• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Table of Contents
 Main














Group Title: Bulletin - University of Florida Agricultural Experiment Station ; 825
Title: Production, planting, and marketing characteristics of the Florida watermelon industry
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00027090/00001
 Material Information
Title: Production, planting, and marketing characteristics of the Florida watermelon industry
Series Title: Bulletin Agricultural Experiment Stations, University of Florida
Physical Description: 27 p. : ill. ; 23 cm.
Language: English
Creator: Wall, G. Bryan ( George Bryan )
Tilley, Daniel S
VanSickle, John
Publisher: Agricultural Experiment Stations, Institute of Food and Agricultural Sciences, University of Florida
Place of Publication: Gainesville Fla.
Publication Date: 1982
 Subjects
Subject: Watermelon industry -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Bibliography: p. 19.
Statement of Responsibility: G. Bryan Wall, Daniel S. Tilley, and John J. VanSickle.
Funding: Bulletin (University of Florida. Agricultural Experiment Station) ;
 Record Information
Bibliographic ID: UF00027090
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 000401514
oclc - 10680520
notis - ACE7362

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Title Page
    Table of Contents
        Table of Contents
    Main
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
Full Text


November 1982


PRODUCTION, PLANTING, AND
MARKETING CHARACTERISTICS OF
THE FLORIDA WATERMELON
INDUSTRY



G. Bryan Wall, Daniel S. Tilley, and John J. VanSickle


/
/
~


Agricultural Experiment Stations
Institute of Food and Agricultural Sciences
University of Florida, Gainesville
F. A. Wood, Dean for Research


Bulletin 825








PRODUCTION, PLANTING, AND
MARKETING CHARACTERISTICS
OF THE FLORIDA WATERMELON
INDUSTRY

G. Bryan Wall, Daniel S. Tilley, and John J. VanSickle


















AUTHORS
Dr. Wall is a postdoctoral associate, Dr. Tilley is a former adjunct
associate professor, and Dr. VanSickle is an assistant professor in the
Food and Resource Economics Department, University of Florida,
Gainesville. Dr. Tilley was also affiliated with the Florida Department of
Citrus.












CONTENTS


Page
Introduction ............................... .......... 1
Total Production ....................................... 2
Intra-Florida Production ................................. 5
Seasonal Price and Shipping Patterns ....................... 5
Prices and Break-Even Costs .............................. 8
M market Share .......................................... 11
Conceptual and Empirical Model of the
Florida Watermelon Industry ................... .. .12
Coefficient Interpretation and Use ................... 13
Im plications ........................................... 16
Conclusions ...................... ...... .. ........... 18
References .......... ............... ............ 19
Appendix ......... ............... ... ............ 21












INTRODUCTION
The state of Florida is a major supplier of spring and summer
watermelons for the domestic U.S. market. Total crop value for
Florida in the 1976 season was well over 25 million dollars, which
represented 4.9 percent of the total fresh vegetable income of the
state that year. For the seven-week period beginning in April,
Florida has averaged over 50 percent of the total shipments of
watermelons each year from 1972 to 1976. There is early season
competition from Mexican imports and late season competition
from other domestic producers, but Florida remains the market
volume leader.
Even as the market volume leader, Florida is not assured prof-
its for its watermelon production. In fact, due to the large volumes
and the rapid and easy entry and exit of producers, watermelon
returns are often depressed relative to competing crops. A
reasonably profitable season encourages entry and increased pro-
duction the following year, with resulting lower prices. In addition,
while south Florida provides the first domestically produced
watermelons of the season, the average statewide price for the
season is well below season opening values. This factor is climate
related: growers in the northern part of the state produce the ma-
jority of watermelons, but they are unable to plant as early as
growers in southern parts of the state because of the cold weather.
As a result, these northern growers are not able to take advantage
of an early harvest, which would net higher market prices. Instead,
the northern area growers must compete directly with domestic pro-
ducers in other states who have a location advantage to the major
eastern and midwestern states.
, The objective of this study is to identify the cobweb pattern of
Florida watermelon production and to clarify those market factors
which result in depressed prices paid to Florida growers in the
northern and panhandle regions of the state. This discussion is in
two sections. The first section presents historical data on Florida
production, prices, and market shares, as well as an estimate of
break-even costs. The second section presents a statistical model of
the Florida watermelon industry; the coefficients of this model are
used to calculate demand flexibilities and to provide estimates of
cobweb-type planting strategies in the Florida watermelon in-
dustry.







TOTAL PRODUCTION
In 1975, Florida produced 76 percent of the U.S. spring produc-
tion and 33 percent of total U.S. production [3] of watermelons.
Yields per acre for Florida producers are consistently greater than
the weighted average of U.S. spring producers and rank behind only
Arizona and the desert-producing areas of California, as shown in
Table 1. Yields however, are highly variable6The lack of irrigation
may be a reasonable explanation for yield variations in some parts
of the state, while local variations in soil fertilizer and physical
characteristics of the soil, rainfall, temperatures, and interactions
of these factors may also account for yield variability. The ease of
entry and exit by producers can also affect statewide yield averages
because of differences in grower skill levels. The number of acres
planted for the Florida crop has shown considerable variation, rang-
ing from 47,000 acres in 1975 to 65,000 acres in 1976, as shown in
Table 24When the acreage data are compared with data depicting
value per hundredweight, shown in Table 3, an inverse pattern
becomes apparent.
$ This inverse relationship between average price and quantity
harvested is shown in Figure 1. On the basis of the data in Tables 2
and 3, it would appear that a small crop in year t, which brings a
reasonably attractive price, induces increased plantings the follow-
ing year. The increased plantings, however, work to reduce the price
paid to growers. This dampened price results in decreased planting
the next year, and the restricted supply works to enhance price at
harvest. This high price, in turn, induces an increased planting the
next year. A possible consequence of the cyclical planting activity is
shown in Table 4, which is a tabulation of total abandoned acreage,
both in absolute terms and as a percentage of total plantings, for the

Table 1. Yields per acre for spring watermelon production.
Crop Year
State 1971 1972 1973 1974 1975 1976 1977 1978
Hundredweight
Alabama 81 79 75 80 60 75 70 64
Arizona 170 195 180 170 140 150 150
Calif. desert 215 195 170 230 185 220 195 215
Florida 150 120 160 150 185 180 175 160
Georgia 79 70 70 120 100 75 85 94
Texas 87 73 83 90 70 70 99 75
Wt. average 130 105 130 129 140 138 138 130
Source: Federal-State Market News Service







Table 2. Planted acreage for spring watermelons by state.
Crop Year
State 1971 1972 1973 1974 1975 1976 1977 1978
Acres
Alabama 700 3500 600 2900 2900 3700 3700 3500
Arizona 1900 3300 900 100 1300 1200 1900
Calif. desert 2500 4300 3800 2700 1800 3000 3500 3700
Florida 52200 61200 54700 50000 47000 65000 65000 59000
Georgia 3000 5000 500 3400 4200 10900 15600 8000
Texas 34500 48000 31000 30000 30000 28000 30000 28000
Source: Federal-State Market News Service

Table 3. Value per hundredweight of spring watermelons by producing
areas.
Crop Year
1971 1972 1973 1974 1975 1976 1977 1978
Dollars/Hundredweight
Alabama 2.30 2.80 4.90 4.50 5.40 3.90 3.85 3.80
Arizona 4.15 2.90 5.50 7.90 5.40 5.70 8.19
Calif. desert 4.01 2.62 4.61 5.48 7.50 6.51 5.30 8.51
Florida 2.72 2.42 3.07 3.28 4.36 2.61 2.97 3.35
Georgia 2.50 2.52' 4.00 3.70 5.00 2.70 2.80 3.80
Texas 3.35 3.61 3.58 4.60 5.77 5.62 4.52 5.24
Wt. average 2.97 2.79 3.29 3.79 4.73 3.28 3.43 4.14
Source: Federal-State Market News Service

major spring producing states. Alabama, California, and Arizona
reported no abandoned acreage in any of the six years, while
Florida, Georgia, and Texas reported substantial abandonment. In
1977, Florida growers left 14,000 acres unharvested, which
represented 21 percent of total plantings. This figure represents the
largest degree of abandonment by Florida growers, although 15 per-
cent of the crop was abandoned the previous year, and over 10 per-
cent was left unharvested in both 1973 and 1974. In the recent past,
both Georgia and Texas have reported abandonment equal to 20
percent of their crop, and in 1976 each reported approximately 11
percent abandonment.
In each of the six years shown in Table 3, the statewide average
value per hundredweight for Florida watermelons was below the
weighted average of other spring producers. This is in spite of the
fact that Florida enters the market first with high early season







Acres
(1000)


64 196 1968 1970 19 1974
1964 1966 1968 1970 1972 1974


1976 1978


Figure 1. Acres planted and value per hundredweight for Florida
watermelons.
Source: Florida Crop and Livestock Reporting Service


Table 4. Abandoned acreage of spring watermelons.
Crop Year


1971 1972 1973 1974 1975 1976 1977
Acres


2100
1000
2700


5100 6000 5500 3400 10000 14000
700 100 200 800 1200 1400
3200 1000 3000 6000 3000 2000


1978


9000
500
4000


Percentage of Total Plantings


4.02
33.00
7.83


8.33 10.97 11.00 7.23 15.38 21.54 15.25
14.00 20.00 5.88 19.04 11.01 8.97 6.25
6.67 3.23 10.09 20.00 10.71 6.66 20.00


Source: Federal-State Market News Service


Value
per CWT


5.00-


4.0-


3.00-


2.00-


1.00-


State


Alabama
Arizona
Calif. desert
Florida
Georgia
Texas



Alabama
Arizona
Calif. desert
Florida
Georgia
Texas


I







prices. In every year except 1971 Florida growers received the
lowest reported price of the six states.

INTRA-FLORIDA PRODUCTION

There has been relatively little change in production shares
among various producing areas of Florida. One of the largest
changes is an 11 percent decrease in production in the north central
area, which encompasses Citrus, Hernando, Lake, Pasco, and
Sumter counties. In addition, the southwest area has experienced a
decrease in production equal to about 5 percent. An increase of
almost twelve percent has been recorded for the southern produc-
tion area of the state, and this area, in conjunction with the western
production area, now produces 66 percent of the total Florida
acreage. In terms of total acres planted from 1960 to 1978, the
northern area has experienced a decline from almost 44,000 acres to
approximately 29,000 acres, although the smallest acreage was
reported in 1974, when less than 23,000 acres were planted.

SEASONAL PRICE AND SHIPPING PATTERNS
Price paid to Florida watermelon producers declines as the
season advances, as shown in Figure 2. Figure 2 reflects the weekly
average prices for the seasons from 1966 through 1977. Although
the graph plots prices for only medium-sized watermelons, the
pricing pattern for small and large melons is similar (Appendix,
Figure A-1 and A-2). The prices shown are f.o.b. prices, and the
season is defined as beginning with the first week in April. Included
in the graph is a lower confidence bound of at 0.05 alpha level, which
can be used as an estimate of a price floor. On the average, only 5
times out of 100 will price, for a given week, fall below this line. Of
course, any change that results in a permanent change in the
market for example, increased transportation costs can per-
manently force the price below this level. The boundary line can also
be used to estimate the degree of price variance. It is interesting to
note the separation between the two curves in the early part of the
season, weeks 1 through 6, which indicates a highly variable price.
The more variable the price, the farther apart are the two lines.
Beyond the 6th week of the season, as the price steadily declines,
the separation between the two curves becomes less divergent,
which indicates less price fluctuation. Traditionally, Florida pro-
ducers have viewed July 4th as the season's end. There is a
moderate amount of shipment after that date, but the Independence
Day target is regarded as the actual termination for Florida












Table 5. Percentage of total Florida production by area as measured by acreage.
Year
...............................................................................................................................................................................


Areaa


1960 1971 1972 1973 1974
Percentage


West
North
N. Central
0) W. Central
E. Central
Southwest
Other
Total


1975 1976 1977 1978


8.59 17.17 16.11 13.76 15.06 12.27 11.82 6.27 8.20
46.37 37.52 44.03 47.54 51.01 50.80 5145 58.82 58.00
18.90 8.68 7.79 7.29 7.53 8.83 10.81 8.63 7.40
7.81 12.67 11.69 12.42 8.66 10.67 8.64 8.24 8.00
.55 .79 .69 .60
10.51 17.17 13.96 12.63 12.70 11.70 7.73 6.47 5.80
7.26 5.99 6.42 6.37 4.04 5.73 10.18 10.88 12.00
100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00


Source: Florida Crop and Livestock Reporting Service
aCounties included in each area are shown in Table A-1.
bProduction included in other counties is composed of plantings that are too small to report on a county basis and does not affect
counties not included in the geographical areas.







Volue
per CWT
($)

6-


5- Average
Price

4-


3- 95% Lower
Bound

2-



O 2 4 6 8 10 12 14 16
Weeks
Beginning Spring Quarter

Figure 2. Average weekly prices and 95% lower confidence limit for
medium-sized watermelons (21-26 Ibs.).

shipments. Independence Day usually occurs around the 14th or
15th week of the season, and the graphs show a noticable price
decline after this point.
Actual shipment totals and percentage of weekly movement in-
dicate that the season generally begins the first week of April and
continues through mid August (Appendix Table A-2). Florida pro-
duction drops rapidly after the 14th week and ends completely by
the 18th week. These data represent the averages of the five produc-
tion years from 1972 through 1976. These data indicate the extent
of Florida's domination of the market. They also show how the
northwest area of Florida must compete directly with other South-
eastern U.S. producers beginning around week 11. The late July and
early August market is supplied by Arizona, California, and
Virginia (which is not shown in the table). In addition, some states,
Missouri and Indiana for example, enjoy a very strong regional
market. This information is presented graphically in Figure 3, and
clearly shows the domination by Florida in weeks 6 through 15.
California's large percentage of the late season market is somewhat
misleading, because at this time the absolute amount of total







---- California
- Florida
...-... Georgia
...-- Mexico
............ South Carolina
------ Texas
/


0 2 3 4 5 6 7 8 9 10 i 12 13 14 15 16 17 18 19 20 21
Weeks
Beginning Spring Quarter

Figure 3. Relative market shares by week.


shipments is small. This information illustrates that the declining
prices of Florida watermelons, between weeks 6 and 11, occurs when
Florida production is increasing. During this period, the northern
and north central Florida producing areas are sending watermelons
to market. The western producing area enters the market around
week 11, at the same time as Georgia, Alabama, and South
Carolina. This increase in marketing volume is added to that coming
from the northern producing areas. It is the rapid price decline after
week 11, associated with the increased marketing, that is respon-
sible, in large part, for the abandonment of potentially productive
fields. The degree of abandonment is, or course, dependent on the
market price and its relation to production costs.

PRICES AND BREAK-EVEN COSTS

Break-even costs have been computed for both the southern and
northern producing area of Florida (Table 6). The range in the
survey data is substantial, and therefore, the estimates are to be
considered indicative of industry-wide costs. These break-even cost
estimates can be applied to Figure 2 to indicate dates of successful
harvest. For example, if the northern areas can produce and market
watermelons for approximately $2.50 per hundredweight, as in-







Table 6. Break-even cost estimates for Florida watermelons by area.
Area
.................................................
Item Southern Northern
Growing cost (dollars/acre) 748 270
Harvest & marketing (dollars/acre) 282 223
Total (dollars/acre) 1030 493

Break-even cost @ 20,000 Ibs/acre
(dollars/hundredweight) 5.15 2.47
Source: Costs and Returns to Florida Growers, unpublished Extension budgets.

dicated by Table 6, and shipments from that area begin around
week 9, one can see that both the average price and the 95 percent
lower bound line is above the $2.50 level for that week. As the
season progresses, both the average price and the lower bound
decrease. There appears to be a leveling off of average price after
week 11, and the lower bound line indicates that average price rarely
falls below $2.00 per hundredweight. By the same token, any grower
in north Florida who cannot produce and harvest watermelons for
less than $3.00 per hundredweight should complete his harvest by
week 9 or 10 at the latest.
A question that producers often ask concerns the degree of
power they feel the farm level buyer exerts on the grower to drive
down prices. While it is to the buying broker's advantage to buy as
cheaply as possible, it is likewise to the grower's advantage to sell
as early as possible. It is not in the broker's best interest to force
prices below production costs, because such an action will, in the
long run, force producers out of the market and thus restrict sup-
plies. During years of large plantings, the buyer can succeed in bid-
ding down the price to break-even levels because he can choose
among a large number of growers for his supply. Conversely, in
times of restricted supplies the broker is not able to do this. In any
case, the limit on what the broker will pay is determined at the ter-
minal markets. Specifically, the Florida f.o.b. price is equal to the
terminal price minus the transportation costs. Table 7 shows the
Baltimore terminal price, by date, for watermelons of equal quality
produced in the various states. It is interesting to note that the ter-
minal market wholesale price is the same, regardless of the state of
origin of the watermelons. Thus on July 8, with the wholesale price
of $5.25 per hundredweight, South Carolina producers will receive a
higher f.o.b. price than Georgia or Florida producers, simply
because they are closer to the market, and transportation costs are
less. This is a major reason for depressed f.o.b. prices in northern







Table 7. Terminal market prices per hundredweight for watermelons by state of origin, Baltimore market.
State of Origin
South North
Florida Georgia Carolina Carolina Maryland Virginia
Date 20 Ib ave. 20 Ib ave. 20 Ib ave. 20 lb ave. 20 Ib ave. 20 Ib ave.
(1976) and larger and larger and larger and larger and larger and larger
April 29 15.00
May 6 10.00
13 7.50
20 6.25
27 5.00
June 3 4.50
. 10 6.50
17 6.00
24 6.00
July 1 6.37 6.37
8 5.25 5.25 5.25
15 3.50 3.50 3.50
22 5.50 5.50
29 5.25
August 5 5.38 5.38 5.38 5.38
12 3.62 3.62
19 3.25 3.25
29 3.50 3.50
Source: Federal-State Market News Service, 1976.
Represents stock of good quality, including but not limited to U.S. No. 1 grade, of mixed varieties but mostly long grays.







Florida. The site prices (Bressler and King, p. 125) paid to Florida
growers are less than the site prices paid to their Georgia and South
Carolina counterparts. Large planting will work to dampen growers'
prices, as previously explained, but as the season progresses, the
loss of the locational and temporal advantages that other domestic
producers enjoy is also a major factor in causing lower prices. Any
fluctuation in consumer demand in the receiving markets, such as
that caused by unseasonable cool weather, can force the terminal
price down, which in turn will result in decreased prices at the farm
level.

MARKET SHARE
It has been shown that although Florida producers have suf-
fered several years of depressed prices, they still control a dominant
share of the domestic market. A measurement of changes in
Florida's market share, by region and cities, is provided in the ap-
pendix Table A-3. The data presented there represent two-year
averages of total unloads over the ten-year span 1965-66 through
1975-76, and the average change in value for each region is an in-
dication of growth or reduction of market share. Both Florida and
Mexico have enjoyed increases in market share in each of the four
regions, with Mexico showing about twice the overall growth of
Florida. In the western markets, Florida is decidedly second to Mex-
ico. In the eastern markets, Florida's growth rate is less than half
that of Mexico's, and approximately one-fourth that of Mexico's in
the mid-western markets. Florida has shown excellent growth in the
southern markets. It would appear from the table that Mexico's
gains have been made at the expense of Georgia and South Carolina
in the eastern markets; Georgia in the midwestern markets; Georgia
and Texas in the southern markets; and California in the western
markets.
This is in keeping with previous data which have shown that
Florida growers have been somewhat affected by early-season Mex-
ican imports, but that on the whole, Florida's market is both well
defined and viable.
The following section provides a conceptual and empirical
model of the planting, harvest, and shipment decisions that make
up the price determination processes.







CONCEPTUAL AND EMPIRICAL MODEL
OF THE FLORIDA WATERMELON INDUSTRY

Florida watermelon growers make two sequential decisions that
key the price determination process. The first decision concerns the
level of planting activity. This decision is hypothesized to be based
on grower expectations for prices for watermelons in the upcoming
season a planning sequence that is a reflection of the cobweb
model discussed previously. In general, it is felt that watermelon
growers, especially in the north Florida area, have a few alternative
crops. Therefore, the decision whether to plant or not to plant is
based solely on their watermelon price expectations. While there are
several alternative expectation hypotheses, the most simple
hypothesis is that future price expectations are some linear com-
bination of recent past prices. The number of lags and the nature of
the weighting system cannot be specified in advance but remain for
empirical determination. The use of lagged price coefficients to cap-
ture the cob-web effect is similar to the use identified by Suits and
further expanded by Waugh [5, 6].
It has also been hypothesized that price expectations are
developed for nominal, as opposed to real, prices. As a result,
Growers suffer from money illusion, at least in the short run. That is,
high nominal prices inspire additional production, even though the
inflation rate may be high, and deflated or real prices are declining.
In addition to price expectations represented by lagged prices, a
time-trend variable is included in the planting decision model to
reflect the overall decline in Florida watermelon acreage that has
been occurring. This decision process is represented by Equation 1
(Qp,) in the model shown in Table 8.
The second decision a watermelon grower faces occurs at
harvest time. As previously stated, each year a portion of planted
acres are not harvested.*The reason for abandonment may range
from weather conditions at harvest, out-of-state competition, labor
and truck availability, and/or low prices. It is Eyypothesized that
quantity harvested is directly related to quantity planted and the
price change from the previous season (Equation 2, Table 8). If the
quantity planted is large, high prices will induce additional
harvesting. If quantity planted is low, high prices will have little im-
pact on harvesting decisions. However, if quantity planted is high,
the grower may face a situation where price is an important factor in
determining how much to harvest. In order to capture these
phenomena, quantity planted and price change are interacted in the
harvesting decision equation (Equation 2) shown in Table 8.







The third equation (QS,) expresses a technical relation between
shipments, harvested acres, yield, and a time trend. The fourth
equation (PF,) describes a behavioral relationship concerning the ef-
fect the quantity shipped from Florida and the quantity shipped
from competing states has on the prices paid to Florida growers.
The model was estimated using two-stage least squares (regres-
sion analysis). All coefficients in the model were found to be of the
expected sign and most had high statistical significance (Table 8).
The structure of the model is nearly recursive, except for the in-
teraction variable' in the harvest equation (Equation 2), which
allows for immediate price feedbacks in the harvest decision.

COEFFICIENT INTERPRETATION AND USE
The coefficients of the model addressed both the entry-exit
issue and the interregional competition questions facing extension
personnel in Florida. Not surprisingly, the model shows that the
present season's plantings are partly determined by the previous
season's prices. This confirms the assumption concerning the
development of price expectations. The duration of this effect is ap-
parently three years. This means that high (low) prices one year will
continue to encourage (discourage) planting for three years, with the
strongest effects coming the first two years following the high (low)
prices. The most recent season appears to have a slightly small ef-
fect on planting decisions than the season two years passed. This
would seem to indicate some form of planting strategy by growers.
Current and lagged response coefficients (ER-,, ER-2, ER-3) indicate
how a percentage change in price effects the percentage change in
planted acres in the following three years. They are computed by
multiplying the coefficient for the appropriate lagged price from
Equation 1 (QP,), Table 8, by a factor equal to the average farm gate
price for the previous years divided by the average acres of water-
melons planted for the previous year; e.g.,

_1 QP, PF,_-
ER- --
OPF,-, QP,

1. Several alternative forms of a polynominal distributed lag were
estimated. The unrestricted form coefficients shown in Table 8 are very
similar to coefficients for a third degree polynominal with current and three
lagged prices included. In the form that is shown, the current price is
assumed to have no effect on current planting. This assumption was
verified by alternative forms that were estimated. The lagged prices were
not found to be pairwise highly correlated.











Table 8. Estimated coefficients and their respective standard errors for linear form of Florida watermelon industry model, 1957 to 1976 data.

Equation Endogenous Variablesb Predetermined Variables Constant
Equation Constant
(PFt)
QP, OHt QSt PF, (QPt) zPFt PFt1 PFt-2 PFt.2 001 YLDt T,
1) QPt -1 16.6478 16.9061 9.8832 -3.6273 34.0954
(8.2280) (8.8028) (8.0713) (.7691) (24.7790)
2) QH, -1 .5272 -16.056 2.1155
(.1480) (6.6472) (3.6255)
3) QSt 88.9088 -1 63.2716 -55.0328 -5333.6
(25.1550) (5.7731) (39.1679) (2560.1)
4) PFt -.00018 1 .000087 5.1098
(.00005) (.000021) (7.2671

aStandard errors are presented in parentheses below the coefficients.
bVariable definitions:
QPt, QHt number of acres of watermelons planted and harvested Florida in year t,
QSt thousands of hundredweight of watermelons shipped from Florida in year t,
PFt, PF,.1, PFt-2 current and prior year farm gate prices for Florida watermelons in year t in dollars per 100 Ibs.,
zPFt = PF PFt-1 change in price from previous year,
QOt thousands of hundredweight of watermelons shipped from other producing states in year,
YLDt yield of watermelons in Florida in year t, 100 Ibs. per acre,
T, time trend, T = 1 for 1953, 2 for 1954, ... 24 for 1976.






The computed response coefficients were:
2.17
(1) ER-, = 16.6478 60.45 = 0.60,

2.17
(2) ER-2 = 16.9061 60.45 0.61, and

2.17
(3) ER-3 = 9.8832 60.45 = 0.35.

Thus a 10 percent increase (decrease) in price can be expected to
result in a 6 percent increase (decrease) in planted acres the next two
seasons and a 3.5 percent increase (decrease) the third season.
These results clearly reveal that the cob-web phenomenon does
exist, and that through the harvest and price equations it becomes a
source of cyclical price variability in the Florida watermelon in-
dustry. For individual producers, an understanding of the lag struc-
ture may allow greater use of countercyclical production decisions.
The question of profit enhancement due to countercyclical reactions
is, of course, dependent on the magnitude of the response.
The coefficients related to the conclusions regarding inter-
regional trade are the quantity coefficients present in the fourth
equation. Shipments from Florida have less of an effect on Florida
prices than do shipments from competing states.
Price flexibilities indicate how a percentage change in
shipments from Florida (Fi) or other states (F2) effects the percent-
age change in Florida prices. These price flexibilities are computed
from the coefficients in Equation 4, Table 8. F, (or F2) is computed by
multiplying the coefficient for Florida (or other states) shipments
of watermelons in the previous years by the average number of
shipments of watermelons in the previous years from Florida (or
other states), and dividing that by the average farm gate price of
watermelons in the previous years; e.g.,

F PF, QS,
F, = -
aQS, PF,

The price flexibilities were computed as:

(4) F, = 0.000180 7713.25 0.640,
2.17

(5) F2 = 0.000087 29051.55 = 1.165.
2.17







Thus, a 1 percent increase in Florida shipments reduces Florida
prices by 0.6 percent, while a 1 percent increase in shipments from
other states would cause a price decline of 1.2 percent in Florida. In
terms of an absolute change in quantity shipped from either region,
the effect on Florida prices would be greater if that quantity comes
from Florida. Quite clearly, Florida prices are influenced by produc-
tion and shipment levels from other states. These results identify
the importance of shipments in the price determination and the
degree to which Florida competes with other regions. This also
means that policy instruments designed to systematically restrict
Florida plantings could effectively enhance prices only if production
from other areas could also be curtailed.
The specification of the harvest equation means that the
marginal propensity to harvest from planted acres (MPH) is price
dependent,2 i.e.,
(6) MPH = 0.5272 PF,
and that the effects of prices on harvest decisions (MPP) depends on
the level of plantings,3 i.e.,
(7) MPP = 16.056 + 0.5272 QP,.
If prices increase, the marginal propensity to harvest from
planted acreage increases as indicated by Equation 6. Equation 7 in-
dicates that if a great many acres have been planted, the marginal
effect of price on acreage harvested increases with acreage planted.
At planting levels below 30,500 acres, computed by making
MPP = 0 (Equation 7) and solving for QP, the price effect
diminishes and becomes negative. In general, MPP has been greater
than zero at all QP,that have been observed. Both of these relation-
ships are reasonable and conform to expectations.


IMPLICATIONS
The implications of this study for Florida producers must be
judged in light of recent production patterns within the state. The
northern and western areas of the state have increased production
to 60 percent or more of the state's total crop (Table 5). Correspond-
ingly, the southern portion of the peninsula is producing a declining
share of total state production (less than 6 percent). From the

9QH,
2. MPH =QH,
aQP,

3. MPP = QH,
aPF,






preceding analysis one can see that this production pattern shift is
not to the advantage of Florida producers. Increased plantings in
the northern and western areas of the state compete directly with
shipments from other states. This competition results in Florida
prices being more dependent on the dominant effect of shipments
from other states.
It can be asserted that prices have generally been higher than
the marginal cost of harvesting melons, and the abandonment that
occurs is due to factors other than price. It is unlikely that Florida
growers, especially in northern parts of the state, would achieve
price enhancement because of an economic abandonment practice.
These conclusions are based upon the observation that abandon-
ment generally occurs when watermelons become available from
production areas north of Florida. Because of a location advantage,
northern producers enjoy lower transportation costs to major
markets than do Florida watermelon growers. Therefore, although
Florida prices are below those in other producing regions, they are
not apparently low enough to compensate for the higher transporta-
tion expense and inconvenience.
The net result is that prices in Florida are largely dependent on
production availability from other areas. Thus, price expectations
based on assumptions about shipments from other states would
seemingly be more reliable. However, because it is not easy to
predict yield, it would be difficult to forecast shipments from other
states. While this conclusion may not be very satisfying, it does
point out the potential price advantages available to Florida
growers should their plantings and production be counter-cyclical to
that from the rest of the United States. It also points out the impor-
tance of early season shipping.
The direct implication of the findings of this study should be of
particular interest for extension personnel and others with an in-
terest in Florida watermelon production. The study indicates that
the pattern of increased production in northern areas of the state
has an adverse price effect, both from the standpoint of additional
tonnage to be marketed and from the dominant impact on price
from competing quantities. Relatively high prices lead to increased
planting and relatively depressed prices lead to reduced plantings.
It appears that each year the f.o.b. price paid to Florida growers is
forced to the level of the variable costs of harvest, a process that
begins when other domestic producers enter the market. Because of
the lower transportation costs to the terminal markets, the produc-
ing areas closer to terminal markets enjoy a location advantage over
Florida. The relatively lower Florida prices for watermelons do not
appear to be low enough to compensate for the increased cost of






transportation.
The net result is that Florida prices are sensitive not only to
Florida production, but also to quantities produced in other
domestic production areas north of Florida. It is possible that
technological improvements resulting in decreased per unit costs
for Florida producers could improve their competitive position.
Although the authors have concentrated on those market factors
which can be readily quantified, there are other, more subjective fac-
tors that have a strong influence on market strength. The
watermelon industry is, for the most part, devoid of any generally
accepted grading standards. As a result, quality is often highly
variable. In addition, there exist little or no generic, or "name
brand", promotion for the watermelon industry. While the Florida
watermelon industry faces problems unique to themselves, the
declining per capital consumption of watermelons [4] indicates the
industry as a whole could benefit from a concerted marketing effort.

CONCLUSIONS
iWatermelon planting in Florida has been shown to exhibit a
cyclical pattern in spite of the fact that there is evidence of
watermelon growers attempting to use the pattern to their ad-
vantage. Those producers in the southernmost part of the peninsula
are the first domestic producers to enter the market; they receive
the corresponding early season price advantage. The northernmost
areas of the state generally face break-even prices due to the large
volume of production in northern Florida and areas north of Florida.
The season average price for the state as a whole is less than the
weighted average of the states of Alabama, Arizona, California,
Florida, Georgia, and Texas. In every year, from 1972 through 1976,
Florida growers received the lowest reported price of these six
spring and early summer-producing states.
The major reason for the large price decline and the resulting
depressed f.o.b. prices in northern Florida is the loss of the tem-
poral market advantage enjoyed by southern Florida and the loca-
tion disadvantage with which the northern producers find
themselves operating. South Carolina watermelon producers, for ex-
ample, are closer to the major east coast terminal markets than
Florida, and thus, for the same quality product, receive a higher per
unit return than their Florida counterpart. Another factor con-
tributing to declining prices is the large volume of planted acreage
in north Florida and a decrease in planted acreage in south
Florida a trend that is, if anything, contrary to what is needed in
order to improve prices.







SIn an attempt to model the planning decision, Florida growers
were shown that the prices for the three previous years are closely
related to production decisions during the current season.


REFERENCES

1. Bressler, R. G. and R. A. King. Markets, Prices, and Interre-
gional Trade. John Wiley and Sons, New York, New York.
416 pp.
2. Florida Crop and Livestock Reporting Service. Vegetable Sum-
mary, various issues. 1222 Woodward Avenue, Orlando,
Florida.
3. Federal-State Market News Service. Marketing Watermelons,
various issues.
4. Showalter, R. K. 1979. Increasing Watermelon Consumption.
Vegetarian Newsletter 79-4, Vegetable Crops Department,
University of Florida, Gainesville, Florida.
5. Suits, Daniel B. 1955. An Econometric Model of the Water-
melon Market. Journal of Farm Economics 37:237-251.
6. Waugh, Fredrick V. 1958. A Theory of Anticipatory Prices. The
American Economic Review 48:188-199.
















APPENDIX




Table A-1. Total acres harvested, by area and county.
Year
................................................................................................................................................................................


Area
West
Calhoun
Holmes
Jackson
Washington
Total


1960 1971


1650
1975
1800


850
2350
2500
2600


1972

800
2330
2790
2950


1973 1974


700
2300
1600
1900


800
2100
1650
2150


1975

450
1600
1400
1900


1976

400
2150
1600
2350


1977

a
1000
1600
1600


1978

a
1800
500
1800


6275 8600 9040 6700 6700 5350 6500 3200 4100


North
Alachua
Bradford
Columbia
Gilchrist
Lafayette
Leon
Levy
Madison
Marion
Putnam
Suwannee
Union
Total


N. Central
Citrus
Hernando
Lake
Sumter
Total


4800 3600 4650 4000 4100
-- a 300 300 300
2600 450 470 450 550
5600 3000 3680 3600 3700
1000 900 1060 1000 1500
-- 400 520 450 300
4700 2200 2740 3300 3200
1700 950 700 750 450
7200 3600 4600 4500 4200
950 a a a 600
3700 2300 3320 2800 2600
-- -- -- a 400

33850 18800 24700 23150 22700


1750
1450
3150
6000
13800


650 770 650 500
400 490 250 250
800 590 500 800
2100 2320 2150 1800
4350 4370 3550 3350


3800
350
900
3100
1500
500
3000
400
3900
a
2700
300
22150


400
250
1000
2200
3850


5100
350
1800
4100
1700
600
3300
600
5700
350
3100
800
28300


1000
300
2100
2200
5600


4100
400
1600
4200
2100
400
4500
700
5500
500
3500
1000
30000


800
350
1150
2100
4400


4100
400
3800
4100
1300
400
4100
1000
4200
300
3700
300
29000


700
350
1000
1650
3700





Table A-1. Continued.
Year
Area 1960 1971 1972 1973 1974 1975 1976 1977 1978
W. Central


DeSoto 600
Hardee 700
Highlands --
Hillsborough 1000
Manatee 700
Polk 2700
Sarasota --


5700


2200
1200
1000
300
400
1000
250
6350


3030
1030
650
250
780
670
150
6560


2600
1200
450
250
1000
550
a


1700
850
400
300
650
400
a


6050 4300


1300
600
500
350
1100
300
500
4650


1950
500
a
350
1550
a
400


1400
900
a
a
1100
450
350


1400
900
a
a
900
450
350


4750 4200 4000


E. Central
Okeechobee
Total


Southwest
Charlotte
Collier
Hendry
Lee
Total


1000
2500
3600
575
7675


1300
2900
3600
800


800
2590
3880
560


1400
1600
2450
700


1150
1700
2200
600


1150
1450
2050
450


a 350


950
1200
1650
450


1100
1400
1900
a


8600 7830 6150 5650 5100 4250 4400


1000
1350
1550
a
3900


Other Counties 6150
State total 73000


3300 3760 3300 1800 2500 5600 5550 6000
50100 56100 48700 44500 43600 55000 51000 50000


Source: Florida Crop and Livestock Reporting Service
a Included in other counties.


Total








Table A-2. Weekly truck shi
pment, carrot equivalent by producing region, 1972-76.


Producing Weeks
Area la 2 3 4 5 6 7 8 9 10 11
Carlot Equivalent
(Percentage)


Alabama
Arizona


California

Floridab
A. 1975-76

B. Other

Northwest


Georgia


South Carolina
Texas

Mexico


Total


1 16 34
(1) (1)
6 45 184 330
(2) (7) (7)

2034
(41)
5 36 130 200 305 433 598 1055 1396 1508 1676
(3) (11) (30) (3) (43) (51) (55) (60) (58) (57) (33)
19
(0)
6
(0)
1 0
9 102 345 677 751 730
(1) (9) (20) (28) (28) (15)
149 282 303 375 402 406 379 363 275 194 115
(97) (89) (70) (65) (57) (48) (35) (21) (11) (7) (2)

154 318 433 575 707 848 1079 1769 2394 2654 4956




Table A-2. Continued.


Producing
Area


Alabama

Arizona

California

Florida
A. 1975-76

B. Other

Northwest


Georgia

South Carolina

Texas

Mexico


Total


Weeks
12 13 14 15 16 17


18 19 20 21 22


Carlot Equivalent
(Percentage)
20 123 142 147 101 72
(2) (3) (4) (4) (5)
96 218 273 264 197 105 47 3 3 0 1
(2) (3) (6) (7) (7) (7) (8) (2) (5) (2)
470 511 456 459 401 244 216 80 55 41 40
(9) (8) (9) (13) (15) (17) (37) (63) (95) (100) (98)

2396 2242 1256 444 128 33 15
(40) (35) (25) (12) (5) (2) (3)
1727 1333 755 310 76 8
(29) (21) (15) (9) (3) (0)
277 386 277 274 143 5
(5) (6) (6) (8) (5) (0)
143 716 973 683 760 426 116 20
(2) (11) (20) (19) (28) (29) (20) (16)
1 70 294 717 721 434 94 3
(1) (6) (20) (26) (29) (16) (2)
800 703 493 279 219 148 102 22
(13) (11) (10) (8) (8) (10) (17) (17)
68 47 34 14 1 1
(1) (1) (1) (0) (0)


5998 6349 4953 3591 2747 1476 590 128 58


41 41


Source: Federal-State Market News Service
aThe first week in April of each year is assigned as week one.
bUntil 1974 Florida shipments were divided into Northwest Florida and other Florida producing areas. Since that time the two have
been combined. The shipments shown under category A represent averages for crop years 1975, 1976, while item B reports
shipments from 1972, 1973, 1974.





Table A-3. Percent of market unloads as select terminal markets for major producing regions for 10-year span, 1965-66
through 1975-76.
Shipping Points
South
Regions and Florida Mexico Georgia Texas Carolina California
Cities 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76
Eastern
Albany, NY 55 61 2 2 5 10 0 0 4 8 0 0
Baltimore, MD 34 51 0 1 20 12 0 0 15 9 0 1
Boston, MA 42 42 0 7 3 11 0 2 22 13 0 0
New York, NY 43 43 1 8 9 6 0 7 17 14 0 0
Philadelphia, PA 40 36 1 6 13 23 0 0 15 10 0 0
Pittsburgh, PA 37 58 1 4 14 2 4 0 13 13 0 0
Providence, RI 52 36 1 3 4 1 0 0 13 48 0 4
S Montreal, Quebec 62 65 1 9 3 3 0 0 16 7 0 0
Ottawa, Ontario 63 77 0 3 11 10 0 0 13 0 0 0
Toronto, Ontario 59 46 0 8 10 9 0 1 16 5 0 0

Midwestern
Chicago, IL 31 29 2 10 2 8 17 27 0 2 0 0
Cincinnati, OH 47 68 0 3 22 14 0 1 2 0 0 0
Cleveland, OH 43 44 1 5 0 9 18 7 3 8 0 0
Detroit, MI 41 50 2 5 10 6 10 10 2 4 0 0
Indianapolis, IN 37 60 0 7 27 2 0 3 0 0 0 1
Kansas City, MO 16 16 1 3 0 0 47 32 0 0 0 0
Louisville, KY 33 57 0 0 17 17 1 0 0 0 0 0
Milwaukee, WI 45 36 0 6 6 17 16 6 0 0 0 0
Minneapolis, MN 30 11 1 4 1 0 55 70 0 0 0 0
St. Louis, MO 24 21 1 4 1 1 31 12 0 0 0 0
Winnipeg, Man. 34 9 0 20 0 0 64 69 0 0 0 0






Table A-3. Continued.

Shipping Points

South
Regions and Florida Mexico Georgia Texas Carolina California
Cities 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76 65-66 75-76
Southern


Atlanta, GA 32 40
Birmingham, AL 35 40
Columbia, SC 51 49
Dallas, TX 6 5
Ft. Worth, TX 2 0
Houston, TX 13 38
Memphis, TN 37 70
Miami, FL 72 45
Nashville, TN 23 64
New Orleans, LA 38 40


0 1
0 0
0 0
2 9
1 15
7 12
1 10
1 6
0 0
0 4


54 0 1
4 0 0
2 0 0
0 92 86
0 97 83
0 78 37
2 1 2
14 0 19
6 5 1
0 7 10


Western
Denver, CO 2 0 13 19
Los Angles, CA 1 4 15 19
Portland, OR 0 0 7 14
Salt Lake City, UT 0 0 10 12
San Francisco, CA 0 0 14 23
Seattle, WA 0 0 6 20
Vancover, BC 0 0 5 43
Source: USDA unload statistics by region, 1967 and 1977.


0 0 63 48 0
0 0 10 15 0
0 0 2 5 0
0 0 4 2 0
0 0 8 4 0
0 0 6 11 0
0 0 0 2 0


0 3 7
0 65 49
0 60 71
0 69 85
0 74 71
0 68 60
0 94 51



































This public document was promulgated at an annual cost of
$2954 or a cost of $1.18 per copy to provide information on
planting and marketing strategies for Florida watermelon
producers.

All programs and related activities sponsored or assisted by the Florida
Agricultural Experiment Stations are open to all persons regardless of race,
color, national origin, age, sex, or handicap.


ISSN 0096-607X


TEACHING\
AESEAPCH
EXIE.M.I
UNVESIY F LOID




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