International Journal of Injury Control and Safety

Promotion

ISSN: 1745-7300 (Print) 1745-7319 (Online) Journal homepage: https://www.tandfonline.com/loi/nics20

Cost of two-wheeler road accidents in India

Sivakumar Balakrishnan & Krishnamurthy Karuppanagounder

To cite this article: Sivakumar Balakrishnan & Krishnamurthy Karuppanagounder (2019): Cost of
two-wheeler road accidents in India, International Journal of Injury Control and Safety Promotion,
DOI: 10.1080/17457300.2019.1576208

To link to this article: https://doi.org/10.1080/17457300.2019.1576208

Published online: 22 Feb 2019.

Submit your article to this journal

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=nics20
INTERNATIONAL JOURNAL OF INJURY CONTROL AND SAFETY PROMOTION
https://doi.org/10.1080/17457300.2019.1576208

Cost of two-wheeler road accidents in India

Sivakumar Balakrishnan and Krishnamurthy Karuppanagounder
a
Traffic and Transportation Planning, Department of Civil Engineering, National Institute of Technology, Calicut, Kerala, India

ABSTRACT ARTICLE HISTORY

Economic loss due to traffic accidents is noticeable in India; however, the country lacks research works Received 4 March 2018
to estimate the magnitude. This study is an attempt through a Stated Preference (SP) survey to know Accepted 27 January 2019
how users value safety and how much they are willing to pay (WTP) for a safer road facility to reduce
KEYWORDS
the risk of their involvement in road traffic accidents. The survey was conducted for two-wheeler riders
Willingness to pay; stated
of selected roads in Calicut City. Attributes selected for SP experimental design includes travel time, preference survey; cost of
travel cost and traffic accidents per year and Binary Logit model was used to determine the WTP val- accidents; binary
ues. It was observed that travel cost, accident rate of the route selected, age, occupation, personal logit model
income and number of household members have significant impact on the decision-making process.
WTP value of a road accident for a two-wheeler user was found to be Rs 0.53/person/trip.

Introduction Ministry or Road Transport and Highways (MoRTH,

2017), India reports the occurrence of about 150,785 traffic
Road accidents are one of the most serious impacts of road
collision-related deaths in the year 2016, which stands the
mobility throughout the world, especially in developing
highest road traffic accident fatality rate worldwide. This
countries. It has been observed in more advanced countries would translate, on an average, into 120 persons killed per
that a global awareness about the financial impact of road million populations in the year 2016 (WHO, 2015). The
accidents have resulted in an increased concern on the effect accident severity has gone up to 29.1 persons killed per 100
of road accidents. This helped in enhancing discussions or accidents in the year 2016. MoRTH (2014) also states that
politics in favour of road safety and thus making the use of cost of road accidents in India come to about 4.7% of GDP
the value of road accident reduction, a key input for evalu- for the year 2009–2010. This shows that road safety is a for-
ation of almost all transportation infrastructure projects and midable challenge for the country. The economic costs of
road safety improvement schemes. traffic accidents turn out to be so high that it becomes eas-
The highest number of road traffic fatalities can be seen ier to justify higher expenditures in promoting road safety.
in the middle-income countries. It is reported that 1.25 mil- Road accidents result in social costs like material damages,
lion people are killed on the world’s roads in a year and medical costs, and loss of productivity and reduce the qual-
fatality (mortality) rates in developing countries are higher ity of life. Information about the cost of traffic accidents
in comparison to those in the developed countries (World will help to evaluate the effectiveness of road safety pro-
Health Organization [WHO], 2015). More than 91% of glo- grams as well as in the proper allocation of the country’s
bal traffic deaths occur in the developing world which inci- limited financial resources for the most effective road acci-
dentally has only 48% of the world’s registered vehicles. dent prevention plans.
Deaths/injuries are critical public health issues which impart The risk value can also be said to reflect how much an
a high cost to a nation through productive life years lost individual is willing to pay to reduce his/her risk to be
(due to premature death and long-term disability) and med- injured or killed in a road accident. One of the powerful
ical resources used (for care-treatment, rehabilitation of approaches in valuing road safety is the Willingness to Pay
injured persons and in pain and suffering of the injured per- approach (WTP). This method helps to understand the pub-
sons, their families and friends). Global Status Report on lic attitude towards road safety and also their willingness to
Road Safety indicates that road accident cost varies from pay to reduce road accidents. WTP is the maximum amount
0.3% of Gross National Product (GDP) in Vietnam to of money an individual is willing to pay to reduce his/her
almost 5% GDP in the USA, which are considerably high risk of premature death or injury from a road accident. The
when compared to the total development aid received by the main assumption of the WTP approach is that individuals
low-income countries (WHO, 2015). Another estimate from are the best judges of their own welfare; hence individual
Organisation for Economic Co-operation and Development preferences must be taken into account in situations involv-
(OECD) countries shows that accident costs come to about ing life and death. WTP for risk reductions will act as a key
2% of GDP (Transport Research Laboratory [TRL], 1995). input in the cost-benefit analysis of road safety improvement

CONTACT Sivakumar Balakrishnan [email protected]

ß 2019 Informa UK Limited, trading as Taylor & Francis Group
2 S. BALAKRISHNAN AND K. KARUPPANAGOUNDER

schemes. Estimation of WTP for accident risk reduction helps to pain and suffering invites a great deal of debate, high-
us to know, how far the people are concerned about safety income countries have been estimating road traffic accident
and the amount they are willing to pay for the reduced risk costs for decades. Various methods that can be used to cost
of road accidents. road traffic accidents have been well documented by the
Studies show that WTP method can produce reasonable TRL (1995). Hills and Jones-Lee (1981) argue that among
estimates of the cost of accidents and thus help to address these methods only the Human Capital (HC) method and
the associated problems (de Blaeij, Florax, Rietveld, & the WTP method appear to be directly relevant. But at the
Verhoef, 2003; Dionne & Lanoie, 2004; Miller, 2000). same time, TRL recommends Gross Output method to cost
Swedish National Road Administration uses Value of road accidents in developing countries as the WTP method
Statistical Life (VOSL) based on WTP approach for estimat- is found to be difficult because it is based on the completion
ing the traffic crash costs (de Blaeij et al., 2003). VOSL is of complex questionnaires relating to perceived risk and
obtained as the product of the marginal value of WTP and payment by individuals to avoid a given hypothetical level
inverse of the risk reduction. Rizzi and de Dios Ort
uzar of risk.
(2003) proposed a stated choice survey, wherein individuals However, in order to monetize the preferences for acci-
are asked to choose among different alternatives, the attri- dent reduction, it should be valued by what a person would
bute levels of which vary according to a statistical design pay for it or it should be expressed as the aggregate of indi-
aimed at maximising the precision of the estimates. The sur- viduals’ willingness to pay for the safety improvements.
vey was conducted among car users of Chile, route choice Hence, for the successful implementation of transportation
experiment was given to the respondents with the attributes safety policies and measures, public preferences towards
like travel time, toll charge and level of risk. Stated choice safety and their readiness to pay for road safety improve-
allows the analyst to mimic actual choices with a high ment schemes have to be considered. This strengthens the
degree of realism, and many studies used this method for need for willingness to pay concept in the costing of road
the valuation of intangibles (de Blaeij, Rietveld, Verhoef, & accidents. Studies show that pain, suffering and lost quality
Nijkamp, 2002; Hojman, Ort
uzar, de D, & Rizzi, 2005; of life for fatalities are best valued using WTP approach
Irag€
uen, & de Dios Ort
uzar, 2004; Louviere, Hensher, & (Ainy, Soori, Ganjali, Le, & Baghfalaki, 2014; Andersson,
Swait, 2000; McFadden, 1998). 2011; Haddak, Havet, & Lefvre, 2014; Irag€ uen & de Dios
This study attempts to examine the suitability of extend- Ort
uzar, 2004).
ing the scope and quality of WTP method for calculating Elvik (1995) mentioned that the cost of an accident
road traffic accident costs in developing countries like India, should reflect the user willingness to pay for reducing their
where there is a large potential for traffic growth and where risk of involvement in a road accident and so WTP method
road traffic accidents are severe. This article reports the should be used for road cost evaluation. Hills and Jones Lee
results of a Stated Preference (SP) survey conducted in (1981) stated that WTP method is more appropriate for
Calicut, India that was designed to provide estimates of the social welfare projects and the value of safety may be esti-
WTP to reduce the chances of being a road accident victim. mated as the amount that individuals are ready to pay for
Given the high proportion of motorcycle accidents in the improvements of road safety. Rizzi and de Dios Ort
uzar
country (28.8% as reported in MoRTH, 2017), the authors (2003) proposed a different approach based on the SP
contribute to the existing literature by analysing the value of method of WTP approach on car drivers in Chile to value
the life of motorcyclists. This study also evaluates how interurban road safety. Respondents were given different
socio-economic factors affect the willingness of motorcyclists route choices based on different levels of three main attrib-
to pay to avoid involvement in accidents and risk of death. utes: travel time, toll charge and level of risk. Their experi-
The objectives of the study may be listed as follows: ment showed that people were sensitive to the risk variable,
thus ‘stating a preference’ for safer routes. Iraguen and

Explore the applicability of WTP concept and to identify Ortuzar (2004) presented WTP for reducing fatal accident
the factors affecting WTP for road-accident risk reduc- risk reduction in urban areas, through an internet-based
tion in developing country like India. web page SP survey. A series of route choice situations

Estimate WTP value for two-wheeler road safety in based on travel time, cost and number of fatal road acci-
Indian context. dents per year were given to the respondents. He found that

Examine the variation of WTP for a reduction in road- WTP increases with the reduction in both time and
mortality risk with socio-economic characteristics. road accidents.
Results of various studies (Andersson, 2011; Haddak et al.,
2014; Hensher, Rose, Ort
uzar, & Rizzi, 2009; Louviere et al.,
Literature review
2000; McFadden, 1998) put in light the significant and posi-
Costing of the road traffic accidents is essential to provide a tive influence of the SP survey on costing of roadway acci-
guideline on the amount of resources to be spent on the dents. As such, the WTP method allows the analyst to mimic
prevention of road traffic accidents or injuries, even though, actual choices with a high degree of realism and hence it can
assigning a monetary value to human life (life-saving or to be taken that it is an appropriate elicitation method for the
the quality of life) is often considered meaningless and eth- valuation of intangibles. Some WTP studies have been con-
ically wrong. Despite the fact that putting monetary values ducted in developing countries too, to estimate accident costs
 INTERNATIONAL JOURNAL OF INJURY CONTROL AND SAFETY PROMOTION 3

(Bhattacharya, Alberini, & Cropper, 2006; Chaturabong, this study, an attempt is made to construct a scenario that
Kanitpong, & Jiwattanakulpaisarn, 2011). reflects the profile of road accidents in developing countries,
Only a few attempts were made in India to estimate the where motorcyclists bear the worst part of road fatalities.
costs of road traffic accidents and most of them were based
on simple economic models. In one of the earlier studies
Srinivasan, Hingorani, and Sharma (as cited in Mohan, Methodology
2002) evaluated road accident costs for Delhi city in 1968 Data required for this project were broadly classified as
and another study by Natarajan (as cited in Mohan, 2002) socio-economic data, travel related data and accident-related
calculated road traffic accident costs for Chennai (Madras) data (Andersson, 2007; Bhattacharya et al., 2006; Haddak
in 1978 based on Insurance data. The first major road user et al., 2014). Accident related data were collected from
cost study published in India in 1982, sponsored by the police records for the period 2014–2017 and face to face
World Bank, included road accidents as a component of interview surveys were conducted for getting the socio-eco-
road user cost (Central Road Research Institute [CRI], nomic and travel-related information. Traffic accident
1982). The costs components were medical expenses, legal related information from police may not be exhaustive as
fees, property damage, insurance costs and loss of output numerous minor accidents goes unreported or victims have
due to death. In all these studies the life expectancy at birth compromising within them. This is a limitation of the acci-
assumed and fatality-injury ratio used were far away from dent data availability in developing countries. The study area
the current situation. Vehicle damage costs were also under- selected was Calicut city in Kerala, India. The surveys were
estimated as most of the ‘minor injury’ and ‘damage only conducted in three main roads of Calicut which includes
road accidents’ have not been recorded perfectly in India. Kunnamangalam – Civil Station route along NH212 (11 km)
Indirect costs such as life cost, permanent disability cost and two major district roads namely, Medical College –
and psychological damage, pain and socio-economic effects Arayidathupalam road (7 km) and Medical College – Mavoor
were not taken into account in all these studies. At the Road (13 km).
same time Bhattacharya et al. (2006), employed stated pref- Socio economic and travel related aspects such as number
erence approach using contingent valuation (CV) to elicit of dependants in the family, marital status, personal monthly
the WTP for reductions in the risk of death in road acci- income, age of the respondent, mode of travel, and so forth,
dents from a sample of pedestrians and two-wheelers in which proved to be affecting the decision making process for
Delhi, India. In their study, the respondents were asked WTP, were included in the questionnaires (Andersson, 2007;
what they would pay to reduce their risk of dying. Bhattacharya et al., 2006; Haddak et al., 2014).
Respondents were given a payment card to select a value or
made to state a bid other than what was shown on the card.
Stated preference survey
Whittington (1998) identified some difficulties of using the
CV method in developing countries. He found that it was SP experiment was designed as a binary route choice for the
not easy for the survey respondents to understand the eco- two-wheeler users. Attributes selected for the SP survey
nomic concept of the valuation process in terms of willing- were finalized based on a pilot survey. The pilot survey data
ness and the ability to pay to reduce their risks. was also used for knowing the suitability of the selected
Establishment of suitable referendum prices is also to be attributes and their levels and also to check whether the
taken care of seriously as the highest price is not too low selected attributes were perceivable to the users. To reduce
and the lowest price is not too high. the complexity of SP experimental design the number of
In a very recent project SafetyCube, authors compared scenarios presented to the respondents was reduced. In the
the WTP method with two alternative approaches: the choice experiment, the attributes selected were travel time,
QALY (Quality Adjusted Life Years, which is based on travel cost and the number of road accidents per year. The
WTP values) approach and the court award approaches SP choice was used to explore preferences of two-wheeler
(Schoeters et al., 2017). They state that compensation pay- for combinations of different levels of travel time, travel
ments to road injuries awarded by courts in most cases are cost and accidents risk reductions. The final SP question-
much lower than the values obtained in the other methods, naire had socio-economic details, travel details and acci-
and are characterized by a huge unpredictability since they dent-related details (past history of road accidents). The
are highly dependent on the judicial system. They also add main survey was conducted for the two-wheeler users of the
that these court award values are not based on individual three road stretches viz., Kunnamangalam – Civil station
preferences, and they are determined ex-post and apply to a route along NH212, and two major district roads namely,
specific case, they cannot be used in a cost-benefit analysis. Medical college – Arayidathupalam road and Medical col-
They conclude that the WTP method is rather limited in lege – Mavoor road.
the specification of injuries and they insist the use of QALY Attribute levels of travel time and travel costs were
instead of WTP for getting values of a larger diversity of selected as a variation of þ25% and 25% from the actual
injury types. values in those routes. The levels selected for accident
In short, the vast majority of WTP studies in the context reduction was 25% and 50% from the actual number of
of road safety have been conducted in high-income coun- road accidents. The attributes and levels chosen are shown
tries; few have been conducted in developing countries. In in Table 1.
4 S. BALAKRISHNAN AND K. KARUPPANAGOUNDER

Table 1. Attributes and levels. binary logit model was developed and the WTP values were
Attributes Level 1 Level 2 estimated. Assume a route is used by N users. If person n
Travel time Reduced by 25% Increased by 25% travels more than once in a reference period, say mn times,
Travel cost Increased by 25% Reduced by 25%
Road accident rate Reduced by 25% Reduced by 50%
this gives rise to mn pseudo-members with a total popula-
tion equal to observations (Hensher et al., 2009). This popu-
lation exactly amounts to the flow on a route in a given
Table 2. Choice set 1. period (say a year). The WTP estimates are a ‘per person
Attributes Presently used mode Alternative per trip’ valuation. To obtain the value of a reduction in
Travel time (minutes) TT 0.75 TT risk of one fatality we have to multiply the WTP per person
Travel cost (Rs) TC 1.25 TC per trip by the total number of person trip generated along
Road accident rate CR 0.75CR
the particular route. Mode-wise annual cost of avoiding a
traffic accident may be estimated based on the ADT value,
There are three attributes with two levels each and hence the number of people travelled mode wise, and per capita
there are a total of 23 combinations, that is, eight combina- trip rate along the route.
tions. Out of the eight treatment combinations, the two
combinations showing (0.75TT, 0.75TC, 0.75CR) and Results and discussions
(0.75TT, 0.75TC, 0.5CR) choices were removed being dom-
inant and finally six alternatives were selected for preparing Detailed discussions on the preliminary analysis, binary logit
SP choice set. These two options showing extreme positive model analysis are given in the following sections.
conditions of reduced travel time, lower travel cost and
reduced traffic accident rates are dominant. Thus, there is 6 Preliminary analysis
choice sets each with two alternatives, one the existing alter-
native and the other, different from the existing alternative Preliminary analysis of questionnaire data and accident data
(A sample choice set is given in Table 2). Finally, an option was carried out to explore socio-economic characteristics, trip
for ranking of the alternatives including the present alterna- characteristics of the sample collected and to get the road acci-
tive was given. The purpose of this was to ensure the con- dent rates of the selected roads. The annual average accident
sistency of the choices made by the respondents. rates obtained from FIR reports for the selected roads are 52,
22, and 7 per million populations for Kunnamangalam-Civil
station road, Medical College-Chevayur road and, Medical
Data collection College-Mavoor road respectively.
Socio-economic characteristics of the respondents include
Face to face interview method was adopted for the survey.
gender, age, marital status, occupation, education, monthly
The stratified sampling method was used for sampling. A
personal income, household size, household monthly
total of 467 two-wheeler drivers were approached for data
income and so forth. About 79% of the sample was male
collection, but only 396 were ready to give the data
(271 out of 344). The two-wheeler respondents mainly were
(84.80%). From this, 52 samples were removed as the
aged between 24 and 40 years (43%) representing the
responses were found to be inconsistent. Inconsistency was
younger age group. About 88% of two-wheeler users
in terms of mismatching of the choices and the rankings
declared that they are married. Nearly 64% of the two-
given by the respondents. Finally, 344 useful samples were wheeler riders claimed that they live in urban area. A great
obtained. On an average 30 min were taken for the survey per majority, 84.5%, said that they are highly worried about the
person. The data was collected from households, offices, shops road accidents and feel insecure about the road accidents;
and some nodal points along the selected road stretches. hence they are ready to pay for the road safety improvement
To clearly convey the hypothetical alternatives with schemes that would decrease their risk of involvement in
reduced time, accident rates and increased cost some exam- the road accident.
ples were given to them like improved roads, grade sepa- Almost 46.8% of the two-wheeler riders had experiences
rated crossroads, flyovers and so forth. The collected data of road accident either directly or indirectly (of family or
was entered in spreadsheets and the preliminary analysis friends). The minimum, maximum and average household
was done. The variables were coded and labelled prior to sizes for the sample are 1.0, 8.00 and 4.22. About 26.8% of
data entry. the riders have monthly personal income between Rs.
10,000 and 20,000. Only 0.5% the riders claimed a monthly
Discrete choice modelling and estimation of cost income above Rs. 2 lakh. The monthly household income
of accident category for 32.7% two-wheeler users was Rs. 20,000 – Rs.
40,000. It can also be noted that 22% of the two-wheeler
One of the most popular and widely used Logit Model users are graduates or postgraduates. A majority (about
derived for the random utility theory is used in this study 36%) two-wheeler users are Govt. employed. Another
for the travel demand analysis. NLOGIT software was used important observation was, irrespective of the differences in
for the modelling of WTP using binary logit model the annual road accident rates along the three roads, a large
(Hensher et al., 2009). Based on the SP data collected a portion of about 53.6% two-wheeler users felt that the
 INTERNATIONAL JOURNAL OF INJURY CONTROL AND SAFETY PROMOTION 5

Table 3. Parameter estimates of the model for two-wheeler users. safer alternatives. But age shows a positive attitude to the
Variables Parameter t statistic existing route, which is in line with the findings of
Gender (0 if male, 1 otherwise) 0.1552 2.506 Andersson (2013) that the effect of age on WTP will be a
Marital status (0 if married, 1 otherwise) 0.2632 3.327
Age (continuous) 0.2908 6.929
way of saving opportunity or about the safety of the new
Personal income (continuous) 0.0794 1.974 unfamiliar alternative.
Accident history (continuous) 0.0651 2.082 In a detailed analysis on the effect of the age group on
Household size (continuous) 0.0419 1.105
Household income (continuous) 0.0703 2.027
WTP, we could find an inverted U- shaped relationship
Travel time (TT) 0.0120 4.228 between age and the willingness to pay. We could see that
Travel cost (TC) 0.0129 2.408 WTP started at a lower value for the age group 18–23,
Accident rate (CR) 0.0068 17.017
Number of observations 4128 increased for the age group 24–40 and then reduced for the
Log likelihood of the estimated model 2203.547 age group 40–60, indicating an inverted U shape where
Log likelihood of constant only model 2861.312 middle-aged respondents tend to be more likely to vote in
Pseudo R2 0.2298
Adjusted R2 0.2242 favour of the WTP. This may be due to the fact that for the
WTP to reduce risk of being a road 0.5271 middle-aged group (24–40) they have more money to spend
accident victim (per trip) because of greater job stability and higher income than
other age groups. Also, better perception of risks due to life
experience (and possibly direct experience of injuries and
present roads are unsafe. The minimum, maximum and
injury costs) may be the reasons why people aged in this
average numbers of dependants for two-wheeler users are
group report a higher WTP. For the group of 18- to 23-
0.0, 5.0 and 1.65, respectively.
years old, lower understanding of risk and less payment
ability translate into low WTP. This is in line with the find-
Binary logit model analysis ings of Aldy and Viscusi (2007).
On the other hand, a different picture is obtained on the
This model is to determine the amount of money that an effect of age on WTP when the entire age group is consid-
individual is willing to forfeit in order to reduce the risk to ered altogether in the whole sample. Age shows a positive
be injured or killed in a road accident. In simple linear affinity towards the existing route indicating a reduction in
models, the WTP is calculated as the ratio of parameter esti- the value of WTP with age. There is also a widespread belief
mates related to one non-monitory attribute (here, number that willingness to pay should decline with age (Andersson
of accidents per year) and one monitory attribute (here, & Treich, 2011). Thus, we can see that the effect of age on
travel cost) holding all else constant. The model was formu- WTP is theoretically ambiguous and there are results indi-
lated using household variables, personal variables and, cating a negative relationship between WTP and age
travel related variables. The final specification of the model (Arthur, 1981; Hammitt, 2007; Hammitt & Graham, 1999).
was arrived based on a systematic process of eliminating In short, the findings in most studies support that VSL fol-
variables found to be statistically insignificant. The choice of lows an inverted U-shape, is declining, or is independent of
the alternate routes was considered as the dependent vari- age (Alberini, Cropper, Krupnick, & Simon, 2004; Alberini,
able. Table 3 presents the results of the binary logit model Hunt & Markandya, 2006; Andersson, 2007; Hammitt &
for two-wheeler users. Liu, 2004; Johannesson et al., 1999; Jones-Lee, Hammerton,
Road accident rate, travel time, travel cost and age are & Philips, 1985; Krupnick, 2007; Viscusi, & Aldy, 2007)
the most significant variables in the model. These parame- In terms of consistency and statistical properties the ‘pseudo
ters in the model are of expected signs. Coefficients of road R2’ index, analogue to the multiple correlation coefficients,
accident rate, travel time and travel costs show logical signs provides a practical and theoretically sound index of the good-
which indicate that an increase in these variables reduces ness of fit (Domencich & McFadden, 1975; Louviere et al.,
the utility. It can be observed that the female respondents 2000; McFadden, 1997). McFadden gives values of 0.2 to 0.4
prefer the existing route. This may be due to the fact that for the ‘pseudo R2’, for an excellent fit. Thus, goodness of fit
they were unsure about the safety of new faster alternatives. as measured by pseudo R2 equal to 0.228 shows a good fit.
Unmarried people attach more utility to the present alterna- The parameters relevant for the WTP calculation are
tive compared to married people. This is an intuitive finding travel cost and the number of road accidents. The WTP
as given by Yannis, Kanellopoulou, Aggeloussi, and value per person per trip is derived by dividing the accident
Tsamboulas (2005), that married people tend to have more parameter (CR) by the Travel cost (TC) parameter
responsibilities (they may have their spouse and/or kids at (Antoniou, 2014; Le et al., 2011). The WTP value of road
home) and so are more willing to seek higher risk reduc- accident for two-wheeler obtained here is Rs. 0.5271/person/
tion. Similarly, the household size has a positive coefficient, trip (US$0.01). Here, in the study, the value of avoiding
indicating lesser affinity towards a new costly alternative. It traffic accident has been estimated based on the average
is very clear that as the number of members increases in a daily two-wheeler traffic count. The average total number of
family, they find it difficult to keep aside more money for people travelling in two-wheeler along the routes in both
an expensive route. The variables, monthly personal income directions together is obtained as 9840 persons, considering
and, household income show a negative utility, indicating an average occupancy of 1.5 persons per two-wheeler.
that an increase in these variables make people think of new Taking per capita trip rate as 1.14 as obtained from a recent
6 S. BALAKRISHNAN AND K. KARUPPANAGOUNDER

unpublished study (‘Mass Transit Feasibility Study for the Traffic and Transportation Division, NIT during the data collection
Calicut City’ by M/S Wilbur Smith Consultants, Bangalore), and analysis stages is greatly acknowledged.
annual costs of accidents were calculated as follows.
9840  1.1.4  365  (0.5271/person/trip) = (Rs. 2,158,171 = ORCID
US$32,496). This value is comparable with the value obtained
by Mohan (2002) considering the consumer price index. Sivakumar Balakrishnan http://orcid.org/0000-0001-6719-0118

Conclusion
References
Most of the researchers calculating the WTP against road
Ainy, E., Soori, H., Ganjali, M., Le, H., & Baghfalaki, T. (2014).
traffic accidents studied the scenarios in developed coun- Estimating cost of road traffic injuries in Iran using willingness to
tries. Similar studies in the context of developing countries pay (WTP) method. PLoS One, 9(12), e112721. doi:10.1371/
are very few. Moreover, the differences with respect to time- journal.pone.0112721
monetary and personal constraints and other socio-eco- Alberini, A., Cropper, M., Krupnick, A., & Simon, N. B. (2004). Does
nomic factors are prominent between developed countries the value of a statistical life vary with the age and health status?
Evidence from the USA and Canada. Journal of Environmental
and developing countries. This article attempts to investigate Economics and Management, 48(1), 769–792. doi:10.1016/
the WTP values against road traffic accident in the context j.jeem.2003.10.005
of a developing country, India. Many families in India get Alberini, A., Hunt, A., & Markandya, A. (2006). Willingness to pay to
financially destabilized once their main earning member reduce mortality risks: Evidence from a three-country contingent
meets with a serious road accident injury or a fatality. valuation study. Environmental & Resource Economics, 33(2),
251–264. doi:10.1007/s10640-005-3106-2
Future education, career opportunities and even survival of
Aldy, J.E., & Viscusi, W.K. (2007). Age differences in the value of stat-
several families have been seriously affected by road acci- istical life: Revealed preference evidence. Review of Environmental
dents. These factors can be included in the costing process Economics and Policy, 1(2), 241–260. doi:10.1093/reep/rem014
only through WTP approach as it is practised in developed Andersson, H. (2007). Willingness to pay for road Safety and estimates
countries. This study is a first attempt of its kind in the of the risk of death: Evidence from a Swedish contingent valuation
study. Accident Analysis and Prevention, 39(4), 853–865. doi:
Indian context. This article has derived the WTP values of
10.1016/j.aap.2006.12.008
accident cost based on an SP survey for the two-wheeler Andersson, H. (2011). Perception of own death risk: An assessment of
users in India. The WTP values of these users per person road-traffic mortality risk. Risk Analysis: An Official Publication of
per trip were obtained through binary logit models. The the Society for Risk Analysis, 31(7), 1069–1082.
WTP value of road accidents for a two-wheeler rider is Andersson, H. (2013). Consistency in preferences for road safety: An ana-
lysis of precautionary and stated behavior. Research in Transportation
obtained as Rs. 0.5271/person/trip and the cost of accident
Economics, 43(1), 41–49. doi:10.1016/j.retrec.2013.01.001
obtained for two-wheelers (US$34,700) is comparable with Andersson, H., & Treich, N. (2011). The value of a statistical life. In A.
that obtained in previous studies. It was also found that de Palma, R. Lindsey, E. Quinet & R. Vickerman (Eds.), A handbook
travel cost and number accidents of the route selected, age, of transport economics (pp. 396–424). Cheltenham, UK: Edward
travel time, travel cost and personal/household income have Elgar.
the significant effect on the decision-making process. Antoniou, C. (2014). A stated-preference study of the willingness-to-
pay to reduce traffic risk in urban vs rural roads. European
Quantification of the parameters that influence the WTP Transport Research Review, 6(1), 31–42. doi:10.1007/s12544-013-
for road accident risk reduction and the WTP values, 0103-3
attempted in this research, could be useful for assisting in Arthur, W.B. (1981). The economics of risks to life. The American
planning/implementation of appropriate traffic safety pro- Economic Review, 71(1), 54–64.
grams. Findings in this study and the methodology used can Bhattacharya, S., Alberini, A., & Cropper, M.L. (2006). The value of
mortality risk reductions in Delhi, India. New Delhi: World Bank
be applied for the identification of risk reduction choice Policy Research Working Paper. doi:10.1596/1813-9450-3995
sensitivity. The model proposed in this study was from a Central Road Research Institute. (1982). Road user cost study in India
specific region of the country and so the existence of people – Final report. New Delhi: Govt. of India.
behaviour referred in this study should be carefully consid- Chaturabong, P., Kanitpong, K., & Jiwattanakulpaisarn, P. (2011).
ered before applying these findings in local, regional and Analysis of costs of motorcycle accidents in Thailand by willing-
ness-to-pay method. Transportation Research Record: Journal of the
national level projects. In future research, the evidence
Transportation Research Board, 2239(1), 56–63. doi:10.3141/2239-07
herein can be disaggregated by road type, distance, and so De Blaeij, A., Florax, R.J.G.M., Rietveld, P., & Verhoef, E.T. (2003).
forth, to provide inputs for various types of analysis like The value of statistical life in road safety: A meta-analysis. Accident:
benefit-cost analysis, in contrast to the aggregated findings Analysis and Prevention, 35(6), 973–986.
presented herein. De Blaeij, A., Rietveld, P., Verhoef, E.T., & Nijkamp, P. (2002). The
valuation of a statistical life in road safety: A stated preference
approach. London, UK: 30th European Transport Forum.
Acknowledgements Dionne, G., & Lanoie, P. (2004). Public choice about the value of a
statistical life for cost-benefit analyses: The case of road safety.
Authors extend their sincere gratitude to the Centre for Transportation Journal of Transport Economics and Policy, 38, 247–274.
Research, Civil Engineering Division, National Institute of Technology Domencich, T.A., & McFadden, D. (1975). Urban travel demand: A
(NIT), Calicut for the immense support given to carry out the present behavioral analysis. Amsterdam, Oxford: North-Holland Publishing
work. Also, the help rendered by Ms. Geethu Ashok, a PG student of Company/American Elsevier Publishing Company, Inc, New York.
 INTERNATIONAL JOURNAL OF INJURY CONTROL AND SAFETY PROMOTION 7

Elvik, R. (1995). An analysis of official economic valuations of traffic Louviere, J.J., Hensher, D.A., & Swait, J.D. (2000). Stated choice meth-
accident fatalities in 20 motorized countries. Accident Analysis & ods: Analysis and application. Cambridge: University Press.
Prevention, 27(2), 237–247. doi:10.1016/0001-4575(94)00060-Y McFadden, D. (1997). Quantitative methods for analyzing travel behav-
Haddak, M.M., Lefevre, M., & Havet, N. (2014). Willingness-to-pay for iour of individuals: Some recent developments. Cowles Foundation
road safety improvement. Paris: Transport Research Arene. Discussion paper No. 474. New Haven, Connecticut: Cowles
Hammitt, J.K. (2007). Valuing changes in mortality risk: Lives saved Foundation for Research in Economics, Yale University.
versus life years saved. Review of Environmental Economics and McFadden, D. (1998). Measuring willingness-to-pay for transportation
Policy, 1(2), 228–240. doi:10.1093/reep/rem015 improvements. In T. G€arling, T. Laitila, & K. Westin (Eds.),
Hammitt, J.K., & Graham, J.D. (1999). Willingness to pay for health Theoretical foundations of travel choice modelling (pp. 339–364).
protection: Inadequate sensitivity to probability? Journal of Risk and Amsterdam: Elsevier Science.
Uncertainty, 18(1), 33–62. Miller, T.R. (2000). Variations between countries in values of statistical
Hammitt, J.K., & Liu, J.T. (2004). Effect of disease type and latency on life. Journal of Transport Economics and Policy, 34(2), 169–188.
the value of mortality risk. Journal of Risk and Uncertainty, 28(1), Ministry of Road Transport and Highways (MoRTH). (2014). Annual
73–95. doi:10.1023/B:RISK.0000009437.24783.e1 report 2012-2013. New Delhi: Government of India. Retrieved from
Hensher, D.A., Rose, J.M., de Dios Ort
uzar, J., & Rizzi, L.I. (2009). http://www.morth.nic.in/showfile.asp?lid=1656
Estimating the willingness-to-pay and value of risk reduction for car Ministry of Road Transport and Highways (MoRTH). (2017). Annual
occupants in the road environment. Transportation Research Part A: report 2016-2017. New Delhi: Govt. of India.
Policy and Practice, 43(7), 692–707. doi:10.1016/j.tra.2009.06.001 Mohan, D. (2002). Social cost of road traffic crashes in India.
Hills, P.J., & Jones-Lee, M.W. (1981). The costs of traffic accidents and Proceedings of First Safe Community Conference on Cost of Injury,
the valuation of accident-prevention in developing countries.
(pp. 33–38). Vibrog, Denmark.
Proceedings of the Summer Annual Meeting, Planning and
Rizzi, L.I., & de Dios Ort
uzar, J. (2003). Stated preference in the valu-
Transport Research and Computation (International), Issue P207,
ation of interurban road safety. Accident Analysis and Prevention,
pp. 113–132. University of Warwick, England.
35(1), 9–22.
Hojman, P., Ort
uzar, J., de, D., & Rizzi, L.I. (2005). On the joint valu-
Schoeters, A., Wijnen, W., Carnis, L., Weijermars, W., Elvik, R.,
ation of averting fatal victims and serious injuries in highway acci-
Johannsen, H., … Daniels, S. (2017). Costs related to serious
dents. Journal of Safety Research, 36, 337–386.
Irag€
uen, P., & de Dios Ort
uzar, J. (2004). Willingness-to-pay for reduc- injuries, D7.3 of the H2020 project SafetyCube. Loughborough:
ing fatal accident risk in urban areas: An internet-based web page Loughborough University.
stated preference survey. Accident Analysis and Prevention, 36(4), TRL. (1995). Overseas road note 10 - costing road accidents in develop-
513–524. doi:10.1016/S0001-4575(03)00057-5 ing countries. Crowthrone, Berkshire, US: Transport Research
Johannesson, M., Blomquist, G.C., Blumenschein, K., Johansson, P.O., Laboratory.
Liljas, B., & O’Conor, R.M. (1999). Calibrating hypothetical willing- Viscusi, W.K., & Aldy, J.E. (2007). Labor market estimates of the
ness to pay responses. Journal of Risk and Uncertainty, 18(1), 21–32. senior discount for the value of a statistical life. Journal of
doi:10.1023/A:1007708326467 Environmental Economics and Management, 53(3), 377–392. doi:
Jones-Lee, M.W., Hammerton, M., & Philips, P. (1985). The value of 10.1016/j.jeem.2006.10.005
safety: Results of a national sample survey. The Economic Journal, Whittington, D. (1998). Administering contingent valuation surveys in
95(377), 49–72. doi:10.2307/2233468 developing countries. World Development, 26(1), 21–30. doi:10.1016/
Krupnick, A. (2007). Mortality risk valuation and age: Stated prefer- S0305-750X(97)00125-3
ence evidence. Review of Environmental Economics and Policy, 1(2), World Health Organization (WHO). (2015). Global status report on
261–282. doi:10.1093/reep/rem016 road safety-2015. Retrieved October 21, 2015, from http://www.who.
Le, H., van Geldermalsen, T., Lim, W.L., & Murphy, P. (2011). int/violence_injury_prevention/road_safety_status/2015/en/
Deriving accident costs using willingness-to-pay approaches – A case Yannis, G., Kanellopoulo, A., Aggeloussi, K., & Tsamboulas, D. (2005).
study for Singapore. Proceedings of the Australian Transport Modelling driver choices towards accident risk reduction. Safety
Research Forum (ATRF), (pp. 28–30). Adelaide, Citeseer, Australia. Science, 43(3), 173–186. doi:10.1016/j.ssci.2005.02.004