PRODUCER PRICE INDEX METHODOLOGY TANZANIA MAINLAND
2019 i
PRODUCER PRICE INDEX METHODOLOGY
ii
Table of Contents
1.0
Introduction ................................................................................................................. 1
1.1
Definition of PPI .......................................................................................................... 1
1.2
Uses of PPI .................................................................................................................... 2
2.1
Scope and Coverage .................................................................................................... 2
3.1
Structure, Composition and Weighting of the Index ............................................. 2
3.1.1
Sampling ............................................................................................................ 3
3.1.2
Rules for Weights Derivation ......................................................................... 3
4.1
Activities (ISIC Classes) and Establishments for Price Collection ....................... 5
4.1.1
Selection of ISIC Classes.................................................................................. 5
4.1.2
Selection of Establishments ............................................................................ 5
4.2
Criteria for Selecting Sampled Products .................................................................. 6
4.3
Aggregation formula and compilation methodology ............................................ 6
4.3.1
Elementary Aggregates Indices ..................................................................... 6
4.3.2
Higher Level Indices ........................................................................................ 7
4.4
Periodicity and Frequency of Rebasing of the Indices........................................... 8
4.5
Data Collection and Pricing Point ............................................................................ 9
4.5.1
Data Collection Procedure .............................................................................. 9
4.5.2
Point in Time Pricing ....................................................................................... 9
4.5.3
Key issues for Product Pricing During Data Collection ............................. 9
4.6
Price Data Editing and Analysis Procedures ........................................................ 10
4.6.1
Data Editing .................................................................................................... 10
4.6.2
Producer Price Index Analysis ..................................................................... 10
iii
4.7 Quality Adjustment .................................................................................................. 11 4.8 Treatment of Missing Prices .................................................................................... 11 4.8.1 Temporarily Missing Prices .......................................................................... 11 4.8.2 Missing Prices for Seasonal Products .......................................................... 11 4.8 Conclusion .................................................................................................................. 12 4.9 Appendices ................................................................................................................. 13 Appendix 1: PPI weights of ISIC Level Four ........................................................................ 13 Appendix 2: PPI weights of ISIC Level Three............................................................ 16 Appendix 3: PPI weights of ISIC Level Two .............................................................. 19 Appendix 4: PPI weights of ISIC broader sectors ..................................................... 19 List of References ....................................................................................................................... 20
1
1.0 Introduction
National Bureau of Statistics (NBS) is the main organization responsible for collecting
statistics to support evidence -based decision making in Tanzania , as stipulated in
Statistics Act Cap. 351 . NBS plays a key role in coordinating the National Statistical
System, ensuring data quality, and maintaining statistical standards. It conducts national
censuses, surveys, and economic reports that guide government planning, policy
development and monitoring.
The methodology for developing Producer Price Ind ex in Tanzania is a complement to
the International Producer Price Index Manual (PPI Manual) published by the IMF in
2004. The purpose of this methodology is to provide a technical experience of how to
compile PPI and disseminate it regularly. IMF AFRITAC provide a technical assistance
project in Tanzania to build the country’s statistical capacity in the PPI compilation.
The concepts and definitions used are consistent with international standards,
specifically the 2008 System of National Accounts (SNA), which provides a comprehensive
framework for compiling national accounts, and the 2004 Producer Price Index Manual:
Theory and Practice, which offers detailed guidelines for the compilation and
interpretation of producer price indices to ensure comparability and reliability of
economic data.
1.1 Definition of PPI
The Producer Price Index (PPI) measures the average change over time in the prices
received by selected domestic producers for the sale of their goods and services at the
first point of transaction. It serves as a key indicator of inflation at the producti on level
and helps in analyzing price trends in various sectors of the economy. The index is
compiled using a representative basket of selected commodities, which reflects the
structure of domestic production. These commodities are carefully chosen based on their
economic significance and are priced on a quarterly basis to ensure accurate and timely
tracking of price movement.
2
1.2 Uses of PPI
The Producer Price Index (PPI), is an important indicator of economic performance that
provides information for economic policy making used by stakeholders to gauge
industrial performance and to forecast future economic performance. Many countries
compile PPI, as it is one of the warning indicators of short-term economic development.
Also, it is used to identify the turning points in economic development at an early stage.
2.1 Scope and Coverage
The scope of the Producer Price Index (PPI) encompasses key sectors of the economy,
including mining and quarrying; manufacturing; and utilities which consist of electricity,
gas, steam and air conditioning supply, as well as water supply, sewerage, waste
management, and remediation activities. The PPI covers both local and export markets,
capturing price changes at the first point of commercial tr ansaction, while excluding
taxes, subsidies, and transportation costs, to reflect pure producer price movements.
The industry classification used in the compilation of the index aligns with the
International Standard Industrial Classification of All Economic Activities, Revision 4
(ISIC Rev. 4), ensuring international comparability. Data for the PPI are collected fro m
selected establishments across the country, representing a wide range of producers in the
covered sectors.
The index is calculated based on a fixed basket of representative products, selected
according to their economic relevance and stability in the market. A designated base year
is used as a reference point for measuring relative price changes over time, allo wing for
consistent tracking and analysis of producer price trends.
3.1 Structure, Composition and Weighting of the Index
The structure and weighting pattern of the Producer Price Index (PPI) were derived from
the 2015 Annual Survey of Industrial Production, using gross output data as the primary
source. The methodology employed a top-down approach to develop the index structure,
determine the weights, and select the product basket. Once the approach was finalized,
3
a fixed basket of representative products was chosen, and 2015 -based weights were
calculated accordingly.
3.1.1 Sampling
A cut-off sampling method was used to select a sample of establishments for inclusion in
the price survey. This method ensures that only establishments contributing significantly
to total production are included, thereby improving efficiency and representativeness in
data collection.
The top-down approach was specifically adopted to maximize the indirect representation
of industries and products within the index. In this approach, relatively minor industries
were progressively excluded at successive levels of industry aggregation (i.e. , at the 1 -
digit, 2-digit, 3-digit, and 4 -digit ISIC Rev. 4 classification levels). However, even with
these exclusions, the full relative value of aggregate outputs at each level was retained.
This ensures that excluded industries still contribute to the overall weighting structure.
At the 4-digit class level of ISIC, the prices of selected commodities that are directly priced
serve as proxies or indicators for related industries within the same classification level
that are not directly priced. The rationale behind this method is tha t unpriced items are
more likely to exhibit similar price movements to closely related products than to
dissimilar ones, thereby preserving the integrity and relevance of the index while
maintaining a manageable data collection process.
3.1.2 Rules for Weights Derivation
The following general rules were applied in the derivation of weights for the Producer
Price Index (PPI), using a top-down approach:
i. At the 1 -digit section level of ISIC, relatively minor 2 -digit Divisions
typically those contributing less than 5% to the total gross output of the
section (subject to variation based on the degree of industry concentration)
were excluded from direct inclusion.
4
ii.
The gross output values of the excluded 2-digit industries were
proportionally redistributed (prorated) among the selected industries
within the same section, ensuring the retention of the section’s full weight
in the index.
iii.
The same elimination and proration process was applied at the 3-digit
Group level, with relatively small industries being progressively excluded.
iv.
The procedure was further repeated at the 4-digit Class level, where
directly priced items were used as proxies for related but unpriced items
within the same class.
v.
The final weighting pattern was expressed in percentage terms,
representing each industry's contribution to the total weighted output used
in the index calculation.
It is important to note that, during the process of eliminating relatively small industries at successive levels of the ISIC classification structure, discretion was applied to account for several important factors. These included: i. Recognizing emerging and growth industries that, while currently small, are expected to expand significantly during the life cycle of the PPI before its next rebase; ii. Selectively retaining unique industries or products whose price movements are not likely to be adequately represented by those of other, more general industries or products; iii. Ensuring higher direct coverage within dominant divisions to improve the representativeness and reliability of the index.
5
4.1 Activities (ISIC Classes) and Establishments for Price Collection 4.1.1 Selection of ISIC Classes The 4 -digit ISIC classes were selected to represent each of the remaining 3 -digit ISIC groups after the elimination of relatively small industries. These selected classes are the industries for which price indicators were directly collected from a sample o f producer establishments. The selected 4 -digit industries not only represent themselves but also indirectly represent the excluded industries, whose weights were reallocated using the top-down allocation approach described above.
4.1.2 Selection of Establishments For the purpose of weight computation, a total of 2,389 establishments from the 2015 Annual Survey of Industrial Production were initially considered. Within each ISIC Rev.4 group, establishments contributing at least 80 percent of gross output were select ed to ensure the sample accurately reflects the industry's production value.
From this process, a final sample of 201 establishments was selected for inclusion in the Producer Price Index, broken down as follows: • 7 establishments representing mining and quarrying, • 186 establishments representing manufacturing, and • 8 establishments representing utilities.
These establishments collectively covered: • 68 activities at the 4-digit ISIC level, • 48 activities at the 3-digit ISIC level, and • 27 activities at the 2-digit ISIC level.
Under this top -down reallocation methodology, the price movements of unpriced industries are assumed to be broadly represented by the price trends of similar priced industries. This approach helps maintain the comprehensiveness and accuracy of the index while keeping the data collection process efficient. 6
4.2
Criteria for Selecting Sampled Products
The following criteria were applied to ensure that the sampled products accurately reflect
producer price movements and maintain consistency over time:
i.
Representativeness: Sampled products are representative of a broader range of
products within each industry in terms of price behavior over time;
ii.
Economic Significance: The selected products are typically the producer’s
highest-selling items in terms of value, ensuring their economic relevance;
iii.
Regular Availability: Products are regularly produced and sold, allowing for
consistent and reliable price tracking;
iv.
Coverage: Three products were selected to represent each ISIC class from each
producer. However, where necessary due to product diversity or market
segmentation a larger number of products was selected to ensure adequate
representation; and
v.
Detailed Specification: A complete specification of each sampled product was
recorded, including all characteristics that influence its price. This includes:
physical attributes (e.g., size, quality, packaging), Unit of measurement,
transaction details, ensuring consistent product quality over time and accurate
measurement of price change.
4.3 Aggregation formula and compilation methodology
4.3.1 Elementary Aggregates Indices The compilation of the Producer Price Index (PPI) follows a multi-level aggregation approach, starting from the most detailed level of classification. The first level of aggregation occurs at the 4-digit ISIC class level, where elementary aggregate (EA) indices are compiled. These EA indices are unweighted and calculated using the Jevons aggregation formula, which is the geometric mean of price relatives. This method is appropriate for capturing relative price changes without the influence of weights at this lowest level.
7
Example of Elementary aggregates Index four-digit level
4.3.2 Higher Level Indices Once the elementary indices are compiled, they are progressively aggregated to higher classification levels which includes 3-digit ISIC group, 2-digit ISIC division, 1-digit ISIC section, and finally, the overall (all-groups) index. At these higher levels, aggregation is performed using the Laspeyres formula, a base - weighted index formula that applies fixed weights derived from the base year (2015). This method ensures that each industry’s contribution to the index reflects its relati ve economic importance during the base period. Examples of higher-level Index two-digit level