THEECONOMIC AND SOCIAL COSTS
OF ROAD ACCIDENTS IN AUSTRALIA:
WITH PRELIMINARY COSTESTIMATES FOR AUSTRALIA 1978
prepared by
A.S.ATKINS
Centre for Environmental Studies
University of Melbourne
Project Sponsor:
Office of Road Safety
Department of Transport Australia
June 1981
i
DEPARTbZNT OF TRANSPORT
O F F I C E O F ROAD SAFETY
I
DOCUMENT RETRIEVAL INFORMATION
Report No.
CR 2 1
'
Date
ISBN
Pages
0 6 4 2 11202 7
June1981
128
T i t l ea n dS u b t i t l e
THE ECONOMIC AND SOCIAL COSTS OF ROAD ACCIDENTS I N AUSTRALIA:
With P r e l i m i n a r y C o s t E s t i m a t e s f o r A u s t r a l i a
i n 1976
Author
Alan S . Atkins
P e r f o r m i n g O r g a n i s a t i o n ( N a m e and A d d r e s s )
Centre f o r EnvironmentalStudies
University of Xelbourne
P a r k v i l l e ,V i c t o r i a ,
3052
Keywords
A c c i d e n tC o s t s ,R e v i e w s / S m a r i e s
Abstract
T h i sr e p o r tc o n t a i n s
a r e v i e wo fr e c e n ta c c i d e n tc o s te s t i m a t i o ns t u d i e s
and
proposes a u n i t c o s t framework which i s c o n s i d e r e d a p p r o p r i a t e f r o m b o t h
conof t h e s o c i a l c o s t s o f
roi
c e p t u a la n de m p i r i c a l
view p o i n t s f o r t h e e s t i m a t i o n
and t o t a l c o s t e s t i m a t e s f o r
a c c i d e n t s i n A u s t r a l i a . A d e t a i l e d s e t o fu n i t
on e x i s t i n g d a t a s o u r c e s ,
supplemt
A u s t r a l i a i n 1978 i s p r e s e n t e d , b a s e d m a i n l y
by a c c i d e n t c l a i m s d a t a
from a sample o f i n s u r a n c e companiesand t h e Motor Acc:
B o a r do fV i c t o r i a .
These p r e l i m i n a r yc o s te s t i m a t e sa r ea l s oc l a s s i f i e d
actor<
t oi n j u r ys e v e r i t y .
Problems a s s o c i a t e d w i t h a c c i d e n t s t a t i s t i c s
and d a t a sou:
are discussed together
w i t h t h e e s t i m a t i o n p r o c e d u r e s . The r e p o r t c o n c l u d e s t k
social costs provide only
minimm e s t i m a t e s o f t h e b e n e f i t s g a i n e d from accidel
r e d u c t i o n ,a n da l s ot h a tt h eu s e
of a v e r a g e c o s t l e v e l s
i s c o n s t r a i n e d by t h e
skewed d i s t r i b u t i o n of m o s ta c c i d e n tc o s tc h a r a c t e r i s t i c s .F u r t h e rr e s e a r c h
directedtowardsrefinementoftheconceptualandempiricalbasesofthese
e s t i m a t e s is recommended.
NOTE :
Tnis r e p o r t i s d i s s e m i n a t e d i n t h e i n t e r e s t ofinformationexchange.
The view
e x p r e s s e d are t h o s e o f t h e a u t h o r ( s )
and do n o t n e c e s s a r i l y r e p r e s e n t t h o s e
of
t h e Commonwealth Government.
I
The O f f i c e o f Road S a f e t y p u b l i s h e s two s e r i e s o f r e p o r t s r e s u l t i n g
from i n t e r
r e s e a r c h a n d external r e s e a r c h , t h a t i s , researchconducted
on behalfof
the
O f f i c e .I n t e r n a lr e s e a r c hr e p o r t s
are i d e n t i f i e d by OR w h i l e e x t e r n a l r e p o r t s
i d e n t i f i e d by CR.
~
~~~
ii
73is studv has benefited from access to t h e statistical c o l l e c t i o n and l i b r a r y
resources of the Office of
Road
Safety and from various O.R.S. s t a f f during
B. Iancashire,
W. Chllaghanand
thecourse of the projectincluding
C. Boughton.
Mr.
Ms.
M.r
The assistance of the k t o r Accidents h a r d of Victoria i n the provision of a
n m k r of special
accident
data
tabulations
under the guidance
of
D.
I k a r s l e y is gratefully acknowledged.
Several of the
rmjor
mtor vehicle
insurers i n Victoria responded t o r e q u e s t s f o r
special tabulations of claims
d a t a , and t h e c w p e r a t i o n of Australian
Associated
k t o r Insurers Ltd.
Insurance Fty. Ltd.,and
(A.A.M.I.), Insurance Council of Australia, R.A.C.V.
t h e State Insurance Office of Victoria is rmch appreciated.
Mr.
Also, i n January 1980 the author was able to undertake useful discussions with
J. O ' h y of the Highway Safety &search I n s t i t u t e of theUniversity
of
Michigan, and
D. Sowrford and
B.M. Faigin of the National Highway
Traffic Safety Administration i n Washington D.C.
M.r
Ms.
Ms.
iii
C O N T E N T S
.
Acknowledgements
L i s t of Figures
L i s t of Tables
1
1
1.1
1
1
Introduction
Scope of Report
?he Concept of Social Cbst and Review
of Recent Cost S t u d i e s
3
iU
U
Intraluction and a t l i n e of Study
1.2
2
I
22
4
4
2.1
'Ihe
Concept
of Soeial Cost
2.2
Economic
Cbnsiderations and me Value
of Life
2.3
Review of Recent Cost Studies
8
11
A Framework for Australian Cost Estimtes
20
3.1
3.2
3.3
4 Proposed Pccident Cost Ramwork
a t l i n e of Cost Conponents and Ihta
Sources
SLnmrrry of E s t i m t i o n Procedures
Conclusions
4
4.1
4.2
and R e c m n d a t i o n s
Conclusions
Recmndations
Bibliography and Ikferences
20
24
46
54
54
57
59
Appendices
Review and a t l i n e of Recent Cbst
Estirration Studies
Supplemntary
A- 2
'hbles
A-3
S t a t i s t iD
c ai ls t r i b u t i o n s
Fitted t o
Accident Cbst Frequencies
A- 1
64
109
120
iv
LIST OF FIGURES
page
Figure Numher
a-1
a-2
a-3
Garma Distribution fitted to Hospital
Claims of the M.A.R. (Victoria)
1977178
123
G
m D i s t r i b u t i o n f i t t e d to Sample
of 1978 Vehicle r)arrage Claims
125
Garrnra D i s t r i b u t i o n f i t t e d t o
1966 Vehicle Darmge Costs
A.C.T.
126
V
LIST OF TABLES
ZBble I@
1
2
3
4
5
G
Page
Average Accident Costs by InjurySeverity
Level: Australia 1978 (10%)
21
surmrr*ry of Total Accident Costs:
Australia 1978 (10%)
22
Suwmry of Average Accident k t s :
Australia 1978 (1WJ)
23
Road AccidentsInvolvingCasualties:
Australia 1975-78
25
Classification of Accidents by Injury
Severity:
Australia
1978
26
Foregone I n c a :
Australia 1978
27
Xoad Accident m t a l i t i e s :
7
& ? a n I n c m by Age Groups:
8
H f e c t of theDiscount
9
Age Distribution of P a d Accident m t a l i t i e s :
10
11
12
13
14
15
16
17
Australia 1978
Fate on Foregone I n c m
29
31
Australia 1978
33
Age Distribution of Road Accident I n j u r i e s :
Australia 1977
33
Foregone Income:
Australia 1978
34
RDad Accident I n j u r i e s :
m i n i n g L i f e t k Income:
I n j u r i e s : Australia 1978
Road Accident
35
Effects of Age Group Weighting on Discounted
I n m s : Australia 1978
35
Foregone .Adjusted Income:
Australia 1978
3G
Foregone Net Income:
Australia 1978
Fatalities:
mtalities:
Foregone Net Adjusted Income:
Australia 1978
38
mtalities
Injury Cost I e v e l s by Vehicle m g e :
A.C.T.
1965/66
38
39
vi
page
Table No
18
19
20
Claim Arising from Raad Accidents:
Year to 30 .June 1978 M.A.B. (Vic.)
40
Hospital, Medical and Otter Costs of
(Vic.)
Accident Chsualties: M.A.B.
41
Total Costs Claimd by Road Accident
I n j u r i e s : M.A.B. (Vic.)
42
Victoria 1977178
43
21
R a d Accident Costs:
22
Preliminary Ektitmte of &tor Vehicle
Accident Repairs: Australia 1977178
44
k t o r Vehicle Accident Repairs:
Australia 1971
44
& t o r Vehicle InsuranceClaims:
Australia 1977178
45
23
34
25
&tor Vehicle Insurance Claims:
2G
B f e c t s of DeletingSelected
27
Annual Qst of Fole Accidents:
28
Average Accident b t by I n j u r y Level
(1979) O x r e n t Resource Cbsts
66
Average Accident Uxts by Injury Level
(1979) %tal Qsts n e t of Qnsmption
67
Average Accident Costs by Injury Level
(1979) %tal (bsts (Direct & Indirect)
68
Estimated Cost ofAccidents
1976 & 1978
71
29
30
31
32
33
1979
Ccst Categories
klbourne
72
73
LoGt Work Efforts of Road Accident
I n j u r i e sa: n a d a
35
65
Work E f f o r t s :
a n a d a 1975
34
51
anada:
Value of Productive Work E€forts:
Chnada 1975
Present Value ofFuture
46
1976
Estimated Social Losses from Raad
Accidents : Japan 1974
75
83
vii
page
‘Igble N3
36
37
E s t i m t e d Lust Incme:
F a t a l i t i e s : Japan
1974
.Average AccidentCnsts
U.S.A. 1975 (%)
Accident
85
by InjuryLevel:
93
38
Abiireviated I n jury Scale ( A . I. S . )
94
39
S m m r y of Lost Incune:
1975
95
Xm-Fatal Injuries
40
Average k c i d e n t Costs by Injury Level:
U.S.A. 1975 (10%)
41
kbad -kcident
Australia 1969
42
43
44
45
Cnmponents:
97
E s t h t e d b k r of Collisions:
Australia 1969
98
Unit Costs of k c i d e n t s by Type:
Australia 19G9
99
Person:
Exmomic Life Cycle of t h e Average
Australia 1969
100
lost Economic Value of Fatalities kt
of Future C o n s q t i o n 1969
100
46
Costs Allocated by k c i d e n t Type:
47
Average “Allocated”Costs
1969
48
AccidentCnsts
1971
1969
46
A3r Accident:
by Severity Type:
47
U.S.A.
48
Average Accident Costs by I n j u r y Level :
U.S.A. 1971 (7%)
106
50
Average Accident Wts by h j u r y
110
51
SeverityLevel:
52
Average AccidentCnsts by I n j u r y Severity
1978 (7%)
Level:Australia
49
53
Australia 1978 (T%)
110
111
Sumrary of T o t a l Accident Costs:
Australia 1978 (13%)
112
viii
l h b l e N3
54
55
56
57
page
Swxmry of Average Accident Costs:
Australia 1978 (7%)
112
3 m m r y of Average Accident Costs:
Australia 1978 (13%)
113
T o t a l Accident Costs by Category and
InjuryLevel:
Australia 1978 (10%)
113
Total Accident U x t s by Category and
InjuryLevel:Australia
58
59
GO
61
63
G4
G5
114
Effects of Age Group Weighting on
Discounted
Incunes:
Australia 1978 (10%)
114
ReminingLifetimeAdjustedIncane:
Injuries:
Australia
1978
115
ReminingLifetime
Injuries:
Australia
Net Income:
1978
115
Remining Lifetime Net Adjusted Income:
Injuries:
Australia
62
1978 (13%)
1978
116
Total Accident Oxts by Category and
Australia 1978 (7%)
InjuryLevel:
116
Approximte Severity Classification
of VictorianInjury h t a 1977/78
117
Foregone I n m f o r 1973 F h t a l i t i e s :
U.S.A. m a r e d with Present Study
118
Productivity
Calculations:
Australia
1967-1978
119
1
1.1
INTRODUCTION
This reprt presents t h e r e s u l t s of a study of the sncial costs
of road accidentsinAustraliacmnissioned
i n 1979 by the
Office of b a d S f e t y w i t h i n t h eb m n w e a l t h
l k p a r t m e n t of
lhe ,min objective of this study is t o review t h e
Transport.
scope of previous w r k on t h e valuation of accident costs with
the aim of extending t h e coverage and lwasuremn~t of road
accident costs i n .Australia t o r e f l e c t a conprahensiveconcept
of social cost and welfare. 'he potentialapplication of such
cost estimtes is to assist g o v e r m n t agencies w i t h the
glanning and evaluation of road s a f e t y progr-s.
w a s envisaged i n i t i a l l y i n two stages i n
of a reviewof
alternative r E t h d o l o g i e fs o qr u a n t i f y i n gt h e
socio-economic
costs o€ road accidents together with a preliminary estirrate of
t h et o t a l
cost of roadaccidents
i n .Australia for a recent
year.
'he
overall
study
which t h e f i r s t s t e e was m i n l yt oc o n s i s t
The second stage envisaged subsequent developnent and
r e f i n e n t of t h e cost framework developed i n the earlier s t a g e
including the estitmtion of separate accident costs f o r a range
of accidenttypes
(each showing d e t a i l s of estimtes f o r t h e
earlier established cost categories).
1.2
SCOPE OF REFORT
The concept of social cost, as it relates t o road accidents, is
consideredin Chapter 2, and t h e d i s t i n c t i o n between f i n a n c i a l ,
econanic and social costs is outlined.
'he use of changes i n cost l e v e l s as masures of social b e n e f i t
i n b e n e f i t c o s t a n a l y s i s is discussed including t h e s i g n i f i c a n t
implications that these applications of accident costs have f o r
t h e d e f i n i t i o n of accident cost concepts. The t h r e ep r i n c i p a l
economic
approaches t o t h e valuation of l i f e problem i nt h e
l i t e r a t u r e are considered,
leading
t o a preference
for
the
h m c a p i t a l approach i n t h e context of accidentevaluation.
The existence of a r e l a t i v e l y large and oftencontroversial
literature i n t h i s area is acknowledged.
2
Chapter 2 alsoincludes
a sunarary review of recent accident
c o s t e s t i t m t i o n s t u d i e s for t i e U.S.A., Australia, *w Zealand,
Japan, and Canada (these studies are reviewed a t greater length
i n Appendix A-1).
While deficiencies i n accidentrecords and cost data are found
source
of d i f f i c u l tiyn
cast e s t i m t i o n
t o t~ a general
resultant cost levels are shown to be mre
procedures,
the
use of alternative cost
subject to v a r i a t i o n r e s u l t i n g f r a n t h e
concepts than frorn estimationaccuracy.
?he s t u d i e s reviewed
are found t o employ a wide range of cost concepts and e s t i m t e d
u n i t cost levels.
Chapter 3 proposes a framwork for the estirration of unit costs
of road accidents
in
Australia,
together
with a set of
preliminaryestimates
of average and t o t a l costs for the year
1978.
An important
characteristic
of t h e cost framework
all
adopted i n t h i s study is the r e q u i m n tt od i s t i n g u i s h
accident costs by i n j u r y s e v e r i t y l e v e l .
The estimationprocedureforforegone
income of f a t a l i t i e s and
casualties, one
of
the key components of accidentcosts,
is
outlinedtogether
w i t h t h e e f f e c t s on t h i s item of changes i n
the discount rate, the age d i s t r i b u t i o n of v i c t i m , and t h e use
of alternative d e f i n i t i o n s of i n m .
These r e s u l t s are
outlined i n s a d e t a i l .
Fstjlmtes of m d i c a lh, o s p i t a l ,
and r e l a t e d costs of road
accidents are presented, based extensively upon dataprovided
by the b t o r k c i d e n t s Board of Victoria. Vehicle d m g e costs
are then considered and t h e basis of estimation is outlined.
The source of a l l other
costs
included
the
inproposed
framework is briefly considered.
Next, a d e t a i l e d sunnary of cost estinationprocedures
traces
through the u n i t cost framework proposed i nt h i sr e p o r t
and
of a l l average cost estimates
for
outlines
the
-derivation
Australia i n 1978 ( o t h e r d e t a i l s of data and e s t i m t i o n s o u r c e s
are contained i n lkble footnotes).
%is section is followed by
calculations of t o t d cost estimtes which shm the e f f e c t s of
excluding certain
cost
categories from t hoev e r a l l
cost
framework.
Finally, a brief
consideration
of the need f o r
f u r t h e r research t o support
the
proposed cost framework
concludeschapter 3.
Chapter 4 presents the conclusions and recornrendations arrived
a t i n this report. %e conclusions incorporate bothconceptual
associated
with the proposed cost
and empirical problem
e s t i m t i o n framework.
The recornrendations relate to f u r t h e r
progressenvisagedforsubsequentdevelopnent
of thepresent
study, together with other recomnendations for further research
i n t o Australian accident costs drawn frcm the r e s u l t s of the
present report.
3
Appendix A-1 contains reviews of the
seven
accident
cost
e s t i m t i o n studiesdiscussed
i n chapter 2. These reviews are
intended t o c o m p l m n t and bring up to date
the
reviews
contained i n t h e earlier study of hterson (1973) and provide
tables s m r i s i n g t h e cost e s t i m t i o n p r o c e d u r e s e q l o y e d .
Supplemntarytables
and c a r p u t a t i o nd e t a i l s are contained i n
Appendix A-2, while Appendix A-3 containspreliminary results
o ft h ef i t t i n g
of g a m m p r o b a b i l i t yd i s t r i b u t i o n s
to several
sets of accident cost data. 'his work has the aim i n t e r a l i a
of providing a generalised basis f o r c l a s s i f y i n g a c c i d e n t costs
i n t o i n j u rsye v e r i tcya t e g o r i e s
based on probability
of
occurrence.
4
2.1
'rm
CONCEPT OF SOCIAL
COST AND ROAD ACCIDENTS
It is the purpose of thisdiscussion
to provide a n o u t l i n e and a
briefdiscussion
of theconcept
of s o c i a l cost, especially i n r e l a t i o n
t o road accidents.
F i r s t it is evident from the titles of this and otherstudies
cost", "societal costs", and "socioconcerned variously
with
"social
economic costs" that these term imply an attempt t o d i s t i n y i s h between
social costs and conventional financia2 costs within
the
economic
I n f a c t there are three d i s t i n c t concepts of cost which
system.
overlap, and are bth a source of confusion and s a controversy i n
publicsectordecision-making,
namely financia2 costs, "real" economic
or resource costs, and social costs.
Financial costs refer to the recorded transactions or accounting
costsassociated
with day to day r e c e i p t s and payments throughoutthe
e c o n m . Fkal economic c o s t s ( s a t i m e s c a l l e d "opportunitycosts")
are
an attemt t o lneasure t h e value of scarce resources produced and
c o n s m d i n the econmy, as r e f l e c t e d i n national incmeaccounts.
The
term "real" is used to mean that the costs are masured i n "constant
dollars" so that t h e f f e c t s
of changes i n the generalpricelevel
between timz periods is eliminated.
Such economic costs exclude some
financialtransactions,for
example the sale and purchase of land or a
wed motor vehicle is a f i n a n c i a l cost to the individual or firm, but is
no~tconsidered t o be a real resource cost i n economic term, since no
new scarce resources are produced or c o n s m d i n the transaction(other
than t h e services of sales people - which docount) and it is excluded
f r m economic costs as a transfer of existing resources.
The concept of s o c i a l c o s t is mre embracing and mre d i f f i c u l t t o
ineasure. Social costs
and b e n e f i t s - r e f e rt o
the value of those
goods and services
generally
provided
by the
public
sector of t h e
econow (i.e. by g o v e r m n t s and publicauthorities)
which have t h e
characteristics that monetary values
for
these comnodities m y not
always be established i n conventionalmrkets,
and t h e i r costs and/or
b e n e f i t sm y
not k confined to the
producer
or consumer.
In the
econo~mes
efficient
resource
allocation
and
private sector of
d i s t r i b u t i o n is theoretically
achieved
by reaction t o p r i c e s and
the absence of c e r t a i nm r k e td i s t o r t i o nf a c t o r s ) .
c o n s m r demnd(in
'Ihe supply
of
goods and services which have "public goods"
characteristics such a s e x t e r n a l i t i e s or interdependencies
cannot
be
e f f i c i e n t l y achieved i n t hper i v ast e c t o r ,
and has become t h e
-
5
responsibility of the public sector of the econom, including t h e supply
of services of roads, education,
justice,
and defence etc.
More
recently the concept of undesirable
"spillovers"
of private economic
(e.g. associated
a c t i v i t y , i n the generation of environtwntalpollution
with the discharge of i n d u s t r i a l wastes t o air and water system;, t h e
noise and d i s r u p t i v e e f f e c t s of urban freeways,etc) has b e e n recognised
a s a further
category
of social
cast
involving
the
icposition
of
external ccsts upon society which are not fully m t by t h e producer or
consumr.
Public
'Wads" such as external costs iuwn s r c i e t y
which are notcarpensatedfor
in p r i v a t e m r k e t s are t h u s also seen as
the responsibility of t h e publicsector.(cf.
Musgrave
and
Musgrave,
1976) .
The costs associated with road accidents belong c l e a r l yi nt h e
sociaZ costcategory,
since theyincludeexternal
or "spillover" cost
e f f e c t s upon others, such as t r a f f i cd e l a y r, e s u l t i n g
fran peak-hour
caused by road accidents ambulance, p l i c e and
t r a f f idci s r u p t i o n
hospital services, and a l s o -use
they include a s i g n i f i c a n t canponent
of non-narket and intangible
costs
such as t h e pain and s u f f e r i n g
delays, and inconvenience t o Pamilies
a r i s i n g frcmaccidents,traffic
and thecomrunity.
However, because mny of thecostsassociated
with
roadaccidents can be m a s u r e d i n t e r n of financial transactions,
such
as vehicle repairs, there has Seen a tendency to focus on t h e financiaZ
c o s t s of roadaccidents,
and t o exclude from consideration any nonmnetary effects.
k r e recently the l i t e r a t u r e concerned w i t h valuation of road
accidents has turned t o the problen of e c o n d c valuation of accident
Faigin,
[1976];
and
the
discussions
in
e f f e c t s ( o i d e W S A , [1972];
Lawson, [1978]; and k n e y , [1978]).
The a t t q t
'
to
incorporate
intangible
and
othed
r i f f i c u lt o
masure non-mrket costs intoaccident cast estimtes h a s resulted in
san? controversy,
generally
w i t h the t h e that otherwise
reliable
estimtes of t a n g i b l e f i n a n c i a l e f f e c t s
of accidents are depreciated by
the addition of canponents with less acceptable or less reliable
values.
There has also k e e n
controversy as t o t h e appropriate
valuation of purely economic costs of accidents. SOm consideration of
t h e economic concepts of cost and value m y be helpful.
Fconanic a n a l y s i s is largely concerned w i t h dollar
values
are
determined i n markets and s i t u a t i o n s Mnere goods and services
exchanged.
me conceptof oaZue i t s e l f h a s several econcmic aspects i n a d d i t i o n t o important non-economic d h n s i o n s . It is not a unique or
absolute masure, but f o r convenience, economic analysis is m t l y
confined t o value-in-exchange,
and t h e resultant p r i c e s frcm
such
of cost has several
exchange (cf. baser, 1947). Similarlytheconcept
aspects, k t costs determined i n mrket exchange are ofmost
interest.
A cost is viewed as a 106s of P ~ S O U P C ~ S(or money, a s a claim on
resources) to obtain s ~ n eequivalent
gain
through
exchange.
nus, a
market p r i c e my sinultaneously
represent
a cost (loss) and revenue
(gain) i n exchange.
6
k o n a n i c costs generallyrefer
t o opportunitycosts,
which simply
mean the
resource
costs involved i n producing any ccmnodity or
service.
Fconomic accounting i n national income and e x p n d i t u r e
exarnple, is intended to show i n resource-cost terns the
accounts,for
total flow of goods and services produced within the tim span and
nation or area covered.
It does n o t necessarily
correspond
with
financial accounting and records. For example, it excludes the sale and
purchase of a l l "transfer" itens.
The relevance of opportunitycost concepts t o t h e presentstudy is
a s follows:accident
"costs" are here t e m d s o c i e t acl o s t s ,
Eaning
t h a t they
include
social and ecOnOmic costs associated w i t h road
accidents. Social costs are here defined t o rrean measurable c o s t s (or
benefits) which m y not be valued i n a conventionalmrket ( i . e . social
cmts m y involvenon-mrketedresources).
They are generallyexcluded
t h e literature relating to accident costs
from nationalaccounts.In
m y writers, including t h e Australian studies of Troy and Wltlin (1971)
and Paterson(1973), have argued f o r a n economic d e f i n i t i o n of accident
costs comparable with national accounting
definitions.
Others
have
aggregated f i n a n c i a l c o s t s and revenues associated with accidents o n l y ,
which f a l l ss h o r t
even of national
accounting
concepts
( c f . Reynolds
[ 19561 ; Japan [JRCTP 19781; &son [ 19751)
me view is adopted here
t b t t h e use of national accounting concepts
as a basis of accident cost
valuation,
although
not
incorrect
i n any technical
sense,
is not
appropriatefor
accident costs because it m yo f t e n
be misleading i n
agplication.
However t h eu n c r i t i c aul s e
of financial costs alone is
considered t o be generally misleading and is rejected. lhe basis €or
t h i s view is that national income cost concepts are mre relevant as
rreasures f o r predominantly
m r k e tcem
dnodities
and resources,
especially those within the p r i v a t es e c t o r of the ecnnoxy, and are not
adequate for the formulation of accident costs and evaluation policy.
.
The i n c r e a s i n g importance of thepublic
sector of t h ee c o n w ,
which is responsible f o r the allocation of resources to roads, defence,
s a f e t y , and other "public goods" areas outside t h e conventional m r k e t
exchange system, has necessarily led t o the d e v e l o p n t of more refined
concepts of social cost and b e n e f i t t o aid public d e c i s i o n a i n g a b u t
resource allocation to and within these a c t i v i t i e s without the guidance
cost analysis, and mre
of conventional m r k e t values.
Benefit
embracing system-planning
techniques,
such
as multiple objective
planning, have been developed to overcome some of the problem of
guidingresource allocation i nt h ep u b l i c
sector of the econoqy. lhese
of nontechniques a l l seek t o simlate thevaluetothecarrmnity
mrketed
benefits
and costs accruing from the
service
of roads,
i r r i g a t i o n water, and social costs such as i n d u s t r i a l and noise
pollutionfor which losses are incurred by sorre members of society but
no c m e n s a t i o n is paid by t h e mrket n-echanisn.
Costsversusbenefits:
An important problem which arises fran the
use of estimtes of road accident costs, apart from t h e issue of cost
concepts, is that of t r e a t i n g changes inaccident
cost l e v e l s as a
measure of benefit, for example inevaluation of safety msures. ?he
practice of i n t e r p r e t i n g changes i n cost l e v e l s as a masure of economic
and social benefit is q u i t e c m n i n mny applications of k n e f i t - c o s t
7
a n a l y s i s , especiallyintransport
economics where theprincipalbenefit
f r m investment i n improved roads is a reduction i n user operating and
travel t i m costs.
In fact, transport
evaluations
also atterrpt t o
of the benefits of improved
measure the contribution to project benefits
road s a f e t y through reductions in .the frequency and cost of accidents.
While various
financial
and narrowly
defined
economic cost
e s t i m t i o q concepts m y iR i n t e r n a l l y consistent and usefulformny
w i t h road accidents, these concepts
c q a r a t i v e purposes
associated
b e c m q u i t e inadequate when t h e r e q u i r e m n t of the cost ~neasuresis to
estirnatetheknefits
to s o c i e t y r e s u l t i n g f r a n changes i n ?;he l e v e l of
accidents or the risks associated with road transport.
Here the
objective
is t o mas;ire society's
valuation
of road
s a f e t y , 33' i n other words, tile aggregate
demnd
for
road safety.In
social k n e f i t - c o s t a n a l y s i s , it is necessary t o masure the wiZZingness
t o pay f o r s m given l e v e l of changes i n road s a f e t y :
t h i s is a
concept which g w s beyond themrketpricevaluationsconsistentwith
as a
national
accounting
(and for which, cost estimtes m sye r v e
reasonable proxy f o r m r k e t v a l u e ) ,
i n t h a t .it incorporatestheconcept
cf consumer'ssurplus
i n a d d i t i o nt om r k e tp r i c e .
b t benefit+ost
evaluations of publicprojects masure social benefits as the change i n
consuer's surplus r e s u l t i n g frm introduction of theplanconcerned.
Theoretical welfare economics h a s reflned the concept of consumer's
surplus a s a masure of social gain sowwhat further,
and 1% defined a
more correct
concept
of
covpznsating u a p i u t i o n as tne generally
masure. For
most
practicalapplications,
however,
appropriatebenefit
t h e difference between ccnyensatmgvariation and consmer's surplus is
not s i g n i f i c a n t , and it is considered acceptable t o use t h e latter
masure.
While the
theoretical
issues
underlying these Easures of
benefit are beyond t h ed i r e c t scope of t h i s report, they are r a i s e d t o
ernphasise that acceptable and consistent accident c o s t estinates are not
are t o be
l i k e l y to myt the mre stringentcriteriarequiredifthey
interpreted as b e n e f iEt a s u r e s
( v i d e Iksgupta and Eearce, [1972];
Harrison, [ 19741; Layard, [ 19721; and Nishan, [ 19711
and
[197G]; and
Winch, [1971] ; fordiscussion
of t h e benefit E a s u r e t a - n t criteria f o r
social k n e f i t - c o s t a n a l y s i s ) .
The mre importantofthese
issues are discussed at greater length
i n t h e reviews of recent cost estinrationstudies
( se e Appendix A-1,
especially lawson,
[1978].
4 m j o r problem encountered ipnr a c t i c e
which influences theperception of thevalue of road s a f e t y is t h a t t h e
infonration on accident costs is rmch mre amenable to r r e a s u r a n t than
the nonl-narket demndside of accidents and roadsafety.
&st empirical
accident cost studies, including the
present r e p o r t , haveproduced a set
of accident social c o s t estimtes, despite an
acceptance
the
of
t h e o r e t i c a l l yc o r r e c t demndvaluationconcept,
namely thevaluation of
changes i n accident risk, which has so f a r proved t o be e x t r e m l y
FJ
d i f f i c u l tt o masure ( c f . F r e a n , [1979];Jones-Lee,[1969]
[ 19741; and Willians, [1979].
8
2.2
FmNObiIC (XINSIDERATIONS AND T I E VALUE OF LIFE
The following
discussion
presents
a brief a s s e s m e n t of t h e
t h e o r e t i c a l and conceptual i s s u e sr e l a t i n g t o attempts to value h m n
l i f e . As noted by
ibboney
(1978), rmt research i n t o the valueof human
l i f e and suffering has been concerned with mortality rather than i n j u r y
andnon-fatal
illness. The need t o considervalues and costs r e l a t i n g
to the value of l i f e and h e a l t h which a f f e c t resource allocation arises
n o t onl-y i n the context ofroadaccidents
and public health s e r v i c e s
( v i d e k n e y , op.cit., Mushkin, [1963],but mre r e c e n t l yi n the policy
area of environmental q u a l i t y , t h e need t o evaluate the e f f e c t s of
health (uide
proposed p u b l iecn v i r o m n t aslt a n d a r d s
upon
community
Freemn,[1979]) lm raisedsimilarissues.
?he valuation of l i f e probletn arises because, whether occurring
e x p l i c i t l y or implicitly,public
and privatedecisions
are mde which
affectexposure of t h e population to r i s k of death and injury. Public
investment i n roads and private investment in
and use of motor vehicles
is just one example.
Otters include
public
health and hospitals,
building
and
industrial
regulations,
crime
prevention,
and mre
r e c e n t l y , e n v i r o m n t a l protection standards (e.g. f o r air q u a l i t y ) .
Tne smio-economicobjective
i n valuing l i f e and s u f f e r i n g is t o
help detennine how much s o c i e t y is prepared to spend on policies t o save
l i f e or reduce risk, m r b i d i t y and injury. Fconomic analysissuggests a
need for consistency i n measuring the opportunity cost of l i f es a v i n g
measures, t h a t is, l i k e lives
should
be valued alike.
The n o m 1
economic concept t o be applied is t h a t t h e c o n s i m r is t h e best judgeof
h i s own u t i l i t y , andof
changes i n w e l f a r e , as measured by canpensating
variation.
This is t e m d the consumers'
sovereignty
approach t o
economic d e c i s i o n 4 i n g (i.e.
t h e consmer
preferences
always
This concept is sometines considered
questionable
i nt h e
dominate).
context of valueof
l i f e , since it involvespriorassumptions
that t h e
risk and
c o n s m r canand does W e whatevervaluejudgemntsinvolving
the value of h i s own l i f e are necessary i n a r a t i o n a l way.
risk, especially stall changes i n risk, is
Individualperceptionof
often held to be d e f i c i e n t , and m yl e a d
t o no resultant change i n
k t t h e c m n i t y as
behaviour i n the face of s l i g h t l yi n c r e a s e dr i s k .
a whole m y respondpositively to any increased risks and for example,
throw its elected g o v e r m n t s has reacted t o increasedmadaccident
f a t a l i t i e s by enforcing
vehicle
safety
regulations
which restrict
p r i v a t e choice, including compulsory wearing of seat belts and
motor
cycle h e h t s . In t h e language of t h e economics of publicfinance, road
s a f e t y appears to have become a merit good (i.e. a c m d i t y notsought
by individualpreference,
but desired by t h e c m n i t y as a whole i n
apparentcontradiction
of individual choice including restrictions on
seat belts, liquor sale, and certain drugs etc.).
In t h e literature it is possible to distinguish t h r e e approaches t o
the value of l i f e p r o b l e m , namely:
( i ) wittingness t o pay, e.g. for r e d u c t i o n i n r i s k ;
( i i ) hman c a p i t a t , o r t h e lost f u t u r e income approach to
and t h e
individual productivity;
9
( i i i ) i m p l i e d uatue of t i f e , implicit i n publicdecision
making.
Frm a technicalpoint of view thewillingness t o p y approach is
consistent with consumrsovereignty.
It requires t h e c o n s m rt o be
willing and a b l et o choose k t w e e n a. reduction i n r i s k (e.g. of death,
and/or
serious
injury)
and a paywnt which would leave his welfare
This conceptnight
have
unchanged, temd his c u r p e n s a t i n g variation.
practicalapplicationfor
example, i n s i t u a t i o n s where the probability
is r e l a t i v e l y mil.
In
of accidental
death
frrxn v a r i o u as c t i v i t i e s
such instances a conceptual"value
of hunan l i f e " mnlltiplied by t h i s
probability would provide a n estimte of t h e max-imum mount a n
individual would 'e prepared t o p a y for
the
snrall
increase
in
probability of survival. 'this muldapplyonly
when the risk concerned
was ,mall (e.g,
10-5 ) ,
If t h e probability of death
increased,
for
s.xample t o 50 %, then individuals would rapidly i n d i a t e i n f i n i t e v a l u e s
of l i f ei ,n d i c s t i n tgh a t
this valuation scale is non-linear.
There
exists a further problem t h a t theoretical relationships (and, a l s of o r
based on classical or
exawle,
accident
reduction
projects)
are
objectiveprobabilities
generally drawn from frequencydistributions,
while t!le actual perception and behaviour o.€ individuals with r e s p e c t t o
r i s k is based on subjective assessmnts of probability. Zkme writers i n
t h i s f i e l d , i n c l u d i n g Mishan (1971) considerthatthesubjectivemasure
of risk is the correct one for use i n economic valuation, even i f it is
subsequently shown to k wrong, for example when t h e i n d i v i d u a ll a t e r
r e g r e t s h i s choice a f t e r a n accident.
Others
(cf.
W n e y , 1978)
judge thebehaviour
of an
consider that although Mishan mycorrectly
that g o v e r m n t s will
individual, he overlooks the mrit g o x i issue:
iatterrpt t oc l o s e
any such gap ' e t w e e n expected and reali.=d u t i l i t y .
Such 'oehaviour suggeststhat 17er11aps individuals i n thesecircumstances
are prepwedto
k directed as t o what t h e i r conpensatingvariation
should be, as i n the example of legislationforculpulsory
wearing of
seat belts.
" t ~ e h m n capital approach h a s a long history i n which one
p a r t i c u l a r concept of the value of l i f e is equated
with
individal
prociuctivity i n the form of a discountedstrean of futureearnings.
In
thecontext
of thepresent
report t h e i n c m stream c o n s i s t s of t h a t
portionoccurring
k t w e e n the time of accidentaldeath
and n o m 1 l i f e
expectancy.
Such a calculation c a n h? expressed as gross l i f e t i m e
earnings, or n e t of future consunption.
I t is generally
considered
( a f t e r Mishan, [1971]) t h a t the grossearnings figure is appropriate for
propxed
"life-saving"
projects.
Vet values are 0nl.j relevant for
determining the h i s t o r i c a l or ear post e f f e c t s of past accidental deaths
on society i n national i n c m t e r n .
The rminshortcoming of t h e h w n c a p i t a l approach is t t l a t it does
not include any measure of consumrsurplusover
and a b v e t h e i n c m
neasure, as does the willingness-to-pay method. It thus yields a value
of lost output which represents only the minimun e s t h t e of what
s o c i e t y should be willing t o pay t o saveanaverage
l i f e . To capensate
for this shortcaning SUE researchers have added subjective estirmtes of
the costs of pain and suffering and grief to their estimates of lost
Output from accidentaldeath
( c f . kwson,[1967]).
Such values m y be
considered"non-econdc" i n t h a t they are notincorporated i n n a t i o n a l
i
~
i
1
I
10
i n c m m a s u r e s , but they clearly affect socio-czonorrdc v e l f a r e i n a way
which should i n principle be quantifiable.
Other problem of the hunan c a p i t a l approach include
misleading
e f f e c t s of age and l i f e - c y c l e characteristics, f o r example, i n net
i n c m calculations
children
and the
elderly
can be shown t o have
negativepresent worth. Such measures are inappropriate and can lead t o
absurdinterpretations
of social preferences.
Also the problem of how
t o attribute an appropriatevalue
t o housewives' services needs to be
f r a n nationalaccountingvalues
is
resolved.Clearlytheirexclusion
sanewhat arbitrary, since
in
the
Australian
Phtional
Accounts
an
estimte for the imputed value of owners'house
rent is included,while
although
the
t h e imputed value of housewives' services is excluded,
latter services are c l e a r l y of value t o both
households
and t h e
cmnity.
The implicit value of l i f e approachsuggests that a study of past
public sector d e c i s i o n a i n g associated with l i f e s a v i n g or accident
r d c t i o n a c t i v i t i e s shouldreveal a d i s t r i b u t i o n of realistic implicit
values of l i f e . For example, i n respect of road safety,publichealth,
fire prevention, and building regulations (cf. Green and Brown, [1978]),
it is reasoned that the implicit values of l i f e , or t h e man of a
d i s t r i b u t i o n of such values, can be interpreted as the appropriate
"public sector value".
This revealedvalueshould
be consistent both
within and between d i f f e r i n gp u b l i cp r o g r m sI.df i v e r g e nvt a l u e s
then
s m investigation is
above or below the man are encountered,
c a l l e df o r
to evaluatethelikely
over- or under-investment i n these
areas. ?hisapproach
is advocated by Harrison (1974) and b n e y (1978)
i nt h e
U.K.,
who proposed appropriateconditions
under which f u t u r e
investmntdecisionsaffectinglife-saving
p r o g r m s could be v a l i d l y
based on a revealed"public
sector value" of l i f e .I np r a c t i c et h i s
estimte might show an excessively wide range ( h n e y cites implicit
values of a s i n g llei frea n g i n g
from $40 million from building
regulations to less than $100 associated with c e r t a i n medical diagnostic
testing).
Court awards for compensation m ya l s oo f f e r
s a w evidence of
implied valuations of l i f e but i n t h e UnitedKingdm
it is considered
that court awards as a r e s u l t of death are often misleading because they
r e f l e c t onlypecuniarylosses
to r e l a t i v e s ,r a t h e rt h a n
a n atterrpt t o
appear
value the l i f e lost. Bowever court awards for non-fatal injuries
t o be mre appropriate and mre recently have incorporated some
assessnentfor
p a i n and suffering.
k g a l awards mynevertheless
be
d i s t o r t e d by any a s s i g m n t of cuZpabiZity. Other problem concern t h e
scm m y be young, SCOE old; some
non-hcmogeneity of l i v e s saved(e.g.
myrecover, SCOE be incapacitated)/.Impliedvalues
my also show that
one f o m of death or r i s k of the S ~ J W
society m y i r r a t i o n a l l yf e a r
as f i r e mre than road accidents).
probability mre than another such
From the viewpointof
economic theory, both the implicit public
measure r e f l e c t some lack of
sector
value,
and the h m nc a p i t a l
consmerpreferenceorientation.Schelling
(1968) noted thisdeficiency
one should
and reasoned that to evaluate a p r o g r m to savelives
ask". .what is it worth t o those who benefit frcm it?. . I ' Mishan (1971)
..
.
..
further developed t h i s approach. In e f f e c t it argues that c o n s u e r sd o
not have a unique value of l i f e , but insteadvalue mall changes i n the
r i s k of death.
The problem remins of tlm to evaluate
these
srall
changes i n risk.
Mishan proposes an
unsentimental
categorisation
of
valuation
purposes:
(i) d i r e c t and voluntary;
(ii)
these risks for
d i r e c t and involuntary;
( i i fi i) n a n c i a l
r i s k t o others; and (iv)
psychic risk to others.
I n theory, the h m n capital approach and reduction i n risk rethcds
should he equivalent, but implied values of l i f e m y r e f l e c t past nonpublic
decisions..
In practice m n y other
r a t i o n a l or inconsistent
issues are important, for example if national i n m masures are
adequate f o r m n i t o r i n g the particular public program objectives, then
the h m c a p i t a l approach is acceptable, but i fp o l i t i c a l
judgerrents
are t o be refined i n allocating resources t o social objectives, then t h e
implied value approach o f f e r s a nethcd of achieving consistency.
.An ethical approach tosocialwelfarefavours
the i n d i v i d u a lt o
judge his own value of reduced r i s k , k t e r r o r s of perception and
l i m i t a t i o n s i n t h e a b i l i t y t o masure r i s k limit its application. Also,
external social costs need t o be accounted f o r i n any such a s s e s m n t .
A generalconclusion a r i s i n g frm this discussion of approaches t o
the social valuation of l i f e (and serious injury) is that i f road
safety,inthe
form of accidentreduction, is considered t o he p a r t l y a
w r i t good, then the h m n c a p i t a l approach t o valuation of l i f e is
favoured,
supplemnted
by m raep p r o p r i a t e
estkmtes of non-market
w i t h t h e concept of nations1welfare which is t o
valuationsconsistent
be mximised i n any evaluation ! d e l .
2.3
A KNIEX OF RECENT ACCIDDlT COST STUDIES
Estimates of the cost of r o a d accidents a t a national l e v e l i n m n y
of the developed economies have been produced since t h e 1950's,
r e f l e c t i n gt h es i d e e f f e c t s
of therapid
grawth a f t e r 1950 i n mtor
vehicle ownership and usage i n thosecountries.
However, i n t e r e s t i n
the econcrnics of r i s k and s a f e t y precededaccident cost work, so that a
theoretical framework w a s available, a t least in principle, for accident
analysis.
Tne purpose of t h i cs r i t i q u e
is to review the principal cost
estimtion studies undertaken i n t h e decade of t h e 1970's.
These
s t u d i e s , seven i n nunber, are considered at greaterlengthin
Appendix
A-1 of t h i s report. Pgpendix A-1 is designedboth
to complewnt and t o
reviewsofnethcdology
i n accident cost
bring up t o date thedetailed
estimtion between 1950
and
1971 contained inchapter
3 of Paterson
,SXE eleven cost
(1973), and includes a review of t h e Fatersonreport.
e s t i m t i o ns t u d i e s
a= reviewed i n t h e Paterson report,including
the
early work by k y n o l d s (1956) and subsequently
Bwson
(1967) i n the
United Kingdom, SCYIE e a r l y s t u d i e s i n t h e U.S.4., and several Australian
studiesincluding the work of Thorpe (1970) i nV i c t o r i a , and the rrajor
study i n the Australian &pita1 Territory undertaken by Troyand
Eutlin
(1971).
12
Tne literature on the valueof
l i f e and its relationship to road
s a f e t y , and public sector decision-making is nOw q u i t e extensive.
Writing i n 1956 Reynolds, i n making one of the f i r s t o f f i c i a l estirmtes
of the costs of roadaccidents
i n t h e United Kingdom, c i t e do n l ys i x
r e l a t d works. In the presentstudy a selectivebibliographycontains
more than f i f t y such references, and an exhaustive l i s t i n g would be rmch
larger.
Fortunately,
however, t h e nunber of innovative contributions is
cost s t u d i e s
much smller than the latter list, and currentaccident
generally follow a c a n p a t t e r n .
It is evident,nevertheless,
that
s m of the m i nd i f f e r e n c e s
between conceptual issues and empirical
s t u d i e s i n accident cost estimation have not been resolved, a f a c t which
is r e f l e c t e d i n continuing controversies i n t h e literature.
'his review s m r i s e s seven
accident
cost estimtion studies,
nanely:
Fox e t a t . (1979), Australia; Lawson (1978) a n a d a ; .Japan
(JRCTP, 1978);
Sherwin (1977), New Zealand;
Faigin
(NHTSA 1976),
U.S.A.;
Paterson
(1973),
Australia;
and NHTSA (1972), U.S.A.
Reviews
of most other mjor cost e s t i m t i o n work up t o 1971 c a n be found i n
hterson (1973),chapter 3.
et
The h k l b u r n e (Australia) roadside pole c o l l i s i o n study byFox
a t . (1979) i s by intention a partial coverage of road accidents i n a
metropolitan area, but i t is a study of mjor size and significance
cost data and developedand
which, i n t e r d i u , collected andanalysed
cast estimates i n benefit<ostevaluations
of
applieddetailedaccident
proposed accident
countermeasure
progrannes.
M t e r reviewing the cost
e s t h t i o n literature, Fox e t at. adopted
the
sccial accounting
framework of accident costs contained i n t h e U.S.A. studyofmigin
(1976).
However, because of the apparently
unresolved
controversy
in
t h e literature as to themthod
of valuationof
foregone income of
accidentvictims,
three sets of averageaccident costs were presented:
current resource costs onZy;
total resource costs net of consunption
expenditure; and total resource costs. All three bases of unit costs
were disaggregated by i n j u r ys e v e r i t y
classes, using a standardised
Abbreviated
Injury
Scale
(AIS) developed i n the U.S.A.
( v i d e '15tble
38 ,p.94). Fox e t a2 collected a wide rangeofdatafrcm
some 879 pole
a c c i d e n t s ,i n c l u d i n gs e l e c t e dm d i c a l
and propertydamge
infomtion.
of
particular i n t e r e s t is the
recognition
that
the
considerable
v a r i a b i l i t y i n accident characteristics required a c l a s s i f i c a t i o n of
ccsts by i n j u r sye v e r i t y .
&cause
accident
data
was not c l a s s i f i e d
according t o injury severity, Fox e t a t . f i t t e d t h e k l b u r n e sample t o
the U.S. - derivedAbbreviatedInjuryScale.
.
The hklburnestudy
produced e s t i m t e da c c i d e n t
f r a n pole accidents s m r i s e d as follows:
Average Cbsts
Costing&thod
per
Per
&rent resource costs:
Total Cbsts n e t of
8,200 consmption:
T o t a l Costs:
a c c i d e nf ta t a t i t y
T
$
3,400
23.1204,600
11,200
costs r e s u l t i n g
Total Cbsts
per annum
w
6,800
7.0
118,500
16.9
13
The study suggestedthat
withannual
social costs of $23 million
per year i n b k l b u r n e r e s u l t i n g f r m pole accidents, a range of accident
to knefitcountermasuresshould be developed and evaluatedaccording
Fox e t a2. a l s po o i n t e tdtohceo n f l i citn
social
cost criteria.
objectivesevident
i n the approach of s a publicagenciesresponsible
forroadside poles: by i n s t a l l i n g o r r e p l a c i n g wles of a non-breakable
costs t oh e
agency of
("non-frangible")
type
t o avoid
average
approximately $300 per
accident,
total
average
social costs of over
$11,ooO are incurred.
The papzr Lawson
by
(1975) contains a set of accident cost
estinates f o r Canada for the
years
1 Y 7 G and 1978, together
with
an
overview and assessment
of
the
theoretical
literature relevant
to
accident cost concepts and estimtes.
Lawson pointsouttnat
road s a f e t y is a s i g n i f i c a n t ccmponentof
b t h public and private expenditure p r o g r m s and therefore the need t o
evaluate the allocation
of resources t o s a f e t y is inescapable. Although
benefit-costanalysis
is i n p r i n c i p l e a s u i t a b l e m t h o d f o r
such s a f e t y
evaluations, Lawson argues that t h e use of estimated c o s t s incurred
through
accidents
instead
of the
theoretically
correct
concept
of
willingness t o pay foraccidentreduction
as a m a s u r e of p r c g r m
benefitsgreatly weakens theeffectiveness
of such evaluations.Social
accounting cost estirrates can onlf consist o f minimum estimtes of the
t r u e l e v e l s of cost which society is willing t o expend i n order t o avoid
accidents.
This importantconclusionappliesnotonlytobenefitxost
f o m r can i n
applications, but also to cost-effectiveness analysis (the
p r i n c i p l e rank accident p r o g r w s withdifferingeffects,
while t h e
latter is appliedtodeterminethe
least-mt m t h c d of achieving some
s i n g l e accidentreductionobjective).
In h i s review of s m twelve cost
s t u d i e s Lawson rejects t h e n e t i n c m concepts used i n e s t i m t i n g
foregone
future
i n c m of a c c i d e nvt i c t i m( e . gi.nt h es t u d i e s
by
Reynolds, [195G];
Lawson, [1967] i tnh e
U.K.;
and
Troy
and
ktlin,
[1971]; and h t e r s o n , [1973], f o r Australia). IIe cites K s h a n (1971)
as resolving t h i s conceptual
debate
by a s s e r t i n g that t h e ez ante
concept of inc-ome, involving a d e f i n i t i o n of society including the
accidentvictim
- (i.e. it includeshis i n c m and consumption) is t h e
appropriate basis of masuremsnt i n assessingforegone i n m r e s u l t i n g
frm accidents.
The n e t i n c m concept measures only what happens t o
national i n c a a f t e r an accident:
it does
not
provide
correct
worth of preventing
information t o d e c i s i o n d e r sa b u tt h et r u e
accidents before their occurrence.
The problem of m a s u r i n g i n t a n g i b l e losses r e s u l t i n g frcm road
accidents is less t r a c t a b l e p, a r t i c u l a r l y
that of pain and suffering.
Lawson concludes that wasuremsnt of thepsychic
and m t i o n a l losses
due to
accidents,
although
a valid social cost, so f a r has eluded
that r e m i n i n g "luss accounting"
acceptable
quantification,
ensuring
methods w i l l produce only minitnun cost e s t h t e s which require m c h
caution i n t h e i r a p p l i c a t i o n t o s a f e t y programre evaluation.
14
Minimun cost estirmtes for Canada for threeaggregattdcategories
of cost only are p r d u c e d i n t h i s paper:
CostCategory
Lost production:
( income)
care:
Average Cost
per accident
c$m
Tot a1
2,500
c$m
180,000
(per fatal)
900
200
100
1,500
1,500
-
4,500
__
Health
PrOpefiY
( a n d mnpensat ion
administration) :
T o t a l Cost
~
The Japan Research Centre for Transport Policy (JRCTP) produced
estixates of "social losses" f m roadaccidents i n Japan for the year
1974.
The JEP study
adopted
a loss accounting
approach
based on
although
it recognised the v a l i d i t y of
"objective social losses",
"subjective" social losses including pain and griefarisingfranroad
accidents,
but
concluded
t h a t they
cannot
yet
be s a t i s f a c t o r i l y
nEasured.
The report then goes on t o estimte accident losses f o rt h e
following cost categories:
Cost Category
Lost n e t income
Average Cost
A$ *
Total Cost
A$m. *
34.3
39,02O$(per f a t a l i t y1), 4 0 2
Medical
470
Vehicle damge
310 (per vehicle)
1,372
Other costs +
170 (per
vehicle)
Total :
930
-(per vehicle)
554
I!
13.5%
33.5%
764
18.70
$4,092rn (10%)
1,770 (per accident)
*Yen valuesconverted t o A$ a t 1974 average r a t e . $ Insurance
administration compmlsed 85% o f o t h e r c o s t s . .d compares with
$9,440 per injury.
15
lhe J R C P study
followed
c l o s e l y the methods of tkwson (19G7),
although estirmtion techniques were o f t e n m d i f i e d by available Japanese
accident and cost data. ?he Japan studychose
t o employ t h e net income
concept i n the calculation of presentvaluesfor
lost f u t u r ei n c m .
nlis was achieved by deducting 4 6 of average adjusted income, t o
excludeconsunptionexpenditwe.
I n addition t o t h i s deduction,average
age group -were multiplied by the
workforce inccm
levels
for
each
proportioa of the workforce t o t h e t o t a l p o p u l a t i o n i n
each age group,
(i.e. workforce participation rates). lliis latter adjustment had t h e
e f f e c t of averaging i n c m s over the total population in
each age group,
a l l remkrs of t h e
and thus imputed t h i s lower i n c mf i g x et o
population whether
housewives,
unemployed, or employed.
Details of
t h e s e calculations are contained i n Appendix A-1.
?he e f f e c t of t h e s e
is t o reduce t h er e l a t i v e
size of foregone i n c m i n t h e
adjus-nts
total cost estirmtes.
Sherwin's paper presents SUE generalis4 t o t a l cost estimatesfor
rcad accidents i n rJew Zealand for 1975, togetherwith
an a s s e s m n t of
t h e e s t i m t i o n task. Total accident costs were estirrated a t k t w e e n NZ
$160 n i l l i o n and NZ $170 millionforthatyear,
or' which foregone n e t
inccm c m p r i s e d a b u t
and property darrage was a b u t 4%.
la,
M Zealand data sources are considered t o k notyetadequate
to
support mre detailedcost
estimates. Sherwin used a modified form of
t h e" l i f e
nodel" a p p r m h developed by Faterson (1973) a s a basis f o r
cmputingthepresentvalue
offoregone
i n c m s . This technique i ~ a sthe
total
population,
and
e f f e c t of averaging i n c m l e v e l s across the
allocates education costs t o the under 19 years age
group.
Sheruin
rejected t h e n e t calculation employed by Faterson t o avoidtheproblem
of interpretingnegativevaluesforchildren
and the aged. A discount
rate of 10% p.a. w a s assumxi i n a l l present
value
calculations
as
r e m n d e d by t h e Treasury L k p r t m n t for public sector evaluation.
Sherwin didnotquote
average accident costs and raised a most
s l g n i f i c a nf et a t u r e
of accident statistics, namely that
they
are
generally characterised by highly skewed d i s t r i b u t i o n s . ?his means t h a t
t h e use of a sknpleaveragecosts
is thereforeoftenmeaningless,
and
m y also be of limited usefulness i n public decision-making.
?he work of m i g i n (1976) resulted i n detailed road accident cost
estimates f o r t i e United States of b r i c a i n 1975,
produced
by t h e
? k t i o n a l Highway T r a f f i cS f e t y" m i n i s t r a t i o n
(NHTSA) withinthe
U.S.
&par-nt
of Transportation. The Faiginstudy completed a p r o g r m of
accident cost r e s e a r c hi n i t i a t e d
by t h e WTSA a f t e r its preliminary
estimtes of the societal cost of road accidents for 1971.
16
Total United States accident
AccidentCategory
cbsts i n 1975 are surnrrarised a s :
Average @st
U.S.$
Fatalities
I nj u r i e s
Property m g e u l l y
287,200
3,200
520
Total Cbst (1975)
U.S.$billion
13.4
12.8
11.4
37.6
lhe mjor feature of t h i s study is t h e developnent of an expanded
range of srxietal cost categories, each c l a s s i f i e d by s i x classes of
the "Abbreviated I n j u r y Scale" (AIS)
injury
severity
according
to
developed by NHTSA with
the
assistance of t h e American k d i c a l
Tnus a detailedrmtrix
of unit accidentcosts was
Association e t aZ.
produced (see Appendix A-1, t a b l e s 37 and 49) with theobjective, i n t e r
evaluations
of s a f e t y
a l i a , of f a c i l i t a t i n g soc,ial benefit+ost
p r o g r m s w i t h d i f f e r i n g e f f e c t s on injury severity.
-le
Faigin
study
involved
a thorough researching of available
accident cost data and other
relevant
studies
to produce w h a t are
probably t h e most detailed and r e l i a b l e estimtes of a c c i d e n te f f e c t s
and costs possible
given
the
limitations
of e x i s t i n ign f o m t i o n .
that there w a s only a slightimproveEntin
the cost d a t a
Faiginnoted
base between 1971 and 1975, and that a s i g n i f i c a n t improvement i n basic
sources w a s necessary to f u r t h e r improve unitcost
estimtes.
Other
f e a t u r e s of t h e 1975 studyinclude
the m i s s i o n of estimatesfor"pain
and suffering" costs which were included i n the preliminary estimtes
for 1971.
Treatnent of foregone income followed the usual form of
calculatingthepresentvalue
of average i n c o m l e v e l s ( f o ~ r median ages
w i t h i n eachagegroup).
Although separate mle and f e m l e income levels
by age groups were calculated, consumption expenditure was not excluded,
and t h e average income levels for those m k r s of the workforce w e r e
imputed t o a l l members of thepopulation
i n each age group. This i s
e q u i v a l e n tt o assuming an opportunitycost
income l e v e lf o r housewives
and those unemployed equal t o t h e average income of
those
i nt h e
it is
workforce. This procedure is also adopted i n the presentstudy:
considered t o be mre appropriate than t h e approachadopted
i n Paterson
(1973) and Japan (1978), f o r example, i n which t h e e f f e c t ofaveraging
workforce incomes across the total population is equivalent t o inquting
a zero income and opportunity cast to those not i n t h e workforce.
The extensive research i n t o the e f f e c t s of (non-fatal) i n j u r i e s
upon work time l o s t and levels of p e m n e n t i n p a i m n t i n the case of
mre serious i n j u r i e s , which was undertaken i n t h e courseof t h e Fkigin
study has produced the most r e l i a b l e and detailed injury cost estimates
presently available.
Expressed as average work days l o s t , or by degree
of
pemnent
i m p a i m n ft o r
each of t h e s i x AIS classes of i n j u r y s e v e r i t yt,h e
injury
severity
e s t i m a t ees m p i l e d
in
t h he i g i n
report
have been
adapted for use i n several other accident studies, including t h e present
study.
17
lhe u n i t cost framework outlined i n t h e F a i g i n report is considered
t o be t h e most appropriate and ccnplete example of the smial loss
accountingapproach
to road accident cost e s t i m t i o n . For this reason,
t h e u n i t cost f r m w o r k adopted i n thepresent
s t u d y closelyfollows
acknowledges t h a t the
Faigin.
However, t h e latter r e p ocr lt e a r l y
concept of s o c i e t a l loss goes k y o n d econonic welPare, and that adequate
quantification of a l l accident effects i n d o l l a r t e r m is t h e r e f x e n o t
possible. Because of these issues F a i g i n emphasised that the t o t a l cost
e s t i m t e s derived f r m theunit
cost f i g u r e s are simply indicators of
the mgnitude of the road accident problem,and
their USE as benefit
masures i n benefit cost analyses is limited because
they
are only
p a r t i a l masures of the willingness-tc-pay criterion.
by
lhe Australianstudy
of t h e cost of road accidentsundertaken
Paterson (1973) p r d u c e d national
cost
e s t k m t e sf o r
1969 f r m d i n
strict national accounting terms.
'Illis s t u a draws extensively upon
u n i t costs and otheraccidentrelationshipsestablished
i n t h e earlier
Australian study by Troy and Butlin (1971), although
t h e Paterson report
developed its okn " l i f e d e l " t o e s t h t e the n e t c o n t r i b u t i o nt o
of t h e c m m i t y , dividedinto
nationalinccm
of theaverwemnber
t h r e e agegroups.
lhe r e s u l t s of t h e B t e r s o n repxt e s t i m t e s are:
Cost Category
Cost per unit
$
Fatalities
Injuries*
Vehicle repairs
(per vehicle)
Other C o s t s
25,300
1,om
Total:
$1,010
(per accident)
220
210
-
Costs
Total
Sm
88.7
90.6
199.7
(
1
18.5%
18.5%
41.5%
101.4
___
__
21.1%
$480.3
lW"
?he report produced disaggregated estimtes for a proportion (68%)
of total accident costs for each state and t e ~ r r i t o r y for 1969, with
costs f o r each region classified according t of i v et y p e s
of accident
( c o l l i s i o n between vehicles;overturned,
or l e f t road;collision
with
p d e s t r i a n ;c o l l i s i o n
withfixedobject;
and a l l other accidents).
As w i t h the presentstudy,
the Bterson report e s t h t e d the total
number of read accidentsindirectlysince
onlycasualtyaccidents
are
recorded .in most Australian states.
Troy and mtlin found t h a t
a p p r o x h t e l y 90% of a l l a c c i d e n t si n t h e AustralianCapitalTerritory
didnotinvolvepersonalinjury,
and ratios for each accident class
Wis of Paterson's estimtes
derived fran the latter studyfo&the
of the t o t a l n m k r of accidents i n 1969'he basis of other cost
e s t h t e s i n t h eP a t e m ns t u d y
is o u t l i n e di n
Appendix A-1.
iYhat is
*
The t e n i n j u r i e s ' is used throughout the presentreport
to mean persons
as a result of road accidents (see also footnote on p.24).
non-fatally injured
18
sanewhat novel i n t h e Patersonreport
is the " l i f e &el'' basis for
calculating the present worth ( a t 5% discount rate) foregone
of
income.
lhis technique r e s u l t s i n an
average
1969 inccm loss or
"value" ol $25,300 f o r a l l a c c i d e n t f a t a l i t i e s , b u t
is further separated
i n t o non-pedestrians w i t h a "value" of $30,500, and pedestrians with a
"value" of $8,700.
The lower figure for lost inccme
of
pedestrians
r e f l e c t st h er e l a t i v e l y
highproportion
of olderpersonsinvolved
in
pedestrianaccidents:
i n t h e econanic l i f e cycleassmed by h t e r s o n ,
persons
over
65 years of age w e r e estimted to
"contribute"
a net
withdrawal of a b u t $1,600 per a n n m , r e s u l t i n g i n a negative "income"
f o r t h i s group. The use of discount rates above 5% i n t h e presentvalue
calculations of the h t e r s o n " l i f e model" (vide l h b l e s 44 and 45, p.100)
would change the net contribution of s m age groups under 19 years from
positive t o negative amunts.
these estimates and a s s m p t i o n s are consistent w i t h t h e
report,
namely that of e x post
c r i t e r i o n adopted i n t h e Paterson
agreernent with the n e t potential contribution of a c c i d e n tv i c t i m
to
national income, t h i s approach c e r t a i n l y does not c o n s t i t u t e a c m n i t y
value, or values,
placed
upon t h e lost l i f e and/or income, as
established by
Mishan
(1971), and discussed by
Lawson
(1978), h n e y
model" assumptions,
the
As a result of t h e s e" l i f e
(1978) e t aZ.
Paterson report is considered to greatly overemphasise t h er e l a t i v e
importance of property damage i n total accident costs, a criticism which
can a l s o be mde of the i n t e r p r e t a t i o n of accident costs by Troy and
Butlin.
While
Tne Bterson report also c o n t a i n s a comprehensive bibliography and
s e l e c t i v e review of accident cost literature up t o 1971, together with a
u s e f uclr i t i q u e
of accident statistics i n Australia.
Although cost
estimtes -were p a r t i a l l ya l l o c a t e d
by states and accidenttypes,
the
report concluded that availabledatadidnotsupportestjlnates
of t h e
separate costs of f a t a lc, a s u a l t y ,
or property darrage accidents.
It
a l s o noted the skewed costdistributionsrevealed
by Troyand
Wltlin,
which lircit the usefulness of average values.
The f i n a l cost estimationstudy
reviewed is t h e 1972 preliminary
Safety
Administration
report of t h e U.S. National Highway Traffic
(NIPTSA).
'his report defined its concept of accident costs as well
beyond financial and mterial loss accountable measures: it adopted t h e
p r i n c i p l e that quantifiable losses are experienced by the c m n i t y
have
values
established
i n mrket
regardless of whether they
transactions. As a result, t h e NHTSA 1972 reportincludedestimatesfor
the value of pain and suffering, c m i t y services, and losses t o
others, i n a d d i t i o n t o t h e usual mterial costs. 'he cost estimates for
the U.S.A. i n 1971 are s m r i s e d a s :
Total Cost
Cost per Unit
Cost Category
( 1971)
F a t a l i t y accident)
iper
Injuryaccident)
iper
Property Eamage Only
Total
accident)
(per
200,700
7,300
300
1,650
11.0 (24%)
27.6 ( 6 % )
7.4 (16%)
46.0 (10%)
19
'Ihe 1972 report
included
mre approxinrate estinrates of certain
costs including
"pain
and suffering" on the basis that a reasonable
estimte of t h i s mgnitude is p r e f e r a b l e t o its omission - which implies
a zero value f o r that item. In t h e view of t h e r e p o r t , t h i s p r a c t i c e
in
t h e past has possibly led s a f e t ya g e n c i e st od i r e c t
a greaterthan
of property darnage,
o p t d l amount of resourcestwardstheprevention
because of t h e d i f f i c u l t i e s experienced i n measuring f a t a l i t y and i n j u r y
casts. But, as i nt h e later m i g i n study, t h i s report Whasises that
f a t a l i t y and injury casts do not purport to establish a unique v a l u e f o r
hunan l i f e .
They c o n s i s t r a t h e r of mininral estirrates of s o c i e t y ' s
?he estirration work i n i t i a t e d
willingness to pay t o avoidsuchevents.
i n t h i s 1972 study was considerably advanced i n the later Faiginstudy
(0p.cit.).
20
A principal
objective
of this
present
study
is t o propose a
framework s u i t a b l feo r
the estimtion of the s o c i a l costs of road
accidents i n Australia, and t op r e s e n t a preliminary set of such cost
estimtes based upon existing
data
sources
consistent
with such EL
framework.
Together with thesepreliminary
estimates a discussion of
existing data sowces and their deficiencies w a s envisaged.
I n t h i s chapter a set of preliminary road accident cost estimates
for Australia i n 1978 is presented,containingbothaverageunit
costs
and t o t a l accident costs c l a s s i f i e d according t o the s p e c i f i c cost
f r m w o r k proposed.
It is convenient to consider i n turn
throughout
t h i sc h a p t e r the relevance of each ccmponent cost categorywithin
this
framework togetherwith
difficulties.
the e s t i m t i o n procedures and attendantdata
on t h a t of the
lhe cost framework adopted is predcaninantlybased
the
National
Highmy
Traffic Slfety
o
f
f
i
c
i
a
l
study
for
Faigin (1976)
Administration, of the U.S. k p a r t r w n t of Transportation.
This latter
study
developed
and published the most comprehensive set of unit
accident cost estimtes i n the literature, and i d e n t i f i e d sorre eleven
categories of "societal", or social and economic, cayts attributed t o
road accidents.
The h i g i n study also undertook d e f i n i t i v e research
i n t o the compositionof
u n i t cost estimationproceduresresultingin
a
seven class disaggregation of a l l accident costs according to injury
Adoption of t h er e s u l t a n t
cost m t r i x ,c o n s i s t i n g
of eleven
severity.
rows and seven columns, is considered to be scmwhat ambitious i n the
l i g h t of existing Australian d a t a sources, but its use as t h e basis of a
set of
preliminary
Australian accident cost estimates is considered
desirable t o r e t a i n t h e
most completeconceptual framework available. A
s i g n i f i c a n t b e n e f i t from the use of a cmnprehensive cost concept is that
it avoids the implicit judgeEnt that a valid cost category which i:s
excluded is a t t r i b u t e d a zero value.
Categorisations of social and economic cc*sts such as t h e h i g i n
cost m t r i x are intended t o go beyond the narrav loss-accounting
procedures of mny earlier studies. However, it is emphasised t h a t t h i s
present
study
both recognises and s u f f e r s from the r e m i n i n g
deficienciesinherentin
these cost f o m l a t i o n s as cited i n Iawson
21
(1978) and acknowledged by Faigin, namely that social c m t estimtes,
however cctrplete, can represent only minimum estimates of the t r u e value
society places upon t h e benefits fran accidentreduction.Thereforethe
use of t h e following cost estimates, however accurate, a s indicators of
the b e n e ~ f i t st o be gainedfranaccidentreduction
is s u b j e c tt o some
limitations.
In
the
present
s t d y the view is taken that it is
d e s i r a b l et o estimte unitaccident
costs according to i n j u r y s e v e r i t y
l e v e l s t o permit any such average cost levels t o be usefully interpreted
The support
for
this
view is t h r e e f o lfdi:r s t l y ,
and applied.
of any accident
application of these estimates itnhaep p r a i s a l
or policy
evaluation
usually
requires such a
prevention
project
d i s t i n c t i o n between t hsee v e r i t y
of accident
effects:
second, the
statistical d i s t r i b u t i o n s of most accident characteristics are generally
highly skewed so that the use of simpleaverages m y be misleading; and
f i n a l l y that t h e nature of t h e accident source data is such t h a t
r e l i a b l e empirical estirration procedures are frequently best achieved
by
undertaking estimtes of costs grouped by severity levels.
costs is contained i n
A r e l a t i v e l y large n u n k r of unitaccident
approach
adopted
i n t h i s presentation
the f o l l w i n g estimtes, and t h e
has been t o show both the e f f e c t s upon costs of differentconceptual
approaches, and a l s o of changes i n the l e v e l s of the key parmters of
the estimationmthod.
.4n overall judgement based on the experience of
tkis present
study
is that i n e f f e c t the minimun l e v e l of d e t a i l
r e q u i r e d t o develop a set ofeven preliminary accident cost e s t i m a t e s i n
this d e t a i l f o r A u s t r a l i a
is very high.
TABLE 1
11
AVERAGE ACCIDENT
COSTS BY INJURY SEVERITY L E V E L
AUSTRALIA 1978: PRELIMINARY ESTIMATES (10%Discount ?,ate)
C O S T
C A T E G O R Y
__
6
~
5
+i tica
POtd
__
I
1.
Foregone Income
Pamily,
2.
~
53,840*
34.950'
19,150*
T
__
~
2
4
rodemte
SeU2'e'
t
?roporty
s
s
8.000*
1.21c
650
365
195
15
s
.13,510*
comnity
__
~
I
~evel:
Abbreviated
Injury
Scale
(AIS)
I
-
50
LO88eS
I
3.
Hospital
670
36,000
11,900
7,lOC
1,900
150
4.
kdical
310
3,120
1,730
1,000
380
75
-
5.
R e h a b i l i t a t i o n etc.
800
3.300
1.320
560
235
50
.
R.
Leg*; 9
1.6.59
1,:00
a00
150
I40
7.
Insurance Adnin.
865
865
865
740
610
8.
Accident Inueetig.
200
260
200
I00
59
9.
Losee. t o others
2,200
%Urt
10.Vehicle
Omge
11. h a f f i c
Eelax
700
I20
60
IO
3,000
4.000
3,000
2,600
1.NO
1.400
60
I60
I 60
I60
60
__
157,085
33,685
57,175
~
-
~
-
14,755
__
5.790
~
1
1571 0 !
1.500
BO
TOTAL
100
1,402
- __
IO
350
I
I
i
2,270
620
22
In the followingdiscussion a consistent set of cost estirrates is
presented throughout based on, i n t e r d i u , the choice of a discount rate
of 10%per annum.
Alternativeresultsusingdiscount
rates of 7% and
cost
item,
but
detailed
results
using
13%=re obtainedforthe
rmin
A-2
(
v
i
d
e
Tables
49
t o 52).
Appendix
the latter ra.tes are contained i n
The f u n b n t a l set of u n i t cost estim%tesf o r road accidents i n
Australia,reflectingtheunderlyingframemrk
of cost categories and
levels adopted i n thepresentstudy,
is contained i n
accidentseverity
Table 1.
Table 1 shows a m t r i x of u n i t or average accident costs for eleven
cost categories, eachseparatedaccording
t os i xc a t e g o r i e s
of i n j u r y
s e v e r i t y , based on t h e AbbreviatedInjuryScale
(AIS) ( v i d e 'Pable 5) and
accidents involvingonlyproperty
darmge. !he unit cost figures shown
i n a b l e 2 i n italic type representdirectunit
cost estimates drawn
from o t h e r s t u d i e s ( m i n l y h i g i n ,
1976). !his s m r y m t r i x shows t h e
existence of a wide v a r i a t i o n i n u n i t cost conponents by accident class,
both i n total and across each row.
The average cost of a s i n g l e
i n 1978, compared with $2,270 for a minor
f a t a l i t y was over$157,080
injury and only $620 f o r an accidentinvolvingonlypropertydamge.
a mximum r a t i o of over250:l
between u n i t cost l e v e l s
Thisreflects
range
of cost l e v e l s associated w i t h road
indicating the extrem?
accidents. Within each injuryseverity
class, it is apparent that the
foregone inccarecomponent
cowrises a mjor proportion of f a t a l i t y and
serious injury costs (rangingfrcm 7% t o 49% respectively). Bble 1
SUMMARY O F TOTAL ACCIDENT COSTS: AUSTRALIA 1978
(10% Discount Rate)
CostCategory
T- -
Fatalities
T
Injuries
Propert)
Damage
Total
Bjor*
Minor*
420.5
im.
45.5
$m.
19.1
485.1
126.2
13.7
5.8
145.6
6.6
57.5
79.6
137.1
Vehicle Darnage
11.1
14.0
129.6
143.6
Other costs
17.6
11.2
61.2
582.0
141.9
295.3
Sm.
ForegoneIncome
Losses to Family
Comnuni ty
Hospital, Medical etc
TOTAL
Im .
322.9
-I"
437.2
143.8
477.6
339.0
,591.1
571.9
23
also includes s e v e r a l non-rket
cost estirrntes,including the valueof
l o s t in(and services) of accidentvictims
t of a m i l i e s and t ot h e
and
ccmrmnity, losses t o employers, and t h e cost of t r a f f i cd e l a y s
congestioncaused by roadaccidents.
?he derivation of each of t h e unit
cost estirmtes is discussed later inthischapter.&foreconsidering
i n mre d e t a i l , a set of t o t a l accident costs
these
unit
costs
(calculated by multiplyingthe
u n i t costs of Table 3 by t h e accident
nmkrs i n Table 5 below) is shown i n Table 2.
Table 2 shows that the e s t i m t e d totaZ cost of r a d accidents i n
Australia was over $1,590 million i n 1978; or equivalent t o nearly 2 %
of G r o s s k s t i c Product i n 1977/78(although,
as indicatedelsewhere
i n t h i s reprt, national income c q a r i s o n s are not
always
the m t
relevant for road accident costs). Foregone i n c m was t h e l a r g e s t cost
category cunprising 31% of t h et o t a l ,
followed by v e h i c l ed m g e
at
30%. k d i c a l and hospital costs combined were r e l a t i v e l y smll a t 9
percent of t h e t o t a l .
By category of accident,fatalitiesrepresented
37%, i n j u r i e s 27%, and vehicles
with
property
d a w g e only were 36
percent of total costs. Based on the s m r y i n 'Jhble 2, an abbreviated
set of u n i t accident costs is shown i n Table 3, both
according
to
average costs perperson (and per vehicleforproperty
damage only) and
per accident.
TAiJLE 3
-
SUMMA2I CF AVERAGE ACCIDENT
COSTS
(lo" DiscountRate)
1
I
T
Cost Category
Property
AUSTB.LI.4 1578
Injuries
MaJor
1
Mlnor
I
,
~
Oawge
Only
Total
~
I
-
$
Tzhsc::
i
? Z iT,YIZX
.25
Foregone Income
Vehicle Oamage
Other Costs
1
8.870
2i0
34,050
1,780
3,000
2.660
11,220
2,730
60
860
4.745
2 . i90
113,510
Losses t o Family,
Comnuni ty
Hedica1,Hospital
1,400
660
270
~
TOTAL
157,085
i i - P E E ACCIDENT
ForegoneIncome
LossestoFamily,
Comnun i t y
Medical .Hospital
v e h i c l e Damage
Other Costs
TOTAL
12,750
128,650
3,830
38,600
16.140 2,020
3.400'
5.380
2,860
178.090
1
I
27,670
620
3.190
4,470
$
300
90
3,930
3.150
39.8UO
1,230
2,010
2,010
950
__
1,060
4.580
6,420
670
510
610
24
The unit cost estimtes i n Table3
represents a consolidation of
a b l e 1 figures and need t o be interpreted
with
m caution,
not
because of u n r e l i a b i l i t y , but because c o n p r i s o n of Table 3 withthe
that average accident
average costs i n Bble 1 d a m n s t r a t e sc l e a r l y
casts change significantlyaccording
to any re*rouping
of accident
classes.
F r a n Brt I1 of this table it can be seen that while the average
cost of any accident in 19'78 was $2,860 t h e cost of a f a t a l a c c i d e n t was
$178,090, v e h i c l e 4 m g e + n l y averaged $1,180, and injury accidents
ranged f r a n $4,580 f o r a minor injuryaccident
to $39,800 f o r a mjor
injury
accident.
When the range of separate costs f o r these two
combined injurygroups
is compared with t h e average cost of a l l injury
that
such
averages
should
be
accidents of $6,420 it is evident
interpreted with caution.
A mjor source of variation i n these cost estimates is thechoice
t o cost categories of foregone
of discount rate. In Tables 1 t o 3 , the w
income and Zosses t o f&Zy
and c o m n i t y are s i g n i f i c a n t l y a f f e c t e d by
changes i n thediscount rate selected. A rate of 10%per annum is used
f o r t h e s e tables, but t h e e f f e c t s of increasing and decreasing this rate
fran 7% t o 13%respectively are shown i n a corresponding set of cost
t a b l e s contained i n Appendix A-2.
The issuessurroundingchoice
of the
discount rate are discussed later i nt h i sc h a p t e r ,b u tt h ee f f e c t s
of
lowering and increasingthe rate, f r a n 7% t o 13, can be surrPlrarised as
r e s u l t i n gi n
a range
of
$1,448 million to $1,850 m i l l i o ni n
totat
a c c i d e n t ' c o s t s , and a rangeof$138,000
to $250,000 i n the average cost
of a f a t a l a c c i d e n t , and a comparablerange of $36,000 t o 546,600 i n t h e
accidents.
Other cast categories are not
u n i t cost of mjor injury
rate although the averageoverall
accident
affected, by thediscount
c o s t shows a range of $2,600 t o $3,300.
3.2
OUTLINE OF O X T COW3NENTS AND DATA SOURCES
The estimtionprocedures and data sources for Bbles 1 and 2 are
now outlined and discussed briefly.
Number of Road Accidents. Qle of the most basic and obvious statistics
requiredforthederivation
of u n i t and total cast estimtes is t h e
n m b e r of road accidents and accident victins.
The Australianwlreau of
Statistics (A.B.S.)
compiles statistics of casualty
accidents
from
reports prepared i n the v a r i oSut sa t e s
and
police
accident
The A.B.S. has recently drawn a t t e n t i o nt ot h ee x i s t e n c e
Territories.
t h e d e f i n i t i o n s and coverage of casualty
of s i g n i f i c a n td i f f e r e n c e si n
accidents between t h e States, and has suspended publication of c e r t a i n
The basic accident
aggregated statistics for. Australia a f t e r 1977.
statistiss of i n t e r e s ti n
the present st@ are the total nmber of
nmber of f a t a l i t i e s and i n j u r i e s , and c e r t a i n
casualtyaccidents,the
details of
c r a s s - A a s s i f i c a t i o n s of the foregoinginformationincluding
*
?he tern "casualty" is used t o i n c l u d ef a t a l
and non-fatalinjury
accidents;"injury"
is used to coveronlynon-fatalinjuryaccidents.
25
the nature of accident, and age and sex d i s t r i b u t i o n of victims.
ktails of published total road accident statistics i n Australia f o r t h e
years 197.5 t o 1978 are shown i n n b l e 4.
T.ABLF: 4
ROAD ACCIDENTS
INVOLVING
-
ACCIDENTS
YEAR
1
1975
1976
1977
1978
of F a t a l
Accidents
Total N o .
of Casualty
Accidents
3,246
3,156
3,161
3,268
65,788
64,282
67.693
71,334
Number
1
j
1
!
j
1
CASUALTIES
T
- AUSTRALIA
CASUALTIES
Fatalities
"
f
3,694
3,583
3.578
95,396
3,705
1
1975-1978
I n.l uri es
89,499
91,818
,
I
Total
91,391
1
101.390
Source: Road h f f i o Accidents Involving Casualties, kcember Quarter 1979,
A.B.S. Canberra (Catutogue No. 9402.0). Explanatory notes t o t h i s
i s some v a r i a t i o n i n t h e d e f i n i t i o n
Dubtication indicate that there
;md reporting of casualty accidents betmen the S t a t e s and k e r r i t o r i e s .
Because the presentstudy adopts a cost estimation procedure
b e d
on injury severity classes
it was necessary to re-classify the casualty
accident
data
of 'Bble 4 according t aop p r o p r i a tien j u r sye v e r i t y
It w a s a l s o necessary to derive estimtes of t h el a r g e and
groupings.
accidents
which involve
vehicle
generally unrecorded nunher of road
and/or
other
property
danage
o d y , without
causing
any i n j u r i e s .
?he
i n j u r y s e v e r i t y c l a s s i f i c a t i o n adopted was the six-category "Abbreviated
Injury Scale" (AIS) which has k e n developed s p e c i f i c a l l y for road
i n t h e United States of .hrica.
The Abbreviated
accident
research
Injury Scale is outlined i n t h e review of Faigin (1976) i n Appendix A-1
(see %ble 3 8 ) .
In the
present
study
an allocation of Australian
accident
nunters
and costs was m dteo
Abbreviated
Injury
Scale
categories, (based on s t a t i s t i c a l a n a l y s i s of a sarnple o f , Australian
t h e U.S.-derived proportions
accident-ccstdistributions,togetherwith
(see page 48 and Appendix A-3), and the
resultant
distribution
of
accident nmhers is shown i n B b l e 5.
ESTIMATED CLASSIFICATION OF ACCIDENTS BY INJURY
SEVERITY
- AUSTRALIA
1978
(Preliminary 0 )
Number o f
Accidents
Number of
Injuries
3,268
143
687
2,736
16,853
47,647
3,705
205
987
3,937
24,187
68,369
5,098
223
1,072
4,268
26,291
74,329
Total casualties
Property Damage Only (P.0.0.)
71,334
485,489
101,390
111,281
922,429
TOTAL
556,823
101,390
1,033.710
Estimated
Accident Classification*
A.I.S. 6 ( f a t a l )
A.1 . S . 5 (Critical:suruiaaluncertain)
A.I.S. 4 (Severe:liff!-threatening)
A . I . S . 3 (Severe:notlife-threatening)
A.I.S. 2 (Modemte)
A . I . S . 1 (Minor)
-
Number o f
Vehicles
Source:
f a t a l and i n j u q , accidenttotals. and t o t a l f a t a l i t i e s and injuriesareobtained
data f c f . l'&k 4 ) ; the total nmber of vehicles involved
i n accidents i s
by subtmcting vehicles in casualty
est-Gnated i n Table 22; P.D.D. vehicles are obtained
by multiplying the number of accidents fcolwrm 1 )
accidents; these latter are estimated
by I . 56, representing the cstirmted mtio of vehicles per casualty accident (based
on
1977 A.B.S. data); the number o f P.D.O. accidents i s obtained by dividing P.D.0;
vehicle numbers by 1.9, representing the estimated
m t i o o f vehicles per P.D.o. accident
( a f t e r Troy and Eutlin, 1071, p . 2 0 2 ) .
f m m A.B.S.
The allocation of accident and casuolty nmbers betveen Abbreviated Injury Scale
(A.I.S.)
classes 1 t o 5 i s based on the 1975 U.S.A. proportions i n these A . I . S . categories after
Faigin (1976 Table 2) together with some adjustments based on the distribution of
accident i n j u r y costs contained i n Motor Accidents Board (M.A.R.) data for Victoria.
Because o f some apparent differences betueen the distributions of accident chnmcteristics
n d e r s were allocated t o A.I.S.
f o r t h e U.S.A. and Victoria, 70% of non-fatal accident
category 1 compared with 85%forthe U.S. distribution: the
remining A . I . S . c h a s e s
2 to 5 were allocated as i n Faigin 119761.
fl These estirmtes by categoryareprovisionalestimatesonly:
a more refined
w i l l require a
classification of Austmlian mad accidentaby injwy severity level
s e p m t e s t u d y based upon impmved
data.
*Abbreviated I n j u r y Scatelsee Tabla 381
27
E s t i m t i o nm t h c d s
are surmarised i n thefootnotesto
'!Able 5.
lhese estinates enploy certain key e s t h t i o n ratios f r m t h e d e t a i l e d
Australian accidentstudy
of Troy and Butlin(1971),
and use is also
made of new data unique i n
A u s t r a l i faotrh se i z e - d i s t r i b u t i o n
of
medical and related accident costs forcasualtiesobtained
from t h e
hbtor Accidents Board of Victoria.
C a l c u l a t i o n s . Che of the mre significantaccident cost
components i n Table 1 is the unit cost of $113,510 representing the
presentvalue
of lost future i n m which w u l d have accrued i n the
'he
period between accidental
death
and n o m 1 l i f e expectancy.
calculation and conceptual basis of t h i s cost ccnponent has k e n t h e
subject of s m controversy i n t h e l i t e r a t u r e , t h e r e f o r e i n the present
report the basis of t h i s estinate and its s e n s i t i v i t y t o changes i n t h e
m i n a s s u r p t i o n s and alternativeconcepts is examined i n saw detail. A
range of foregone inl e v e l s for road accident fatalities i n 1W8 is
F.
presented i n %ble
ForegoneIncome
TABILE I;
Table 6 shows t h e present value i n 1978 d o l l a r s of average i n c m
foregone f o r each age group and an o v e r a l lt o t a l , for each of s i x sets
of discount rate and i n c m productivity rate assunptions. ?he estirmtc
of $113,510 i n Table 1 is drawn from t h e estimtes i n t h e f i f t h
colm
of '!Able 6
( i . e . 1G%and
?he r e s u l t a nft i g u r e
is a "weighted"
t o t a l of t h e lost i n c m f o r each of t h e agegroups shown i n t h e €irst
column of the
table.
Tfie E t h d of a m p u t a t i o n i s similar t ot h a t
modified
to comply with
Australian
data
ernloyed i n Faigin (197F),
In addition t o t h e need for data on average
annual
incm
sources.
levels by age groups ( u i d e Table7)
t h e figures i n Table 6 incorporate
t h e e f f e c t s of the age and sex d i s t r i b u t i o n s of the particular sample of
rcad a c c i d e n tf a t a l i t i e si n
1978, t h a t is theforegone
i n c e figures
will vary withthe
age/sex d i s t r i b u t i o n of road casualties a s well as
3%).
28
'Ihis f a c t is of particular significance because
victims
are
the age and sex characteristics of road accident
s i g n i f i c a n tdl iyf f e r e n t
from the national and state population
averages.
?he foregone income f i g u r e sf o r
each separate age group i n
Table 6 s h m t h e average i n c m loss f o r persons of median age i n each
group
(e.g.
$163,101
is the
present
value
at 10%of averagefuture
i n c m f o r a person of 25 years).
with changes i n income.
I n a,ddition totheforegoingsources
of variation i n Table 6, the
conceptof income adopted f o r these calculations also has a s i g n i f i c a n t
e f f e c t on theresultant
estimate. Because of t h en u n b r of parameters
capable of variation,
together
with
the
relative
complexity
of the
calculation mthod, the effects
ofchanges
i n these key parameters upon
the r e s u l t e s t i m a t e s of the present value
of lost i n c a w are explored i n
sorre detail.
Ihe principles
underlying
the concept of foregone gross i n m
adopted i n t h i s present study as an indicator of social loss follow
of which is
closelythe approach of Mishan (1971), a usefuldiscussion
contained i n the recent Cdnadian research of Lawson (1978).
In this
approach the level of personal gross i n c m foregone (or its gross
is taken t o be t h e loss to the camunity from
productionequivalent)
accidentaldeath
or seriousinjury,
where the society or m n i t y is
defined so as t o incZude the personkilled or injured. 'he a l t e r n a t i v e
concept of n e t income, or net
production,
whilst
consistent
with
national income d e f i n i t i o n s ( i n t h e sense that i t measures t h e change i n
national i n c e which occurs as a r e s u l t of road accidents) is not
of s o c i a l loss frcm road
considered t o be appropriate to theconcept
accidents, nor to that of gainsfranaccidentprevention.
The e f f e c t s
of thesediffering
i n c m d e f i n i t i o n s upon theresultantaccident
cost
e s t h t e s are outlined below.
The b i c data required t o undertakeforegone income calculations
are w a n incorre per c a p i t a c l a s s i f i e d by age groups, contained i n 'Ctble
7 , and the
age-distribution
of road accident casualties.
The 1978
i n c m f i g u r e s by age are based on a 1973-74 surveyundertaken by t h e
Australian Wlreau of Statistics and updated by priceindicesforthe
present study.
&re recent
survey
data
r e l i a b i l i t y of the income estimates.
would
improve the
overall
29
TABLE 7
It
. "
3,021
4,319
4,785
5,150
4,805
3,852
Source:
MEAN INCOME by AGEGROUPS
AUSTRALIA 1978
__
Age Group
Mal e
(Years)
$
i
4
15-19
20-24
25-34
3,520
8,663
!
5,231
35-44
45-54
55-59
60-64
65+ 3,285
I
I
Female
I
12,275
13,142
13,160
~
i1
11,409
10,893
5,179
based on 1973/74 f i g u r n s ?.+om the S u m e y gf I%-ome ,5ist%2wtiori 15731974. A u s t m l i a n Bumau of S t a t i s t i c s : iii increnenkd by a f a c t o r of
1 . 8 4 3 i r e p r e s e n t i n n t h e ircrense i n Avemoe V e e k h Earninos b e t m e n
19?3/74 and calend& lt;?a! f o r b o t h male i n , fern&
incomzs and ( i i l
fernale incomes were i n m e r e n t e d by a f3irther 255 t o allow f o r t h e relat i v e r i s e i n female i n c o r ~ sns LZ proportion of male incomes betieen
1373/74 and 1978 i e. g . figures f o r earnings r,ne n o t a-xriiable, b u t f e male wage m t e s rose .fmv 8 5 % of t h e . m l e p a t e i n 1 9 7 3 t o 94% i n 1 9 7 7 1 .
-
Choice of Discount b t e .
The choice of discount rate h a s a most
s i g n i f i c a n t e f f e c t on the presentvalue
of foregone i n c m as shown i n
calculations of 'Pable 6. In t h i s table
the
discounted
present
value
projectedfuture i n c m is incremnted by an annual labourproductivity
rate, in addition to the discounting calculations.
lhese w
t o rates have
f i g u r et :hhei g h etrh e
opposing e f f e c t s upon the foregone i n c m
discount rate the lower t h e presentvalue of foregone income; while t h e
rate (or increase i n real inccme) operates t o
annual
productivity
increase i n m .
Tne e f f e c t s of varyingthese
w
t o p u n t e r s is sham i n B b l e 6
whichshows
three discount rates (7%, 1% and 13%)
combined with three
productivity rates (2.5%, 3% and 3 . 5 % ) . 'Ihe present-value
algorithm
used i n these c a l c u l a t i o n sn t i l i s e s
the f a c t h a t f, o r
example, 10%
discount together with a 3%'productivity increase is equivalent t o an
effective
discount rate of
c o r r e c t simply t o s u b t r a c t
i n these calculations).
[=1.
io
o r 6.8%. (It is therefore not
the p m d u c t i v i t y rate frcm thediscount
rate
30
'Ihe preferred i n c m productivity rate f o r t h i s study is 3 % per
annun ( v i d e Appendix A-2, mble 65 for t h e basis of theproductivity
estirrate based on Australian data over the period 1967 t o 1978).
of discount rate i n
The issuessurrounding
the appropriatechoice
of
b e n e f i t s o s t a n a l y s i s are somewhatcomplexand
are alsothesubject
continuing
controversy
(vide:
Treasury [1978];
Layard
[19721;
Dasgupta and Pearce [1972]. Die basis of t h e approach adoptedhere is
outlined
briefly.The
need t o discount
future
(projected)
streams of
costs and revenuespublic and p r i v a t e sector investrent analysis arises
becausebenefits (i.e. revenues, etc.) and casts occurring i n d i f f e r e n t
t h periods are valued d i f f e r e n t l y by t h e c-ity.
This is sometires
t e m d a positive rate of time preference i n t h ee c o n m i cl i t e r a t u r e ,
implying t h a t individuals tend t o prefer consumption i n the c u r r e n t time
period(tosaving
income) r a t h e r t h a n t od e f e r it t o scm time i n the
future.
Thus t h e discount rate "r" whic.h equates s m given d o l l a r
a m u n t one year hence with the same m u n t i n the present represents the
time-preference rate f o r that individual.
Because of theexistence
of
e x t e r n a l i t i e s and interdependencies associated w i t h the benefits and
costs to the c m n i t y , from public
sector
investment
(or, what are
termed the public or "social" goods characteristics of public
:
vide Musgrave 8 hbsgrave [ 19761 chapter 3 ) , the true
inves-nt
c m n i t y rate of private time preference, or Social
Time-Preference
w t e of discount, cannot be derivedreadily from individual rates. For
p r a c t i c a l purposes t h e Social Tim-Preference (or STP) rate of discount
is generallytaken
t o be s m amlgam of theimplicit rates determined
by voting practices and political d e c i s i o n a k i n g .
A s i g n i f i c a n t problem i n s e l e c t i n g a rate of discount is t h a t
e s t h t e s based o n . the Social TimPreferenceconcept
are considered
l i k e l y t o d i v e r g e from rates based on the principal alternative discount
Opportunity
Cost of capital (or SO2
r a t e concept, namely t h e Social
rate).
?he Social @portunity Cost rate is intended t o measure t h e
marginal productivity of capitalinvestment,
or i n effectthepresent
cost t os o c i e t y of divertingresources
away from i n v e s t m n t i n p r i v a t e
Under
s e c t o r projects, to public (i.e. government) sector projects.
certain theoreticallyidealconditions,
these w
t o rates-STP and SOCare
equivalent. In practice theydiverge, it beinggenerally a s m e d t h a t a
hzb.1 Time m'eference rate will be less than the Social @portunity
Qst rate of discount.
More -lex
issues concerning the choice
between STP and SO2 rates of discount are considered at length i n the
l i t e r a t u r e , and t h e m i n issues are outlinedin
s m r y i n the draft
prepared
by the Federal Treasury Depar-nt
discount rate guidelines
( v i d e : Treasury [1978]). I n t h e latter Treasury document, a case is
argued for a choice of discount rate based on the Social @portunity
Qst concept rather thanSocial
Time Preference, and t h er e c m n d e d
of publicprojects
is
discount rate for use i nb e n e f i t 4 o s ta n a l y s i s
lo%,withsensitivity
tests of the r e s u l t s t o be capleted at rates of
7% and 13%per annum.
In t h e i n t e r e s t s of consistency and carparability,theaccident
those s p e c i f i c cost components
have been estimated, using
the
affected by discountingcalculations)
However
truo bases of
r e c m n d e d rates of lo%, 7% and 13%per annum.
cost estimtes i n t h i s presentstudy(or
31
disagreemnt
with
the
conclusions
of t h e Treasury d r a f td o c m n t
as
generalprescriptions
can be noted: f i r s t , t h a t t h e argument i n favour
of an S0c-bas.d discount rate m y givetoo rmch weight t o t h e issue of
d i s p t a c e m e n t or the o p t k l d i s t r i b u t i o n of t o t a l investmznt between the
private and p u b l i cs e c t o r s
of t h e econom i nt h e current tim period,
ratherthan
to a m a s u r e of the c m n i t y ' s collectivepreferencefor
public goods which are not o p t i m l l y provided by t h e p r i v a t e s e c t o r .
Second, i n t h e case of public sector provision of carmodities such
as improved road safetythe
use of a higherdiscountrate
is s e e n t o
discriminatenarkedlyinpractice
between those accident cost lwasures
which i n v o l vfea t a l i t i easn d / o r
mjor i n j u r i e s , and those i n which
property danrage costs predominate.
Inevitably, t h i s also raises " v a l u e
of l i f e "i s s u e s
which are notreadilyresolved,
such a s whether it is
f e a s i b l e or desirable i n publicdecision-making
to t r a d e 4 f fa c c i d e n t
mre
prevention
(especially
death
and s e r i oi n
u jsu a
r yg)a i n s t
conventional
masures
of s o c i a l and economic benefit ( v i d e Freemn
[1979], Schelling [ 19681, and l"isllan [1971]).
TAt31,E 3'
~~
THE EFFECT O F THEDISCOUNT
RATE ON FOREGONEINCOME
( a t 37;p.a. p r o d u icnt icvr ietays e )
For*
Fore
Discount
. .
. .
Fatalities
Rntp l n a \
nurles
$
$
-
*
Australian
Population
Aver-
8
219,791
225,309
170,787
162,264
163,548
121,428
113,510
86,641
112,004
81,729
84,076
60,762
74,965
72,096
51,853
56,835
53,768
38,246
Differences i n incoxe figures bet-deen accident victims and the
national population average r e f l e c t diffeFing age d-Lstributions.
I n suxmry tile vieu x i ~ p t e di l l t i l i s pr,-sc:1 r e p r t 1s tlut tile need
i n publicsectorevaluation
outweighs a r g u m n t s f o r
forcomparability
theuse
of a lower discount rate s p e c i a l l y determined f o rr c a ds a f e t y
evaluations.Therefore,
a discount rate of
p.a. is employed i n t h e
body
of
the text.
main average and total a c c i d e n tc o s tt a b l e si nt h e
However a f a i r l y comprehensive range of t a b l e s showing thedivergent
cost estivates obtained a t rates of 7% and 13%is contained i nt h e
results are
appendices
and,
where s i g n i f i c a nt th, e a
s el t e r n a t i v e
referenced i n the t e x t .
The o v e r a l l result of t h i si n v e s t i g a t i o n
of
discount rate e f f e c t s is a p r o l i f e r a t i o n of tables i n this report which
are needed t d
o i s p l a tyh e
range and s e n s i t i v i t y of t h em i n
cost
estimtes t o changes i n t h e d i s c o u n t r a t e t o g e t h e r w i t h o t h e r e s t i m t i o n
parameters.
la
32
In mble 8 , the e f f e c t s of varying the discount
rate f m 5%t o 20%
per annm upon the estimated reraininglifetime
income which would be
lost as a result of premturedeathin
a road accident is shown f o r
three separate age-groups sample populations.
Table
8 compares
estimtes of the presentvalue
of t o t a l i n m foregone between age a t
time of accident and n o m 1 l i f e expectancy, weighted by t h e age group
of t h e sample of the
population
recorded
i n (i) road
accident
f a t a l i t i e s ; (ii) non-fatalinjuries;
and ( i i i ) by t h e total population
Thus, two types of variation are shown:
within each
age-groups.
column, the 1978 presentvalue of r e i n i n g i n c m can be compared; and
across the three columns, the e f f e c t s of the d i f f e r i n g age groups is
compared.
Ingeneral Table 8 shows that the presentvalue of foregone i n m
f o ra c c i d e n tf a t a l i t i e s
and injuries is comparable, but the equivalent
results for h s t r a l i a n averagepopulation
are approximtely 30% lower.
'Ihe e f f e c t s of age group weighting are discussed
further
below. 'Ihe
of
t
h
i
s
table
is
the
significant
decline
i n the
principal outcome
presentvalue
of foregone income, as t h e discount rate is increased.
For f a t a l i t i e s , foregone inccme a t a 7% discount rate is 43% greater
than a t a 10%discount rate, and at 13%t h i s v a l u e f a l l s 24% belw the
10%level. Similar (thoughnot
identical)differences
occur forinjury
age groups.
'he e f f e c t s of these
and average
Australian
population
discount rate variations between 7%, 10% and 13%upon unit(average)
accident costs can be seen by comparing Table 1 with Thbles 50 and 51.
The f i r s t three entries i n rows 1 and 2 of these tables (mrked
with an
a s t e r i s k ) are dependent on the foregone i n c m calculations:
these
item a l s o account for a mjor share of v a r i a t i o ni nt h eo v e r a l l
cost
estimtes.
Age Distribution of F a t a l i t i e s and I n j u r i e s . Details of the age and sex
d i s t r i b u t i o n s of road a c c i d e n t v i c t i m are contained i n 'Ihbles 9 and 10.
33
TMLE 9
AGE DISTRIBUTION OF ROAD ACCIDENTFATALITIES
AUSTRALIA 1978
Age Group
(Years)
0-4
5-6
7-16
17-20
21-29
30- 39
40-49
50-59
60+
Not s t a t e d
Total
Source:
f
Total
199
5 89
167
293
160.1)
193
167.7)
211
342
(66.6)
132
157
148
101
92
1 79.01
( 8 2 . SI
(74.4)
106
262
156. 6 )
9
2632
(39.9)
(21.0)
(17.1)
125.6)
(32.3)
133.4)
143.4)
1
(71.0)
~
1073
L
(29.0)
104
(2.81
50
331
746
864
394
285
(I.+)
(8.9)
(20.1)
123.3)
(10.6)
( 7.71
317
604
(16.7)
10
10.3)
j
3705
1103.01
j
(8.6)
~
~
~
O f f i c e of Road S a f e t y ( O X ) Working Document No. 16 ( 3 / 8 0 1 ( A . R . S .
Datal
The recordedage
distribution ranges for road accident statistics
d i f f e r frm the age ranges f o r which i n c m data are recorded, and some
re-estirration w a s required to derive t h e i n c m l e v e l s f o r i n j u r i e s
in
lkble 11.
TLBLE 10
AGE DISTRIBUTION OF ROAD ACCIDENT INJURIES
AUSTRALIA 1977
Age Group
Males
%
Females
%
I
Total
~
(Years)
0-4
5-6
7- 16
17-20
21-29
30- 39
40-49
50-59
60+
Not stated
1474
950
6377
14614
14426
6091
3951
3388
3328
2 184
Total
56783
(54.8)
153,21
I
i58.5J
168.71
166.5)
(60.01
(56.91
(55.11
(51.31
158.8)
(62.0)
Soume: Road !l f f i c A c c i d e n t s ,
9403.0
~
!
i
1%)
1217
(45.2)
,
656
4526
6661
7273
4063
2989
2762
3158
1528
(40.81
,
(41.2)
6940
6 150
6486
3712
34833
138.0)
91616
(December&uarter
( d l . 5J
(31.3)
133.5)
140.01
143.11
(44.9)
(48.7)
19771,
2691
1606
10900
21278
21698
10 155
B.S.
(2.91
11.81
(11.9)
(23.2)
123.71
(11.1)
(7.6)
(6.71
(7.1)
(4.1)
(100.0)
(Catalogue No.
34
Separate average i n c m estimtes were requiredforcalculationof
t h e foregone i n m of i n j u r i e s , and these i n c m levels are shown i n
Table 11.
TABLE 11
AVERAGE INCOME OF ROAD ACCIDENT INJURIES
AUSTRALIA 1978
Age
Group
Per cent of
17-64 years
total
(2)
17-20
21-29
30-39
40-49
%
(17.7)
36.2
17.0
11.6
10.3
50-59
60-64
i:
(1)
(2)
(31
(7.2)
Proportion:
Mal e
Females
%
68.7
66.5
60 .O
56.9
55.1
51.3
%
31.3
33.5
40 .O
43.1
44.9
48.7
rlJ
AverageAnnual
Income
131
Male Female
$
5,064
9,955
12,666
13.150
12,079
10,893
WEIGHTED AVERAGE 1978 INCOME per INJURY
;
$
3,684
4,827
4,531
4,949
4,886
3,852
I
deighted
Average
Income
L
$
4,632
8,237
9,412
9,615
8,849
7,464
$7,963
CaZculated on the basis o f mzle/fermleweightedaverage
f o r each age p
u
p
:
the overall weighted average i s based on the pmportion of each age group
t o t h e 17-64 years total (see note 2).
These percentagesare based on 1977 Road Accident S t a t i s t i c s , exqluding
thosewith age notspecified,
and adjusted ( i )toincludethe
17-20
years group(assumed t o be i n the workforce); and (ii)to include twot h i r d s o f t h e 6 0 t age group (comprising 60-64 years group).
Adjusted t o conform w i t h age group8 forinjur,yaccidents(bylinearinterpolation between medim incomes f o r each g
roup, and recalcuktion of averages)
The foregone i n m c a l c u l a t i o nf so r
r a d a c c i d e ni nt j u r i e s ,
comparable with the f i g u r e s for f a t a l i t i e s i n Table 6 , and based on the
i n m and age distribution data i n Thbles 7 and 11, are shown i n Table
12.
35
TABLE 12
t
TABLE 13
~~~~~
THE EFFECTS O F AGE GSiILIP UEIGHTING ON 0ISCOUN-ED
1hCOMES"AUSTKAL:A
Gmup
I
0-4
58.617
5-6
7-16
74,520
7-20
'1-29
71 ,399
105,931
168,MO
183,051
173,899
143,564
10-39
56,283
80-49
33,364
0-59
SO+
AVEWGE TOTN
INCOFYS
conpared:
t
e a t k e d i a n Ages:
Age
84,755
47,960
I
27,387
33,358
49,492
66,588
65,395
I
16,960
Hale:
Female :
Total :
~
TotalI:
Weighted by
Fatalities
1978
10: Discount; 3% P m d u c t i v i t y
Total11:
Weighted by
Injuries
5
I
6
41,801
44.493
43.408
57,704
83,424
55,s:
I
T0t.l 111:
Population
We7ghted
1
148,721
163,101
146.084
115.389
82.525
139,056
144,022
130,469
105,973
i
i
52.797
70,431
122.315
125,373
121 ,160
!
101,272
67.571
34,513
61,675
32,866
,
59,222
31.946
140.845
45,788
146,635
55,233
i
117,519
113,510
112,004
81 -729
I
36
rile significance of d i f f e r e n c eitsnh e
age d i s t r i b u t i o n s of
with
t h e population
accident fatalities and i n j u r i e s when conpared
average age is r e f l e c t e di n
the respectiveweighted-averagediscounted
i n c m v a l u e s i n 'hble 13 (these f i g u r e s i n d i c a t e t h e basis of the ' h b l e
8 estinates).
I t is evident that there is only a s l i g h tv a r i a t i o n
between the
are
averageforegone incomes f o r f a t a l i t i e s and injuries (although there
mre mrkedvariations
between p a r t i c u l a r age groups),but
b t h these
levels are over 30% greaterthanthepopulationaverage
income. lhese
d i f f e r e n c ems i n lrye f l e ct htdei s p r o p o r t i o n a t e llya r gnem b e r
of
younger mles in
the
accident
statistics, conpared ttohaev e r a g e
population.
The E f f e c t s of GPOSS, Net, and Adjusted Income Concepts. To ccmplete
t h i sa n a l y s i s
of theforegone
i n m e s t i m t e s ,t h ee f f e c t s
upon t h e
the income concept i n t h e
presentvalue
offoregone
i n m ofchanging
is o u t l i n ei n
d
the
foregoing incorre c a l c u l a t i o nf sof ar t a l i t i e s
following tables. A canparable set of t a b l e s f o r road a c c i d e n t i n j u r i e s
is contained in Appendix A-2 (vide: Tables 59 t o 61).
Adjusted Income. The e f f e c t s of averaging gross i n m l e v e l s across
the total population in each
age group is shown i n Table 14.
TABLE 14
FOREGONE ADJUSTED
1
Age
i Gmup
Nunkr o f
Fatalities
I
~
0-4
5-6
7-16
17-20
746
21-29
864
30-39
394
4 0 -24895
50-59
317
60t
104
50
331
604
-r
Presen
Iscount:7%
rnductivity:
2.5%
64.892
89,746
111.658
181,757
196.410
158.647
106,743
48.436
17,736
INCOME^":
'alue of l n u
7%
3%
ROAD ACCIDENTFATPLITIES
F o r e g o n e( A v e
7%
3.5%
- ALUTW\LIA 1978
ed over P o p u l
10%
2.5%
ion)
I
10%
3.0x
101
3.5%
t7 I I I I
75.803
103.551
125,074
196.756
209.814
166.832
110,468
49.300
17.903
88.772
119,787
140,483
213.566
224,605
175,640
114.377
50.184
18.071
28.442
42.169
62.089
122,592
140.886
122.255
88.883
43,954
16,835
32,666
47.843
68,385
130,604
148,630
127,611
91,664
44,693
16.988
37,593
54.395
75,490
139,451
157.075
133,337
94.574
45,449
17.143
136,990
147.151
91,042
96.101
101,654
AVERAGE
INCOMI
FOREGONE: p e p
fatalit# Uqiu&
for W0nkf-e
partidpution)
127.870~
d
the No,
=h age 5
37
lhese adjusted i n c m levels are approximtely 85 % of the foregone
gross i n c a w levels i n 'hble 6.
'Ihey reflect t h e factoring of i n c m s
by the average 1978 workforce p a r t i c i p a t i o n rate of 79 % for mles, and
44 % for fHllales.
?he e f f e c t of the workforce p a r t i c i p a t i o n a d j u s h n t s t o i n c m s i n
Table 14 is equivalent to assuming a zero i n c m f o r a l l persons between
17 and 65 years of age who are n o t i n t h e wrkforce,
including
housewives as well as tile unemployed. It h a s t h e effect of a t t r i b u t i n g
t h e lwer populationaverage
of i n c m s t o a l l people i n the accident
sample.
This approach was e x p l i c i t l y employed i n the Japan (1978)
study, and t h ee f f e c t s of thelkterson
(1973) " l i f e x y c l e model" appear
t o k very similar ( v i d e Appendix A-1, Tables 35,43 Er 44). In c o n t r a s t ,
t h ei n c m
assumptions
underlying
Tables
6 and 13 assume that a l l
rmnkrs of theaccident sample earn the averageworkforce incorre levels
of
each
age group.
Thus both housewives and those unemployed are
a t t r i b u t e d an opportunity c m t i n c m l e v e l equal t o t h e workforce
average.
I n t h e present study, t h i s latter assunption is preferred
since it is considered t o more accurately reflect the comrmnity's
valuation i n i n m term of such a c t i v i t i e s .
Net Income. 'k show t h e e f f e c t s on t h e e s t i m t e of using t h e net income
concept i n calculatingforegone
i n c m as employed i n the cost s t u d i e s
of Dawson (1967) Troy and Butlin (1971), Paterson
(1973),
and Japan
(1978) ( v i d e Appendix A - l ) , an average
level
of private
consunption
expenditure of 31 % of household i n c m was derived frcm t h e 1977/78
AustralianIktional
Accounts, and a l l i n c m l e v e l s i n lkble 7 were
reduced by 31 % beforecalculation of the presentvalue of figures. The
results of these assumptions for net
income are contained i n Table 15.
F i n a l l y , i n t h i s series of foregone inccm c o w r i s o n s , the results
of assuning b t h : ( i ) lower i n c m s averaged
over
the
population,
and
(ii) of excluding consumption expenditure from future i n c m are shown
i n Table 16.
An equivalent set of tablesshaving
the e f f e c t s of these changed
assumptions u p n i n j u r i e s is contained i n Appendix A-2 ( v i d e Tables 59
t o 61).
38
TABLE 15
-1
FOREGONE N E T I N C O M E + : ROAO ACCIDENT FATALITIES
hge
Gmup
-
AUSTRALIA 1978
Present Value of Awrage l n c m Less Consunption expenditureat:-
Nlmber O f
Fatalities(')
IL1
f
10%
3 .O%
''I
1
10%
3.51
7 7 104
0-4
5-6
7-16
17-20
21-29
30-39
40-49
50-59
60t
864
50
331
746
64.253
67,667
394
285
31 7
604
43,904
VERAGE NET INCOME
(4
y:
OREMNE p r f a t a l i t50.301
Based on 321 o f groa
r
t
Age
Gmup
57,215
42,279
22.928
11.170
1
30.121
38.63415.788
47.227
69,658
72,528
60,402 43.258
23,387
11,275
~~
Fatalities(1)
23,510
43.281
47,825
24,589
20,559
10.603
37.221
35,187
33,336
14.90.'
20.311
28.513
49,215
53,556
47.467
37.020
21,347
10.797
,me.
FOREUXlE NET ADJUSTED I N M *
Number of
44,735
53,104
75,747
77,920
63,850
45,616
23.858
11,381
54.104
46,902
12.957
17.886
25.858
46.095
50,561
45.287
35,775
20.948
10.699
11,292
35,334
- ROAO ACCIDENT FATUITIES - AUSTRALIA
1978
Present Value of Adjusted I n c m Net: o f C o n s w t at:(')
ion
101
7%
7%
2.5%
3.51
3%
-7%(Disc)
2.5%(pmd)
T
(31
101
3.55
I -3"- 1
104
331
864
394
285
317
0-4
5-6
7-16
17-20
21-29
30-39
40-49
50-59
60+
fatality:
50
746
604
20,116
27.821
34,614
56,344
60.888
(9,180
33.091
15,013
5.498
1 5 , 6 1472 , 4 6379 , 6 4 0
23.498
32.101
38,772
60.994
65.043
51.717
34.246
15.281
5.550
27,519
31,134
13.549
66.205
69.628
54.448
35.458
15.555
5,602
f
8.817
13.072
19.247
38.003
43.675
37,898
27.555
13.624
5.219
28,223
10;126
14,831
21.199
40,487
46,076
39,559
28.41 7
13,853
5,266
11,653
16,862
23,401
43,230
48,694
41,331
29.319
14.087
5.314
29.792
31,512
t1
pti
oti
39
Medical and Retated
Accident
Costs.
Perhaps t h e m j o r shortcoming
available
for
a f f e c t i n g t h e range and q u a l i t y of A u s t r a l i a n data
estirmtion of accident costs is t h e absence of any recent study
of Troy and W l t l i n (1971) i n which a
comparable t ot h ep i o n e e r i n gw r k
v i r t u a l census of road
accidents
w a s undertaken in
the
Australian
Capital Territoryduring
1965/66. %is ,mrk, which established a world
accident
research yielded, i n t e r a l i a , a j o i n t
bench-rrark i n road
d i s t r i b u t i o n of m d i c a l and vehicle darrage casts. Such i n f o m t i o n is
v i t a l i n order t o relate injury s e v e r i t y l e v e l s t o o t h e r i q m r t a n t c o s t
categories,
especially
In a mre recentAustralian
vehicle d m g e .
s t u & by Paterson (1973) t h irse l a t i o n s h i p
w a s referred
to
but
no
estilnates by accident severity were presented (although other
parmters
d r a w f r a n the Troy
and
B t l i n study w e r e e x t e n s i v e l yu t i l i s e d ) .
Tne
r e s u l t a nctr o scsl a s s i f i c a t i o n
of injury and vehicle repair costs
compiled by Troyand Butlin is smrised i n mble 17.
T,IBLF 17
INJURY COST LEVELS by VEHICLE WK4GE: A . C . T .
( U n i t s are:
1965/66
nuder of collisions)
Cost lntervals for Pep~onaI I n j u r y
VEHICLE
OAPAGE
$0
10.~200
163
10
1,367
819
259
75
42
13
48
86
65
28
12
10
$400$l.WO
$2TOo
51.000$2,000
143000- 66,WO$6,000
S8.WO
S2.000$4,000
s
I
I
~
0-200
200-400
400-600
600-800
800-1000
1000-1200
1200-1400
1400-1600
1600-1800
1800-2000
,
'
'
,
1
~
i
4
4
4
~
I
i
5
7
3
1
4
I
8
6
2
1 2
1
;
i
7
l3
12
1
,
1
i
9
~
6
7
8
2
:
.
1
.
i
r8,000+
1
2
6
1
1
1
1
'
.
,
1
1
2
2
1
1
1
,
I
.
I
,
.
2
1
1
.
.
,
~
'
1
2
Total
.
,
778
1.428
943 2
355
120
64
26
12
13
8
11
Hecause of its r e l a t i v e age and r e s t r i c t e d sample region t h ec a s t
data i n %ble 17 is consideredinadequa-te by i t s e l f f o r t h e purposes of
the
present
study.
Fortunately,
a valoable
source
of r e d i c a l and
related data from road accident cases is ncav being compiled by t h e k t o r
Accidents Board of Victoria (M.A.B.), an agency which h a s operated i n
in
Victoriasincetheintroduction
of a "no f a u l t " compensationsystem
t h a t State i n 1974. its yet it is not p s i b l e tocrass-classify
both
m d i c a l and property danage a r i s i n g from accidents,but
a valuableset
of individualfrequencydistributions
of m d i c a l ,h o s p i t a l
and r e l a t e d
accident casts h a s been specially tabulated for the present study by t h e
b t o r Accidents Board.
40
&tails of scire $19 million paid in
claim by the b t o r Accidents Board
of Victoria i n respect of hospital,medical,
ambulance, loss of income,
and certainother
costs r e s u l t i n g fran road accidentsoccuring
i n the
year to 30th June, 1978 are considered i n 'Ihbles 18 t o 21.
TABLE 18
I
1
CLAIMS
ARISING
F,P.OM ROAD ACCIDENTS
IN
YEAR TO JUNE 1978
ACCEPTED BY THE MOTOR ACCIDENTS BOARD, VICTORIA.
I
I
Type of
Accident
I,
Number of
I
Amounts Paid
Hospital
$m
79 7
Fatality
Major Injury
Minor Injury
10,863
19,606
TOTAL
31,266
Medical Ambulance
$m
$m
1
income
Of
$m
1
Other*l T o t a l
Bm
0.520
0.171
0.206
1.973
0.178
0.603
9.945
0.289
4.121
0.006
10 .322
2225
0.942
4.449
e:
based on B u l l e t i n o f S t a k i s t i c s
1.023
1I
$m
1.137 0.050 0.074
0.322
17.121 0.479
0.703 0.024
18.962
J
o f Persons Killed o r Injured i n Road
AccidentsOccurring ( i n ) Year Ended 30th J m e 1978 ( f o r whichclaims
were registered with the Board): Motor Accidents Board o f Victoria,
J m e 1920; together with a d d i t i o n a l (unpublished) computer tabulations
providing frequency distributions of claims
by categories.
Note: that these figures cover claims accepted
by the M.A.B. up t o
11th November 1978 only l i t i s estimated that subsequent Major Injury
claims will comprise about I O per cent of the
above t o t a l ) .
4 "Other" costsincludechemist,dental,fmeral,housekeeping,physiotherapy,etc.
-
41
A s m r y of these
payrents
by category and type of accidents i s
contained i n Table 18. Tables 19,'and 20 show frequency d i s t r i b u t i o n s
for cost categories and total claim respectively. while %ble
21 shows
f u r t h e r d e t a i l s of t o t a l cast by category.
TABLE 19
r
1
HOSPITAL, KOICAL 6 OTliER C O S S O F WAC A t C I E l l T CASUALTIES
VICTORIA 1977-78 : MITOR ACCIDLH:
!l i
cost
1"teT"al
:,+nor injr-Lea'
a j o , i",;..i..'
0-5W
500-1000
1MI0-15W
1500-2000
2000-2500
25W-3x4
3l00-35W
3%
llunbcr
of C l a i m
+
I
(21
~
r
~
41
LIwl
~
rota1
IUnbCl
rotat
1ayIRCIIt
,f Claim
'aymnt
I
.
~
Z.0.
0,ZDB
E
0.174
n.9.
0.920
0.759
0.638
l,M7
514
340
208
161
159
m1
(31
LOSS OF I N C O t f
KOlCAL
HOSPITAl
L
BOAR0
I
c
0.587
0465
0.440
0.516
5.826
,:,a.
s,jas
".a.
724
137
18
9
5
4
i
I.O*S
, 1,72J
1.227
0.195
65
c
Im
0.27d
0.112
0.039
0.026
0.016
0.017
..C.
0.006
9.024
0.353
659
3.38(;
0.402
0.475
270
0.338
0.354
S.306
1.397
~
6.0.
,
!
0.377
0.066
0.021
0.007
0.016
0.W6
0.011
__i
0.479
0.503
-
Fatalities:
TOlAL A l l Claim.
255
,/.a_
+
0.520
__
1
i
42
?he Victorian M.A.B. data i n these tables are of considerable irqmrtance
to the present study
since theyprovide
the principal estinmtion basis
€or the hospital and medical costs i n Table 1 and i n later tables.
TABLE 30
TOTAL COSTS CLAIMED BY ROAD ACCIENT INJURIES
VICTORIA 1977-78 (MOTOR ACCIDENTS BOARD)
Number of
claims
CostInterval
J
Total
Payment
%
a*
(‘%tinorInjuries“)
0-100
19,606
64.3 3.9
0.703
18.7
-
1.253
(“Mzjor I ~ ” u i e s ” )
100-500
0-500
500- 1000
1000- 1500
1500-2000
2000-2500
2500-3000
3000-3500
3500 +
25,307
1,575
82 7
528
360
258
225
1,369
Sub-total
k j o r Injuries:
I O ,35.7
863
(all
TOTAL injuries)
Source:
30.469
83.1
5.2
2.7
1.7
1.2
0.8
0.7
4.6
1.956
1.122
1.014
0.925
0 .804
0.706
0.730
10.567
7.0
11.0
6.3
5.7
5.2
4.5
4.0
4.1
i
$
36
220
-
77
712
1226
1753
2233
2735
3245
7608
+
59.3
17.121
100.0%
:
Average Payment
17.824
100.0%
Bulletin of S t a t i e t i c s of Pereons Killed O P Injured i n Road Accidents
30th J m e 1978, Motor Accidents Board, Victoria,
Jtme 1980, together with Lmpublished supplementary computer tabulations
of prequemy distribution by coat intsrunl supplied by the B w r d .
These f i g w e s cover claims accepted by the 4I.A. B . up t o 11th November
1978
it i s estimated t h a t subsequent major injury claims w i l l
increase the above total by nwpe than 10%.
(*I Paymnta by the M.A.B. includeclaims for Hospital, Medical,
Ambulance, hoes of Income, and Other costs (Chemist, Dental,
Housekeeping, Physiotherapy, etc.) arieing from m a d accidents.
** Minor Injury claims m e definedasnon-fatal
claims for which
total paynrsnts are less than $100.
- Oocurrinq Year Ended
e:
These a c c i d e n tc m tl e v e l s
and t h e i r frequency d i s t r i b u t i o n s are
used t o estirrate c o s t l e v e l s , and t o allocate medical and r e l a t e d costs
according t o t h e i n j u r y s e v e r i t y grouping.illthough
thisVictoriandata
is of considerablevalue i n t h e p r e s e n t e s t h t i o n
task there are scme
s i g n i f i c a n t a s s m p t i o n s i m q l i c i t i n t h i s use of M.A.B. data.
pI a
nr t i c u l a r ,
it is a s s m d that
the
observed
frequency
d i s t r i b u t i o n s of a c c i d e n tc l a i m
i n ascendingorder
of average casts
provide a reasonableindication
of i n j u r y s e v e r i t y . 'his r e l a t i o n s h i p
has been investigated i n t h e work of Fox e t aZ. (1979) i n t h e course of
a mjor study of poleaccidents
i n &lbourne i n which surveydata was
conparedwith M.A.B.cost c d i n g s . Because t h e f i t o r k c i d e n t s b a r d is
dependent upon accident wdical records cowiled using the International
C l a s s i f i c a t i o n of Diseases (I.C.D.) code as a guide,conversion
t o an
i n j u r y s e v e r i t y assessrent presents m n y difficu1r;ies.
An attempt t or e - c l a s s i f yt h e
I.C.D.
code t o conform withthe
Abbreviated Injury Scale groupings is contained i n Table 63 i n Appendix
A-2.
This mrk is not
ccnplete
and farther
research
is needed to
e s t a b l i s ht h ef e a s i b i l i t y
of deriving a s u i t a b l ei n j u r ys e v e r i t y
scale
foraccidentrecording
i n Australia. In thepresentstudy,the
M.A.B.
cast r e l a t i vdei s t r i b u t i o n s
were used t o modify for Australian
conditions the groupings for
U.S.A. d a t a corrpiled by Faigin (1976).
VehicZe Damge.
A f u r t h e r problem of significance i n these estimtes
concernstheclassification
of vehicle damage costs by i n j u r ys e v e r i t y
l e v e l . Total vehicledamgecosts
wm obtained fran 1978 statistics of
motor vehicleinsurance
claim of $545 m i l l i o n i n a b l e 24. However,
because of the need t o estirrate dannge t o uninsured vehicles, estimted
by 'Ihorpe (1970) t o be as high as 35 % of t o t a l motor vehicles, 1971
data fran the A.B.S. Survey of b t o r Vehicle usage w a s projected forward
t o 1978, showing the estimated nunber, cost and range of vehicle
repairs. These results are contained i n Tables 22 and 23.
44
TAHIJF: 22
PRELIMINARY
ESTIMATE OF MOTOR VEHICLE ACCIDENT REPAIRS
AUSTRALIA 1977-78
I
i
Not Reported
Total
No.
cost
Reported*
Total
No.
cost of
Repairs
cost
141,000
271,000
331,000
$m
10.6
81.3
314.5
192,000
71,000
28,000
743,000
406.4
291,000
$
under $100
$100-$500
$500+
TOTAL
Estimated"Excess"
paid by owners:
~~~~
~~~~~~
Source:
~
Total
Total
cost
No.
1
$m
j
14.4
21.3
26.6
333,000
342,000
359,000
62.3
1,033,710
1
1
j
~
66.9
$m
25.0
102.6
341.1
468.7
535.6
~~
based on an a p p m z i m a t e update o f 1971 data f m m t h e Motor Vehicle Usage
Survey (see T a b l e 231 by incrementingaccident nwnbersby
theincrease i n
motor vehicles on r e g i s t e r betdeen 1.972 and 1978 iadfusted f o r tha casualty
a c c i d e n t r a t e l , and the mean c o s t o f r e p a i r s b y t h e i m p l i c i t p r i c e
indez
f o r GrossNationalExpenditureoverthisperiod.
The d i s t r i b u t i o n o f
repair costs beimeen mnges
uas aZso adjusted.
'%xcess" payments assume t h a t 908 of reported claims require the
mer to
pay t h e f i r s t $100 o f eachcZaim.
*'LQeported"claimsarethosereported
to i n s u m c e companies.
TABLE 23
II
I
c o s t of
I
REPoRTEod
$
NOT REPORTED
I
Cars &
Stn Wagcns
Cars &
Stn Wagons
Repairs
S:-
-
MOTOR VEHICLE ACCIDENT
REPAIRS
AUSTRALIA 1971
(YEAR t o 30th SEPTEMBER 1971)
I
TOTAL
Cars &
Stn Wagons
Total*
1
under 50
50-200
200+
93,304
130.666
288,919
190.260
344,118
TOTAL
572.482
689,366
154,814
248,118
65,648
14,302
178,341
74,932
16,228
214,583
234,765
269,501
807,247
309.007
419,049 354,567
230.811 204,562
958,867
Surveu o f Motor Vehicle Usage f o r twelve months ended 3 0 t h September,
1971, A . B . S . Crmberm, 1973.
6 Reported t o inswanLIB company.
* Excluding nwtor cycks.
45
claim, both comprehensiveand
D e t a i l s of motor vehicleinsurance
“thirdparty”(i.e.personalinjury)
claim are shown i n l k b l e 24 which
also containsthe
slmre of insurance costs allocated t o m t o rv e h i c l e
clad.
‘rmm 24
MITOR VEHICLE INSURANCE CLAIlls e t c .
AUSTPALIA 1977-78
Premium
State
M.V.
Coqrehenrive
t
N.S.W.
Victoria
Queensland
South Australia
Uestem Aurt.
Tasmania
R.C.T.
N.T.
545.4
TOTAL
577.4
canpu1saly
Third Party
*n
Im
35.5
216.4
99.7
:ii:o”
6 2 .O
64.2
20.2
9 . 1 10.3
3.6
780.6
Manageuent Expenses
Claim
42.4
60.5
27.2
10.6
10 .2
4.2
I
1
I
,
~
,
II . Y .
Conprehenrive
Rllocated to:-
Conpdlsory
T h i r d Party
235.4
131.5
68.6
37.6
44.1
14.6
3.2
.
art M.Y
45.1
2.5
Co
Sm
132.1
238.2
160.6
141.9
37.9
57.5
38.9
33.4
11.3
6.4
~
5.5
~
~
15.9
11.4
9.4
2.7
1.8
0.1
562.3
A sample of motor vehicleproperty
damage c l a i m w a s undertaken
during this present study, including
s a of the principal mtor vehicle
A.A.M.I.Ltd.,
the State
i n s u r e r s i n the State of Victoria(including
Insurance Office, and also theinsuranceindustrygroup,theInsurance
Council of Australia).
Tabulations
of vehicle danrrge claim w e r e
receivedcovering a proportion of t o t a l claim paid. When a s u f f i c i e n t
coverage is achieved it w u l d be p o s s i b l et o
applytheseproportions
danrrge accident
derived f m n A u s t r a l i a n d a t at or e d i s t r i b u t ep r o p e r t y
costs by i n j u r y s e v e r i t y i n ’Pable 1 e t a2 t o r e p l a c e the d i f i e d U.S.A.
a l l o c a t i o n s used i n t h e current Australian estimtes.
M O T O R VEHICLE INSURANCE CLAIMS: 1 9 7 9 *
( C a r sa n dS t a t i o n
Wagons)
C1 aim
Size
8
less than
200
200-300
300-400
400-500
500-600
600-700
700-800
800-900
.900-1000
1o o o t
TOTAL4 6:
Number
of Claims
(No.)
Amount
o f Claims
Average
C1 aim
8m
$
2.205
75
243
344
29,347
4,880
3,449
2,696
2,061
? ,693
1,325
1,112
869
1.186
1.187
1.203
1.124
1.092
0.990
0.951
0.821
855
945
6,856
14.499
2,115
2 5 . 2 4 95 4 , 2 8 8
446
546
645
747
5
* Excluding "excess" payments which are estimated a t $127 p e r vehicle,
The data i n l k b l e 25 relates topropertydamge(vehiclerepairs)
only, and is compiled independently of t h e i n j u r y costs.
3.3
hJJMMARY OF ESTIMATION PROCEDUMS
In t h e following section t h e methaiofderivation
of t h e unit cost
estimtes presented i n Table 1 (and r e l a t e d cost Tables) is outlined i n
s m r y form. Tne e f f e c t s of deletingcertain cost components from the
o v e r a l l c o s t framework a r e then examined.
In sorre instances d e t a i l s of theestimtionprocedures
and data
sources used to compile cost itens are contained i n theexplanatory
estimtes of Foregone income,
footnotes to selectedtables,including
Medical and Hospital costs, and Vehicle Lkzrnage i n Tables 5 t o 7, and 11
to 16 of section 3.2 a b v e .
The accident cost categories of
Table
order.
1 are now considered i n
(1) Foregone Income: theseestimates are based on the presentvalue of
i n c c m estimtes f o r f a t a l i t i e s and i n j u r i e s i n Tables 6 and 12 together
with the degree of i m p a i m n t or l o s t work associated with each l e v e l of
t h e Abbreviated Injury Scale (AIS) for
road
accidents
AIS
?he
47
proportions of work-tim lost or p e m n e n t impaimnt are draw frcm t h e
study of these corrponents by h i g i n (1976)
and
are
d e t a i l e d U.S.
consideredappropriate
for Australian road accidentconditionsin
the
absence of a n equivalent local m d i c a l assesment (vide Appendix A-1,
e s p e c i a l l y Tables 37 and 38).
Details of t h e Foregone Income
c a l c u l a t i o n s by AIS l e v e l are as follows:
Fstimted
m y s Lost
Value of
LostInccm
1978
$Imaimnt
Estimted
1978 Loss
(per day
or p.a.)
1 Minor i n j u r i e s
2 Moderate
3 Severe-I
1.6
21
39
$
31
31
31
50
650
1,210
4 Severe-11
5 Critical
G Fatal
25%
57%
12,004
12.004
113,510
23,000
63,840
113,510
AIS
Level
or
100%
$
?he estimted lass per day of $31 i n 1978 is based on t h e income
f i g u r e s of Table 11 divided by annual work days. me 25% and 57% l e v e l s
of + m n e n t i m p a i m n t a s s o c i a t e d w i t h AIS l e v e l s 4 and 5 respectively
are appliedtothepresentvalue
of inccrre f i g u r e s ( a t 1% discount, 3%
productivity)fran
Table 12. The f a t a l i t yi n c mf i g u r e
is f r m Table
6.
This cost
is intended t o
category
(2)
Family, C o m n i t y Losses:
accidentcaused
losses
for uurk
representthe
cost
value
of
opportunity
and services performed outside t h e n o m 1 working week:
( a ) for t h e
family and home, and (b) forvoluntary
services t o the c m i t y . The
opportunity cast m t h c d of valuation is p r e f e r r e dt or e p l a c e n e n tc o s t ,
because the f o m r is consistentwiththemrket
inconcept.
?he
average time devoted t oi d e n t i f i e df u n c t i o n s
(e.g. ham? mintenance,
household t a s k s , care and upbringing of c h i l d r e n , e t c ) w a s e s t i m t e d for
t h e U.S. by m i g i n (1976) as 10hours per week for ham? and family, and
2 hours per w e e k forvoluntary c m i t y a c t i v i t y ,r e p r e s e n t i n g
30% of
t h e working week.
This time-based estimate is alsosupported by other U.S. s t u d i e s
of householdproduction
which suggest that theequivalentmrketvalue
r e p r e s e n t s a b u t 25% of t o t a l household i n c m i n N t i o n a l i n c m term
( v i d e Gronau [1W3], and migin[1976]).
me proportion of 30% of m r k e t i n c m derived by m i g i n is thus
adopted a s the basis of thecategory 2 estiwtes i n 'Itible 1 which thus
represent 30% of the last incam? costs i n row 1.
Categories ( 3 ) , ( 4 ) , and ( S i :
Hospital, Medical and Rehabilitation
c o s t s : These cost estirrates are based on the frequency distributions of
1977/78 cost data frm the k t o r Accidents b r d of Victoria (vide:
48
Tables 18 to 21) tcgether with the e s t i m t e dp r o p o r t i o n s
n u n k r s i n each AIS injury severity l e v e l .
of c a s u a l t y
PS indicated .in the
footnotes
t o Table 5, the
proportion
of
accident cases i n each of t h e (non-fatal) AIS l e v e l s 1 t o 5 is based on
a modification of the U.S.A.
proportionsderived
by Faigin ( op.cit.)
r e s u l t i n g from an inspection of t h e M.A.B. d i s t r i b u t i o n s of hospital and
m d i c a l costs for t h e State of V i c t o r i a ,s m m r i s e d
i n Tables 19 and
85% of a l l injury accident cases t o
20. lhe U.S. proportionsallocate
AIS level 1 (minor i n j u r i e s ) , whereas t h i s seem toohigh
a proportion
of theAustralianinjuries,
based on t h e M.A.B. data and assuming cost
l e v e l s a s a proxy for
injury
severity.
PIUS f o r the Australian
proportionsforinjuryseverity,
AIS level 1 was set a t 7a, and the
remining proportions for classes 2 t o 5 w e r e calculated as f o r the U.S.
estirmtes i n Ehigin [ o p . c i t . ] .
Although t h i s adjustment is sarewhat
a r b i t r a r y , it is mre consistent with Australian cost data both for
Victoria i n 1977/78, and with the Australian Capital Territory study
for
1965/66 ( v i d e Table 17), and m y reflect
under-reporting
of minor
casualty accidents i n Australia q a r e d with U.S. data. lhe r e s u l t s of
t!lese a d j u s t m n t o
s
AIS class proportions are swnrarised below,
together
with
the r e s u l t a n t average cost ranges
based
on these
proportions.
est.
U.S.A.
AIS
%
Class
AUST
%
1 (minor)
(moderate)
2
3 (severe-I)
4 (severe-I I
5 (critical) '
Total
85.08
12.31
2.01
0.50
0.10
100.00
.
70.00
24.77
4.03
1.00
0.20
100.00
lhe Range and Average Costs f o r t h e s e AIS classes are:
(from M.A.B. data)
Hospital
Average
Medical
(Range)
Average
$
(Range)
$
0-$707
158
75
$707-$5000
1,919
$5000-$10,500
7,100
$10,500-$30,000
11,928
36,000
$30,000
Rehabilitation
382
$768-$1468
1,015
1,738
t
3,123
Average
(Range)
$
0-$199
0-$143 50
$199-$768$143-$431
235
$431-$864
561
$1468-$2472
$864-$2214
1,322
$2472 t
2,214
$2214
+
.
49
Ihe Australian AIS c w l a t i v e proportions were a p p l i e tdo
the
frequencydistributions
of MAR c l a h f o rh o s p i t a l ,m d i c a l
and other
costs, respectively ( v i d e Table 19), giving, for example, i n t h e case of
hospital costs t h e f i r s t 70% of claim numbers, ranging f m zero to $707
i n amount, w i t h an average cost of $158 were taken t or e p r e s e n t minor
i n j u r i e s (AIS level 1). Similarly the next 24.8% (or from 70.0% to
94.8% of claimaccordingtocurmlativefrequencies)ranging
frcm $707
of $1919 which represents
t o $5000 i n m t , shwed a n averagevalue
i n Table 1. Medica2 and Other (Rehabilitation)
average
AIS level
3
costs were determined by injuryseveritycategory
i n t h e same way, thus
yielding t h e unit cost e s t h t e s f o r rws 3,4, and 5 of Table 1. 'he
v a l u e sf o rf a t a l i t i e s
(AIS level 6) are from Table 21, where the entry
i n column 1 of row 5 represents funeral costs.
I t is evident from the relationship between the average costs by
injury level, and the respective range of c a s t s for each category,that
t h e frequency d i s t r i b u t i o n of M.S.B. data is s i g n i f i c a n t l y skewed (such
that theaverage value for each range is mch closer to one of ths range
b u n d a r i e s than to the midpoint).
I n Appendix A-3, The preliminary results of an attempt t o f i t a
s u i t a b l e t h e o r e t i c a l probabilitydistribution
t o M.A.B. medical d a t a ,
and otheraccident costs arepresented.
?his work is at an early stage
but e a r l y results suggest
that
a range
accident
of
costs and
c h a r a c t e r i s t i c s m y be usefullyrepresented
by t h e g a m m d i s t r i b u t i o n .
bkch a result, i f confirmed, would k very useful i n t e ra l i a
i n reclassifyingaccidentcostsinto
any desired set of accident classes
h d on probability of occurrence (e.g. i n a specially definedinjury
c l a s s i f i c a t i o n for
Australian
conditions), as well as providing a
valuable slnmary of respective accident characteristics.
llle preliminaryestimatesforkbspital,kdical
and Other related
accident casts i n Table 1, however, are kased d i r e c t l y on the M.A.B.
d a t a r e f e r r e d t o above, and not upon theoretical d i s t r i b u t i o n s f i t t e d t o
these data ( v i d e Appendix A-3).
Legal and Courtcosts:
?his cost category is as defined i n Faigin
( o p . c i t . ) , representing the economic resource costs conswed as a result
the costs of bothpublic and p r i v a t e l e g a l
ofroadaccidents,including
a c t i v i t y generated by accidents.
(6)
No recent Australian data were a v a i l a b l e t o t h e p r e s e n t s t u d y ,
and
the f i g u r e si n row 6 of Table 1 are (rounded) estimtes fran t h e U.S.
1975 cost Study by Fkigin (op.cit.).
.Australian researchers i n
t h i s f i e l d have suggested a relationship between legal and court costs
and mtorvehicleinsurance
claim paid ( v i d e lhorpe[1970], Troy and
Butlin [1971], and Paterson[1973]).
Based on t h e i r 1965/66 study i n
the Australian&pita1lkrritory,
Troyand
k t l i n suggested that up t o
30% of t h i r p
d a r t y(,i . e .
personal injury)
insurance
claim i n t h e
A.C.T.
were forlegalexpenses.
?he Bterson (1973) study
adopted
a
l m r proportion of 25% i n its 1969 accident cost estimtes r e s u l t i n g i n
a 1969 u n i t legal cost per c o l l i s i o n of $72 (or $88 including police and
court costs vide Table 40). I n the present study t o t a l e s t i m t e d l e g a l
and court casts (based on t h e U.S.4erived unit casts) are $35.2 million
50
(fran Table 52), which represents only 6.3% of t h e t o t a lt h i r dp a r t y
payments i n 1978 of $562 m i l l i o n ( f r m Table24).
lhese costs represent
$63 p x accident, $364 per c a s u a l t y accident (and $256 per c a s u a l t y ) ,
and are much lower t h a n the figures suggested by Paterson e at l .
u n t i l more i n f o m t i o n i s
b r e o v e r , they are preferred to t h e f o r " ,
Australian
research
on t h e grounds that it
by
further
provided
specific
seem l i k e l y that a significantproportion
of the large third-party
claimpayout relates to courtvaluation of "pain and suffering" arising
f r a n read accidents, and associated
legal
costs.
This latter cost
category is s p e c i f i c a l l y excluded frm t h e cost framework proposed i n
t h i sp r e s e n t report ( m i n l y because no analysis of these payments has
been undertaken).
lhus the U.S.-based unit costs providethe cost e s t h t e s f o r row
6:
Clearly t h i s cast estimte would be greatly improved by
investigation of t h i r d p a r t y claim payments.
t h e resource costs of insurance
(7)
insurance
Administration:
mnagement
expenses
associated
w i t h motor vehicle comprehensive
insurance are derived frm thepublished
totals i n Table 24, together
with
the
injury
severity a l l o c a t i o n s (AIS levels)
derived
i n the
hklbournestudy by Fox e t a t . (1979) (uide Table 29). 03st f i g u r e s from
t h e Fox e t a t . study w e r e increased by a f a c t o r of a b u t 2.5 t o agree
with the total f o r mnagement expenses i n Table 24.
( 8 ) AccidentInvestigation:
these costs representthe
resource costs
r e l a t i nit n
o
gv e s t i g a t i o n
both
ofcasually
and property darrage
accidents. me e s t h t e s i n row 8 represent a modifiedversion of the
Fox e t a l . (1979) cost levels for this category, adjusted
t o relate mre
closely to t h e b i g i n i n j u r y s e v e r i t y l e v e l s .
(9)
Losses t o Others:
t h i s cost category is a non-rrarket estimte
similar i n concept to category 2 ( l o s s e st of a m i l y ,c a m u n i t y ) .
I t is
defined t o cover losses caused to employers and o t h e r sr e s u l t i n g from
costs, tirw spent i n
road
accidents,
including
labour
replacewnt
v i s i t i n g ,t r a n s p o r t , h a care, etc. I t is thus the opportunity cost of
The estimates i n
t h e time s p e n t i n t h e s e a c c i d e n t + e n e r a t d a c t i v i t i e s .
Table 1 are based on the U.S. N.H.T.S.A.
study (1971) proportionsfor
these it-,
and represent the following percentages of lost i n c a w i n
row 1: AIS 1:20%; AIS 2 t o 3:10%; AIS 4 t o 5:2.5%; and AIS 6 : l . s .
(10) Vehicle Rznazge: t h i s is a d i r e c t cost of mtor vehicle accidents,
nanEly vehicledamge
repair costs, and represents one of t h e major
conponents i n total accident costs. The a l l o c a t i o n s ofaverage costs t o
AIS classes was based on application of t h e AIS class proportions (i.e.
c m u l a t i vper o b a b i l i t i e st)o
a frequency d i s t r i h t i o n of insurance
claiffi (e.g. shown i n total i n Table24,
and a sample
vehicledamge
frequency d i s t r i b u t i o ni n Table 25). lhe c a l c u l a t i o n sa l s o assme that
vehicle danrage generally increases with accident
injury
severity
(although the W O vehicle danage cost f o r critical injury accidents is
higher than t h e $3000 average for f a t aal c c i d e n t s
i n Table 1. This
difference is not
very
s i g n i f i c a n t because of t h er e l a t i v e l y
all
nmber of vehicles i n AIS class 5 conpared t o class 6 : vide Table 5).
51
(11) T r a f f i c Delay: theseunit costs are takendirectly from t h e U.S.
study of m l g i n ( o p . c i t . ) , and are based on calculations of the tM
lost i n person hours i n the p r o p r t i o n of accidentsoccurringin
urkn
week-day peak-hour t r a f f i c , and a value-of-time cost per person-hour of
$2.63.
Trle U.S. proportions, based on 1973 accidents, shou t h a t
approximately 26% of a l l accidents
occur
d u r i n g peak hour t r a f f l c ,
i n c l u d i n g 18%of f a t a la c c i d e n t s and 2G% of i n j u r y and p r o p e r t y d m g e
onlyaccidents.
’Ihus theaverage t r a f f i c delay cost of v e h i c l e 4 m g e only and less serious i n j u r y accidents is higher t!xn t h a tf o r
mre
severeinjury
and f a t a la c c i d e n t s ,r e f l e c t i n g
the higherproportion of
the f o m r occurring i n urban pcak hours.
mese f i g u r e sa r ep r e f e r r e d
t o the Australian estimates of Fox e t a l . ( v i d e a b l e 29) which relate
s p e c i f i c a l l yt os e r i o u si n j u r ya c c i d e n t s .
However a r e l a t i v e l ym d e s t
study k e d on Australian data would improve these e s t h t e s .
E f f e c t s o f Deleting Selected Costs
I n a b l e 1 the estimtes f o r c e r t a i n cost categories are shown i n
i t a l i c type, generallysignifying that no .Australiandata
is presently
availabletoprovide
a basis f o r cost e s t i x a t i o n .m e s e
rows include
Family d Community Losses,Legal
4 Court Costs,AccidentInvestigation,
Since these cost estiinates
Losses t o Others and T r a f f i c Delay Costs.
are largely tused on U.S. experience,notably t h e 1976 study of h i g i n ,
i t is of S Q T ~ interest t oi n v e s t i g a t e
the e f f e c t on u n i t costs (and
their r e l a t i v i t y by AIS l e v e l ) and on t o t a l costs of deletingthese
item;.
me e f f e c t s of excluding these five cost categories
total unit costs and t o t a l costs is shown i n B b l e 2G.
Table 2G:
EFFKTS OF DELETING SfLECIW COST CATECfXES*
AI s
Before
Level
Average Bst
$
Total Cost
.After
$
G
157,085
119,155
5 133,F85 111,125
4
46,815
57,175
13,210
3
14,755
2
5,175
5,790
1
1,895
2,270
P.D.O.
620
450
Total
[-I
[-I
*
i n ’Pable 1 upon
After
Before
582
27
56
58
140
155
572
1,591
*
441
23
46
52
125
130
415
1,232
The “before“ columns arefromTables
1 and 52, and t h e ’ h f t e r ”
exclude cost c a t e g o r i e s 2,6,8,9 and 11 fromTable 1 .
colms
It can be Seen t h a t t h e p r i n c i p a le f f e c t
of excluding these f i v e
cost categories is t o reduce the relative s i g n i f i m c e of f a t a l
accidents carpared t o other casualty accidents, since the
largest change
is to the ccst of a f a t a l i t y . %tal accident costs are reduced by $359
million, or 23% a s a result of these cost exclusions.
52
I t is considered that it is mre desirable to improve t h e basis of
estimation of these cost categories,thuspreserving
a mre s y s t e m t i c
concept of social cost, rather thanexclude any oP the c a t e g o r i e s c i t e d
abve
.
!TheNeed
f o r FurtherResearch
?he sccio-economic cost Pramework adopted for this studyconsists
of accident-caused losses t o economic and social welfare which are
reasonably c m n s u r a b l e .
The framework is thus an extension of the
"loss accounting"approach
(e.g.of
Reynolds [ 19561, and Dawson [1967]
i n the United Kingdom and Japan [JRCTP, 19781) by progressively
broadening t h e d e f i n i t i o n of cost f r m mre narrowly
defined
to that of
financial/accounting losses r e s u l t i n g frm a c c i d e n t s ,f i r s t
economic resource costs (comparable with nationalaccounting measures),
eventuallyleadingto
a socic-economic masure of those costs borne by
t h e t o t a l comnunity which takes f u l l y i n t o account e x t e r n a l i t i e s ,
interdependencies, and other non-market e f f e c t s of accidents which are
reasonablycapable o f comensuratevaluation.
lhus sare cost categories are includedbecause a judgement is m d e
that there is sufficient
acceptance
of the concept and basis of
valuation even
though
t h e s t i m a t em
s y
be empirically rough
(e.g.
familylosses, t r a f f i c d e l a y costs, e t c ) . However certain other c o s t s ,
such as pain and suffering a r i s i n g f r m accidents, are excludedonly
because no generally
accegted
basis
of valuation has yet been
established(althoughthe
courts &e judgements a b u t the dollarvalue
of compensation f o r p a i n and suffering and other non-market q u a n t i t i e s ,
r e s u l t i n g i n t h e allocation of l a r g e sum i n ccopensationforaccident
losses.
Such a settlement i n N.S.W. exceeded $1 million i n t o t a li n
1980 [National Times, &pt 21, 1980, page 241 and it is thus apparent
that t h e basis and relative amounts of suchjudgementswarranted f u r t h e r
study). ?he r e l i a b i l i t y of mny of the cost components i n ' h b l e 1 would
be improved both by f u r t h e r research into existing data sources, and the
c o l l e c t i o n of relevant new i n f o r m t i o n by survey.
Both the separate and j o i n td i s t r i b u t i o n s
of vehicle darrage and
to f u r t h e r substantiate the
medical costs require additionalresearch
assumptions d e a b u t i n j u r ys e v e r i t yc l a s s i f i c a t i o n
i n thecourse oP
these
estimates.
There has been no
study
since Woy and Butlin
(197l)which has attempted t o measure these j o i nrte l a t i o n s h i pisn
Australia, such a s t h e U.S. uork of Flora, Bailey, and O ' k y (1975) and
Marsh, Kaplan and Kornfield (1977) on the financial consequences
of
vehicle accidents i n Michigan.
OtherUnitCost
Estiktes.
As discussed earlier, several of t h e cost
item contained i n t h e unit cost framework of B b l e 1, (including
directly
established
frcm
categories 2,5,6,8,9, and 11) w e r e not
Australian data
sources
(in
several cases the U.S.A.
relationship
or slightly
modified).
derived by m i g i n were a p p l i eddi r e c t l y ,
Improved local estimtes would result frcm f u r t h e r research i n t o cost
categories such as t r a f f i cd e l a y ,l e g a l
and courtcosts,
and accident
'be concept of l o s s e s t o family and community f o r
investigation.
example is considered to be an jnportant and validcategory of social
53
cost,but
t h e e s t i m t i o n basis ( 3 0 % of foregone income, a f t e r h i g i n )
needs f u r t h e r a s s e s s n e n t and i n t e r p r e t a t i o n to c a l i b r a t e an acceptable
Australian cormunity valuation of t h i s item. Similarly, t r a f f i c deZay
costs could be mre accuratelycalculated
for Australianaccident
and
i n a separate study.
!be problem of assessing a
t r a f f i cc o n d i t i o n s
valuation for pain and s u f f e r i n g r e s u l t i n g fran road accidentscould be
approached by establishing its value a t some acceptableproportionfor
foregone i n c m :
again, the validation of such a social value would
require a separate study.
me u n i t cost f o n m t of 'Ihble 1 is considered to represent t h e
minimun set of cost categoriesfor
which accidentcosts
need to be
derived.
I t is, i n t h e . v i e w of t h e presentauthor,
less misleading t o
present a conceptually conplete set of cost estimtes c o n t a i n i n g d i r e c t ,
i n d i r e c t and "translated" k t c a m a r a b l ef o r e i m cost estimates. rather
54
4.1 ConcZusions
The proposal for this study envisaged
that saw d i f i c a t i o n of the
"loss accounting" approach t o a c c i d e n t cost measuremnt would prove
acceptable forAustralianapplication.If
this w e r e s o , m t of the
subsequent research and investigation would have concentrated on the
refinement of estimates and assessnent of the adequacy of relevant
estimtes.
Australian data sources in supporting these
I
In t h e l i g h t of the
foregoing
review
inchapters
2 and 3 (and
Appendix A-1) of theconceptual
and empirical literature on road
accidents, the presentstudystronglysupports
the view t h a t the
loss accounting
approach
t aoc c i d e n t
cost measurement - whilst
u s e f u li n s c l l ~areas of accidentpolicy - is of limited usefulness
i n respect of its principalobjective:theevaluation
of accident
reduction p r o g r m s .
Moreover the
debates
which have a r i s e n
or "intangible"
concerningtheacceptability
of c e r t a i n non-rket
c o s t conponents, such as pain and suffering, losses t o f a m i l i e s , the
CMlnUnity, employers, t r a f f i c congestiondelaysarising
f m road
a c c i d e n trse, f l etchut en s a t i s f a c t o rnya t u r e
of f i n a n c i a l or
economic resource-cost e s t h t e s as a masure of t h e social b e n e f i t s
t o be gained f r m accident reduction.
In t h i s respect t h i s conclusion f o l l m sc l o s e l y
that of Lawson
(1978) i n Canada, and theconceptualrequiremntsin
Mishan (1971),
with some modifications as t o t h e scope for
developrent
and
application of e x i s t i n g cost framework ( v i d e Chapter 2, section
2.3; and Appendix A-1).
I t is thereforeconsideredthatthe
"loss accounting"approach
to
accident cost measuremnt, whether f i n a n c i a l , or consistent with the
opportunity cost concepts and national i n m accounts, is
inadequate f o r the purposes of, accident policy evaluation.
?he conceptually correct measure of w h a t the cmmunity would be
willing to pay foraccidentreduction
is a direct measure of t h e
demand f o r such a benefit(foraccidentreduction).
The economic
resource costs of road accidents
represent
only
the
minim
estimates of what the camunity would be willing to pay for accident
measures, and attempts tpol a c e
a value on
the range
of
intangible accident effects, including pain
and s u f f e r i n g r e f l e c t an
a t t e n p t to bridge t h e gap between financialcosts
and the "social
value" of accidents.
55
However, the view of Mishan (op.cit.) that only d i r e c ta t t e n p t st o
measure t h e denrand f o r s a f e t y and accidentreduction are acceptable
is notsupported i n t h e presentstudy.
bst such valuation a t t e n p t s
have had limited success i n terns of e r r p i r i a l estimtes of
a c c e p t a br el el i a b i l i t y ,
comx?nsurate with
other
casts ( v i d e
Blonquist
[1979];
Freenran [1979]; Jones-Lee [1974]; and 'Ihaler
Therefore
until a breakthrough occurs i n t h e
and b s e n [1975]).
area of d i r e c t wasuremnt of the aggregate dernand f o r r e d u c t i o n i n
and refine t h e
r i s k it is considered desirable to continue to extend
and
previous "economic" cost frameworks of Bwson (196"), ,Troy
t h e expanded
Rutlin (1971) and Paterson (1973), f o r example,towards
social (or "societal") cost framwork
proposed
i n Fkigin
(1976).
This latter cost f r e w r k , which f o m the basis of t h a t proposed
i n the pEsentstudy,incorporates
a l l identified
resource
costs
which are displaced as a result of road accidents,togetherwith
certain nonmarket and intangible accident "costs" which r e f l e c t t h e
attenpt to measure t h e required e x t r a conpensation above m r k e t cost
l e v e l s which the c a m u n i t y w u l d be willing to allocate to achieve
research is considered t o be
accident
reductions.
Further
worthwhile t o advance b o t ht h en o n m r k e t
and e n p i r i c a l W o n e n t s
of this accident cost framework.
So=foregone
problerrs m i n concerning the concept and e s t i m t i o n b a s i s of
i n c m component of accident costs. Although
often
the
the
unintended and generallyinappropriateeffects
of t h e various"net"
and "adjusted" i n c m concepts a p p l i eidn
earlier studies
(as
discussed
in
chapter
3 , section 3.2, and
Appendix
A-1) can be
readilyavoided,severalquestions
are also posed by t h e e f f e c t s of
t h e age and sex d i s t r i b u t i o n s of accidentvictims, and thediscount
rate selected, upon t h e r e l a t i v e importance of foregone i n c m i n
t o t a la c c i d e n t costs. The f i g u r e s i n Table 13 i n chapter 3 suggest
that the c m i t y apparently values the last i n c m of a road
as one-third
higher
than
that of an average
a c c i d e nfta t a l i t y
population member, although t h i s simply r e f l e c t s t h e higher
proportion of younger mles i n t h e a c c i d e n t sample. C e r t a i n e t h i c a l
o
r
a
l considerations m y intrude:
does s o c i e t y r e a l l yv a l u e
and m
l i v e s savedconsistentwithpresentvalues
by age? For example, i n
the calculus of section 3.2, the l i v e s of t h e elderly, who are
disproportionately
involved
i n pedestrian
accidents,
are still
" w r t h " less i n i n c m terns than younger people.
Whilst t h e
foregone inc a l c u l a t i o n s m y r e f l e c t an unsentimental assesscent
based upon individual cases, t h e camunity a s a whole m yc o n s i d e r
t h a t rnoratty a l l l i v e s are equal i nt h e i r
right t o be saved,
age or economic or social s t a t u s .
regardless of d i f f e r e n c e isn
Further
research
into
the
implications
for
accident
prevention
policy of t h e e f f e c t s of and attitude t o t h e age d i s t r i b u t i o n f a c t o r
m y be warranted t o c l a r i f y these issues.
In themeantim a reasonableinterpretation ofTable
13 values for
foregone insuggests that accidentsresulting i n f a t a l i t i e s or
serious injury are of greater relative inportance t o s o c i e t y than
other categories of social cost such as propertydamgeaccidents.
'Ihis d i s t i n c t i o n b
e
c
m of even greater significance when accident
costs are disaggregatedaccording t oi n j u r ys e v e r i t y .
lhe r e l a t i v e
56
total cost
irrportance of these individual cost item ttoh e
framswork should be borne i n mind, however, i n considering the need
fordata improvements. In Table 2 of chapter 3 , the w
t o categories
offoregone i n c m and losses to family and the c c m u n i t y , together
representover
51 % of total accident costs, vehicle danage is 28
percent and the insuranceadministration ccnponent i n "other" costs
is 5.2
'Ihese fourcategoriestogethercomprise
84 % of t o t a l
accident costs, and v a r i a t i o n s i n theirestimationprocedures
and
sources are l i k e l y t o overshadow refinements t o o t h e r r e m i n i n g cost
categories
including
medical
costs, legal, court, and m r g e n c y
service c o s t s , and t r a f f i c d e l a y .
%.
Conclusions i n respect of t h e adequacy of existingAustraliandata
each of t h e cost categories
sources are b r i e f lsyt a t efdo r
considered (together with respective t e x t references).
Foregone income r e s u l t i n g from f a t a l i t i e s and i n j u r i e s : a more
recent incanes survey would improve the
present
estinmte
(Chapter3,p28);
Family and C o m n i t yl o s s e s :
estimated at 30% of ( i )a f t e r
Faigin (1976) :
further
research
is needed to c a l i b r a t e a
current valuation for Australian conditions
(3.2p.53)
Medical,Hospital
and relatedcosts:
'[he b t o r Accidents Board
i n V i c t o r i a is t h e only known source ofsuch medical cost data
i n Australia a t present, a major survey by States would be
required to f u r t h e r -rove
these estimates (p.43);
(see also
item i v ) ;
Legal and Court costs:
no s t u d i e s of Australian
data
are
readilyavailable:
a survey
of
court award claim paid by
State insurance officesundertakingthirdparty
mtor vehicle
insurance would be required (p.53); Such a studyshould also
provide data on m d i c a l costs etc, and any canpensationfor
pain and suffering;
Insuranceadministration:
adequatesource data is collated by
the Australian Bureau of S t a t i s t i c s (p.45);
Accident
Investigation
(including
the
cost of emergency
is onlyavailable from special s t u d i e s :t h i s
services):data
item is r e l a t i v e l y -11
and a sample survey would be needed t o
r e f i n e the estimates (p 53);
Losses t o others:
no data is available frcm Australian
s t u d i e s , and t h e U.S. data was applied:
a scciceconomic
surveyoftheworkforce
and industryeffects
of accidents is
required @I 50);
Vehicte h g e :
t h e fundamental statistics f o rt h i se s t i n m t e
are not available (since property
danage only accidents are not
data
was
recorded i n m t States) and 1971 A.B.S. survey
required
for
these
estirrates.
General improvemnt itnh e
57
statistical base and consistency i n accident
recording
is
required, but inthemantimesurveys
ofccmprehensive
mtor
vehicle insurance claim are considered t o k the most f r u i t f u l
source of improved estimates, tcgether withappropriate sample
estimtes fran future surveys
of m t o r v e h i c l e usageconducted
by A.B.S.
E s t k m t e s r e l a t i n g t o uninsured
vehicles
and the
"excesst1
proportion
of
c l a i mp a i d
by owners would also be
derived frcm these surveys (p 43);
(ix1
T r a f f i c detay: m d i f i e d U.S. data were applied for thepresent
a
fairly
-11-scale
study
of
the
estimates:
metropolitan/other urban/rural d i s t r i b u t i o n of t r a f f i c and road
accidents by states, and data on the appropriate distributions
of t r a f f i cf l a w s ,
would provide more refined tim and cost
estinnttes for Australian conditions (p 53);
(x )
Pain and S u f f e r i n g : no Australianstudies are available and no
costestimate was included i n thesepreliminaryestimates,but
tulo approaches are considered warranted to f a c i l i t a t e e s t i r r a t e s
for Australia (p 53) :
(a)
a survey
of
t h e ccrrposition of third
party
(personal
injury) corrpensationawards, with the assistance of State
g o v e r m n ti n s u r a n c eo f f i c e s
(b)
(see item i v , above);
and
an associated
theoretical
study t o assess whether it
would be appropriate to r e l a t e " p a i n and suffering" costs
levels
derived
from (a) a s a proportion of foregone
incm
.
F i n a l l y , it is concluded that the framework of unit costs i nB b l e
1
proposed i nt h i ss t u d y ,f r a n
which the preliminary estimates of total
accident costs were derived
provides
a set of minimm s o c i a l cost
estimtes f o r Australia capableofusefulapplication
i n the evaluation
of road s a f e t y programnes.
However these cost estimates need careful
q u a l i f i c a t i o n i n use and f u r t h e r r e f i n e m n t of individual estimtes. In
particular, the retention of a c l a s s i f i c a t i o n of costs
according
to
injury
severity
which r e f l e c t s Australian conditions is considered
necessary t o pxnit t h e use of t h i s cost framework t of a c i l i t a t e
comparison of a l t e r n a t i v e a c c i d e n t r e d u c t i o n p r o g r m s .
4.2
Recommendations
These recawendations arise frcm the issues covered i n t h e
c o r n of t h i s presentreport,
but also r e f l e c t t h e view t h a t
f u r t h e r work might best proceed by varying the i n i t i a l l y
envisaged
content
of later stages of this
study.
E!ased on
acceptance of the cost framework proposed in the foregoing,
it
is r e c m n d e d that:
(i)
the
preliminary
estimates
of unit and t o t aalc c i d e n t
costs for Australia i n 1978 should b? f u r t h e r developed
and standardised
to
form t h e basis of a regularly
produced set of social accident cost estimates t o guide
58
accident research and policy;
(ii) improvement across a broadrange
of accidentrecords and
statistics relevant to these estimtes is a necessary but
long-term program?; i n the mantime a range of surveys
and s t u d i e s of individual cost conpnents is required t o
achieve wre i r m d i a t ei q r o v e m n ti nt h e
accuracy and
application of these estinrates;theseinclude:
of mdica1,hospital and related
furtheranalysis
costs of m t o r v e h i c l e a c c i d e n t s from the records
of the b t o r Accidents Board of Victoria;
a survey of t h i r d party(personalinjury)
claim
on one or mre State g o v e r m n t i n s u r a n c e o f f i c e s ,
and court costs, court awards
t o deterrrinelegal
for pain and suffering, and other such cost i t e m ;
f u r t h e r survey and analysis of conprehensive mtor
vehicle insurance clains t o inprove the estimates
of v e h i c lper o p e r tdym g e
costs, and t h e i r
statistical distributions (including consideration
of t h e "excess" pxid by owners, and t h e problemof
uninsured vehicles);
a series of r e l a t i v e l y mall socio-economic
value
of c e r t a i n
s t u d i e s of the non-mrket
i n Australian
conditions,
accident
effects
losses a t t r i b u t e d to families
includingaccident
and c m n i t y , employers and i n d u s t r yt ,r a f f i c
delays, and t o p a i n and suffering;
a study of the need f o r and f e a s i b i l i t y of
d e r i v i naginn j u rsye v e r i tcyl a s s i f i c a t i o n
of
accident cost data
for
Australian
conditions,
together
with
an
appraisal
of its use i n the
evaluation of road safety p r o g r m s .
of more detailedaccident cost estimtes
( i n t h e f o m t of the present report) by type of accident
as envisaged i n subsequent
and by region(urban/rural)
stages of t h i s studyshould
be proceeded with according
to prescribed areas and c l a s s i f i c a t i o n s .
(iii) theproduction
59
B1BLIQ;RAPHY AND
Acton,
(1976)
"Measuring
the k n e t a r y Value of Life
The RAND Corporation,
Santa
hbnica,
California
(paper t o Cbnference on The Value of k
n Life,July
pp.43.
J.P.
Saving
FVogram"
1976)
(1979)
"?he Value of Life Saving:
Implications
of
Consumption Activity" i n Journal o f P o l i t i c a l Economy V o l . 87
f.Jo.3, pp. 510-558.
Blmquist,
Bolnn, P.
G.
(1971)
"An Approach t o t h e Problem of F s t ~ m t i n g k w n d f o r
(Eds)
The
Public Goods" i n Bohrn, P. and Kneese, A.V.
Economics of E n v i r o m n t ,
.%rtin, London.
.
St.
B
r
m
, J.
(1978)
"Trying
EcO?WmiCs V O ~9.(pp.
Conley, B.C.
(1976)
t o Value
a
91-100). .
"The Valueof
Life" i n J o u m l of
H m n L i f ei nt h e
i n American Economic Review, Vol. 61. No. 1.
Cooke,
Public
Demnd forSafety"
(pp. 45-55).
P.J.
(1978).
"The Value of Hwan Life i n t h e Derrand f o r
Vol
68
(Sept.
Safety: Cmment" i n American Economic Review,
1978) pp. 710-721.
Casypta,
A.K. and Pearce D.W.
kcmillan, London.
(1972).
Cost B e n e fAi tn a l y s i s ,
Cawson,
R.F.F.
(1967) Cost of Road Accidents i n Great Britain,
ResearchLaboratory, Crowthorne.
%ad
Departrent of Transportation (U.S.) (1975). ProceedingsoftheFourth
July 14-16, 1975,
International Congress on AutomotiveSafety,
Washington D.C. a m e t i n g of t h e National Motor Vehicle Safety
AdvisoryCouncil
(a group t o advise U.S. D.O.T.,
administered
by the I k t i o n a l Highway Safety
pdvisory
Cbuncil);
theme:
"Considerations i n Determining P r i o r i t i e s m n g hbtor Vehicle
S a f e t y Standards:Effects
of SafetyStandards - Beneficial and
&asurement and @antifieation of Effects Adverse;
Methodological
Issues;
and Qnsiderations
Social and
P o l i t i c a l " , pp. 1003.
-
Dodge,
L. (1976)
"Politics and Benefit/&&
Analysis i n NHTSA
Rulermking" , Presentation t o (U.S .) Iktional&torVehicle
Safetypdvisory Cbuncil Seminar on k b l i c R l i c y , Eblitics and
&tor Vehicle
Fafety
Standards,
(14 July
1976);
(pp.
9)
Author is Special Cbunsel, &use Qmnittee on I n t e r s t a t e and
Foreign Camzrce (U.S. Congress).
Dyson, R.B. (1975). "Safety Versus Savings: An &say on theFallacy of
i n Proceedings o f 4 t h I.C.A.S....
Emncmic Costs ofAccidents"
(pp. 145-153).
60
ttoad Safety"
(1975)
The
Road
AccidentSituation
in
A u s t r a l i ai n
1975, (A reporttotheAustralian
Ninister for
,Transport by the Expert Group on Road S a f e t y ) ;P a r l i a m n t a r y
pitper No. 274, Canberra, 1976.
"Expert Group
on
Faigin, B.M.
(1975)
"Societal
Costs
of b t o r Vehicle
Accidents
for
A krspective on the b j o r Issues and
Benefi~t-stAnalysis:
SOE
Recent Findings"
i n Proceedings o f
the Fourth
International Congress on A u t o m t i v eS a f e t y , July 14-16, 1975,
U.S. b p a r t m n t of Transwrtation.
National Highway T r a f f i c
@p. 155-172).
SafetyPdministration, Washington D.C.
1975 Societal
Costs
o f Motor Vehicle
Accidents,
NHTSA, Washington, D.C., December 1976.
Faigin, B.M. (1976)
U.S. D.O.T.,
Elsevier,
Fisher, G.H. (1973) CostConsiderations i n SystemsAnalysis,
New York (especiallyChapter
3:
"Concepts of Fconomic @st"
pp. 24-62).
Fleischer, G.A.
and Jones, G.P. (1975).
"Cost-Benef it and &stBfectivenessAnalysis
i n Determining Priorities arlmng k t o r
Vehicle Safety Standards, Frograrrs and Projects" i n Proceedings
of 4th I.C.A.S....(pp.
173-192).
Flora,
Bailey, J., and O ' l h y , J. (1975)
The Financial
Consequences of Auto Accidents HIT LAB Reports Vol. 5 No. 10,
Highway Safety&search
I n s t i t u t e , University of Uichigan, Ann
Arbor (pp. 7).
<J.D.,
Fox, J.C.,
Good, M.C. and Joubert, P.N. (1979) C o l l i s i o n sw i t h U t i l i t y
Poles, University of k l b o u r n e f o r O f f i c e o f Road Safety, Dept.
of Transport ( C R 1).
Fraser, L.M.
(1947)
Economic Thought and Language, A.C.
Reerran, A.M. (1979) "The Value ofLongevity",chapter
i n The B e n e f i t s on EnvironmentalImprovement,
Resources f o r "ne Future, hltimre.
Black,bndon.
7(pp.
165-194)
Johns Hopkins for
Gates, H.P. (1975)
"Review and Critique of W t i o n a l Highway T r a f f i c
Safety
Administration
Revised
Restraint System Cost Benefit
(pp. 209-234).
Analysis i n Proceedings of 4th I.C.A.S.
and Brown R. (1978) "LifeSafety:
W
h
a
t is it and h w mch
is it worth", (U.K.) Building
Research
Estabtishment
(Fire
Research Station) current paper, CP52/78 (pp.7).
Green, C.H.
Gronau, R. (1973) The Measurement of Economic and SocialPerformnce,
k t i o n a l Bureau of Economic Research, New York.
Harrison, A.H. (1974)
Landon.
The Economics of TransportAppraisal,
Qmklm,
61
Henderson, M. (1978).
"The Value
considerationsinTrafficSafety"
Feb 1975) pp. 19-23.
Japan&searchOentreforTransport
Road Accidents(pp.55).
of
I I m n Life:
Cbst-knefit
i n Search Vol. 6, 1-2 (Jan-
a l i c y (1978)
S o c i a la s s e sf r o m
Joksch, H.C. (1975)
"A Critical Appraisal
of
the a p p l i c a b i l i t y of
benefit-cost analysis t o highway t r a f f i cs a f e t y "
i n Accident
Analysis and Preuention, Vol. 7 (pp. 133-153).
Jones-he, M. (1969) "Valuation of Reduction i n Probability of Death by
Road Pccident" i n Journal o f Transport Economics and Policy
Vol. 3, No. 1 (January 1969) pp. 3 7 4 7 .
Jones-Lee,(1974)
"?he value of changes i n the Probability of B a t h or
Injury" i n Journal of P o l i t i c a l Econony V01.12 (pp. 249-256).
Jones-Lee, M.W. (1979a) "Trying t o Value a Life: why E
r
m does not
sweep clean" i n Jou&t o f Public Economics, Vol. 12 (pp. 249256).
Jones-Lee, M.W. (1979b) "Trying t o Value a Life:
PublicEconomics", Vol 12 (pp.259-262).
A reply" i n Journal o f
Iieeney, R.L. and Rdiffa, H. (1976) DecisionswithMultipleObjectives,
John Wiley & Sons, New York.
Ker,
I.K. (1978)
Value and Value
Judgements
i n Road. Accident
Director
General of
Evaluation, Draft Report, Office of the
Transport, Western Australia (pp.46)
.
Lane,
J.C.
(1964) "The b n e y Value of a hhn", paper t oS e t i o n
J
Australian and Mw Zealand Associationfor the Pdvancerent of
20 January 1964, (knkn-a A.C.T.
(pp
Science,
37th
Congress,
9).
l a n e , J. C. (1968) "Safety i n Transport", Pansport Section, Dept. of
Civil
Engineering,
University
of &lbourne
( l e c t u r e s on
TbtionalTransport€blicyOctober,
1968) pp. 19.
(1978) "The Bsts of Road Accidents and theirApplication
i n F%onomic Evaluation of S f e t y P r o g r m s " , Proceedings of
Annual Conferenceof
Roads and TransportationAssociation
of
Canada, (Sept. 18-2, Ottawa. (pp. 30) .
Lawson, J.J.
Layard, R. (ed)
marsh,
(1972)
Cost-BenefitAnalysis,
Penguin, H a m n d s w r t h .
J.C.,
Kaplan, R.J.
and Kornfield, S.M. (1977).
Financial
Consequences
Serious
of
Injury,Highway
Safety Research
I n s t i t u t eU
, niversity
of Michigan ( F i n a l k p r t ) Ann Arbor.
UM-HSRI-77-27.
62
Mishan, E.J. (1971 ( b ) ) "Evaluation of Life and Limb" A Theoretical
Approach" i n Journal of P o l i t i c a t Economy, V o l 79 (July1971),
pp. 687-705.
fishan,
i n Times
(1976)
" m a r r e l l i nwgitthQ
hel a n t i f i e r s "
Literary Supplement 8 October 1976 p. 1282.
E.J.
and Graybill, F.A. (1963) I n t r o d u c t i o tntoh e
Theory of
S t a t i s t i c s , Mdraw Hill-Kogkkusha, New York [espec. chapter 6 ,
p126 f f ]
kod, A.M.
.
Mooney, G. (1978)
" W n Life and a f f e r i n g " , Qlapter 6 i n Rarce
(1978) pp. 120-139.
and
-rave
P.B. (1976). Public Finance i n Theory and
Practice, (Second Edition) McGraw-Hill Kogakusha, Tokyo.
Musgrave, R.A.
i n Journal of PotiticaZ
(1962) "Health as anInvestment"
Economy, Vol. 70 (pp. 129-157)
Mushkin, S.J.
N.B.T.S.A.
(1972) Societal
Costs
of
PreliminaryReport, U.S.
COT,
Motor
Vehicle
Accidents:
Washington, D.C.
Nicholas Clark & Associates (July 1979) "Study Design f o r t h e E v a l u a t i o n
of the
Effectiveness
of Miters
Type Projects:
S m r y
Internal Research Report for Office of Road Safety.
-
"Evaluating b t o r
i n Proceedings of 4th
O'Neill,
B., Kelley, A.B.,
and Wong, J.
VehicleSlfetyRrfomnceStandards",
I.C.A.S.
(pp. 551-560).
(1975)
Paterson,
J. (John
Paterson
Urban Systems 1973), The Cost of Road
Accidents i n Relation t o Road Safety, Report No. NR/23 f o r t h e
Printer, Chnberra.
ikpt. of Transport,
Govt.
The ValuationofSocialCost,
Fearce, D.W. (ed).(1978).
Unwin , London.
George Allen &
made, E.S. (1970) "Cost-Effectiveness:
Som "rends i n Analysis" The
RAND Corporation, Santa k n i c a , (distributed by U.S. Clearing
house) pp .26.
"%e @ s t of Faad Pccidents" i n Journal of the
Royal S t a t i s t i c a l S o c i e t y , Vol. 119 No. 4 (pp. 393-408).
h y n o l d s , D.J.
(1956)
Saunders, A.B. and Benson, D.A. (1975) "The Practical
Application
of
SocialCosting i n &ad S a f e t y m l i c y W i n g " i n Proceedings of
4 t h I.C.A.Sm...(p
591-604).
.
(1968) "'he Life You Save bhy Be Your ckun" (pp 127-176),
i n Problems i n Ilcbtic
Expenditure
Analysis
(ed) S.B. Chase,
Schelling, T.C.
Bookings Institution,
S e l f , P. (1975).
Washington D.C.
Econocrats and thePolicyProcess,
MacMillan, London.
63
Sherwin,
M.A.
(1976) "Road Accident
Costs"
Accidents:Papers
and Reportfrom
Accidents, ivbrch, 1977,Wellington,
b a r d , 1978) (pp. 7-31)
.
i n !%e Cost of Road
a Workshop on Cost of Road
New Zealand (IktionalWads
Smeed, K.J. (1972) "The Usefulness of F o m l a e i n TrafficEhgineering
and Road Safety" i n Accident AnaZysis and Prevention, Vol. 4
Pergamon Press (pp. 303-312).
Spiegel,
M.R.
(1961)
Theory and 2robZem
Publishing 8. New York. [chapter 51.
Struble,
D.,
Peterson, R. Wilcox, B., and R i e d m n , D.
(1975).
"Societal bsts and their k d u c t i o n by Safety System" i n
Proceedings of 4th I.C.A.S.. . (pp. 695-779).
of
Statistics,
&ham
Thaler & Fbsen (1975) i n HotisehoZd Production and Consumption,
N.E.
Terleckyj(ed). , Studies i n Income and Wealth, Vol. 40, by the
Conference i n I n c a and Wealth I k t i o n a l Eureau of F c o n d c
Research, New York.
Thorn, H.C.S.
(1958) "A Note on t h e Carma Distribution", MonthZy Weather
117-122).
Review U.S. kpartrrent of Cmmrce(pp.
Treasury
kpartment
[1978]
"Discounting
i n Australian mblic Sector
ProjectAnalysis" Draft Fapr, C a n k r r a ACT pp.61.
Thmon,
J.M.
(1974)
brmondsworth.
Economics,
Modem
Transport
Penguin,
lhorpe, J.D. (1970) "Estirmted Cast of Road Accidents i n Victoria 196667" i n AustraZian Road Research Vol. 4 No. 3 ( k r c h 1970)pp.
55-70.
T r i l l i n g , D.R. (1978) "A Qst-EffectivenessEvaluation
of Highway Safety
CMmter-nEasures" i n T r a f f i c QuarterQ, Vol. 78 No. 1 (January
1978) pp. 41-66.
and b t l i n , N.G.
hlelburne, (pp.292).
Troy, P.N.
(1971)
The
Cost
of CoZZisions,Cheshire,
Willians, A. (1979) "A NDte on 'Trying t o Value a Life'" i n J o u m t o f
Public Economics, Vol 12. (pp. 257-258).
Wilson, R.
(1979)
"Analyzing
the h i l y Risks of Life" i n TechnoZogy
February
1979,
Revim, Massachusetts I n s t i t u t e of Technology,
(pp.41-45).
Winch,
D.M.
(1971)
AnaZyticaZ
Ha.mndsworth, U.K.
Vetfare
Economics,
Wnguin,