Content uploaded by Yingbin Feng
Author content
All content in this area was uploaded by Yingbin Feng on Jan 22, 2020
Content may be subject to copyright.
The International Journal of Construction Management (2011) Vol. 11 No. 3, 79-92
COSTS OF CONSTRUCTION ACCIDENTS TO
SINGAPORE CONTRACTORS
Evelyn Ai-Lin TEO1 and Yingbin FENG2
1 Department of Building, National University of Singapore, 4 Architecture Drive, Singapore
117566. Email: bdgteoal@nus.edu.sg
2 Email: fengyingbin@nus.edu.sg
Abstract
Occupational accidents may incur considerable financial losses for companies. The
purpose of this study was to investigate the financial losses incurred by building
contractors in Singapore due to construction accidents. A questionnaire survey was
used to collect data for this study. The results show that the average insured accident
costs, average uninsured accident costs and average total accident costs account for
0.15%, 0.1% and 0.25% of the contract sum of a project, respectively. The ratio
between insured and uninsured accident costs for building projects was found to be
1.5:1. Furthermore, company size, project size, and percentage of work completed by
subcontractors were found to have impact on the size of uninsured costs of accidents.
In light of this study, Singapore contractors may be more conscious about the
financial implications of construction accidents, which would in turn motivate
contractors to prevent construction accidents.
Keywords
Construction, accidents, safety, costs, Singapore.
INTRODUCTION
In the “Workplace Safety and Health (Incident Reporting) Regulations 2006” of
Singapore (MOM, 2006), an accident is defined as any unintended event which causes
bodily injury to a person and a workplace accident is any accident occurring in the
course of a person’s work. Various losses are incurred by the injured worker(s) after
the occurrence of an accident. These losses may include costs to victims and their
families, to employers and to society (Davies and Teasedale, 1994). The European
Agency for Safety and Health at Work estimated that 4.6 million workplace accidents
happened every year in the EU resulting in 146 million lost working hours (EU
OSHA, 2001). This translates to approximately 2.6–3.8% of the collective EU gross
national production (GNP) lost every year.
Workplace accidents in the construction industry may also cause considerable
financial losses for individual contractors. According to Levitt et al. (1981), accidents
costs in construction companies in USA were found to be as high as 3% of the total
construction project costs (10% of labour costs). The costs of accidents have long
been regarded as a motivating factor for improving safety performance (e.g., Heinrich
1931; Levitt 1975; Lingard and Rowlinson 2005). U.S. Department of labor (1955)
argued that the main driving force behind the industrial safety movement is the fact
80 Teo and Feng
that accidents are expensive, and substantial savings can be made by preventing them.
Thus, preventing workplace accidents should make good economic sense for
contractors (Dorman, 2000).
This study aims to examine the costs of construction accidents to Singapore
contractors. The specific objectives are: (1) to estimate the accident costs of building
projects; (2) to investigate the components of uninsured accident costs of accidents to
contractors; and (3) to identify the factors influencing the uninsured costs of accidents.
In this study, the costs of workplace accident are confined to the financial losses of a
contractor. Social costs, such as costs associated with human feelings and impact on
family and society, are not included in this research. This is because social costs are
not based from the contractor’s point of view, but on society’s point of view. Other
ancillary costs arising from the accident which are difficult to quantify, such as
damage to company reputation and morale of employees are also excluded in this
study.
LITERATURE REVIEW
The study of costs of accident was pioneered by Heinrich (1931) more than 80 years
ago. He classified the costs as direct and indirect costs, and concluded that indirect
costs were significant as they accounted for as much as four times the direct costs of
accidents (Heinrich 1931). Simonds and Grimaldi (1956) proposed an alternative
approach by dividing the costs into insured and uninsured costs. Simonds and
Grimaldi (1956) criticized Heinrich’s (1931) definition of indirect costs, arguing that
many such costs, for example the overhead cost of insurance, are direct since they
appear in a firm’s financial accounts. Although not all later researchers were
persuaded to change their jargon to insured costs and uninsured costs as proposed by
Simonds and Grimaldi (1956), some of them were prompted to re-define the direct
and indirect costs as insured and uninsured costs (Head and Harcourt, 1997).
The categorization of accident costs into direct and indirect costs or insured and
uninsured costs implies that focus only on the direct costs may fail to reveal the true
losses to employers due to an accident. Many of the losses incurred by an accident are
“hidden” and difficult to quantify. These “hidden” costs may be significant, and some
may be particularly prominent in the construction industry. For example, there are
heavy penalties for time-overruns on construction projects (Lingard and Rowlinson,
2005). The Accident Cost Iceberg proposed by Bird (1974) showed that the proportion
of hidden costs could be much larger than the costs directly related to the accident.
The indirect cost theory of workplace accidents developed by Brody et al. (1990)
suggests that the identification of indirect costs will motivate cost-minimizing firms to
increase investment in accident prevention to improve the safety performance of
building projects.
In addition to traditional classification of accident cost as direct (insured) and indirect
(uninsured) costs, several researchers proposed different accident cost typologies
based on the specific characteristics of the accident costs. For example, in the cost
typology proposed by Riel and Imbeau (1996), health and safety costs are classified
into three categories: insurance-related costs; work-related costs; and
perturbation-related costs. They are also classified as quantifiable, irreducible and
81
Costs of Construction Accidents to Singapore Contractors
intangible costs in this typology. Rikhardsson and Impgaard (2004) argued that the
traditional cost components are rather difficult for management to use, as they require
a number of definitions and clarifications before use including asset specifications and
income definitions. Thus, they categorized accident costs as time, materials and
components, external services and other costs. These categories reflect traditional
accounting classifications in accounting systems, thus they are believed to be simpler
to apply by managers. Despite the debates on various typologies of accident costs, the
consequences or cost components of accidents seem to be consistent among literature.
In order to understand the components of accident costs, a number of past studies
have been examined. Table 1 lists some of the accidents cost research undertaken
since 1931. These nine studies give a comprehensive representation of direct (insured)
and indirect (uninsured) accident costs components. The direct accident costs are
those actual cash flows that can be directly attributable to or associated with injuries
and fatalities (Everett and Frank Jr. 1996; Hinze 1997). The insured accident costs to
contractors are typically the costs covered by work injury compensation insurance
policies. These costs may be measured by the insurance premium paid by contractors
(Simonds and Grimaldi, 1956; Laufer, 1987; Leopold and Leonard, 1987; Monnery,
1999).
Table 1: List and Summary of Previous Accident Costs Research
Reference
Cost typology
Heinrich
(1931)
Direct costs:
1) Compensation; 2) Medical aid
Indirect costs:
1) Cost of lost time of the injured employee; 2) Cost of time lost by other employees;
3) Cost of time lost by foremen; 4) Cost of time spent by first aid attendants; 5) Costs
due to damage to machines, tools or other property; 6) Incidental costs due to
interference with production; 7) Costs to employers under employee welfare systems;
8) Costs to employers in continuing the wages to the injured employee; 9) Costs due
to loss of profit on the injured employee’s productivity; 10) Costs that occur in
consequence of weakened morale due to the accident; 11) Overhead costs per injured
employee.
Simonds and
Grimaldi
(1956)
Insured costs:
1) Net insurance premium
Uninsured costs:
1) Cost of wages paid for working time lost by workers who were not injured; 2) The
net cost to repair, replace, or straighten up material or equipment that was damaged in
an “accident”; 3) Cost of wages paid for working time lost by injured workers, other
than workmen’s compensation payments; 4) Extra cost due to overtime work
necessitated by an “accident”; 5) Cost of wages paid to supervisors while their time is
required for activities necessitated by the injury; 6) Wage cost due to decreased
output of injured worker after return to work; 7) Cost of learning period of new
worker; 8) Uninsured medical cost borne by the company; 9) Cost of time spent by
higher supervision and clerical workers on investigations or in the processing of
compensation applications forms; 10) Miscellaneous unusual costs.
Laufer
(1987)
Insured costs:
1) Net insurance premium
Uninsured costs:
1) Costs due to labor lost time, including injured workers, other workers, replacement
worker, foreman, clear-up and administration; 2) Costs due to complementary wages
to the injured; 3) Cost due to property accidents.
82 Teo and Feng
Table 1: List and Summary of Previous Accident Costs Research (cont’d)
Reference
Cost typology
Leopold and
Leonard
(1987)
Insured costs
1) Net insurance premium
Uninsured costs
1) Lost labor; 2) Continuing payments to injured worker after accident; 3) Insurance
costs; 4) Damage to equipment; 5) Legal costs
Brody et al.
(1990)
Direct costs:
1) Fix insurance costs; 2) Variable insurance costs.
Indirect costs:
1) Wage costs; 2) Material damage; 3) Administrators’ time; 4) Production losses; 5)
Other costs; 6) Intangible costs
Hinze
(1991)
Direct costs:
1) Costs reimbursed by Worker’s Compensation Insurance
Indirect costs:
1) Cost of injured worker; 2) Cost of injured worker’s crew; 3) Costs associated with
obtaining medical help; 4) Costs of other crews; 5) Costs of equipment and material
damage; 6) Costs of supervisory staff; 7) Other costs
Everett and
Frank Jr.
(1996)
Direct costs:
1) Benefits paid to injured workers by Workers’ Compensation Insurance
Indirect costs:
1) WCI carriers’ overhead and profit; 2) Claims costs; 3) Other costs, including loss
of productivity, disruption of schedules, administrative time for investigations and
reports, training of replacement personnel, wages paid to the injured workers and
others for time not worked, cleanup and repair, adverse publicity, and equipment
damage.
Head and
Harcourt
(1997)
Direct costs:
1) Those paid by the Accident Rehabilitation, Compensation, and Insurance
Corporation’s Employers’ Account
Indirect costs:
1) Indirect community costs, including accident investigations by OSH, and social
welfare benefits; 2) Indirect employer costs, including productivity losses, accident
investigations, legal penalties, recruitment, selection, and training; 3) Indirect
employee costs, including health and medical services, partial loss of earnings, and
full loss of earnings.
Monnery
(1999)
Insured costs:
1) Insurance premiums
Uninsured costs:
1) Cost of absentees time;2) Cost of other person’s time; 3) Travel to hospital; 4)
Replacement labour; 5) Machine breakdown; 6) Opportunity costs (Financial costs).
Different definitions exist for the indirect (uninsured) costs of accidents, but in
general they are regarded as consisting of all the costs that are not covered by
worker’s compensation insurance (Hinze, 1991). The components of uninsured
accident costs originate from studies that have been focused on accident costs in
various industries (e.g., construction, furniture, forestry, chemistry, cleaning service,
financial service, and manufacturing). Nonetheless, as shown in Table 1, the
components of uninsured accident costs from various industries demonstrate strong
similarities. Based on the literature review (see Table 1), a set of components of
uninsured accident costs in construction environment was identified. The uninsured
costs of accidents comprise the following 14 possible components:
(1) Compensation for the injured worker that is not covered by insurance policy
(Simonds and Grimaldi, 1956);
(2) Lost productivity due to the injured worker (e.g., Heinrich, 1931; Simonds and
Grimaldi, 1956; Hinze, 1991);
83
Costs of Construction Accidents to Singapore Contractors
(3) Lost productivity due to crew of injured worker (e.g., Heinrich, 1931; Hinze,
1991; Monnery, 1999);
(4) Lost productivity due to other workers in vicinity of accidents (e.g., Heinrich,
1931; Laufer, 1987; Hinze, 1991);
(5) Losses due to replacement of the injured worker (e.g., Laufer, 1987; Everett and
Frank Jr., 1996; Monnery, 1999);
(6) Lost productivity due to the investigation or inspections as a result of the injury
(Simonds and Grimaldi, 1956; Head and Harcourt, 1997);
(7) Cost of supervisory or staff effort (e.g., Heinrich, 1931; Simonds and Grimaldi,
1956; Hinze, 1991);
(8) Losses due to damaged equipment or plant, property, material or finished work
due to the accident (e.g., Heinrich, 1931; Brody et al., 1990; Hinze, 1991);
(9) Cost of transporting injured worker (e.g., Simonds and Grimaldi, 1956; Hinze,
1991; Monnery, 1999);
(10) Consumption of first-aid materials in the accident (Hinze, 1991; Head and
Harcourt, 1997);
(11) Additional work required as a result of the accident (e.g. cleaning, additional
barriers and so on) (e.g., Simonds and Grimaldi, 1956; Laufer, 1987; Everett and
Frank Jr., 1996);
(12) Fines and legal expenses (Leopold and Leonard, 1987; Head and Harcourt,
1997);
(13) Losses due to Stop Work Orders (SWO) issued to the project (disruption of
schedules) (Brody et al., 1990; Everett and Frank Jr., 1996);
(14) Additional benefits to the injured worker beyond the Work Compensation Act
(WCA) (Heinrich, 1931).
METHODOLOGY
Instrument
A questionnaire was designed with the objective of collecting information about the
costs of construction accidents to building contractors. The questionnaire consists of
two parts. The first part collects the information about the general characteristics of
the project, e.g., company size, project size, duration, man-days worked, height of
building, type of project, type of client and so on. The second part aims to collect the
information based on a job related accident that happened in the project. For more
than one accident in the project, the respondents were required to photocopy this
section for other accidents. Based on literature review, 14 possible uninsured cost
items were identified to be related to the occurrence of the accidents (see Literature
Review Section). Respondents were required to provide information or estimate the
costs of these 14 cost items based on the questions in the questionnaire.
A pilot study was conducted by means of five structured interviews using the
questionnaire to pre-test the instrument and validate the questions. The interviewees
comprised three project managers and two safety officers from five different
completed building projects in Singapore. All the three project managers have more
than 15 years of experience in construction industry, and both the two safety officers
are registered Workplace Safety and Health Officers (WSHO) with MOM and have
more than 10 years of experience in construction safety. The pilot study led to some
adjustments to the questionnaire: (1) some changes in the wording of the
84 Teo and Feng
questionnaire were made to avoid confusion; and (2) some other possible costs items
incurred by work injuries were added to the questionnaire.
Data Collection
Personal interviews using a questionnaire were used to collect data for this study. The
population consists of all building contractors in Singapore. The sampling frame is a
list of general building contractors who were registered with the Building
Construction Authority (BCA) of Singapore. 234 general building contractors
belonging to the grades A1, A2, B1, and B2 under Construction Work Heads CW 01 –
General Building category were selected for the data collection. The contractors
belonging to the C1, C2, and C3 categories (with tendering limit of S$5 million and
below) were excluded from the sampling frame of this study. This is because,
according to the practices of the Singapore construction industry, small general
building contractors (C1, C2, and C3) usually perform as sub-contractors of building
projects and it is not possible to acquire complete information about the whole
building project from sub-contractors (Teo and Feng 2010).
Since the sampling frame is naturally stratified by BCA Grade, a stratified sampling
method (see Tan, 2008) was adopted for this study. To ensure representative data,
samples from homogeneous strata were randomly selected. In the first stage, 50% of
contractors under each BCA Grade were randomly selected from the sampling frame,
and, in total, 117 building contractors were randomly selected from the sampling
frame. All these randomly selected contractors were contacted via telephone or E-mail
to request their participation in this study. During the call or Email, an opportunity to
interview the project manager or safety officer in any of the company’s completed
building projects was requested. In the second stage, one to three projects from each
contractor prepared to participate in this study were selected as part of the sample
based on the inclusion criteria. The criteria to select projects for this study include: (1)
the projects should have been completed within the past three years; and (2) the safety
personnel or the project managers, such as safety managers, safety officers or project
managers, must be willing to participate in this study.
Face-to-face interviews were conducted upon being granted the opportunity to
interview the project manager or safety officer. The interviewees were requested to
recall or review the historical records of the project, or provide their estimation
whenever the records were unavailable to complete the questionnaire. In most cases,
the project manager or safety officer did not answer all the questions by
himself/herself. Thus, he/she had to consult other project personnel such as quantity
surveyors and safety supervisors, or the personnel in the head office who were in
charge of Workplace Safety and Health (WSH) issues during the interview.
Data Sample Characteristics
Out of the 117 contractors contacted, 23 participated in this study representing a
response rate of 20%. The relatively lower response rate may impact the
representative nature of the contractors selected. However, this was minimized by the
stratified sampling method and the random selection process (see Data Collection
Section). The distribution of the 23 contractors is shown in Table 2. The response rate
ranges from 14% to 29% among different BCA grades. Table 2 shows that the
response rate of large contractors (grade A1 and A2) is higher than that of smaller
85
Costs of Construction Accidents to Singapore Contractors
contractors (grade B1 and B2). The grade B2 contractors have the lowest response
rate (14%) among the four grades. The relatively lower response rate in grade B1 and
B2 contractors may be attributed to the fact that a considerable part of their contracts
are subcontracts, especially for B2 contractors. Thus, it is possible that there were no
building projects completed by some small companies as the main contractor within
the past three years.
Table 2: Distribution of Contractors
BCA Grade
A1
A2
B1
B2
Tot al
Sampling frame
18
14
28
57
117
Sample contractors
5
4
6
8
23
Response rate
28%
29%
21%
14%
20%
The 23 contractors provided information from 47 completed building projects. The
distributions of the sample projects are shown in Table 3. The types of projects
comprise commercial building (10.6%), residential building (63.8%), office building
(12.8%) and industrial building (12.8%). The contract value of most projects (83%)
ranges from SGD 10 million to SGD 100 million. 83 percent of the projects are from
private sector, and 17 percent are from public sector. The sample projects are evenly
distributed among the four BCA grades. The profile of the projects suggests that the
data were collected from a wide range of building projects with a focus on residential
(63.8%), middle-size (83%) and private building projects (83%). The interviewees
consist of 42 project managers and 5 safety officers. Most of the interviewees are
project managers, and have more than 10 years of experience in the construction
industry. The average working experience of the interviewees was 13 years, and the
minimum working experience was 7 years.
Table 3: Characteristics of Sample
Profile
Number
Percentage*
Project Type
Commercial building
5
10.6
Residential building
30
63.8
Office building
6
12.8
Industrial building
6
12.8
Project Size (Singapore Dollars)
Up to $10 mil
5
10.6
> $10 mil ≤ $50 mil
29
61.7
> $50 mil ≤ $100 mil
10
21.3
> $100 mil
3
6.4
Type of Client
Private
39
83.0
Public
8
17.0
Height of Building
Up to 5 stories
16
34.0
> 5 ≤ 10 stories
13
27.7
> 10 ≤ 15 stories
8
17.0
More than 15 stories
10
21.3
Firm’s BCA grade
A1
12
25.5
A2
12
25.5
B1
11
23.4
B2
12
25.5
*Rounding-off error may have occurred.
86 Teo and Feng
RESULTS AND DISCUSSION
Estimation of Accident Cost for Building Projects
The Total Accident Cost (TAC) is the sum of Insured Accident Cost (IAC) and
Uninsured Accident Cost (UAC) for a project. Three dimensionless quantities, the
Total Accident Cost Ratio (TACR), Insured Accident Cost Ratio (IACR) and
Uninsured Accident Cost Ratio (UACR) were used to enable comparison of the level
of accident cost among projects of different sizes. TACR, IACR and UACR were
therefore defined as follows:
%100
)( X
mContractSu
TACentCostTotalAccid
TACR =
%100
)( X
mContractSu
IACidentCostInsuredAcc
IACR =
%100
)( X
mContractSu
UACtccidentCos
UninsuredA
UACR =
where, TAC = IAC + UAC, IAC is the insurance premium paid by the project, and
UAC is the sum of the following 14 uninsured accident cost items:
Compensation for the injured worker that are not covered by insurance policy
(UC1);
Lost productivity due to the injured worker (UC2);
Lost productivity due to crew of injured worker (UC3);
Lost productivity due to other workers in vicinity of accidents (UC4);
Losses due to replacement of the injured worker (UC5);
Lost productivity due to the investigation or inspections as a result of the injury
(UC6);
Cost of supervisory or staff effort (UC7);
Losses due to damaged equipment or plant, property, material or finished work
due to the accident (UC8);
Cost of transporting injured worker (UC9);
Consumption of first-aid materials in this accident (UC10);
Additional work required as a result of the accident (e.g. cleaning, additional
barriers and so on) (UC11);
Fines and legal expenses (UC12);
Losses due to Stop Work Orders (SWO) issued to the project (disruption of
schedules) (UC13); and
Additional benefits to the injured worker beyond the Work Compensation Act
(WCA) (UC14).
The survey result shows that the average insured accident costs of the 47 building
projects were 0.15% of contract sum and the average uninsured accident costs were
0.1% of contract sum. The survey result further reveals that not all the above 14 cost
items were encountered by each of the accidents. Figure 1 shows how often the 14
cost items were encountered in connection with the 168 MOM reportable accidents
collected. It was found that there are large variations in the frequency in which these
items were involved in the 168 accidents. As shown in Figure 1, the items with
87
Costs of Construction Accidents to Singapore Contractors
relatively lower incidence rate (less than 30% of total accidents) include:
compensation for the injured worker that are not covered by insurance policy (UC1);
lost productivity due to other workers in vicinity of accident (UC4); fines and legal
expenses (UC12); losses due to SWO issued to the project (UC13); and additional
benefits to the injured worker beyond WCA (UC14). The items with relatively higher
incidence rate (more than 70% of total accidents) include: lost productivity due to
crew of the injured worker (UC3); cost of supervisory or staff effort (UC7); cost of
transporting injured worker (UC9); and consumption of first-aid materials (UC10).
Figure 1: Occurrence of Uninsured Accident Cost Items
Total accident costs (TAC) of building projects comprise the insured accident costs
(IAC) and uninsured accident costs (UAC). Among the 47 building projects examined,
there is a large variation in TACR, which ranges from 0.12% to 0.83%. This variation
is largely the result of variations in safety performance of the 47 building projects.
The average total accident costs for contractors account for 0.25% of total contract
sum of a building project in Singapore. This result indicates that accidents are costly
and substantial savings could be made by prevention. This result is not much different
from Tang et al.’s (1997) study, where the average accident loss ratio (equivalent to
TACR in this study) was 0.31% of contract sum of a building project in Hong Kong.
The difference between the two figures (i.e., 0.25% and 0.31%) may be explained by
the differences in compensation required by legislation, wage level, and price level
between Singapore and Hong Kong.
Estimation of Uninsured to Insured Accident Costs Ratio
Based on the definition, insured costs of accidents are explicit and easily ascertained
by employers. However, the uninsured costs remain, for the most part, either hidden
UC14
UC13
UC4
UC12
UC6
UC5
UC2
UC8
UC11
UC10
UC3
UC
7
UC9
UC1
Incidence rate between 30% and 70%
Incidence rate >70%
Incidence rate <30%
Legend :
88 Teo and Feng
or attributed to other accounting ledgers (Brody et al., 1990). To show the magnitude
of uninsured accident costs to employers, the ratio between insured accident costs and
uninsured accident costs of building projects was examined.
In this study, the average insured accident cost was estimated to be 0.15% of the total
contract sum of a building project, and the average uninsured accident cost was
estimated to be 0.1% of the contract sum. Thus, a ratio between insured accident cost
and uninsured accident cost was obtained, in the order of 1.5:1. This result shows that
the uninsured accident costs account for about 67% of the insured accident costs. It
suggests that the uninsured accident costs are substantial for building projects.
Therefore, the focus on the perceived or explicit costs of accidents fails to show the
“true reality” of accident costs. This finding is consistent with the findings of many
studies (Head and Harcourt, 1997; Everett and Frank, 1996; Hinze, 1991; Leopold
and Leonard, 1987; Heinrich, 1931) that the indirect (uninsured) accident costs are
significant and should be paid attention.
However, there is no generally accepted ratio between direct and indirect costs of
accidents. For example, the ratios between direct (insured) and indirect (uninsured)
costs suggested by various authors vary between 1:4 (Heinrich, 1931), 1:2 (Simonds
and Grimaldi, 1956), 1: 1.6 (Levitt, 1975), 1: 1 (Rinefort, 1976), to 1:0.3 (Laufer,
1987). The variation of these results may be explained by the following reasons:
Firstly, there exist different definitions and components of direct and indirect accident
costs, or insured and uninsured accident costs.
Secondly, the direct or insured accident costs vary greatly with the different work
injury compensation and insurance policies in different countries/regions.
Thirdly, since most of indirect or uninsured costs represent those intangible costs, the
data collected for indirect or uninsured costs may not be as reliable as those direct or
insured costs. The accuracy of the data depends largely on the quality of the survey
and estimation methods of estimation.
Finally, the studies which came out with a “magic” ratio between direct (insured)
costs and indirect (uninsured) costs of accidents were conducted in different industries
such as construction, manufacturing, chemistry, and forestry. The nature of different
production systems in different industries may also explain part of the variation in the
cost ratio.
Factors Influencing the Uninsured Accident Costs
Pearson correlation coefficient was used to test whether the uninsured accident costs
(UAC) of building projects are influenced by project characteristics, such as project
size (PS), company size (CS), project duration (PD), height of building (HB), and
percentage of work completed by sub-contractors (SUB). The results of correlation
analysis presented in Table 4 show that the uninsured accident costs were significantly
(p<0.05) associated with company size (r=0.3), project size (r=0.389), and percentage
of work completed by subcontractors (r=0.344).
89
Costs of Construction Accidents to Singapore Contractors
Table 4: Correlations between Variables
CS
PS
PD
SUB
HB
UAC
CS
PS
PD
SUB
HB
UAC
1.000
.593**
1.000
.621**
.630**
1.000
.561**
.517**
.397**
1.000
.331*
.376**
.416**
.275
1.000
.300*
.389*
.066
.344*
.248
1.000
** p<0.01; *p<0.05
The results of correlation analysis (see Table 4) show that the uninsured accident costs
are significantly (p<0.05) and positively related to company size (r= 0.300), which
was measured by the company’s BCA Grade. This finding indicates that more
uninsured costs would be caused by accidents in larger contractors. It is in line with
the findings of other studies. The impacts of company size on the size of total safety
costs were demonstrated in the research by Rinefort (1976), who investigated and
compared the quantitative effects of safety control activities on work injury costs in
large-size, medium-size, and small-size companies. The results of Rinefort’s (1976)
study indicated that the variation of the effects of safety control activities on work
injury costs could partly be explained by differences in company size. The argument
by Rikhardsson and Impgaard (2004) further supports the influences of company size
on uninsured accident costs: “in larger companies the Occupational Health and
Safety department is a staff function manned with a number of specialists and
secretaries and functions under numerous policies, rules and regulations. Thus, when
an accident occurs in larger companies more formal activities are initiated than in
smaller companies. There are more people involved, there are more internal
administrative processes that have to be complied with and more organizational levels
have to be informed.” (p. 179)
Company Size
The results also (see Table 4) show that the uninsured accident costs are also
positively correlated with project size (r= 0.389, p<0.05), which was measured by the
contract sum of the project. This result suggests that the uninsured accident cost tends
to be higher in larger projects. This finding is consistent with Hinze’s (1991) study,
which found that the cost ratios between indirect and direct costs tend to increase with
project size. Hinze (1991) further explained that “larger projects generally employ
greater numbers of workers resulting in work being performed in more crowded
conditions. An injury would naturally be expected to have a broader indirect cost
impact on a larger project. Larger projects are also associated with deeper hierarchy
structures in which greater numbers of administrative and supervisory personnel
become involved with injury reporting and accident investigations. It can be
concluded that project size does have a significant role in influencing the cost ratios of
injuries (p. 9-10)”.
Project Size
As shown in Table 4, the uninsured accident costs were found to be positively related
to the percentage of work completed by subcontractors (r= 0.344, p<0.05). This result
suggests that the more work completed by subcontractors, the higher the uninsured
Percentage of Work Completed by Subcontractors
90 Teo and Feng
accident costs. The involvement of more employers in the construction site may
explain some of the variations. The involvement of more subcontractors in the project
tends to increase the levels of management. It seems that, in a construction site with
more subcontractors, more people will be involved in administration, communication,
investigation and inspection processes when an accident occurs. Thus the costs
incurred in these processes due to the occurrence of an accident tend to be relatively
higher if subcontractors undertake more work.
Moreover, the influence of the percentage of work undertaken by subcontractors on
the magnitude of uninsured accident costs could be partly explained by the findings of
Hinze (1991) that the cost ratio between uninsured and insured costs varies with
different types of contract between main contractors and subcontractors. Hinze (1991)
made a comparison of the cost ratios between uninsured and insured costs on different
contract types such as lump sum contracts and cost reimbursable contracts and found
that in terms of medical case injuries, the cost ratios are significantly higher on cost
reimbursable contracts. Hinze (1991) argued that a poorly managed cost reimbursable
contract provides an inherent incentive to increase costs.
CONCLUSION
This study examined the financial costs of accidents to building projects in Singapore.
The results show that the average insured accident costs, uninsured accident costs, and
total accident costs of building projects account for 0.15%, 0.1%, and 0.25% of
contract sum, respectively. This finding implies that the accident costs are substantial
for contractors and that potential savings could be achieved by preventing the
occurrence of construction accidents.
The uninsured cost items incurred by construction accidents were examined. The
results show that there were large variations in the incidence of each cost item caused
by the accidents. Lost productivity due to the crew of the injured worker, cost of
supervisory or staff effort, cost of transporting the injured worker, and consumption of
first-aid materials in the accident were among the highest incidence items of the
uninsured accident costs.
Furthermore, a ratio between the insured and uninsured accident costs for building
project in Singapore was obtained to be 1.5:1. This finding implies that the uninsured
accident costs are substantial for building projects and should be paid much attention
to. The comparison between this result and previous studies suggests that the there is
no constant ratio between direct (or insured) and indirect (or uninsured) costs of
accidents. The ratio may vary with different industries, countries/regions, companies,
projects, and definitions of direct (or insured) and indirect (or uninsured) accident
costs.
Finally, the factors that may impact the size of uninsured accident costs were
examined. Three factors were identified to impact the indirect accidents cost. The
results imply that more uninsured cost could be incurred in larger companies, larger
projects, and in projects with more work undertaken by subcontractors.
In light of this study, Singapore contractors may be more conscious about the
91
Costs of Construction Accidents to Singapore Contractors
financial implications of construction accidents, which would in turn motivate the
contractors to initiate efforts to prevent construction accidents. However, two
limitations of this study need to be highlighted. The first limitation lies in the ability
to generalize these findings. The findings were reached based on the information of
47 building projects from 23 building contractors. The profile of the projects (see
Table 3) suggests that the data were collected from a wide range of building projects
but with a focus on residential (63.8%), middle-size (83%), and private building
projects (83%). The findings are grounded in such data and hence generalizations to
other populations may need meticulous interpretation.
Secondly, the losses of contractors due to construction accidents were confined to the
financial losses in this study. Other ancillary costs arising from the accident which are
difficult to quantify, such as damage to company reputation and morale of employees
were not included in the indirect cost of accidents. The size of uninsured accident
costs would be substantially increased if these ancillary costs were considered. Thus,
future studies could be conducted to investigate the ancillary costs such as damage to
company reputation, damage to the relationship with the clients, damage to the morale
and job satisfaction of workers, and so on.
ACKNOWLEDGMENT
The authors gratefully acknowledge the National University of Singapore for
providing funding for this research effort (RP296-000-130-112).
REFERENCE
Bird, F.E. (1974). Management Guide to Loss Control. Atlanta: Institute Press.
Brody, B., Letourneau, Y. and Poirier, A. (1990). An indirect cost theory of work
accident prevention. Journal of Occupational Accidents, 13, 255-270.
Davies, N.V. and Teasedale, P. (1994). The Costs to the British Economic of Work
Accidents and Work-Related Ill-Health. Health and safety Executive, HMSO,
London.
Dorman, P. (2000). The economics of safety, health and well-being at work: an
overview. In Focus Program on SafeWork, International Labor Organization. The
Evergreen State College. From http://www.ilo.org.
EU OSHA (2001). EU European Agency for Health and Safety at Work. Economic
Impact of Occupational Safety and Health in the Member States of the European
Union. European Agency for Safety and Health at Work.
Everett, J.G. & Frank Jr., P.B. (1996). Costs of accidents and injuries to the
construction industry. Journal of Construction Engineering and Management.
122(2), 158-164.
Head, L. and Harcourt, M. (1997). The direct and indirect costs of workplace
accidents. Proceedings of the 11th AIRAANZ Conference, 30 January – 1 February
1997, Brisbane, Australia.
Heinrich, H.W. (1931). Industrial Accident Prevention. New York: McGraw-Hill.
Hinze, J. (1991). Indirect Costs of Construction Accidents. A report to the
Construction Industry Institute, The University of Texas at Austin.
Hinze, J.W. (1997). Construction Safety. New Jersey: Prentice-Hall.
Laufer, A. (1987). Construction accident cost and management safety motivation.
92 Teo and Feng
Journal of Occupational Accidents, 8, 295-315.
Leopold, E. and Leonard, S., (1987). Costs of construction accidents to employers.
Journal of Occupational Accidents, 8, 273-294.
Levitt, R.E. (1975). The Effect of Top Management on Safety in Construction.
Technical Report No. 196, The Construction Institute, Stanford University.
Levitt, R.E., Parker, H.W. and Samuelson, N.M. (1981). Improving Construction
Safety Performance: The user's role. Technical Report No. 260, Stanford
University.
Lingard, H. and Rowlinson, S. (2005). Occupational Health and Safety in
Construction Project Management. New York: Spon Press.
MOM (Ministry of Manpower) (2006). Workplace Safety and Health (Incident
Reporting) Regulations 2006.
http://www.mom.gov.sg/Documents/safety-health/WORKPLACE%
20SAFETY%20AND%20HEALTH%20_INCIDENT%20REPORTING_%20REG
ULATIONS%202006.pdf. (accessed in July 2008).
Monnery, N. (1999). The costs of accidents and work-related ill-health to a cheque
clearing department of a financial services organization. Safety Science, 31,
59-69.
Riel, P.F. and Imbeau, D. (1996). Justifying investments in industrial ergonomics.
International Journal of Industrial Ergonomics, 18, 349-361.
Rikhardsson, P.M. and Impgaard, M. (2004). Corporate cost of occupational
accidents: an activity-based analysis. Accident Analysis and Prevention, 36,
173-182.
Rinefort, F.C. (1976). A Study of Some of the Costs and Benefits Related to
Occupational Safety and Health in Selected Texas Industries. Ph.D. Thesis,
Texas A&M University.
Simonds, R.H. and Grimaldi, J.V. (1956). Safety Management: Accident Cost and
Control. Richard D. Irwin, Homewood, Ill.
Tan, W. (2008). Practical Research Methods (3rd ed.). Pearson Prentice Hall.
Tang, S.L., Lee, H.K. and Wong, K. (1997). Safety cost optimization of building
projects in Hong Kong. Construction Management and Economics, 15, 177-186.
Teo, A.L. and Feng, Y. (2010). The moderated effect of safety investment on safety
performance for building projects. The International Journal of Construction
Management, 10 (3), 45-61.
U.S. Department of Labor, Bureau of Labor Standards. (1955). Safety Subjects.
Bulletin No. 67, U.S. Government Printing Office, Washington, D.C..
