Volvo Chair Professor, Transportation Research and Injury Prevention Programme, Indian Institute of Technology Delhi, New Delhi,

India
(Received March 30, 2009)
Download full-t ext PDF Download cit at ion Copy link

1. INTRODUCTION estimates suggest that the actual number of vehicles

use may be about 60-70% of the official number 1
The number of vehicles registered in India is shown Table 2 shows the sales of motor vehicles in India in 199
in Table 1. These data show that the total number of ve- and 2007. The sales figures also show an average annu
hicles increased from 37 million in 1997 to 73 million in increase of 10-12% per year.
2004. This represents an annual average growth rate of Table 3 shows the number of road traffic fatalit
about 11% for cars and motorised two-wheelers and 7% and the population of India from 1997 to 20073. The to
for trucks and buses. However, these numbers are probably number of fatalities increased at an average rate of ab
overestimates as personal vehicle owners register their 4% per year in the period 1997-2003 and the rate has
vehicles and pay the road tax once when they buy the ve- creased to 8% per year since then. The number of fata
hicle and are not required to pay an annual tax. Because ties per million population remained around 79-83 in t
of this, a large number of vehicles remain on the official period 1997-2003 and has since increased to 101. Traf
record even when they are not in use any more. Recent fatalities per unit population has been taken as an indic

Table 1 Motor vehicle registration in India

Year MTW* Cars/Jeeps Trucks Buses Others** Total
1997 25,729 (69)*** 4,672 (13) 2,343 (6) 484 (1.1) 4,104 (11) 37,332 (100)
2004 51,922 (71) 9,451 (13) 3,749 (5) 768 (1.3) 6,828 (09) 72,718 (100)
Growth/year (%) 10.6 10.6 6.9 6.8 7.5 10.0
* Motorised two-wheelers
** Others include tractors, trailers, three wheelers and other miscellaneous vehicles which are not separately classified.
*** Numbers in parentheses represent row percentages.
Source: Ministry of Road Transport and Highways, Delhi

Table 2 Vehicle sales in India

Year Motorised two-wheelers Three-wheelers Cars Commercial vehicles Total
1997 2,885,004 (79) 216,729 (06) 396,450 (11) 169,937 (5) 3,668,120 (100)
2007 7,416,191 (78) 380,663 (04) 1,273,893 (13) 479,593 (5) 9,550,340 (100)
Growth/year (%) 10 6 12 11 10
Source: Society of Indian Automobile Manufacturers, Delhi

Table 3 Road traffic fatalities in India

Year Fatalities Population (million) Fatalities/million persons

1997 77,000 955 81
1998 79,900 971 82
1999 82,000 987 83
2000 78,900 1,002 79
2001 80,900 1,027 79
2002 84,059 1,051 80
2003 84,430 1,068 79
2004 91,376 1,086 84
2005 98,254 1,103 89
2006 105,725 1,120 94
2007 114,590 1,136 101
Source: National Crime Records Bureau, Delhi

IATSS RESEARCH Vol.33 No.1, 2009

Download full-t ext PDF Download cit at ion Copy link

tor of the health burden of road traffic crashes on society Table 4 Traffic fatalities by category of road
at the city, regional, or national level. At the individual user in Delhi and selected locations
level, what is of consequence is the risk of injury per trip, on national highways
and the total number of trips is proportionate to the popu-
Location (%)
lation. Therefore, traffic fatalities per unit population can Type of road user Delhi Highways*
be taken as a rough indicator of risk faced by individuals. 2001-2005 1999
The risk of being involved in a fatal road traffic crash has
Truck 2 14
obviously been increasing for Indian citizens over the Bus 5 3
past few years. While some of this increase can be attrib- Car 3 15
uted to increase in the number of motor vehicles per cap- Three-wheeled scooter taxi 3 -
ita in India, however, increasing vehicle ownership need Motorized two-wheeler 21 24
not result in increased fatality rates if adequate safety Human and animal powered vehicle 3 1
measures are implemented 4. Bicycle 10 11
Pedestrian 53 32
2. CRASH PATTERNS Total 100 100
* The data are for 11 selected locations, and thus might not be represen-
Details of traffic crashes are not available at the na- tative for the entire country. (Tractor fatalities are not included.)
tional level. While the official road traffic fatality data
may be close to the actual number, the injury data are gross
underestimates5. In this report only fatality data are used 2.2 Age and gender
for analysis as non-fatal data may suffer from many bi- Figure 1 shows the distribution of road traffic fa-
ases. talities in 2007 by age group and gender3. In 2007, only
15% of the victims were females. This is partly because
2.1 Road user category of the low representation of women in the Indian work-
Official road traffic crash data do not include fatali- force and exposure on roads. Children aged 14 years and
ties by road user category in India. Such data are only younger comprise only 6% of the fatalities, though their
available from a few cities and research studies done on share in the population is 32%. The proportion of fatali-
selected locations on rural highways. Table 4 shows traf- ties in the age groups 15-29 and greater than 60 years is
fic fatalities by category of road users in Delhi (capital
city of India) and selected locations on national high-
45,000
ways4,6. These data show that car occupants were a small Female Male
proportion of the total fatalities, 3% in Delhi and 15% on 40,000
rural highways. Vulnerable road users (pedestrians, bicy- 35,000
clists, and motorized two-wheeler riders) accounted for
Number of Fatalities

30,000
84% of deaths in Delhi and 67% on highways. This pattern
25,000
is very different from that obtained in all high-income
countries. The low proportion of car occupants can be 20,000
explained by the low level of car ownership at 7 per 100 15,000
persons as compared to more than 50 per 100 persons in
10,000
most high income countries. At present levels of growth
5,000
in vehicle ownership in India, vulnerable road users are
likely to remain the dominant mode for the next few de- 0
15 - 2 9 30 - 44 4 5 - 59 > 59
0 -14
cades. The incidence of road traffic fatalities can only be Age [ years]
controlled in the coming years if road safety policies put Source: National Crime Records Bureau, Delhi
a special focus on the safety of vulnerable road users.
Fig. 1 Traffic fatalities by age and gender, India 2007

76 IATSS RESEARCH Vol.33 No.1, 2009

Download full-t ext PDF Download cit at ion Copy link

similar to their representation in the population, but the under the influence of alcohol. Evidence for increas
middle-age groups 30-44 and 45-59 are over represented use of alcohol comes from a hospital study in Delhi whe
by about 70%. The low representation of children (2 fatali- 29% of the riders of motorized two-wheelers admitted
ties per 100,000 persons) is curious because a significant alcohol consumption before the crash7. In Bangalore
number of children walk and bicycle to school unescort- hospital-based study showed that alcohol was involved
ed, both in urban and rural areas. Though the exposure 22% of nighttime crashes, and that 35% of random
numbers for India are not available, children’s presence checked drivers on the road at night were under the infl
on the road unsupervised is not insignificant. The reasons ence of alcohol5.
for the low involvement rate of children needs to be in-
vestigated. 2.4 Fatalities in cities with population greater than
million persons
2.3 Time of day Figure 3 shows the fatality rates for cities with po
Figure 2 gives the proportion of fatalities by time of ulations greater than 1 million persons for the years 20
day in 35 large cities of India (population > 1 million) and 20073. Delhi had the highest number of fatalities
and in the rest of the country including rural roads. In the 2007 (1,789) with a rate of 140 per million population. T
period 09:00 to 21:00 the proportions remain high and lowest rate was in Kolkata (35) and the highest in Ag
similar both in the large cities and elsewhere. In the late (386), with an overall average of 122 fatalities per m
night hours (21:00-24:00) traffic volumes are much lower lion persons for all these cities. In this period of six year
than the peak day time rates1 but the fatality rates do not only eight of the 35 cities did not experience an increa
reflect this. In the early morning hours (00:00-06:00) the in fatality rates. The highest increase was 433% in Asa
proportions are much lower in the large cities, but rela- sol. Since a vast majority of the victims in these cities a
tively higher in the rest of the country. It is possible that vulnerable road users, one possible cause could be i
since the rest of the country includes national highways, creases in vehicle speeds. The probability of pedestri
the commercial goods traffic on those highways may ac- death is estimated at less than 10% at impact speeds of
count for this. In the absence of more detailed epidemio- km/h and greater than 80% at 50 km/h, and the relati
logical data we can only surmise that the high rates at ship between increase in fatalities and increase in imp
night could be due to higher speeds of vehicles when traf- velocities is governed by a power of four 8,9. Small i
fic volumes are lower and/or higher frequency of driving creases in urban speeds can increase death rates drama
cally.
18
Million + cities Rest of India 2.5 Fatalities on rural highways
16
Detailed data are not available at the national
14
state level for crashes on national highways. A study co
12 lected data on modal shares, vehicle speeds, and traf
10 crashes on selected locations on national and state hig
%

8 ways around the country in the late 1990s6. Table 4 show

6 the type of road users killed on highways. The study r
4 ported that trucks were the striking party in 65% of fa
2 crashes. Other studies report that majority of the crash
0 involved buses, 25% of the victims were pedestrian
00-03 0 3 -06 0 6 -09 0 9 -1 2 1 2 -1 5 15 -1 8 18 -2 1 21 -24 rear-end crashes comprised 40% of total crashes and t
Time [hours]
crashes were increasing at a rate of 3.9% per year10-13
Source: National Crime Records Bureau, Delhi
study of road traffic crashes on a National Highway in
Fig. 2 Road traffic accident proportions (%) by time southern state of Kerala reported that heavy vehicles h
of day in 35 cities with more than 1 million pop- a high involvement, and pedestrians and cyclists we
ulation and those in the rest of India in 2007 28% of the victims14. The most important finding of th

IATSS RESEARCH Vol.33 No.1, 2009

Download full-t ext PDF Download cit at ion Copy link

400
2001 2007
350

300
Fatalities per million

250

200

150

100

50

Varanasi

Vijayawada
Surat
Vadodara

Kochi
Madurai

Meerut
Faridabad
Jabalpur

Allahabad

Jaipur
Mumbai

Lucknow
Patna

Nagpur
Ludhiana

Bangalore
Hyderabad

Nasik
Asansol
Kolkata

Jamshedpur

Agra
Amritsar

Bhopal

Coimbatore
Rajkot

Vishakhapatnam
Indore
Delhi (City)

Pune

Chennai

Kanpur
Dhanbad

Source: National Crime Records Bureau, Delhi

Fig. 3 Traffic fatality rates in cities with populations of at least one million, 2001 and 2007

study is that the fatality rate per volume is more than traffic fatalities in India. If we assume that the present
three times higher on the four-lane section than on two- growth rate of 8% per year declines in a linear manner to
lane sections. The construction of four-lane divided high- 0% by 2030, then we can expect about 260,000 fatalities
ways (without access control) does not seem to have re- by 2030. Neither of these projected dates (2042 and 2030)
duced fatality rates, and vulnerable road users still account can be accepted as road safety goals for the country.
for a large proportion of fatalities. There is a clear case for An earlier report co-authored by the present author
redesign of intercity roads with separation of slow and has a more detailed analysis of the road traffic situation in
fast modes. The need of road users on local short distance India and possible countermeasures4. In summary, road
trips will have to be accounted for. Solutions for many of safety policies in India must focus on the following is-
these issues are not readily available and research studies sues to reduce the incidence of road traffic injuries: pe-
are necessary for the evolution of new designs. destrians and other non-motorist in urban areas; pedestri-
ans, other non-motorists, and slow vehicles on highways;
3. SUMMARY motorcycles and small cars in urban areas; over-involve-
ment of trucks and buses; night-time driving; and wrong-
Road traffic fatalities have been increasing at about way drivers on divided highways. There is an urgent need
8% annually for the last ten years and show no signs of to revamp police data collecting procedures so that neces-
decreasing. Two modelling exercises have attempted to sary information is available for scientific analysis. India
predict the time period when we might expect fatality rates specific countermeasures will be possible through con-
to start to decline in a range of countries9,15. Cropper and tinuous monitoring and research, which will require the
Kopits predicted that fatalities in India would reach a total establishment of road safety research centers in academic
of about 198,000 before starting to decline in 2042 and institutions and a National Road Safety Board that could
Koornstra predicted an earlier date of 2030 for the peak help move toward a safer future as outlined above.

78 IATSS RESEARCH Vol.33 No.1, 2009