bls_1256_1959.pdf

mwm

CONSUMER PRICESIN THE UNITED STATES ,

1953-58Price Trends and Indexes

Bulle tin No. 1256\ UNITED STATES DEPARTMENT OF LABOR

) James P. Mitchell, SecretaryBUREAU o f l a b o r s t a t is t ic s

. A ' Ewan Clague, Commissioner

Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis

U. S. DEPARTMENT OF LABORB U R E A U O F L A B O R S T A T I S T I C S

W a s h i n g t o n 25, D. C.

Errata Sheet

ELS Bulletin 1256, Consumer Prices in the United States,1953- 58.

Index numbers (191:7-1:9 = ICO) appearing in Tables B -2 , B -l i , B-5, and C -2 have been revised, as follows:

Tables: B-2, B-U, B-5 . C-2 C-2City: U.S. City A'verage Los Angeles Los AngelesGroup: Medical care All items Medical care

Revised Revised RevisedDate index index index

1958: Average T O i ^ u nroJune lhli.2 125.5 11:3.8July Hi 5.0 125.7 lUli.3Aug. 11:5.3 125.5 Hill.5Sent. Ui6.5 126.0 ll|li.6Oct. Hi 7.1 125.9 11,5.8Nov. lhl.h 126.5 11:5.8Dec. 11:7.6 126.5 11:5.8


CONSUMER PRICES IN THE UNITED STATES1953-58

Price Trends and Indexes

Bulletin No. 1256UNITED STATES DEPARTMENT OF LABOR James P. Mitchell, Secretary

B U R E A U O F LA B O R S TA T IS T IC S Ewan Clague. Commissioner

For sole by the Superintendent of Documents, U .S . Government Printing O ffice , Washington 2 5 , D.C. Price 65 cents


Preface

n ils b u lle tin presents an a n a lysis o f the trend o f r e t a il p rio e s fo r the 6 -ye a r period 1953-58, ta b le s o f indexes and re la te d data fa r these and e a rlie r ye a rs , and a d e s c rip tio n o f the 1952 re v is io n o f the Consumer Price Index. B u lle tin 1165 covered the y e a n 1949-52. In 1953, the Consumer Rrioe Index was rebased from 1935-39=100 to 1947-49=100, and indexes fo r e a rlie r years were converted to the new base to p ro vid e h is t o r ic a lly eoaparable data. T h is b u lle tin presents ta b le s o f indexes on the new base not o n ly fo r the yea rs under study (1953-58), but a lso fo r pre viou s ye a rs .

A number o f new index groupings were established a t the tin e o f the 1952 re v is io n and subsequently. Data fo r these new ly constructed groups are summarised in the ta b u la r summary.

The b u lle tin was prepared by members o f the s ta ff o f the Bureau's D iv is io n o f P rices and Coat o f L iv in g , under the general d ire c tio n o f A lla n D. S e a rle , A ctin g C h ie f, Index Methodology and R eporting Branch. Preparation o f the b u lle tin waa the re s p o n s ib ility o f Joseph H. Freeman, assisted b y C a rlyle P. S ta llin g s , E lisa b e th P. Deutermann, and Charles H. Lewie.

i i


Contents

US

Preface....................................................... ii

Price trends.................................................................. . 1Background........ ............................. .............. ............ 1The economy and pricess 1953-58.**,.................. ............. ...... . 1Special groupsi 1953-58........................ ......................... . 4Major groups of goods and services: 1953-58...... ...................... . 5

Retail prices of food.................. ............................... 5Meats, poultry, and fish.......... .......................... . 5Fruits and vegetables.......... 7Dairy products................................................ 8Cereals and bakery products...... ........................... 8Other foods at boos.............. ............................... 8

Housing.................. 8Apparel......... 9Transportation. ....... ............................................. . 10Medical care.......................................................... 12Personal care............ 13Reading and recreation.................... ..... .................... 13Other goods and services................ ......................... ..... 13

History, scope,and meaning of the Consumer Price Index..................... 14Revision of 1934-36........................................................ 14World War II............................................................... 14Postwar revision........ 14Description of old, adjusted, aid revised series.............. 15The current index.................. ..... .............. .......... 16

Technical notes.............. 19The comprehensive revision of 1952........................................ . 19

Revision of housing concept................................................. 19Revising the index........................................... ......... 19

Revision of city sample........................ ..........*.... . 20Converting expenditure data to index weights.................. ..... . 20

General description........................................... 201952 weights............. 21Food and alcoholic beverages......... ..................... 21Automobiles....................................................... 21Television.......... 22Other goods and services*............. ............... 22

Item sample......................... 22CXitlat sample.................... 22Population weights........................... 23

Some limitations.............................. .... 23Limitations on use............. 23Limitations of measurement........................ 23

Relative importance of Consumer Price Index c o m p o n e n t s . ........... 24Correction procedure for published price indexes and prioea............ . 27

Brices or indexes for individual items............. 27Subgroup, group, or all items indexes........ ...... ............ ...... . 27

iii


Tables

Page

1, Comparison of Old, Adjusted, and Current Consumer Price Index Series............. 17

2. Family Expenditures and Relative Importance of Consumer Price Index Components,Selected Periods, 1917-58............................ ......... . 26

Charts

1. Consumer Price Index Selected Croups, Annual Averages, 1935-58......... 2

2. Consumer Price Index Major Groups of Goods and Services, Ninthly,Deoenber 1952-Deoember 1958................................ . 6

3. Consumer fries Index New and Used Cars, Ninthly, January 1953-Deceaber 1958...... 11

Tabular sunmary............... ...................... ................... . 28

Bibliography ... ............ ............. ........ 123

i v


CONSUMER PRICES IN THE UNITED STATES, 1953-58

Price Treads

In the 12 years preceding 1953, the United States was engaged in World War II and the Korean conflict and, as a result, experienced several periods of severe price inflation* Between 1940 and 1952, annual average consumer prices rose nearly 90 percent (chart 1 )

As the United States increased its armed strength and later became involved in World War II, consumer prices climbed rapidly upward* From spring (March) 1941 through May 1943, the Consumer Price Index advanced 24 percent* However, toward the latter part of this period, the Government's economic stabilization program began to take effect and the rate of advance was retarded* Because of the success of the stabilization program, the period from May 1943 to June 1946 was one of relatively moderate price advance* Alter July 1946, however, controls were removed and prices soared, rising at the sharpest rate ever measured by the CPI* Largely responsible for this rise was the pent-up demand for consumer goods growing out of wartime shortages, supported by the increase in incomes and purchasing power generated by Government war expenditures* This postwar price advance lifted the CPI level more than 30 percent in a little over 2 years (June 1946-August 1943)*

The inflation was temporarily halted by the recession of late 1943-49* More plentiful supplies and more cautious consumer buying helped check the price rise* As in the previous recession (1937-38 ) 2/, however, the CPI declined only moderately* Between the September 1948 index peak and February 1950, the subsequent low for this period, consumer- prices in general fell 4 percent, primarily because of price decreases of 9 and 8 percent for food and apparel, respectively* Price movements were mixed, however; partially offsetting these declines was a steady increase in the costs of the services, including rent, which responded promptly to the relaxation of wartime rent controls*

With the outbreak of the Korean conflict in June 1950, prices moved quickly upward with all groups of goods and services joining in the rise* Between June 1950 and February 1951, the CPI advanced 8 percent or at a rate of 1 percent per month* After February 1951, the effects of the General Ceiling Price Regulation (GCPR) of January 25, 1951, placing controls over most prices, began to be felt and, from February through December 1951, the rate of advance was slowed, prices rising about 3 percent or 0*3 percent per month*

The year 1952 was one of comparative price stability* With the immediate crisis past by early 1952, it was possible to relax price controls gradually* Despite the loosening of controls, the index rose less than 1 percent over the year, December 1951-December 1952*

The Economy and P r ^ s.l. 1953-58

During the years 1953 through 1958, America experienced a peacetime economic expansion, twice interrupted by recessions once in 1953-54 and again in 1957-58* Over the entire 6-year period (December 1952-December 1958), the CPI rose 8*4 percent* During the first 3 years, there was almost complete price stability, but during the last 3-year period there was a renewed price rise* Between December 1952 and December 1955, the index advanced only 0*5 percent* j^ From December 1955 to December 1958, however, it rose 7*8 percent*

JJ See Bureau of Labor Statistics Bulletins 710, 966, and 1165 for detailed analyses of price trends during the years 1941-52*

2J For a discussion of the 1937-38 recession and the CPI, see The Consumer Price Index in the Business Cycle, by Ewan Clague (in Monthly Labor Review, June 1958, pp* 616-620)*

l


In J u ly 1953, the Korean truce was signed and w ith i t cane a sharp reduction in defense expenditures. J u ly 1953 narked the peak o f the Korean period business cycle 2/ and inm ediately th e re a fte r the "adjustment" o r "recession" o f 1953-54 ensued.

A8 in the short recessions o f 1937-38 and 1948-49, during the 1953-54 economic downturn, in d u s tria l production declined sh a rp ly (about 10 percent} but no such sharp declin e was evid en t in the general le v e l o f consumer p ric e s , j j Food p rice s weakened n o tice a b ly in la te 1954 and during 1955, but the re s t o f the index was q u ite firm throughout the whole p eriod . One o f the p rin c ip a l fa c to rs responsible fo r the seeming in s e n s it iv ity o f the o v e ra ll index to the 1953-54 downturn was the uninterrupted r is e in the p rices o f s e rvic e s .

During 1955, consumer prices demonstrated their insensitivity and tendency to lag on the "upturn" just as on the "downturn," during sudden changes in economic cycles. The Federal Reserve Board index of industrial production climbed steadily in the fall of 1954, reaching its former (July 1953) peak by the spring of 1955, and then continued upward to establish further highs in the fall of 1955. The CPI, however, rose only 0.3 percent from December 1954-December 1955* In the face of steady increases in the price of services (up 2 percent over the year), it was primarily the drop in food prices that held the total index virtually stable. Price declines for durables, e.g., automobiles, appliances which were becoming increasingly subject to discounting also contributed toward this stability.

The year 1956 saw the end of the relatively long period of price stability that began in 1952. The Consumer Price Index declined 0.1 percent in January of 1956 and held even in February, but during the last 10 months of the year it rose 3 percent. Only in August of that year when food prices dropped sharply from their seasonal peak, was the rise interrupted. The 0.2-percent drop in August was to be the last decline in the CPI for the next 2 year3.

The renewed vigor of the climb in consumer prices in 1956 was, to a large extent, based upon two price developments: First, food prices, which in February 1956 had reachedtheir lowest point since December 1950, moved rapidly upward; and second, the prices of durable commodities, which had been declining since 1952, held relatively firm during most of 1956 and then registered their usual fall spurt with the introduction of higher priced, new~model cars in October. The relative firmness in durables in 1956, before the seasonal spurt, was in part due to a steadying in appliance prices. These factors added to the continuing increases in the cost of services, upset the balance that had been responsible for the stability in the CPI over the preceding years, and initiated a rise that continued through July 1958 despite the occurrence of the recession of 1957-58. Services and food played a dominant role in the continuing climb of the CPI through 1957 and the spring of 1958.

August 1957 marked the peak of another business cycle. The economy then turned downward and the recession of 1957-58 followed. In the 6 months from October 1957 to April 1958, unemployment nearly doubled, and the Federal Reserve Board index of industrial production fell more than 10 percent. In April, the seasonally adjusted rate of unemployment stood at 7.5 percent of the civilian labor force, the highest rate in the postwar period. Yet in these 6 months, the CPI rose 2 percent. 2/ Although as already mentioned in the discussion of prior recessions, consumer prices normally lag and are little affected by short-nan cyclical changes, regardless of intensity, there was a significant difference this time. During the previous periods of economic decline, any weakness in prices tended to be most evident in the food category. During the 1957-58 downturn, the food component played a major role in forcing the index upward in the face of the business decline. This reversal of the usual pattern was due primarily to two distinct causes: First, the effects

2/ Defined by the National Bureau of Economic Research. j j Clague, op. cit., Monthly Labor Review, June 1958.y ibid.

3


of abnormally bad weather during the winter of 1957-58, which destroyed much of the fresh fruit and vegetable crops and hence forced prices of these items upward; and second, longer range factors which encouraged withholding of cattle and hogs from market and thus pushed up meat prices* 6/ Contributing to the rise in the index was the upward pressure of these food prices during this period coupled with the continued climb of the services group*

The CPI reached an all-time high (123*9) in July of 1958* Then in August as the result of a sharp drop in food prices, the index registered its first decrease since August 1956, dropping 0*2 percent* For the next 2 months (September and October), prices on the average remained unchanged due to the effect of further declines in the foods group offsetting increases in other components* In November, prices rose 0*2 percent to equal the July high, because of the seasonal effect of the introduction of higher priced, new-model automobiles But at the year*s end, December 1958, the CPI dropped back 0.2 percent as the result of a further decline in food and minor decreases in several other groups*

Over the second half of 1958 (since June), consumer prices registered no net change; temporarily, a new price stability had been established*Special Groups; 1953-58

The "all items" CPI is subdivided according to two systems: (l) commodities versusservices, i.e., the so-called "special groups," and (2) specific end use, i.e*, the so-called "major groups" (food, housing, apparel, etc*)* In the special groups, the two broad categories, commodities and services, are further divided into such subcategories as nondurable commodities, nondurable commodities less food, durable commodities, etc*; services are divided into household operation services, transportation services, medical care services, etc* 2/

The special groups originated out of the need for groupings of CPI items that could be more readily related to trends of other economic indicators, e*g., wholesale or primarymarket prices, the Department of Commerced general business indicators, the Federal Reserve Board index of industrial production, etc. Price movements among the principal special groups core shown in chart 1, which plots the movement of selected special group indexes from the year 1935 (the first year for which such indexes are available) forward.

The most significant movements among the special groups over the years 1953-58 were the strong, steady rise of prices in the services category and the weakness of durable commodities* During the previously described periods of wartime and immediate postwar inflation, services, though moving upward, did not register increases of the amplitude recorded for commodities--between 1940 and 1952, peryices rose 48 percent as against an increase of 114 percent for commodities However, in the 6 years, December 1952 through December 1958, while commodities were rising only 4 percent, the cost of consumer services rose 18 percent* In this same 6year period, prices of durable commodities fell 1 percent*

All major components of the services group contributed to its sustained upward movement* Over the period December 1952 through December 1958, rents, which had already been partially freed from controls prior to 1953 and which were almost completely decontrolled after the summer of 1953, rose 15 percent. Transportation services advanced 23 percent, led by a 32-percent increase in public transportation (mostly local transit fares)* Medical care services increased 25 percent with all its component elements contributing,i.e*, professional services (doctor, dentist, optometrist, etc*), hospital rates, and group hospitalization insurance* These individual items along with others are discussed in more detail in the following sections of the bulletin.

6/ Fo r more d e ta il see sectio n on Foods, p. 5#2/ Indexes fo r "special groups" from 1935 to 1958 are shown in the ta b u la r summary,

ta b les B4 and B5*

4


The almost uninterrupted rise in the services group (it advanced every month during the 6-year period except for March 1956 when it dropped 0.1 percent, and June and September 1958 when it held even) was in large part responsible for the insensitivity of the CPI to the two economic recessions that took place during the 6 years* Because many services are strongly influenced by custom (e.g., various professional and personal care services) and a number of others are specifically regulated by public authority (e.g., gas and electricity, public transportation, telephone, postage, etc.), they are not readily affected by short-run changes in the economy.

Durable commodity prices declined about 1 percent over the 6 years, December 1952-December 1958. During the first 3 years, December 1952-December 1955, they fell almost 8 percent and were one of the factors which helped offset the increase in service prices, thus contributing to the stability of the overall index during those years. The story of the decline in durables prices centers primarily around the drop in automobile prices during the years 1953 through 1955, and the decrease in appliance prices during most of the 6-year period ending December 1958. This story involving consideration of the application of discounting techniques in the sale of new automobiles and the emergence of the so-called ''discount house" in appliance merchandising is discussed in later sections dealing with these specific items.

The nondurables less food group encompassed such a diversity of price movements among its various items during the years 1953-58 that their story is best told in the discussion of individual "items" in the sections that follow. Food is in itself a major group and is analyzed separately in detail.

Major Groups of Goods Sargisesi__1222::2gRetail Prices of Food* Retail prices of food moved upward between December 1952 and December 1958 but exhibited marked seasonal price fluctuations. In December 1958, they averaged 4.3 percent higher than in December 1952, and 1.8 percent above the post-Korean peak reached in August 1952. By July 1958, they had reached a new peak 4*4 percent above August 1952. However, between July and December 1958, food prices fell 2.5 percent. The trend of food prices was slightly downward during 1953, 1954* and 1955, and along with the decline in the prices of durable commodities, helped partially to offset the upward pressure of rising service costs on the overall index during these 3 years# During 195o, 1957, and 1958, however, the trend was upward# Prices of food as a whole reached their lowest point for this period in February 1956 (6#7 percent below August 1952) and then began a fairly steady rise#

The following tabulation gives percent changes for food and its major components for selected periods:

Aug. 1952 Aug. 1952 Feb. 1956 Dec. 1952to to to ' to

Group J22S Feb. 1956 Hites,* 122STotal food-------------------------- + 1.8 - 6.7 + 11.9 + 4.3

Food at home---------------------- + 0 .2 - 8.1 + 12.5 + 2.6Cereals and bakery products----- - + 14.0 + 5.3 + 6.9 + 13.8Meats, poultry, and fish ---- - - - 5.4 - 21.6 + 27.4 0Dairy products ----- ---- + 3.0 - 3.3 + 4.8 + 1.4Fruits and vegetables--------------- + 1.2 - 4.5 + 16.4 + 3.7Other foods at hone------------- - 2.1 - 3.1 + 2.0 + 0.1

Meats. PpwltFYt ft*** Fish. Average prices of meats, poultry, and fish declined sharply from their August 1952 peak (19.4 percent above the 1947-49 average) to a level 22 percent lower by March 1956, and then rose markedly during 1956, 1957, and the first 7 months of 1958. By July 1958, they were back up to 19.2 percent above the 1947-49 average

5


CONSUMER PRICE INDEXMajor Groups of Goods and Services

INDEX (1947-49 = 100)December 1952-December 1958

1952 1953 1954 1955 1956 1957 1958UNITED STATES DEPARTMENT OF LABORBUREAU OF LABOR STATISTICS


and only 0.2 percent below the August 1952 peak. From. July to December 1958, however, they dropped 5.2 percent and thus, at the year*s end, the meat, poultry, and fish index was 5.4 percent below the August 1952 peak and at exactly the same level (113*0) as it was in December 1952.

The behavior of prices was different for beef and veal, pork, and poultry. Beef ani veal had quite sharp and contrasting upward and downward movements, while poultry moved rather steadily downward. At the beginning of 1953, beef and veal prices were declining rapidly from 1951 levels, which were more than 30 percent above the 1947-49 average.Record production caused prices to continue steadily downward through March 1956, but, as farmers cut down on their herds in response to the low prices, the number of cattle available for market decreased. Because of the resultant short supplies, prices turned up sharply, rising more than 36 percent from March 1956 to June 1958. From the June 1958 peak, beef and veal prices fell for 3 months and then rose during the last 3 months of the year; at the end of 1958, they were 1.3 percent below the June level.

Pork prices, in contrast, began a substantial rise in 1953 in response to decreased production, reaching a peak in May 1954 of 22 percent above the 1947-49 average, at a time when beef and veal prices were below the 1947-49 base. As production increased under the encouragement of large supplies and low prices of corn for feed, pork prices dropped sharply. By December 1955, they were 29 percent below the May 1954 peak. They turned up in 1956 and 1957 and rose sharply in 1958 through July. However, during the last 5 months of 1958, pork prices dropped 9*4 percent.

Poultry prices responded to record supplies resulting from the rapid growth of the commercial broiler industry, declining more than 30 percent from December 1952 to December 1957. In the first half of 1958, poultry prices turned up despite record supplies because of demand for poultry as a substitute for high priced red meats. In the last half of the year, however, these record supplies made themselves felt and poultry prices declined in every month except one.

Fruits and Vegetables. Fruits and vegetables, as usual, exhibited sharp upward and downward price fluctuations depending upon the availability of supplies. During 1953, 1954, and 1955, price changes were generally seasonal in character, but unfavorable weather caused shortages in 1956 and again in 1958 and unusually sharp price rises occurred.Average prices were up 3*7 percent from December 1952 to December 1958.

Prices of fresh fruits and vegetables during 1953-55 averaged somewhat below 1952, reaching summer peaks each year about 20 percent above the 1947-49 base. In July 1956 and July 1957, they reached peaks considerably higher, while freezing weather in Florida in December 1957 brought especially sharp price rises and seasonal peaks occurred several months early in 1958.

The precipitate drop in the price of potatoes from $1.31 for 15 pounds in July 1952 to 64J cents in March 1954 was one of the major influences in the 1953 decline. With adverse weather in growing areas, potatoes jumped to a level in May 1955 almost as high as in July 1952, and to $1.74 for 15 pounds in July 1956, the highest price on record.

Prices of other fresh fruits and vegetables responded to varying supplies. Tomato prices jumped 29 percent in the month of March 1956 to an average of 42 cents a pound, dropped to less than 17 cents in September 1956, and reached 46 cents a pound in April 1958. Green beans were very high priced and unavailable in many stores in early 1958 as a result of damage to the Florida crop.

Apples rose to a peak level of 23 cents a pound in 1957. Oranges were unusually high priced in 1958 because of the serious damage to the Florida crop. In July 1958, they sold at 80 cents a dozen and reached 87 cents a dozen in September and October before dropping off. Beaches were scarce in 1955 because of virtual destruction of that year *s southeastern crop. In July 1955, they averaged 31 cents a pound compared to only 17 cents

7


ig"July 1958, due to a good crop in the latter year.

Processed fruits and vegetables exhibit less marked price fluctuations than the fresh items. They decreased slightly during 1953-54 in response to large packs, began a steady rise in early 1955 which lasted well into 1956* and then turned down until mid-1957.In 1958, prices of frozen orange juice rose markedly as a result of persistently strong demand in the face of shortages of fresh oranges, reaching 29 cents for a 6-ounce can.Canned orange juice also increased in price but by a smaller amount. In December 1958, frozen fruits and vegetables averaged 22 percent above December 1952, and canned fruits and vegetables, 10 percent.

Dairy Products- Dairy products dropped 8.7 percent in price from December 1952 to June 1954# with a sharp drop in early April 1954 as a result of the reduction in support prices from 90 to 75 percent of parity, and the increase in production. Prices rose in subsequent years, partly as a result of Government efforts. In December 1958, prices averaged 11.1 percent above June 1954*

i

Cereals and Bakery Products. Unlike other food groups, cereals and bakery products moved upward in price by small amounts but rather steadily, from December 1952 to December 1958. The total increase over the period was 13.8 percent. The steady rise was largely the result of higher manufacturing and operating costs.

Other Foods at Home. The other foods at home group, which includes coffee and other beverages, fats and oils, sugar and sweets, eggs, and other miscellaneous foods, showed fairly sharp price fluctuations over the period due to changes in the prices of eggs, but relatively little distinct trend. From December 1952 to December 1958, average prices rose only 0.1 percent.

Egg prices dropped sharply during 1954 as a result of abundant supplies. Egg production in 1954 was nearly double that of 1935-39# reflecting developments in nutrition and methods of raising poultry, and rose to even higher levels in following years. In January 1955# prices averaged only 52 cents a dozen, and in June 1957, 48 cents a dozen.By June 1958, the low point for the year, prices had increased somewhat to 55 cents a dozen.

From an average price of 87 cents a pound for bag and can coffee combined in December 1952, prices rose sharply to an all-time high of $1.23 a pound in August 1954 because of frosts and drought in Brazil. As a result, consumers shifted to instant coffee, tea, and other substitutes, and prices declined, by June 1955# to an average of 89 cents or a little less than $1 a pound for vacuum pack coffee. Greatly increased planting of trees was reflected in price decreases beginning in the second half of 1957 and by December1958, coffee prices had declined below the December 1952 level.

Housing- The CPI housing group includes residential rents and all items of expense connected with the acquisition and operation of a home. Subgroups of this index are rent, other shelter costs, gas and electricity, solid fuels and fuel oils, housefurnish-ings, and household operation. Due to the comprehensive nature of the housing group, whichencompasses a variety of durable and nondurable commodities and services, there are often important differences in price movements within the group.

Two-thirds of the items included in housing costs in the CPI are consumer services, such as rent, charges involved in purchasing a home, and home maintenance and repair items. Their steady upward trend was largely responsible for the 10-percent increase in total housing costs during the 6 years, December 1952 through December. 1958. Housing costs to the consumer rose 2.1 percent in 1953 (December to December) but increased less than 1 percent per year in 1954 end 1955. The climb accelerated in 1956, and the following year, and leveled again through 1958.

Underlying the sharp increase in housing costs in the beginning of the 6-year period was the rise in residential rents, as Federal wartime controls were completely lifted in July 1953 and rapid local decontrol followed. The increase in rents between December

8


1952 and December 1953 was equal to more than one-third (38 peroent) of the 14.9-percent increase which occurred over the entire 6-year period. The 5.7-percent advance in rents between December 1952 and December 1953 was followed by an average annual (December to December) increase of 1.7 percent in the next 5 years.

On a national average, the cost of buying a home rose moderately during the 1953-58 period, although about a third of the index cities reported lower costs. For the Nation as a whole, the 6-year (December 1952-December 1958) increase in mortgage interest rates was about 14 percent, and a very large part of this occurred between mid-1956 and raid-1957, in response to a tighter Government credit policy.

Homeowners' maintenance and repair expenses climbed a total of 17 percent over the 6 years, December 1952 to December 1958. The services associated with maintenance and repair, such as repainting, refinishing floors, and reshingling roofs, advanced more than twice as much as did the prices of commodities used in maintaining and repairing a house (house paint, porch flooring, water heaters, kitchen sinks, and faucets).

From December 1952 to December 1958, the family bill for gas and electricity rose11.9 percent. The gas increase, however, was five times larger than that of electricity,20 percent and 4 percent, respectively. Although prices of home fuels naturally vary with the seasons, the trend has been upward. Coal prices were higher, with bituminous rising considerably more than anthracite. The fuel oil price increase, a total 13.5 percent over the period, was precipitated by the Suez crisis in the summer of 1956.

In contrast to most housing expenses, prices of housefurnishings were lower in December 1958 than in December 1952, by 4.3 percent. They were relatively stable during1953 but declined steadily in the next 2 years with intensified competition and price cutting in the retail appliance market. Brices steadied in 1956 and 1957 and again drifted lower in 1958.

The price fluctuations of individual housefurnishings items were varied. The major cause of the general decline was the drop in appliance prices (15 percent) over the 6-year period, December 1952 through December 1958. This trend was due primarily to the radical change in retail merchandising of appliances. Discount houses made inroads into the market and competition for customers became severe. Regular outlets that did not give reductions from list prices were caught, at first, between the discount houses and the need to maintain their prices under so-called "fair trade" statutes. From 1955 to 1957, manufacturers gradually dropped fair trade pricing and discount competition intensified. Appliance prices steadied somewhat during the latter part of 1956 and in 1957, but fell again during the first half of 1958.

The furniture and bedding index decreased 0.3 peroent from December 1952 to December 1958. Slices for these items declined 3 percent from December 1952 to June 1955 and showed a subsequent increase of 4.3 percent from March 1956 to June 1957, partially in response to higher labor and material costs (especially steel), and freight rates. The pressures of consumer demand arising during the general boom in the sales of household durables also were partly responsible for the price increases during this period.

Until 1955, textile housefurnishing prices fell from peaks reached during the Korean conflict, led by a 19-percent drop in sheet prices between the end of 1952 and September 1955. Textile items strengthened slightly the next year and remained fairly steady through 1958. Floor coverings rose in price through 1957 as material and other costs increased, though prices in 1958 were reduced.

Apparel, December 1952, apparel prices had declined 3.8 percent from the high point in the fall of 1951 during the Korean conflict. Only relatively small changes in prices occurred during the following 6 years. There was a downward drift in prices for about 2& years, except for the usual fall increases. This slow downtrend was reversed in the fall of 1955 and by October 1956, apparel prices were 2.1 percent higher than in

9


October of the preceding year* Some lesser increases occurred during 1957 but in 1958 apparel prices were almost stable (except for seasonal influences), primarily due to an adequate supply of clothing and limited increases in manufacturing costs* Over the 6-year period from December 1952 to December 1958, the average increase for apparel was 2*3 percent, the smallest Increase shown for any of the major groups*

Among the apparel subgroups, the only significant departures from this general trend were for prices of footwear and of apparel made of rayon or other manmade fibers*

Footwear prices, after declining 4*8 percent from a high point in October of 1951 to a low in January 1953, edged up slowly until the fall of 1955* From September 1955 to December 1956, prices rose 7*0 percent, reflecting higher wholesale prices for leather and higher operating costs* During the next 2 years, the rate of increase was considerably slower* In December 1958, footwear prices were 3*2 percent higher than in December 1956, for a total increase over 6 years (December to December) of 14 percent*

In contrast to footwear, prices for apparel made of rayon and other manmade fibers declined slowly but almost continuously over the 6 years from December 1952 to December 1958, aside from some fluctuation due to seasonal factors* Although most of these commodities participated in the decline for a large part of the time and were lower in December 1958 than in December 1952, the major factor was the persistently lower prices for nylon hosiery* By the end of 1958, nylon hose prices were 15 percent lower than 6 years earlier* In addition, rayon and nylon slips and nightgowns had only small price increases during the upturn in 1955 and then remained stable or were further reduced* Since women1 s and girls1 apparel of manmade fibers are more important in the index than similar apparel for men and boys, the lowered prices for these articles was the main reason for a small decrease for women *s and girls* apparel for the 6-year period* All men*s and boys* apparel increased 0*9 percent, and all women*s and girls* decreased 0*2 percent from December 1952 to December 1958*

Brices of wool and cotton apparel went down slowly during 1953 and 1954* Both classes then showed general increases of moderate amounts until 1958, when some decreases were reported* From December 1952 to December 1958, the increase for wool garments amounted to 2*6 percent and for cotton apparel, 3*9 percent*

Transportation. The transportation group includes both public and private transportation* Public transportation is comprised of city bus, streetcar, and subway fares, and railroad coach fares* Private transportation includes prices paid for new and used cars, gasoline and motor oil, tires, automobile repairs, insurance, and registration fees*

The overall transportation index remained relatively stable from 1952 to mid-1956 and then moved firmly upward to the end of 1958* Over the 6 years from December 1952 to December 1958, the cost of public transportation increased about 32 percent, while the price of private transportation rose a little more than 9 percent*

The advances in local transit and railroad fares resulted from a combination of increased operating costs and a decline in patronage* Nationwide population movements to suburbs and a concomitant preference for private transportation caused a decline in the use of existing methods of service so that companies were unable to sustain profitable operations without substantial fare increases*

Within the private transportation subgroup, dealers* selling prices of new cars averaged 13*5 percent higher in December 1958 than in December 1952. (See chart 3.) This price increase did not reflect higher prices resulting from voluntary quality upgrading by consumers, such as the increasing preference for 8-cylinder cars, automatic transmissions, etc* It did, though, reflect price increases attributable to the introduction of certain design and styling features, such as wider windshields, increased use of chrome, etc*

IO


Chart 3.CONSUMER PRICE INDEX

New and Used CarsJanuary 1953 - December 1958

UNITED STATES DEPARTMENT OF LABORBUREAU OF LABOR STATISTICS


Each fall, as new model cars were brought on to the market, the new car index registered a seasonal increase* Then as the model year progressed, dealers tended to grant price concessions* Reported prices after June 1954 reflected the growing practice of larger price concessions as the model year progressed*

In the 6-year period (December 1952-December 1958), the volume of new car purchases moved inversely to price* Sales of new automobiles rose steadily to a record high in 1955, a year in which production reached a peak of 8 million cars* However, prices of new automobiles declined 2*8 percent between December 1952 and December 1955* In 1956, though, purchases were not quite as great as in the year before* Demand also dropped sharply in the winter of 1957-58. As of December 1958, however, prices of new cars were 16.7 percent higher than in December of 1955*

Used car prices first appeared in the revised index in January 1953* They declined sharply until March 1956 when prices were a little more than 30 percent below the January 1953 level. This drop occurred as automobile ownership grew from an average of slightly less than one car per household in 1952 to 1*12 in 1957, 8/ and was indicative of a growing consumer preference for new rather than used cars* From the summer of 1956 to autumn 1957, prices strengthened, declined in the disappointing market of the following winter, and resumed their climb at a more rapid rate to the end of 1958* At this point, used car prices were still 10 percent below their level at the beginning of 1953*

Rising labor and other operating costs were mainly responsible for the 21 4-percent increase in automobile repair charges over the 6-year period (December 1952-December 1958)* A small rise in prices of tires despite a much larger increase in wholesale prices was due to an extremely competitive market to which retailers responded with trade-in allowances, "sale" prices, and discounts from list prices* Auto insurance rates increased moderately until 1958, when the rising cost of claims precipitated a 10-percent rate increase in the first quarter (from December 1957 to Ifarch 1958). Registration fees averaged 18*9 percent higher in 1958 than in 1953, with States increasing their charges as new cars became heavier and more powerful*

Gasoline and motor oil, the most important of the automobile operating expenses, were both considerably higher in December 1958 than they were 6 years earlier. Oil prices were persistently strong but the gasoline prices moved erratically as a result of tax changes political disturbances in the Middle East* The Federal gasoline tax of July 1, 1956, raised average prices more than 3 percent and State increases were comparable. When the Suez Canal closed in November of 1956, American producers geared production to satisfy probable European requirements* Then when the Canal opened sooner than expected, stocks were overexpanded* Also at that time, there occurred a reduction in the growth rate of U*S* consumption. As a result, price wars flared in many areas of the country as operators tried to move their stocks*

Modi cal Care, In the 6 years, December 1952 to December 1958, the medical care index increased 23*5 percent* During this period* all of the medical care items rose, although the degree of price movement varied widely* In December 1958, group hospitalization premiums, reflecting the advance in hospital rates, were higher by 49*1 percent than 6 years earlier; whereas the optometric examination and eyeglasses advanced only 5*2 percent. Hospital rates increased 40.2 percent during the same period, as labor and operating costs continued to mount* Physicians * and dentists1 fees also increased 21*3 percent and 16*1 percent, respectively. Prices of prescriptions and drugs rose 11.9 percent as higher production and distribution costs were passed on by the manufacturers*

When price changes for the various physicians1 services are compared, the

8/ Estimated from Automobile Facts and Figures, Mctor Vehicle Registrations in U.S*, (p* 18), 1958 Edition Automobile Manufacturers Association*

I2


differences are not quite as marked, ranging from a rise of 26*1 percent for an obstetrical case to 6.1 percent for an appendectomy. Physicians* office and house visits advanced 22.3 percent and 23*5 percent, respectively, and a tonsillectomy was higher by 14*9 percent.

In the 6-year period (December 1952-December 1958), the annual (December-to- December) increase for the medical care index averaged 3*9 percent. The smallest increase was 2.2 percent in 1954* due to the smaller-than-average increase for medical services other than hospital rates and group hospitalization premiums. The largest annual increase was reported in 1958 with group hospitalization higher by 10.8 percent and hospital rates by 3#9 percent, resulting in an overall annual, increase of 4*6 percent for medical care.

Family expenses for personal care, which include such items as men's haircuts, beauty shop services, and toilet articles, advanced a total of 15 percent from December 1952 to the end of 1958. An important factor in the persistent price advance was the fact that half of the personal care items are services. The personal care price trend generally paralleled that of all services in the CPI for much of the 6-year period.A major cause of the price increase for personal care was the advance in men's haircuts, which were priced 24 percent higher in December 1958 than 6 years earlier. Beauty shop services rose about 19 percent during the same period, reflecting price increases for such items as shampoo and wave set. Toilet goods price movements were mixed but in general they were higher.

RsfitfUftfl ftrci Recreation. The cost of reading and recreation to the urban family rose only 8.2 percent between December 1952 and the close of 1958, slightly less than the average of all consumer prices. Prices fell slightly from December 1952 to the summer of 1955 and remained fairly stable until the following summer, but in the next 2- years (June 1956 to December 1958), they rose almost 9 percent. Within the group, there were divergent movements: Commodities declined while services increased. Prices of televisionsets, affected by competition between discount houses and regular outlets, dropped almost 7 percent between December 1952 and the end of 1958. In contrast, movie admissions rose28.9 percent in the same period. To combat declining attendance, theaters introduced special features and improved screens and techniques. To offset the resultant increased costs, they raised prices. Newspaper prices increased sharply after remaining relatively stable for a number of years. By December 1958, newspaper prices had increased 19*7 percent over the December 1952 levels.

Other Goods and Services. During most of the 6-year interval, from December 1952 to December 1958, other goods and services, which consist primarily of tobacco and alcoholic beverages, moved similarly to the total CPI, increasing almost 10 percent. Prices of alcoholic beverages did not advance as rapidly as prices of tobacco products. Cigarette prices rose 16.3 percent, reflecting industrywide price increases in February 1953 and June 1957. Higher manufacturing costs and increased State taxes contributed to price increases for both alcoholic beverages and tobacco.

I3


H is to ry, Scope , and Meaning o f the Consumer P rice Index

The Bureau o f Labor S ta tis tic s ' Consumer P rice Index measures changes in re t a il p ric e s o f goods and se rvice s purchased by c it y wage-earner and c le ric a l-w o rk e r fa m ilie s .I t is an index o f p rice change, and th e refore does not in d ica te changes in the kinds and amounts o f goods and se rvice s fa m ilie s buy o r the to ta l amounts fa m ilie s spend fo r liv in g , o r d iffe re n ce s in p rice le v e ls o r l iv in g costs in d iffe re n t places.

The Consumer P rice Index in ro u gh ly i t s present form was f i r s t published in 1921 w ith data back to 1913* The basic features o f the e a rly 11 c o s t-o f-liv in g ' in d e x, as the s e rie s was then known, included: ( l) An index designed to measure changes in p rice s paidb y a sp e c ific population group urban wage-earner fa m ilie s ; (2) a n a tion a l " a ll items'* in d e x, published re g u la rly , based on a c it y sample re p resen ta tive o f a l l U .S. c it ie s included in the scope o f the measure ( in it ia lly the index was lim ite d to la rg e c it ie s ) ; (3) a w eighting stru ctu re which represented the re la tiv e importance o f item s in the ty p ic a l wage-earner and c le ric a l-w o rk e r fa m ily consumption p a ttern as d erived from consumer expenditure su rve ys; and (4) system atic c o lle c tio n o f p ric e s from r e t a il stores and se rvice establishm ents t y p ic a lly patronized by the "index fa m ilie s" .

The CPI, as published in 1921, was based on data developed from surveys o f fa m ily expenditures during the years 1917 to 1919* D e ta ils o f cost o f goods in the fa m ily "market basket" were obtained from 12,100 wage-earner and c le ric a l-w o rk e r fa m ilie s in 92 in d u s tria l centers* P rice data were co lle cte d re g u la rly fo r about 145 commodities, and se rvic e s and indexes were constructed fo r 32 c it ie s and fo r the U .S. as a whole. Two other c it ie s were added la te r and the 3 4 -c ity sample was maintained u n t il the re vis io n o f the index in 1952. The base period o f the e a rly CPI was 1913=100, la te r changed to 1923-25=100.

R evision o f 1934-36. In 1934 through 1936, the Bureau undertook a comprehensive fa m ily expenditure su rvey and In s titu te d marked improvements in methods o f c o lle c tin g r e t a il p rice s. The su rvey covered buying patterns o f n e a rly 14,500 fa m ilie s in 42 c it ie s , each w ith a popula tio n o f 50,000 o r more. The 1934-36 su rvey revealed th a t expenditures fo r food, housing, and c lo th in g had a l l changed m arkedly in the kinds o f goods purchased and in d o lla r volume as a percent o f to ta l fa m ily Income and expenditure. The re s u lts o f the survey were used to rew eight the in d e x. Sim ultaneously, new commodities were added and the base period sh ifte d to 1935-39=100. A s ig n ific a n t development was the in it ia t io n o f p ric in g according to predetermined q u a lity d e sc rip tio n s. The work o f re vis io n was completed in 1940.

W orld War I I . During World War I I , the Bureau attempted to take account o f the e ffe c t o f shortages, ra tio n in g and p ric e c o n tro l, and q u a lity d e c lin e s, on r e t a il p ric e s . Proper allowance fo r q u a lity change could not be made, however. No s a tis fa c to ry method e xiste d f o r measuring "black market" p ric e s , which apparently were frequent during th a t p e rio d . The Bureau was able to re vis e the l i s t o f a rtic le s p ric e d , however, and take cognizance o f the e ffe c t o f ra tio n in g b y s h iftin g the importance o f weights f o r c e rta in fo o d stu ffs (meat and sugar, fo r example). Wartime product sp e c ific a tio n s replaced those o f peacetime, c o lle c tio n o f re n t inform ation d ire c t ly from tenants was in s titu te d , and oth er changes were made to meet the w art line s itu a tio n .

Postwar R e visio n . In 1949, the Bureau o f Labor S ta tis tic s commenced a 3 -yea r program to re vis e the in d e x. Before th is was completed, however, the Korean c o n flic t occurred in 1950 and i t was decided th a t immediate changes in the index were necessary, even though based on Incomplete survey re s u lts . Consequently, an "interim adjustment" was made, using inform atio n then a va ila b le and embracing (1) re vis io n o f c it y population weights using 1950 Census

14


data; (2) co rre ctio n o f the re n t index fo r the "new u n it b ias," 2/ to take account o f the re n t d iffe re n tia l between housing constructed prewar and postwar (Vorld War I I ) ; (3) additio n o f soae 25 new ite a s ; and (4) re vis io n o f w eights fo r groups o f ite a s and fo r sone co a n od ities.

Meanwhile, the 3 -yea r comprehensive re vis io n program continued and was completed in December 1952. The index base was sh ifte d to 1947-49=100. New w eights representing 1951-52 spending p a tterns were derived from the 1950 consumer expenditure study* T h is re v is io n o f the market basket was the f i r s t basic step in re vis io n o f the CPI.

An in te g ra l p a rt o f th is re vis io n was a complete reexam ination o f the c it ie s chosen fo r in c lu sio n in the to ta l C PI. The form er index represented o n ly la rg e c it ie s , but the re vise d index is based on a s c ie n tific a lly chosen sample o f 46 urbanised areas, includin g la rg e , m edium -site, and small c it ie s w ith more than 2,500 p op ulation. IQ/

The fo llo w in g ta b u la tio n shows the expenditure bases fo r weights in the o rig in a l index and in each m ajor re vis io n and the period during which these weights were used:

Date o f expenditure WffTpy______

Period used in index

O rig in a l index----------------------------------------- 1917-191940 1/-------------------------------------------------- 1934-361951 V -------------------------------------------------- 1947-49 U1953------------------------------------------------------- 1950

1913-39 1935-52 2/ 1950-52 1953 to date

1/ In the 1940 re vis io n , a ll-ite m s indexes fo r the ye a rs 1925-29 were re vise d re tro - a c t iv e ly b y re ve ig h tin g group indexes w ith weights calculated as averages o f group expenditu re s derived from the surveys o f 1917*19 and 193-4*36; a ll-ite m s indexes fo r the yea rs 1930*34 were re vise d re tro a c tiv e ly b y rew eighting group indexes w ith weights derived from the 1934-36 su rve y; fo r the ye a rs 1935*3% indexes fo r a ll item s and m ajor groups were comp le te ly recalculated w ith re vise d weights derived from the 1934*36 su rve y.

2/ Weights based on the 1934-36 expenditure survey were used in the "old s e rie s index through June 1953*

Jj In the 1951 re vis io n (in te rim adjustm ent), a ll-ite m s and group indexes fo r January to December 1950 were re vise d re tro a c tiv e ly ; indexes fo r re n t and a l l item s were corrected fo r the new u n it b ia s from 1940.

lJ 7 c it ie s only*D escrip tion o f Old* Adjusted. and Revised S e rie s, The "old s e rie s91 index measured average p ric e changes through June 1953 w ith weights based on 1934-36 spending p a tte rn s. The "adjusted s e rie s 91 index measured average p rice changes from 1950 through 1952 w ith w eights based on estim ated 1949 expenditures, w ith a re tro a c tive c o rre ctio n fo r the new u n it b ia s in the re n t index back to 1940. Both the old and adjusted se rie s were published fo r the period 1940-52. The "revised se rie s," which is the index c u rre n tly published, measures p ric e changes from 1953 to date w ith w eights based on 1950 expenditure patterns from the expendi-

2/ The rent component had understated the actual ris e in re n ts a fte r 1940 because the index techniques fa ile d to re fle c t the s ig n ific a n t d iffere n ce between re n ts fo r new d w ellin g s when they f i r s t came on the re n ta l market and those o f comparable dw e llin gs a lrea d y on the market. In a market fre e from c o n tro ls there is no consistent d iffe re n tia l between the "old" and "new" u n its , but under the p ro vis io n s o f re n t c o n tro l, rents fo r o ld e r u n its were co n tro lle d w h ile most new u n its were exempt. See In te rim Adjustment o f CPI (BLS B u ll. 1039)#

10/ F o r a more d e ta ile d d iscu ssion , see S election o f C itie s fo r Consumer Expenditures S u rvey, 1950 (in M onthly Labor Review, A p r il 1951, pp# 430-436; a lso issued as R eprint R. 2060).

15


tu re su rve y, adjusted f o r income and p ric e change to 1952* A continuous s e rie s extending back to 1913 was achieved by lin k in g the re vise d se rie s to the adjusted se rie s a t December 1952 The old se rie s vas discontinued in June 1953 Table 1 presents a comparison o f the cu rren t index w ith e a r lie r series*

The Current Index* The current index is based on p ric e s o f some 300 item s which were sele cted to represent average p rice movements o f a ll goods and se rvice s bought b y the fa m ilie s o f urban wage earners and c le r ic a l w orkers, as shown by the Survey o f Consumer Expenditures in 1950* P ric e s are co lle cted from a sample o f r e t a il sto res and se rvic e establishm ents in the 46 urbanised areas and sm aller c itie s # The establishm ents where p ric e s are obtained are those most fre q u e n tly patronised b y wage-earner and c le ric a l-w o rk e r fa m ilie s and represent the im portant types o f m erchandising operations, such as chain, Independent, department, and s p e c ia lty sto re s, and others* Rental data are obtained from tenants liv in g in u n its represe n ta tive o f a ll re n ta l dw ellings in the urbanised areas*

The goods and se rvice s Included are priced in accordance w ith d e ta ile d sp e cifica tio n s which define q u a lity* To the exten t p o ssib le , p ric e s are obtained fo r commodities o f constant q u a lity from month to month so th a t the index shows p ric e change o n ly and is not affected by q u a n tity o r q u a lity d iffe re n c e s. U / In a d d itio n to v e r ify in g the product d e s c rip tio n , the Bureau's agents check on discounts o r tra d e -in allowances to obtain re a lis t ic net prices* These net p ric e s paid by customers, in c lu d in g sales and excise ta xe s, are used to calculate the CPI*

The na tion a l index (U.S. c it y average) includes p ric e s from the 20 la rg e urbanized areas fo r which separate indexes are published and fo r 26 a d d itio n a l medium-size and sm all c itie s * 12/ For the 5 la rg e s t c it ie s , indexes are published each month fo r a ll m ajor groups and subgroups o f item s (see Tabular Summary tables C-2 and C-3)* Fo r the oth er 15 la rge c it ie s , indexes are published m onthly fo r food and it s subgroups and q u a rte rly fo r a ll items, in c lu d in g food* The p u b lica tio n schedule fo r the c it y a ll-ite m s indexes is as fo llo w s :

Monthly a^?*J u ly , Oct*F e b ., May, A u g ., Nov.

Mar*, June, Sept*, Dec*

Chicago D e tro it Los Angeles New York P h ila d elp h ia

Boston Kansas C ity M inneapolis P ittsb u rgh Portland

ClevelandHoustonScrantonS eattleWashington, D* C.

A tla n ta Baltim ore C in c in n a ti St* Lou is San Francisco

Indexes fo r selected items and fo r special groupings o f item s are published f o r the months o f March, June, September, and December based on data fo r c it ie s priced in these months*

. u / See Average R e ta il P ric e s: C o lle c tio n and C alcu la tion Techniques and Problems(BLS B u ll. 1182).

12/ See footnote 10.

16


T a b le 1 . Com parison o f O ld , A d ju s t e d , and C u rren t Consumer P r i c e In d ex S e r ie s

Item O ld In d ex A d ju s te d In d ex C u rren t In d ex

B ase p e r i o d - - ---------- 1935 -39= 100 1 9 3 5 -3 9= 1 0 0 194 7 -4 9= 1 0 0B a s is o f in d e x w e ig h t s :

E x p en d itu re w e ig h ts ------------ -A verage fa m ily e x p e n d itu r e s d e r iv e d

from 1 9 3 4 -3 6 Survey o f Money D is bursem ents o f Vage E arn ers and C le r i c a l W orkers in 42 L a rge C i t i e s *

R e la t iv e w e ig h ts o f O ld In d ex a d ju s t e d t o p o stw a r p a t t e r n b y e s t im a t e s b a se d on S u rv eys o f Incom e and E x p e n d itu re s in 7 C i t i e s , 1 9 4 7 -4 9 and a p p r o p r ia te p o stw a r d a ta from o t h e r s o u r c e s , su ch a s r e c e n t fo o d con su m p tion s u rv e y s b y U .S . D epartm ent o f A g r i c u l t u r e , and from t r a d e and o f f i c i a l s o u r c e s on p r o d u c t io n , m a r k e t in g , s a l e s , e t c .

A verage fa m i ly e x p e n d itu r e s d e r iv e d from 1950 Consumer E xp en d itu re S urvey in 91 C i t i e s , a d ju s te d t o r e f l e c t th e 1952 e x p e n d itu r e p a t t e r n r e q u ir e d t o m aint a i n th e l e v e l o f l i v i n g c h a r a c t e r i s t i c o f u rban w a g e -e a rn e r and c l e r i c a l - w ork er fa m i l i e s *

P o p u la t io n w e ig h ts E stim a ted 1942 p o p u la t io n s o f l a r g e c i t i e s *

1950 p o p u la t io n s o f la r g e c i t i e s * 1 95 0 p o p u la t io n s o f a l l u rban p la c e s *

P o p u la t io n C o rera ge o f E x p e n d itu re S u rvey 1 /

F a m ily s i s e - ---------------------------- E m p loy m en t-occu p a tion o f c h i e f

e a r n e r o r head o f fa m ily *

2 o r more p e rs o n s Wage e a rn e r o r s a la r ie d c l e r i c a l

w ork er

Same a s o ld in d e x Same a s o l d in d e x . D o.

L ength o f em ploym ent* 1 member, a t l e a s t 1 ,0 0 8 h o u rs sp re a d o v e r 36 w eek s .

Head o f f a m i l y , 26 w eeks* No s p e c i f i c re q u irem en t b u t m a jo r p o r t i o n o f incom e o f f a m i ly head must be from em ploym ent a s wage e a rn e r o r s a la r i e d c l e r i c a l w orker*

Minimum fa m ily incom e o f $500 In th e su rv e y y e a r and e a rn in g s o f c h i e f e a r n e r a t l e a s t $300* C h ie f e a r n e r , s a la r i e d c l e r i c a l w o r k e rs , e a rn in g l e s s than $ 2 ,0 0 0 d u r in g th e su rv ey y e a r o r l e s s than $200 d u r in g any 1 month* No u p p er l i m i t a t i o n on wage e a r n e r s o r t o t a l e a rn in g s o f a l l members o f th e fa m ily com bined* No m ore than o f incom e c o u ld be from i n t e r e s t , d iv id e n d s , r e n t s , g i f t s , incom e in k in d , e t c .

F a m ily incom e u n d er $ 1 0 ,0 0 0 a f t e r t a x e s in th e s u rv e y y e a r* No lo w e r in com e l i m i t , e x c e p t t h a t f a m i l i e s w h ich had no in com es from w ages o r s a l a r i e s w ere e x c lu d e d .

Same a s a d ju s t e d in d ex*

E conom ic l e v e l No r e l i e f f a m i l i e s e i t h e r on d i r e c t o r w ork r e l i e f *

No e x c lu s i o n f o r r e c e i p t o f r e l i e f , a s s u c h , b u t o n l y f a m i l i e s w ith wage o r s a la r y e a r n in g s in c lu d e d *

Do*


T ab le 1* Comparison o f O ld , A d ju sted , and C urrent Consumer P r ic e Index S e r ie s - Continued

Item Old Index A d ju sted Index Current Index

C ity C overage

________________________________ 34 la rg e c i t i e s none l e s s than 50,000 Same a s o ld in d e x . 46 c i t i e s rang ing in s i s e from M a d ill ,p o p u la tio n ; o n ly 1 w ith 1950 p o p u la - O kla , (abou t 2 ,5 0 0 p o p u la t io n ) t ot io n o f l e s s than 1 00 ,0 0 0 , (56 c i t i e s New York C ity .f o r f o o d ) .

Food and f u e l p r ic e d m onthly In a l l _ - T r ----- ,----- ,-------------- t---------------- ,----- -------- fitml mvu4 m fm o m a l 4fmnm w w p wllu A luN vOXev vOU J v^Moc i t i e s . p r ic e d m onthly in a l l c i t i e s .

Other com m odities and s e r v ic e s p r ic e d __ ___________________ - __- _____ ___________ Other com m odities and s e r v ic e s p r ic e dm onthly in 10 c i t i e s , q u a r te r ly in m onthly in 5 la r g e s t c i t i e s , q u a r te r ly2 4 . in 15 la r g e c i t i e s and 26 m edlum -sise

and sm all c i t i e s .

Commodity Coverage

Number o f item s p r i c e d - ----------- 200 225 310Food - ----------------------- 51 60 87Housing --------- 46 52 66

S h e lte r ------------------------------------ - 1 1 15

House purchase and upkeep-X

14Gas and e l e c t r i c i t y --------- - 2 3 3S o lid f u e l s and f u e l o i l -------- 8 8 7H ou se fu m ish in gs - - - - - 25 29 30H ousehold o p e ra t io n 10 11 11

A pparel --------------------- ------------ 62 66 73T ra n sp orta tion --------- 10 10 18M edica l ca re --------------- --------- 15 16 18P erson a l ca re --------- ------------- - 11 12 13Reading and r e c r e a t io n ------- -- - 2 5 31Other goods and s e r v ic e s ------------ 3 4 4

P ublish ed group in d exes F ood , r e n t , a p p a re l, f u e l , h ou se- Same a s o ld in d e x . F ood , h ou sin g , a p p a r e l, t r a n s p o r ta t io n ,fu r n is h in g s , and m isce lla n eou s goods m ed ica l c a r e , p e rso n a l c a r e , rea d in gand s e r v ic e s . and r e c r e a t io n , and o th e r goods and

s e r v ic e s .Im portant ch an ges:

I ?m 4 4 n k e 4 a /4 4 a tia w m enny> m 4 a * P m m fa n w t t n t t r a a I a r\T*4 a a /44 * 0 w lM clw u U b O H A V O BHHW p r i v O h m v O W Q v

a s fo o d consumed a t home.i B o w a U i a U b 1 U O U 0 p i l v V U e

Used ca rs Estim ated t o have same p r ic e movement ^ d o Used ca rs p r ic e d .a s new o a rs .

H ousing:Rent No adjustm ent f o r new u n it b ia s . A d ju sted f o r new u n it b ia s from 1940, No adjustm ent need ed .Home-ownership c o s t s - --------- Home purchase n ot Inclu ded in in d e x . Same as o ld in d e x . Home purchase in c lu d e d . Home m ainte

Maintenance c o s t s estim ated t o have nance item s p r ic e d and purchase p r i c esame p r ic e movement as r e n ts . o f home rep resen ted by d i r e c t p r i c in g .

1 / F or the o ld in d e x , exp en d itu re surveys were lim ite d t o w age-earner and c le r lc a l* w o r k e r f a m i l i e s . F or the a d ju s te d and cu rre n t in d exes the surveys In clu ded a l l urban consum ers. The p o p u la tion coverage in form a tion presen ted h e r e , how ever, r e f e r s o n ly t o the w age-earn er and c le r ic a l -w o r k e r fa m il ie s o f the su rveys .


Technical Motes

Thft ftrngton gf. 12gRevision f Hnnaing Concent. Although the current (revised) index is not fundamentally different Aron prior series, important changes in the pattern of living in the U.S. between 1939 and 1950 Bade certain conceptual revisions desirable. In that period, the expenditures of city-workers' families for housing changed radically, requiring revision of the housing concept. When the index was initiated, the identification and classification of expenditures for housing were relatively simple. Families either owned or rented their dwellings (shelter only) and paid for fuel and furnishings separately. But as the manner of living altered, housing became an ever-changing group of goods and services. According to early family living studies, most city workers rented their homes and the purchase price of homes and payments on mortgage principal were treated as savings. By 1950, the increased proportion of homeowners among wage-earner families required a revised measurement of homeowner shelter costs. In the light of the revised housing concept, the emission of home purchase from the index was no longer valid.

In the early versions of the index, prloes of homeowner's maintenance items were assumed to move like rents, on the grounds that owner housing had been in competition with rental housing, and many factors which affected rents affected owner maintenance costs similarly. The assumption that rents and maintenance costs moved together became invalid during World War II as rents cams under control but home owners' costs in general did not. Therefore, in revising the index, the Bureau broadened the housing definition to include all items of expense connected with the acquisition and operation of a home. The purchase of a home was included in the weights and pricing of both beam purchase and maintenance items was instituted.

The revised housing ooncept led to seme changes in classification of items in the index, frier to the revision, the individual items of housing expense, Insteed ef comprising a separate housing index, appeared in four of the six major components of the CPI: rent} fuel, light, and refrigeration} housefttrnishings} and miscellaneous goods and services. The rent component included expenditures by renters for shelter and other items in the rental price, combined with expenses of owners for mortgage interest, taxes, insurance, repairs, ground rent, and financing charges in connection with sale or purchase. With the 1953 revision, the calculation of a housing index was started. Subgroups of this index now include shelter, gas and electricity, solid fuels and fuel oils, housefurnishings, and household operation. The weights for the new shelter component include residential rents, home purchase and upkeep, and housing away from home.

The rent segment measures the effect of price change on the cost of renting dwelling units in proportion to the number of tenants among city-workers' families in 1952. It reflects changes in contract rents charged for dwellings which represent all types of family rental units. Rent information is obtained directly from the tenant by personal interview.

The homeowner segment measures the effect of price change on the costs of acquiring and operating homes in current markets at the ownership level in 1952. These costs consist of such items as financing services required to buy similar qualities of housing under the same mortgage risk conditions as existed in the reference year} the same level of protection against fire and other hazards as existed in 1952 for a similar quality of house and typo of risk} property tax payments levied on the same quality of beams owned in 1952; and maintenance and repairs at the same rate as observed in 1952.Revising the Index. One of the first steps in the revision program was collection of information on rent and housing characteristics for a large sample of dwelling units. These dwelling unit surveys served two purposes: (1) They provided addresses from which a representative sample of households could be selected for a detailed study of family expenditures,

19


and (2) they provided the baeie data needed to ealeolate and sake the necessary correction for the understatement in the rent index. 13/

The f sally expenditure data for the 1952 index revision were obtained during the early souths of 1951. Bureau field interviewers obtained Aron each family in the staple a comprehensive accounting of its lncoaw, savings, and expenditures during the calendar year 1950. The data obtained included the kinds and amounts of goods and services bought, the anounts spent for each iteei, and where purchases were aade. 14/ Price Information was also obtained for an expanded list of goods and services in additional types of stores, to provide data on comparative price movements required for selection of iten and store samples.

Revision of City Sample. By the addition of small and medium-size cities, the city sample was made representative of all urban places. Selection of the cities to be included in the expenditure survey was based on two major considerationss (1) To provide a good sample of the total urban population on which estimates of the U.S. urban spending pattern could be based, and (2) to make possible reasonably reliable estimates of expenditure weights for any city in the D.S.

Thirteen large urbanised areas 15/ having a total population of 35.5 million were included because of their extreme importance from any point of view. The remainder of the 46 cities were chosen by scientific sampling methods Aram among 2,785 cities with a total population of 55.2 million. In selecting the sample, consideration was given to temperature, population density, community income level, and (in the case of the smallest cities) distance to a market center.

To expand the sample for expenditure data, 50 additional cities were selected.Of the total of 96 cities in the sample, 6 had been surveyed during the previous 3 years and were not included in the 1950 program. Only the national sample of 46 cities is included in the revised Consumer Price Index. 16/

Converting Expenditure Data to Index Weights. The consumer expenditure surveys yielded expenditure data which could serve adequately as a firm basis for weights for the national "all items11 CPI and for the various groupings within the TI.S. index. The same reliability attached to the weights for the large cities.

General Description. As usual with samples, however, the finer the detail required, the more sampling error, other things being equal. Thus, the expenditure data for a group of items for any one small city would have mere sampling error than for the U.S. as a whole. This error was reduced by peeling the data for a number of related cities and using the average expenditures for the city-groups to develop weights which were then used for each city in the group.

Under this general procedure, the first step in the adjustment of the survey data was the classification of items into three broad categories of consumption which previous studies had indicated were related to level of income. These groupings were: (1) food and alcoholic beverages, (2) housing, and (3) all other goods and services. Within each of these categories, items were further classified into subcategories, groups, subgroups and item classes.

13/ See footnote 9.14/ Methodology of the Survey of Consumer Expenditures in 1950, Helen Humes Lamale,

Bureau of Labor Statistics, U.S. Department of Labor, in conjunction with Study of Consumer Expenditures, Incomes, and Savings, University of Pennsylvania, 1959.

15/ Two urbanised areas, Hew Tork City and northeastern Hew Jersey, were later combined for use In the index.

16/ See footnote 10.

20


Beginning with the saalleat classification:1. The relative importance of each item in total expenditure for mall groups

of related items was calculated for each city.2. Distributions of cities by these relative importance figures were examined

for similarities or differences in expenditure patterns.3. Expenditure data for eities having similar family expenditure distributions

were averaged for each city grouping separately and the resulting average relative importance was used for each city in the group. The average was assumed to give a more representative pattern and minor intercity differences were regarded as sampling errors. Where the range was extremely large, several city groupings were made. The objective was to minimise the adjustments due to sampling and reporting errors, but to preserve city differences which were felt to represent real differences in consumption patterns by city.This same procedure was used at successively higher levels, subgroups, groups, etc.,

up to but not including the three major categories of consumption food, housing, and all other goods and services, which ware derived by procedures to be discussed later.

1952 Weights. The Consumer Expenditure Survey data relate to the year 1950, but the new weights were not introduced into the CPI until 1952. Accordingly, two broad types of adjustment were needed: (1) to take account of changes in basic spending patterns between 1950 and 1952, and (2) to take account of price changes between 1950 and 1952.

The first step in estimating 1952 expenditures was derivation of estimated family income for each city. Family income for 1952 was estimated from a straight-line regression of 1950 income of wage-earner and clerical-worker families and gross weekly earnings of production workers in manufacturing industries. Two sets of regression coefficients were calculated, one for use with cities having populations of 240,000 and over and another for cities of under 240,000 papulation. The equations thus developed were then applied to 1952 earnings data to obtain estimated family income by city in 1952.

A number of aspects of the method of converting statistics from the expenditure survey to weights in the Consumer Price Index have been described in various articles in the Monthly Labor Review and in BLS Bulletin 1165, Consumer Prices in the U.S., 1949-52.The following discussion, therefore, deals with certain aspects ef the weight revision which are of general interest and have not been treated elsewhere. The method of deriving the weights far housing are covered in considerable detail in two issues of the Monthly Labor Review. 22/

Food and Alcoholic Beverages. Expenditures for total food and alcoholic beverages were derived from the 1950 expenditure survey, adjusted for changes in consumption between 1950 and 1952 on the basis of observed relationships between food expenditures and net money income. More specifically, the 1952 estimated expenditures for food were computed by region and city-sise class from regression equations of food expenditures and net money income in 1950. From these equations and 1952 estimated income, derived as described above, 1952 estimated food expenditures were obtained. In addition, a variety of data were used to correct for the under-reporting of expenditures for alcoholic beverages.

Automobiles. Automobile purchases data reported in the expenditure survey indicated abnormally high rates of acquisition, probably resulting from advanced buying because of the Korean conflict and delayed buying caused by World War II shortages. It was decided, therefore, to adjust the reported data on purchases to eliminate the effect of unusually high purchase rates. For new and used cars separately, a normal purchase rate was calcu-

17/ Housing Costs in the Consumer Price Index (in Monthly Labor Review, February and April 1956, pp. 189-196 and 442-446).

21


lated from the change in percentage of families owning cars between 1935 and 1950* Annual rates of increase in ownership were calculated separately for each region (South, North Central-Northeast, and West). The number of initial buyers was calculated as the difference between the number of owners at the beginning and the end of the year. Replacement rates derived from the expenditure data were applied to the ownership figures to obtain an estimate of replacement buyers. 18/ The average number of cars bought per family was multiplied by the average price to obtain the expenditure weight.

Television. Expenditures for television sets in 1952 were estimated by differing techniques, depending upon the availability of data. For cities in which television transmitters were located, data on the number of sets in use were available from industry sources in addition to the data on ownership and purchases from the BLS expenditure survey of 1950. From the data on sets in use for various dates, rates of sales and ownership in 1950 were derived. Examination of both sets of data disclosed consistent relationships between sales (or purchases) during the year and ownership at the end and the beginning of the year.Where available, the more comprehensive industry figures were used as a basis for adjusting the BLS survey data. The corrected 1950 purchase rates were projected to 1952 by the ratio of sales in 1952 to sales in 1950, derived from the industry data on the number of sets in use.

Other Goods and Services. For this category, estimates of 1952 expenditures were made on the basis of 1950 expenditure-income ratios for six regional groupings and three metropolitan areas. The ratios were applied to the estimates of 1952 Income minus expenditures for food, housing, automobiles, and television sets. Weights in finer detail subcategories and items were developed from the relative importance of the subcategory or item, using expenditure data from the survey.

Item Sample. The expanded sample of goods and services includes approximately 300 items, most of which represent a group of items which are considered to be in the same "price family." The classification by price family is based on similarity in physical characteristic, function, and other factors affecting price movements. Appendix table A-3 shows the items currently priced for the index and the items represented by each priced item. The appropriateness of these groupings was checked by comparing the prices of the several items over a considerable period of time to make sure that their movements conformed generally to the trend in prices of the sample commodity.

Outlet Sample. Prices are collected periodically from a sample of stores and other establishments where urban wage-earner and clerical-worker families buy goods and services. These include chain and independent groceries, department stores, drug stores, and other specialty stores. Prices obtained from these sources are averaged for each city*

For cities added to the CPI in 1952, samples of independent food stores were selected from listings of all outlets in which wage earners reported food purchases in each city, stratified by type of store in terms of foods sold, importance of store in wage-earner purchases, and geographic location within the city. All important chain store systems are included in the sample for each city. Prices obtained from chain and independent stores are averaged separately and combined with weights based on sales volume data.

Samples of rental units were selected by probability sampling procedures, from block listings of the rental dwelling units in each city, stratified according to block density and location within the area.

For goods and services other than foods and rents, outlets are selected to include the establishments more frequently patronised by wage-earner and clerical-worker families,

18/ Adjustments ware also made to account for sales without repurchase.22


and which represent all inport ant typo* of nor chandi ling operations. In the newly added cities, the importance of individual stores in family laying was established by asking families where they bought the goods and serrices reported as expenditures in 1950* Stores of different types mentioned most frequently were selected after checking with local business associations.

In small cities, where mail-order buying is important, prices obtained from catalogs are included in the calculation of price change for the city. Also, for small cities, an appropriate part of price changes for commodities bought "out of town," such as furniture, is represented by the price change in large cities. Thus, the movements of prices in small cities included in the national average are based on prices from outlets in the sample cities, from mail-order catalogs, and from outlets in large marketing centers, weighted in accordance with the importance of such buying as reported in the 1950 survey.

Population Weights. City indexes are combined into the TJ.S. index by use of weights proportionate to the population of wage-earner and clerical-worker families in cities represented in the index. Each of the 12 largest urbanised areas has a weight proportionate to its own population. These 12 cities combined have about two-fifths of the total weight. The other nine large cities combined represent a class of cities having about one-fifth of the urban population and were, therefore, assigned about one-fifth of the weight. In addition to its own population, each citys weight represents all other cities with similar classification characteristics. Similarly, the 9 medium-sise oities have about one-fifth and the 16 small cities about one-fifth of the weight. The importance of cities in the index is based on the 1950 Census of Population.So Limitations.

T.

To gain about the ease degree o f accuracy throughout the in d e x, the number o f p ric e obeerrationa obtained fo r a^y item la determined by it a p ric e v a r ia b ilit y and ita Importance in the to ta l index. F o r example, pricea o f fre sh vegetablea, which are importa n t in the fa m ily food budget, change fre q u e n tly and have d iffe re n t aeaaonal patterns in d iffe re n t c it ie s ; th e y d if fe r from item to item and from sto re to sto re . To measure the average change in pricea o f a l l vegetables s a tis fa c to r ily , a la rg e number o f them must be p riced in a f a i r l y la rg e number o f stores each month in e ve ry c it y . On the other hand, men1 a n a tio n a lly a dvertised brand s h irts can be priced le ss fre q u e n tly and in fe ve r sto re s. S h irt p ric e s do not change ofte n , and s h irts o f comparable q u a lity s e ll fo r about the same p ric e in almost a l l sto re s.

The Bureau employs c a re fu lly tra in e d in te rvie w e rs in order to keep re p o rtin g e rro rs to a minimum. I t also takes precautions to prevent month-to-month p ric in g e rro rs which m ight s e rio u s ly a ffe c t the in d e x. S e llin g p rice s are recorded as marked o r as re ported by the merchant, in clu d in g sale p ric e s , discounts, tra d e -in allowances, premium p ric e s , and sp ecia l charges. Sale p ric e s fo r merchandise u s u a lly priced fo r the index are used i f the sale is o f s u ffic ie n t dura tion to a llo w most consumers to taka advantage o f the re d u ction s. Discounted p rice s are used i f th e y apply g e n e ra lly to a l l o r a m a jo rity o f customers; they are not used i f o ffe re d to sp e cia l types o f buyers o n ly, o r i f affected p rim a rily by the bargaining s k ills o f the buyer.

The agent is In stru cted to re p o rt p ric e s fo r the same sp e cifie d item as th a t fo r which p ric e s were reported in the preceding p e rio d . When th is is not p o ssib le , a su b stitu tio n is made o f another a rt ic le th a t f a lls w ith in the q u a lity lim its defined b y the s p e c ific a tio n s

One o f the more d if f ic u lt problems in the index c a lc u la tio n is the measurement o f q u a lity changes. In order to elim inate the e ffe c t o f m ajor changes in q u a lity o f the products, d e ta ile d sp e c ific a tio n s are drawn up and used b y the Bureaufs f ie ld agents in comparing p ric e s o f 1 month w ith those o f the n ext. The sp e c ific a tio n s provide a d e ta ile d d e sc rip tio n o f the a rt ic le , in clu d in g q u a lity fa c to rs associated w ith p ric e as w

bls_1256_1959.pdf

Documents

s o f indexes

tin e o f

u o f la b o r s ta

ff o f

coat o f

secretarybureau o f

members o f

lysis o f