Content uploaded by Benedict C. Posadas
Author content
All content in this area was uploaded by Benedict C. Posadas on Aug 15, 2018
Content may be subject to copyright.
Economic Impacts of Mechanization or
Automation on Horticulture Production Firms
Sales, Employment, and Workers’ Earnings,
Safety, and Retention
Benedict Posadas
1
ADDITIONAL INDEX WORDS. wholesale nursery, greenhouse, southern states, multiple
linear regression
SUMMARY. Using a socioeconomic database collected by face-to-face interviews of
nurseries and greenhouses, empirical models were estimated to measure the
economic impacts of mechanization or automation on annual gross sales, annual
employment, and workers’ earnings, safety and retention. The survey was con-
ducted among 215 randomly selected wholesale nurseries and greenhouses located
in eight southern states from Dec. 2003 to Nov. 2009. The level of mechanization
or automation (LOAM) observed among the participating wholesale nurseries
and greenhouses averaged 20% of the major tasks performed by workers. Nurseries
and greenhouses that reported greater annual gross sales demonstrated higher levels
of mechanization, implying economies of scale associated with technology adoption
by these wholesale horticulture production firms. The increase in total workers’
earnings associated with improved mechanization indicated that nurseries and
greenhouses were able to pay their workers higher wages and salaries. The increased
levels of mechanization produced neutral effects on employment and raised the value
of the marginal productivity (VMP) of labor, implying that technology adoption by
wholesale nurseries and greenhouses did not displace any worker but instead
improved total workers’ earnings. Growers that reported higher levels of mechani-
zation hired fewer new workers with basic horticultural skills, especially among
horticultural firms which operated both nursery and greenhouse enterprises. The
length of training period for basic horticultural skills was not influenced by the level
of mechanization, but was significantly extended when nurseries or greenhouses
hired more new workers without basic horticultural skills. The number of work-
related injuries increased as a result of improvements in mechanization, which
primarily consisted of back strains, cut fingers, shoulder and ankle strains, and
eye injury. The workers’ retention impact (WRI) of the level of mechanization turned
out to be neutral or indeterminate since almost all of their workers were with them
during the past 2 years before conducting the interviews. Overall, advances in
mechanization or automation generated enhancing effects on the annual gross sales of
horticultural production firms, enabled them to retain and pay better wages for their
workers, hired fewer new skilled workers, and reported more work-related injuries.
With the tightening in the
regulations regarding mi-
grant workers, the nursery
and greenhouse industry is facing a
critical shortage of labor (Posadas
et al., 2004). Migrant labor issues are
a major concern facing agriculture
and especially the horticultural indus-
try in the United States (Bellenger
et al., 2008).
Workers in this industry perform
varied functions and are subjected to
different working conditions (O*Net
Online, 2012). Posadas et al. (2008)
reported that at least 8 of the 15 major
tasks were performed by workers with
significant number of nurseries using
mechanized or automated systems in
media preparation, filling containers
with substrates, moving containers
from potting to transport, transport-
ing containers to field, plant pruning,
and fertilizer, pesticide, and irrigation
application. Six of the 10 major tasks
were performed by workers employed
by a significant number of greenhouse
operations with mechanized or auto-
mated systems in media preparation;
filling containers with substrates; en-
vironmental control; and fertilizer,
pesticide, and irrigation application.
Very few nurseries or greenhouses
were using mechanized or automated
systems in cutting and seed collection
and preparation; placing plant liners;
sticking cuttings and planting seed;
harvesting and grading production;
spacing of plants and containers; re-
moval, picking up, loading, and placing
of plants; and jamming of plants for
winter protection.
The nursery and greenhouse in-
dustry is often described as one of the
fastest-growing sectors of U.S. agricul-
ture and is inherently labor intensive
(Regelbrugge, 2007) with greater than
40% of production costs consisting of
labor costs (Mathers et al., 2010).
Hodges et al. (2011) estimated the
total economic impact of the U.S.
green industry at $175.26 billion
representing 0.76% of the national
gross domestic product in 2007. The
U.S. green industry generated a TEI
of 1.95 million jobs, labor earnings
impact of $53.16 billion, and value-
added impact of $107.16 billion.
To sustain robust growth in the
industry, continuous improvements
in the skills of the workforce and their
year-round availability are necessary.
Many jobs in the industry require
large amounts of stooping, lifting of
heavy containers, and exposure to
chemicals, dust, and plant materials
(Bureau of Labor Statistics, 2012b).
These tend to be relatively low-paying
jobs, with median wages in 2010
amounting to $8.98 per hour or
Production
and Marketing
Reports
388 •June 2012 22(3)
$18,690 per year (O*Net Online,
2012), making it difficult for man-
agers to compete for and retain
workers in currently tight domestic
labor markets. Many commercial op-
erations have employed immigrant
labor, which is mostly less skilled, to
meet their rising labor requirements.
The nursery migrant workforce are
employed, on average 6 months, and
most stayed for 10 months (Mathers
et al., 2010). In the long run, there is
a need to increase the skill level of
these migrant workers to improve
wage rates, recruitment, and reten-
tion of workers.
Mechanization of an operation
can provide mechanical power, speed,
repetition, safety, and a greater po-
tential for consistency and quality
control. Mechanization is normally
defined as the replacement of a human
task with a machine (Giacomelli,
2002). Automation includes these
attributes, but with greater flexibility
and, potentially, some automated de-
cision making (Giacomelli, 2002). But
true automation encompasses more
than mechanization. Automation in-
volves the entire process, including
bringing material to and from the
mechanized equipment. It normally
involves integrating several opera-
tions and ensuring that the different
pieces of equipment communicate
with one another to ensure smooth
operation. Many times, true automa-
tion requires reevaluating and chang-
ing current processes rather than
simply mechanizing them (Porter,
2002). The possible benefits associated
with automation were summarized
by Ling (1994) as follows: reduce
manual labor requirement, improve
production quality, eliminate haz-
ardous working conditions, reduce
production costs, increase market
value, and improve professional es-
teem. Simonton (1992) concluded
that the benefits and incentives to
automate are significant and in-
clude improving the safety of the
work force and the environment,
along with ensuring sufficient pro-
ductivity to compete in today’s global
market.
Given the above-mentioned ex-
pected benefits and the tightening
labor markets faced by the nursery and
greenhouse industry, this article evalu-
ated the economic impacts associated
with mechanization and automation by
using socioeconomic databases col-
lected in previous surveys. The specific
objective of this article was to measure
the economic impacts of mechaniza-
tion or automation on the horticulture
firms’ total revenues (TR), annual
employment, and workers’ earnings,
skills, training, safety, and retention
rates.
This article is a spatially, tempo-
rally, and analytically expanded ver-
sion of an earlier article which covered
87 growers located in the three north-
ern Gulf of Mexico states (Posadas
et al., 2008). With the data collected
by face-to-face interviews of nurseries
and greenhouses, multiple linear re-
gression analysis was applied to esti-
mate empirical models to measure the
socioeconomic impact of automation
or mechanization on annual gross sales;
annual employment; and workers’
earnings, safety, and retention. In this
article, there are more southern states
included in the study (eight vs. three
states), covered a longer period (2003
to 2007 vs. 2003 to 2009), and in-
cluded more producers (87 vs. 215
growers). Because of the longer pe-
riod covered in the study, this article
used deflated values of annual gross
sales and total workers’ earnings and
hourly wage rates and also added the
interview date as an explanatory vari-
able. Additional variant models were
estimated where the number of full-
time equivalent (FTE) workers was
segregated into permanent workers
(PW) and part-time workers (PTW).
The segregation allowed for the com-
parisonoftheVMPofthePWand
PTW.
Empirical models
To evaluate the economic impacts
of mechanization or automation, em-
pirical models were estimated for the
horticulture production firms’ TR, an-
nual employment, workers’ earnings,
safety, skill levels, and retention rates.
The general hypothesis of the empirical
models is that if the estimated coeffi-
cient, slope, or first derivative of the
estimated empirical equation with re-
spect to the average level of mechani-
zation is not statistically different from
zero, then mechanization has a neutral
effect on the designated economic
variable. The economic impacts of
the other variables included in each of
the models were also measured by the
same procedure.
TheaverageLOAM(AVELOAM)
of all the identified major tasks per-
formed by workers in each nursery or
greenhouse was used in the empirical
models instead of the specific LOAM
of each individual task identified in
the survey. The use of the specific
LOAM in each individual task per-
formed by workers in each nursery or
greenhouse resulted to errors in es-
timation because there was insuffi-
cient number of observations. The
formula used in estimating the aver-
age level of mechanization of all the
identified major tasks performed by
workers in each nursery or green-
house is as follows:
AVELOAM = P19
n=1 LOAM
N;
where LOAM = level of mechanization
or automation in each specific task
performed by workers in each nursery
or greenhouse and N=numberof
tasks performed by workers in each
nursery or greenhouse. The specific
tasks performed by workers in nurser-
ies and greenhouses that were included
in the survey are listed in Table 1.
The TR of nurseries and green-
houses were derived from the mid-
point of the annual gross sales category
reported by each participating nursery
or greenhouse operation. The TR
were deflated by the consumer price
index (CPI, 2009 = 100) to convert
Units
To convert U.S. to SI,
multiply by U.S. unit SI unit
To convert SI to U.S.,
multiply by
0.01 % gg
–1
100
10 % gkg
–1
0.1
10 % gL
–1
0.1
10 % mgg
–1
0.1
10 % mLL
–1
0.1
1 % mL/100 mL 1
0.4047 acre(s) ha 2.4711
Coastal Research and Extension Center, Mississippi
State University, 1815 Popps Ferry Road, Biloxi, MS
39532
1
Correspondingauthor. Email: benp@ext.msstate.edu.
•June 2012 22(3) 389
values to a recent base year. The CPI
measures the changes in the prices
paid for a representative basket of
goods and services (Bureau of Labor
Statistics, 2012a). The marginal reve-
nue impact (MRI) of mechanization
or automation was expected to be
positive, indicating that horticulture
production firms that experienced
higher levels of production or sales
wouldalsodemonstrateadvanced
LOAMs. To test the MRI hypothesis,
the TR empirical model was estimated
using the following formulation:
TR = B0+B
1AVELOAM + B2FTE
+B
3ACRE + B4YEARS
+B
5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE + E;
where B
0
=constantterm,B
i
=re-
gression coefficients, FTE = full-time
equivalent workers (number), ACRE =
area in production (acres), YEARS =
period since establishment (years),
NURSERY = nursery-only operations,
GREENHOUSE = greenhouse-only
operations, PERCENT = area used in
production (percent), DATE = date of
interview, and E = error term. This
empiricalTRmodelissimilarto
Posadas et al. (2008) except for the
use of deflated values, the addition of
the interview date, and the estimation
of an additional variant model where
the number of FTE workers was seg-
regated into PW and PTW. The seg-
regation of FTE workers allowed for
the comparison of the VMP of the PW
and PTW. The regression results with
the separation of FTE workers into
PW and PTW in all the empirical
models were not presented in table
form but were cited in the text as part
of the overall discussion.
The total earnings of workers
(TWE) were derived from the total
annual man-hours employed (TMH)
multiplied by the reported hourly
wage rate. The TWE was deflated by
the CPI with the year 2009 as base
year. The marginal workers’ earnings
impact (MWEI) was expected to be
positive, indicating that the VMP of
labor was enhanced as a result of
mechanization or automation. The
MWEI hypothesis was tested using
the following empirical model:
TWE = B0+B
1AVELOAM + B2FTE
+B
3ACRE + B4YEARS
+B
5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE + E:
This empirical TWE model is sim-
ilar to Posadas et al. (2008) except with
the use of deflated values of workers’
earnings, addition of the interview date
as explanatory variable, and estimation
of a variant model where the number of
FTE workers was broken down into
PW and PTW. The separation of the
number of FTE workers allowed for
the comparison of the marginal earn-
ings of the PW and PTW.
Annual employment was mea-
sured in terms of the number of FTE
workers, which was equal to the sum
of the number of PW and one-half the
number of PTW. The total man-
hours employed were computed from
the number of FTE workers multi-
plied by the number of working hours
each month. The TEI was expected to
be negative, indicating a labor-saving
characteristic of automation or mech-
anization. The TEI hypothesis was
evaluated using the following empir-
ical model:
FTE;PW;PTW or TMH
=B
0+B
1AVELOAM
+B
2DEFWAGER + B3ACRE
+B
4YEARS + B5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE + E;
where DEFWAGER = deflated hourly
wage rate (dollars). The empirical FTE
workers, PW, PTW, or TMH model is
comparable to Posadas et al. (2008)
with the addition of the interview date
as explanatory variable, a model varia-
tion where the dependent variable FTE
workers was broken down into PW and
PTW, and use of deflated wage rate.
The division of FTE workers allowed
Table 1. Proportion of the major tasks performed with some form of automation
or mechanization by workers employed in nurseries and greenhouses by type of
operation. The respondents were asked to describe, in percentage of terms, the
level of automation or mechanization in each of the major tasks performed in
their respective nursery or greenhouse operations.
Workers’ tasks
Proportion of
tasks in
nurseries (%)
Proportion of
tasks in
greenhouses (%)
Media preparation 28.44 25.82
Filling containers with substrate 28.54 34.28
Cutting and seed collection N/A
z
0.28
Cutting and seed preparation N/A 2.66
Placing plant liners, sticking cuttings and
planting seed
8.76 12.66
Environmental control N/A 47.16
Moving containers from potting to transport
vehicle for movement within the nursery
14.92 N/A
Transporting containers to field in nurseries 31.91 N/A
Removing containers from transport vehicle
and placing in the field
2.78 N/A
Spacing of plants and containers 4.02 N/A
Harvesting and grading production N/A 0.00
Picking plants up and loading onto transport
vehicle at time of sale
10.61 N/A
Removal of plants from transport vehicle and
placing in holding area awaiting shipment
7.75 N/A
Picking up plants from holding area/transport
trailers and loading onto delivery vehicles
11.23 N/A
Jamming plants for winter protection 0.00 N/A
Plant pruning 13.10 N/A
Fertilizer application 16.12 39.45
Pesticide application 24.70 30.63
Irrigation application and management 51.81 56.82
All workers’ tasks included in the survey 17.51 25.48
The socioeconomic survey was conducted among randomly selected wholesale nurseries and greenhouses located
in eight, selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
Not applicable because this question was not asked to this group.
390 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
for the comparison of the impacts of
mechanization on the employment of
PW and PTW.
Workers’ skills (WS) were mea-
sured in terms of the percentage of
new workers hired having basic hor-
ticultural skills. The WS impact (WSI)
was expected to be negative, indicat-
ing reduced requirements for manual
workers arising from automation or
mechanization. The WSI hypothesis
was evaluated using the following
empirical model:
WS = B0+B
1AVELOAM + B2FTE
+B
3ACRE + B4YEARS
+B
5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE
+B
9RETURN + B10MEDIUM
+B
11LARGE + B12 SUPER + E;
where RETURN = workers who were
employed in the same nursery the pre-
vious year (percent), MEDIUM = op-
erations with annual sales between
$250,000 and $499,999, LARGE =
operations with annual sales be-
tween $500,000 and $999,999, and
SUPER = operations with annual sales
of $1,000,000 and above. The size of
the nursery and greenhouse opera-
tions was measured by the reported
annual gross sales. The dummy vari-
ables representing the various sizes of
the nursery and greenhouse operations
were based on the annual gross sales
reported by the wholesale growers.
These annual gross sales categories
were based on the suggestions made
by Hoppe et al. (2007) which included
the following: less than $250,000,
$250,000 to $499,999, $500,000 to
$999,999, and more than $1,000,000.
The empirical WS model is analogous
to Posadas et al. (2008) with the
addition of the interview date, the
three farm sizes as explanatory vari-
ables, and a variation where the in-
dependent variable FTE workers was
broken down into PW and PTW.
Training time (TT) was deter-
mined by the length of the basic train-
ing period for the newly hired workers.
The workers’ TT impact (TTI) was
indeterminate depending on the need
for increased training in the han-
dling of specialized equipment and
the lower requirement for manual
workers as a result of automation or
mechanization. The TTI hypothesis
was tested using the following empir-
ical model:
TT = B0+B
1AVELOAM + B2FTE
+B
3ACRE + B4YEARS
+B
5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE
+B
9WS + B110RETURN
+B
11MEDIUM + B12 LARGE
+B
13SUPER + E:
The empirical TT model is equiv-
alent to Posadas et al. (2008) with the
addition of the interview date, the
three farm sizes as explanatory vari-
ables, and a variation where the in-
dependent variable FTE workers was
separated into PW and PTW.
Workers’ safety was measured in
terms of the number of man-hours lost
(MHL) due to work-related injuries
and number of work-related injuries
reported (WRIR) the year before the
interviews were conducted. The work-
ers’ safety impact (WYI) was expected
to be positive because automation or
mechanization would eliminate haz-
ardous working conditions. The WYI
hypothesis was tested using the follow-
ing empirical models:
MHL or WRI
=B
0+B
1AVELOAM
+B
2FTE + B3ACRE + B4YEARS
+B
5NURSERY
+B
6GREENHOUSE
+B
7PERCENT + B8DATE
+B
9WS + B10TRAIN1
+B
11TRAIN2
+B
12MEDIUM + B13 LARGE
+B
14SUPER + E;
where TRAIN1 = workers trained
on chemical and pesticide application
(percent) and TRAIN2 = workers
trainedonbasichorticulturalskills(per-
cent). The empirical MHL and WRIR
models are comparable to Posadas et al.
(2008) with the addition of the in-
terview date and the three farm sizes
as explanatory variables, and a model
variation where the independent vari-
able FTE workers was segmented into
PW and PTW.
Workers’ retention rates were
expressed as a percentage of the work-
ers who were employed in the same
nursery or greenhouse for the past 2
years before the interviews. Gabbard
and Perloff (1997) reported that
farmworkers are more likely to
return to employers who offer ben-
efits, pay by the hour, provide good
working conditions, and hire directly.
The WRI was expected to be positive
because automation or mechaniza-
tion would improve professional es-
teem and work satisfaction as a result
of better and safer working condi-
tions. The WRI hypothesis was
tested using the following empirical
model:
RETURN = B0+B
1AVELOAM
+B
2FTE + B3ACRE
+B
4YEARS
+B
5NURSERY
+B6GREENHOUSE
+B
7PERCENT + B8DATE
+B
9REST + B10HOUSING
+B
11INSURANCE
+B
12RETIREMENT
+B
13MEDIUM
+B
14LARGE
+B
15SUPER + E;
where REST = workers with access
to rest and lounging areas (percent),
HOUSING = workers provided
with housing benefits (percent),
INSURANCE = workers provided
with medical and dental insurance
(percent), and RETIREMENT =
workers provided with retirement
benefits (percent). The empirical
WRI model is analogous to Posadas
et al. (2008) with the addition of the
interview date and the three farm
sizes as explanatory variables, and a
model variation where the indepen-
dent variable FTE workers was di-
vided into PW and PTW.
The empirical models were es-
timated by using the multiple linear
regression method. All the regres-
sion analyses were performed by
using EViews 6 (Quantitative Micro
Software, Irvine, CA). The descrip-
tive statistics about mechanization,
socioeconomic characteristics, and
percentage distribution of nurseries
and greenhouses by annual gross
salesandtypesofoperationswere
computed by using SPSS (version
19.0 for Windows; IBM Corporation,
Armonk, NY).
•June 2012 22(3) 391
Data collection and analysis
The face-to-face socioeconomic
survey of wholesale nurseries and green-
houses in eight southern states—
Mississippi, Alabama, Louisiana, Flor-
ida, Tennessee, South Carolina, North
Carolina, and Georgia—was conduct-
ed between Dec. 2003 and Nov. 2009
(Fig. 1). This length of time was re-
quired due to the distance traveled to
complete the surveys, and the availabil-
ity of the growers to meet one-on-one
with the survey administrator. Official
lists of certified nurseries were retrieved
from Mississippi Department of Agri-
culture and Commerce (2003), Ala-
bama Department of Agriculture and
Industries (2004), Louisiana Depart-
ment of Agriculture and Forestry
(2003), South Carolina Department
of Agriculture (2006), Florida Depart-
ment of Agriculture and Consumer
Services (2005), North Carolina De-
partment of Agriculture and Consumer
Services (2008), Georgia Department
of Agriculture (2007), and Tennessee
Nursery and Landscape Association
(2006).
Only wholesale growers operat-
ing throughout the seven states, and
northern Florida, were included in
the selection of survey participants.
In northern Florida, nurseries were
randomly selected from the listing
using only the nurseries in counties
from Alachua County and north. The
wholesale growers in each state in-
cluded in the survey were identified
and numbered from 1 to N.Using
Excel (Office 2003; Microsoft Corpo-
ration, Redmond, WA), 50 random
integers were individually generated
from 1 to N,whereN= the number
of wholesale growers in each state.
Individual letters of introduction
were sent to the 50 selected nurseries
and greenhouses in each state in
advance. Follow-up telephone calls
were made to each of the nurseries
and greenhouses selected to deter-
mine their willingness to participate
and their availability for the inter-
views. All personal interviews were
conducted by the research associate
hired for this purpose by the Missis-
sippi State University Coastal Re-
search and Extension Center. The
respondents to the survey were the
owners or operators of the selected
nurseries and greenhouses. These se-
lected growers were contacted via
mail and were asked to return a pre-
paid postcard indicating their willing-
ness to participate in the survey.
Those nurseries indicating a willing-
ness to participate were then contacted
by phone, and interviews scheduled.
A total of 215 personal interviews were
completed with wholesale nurseries
(N= 88), greenhouses (N= 52) and
mixed nursery/greenhouse operations
(N= 75) in Mississippi (32), Louisiana
(29), Alabama (26), Florida (27),
Tennessee (17), South Carolina (30),
North Carolina (30), and Georgia (24).
The socioeconomic panel data
consisted of variables dealing with
labor, technical, and economic infor-
mation about the nurseries and green-
houses in the eight southern states.
The workers’ demographic character-
istics included among others race, age,
gender, and formal education com-
pleted. The operational characteris-
tics included but not limited to labor
use, growing area, number of green-
houses, nursery operations, and an-
nual gross sales. Previous reports
using the above-mentioned databases
covered the socioeconomic charac-
teristics of workers and working con-
ditions (Posadas et al., 2005b, 2010b),
operational characteristics (Posadas
et al., 2010a), socioeconomic de-
terminants of technology adoption
(Posadas et al., 2005a), and current
mechanization systems (Coker et al.,
2010). Additional reports will be
forthcoming covering all the partici-
pating nurseries and greenhouses in
the eight southern states included in
the survey.
Fig. 1. Map showing all of the randomly selected wholesale nurseries and greenhouses in eight selected southern U.S. states that
participated in the socioeconomic survey from Dec. 2003 to Nov. 2009 by type of operation.
392 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
The empirical models described
above were estimated using the cur-
rent socioeconomic datasets presented
by type of operations in Tables 2–4.
The types of horticultural operations
included nursery-only, greenhouse-
only, and mixed operations. Mixed
operations are horticultural farms,
which operate both nurseries and
greenhouses. Dummy variables repre-
senting nursery-only and greenhouse-
only operations were included in the
models to differentiate them from
mixed-type operations.
Mechanization of workers’
tasks
The nursery mechanization or
automation index (NMAI) could be
defined as a measure of the level of
automation or mechanization cur-
rently being practiced in each nursery
or greenhouse included in the re-
gional survey. The NMAI shows the
extent to which nurseries have cur-
rently automated or mechanized the
various tasks involved in the produc-
tion of horticulture products (Posadas
et al., 2008). The AVELOAM observed
among the participating nurseries and
greenhouses was 20.3% with signifi-
cant differences observed among
nursery-only (17.6%), greenhouse-
only (24.9%), and mixed operations
(20.3%, Table 2). The average NMAI
reported by Posadas et al. (2008) was
20% for all operations, 13% for nurs-
ery-only, 28% for greenhouse-only,
and 19% for mixed operations. There
were 15 major tasks included for
workers in nursery operations and ten
major tasks for workers in greenhouse
operations (Table 1). The current
mechanization systems observed
among participating wholesale nurser-
ies and mixed operations were de-
scribed by Coker et al. (2010).
On average, 17.5% of the major
tasks in nursery operations were per-
formed by workers with some form of
mechanization or automation (Table
1). The top five major tasks performed
by nursery workers with significant
levels of mechanization included irri-
gation application and management
(51.8%), transporting containers to
field in nursery (31.9%), filling con-
tainers with substrate (28.5%), media
preparation (28.4%) and pesticide ap-
plication (24.7%). The second tier of
five major tasks carried out by nursery
workers with some level of mecha-
nization were fertilizer application
(16.1%), moving containers from
potting to transport vehicle for move-
ment within the nursery (14.9%), plant
pruning (13.1%), picking up plants
from holding area or transport trailers
and loading onto delivery vehicles
(11.2%), and picking plants up and
loading onto transport vehicle at time
of sale (10.6%). The lowest levels of
mechanization were observed among
the third cluster of five major tasks
performed by nursery workers which
included placing plant liners/sticking
cuttings/planting seed (8.8%), re-
moval of plants from transport ve-
hicle and placing in holding area
awaiting shipment (7.8%), spacing
of plants and containers (4.0%), re-
moving containers from transport
vehicle and placing in field (2.8%),
and jamming plants for winter pro-
tection (0%).
Workers in greenhouse opera-
tions accomplished 25.4% of their
tasks with some form of mechaniza-
tion or automation (Table 1). The top
five major duties done by greenhouse
workers with considerable levels of
mechanization included irrigation
application and management (56.8%),
environmental control (47.2%), fertil-
izer application (39.4%), filling con-
tainers with substrate (34.3%) and
pesticide application (30.6%).The low-
est five major responsibilities under-
taken by greenhouse workers with
limited levels or no mechanization
were media preparation (25.8%), plac-
ing plant liners/sticking cuttings/
planting seed (12.7%), cutting and
seed preparation (2.7%), cutting and
seed collection (0.3%), and harvesting
and grading production (0%).
Marginal revenue impact
There were wide variations in the
annual gross sales of participating
nurseries and greenhouses. Majority
of the wholesale horticulture produc-
tion firms (55.5%) reported annual
gross sales below $250,000. Less
than one-fifth (19.4%) had annual
gross sales between $250,000 and
$499,999. About 10.4% of the par-
ticipating nurseries and greenhouses
Table 2. Total number of operations, average level of mechanization, and percentage of distribution by annual gross sales of
nurseries and greenhouses by type of operation.
Characteristic
All nursery and
greenhouse operations
Nursery-only
operations
Greenhouse-only
operations
Mixed nursery and
greenhouse operations
Total number of operations (number) 215 87 53 75
Average level of mechanization (%)
z,y
20.34 17.58 a 24.94 b 20.29 ab
Total deflated gross annual sales ($)
y,x,w
605,334.33 518,166.26 a 293,633.37 a 931,253.13 b
Nurseries and greenhouses
with annual sales below $250,000 (%)
55.5 25.6 16.1 13.7
Operations with annual sales
between $250,000 and $499,999 (%)
19.4 10.0 4.3 5.2
Operations with annual sales
between $500,000 and $999,999 (%)
10.4 0.5 4.3 5.7
Operations with annual
sales $1 million and above (%)
14.7 5.2 0.0 9.5
These economic averages and percentages were computed from the results of the socioeconomic survey, that was conducted among randomly selected wholesale nurseries and
greenhouses located in eight, selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
Significantly different by type of operation at P£0.05 using analysis of variance (ANOVA).
y
Values in the same row with different letters are significantly different by type of operation at £0.05 using Duncan’s multiple range test.
x
Significantly different by type of operation at P£0.01 using ANOVA.
w
Annual gross sales were deflated by the consumer price index at 2009 prices.
•June 2012 22(3) 393
generated annual gross sales between
$500,000 and $999,999. About 14.7%
of these horticulture firms achieved
annual gross sales of $1,000,000 and
above, as shown in Table 2. The annual
gross sales of the participating whole-
sale growers averaged $563,981 per
operation. Significantly different annual
gross sales were reported by type of
operation with the mixed operations
averaging higher gross annual sales
than the nursery-only and greenhouse-
only operations.
The estimated TR model ex-
plained 89% of the variations of
the deflated TR of the participating
nurseries and greenhouses, as shown in
Table 5. The two explanatory variables
which exerted significant impacts on
deflated TR were the average level of
mechanization ($3824.63) and num-
ber of FTE workers ($88,904.79).
However, Posadas et al. (2008) re-
ported three explanatory variables
that had significant effects on annual
gross sales: the average level of mech-
anization ($4899.62), number of FTE
workers ($69,251.62), and acres in
production ($958.95). The average
area in production in the eight south-
ern states, which was 14.94 acres, was
not considerably different from the
average 13.0 acres reported in the
three northern Gulf of Mexico states
(Posadas et al., 2010a). Attempts to
estimate the TR model in quadratic
form did not improve the predictive
and explanatory properties of the re-
gression results.
The MRI of the average level
of mechanization, as expected, was
positive, indicating that mechaniza-
tion or automation had consider-
able enhancing effects on annual gross
sales. However, the positive MRI did
not specify the net effects on net
revenues above total costs of pro-
duction. The TR model results fur-
ther suggested that an additional
FTE worker was associated with an
increase in annual gross sales by
$88,904.79. When the number of
FTE workers was separated into its
two components, the VMPs of the
PW and PTW were estimated. The
VMP of an additional PW employed
in nurseries and greenhouse was about
the same ($87,910.77) as an addi-
tional FTE worker. When nurseries
and greenhouses employed PTW in
their operations, the VMP of an extra
PTW averaged $47,711.78.
Marginal workers’ earnings
impact
The TWE averaged $175,272.57
per operation with significant varia-
tions by type of operations, as shown
in Table 3. Workers in wholesale op-
erations with annual sales $1,000,000
and above received significantly higher
total annual earnings. The total
workers’ earnings comprised on aver-
age, 29.1% of the annual gross sales
reported by nurseries and greenhouses.
There were no significant variations in
the ratios of the total annual workers’
earnings to total annual gross sales by
type of operation.
About 94% of the differences in
the deflated total workers’ earnings
were explained by the variations in the
independent variables included in the
total workers’ earnings model. Three
independent variables explained signif-
icantly the differences in total workers’
earnings, as Table 5 shows. The in-
dependent variables were the average
level of mechanization ($1830.64),
number of FTE workers ($21,980.17),
and area in production ($742.08). The
same three variables were reported by
Posadas et al. (2008) which exerted
Table 3. Selected economic and technical characteristics of nurseries and greenhouses by type of operation.
Characteristic
All nursery and
greenhouse operations
Nursery-only
operations
Greenhouse-only
operations
Mixed nursery and
greenhouse operations
Full-time equivalent (FTE)
workers (number)
z,y
6.52 5.15 a 3.83 a 10.01 b
Average FTE workers
(workers/acre)
x
1.21 1.32 1.45 0.91
Total annual man-hours
employed (hours)
z,y
16,830.25 13,571.20 a 9,589.90 a 25,415.48 b
Average annual man-hours
employed (hours/acre)
y,x,w
5,965.23 5,291.08 a 8,700.17 b 4,895.38 a
Wage rate ($/hour)
y,w,v
8.12 8.51 b 8.04 ab 7.74 a
Total deflated annual
workers earnings ($)
y,w,v
175,272.57 150,640.48 ab 102,249.86 a 243,440.91 b
Ratio of total deflated annual workers
earnings to deflated
annual gross sales (%)
v
29.06 28.62 32.96 27.21
Average deflated total annual
workers earnings ($/worker)
v
19,736.57 20,162.67 19,361.88 19,736.57
Area in production (acres)
y,x,w
14.94 18.93 b 1.97 a 18.72 b
Area used in production (%)
z,y
60.68 67.36 b 45.30 a 62.87 b
Average area in production
(acres/operation.)
y,x,w
2.82 3.55 b 1.55 a 2.84 b
Period since establishment (years)
y,w
24.42 23.84 ab 20.33 a 28.04 b
These economic and technical averages and percentages were computed from the results of the socioeconomic survey, that was conducted among randomly selected wholesal e
nurseries and greenhouses located in eight, selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
Significantly different by type of operation at P£0.01 using analysis of variance (ANOVA).
y
Values in the same row with different letters are significantly different by type of operation at £0.05 using Duncan’s multiple range test.
x
1 worker/acre = 2.4711 workers/ha, 1 h/acre = 2.4711 hha
–1
, 1 acre = 0.4047 ha.
w
Significantly different by type of operation at P£0.05 using ANOVA.
v
Economic values were deflated by the consumer price index at 2009 prices.
394 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
significant influences on the total
workers’ earnings, namely average level
of mechanization ($1608.88), number
of FTE workers ($18,650.94), and area
in production ($811.95).
The MWEI on the level of mech-
anization, as expected, was positive,
indicating that mechanization or au-
tomation had enhancing effects on
workers’ earnings. The marginal earn-
ings of an extra FTE worker were
$21,980.17 as compared with the
average earnings of an FTE worker
which were $19,736.57. There were
no significant differences in the average
workers’ earnings per FTE worker by
type of operation (Table 3). In addi-
tion, the cultivation of an additional
acre to horticulture production raised
workers’ earnings by $742.08.
When the number of FTE workers
was separated into its two components,
the regression results showed that, as
expected, the marginal earning of an
additional PW was about the same
($20,791.65) as an additional FTE
worker. The marginal earning of an
extra PTW was $15,138.89. The
VMP of additional PW and PTW
were $87,910.77 and $47,711.78,
respectively. However, these regres-
sion results are not sufficient to
draw any conclusions regarding hir-
ing decisions involving PW and
PTW.
Marginal employment impact
The number of FTE workers
employed by participating nurseries
and greenhouses averaged 6.52 work-
ers per operation with significant
variations by type of operation, 5.15
workers for nursery-only, 3.83 workers
for greenhouse-only, and 10.01 workers
for mixed-type operations, as shown
in Table 3. On a per acre basis, the
number of FTE workers averaged
1.21 for all types of operations with-
out any significant variations by type
of operation. The regression equation
describing the decisions involving the
number of FTE workers hired by the
wholesale growing operations account-
ed for 41% of the decision-making
process. Three of the independent vari-
ables included in the employment
model exerted significant influences in
the hiring decisions made by partici-
pating nurseries and greenhouses, as
shown in Table 5. The significant ex-
planatory variables were area in pro-
duction (0.12), nursery-only operation
(–6.09), and greenhouse-only oper-
ation (–4.56). The estimated regres-
sion coefficient of the average level of
mechanization turned out to be statis-
tically insignificant. Comparable re-
gression results were arrived at among
the growers in the three northern
Gulf of Mexico states (Posadas et al.,
2008).
The total annual number of
man-hours employed by each partici-
pating horticulture production firm
averaged 16,830.25 h with signifi-
cant variations by type of operation
(Table 3). On a per acre basis, the
total number of man-hours employed
averaged 5,965.23 h/acre with the
greenhouse-only operations using
significantly more labor input per acre
than the nursery-only and mixed-type
operations. About 49% of the differ-
ences in the hiring decisions dealing
with the number of man-hours were
explained by the estimated regression
equation (Table 5). Three of the
explanatory variables played critical
roles in the hiring decisions made by
Table 4. Workers training, safety, and benefits by type of operation.
Characteristic
All nursery and
greenhouse operations
Nursery-only
operations
Greenhouse-only
operations
Mixed nursery and
greenhouse operations
Workers who were employed in the
same nursery or greenhouse
the previous year (%)
88.90 91.63 86.67 87.32
New workers with basic
horticultural skills (%)
z,y
58.44 73.09 a 65.82 a 34.12 b
Length of annual training period (d)
z,y
5.65 2.25 a 2.27 a 11.88 b
Workers trained on chemical
and pesticide application (%)
26.48 24.69 25.10 29.60
Workers trained on basic
horticultural skills (%)
z,y
31.19 22.82 a 19.90 a 49.33 b
Total annual reported
work-related injuries (number)
0.74 0.64 0.74 0.86
Total annual man-hours lost (hours) 15.52 12.87 5.98 25.08
Workers with access to rest
and lounging areas (%)
y,x
96.10 100.00 b 88.24 a 97.18 b
Workers provided with
housing benefits (%)
z,y
14.06 8.98 a 7.79 a 24.14 b
Workers provided with medical
and dental insurance (%)
8.87 7.67 8.19 10.70
Workers provided with
retirement benefits (%)
z,y
7.59 3.07 a 3.19 a 15.77 b
Workers with access to sanitation
facilities and drinking water (%)
97.03 100.00 96.08 94.29
These training, safety, and economic averages and percentages were computed from the results of the socioeconomic survey, that was conducted among randomly selected
wholesale nurseries and greenhouses located in eight selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
Significantly different by type of operation at P£0.01 using analysis of variance (ANOVA).
y
Values in the same row with different letters are significantly different by type of operation at £0.05 using Duncan’s multiple range test.
x
Significantly different by type of operation at P£0.05 using ANOVA.
•June 2012 22(3) 395
the participating nurseries and green-
houses, namely area in production
(356.52), nursery-only operation
(–15,574.71), and greenhouse-only
operation (–11,517.46). The esti-
mated regression coefficient of the
average level of mechanization was
not statistically significant (Table 5).
Similar regression outcomes were
generated among the growers in the
three northern Gulf of Mexico states for
this empirical model, as follows: acres
in production (343.73), nursery-only
operations (–16,682.48), and green-
house-only operations (–11,663.73).
The employment impact of the
level of mechanization was neutral,
which is contrary to the expected
labor-saving characteristic of automa-
tion or mechanization. Both the num-
ber of FTE workers and the number of
man-hours employed were not signif-
icantly affected by the average level of
mechanization. When the number of
FTE workers was segregated into PW
and PTW, the same results were ob-
served. Both the numbers of PW and
PTW were not significantly influenced
by the average level of mechanization.
The best possible explanation of these
results is that the participating nurser-
iesandgreenhouseswereabletouse
existing labor inputs more efficiently
with any improvements in mechaniza-
tion or automation. The International
Labor Organization (2012) reported
that the increase in mechanization and
automation often speeds up the pace
of work and at times can make work
less interesting.
The number of acres placed in
production exerted positive effects on
man-hours employed and the number
of FTE workers. Each added produc-
tion acre required an additional 0.12
FTEworkeror356.52h.Thetwo
dummy variables representing nursery-
only and greenhouse-only operations
applied negative impacts on number of
FTE workers and number of man-
hours employed. The mixed operations
significantly employed more workers
than the nursery-only or greenhouse-
only operations.
Workers’ skills impact
The percentage of new workers
with basic horticultural skills who were
hired by the participating nurseries
and greenhouses averaged 58.4%
(Table 4). There were significant dif-
ferences in the hiring decisions made
by various types of operations. About
Table 5. Factors influencing the annual gross sales, workers total earnings, number of full-time equivalent (FTE) workers, and total man-hours employed in nurseries and
greenhouses.
Independent variable
Deflated total
revenues (2009 dollars)
Deflated total workers’
earnings (2009 dollars) FTE workers (no.)
Total man-hours
employed (hours)
Coefficient SE Coefficient SE Coefficient SE Coefficient SE
Constant term 33,685,256 28,187,941 –5,283,098 6,393,971 –707.67 737.15 2,279,864 1,835,665
Average level of
mechanization (%)
3,824.63* 1,862.04 1,830.64** 374.71 –0.00 0.04 81.55 121.17
FTE workers (number) 88,904.79** 9,208.17 21,980.64** 666.04 N/A
z
N/A N/A N/A
Deflated wage rate ($/hour) N/A N/A N/A N/A 0.85 0.50 2,136.88 1,247.43
Area in production (acres)
y
–26.35 930.42 742.08** 157.67 0.12** 0.01 356.52** 38.44
Period since establishment (years) 544.91 1,344.89 –313.39 364.37 0.04 0.04 98.90 104.64
Nursery-only operation 81,529.27 79,983.82 21,438.78 14,286.73 –6.09* 1.59 –15,574.71** 3,977.82
Greenhouse-only operation –75,406.65 68,265.15 5,420.87 16,272.17 –4.56* 1.85 –11,517.46* 4,595.55
Area used in production (%) –28.65 613.41 18.58 191.49 0.01 0.02 16.18 55.29
Interview date –46.10 38.47 7.15 8.73 0.00 0.00 3.10 2.50
Included nurseries and
greenhouses (number)
202 — 172 — 177 — 172 —
R-squared 0.89 — 0.94 — 0.41 — 0.49 —
Standard error of regression 337,802.10 — 70,569.10 — 8.26 — 20,322.67 —
F statistic 188.24** — 294.49* — 15.16** — 19.63** —
Durbin–Watson statistic 2.00 — 1.84 — 1.84 — 2.27 —
The multiple linear regressions used data collected from the socioeconomic survey of randomly selected wholesale nurseries and greenhouses located in eight selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
Not applicable in this model.
y
1 acre = 0.4047 ha.
*,**Statistically significant at P£0.05 or 0.01, respectively, using Tvalue or F statistic.
396 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
32% of the decisions to hire new
workers with basic horticultural skills
were explained by the explanatory vari-
ables included in the model (Table 6).
The estimated regression equation
showed that four explanatory variables
exerted significant influence on the
decisions to hire new workers with
horticultural skills. The significant de-
terminants included the average level
of mechanization (–0.49), nursery-
only operations (27.84), greenhouse-
only operations (24.48), and area used
in production (–0.28). Only three
significant variables were reported by
Posadas et al. (2008) among the
growers in the three northern Gulf of
Mexico states, namely average level of
mechanization (–1.68), nursery-only
(53.09), and greenhouse-only (74.80)
operations.
Empirical results showed that
nursery-only and greenhouse-only
operations were more inclined to hire
new workers with basic horticultural
skills than mixed nursery and green-
house operations. Operations which
were using more of the existing acre-
age to horticulture production tended
to employ lesser new workers with
basic horticultural skills. The WSI of
the level of mechanization, as expected
was negative, indicating reduced re-
quirements for manual workers arising
from automation or mechanization.
The regression results suggested that
a 10% increase in the level of mechani-
zation reduced the hiring of new
workers with horticultural skills by
4.9%. In addition, when the number
of FTE workers was separated into its
two components, there were no signif-
icant effects registered by the number
of PW and PTW on the percentage of
new workers with basic horticultural
Table 6. Factors affecting the hiring of new workers with basic horticultural skills, length of basic training period for new
workers employed, man-hours lost (MHL) due to work-related injuries, and number of work-related injuries reported
(WRIR) by the nurseries and greenhouses.
Independent variable
New workers hired with
basic horticultural
skills (%)
Length of basic
training period
for new workers (d)
MHL because of
work-related
injuries (hours) WRIR (no.)
Coefficient SE Coefficient SE Coefficient SE Coefficient SE
Constant term –6,097.23 3,770.37 870.18 1,051.24 –1,088.92 3,438.08 –517.37* 168.27
Average level of
mechanization (%)
–0.49* 0.22 –0.04 0.07 –0.31 0.24 0.03** 0.01
Full-time equivalent
workers (number)
–0.17 0.30 –0.12 0.08 6.69** 0.46 0.09** 0.02
Area in production (acres)
z
–0.12 0.07 0.06** 0.01 0.27* 0.08 –0.00 0.00
Period since establishment (years) –0.33 0.30 0.00 0.05 –0.28 0.21 –0.01 0.01
Nursery-only operation 27.84* 9.25 –6.80* 2.83 5.04 8.62 –0.01 0.42
Greenhouse-only operation 24.48* 9.77 –8.54* 3.85 7.84 9.18 –0.04 0.44
Area used in production (%) –0.28* 0.12 –0.16** 0.06 –0.00 0.10 0.00 0.00
Interview date 0.01 0.10 –0.00 0.00 0.00 0.00 0.00** 0.00
Workers who were employed
in the same nursery or
greenhouse the
previous year (%)
0.03 0.29 0.07 0.05 –0.34 0.19 –0.01 0.00
New workers with basic
horticultural skills (%)
N/A
y
— –0.13* 0.04 –0.06 0.07 0.00 0.00
Workers trained on chemical
and pesticide application (%)
N/A — N/A — 0.26* 0.09 –0.01* 0.00
Workers trained on basic
horticultural skills
N/A — N/A — –0.11 0.09 0.01** 0.00
Operations with annual sales
between $250,000
and $499,999
–0.60 9.19 –1.05 2.51 –15.05 9.06 0.17 0.44
Operations with annual sales
between $500,000
and $999,999
–14.12 13.10 6.34 7.41 –35.64** 12.12 –0.16 0.59
Operations with annual sales
$1 million and above
9.54 13.89 –2.08 4.87 –83.21** 15.09 –0.10 0.73
Included nurseries and
greenhouses (number)
170 — 161 — 169 — 169 —
R-squared 0.32 — 0.27 — 0.75 — 0.33 —
Standard error of regression 38.98 — 15.13 — 37.61 — 1.84 —
F statistic 6.17** — 4.26** — 31.34** — 5.18** —
Durbin–Watson statistic 1.85 — 1.85 — 2.00 — 2.03 —
The multiple linear regressions used data collected from the socioeconomic survey of randomly selected wholesale nurseries and greenhouses located in eight selected southern
U.S. states from Dec. 2003 to Nov. 2009.
z
1 acre = 0.4047 ha.
y
Not applicable in this model.
*,**Statistically significant at P£0.05 or 0.01, respectively, using Tvalue or F test.
•June 2012 22(3) 397
skills who were hired by the participat-
ing nurseries and greenhouses.
Workers’ training time impact
The length of the basic training
period for newly hired workers aver-
aged 5.65 d with significant variations
among various types of operations, as
Table 4 shows. About 27% of the
differences in the number of training
days were explained by the estimated
equation shown in Table 6. Five
explanatory variables have significant
effects on the decisions involving the
length of the training period. The
significant variables were area in pro-
duction (0.06), nursery-only operations
(–6.80), greenhouse-only operations
(–8.54), area used in production
(–0.16), and new workers with basic
horticultural skills (–0.13). These re-
sults included more variables than the
growers in the three northern Gulf of
Mexico states which included only
the new workers with basic horticul-
tural skills (–0.27).
The workers’ TTI of the level of
mechanization was neutral, implying
that its impacts depend on the need
for increased training in the handling
of specialized equipment and the
lower requirement for manual workers
as a result of automation or mechani-
zation. The acres in production had
positive impact on training period
since additional acreage placed under
production required more man-hours,
as illustrated in the employment im-
pact model. The nursery-only and
greenhouse-only operations tended
to spend fewer training days for newly
hired workers since these operations
were more inclined to hire new
workers with basic horticultural skills
than mixed nursery and greenhouse
operations. Since those operations
using more of the existing acreage to
horticulture production tended to
employ fewer new workers with basic
horticultural skills, they devoted lesser
number of days training them. In
addition, when the number of FTE
workers was separated into its two
components, the number of PTWs re-
gistered a significant negative effect
(–0.61) on workers’ TT.
Workers’ safety impact
Workers’ safety was measured in
terms of MHL due to work-related
injuries and number of WRIR. The
number of WRIR the year before
they were interviewed averaged 0.74
injuries per operation with no signif-
icant differences observed among
types of operations (Table 4). The
number of MHL due to work-related
injuries averaged 15.52 h per opera-
tion across the three types of opera-
tions. About 26.4% of the workers in
all three types of operations were
trained in chemical and pesticide ap-
plication. About 31.2% of workers
were trained on basic horticultural
skills with more workers’ training con-
ducted by the mixed-type operations.
The regression results of the
WYI models indicate that 75% and
33% of the variations in MHL and the
number of injuries were explained by
the independent variables included in
the models, respectively (Table 6).
The significant variables affecting the
number of MHL due to work-related
injuries were the number of FTE
workers (6.69), area in production
(0.27), workers trained on chemical
and pesticide application (0.26), op-
erations with annual sales between
$500,000 and $999,999 (–35.64),
and operations with annual sales be-
tween $1,000,000 and above (–83.21).
For the number of work-related in-
juries, the significant determinants were
the average level of mechanization
(0.03), the number of FTE workers
(0.09), date of interview (0.00),
workers trained on chemical and pes-
ticide application (–0.01), and workers
trained on basic horticultural skills
(0.01). In comparison, only the num-
ber of FTE workers significantly af-
fected the number of MHL (9.62) and
number of injuries (0.32) reported by
the growers in the three northern Gulf
of Mexico states.
The WYI was expected to be
positive because automation or mech-
anization would eliminate hazardous
working conditions. However, the
number of WRIR by participating
growers increased as a result of the
improvements in the average level of
mechanization. The International La-
bor Organization (2012) reported
that many workers suffer from injuries
and diseases that result from manual
work and the increased mechaniza-
tion of work. One of the results of
manual work, as well as the increase in
mechanization, is that more and more
workers are suffering from back aches;
neck aches; sore wrists, arms and legs;
and eyestrain. The most commonly
reported injuries among the partici-
pating nurseries and greenhouses
were primarily back strains, cut fin-
gers, shoulder and ankle strains, and
eye injury.
Each FTE worker added to the
labor force led to an additional 0.09
work-related injury and 6.69 h lost as
a result of these injuries. The percent-
age of workers trained on chemical
and pesticide application had direct
effects on MHL. An increase in the
area of production by an acre led to
a rise in the MHL by 0.27 h due to
injuries. Over the 6 years when the
personal interviews were conducted
with the nurseries and greenhouses,
a gradual increasing trend in the
number of work-related injuries was
observed. The increase in the percent-
age of workers trained on basic hor-
ticultural skills caused the number of
work-related injuries to gradually es-
calate. The increase in the percentage
of workers trained on chemical and
pesticide application drove the num-
ber of work-related injuries to fall but
may have caused the MHL to rise.
The wholesale operations with annual
sales above $500,000 reported lower
number of MHL due to injuries.
When the number of FTE workers
was separated into its two compo-
nents, the regression results showed
that the work-related injuries were
accounted for by the 0.09 injuries
reported for every additional PW
employed.
Workers’ retention impact
Nursery and greenhouse growers
can retain their current workers by
maintaining good working condi-
tions, providing workers’ benefits,
and improving productivity through
the adoption of mechanized produc-
tion systems. The lack of field sanita-
tion on agricultural job sites increased
the probability of workers reporting
gastrointestinal disorders (Frisvold
et al., 1987). The percentages of
workers with access to sanitation facil-
ities and drinking water and torest and
recreational areas averaged 97.0% and
96.1%, respectively (Table 4). On the
other hand, low percentages of the
workers were provided with housing
benefits (14.0%), medical and dental
insurance (8.9%), and retirement ben-
efits (7.6%).
Exceedingly high workers’ reten-
tion rates were observed among the
participating wholesale operations av-
eraging 88.9%, with no significant
variations among various types of
398 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
operations, as shown in Table 4. How-
ever, the regression results of the re-
tention model showed that only 12%
of the variations in retention rates was
explained by the independent variables
and that the F test indicated that the
estimated equation was not statistically
significant (Table 7). The WRI of the
level of mechanization was expected to
be positive, but it turned out to be
neutral. It seemed that no significant
variations in retention rates were ob-
served among the participating opera-
tions since almost all of their workers
were with them during the past 2 years
before conducting the interviews. The
same neutral impacts were observed
when the number of FTE workers was
segmented into its two permanent and
part-time components. In contrast,
the retention of workers in the three
northern Gulf of Mexico states
(Posadas et al., 2008) was significantly
affected by greenhouse-only opera-
tions (14.94), workers with access to
rest and lounging areas (0.78), workers
provided with housing benefits (0.17),
and workers provided with retirement
benefits (–0.23).
Conclusions
About one-fifth of the major
tasks performed by workers in nurs-
eries and greenhouses in the eight
southern states that participated in
the face-to-face interviews were per-
formed with some form of mechani-
zation or automation. About 17.5% of
the major tasks in nursery-only opera-
tions were performed by workers with
some form of mechanization or auto-
mation. The top five major tasks per-
formed by nursery workers with
significant levels of mechanization
included irrigation application and
management (51.8%), transporting
containers to field in nursery (31.9%),
filling containers with substrate (28.5%),
media preparation (28.4%), and pesti-
cide application (24.7%). Workers in
greenhouse-only operations accom-
plished 25.4% of their tasks with
some form of mechanization or auto-
mation. The top five major duties
done by greenhouse workers with
considerable levels of mechanization
included irrigation application and
management (56.8%), environmental
control (47.2%), fertilizer application
(39.4%), filling containers with sub-
strate (34.3%), and pesticide applica-
tion (30.6%).
There was wide disparity in the
annual gross sales reported by the
horticulture firms in the eight southern
states. Majority of the growers (55.5%)
were operations, which generated an-
nual gross sales below $250,000. Less
than one-fifth of the horticulture oper-
ations grossed between $250,000 and
$499,999 per year. One out of 10 of
the nurseries and greenhouses earned
between $500,000 and $999,999.
About 14.7% of the horticulture firms
sold horticulture products and services
valued at $1,000,000 or more per year.
Nurseries and greenhouses in the
eight southern states that reported
higher levels of annual gross sales
demonstrated higher levels of mech-
anization or automation, implying
economies of scale associated with
technology adoption by these whole-
sale horticulture production firms.
The two explanatory variables that
exerted significant impacts on de-
flated TR were the average level of
mechanization ($3824.63) and num-
ber of FTE workers ($88,904.79).
However, in the three northern Gulf
of Mexico states, three explanatory
variables were reported to have sig-
nificant effects on annual gross sales,
namely the average level of mechani-
zation ($4899.62), number of FTE
workers ($69,251.62), and acres in
production (958.95).
The increase in total workers’
earnings associated with improved
mechanization indicated that nurseries
and greenhouses were ableto pay their
workers higher wages and salaries. The
overall ratio between the total annual
workers’ earnings and the total annual
gross sales was 29%. When the num-
ber of FTE workers was separated into
its two components, regression results
showed that the marginal earning of
an additional PW was $20,791.65 and
that of a PTW was $15,138.89. It
should be noted that the values of the
annual marginal productivity of the
PW and PTW were estimated to be
Table 7. Factors affecting the retention rate of workers who were employed in the
same nursery or greenhouse the previous year.
Independent variable
Workers’ retention rates (%)
Coefficient SE
Constant term 117.73 1,418.15
Average level of mechanization (%) 0.14 0.11
Full-time equivalent workers (number) 0.20 0.12
Area in production (acres)
z
–0.03 0.01
Period since establishment (years) 0.09 0.08
Nursery-only operation 2.05 3.15
Greenhouse-only operation 2.30 3.78
Area used in production (%) –0.11 0.06
Workers with access to rest
and lounging areas (%)
0.33 0.21
Workers provided with housing benefits (%) 0.07 0.03
Workers provided with medical
and dental insurance (%)
0.23* 0.08
Workers provided with
retirement benefits (%)
–0.17* 0.06
Workers with access to sanitation
facilities and drinking water (%)
–0.48* 0.23
Interview date –0.00 0.00
Operations with annual sales
between $250,000 and $499,999
–3.50 4.36
Operations with annual sales
between $500,000 and $999,999
–12.14 4.71
Operations with annual sales $1 million and above –13.38 4.97
Included nurseries and greenhouses (number) 194 —
R-squared 0.12 —
Standard error of regression 20.40 —
F statistic 1.55 —
Durbin–Watson statistic 1.81 —
The multiple linear regression used data collected from the socioeconomic survey of randomly selected wholesale
nurseries and greenhouses located in eight selected southern U.S. states from Dec. 2003 to Nov. 2009.
z
1 acre = 0.4047 ha.
*Statistically significant at P£0.05 using Tvalue.
•June 2012 22(3) 399
$87,971 and $47,712, respectively.
Three independent variables explained
significantly the differences in total
workers’ earnings including the aver-
age level of mechanization ($1830.64),
number of FTE workers ($21,980.17),
and area in production ($742.08). The
same three independent variables
exerted significant influences on the total
workers’ earnings in the three northern
Gulf of Mexico states, namely average
level of mechanization ($1608.88),
number of FTE workers ($18,650.94),
and area in production ($811.95).
The number of workers or man-
hours hired by the horticultural
firms in the eight southern states aver-
aged 1.21 FTE workers per acre or
5965.23 h per acre with significant
differences by type of operation. The
increased levels of mechanization
produced neutral effects on employ-
ment and raised the VMP of labor,
implying that technology adoption by
wholesale nurseries and greenhouses
did not displace any worker but in-
stead improved total workers’ earn-
ings. When the number of FTE
workers was segregated into its com-
ponents, both the numbers of PW
and PTW were not significantly influ-
enced by the average level of mechani-
zation. The best possible explanation
of these results is that the participating
nurseries and greenhouses were able to
use existing labor inputs more efficiently
with any improvements in mechaniza-
tion or automation. Three independent
variables exerted significant influences
in the hiring decisions made by par-
ticipating nurseries and greenhouses,
namely area in production (0.12),
nursery-only operation (–6.09), and
greenhouse-only operation (–4.56).
Comparable regression results were
arrived at among the growers in the
three northern Gulf of Mexico states.
About 58.4% of the new workers
hired by nurseries and greenhouses in
the eight southern states had basic
horticultural skills with the mixed-
type operations hiring more less-
skilled workers. Significant advances
in mechanization have considerable
implications on the skill levels of
newly hired workers. Growers that
reported higher levels of mechaniza-
tion hired fewer new workers with
basic horticultural skills, especially
among mixed nurseries and green-
houses. Regression results showed
that the average level of mechaniza-
tion (–0.49), nursery-only operation
(27.84), greenhouse-only operation
(24.48), and area used in production
(–0.28) exerted significant influence
over the hiring of workers with basic
horticultural skills. On the other
hand, only three significant variables
were reported among the growers in
the three northern Gulf of Mexico
states, namely average level of mech-
anization (–1.68), nursery-only oper-
ations (53.09), and greenhouse-only
operations (74.80).
Horticulture operations hiring
fewer new workers with basic horti-
cultural skills spent fewer training
days providing them with basic hor-
ticultural skills. The length of training
period for basic horticultural skills was
not influenced by the level of mecha-
nization, but was significantly extended
when nurseries or greenhouses hired
more new workers without basic horti-
cultural skills. Five explanatory variables
have significant effects on the decisions
involving the length of the training
period. The significant variables were
area in production (0.06), nursery-only
operation (–6.80), greenhouse-only
operation(–8.54),areausedinpro-
duction (–0.16), and new workers with
basic horticultural skills (–0.13). These
results included more variables than
the growers in the three northern Gulf
of Mexico states which included only
the new workers with basic horticul-
tural skills (–0.27).
The WYI was expected to be
positive because automation or mech-
anization would eliminate hazardous
working conditions, as reported among
the growers in the three northern Gulf
of Mexico states. However, the num-
ber of WRIR by participating growers
increased as a result of the improve-
ments in the average level of mecha-
nization. The significant variables
affecting the number of MHL due
to work-related injuries were the
number of FTE workers (6.69), area
in production (0.27), workers trained
on chemical and pesticide application
(0.26), operations with annual sales
between $500,000 and $999,999
(–35.64), and operations with an-
nual sales of $1,000,000 and above
(–83.21). For the number of work-
related injuries, the significant de-
terminants were the average level of
mechanization (0.03), the number of
FTE workers (0.09), date of interview
(0.00), workers trained on chemical
and pesticide application (–0.01), and
workers trained on basic horticultural
skills (0.01). In comparison, only the
number of FTE workers significantly
affected the number of MHL (9.62)
and number of injuries (0.32) report-
ed by the growers in the three north-
ern Gulf of Mexico states.
The WRI of mechanization or
automation was neutral since most of
the workers were with the participat-
ing nurseries and greenhouses during
the past 2 years before conducting the
interviews. This result was in contrast
with the findings among the workers in
the nurseries and greenhouses located
in the three northern Gulf of Mexico
states where workers’ retention was
significantly affected by greenhouse-
only operations (14.94), workers with
access to rest and lounging areas
(0.78), workers provided with housing
benefits (0.17), and workers provided
with retirement benefits (–0.23).
Literature cited
Alabama Department of Agriculture and
Industries. 2004. Certified nurseries. Ala-
bama Dept. Agr. Ind., Bur. Plant Ind.,
Montgomery, AL.
Bellenger, M., D. Fields, K. Tilt, and D.
Hite. 2008. Producer preferences for mi-
grant labor and the wage, hours, and gross
sales effects in Alabama’s horticulture in-
dustry. HortTechnology 18:301–307.
Bureau of Labor Statistics. 2012a. Con-
sumer price index. 14 Feb. 2012.
<http://www.bls.gov/cpi/home.htm>.
Bureau of Labor Statistics. 2012b. Occu-
pational outlook handbook, 2010–11
edition: Agricultural workers. 8 Feb.
2012. <http://www.bls.gov/oco/
ocos349.htm>.
Coker, R.Y., B.C. Posadas, S.A. Langlois,
P.R. Knight, and C.H. Coker. 2010.
Current mechanization systems among
nurseries and mixed operations. Missis-
sippi. Agr. For. Expt. Sta. Bul. 1189.
Florida Department of Agriculture and
Consumer Services. 2005. Certified nurs-
ery and stock dealer list. Florida Dept.
Agr. Consumer Serv., Bur. Plant Ind.,
Tallahassee, FL.
Frisvold, G., R. Mines, and J.M. Perloff.
1987. The effects of job sites sanitation
and living conditions on the health and
welfare of agricultural workers.
CUDARE Working Paper 431. Dept.
Agr. Res. Econ., Univ. of California,
Berkeley, CA.
Gabbard, S.M. and J.M. Perloff. 1997.
The effects of pay and work conditions on
farmworker retention. Ind. Relat. 36:
474–488.
400 •June 2012 22(3)
PRODUCTION AND MARKETING REPORTS
Georgia Department of Agriculture.
2007. Live plant licenses. Georgia Dept.
Agr., Bur. Plant Ind., Atlanta, GA.
Giacomelli, G. 2002. Greenhouse struc-
tures. Paper no. E-125933-04-01. Dept.
Agr. Biosystems Eng., Univ. of Arizona,
Tucson, AZ.
Hodges, A.W., C.R. Hall, and M.A.
Palma. 2011. Economic contributions of
the green industry in the United States,
2007. Southern Coop. Series Bul. No. 413.
10 Dec. 2011. <http://www.fred.ifas.
edu/economic-impact/pdf/US-green-
industry-in-2007.pdf>.
Hoppe, R.A., P. Korb, E.J. O’Donoghue,
and D.E. Banker. 2007. Structure and
finances of U.S. farms: Family farm report,
2007 edition. EIB-24. U.S. Dept. Agr.,
Econ. Res. Serv., Washington, DC.
International Labor Organization. 2012.
Your health and safety at work. Ergonom-
ics 8. <http://actrav.itcilo.org/actrav-
english/telearn/osh/ergo/ergoa.htm>.
Ling, P.P. 1994. From mechanization to
the information highway. Greenhouse
systems: Automation, culture and envi-
ronment. Proc. Greenhouse Systems Intl.
Conf. p. 5–7.
Louisiana Department of Agriculture and
Forestry. 2003. Nursery certificate listing.
Louisiana Dept. Agr. For., Hort. Quar-
antine Programs, Baton Rouge, LA.
Mathers, H.M., A.A. Acuna, D.R. Long,
B.K. Behe, A.W. Hodges, J.J. Haydu,
U.K. Schuch, S. Barton, J.H. Dennis,
B.K. Maynard, C.H. Hall, R. McNeil, and
T. Archer. 2010. Nursery worker turnover
and language proficiency. HortTechnology
45:71–77.
Mississippi Department of Agriculture
and Commerce. 2003. Directory of Mis-
sissippi certified nurseries and nursery
dealers. Mississippi Dept. Agr. Commerce,
Bur. Plant Ind., Mississippi State, MS.
North Carolina Department of Agricul-
ture and Consumer Services. 2008. Di-
rectory of certified nurseries and plant
collectors. North Carolina Dept. Agr. Con-
sumer Serv., Plant Ind. Div., Raleigh, NC.
O*Net Online. 2012. Summary report
for: 45-2092.01: Nursery workers. 14
Feb. 2012. <http://www.onetonline.
org/link/summary/45-2092.01>.
Porter, M. 2002. Automation vs. mechani-
zation.5May2008.<http://www.gpnmag.
com/Automation vs. Mechanization–
article3017>.
Posadas, B.C., G.B. Fain, C.H. Coker,
P.R. Knight, C.D. Veal, and R.Y. Coker.
2004. Socioeconomic survey of nursery
automation. Proc. Southern Nursery
Assn. Res. Conf. 49:306–309.
Posadas, B.C., P.R. Knight, C.H. Coker,
R.Y. Coker, and S.A. Langlois. 2008.
Socioeconomic impact of automation on
horticulture production firms in the
northern Gulf of Mexico. HortTechnol-
ogy 18:697–704.
Posadas, B.C., P.R. Knight, C.H. Coker,
R.Y. Coker, and S.A. Langlois. 2010a.
Operational characteristics of nurseries
and greenhouses in the northern Gulf of
Mexico states. Mississippi Agr. For. Expt.
Sta. Bul. 1184.
Posadas, B.C., P.R. Knight, C.H. Coker,
R.Y. Coker, and S.A. Langlois. 2010b.
Socioeconomic characteristics of workers
and working conditions in nurseries and
greenhouses in the northern Gulf of Mex-
ico states. Mississippi Agr. For. Expt. Sta.
Bul. 1182.
Posadas, B.C., P.R. Knight, C.H. Coker,
R.Y. Coker, S.A. Langlois, and C.D. Veal.
2005a. Levels of technology adoption
among horticulture firms in the northern
Gulf of Mexico. Proc. Southern Nursery
Assn. Res. Conf. 50:365–368.
Posadas, B.C., P.R. Knight, C.H. Coker,
R.Y. Coker, S.A. Langlois, and C.D. Veal.
2005b. Socioeconomic characteristics of
horticulture firms in the Gulf South. Proc.
Southern Nursery Assn. Res. Conf.
50:348–350.
Regelbrugge, C.J. 2007. American agri-
culture and immigration reform: An in-
dustry perspective. 5 May 2008. <http://
www.usda.gov/oce///forum/
2007%20Speeches/PDF%20speeches/
CRegelbrugge.pdf>.
Simonton, W. 1992. Automation in the
greenhouse: Challenges, opportunities,
and a robotics case study. HortTechnol-
ogy 2:231–235.
South Carolina Department of Agricul-
ture. 2006. Nursery directory. South Car-
olina Dept. Agr., Columbia, SC.
Tennessee Nursery and Landscape As-
sociation. 2006. Nursery buyers guide.
Tennessee Nursery Landscape Assn.,
McMinnville, TN.
•June 2012 22(3) 401
