Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Tag Archives: Methodology

Learning about the Producer Price Index Special Groupings

Inflation has been in the news—and on all of our minds—in recent months. Indeed, BLS price indexes have been prominent in nearly every inflation story. You may be familiar with the Consumer Price Index, which measures changes in the prices consumers pay for goods and services. You may be less familiar with the Producer Price Indexes (PPI), which measure price changes from the perspective of businesses.

The headline PPIs are the Final Demand-Intermediate Demand (FD-ID) Indexes. These indexes measure changes in the prices businesses receive for goods, services, and construction sold to end users (final demand) and to other businesses as inputs to production (intermediate demand). From July 2021 to July 2022, the Producer Price Index for Final Demand rose 9.8 percent.

BLS recently introduced two special grouping indexes that break government purchases into defense and nondefense purchases. I’ll say more below about these new groupings, but let’s first look at the variety of PPI data we publish.

PPI Basics

The PPI measures price change for more than 10,000 products and groups of products. We organize these items in many ways to show price change for categories of interest. We group the FD-ID indexes by the type of product sold. Within final demand, the three main indexes are final demand goods, final demand services, and final demand construction. These indexes measure price changes for products intended for personal consumption, capital investment, government, and export.

The main intermediate demand indexes are processed goods for intermediate demand, unprocessed goods for intermediate demand, and services for intermediate demand. These indexes measure price changes for business-to-business sales of fabricated goods, unfabricated goods, and services used as inputs to production.

In addition to the breakouts highlighted in the PPI news release, BLS also publishes special grouping indexes that further divide final demand and intermediate demand in useful ways.

The most popular special indexes are often called “core” indexes. These indexes remove historically volatile components, which may make it easier to understand the underlying rate of inflation. For example, you may have seen news stories that mention the special index for final demand less foods, energy, and trade services.

Indexes for Type of Buyer

A second set of special grouping final demand indexes organizes PPI data by type of buyer, based on Gross Domestic Product categories. These categories include personal consumption, private capital investment, government purchases, and exports. (The PPI measures price changes for domestic producers and thus excludes import prices.) Organizing final demand by type of buyer, rather than type of product, helps us identify when different categories of buyers experience differing rates of inflation.

The chart below shows the special grouping indexes for final demand by buyer type from January 2019 through July 2022. (For comparison purposes, these indexes are rebased to 100 in January 2019.) The chart highlights how different end-use buyers can experience different rates of inflation.

Producer Price Indexes for final demand by type of buyer, January 2019 to July 2022

Editor’s note: Data for this chart are available in the table below.

At the beginning of the COVID-19 pandemic, prices for exports and government purchases initially dropped more than prices for personal consumption and capital investment. From January 2020 through April 2020, prices for exports decreased 3.0 percent and prices for government purchases fell 4.1 percent. At the same time, prices declined 1.6 percent for personal consumption and 0.7 percent for private capital investment. From April 2020 through July 2022, prices jumped 26.9 percent for exports and 26.8 percent for government purchases, compared with increases of 18.1 percent for personal consumption and 19.5 percent for private capital investment.

Indexes for Government Purchases of Goods and Services

Now let’s examine the new special grouping indexes for government defense and nondefense purchases. The most highly weighted items in the defense index include military aircraft; engineering services; jet fuel; search detection, navigation and guidance systems and equipment; and machinery and equipment wholesaling. The most highly weighted items in the nondefense index consist of new school building construction, portfolio management, commercial electric power, diesel fuel, and property and casualty insurance.

The chart below presents the new government purchases indexes from January 2019 through July 2022. (Again, for comparison purposes these indexes are rebased to 100 in January 2019.) Although the two indexes move similarly, there is one notable difference. Prices for defense purchases fell more sharply than prices for nondefense purchases at the start of the COVID-19 pandemic. Both indexes moved higher beginning in mid-2020, but prices for government defense purchases have increased more slowly than for nondefense purchases.

Producer Price Indexes for government purchases, defense and nondefense, January 2019 to July 2022

Editor’s note: Data for this chart are available in the table below.

Indexes for Distributive Services

A third set of special grouping indexes are for distributive services. These indexes are particularly relevant now, given recent issues in U.S. and global supply chains. These indexes measure price changes for services associated with distributing goods sold to final demand and intermediate demand. Distributive services include transporting, storing, and reselling goods.

The chart below presents the indexes for final demand distributive services and intermediate demand distributive services from January 2019 through July 2022. The final demand distributive services index measures price changes for distributing goods sold to final demand (personal consumption, capital investment, government purchases, and exports). The intermediate demand distributive services index measures price changes for services associated with distributing goods sold to other businesses as inputs to production.

Producer Price Indexes for distributive services, January 2019 to July 2022

Editor’s note: Data for this chart are available in the table below.

Both distributive services indexes rose sharply from 2019 through 2022. Notably, neither index showed the dip at the beginning of the COVID-19 pandemic that happened with many other BLS price indexes. The intermediate demand distributive services index rose nearly 30 percent since March 2020, compared with approximately 25 percent for final demand distributive services. The larger increase in prices for business-to-business distributive services potentially highlights supply chain issues early in the production pipeline.

These are just some of the special groupings available for the PPI. You can see the full list of special grouping final demand and intermediate demand indexes in table 1 of the PPI news release.

Producer Price Indexes for final demand by type of buyer
MonthFinal demandExportPersonal consumptionGovernment purchasesPrivate capital investment

Jan 2019

100.000100.000100.000100.000100.000

Feb 2019

100.343100.000100.345100.448100.086

Mar 2019

100.857100.442101.034101.166100.172

Apr 2019

101.542100.795101.724101.614100.516

May 2019

101.628100.618101.810101.794100.774

Jun 2019

101.542100.530101.897101.345100.430

Jul 2019

101.714100.618102.069101.794100.860

Aug 2019

101.799100.353102.241101.704100.688

Sep 2019

101.37199.735101.897101.345100.000

Oct 2019

101.79999.912102.414101.435100.430

Nov 2019

101.37199.823101.810101.256100.430

Dec 2019

101.457100.088101.810101.435100.602

Jan 2020

101.971100.707102.328101.883100.774

Feb 2020

101.457100.000102.069100.807100.172

Mar 2020

101.20099.558101.810100.000100.086

Apr 2020

100.00097.703100.69097.668100.086

May 2020

100.51497.968101.46697.84899.742

Jun 2020

100.85798.410101.63898.744100.172

Jul 2020

101.45798.763102.24199.641100.688

Aug 2020

101.54299.117102.328100.000100.516

Sep 2020

101.71499.735102.50099.910100.344

Oct 2020

102.399100.618103.190100.179101.032

Nov 2020

102.228101.148102.759100.538101.032

Dec 2020

102.314102.208102.586101.166100.946

Jan 2021

103.599104.417104.052102.332101.290

Feb 2021

104.456105.389104.828103.408101.978

Mar 2021

105.398106.979105.862104.843102.064

Apr 2021

106.512108.657106.897105.650103.181

May 2021

107.541110.777107.500106.996104.557

Jun 2021

108.483111.749108.534108.072104.901

Jul 2021

109.532112.973109.543108.996106.063

Aug 2021

110.330113.887110.360109.641106.740

Sep 2021

110.639113.716110.711110.048107.269

Oct 2021

111.480114.644111.262111.408109.044

Nov 2021

112.331116.096111.962112.271109.996

Dec 2021

112.548116.431112.008112.168111.042

Jan 2022

114.031117.875113.262114.346113.212

Feb 2022

115.322119.163114.484116.048114.586

Mar 2022

117.687122.023116.866119.236115.960

Apr 2022

118.414123.824117.125121.090117.153

May 2022

119.489124.885118.190123.078117.801

Jun 2022

120.793125.305119.741124.975118.516

Jul 2022

120.223124.008119.038123.879119.514
Producer Price Indexes for government purchases, defense and nondefense
MonthGovernment purchases, defenseGovernment purchases, nondefense

Jan 2019

100.000100.000

Feb 2019

100.672100.380

Mar 2019

101.344101.045

Apr 2019

101.631101.614

May 2019

101.919101.804

Jun 2019

101.248101.519

Jul 2019

101.536101.994

Aug 2019

101.344101.899

Sep 2019

100.960101.614

Oct 2019

101.536101.425

Nov 2019

101.440101.235

Dec 2019

101.631101.330

Jan 2020

102.303101.709

Feb 2020

100.576100.950

Mar 2020

99.424100.285

Apr 2020

96.16198.291

May 2020

96.06598.575

Jun 2020

97.60199.240

Jul 2020

98.273100.285

Aug 2020

98.369100.855

Sep 2020

98.081100.760

Oct 2020

98.464101.045

Nov 2020

98.752101.330

Dec 2020

99.808101.899

Jan 2021

100.864103.134

Feb 2021

101.823104.179

Mar 2021

102.975105.793

Apr 2021

103.935106.458

May 2021

105.278107.882

Jun 2021

105.950109.117

Jul 2021

106.656110.179

Aug 2021

107.261110.821

Sep 2021

107.808111.176

Oct 2021

108.995112.638

Nov 2021

109.702113.563

Dec 2021

109.312113.556

Jan 2022

111.486115.760

Feb 2022

113.509117.346

Mar 2022

117.406120.283

Apr 2022

119.475122.065

May 2022

121.279124.189

Jun 2022

122.261126.568

Jul 2022

121.121125.379
Producer Price Indexes for distributive services
MonthFinal demand distributive services Intermediate demand distributive services

Jan 2019

100.000100.000

Feb 2019

100.256100.000

Mar 2019

100.683100.654

Apr 2019

101.537101.552

May 2019

101.281101.307

Jun 2019

101.281101.307

Jul 2019

101.281101.471

Aug 2019

102.050102.451

Sep 2019

101.366103.023

Oct 2019

102.391102.533

Nov 2019

100.939102.288

Dec 2019

101.110102.941

Jan 2020

101.366103.350

Feb 2020

101.025103.105

Mar 2020

101.964103.268

Apr 2020

102.818103.023

May 2020

102.135103.186

Jun 2020

101.964103.676

Jul 2020

102.818103.758

Aug 2020

102.904105.147

Sep 2020

102.733106.454

Oct 2020

104.611107.598

Nov 2020

103.672107.353

Dec 2020

102.989109.069

Jan 2021

103.672110.376

Feb 2021

104.526111.111

Mar 2021

105.124112.337

Apr 2021

107.515114.951

May 2021

108.881117.075

Jun 2021

109.564119.690

Jul 2021

111.207120.672

Aug 2021

113.412122.481

Sep 2021

113.825122.740

Oct 2021

115.137123.395

Nov 2021

116.470123.114

Dec 2021

117.774124.770

Jan 2022

119.292126.529

Feb 2022

121.753127.859

Mar 2022

125.159132.128

Apr 2022

126.081133.998

May 2022

126.610134.707

Jun 2022

126.983133.757

Jul 2022

127.362133.879

A Blueprint for Modernizing the Consumer Price Index

At BLS, we never stop improving. We highly value any input from our data users, technical advisors, and other experts that helps us improve our high quality economic statistics. On May 3, 2022, we welcomed the latest evaluation of one of our statistical programs from the National Academy of Sciences, Committee on National Statistics (CNSTAT): Modernizing the Consumer Price Index for the 21st Century.

As we first reported in December 2019, CNSTAT convened an expert panel to study the Consumer Price Index (CPI). In our update in August 2021, we shared the panel membership and described the public meetings where the panel gathered information for its report. Now that the panel has completed its report, we share our plans to address their recommendations:

  • Adopt an alternative data strategy that significantly expands the use of new data sources and collection methods.
  • Improve the timeliness and quality of market basket weights in the CPI.
  • Continue research to enhance and inform the public’s understanding of consumer price change for shelter and medical care.
  • Calculate income-based CPIs and address methodology limitations.
  • Collaborate across the federal and international statistical system.

Use Alternative Data Sources

The chapter on modernizing elementary indexes focused on alternative data sources. Using data from sources beyond traditional surveys is a theme throughout the report. The recommendation to develop a household scanner data program would be a long-term strategy to address many challenges of calculating an accurate and timely CPI. BLS agrees with the panel to seek new data sources to improve every aspect of price index calculation: prices, expenditures (including the quantities purchased), quality adjustment, modeling, estimation, and imputation. Doing so will enable BLS to improve and expand the data we produce and provide users the data they need when they need it.

Even before the CNSTAT report, BLS has been busy building a pipeline of alternative data sources and improving our estimation methods for the CPI. Increasing the focus on alternative data should generate a steady flow of new data sources. This focus also will improve our ability to collect data through a variety of methods and give us new opportunities to address quality change. Consistent with our values to provide accessible information, we will keep you informed about new data sources and methods through the BLS website.

Improve Timeliness and Quality of Market Basket Weights

Beginning in January 2023, BLS plans to update market basket weights in the CPI annually, using 1 year of data. This change will immediately improve the timeliness of the market basket. BLS will continue our efforts to collect and process data more quickly to calculate the CPI using the most recent spending information.

BLS uses several data sources to adjust data collected in the Consumer Expenditure Surveys to calculate the CPI market basket weights. BLS will analyze the feasibility of using business data from the Bureau of Economic Analysis to adjust for categories consumers are reluctant to report, such as alcohol and tobacco. BLS plans to research alternative data sources to improve expenditure estimates when information from respondents is missing or aggregated.

BLS continues to believe collecting data directly from consumers is important to achieve our measurement objective. We are conducting research on a Household Cost Index, which requires household-level expenditure estimates to calculate household-specific indexes. As the panel notes, indexes for specific populations also require linking expenditures with information about households. Given current resources, we do not plan to expand our use of data from other sources in the next few years to supplement data collected in the Consumer Expenditure Surveys. In the future, BLS could pursue a household scanner data program to address the concerns the panel raises regarding the Consumer Expenditure Surveys.

Modernize Shelter

BLS is exploring alternative data sources to supplement rents collected in our housing survey and improve imputation of rental equivalence estimates for owner-occupied housing . We will continue to produce research indexes that meet user needs. BLS plans to publish research on a rent index focused on new tenants. Future research will target alternatives to rent data as a proxy for rental equivalence in predominantly owner-occupied areas and alternatives to the rental equivalence approach for high-end properties.

All BLS consumer indexes currently use a rental equivalence approach to target a cost-of-living measurement objective. Research indexes based on occupancy (renter and owner) will provide users with more insight. Some users need indexes for certain populations. As already mentioned, BLS will continue to research a Household Cost Index that uses a payments approach for owner-occupied housing. Some of these research indexes may ultimately be “promoted” to official status.

Modernize Medical Care

BLS uses an indirect method to price health insurance because directly pricing health insurance premiums is difficult. We have confirmed the retained earnings data incorporate rebates and will pursue further improvements to the indirect approach. We are pursuing implementation of claims data for physician’s and hospital outpatient services and will monitor hospital price transparency data as a possible data source in the future. Research comparing the indirect and direct methods is well underway and will be published initially as a research paper.

Calculate Supplemental Population Price Indexes

BLS continues research on producing price indexes by income groups. While BLS recognizes the limited benefit of reweighting the market basket to create indexes for particular population groups, we believe indexes for renters and owners will provide more insight into measuring price change for shelter. BLS will continue to seek cost-effective methods to study household behaviors and seek resources to collect household scanner data linked with demographics.

Collaborate with Other Statistical Organizations

Another theme throughout the report is communication and collaboration among statistical agencies. The panel recommended expanding collaboration, especially in research and data sharing. As the complexity of data sources and methods increases, BLS also needs to communicate with stakeholders to maintain transparency. Our practice is to announce on the BLS website in advance any changes to our data sources or methods. We will continue to share research index results to document the impact of these changes. BLS is looking into new ways of sharing data and improving transparency.

We value our partnerships with other agencies in the federal and international statistical community. In June 2022, we will share the CNSTAT recommendations with the Federal Economic Statistics Advisory Committee and discuss our plans. We will continue to seek out new opportunities to connect and collaborate with colleagues in the government, academic, and private sectors as we improve our statistics. We also will ensure our staff has the skills to innovate the modern methods of the future. In the last few years, BLS developed an in-house Data Science Training Program designed to bring awareness and improve the skills of BLS staff in key areas of data science. This annual program introduces a new cohort of BLS staff to these concepts, with plans to scale for larger cohorts in the future and include more specialized learning streams.

It is an exciting time to produce economic statistics. Their importance is paramount, and the opportunities have expanded to improve their accuracy, relevance, and timeliness. The CNSTAT’s latest report on the CPI is a valuable guide to help us keep improving and continue to produce gold standard data well into the future.

Measuring Changes in Shelter Prices in the Consumer Price Index

Shelter costs are the largest regular expense for most households. That makes them a topic of considerable interest to users of Consumer Price Index (CPI) data. The U.S. city average for shelter increased 5.1 percent from April 2021 to April 2022. Its two main components, owners’ equivalent rent of residences and rent of primary residence, each increased 4.8 over the year. (Lodging away from home is the other component of shelter, and lodging prices rose 19.7 percent from April 2021 to April 2022.)

Consumer Price Index for All Urban Consumers, all items and shelter, January 2012 to April 2022

Editor’s note: Data for this chart are available in the table below.

Because of their large weight in the CPI market basket—nearly a third—the indexes for owners’ equivalent rent and rent can have a large impact on the overall inflation estimate. There is also a lot of misunderstanding about these shelter indexes, and so it is worth taking a few minutes to get a clear understanding of what they measure.

Owners’ equivalent rent is the larger of these two components, at nearly one quarter of the consumer market basket, or weight, in the CPI. It represents the implicit amount an owner of a housing unit would have to pay in rent to live in the unit, assuming it was leased instead of owned. The expenditure weight for owners’ equivalent rent in the CPI is based on a question in the Consumer Expenditure Survey. That question asks homeowners, “If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?” The role of this question can be easily misunderstood by even sophisticated users of BLS data. That has contributed to a common misconception: the mistaken belief that the price observations used for owners’ equivalent rent in the CPI are also from homeowner estimates of their home’s rental value. In fact, the sample of prices used in the owners’ equivalent rent index comes from observations of rent collected in our monthly survey of housing prices, but with utilities and other similar charges removed.

Why don’t we just measure changes in home values in the CPI? It’s because a home isn’t just a consumption item for the owner. It is also an investment, often the largest investment many people will make in their lives. The concept in the CPI—and in the economic statistics programs of most other nations—is to treat owned housing as a capital or investment good, distinct from the shelter service it provides. We treat spending to buy and improve houses and other housing units as investment and not consumption in the CPI. Mortgage interest costs, property taxes, real estate fees, most maintenance, and all improvement costs are part of the cost of the capital good and are also not treated as consumption items. These nonconsumption costs of owned housing are out of scope for the CPI under the cost-of-living framework that guides the index.

Some people have noted that the CPI index for rent (which represents just over 7 percent of the weight of the CPI) is not rising as fast as some other measures, notably those published by firms in the real estate industry. One reason for this is that over 80 percent of rental units in the CPI sample each period have tenants who continue to rent the same unit. Landlords often raise rents when a unit is vacated by a prior tenant and a new tenant moves in. In some cases, the rent paid by tenants with multi-year leases increases periodically—and automatically, by the CPI itself—through an escalation clause in the lease agreement that cites the CPI for this purpose.

Because rents for existing tenants change in line with the terms of leases and rental agreements, and many leases are for 12 months, existing tenants typically do not face price change within the 12-month period of the lease. This is called a “sticky” price. Because of this, the process used to calculate the indexes for rent and owners’ equivalent rent differs from the process used to calculate the rest of the CPI. Most prices are collected either monthly or every 2 months, but rent prices are collected every 6 months. In effect, this means price increases for shelter can sometimes take longer to appear in the CPI than in some other data sources.

We are always working to improve the accuracy of the CPI, and that includes our shelter indexes. We asked for expert opinion from the National Academy of Sciences, Committee on National Statistics, on better ways to measure price change for these important items. The committee recently published their report, “Modernizing the Consumer Price Index for the 21st Century.” The report endorsed the use of owners’ equivalent rent in the CPI and recommended that, “BLS should continue using rental equivalence as the primary approach to estimating the price of housing services for owner-occupied units.”

I will say more about the report from the Committee on National Statistics soon. In the meantime, we will consider all the recommendations of this distinguished group as we plan future improvements to the CPI.

You can read more about shelter in our factsheet for rent and owners’ equivalent rent. We also have more technical details in the Handbook of Methods.

Consumer Price Index for All Urban Consumers, all items and shelter, January 2012 to April 2022
MonthAll itemsShelterRent of primary residenceOwners’ equivalent rent of residences

Jan 2012

100.000100.000100.000100.000

Feb 2012

100.440100.205100.182100.102

Mar 2012

101.203100.472100.332100.295

Apr 2012

101.509100.638100.469100.465

May 2012

101.390100.799100.589100.560

Jun 2012

101.241100.999100.657100.660

Jul 2012

101.076101.179100.929100.837

Aug 2012

101.639101.350101.150101.097

Sep 2012

102.092101.511101.438101.322

Oct 2012

102.052101.737101.937101.538

Nov 2012

101.569101.804102.193101.735

Dec 2012

101.295101.922102.477101.880

Jan 2013

101.595102.213102.711102.077

Feb 2013

102.427102.481102.926102.249

Mar 2013

102.695102.720103.146102.384

Apr 2013

102.588102.848103.209102.542

May 2013

102.771103.097103.432102.701

Jun 2013

103.017103.340103.566102.888

Jul 2013

103.058103.554103.790103.045

Aug 2013

103.182103.779104.187103.355

Sep 2013

103.302103.905104.432103.570

Oct 2013

103.036104.053104.752103.839

Nov 2013

102.825104.285105.038104.149

Dec 2013

102.816104.509105.422104.415

Jan 2014

103.199104.852105.666104.646

Feb 2014

103.581105.113105.828104.815

Mar 2014

104.248105.512106.120105.056

Apr 2014

104.591105.696106.358105.227

May 2014

104.957106.036106.595105.411

Jun 2014

105.152106.252106.832105.604

Jul 2014

105.111106.567107.192105.844

Aug 2014

104.935106.787107.502106.124

Sep 2014

105.014106.979107.871106.380

Oct 2014

104.751107.224108.254106.667

Nov 2014

104.185107.399108.695106.969

Dec 2014

103.594107.543108.987107.140

Jan 2015

103.107107.932109.258107.403

Feb 2015

103.555108.247109.575107.632

Mar 2015

104.171108.628109.862107.885

Apr 2015

104.383108.871110.044108.142

May 2015

104.915109.101110.295108.353

Jun 2015

105.282109.454110.600108.720

Jul 2015

105.289109.886111.011109.018

Aug 2015

105.140110.096111.390109.325

Sep 2015

104.977110.379111.871109.664

Oct 2015

104.929110.648112.306109.963

Nov 2015

104.708110.818112.653110.260

Dec 2015

104.350111.000112.995110.509

Jan 2016

104.523111.434113.305110.795

Feb 2016

104.609111.802113.605111.031

Mar 2016

105.059112.101113.882111.250

Apr 2016

105.557112.353114.148111.546

May 2016

105.984112.781114.482111.890

Jun 2016

106.332113.231114.818112.249

Jul 2016

106.160113.510115.190112.574

Aug 2016

106.258113.834115.599112.942

Sep 2016

106.513114.165116.005113.367

Oct 2016

106.646114.543116.563113.751

Nov 2016

106.480114.757117.024114.167

Dec 2016

106.515115.016117.469114.458

Jan 2017

107.136115.389117.753114.717

Feb 2017

107.473115.736118.042114.947

Mar 2017

107.560115.972118.297115.127

Apr 2017

107.879116.233118.533115.321

May 2017

107.971116.546118.883115.529

Jun 2017

108.069116.916119.246115.874

Jul 2017

107.995117.102119.579116.186

Aug 2017

108.318117.589120.086116.629

Sep 2017

108.892117.859120.392116.974

Oct 2017

108.823118.253120.871117.386

Nov 2017

108.825118.386121.324117.733

Dec 2017

108.761118.701121.803118.092

Jan 2018

109.354119.071122.146118.391

Feb 2018

109.850119.356122.336118.563

Mar 2018

110.098119.826122.571118.878

Apr 2018

110.536120.167122.913119.194

May 2018

110.996120.638123.195119.468

Jun 2018

111.172120.877123.516119.779

Jul 2018

111.180121.219123.917120.129

Aug 2018

111.242121.574124.421120.514

Sep 2018

111.371121.734124.763120.799

Oct 2018

111.568122.001125.188121.205

Nov 2018

111.194122.224125.708121.633

Dec 2018

110.839122.500126.037121.899

Jan 2019

111.050122.911126.340122.185

Feb 2019

111.520123.376126.633122.505

Mar 2019

112.149123.869127.084122.830

Apr 2019

112.743124.313127.536123.190

May 2019

112.983124.676127.790123.463

Jun 2019

113.005125.113128.300123.861

Jul 2019

113.194125.442128.672124.179

Aug 2019

113.188125.656129.073124.542

Sep 2019

113.277126.005129.537124.906

Oct 2019

113.536126.081129.865125.222

Nov 2019

113.475126.280130.307125.597

Dec 2019

113.372126.476130.683125.894

Jan 2020

113.812126.982131.085126.273

Feb 2020

114.123127.454131.392126.523

Mar 2020

113.875127.596131.743126.785

Apr 2020

113.114127.559131.982126.973

May 2020

113.116127.851132.244127.237

Jun 2020

113.735128.067132.431127.379

Jul 2020

114.310128.368132.686127.653

Aug 2020

114.671128.532132.878127.889

Sep 2020

114.830128.579133.058128.015

Oct 2020

114.878128.640133.332128.347

Nov 2020

114.808128.704133.496128.454

Dec 2020

114.916128.809133.658128.625

Jan 2021

115.405129.037133.775128.810

Feb 2021

116.036129.321133.963129.090

Mar 2021

116.858129.760134.148129.337

Apr 2021

117.819130.245134.361129.564

May 2021

118.763130.677134.652129.920

Jun 2021

119.867131.372134.969130.363

Jul 2021

120.443131.997135.215130.757

Aug 2021

120.692132.182135.697131.151

Sep 2021

121.020132.641136.296131.721

Oct 2021

122.025133.121136.932132.368

Nov 2021

122.625133.642137.566132.989

Dec 2021

123.002134.131138.111133.505

Jan 2022

124.037134.667138.812134.075

Feb 2022

125.170135.454139.545134.649

Mar 2022

126.841136.244140.110135.204

Apr 2022

127.549136.941140.835135.764

Measuring Consumer Prices for New Vehicles More Accurately

In May 2020, we announced a new research index designed to improve the way we measure consumer price changes for new vehicles. Part of the Consumer Price Index (CPI) program, the research index uses transaction data on vehicle purchases. BLS obtains the data from J.D. Power. After carefully studying the data, we now plan to implement the new data source into the official CPI, effective with the April 2022 release, which we will publish on May 11, 2022. We also will update Measuring Price Change in the CPI: New vehicles factsheet at that time.

The research index departs from the traditional survey methods and data sources we have used in the CPI. The traditional methods sample vehicle dealers and the makes, models, and features of the vehicles they sell. That’s why we took a very deliberate approach before we incorporated the new data into the official CPI. We discuss details of our methods and research in “A New Vehicles Transaction Price Index: Offsetting the Effects of Price Discrimination and Product Cycle Bias with a Year-Over-Year Index.”

Leading up to March 2020, the movements of the research index for new vehicles were similar to the official index for new vehicles. Since March 2020, the indexes began to diverge, with the research index showing faster price increases that the official index did not reflect.

Editor’s note: Data for this chart are available in the table below.

Since the COVID-19 pandemic began in March 2020, the economy experienced significant disruption. This is especially true of the automobile industry, which saw fluctuations in the supply chain, employment, and consumer demand. Combining these shifts in the economy with the changes in methods represented by the research index made it challenging to assess our results.

While the research index and the official new vehicles index maintained similar trends, the recent divergence between them provided an opportunity to assess the robustness of the two approaches.

We weighed several factors in deciding whether to incorporate the J.D Power data on new vehicle prices into the official CPI. The new data include records of the prices paid during hundreds of thousands of transactions each month. That dwarfs the roughly 500 prices collected using traditional CPI methods. The larger dataset allows us to estimate price changes more precisely. As a result, the research index has a much lower standard error than the official new vehicles index.

Editor’s note: Data for this chart are available in the table below.

Because the research data reflect actual transactions, the shift in consumer preference from cars to other types of vehicles, such as trucks, is built into the data. This differs from the official index, which has maintained a roughly equal weight between cars and trucks.

In addition to the quantitative evaluations, BLS continued to ask for feedback on the research index through our website and by consulting with other statistical agencies. We received positive feedback and no major concerns, and we remain confident the research index is statistically sound. For these reasons, we have decided to incorporate the new data source and methods for new vehicles in the official CPI.

In some ways, the past 2 years have been an unprecedented time for statistical measurement, but in other ways business at BLS has continued as usual. When the COVID-19 pandemic began in March 2020, BLS ceased in-person data collection for the CPI and other programs. We collected more data online, by telephone, and through video. While the pandemic affected data collection, we continue to publish data on schedule. We also continue to assess our methods and seek ways to improve the quality of our data. Improving our methods for collecting price data for new vehicles is another step forward in innovating and improving the CPI.

Comparison of research index and official index for new vehicle prices
MonthResearch indexOfficial index

Jan 2018

100.000100.000

Feb 2018

100.02699.871

Mar 2018

99.48199.817

Apr 2018

99.84299.370

May 2018

99.58399.560

Jun 2018

99.49999.705

Jul 2018

99.84199.681

Aug 2018

99.94299.424

Sep 2018

99.91799.129

Oct 2018

99.88699.043

Nov 2018

100.07299.204

Dec 2018

99.47199.409

Jan 2019

99.969100.043

Feb 2019

100.295100.157

Mar 2019

100.182100.539

Apr 2019

100.487100.574

May 2019

100.659100.452

Jun 2019

100.362100.287

Jul 2019

100.484100.027

Aug 2019

100.08999.633

Sep 2019

100.20399.224

Oct 2019

100.44399.136

Nov 2019

99.96599.138

Dec 2019

99.91299.472

Jan 2020

100.486100.175

Feb 2020

100.843100.549

Mar 2020

101.301100.087

Apr 2020

102.431100.008

May 2020

102.842100.154

Jun 2020

103.653100.076

Jul 2020

104.047100.549

Aug 2020

104.142100.284

Sep 2020

103.895100.249

Oct 2020

103.841100.653

Nov 2020

103.977100.726

Dec 2020

103.781101.425

Jan 2021

104.818101.620

Feb 2021

105.652101.714

Mar 2021

106.308101.582

Apr 2021

107.156101.971

May 2021

111.189103.502

Jun 2021

113.656105.341

Jul 2021

114.795106.944

Aug 2021

115.205107.930

Sep 2021

115.942109.013

Oct 2021

118.107110.566

Nov 2021

118.980111.915

Dec 2021

120.336113.373

Jan 2022

121.230114.005

Feb 2022

122.481114.308
Comparison of 12-month standard errors for the research index and official index for new vehicle prices
MonthResearch indexOfficial index

Jan 2019

0.110.61

Feb 2019

0.100.58

Mar 2019

0.110.42

Apr 2019

0.110.43

May 2019

0.130.43

Jun 2019

0.120.46

Jul 2019

0.140.43

Aug 2019

0.130.42

Sep 2019

0.140.42

Oct 2019

0.120.45

Nov 2019

0.160.46

Dec 2019

0.160.40

Jan 2020

0.150.41

Feb 2020

0.150.38

Mar 2020

0.140.33

Apr 2020

0.150.36

May 2020

0.130.40

Jun 2020

0.130.36

Jul 2020

0.150.48

Aug 2020

0.140.47

Sep 2020

0.130.60

Oct 2020

0.130.64

Nov 2020

0.120.65

Dec 2020

0.120.62

Jan 2021

0.130.62

Feb 2021

0.130.66

Mar 2021

0.120.67

Apr 2021

0.110.60

May 2021

0.120.59

Jun 2021

0.130.63

Jul 2021

0.130.61

Aug 2021

0.130.69

Sep 2021

0.130.74

Oct 2021

0.160.72

Nov 2021

0.150.55

Dec 2021

0.150.69

Jan 2022

0.140.67

Feb 2022

0.140.68

What’s Going on with the Large Revisions in Seasonally Adjusted Employment Estimates?

When BLS released The Employment Situation for January 2022 on February 4, we reported larger-than-normal revisions to the seasonally adjusted monthly nonfarm employment estimates from the Current Employment Statistics survey. Let’s take a closer look at the seasonal adjustment process and explain why the revisions were unusually large.

BLS seasonally adjusts data to account for recurring seasonal patterns in a time series. It allows our data users to analyze the underlying trends without the influence of seasonal movements. Seasonal movements might entail hiring extra retail trade workers around the December holidays. Seasonal movements also may entail workers leaving payrolls, such as school bus drivers when the schoolyear ends. These types of movements happen around the same time every year and affect about the same number of workers each year. Because the seasonal movements in nonfarm employment typically don’t change by much year to year, revisions to over-the-month changes from seasonal adjustment are typically small. However, data users may have a hard time seeing seasonal patterns when large events like strikes, hurricanes, or recessions occur. The unprecedented changes brought on by the COVID-19 pandemic make it even more difficult to determine the seasonal patterns.

Every year, during the employment benchmarking process, BLS staff takes a more comprehensive look back at the seasonal adjustment of time series to account for changes in seasonal patterns of nonfarm employment. Months in which large, irregular events occur are treated as outliers, which means that the abnormal change will not be treated as a new seasonal pattern. When we seasonally adjusted the data for 2020 after 2020 had ended, we treated several months individually as outliers. That works well for short-lived events, like strikes or a weather-related event. At that time, with very few observations after the start of the pandemic, that type of individual outlier detection was appropriate.

For example, the economy lost a massive number of jobs in March and April 2020 at the start of the pandemic. If we had not treated those months as outliers, or irregular events, the seasonal adjustment model would have expected massive employment losses as a new seasonal pattern for March and April 2021. That would not have been appropriate because those huge losses in March and April 2020 were not likely to recur. If the model had included the large March and April 2020 employment losses, when large job losses did not occur in March and April 2021, the model would have shown large job gains for those months on a seasonally adjusted basis. That would have distorted the underlying employment trend.

We now have more data observations after the start of the pandemic, and we benefit from the hindsight of our normal annual review of the data, which encompasses the last 5 years of data. We incorporated two additional outlier types that better isolated the initial losses in employment from the pandemic, while simultaneously accounting for new seasonal patterns that we have detected. One new type of outlier accounts for a temporary change in the level of the series. For example, drinking places that serve alcoholic beverages experienced a significant employment decline in April 2020 but have gained jobs since then, even if in fits and starts. In this case, we treated April 2020 as a temporary change outlier, which will adjust the months from April 2020 forward to account for both the large drop and subsequent recovery in employment this industry has experienced. This will better account for this temporary period of readjustment as employment levels recover towards what they were before the pandemic.

Employment in drinking places, alcoholic beverages, January 2020 to December 2021, seasonally adjusted

Editor’s note: Data for this chart are available in the table below.

The other new type of outlier accounts for a permanent shift in the level of a series. An example of this is nursing care facilities, which continued to experience employment declines after April 2020. In this case, April 2020 was treated as a level shift outlier, which resulted in subsequent months being adjusted to better reflect the continued lower level of employment.

Employment in nursing care facilities, January 2020 to December 2021, seasonally adjusted

Editor’s note: Data for this chart are available in the table below.

Incorporating these additional types of outliers stabilized the normal seasonal patterns and smoothed out the initial large swings in the seasonally adjusted data. These changes resulted in a more accurate payroll employment series and will allow users to better differentiate longer-term trends from seasonal movements.

These changes resulted in some large revisions to our monthly seasonally adjusted data for 2021, although the revisions partly offset each other. For example, total nonfarm employment in November and December 2021 combined is 709,000 higher than previously reported, while employment in June and July 2021 combined is 807,000 lower. The change for all of 2021 is just 217,000 higher than previously reported. Although these revisions are larger than usual, trends in the data emerge more clearly.

Want to really dig into the details of seasonal adjustment for the nonfarm employment data? See our specification files and technical documentation.

Employment in drinking places, alcoholic beverages, January 2020 to December 2021, seasonally adjusted
MonthPreviously publishedRevised

Jan 2020

417,200419,800

Feb 2020

423,700426,400

Mar 2020

381,800383,600

Apr 2020

85,60084,800

May 2020

145,400140,600

Jun 2020

233,800225,300

Jul 2020

238,700226,200

Aug 2020

250,800243,800

Sep 2020

266,200260,000

Oct 2020

289,700278,500

Nov 2020

280,900281,300

Dec 2020

243,000239,700

Jan 2021

248,300248,700

Feb 2021

282,200280,700

Mar 2021

303,900298,300

Apr 2021

311,700309,900

May 2021

318,300322,900

Jun 2021

322,000331,200

Jul 2021

329,700336,800

Aug 2021

329,500342,900

Sep 2021

343,900350,600

Oct 2021

345,500354,200

Nov 2021

347,900365,000

Dec 2021

344,500363,800
Employment in nursing care facilities, January 2020 to December 2021, seasonally adjusted
MonthPreviously publishedRevised

Jan 2020

1,586,5001,585,100

Feb 2020

1,585,7001,584,800

Mar 2020

1,581,6001,581,200

Apr 2020

1,534,3001,537,200

May 2020

1,502,6001,508,000

Jun 2020

1,487,3001,490,000

Jul 2020

1,469,7001,470,600

Aug 2020

1,459,2001,463,300

Sep 2020

1,456,6001,454,500

Oct 2020

1,451,5001,447,800

Nov 2020

1,439,0001,437,200

Dec 2020

1,433,4001,430,900

Jan 2021

1,415,0001,416,700

Feb 2021

1,404,2001,406,300

Mar 2021

1,400,5001,403,600

Apr 2021

1,382,2001,388,700

May 2021

1,375,0001,383,700

Jun 2021

1,371,1001,376,300

Jul 2021

1,371,0001,373,700

Aug 2021

1,365,3001,367,700

Sep 2021

1,349,8001,346,700

Oct 2021

1,359,1001,349,200

Nov 2021

1,350,9001,346,000

Dec 2021

1,345,7001,345,100