The U.S. Bureau of Labor Statistics: Consumer and Producer Price Indexes

How they are constructed and how they are used in contract escalation

The U.S. Bureau of Labor

Statistics:

Consumer and Producer Price

Indexes

1

Annual MeetingApril 7 - 10, 2013Orlando, Florida

U.S. Bureau of Labor Statistics: Consumer and Producer Price

Indexes

How they are constructed and how they are used in contract escalation

Michael W. Horrigan, Ph.D.Associate Commissioner

Office of Prices and Living Conditions

Annual MeetingApril 7 - 10, 2013Orlando, Florida 2

Outline

Overview of the BLS

Consumer Price Index for All Urban Consumers (CPI-U)

Producer Price Index

Using Price Indexes to Escalate Contracts

CPI-U, CPI-W and Social Security, and the CPI-E

Chained CPI-U

3

Overview of the BLS

The U.S. Bureau of Labor Statistics (BLS) collects, analyzes, and publishes a vast array of data on our economy and our society.

We provide critical information on inflation, employment, unemployment, and productivity that are used by government policy makers, the media, workers, business leaders, consumers, and job seekers.

4

BLS Programs

Consumer Price Index

Employment Cost Index

The Employment Situation

Civilian Unemployment Rate

Nonagricultural Payroll Employment

5

Producer Price Indexes

Productivity and Costs

Real Earnings

U.S. Import and Export Price Indexes

Principal Federal Economic Indicators produced by the BLS:

Overview of the Consumer Price Index (CPI) Program

Goal of the CPI Scope / Coverage Classification System Weighting Sampling Data Collection Estimation Publication

3

Goal and Scope of the CPI

The goal of the CPI is to approximate a cost of living index.

Cost of living is a theoretical concept. The CPI seeks to measure the change in the cost of living by measuring the change in prices that consumers pay for a market basket of goods and services.

The CPI reflects prices paid by consumers in urban areas of the U.S. for a market basket of goods and services.

Any item purchased for consumption, goods or services, is potentially eligible for pricing.

7

Classification system

In the CPI, the consumer market basket is a categorization of goods and services with corresponding weights. In the U.S. there are 211 categories of items (goods and services) that are aggregated into eight major groups:

Food And Beverages

Housing Apparel Transportation Medical Care Recreation Education and

Communication Other Goods and

Services

8

Weighting

The Consumer Expenditure Survey (CE)—source of weighting for CPI--consists of two separate surveys:

Quarterly interview survey—asks consumers about major purchases over 5 consecutive quarters

Diary survey—asks consumers to keep diary of frequently purchased items over two weeks

Used to create expenditure weights to construct the CPI market basket.

Weights are updated every two years. Current CPI weights are based on consumer expenditures in 2009-2010.

9

Sampling / Geography

The CPI reflects prices paid by consumers in urban areas of the U.S.

Based on Census and OMB definitions, 87 geographic areas are selected to represent the urban population.

Referred to as Primary Sampling Units (PSUs)

Represent 38 distinct geographical units: 31 large cities plus a combination of the remaining 56 smaller areas into 7 geographical units

10

Data collection / Outlets

The geographic sample and market basket form the component cells that are used to build the CPI.

Item categories x Geography Units =

211 item categories x 38 geography units = 8018 cells Indexes calculated for each cell

The Telephone Point-of-Purchase Survey (TPOPS) of households is used to create the frame of outlets (stores, medical offices, web sites) for the collection of data in each of the 87 PSUs.

12

Data Collection / Outlets

BLS Field Representatives visit outlets and use Computer Assisted Data Collection (CADC) to select and price items using probability sampling.

Items are described completely in terms of price determining characteristics using the CPI Checklist.

Sample rotation allows the sample of specific items in the CPI to stay up-to-date.

Each outlet/item sample is replaced every four years, 1/8th of the sample every 6 months.

13


Based on the TPOPS survey, the commodity product or service line(s) (e.g., apples) to be priced in each outlet is known in advance of visiting the outlet.

During the initiation interview at an outlet, a process known as disaggregation is performed to determine the exact items to be priced within these product or service lines.

The exact items are chosen using statistical methods that give individual items a chance to be selected proportional to their sales for that particular product or service line at the selected outlet.

14


CPI price data are collected throughout the entire month. The month is divided into three pricing periods, with field representatives required to collect data during each period.

The price sought in the CPI is the retail, transaction price paid by the consumer, including sales and excise taxes.

Each month a field staff of about 350 part-time economic assistants and 100 full-time economists collects prices for over 83,000 individual items based on personal visits to more than 23,000 outlets in 87 cities.

15

Data Collection / Housing

A survey of 32,000 housing rental units tracks the rate of inflation in housing services.

Provides data for measuring changes in shelter costs for consumers who rent (about 32 percent of all consumers).

Also provides data for measuring changes in shelter

costs for consumers who own their own homes (about 68 percent of all consumers).

This latter index is estimated based on the concept of rental equivalence or the market rent that would be charged for these owner occupied dwellings if they were rented.

16

Estimation

The CPI-U is constructed in two stages. In Stage 1 indexes for each item/area cell are

constructed from a sample of prices. – Types of steak in Chicago

In Stage 2 those indexes are aggregated across item/areas.

– Steaks in Chicago, Hamburger in Chicago, Steaks in LA, etc.

The CPI uses either a geometric mean or a Laspeyres formula for calculating inflation in stage 1 and a Laspeyres formula for indexes in stage 2.

17

Estimation

Consumers

assumed to

substitute among types of

steak when prices

change

No substituti

on is assumed between

hamburger and steak

Estimation

The geometric mean formula implicitly assumes a degree of substitution among the items used in calculating a price index.

The Laspeyres formula does not allow any such substitution and by construction is an upper bound to the change in the cost of maintaining a standard of living.

Substitution among different types of steak in stage 1 Does not assume substitution of hamburger for steak in

stage 2 Note: The Laspeyres formula is used in Stage 1 for certain

items such as surgery (consumers do not substitute among different kinds of surgery)

19

Estimation/Publication

Quality adjustment

Direct comparison, Imputation, Hedonic regression

Seasonal Adjustment

Publication

CPI-U All Items U.S. City Average CPI for urban consumers

CPI breakdowns for food, energy, and all items less food and energy often receive attention.

CPI-W, CPI-E, and Chained CPI-U

20

PublicationPublication

Tools for Dissemination of CPI Data CPI news release available to public

at 8:30 a.m. on release day Press Release Detailed Report: Electronically

Available Telephone 202 691 7000 Available online @ (www.bls.gov/cpi

) Email requests: [email protected]

http://www.bls.gov/cpi

Outline

Overview of the BLS





Chained CPI-U

22

Goal of the PPI

The goal of the PPI is to measure the change in revenue received by a producer for the output produced using a fixed stock of labor, capital, and technology.

What does the PPI measure?

PPI measures changes in net revenues received by producers of goods or services for a specific item.

Price-determining variables could include: Kind of buyer Product or service Transaction terms Time of purchase

Sales promotion techniques such as rebates or financing plans are reflected, as they affect the net proceeds of the producer.

Changes in excise taxes are not reflected.24

PPI Coverage PPI data include the output of all industries in the goods-

producing sectors of the U.S. economy such as mining, manufacturing, agriculture, and construction, as well as goods competitive with those made in the producing sectors, such as waste and scrap materials.

PPI is continually expanding coverage of service industries. The program covers a majority of the service sector's output, publishing data for selected industries in various industry sectors including:

25

wholesale and retail trade transportation and warehousing information finance and insurance real estate brokering

rental, and leasing

professional, scientific, and technical services

administrative, support, and waste management services

health care and social assistance

accommodation

PPI Coverage

26

Actual coverage of in-scope marketed output of the U.S. economy by PPI:

Sector ’07 GDP Share PPI Coverage (FY2012)

GoodsServicesTotal

27.3 % 90.0 % 72.7 % 78.2 % 100.0 % 81.8 %

PPI publication structures

Three main PPI publication structures:

Industry-based (NAICS)– Over 600 industry price indexes and over 4,900 specific

product line and product category sub-indexes

Commodity-based (PPI)– Over 4,700 commodity price indexes organized by type

of product and end use

Stage-of-processing based– Aggregate price indexes organized by commodity-

based processing stage—Crude Materials for Further Processing; Intermediate Materials, Supplies, and Components; and Finished Goods

27

Producer Price Indexes are subject to change after being published: After an index is first published, it is subject to

recalculation to take into account late survey reports and corrections by respondents.

Every index is recalculated on a systematic basis—four index months after being first published. Previously published seasonally adjusted indexes are also subject to change in January when new seasonal factors are calculated and applied to the most recent five years of data.

PPI data revision

28

Publication

Over 1,000 indexes for specific outputs of industries in the services sector and other sectors that do not produce physical products;

Several major aggregate measures of price change organized by commodity-based, industry-based and stage of processing.

PublicationPublication

Tools for Dissemination of PPI Data PPI news release available to

public at 8:30 a.m. on release day Press Release Detailed Report: Electronically

Available Telephone (202) 691-7705 Available online @ (

www.bls.gov/ppi)

http://www.bls.gov/ppi

Using the Producer Price and Consumer Price Indexes for

Contract Escalation

The Producer Price Index (PPI) measures the average change over time in the selling prices received by domestic producers of goods and services www.bls.gov/ppi/

The Consumer Price Index (CPI) measures the average change over time in the prices paid by urban consumers for a defined market basket of consumer goods and services www.bls.gov/cpi/

31

http://www.bls.gov/ppi/

http://www.bls.gov/cpi/

Hundreds of billions of dollars in contract values are tied to Producer Price Indexes (PPI) through price escalation clauses, which are common in both government and private sector contracts.

These price escalation clauses are used to protect both the buyer and the seller from unanticipated surges or drops in prices.

PPI for escalation

32

Private firms use PPI data to compare changes in material costs they incur against changes in the PPI for the material in question. Private firms also compare changes in the prices they charge with changes in the PPI

PPI for escalation

33

Generally, an index should be chosen that represents the costs for providing a particular product or service, rather than indexes for the products or services themselves.

For example, if an apparel manufacturer is contracting for long-term purchases with a producer of finished fabrics, it would be more advisable to tie the escalation clause to a PPI for synthetic fibers rather than to a PPI for a type of finished fabric.

PPI for escalation

34

While detailed indexes may target costs more specifically, such indexes are more likely to be discontinued by BLS, or to have occasional gaps in availability.

Contracts should provide for these contingencies, and may minimize them if they cite only the higher-level categories.

In addition, because of the unavailability of certain indexes, proxies must sometimes be chosen to estimate price movements for some series.

PPI for escalation

35

General guidelines (similar to those for CPI escalation):

ESTABLISH the base selling price subject to escalation.

SELECT an appropriate index or indexes.

IDENTIFY clearly the selected index and cite appropriate source.

SPECIFY that the selected index will not be seasonally adjusted.

PPI guidelines for escalation

36

General guidelines (continued):

STATE the frequency of price adjustment.

PROVIDE for missing or discontinued data.

SPECIFY that calculations of price adjustments shall always use the latest version of the PPI data published.

AVOID locking indexes used for escalation into any particular reference base periods.

DEFINE the mechanics of price adjustment.

PPI guidelines for escalation

37

Calculating a contract escalation using the PPI

Escalation agreements using the PPI usually involve changing the contract’s base payment by the percent change in the level of the PPI between the reference period and a subsequent time period.

Again, this is calculated by first determining the index point change between the two periods and then the percent change.

Let’s look at the percent change from January 2011 to 2012…

Series Id:Industry:Product:Base Date:

Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

2007 111.0 111.5 111.7 112.0 112.1 112.0 112.1 112.3 112.5 112.7 113.0 113.12008 113.8 114.3 114.6 115.1 115.8 116.4 117.9 118.3 118.8 119.4 119.9 120.02009 120.5 120.4 120.4 120.3 120.2 120.2 120.3 120.2 120.3 120.1 120.2 120.32010 120.2 120.2 120.2 120.4 120.4 120.3 120.5 120.6 120.8 120.8 120.9 121.12011 121.7 122.0 122.4 122.9 123.2 123.5 123.8 123.9 124.2 124.3 124.5 124.62012 125.2

December 2003=100

*All indexes are subject to revision four months after original publication.

Producer Price Index Industry Data, Original Data ValuePCU333---333---Machinery manufacturingMachinery manufacturing

Computation of percent change:

PPI for Jan 2012 125.2

Less PPI for Jan 2011 121.7

Equals index point change 3.5

Divided by previous period PPI 3.5 / 121.7

Equals 0.029

Results multiplied by 100 0.029 X 100

Equals percent change 2.9

Escalation example: Calculating a percent

change

* PPI for machinery manufacturing for January 2011 and 2012In this example, first-published (preliminary) indexes are used

39

Percent change formula

PPI for current − PPI for previous

period period

X 100 PPI for previous period

40

41

Escalation example: Applying a percent

changeIf a 1-year contract has a base price of $15,000

and begins December 2010:

Base price 15,000

Dec. 2011 percent change divided by 100 2.9 / 100

Equals 0.029

Multiply base price by percent change 15,000 X .029

Equals

435

Results plus base price 15,000 + 435

Equals adjusted price

15,435

* PPI percent change for machinery manufacturing for December 2011In this example, first-published (preliminary) indexes are used

New PPI series retrieval tool

Go to PPI home page www.bls.gov/ppi Look for ‘Guide to Retrieving PPI data’ Find the series code you want

– Click on ‘PPI Industry Data’ tab or ‘PPI Commodity Data’ tab

– Click on ‘text’ or ‘excel’ to get the list of available indexes

– Find your index (there are thousands!)– Gulp / write it down!

Under ‘Get PPI Data’ click on:– ‘Click here for Series Report (opens in a new window)’– Enter your series ID– Choose your options and get your data!

42

http://www.bls.gov/ppi

Using the CPI for Contract Escalation

Advice is similar to that of using PPI for contract escalation Use Not seasonally adjusted data Generally CPI-U for all items is used CPI-U for published item strata do not go in and out

of publication

43

Outline

Overview of the BLS





Chained CPI-U

44

The CPI-U, CPI-W, and CPI-E

The CPI-U, CPI-W, and the CPI-E, which are defined precisely in the next few slides, all use the same price index formulas. Substitution allowed in stage 1 No substitution allowed in stage 2

The CPI-U, CPI-W, and the CPI-E differ in terms of consumer expenditure weights and the target population groups associated with these weights. U – all urban consumers W – urban consumers in wage and clerical households E – urban consumers in elderly households

45

CPI-U and the CPI-W

The CPI for all urban consumers, CPI-U, represents the spending habits of about 88 percent of the population of the U.S.

The CPI for Urban Wage Earners and Clerical Workers, CPI-W, represents the spending habits of about 28 percent of the population of the U.S.

More than 50 percent of total income must come from earnings in wage or clerical occupations.

46

Social Security

Social Security benefits are revised annually based on changes in the CPI-W. In 2013, about 58 million Americans received about

$821 billion in Social Security payments.

Why does the Social Security Administration use the CPI-W to adjust Social Security payments?

At the time when Social Security payments were first indexed using the CPI in 1975, the CPI-U did not exist as an index.

47

Social Security

When the CPI-U was introduced in 1978, it replaced the then current CPI with one based on a representation of all Urban Consumers instead of Urban Wage Earners and Clerical Workers.

The CPI-W is essentially the index that is a continuation of the CPI prior to 1978, and given its use since 1975, the decision was made by the Social Security Administration to continue the use of the CPI-W as the basis for adjusting Social Security benefits.

48

Social Security Cost of Living Adjustments, 2008-2012

Year % change from previous peak

Official COLA

2008 5.8% 5.8%

2009 -2.1% 0.0%

2010 -0.6% 0.0%

2011 3.6% 3.6%

2012 1.7% 1.7%

49

CPI-E

The experimental CPI for the Elderly, CPI-E, representing Americans ages 62 and older, was developed in 1987 at the request of Congress. The series was reconstructed back to December 1982.

For the experimental CPI-E, we calculate weights for a subset of households where the reference person or their spouse is age 62 or older.

Coverage is about 16 percent of the population.

50

CPI Relative Importances, December 2012 CPI-U CPI-W CPI-E

All items 100.0 100.0 100.0Food and beverages 15.3 15.9 13.1Food at home 8.6 9.4 8.0Food away from home 5.7 5.7 4.3Alcoholic beverages 0.9 0.9 0.7

Housing 41.0 39.9 45.5Shelter 31.7 30.6 35.2Rent of primary residence 6.5 9.0 3.9Owners' equivalent rent 24.0 20.9 30.1

Fuel oil 0.2 0.2 0.4Apparel 3.6 3.6 2.4Transportation 16.8 19.0 14.8Motor Fuel 5.5 7.0 4.2

Medical Care 7.2 5.8 11.7Medical Care Commodities 1.7 1.3 3.1Medical Care Services 5.4 4.4 8.6

Recreation 6.0 5.5 5.4Education and Communication 6.8 6.8 3.8College Tuition and Fees 1.7 1.4 0.4

Other Goods and Services 3.4 3.5 3.3Tobacco & smoking products 0.8 1.2 0.6

1123

Limitations of the CPI-E

Scope Includes spending by non-Social Security recipients.

Does not include surviving spouses less than 62 or their minor children receiving Social Security.

Does not include elderly living with families (e.g., their children) where the reference person and spouse are younger than 62.

Geography Uses geographic sample for the CPI-U, not designed

specifically to reflect where older Americans live.

52


Weighting CPI-U based on 76,000 interviews.

CPI-W based on 22,500 interviews.

CPI-E based on 19,400 interviews.

Both the CPI-W and CPI-E have higher variances than the CPI-U.

Retail Outlets Uses the retail outlet frame for the CPI-U; the sample

is not designed to reflect where older Americans shop.

53


Collecting the right item and the right price Uses the prices collected for items selected to

represent the purchasing patterns of the CPI-U.

The items and prices may not be representative of the purchasing patterns of older urban consumers.

For any item category (apples), the retail stores cannot reliably report the revenue they receive from sales of different apple types to older urban consumers.

Senior citizen discount rates are far more prevalent for the older American population than the urban population as a whole.

54

What is the C-CPI-U?

The Chained Consumer Price Index for All Urban Consumers (C-CPI-U).

First produced in 2002, data back to January 2000 (December 1999=100).

The final C-CPI-U is designed to be a closer approximation to a cost-of-living index in that it reflects changes in consumer spending patterns across CPI item categories every month.

Differs from the CPI-U in both weighting and formula.

It is released monthly with the CPI-U and CPI-W, but is subject to two annual revisions before becoming final.

55

The Chained Consumer Price Index (C-CPI-U)

Updated data from the Consumer Expenditure Survey are used to create weights that reflect consumer spending each month, allowing a chained index that has weights that reflect consumer spending in the current month and in the previous month.

These final expenditure data are available only with a 1-2 year time lag, so the index must be estimated and subsequently revised.

56

Chaining in the C-CPI-U

Each monthly change in the C-CPI-U uses prices and spending weights from the current and previous month.

In contrast, the CPI-U and CPI-W use spending weights for a base period, currently 2009-2010.

In those indexes, price levels in the current month are compared to the levels in the base period.

That base period is only updated every two years.

57

Chaining in the C-CPI-U

The C-CPI-U is referred to as “chained” because BLS computes the aggregate final index by linking together month-to-month changes.

For example, the percentage change in the C-CPI-U between January and March is the percentage change from January to February multiplied (chained) by the percentage change from February to March.

58

Substitution in the C-CPI-U

The C-CPI-U, CPI-U, and CPI-W each allow for substitution within most item categories.

Example: when Gala apples go on sale, some consumers switch from Fuji apples to Gala within the Apples item category.

The C-CPI-U also captures substitution across item categories.

Example: when the relative price of apples increases, consumers shift some apple purchases to other categories of fruit or food.

59

How the C-CPI-Umeasures substitution

The amount of substitution reflected in the C-CPI-U depends on the observed changes in consumer purchasing patterns when relative prices change.

The BLS does not specify which item categories are substitutes; all of the 211 item categories are treated as potential substitutes for one another.

Historically, the key contributors to the difference between the CPI-U and C-CPI-U have been in energy and high-tech areas such as gasoline and televisions.

60

Preliminary C-CPI-U Estimates

How does BLS construct initial estimates of the C-CPI-U given that the expenditure weights are only available with a lag?

A geometric mean formula is used to estimate the initial C-CPI-U.

Revisions to initial C-CPI-U indexes have often been significant.

Current research is focused on reducing the difference between the initial estimate and the final index.

61

The C-CPI-U: Publication

Published C-CPI-U values begin with January 2000 (December 1999=100).

Final indexes are available through 2011; the 2012 indexes will become final in February 2014.

Unlike the CPI-U and CPI-W, C-CPI-U series are not seasonally adjusted.

62

Components of the C-CPI-U

The C-CPI-U is constructed using the same set of 211 items x 38 areas basic component indexes as the CPI-U and the CPI-W.

BLS publishes fewer C-CPI-U indexes than for other two index series, however.

27 lower-level C-CPI-U series are published in addition to the All-items, U.S. City Average.

Only U.S.-level C-CPI-U series is published.

The set of published C-CPI-U series is shown on the next slide.

63

Percentage change Dec 1999 to Dec 2012 for CPI-E (38.0%), CPI-W (36.8%),

CPI-U (36.4%), and C-CPI-U (31.6%), December 1999=100

65

The U.S. Bureau of Labor Statistics: Consumer and Producer Price Indexes

Documents

Transcript of The U.S. Bureau of Labor Statistics: Consumer and Producer Price Indexes