overlap method. This method is applied when the change in the price of the new product is due entirely to the change in specification and the value of the quality change is unknown.
where:
 |
The price of the old product in the base period |
 |
The price of the new product in the previous month |
 |
The price of the old product in the previous month |
6. Publicity and transparency
6.1. Announcement of the results
The PPI data is firstly released at the regular press conferences of NSI 40 days after the end of the reference month. The announcement includes the base year Producer price indexes as well as indexes according to the precious month and indexes according to the same month of the previous year. Thereupon all users have access to the final data thought Statistical Services Division in the NSI.
6.2. Publications
Data is published in monthly bulletin “Statistical Journal” In the bulletin the published tables contain Producer price indexes at total industry level, by sections and by 2-nd digit level of NACE.BG
6.3. Electronic access
Every month PPIs are posted on the NSI website. The methodology for compilation of the Producer price indexes on domestic market also is posted in NSI WEB-site.
Published on 11.11.2003
