• Overview
• Concepts
• Data sources
  • PPI sampling frame
  • Alternative data sources
  • Confidentiality of data sources
• Design
• Calculation
• Presentation
• History
• More Info

• Methods Home
• About
• Programs
• Glossary
search

Handbook of Methods Producer Price Indexes Data sources

Producer Price Indexes: Data sources

The Producer Price Index (PPI) is a voluntary survey, and although some companies decline to participate during survey initiation and others may drop out of the program over the course of the timeframe of the sample, respondent cooperation is paramount to the success of the survey and remains steady. The PPI sampling process is divided into two stages. During the first stage, samples of producing establishments within an industry are selected from the universe of establishments for the industry. The sampled establishments are then contacted by the Bureau of Labor Statistics (BLS) representatives, who attempt to solicit voluntary cooperation. If a sampled establishment agrees to provide information for the PPI survey, a second stage of sampling follows. During this second stage, the BLS representative obtains detailed product and transaction information for sampled items, including prices, describing a group of transactions that represent the full production of the firm. For a more detailed description of the PPI sampling process, see the design section of this PPI handbook.

Following initiation into the PPI survey, reporting establishments that agree to supply prices during the initiation process provide updated pricing information on an agreed-upon schedule, usually monthly, but sometimes less often. The PPI program obtains price updates through the BLS secure Internet Data Collection Facility (IDCF), permitting respondents to provide updates via the internet. Additional information for PPI survey respondents is available at the PPI respondents webpage.

All reported prices are analyzed using PPI computer systems, and price changes determined to be unusually large are flagged for further review by PPI analysts who verify the information before it is used in the calculation of indexes. These verifications include ensuring there were no material changes to product specifications, terms of transactions, or product substitutions. If any of these changes occurred, the reported price will be subject to a quality adjustment, which ensures comparability between the current and previous prices reported for the products. For a more detailed description of this process, see the calculation section.

PPI sampling frame

The universe for the PPI survey consists of roughly 5.0 million establishments comprising the covered portions of the mining, manufacturing, forestry, utility, services, and construction sectors. The PPI program constructs this list using Unemployment Insurance (UI) files taken from various state agencies. Supporting information and alternative frames may be obtained from other sources if the PPI deems these sources to be more accurate. The PPI program updates its samples on a regular basis to better represent current industry production. Over time, industries commonly experience changes in the universe of producers, as new establishments open and others close. Additionally, the various products produced by an industry change over time. Lastly, updating the sample of producers is vital to the maintenance of the PPI survey because over time some businesses that initially agree to cooperate withdraw from the survey.

Alternative data sources

The PPI program uses price data from alternative sources if these other sources provide the best information available and are suitable for PPI methodology. For example, the PPI program uses data from the Department of Agriculture for the calculation of many commodity-based farm product indexes. Occasionally, published prices or prices obtained from purchased datasets are employed by the program. These data can be used to calculate indexes alone or they can be blended with survey responses. The PPI program continues to seek out and evaluate potential sources of alternative data for use in index calculation. As part of the sample-update process for an industry, PPI conducts an industry study, including an environmental scan to identify alternative producer price information for that industry. To date, this ongoing effort has identified a limited availability of secondary source producer price information. Lastly, secondary data sources are also used when developing hedonic models for quality adjustment. (Additional details are available on the PPI quality adjustment methods page.)

To further mitigate duplication, PPI may share price information with other BLS price programs when circumstances permit. To increase efficiency and reduce overall respondent burden, the Consumer Price Index (CPI) program, PPI, and the International Price Program (MXP) may share resources to collect pricing information from respondents that are selected for inclusion in multiple surveys. In these cases, prices for the same product or service may be used by more than one price program; however, each program determines appropriate weighting according to its own established methodology. All information shared across programs is used for statistical purposes only and is protected under the BLS confidentiality pledge.

When survey respondents are unable to provide pricing information on a timely basis, PPI may use alternative data sources in some initial estimates. In these situations, initial prices are often those quoted on organized commodity exchanges or at central markets. These prices may be updated later with actual prices from survey respondents. This practice is used most often for farm and energy products.

Confidentiality of data sources

The publication of company-specific data in identifiable form is prohibited in the statistical and research work of BLS. Data from firms participating in the PPI program are protected to ensure respondent confidentiality even within BLS, so only those few staff members with a need to know can identify a respondent. Furthermore, BLS has publication criteria that prevent the inadvertent disclosure of a respondent’s identity to the public through movements in a published index. For more information about respondent confidentiality protections at BLS, visit the Confidentiality and Disclosure webpage.