Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Rolling review of the producers price index

This chapter describes the steps we take to develop or review a price index. We review the producers price index (PPI) on a rolling basis to ensure the index baskets remain representative of the actual purchasing and selling patterns of industries.

The main steps we take are:

Establish the conceptual basis and coverage of the index

The conceptual basis for a price index provides the foundation on which we build the index structure. It establishes the scope and coverage of an index, and outlines the key terms and methods. For example, the conceptual basis establishes the range of businesses from which we collect prices, definitions of the inputs and outputs we price, and key practices such as the treatment of taxes.

Using a well-defined conceptual basis ensures that the prices we collect over time are consistent, and our indexes have a clear interpretation. Without this consistency, the results will lack meaning and may be used inappropriately. For example, customers of the PPI inputs indexes might index costs in a supply contract using these indexes without realising they exclude labour and capital items.

See Conceptual basis of the producers price index for more information

Collect income and expenditure information from businesses

We carry out considerable research to derive income and expenditure values in sufficient detail for index weights to be accurately calculated, and appropriate items to be selected for pricing.

The main data sources we use are:

  • Commodity Data Collection Survey
  • other Statistics NZ surveys
  • import and export statistics
  • parliamentary reports and company reports
  • information from government departments, and professional and business organisations
  • ad hoc surveys and discussions with businesses.

The Commodity Data Collection Survey is a dual-purpose survey that collects income and expenditure information by commodity from a representative sample of businesses in each industry of the economy. In addition to being the main source of PPI weights, this survey provides the information we need to update income and expenditure commodity proportions, as used in the balancing of the current price annual national accounts.

Establish the commodity groups

Under the index structure, there are two levels of commodity groups – NA06CC commodities, and representative commodities.

To determine the weights of the NA06CC commodity level, we use the supply and use reconciliation carried out annually as part of the national accounts. By having this level, we can use the supply and use reconciliation to update the weights of the NA06CC commodities and industry levels annually. We have been using this method since 2011 to ensure our indexes remain relevant and up-to-date.

Representative commodities are designed to contain a selection of prices for similar types of goods and services. The commodities are loosely based on the international standard commodity classification, Central Product Classification.

However, the relative importance of each classification code determines the way that we split the income and expenditure of the industry into representative commodities. Categories with large incomes or expenditures may be split into two or more categories, while categories with small incomes or expenditures may be aggregated. Insignificant categories may be excluded altogether.

In addition, industry outputs may be sold both domestically and exported, while industry inputs may be sourced from domestic production and imports. Therefore, three separate representative commodities are possible:

  • commodity domestic sales
  • commodity export sales, and
  • commodity import purchases.

See Structure of the producers price index for more information about the commodity groups 

Select the items to be priced

The number and description of items we select to price in any commodity group depends on the importance of that commodity group in relation to total expenditure or income. Typically, priced items are representative of a broad category or group of items where small weights have been added together.

We select items by using statistical techniques or subjective methods such as purposive sampling, which relies on a wider range of information and judgement from the development team. When we use purposive sampling it is to ensure the respondent load on surveyed businesses is kept to a minimum.

Determine the index weights

The weighting process is an arithmetic procedure that enables us to express the dollar value of expenditure on, or income from, particular items or groups of items as index weights. These weights sum to a common total in every index. We calculate the weights at each level of the index according to the relative importance of commodities and businesses within an industry or industry group.

Often we derive an expenditure weight for a commodity or item that is too low in value to be worth obtaining an explicit price. This weight is added to the expenditure of a similar commodity item. This has the practical effect of imputing a price movement based on the movement of the similar commodity/item.

Document the weighting and pricing methodology

We create and maintain comprehensive documentation as part of the review. Its primary purpose is to inform customers and future developers about the nature of the measures produced.

Typically, a price index summarises a considerable amount of information and presents it as a single numeric series. In order to understand the meaning of this abstract set of numbers, customers need accessible documentation that explains both the index methodology and practices. This documentation is particularly important for analysts interested in time series analysis. Such documentation is the key to understanding an index’s coverage and quality many years after it has been produced. This documentation generally includes:

  • an explanation of the conceptual basis of the indexes
  • analysis of data sources used for both weights and prices, including a discussion of quality and coverage
  • details of pricing methodologies, including strengths and weaknesses
  • details of baskets produced, including a comparison with previous information to explain the impact of structural change in the industry.
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+