September 1970
Ag. Econ. Report 10
Impact of Mechanical Harvesting
on the Demand for Labor
in the Florida Citrus Industry
Department of Agricultural Economics
Florida Agricultural Experiment Stations
Institute of Food and Agricultural Sciences
University of Florida, Gainesville
J. Kamal Dow
HUME LIBRARY
i A R 9 1d/I
I.F.A.S. Univ. of Florida
I yl _--- -- I rl
r r I,
FOREWORD AND ACKNOWLEDGMENTS
The research on which this report is based was possible thanks to
the sponsorship of the Manpower Administration Office, United States
Department of Labor. The opinions and conclusions, however, are those
of the author and do not necessarily represent the Department's official
opinion or policy.
The author wishes to acknowledge the suggestions and editorial
assistance of Dr. Donald L. Brooke as well as the collaboration of all
the personnel of the Lake Alfred Citrus Experiment Station.
Appreciations are also extended to all the persons in the United
States Department of Labor, United States Department of Agriculture and
Department of Agricultural Economics, University of Florida who read the
preliminary report and offered valuable suggestions and comments concerning
its form and content. The help of all the personnel that assisted with
computational and secretarial duties is sincerely appreciated.
This report is a slightly shortened version of the original study
submitted to the United States Department of Labor.
CONTENTS
Page
FOREWORD AND ACKNOWLEDGEMENTS ...... . . i
INTRODUCTION . . . . ... . 1
Importance and Purpose of the Study . . .. 1
Summary of Results . . . .... .. 2
Implications . . . . 4
I. THE CITRUS INDUSTRY AS AN EMPLOYER OF AGRICULTURAL LABOR 7
Employment Patterns . . . . 7
General Characteristics of the Citrus Harvesting Labor
Force . . . . . . 9
Skill Structure of the Harvesting Labor Force . .. 13
Productivity and Labor Requirements . . 13
II. PRODUCTION AND UTILIZATION OF FLORIDA CITRUS . ... 16 *
Production . . . .. . 16
Utilization of Florida Citrus . . ... 17
III. MECHANICAL HARVESTING SYSTEMS. BACKGROUND AND ECONOMIC
COMPARISON . . . . ... ... 22
Background . . . . ... . 22
Air Blast System . . . . ... .. 23
Mechanical Shaker and Catching Frame . .... .. .24
Shaker, Windrower and Pick-up Machine . ... .25
Economics of Mechanical Harvesting . . ... .26
IV. IMPACT OF MECHANIZATION ON LABOR REQUIREMENTS . .. .40
Projections of Total Production . . ... .41
Extent of Mechanization . . . ... 42
Calculating Labor Requirements with Mechanization 45 -"
Seasonality of Production and Mechanization . ... 48
Total Labor Requirements . . . ... 56 -f
Skill Requirements . . . .... ... 61
V. SOME GENERAL CONSIDERATIONS . . . ... 66
Some Supply Considerations . . . . 66
Opportunities for Off-season Employment . ... .68
Topics for Additional Research . . .. .72
iii
CONTENTS (Continued)
Page
APPENDIX . . . . ... . .... 78
BIBLIOGRAPHY .......................... 84
IMPACT OF MECHANICAL HARVESTING
ON THE DEMAND FOR LABOR IN THE
FLORIDA CITRUS INDUSTRY
by
J. Kamal Dow
INTRODUCTION
_IE0.ortance and Purpose of the Study
The Florida citrus crop is expected to grow from its present
potential of about 190 million boxes per year to almost 300 million
boxes per year a decade from now. 1if the present systems of manual
harvesting were continued, the labor requirements of the industry by
1980 would be over 50 percent higher than they are now. Current trends
indicate that the labor force available for agriculture will not be able
to satisfy those requirements. Furthermore, the migrant labor force, a
major factor in peak harvesting periods, is actually declining
[Because of the above, increasing emphasis has been placed on the
development of mechanical systems for harvesting citrus. The results
obtained so far indicate that mechanical harvesting of citrus for proc-
essing, as distinguished from fruit for table use, 'may become economic-
ally feasible in the near future. The adoption of the new mechanical
systems will bring substantial changes in the labor market.] Advance
knowledge of these changes will help in preventing the hardships that
might otherwise develop as a result of the mechanization process. It is
the main purpose of this study to explore the impact that the adoption
of mechanical citrus harvesting systems will have on the demand for
labor of different skill levels as well as the implications of that
impact.
Summary of Results
[Comparison of the different mechanical systems with the manual indi-
cated that should the present trends in wage rates, production and citrus
prices continue, it will be economically advantageous to use mechanical
harvesting systems by 1975 or before Due to the cautious attitude of
growers toward adopting new methods and to the fact that not too much
difficulty has been experienced so far in recruiting workers, it is anti-
cipated that the mechanization process will be slow over the next five
years and only 10 percent of the industry will be mechanized by the 1975-76
season. Mechanization should proceed at a faster pace after 1975. It
is anticipated that by 1980 the degree of harvesting mechanization will
range from 20 to 30 percent of the grapefruit to 50 to 60 percent of the
early and midseason oranges.
[One of the effects of mechanization will be to reduce the number of
workers needed to harvest a certain size crop Projected increases in
production and the low level of mechanization anticipated indicate that
the total harvesting labor requirements will continue to increase until
1975. Under the assumptions of the study, an average of about 25,000
workers will be needed that season during the peak months of January and
February as compared with about 23,000 during the 1969-70 season. Only
during one peak week of the 1969-70 season were 25,000 workers required.
The impact of mechanization will be felt by the labor market beginning
in 1975. In spite of the much higher expected level of production, fewer
workers will be required to harvest the 1980-81 crop than either the
1969-70 or the 1975-76 crops.
The most important impact of mechanization, however, will be on the
skill structure of the labor force. It is expected that between now and
1980 the demand for less skilled workers will be reduced by about 15
percent during the peak months while that for skilled and semi-skilled
workers will almost double. Although the demand for harvesting labor
will remain highly seasonal, it seems that enough opportunities will
exist for the employment of the skilled labor during the off-season.
Such is not the case for the less skilled workers and efforts will have
to be increased to find job opportunities for them whether it be in the
industry, in other agricultural industries, or in non-agricultural
activities.
It is expected that the results of this study will assist:
1. Federal, state and local government". Officials concerned
with manpower problems, adjustment of human resources, job
placement and labor policy making in general will benefit
from knowing the changes that will take place in the labor
market. Such knowledge will assist them in planning
programs aimed at improving the welfare of the labor
sector of the economy. Training programs for the unskilled
can be better structured if there is knowledge about the
demand for the different skills in the industry. This will
aid in reducing unemployment and underemployment by adjusting
the future labor supply to the future needs.
2. Management. The results as well as the methodology developed
in the study can be useful in making decisions concerning the
combination of hand and mechanical harvesting to fit different
conditions in the labor market. In addition, planning future
employment policies will be made easier when those conditions
are known. The availability of more stable employment for
some of the workers will result in changes in the traditional
employer-employee relationship that exists between pickers
and management. Employers will have to recognize that perma-
nent more-skilled workers will demand higher wages, better
working conditions, and fringe benefits. The possibility of
workers joining new or existing unions should not be dis-
counted by management.
Implications
Technological changes that substitute capital for labor always have
an impact on the labor market and unless preventive measures are taken
they usually affect adversely some of the workers involved. The magnitude
of the impact depends on how considerable the change in the capital-labor
ratio is. For a given change, however, it is the characteristics of the
workers affected along with the general market conditions which will
ultimately determine the adjustments that will take place in the level
of employment.
Mechanization will affect more adversely those workers whose mobility
is lower. Major determinants of labor mobility are age, education, race,
migratory status and work experience in other fields. These factors
affect the ability of workers to adjust to changes in the demand for
their services.
Younger workers should be less affected by mechanization since it
is easier to train them in the new skills demanded by the adoption of
mechanical systems. Similarly, it is also easier to train them for
activities other than those in the citrus industry. Over 60 percent of
the citrus pickers surveyed during the 1967-68 season were 44 years or
younger, while over 40 percent were 34 years or younger. Thus, it appears
that age would be a favorable factor in the implementation of manpower
training programs for citrus pickers.
A worker's occupational mobility is greater the higher his level of
education. Nearly 50 percent of the citrus pickers surveyed had less
than a fifth grade education. Although this indicates a low educational
level for pickers in general, the younger groups were better educated.
Education, coupled with age, will probably result in one group of younger,
better educated, workers which will be able to adjust to the new skill
requirements and another group of the older less educated which will fill
positions for which few or no skills are required.
Since some degree of discrimination exists in non-agricultural
employment, inter-industry mobility is limited among black and other
minority groups in the farm labor force. This fact is reflected in the
figures on citrus picking experience. While almost 50 percent of the
non-white males had 10 or more years experience, the figure for white
males was slightly over 20 percent. There were no white women with more
than 10 years of picking experience while for non-white women the figure
was 29 percent. Educational levels seem to be lower among non-whites,
which tends to add to the problem. The age distribution of the non-
white pickers was not significantly different from that of the white
pickers.
The composition of the citrus picking labor force as it relates to
the migratory status of the workers varies throughout the season. The
tendency is toward an increase in the percentage of interstate migrants
as the season progresses toward its peak and the need for pickers increases.
A reduction in the total number of workers required throughout the season,
and a smoothing of the peak periods will therefore affect interstate
migrant workers first.
Mechanization will reduce, but will not eliminate, the need for
migration in order to find supplementary work during the off-season.
Because of their unstable residential situation, migratory workers might
not be able to take advantage of the same type of programs that would
benefit local non-migratory workers. This fact, as well as the differences
in characteristics that exist among workers, should be taken into account
when implementing training and other types of programs.
The results of the study are encouraging in the sense that the course
that mechanization is expected to take seems to provide enough time for
implementing programs designed to ease the hardships usually associated
with transitions of the type dealt with in this study. Based on those
results it appears that efforts should be concentrated first on estab-
lishing training programs aimed at increasing skill levels to match those
required by the industry in the future, and secondly, on finding off-
season employment for those workers whose characteristics prevent them
from acquiring new skills and enjoying year-round employment in the citrus
industry.
CHAPTER I
THE CITRUS INDUSTRY AS AN EMPLOYER
OF AGRICULTURAL LABOR
Employment Patterns
The citrus industry employs more farm labor than any other agri-
cultural activity in the state of Florida. During the peak employment
period of the 1967-68 season, 30,673 or almost 40 percent of the 78,037
seasonal workers employed in Florida agriculture were employed by the
citrus industry.
Like most activities in specialized commercial agriculture the
need for labor in the citrus industry is highly seasonal. Figure 1 shows
employment in citrus during the 1968-69 season. Total employment varied
from 8,614 during the late part of September 1968 to 30,918 workers in
late January 1969. The number of workers used in harvesting ranged from
a low of 147 to a high of 23,823. The peak and low periods in harvesting
employment coincide with those of total employment.
The patterns illustrated in that figure are typical of Florida citrus
activities; there is a period of about 90 days, usually centered in late
August, where the need for harvesting labor is practically non-existent.
Starting in late October employment rises continuously and reaches a
peak usually in late January. It then starts falling until late March or
early April when the harvesting of most early and midseason varieties
has been completed. It is during this lull between the early and mid-
season and the late harvest when a critical period occurs; many unemployed
or underemployed workers leave the state thus making it harder for
FIGURE 1
FLORIDA AGRICULTURAL EMPLOYMENT IN THE CITRUS INDUSTRY
1968-69 SEASON
SOURCE: Farm Labor Statistics Department, Florida Industrial Commission,
employers to recruit pickers for the Valencia harvest. During the high
crop seasons of 1966-67 and 1968-69 it was necessary to bring in foreign
workers in June and July. The peak of the Valencia season usually occurs
during the month of May and after that employment drops steadily until
the end of the season.
Harvesting activities take up a large share of the citrus labor
force. The remainder is absorbed in production and maintenance activities
such as grove care, spraying and fertilizing. As can be observed in
Figure 1 non-harvest employment does not present seasonal characteristics
to the same extent as harvest employment. The number of workers engaged
in non-harvest activities varied from a low of 7,029 to a high of 9,446.
The peak occurs during the post-harvest period while the period of lowest
employment is during the winter months when harvesting labor requirements
are high. This permits movement of workers from one activity to the other
although, as the figures suggest, the amount is rather small when
compared with total labor requirements.
General Characteristics of the Citrus Harvesting Labor Force
Comprehensive statistics concerning the main characteristics of the
labor force employed in harvesting citrus are not available. The Florida
Industrial Commission in cooperation with the Bureau of Employment
Security, U. S. Department of Labor has undertaken the job of surveying
some of those characteristics. Two surveys of citrus harvest labor have
been made available so far; a sample of 55 crews covering a total of 512
workers are used in the 1967-68 Survey. Some of the results of this survey
are summarized below.
Race, sex and migratory status. Eighty five percent of the citrus
pickers in the Survey were non-white and 90.8 percent were males. The
ratio of men to women was the same for whites and non-whites, about 9
to 1. The bulk of the picking labor force was composed of non-white males
which accounted for 77.5 percent of the total. Slightly over two thirds
were non-migrating residents of Florida while 29 percent were interstate
migrants; only 2.5 percent were intrastate migrants. Two thirds of the
interstate migrants were non-white. The percentage of interstate workers
increases as the season progresses, however, The 1966-67 Survey pointed
out that the proportion of interstate workers had increased to 46 percent
in March and reached 50 percent in May.
Age and educational background. Ages of the citrus pickers ranged from
14 to 72 but almost two thirds of the sample fell into the 25-54 age
bracket. There were more workers (23 percent) in the 35-44 age group
than in any other; 18.6 percent of the workers were 24 years of age or
younger. Local workers were in general younger than interstate migrants.
Almost two thirds of the interstate migrants surveyed were 35 years or
older, compared with 45.4 percent of the local pickers. There was no
significant difference in the age distribution of whites vs. non-whites.
Nearly one half of the pickers interviewed had a fifth grade educa-
tion or less (10.9 percent had less than a third grade education and 37.7
percent were in the grades 3-5 bracket); 27.7 percent were in the
grades 6-9 group; 23.1 percent in the grades 10-12; and 0.6 percent received
more than a high school education. The younger groups, as would be
expected, were generally better educated than the older ones. This was
particularly noticeable in the 18 to 24 age bracket. In this group 86.1
percent of all men had better than a fifth grade education as compared
with 58.1 percent for the 25-34 bracket, and only 16.9 percent for the
55-64 group. Seventy five percent of the men in the 14-17 group had
better than a fifth grade education. This group was affected, of course,
by the fact that many of its members were still attending school and had
not completed their education.
The level of education was higher among white workers. The per-
centages of workers with more than a fifth grade education were 66.2 for
white males and 48.6 for non-white males. The difference was even wider
for female workers. While 77.8 of the white females had more than a
fifth grade education, only 47.4 of the non-white women had similar
school training. There are no data available for comparison of educa-
tional levels of migrants vs. non-migrants.
Mobility and citrus experience. According to the Survey, 88.5 percent
of the workers interviewed had worked for only one employer during the
season. The Survey points out that this figure is high because most of the
interviews took place near the beginning of the season. Turnover increases
as the season progresses and the average turnover rate is much higher than
the Survey indicates. This was illustrated in the previous survey (1966-67)
when only 16 percent of the December sample had worked for more than one
employer, but this percentage increased to over 50 percent for the May
sample.
Over 80 percent of the pickers in the sample had two or more years
experience and over 40 percent had 10 years or more. Experience was
greater among non-whites and Florida residents than among whites and
interstate migrants.
Average employment and earnings. Pickers worked an average of 6.1 hours
per day 3.5 days per week for an average of 21.4 hours per week. The
average number of weeks worked during the season was 5.1 per picker.
Employment was higher on Tuesday and Wednesday and lower on Monday,
Thursday and Friday. The Survey points out that the total man-weeks
for the season was determined by a count of all workers who worked at
all during any week. For example, a picker who joined a crew and worked
for only one day was counted as a man-week. Since the average number
of hours per week and days per week was computed by dividing total man-
hours and total man-days, respectively, by total man-weeks the results
tend to be biased downward.
Citrus pickers are paid on a piece rate basis. Prevailing rates
during the 1967-68 season were 30 cents for oranges, 20 cents for grape-
fruit, 85 cents for tangerines and 45 cents for tangelos and murcotts.
The distribution of average hourly earnings showed that 27.1 percent of the
workers earned $2.50 per hour or more whereas only 2.1 percent received
less than $1.15 per hour. The average hourly earnings for all types of
fruit was $2.36 per hour. Average weekly earnings amounted to $50.60 but
over 50 percent of the workers earned less than $30.00 per week. Low
weekly earnings are a result of working only a few days per week, not of
low hourly earnings. This seems to indicate that a large number of
workers consider citrus employment as a means to achieve a certain target
income rather than as a desirable full time occupation. Once their
target income for the week is achieved they seem to lose their motivation
for working. This hypothesis is supported further by the fact mentioned
above that employment is higher on Tuesdays and Wednesdays and lower on
Monday, Thursdays and Fridays.
Skill Structure of the Harvesting Labor Force
The statistics published by the Florida Industrial Commission provide
a breakdown of the harvesting labor force between pickers and non-pickers.
The pickers can be assumed to be relatively less skilled workers. Non-
pickers include mainly tractor drivers, goat truck operators and crew
leaders, that is, workers with a certain degree of mechanical or super-
visory ability. They can be considered as being semi-skilled and skilled.
The ratio of skilled and semi-skilled to relatively less skilled workers
remains fairly stable throughout the harvesting period as the figures in
Table 1 show for 1968-69. A least squares line was fitted to the data
available. After eliminating the slack period (July 31 to October 15)
there were 64 observations remaining between October 31, 1966 and May 31,
1969. The following equation was obtained:
Y = 185.90 + 0.13 X (R2 = .82) (I-a)
where Y is the number of non-pickers and X is the number of pickers. In
seven of the nine harvesting months the X ranged from around 12,000 to
around 20,000 which gives ratios of 6.9:1 and 7.2:1 pickers to non-
pickers, respectively. It can be said that on the average the present
manual harvesting system takes one skilled or semi-skilled worker for
each seven less skilled workers.
Productivity and Labor Requirements
The Florida Industrial Commission Survey estimated the average pro-
ductivity of pickers for different fruit varieties. Production rates in
boxes per picker hour were 7.4 and 6.8 for early and midseason oranges
for 1966-67 and 1967-68, respectively, 7.3 and 7.2 for valencias and 11.0
and 10.9 for grapefruit. Production rates can be used to calculate the
number of man-hours needed to harvest a given amount of fruit. It is
TABLE 1
CITRUS HARVESTING. RATIO OF PICKERS TO NON-PICKERS
1968-69 SEASON
Period Number of Ratio
Pickers Non-pickers
10-31-68
11-15-68
11-30-68
12-15-68
12-31-68
1-15-69
1-31-69
2-15-69
2-28-69
3-15-69
3-31-69
4-15-69
4-30-69
5-15-69
5-31-69
6-15-69
6-30-69
7-15-69
Average
4,011
7,289
11,651
15,569
17,460
19,652
21,206
20,343
19,651
18,382
15,132
14,180
15,149
15,486
14,567
13,749
12,381
7,670
498
1,104
1,491
1,962
2,110
2,562
2,617
2,629
2,601
2,439
2,087
1,903
2,110
2,118
2,099
1,926
1,709
1,000
8.0:1
6.6:1
7.8:1
7.9:1
8.2:1
7.7:1
8.1:1
7.7:1
7.6:1
7.5:1
7.3:1
7.4:1
7.2:1
7.3:1
6.9:1
7.1:1
7.2:1
7.7:1
7.5:1
SOURCE: Farm Labor Statistics
Tallahassee, Florida.
Department, Florida Industrial Commission,
important, however, to know how many different workers are needed during
a specified period of time to harvest that fruit. The relationship
between man-hours and number of workers depends on the average number of
hours worked per picker as well as on the turnover rate.
Based on employment figures published by the Florida Industrial
Commission some equations for determining labor needs were developed by
Leo Polopolus of the Economic Research Department, Florida Citrus
Commission.- Using linear regression techniques, he estimated the
relationships between weekly boxes harvested and number of harvest
workers. The following different relationships for the different sub-
seasons were found:
Y = 444 + 4.634 X (Early season) (I-b)
Y = 9,747 + 1.7222 X (Midseason) (I-c)
Y = 3,383 + 2.698 X (Late season) (I-d)
where X is weekly boxes harvested in thousand and Y is number of harvest
workers on an average daily employment basis.
Applying these equations to the 1968-69 production data gives a
peak harvesting employment of 23,261 workers. The actual figure was
23,823 or less than 2 percent different from the predicted value.
Subject to some qualifications to be discussed later, the above equations
can be used with confidence to determine harvesting labor requirements
under the manual system.
1/Leo Polopolus, A Mathematical Determination of Weekly Harvest
Labor Requirements for Florida Citrus. Economic Research Department,
Florida Citrus Commission, University of Florida, August 1968, 36 pages.
(Unpublished paper.)
CHAPTER II
PRODUCTION AND UTILIZATION
OF FLORIDA CITRUS
The transition between the employment patterns discussed in Chapter I
and those projected later in the study needs to be viewed in the light of
certain characteristics of the citrus industry. Such characteristics as
production and utilization will exert great influence in the changes
that will take place in the demand for labor.
Production
Florida is the leading citrus-producing state in the nation. During
the 1968-69 season it produced 60 percent of all the oranges, 77 percent
of all the grapefruit and 83 percent of all the tangerines produced in
the United States.
Florida oranges fall into three general categories depending on the
time they reach maturity: Early varieties which are harvested during
late October and November; midseason varieties which are harvested starting
in December and lasting until the middle of March; and late varieties
which are harvested from March until the early part of the summer. This
distinction is important for this study not only because of the influence
of maturity date on the seasonal labor requirements but also because the
late varieties present more complex problems for mechanization. In
grapefruit, the important distinction for mechanization purposes is between
the seedy and the seedless varieties because of their different final
utilization.
Table 2 shows the volume of production of the main Florida citrus
from 1950-51 to present. If the latest and earliest five-year period
averages are compared, an increase of over 40 percent in production can
be observed for oranges. Both the early-midseason and the late groups have
increased production although the former has done it at a slightly faster
rate than the latter. The increase in orange acreage discussed above seems
to indicate that the upward trend in production should continue in the
future. Projections of future production will be discussed in more
detail in a later chapter.
Production of grapefruit has remained at a relatively stable level
over the last two decades, but the variety composition of that production
has changed. There has been a substantial decrease in the production of
seeded grapefruit that has been offset by an increase in production of
the seedless varieties. Grapefruit production should show an increasing
tendency in the near future due to the substantial increases in plantings
that have taken place during the last five years.
Tangerine production has shown a slight downward trend during the
last twenty years. Average annual production for the 1965-69 period was
4.0 million boxes compared with 4.8 for the 1951-55 period. This down-
ward trend should be reversed in the near future due to the increased
plantings of the last five years.
Utilization of Florida Citrus
Prior to World War II over 80 percent of the total Florida orange
crop was shipped to the fresh market. Since then, larger volumes of
production plus the creation of new types of canned and processed citrus
TABLE 2
FLORIDA PRODUCTION OF ORANGES, GRAPEFRUIT AND TANGERINES
1950-51 to 1968-69
/ Oranges Grapefruit Tangerines
Crop Season Early &
Midseason Late Total Seeded Seedless Total
- - - -Million Boxes- - - -- - -
1950-51 36,800 30,500 67,300 17,400 15,800 33,200 4,800
1951-52 43,800 34,800 78,600 18,300 17,700 36,000 4,500
1952-53 42,300 29,900 72,200 15,400 17,100 32,500 4,900
1953-54 50,200 41,000 91,200 20,100 21,900 42,000 5,000
1954-55 52,000 36,400 88,400 14,300 20,500 34,800 5,100
1955-56 51,500 39,500 91,000 17,700 20,600 38,300 4,700
1956-57 54,300 38,700 93,000 14,800 21,600 37,400 4,800
1957-58 52,700 29,800 82,500 13,500 17,600 31,100 2,100
1958-59 47,100 38,900 86,000 14,600 19,600 35,200 4,500
1959-60 49,000 42,500 91,500 10,400 20,100 30,500 2,800
1960-61 51,000 35,700 86,700 12,400 19,200 31,600 4,900
1961-62 56,900 56,500 113,400 11,200 23,800 35,000 4,000
1962-63 45,500 29,000 74,500 10,000 20,000 30,000 2,000
1963-64 27,800 30,500 58,300 6,600 19,700 26,300 3,600
1964-65 46,400 39,800 86,200 10,200 21,700 31,900 3,900
1965-66 51,500 48,900 100,400 11,200 23,700 34,900 3,600
1966-67 78,200 66,300 144,500 13,500 30,100 43,600 5,600
1967-68 55,900 49,100 105,000 9,200 23,700 30,900 2,800
1968-69 74,200 61,000 135,200 12,200 28,000 40,200 4,300
Preliminary
SOURCE: Florida Agricultural Statistics, Florida Department of Agriculture, Tallahassee,
products have contributed to increasing the percentage of the orange crop
that is processed before being marketed. At the present time over 80
percent of the Florida orange production is utilized by processors. The
figures in Table 3 clearly show the tendency to process a larger part of
the orange crop. This trend should continue in view of the likely
increase in future production and the failure of fresh orange consumption
to keep up with it. The proportions of fruit marketed fresh and processed
do not differ significantly between the early-midseason and the late
varieties. During the last ten years, on the average, 78.4 percent of
the total production of early and midseason oranges has been processed
compared with 83.5 percent of the late varieties.
Grapefruit is not processed to the same extent that oranges are
and there is no clear trend. Approximately the same numbers are being
processed now as were during the late forties. Over 80 percent of the
acreage planted over the last five years belongs to the seedless varieties.
Only about 50 percent of seedless grapefruit goes to processors as
compared with 90 percent of the seeded type. Thus, the increased number
of seedless that goes to processors (due to increases in production) has
offset the effect of the decrease in production of seeded varieties.
Unless consumption of fresh grapefruit increases in the future larger numbers
will have to be utilized by processors.
The number of tangerines used by processors has fluctuated widely
from year to year. It was as low as 397,500 boxes in 1962-63 and as high
as 1,587,000 boxes in the 1960-61 season. The number of boxes going to
the fresh market has not varied as widely and, except for the years of low
production, it has fluctuated between 3.5 and 4.5 million boxes per year.
TABLE 3
UTILIZATION OF FLORIDA CITRUS
1950-51 to 1967-68
Season Percent of Crop Used by Processors
Oranges Grapefruit Tangerines
- - - Percent - - -
1950-51 61 53 20
1951-52 59 37 14
1952-53 62 45 22
1953-54 68 47 21
1954-55 68 44 21
1955-56 71 48 20
1956-57 73 51 26
1957-58 78 53 17
1958-59 80 50 35
1959-60 76 47 17
1960-61 80 51 32
1961-62 81 47 31
1962-63 84 53 20
1963-64 78 44 32
1964-65 80 50 27
1965-66 83 56 19
1966-67 86 60 19
1967-68 85 55 24
SOURCE: Florida Canners Association, Statistical Summary, Winter Haven,
Florida, Published annually.
21
This seems to indicate a rather stable market for the fresh product.
Unless demand for fresh tangerines increases in the future, larger
amounts will have to be processed in order to take care of the projected
increases in production.
CHAPTER III
MECHANICAL HARVESTING SYSTEMS
BACKGROUND AND ECONOMIC COMPARISON
Background
Harvesting is the only stage in the Florida citrus industry in
which mechanization has not kept pace with other crops. Machines are
used for cultivating, fertilizing, spraying and tree shaping, not to
mention the operations that take place after the fruit leaves the grove.
The harvesting operation, however, is still done entirely by manual means.
Many attempts have been made to develop mechanical aids to help
increase the productivity of the picker. Some of these are of a very
simple nature such as clipping devices attached to the end of a long
pole. Others like the man-positioning machines are of a more sophisti-
cated nature. That type of aid has not been successful because its high
capital cost can not be offset by the gains in productivity.
True mechanical harvesters can be classified as belonging to either
of two classes: contact removal harvesters or mass removal harvesters.
Contact removal harvesters have not been very successful because of their
removal potential being limited to the outer 18 inches of the canopy
thus leaving up to 40 percent of the fruit on the tree. Although they
are not economically feasible at the present time, contact removal type
harvesters seem to offer possibilities for use with the late oranges
because of their dual crop characteristic. By making direct contact with
the fruit, the ripe fruit can be removed while leaving the young fruit
on the tree.
Mass removal systems do not present as many problems as contact
systems and they seem to be at a more advanced stage of development.
It is the general consensus that systems of this type will be the first
ones to be used extensively in Florida citrus. The operation of these
systems has been satisfactory in all citrus except for valencia oranges
where some reduction in the following year's crop takes place due to the
presence of the young fruit at the time of harvesting. It is expected
that some abscission chemicals will be developed which will alleviate
this problem in the future. Three mass removal systems were studied
and compared with the manual system:
Air Blast System
This system uses an oscillating blast of high velocity air that
shakes the tree limbs and removes the fruit which falls on an attached
catching frame. A conveyor moves the fruit from the catching frame to
a basket; once the basket is full it is unloaded into a goat truck
which takes it to the roadside trailer.
One air blower, two tractors, two catching frames and the services
of a highlift truck are required for the operation of this system.
Personnel requirements are two tractor drivers, one blower operator,
one driver for the highlift truck and two unskilled workers to pick
up oranges that fall outside the frames or truck and keep the system
clear of branches and any large foreign material. Future design
improvements are expected to eliminate the blower operator and the non-
skilled workers.
At its average speed of one-fourth of a mile per hour, and taking
into account the time it takes for the machine to turn from one row to
the next, the rate of harvesting of this system is about 30 trees per
hour.- The rate of removal without the use of abscission chemicals is
about 80 percent of the fruit on tree. Some pruning of the trees is
necessary in order to operate the system smoothly; cost of shaping the
trees was taken to be $0.24 per tree per year. Some fruit needs to have
the stems detached and there is slight fruit damage; the cost of stem
removal and fruit damage was assumed to be $0.04 per box. This system
offers the best labor saving potential of all three. Under present con-
ditions, and based on average yields for older trees, each worker employed
with this system will do the work of 3.7 average manual pickers. With
the improvements expected for 1980 each worker employed with an air blast
system will replace 7.4 pickers. The labor replacement potential varies
directly with yields. It can be calculated using the figures in Table 8,
Chapter IV and the productivity rates of manual pickers given in Chapter
II. For a successful operation this system requires very flat terrain
for stability and long rows of trees in order to minimize the time spent
turning from one row to the next.
Mechanical Shaker and Catching Frame
This system consists of two tractor-drawn catch frames each equipped
with a limb shaker to remove the fruit from the trees. The catching
frames automatically collect the harvested fruit and store it in a bin
equipped to unload into a conventional lift body grove truck which
delivers the fruit to a roadside trailer. Each truck can serve two
-/Harvesting rates for all systems are based on an average distance
between trees of 20.5 feet.
systems of shakers. The personnel requirements are two tractor driver-
operators, one unskilled man to pick up fruit from the ground and remove
branches and foreign material, and a highlift truck driver. Future
developments will probably eliminate the need for the unskilled worker.
The rate of harvesting of this system is 15 trees per hour and the
efficiency without abscission chemicals is 90 percent of the fruit on
tree. Pruning of the trees is necessary and some holes have to be
opened in the canopy in order to shake the tree better; the cost of
shaping the trees was taken to be $0.30 per tree per year. There is
less fruit damage than in the air blast system and less fruit with stems
attached; the cost of detaching stems and damaged fruit was assumed to be
$0.02 per box. Under average conditions in the older groves, each worker
employed in connection with this system can presently replace 3.2 average
pickers. Potential improvements in the system may raise this figure to
4.5 by 1980.
Shaker, Windrower and Pick-up Machine
This system uses the same type of mechanical shaker described above
but it shakes the fruit to the ground where it is windrowed by mechanical
means. A pick-up machine later picks up the fruit and unloads it auto-
matically into a highlift truck which takes it to the roadside trailer.
Each windrower and pick-up machine can take care of the fruit shaken by
five shakers; the same is true for the truck.
If one system is defined as five shakers, one windrower, one pick-
up machine and one highlift truck, then the rate of harvesting is 43 trees
per hour. The efficiency without the use of abscission chemicals is 90
percent of the fruit on tree. Personnel required are five tractor
driver-shaker operators, one windrower operator, one pick-up machine
operator, one highlift truck driver and two unskilled workers to pick up
missed fruit and remove foreign material. Future developments will
eliminate the need for the last two workers. Some pruning of the trees
is necessary although less than in either of the first two systems since
no catching frame is required; the cost of pruning the trees was assumed
to be $0.15 per tree per year. Fruit damage is higher than in the other
two systems; cost of damaged fruit was assumed to be $0.06 per box. This
system is suitable for use in those instances where the shape of the
trees makes it difficult to use catching frames. Present and potential
labor replacement characteristics for this system are the same as for the
shaker and catching frame system.
Economics of Mechanical Harvesting
It seems reasonable to assume that in order for the mechanical
systems to be adopted, they must compare favorably in economic terms
with the manual system. Since the direct and indirect input requirements
and the harvesting efficiency of all systems are known it is possible
to make a comparison of their economic efficiency with that of the
manual system.
In order to make the comparison, the harvesting operation will be
considered as a separate economic enterprise. The decisions concerning
it will be considered as being independent of prior decisions as to how
much to plant, what varieties and methods to use, amount of fertilizing,
spraying, etc. Therefore, the decision-maker will face a situation in
which he has a certain amount of fruit on the trees and he will try to
maximize his profits from picking and selling it. By harvesting operation
is meant the process that takes place from the time the decision to harvest
is made until the fruit is loaded onto a roadside truck. A computerized
budgeting model in which all prices could be varied was used for the
comparison of the systems.
All three mechanical systems under consideration have design
characteristics which do not provide them with much flexibility as to
the area covered or number of trees harvested per unit of time. It
follows then, that yield becomes a most important parameter. The per
unit cost of mechanical harvesting will vary inversely with yield levels
while the per unit cost of manual harvesting is, for all practical
purposes, independent of yield. Similarly, the price of the fruit is a
very important variable in determining the advantage of one system over
another. Since the mechanical systems leave part of the fruit on the
tree, it follows that high fruit prices tend to favor the manual system
and vice versa. Finally, wages play an important role in comparing
economic efficiencies. Since the ratio of labor to other inputs is
relatively higher for the manual system, it follows that increases in
wages will tend to favor the mechanical systems.
From the above, it becomes apparent that whether the citrus har-
vesting operation is mechanized or not will depend as much on consider-
ations of yield, fruit prices and wage levels as it does on the Charac-
teristics of the equipment. Therefore, in comparing the systems it is
necessary to make certain assumptions regarding the values that those
variables might take in years considered.
Starting with the 1965-66 season, the Florida Crop and Livestock
Reporting Service began estimating orange yields in boxes per tree by
age classes. The results are shown in Table 4 both for the early and
midseason and for the Valencia varieties. The four seasons for which
data are available include a bumper crop year (1966-67), a low year
(1967-68) and two normal years in which the yield value closely
approached the four-season average. Since there is no reason to do
otherwise, the four-year averages will be used in the comparison.
Data for prices at roadside are not available. There are, however,
data on "delivered in" prices, that is, the price that processors of
fresh concentrated orange juice pay for fruit delivered at their plants.
Approximately 80 percent of all oranges processed goes into fresh concen-
trated orange juice. Roadside prices will be assumed to equal "delivered
in" prices minus the average cost of hauling from the roadside to the
2/
plant. A time series of average hauling costs is available.-
A regression model was developed in order to investigate the rela-
tionship between "delivered in" prices of oranges and other variables
which a priori were believed to be relevant. The first explanatory
variable chosen was the size of the crop which was expected to be inversely
related to price. Beginning of season inventory of fresh concentrated
orange juice was the second variable used since it seems reasonable to expect
that the size of the inventory of the product will have an influence on
the price that processors are willing to pay for oranges. Time was the
last variable included in the model. It was expected to reflect changes
-/A. H. Spurlock, Costs of Picking and Hauling Florida Citrus Fruits,
Department of Agricultural Economics, University of Florida. Published
annually.
TABLE 4
FLORIDA ORANGES
AVERAGE YIELDS BY AGE CLASSES 1965-66 to 1968-69
Crop 4-9
Year Years
1965-66
1966-67
1967-68
1968-69
Average
1965-66
1966-67
1967-68
1968-69
Average
.5
1.2
1.0
1.1
1.0
10-14 15-24
Years Years
- -Boxes Per Tree- -
Early and Midseason
1.4 3.7
3.0 5.7
1.6 3.4
2.9 4.3
2.2 4.3
Valencias
3.1
4.2
2.6
3.4
3.4
SOURCE: Florida Crop and
Orlando, Florida.
Livestock Reporting Service,
25 Years
& Older
5.1
7.0
4.0
5.1
5.3
4.0
5.7
3.2
4.2
4.3
in general price levels and demand conditions. Several models were
tested and the best fit was obtained by using the following equation:
Log Y = 3.5797 1.6435 Log X1 + 0.4208 Log X2 0.3153 Log X3 (III-a)
where
Y = Price paid by processors of F.C.O.J. in dollars per box.
X1= Total production of oranges in million boxes.
X2= Coded variable indicating season.
X3= Beginning of season inventory of fresh concentrated orange
juice in million gallons.
The regression coefficients had the signs expected a priori and
were all statistically significant at the 99 percent level of probability.
2
The multiple correlation coefficient was R = 0.9079, which means that
90 percent of the variation in prices was explained by the variables
included in the equation. The data used in the regression are shown in
Table 5.
The limitations of the above model are obvious. It is not intended
to be a model for price forecasting but rather a structural relationship
among several relevant variables to be used to derive an idea about the
order of magnitude of the price, given certain assumed conditions.
Holding the inventory level constant at its mean value, the equation
shows that an orange crop of 180 million boxes in the 1975-76 season will
yield a price of $1.27 per box. This result seems to agree with
industry estimates. The DARE report estimates that "if a 180 million box
crop is realized in 1975, a price in the neighborhood of $1.25 per
box might be expected."!/ A crop of 220 million boxes in 1980, will,
-/DARE Report 1965, University of Florida, Institute of Food and
Agricultural Sciences. Gainesville, Florida.
TABLE 5
TOTAL PRODUCTION OF ORANGES, BEGINNING OF SEASON INVENTORY OF FRESH
CONCENTRATED ORANGE JUICE AND DELIVERED-IN PRICE PAID BY PROCESSORS
SEASONS 1952-53 THROUGH 1967-68
Season Production Beginning of Season Price Paid
Inventory
Million Boxes
72.3
91.3
88.4
91.0
93.0
82.5
86.0
91.5
86.7
113.4
74.5
58.3
86.2
100.4
144.5
105.0
Million Gallon
F.C.O.J.
8.29
4.55
11.15
8.92
12.95
10.95
6.45
16.32
9.52
13.42
33.75
15.40
10.14
21.81
11.99
26.64
1952-53
1953-54
1954-55
1955-56
1956-57
1957-58
1958-59
1959-60
1960-61
1961-62
1962-63
1963-64
1964-65
1965-66
1966-67
1967-68
$ Per Box
1.81
1.62
1.72
2.29
1.74
2.56
3.50
2.54
3.47
2.26
2.71
5.25
3.37
2.28
1.29
2.76
SOURCE: Florida Canners Association, Statistical Sumnary, Winter Haven,
Florida, Published annually.
according to the equation, produce a price in the neighborhood of $1.00.
Although this price seems low now, it seems reasonable to expect the
price of oranges to decline gradually as the crop size increases, other
things being equal.
Table 6 and Figure 2 illustrate the behavior of the labor and non-
labor components of the cost of picking Florida oranges from 1950-51 to
present. The tendency of the labor costs has been one of continuous
increase which has become particularly sharp over the last few years.
Labor costs per box increased 60 percent between the 1961-62 and the
1967-68 seasons, or at an annual cumulative rate of 8.2 percent. Increas-
ing employment opportunities outside of agriculture and the consequent
expected reduction in the agricultural labor force suggest that the above
trend is likely to continue in the future. Since skilled workers are
and will continue to be relatively scarcer than unskilled workers, their
wages should also be expected to rise as the demand for their services
increases as a result of mechanization.
The results of the comparisons of the systems are illustrated in
Figures 3, 4 and 5. The difference between the per-box price of oranges
at roadside and per-box harvesting costs, or "harvesting profit" has been
plotted against per-box price of fruit at roadside under two different
wage rates and two different yield levels.
4/
Since time was included as a variable in the regression equation,
the prices obtained are current prices. An implicit assumption is that
the average increase in the price level will remain the same as during
the observed period (1952-68).
TABLE 6
FLORIDA AVERAGE COST OF PICKING ORANGES,*
1950-51 to 1967-68
Labor as Per-
Year Total Cost Labor Other Labor as er
cent of Total
- - -- -cents per box -- - -- %
1950-51
1951-52
1952-53
1953-54
1954-55
1955-56
1956-57
1957-58
1958-59
1959-60
1960-61
1961-62
1962-63
1963-64
1964-65
1965-66
1966-67
1967-68
28.36
28.43
28.90
29.09
28.92
30.49
30.80
33.14
32.96
33.99
34.96
33.79
39.57
43.04
43.43
46.12
46.25
54.09
22.33
23.69
24.65
24.78
24.10
25.38
25.95
26.83
26.77
26.89
27.69
28.13
31.52
34.05
36.02
29.01
29.57
45.71
6.03
4.74
4.25
4.31
4.82
5.11
4.85
6.31
6.19
7.10
7.27
5.66
8.05
8.99
7.41
7.11
6.68
8.38
78.7
83.3
85.3
85.2
83.3
83.2
84.3
81.0
81.2
79.1
79.2
83.2
79.7
79.1
82.9
84.6
85.6
84.5
SOURCE: A. H. Spurlock, Costs of Picking and Hauling Florida Citrus
Fruits, Department of Agricultural Economics, University of
Florida. Published annually.
*To roadside.
FIGURE 2
TOTAL, LABOR AND NON-LABOR COST OF PICKING FLORIDA ORANGES
1950-51 to 1967-68
TOTAL COST
2
LABOR
1950-51 51-52 52-53 53-54 54-55 55-56 56-57 57-58 58-59 59-60 60-61 61-62 62-63 63-64 64-65 65-66 66-67 67-68
SEASON
SOURCE: A. H. Spurlock, Costs of Picking and Hauling Florida Citrus Fruits. Department of Agricultural Economics,
University of Florida. Published annually.
__
FIGURE 3
DIFFERENCE BETWEEN ROADSIDE PRICES AND HARVESTING COSTS FOR DIFFERENT
SYSTEMS AND FRUIT PRICES WITH GIVEN WAGE LEVELS AND YIELD
'2.0
o ^*
I0
'.4
w
Cd
-4
p-
015
0
1-4
o 1.5-
a
w
*u
*r(
1-1
C(
pa
0}
>-i
S-- 1966-67 Average Price
A.A I
= Is a ---- i I a a I -I -- -
'1.20 1.30 1.40 1.50 1.60
1967-68 Average Price I
* I I
1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70
Price of Fruit at Roadside (dollars per box)
Notes: Wage levels assumed here were those listed in Appendix Table 3.
Yield of 5.3 boxes per tree assumed.
019
2.80 2.90 3.00
)
I
!
FIGURE 4
DIFFERENCE BETWEEN ROADSIDE PRICES AND HARVESTING COSTS FOR DIFFERENT
SYSTEMS AND FRUIT PRICES WITH GIVEN WAGE LEVELS AND YIELD
- -.
-.
- .-
1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10
2.20 2.30 2.40 2.50 2.60 2.70
Price of Fruit at Roadside (dollars per box)
Notes: Assuming wages 50 percent higher than those assumed for Figure 4.
Yield of 5.3 boxes per tree assumed.
gSa"
p~c I
FIGURE 2
DIFFERENCE BETWEEN ROADSIDE PRICES AND HARVESTING COSTS FOR DIFFERENT
SYSTEMS AND FRUIT PRICES WITH GIVEN WAGE LEVELS AND YIELD
2.OF
'-sje
-I--. ~
-. -I-
~. -~
- # I I I I I I I I p I I
1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30
Price of Fruit at Roadside (dollars per box)
Notes: Assuming same wage levels as for Figure 5.
Yield of 4.3 boxes per tree assumed.
2.40 2.50 2.60 2.70
1.0O-
0.5
Figure 3 shows a comparison of the harvesting profits with a given
yield of 5.3 boxes per tree and the currently prevailing wage levels
listed in Appendix Table 3. It can be observed that except for low
fruit price (below $1.40 per box) the manual system is preferable to
either one of the systems using shaking devices. The air blast system
appears to be more profitable for fruit prices of up to $2.40 per box.
The average price for the 1967-68 season of over $2.60 per box does not
provide the incentive to mechanize. The 1966-67 price of $1.18 per box,
on the other hand, would have favored the use of mechanical systems.
Comparison of Figure 4 with Figure 3 illustrates the impact of
wages on the relative profitability of the systems. With everything
else the same as in Figure 3, wages have been raised by 50 percent in
Figure 4. If the past trend continues, this level of wages may corres-
pond to the situation around 1975. Figure 4 shows that the effect of
higher wages is to make the air blast system more profitable than the
manual for all fruit prices considered. The shaker systems, in the
other hand, compare favorably with the manual for all prices below $2.50
per box. Roadside prices of over $2.50 in 1975 are only a remote pos-
sibility. If the production projections are realized, such a price could
only be a result of a freeze or other unusual circumstance. Under normal
conditions, thus, it seems that there will be an incentive for mechani-
zation in the near future.
Comparison of Figure 5 with Figure 4 illustrates the effect of yield
on the relative profitability of the systems. The only difference between
the two figures is the yield level assumed (4.3 boxes per tree in Figure 5
and 5.3 boxes per tree in Figure 4). It can be observed that point E,
39
which marks the fruit price at which the profitability of the manual and
the shaker systems is the same, moves to the left with a decrease in
yield. In other words, other things being equal, lower yields require
lower fruit prices before the incentive to mechanize exists.
CHAPTER IV
IMPACT OF MECHANIZATION ON LABOR REQUIREMENTS
The trends observed in fruit prices and costs of harvesting labor
were discussed in Chapter III. Given those trends and the projected
levels of production to be discussed later it seems that incentives to
mechanize citrus harvesting will be present in the very near future.
Many industry leaders are already aware of this fact. On the other hand,
the engineering tools have been substantially developed over the last
decade, and the state of the art is expected to improve at an increasing
rate over the next few years. It appears that mechanization of citrus
harvesting is on the horizon. Its arrival should have a substantial
effect on the labor market.
The impact of mechanization on the demand for labor will be two-
fold. In the first place it will reduce the number of workers required
to harvest a certain size crop. Secondly, it will change the skill
structure of the labor force demanded. A higher percentage of skilled
workers will be needed to harvest a crop by mechanical means than would
be the case if the harvesting was done by conventional manual methods.
In order to determine the total number of workers required for
citrus harvesting in any particular year, several factors have to be
considered. The size of the crop to be harvested and the percentage of
the crop that will use mechanical harvesters have to be determined and
the type or types of systems to be used must be known or assumed. Once
the number of workers is determined, it is possible to break down the
total in terms of skill categories.
Projections of Total Production
The production of citrus for any particular year is difficult to
forecast since it depends on many factors. Some of these factors are:
economic abandonment of trees, availability of harvesting labor, intensity
of grove production practices, and weather conditions. The range within
which production in any year may fall is a very wide one. Potential pro-
duction, on the other hand, is relatively easy to calculate. It takes
five years for a tree to come into bearing and most of the production
comes from trees that are 10 years or older. Since there are figures
concerning tree numbers plus historical information on yields by age
groups, potential future production can be established with a certain
degree of accuracy, the factors mentioned above notwithstanding.
Estimates of potential future production are available from several
sources. The Florida Crop and Livestock Reporting Service has made some
calculations in that respect. The Economic Research Department of the
Florida Citrus Commission made a study of potential production using
1/
random sampling techniques.- The DARE Citrus Subcommittee made potential
production projections based on estimates provided by each individual
county. Projections from the different sources do not vary significantly
1/
L. Polopolus and W. B. Lester, Estimation of Florida's Orange
Production Over the Next Fifteen Years by the Random Sampling Technique,
Economic Research Department, Florida Citrus Commission, University of
Florida, Gainesville. September 1968. (14 Pages, Unpublished.)
and those published by the 1969 DARE Report will be used here. Those
projections are shown in Table 7.
Citrus production is expected to increase by 37 and 62 percent of
its present level by 1975 and 1980, respectively. In terms of number of
boxes, most of the increase will be in round oranges. Including temples,
the orange crop will increase by 56.8 and 91.8 million boxes by the
1975-76 and 1980-81 seasons, respectively. Some increase is expected in
grapefruit production as well as in the other citrus fruits. Round
oranges and grapefruit are by far the most important types of fruit for
the purposes of this study. Not only are they the major users of har-
vesting labor but the mechanical systems developed thus far are expected
to be used primarily on them in the near future. As the harvesting of
oranges and grapefruit becomes more mechanized, however, the relative
importance of the other citrus varieties as users of less skilled labor
will increase, and their trends should be carefully considered when the
total number of pickers is to be estimated.
Extent of Mechanization
With presently available knowledge, it is extremely difficult
to make predictions about the extent to which the industry will mechanize
during the next decade. Although the basic engineering technology has
been developed, it is difficult to estimate the willingness of the people
in the industry to undertake large investments in equipment which is to
a certain extent unknown to them. In the case where there was a general
awareness of the advantages of mechanization, and willingness to replace
capital for labor, there would still be a lag between the time when the
demand for the machinery appears and the time when large-scale production
TABLE 7
FLORIDA. CURRENT AND POTENTIAL PRODUCTION OF CITRUS
Type of Fruit
Oranges
Temples
Grapefruit
Tangerines
Tangelos
Murcotts
Lemons
TOTAL
1968-69
130.7
4.5
40.2
4.3
1.8
1.0
182.5
Season
1975-76
- Million Boxes -
185.5
6.5
44.0
5.8
4.5
1.5
2.5
250.3
1980-81
219.0
8.0
50.0
7.2
6.0
2.0
3.5
295.7
SOURCE: DARE Report 1969, University of Florida, Institute of Food
and Agricultural Sciences, Gainesville, Florida, October
1969. p. 9.
takes place. Assumptions about the percentage of fruit that will be
harvested mechanically in 1975 and 1980 belong in the category of
"educated guesses".
For the purposes of this study it will be assumed that 10 percent
of the Florida citrus crop will be mechanically harvested by 1975. This
assumption is the result of interviews with several of the engineers
involved in the development of the mechanical systems as well as with
people in the industry familiar with current progress in the field. Ten
percent mechanization of citrus harvesting by 1975 is considered reason-
able by most researchers in the field. Projecting the extent of mechani-
zation to 1980 is still more difficult. Here, the author relied mainly
on his own judgment and conversations with the engineers. The assumption
was made that once the critical period is overcome by 1975 and 10
percent of the industry is sold on mechanical harvesting, the substi-
tution of capital for labor will proceed at a much faster rate during
the following years. In other words, the period from 1975 to 1980 will
be the period of fastest growth in citrus harvesting mechanization.
This type of pattern is typical of most technological changes and is
consistent with the consensus of the DARE Engineering and Equipment
Committee:
Industry acceptance of mechanical harvesting for
processing fruit is expected to evolve over
several years, possibly gaining a firm foothold
by 1975. With adequate developmental effort the
main expansion period is expected to begin by 1975.
.'DARE Report 1969 University of Florida, Institute of Food and
Agricultural Sciences. October 1969. p. 145.
Based on the above, it has been assumed that by 1980, approximately
50 to 60 percent of all early and midseason oranges, 40 to 50 percent of
all Valencias, and 20 to 30 percent of all grapefruit will be harvested
mechanically. The difference between Valencia and the other oranges is
due to the fact that less progress has been made in the development of
a system that can harvest dual crop varieties efficiently. The lower
percentage for grapefruit is due to the fact that a higher portion of
the grapefruit crop goes to fresh use. Harvesting systems for tangerines
and other specialty citrus have not been developed to the point where it
would be safe to assume their eing used to any great extent by 1980. In
addition, tangerines and other citrus are only a minor item when compared
with oranges and grapefruit. Only the last two will be considered as
harvest mechanizable by 1975 and 1980.
Calculating Labor Requirements with Mechanization
The different mechanical systems and their personnel requirements
were described in Chapter III. The number of man-hours required to
harvest one box of fruit has been calculated for the different systems
and for two different yield levels and the results are shown in Table 8.
The first column in Table 8 shows the present man-hour requirements,
which will be assumed to prevail until 1975-76. The second column shows
the potential man-hour requirements provided that some of the labor
presently required is eliminated; it is assumed that this will be the
case by 1980.
Since the man-hour requirements are not the same for all systems,
the total labor needs will depend among other things on the proportions
in which the systems will be combined. Different groves have different
TABLE 8
PRESENT AND POTENTIAL MAN-HOUR REQUIREMENTS PER BOX
WITH DIFFERENT YIELD LEVELS
Man-hour Requirements
System Present Potential
- - -Man-hr/box- - - -
Yield Level = 5.3 boxes/tree
Air Blast .038 .019
Shaker & Catching Frame .044 .031
Shaker-Windrower-
Pick-up .044 .031
Average .042 .027
Yield Level = 4.2 boxes/tree
Air Blast .047 .023
Shaker & Catching Frame .054 .039
Shaker-Windrower-
Pick-up .054 .039
Average .052 .034
Since there is a direct relationship between yields and profit
margins, it is reasonable to assume that mechanical harvesting will be
directed primarily toward the older more productive trees, at least
during the first years. Advances in technology and/or scarcity of labor
may later serve as incentives for mechanization of younger trees. Simi-
larly, the younger trees, being smaller, are easier to harvest by hand.
Hence, the tendency will be toward using as many of them as possible for
obtaining the fruit that will be shipped for fresh use. It is assumed
that 75 percent of the fruit harvested mechanically comes from trees 25
years or older and the other 25 percent from trees that are between 15
and 24 years of age. This assumption is consistent with projected tree
population and production figures.
The proportions in which Valencias and the other varieties will be
harvested at any particular time are based on projected seasonal distri-
bution and mechanization patterns to be explained in the next section.
Based on them and on the tree age assumptions mentioned above, the
average number of boxes mechanically harvested per worker per week has
been calculated for each month, and for the two peak periods of the
1975-76 and the 1980-81 seasons. The results are shown in Table 9. It
can be noted that as the season progresses and the percentage of lower
yielding Valencia trees increases, the number of boxes harvested per
worker decreases.
Seasonality of Production and Mechanization
Citrus production in Florida follows a seasonal pattern which is,
for obvious reasons, similar to that shown by citrus employment and
discussed in Chapter I. Based on production data for four years
characteristics such as topography, length of rows, spacing between
trees, tree trunk shape, etc., which make a certain system more desirable
than the others for a particular grove. For the purposes of this study
it will be assumed that all three systems will be used in equal propor-
tions. The other variable affecting the manpower requirements is yield.
Since yield depends on the age of the tree, and since Valencias have
lower yields than early and midseason varieties, the man-hour require-
ments will depend on the age of the trees being harvested and on the
proportion of the crop being harvested that is made up of Valencias.
The following formula can be used to calculate the average number of
boxes harvested mechanically per worker in a 40-hour week:
B = 40 [01 P1 + 02 (1 P)] [1 P + [V P + V2 (1 P)] P (IV-a)
where:
B = Average number of boxes harvested mechanically per worker in
a 40-hour week.
01 = Average number of boxes per man-hour of early and midseason
oranges or grapefruit harvested mechanically from trees 25
years of age or older. Calculated from Table 8.
02 = Same as 01, but from trees 15-24 years of age.
V2 = Equivalent of 01 for Valencias.
V2 = Equivalent of 02 for Valencias.
P, = Proportion of mechanically harvested fruit cominggfrom trees
25 years of age or older.
P = Proportion of the mechanically harvested fruit which is made
of Valencias.
TABLE 9
MECHANICAL SYSTEMS. NUMBER OF BOXES HARVESTED WEEKLY PER WORKER
ASSUMING THAT THE THREE SYSTEMS ARE EMPLOYED IN EQUAL PROPORTIONS
Boxes harvested weekly per worker
Period 1975-76 1980-81
September 911 1,410
October 911 1,410
November 911 1,410
December 911 1,410
January 911 1,410
Early Peak 911 1,410
February 911 1,410
March 882 1,372
April 876 1,296
Late Peak 862 1,283
May 862 1,283
June 862 1,277
July 853 1,289
Based on 40 hours per week operation.
Based on 40 hours per week operation.
Polopolus estimated the seasonal utilization of Florida citrus on a weekly
3/
basis.- Although a particular year's crop may be earlier or later than
average, the general shape of the distribution will tend to be the same.
Thus, once the size of the crop is estimated, weekly utilization in boxes
can be estimated with a certain degree of accuracy. Similarly, the average
number of boxes utilized per week for any particular month can be calcu-
lated. Showing one average weekly figure for each month simplifies the
presentation without any loss of perspective of the seasonality of produc-
tion. Weekly rather than monthly averages, however, must be used in order
to be able to make use of equations I-b, I-c, and I-d. Projected distri-
butions by varieties are shown in Table 10. Average weekly figures in
boxes are given for each month in the season and for the two peak weeks.
Once the seasonal distribution of the projected production has been
determined the ne:t step is to determine what part of that production is
available for mechanization and how much will actually be harvested
mechanically. Since the mechanical systems can only be used on fruit that
will be processed, the amount of fruit that is picked for fresh use must
be subtracted from total production in order to determine how much fruit
is available for mechanical harvesting. The percentage of fruit that is
harvested for fresh use is higher than the percentage of fruit actually
sold in the fresh market since some of the fruit is rejected at the
packing house. The actual percentage destined for processing when har-
4/
vested has been calculated by Whitney- using the following formula:
3/
Polopolus, Leo. A Mathematical Determination of Weekly Harvest Labor
Requirements for Florida Citrus. Economic Research Department, Florida Citrus
Commission, University of Florida, Gainesville, August 1968. (Unpublished).
4/
Whitney, Jody. Importance of Citrus Harvesting Mechanization. Citrus
Experiment Station, Lake Alfred, Florida. (Unpublished).
TABLE 10
PROJECTED DISTRIBUTION OF FLORIDA CITRUS PRODUCTION
BY MONTH AND PEAK WEEKS ON AN AVERAGE WEEKLY BASIS, 1975-76 and 1980-81
Period Average Weekly Production
Early & Mid- Tangerines
season oranges Valencias Grapefruit Tangerines & Murcotts
1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81
- - - - Million Boxes - -- -- - - -
September .07 .08
October .38 .46 .65 .74 .01 .01 .03 .04
November 1.58 1.91 1.10 1.25 .21 .26 .30 .41
December 4.59 5.55 1.25 1.41 .72 .90 .39 .51
January 7.43 8.99 .24 .28 1.43 1.62 .31 .39 .19 .24
Early Peak 8.85 10.70 .32 .37 1.55 1.77 .07 .08 .30 .39
February 7.08 8.56 .69 .80 1.50 1.70 .05 .07 .25 .33
March 2.73 3.30 2.28 2.62 1.82 2.07 .02 .02 .16 .21
April .40 .48 4.85 5.57 1.41 1.61 .06 .07
Late Peak 7.20 8.28 .97 1.11 .01 .02
May 6.57 7.55 .65 .74 .01 .01
June 4.37 5.02 .25 .28
July 1.19 1.37 .10 .11
Calculated from production projections in Table 7 and seasonal distributions as calculated by
Polopolus.
PC = 100 11 ] (IV-b)
where:
PC = Percent of total crop which was meant to go for processing
when harvested.
F = Percent of total crop which was sold as fresh fruit.
PK = Percent of fruit picked from packing house which was sold for
fresh use. (Percent Packout.)
The F and PK values used by Whitney in his paper indicate that 30
percent of all oranges and 60 percent of all grapefruit are harvested
for fresh use. These percentages will be applied to the projected levels
of production. Average weekly production by month and the quantities
harvested for fresh use are shown in Tables 11 through 13.
The approximate percentages of the total annual crop that will be
mechanized have been determined earlier in this chapter. Some criteria
are needed now in order to determine how those quantities will be dis-
tributed in a way that is consistent with the seasonal nature of pro-
duction. The main criterion used was to make use of the entire equipment
and the personnel that operates it for as long a period as possible. Due
to the seasonality of production, however, this period turned out to be
only about six months, December through May, for the 1980-81 season.
Nine months of full use were possible in the case of the 1975-76 season
due to the low mechanization assumption. Because of its larger impact on
the labor market, however, the former case is of more interest for this
study. Tables 11 through 13 show the amounts of the fruit harvested for
processing that will be picked by manual and mechanical means. The last
four columns in each table show the totals for mechanically and manually
harvested fruit.
TABLE 11
EARLY AND MIDSEASON ORANGES. AMOUNTS TO BE HARVESTED
MECHANICALLY AND MANUALLY ON AN AVERAGE WEEKLY BASIS, 1975-76 and 1980-81
Average Hand Harvested
Average Hand Harvested Harvested for Processing
Period Weekly for Total Hand
Production Fresh Use Hand Mechanically Harvested
1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81
- - - - Million Boxes - - - - -
October .38 .46 .11 .14 .27 .32 .11 .14
November 1.58 1.91 .47 .41 .56 .55 1.50 1.03 41
December 4.59 5.55 1.37 1.66 2.67 .39 .55 3.50 4.04 2.05
January 7.43 8.99 2.23 2.70 4.65 2.49 .55 3.50 6.88 5.49
Early Peak 8.85 10.70 2.65 3.21 5.65 3.99 .55 3.50 8.30 7.20
February 7.08 8.56 2.12 2.57 4.41 2.49 .55 3.50 6.53 5.06
March 2.73 3.3G .71 .85 1.77 .25 2.45 2.48 85
April .40 .48 .12 .14 .11 .17 .34 .23 .14
SEASON
TOTAL* 105.00 127.00 31.50 38.10 61.55 24.44 11.95 64.46 93.05 62.54
*Totals are based on the following numbers of weeks at each month's rate: October 2; November and
January 5; December, February and March 4; April 3. All figures rounded to nearest hundredth.
TABLE 12
VALENCIA )RANGES. AMOUNTS TO BE HARVESTED
MECHANICALLY AND MANUALLY )N AN AVERAGE WEEKLY BASIS, 1975-76 and 1980-81
Average Hand Harvested
Average Hand Harvested Harvested for Processing
Period Weekly for arrested for ProcessiTotal Hand
Production Fresh Use Hand Mechanically Harvested
1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81
- - - - Million Boxes - - - - -
January .24 .28 .07 .08 .17 .20 .24 .28
Early Peak .31 .36 .09 .11 .22 .25 .31 .36
February .69 .80 .21 .24 .48 .56 .69 .80
March 2.28 2.62 .68 .79 1.32 .88 .28 .95 2.00 1.67
April 4.85 5.57 1.46 1.67 3.04 1.05 .35 2.85 4.50 2.72
Late Peak 7.20 8.23 2.16 2.47 1.53 2.61 .51 3.15 6.69 3.08
May 6.57 7.55 1.97 2.27 4.09 2.13 .51 3.15 6.06 4.40
June 4.37 5.02 1.31 1.51 2.51 1.01 .55 2.50 3.82 2.52
July 1.19 1.37 .36 .41 .22 .26 .61 .70 .58 .67
SEASON
TOTAL* 87.00 100.00 26.10 30.00 51.19 28.25 9.71 41.75 77.29 58.25
January 2; February,
hundredth.
*Totals are based on the following numbers of weeks at each month's rate:
March, April, June and July 4; May 5. All figures rounded to the nearest
TABLE 13
GRAPEFRUIT. AMOUNTS TO BE HARVESTED
MECHANICALLY AND MANUALLY ON,AN AVERAGE WEEKLY BASIS, 1975-76,and 19-80-81
Average Hand Harvested
Average Hand Harvested Harvested for Processing
Period Weekly for Hvested for Total Hand
Production Fresh Use Hand Mechanically Harvested
1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81 1975-76 1980-81
- - - - Million Boxes -- - - - - -
September .07 .08 .05 .05 .02 .03 .05 .05
October .65 .74 .42 .44 .10 .13 .30 .52 .44
November 1.10 1.25 .72 .75 .25 .12 .13 .38 .97 .87
December 1.25 1.41 .81 .85 .31 .18 .13 .38 1.12 1.03
January 1.43 1.62 .93 .97 .37 .27 .13 .38 1.30 1.24
Early Peak 1.55 1.77 1.01 1.06 .41 .33 .13 .38 1.42 1.39
February 1.50 1.70 .98 1.02 .39 .30 .13 .38 1.37 1.32
March 1.82 2.07 1.18 1.24 .51 .35 .13 .38 1.69 1.59
April 1.41 1.61 .92 .97 .36 .26 .13 .38 1.28 1.23
Late Peak .97 1.11 .63 .66 .21 .07 .13 .38 .84 .73
May .65 .74 .42 .44 .10 .13 .30 .52 .44
June .25 .28 .16 .17 .09 .11 .16 .17
July .10 .11 .07 .07 .03 .04 .07 .07
SEASON
TOTAL* 44.00 50.00 28.60 30.00 10.33 7.88 5.07 12.12 38.93 37.88
2: October,
figures
*Totals are based on the following numbers of weeks at each month's rate: September
December, February, March, April, June and July 4; November, January and May 5. All
rounded to nearest hundredth.
Total Labor Requirements
Given the amounts of fruit to be harvested both manually and
mechanically total employment for each period can be estimated. The
labor requirements for harvesting fruit manually can be calculated with
the help of equations I-b, I-c and I-d. In the case of mechanically
harvested fruit, the labor requirements are found by dividing the figures
on Table 9 into the corresponding number of boxes to be harvested. The
total amounts of fruit to be harvested and the number of workers required
in each case, as well as the total labor requirements, are shown in
Tables 14 and 15.
Table 16 shows a comparison between the labor requirements esti-
mated for 1975-76 and 1980-81 and those for the 1969-70 season on the
basis of the October 1969 crop estimate. A comparison of the three
columns will show the impact that the adoption of mechanical systems
will have upon the demand for harvesting workers. Even in 1975-76 with
only about 10 percent of the citrus crop being mechanized, the increase
in labor requirements is lower throughout the season than the corres-
ponding increase in crop size. Employment during the peak week of the
season is expected to reach 28,410 workers in 1975-76 compared with
25,340 estimated for the 1969-70 season or an increase of 12 percent in
the number of workers required to harvest a crop 30 percent larger.
The difference between the percentage increases in crop size and
in the number of workers required is much wider in the case of the 1980-81
season, as would be expected. Labor requirements during the peak week of
the season are only 13 percent higher than in the 1969-70 season in spite
of a crop over 50 percent larger. It was pointed out earlier in this
TABLE 14
ALL CITRUS. TOTAL HARVESTING LABOR REQUIREMENTS
MANUAL AND MECHANICAL, 1975-76
Manually Harvested Mechanically Harvested
Workers Workers Total Number of
Period Production Needed Production Needed Workers Needed
Mil. boxes Workers Mil. boxes Workers Workers
September .05 680 .02 20 700
October .67 3,550 .40 440 3,990
November 2.51 12,070 .68 750 12,820
December 5.27 18,820 .68 750 19,570
January 8.92 25,110 .68 750 25,860
Early Peak 10.40 27,660 .68 750 28,410
February 8.89 25,060 .68 750 25,810
March 6.35 20,510 .66 750 21,260
April 6.07 19,760 .65 750 20,510
Late Peak 7.54 23,730 .64 750 24,480
May 6.59 21,160 .64 750 21,910
June 3.98 14,120 .64 750 14,870
July .65 5,140 .64 750 5,890
Production figures rounded to the nearest hundredth.
Workers figures rounded to the nearest ten.
TABLE 15
ALL CITRUS. TOTAL HARVESTING LABOR REQUIREMENTS
MANUAL AND MECHANICAL, 1980-81
Manually Harvested
Workers
Period Production Needed
Mil. boxes Workers
September .05 680
October .63 3,190
November 1.95 9,480
December 4.49 17,480
January 7.64 22,900
Early Peak 9.42 25,970
February 7.58 22,800
March 4.34 15,090
April 4.16 14,610
Late Peak 5.83 19,110
May 4.77 16,250
June 2.69 10,640
July .74 5,380
Mechanically
Production
Mil. boxes
.03
.62
1.88
3.88
3.88
3.88
3.88
3.78
3.57
3.53
3.53
2.61
.74
Harvested
Workers
Needed
Workers
20
440
1,330
2,750
2,750
2,750
2,750
2,750
2,750
2,750
2,750
2,040
570
Total Number of
Workers Needed
Workers
700
3,630
10,810
20,230
25,650
28,720
25,550
17,840
17,360
21,860
19,000
12,680
5,950
Production figures rounded to the nearest hundredth.
Workers figures rounded to the nearest ten.
TABLE 16
COMPARISON OF PRESENT AND FUTURE LABOR REQUIREMENTS
UNDER ASSUMED MECHANIZATION INTENSITY
Period
September
October
November
December
January
Early Peak
February
March
April
Late Peak
May
June
July
1969-70"
730
3,800
11,980
19,030
22,970
25,340
23,130
18,220
16,660
19,460
17,490
12,340
5,890
Labor Requirements
1975-76
- -Workers- -
700
3,990
12,790
19,570
25,860
28,410
25,810
21,260
20,510
24,480
21,910
14,870
5,890
Based on utilization
Florida.
estimated by Citrus Industrial
Council, Lakeland,
1980-81
700
3,630
10,810
20,230
25,650
28,720
25,550
17,840
L7,360
21,860
L9,000
12,680
5,950
-----------
chapter that the period from 1975 to 1980 would be a critical one for
mechanization. This will be reflected in the labor market as can be
observed by comparing the last two columns in Table 16. In spite of an
increase of almost 20 percent in the size of the crop between the 1975-76
and the 1980-81 seasons, the number of harvesting workers required shows
a significant drop in every period with the exception of the peak months
(January and February) where the differences are negligible and December
which shows a slight increase.
The figures in Table 16 show that the use of labor in citrus har-
vesting will remain highly seasonal in spite of mechanization because of
the seasonal characteristics of production. The need for variable
numbers of workers arises from the fact that building up a mechanical
capacity to take care of the peaks would imply having part of the
equipment and the skilled personnel that operates it idle during most of
the year. A feasible improvement in the pattern of labor usage shown in
Table 16, however, would be to reduce the difference between the number
of workers required during the peak months of January and February and
the other months of the season and to increase labor requirements in
May. It is not difficult to conceive of an arrangement by which the
operators of mechanical harvesters would work longer hours during the
peak months in exchange for time off during the slack period, since
this would insure them of stable employment.
Based on the projected production pattern, it is possible to increase
the number of boxes harvested mechanically by 50 percent during January
and February and by 25 percent in May without increasing the percentage
of mechanically harvested fruit beyond reasonable limits. Furthermore,
the peak weeks might be eliminated if the industry would be willing to
spread mechanical harvesting evenly over the peak months. This may
imply a small sacrifice in quality by picking some of the fruit before
and some after the time when it is regularly picked. With labor
scarcity the industry may have to make that sacrifice for the sake of
harvesting efficiency. Assuming 60 hour work weeks during January and
February, 50 hour work weeks in May, and the even spreading of harvesting
in order to eliminate the peaks, the labor requirements would be as
shown in Table 17. Under these assumptions it can be observed that the
larger 1980-81 crop will, in general, require a few workers less to har-
vest it than the 1969-70 crop. Comparing the 1980-81 requirements with
those of 1975-76, however, a significant reduction in numbers can be
observed. It seems then that the impact of mechanical harvesting on
the total labor requirements will not be felt until 1975 and later.
During the period 1975-1980, which has been assumed here to be the crit-
ical period for mechanization, there will be a reduction in the number
of workers of between 10 and 15 percent. These figures are by no means
indicative of a great impact such as occurred in other industries, the
reason being that the increases in labor productivity brought about by
mechanization will be partially offset by increases in total production.
Skill Requirements
The real impact of mechanization will be reflected on the skill
structure of the labor force. As was pointed out earlier the bulk of
the harvesting labor force used now is made up of relatively unskilled
labor. At the present time, approximately one out of every eight workers
can be classified as being skilled or semi-skilled. These are the crew
TABLE 17
COMPARISON OF PRESENT AND FUTURE LABOR REQUIREMENTS ASSUMING ELIMINATION
OF PEAK WEEKS, A 60 HOUR WORK WEEK DURING JANUARY AND FEBRUARY
AND A 50 HOUR WORK WEEK IN MAY
Labor Requirements
Period
1969-70 1975-76 1980-81
- - Workers - - -
September 730 700 800
October 3,800 3,990 3,630
November 11,980 12,790 10,810
December 19,030 19,570 20,230
January 22,970 25,270 22,310
February 23,130 25,220 22,210
March 18,220 21,260 17,840
April 16,660 20,510 17,360
May 17,490 21,480 16,870
June 12,340 14,870 12,680
July 5,890 5,890 5,950
leaders who must have some supervisory ability and the highlift truck
and tractor operators. By 1980 the required ratio of skilled and semi-
skilled workers to relatively less skilled will increase to one out of
every four. The total labor requirements by skill types for the three
seasons being considered in the study are shown in Table 18. The total
number of skilled and semi-skilled workers is the sum of those required
for manual harvesting as calculated from the equation (I-a) and those
required to operate the mechanical systems.
It can be observed from Table 18 that the effect of mechanization on
the labor requirements will be two-fold. First, there will be a reduction
in the total numbers of unskilled workers needed; the number of unskilled
workers displaced varies according to the month, but is as high as 3,250
workers in the month of February when comparing 1980-81 with 1969-70
and as high as 5,850 workers in the month of May when comparing the
1980-81 with the 1975-76 season. Secondly, there is a substantial
increase in the number of skilled and semi-skilled workers required; the
number of these workers more than doubled in several cases from 1969-70
to 1980-81.
These results imply that in order to avoid a situation where
unemployment exists in one sector of the citrus harvesting labor market
wt.:le there is scarcity of qualified personnel in the other sector of
that market, efforts must be made to upgrade the skills of the workers
in order to facilitate mobility from one sector to the other. What is
encouraging about these results is that if the necessary steps are taken,
no hardship is anticipated in terms of unemployment resulting from
mechanical harvesting. There seems to be reasonable time available
TABLE 18
COMPARISON OF PRESENT AND FUTURE LABOR REQUIREMENTS BY SKILL TYPES
UNDER THE ASSUMPTIONS OF TABLE 24
Number of Workers Required
Month Relatively Less Skilled Skilled or Semi-skilled
1969-70 1975-76 1980-81 1969-70 1975-76 1980-81
-----------------------------Workers--------------------------------
October 3,200 2,980 2,660 600 1,000 970
November 10,440 10,520 8,230 1,540 2,300 2,580
December 16,680 16,490 15,310 2,350 3,080 4,920
January 20,170 21,540 17,150 2,800 3,730 5,160
February 20,310 21,490 17,060 2,820 3,730 5,150
March 15,960 17,990 13,290 2,260 3,270 4,550
April 14,580 17,320 12,770 2,080 3,190 4,590
May 15,310 18,180 12,330 2,180 3,300 4,540
June 10,760 12,330 9,250 1,580 2,540 3,430
July 5,050 4,380 4,600 840 1,510 1,350
65
for the industry and Government to take the necessary steps to trans-
form the structure of the labor force in order to adapt it to the
future needs of the industry.
CHAPTER V
SOME GENERAL CONSIDERATIONS
As the name of this study suggests, its main purpose was to explore
the impact of the adoption of mechanical harvesting equipment on the
demand for labor in the citrus industry. Consequently, most of the
research time was used in studying the economic and physical features of
the different harvesting systems, the state of the art in general and the
possible mechanization patterns. Some assumptions were made about the
future behavior of the other variables that affect employment of har-
vesting labor and the final product was a set of estimates of the future
harvesting labor requirements both total and by skill categories. There
are certain considerations, however, that even though they are not
directly related to the demand side, will help to make better use of the
results of the study and may prove useful for policy purposes. These
considerations will be treated in this chapter under three separate
headings: some supply considerations, opportunities for off-season
employment, and topics for additional research.
Some Supply Considerations
One of the reasons why the adoption of mechanical harvesting methods
has not taken place is that so far the citrus industry has been able
to recruit the necessary harvesting labor among both Florida residents
and migrants from neighboring states. This labor has been available at
wages that, even though they have been increasing rapidly, are still
competitive with what per box costs would be for the mechanical systems.
Considering the anticipated increase in citrus production, however, this
situation is not likely to continue into the future. All agricultural
labor in Florida is becoming scarce. Workers leaving the agricultural
labor force are not being replaced at a comparable rate by people
entering the labor market.
Several factors contributing to this situation are listed in the
Florida DARE Report of 1969-/: the fact that we live in a more affluent
society with more alternatives for people; the broadening of the horizons
of the rural population caused by improved transportation and communi-
cation systems and better opportunities for education and training in
other fields; the widespread efforts by many groups to divert workers
from the migratory stream, and the publicity on undesirable aspects of
agricultural work. Another important factor which affects the supply of
hired agricultural labor is technological change and its consequent labor
displacement effect. Although this may seem paradoxical at first, when
the change has the effect of reducing the span of time in which the
worker can be effectively employed the result may be to cause him to
move out of the agricultural labor force entirely rather.than make him
available for employment in other sectors of agriculture.
The average number of workers on Florida farms has been declining
continuously. During the period 1960-64 an average of 120,000 workers,
including family and hired labor, were working on Florida farms. This
/ DARE Report 1969, University of Florida, Institute of Food and
Agricultural Sciences. October 1969. p. 152.
figure had declined to 116,000 by 1967 and in 1969 there were 112,000
farm workers.- This tendency is expected to continue:
Unless substantial changes occur in present trends
regarding non-agricultural employment opportunities,
total U. S. unemployment or wage levels, the supply
of domestic agricultural workers will probably decline
during the next decade. It is expected that the
number of Florida residents available for agricultural
labor will drop to about 90,000 by 1980. 3
Since the declining tendency of available agricultural labor is
characteristic not only to Florida but is rather widespread across the
nation, and the South in particular, increasing difficulty will be encoun-
tered in recruiting labor from the neighboring states.
The implication of these trends in agricultural labor supply is to
reiterate what was said at the end of Chapter IV: If the assumptions of
this study prove to be right, there will be a minimum of hardship on the
labor force, provided the necessary steps are taken towards the
training of part of the unskilled workers.
Opportunities for Off-season Employment
The results of Chapter IV indicate that even after the industry is
partly mechanized, the harvesting labor requirements are likely to remain
highly seasonal. Traditionally the industry has adjusted its recruiting
practices to the seasonality of production and only a very small minority
of harvest workers enjoy year-round employment within the citrus
Farm Labor Reports, United States Department of Agriculture,
Statistical Reporting Service, Crop Reporting Board, Washington, D. C.
Several issues.
/DARE Report 1969. p. 152.
industry. Since any worker regardless of skill can pick citrus, the
industry has been able to obtain labor on a "hire them when you need
them" basis.
With the advent of mechanization, harvesting concerns will be faced
with a different situation. They will find themselves competing for
skilled labor; the operator of a mechanical harvester must be able not
only to operate specialty equipment but he also needs to be able to do
some trouble shooting, minor repairing and maintenance on the machine.
Workers meeting these qualifications will not be easy to find for
temporary jobs. Not only will the skilled workers be demanding some
employment security but the producers and owners of the machinery will
find it more desirable to trust expensive equipment to a permanent
employee. In other words, the industry will find it both desirable and
necessary to provide year-round employment for the operators of mechanical
harvesting equipment.
The only possibilities for off-season employment within the industry
would be in production activities, or in new activities associated with
mechanization. The number of workers employed in production activities
during the 1968-69 season was shown in Figure 1, page 8, by the broken
line labeled "non-harvest". It varied from 7,029 workers during the
second week of March to 9,446 workers during the last week of August, a
difference of 2,417 workers. With the exception of the years following
a freeze, in which extra labor is required for replacement of trees and
other extraordinary activities, the figures for "non-harvest" workers
have not shown any significant variation over the last ten years in
spite of the increase in planting. Even if future increases in planting
were to increase the use of non-harvest labor, the increase should be
expected to be small due to the off-setting effect of improved technology
in production practices, and the difference between the trough and the
peak would remain in the same order of magnitude. Since the months of
low employment in non-harvest activities coincide with the peak employment
months in harvesting and vice versa, it is evident that some labor can be
and presently is transferred from one activity to the other.
The adoption of mechanical harvesting will create additional employ-
ment opportunities in the industry. Qualified personnel will be required
for the repair and maintenance of the equipment. Although part of these
activities will have to be carried on during the harvesting season, it
is reasonable to assume that major overhauls and preventive care will be
done during the off-season whenever possible. With over 2,000 different
machines, not counting trucks, operating by 1980, there will be oppor-
tunities for employment for about 400 to 500 workers in repair and
maintenance activities during the off-season. Additional employment
opportunities will exist during the off-season due to the necessity of
conditioning the trees for mechanical harvesting. Trees will have to be
hedged and skirted, and/or openings must be made in their canopies in
order to permit the operation of the machinery. So far, the only tree
conditioning that has been done was in connection with one of the experi-
mental systems; the operation involved 3 minimum wage workers, one
foreman and one hand saw. That operation could be almost completely
mechanized and on a commercial basis the labor requirements could be
reduced to the point where one man could take care of about 5 trees per
hour or more as compared to 2 in the above mentioned case. Based on a
total of around 30 million trees to be harvested mechanically by 1980,
and assuming that each tree has to be conditioned every five years, and
that most of the conditioning is done in the 20 weeks when harvesting
activities are lower, then, around 1,500 workers would be needed during
the off-season.
Between production, repair and maintenance of machinery and tree
conditioning activities it seems that there will be enough off-season
job opportunities for the industry to be able to offer year-round employ-
ment to its skilled operators without any financial strain.
Unskilled workers have traditionally been hired on a seasonal basis.
Many of them are brought from their homes in the neighboring southern
states during the peak months of the season and returned when work oppor-
tunities become scarce. Many others are part of the migratory current
that originates in Florida and follows the eastern seaboard as far north
as New Hampshire. Still others are able to find employment in other
activities within the state. From a welfare standpoint it would be
desirable to be able to assure all workers year-round employment. This
would not only offer them the opportunity to improve their standard of
living but would provide the industry with a more productive and reliable
labor force. It would be beyond the scope of this study to undertake the
task of exploring in detail the possibilities for off-season employment
for all workers. This would involve not only an interdisciplinary study
with many of the social sciences involved, but it would also imply a
study at the county level of all the activities and industries that could
offer employment during the summer months. That difficulties will be
encountered in providing year-round employment for all citrus workers,
however, can be observed in the employment indexes illustrated in Figures
6 and 7. The other important crops in the state, from the standpoint of
labor use, exhibit seasonal characteristics very similar to those of
citrus. Non-agricultural employment, illustrated in Figure 7 does not
seem to offer better alternatives with the-exception of contract con-
struction. Although it is true that the groups shown correspond to a
very broad classification and that certain sub-sectors within a group
may present different characteristics, it is also true that there is much
more labor mobility within a group than between groups.
Topics for Additional Research
This study was intended to be, and is in fact, of a partial nature.
First it is partial in the sense that it looks only at the demand side;
it is also partial since citrus employs only about one third of the total
hired agricultural labor force. As such it should be considered as a
starting point toward more comprehensive research in the agricultural
labor field.
One of the areas in which work is needed is in estimating the total
future labor requirements for Florida agriculture. Technological changes
are taking place in most other crops, particularly vegetables and sugar
cane which will cause changes in per unit labor requirements, while
changes in total requirements will take place due to changes in the
levels of activities. Research along these lines should answer, in a
state-wide all-crops fashion, the same questions that this study answers
for the citrus industry.
Research on the impact of technology on labor requirements is needed
not only on a state basis but on a regional basis as well. Because of
FIGURE 6
FLORIDA MONTHLY EMPLOYMENT INDEXES FOR SELECTED CROPS
Mixed
Citrus Vegetables
100
1 / 50
J F M A M J J A S O N D
Sugar Cane t
J F M A M J J A S T N
._.-- Tomatoes
J F M A M J J A S O N D J F A MJJASO N
Strawberries Other
J F M A M J J A S O N
SOURCE: 1968 Annual Farm Labor Report, Florida Industrial Commission,
Tallahassee, Florida.
J F M A M J J A S N D
FIGURE 7
FLORIDA. MONTHLY NONAGRICULTURAL EMPLOYMENT INDEXES
BY INDUSTRIAL GROUPS (5-yr. Average)
Manufacturing
Mining
105
100
95
J F M A M J J AS N D
105
100
Contract Construction
105
100
95
J F M A M J J AS ND
Transportation,.Communi-
cations, and
Public Utility
J F M A M J J A S O N D
J F M A M J J A S O N D
FIGURE 7 (Cont.)
Services
(Miscellaneous)
105
100
95
J F M A M J J A S O N D
J F M A M J J A S O N D
Government
J F M A M J J A S 0 N D
JFMAMJJASOND
SOURCE: Florida Employment Statistics, Florida Industrial Commission,
Tallahassee, Florida.
Trade
its nature, the availability of migrant workers is dependent not only on
the conditions in Florida, but also on any developments that take place
in the eastern seaboard states from Georgia to New Hampshire. Any tech-
nological labor saving improvement that takes place in those states will
have an impact on the availability of migrant workers to Florida agri-
culture and vice versa.
More sociological and psychological research concerning the charac-
teristics of workers and their motivation would improve the knowledge
necessary to estimating the changes that will take place in the size of
the available agricultural labor force. The future supply of agricultural
labor will in a way affect the demand for that labor, the reason being
that employers have to adjust their activities to the availability of
inputs. If a shortage of labor becomes persistent the tendency will be
toward structuring agriculture in favor of those activities that can be
more readily mechanized. Those activities where labor saving techniques
are difficult to adopt will tend to disappear. Such may be the case with
strawberries, and for that matter, with fresh citrus. Changes in the
structure of agriculture will in turn bring about changes in the demand
for labor. Knowledge of future total demand and local supply will also
assist in making decisions concerning policy toward migrant labor.
Also needed is research on the opportunities for off-season employ-
ment for the unskilled workers. Whether it would be possible for them
to find work given their present skills is a question that needs to be
answered. What would be the possibility of upgrading their skills, what
would be the optimum way of doing it and what it would cost also need
to be investigated. Finally, whether the workers would be willing to
77
return to the citrus harvesting activities after working during the
off-season in activities that require higher skills is something that
should be inquired. As more opportunities are opened outside agri-
culture, and more workers strive for better working and living conditions,
the problem of finding year-round employment for its workers may very
well turn out to be the most crucial one that the citrus industry will
have to face in the future.
APPENDIX
APPENDIX TABLE 1
CHARACTERISTICS OF THE DIFFERENT TYPES OF EQUIPMENT
Systems Initial Useful Salvage Fuel
and Units Cost Life Value Consumption
Equipment C t S f
I. Air Blast
1. Tractors
2. Blower
3. Catchers
4. Highlift
truck
II. Shaker
Catching-frame
1. Tractors
2. Catchers
3. Highlift
truck
III. Shaker-Windrower-
pick-up
1. Tractors
2. Shakers
3. Windrower
4. Pick-up
machine
5. Highlift
truck
dollars
67,000
10,000
50,000
1 7,000
1/2
37,189
10,760
22,929
3,500
94,000
30,000
25,000
2,000
1 30,000
years dollars
10 2,500
5
5
10
5
5
10
5
4
5
5,000
2,000
2,690
2,293
1,000
7,500
2,500
200
3,000
gallons/day
30
320
15
41.2
120
1 7,000 5
2,000 15
APPENDIX TABLE 1 (Extended)
Price of Harvesting Harvesting Relative
Fuel Rate Efficiency Quality of
P Picked Fruit
Percent
In Between
Highest
Lowest
$/gal.
trees/hr.
29.5
.135
.235
.235
15.0
.135
.235
43.0
.135
.235
.235
APPENDIX TABLE 2
LABOR AND NON-LABOR COSTS OF OPERATING DIFFERENT MECHANICAL SYSTEMS
Cost Item Depreciation Interest, Taxes Repair Lubricants
Housing and and
Insurance Maintenance
System D I R L
- - dollars per year - - -
Air Blast 10,750 2,875 10,050 1,676
Shaker and
Catching Frame 5,434 1,561 5,578 148
Shaker-Windrower
Pick-up 13,600 3,940 14,100 515
-Based on 135 8-hour days of operation per harvesting season.
2/Based on wages listed in Appendix Table 3.
- Based on wages listed in Appendix Table 3.
APPENDIX TABLE 2 (Extended)
Annual Non-Labor Costs Labor Costs
Fuel Total Per Hour- Total Per Hour/
Consumption 2/
F Wl2
-- - dollars $/hr. dollars $/hr.
(yr.) (yr.)
11,175 36,526 33.82 11,722 10.85
989 13,710 12.69 7,640 7.07
3,425 35,580 32.94 26,355 24.40
APPENDIX TABLE 3
TYPICAL WAGES USED IN COMPUTING LABOR COSTS*
Class
Non-skilled worker
Tractor driver
Blower operator
Highlift truck driver
Tractor driver-shaker operator
Pick-up machine operator
Windrower operator
dollars per hour
1.30
1.65
1.65
2.50
2.50
2.50
2.50
*At prevailing 1969 wage levels.
Wage
BIBLIOGRAPHY
1. American Society of Agricultural Engineers. Agricultural Engineers
Yearbook, Fourteenth Edition, 1967.
2. Camp, A. F., Homer E. Hooks and L. G. MacDowell. Citrus Industry
of Florida, Department of Agriculture, Bulletin No. 2, R-July 1961.
3. Cargill, B. F. and G. E. Rossmiller (ed.). Fruit and Vegetable
Harvesting Mechanization Technological Implications, RMC Report
No. 16, Rural Manpower Center, Michigan State University, East
Lansing, 1969.
4. Institute of Food and Agricultural Sciences. 1969 DARE Report,
University of Florida, Gainesville, October 1969.
5. 1965 DARE Report, University of Florida, Gainesville,
1965.
6. Davis, Velmar W. Economic Implications of Future Horticultural and
Engineering Technology in the Production of Fruits and Vegetables,
paper presented in the Colloquium on Agricultural Manpower Impli-
cations in a Changing Fruit and Vegetable Industry, Michigan State
University, July 21-23, 1969.
7. Florida Canners Association. Statistical Summary, 1967-68, Winter
Haven, Florida, 1969.
8. Florida Citrus Mutual. Annual Statistical Report, 1967-68 Season,
Compiled by Statistics and Economics Division.
9. Florida Crop and Livestock Reporting Service. Commercial Citrus
Inventory, Orlando, September 1968.
10. Florida Department of Agriculture. Citrus Summary 1968, Florida
Agricultural Stat.istics, Tallahassee, 1968.
11. Florida Industrial Commission. Annual Farm Labor Report 1968,
Tallahassee, Florida, June 1969.
12. Florida Employment Statistics, Monthly Report
Tallahassee, Florida.
13. _. Florida 1967 Farm Labor Report, Tallahassee, Florida.
14. Survey of Citrus Harvest in Florida, 1966-67,
Tallahassee, Florida.
BIBLIOGRAPHY (Continued)
15. Survey of Citrus Harvest in Florida, 1967-68,
Tallahassee, Florida.
16. Hall, Carl W. Crystal Balling The Engineering Prospects, paper
presented in the Colloquium on Agricultural Manpower Implications
in a Changing Fruit and Vegetable Industry, Michigan State Uni-
versity, July 21-23, 1969.
17. Maddox, James G. The Advancing South, Manpower Prospects and
Problems, New York: The Twentieth Century Fund, 1967.
18. Polopolus, Leo and W. B. Lester. Estimation of Florida's Orange
Production Over the Next Fifteen Years by the Random Sampling
Technique, Economic Research Department, Florida Citrus Commission
University of Florida, Gainesville, September 1968. (14 pages,
unpublished).
19. Polopolus, Leo. A Mathematical Determination of Weekly Harvest
Labor Requirements for Florida Citrus, Economic Research Depart-
ment, Florida Citrus Commission, University of Florida, August
1968, 36 pages. (Unpublished).
20. Spurlock, A. H. Costs of Picking and Hauling Florida Citrus Fruits,
1950-1968, Annual.
21. U. S. Department of Labor. Report of Employment 1963-68, Wash-
ington, D. C.
22. Whitney, Jody. Importance of Citrus Harvesting Mechanization,
Citrus Experiment Station, Lake Alfred, Florida. (Unpublished).
|