• TABLE OF CONTENTS
HIDE
 Front Cover
 Errata
 Table of Contents
 Summary
 Introduction
 Procedure
 Relative importance of selected...
 Relative importance of selected...
 Consumption patterns for rival...
 Estimating the elasticity of demand...
 Behavior of retail prices for citrus...
 Competitive relationships of specified...
 Acknowledgement
 Appendix I. Derivation of the demand...
 Appendix II. Effect of selling...














Group Title: Bulletin University of Florida. Agricultural Experiment Station
Title: Economic relationships involved in retailing citrus products
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00026822/00001
 Material Information
Title: Economic relationships involved in retailing citrus products
Series Title: Bulletin University of Florida. Agricultural Experiment Station
Physical Description: 88 p. : ill. ; 23 cm.
Language: English
Creator: Powell, Levi A
Godwin, Marshall R ( Marshall Reid ), 1922-
Publisher: University of Florida Agricultural Experiment Station
Place of Publication: Gainesville Fla
Publication Date: 1955
Copyright Date: 1955
 Subjects
Subject: Citrus juices -- Marketing -- Florida   ( lcsh )
Citrus juices -- Marketing -- Tennessee   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: by L.A. Powell, Sr. and Marshall R. Godwin.
General Note: Cover title.
General Note: "This is a report of one phase of Southern Regional Project SM-4, Bankhead-Jones Act, in cooperation with the Texas Agricultural Experiment Station, College Station, Texas."
 Record Information
Bibliographic ID: UF00026822
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - AEN7467
oclc - 18282547
alephbibnum - 000926767

Table of Contents
    Front Cover
        Front Cover
    Errata
        Page 1
    Table of Contents
        Page 2
    Summary
        Page 3
        Page 4
        Page 5
        Page 6
    Introduction
        Page 7
        Page 8
        Page 9
        Page 10
    Procedure
        Page 11
        Page 12
    Relative importance of selected fruits in the different income areas in terms of quantity and value
        Page 13
        Page 14
        Page 15
        Page 16
    Relative importance of selected Processed juices in the different income areas in terms of quantity and value
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Consumption patterns for rival products
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
    Estimating the elasticity of demand for fresh oranges
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Behavior of retail prices for citrus products
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
    Competitive relationships of specified commodities as reflected by relative prices and relative quantities purchased
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
    Acknowledgement
        Page 76
    Appendix I. Derivation of the demand function for fresh oranges
        Page 77
        Page 78
        Page 79
        Page 80
    Appendix II. Effect of selling at the weighted average price for the market on total sales of a commodity under assumed conditions of demand
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
Full Text



Bulletin 567 August 1955


UNIVERSITY OF FLORIDA
AGRICULTURAL EXPERIMENT STATIONS
J. R. BECKENBACH, Director
GAINESVILLE, FLORIDA

(This is a report of one phase of Southern Regional Project SM-4,
Bankhead-Jones Act, in cooperation with the Texas Agricultural
Experiment Station, College Station, Texas)



TECHNICAL BULLETIN





Economic Relationships Involved in

Retailing Citrus Products


By L. A. POWELL, SR., and MARSHALL R. GODWIN

Million Boxes Million Boxes
200 200
| Other Areas

160 t Calfornia 160
Florida


120 120



80 80



40 ..'.'.' -'.,-' ' 40



0 0
1924 1926 1928 1930 1932 1934 1936 1938 1940 1942 1944 1946 1948 1950 1952
1925 1927 1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949 1951 1953
Fig. 1.-United States production of citrus fruit, 1924-25 to 1952-53.



Single copies free to Florida residents upon request to
AGRICULTURAL EXPERIMENT STATION
GAINESVILLE, FLORIDA






ERRATA
Folio line, all odd pages:
Economic Relationships in Retailing Citrus Products

Page 16, first line, second paragraph under Ranking by Value:
As the second most important fruit with reference to value,
Page 17, lines 23 and 24:
come areas. Pricing of Texas and Florida grapefruit was com-
paratively the same.
Bottoms of charts, Figs. 21 and 22 (pp. 44 and 46):
-(given figure)
Assumed demand function: x = ay
Page 86, line 10:
if K is negative, and a positive and a negative term only if K













CONTENTS
Page

SUM M ARY .................3... .. ........ ... ... ....... ..... -- ---- .....- ..-- 3

IN TRODUCTION ..... -....... ....... ......--- ... .. ........ ... ................. 7

PROCEDURE .- -- --.............-- --------------------- ---- -------........ ----.---- 11

RELATIVE IMPORTANCE OF SELECTED FRUITS IN THE DIFFERENT INCOME
AREAS IN TERMS OF QUANTITY AND VALUE ................----................ 13
Ranking by Quantity ...... -...----....----.-------. ........ ..... ----------13
Ranking by Value ...................................... .. ... ....... ...... 16
P prices ........................... .... ......- ..... ....... ... 17

RELATIVE IMPORTANCE OF SELECTED PROCESSED JUICES IN THE DIFFERENT
INCOME AREAS IN TERMS OF QUANTITY AND VALUE .................... 17

CONSUMPTION PATTERNS FOR RIVAL PRODUCTS ....................--..-----.............. 23
Oranges, Grapefruit and Apples ....--- -- ----............. .....-........-- .... -- 23
Oranges, Orange Concentrate and Orange Juice ............................... 26
Orange Juice, Orange Concentrate and Orangeade ...... ................... 32
Orange Juice, Blended Juice and Grapefruit Juice ........-...--- ......-... ---37

ESTIMATING THE ELASTICITY OF DEMAND FOR FRESH ORANGES ................ 41

THE BEHAVIOR OF RETAIL PRICES FOR CITRUS PRODUCTS ....... ................. 48
Distribution of Sales by Selected Price Intervals ............- ..-............ ---49
Relationship Between Price Dispersion and Sales ...........--................ 55

COMPETITIVE RELATIONSHIPS OF SPECIFIED COMMODITIES AS REFLECTED
BY RELATIVE PRICES AND RELATIVE QUANTITIES PURCHASED ........... 58
Oranges Apples ...-............ .....-............ -- --- --- -----------............ 61
Grapefruit Oranges ........ ... --..- ..-..-. ... .--.......- .. ..........- ... 63
Oranges Orange Concentrate .... .............................. ............. 66
Other Product Combinations ......._ ................ . ........ ............... 71

ACKNOWLEDGMENTS -....... --............ ...........----...........................-- ..-. 76

APPENDIX I .....-......-- -...---- ... ...-- .----- ........-.-...-....----..--------- -..----. ............... 77
A PPENDIX II ......... -............... .......... ..... .. .. ... .. ............ 81









SUMMARY
This study was undertaken to analyze the consumer purchas-
ing pattern for certain citrus and non-citrus products, as well
as other demand relationships, by using aggregate retail sales
data. The data used were obtained through a survey of 18 retail
food stores in Jacksonville, Florida, during portions of the 1949-50
season, and in nine stores in Memphis, Tennessee, during the
first half of 1951. Sample stores were selected with the view of
obtaining representation from each category of those predomi-
nately patronized by one of three broad groups; low, medium or
high income consumers.
In terms of quantity (pounds), oranges ranked first, com-
prising well over 40 percent of the total sales volume of the four
fresh fruits considered in the study (oranges, apples, grapefruit,
tangerines) in the low, medium and high income areas of both
Jacksonville and Memphis. From the standpoint of value, how-
ever, apples were the leading item in all income areas of Jack-
sonville. In the low income area of Memphis, the value of apple
and orange sales was about equal, but the total value of oranges
exceeded that of apples in the other two income areas. Grape-
fruit was the third ranking commodity on the basis of both
quantity and value in all three income areas of Jacksonville. The
same ranking with regard to value occurred in Memphis. The
ordering with respect to quantity presented a different picture.
Apple sales were larger in the low, approximately equal to in
the medium, and less than grapefruit sales in the high income
area of Memphis.
Orange juice products enjoyed considerable favor among con-
sumers of all income levels. Orange juice headed the list of nine
juice products under consideration in value and volume in the low
and medium income areas of Jacksonville. Orange concentrate
occupied first place in the high income area. Apparently growing
consumer acceptance caused orange concentrate to dominate
juice sales in both medium and high income areas of Memphis.
Tomato juice sales slightly surpassed orange concentrate sales
in the low income area of this city, but only in terms of volume.
An examination of the weekly proportional purchases of se-
lected combinations of certain commodities, three at a time,
disclosed a tendency for relative consumption to be associated
with price and the level of consumer income. For example, in
the fresh fruit group the proportion of sales accounted for by







4 Florida Agricultural Experiment Stations

grapefruit tended to vary directly with the level of income.
Apples, on the other hand, were more popular among lower in-
come consumers. Differences among income areas in the con-
sumption pattern for the three major orange products-fresh
oranges, orange concentrate, and orange juice-were quite pro-
nounced. The proportion of the combined purchases of these
three products attributable to orange concentrate varied directly
with the level of consumer income, while the proportions ascrib-
able to fresh oranges and orange juice were comparatively
greater in the low than in the higher income areas.
Probably the effect of price and the level of income upon the
purchasing pattern for three rival products is best exemplified
by a comparison of the relative sales of orangeade, orange juice
and orange concentrate. Orange concentrate was the highest,
orangeade the lowest, and orange juice the intermediately priced
item. Sales of the two lower-priced products, orangeade and
orange juice, were relatively larger as the level of income de-
clined. The reverse was generally true for the more expensive
commodity, orange concentrate.
An analysis of the demand for fresh oranges yielded the fol-
lowing estimated coefficients of elasticity with respect to price
in the low, medium and high income areas, respectively: Jack-
sonville: 1.206, 1.329, 0.860-Memphis: 1.782, 1.665, 1.143.
These results indicate that the demand for fresh oranges was
slightly elastic in all cases except for the high income area in
Jacksonville. The possibility that orange concentrate was in-
creasingly regarded as a fair substitute for fresh oranges may
partially account for the slightly higher coefficients derived in the
latter study.
The manner in which sales were distributed according to price
suggests considerable diversity in the actions of retail store
operators during periods when supply and demand conditions
were operating to establish a new price level for either of the
several products under consideration. While at all times sales
were distributed over a wide range of prices, during periods of
price adjustment the distributions of individual products ex-
hibited pronounced multi-modal tendencies. However, under
stable price conditions the distribution of sales tended to become
uni-modal. Indications are that under conditions of elastic
demand, variations among retailers in the price charged for the
same product may well result in a lower sales volume than would







Economic Relationships in Retailiny Citrus Products 5

be the case if all retailers sold at a price near the mean for the
market as a whole.
As in the demand analysis, devices used to test for competi-
tive relationships between goods can provide only approxima-
tions. Statistical analyses that were made consisted of (a) the
"rough test" which involved a comparison of the coefficients of
variation of the price and quantity ratios of pairs of products, and
(b) the regression of quantity ratios on price ratios, with time
included in the equation as an adjustment factor for changes in
the seasonal relationship. These analyses supplemented by
charts-which in some cases depicted a pronounced inverse rela-
tionship between the price ratios and quantity ratios of a given
pair of commodities-gave substantial support to the notion that
such items were substituted in consumption in response to
changes in their relative prices. In other cases inconsistencies
in the analyses precluded the establishment of a definite associa-
tion; i.e., no conclusion could be drawn other than that the rela-
tionship was uncertain.
In most instances the analysis left the impression that apples
and fresh oranges were probably directly competitive. Generally
the results indicated that grapefruit and oranges also were sub-
stitutes, but in some cases the relationship was doubtful.
By the time the Memphis phase of the study was undertaken,
orange concentrate had become well established in the market.
According to the results of the analysis, apparently some sub-
stitution was effected between oranges and orange concentrate
by consumers of that city.
Some of the other pairs of commodities that appeared to be
probable substitutes were oranges-orange juice, orange concen-
trate-orange juice, and orange juice-grapefruit juice.












Economic Relationships Involved in

Retailing Citrus Products

By L. A. POWELL, SR., and MARSHALL R. GODWIN

INTRODUCTION
As members of a competitive economy, producers of citrus
products are continually confronted with the necessity of modi-
fying their activities in conformance with changes in the wants
and desires of the ultimate consumer. For this reason, the citrus
industry is in perpetual need of information relating to the
consumer demand for its products-of shifts in preferences and
purchasing patterns, variations in wants among segments of
the national market, and the effects of innovations on the mar-
ket structure and buying habits.
The dynamic nature of the citrus industry makes it increas-
ingly difficult for those concerned to do an efficient job of mar-
keting and emphasizes the need for a better understanding of
marketing conditions. If available, the proper interpretation
and use of this knowledge would enable the producer, the fresh
fruit shipper, the processor and the distribution agencies to
accomplish effectively the complex operations of moving an an-
nually augmented output into consumption channels.
Few segments of agriculture can match the growth in pro-
duction achieved by the citrus industry during the past two
decades. Even with adverse climatic conditions resulting in the
virtual elimination for a period of Texas as a producing area, the
national production of citrus has steadily and swiftly mounted,
(Fig. 1). Practically all of the increase stems from the phe-
nomenal growth of the citrus industry in Florida. Considering
the fact that the preponderance of the citrus groves in Florida
are relatively young, and that substantial acreages have not as
yet reached bearing age, it would appear that only the most
severe price or weather conditions could forestall a continued and
rapid rise in the volume of citrus that the population must be
induced to consume.
Although production has been increasing steadily, the intro-
duction and perfection of new methods of utilization, from time
to time, have done much to alleviate the pressure of an expand-
ing supply. Frozen orange concentrate represents the most re-








8 Florida Agricultural Experiment Stations


cent accomplishment of this nature. An idea of the impact of
concentrate upon the citrus industry in Florida can be obtained
from Figure 2. Within a span of less than four years the volume
of Florida oranges consumed in this form has tripled. The
1952-53 rate of consumption was slightly more than three mil-
lion boxes of fresh fruit equivalent per month. While the avail-
able data suggest that further increases in the consumption of
oranges in the form of frozen concentrate will be at a some-
what slower rate than that experienced during the past, there is
every reason to believe that this outlet will continue to absorb
a major share of the Florida orange crop.

Th iuanr. . .diI . . i. i i l i I
of Boxe





2500.
-- Computed trend In consumpHoi *


2000-


1500.
150. '- Actul estimated consumption*


1000.



500-



Oct. Jon. Apl. July Oct. Jan. Apr. July Oct. Jon. Apr. uly Oct. Jan. Moy
1949 1950 1951 1952 1953
SSource "Camun er Fruit and Juice Purchase", U.S.D.A., Bureau of Agrlcultuml Economics, Wahlngtcn, D. C.
"** Uing the following fonnul: y' 1015.804 43.317x + 5.337x2 .07434x3
where y = computed monthly consumptlon
x = month number beginning with Octobr 1949 or month 1.

Fig. 2.-Trend in national consumption of orange concentrate,
October 1949-May 1953.

The development of frozen orange concentrate has done much
to alter the historic relationships which have served previously
as a guide for the actions of the various elements of the in-
dustry. The expansion in consumption of this product has not







Economic Relationships in Retailiny Citrus Products b

taken place without repercussions elsewhere in the industry.
Not the least of these has been a diminution in the consumption
of other processed citrus products. In spite of an expanding pro-
duction of citrus fruit and a growing population, the national
consumption of canned single-strength orange and grapefruit
juice has begun to decline (Figures 3 and 4). Thus, a technologi-
cal development which represents a veritable windfall to the
industry as a whole has brought with it problems of adjustment.

Thoum.nd" 1 1 1 1 1 1 1 1 . . . . . . .
of Boam
1400-

Computed trend In consumption"


Actual estimated cor.umption*

600
400 -

200

Oct. Jan. Apr. 9 July Ot. Jn. Apr. July Oc n. Jun. p ly Oct. Jan. May
1949 1950 1951 1952 1953
*Souce: "Consumer Fruit and Juice Purchases", U.S.D.A., Bureauof Agricultural Economics, Wcshington, D.C.
** Using the following formula: y' I 219.087723 2.675616x
where y' = computed monthly cornumption
x = month number beginning with October, 1949 or month 1.

Fig. 3.-Trend in national consumption of canned orange juice,
October 1949-May 1953.

It is evident that Florida citrus producers and the distribution
and processing agencies which stand between them and ultimate
consumers must operate in an atmosphere of continual change.
The successful development of an effective marketing program
under such conditions will necessitate a comprehensive appraisal
of the changing nature of the market structure. It is believed
that some contribution to this end can be achieved by an exami-
nation of citrus marketing at the retail level. Through such
studies, insights can be gained concerning consumer purchasing
patterns, quantitative responses to price changes, and competi-
tive relationships between products. It is with these general
economic areas that this study is concerned.
However, the vastness of the problem and the complexities
involved in obtaining adequate data precluded the possibility








10 Florida Agricultural Experiment Stations

of dealing either exhaustively or conclusively with these funda-
mental problems. Consequently, this report represents only a
beginning step in this direction. The specific objectives toward
which this study was directed were to:

Thoursands
of Boxes
1400-

1200. Computed trend in
conzumptlion*
1000 .
S \ Actual etlmated consumption*

600.

400-

200-

Oct. Jan. Apr. July Oct. Jn. Apr. July Oct. Jan. Apr. July Oct. Jan. May
1949 1950 1951 1952 1953
Source: "Consumer Fruit and Juice Purchases", U.S.D.A., Bureau of Agricultural Economics, Washington, D.C.
"* Uing the following formula y' 842.7889 + 25.1752x .594695x2
where y' computed monthly consumption
x 0 month number beginning with October 1949 or month 1.

Fig. 4.-Trend in national consumption of canned grapefruit juice,
October 1949-May 1953.

1. Determine the relative importance of citrus and competitive
products among consumers of different income groups.

2. Establish the general relationship between price and pur-
chasing pattern at different income levels for the following
groups of rival products:

oranges-grapefruit-apples
oranges--orange concentrate-orange juice
orange concentrate-orange juice-orangeade
orange juice-blended juice-grapefruit juice.

3. Estimate the price elasticity for fresh oranges at dif-
ferent income levels.

4. Examine the temporal changes in the range of prices of
selected citrus products and the distribution of sales by price
intervals.

5. Test for the existence of substitution between chosen com-
modities included in the study.







Economic Relationships in Retailiny Citrus Products 11

PROCEDURE
Since the proposed study was designed to deal with consumer
preferences and the competitive relationships between citrus
and rival products, retail stores were chosen as a suitable source
of information; especially since they provide the setting in which
the final choices of consumers are effected.
The project throughout was of a dual nature involving data
collected in Jacksonville, Florida, for the periods October 31
through December 16, 1949, and January 2 through June 23,
1950; and in Memphis, Tennessee, during the period January
16 through June 12, 1951.
In order that the purchases of consumers of different income
levels might be examined, each city was divided into three gen-
eral strata; namely, low, medium and high income areas. Eco-
nomic characteristics prevailing within various sections of each
city served as a basis for delineating the income areas. These
criteria were comprised largely of information concerning real
estate rentals, furnished by local newspapers, and impressions
formed after conferring with Chamber of Commerce officials,
store managers and other individuals familiar with relevant con-
ditions.
The stores chosen in each city unavoidably consisted of a
judgment sample. Other sampling techniques were found to be
impractical, as the selection was influenced by the cooperation
of store operators and because the division of the cities into
income areas could not be done with a great deal of precision.

TABLE 1.-SIZE AND DISTRIBUTION OF STORES BY INCOME AREAS,
JACKSONVILLE, FLORIDA.

Low Medium High
Store Size Income Income Income
Area Area Area

Large supermarkets ....--.....-- ....--..-.. ...--..- -... 2 1 2
Supermarkets .-.. --......- -- -----------.. --..-... 1 2 1
Medium size stores ........-.. ----------- ..- 2 2 1
Small stores ............ ------........ 2 1 1


In Jacksonville, Florida, the size of store, type of manage-
ment and location were given as much consideration in drawing
the sample as circumstances permitted. Hence, stores ranging







12 Florida Agricultural Experiment Stations

in size from small corner groceries to large supermarkets, in-
cluding those under independent and chain management, were
sampled. Furthermore, an attempt was made to give each in-
come area equal representation in the sample.
The stores surveyed in Jacksonville were distributed according
to size among the three income areas, as shown in Table 1. Floor
space and the volume of sales were used in classifying the stores
as to size.
A slightly different procedure was followed in obtaining the
sample of Memphis stores. Experience gained in the Jackson-
ville study indicated that more homogeneity with regard to
store size and type of management would be desirable. Wide
variations in the results obtained from individual retail outlets
suggested that a group of stores with rather similar pricing
policies and merchandising techniques would yield aggregative
data more meaningful for purposes of the analysis. Accord-
ingly, all stores in the Memphis sample were supermarkets-
five belonging to one chain and three to another, with one inde-
pendent. Some of the supermarkets were moderate in size while
others were very large, as designated in Table 2.

TABLE 2.-SIZE AND DISTRIBUTION OF STORES BY INCOME AREAS,
MEMPHIS, TENNESSEE.

Low Medium High
Store Size Income Income Income
SArea Area Area
Large supermarkets .- ....... .-............... ....--- 2 2 2
Supermarkets ................... .... ... .....-- .......----2 1


Since no criteria were available for identifying those products
that compete most directly with citrus, the commodities included
in the study were, of necessity, rather arbitrarily selected. The
exploratory nature of this investigation precluded any notion that
the following list of commodities finally decided upon was ex-
haustive or ideal. The products dealt with in the survey were:

Fruits Processed Juices
Oranges Orange concentrate Tomato juice
Apples Orangeade Pineapple juice
Grapefruit Orange juice Prune juice
Tangerines Grapefruit juice Apple juice
Orange-grapefruit blend







Economic Relationships in Retailiny Citrus Products 13

The method of collecting the data was the same in both cities.
In obtaining the information, weekly visits were made to each
store by two enumerators.' Once established, the schedule of
visits was adhered to as closely as possible. Upon each store call,
an inventory was taken and recorded for each product included
in the study. All stock receipts of the respective products that
had been received subsequent to the last visit were also listed.
Thus, weekly sales volume was determined by subtracting the
ending inventory for the current week from the sum of stock re-
ceipts and the ending inventory of the preceding week. When
the price of a product was changed during the week, an inven-
tory was taken, making it possible to determine the quantities of
the products which sold at different prices for that week.
Through an arrangement with the store managers, records of
weekly stock receipts were kept and an inventory was recorded
at the time of each price change. The practice of visiting the
stores once a week necessitated this cooperation from the man-
agement.

RELATIVE IMPORTANCE OF SELECTED FRUITS IN THE
DIFFERENT INCOME AREAS IN TERMS OF
QUANTITY AND VALUE
In studying the competitive relationships between components
of a group of rival products, it is beneficial to know some-
thing of their relative importance to consumers. Such informa-
tion serves as a rough guide in isolating certain commodities
between which competition is most likely to occur. Further-
more, it may be useful as an aid in determining products or
different forms of a product that are most acceptable to con-
sumers in various segments of the market.

RANKING BY QUANTITY
A comparison of the data in Table 3 shows that oranges were
the leading product and also of about equal importance in all
three income areas in Jacksonville. In each case orange sales
represented something over 40 percent of the total volume of
the four fresh fruits considered in this study. The second rank-
ing fruit in each income area was apples. Pencentage-wise, apple

1 Because of the abnormal effect of the holiday season upon sales,
stores were not visited in Jacksonville during the last two weeks of De-
cember 1949.







14 Florida Agricultural Experiment Stations

sales were highest in the medium income area, slightly less in the
low income area and least in the high income area. Grapefruit
were third in order. Proportionately, however, grapefruit sales
were almost twice as large in the high income area as in the
low and medium income areas. Tangerines were relatively un-
important, comprising from 3 to 5 percent of combined fruit
sales, and varying inversely with the level of income.

TABLE 3.-THE RELATIVE IMPORTANCE OF SALES OF SELECTED FRUITS IN
VARIOUS INCOME AREAS
(18 Retail Food Stores, Jacksonville, Florida, October 1949-June 1950).

Total Percent Price Total Percent
Product Quantity of Total Per Value of Total
(pounds) _Pound I
Low Income Area

Oranges .......... 145,785.1 45.1 $0.0585 $ 8,531.60 31.7
Apples .......... 118,818.0 36.7 0.1236 14,680.98 54.6
Grapefruit ...... 42,371.1 13.1 0.0582 2,467.88 9.2
Tangerines ... 16,524.0 | 5.1 0.0726 1,199.87 4.5
1 1
Total ...... 323,498.2 100.0 1 $26,880.33 100.0
Medium Income Area

Oranges ....-.... 121,926.8 43.2 $0.0580 $ 7,073.33 29.4
Apples ........... 109,702.9 38.9 0.1236 13,554.33 56.4
Grapefruit .... 39,948.6 14.1 0.0611 2,442.40 10.2
Tangerines ..-. 10,685.9 3.8 0.0890 951.41 4.0
Total ..-.. 282,264.2 100.0 ____$24,021.47 100.0
High Income Area

Oranges ........- 168,287.7 45.5 $0.0621 $10,455.00 34.3
Apples .--......... 116,724.3 31.5 0.1224 14,290.46 46.8
Grapefruit .... 73,961.2 20.0 0.0629 4,650.95 15.2
Tangerines ... 11,230.3 3.0 0.0990 1,111.95 3.7
Total ........ 370,203.5 100.0 _1 $30,508.36 | 100.0

Except for minor differences, the relative importance of orange
sales in Memphis closely paralleled results obtained in Jackson-
ville (Table 4). Comparable figures were slightly higher
throughout in Memphis than in Jacksonville, approaching 50
percent of total volume in the high income area. The move-
ment of California oranges tended to increase progressively with
income, which may partially account for this disparity. Further-
more, since the Memphis study did not include data collected








Economic Relationships in Retailiny Citrus Products 15

during the months of November and December, as was the case
in Jacksonville, one would expect some dissimilarities to appear.

TABLE 4.-THE RELATIVE IMPORTANCE OF SELECTED FRUITS IN
VARIOUS INCOME AREAS
(9 Retail Food Stores, Memphis, Tennessee, January-June 1951).
Per-
Total Percent Price Total cent
Product Quantity of Total Per Value of
(pounds) Pound Total
Low Income Area

Oranges, Tot.... 121,985 45.6 $0.0711 $ 8,671.26 39.7
Florida ......... 112,996 42.2 1 0.0690 7,798.32 35.7
California .... 8,989 3.4 0.0971 872.94 4.0
Apples ............. 78,847 29.4 0.1105 8.714.39 39.9
Grapefruit, Tot. 53,530 20.0 0.0650 3,482.00 15.9
Florida .......... 49,373 18.4 1 0.0648 3,197.57 1 14.6
Texas ........... 4,157 1.6 0.0684 284.43 1.3
Tangerines ...... 13,363 5.0 0.0743 992.79 4.5

Total ...... 267,725 100.0 $21,860.44 100.0
Medium Income Area

Oranges, Tot..... 97,381 45.9 $0.0721 $ 7,021.88 39.3
Florida ........ 87,335 41.2 0.0685 5,982.11 33.5
California .... 9,507 4.5 0.1034 982.60 5.5
Arizona ........ 539 0.2 0.1061 57.17 0.3
Apples ............. 54,960 25.9 0.1166 6,408.21 35.9
Grapefruit, Tot. 54,388 25.7 C.0730 3,971.41 22.2
Florida .......... 48,915 23.1 0.0732 3,582.21 20.0
Texas ........... 5,473 2.6 0.0711 389.20 2.2
Tangerines ........ 5,358 2.5 0.0879 471.06 2.6

Total ...... 212,087 100.0 $17,872.56 100.0
High Income Area

Oranges, Tot.... 91,935 49.5 i $0.0736 $ 6,764.71 42.5
Florida .......... 79,988 43.1 0.0690 5,520.67 34.6
California ...... 10,948 5.9 0.1059 1,158.91 7.3
Texas ............ 460 0.2 0.0580 26.68 0.2
Arizona ........ 539 0.3 0.1084 58.45 0.4
Apples ............. 39,522 21.3 0.1273 5,031.33 31.6
Grapefruit, Tot. 53,3t6 28.7 0.0757 4,035.66 25.3
Florida .......... 46,342 25.0 0.0758 3,514.90 j 22.1
Texas ............ 6,896 3.7 0.0745 514.01 3.2
Arizona ....... 68 .. 0.0993 6.75
Tangerines ...... 855 i 0.5 0.1142 97.66 0.6

Total ........ 185,618 100.0 $15,929.36 100.0

Although apples occupied second place with respect to volume
in the low and medium income areas in Memphis, they were out-
ranked by grapefruit in the high income area. The positions of
the two fruits were reversed in the order of importance, with








16 Florida Agricultural Experiment Stations

grapefruit ranking third in the low and medium income areas
and second in the high income area. This divergence between
the Memphis and Jacksonville results was influenced possibly
by the sale of Texas grapefruit in Memphis,2 which was pro-
hibited in Jacksonville,3 and because of relatively lower grape-
fruit sales during the first two months of the Jacksonville survey.
In Memphis, tangerines again were of little significance. The
decline in sales as income increased was more pronounced than
in Jacksonville, falling from 5 percent to 0.5 percent from the low
to the high income areas, respectively.

RANKING BY VALUE
Unlike the ranking by quantity, the value of apple sales in
Jacksonville exceeded sales of the other fruits by a considerable
measure. In the low and medium income areas, apples con-
stituted over 50 percent of the total value of the four fruits sold
and 45 percent in the high income area (Table 3).
As the second most important fruit with refenence to value,
oranges accounted for 34, 32 and 29 percent of fruit receipts
in the high, low and medium income areas, respectively. Grape-
fruit ranked third on a value as well as a volume basis. The
percent of total fruit sales attributable to grapefruit in the high
income area was approximately 11/ times the corresponding
percentages in the low and medium income areas. Tangerines
were responsible for roughly 4 percent of the total value of the
four fruits sold in each division of the market.
Even though apple and orange sales were of equal value in the
Memphis low income area stores, revenue from oranges was
higher in the other two areas (Table 4). Moreover, the rela-
tive value of apples appeared to decrease with an increase in the
income status of consumers, while the returns from oranges
increased.
Next in line of importance were grapefruit. Data from both
Memphis and Jacksonville indicated that fresh grapefruit sales
were positively correlated with level of consumer income. How-
ever, the value of grapefruit relative to the other fruits under

Sales of Texas grapefruit were probably much lower than normal
because the 1949 and 1951 freezes had drastically reduced the supply of
Texas fruit.
"8 A quarantine is imposed by the State Plant Board of Florida to prevent
the introduction of dangerous citrus diseases into the state.
For particulars see Special Bul. No. 20, State Plant Board of Florida,
Office of Plant Commissioner, Gainesville, Fla.







Economic Relationships in Retailiny Citrus Products 17

consideration was considerably higher in Memphis than the latter
city. The longer time period covered in the Jacksonville study
and the inclusion of Texas grapefruit sales in the Memphis data
would of course have some bearing upon the outcome.
The proceeds from tangerines were negligible in the high in-
come area stores of Memphis, but were somewhat larger in the
other two areas.
PRICES
A comparison of the prices of the different fruits shows that the
price of both oranges and grapefruit averaged about 6 cents per
pound in all three income areas of Jacksonville. The price of
apples was twice this amount, running a little better than 12
cents per pound in each area. Tangerines averaged roughly 7, 9
and 10 cents per pound in the low, medium and high income
areas.
Florida oranges were priced at approximately 7 cents per
pound for the period of study in Memphis, or nearly a cent
higher than for the previous season in Jacksonville. California
oranges sold in Memphis were several cents higher than Florida
oranges, averaging in the neighborhood of 10 cents per pound.
Florida grapefruit were about 6.5 cents per pound in the low
income area and 7.0 to 7.5 cents in the medium and high in-
come areas. Prices of Texas and Florida fruit was comparatively
the same.
Tangerines displayed a wide difference between income areas,
ranging from about 7.5 cents per pound in the low to nearly
11.5 cents in the high income areas.

RELATIVE IMPORTANCE OF SELECTED PROCESSED
JUICES IN THE DIFFERENT INCOME AREAS IN
TERMS OF QUANTITY AND VALUE
Data relating to both total quantity and value of each pro-
cessed juice dealt with in the study are given in Tables 5
and 6. The prices shown are for an equivalent number 2 con-
tainer of each product.
The relative importance of each product may be determined
by comparing the proportional values relating to the quantity
and proceeds from each item sold.
An examination of the juice sales in the two cities reveals
the particularly interesting behavior of orange juice and orange
concentrate. Orange juice sales were inversely correlated with








18 Florida Agricultural Experiment Stations

consumer purchasing power in each city; whereas, orange con-
centrate displayed a strong positive correlation.

TABLE 5.-THE RELATIVE IMPORTANCE OF SALES OF SELECTED CANNED
JUICES IN VARIOUS INCOME AREAS
(18 Retail Food Stores, Jacksonville, Florida, October 1949-June 1950).
No. of Per- Price I Per-
Cans cent Per Total cent
Product (No. 2 of No. 2 Value of
_Equiv.) Total Equiv. Total
Low Income Area

Orange juice ......... 19,275.3 23.6 $0.1515 $ 2,920.07 24.7
Orangeade .......-...... 13,542.2 16.6 0.1098 1,487.27 12.6
Tomato juice .......... 11,878.9 14.6 0.1235 1,467.05 12.4
Grapefruit juice ...... 11,561.7 14.2 0.1543 1,783.42 15.1
Pineapple juice ....-. 8,859.0 10.9 0.1674 1,482.90 12.5
Prune juice .........- .. 5,368.2 6.6 0.1711 918.25 7.8
Blended juice .......... 4,654.9 5.7 0.1394 648.80 5.5
0. concentrate ...... 3,965.2 4.9 0.2063 818.14 6.9
Apple juice .............. 2,381.7 2.9 0.1264 301.09 2.5

Total ... ......... 81,487.1 100.0 $11,826.99 100.0
Medium Income Area

Orange juice .......... 17,706.2 21.9 $0.1591 $ 2,716.08 22.3
Tomato juice .......... 14,549.4 18.0 0.1322 1,922.79 15.8
0. concentrate ...... 8,913.7 11.0 0.2107 1,878.16 15.4
Grapefruit juice ...... 8,852.8 10.9 0.1506 1,333.60 10.9
Orangeade ...........-- .. 7,578.4 9.4 0.1021 773.70 6.3
Pineapple juice ...... 6,371.1 7.9 0.1600 1,019.49 8.4
Prune juice ....-....... -6,275.0 7.7 0.1706 1,070.35 8.8
Blended juice ......... 5,612.0 6.9 0.1462 820.38 6.7
Apple juice ............ 5,064.4 6.3 0.1302 659.29 5.4

Total ............ 80,923.0 100.0 $12,193.84 100.0
High Income Area

O. concentrate ...... 37,152.6 29.0 $0.2124 $ 7,891.78 37.8
Tomato juice ......... 26,009.0 20.3 0.1297 3,373.96 16.2
Orange juice ............ 17,275.1 13.5 0.1496 2,584.14 12.4
Grapefruit juice ...... 10,793.3 8.4 0.1495 1,613.42 7.7
Pineapple juice ...... 9,007.8 7.0 0.1686 1,518.75 7.3
Apple juice .............. 8,972.8 7.0 0.1290 1,157.54 5.5
Prune juice ....--....... 8,005.6 6.3 0.1730 1,384.59 6.6
Orangeade .............. 5,565.2 4.4 0.1042 580.16 2.8
Blended juice ......... 5,297.1 4.1 0.1437 761.17 3.7

Total ............... 128,084.5 100.0 $20,865.51 100.0

Judging from the change in the distribution of sales of the
processed products, orange concentrate apparently enjoyed in-
creasing consumer approval over the lapse of time between the
two surveys, while orange juice experienced a substantial rela-







Economic Relationships in Retailiny Citrus Products 19

tive decline in favor. However, one should not overlook the fact
that orange concentrate as well as orange juice was several cents
lower for an equivalent container in Memphis than for the
previous season in Jacksonville. This change in the pattern
of consumption between the two cities may be evidence of a shift
in demand, and possibly, that orange concentrate, as compared

TABLE 6.-THE RELATIVE IMPORTANCE OF SALES OF SELECTED CANNED
JUICES IN VARIOUS INCOME AREAS
(9 Retail Food Stores, Memphis, Tennessee, January-June, 1951).
No. of Per- Price Per-
Cans cent per Total cent
Product (No. 2 of No. 2 Value of
SEquiv.) Total IEquiv. Total
Low Income Area

Orange juice ..... 14,723.6 16.4 $0.1139 $ 1,677.42 14.7
Grapefruit juice .. i 11,266.4 12.5 0.1050 1,183.11 10.3
Blended juice .......... 5,635.6 6.3 0.1138 641.43 5.6
Tomato juice .......... 19,484.4 21.7 0.1106 2,155.29 18.8
Pineapple juice ...... 8,497.8 9.4 0.1503 1,277.51 11.2
Prune juice ........... 2,039.1 2.3 0.1960 399.61 3.5
Apple juice ........ ..... 1,720.4 1.9 0.1260 216.85 1.9
O. concentrate ...... 18,588.9 20.7 0.1621 3,014.13 26.4
Orangeade ........... 7,876.3 8.8 0.1099 865.73 7.6

Total ........... .. 89,832.5 100.0 $11,431.08 100.0
Medium Income Area

Orange juice .......... 8,946.4 10.0 $0.1268 $ 1,134.01 9.5
Grapefruit juice .... 7,645.2 8.5 0.1127 861.38 7.2
Blended juice ......... 4,150.4 4.6 0.1224 507.97 4.2
Tomato juice .......... 17,862.7 19.9 0.1186 2,118.75 17.7
Pineapple juice ...... 5,258.0 5.8 0.1642 863.49 7.2
Prune juice ....-......... 2,724.0 1 3.0 0.1920 523.14 4.4
Apple juice ........... 2,131.4 2.4 0.1280 272.88 2.3
0. concentrate .... 36,363.2 40.5 0.1426 5,183.70 43.3
Orangeade ............ 4,781.4 5.3 0.1042 498.03 4.2

Total .............. 89,862.7 100.0 $11,963.35 100.0
High Income Area

Orange juice .......... 6,382.8 8.7 $0.1348 $ 860.54 8.2
Grapefruit juice .... 7,133.8 9.7 0.1104 787.35 7.5
Blended juice .......... 3,519.7 4.8 0.1221 429.86 4.1
Tomato juice .......... 13,889.8 18.9 0.1200 1,666.21 15.9
Pineapple juice ........ 3,818.8 5.2 0.1722 657.78 6.3
Prune juice ............ 3,635.2 4.9 0.1913 695.58 6.6
Apple juice ........ 2,524.6 3.4 0.1361 343.63 3.3
0. concentrate .... 30,960.0 42.1 0.1572 4,867.46 46.4
Orangeade ............ 1,702.6 2.3 0.1063 180.99 1.7

Total .-...... ...... 73,567.3 100.0 $10,489.40 100.0







20 Florida Agricultural Experiment Stations

with orange juice, has a greater cross-elasticity of demand for
other juices.
Tomato juice seemed to suffer little from the increase in con-
centrate sales. Out of nine juices, tomato juice was respon-
sible for 18 to 20 percent of total volume in the medium and
high income areas in both surveys and even gained in importance
in the low income area of Memphis as compared with Jackson-
ville. Prices were slightly lower, though, in the Memphis stores.
Grapefruit juice comprised a larger share of the juice pur-
chases of low income than higher income consumers in both cities.
A comparison of the Jacksonville and Memphis data shows that
grapefruit juice declined in relative importance for the low and
medium income groups in the latter study, but increased slightly
in the high income area.
Consumption of orangeade by low income consumers was es-
pecially high in the Jacksonville stores, accounting for almost
17 percent of the total volume of juices and ranking second
among all processed products considered. Generally speaking,
the proportional purchases of this product were about twice as
great at all income levels in Jacksonville as in Memphis. In
both cities, the relative importance of orangeade varied inversely
with the income status of consumers.
For the most part, the percentage of total sales accounted
for by the other miscellaneous juices was less in Memphis than
in Jacksonville.
The above analysis has pointed up the remarkable growth in
importance of orange concentrate as a beverage. The fore-
going results give the impression that orange juice and orange-
ade have borne the brunt of the telling force of competition from
orange concentrate. Although other miscellaneous juices were
affected, they were influenced to a lesser degree.
Finally, an observation of these relationships suggests that
price and income status had an important bearing upon the rela-
tive consumption of processed citrus products. Individuals from
low income areas consumed comparatively more orange juice
than those in the higher income areas; the reverse was true for
orange concentrate. A comparison of the prices of these two
products shows that in Jacksonville orange juice averaged 15
to 16 cents per equivalent number 2 container for the three
income areas, against an average of 21 cents for orange concen-
trate. Comparable figures from Memphis were 11 to 13.5 cents
for orange juice and 14 to 16 cents for orange concentrate.








Economic Relationships in Retailiny Citrus Products 21


Percent Low Income Area


75


50


25



2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks
Percent Medium Income Area


75


50







2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks






















Fig. 5.-Proportional distribution of quantities of oranges, grapefruit
Percent sold within each Income Area


75 1949 and January 2-June 23





25 '



2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks

Oranges Grapefruit* Apples*
'Converted to pounds for comparison.

Fig. 5.--Proportional distribution of quantities of oranges, grapefruit
and apples sold within each income area (Jacksonville, October 31-December
16, 1949, and January 2-June 23, 1950).









22 Florida Agricultural Experiment Stations

ORANGES
GRAPEFRUIT APPLES

.09- -.15

.08---- . .14


.076 \ ./-.1




.04 "- -.10

.03 .09
Low Income Area
T I I I I I I I I I I I I II I .' I TI
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
19

.07
*7- / \ 18



.06



.05.\ .13
\ .... .. .... 1 2
.12
.. 11
.04 ,' 10
"Medium Income Area
0-- I I I , I I , I .... I O
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
.17

.08 ..- 16

.07 / -.15

.06 .14

.05-\ 13

.04- .12

.03- --- -.11
.10
High Income Area
II I i I i I a I I i i 0

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
Oranges ---- Grapefruit -- Apples .........
Price per Pound
Fig. 6.-Weekly weighted average prices of oranges, grapefruit and
apples by income areas (Jacksonville, October 31-December 16, 1949, and
January 2-June 23, 1950).







Economic Relationships in Retailiny Citrus Products 23

The large sales of orangeade in the low income area of Jack-
sonville, where the price of this product was low relative to the
price of orange juice, also demonstrate the effect of price appeal
among low income consumers.

CONSUMPTION PATTERNS FOR RIVAL PRODUCTS
Substitution in consumption engendered by price differences
plays an important role in establishing the demand for products
which generally serve a common use. Some idea of the interre-
lated demand for rival products can be gained by examining the
shifts in the pattern of consumption associated with changes in
relative prices over time. If inter-income area comparisons are
made, these results may also be used to detect any tendency for
the pattern of consumption to vary with the level of consumer
income. As a method of illustrating the foregoing relationships,
selected commodities were arranged in groups of three and an
intra- as well as an inter-income area comparison made of the
weekly proportional sales of each item.

ORANGES, GRAPEFRUIT AND APPLES
Figure 5 shows that the quantity of oranges purchased in
each income area of Jacksonville was relatively high for the
first and last few weeks of the period, as compared with grape-
fruit and apples. During the middle of the period, however,
grapefruit and apple sales increased proportionately at the ex-
pense of oranges. Quite likely, the sharp rise in orange prices
at that time was largely responsible for this development (Figure
6). Grapefruit and apple prices provided other contributing fac-
tors, because the increases in the prices of these two fruits for the
corresponding period were not nearly so pronounced as that of
oranges.
As for the effect of income upon the consumption pattern for
these three fruits, it is seen that the relative consumption of
oranges was roughly the same in each income area. Grapefruit
were more important in the high income area, as compared with
low and medium income areas, and apples were more popular in
the latter two areas.
Results obtained by applying the same technique to the Mem-
phis data are depicted in Figure 7. While the change was less
conspicuous than in Jacksonville, orange sales in the Memphis
stores were depressed during the period extending from the latter








24 Florida Agricultural Experiment Stations


Percent L ..
S ................ Income Area............

75






25


0
I . .. I '1 I I "'t" '" I i i ......
2 4 6 8 10 12 14 16 18 20
Weeks
"Perce :: .:..:.: Medium Incomc A ::












2 4 6... 1 1 48
75 7:A'Y........ .... ...


.50


25



2 4 6 8 10 12 14 16 18 20
Weeks

. . . . i Fi i:. iiiiii~.... .. ...... .
...o...........








2'5


0
2 4 6 8 10 12 14 16 18 20
Weeks

SOranges* [lI[[ Grapefruit* :: Apples*

*Converted to pounds for comparison.

Fig. 7.-Proportional distribution of quantities of oranges, grapefruit
and apples sold within each income area (Memphis, January 16-June 12,
1951).








Economic Relationships in Retailiny Citrus Products 25



ORANGES
GRAPEFRUIT APPLES
.08 .13


.07 -.12


", 06\\ -


.05 -.10

= 5 Low Income Area
0 i I 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS
.14
.08 /

/ .13

.07 // \

S~-/\ ^ .-.1 / -.12

06
"' % ... ... "", 11

.05 -
Medium Income Area
7I I - 1 - 1 -----,, i I I -- i il -- l i-- l -- l -- l -- r1- o
C II i I I i T
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS


.11 ,-...... 1i
10 .1
.14




".07 1



.04 High Income Area

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS

Oranges -- - Grapefruit ---- Apples -.-..--..
Price per Pound


Fig. 8.-Weekly weighted average prices of oranges, grapefruit and apples
by income areas (Memphis, January 16-June 12, 1951).






26 Florida Agricultural Experiment Stations

part of January into March of 1951. Following this interval,
orange sales rose relative to the sales of the other two fruits
included in the comparison.
Orange purchases exhibited a general tendency to decline pro-
portionately (Figs. 7 and 8) as the price of oranges rose relative
to the price of grapefruit and apples. This relationship is more
evident for periods when orange prices were relatively high for
a span of several weeks; for example, weeks 4 through 10.
The purchases of the three fruits in each income area of
Memphis followed the same general pattern as in Jacksonville for
comparable portions of the season. Grapefruit sales provided the
most noticeable exception by commanding a larger share of total
sales in each income area of Memphis. Oranges appeared to
enjoy a slightly higher standing in Memphis, but apples were
noticeably less popular. A comparison between income areas
shows oranges to be responsible for about the same proportion
of total sales in each area, with the share attributable to apples
decreasing and to grapefruit increasing with an increase in the
level of income.

ORANGES, ORANGE CONCENTRATE AND ORANGE JUICE
Changes in the relative consumption of oranges, orange con-
centrate and orange juice in Jacksonville are shown in Figure 9.
The shift in consumption of oranges and orange concentrate,
part of which is presumably seasonal, is most apparent in the
high income area. Proportional purchases of oranges would
probably have been higher during the middle of the period had
it not been for the high price of oranges prevailing during
that time (Figure 10). Orange consumption was high in all three
income areas relative to concentrate and orange juice. This was
especially true during the first six weeks of the study when
orange prices were low.
Orange juice assumed much the same pattern in the low in-
come area that was followed by orange concentrate in the high
income area. The consumption of orange juice increased pro-
portionately toward the end of the season, while orange con-
sumption decreased.
Orange juice purchases were important also in the medium
income area. The consumption of this item showed a relative
increase during the middle of the period, following a rise in
the prices of oranges and orange concentrate, and during the








Economic Relationships in Retailiny Citrus Products 27



Percent Low Income Area ...:. .. .





05








2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks
Percn ...I M .dl rm Inco."me Ar I. ,
"0














2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks

















Weeks

Oranges* Orange Concentrate* Orange Juice*
'Converted to No. 2 equivalent in juice.

Fig. 9.-Proportional distribution of quantities of oranges, orange con-
centrate and orange juice sold within each income area (Jacksonville, October







31-December 16 1949 6 18 950).
I ... ........ .. . .













2 4 6 8 10 12 14 16' 18 20 22 24 26 28 30








28 Florida Agricultural Experiment St tions


ORANGE JUICE
ORANGE CONCENTRATE ORANGES
.23
S22 .18
.21


.17-. / -.15
.16 .. 14
Isr / ,
.15


.15 / .-. -13
.14 /.12
.13.
.12 -\ 11
.11 Low Income Area. 10

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS

.229
.21 18

.19/ \ /
.18 / / 15
.17 / -.14
".16 / ..----- ...1... 3
15 -\ "- "" 12
14 .11
13 .10
.12- -.09
11 Medium Income Area .08
o~L-rI , I ,13 , , , | | , 7-"
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS

23 .20
.22 /1

.2019 .16
.191 .16
.18 .... ,'-- 14
"lL--. / .. ... 13
.15 .1
... ..-------... ... .

14 .11
.10
S.- .09
.12- --.0
.11 High Income Area .08
......... I "..........................-"-o
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
Oranges ---- Orange Concentrate Orange Juice ---........

Price per No. 2 Equivalent

Fig. 10.-Weekly weighted average prices of oranges, orange concen-
trate and orange juice by income areas (Jacksonville, October 31-Decem-
ber 16, 1949, and January 2-June 23, 1950).







Economic Relationships in Retailiny Citrus Products 29

last several weeks when consumers were faced with a diminish-
ing supply of the fresh product.
The difference in the pattern of consumption of these three
products between the income areas apparently was strongly
influenced by level of consumer income. As the level of income
fell, fresh oranges absorbed a larger share of the total sales
of the three commodities. As for the two processed products,
orange juice reacted in the same manner as oranges, but the
purchases of orange concentrate showed a decided tendency to
vary directly with the income status of consumers.
The distributional purchases of oranges, orange concentrate
and orange juice presented in Figure 11 relate to the data ob-
tained in Memphis. Over the 20-week period, prices evidently
had an important bearing upon the relative sales of the three
products. For the approximate periods, weeks 4 through 10 and
17 through 20 (Figure 12), when orange prices were high in
each income area, the percentage of sales accounted for by
oranges declined, that by orange concentrate increased. This
situation was not as manifest in the low income area, however,
where the price of orange juice had undergone an increase that
extended over the latter period. The sharp drop in the price of
orange juice and the corresponding increase in orange juice pur-
chases in the low income area from weeks 4 through 10 further
suggest that consumers in this area of Memphis were more re-
sponsive to changes in the price of this item. Apparently, con-
sumers in the medium and high income areas did not attach as
much importance to this product.
The contrast provided by Figures 9 and 11 gives some idea
of how readily consumers with moderate incomes have adopted
orange concentrate. In the low and medium income areas of
Jacksonville, orange juice sales dominated the purchases of
orange concentrate. On the other hand, orange concentrate
had a slight edge over orange juice in the low income area of
Memphis, and an overwhelming one in the medium income area.
Even in the high income area, orange concentrate displayed
some increase as compared with Jacksonville.
At first glance, one might conclude that the level of income
had little effect upon the proportional purchases of oranges,
orange concentrate and orange juice in the medium and high in-
come areas of Memphis. However, considering the fact that more
oranges from other areas were consumed relative to Florida
oranges in the high income area, and assuming that orange con-








30 Florida Agricultural Experiment Stations



Per en t 1m ........ ... ................ .... ..... ... . .. ............. ........... ."" .":












Low Income Area

2 4 6 8 10 12 14 16 18 20
Weeks



75


Si





........ .. ... .edium Income Area .... . ''.

2 4 6 8 10 12 14 16 18 20
Weeks













... HIh come Area
2 4 6 8 10 12 14 16 18 20
Weeks

P Orange* IIIIIlI .range Cocet ntrate* i........ Or..ange Juie
Converted to No. 2 equivalent in juice.


Fig. 11.-Proportional distribution of quantities of oranges, orange con-
centrate and orange juice sold within each income area (Memphis, January
16-June 12, 1951).
16-June 12, 1951).









Economic Relationships in Retailiny Citrus Products 31



ORANGES
ORANGE CONCENTRATE ORANGE JUICE
.19.13


18 -, .13

\.17



.16 /

Low Income Area
I 1 . . 1 1 0
01 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS

S20 16
S19 ---- ..15

.18- / \ .14
.17 ,. A .13
.16 .. "' .12




Medium Income Area
V III I I i l IIl I TO
01 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS

.20 .15
S19-



15 ," -. .1
.17


.113

.13 -
High Income Area
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
"WEEKS
Oranges ---- Orange Concentrate Orange Juice ........
Priceper No. 2 Equivalent

Fig. 12.-Weekly weighted average prices of oranges, orange concen-
trate and orange juice by income areas (Memphis, January 16-June 12,
1951).







32 Florida Agricultural Experiment Stations

centrate is a better substitute for juice type oranges from
Florida, the results are not too inconsistent with those obtained
in Jacksonville. Under these conditions a higher rate of sub-
stitution of orange concentrate for Florida oranges by high in-
come consumers would tend to be counterbalanced by their
higher purchase rate for California oranges. This could partially
explain why, in comparison with the medium income area, total
orange sales were comparatively larger relative to the combined
purchases of processed orange products in the high income sec-
tor of the market.
Without doubt, a large share of the responsibility for the
phenomenal growth in the utilization of orange concentrate as
a beverage-as evidenced by the Memphis data-can be traced
to its price behavior. During the course of the Jacksonville sur-
vey the price of orange concentrate ranged in the vicinity of 21
cents per equivalent number 2 container. By the following sea-
son in Memphis, the price had dropped to a level of 14 to 16
cents. Although there was a corresponding decline in the price
of orange juice, the drop in concentrate prices was more pre-
cipitous, thereby reducing the approximate price spread between
the two products from 6 cents to 3 cents.

ORANGE JUICE, ORANGE CONCENTRATE AND ORANGEADE
The most pronounced adjustments in relative consumption of
these three products as determined from the Jacksonville study
occurred in the low income area (Figure 13). A glance at the
figure shows that the relative consumption of orangeade in-
creased remarkably toward the latter part of the period. Orange
juice, on the other hand, showed a marked decrease. An appeal
to logic would suggest that the stability of orangeade prices
and the upsurge in the price of orange juice, portrayed in Figure
14, were to some extent accountable for this consequence. This
lends weight to the notion that price becomes more and quality
of substitutes less significant as consumer income declines.
The comparative sales of orange juice, orange concentrate
and orangeade in Jacksonville provide an excellent example of
how the purchasing pattern of consumers changes from one in-
come level to another. Needless to say, the higher priced prod-
ucts were relatively more important in the high income area,
while the reverse was true for the lower income areas.








Economic Relationships in Retailiny Citrus Products 33



Percent













2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks



75


50



2Mediuml Income Area (


2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks
Percent -

















Orange Juice* Orange Concentrate* Orangeade*
*Converted to No. 2 equivalent in juice.


Fig. 13.-Proportional distribution of quantities of orange juice, orange
concentrate and orangeade sold within each income areq (Jacksonville,
October 31-December 16, 1949, and January 2-June 23, 195022 24 26
October 31-December 16, 1949, and January 2-June 23, 1950).








34 Florida Agricultural Experiment Stations



ORANGEADE
ORANGE JUICE ORANGE CONCENTRATE
.18 i
17 ..23
16.22
.15 .22
.14 20
.13.
12 \ 19

S. ... ..*- ** ... 18
10 *-.. ....*-....." "-***,.. **-...-. ......*** /-.. - .... ,.,
S. =017
= = Low Income Area
I I I I. | I I I I I I I .I . . I I I --^
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
I
.16, \ .. ,,
..215
.14.21
.13 -. 20

.12 / -.19
.13
11-
.1 -...- **.***. *-. .* 18

.09 .. -. 17
0
08 Medium Income Area

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
.18 I
".17 .23
.16 ~~ .22

15 // ""
.14
.13-
.12 19


"10V . .'" ......' ... ......
S.09 High Income Area

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS
Orange Juice ---- Orange Concentrate-- Orangeade ..------
Price per No. 2 Equivalent



Fig. 14.-Weekly weighted average prices of orange juice, orange con-
centrate and orangeade by income areas (Jacksonville, October 31-December
16, 1949, and January 2-June 23, 1950).









Economic Relationships in. Retailiny Citrus Products 35



Perceat.. ... ..


75


50


25


0 Low Income Area
2 4 6 8 10 12 14 16 18 20
Weeks
Percent


75


50


25














o 0 nerde o Niotdium Income Areaeii
2 4 6 8 10 12 14 18 20
Weeks
Percent


75


50


25



2 4 6 8 10 12 14 16 18 20
Weelk

Orange Jice* Orange Concentrate* M US Orangeade*

*Convenrted to No. 2 equivalent in juice.



Fig. 15.-Proportional distribution of quantities of orange juice, orange
concentrate and orangeade sold within each income area (Memphis, January
16-June 23, 1951).








36 Florida Agricultural Experiment Stations





ORANGEADE
ORANGE JUICE ORANGE CONCENTRATE
.13- -------- .18
I--\

V. 17






V/ \.!
.,-, .". .




.10 /

Low Income Area
"1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS
.16
15

14 15

13 /
12 .-. 14
112
10 ---........ -. -------.--------"-""- .13

.09 .125
.09 Medium Income Area 25
o- I r I I I I II I I I I l 0T
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16. 17 18 19 20
WEEKS
15

.14 --.19
.13 .18

.12- .17
111
10 "---.< ... y^ -. 15

.09 -.14
..09 13

SHigh Income Area
I'I I I I I I I o0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS
Orange Juice Orange Concentrate Orangeade ........

Price per No. 2 Equivalent



Fig. 16.-Weekly weighted average prices of orange juice, orange con-
centrate and orangeade by income areas (Memphis, January 16-June 12,
1951).







Economic Relationships in Retailiny Citrus Products 37

With the exception that concentrate w s much more important,
the relationships with regard to the cons mption of orange juice,
orange concentrate and orangeade in he low income area of
Memphis tended to bear out the result obtained in the corre-
sponding area in Jacksonville (Figure 1 ). Purchases of orange
juice and orangeade reacted notably to changes in the prices of
these two products, which were of greater magnitude than fluc-
tuations in the price of concentrate ( figure 16). Moving in-
versely with price changes, orange ju ce sales were relatively
higher in the first than the latter pa t of the period. Tem-
porally, orangeade sales followed a re erse trend.
In both the medium and high inco e areas, orange concen-
trate claimed the lion's share of total s les. Although no trends
in consumption were apparent in these two areas, a close exam-
ination of Figures 11 and 12 leaves t e impression that there
was some response to price movements
Differences in the pattern of consu option for the three prod-
ucts between income areas were not a striking in Memphis as
in Jacksonville. There was reason t expect this though, as
price differentials between the respe tive products were only
about one-half as much. In Jacksonville the price spread, first
between concentrate and orange jui e and secondly between
orange juice and orangeade, average 5 or 6 cents. In both
cases the spread was reduced to 2 or 3 cents in Memphis.

ORANGE JUICE, BLENDED JUICE A D GRAPEFRUIT JUICE
The proportional division of the p rchases of orange juice,
blended juice and grapefruit juice wi hin each income area of
Jacksonville is illustrated in Figure 1 Aside from short-time
fluctuations, there was no evidence of well-defined trend toward
a shift in relative consumption from any one of the products
to another. All three commodities maintained practically the
same relative importance.throughout he period. Likewise, the
general price movement for each pr duct was similar (Figure
18). But on occasions when the price of an individual item was
out of line with the others, the disparity was reflected in sales
of that product.
The distribution of the proportion purchases of the products
was essentially the same in each inco e area; i.e., there was no
inclination for the pattern of consum tion to vary with the level
of income.








38 Florida Agricultural Experiment Stations

Percent ji ow Income Ar J



Percent









4 6 8 10 12 14 16 t1 20 22 24 26 28 30
Weeks
Percent Medium Income Area








25 -


0 r
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Weeks
erce nt i High Income Ara













2 4 6 8 10 2 14 16 18 20 22 24 26 28 :30
Weeks

Orange Juice* Blended luice* Grapefruit Juice*
"*Converted to No. 2 equivalent in juice.
Fig. 17.-Proportional distribution of quantities of orange juice, blended
juice and grapefruit juice sold within each income area (Jacksonville,
October 31-December 16, 1949, and January 2-June 23, 1950).









Economic Relationships in Retailiny Citrus Products 39
ORANGE JUICE
BLENDED JUICE
GRAPEFRUIT JUICE

17 \ .-.... 17
16% -, .16
," '---.


.1 / - 15






11- Low Income Area I


2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS


.16 .- .16






12 .12
.11 .11

Medium Income Area
~L ~ I ,I i ] | i i ... . i I 1 |. I 0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS



.17\ .17



.14 .14
I/



















.12 .12


.11 .11
SMeHigh Income Area

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
WEEKS

Orange Juice---- Blended Juice Grapefruit Juice .

Price per No. 2 Equivalent
Fig. 18.-Weekly weighted average prices of orange juice, blended
juice and grapefruit juice by income areas (Jacksonville, October 31-
December 16, 1949, and January 2-June 23, 1950).








40 Florida Agricultural Experiment Stations


Percent -.:."
...........w ncome Area

75 :: :


50


25



2 4 6 8 10 12 14 16 18 20
Weeks
Pert ent
ceant g. uit's':.".::: o Med in Income Area ::::::::

75


50


2S



2 4 6 8 10 12 14 16 18 20
Weeks










*Converted to No. 2 equivalent in juice.
....juice and grapefruit juice sold within each income Arearea (emphis, January
7516June 12
,.. .. .. ...... ...










2 4 6 8 10 12 14 16 18 20
Weeks

~~ Orange Juice I Blended juice* Grapefruit Juice

*Converted to No. 2 equivalent in juice.

Fig. 19.--Proportional distribution of quantities of orange juice, blended
juice and grapefruit juice sold within each income area (Memphis, January
16-June 12, 1951).







Economic Relationships in Retailiny Citrus Products 41

Of all the commodity groups for which inter-city comparisons
were made, the combination under present discussion exhibited
the closest agreement. Notwithstanding the fact that there
were sizeable fluctuations in the relative consumption of the
commodities in the low income area of Memphis-attributable
no doubt to the wide swings in prices (Figures 19 and 20)-
the relative distribution of sales in all income areas of each city
was very similar.
Apparently these three products were associated one with
the other in much the same manner by consumers of different
income levels.

ESTIMATING THE ELASTICITY OF DEMAND FOR
FRESH ORANGES
Citrus producers have learned from experience that large crops
are accompanied by relatively low prices. Furthermore, in the
disposal of the crop for a given season, the price received for a
unit of the product may undergo considerable change as the
season progresses. With producers having nothing more than
a vague idea of what the demand conditions for a season will
be, attempts to establish the best marketing procedure reduce to
a process of trial and error. Then, there is no assurance that the
program adopted by the industry will maximize returns for the
season's production. A large element of uncertainty with re-
gard to maximizing returns could be avoided if more was known
about that aspect of demand having to do with consumer sensi-
tivity to price-or in economic terminology, price elasticity of
demand.
To facilitate interpretation of the following discussion, it is
appropriate at this point to comment briefly about the concept
of demand. Conditions of demand is a collective term relating
to all of the complex relationships affecting a consumer's de-
cisions and finally determining the quantity of a good that he
would be willing to purchase at various prices at a certain time
and place. In demand analysis this category of conditions is
taken as given, which implies that the individual's tastes and
preferences are fixed for the time being, the prices of important
rival products are known and unchanging, and that his real in-
come (command over goods and services in general) is constant.
The purchases of varying magnitude that the consumer would
make in response to various prices for the product, and subject to
the same set of demand conditions, are commonly referred to








42 Florida Agricultural Experiment Stations





BLENDED JUICE
GRAPEFRUIT JUICE ORANGE JUICE

12 .... / ------. \*13


12 12


.10 .11


09 Low Income Area

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS




13







.10 -.113
-.12


10
Medium Income Area
Q-C -I -- -I -- l -- -- l --[ -[ -i --[ -I -- l -- i i-- l -- -- l i-- -r-
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS




""14 \ .14

..........




Orange Juice---- rapefit Juice Blended Juice --
.10- -.12
.09-
^ High Income Area

1 2 3 4 S 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
WEEKS
Orange Juice ---- Grapefruit Juice--- Blended juice ...

Price per No. 2 Equivalent

Fig. 20.-Weekly weighted average prices of orange juice, grapefruit juice
and blended juice (Memphis, January 16-June 12, 1951).







Economic Relationships in Retailiny Citrus Products 43

as his demand for that item. The summation of the demands
of all individuals in a market at a given time gives rise to the
market demand. Demand is customarily thought of as referring
to the quantities of a product that would be taken at any price
falling within the range of relevant prices, and not to the amount
that would be purchased at a specific price. It is more con-
venient to think of the latter as the quantity demanded at a
particular price. It follows that as one moves along a typical
demand curve, the quantity demanded decreases as the price
rises and increases as the price falls, while the demand, as such,
is unchanged.
If conditions of demand undergo a change, the position of the
demand curve is shifted and the quantity taken at each individual
price increases or decreases, depending upon whether demand
has risen or fallen. It is readily apparent that to isolate the
effect of price upon the quantity of oranges purchased by con-
sumers, which is the primary concern of this section, it is de-
sirable that demand conditions remain stable. Otherwise, the
results would not reflect the effect of price alone, but would
also include the effects of factors that bring about shifts in con-
sumer demand.
In this study, as usually is the case, it was impossible to obtain
data that would ideally satisfy the requirements of theoretical
demand analysis. Hence, it was necessary to make certain as-
sumptions about the nature of the conditions and to improvise
adjustments for the data in order that they might approximate
the ideal. Since both the Jacksonville and Memphis studies cov-
ered relatively short periods, it is reasonably safe to assume
that in neither case did consumer tastes and preferences or
income change appreciably. An index of seasonal variation com-
puted from the weekly shipments of fresh oranges was used to
adjust the data for the effect of the changing array of sub-
stitute commodities throughout the season and the seasonal
changes in the purchasing habits of consumers.
For the demand analysis, adjusted total weekly sales of oranges
were taken as dependent upon the weekly average price of
oranges. These data were converted to logarithms from which
demand curves of constant elasticity were derived; i.e., the
elasticity of demand was the same for each point on a curve./
The demand curves for fresh oranges, as derived from data ob-
tained in the low, medium and high income areas of Jacksonville,
See Appendix 1 for a more detailed account of the analysis.











Price Low Medium High
090

.080 ..


070 \.


060


.050


.040 -


.030

.020.
Assumed demand function y = ax-1.206 Assumed demand function: y = ax-1329 Assumed demand function: y ax- 860
T I I I I I T T I I I I T I I I I -r
25 35 45 55 65 75 85 20 30 40 50 60 70 80 30 40 50 60 70 80 90 .
Volume of Sales (hundreds of pounds)

Fig. 21.-Demand curves for fresh oranges derived from weekly sales data in the low, medium and high income
areas (Jacksonville, October 31-December 16, 1949, and January 2-June 23, 1950).







Economic Relationships in Retailiny Citrus Products 45

are shown in Figure 21. An estimated coefficient of elasticity
of 1.206 was found for low income consumers in Jacksonville for
the 1949-50 season. This is interpreted to mean that a 1 percent
increase (or decrease) in the price of fresh oranges was as-
sociated with an approximate decrease (or increase) of 1.2 per-
cent in quantity purchased weekly. The coefficient of elasticity
of demand for the medium income group of consumers was found
to be 1.329. High income consumers in Jacksonville appeared
to have the lowest elasticity of demand for fresh oranges. The
estimated value of the elasticity coefficient for this group was
0.860. These figures indicate that the demand for fresh oranges
was elastic among low and medium income consumers but in-
elastic among high income purchasers. This suggests that with-
in the relevant segment of the demand curve a decrease in the
price of oranges (other things equal) would increase total reve-
nue from orange sales at the retail level in the low and medium
income areas, but would bring about a decrease in total revenue
in the high income area.
When applied to the Memphis data, the analysis yielded a
higher coefficient of elasticity for each income area than was
obtained for its respective counterpart in Jacksonville. (See Fig-
ure 22 for the Memphis demand curves.) However, the measure
of the demand elasticity of high income consumers relative to
that of low and medium income consumers. in Memphis- was re-
markably similar to results obtained in Jacksonville. The co-
efficients as derived from the Memphis data were 1.782, 1.665 and
1.143 for the low, medium and high income areas, in the order
mentioned. These coefficients indicate that the demand for
fresh oranges in Memphis, during the latter part of the 1951
season, was elastic in the low and medium income areas and
slightly elastic in the high income area. Offered as a partial rea-
son for the higher elasticity figures in Memphis is the increasing
acceptability of orange concentrate as a substitute for fresh
oranges. As a rule, the more numerous and better the substi-
tutes available to replace a particular product, the higher will
be the elasticity of demand for that product.
A more recent study, conducted under semi-controlled con-
ditions during the spring of 1952, furnishes a source of infor-
mation for checking the reliability of results obtained in Jack-
sonville and Memphis.5 Findings from the three analyses provide
S"Customer Response to Varying Prices for Florida Oranges" by Mar-
shall R. Godwin; Florida Experiment Station Bulletin 508; December, 1952.










Price Low Medium High


.080



.075-








.065




. 060



S055d Assumed demand function: y =ax- "665 Assumed demand function y = ax- I. C4
T I I I I I I I I
35 45 55 65 75 30 40 50 60 70 30 40 50 60 70
CO
Volume of Sales (hundreds of pounds)

Fig. 22.-Demand curves for fresh oranges derived from weekly sales data in the low, medium and high income areas
(Memphis, January 16-June 12, 1951).







Economic Relationships in Retailiny Citrus Products 47

an interesting comparison. Data from the 1952 survey yielded
an elasticity coefficient of 1.160, which is remarkably close to
the values of the coefficients obtained in the two earlier studies.
Although the demand for fresh oranges appears to be slightly
elastic at the retail level, indicating that a reduction in price
would result in an increase in total revenue to the retailer, this
does not mean that the same situation would occur at the
grower level. A hypothetical example will illustrate this point.
Assume that the elasticity of demand at the retail level is 1.2,
which means that if the price were changed by 1 percent, the
quantity taken would vary inversely by 1.2 percent. If, on the
average, marketing charges are relatively stable and at a cer-
tain time represent, say 70 percent of the retail supply price, the
elasticity of demand at the grove level would be only 0.36 as
compared with 1.2 at retail.6
Provided that, under these conditions, growers were forced
to rely entirely upon the fresh market as an outlet for their
fruit, the marketing of a continuously augmented production in
fresh form would lower, rather than raise, gross revenue to
the industry. This explains why the development of new mar-
keting outlets, through the introduction and perfection of pro-
cessed citrus products, has done much to relieve the downward
pressure on citrus prices.
Another example will demonstrate how an inelastic demand at
the grove level can precipitate large and abrupt changes in price
to the grower. The appearance and disappearance of substitute
fruit products of a seasonal nature, changes in prices of sub-
stitutes during the season, temperature changes, etc., create
changes in conditions of demand and thus give rise to shifts
in consumer demand for fresh oranges. Suppose that as a result
of such a short-run shift, the demand for oranges rose by 30
percent, i.e., 30 percent more oranges would be taken at each
price than had formerly been purchased. Assume also that the
supply marketed, because of the rise in demand, was increased
by 32 percent rather than 30 percent. Accepting, for purposes
of analysis, the elasticity coefficient of 0.36 at the grove level

Taking unity as the retail value of a box of fruit, a 1 percent change
in price would be equivalent to .01/1 = 1%. But, with the grower receiv-
ing only 30 percent of the retail price, the change in price at the grove
would be .01/.3 = 3.3%. Therefore a 3.3 percent change in price at the
grove would give rise to only a 1.2 percent change in the quantity taken
at the retail level, which is the same as saying that the elasticity of
demand at the grove level is approximately 1.2/3.3 = .36.







48 Florida Agricultural Experiment Stations

derived in the foregoing example, this would be indicative of a
0.36 percent change in the quantity taken as a result of a 1
percent change in price. Or stated in another way, this would
mean that consumers could be induced to increase their pur-
chases by 0.36 of 1 percent, only if the price were reduced by
1 percent. Thus, if the supply marketed were increased by 32
percent to satisfy a demand rise of 30 percent, the excess 2
percent marketed would effect a 5.6 percent price decline at the
grove.' Conversely, if the quantity marketed fell short of the
increased demand by 2 percent, the price would rise by approxi-
mately 5.6 percent.
The examples given above illustrate why a knowledge of the
elasticity of demand for citrus and a better understanding of the
purchasing habits of consumers are essential to the development
of a revenue maximizing marketing program for the citrus in-
dustry.

BEHAVIOR OF RETAIL PRICES FOR CITRUS PRODUCTS
Price manipulation is the means by which retail store opera-
tors attempt to maintain favorable inventory positions and the
device by which the flow of goods into the hands of consumers
is controlled in accord with current conditions of supply. How-
ever, the pricing practices followed in the marketplace are fre-
quently at variance with conditions that achieve these ends.
Much of this stems from the different attitudes that retailers
have regarding pricing techniques. Some retailers approach
pricing from the standpoint of historical cost; some insist on
comparatively fixed margins; while others attempt to make
adjustments to conform with the equilibrium price of the mar-
ket. Obviously these categories are not mutually exclusive.
One may actually find retailers following any one or a combi-
nation of these pricing policies.
Since actions on the part of retailers in this area have a ma-
terial effect on inventories and the rate at which products flow
into consumption channels, it is not inappropriate to examine the
price behavior for some of the more important citrus products
to the extent possible from the data obtained for this study.

If a 1 percent change in price is accompanied by .36 of 1 percent change
in the quantity taken, then an approximate change of 2/.36 = 5.6 percent
in price would be required to accommodate a marketing excess of 2
percent.







Economic Relationships in Retailiny Citrus Products 49

DISTRIBUTION OF SALES BY SELECTED PRICE INTERVALS
The distribution of sales of selected citrus items by bi-weekly
periods in both Jacksonville and Memphis is given in Figures
23 through 30.8 These distributions reveal clearly that the prices
which customers pay for individual citrus products vary con-
siderably over time and among retail stores during any given
period. Fresh oranges were sold in Jacksonville at prices ranging
from 1 to 10 cents per pound, and in Memphis at prices ranging
from 5 to 9 cents per pound; orange concentrate prices ranged
from 22 to 33 cents per 6-ounce can in Jacksonville and from
11.5 to 27 cents in Memphis; prices for orange juice varied from
21 to 55 cents per 46-ounce can in Jacksonville and from 24.5
to 47 cents per can in Memphis; and canned grapefruit juice in
46-ounce containers varied in price from 23 to 49 cents in Jack-
sonville and from 22.5 to 40 cents in Memphis. In each instance
part of the range in prices may be logically ascribed to variations
in quality of the product among the retail outlets studied. How-
ever, there is little question that much of the variation reflects
differences among retailers regarding their merchandising and
pricing policies.
As a general rule, price ranges for specific products were wider
in the Jacksonville than in the Memphis stores. This difference
may be attributed primarily to the fact that the stores included
in the Jacksonville phase of the study covered a wide range of
sizes and managerial arrangements, while the group studied in
Memphis was relatively homogeneous in this respect.
Certain of the products included in this analysis provide ex-
cellent illustrations of the behavior of prices and sales during
periods when price movements are taking place. During the be-
ginning weeks of the study in Jacksonville it is evident that the
retail price of canned orange juice was in the process of adjust-
ment to lower levels (Figure 23). The adjustment process was
characterized by an extremely wide range in retail prices and
comparatively small concentrations of sales at any particular
price level during the first six weeks of the study. In the 4th
period, price reduction on the part of some retailers was evi-
denced by the heavier concentration of sales at the lower price
levels. During the 5th period there was some evidence of an
upward shift in prices. By the 6th period, the price range
s Numbers appearing in the charts express the percent of total sales
occurring during a given bi-weekly period and within a particular price
interval.







50 Florida Agricultural Experiment Stations

narrowed somewhat when some retailers made upward revisions
in their orange juice prices and by the 7th period a condition of
comparative stability was reached.

Cents Per
Can -
53-56.9 0.1 0.2 0. O
0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1' 0.2 0.1 0.1 0.1
49-52.9 I
0.9 1.0 0. 6 0.1 0.2 0.2 0.1 0.1 0.2* *
45-48.9
I 1
1.4 0.1 0.2 0.4 0.8 1.1 0.5 0.4. 0.1
41-44.9

37-40.9 0.1 0.2 0.6 1.7 3.0 3.3 3. 4. 2 2 .

.233-36 1.2 0.9 0.5 2.4 4. 3.5 3.0 2. 2.2 0.4 0.4 3. 4.0
33-36.9

29-32.9 0.8 1.- 0.1 3.1 3,2 1.5 0.2 0.1 1.0

2.0 1. I .6
25-28.9 2.o 1.1 oI o.6
0.3 0.3 1.6 2.2 1.0
21-24.9
1 2 '3 4 5 6 7 8 9 10 11 12 '13 '14 15

Period
SLess than 1/10 of 1 percent.

Fig. 23.-Proportional distribution of total sales of orange juice in
46-ounce containers by price and bi-weekly periods (Jacksonville, October
31-December 16, 1949, and January 2-June 23, 1950).

In several other instances the data suggests wide variations
in the actions of retail store operators during periods when
supply and demand conditions were operating to establish a
new price level for the products under consideration. This situa-
tion is manifested by the distribution of sales into two or more
almost equally important price groups, as contrasted with the
tendency for sales to be largely concentrated in a single price
group under relatively stable market conditions. Examples of
this tendency toward a multi-modal distribution of sales among
several price levels may be noted in the case of the upward
adjustment of fresh orange prices in Jacksonville extending
from period 4 through period 6 and in the subsequent down-
ward price movement beginning in Period 11 (Figure 24). Simi-
lar price behavior during periods of instability may be observed
by an examination of the proportional distribution of the sales









Economic Relationships in Retailiny Citrus Products 51


Price Per Pound, I i I I


.10-.109 *


.09-.099 0.1 0 1 0.1 0.1 0.1 .1 I.

.08-.08 0.6 0.1 l 0.1 0.5 1.1 0.8 0.9 0.6 0.3 0.2 0.2 0.3

.7 0 0 1 0 3 0.1 06 2.0 3.8 3.6 4.3 3.9 2.0 0. 0.7 1.8 2.9


0 4 0 2 0 1 0.8 1.7 2.8 0.8 0.9 0.5 0.4 1.3 0.4 0.7 1.0 0.3
.06-.069

1 5 0.8 0.5 2.1 .7 0 3 0.1 0.1 1.9 5.3 .3 2.1 0.2
.05-059 0 3

04 .0 4.9 2.2 1. .3 0.3 *
.04-.0 4 6 2

03-.03 7 .4 87 0.2 0.1 *
.03-.039

.02-.029 0. I


1 1 2 1 3 4 5 6 7 1 8 1 9 10 11 1 12 13 1 14 1 15

Period
Less than 1/10 of 1 percent.

Fig. 24.-Proportional distribution of total sales of fresh oranges by
price and bi-weekly periods (Jacksonville, October 31-December 16, 1949,
and January 2-June 23, 1950).


Cents Per
Can

1 0.1 0.2 0.3 0.2 0.1 0.1 0.1 0.1 0.1
47-50.9.

3 0.2 0.3 0.2 0.4 0.2 0.3 0.3 0.2 0.1 0.1
43-46.9.

0.1 0.1 0.5 1.2 1 .2 1 1.5 1. 0 .0 1.0 0.9 0.4 3
39-42.9

0.4 0.6 0.6 1.8 2.1 3.5 4.9 5.7 4.9 .2 9 5.7 .r 16 8 1 4
35-38.9

31-34.9 0 0.7 1. 0.7 0.7 0.2 *1.2 0.


27-30.9 7.8 4.2 5.2 2.3 2 1
0.0
23-26.9 8 -9 I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Period

Less than 1/10 of 1 percent.

Fig. 25.-Proportional distribution of total sales of grapefruit juice in
46-ounce containers by price and bi-weekly periods (Jacksonville, October
31-December 16, 1949, and January 2-June 23, 1950).









52 Florida Agricultural Experiment Stations


Cents Per
Can I

0.0 0.9 0.9 1. 0 1.0 0.2
32-33.9

0.8 0.6 0.2| 4 2.4 3 .0 2.7 24 2.2 0 0 2 0.1 0.2
30-31.9 a


.3 2.1 0.1 0.1 0.4 0.4 1.8 1.9 1.2 1.8 4.4 50 o .0
28-29.9


0.9 3.0 4.1 0.9 4.1 2.1 0.4 0.1 0.2
26-27.9


0.2 1 0.3 .0.6 0.9 0.5 0.1 0.1 0.3 0.2 0.1 2.4 0.4 2.3 3.7
24-25.9


0.7 4.0 0.2 06 0.4 0.2
22-23.9 I- I ,I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Period
Less than 1/10 of 1 percent,

Fig. 26.-Proportional distribution of total sales of orange concentrate
in 6-ounce containers by price and bi-weekly periods (Jacksonville, October
31-December 16, 1949, and January 2-June 23, 1950).




Cents Per
Can
0.3 0.2 0.2 0.1 0.1 0.3 0.2 0.1 0.1
45-48.9

41-44.9 1 I 0. 0.1 0.1 0.1 0.1 0.1 0.2


37-40.9 0.1 0,1 0.2 0.4 0.3 0.1 0.3 0.1

0.0 0.1 0.0 3.7 0.1 2.
33-36.9. 0 6.

29-32.9 .L 1 2.7 2.9 1.66 0 .

0.1 0.1 0.8
25-28.9

21-24.9 1 4 4 .9
1 2 3 4 5 6 7 8 9 10
Period
Less than 1/10 of 1 percent.

Fig. 27.-Proportional distribution of total sales of orange juice in 46-
ounce containers by price and bi-weekly periods (Memphis, January 16-
June 12, 1951).







Economic Relationships in Retailiny Citrus Products 53

of canned grapefruit juice (Figure 25) and frozen orange con-
centrate (Figure 26) in Jacksonville.
Since the Memphis stores were selected with the idea of
obtaining the maximum degree of homogeneity with regard
to store types and managerial practices, the distribution of sales
among several price levels tends to reflect variations in the
short-run merchandising practices of the several retailing or-
ganizations. For example, in a sales promotional program of
one chain, orange juice was featured at an especially low price
as a customer attraction during a period of comparative price
stability for this product in the market as a whole. This re-
sulted in a pronounced bi-modal distribution in the sales of
canned orange juice during the first five periods in which data
were obtained (Figure 27). Effects of special promotional activi-
ties on the distribution of sales may also be noted for grapefruit
juice in the second bi-weekly period (Figure 28).
Sales of frozen orange concentrate in Memphis stores were
divided between two distinct groups during the first 16 weeks of
the study (Figure 29). To a large extent this is a reflection
of price differences associated with either real or imaginary
variations in the quality of frozen orange concentrate, and stems
from fundamental differences in policies followed by packers
and distributors of this product. Certain brands of frozen con-
centrate are advertised extensively and on a national basis, while
others (including packer, distributor and the private brands
of chain stores) are brought to the attention of the customer
through local advertising or, more commonly, through specific
merchandising activities within the retail store. The latter
group depends primarily on price as the major selling point,
while the former group commands premium prices due to prefer-
ences created by a sustained program of advertising designed
to make and keep a segment of the market brand conscious.
During the 9th and 10th bi-weekly periods covered by the
Memphis study, the bi-modal distribution of frozen orange con-
centrate sales tended to disappear. Consumer purchases during
these two periods were rather well distributed over the prevailing
price range, which was characterized by an approximate differ-
ence of 12 cents between the lowest and highest priced unit.
This condition is the result of short-run variations in the mer-
chandising policies of the organizations included in the study.
Both advertised and non-advertised brands were featured as
"specials" in some stores, while others of each type remained









54 Florida Agricultural Experiment Stations


Cents Per
Can

0.5 0.6 0.6 0.5 0.3 0.3 0.1 0.1
39-42.9


0o.1 0.3 0.2
35-38.9.


0.1 0.9
31-34.9



27-30.9 r
*5.2 I 2.1 2.8 B.1 5.3 *o 5_5 S.7



2 6 4 8 .4 1.3 8.0 0.2 3.2
23-26.9 2



19-22.9
1 2 3 4 5 6 7 1 8 9 10

Less than 1/10 of 1 percent. Period

Fig. 28.-Proportional distribution of total sales of grapefruit juice in
46-ounce containers by price and bi-weekly periods (Memphis, January 16-
June 12, 1951).





Cents Per
Can

0.2 O. 0. 0 0.4 0.5 0.7 0.6
26-27.9
2 -2 .9 1.4 1. 1.8 2.o 1.9 1.5 1.0 2.1 2.1

4.8 4.1 2.5 2.5 1.7 1.9 2.2 2.3 2.3 1.3
22-23.9
0. 0I1" 0.1 0.3 0.1 0' 0.3 1.4
20-21.9
18-19.9 15 1.9 1. 3.6 43 2.7 1.0
0.7 5. 2.1 4.4 1.1 0.8 2.0
16-17.9


12-13.9
0.3 0.8
10-11.9
1 1 2 1 3 1 4 1 5 1 6 1 7 8 9 10
Period
Less than 1/10 of 1 percent.

Fig. 29.-Proportional distribution of total sales of orange concentrate
in 6-ounce containers by price and bi-weekly periods (Memphis, January 16-
June 12, 1951).







Economic Relationships in Retailiny Citrus Products 55

at the normal market price. As a result, the distinct price
cleavage between the two groups became less evident.

RELATIONSHIP BETWEEN PRICE DISPERSION AND SALES
The material presented above demonstrates the existence
of relatively large differences in the retail prices of specific prod-
ucts among stores in the same market. Under certain condi-
tions, a system of widely dispersed prices tends to reduce the
quantity of a product that is sold through a group of retail outlets
in a market.
Abstracting from the fact that an individual retailer cannot
act independently but must consider the effects of his actions
upon other retailers and realize that their actions affect the de-
mand for his goods, it can be shown that in some instances more
of a product could be sold through a market if a single price
prevailed than if each retailer attached a different price to the
commodity. Suppose that each store in a market is faced with
a demand curve of equal elasticity for a particular product, that
the demand curves are all elastic in the segment which includes
the price range of the product in the market, and that the co-
efficient of elasticity is constant over this range of prices. If
under these circumstances each retailer would change his price
to conform with the current weighted average price for the
market, the quantity of the product sold and total revenue for
the market would be increased." Thus, more of the product
could be moved through the market while at the same time con-
sumers as a group would be paying the same average price that
they had previously. A demand of unitary elasticity would
leave the quantity sold and total revenue unchanged. On the
other hand, an inelastic demand would actually reduce sales and
revenue.
The significance of this relationship to citrus producers is
obvious. If it can be established that the demand for oranges
is elastic at the retail level, it seems that it would be in the
interest of producers to encourage retailers to narrow the range
of prices about the mean price for the market. In this way
more oranges could be sold in fresh form without necessarily
lowering the average retail price level for fresh fruit. Two
examples presented in Tables 7 and 8 have been prepared to
illustrate this point. The total sales of fresh oranges for the

g Algebraic proof of this contention is given in Appendix II.








Price Per Pound I I I


.09-.099 0.2 0.3

0.1 0.2
.08-.089 J


.07-.0793.1 6.5 8.5 7.4 4.8 5.1 2. 5 .2

7.7 2.7 1.0 2.4 5.9 2.6 4.8 6 1.4
.06-.069

0.6 0.6 0.3 0.1 1,7 3.2 2.5 0.1
.05-.059

0.3 0.3
.04-.049


.03-.039
1 2 3 4 1 5 6 7 8 9 10
Period
Less than 1/10 of 1 percent.

Fig. 30.-Proportional distribution of total sales of fresh oranges by price and bi-weekly periods (Memphis, January
16-June 12, 1951).








Economic Relationships in Retailiny Citrus Products 57

TABLE 7.-ACTUAL SALES OF FRESH ORANGES AND HYPOTHETICAL SALES
COMPUTED BY USING THE WEIGHTED AVERAGE PRICE FOR THE MARKET,
JACKSONVILLE, FLORIDA.* (OCT. 31-DEC. 16, 1949 AND JAN. 2-JUNE 23,
1950)
Hypo-
thetical Percent Change in
Price Total Percent Sales at of Total Percent-
Interval Sales of Wt. at Wt. age
(Cents) Total Average Average Distr.
Price* Price** of Sales
pounds pounds
1 1.9 72.0 .02 11.8 .01 .01
2 2.9 2,676.0 .62 861.6 .20 .42
3 3.9 73,414.6 17.05 36,968.2 8.59 8.46
4 4.9 58,277.1 13.54 40,981.6 9.52 4.02
5 5.9 99,164.3 23.04 91,047.7 21.15 1.89
6 6.9 52,325.9 12.15 59,988.3 13.93 + 1.78
7- 7.9 117,674.1 27.34 163,158.5 37.90 + 10.56
8 8.9 24,534.9 5.70 40,175.4 9.33 + 3.63
9 9.9 2,281.5 .53 4,331.1 1.01 + .48
10-10.9 56.3 .01 122.1 .03 + .02


Totals 430,476.7 100.00 437,646.3 101.67 + 1.67

-1.329
Assuming the demand function to be: X = ay
** Expressed as percent of actual sales.


TABLE 8.-ACTUAL SALES OF FRESH ORANGES AND HYPOTHETICAL SALES
COMPUTED BY USING THE WEIGHTED AVERAGE PRICE FOR THE MARKET,
MEMPHIS, TENNESSEE.* (JAN. 16-JUNE 12, 1951)

SHypo- I
Per- thetical Percent Change in
Price Total cent Sales at of Total Percentage
Interval Sales of Weight Iat Wt. IDistr. of
(Cents) Total Average Average Sales
SPrice* Price**
pounds pounds
4 -4.9 1,530 .56 749.3 .27 .29

5- 5.9 24,730 8.98 16,915.6 6.15 2.83
6 6.9 112,516 40.88 101,643.6 36.93 3.95

7 7.9 133,996 48.69 153,616.4 55.82 + 7.13
8 -8.9 1,095 .40 1,546.2 .56 + .16

9 -9.9 1,350 .49 2,294.1 .83 + .34


Total 275,217 100.00 276,765.2 100.56 + .56

-1.665
Assuming the demand function to be: X = ay
** Expressed as percent of actual sales.







58 Florida Agricultural Experiment Stations

sample stores in Jacksonville and Memphis were recomputed
using the weighted average prices and assuming a constant elas-
ticity coefficient of 1.329 and 1.665, respectively, for each
group of stores as derived for the medium income areas in the
demand analysis of the preceding section.10
Although the quantity sold will be larger in any event if the
demand is elastic and all of the prices for the product are
changed to agree with the weighted average price for the mar-
ket, the extent of the price range and the distribution of sales
over this range will affect the degree to which sales increase. In
Memphis the demand for oranges was more elastic than in
Jacksonville. Yet, when the data were treated as if the weighted
average price had prevailed in the market, indications are that
sales in Jacksonville would have been increased by 7,170 pounds
or 1.67 percent, while sales in Memphis would have been in-
creased by only 1,548 pounds or 0.56 percent. The wider price
range in Jacksonville, coupled with the fact that there were size-
able concentrations of sales over a large portion of this price
range, more than compensated for the lower demand elasticity in
this city.

COMPETITIVE RELATIONSHIPS OF SPECIFIED COMMODI-
TIES AS REFLECTED BY RELATIVE PRICES AND
RELATIVE QUANTITIES PURCHASED
Since the marketing of citrus fruit is characterized by the
use of several product forms which may be substituted rather
freely in consumption, producers are confronted with the matter
of ascertaining the most profitable utilization of the entire crop.
However, substitution may take place not only between the
various citrus products but also between citrus and certain non-
citrus items, all of which contribute to the complexity of the
problem.
While it is recognized that prices of all products utilized for
the same general purpose affect the consumption of a particular
commodity of that group, the analysis in this section is confined
to an examination of the relationship between price and quantity
ratios of selected pairs of items." The behavior of the relative
"10 See Appendix II, Tables 3 through 6 for computation of weighted
average prices and hypothetical sales at these prices.
"1The ratios for two commodities A and B are expressed as follows:
price A quantity A
Price ratio, --- ; quantity ratio, ----- .
price B quantity B







Economic Relationships in Retailiny Citrus Products 59

consumption of two commodities in response to changes in their
relative prices is assumed to serve as a fairly sensitive indicator
of the presence (or absence) of competition between them; al-
though the relationship, admittedly, involves the influence of
other prices, the level of consumer income, and other economic
factors.
Price and quantity ratios of the designated pairs of com-
modities were subjected to two analyses. The "rough test",12 in-
volving a comparison of the relative variation of the two series
of ratios for each pair of commodities, was first applied. The
rationale behind the use of this technique assumes that if two
commodities are highly competitive, a slight change in their rela-
tive prices would promote a substantial change in relative con-
sumption. On the other hand, if the items were complementary,
i.e., tended to be consumed jointly, a pronounced change in the
price ratios would be required to alter the quantity ratios to an
appreciable extent.
In the second part of the analysis, the regression of the
logarithms of the quantity ratios on the logarithms of the price
ratios was computed to determine the percentage changes in
the quantity ratio associated with a 1 percent change in the
price ratio.13 The factor time was included in the equation to
adjust for changes in the quantity ratios arising from seasonal
influences. A negative regression coefficient is indicative of
substitution in consumption. Although necessary, this condi-
tion is not sufficient. It is conceivable that a fall in the price of
one of a pair of commodities could lead to a rise in the con-
sumption of that commodity with no change in the absolute price
and consumption of the other item. 14 In this event, a negative
Henry Schultz, The Theory and Measurement of Demand, (Chicago,
Illinois, The University of Chicago Press, 1938) pp. 571-572.
"13log X lg a b log y + c t'
or
X = a yb et
where X = quantity ratio
y = price ratio
t = time in weeks using the mid-week of the series as the
origin
t' = t (log 10e)
The regression coefficient, b, denotes the percentage change in the
quantity ratio associated with a one percent change in price ratio.
The coefficient, 0 = c, when multiplied by 100 indicates the percentage
change in the quantity ratio associated with time.
For a critical treatment of the limitations of this technique see
Irving Morrissett, "Some Recent Uses of Elasticity of Substitution-
A Survey", Econometrica, Vol. 21, No. 1, January, 1953.
In this case, the analysis would yield an estimate of the price elas-
ticity of demand for the commodity undergoing the price change.










60 Florida Agricultural Experiment Stations



QUANTITY RATIO PRICE RATIO

220 7$

200 70



160- / \ 60

""\0/ \ /

120 50





60 5

Q 3D












IM -
20 -25

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25




















4C r 30
20--









I i I I I I L L L LI I I I I
'---























220 /7 P rI \ t
S20

TR 60


















160 -

140 50
120 -

I 2 3 4 5 6 7 1 9 10 1 12 I3 14 15 16 17 B 119 20 21 22 23 24 2

27-














Fig. 31-Prices and purchases of oranges relative to apples (both












converted to pounds for comparison) in low (top), medium (center) and
high (bottom) areas (Jacksonville, October 31-December 16, 1949, and
January 2-June 23, 1950, except first two and last three weeks omitted).
12/ -




122 -- 3a







Economic Relationships in Retailiny Citrus Products 61

regression coefficient would obtain, but one could hardly consider
the commodities as good substitutes. It would be more reason-
able under these circumstances to conclude that the commodities
were independent.
In addition to the analyses mentioned above,15 the movement of
relative prices and relative purchases of specified pairs of items is
examined graphically. In certain cases the depicted relation-
ships supported by the regression analyses indicate that the
relevant commodities are competitive.

ORANGES APPLES
Changes in relative prices and relative purchases of oranges
and apples in Jacksonville are portrayed in Figure 31. Toward
the first of the period the price of oranges was about 35 percent
of the price of apples. With this price relationship in effect,
consumers were purchasing about 1.4 to 1.8 pounds of oranges
to each pound of apples. During the following weeks the price
of oranges rose relatively, to about 55 to 60 percent of the price
of apples. At this level, about 3/4 to 1 pound of oranges was
purchased per pound of apples. Orange prices fell to 45 percent
of apple prices during the last few weeks of the period and
orange purchases relative to apple purchases rose to the ap-
proximate level that prevailed at the beginning of the study,
even though relative prices had been lower at that time. Perhaps
seasonal factors were responsible for this development.
To facilitate comparison, relative prices and purchases of
oranges and apples in the medium and high income areas of
Jacksonville are given in the same figure. The inverse rela-
tionship between the price and quantity ratios suggests that
these two products may be competitive to a certain degree. Re-
sults of the "rough test" show that the coefficients of variation
of the quantity ratios exceed the coefficients of variation of the
price ratios in each case. The regression analysis yielded co-
efficients of -1.4857, -1.7156 and -.8563 for the low, medium
and high income areas, respectively. Each regression coefficient
denotes the percentage change in the quantity ratio associated
with a 1 percent change in the price ratio. For example, in the

"1 Results of the "rough test" and the regression analysis for the product
combinations referred to in this bulletin, as well as a number of other
miscellaneous product combinations, are contained in a separate mimeo-
graphed Statistical Supplement which can be obtained by writing the Office
of the Director, University of Florida Agricultural Experiment Stations,
Gainesville, Florida.










QUANTITY RATIO PRICE RATIO


I r
210







0 \/

10 -



2 TIII I I 0
00 / /
"I \ / I










o I -
S I I I I I I I I Ir II I,rI t 0




















2 3 4 3 6 7 B 9 10 11 12 13 1 15 16 17 17 Ia I






Fi 3 .-- ries and ra s
240 -\ \0

1 \ / \
00 0



20.





150 0


4 2 3 04 5 6 7 B 9 o 11 10 1 14 15 16 107 18 1



















W3Wlr

Fig. 32.Prices and purchases of oranges relative to apples (both
converted to pounds for comparison) in low (top), medium (center) and

first week omitted).
first week omitted).







Economic Relationships in Retailiny Citrus Products 63

low income area the ratio of orange purchases to apple pur-
chases changed inversely by approximately 1.5 percent for a 1
percent change in the ratios of price of oranges to price of
apples. The regression coefficients for the medium and high
income areas are to be interpreted similarly. The high corre-
lation of the price and quantity ratios and the size of the co-
efficients (-1.4857 and -1.7156) obtained for the low and
medium income areas suggest that consumers in these income
categories may consider oranges and apples as competitive prod-
ucts. This implies that oranges would tend to be substituted for
apples, or vice versa, dependent upon the direction of change in
the relative price of the two commodities. Although the corre-
lation between the two ratios is fairly high in the high income
area, the regression coefficient (-.8563) seems rather small for
supporting a contention that the commodities are very compe-
titive.
The relative purchases and relative prices paid for oranges
and apples in the three income areas of Memphis are shown in
Figure 32. It will be noted that for comparable relative price
levels, generally more oranges were purchased per pound of
apples by similar income groups in Memphis than in Jackson-
ville. The coefficients, -1.6799 for the low, -1.1436 for the
medium and -1.4538 for the high income area, obtained from
the regression analyses for Memphis differed slightly in magni-
tude and order from those derived from the Jacksonville data.
The lower correlation between the two variables as determined
from the Memphis data, especially from those obtained in the
low and medium income areas, raises some question as to the
confidence that can be placed in the estimates. Taken alone,
however, the size of the regression coefficients lends support to
the intuitive feeling that the two products share a competitive
relationship.
GRAPEFRUIT ORANGES
The relationships depicted in Figure 33 leave the impression
that relative prices of grapefruit and oranges had a decided
influence upon relative purchases of the two products in Jack-
sonville. A comparison of relationships existing for the first
and last few weeks of the period, however, demonstrates the
additional effects of seasonal forces such as availability of each
product, quality, etc. A similar observation may be made about
the pattern of behavior between relative prices and relative pur-









64 Florida Agricultural Experiment Stations


QUANTITY RATIO PRICE RATIO




-120




30 1











I I I I I I L / ' I- _
/

























I -60
Fig. 33.-Prices and purchases of grapefruit relative to oranges (both









































concerted to pounds for comparison) in low (top), medium (center) and
high (bottom) income areas (Jacksonville, October 1-December 16, 1949,















and January 2-June 23, 1950, except first two and last three weeks omitted).
""2 i 5 6 ( S a 10 li 2 13 I 5 1( 17 18oIA 1O 20 A1 22 23 24 21































and January 2-June 23, 1950, except first two and last three weeks omitted).









QUANTITY RATIO NICE fP lIO




S110
I \ \ j\
7I \

/ \ I

60 \ 1f














20/ \ 0






110- In

--- 1 -1\0
I I I I I I I I I I I I I I T

I I

/ \



\\ _~ -11







50 4\/1 Quant ty grtelo 5



S- 85

D20 -P0
2D IW


110

1I .1 130
Price t r t org t








first week omitted).






10 .

-I I .- I ___ I I I I I I I I
1 2 1 4 5 6 7 B 9 io 11 12 13 1 I5 16 17 IB I1
WEEKS


Fig. 34.-Prices and purchases of grapefruit relative to oranges (both
converted to pounds for comparison) in low (top), medium (center) and
high (bottom) income areas (Memphis, January 16-June 12, 1951, except
first week omitted).







66 Florida Agricultural Experiment Stations

chases of the two products in Memphis as shown in Figure 34.
It is interesting to note that for a given general level of rela-
tive prices, the higher the income the larger the quantity of
grapefruit purchased per unit of oranges in both cities. But
oddly enough, for a given price relationship, purchases of grape-
fruit relative to oranges were generally higher in Memphis than
for corresponding income areas in Jacksonville. The growing
acceptance of concentrate and its influence upon fresh orange
consumption may have been partially responsible for this phe-
nomenon. The regression coefficients which express the per-
centage change in the ratio of grapefruit purchases to orange
purchases for a 1 percent change in the ratio of their prices are
as follows for both cities ordering from the low through the high
income area-Jacksonville: -1.6407, -2.2875, and -1.0206;
Memphis: -1.0288, -1.9025 and -2.1414. On the basis of
these coefficients, competition between grapefruit and oranges
appeared to be stronger among low and medium income con-
sumers of Jacksonville, but among medium and high income con-
sumers of Memphis. The weaker correlation found for each
income area in the Memphis analysis tends to impair this ap-
parent inconsistency.

ORANGES ORANGE CONCENTRATE
Because orange concentrate was a relatively new product dur-
ing the period of the Jacksonville study, not enough time had
elapsed for a normal competitive relationship to be established
between it and fresh oranges. Apparently, however, this rela-
tionship was tending toward stability by the following year,
as evidenced by the Memphis data (see Figures 35 and 36).
Orange prices rose relative to the price of concentrate and
orange purchases relative to concentrate purchases fell consid-
erably by the time of the latter study. The largest adjustment
in quantity of oranges purchased per unit of orange concentrate
occurred in the low and medium income areas, as shown by a
comparison of the charts for the two areas in each city.
Normally, about 15 units of oranges were purchased per one unit
of orange concentrate in the low income area of Jacksonville.
This ratio fell to approximately 2.5 to 1 in the low income area
of Memphis. This demonstrates the growing acceptance of
orange concentrate among all income classes.
Because orange concentrate had become better established
in the market by the time the Memphis phase of the study was










Economic Relationships in Retailiny Citrus Products 67

QUANTITY RATIO PiICE RATIO


5000 -Price Rio /R '

I i v \
I V \/ \
4000 \ 70






SQotity Ratio

1oM so



0 I I I I I I L I I I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1. 19 20 21 22 23 21 25
1200 I 1 1 I I II

1100 Price Ratio //

I ODO 75

900 \ 70

oo600. 65

4 ., \ ....


500 50





I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

700 - 40
1 2 31 J 6 9 1. lI U 13 1 1J 11 17J 1 1 V I


i0 Prie Ratio 9/ '


400 \ 70







300. 60 V ~ \< O ll _





I1 21 31 4, 51 61 71 8, 9- 101 ]1 12 '13 14, '15 116: 117. 118 119' 120 121 122 123 124 L25
WEEKS
Fig. 35.-Prices and purchases of oranges relative to orange concen-
trate (both converted to equivalent No. 2 cans of juice) in low (top),
medium (center) and high (bottom) income areas (Jacksonville, October 31-
December 16, 1949, and January 2-June 23, 1950, except first two and last
three weeks omitted).










68 Florida Agricultural Experiment Stations



QUANTITY RATIO PRICE RATIO
500 -
Price Ratio
A / I h"0
/ \ \

400 V



300-
#11 'i I I -

II \ \ \ 0




100-




I 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
\ it I I I
160- 160

140 Price Ratio 140

120 \ ~ / 120

100 \ 100

BO 80
00
60 60

40_ QuMtity Ratio 40

20 20


1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19

130

Price Rotio

200 11 120


S\ 10

\ // 00
IRS


100 90'

Ouortity Ratio

501 80
50,. 8
0T .

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
WEEKS

Fig. 36.-Prices and purchases of oranges relative to orange concen-
trate (both converted to equivalent No. 2 cans of juice) in low (top),
medium (center) and high (bottom) income areas (Memphis, January 16-
June 12, 1951, except first week omitted).









QUANTITY RATIO PRICE RATIO



140
600
130

120
500 Pric Rati.







80
woo .


/


/ ity Rti \ 60
200
50
O I I I I I I I I I i i l i i i i i i I i i i T o
1 2 3 4 5 6 7- 8 9 10 11 12 13 14, I5' 16 17 18 119 20 21 22 23 .24 25
SI I I I I I I
1000- / 105

0 Price Ratio \
900 / \
0/0 100





600 /\
N' 95




500 -
I s

400 J
75
30 O l^ I 't7o






If 21 31 41 5 61 7' 2 91 10111 12 13 14.15 16 17'1 1.19 ,20 21221 i23 24'25
11 P R 65



100





S0 I II
II 21 31 4 5( 61 7, 8 91 10 II 12 13 14 15 16 17 18 119 ,20 21 22 23 24 25


"Fig. 3.-Prices and purchases of oranges relative to orange juice (both130



00 high (bottom) income areas (Jacksonville, October 31-December 110

800om i

700 I 90

600 U 80

500- ^ 70

/ Ul' ^ _so






II 21 3| 4| 6J 71 8 9 10 11 1 13 14 15 16 17 18 I19 2 21 22 23 24 2
WEEKS


Fig. 37.-Prices and purchases of oranges relative to orange juice (both
converted to equivalent No. 2 cans of juice) in low (top), medium (center)
and high (bottom) income areas (Jacksonville, October 31-December 16,
1949, and January 2-June 23, 1950, except first two and last three weeks
omitted).










70 Florida Agricultural Experiment Stations




UANTITY RATIO PRICE RATIO
1000 IT


/ \/ lea ll/
am 1710

















I 2 3 4 5 6 7 0 9 10 II 12 13 14 15 16 17 iI 19
700











A/ / \ / \
600 2 1 A 10
I I I











1

1 \*I








\ e / 110
200 .I










201



7t 150
1oo








"00 / 10

















103
converted to equivalent No. 2 cans of juice) in (top), medium (center)



and high (bottom) income areas (Memphis, January 16 -June 1 1951, ex-
3100 1/50








Economic Relationships in Retailiny Citrus Products 71

undertaken, coefficients of regression derived from these data
probably provide better indicators of the competitive relation-
ship between the two products. Although the coefficient for each
income area in this city exceeded unity, the highest value,
-2.6714, was obtained in the medium income area. This sug-
gests that these consumers showed most response to a change
in the relative price of oranges and orange concentrate by way
of substituting one product for another.

OTHER PRODUCT COMBINATIONS (ORANGES ORANGE JUICE,
ORANGE CONCENTRATE ORANGE JUICE, ORANGE JUICE -
GRAPEFRUIT JUICE)
The remainder of the more important product combinations
listed above are considered collectively. In most cases there
was an inverse relationship between price and quantity ratios,
especially when the price ratio moved to a new level which was
generally maintained for a period of several weeks (see Figures
37 through 42). The most notable exception to the usual asso-
ciation occurred for orange concentrate and orange juice in Jack-
sonville where the price ratio of the two products was relatively
stable (Figure 39).

TABLE 9.-REGRESSION COEFFICIENTS FOR SELECTED COMBINATIONS OF
CITRUS PRODUCTS.

Product Combination Income Areas
i Low Medium High
Jacksonville, Florida

Oranges-Orange Juice ....... -1.0459 1.8782 .6455
Orange Concentrate-Orange
Juice ................................... ----- -- 2.2417 4.3119 .0156
Orange Juice-Grapefruit Juice.... 2.0346 3.1498 1.1783

Memphis, Tennessee

Oranges-Orange Juice ...........- 2.3770 1.3500 1.8972
Orange Concentrate-Orange
Juice ....................---------................. - 3.1762 3.6542 .1730
Orange Juice-Grapefruit Juice .... 4.7141 2.1929 2.3420


The regression coefficients contained in Table 9 imply that
oranges and orange juice were probably competitive in the low











QUANTITY RATIO PRItE RATIO







20 -20







25
20







.Pk, R. o




5 5
7 2 3 4 5 6 7 8 'S 10 11 TI2,5 1 14,112 1^ 1 A 19 1 20,121 I[2 123 '24 ]21'




100 20







60
80 .Price Ratio















0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 21 22 23 24 25



35
20


30




20
'5
151
10









1 2 3 4 5 6 7 8 9 10 11 12 13 1 115 16 17 19 20 21 22 23 24 25
WEEKS


Fig. 39.-Prices and purchases of orange concentrate relative to orange
juice (both converted to equivalent No. 2 cans of juice) in low (top), medium
(center) and high (bottom) income areas (Jacksonville, October 31-Decem-
ber 16, 1949, and January 2-June 23, 1950, except first two and last three
weeks omitted.)











Economic Relationships in Retailiny Citrus Products 73



CQUAN0 7 RATIO PRICE RATIO

275 210

250 no7





175 17o

50 1620

23, 7

7I5 ( \ i \ .1
75 -- 735





25 120
0 1
so A To 1


1 2 3 4 5 6 7 8 10 0o 12 13 14 15 M 17 8s 1I


950

375 140





B oo
725 -13

650 1

Ai
575 P 7 --1---- 1

W 7 I ./ -



















"455 \ 7 7 '
350















300 105
750 I00

















1 2 3 4 5 6 7 9 0 11 12 13 14 15 16 17 18 19
WEEKS
530 -- -'

7- 7












Fig. 40.-Prices and purchases of orange concentrate relative to orange

juice (both converted to equivalent No. 2 cans of juice) in low (top),
750 700











medium (center) and high (bottom) income areas (Memphis, January 16-
June 12, 1951, except first week omitted).









74 Florida Agricultural Experiment Stations


QUANTITY RATIO FRICE RATIO




150
25D


O R..y 125



















350 \














1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16. 17 10 19 20 21 22 23 24 25
0 1y300 R.1n oo
220,
200

S75












2 3 4 5 A ? 87 0 0 10 Ii 1 ,2 13 14 15 16 17 18 19 20 21 22 23 24 2
3 05



















Fig. 41.-Prices and purchases of orange juice relative to grapefruit
juice (both converted to equivalent No. 2 cans) in low (top), medium
(center) and high (bottom) income areas (Jacksonville, October 31-Decem-
ber 16, 1949, and January 2-June 23, 1950, except first two and last three

weeks omitted).
weeks omitted).









QUANTITY RATIO PRICE RATIO

140
350






S120
200 30



50 \ 10




I / \






















909
70 \


















0
114


1 3 4 5 6 7 B 9 10 11 12 13 14 15 16 17 18 19
















10130







S70 I ,
150
i/I P.t.R tt

100 ~ It

,70

060

























40 .f
i 2 3 4 5 6 7 98 10 11 12 13 14 13 16 17 8 19



























Fig. 42-Prices and purchases of orange juice relative to grapefruit
50 /





60 -






1 2 3 4 5 7 8 9 10 11 12 13 4 35 16 17 10 I




(center) and high (bottom) income areas (Memphis, January 16-June 12,
1951, except first week omitted).







76 Florida Agricultural Experiment Stations

and medium income areas of Jacksonville and in all income areas
of Memphis. In the low and medium income areas of both
cities, orange concentrate and orange juice appeared to be con-
sidered as substitutes. There is no evidence of this however,
in either of the two high income areas. A competitive relation-
ship seemed to exist between orange juice and grapefruit juice
for consumers of each income level in both cities.

ACKNOWLEDGMENTS
The writers are very appreciative of the cooperation received from the
management of the following firms who, through their interest and
assistance, contributed much toward the accomplishment of this survey:
Jacksonville phase of the study: The Great Atlantic and Pacific Tea
Company, Daylight Super Markets, Winn and Lovett Grocery Com-
pany, Setzers Super Stores, C and C Grocery Company, Farris Super
Market, Family Food Market, ABC Super Market, Pridgen's Food
Store, Fairfax Market, Meide's Market.
Memphis phase of the study: The Great Atlantic and Pacific Tea Com-
pany, Kroger and Company, John Gray and Son.
Particular recognition is accorded Dr. H. G. Hamilton, head, Depart-
ment of Agricultural Economics, University of Florida, for his efforts in
initiating and developing the project and his guidance in planning and
performing the subsequent work.
The tedious task of working out arrangements with the cooperating gro-
cery firms and collecting the data in the field was accomplished by Mr. Tall-
madge Bergen, Florida Agricultural Experiment Stations, and Messrs. K. A.
Fugett and H. M. Mayes, Jr., Texas Agricultural Experiment Stations.
The writers also extend their thanks to the many other persons who
participated in the preparation and completion of this work.







Economic Relationships in Retailiny Citrus Products 77


APPENDIX I

DERIVATION OF THE DEMAND FUNCTION FOR
FRESH ORANGES
The demand analyses made for fresh oranges involved several
underlying assumptions. Since no provision was made for
keeping a customer count when the studies were being conducted,
the number and class of customers patronizing each store was
assumed to be constant for the duration of the respective periods.
Moreover, as each study was confined to a portion of a particular
season, it appeared reasonable to assume that the general pur-
chasing power of consumers was relatively constant. Making
use of a more rigorous assumption, the share of income devoted
to the purchase of oranges and close substitute commodities was
considered to represent only a minor part of the consumer's
budget and hence the income effect of a price change in these
commodities was assumed to be negligible.
For several reasons, the original data were not in satisfactory
form for analysis. The problem of selecting a representative
price for the product in a given income area was dealt with by
using the weighted average price; i.e., each price in effect during
a week in the stores of a particular income area was weighted
by the quantity of oranges sold at that price. Choosing the
weighted average price for the analysis introduced new difficul-
ties. In some instances, the process of computing the weighted
average price for oranges in an income area gave rise to apparent
price changes, which in fact were not reflections of actual changes
in the prices of the product. Frequently, a store would offer
several lots of oranges at different prices. Even if the price
of each lot had remained the same for a period of several weeks,
the weekly weighted average price of oranges would have varied
as consumers shifted their purchases between lots from one
week to the next. The same principles applied to the problem of
aggregating the data for an income area, as the sales made by
each store may well have fluctuated under a condition of rela-
tively stable prices. Another matter that had to be considered
was the fact that purchases varied from week to week because
of factors other than price. Weather conditions, holidays, avail-
ability of the product in the various outlets, price and the
availability of possible substitutes, the promotional emphasis








78 Florida Agricultural Experiment Stations

used in selling the product and the periodicity of consumer in-
come receipts served to alter weekly purchases of oranges.
For the foregoing reasons, it was deemed more feasible to
use three-week moving averages of the prices and quantities in
order to smooth out some of the disturbing elements inherent
in the original data. In the analysis of the Jacksonville data,
the first two and the last three observations in each of the
series of moving averages were omitted. This step was taken
because certain difficulties had to be ironed out before reliable
data could be collected at the beginning of the study and because
orange selections became rather limited or not even available
in some stores at the end of the season.

TABLE 1.-INDEX OF SEASONAL VARIATION USED IN ADJUSTING THE
QUANTITY OF ORANGES SOLD IN BOTH JACKSONVILLE, FLORIDA, AND
MEMPHIS, TENNESSEE.

Period Index Period Index

1 101.1* 12 115.7
2 114.2 13 114.4
3 109.1 14 111.0
4 105.6 15 109.0
5 99.7** 16 105.9
6 105.0 17 105.5
7 111.4 18 100.1
8 113.6 19 95.2
9 109.2 20 88.5
10 110.2 21 80.8*
11 113.1 22 72.9**

Inclusive of the series of index numbers used to adjust the Jacksonville Data.
** Inclusive of the series of index numbers used to correct the Memphis Data.

To cope with the problem of seasonal shifts in consumer de-
mand for oranges, an index of seasonal variation was computed
from the total weekly shipments of fresh Florida oranges. Be-
fore calculating the index, these data, which consisted of weekly
shipments from the 1944-45 season through the 1950-51 season,
were summed over seasons by weeks and a three-week moving
average was computed for this series. The index of seasonal
variation shown in Appendix Table 1 was obtained by dividing
each observation in the series of moving averages by the arith-
metic mean of the series. The moving averages of store sales
data were then divided by the index for corresponding periods
of the season in an effort to eliminate the effects of seasonal
variation upon purchases. The adoption of this technique of







Economic Relationships in Retailiny Citrus Products 79

correcting for seasonal shifts in consumer demand seems justi-
fiable because of the manner in which citrus is marketed. For a
particular year the total production of citrus is given, and, ex-
cept for differences in the maturing seasons of certain varieties,
could be marketed rather quickly, or the process extended over a
relatively long period. Viewed in this light, one might look
upon changes in weekly shipments over the season as reflection
of shifts in consumer demand. The availability of fresh fruit
and vegetables, of a seasonal nature, would no doubt contribute
heavily to shifts in the demand for oranges. Hence, the index
when applied to the data would tend to adjust for the changing
array of substitute commodities over the season and changes
in the seasonal purchasing habits of consumers as referred to in
the text.
In subjecting the data to the demand analysis, the demand
function was assumed to be of the type
x = a yb,
where a and b are constants and y and x indicate prices and
quantities, respectively. When the data were converted to
logarithms, the demand function was transformed into the linear
form,
Log x = Log a + b log y.
The value of b for this type of function is identical with the
estimated coefficient of price elasticity of demand. A function
of constant elasticity was selected because it gave, in most cases,
a fairly good fit to the data and because of the narrow range
in market prices. Furthermore, it was more expedient to use
the same type function in each of the analyses for the sake
of consistency in making between-group comparisons of the
elasticity coefficients. Quantity was selected as the dependent
variable because it was felt that the price series contained less
error and because in the short run the consumer must take the
retail price as given. The derived demand equations for the
respective income areas of Jacksonville and Memphis are given
in Appendix Table 2.










TABLE 2.-CHARACTERISTICS OF THE DEMAND FUNCTIONS FOR FRESH ORANGES, JACKSONVILLE, FLORIDA, 1949-50, MEMPHIS,
TENNESSEE, 1951.
x' = Logs. of three-week moving averages of weekly sales in pounds (adjusted for seasonal variation)
y' = Logs. of three-week moving averages of weekly prices per pound (Figures in parentheses are standard errors)
.----------------------------------
Equation Income Equation Descriptive Constants
Number Area Constant Elasticity Adj. Standard Adj. Index of
Term y' of Demand i Error of Est.* Correlation*
Jacksonville, Florida, (Oct. 31-Dec. 16, 1949 and Jan. 2-June 23, 1950)

Low x' 4.609402 -1.206341 1.206341 .026570 .966996
(.068960)
Medium x' = 4.620807 1.329056 -- 1.329056 .062285 .857116
( .170770) 1

High x' 4.420642 .859643 .859643 .024544 .950013
(.061158)

Memphis, Tennessee, (Ja 16-June 12, 1951)

Low x' = 5.233062 -1.782313 1.782313 .037839 .718660
(.406753) I

Medium x' = 5.023138 -1.664995 1.664995 .021279 .925423
( .164539) [
High x' = 4.550821 -1.143151 1.143151 .033278 .664716
( .300411)

Corrected for the number of constants in the equation.







Economic Relationships in Retailiny Citrus Products 81


APPENDIX II

EFFECT OF SELLING AT THE WEIGHTED AVERAGE
PRICE FOR THE MARKET ON TOTAL SALES OF A
COMMODITY UNDER ASSUMED CONDITIONS
OF DEMAND'

Given a group of retailers offering a product at different
prices, one is sometimes confronted with the question of whether
total sales within a market would be larger or smaller if all
were to sell at the weighted average price. Restriction of the
problem further to a case where each retailer is confronted
by a demand curve of constant elasticity (p), and which differs
from the curves of other retailers only in scale (-i), the answer
to this question under varying conditions of demand is obtained
as follows:

Given a system of equations

qi= ip i = 1 .... n (1)

Define a quantity

^ so0 = cgo
iL1 i=1
a a



"QO p CC 1 (2)


and a quantity
a n
"q. = 1 (3)
i=1 il =1


Pi = Po + X

where
n a
"E qi 1 = ,i( o + x) = 0 (4)
=1i=

SThe authors are grateful to William G. O'Regan, Assistant Professor
of Agricultural Economics and Consultant, Statistical Laboratory, Univer-
sity of Florida, for the development of the proof presented in this section.







82 Florida Agricultural Experiment Stations

which implies n n
"Z q9p, E %.(Po + xI)
i =1 i = i=1 (5)
0 n n
Mq:L E q.i
1=1 i=1

If it can be shown that the nature of the inequality sign in
Fr() = % o (6)

subject to the side condition
n +
i=1

is determined by the magnitude of /f, then it follows that moving
to po from pi will increase, have no effect upon, or decrease

n
S q, depending upon the magnitude of /.
i=1

If it can be shown that F(x), subject to the side condition
f(x), is always equal to zero when 3 = (-1), has a stationary
point at zero when Xi = 0, and that this stationary point is a
maximum when p> (-1) and a minimum when pf< (-1), the
point will be proven.

when = (-1)
a n
t 0x) r OC1 i ZCJ + Xi)-1
i=1 i=1


f(x) = WC (Po + X)- i = o
i=1



n r c ix
f(x) = E o
i=1 po + X J

n -1 n o
z C PO + Xi) ---
i=1 i=1 PO + Xi


POcc = i Co Po(+
PO +I PO PO(PO + Xi)







Economic Relationships in Retailiny Citrus Products 83


E r (
i=1 P+ Xi P i=1 o+ Xi

n n n [ Z
F(x) = --i2 c i i + -1. 0 (8)
Po i= P0 i1 P i=1 iP + i


n
So, given (7), when / = (-1), : qi does not vary when ps's
change to po. i=1


Define


G(z) = 7(z) + ef(x)

n n n"'
"a aI Ea Po +* + Z. (or O + )


Q = AQcj(p + xj -' + e[X8 (po + X + g(po + x (10)


a a (11)
i-1

Setting these equal to zero and solving gives



PO

F(x) has a stationary point at X = O, where F(0) = O (13)


Is this a maximum, minimum, or neither? Hancock1 shows
that the answer to this question may be determined by inspect-
ing the signs of the polynomial in x generated by the expansion
of the determinant D.

1 Theory of Maxima and Minima, Harris Hancock, Ginn and Company,
New York, 1917, pp. 96-116.







84 Florida Agricultural Experiment Stations

(11 X ) 012 .

G21 (G22 X ) Gn f2

"31 13

D = (14)

.

(G. _2 (= u
81 n2 0-^ in

f1 f2 ^ 0

where

S = : h a 0 (15)
nhh = 0' e =


'hJ = jh = =0, (16)


f fi
J s = 0 (17)

and if these signs are unchanging, F(x) is a maximum at F(0)
and if the signs are alternating, F(x) is a minimum at F(0).

2- 8 -1) j (po+ xj-2 + ex( 1) (p + x/- +

S xi'o + 1 + 0 (Po + x (18)

2G =0



dX .1 + X 0 (19)


Evaluating these derivatives at Xi = 0 and e gives
Po







Economic Relationships in Retailiny Citrus Products 85




2



-a XI = 0, e = 0 (21)



A= f (22)
xjI Ix =0


The determinant then becomes

(K 1 \) 0 0 0 1
o (Ko x) o o C2
o o (K a X) .3
D p 0 (23)

0
.(K<^n- X) 'n

CC1 3 n 0



I -= ( p + 1)P-2 (24)
K > 0 when B < (-1)
K < 0 when > (-1) (25)

Rewriting the determinant of (23)

P11 0 0 .... o 0
S P 22 .... 0 M2
0 0 P33 .... o
20
D = o 0 (26)




c1 c2 13 .... n 0


Pii Ki X (27)







86 Florida Agricultural Experiment Stations

and recalling the rules for expanding a determinant, we note
that the expansion of (26) will contain exactly n terms, each
of which will contain an c2 and each of which will contain
n 1 of the Pii's. The subscripts of the omitted Pii in each
term will correspond to the subscript of the included -i. The signs
of the terms generated by the expansion will always be negative.
Therefore, the signs of the polynomial in A will be unchanging
if P1i contains two negative terms and alternating if Pii contains
a positive term. Now, Pi will contain two negative terms only
if K is native, and a positive and a negative term only if K
is positive.

Therefore, when

/3> (-1), K < O, F(x) is a maximum at F(0) = 0.

3 < (-1), K > O, F(x) is a minimum at F(O) = 0.
Which, if (1) be considered demand equations, indicates that
in an elastic situation moving from individual prices to weighted
average price will increase the total quantity demanded in all
cases except that one in which each individual price is equal to
the weighted average price. In the inelastic case such a move-
ment decreases total quantity except when each individual price
is equal to weighted average price.

SUPPLEMENTARY MATERIAL FOR TEXT TABLES 7 AND 8, PAGE
57, IS CONTAINED IN THE FOLLOWING TABLES.
TABLE 3.-DISTRIBUTION OF THE TOTAL SALES OF FRESH ORANGES BY PRICE
INTERVALS (JACKSONVILLE, FLORIDA, OCT. 31-DEC. 16, 1949 AND JAN. 2-
JUNE 23, 1950).

Price Interval Total Sales Total Value*
(Cents per Lb.) (Pounds) (Dollars)

1 -1.9 72.0 1.08
2 2.9 2,676.0 66.90
3 3.9 73,414.6 2,569.51
4 4.9 58,277.1 2,622.47
5 5.9 99,164.3 5,454.04
6 6.9 52,325.9 3,401.18
7 7.9 117,674.1 8,825.56
8 8.9 24,534.9 1 2,085.47
9 9.9 2,281.5 1 216.74
10 -10.9 56.3 i 5.91

Totals 430,476.7 1 25,248.86
Weighted Average Price = 5.865 cents

Price multiplied by quantity with the price taken as the mid-point of the price
interval.








Economic Relationships in Retailiny Citrus Products 87













TABLE 4.-DISTRIBUTION OF THE TOTAL SALES OF FRESH ORANGES BY PRICE
INTERVALS (MEMPHIS, TENNESSEE, JAN. 16-JUNE 12, 1951).

Price Interval Total Sales Total Value*
(Cents per Lb.) (Pounds) (Dollars)

4 4.9 1,530 68.85
5 5.9 24,730 1,360.15
6 6.9 112,516 7,313.54
7 7.9 133,996 10,049.70
8 8.9 1,095 93.07
9 9.9 1,350 128.25

Totals 275,217 19,013.56
Weighted Average Price = 6.909 cents

Price multiplied by quantity with the price taken as the mid-point of the price interval.






TABLE 5.-METHOD OF COMPUTING THE HYPOTHETICAL SALES OF FRESH ORANGES AT THE WEIGHTED AVERAGE PRICE FOR THE
MARKET UNDER ASSUMED DEMAND CONDITIONS, JACKSONVILLE, FLORIDA, OCT. 31-DEC. 16, 1949 AND JAN. 2-JUNE 23, 1950.
(1) (2) (3) (4) (5) (6) (7)
Constant Term, a, Hypothetical Sales
Price Total Sales Log. Price Log. Price Anti-log. in the Equation in Pounds
(Cents)* (Pounds) Col. (1) X 1.329** Col. (4) Col. (2) X Col. (5) Col. (6) 10.496t

1.5 72.0 0.17609 0.234024 1.7140 123.4 11.8
2.5 2,676.0 0.39794 0.528862 3.3795 9,043.5 861.6
3.5 73,414.6 0.54407 0.723069 5.2853 388,018.2 36,968.2
4.5 58,277.1 0.65321 0.868116 7.3810 430,143.3 40,981.6
5.5 99,164.3 0.74036 0.983938 9.6369 955,636.4 91,047.7 .
6.5 52,325.9 0.81291 1.080357 12.033 629,637.55 59,988.3
7.5 117,674.1 0.87506 1.162955 14.553 1,712,511.18 163,158.5
8.5 24,534.9 0.92942 1.235199 17.187 421,681.33 40,175.4
9.5 2,281.5 0.97772 1.299390 19.925 45,458.89 4,331.1
10.5 56.3 1.0211893 1.357161 22.759 1,281.4 122.1

Using the mid-point of the price interval as given in Appendix Table 3.
1.329
"** Where 1.329 is the exponent ef y in the assumed demand function X = a y
1.329
t 10.496 = 5.865 where 5.865 is the weighted average price as computed in Appendix Table 3.

TABLE 6.-METHOD OF COMPUTING THE HYPOTHETICAL SALES OF FRESH ORANGES AT THE WEIGHTED AVERAGE PRICE FOR THE
MARKET UNDER ASSUMED DEMAND CONDITIONS, MEMPHIS, TENNESSEE, JAN. 16-JUNE 12, 1951.
(1) (2) (3) (4) (5) (6) (7)
S1 Constant Term, a, Hypothetical Sales
Price Total Sales Log. Price Log. Price Anti-log. in the Equation in Pounds
(Cents)* (Pounds) Col. (1) X 1.665" Col. (4) Col. (2) X Col. (5) Col. (6) 24.982t

4.5 1,530 0.65321 1.087595 12.235 18,719.6 749.3
5.5 24,730 0.74036 1.232699 17.088 422,586.2 16,915.6
6.5 112,516 0.81291 1.353495 22.568 2,539,261.1 101,643.6
7.5 133,996 0.87506 1.456975 28.640 3,837,645.4 153,616.4
8.5 1,095 0.92942 1.547484 35.276 38,627.2 1,546.2
9.5 1,350 0.97772 1.627904 42.453 57,311.6 2,294.1

Using the mid-point of the price interval as given in Appendix Table 4.
-- 1.665
** Where 1.665 is the exponent of y in the assumed demand function X = a y
1.665
t 24.982 = 6.909 where 6.909 is the weighted average price as computed in Appendix Table 4.





University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs