Are truck drivers underpaid? - Applied Economics Letters, 2005, 12, 13-18
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
Applied Economics Letters, 2005, 12, 13–18
Are truck drivers underpaid?
Dale Belmana and Kristen Monacob*
a
School of Labour and Industrial Relations, South Kedzie Hall,
Michigan State University, East Lansing, MI 48824-1032, USA
b
Department of Economics, California State University Long Beach,
1250 Bellflower Blvd, Long Beach, CA 90840, USA
There is an emerging debate over whether truck drivers are ‘underpaid’
given their human capital and working conditions. Using data from the
2000 Current Population Survey, the pay differentials between truck
drivers and other blue collar occupations are investigated. While truck
drivers appear to receive a small premium in their hourly wage compared
to workers with similar skill requirements, they receive substantial pre-
mium in their weekly wage. This weekly wage differential is primarily
the result of a substantially longer work week.
It is well established in the literature that truck adequately compensated for their work. The view
driving became a less desirable occupation after that drivers’ pay is too low is common currency
deregulation in the late 1970s, when wages plum- among those who drive, but such views are not
meted significantly.1 Average hourly wages for limited to drivers. Belzer (2000) argues that since
employee truck drivers fell from $16.00 to $12.72 deregulation the non-union sector in trucking has
over the period 1979 to 1993, a decrease of 20.5%.2 taken on many of the characteristics of sweatshops:
After bottoming out in 1993, hourly wages below subsistence wages, overwork, and unpleasant
rebounded somewhat and reached $13.70 in 2000, a and unhealthy working conditions. Others, such as
7.7% increase. Corsi (2001) and Beilock (2001, 2003), argue against
Truck drivers face demanding working condi- such conclusions. In particular, Beilock notes that
tions, especially long-haul truck drivers (Belzer, drivers’ earnings are closely aligned with those of
2000; Belman et al., forthcoming). Drivers’ hours individuals with similar education levels.
of work are governed by Federal Hours of Service How might this issue be addressed? Economic
Regulations which limit drivers to 60 hours of work theory suggests that wages are largely determined
in a seven day period. It is widely acknowledged in by an individual’s productivity, however, pay is not
the industry that long-haul drivers routinely violate determined solely by productivity. The theory of
these regulations (Belman et al., 2004; Braver et al., compensating differentials implies that pay is influ-
1992; Beilock, 1995) and often work considerably enced by working conditions. Employers who pro-
longer than 60 hours per week. vide substandard working conditions will be unable
In combination with the long hours of work, to attract a sufficient supply of workers at the going
the post-deregulation decline in driver’s real earnings wage rate, which compels them to offer an increment
has fuelled discussion over whether drivers are to pay to attract workers who would otherwise
*Corresponding author. E-mail: kmonaco@csulb.edu
1
Research on deregulation’s impact on the trucking labour market includes Belzer (1995), Rose (1987), Hirsch (1988, 1993),
Hirsch and Macpherson (1998), and Belman and Monaco (2001).
2
All figures are calculated in 2000 dollars using data from the Current Population Survey. Data from the May Survey is used
in 1979 and from the Outgoing Rotation Groups files after 1979.
Applied Economics Letters ISSN 1350–4851 print/ISSN 1466–4291 online # 2005 Taylor & Francis Group Ltd 13
http://www.tandf.co.uk/journals
DOI: 10.1080/1350485042000291411Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
14 D. Belman and K. Monaco
choose jobs with better conditions. As labour mar- We estimate this model using in turn log hourly
kets are characterized by a certain level of fluidity, earnings and log weekly earnings as the dependent
this ‘compensating differential’ applies not only to variable, and report results for each of these respec-
similar jobs within an industry but across industries tively in Tables 1 and 2.5 We use the log wage
and occupations. Given the demanding nature of estimates to construct predictions of the hourly
truck drivers’ jobs, we would expect that if truck dri- wage by occupation, and report these in the tables.
vers were underpaid their wages would resemble The predictions are standardized for human capital
those of workers with similar human capital working by calculating the predicted wage at sample mean
under better conditions. human capital characteristics.
One way to determine whether the market is For each table, we display the occupational title in
compensating drivers for their working conditions column 1, followed by the occupation code used in
is to isolate the component of individuals’ wages the CPS in column 2, the actual mean hourly earnings
associated with their occupation and compare this by occupation (unadjusted for human capital or other
to that of similar occupations.3 This is done by characteristics) in dollars in column 3, and by the
estimating a wage model with an indicator variable earnings predicted by the regression model (calcu-
for each occupation but no overall intercept. The lated at the mean human capital characteristics),
coefficient on the occupation indicator variable is also in dollars, in column 4. In the next column, we
the ‘occupational wage component,’ which reflects show the difference between the earnings of drivers in
the market valuation of the advantages and disadvan- the trucking industry and each of the other occupa-
tages of that occupation. The estimated magnitudes tions, including truck driving outside of the trucking
of the occupation components can be compared industry, again in dollars. In the sixth column we
between occupations and judgments made about the present the t-statistic that indicates whether the
appropriateness of the implied wage differentials. difference is statistically significant. The last column
This approach is necessarily somewhat judgmental shows the sample size for the occupation, that is the
as there is seldom a perfect fit between occupations. number of individuals in the sample in each occupa-
Nevertheless, it provides a workable means of explor- tion. In both tables, we restrict ourselves to occupa-
ing the issue of driver underpay.4 tions that have at least 80 observations to reduce the
We use data from the year 2000 Current effects of sampling variability. We believe that the
Population Survey Outgoing Rotation Groups files occupations in question are a representative subset
to estimate a conventional human capital wage equa- of the blue-collar occupations in our data and in
tion with controls for educational attainment, age, that sense the results in Tables 1 and 2 characterize
gender, race and ethnicity, union membership, region the patterns we found in the larger set of occupations.
of residence and residence in a metropolitan area. We Table 1 shows that the sample mean hourly
limit our sample to those individuals employed in the earnings for truck drivers employed in the trucking
private sector only and exclude individuals who are industry is $15.67 while the earnings predicted at
self-employed. The 99 821 individuals in the resulting sample mean human capital characteristics are
data were engaged in 484 different three-digit private $13.64. A natural initial comparison is with truck
sector occupations according to the CPS definitions. drivers employed outside of the trucking industry.
The model includes indicator variables for each of This ‘occupation’ is comprised of drivers whose
these 484 occupations. To better understand the work is similar to that of drivers within motor freight,
occupational effects of employment in trucking, we but it also includes local delivery drivers in light
separate truck drivers employed in the trucking trucks, construction drivers and others whose work
industry from those employed in other industries. and working conditions are quite different from that
3
An alternative would be to incorporate measures of working conditions into a conventional human capital model of wages.
Although the numeric ratings for occupational characteristics are available from the Dictionary of Occupational Titles or
from ONET, the incorporation of these measures into wage regressions has been problematic. Because of all these issues, few
articles in the compensating differentials literature incorporate direct controls for occupational characteristics.
4
Further explanation and an example of this approach is provided in Belman et al. (2002).
5
Hourly earnings are calculated as the ratio of average weekly earnings to actual hours last week. Actual rather than average
hours was used because employees who work irregular schedules report a value of ‘variable hours’ for average hours and this
precludes calculating an hourly wage. As many truck drivers work irregular schedules, use of average hours would exclude
large numbers of truck drivers. Although actual hours is a noisy representation of average weekly hours for an individual, the
estimates of the hourly wage should be reasonably accurate given the size of the sample used in this study.Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
Table 1. Comparing hourly earnings across occupations
Are truck drivers underpaid?
Adjusted for human capital
Population
occupation Hourly earnings Predicted Difference from t-stat. for
code (unadjusted) earnings drivers in trucking difference N
Motor Vehicle Operators
Truck drivers-trucking industry 804 $15.67 $13.64 1308
Truck drivers-other industries $13.11 $12.06 $1.57 2.910 1442
Transportation Operatives
Supervisors, motor vehicle opers 803 $17.29 $13.98 $0.34 0.374 82
Driver-sales workers 806 $14.35 $12.24 $1.40 1.948 155
Bus drivers 808 $12.08 $11.36 $2.27 3.677 276
Taxi cab drivers/chauffeurs 809 $11.21 $9.94 $3.70 6.063 215
Operatives
Punching & stamping press oper 706 $12.33 $11.86 $1.78 2.200 90
Grinding & polishing machine oper 709 $12.52 $12.31 $1.32 1.610 94
Printing machine operators 734 $14.99 $13.42 $0.21 0.319 287
Textile sewing machine operators 744 $8.56 $10.14 $3.49 6.400 423
Painting and spraying machine oper 759 $13.78 $13.30 $0.34 0.464 168
Material Moving Operators
Operating engineers 844 $17.53 $15.75 $2.11 2.778 215
Crane and tower operators 849 $16.26 $13.62 $0.02 0.020 75
Excavating & loading machine oper 853 $16.38 $15.50 $1.87 2.042 95
Industrial truck/tractor equip oper 856 $12.58 $11.93 $1.70 2.927 485
Labourers
Stock handlers and baggers 877 $9.39 $9.61 $4.03 7.872 635
Freight & material handler, nec 883 $11.81 $11.30 $2.34 4.157 522
Labourers, ex construction (8769) 889 $11.65 $11.07 $2.56 4.876 1081
Precision Production
Automobile mechanics 505 $15.08 $13.77 $0.13 0.220 661
Bus, truck, and engine mechanics 507 $16.15 $14.85 $1.22 1.797 341
Aircraft engine mechanics 508 $18.74 $15.11 $1.47 1.716 108
Heavy equipment mechanics 516 $16.29 $14.79 $1.16 1.511 174
HVAC mech 534 $16.97 $14.91 $1.28 1.819 277
Carpenters 567 $16.38 $14.85 $1.22 1.990 1021
Electricians 575 $19.46 $16.42 $2.79 4.231 728
Plumbers, pipefitters & steamfitters 585 $18.28 $15.78 $2.14 3.178 468
Roofers 595 $15.89 $14.58 $0.94 1.172 150
Machinists 637 $16.57 $14.53 $0.90 1.411 509
Butchers and meat cutters 686 $12.38 $12.57 $1.07 1.658 275
Bakers 687 $11.27 $11.19 $2.44 3.563 152
Earnings estimates constructed from a log wage equation estimated with the 2000 Outgoing Rotation of the Current Population Survey. The wage
difference is calculated as (estimated earnings of drivers – estimated earnings of other occupation).
15Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
Table 2. Comparing weekly earnings across occupations
16
Weekly Earnings
–
Adjusted for Human Capital
–
Population
occupation Weekly earnings Predicted Difference from t-stat. for Number of
code (unadjusted) earnings truck drivers difference observations
Motor Vehicle Operators
Truck drivers-trucking industry 804 $708.54 $603.01 1986
Truck drivers-other industries $554.77 $501.88 $101.13 4.105 2032
Transportation Operatives
Supervisors, motor vehicle opers 803 $779.94 $615.26 $12.25 0.289 106
Driver-sales workers 806 $614.17 $527.76 $75.25 2.245 206
Bus drivers 808 $421.40 $457.35 $145.67 5.300 383
Taxi cab drivers/chauffeurs 809 $484.80 $422.20 $180.82 6.438 362
Operatives
Punching & stamping press oper 706 $496.27 $468.49 $134.52 3.793 118
Grinding & polishing machine oper 709 $541.34 $501.46 $101.55 2.749 116
Printing machine operators 734 $617.18 $536.87 $66.14 2.234 370
Textile sewing machine operators 744 $334.91 $418.27 $184.74 7.455 528
Painting and spraying machine oper 759 $557.36 $525.21 $77.80 2.390 220
Material Moving Operators
Operating engineers 844 $709.83 $635.73 $32.72 0.962 289
Crane and tower operators 849 $701.10 $557.97 $45.05 1.081 90
Excavating & loading machine oper 853 $645.86 $618.52 $15.51 0.383 151
Industrial truck/tractor equip oper 856 $519.41 $471.80 $131.21 5.093 642
Labourers
Stock handlers and baggers 877 $346.87 $377.27 $225.74 9.947 1555
Freight & material handler, nec 883 $454.71 $445.31 $157.71 6.306 856
Labourers, ex construction (8769) 889 $442.60 $424.82 $178.19 7.724 1637
Precision Production
Automobile mechanics 505 $634.97 $552.78 $50.24 1.849 1104
Bus, truck, and engine mechanics 507 $685.64 $601.36 $1.65 0.054 431
D. Belman and K. Monaco
Aircraft engine mechanics 508 $806.37 $610.29 $7.28 0.188 130
Heavy equipment mechanics 516 $710.87 $618.19 $15.18 0.435 206
HVAC mech 534 $693.86 $594.45 $8.56 0.274 424
Carpenters 567 $634.02 $563.90 $39.11 1.470 1890
Electricians 575 $815.32 $645.98 $42.97 1.476 987
Plumbers, pipefitters & steamfitters 585 $732.71 $622.91 $19.90 0.665 690
Roofers 595 $557.59 $558.77 $44.24 1.269 254
Machinists 637 $680.48 $577.62 $25.39 0.893 599
Butchers and meat cutters 686 $508.31 $511.01 $92.00 3.179 370
Bakers 687 $437.58 $419.44 $183.57 6.176 227
Earnings estimates constructed from a log wage equation estimated with the 2000 Outgoing Rotation of the Current Population Survey. The wage
difference is calculated as (estimated earnings of drivers – estimated earnings of other occupation).Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
Are truck drivers underpaid? 17
of freight drivers. Average hourly earnings for these they also involve exposure to weather and hazardous
drivers are $13.11, while the predicted hourly earn- conditions and may involve periods of time away
ings are $12.06, $1.57 less per hour than drivers in from home. Our estimates suggest that two of the
motor freight. Motor freight drivers earn a premium three construction occupations are paid more than
of 13.1% over employees with similar skills doing drivers (operating engineers and excavating machine
similar work, but who may be less likely to work operators), while tower and crane operators have
extended hours or be away from home for extended similar hourly earnings (crane and tower operators).
periods. Industrial truck and tractor equipment operators,
Comparing the earnings of drivers in motor freight individuals who operate trucks and other moving
with those of other motor vehicle operatives again equipment on factory premises and whose jobs
indicates that motor freight drivers earn a premium involved skills similar to those of truck drivers, are
over other drivers. Drivers in motor freight earn paid $1.70 per hour less than drivers. The lower pay
$1.40 (11.4%) more than driver sales workers, $2.27 is consistent with a downward adjustment in compen-
(20.1%) per hour more than bus drivers, and $3.70 sation for the better conditions of industrial truck
(37.2%) more than taxi drivers; each difference is operators.
significant at the 5% or better level.6 Finally, precision production occupations are blue-
The balance of Table 1 compares drivers in motor collar jobs that require advanced vocational training,
freight with non-transportation blue-collar occupa- often three or four year apprenticeships. Some
tions. Individuals in the first group, production of these occupations involve exposure to weather,
operatives, are engaged in repetitive operations, typi- various hazards in construction sites, and may also
cally in a factory or other production facility. Similar require that workers spend part of the year away
to truck driving, these jobs normally require no from home. The predicted earnings of most of these
more than a high-school degree and no more than occupations are $1.00 to $2.00 per hour more than
a few months of occupation-specific training and those of truck drivers, and most of these differences
on-the-job experience. They provide better working are statistically significant. The occupations for which
conditions than truck driving, at least with respect to earnings are not significantly higher than those of
hours of work and rest and time spent at home. The drivers are those that do not involve extended
wages predicted by our regression model indicate advanced vocational training (auto mechanics and
that the hourly earnings of truck drivers in motor roofers) or trades that formerly involved advanced
freight are similar to those of painting and printing skills but that have evolved into production jobs in
machine operators but significantly greater than factories (butchers, bakers and some machinists).
that of sewing machine, punch press and grinding The pattern that emerges from the hourly wage
machine operators. data indicates that truck drivers earn more than
The educational and training requirements for workers in manual occupations with similar training
labourers, manual workers involved in operations requirements that do not entail the hours or working
such as stock and freight handling, are similar to conditions associated with truck driving. In contrast,
those of truck drivers but labourers are less likely drivers often earn considerably less than manual
to control expensive or powerful machinery. occupations that involve advanced vocational train-
Although labourers’ duties often involve extended ing and that may involve less favourable working
periods of heavy work, they are not characterized conditions.
by the extended working time and time away from The favourable position of truck drivers relative to
home issues that characterize the work of truck occupations with similar training requirements
drivers. Consistent with the lesser requirements of becomes more pronounced when weekly earnings
the labourers’ position, most labourers are paid are considered (Table 2). Drivers in motor freight
considerably less, between $2.34 and $4.03, than earn considerably more per week than drivers outside
truck drivers, and these differences are statistically of motor freight, transportation operatives, and
significant. labourers. Drivers’ weekly pay is similar to that of
Material moving occupations involve the operation material moving operatives, including operating
of heavy equipment. Proficiency in these occupations engineers and excavating machine operators, and
may require extended formal and on-the-job training; of precision production workers including aircraft
6
A driver sales worker is typically a truck driver who delivers and positions goods. Soft drink and bakery drivers, who unload
their trucks and place the goods on store shelves are driver sales workers.Downloaded By: [Michigan State University Libraries] At: 16:32 15 November 2007
18 D. Belman and K. Monaco
engine mechanics, plumbers and carpenters. On a References
weekly basis, the earnings of drivers closely approach Beilock, R. (1995) Schedule-induced hours-of-service and
those of electricians. speed limit violations among tractor-trailer drivers,
In combination, these results imply that truck Accident Analysis and Prevention 27, 33–42.
Beilock, R. (2001) Review of: sweatshops on wheels:
drivers are paid somewhat more per hour than
winners and losers in trucking deregulation,
those employed in occupations with similar skill American Economic Review, 39, 1264–5.
requirements but better working conditions. Beilock, R. (2003) The elusive sweatshop, Journal of the
However, the longer weekly work hours of truck Transportation Research Forum, 153–65.
drivers allow them to earn as much per week as Belman, D. L., Monaco, K. A. and Brooks, T. J. Sailors
of the Concrete Sea: A Portrait of Truck Drivers Work
blue-collar workers in occupations requiring consid- and Worklives, Michigan State University Press, East
erably more skills and occupational training. An Lansing, forthcoming
interpretation of these results is that truck drivers Belman, D. L., Heywood, J. S. and Voos, P. (2002) Public
engage in a Faustian bargain in which they accept sector earnings comparability: alternative estimates
long hours of work in return for earning as much for the U.S. Postal Service, Relations Industrielles-
Industrial Relations, 57, 687–711.
as workers engaged in manual occupations that Belman, D. L. and Monaco, K. A. (2001) The effects
require more training and skills. of deregulation, de-unionization, technology, and
Nevertheless, it is possible that markets do not human capital on the work and work lives of truck
compensate drivers for the full value of the disutility drivers, Industrial and Labour Relations Review, 54,
502–24.
associated with their working conditions. Economic
Belzer, M. H. (1995) Collective bargaining after deregula-
theory proposes that compensating differentials are tion: do the teamsters still count?, Industrial and
driven by the movement of marginal workers out of Labour Relations Review, 48, 636–55.
employment with worse conditions and into employ- Belzer, M. H. (2000) Sweatshops on Wheels: Winners and
ment with better conditions. Firms seeking to staunch Losers in Tricking Deregulation, Oxford University
Press, Oxford.
the flow of workers away from worse conditions, Braver, E. R., Preusser, C. W., Preusser, D. F., Baum,
must increase their pay, thereby generating a suffi- H. M., Beilock, R. and Ulmer, R. (1992) Long hours
cient compensating differential. This mechanism and fatigue: a survey of tractor-trailer drivers, Journal
relies on ease of movement between markets. Differ- of Public Health Policy, 13, 341–66.
ences in the hours of work regulations between truck- Corsi, T. (2001) Book Review: Sweatshops on Wheels:
Winners and Losers in Trucking Deregulation, Journal
ing and the balance of the economy can impede such of the Transportation Research Forum, 40, 147–9.
movement. Movement from a 60-hour week in truck- Hirsch, B. T. (1988) Trucking regulations, unionization,
ing to a job with similar pay and somewhat better and labour earnings: 1973–1985, Journal of Human
conditions outside of trucking would be accompanied Resources, 23, 296–319.
Hirsch, B. T. (1993) Trucking deregulation and earnings:
by a one-third reduction in weekly take home pay.
is the union premium a compensating differential?,
The reduction in income associated with migration Journal of Labour Economics, 11, 279–301.
out of the industry serves to restrain the arbitraging Hirsch, B. T. and Macpherson, D. A. (1998) Earnings
of pay and working conditions. Although the and employment in trucking: deregulating a naturally
results of this inquiry suggest that driver pay is not competitive industry, in Regulatory Reform and Labour
Markets (Ed.) J. Peoples, Kluwer Academic
greatly out of line with the nature of their job, the Publishers, Norwell, MA.
impediments to the smooth functioning of the Rose, N. L. (1987) Labour rent sharing and regulation:
market suggests that moderate underpay may exist evidence from the trucking industry, Journal of
and persist. Political Economy, 95, 1146–78.You can also read
