Infrastructure                                       184
aX,oNAL                   __        _                                        June 2001
^,   , ,    ,   Findings reports on ongoing operational, economic, and sector work carried out by the World Bank
and its member govemments in theAfrica Region. Itispublishedperiodicallybythe Knowledge and
Leaming Center on behalf of the Region. The views expressed in Findings are those of the author/s
j WORRLD BA N K               and should not be attributed to the World Bank Group.
Urban Transport Dysfunction
IVON AN9                  and Air Pollution in Dakar
Study Conclusions
his note presents the main  urban mobility in the Dakar ag-
T   conclusions of an on-site  glomeration. This project is to be
study of urban transport dys-  implemented under the auspices
function and air pollution in the  of the World Bank.
~~~ ~~Dakar agglomeration carried out
from August to November 1998.
On December 17 and 18, 1998, the
findings were discussed at a na-    The study's goal was to define the
tional seminar on urban transport  main causes of transport dysfunc-
and air quality organized jointly by  tion, quantify them, express them
CETUD (The Executive Council for  in figures, and recommend a suit-
Urban Transport in Dakar) and the  able framework for monitoring
World Bank. This formed part of  changes in the parameters. The
the Sub-Saharan African air-qual-  area studied is the Dakar agglom-
!  s             ~~~~~ity initiative that was launched  eration, which has more than 2.2
jointly in 1998 by the Urban Trans-  million inhabitants, 1.2 million of
port Component of the Sub-Sa-  whom live to the east of the city
haran Africa Transport Policy Pro-  itself, in the Pikine-Guediawaye
gram (SSATP) and the World Bank  Rufisque area.
r                           Institute (formerly the Economic
Development Institute). The report  Problems
on this seminar (SSATP Working    The fundamental problem, as in
Paper No.38, February 1999) is
available from the SSATP Coordi-  most large cities, is that the resi-
nation Unit. This note was origi-  dential location is not coordinated
nally published as SSATP Techni-  with work place location, and ev-
cal Note No. 19.                 ery day this leads to large-scale
As regards the particular prob-  movements of population. Viewed
As regards the particular prob-  oeal hrfr,tepolmi
lem of air pollution caused by ur-    all   to  the qsonlof it
_~ ~~~~~~~bntasot h  td  n  h   also linked to the question of city
_                    ~~~~~~~ban transport, the study and theplnigadthsoaln-er
_                                            ~~~~~~~~~~~~~~~~planning, and thus to a long-term
U  recommendations formulated at  view of the development of the city
the seminar made it possible to
define the principles on which an
action plan could be based. The    The map provides a clear picture
atnl a n could be. based Ter   of these problems, and shows the
lattr iscretybigcniee   heavy flows on Dakar's east-west
for financing as part of the prepa-
ration of a project for increasing  corridors, especially during rush
hours.



COMPOSED VOLUMES ON AUTO NETWORK                        These social costs are borne by
CLI'SSPECIFIC AI r[ voC UMES        L          the local authority, rather than di-
'7OL'CC0C  rectly by the individual users (even
"' ('6<  though it is user demand that
causes transport dysfunction).
The importance of evaluating
externalities
It is important to make an accu-
rate evaluation of the costs of dys-
function (or transport externali-
SCILL: 1715  ties), and to decide where the re-
sponsibilities lie, so that these
costs can be properly allocated;
however, this sort of evaluation
also plays a major role in provid-
C OCCO    maling "decision support" when trans-
IV .111i RMIILkI  e<;I;I", vn<uInI.sod!)Sa              sR4S< 0l MIC  port systems  are  selected. Failing
T. 7   TRI 2BL 1 to include all the social costs in the
Definitions                        equation means that the real cost
Added to this are the following                                     of a system will be underestimated,
problems:                            However, what exactly is "trans-  and that a given system may be
theincreaseinpport dysfunction"?   favored over another that would
* the mncrease in vehicle ownership   pr  yfnto"
* the inadequacies of the infra-    In reality, it consists of "costs  have generated fewer external
structure in general, and inter-  arising from situations in which a  costs (by creating less pollution, for
sections in particular, since they  transport user does not pay for all  example).
are un e to cthe costs generated by the trans-    Thus, an evaluation comparing
are unable to cope with this in-
creased traffic.                 port service (or else does not re-  a large-capacity bus project and a
ceive all the anticipated benefits).  suburban rail project could lead to
This leads to the following kinds    Overall, transport costs are of  a decision in favor of the latter be-
two types:                         cause it was demonstrably safer
of dysfunction:
r   those that appear evident - i.e.,  (resulting in fewer accidents and
in roadesafety pnjuroblems rest i  the operating costs generated by  traffic jams) and created less pol-
n death, ijury, and damage to   the activity (fuel, staff, etc.) and  lution.
prpet                          its infrastructure (roads), these
* congest-ion (and, as a result, time itinrsucre(od)the
wasted in traffic jams)            largely being covered through   Ways of measuring dysfunction
wastediin ttaxation; and,                                       .       The following are the generally
* air pollution                    * costs generated by dysfunction;  accepted methods of calculating
* noise pollution.                  these are less evident, but con-  external costs:
sist of accident-related costs (i.e.,
Transport demand                     health care, the value of lives  *ot avod; cos onsiststof
According to a 1998 study, daily    and environmental costs (atmo-    identifying and quantifying the
commutes in the Dakar agglomera-    spheric pollution and noise pol-    costs of measures for reducing
tion total 4.3 million, 70 percent    sutio  aosts aud  by con-    externalities;
of which are made by mass tran-    gestion, represented by the time   * the damage costs evaluation
sit, which therefore plays a major    wasted by passengers when us-    method; this consists of quanti-
role, and has been singled out as    ing public transport).            fying the observable impact (fa-
being responsible for much of the                                       talities, health care, and produc-
dysfunction.                                                           tivity losses);



* the hedonic method; this con-                           Estimated costs of the impact of dysfunction in Dakar
sists of assessing the difference
sistsofassessingthe Type of dysfunction  Number  Unit cost  Total cost  % of GNP
in value between a good located                                             (CFAF)         (CFAF)       96
in a place subject to dysfunc-    Bodily injury                                                      2,372.7
tion (i.e., pollution and noise)    Fatalities, per year'             179  6,600,000    1,181,400,000
and the same good in an un-    Serious injuries, per year            1,053   800,000    842,400,000
and the same good in an un-       Total                         ~~~~~~~~~~~~2,023,800,000  0.1
polluted place;                     Congestion
* the stated preference method, in    Hours wasted, per year      224,400,000    185   41,401,800,000    1.7
Pollution
which interviewees who are vic-    Number of inhabitants affected by
2 pollutants (PM10& Pb)2      2,100,000    30,000   63,000,000,000  2.7
tims of transport dysfunction       Total cost                                       108,449,400,000    4.6
are surveyed to find out how
much they would be willing to        Estimated by means of the "lost income" method; it should be noted that the 'willingness to pay" method results
t not be subjected   in figures that are from 25 to 30 times larger
pay to not be subjected to it, or   2 Estimated on the basis of figures from a study made in Jakarta (Urban Air Quality. World Bank Paper NO 379, 1997.)
to receive some form of com-
pensation .
Thenstatedn. preferene method         Congestion and traffic jams         * comparison with European stan-
The stated preference method 
produces results closest to eco-   * using the EMME2 model, calcu-    dards; measurements  have
nomic values because it reflects    lation of the hours wasted in ve-    shown that the noise level dur-
market prices; however, it requires    hicle trips, measured according       ing the morning rush hour is al-
more resources than were avail-    to the difference between travel-    ways higher than  acceptable (65
able for this study.                   ing times on a congested system       dBA), on both primary and sec-
and  on  an uncongested system;    ondary routes; at other times,
calculaton ofuthesuni coster    however, levels on secondary
Methodology adopted                     waste ou of          theu   costp    routes are acceptable.
wasted hour, from which the cost
The damage costs evaluation           of total hours wasted can be de-    The models used to evaluate
method was adopted, since it can       duced;                               these costs make it possible to
be applied even when the users in                                          monitor trends over time, or to pro-
question have little or no appre-                                           duce  simulations  based  on
ciation of the extent of the dam-    Airpollution                          changes in transport-system  pa-
age caused, or else miscalculate it.  . calculation of quantities of pol-  rameters.
The methods of evaluation, by type    lutants emitted by traffic (CO,
of dysfunction, were as follows:       VOCs, and NOx), and their con-  Results
centrations, based on the char-
acteristics of traffic in Dakar as
Bodily injury                         defined by a model (EMME2), and   of the calculations and estimates.
* gathering of accident statistics,    on information on the emission   It shows that the costs of the main
classified by degree of serious-    factors for a given vehicle fleet;   sources of dysfunction are equiva-
ness;                               * comparison of these levels of pol-  lent to about 5 percent of GNP;
* calculation of unit costs (by type    lution with WHO standards; it  however, it should be noted that
of accident) and of the cost of    would be helpful if these data  the pollution costs are estimates
human losses, from  which the         could be calculated on the basis   extrapolated from  studies that
total cost can be deduced;            of samples collected by monitor-  were made in Asia (in Jakarta) over
* definition of indicators for acci-    ing stations - in Dakar, no such   a number of years, and from which
dose-effect curves have been de-
dents and fatalities caused by       measurements have so far been   rivedfthrureessio n ali
public transportation, and estab-    made;                                rived through regression analysis
lishment of the relationship of                                           Moreover, the costs as estimated
are not entirely due to transport,
tehilese figures the number of kmtNoise                                   since there are other sources of
vehicles x the number of km trav-
eled (as ascertained, adjusted, or  * calculation of the noise level gen-  pollution in Dakar (e.g. electric
simulated by means of a traffic    erated by traffic (using data from   power plants, a cement factory,
model);                               EMME2 and a computer model  various industries, etc.).
prepared by the consultant);



Number of persons> Long-term standard  - improving fuel
Period of one day               quality, and intro-  The Urban Transport Component
ducing tax incen-
{400000f,.  _ _ _ _   __ __  _ j           tives for the use of
1200000                             f --        cleaner fuels.        The main objective of SSATP's
Io                      .           . e f        L -u-u',  *  REF I  Urban Transport Component is to
800                                            -Operational mea-
a ' 800000 -0 00 0   l  __  .__    =    '*olmeaVEH             promote reforms in transport
IJJ 6(0000 ,~E              .URB   sue  f               policy. It seeks to achieve this
400000UR                                        ue  frices
4200000o                                                             ' 11 E----. objectiveing the pollution-  by undertaking activities
-                                           btmntef-that will improve institutional and
CO    NOx   H-C   S02   PMIO            ciency of public  regulatory frameworks and road
transport through   safety, particularly with regard to
Key REF: base scenario VEH: measures applicable to vehicles; URB planning   changes in the pa-
OPE operational measures.                            rameters of supply.    udtes  ofety, byncrring out
Thiscatgoryin-studies of microentreprises and
A strategy for combating air         clude                            launching an air-quality manage-
pollution caused by transport         uds.                            ment initiative, and by strengthen-
Air pollution caused by urban   -rorderiztiono achiev   otimumt in-  ing local capacity. Currently, the
transport can be reduced through      order to achevr   opthmum in-  Component has 18 member
measures in a number of areas:        ous         opes   of t hensport;countries.
institutional, technical, or organi-   ous modes of transport;          This article wasfirstpublished
zational. All of these require forms    - actilon for easig the flow of ve-  as SSATP Technical Note No. 19,
of investment that differ as to cost,  hclrtafc(rogatan_May 1999. For more information,
difficulty of implementation, and     port plan, street improve-  contact Mr. Patrick Bultynck,
timing.                               ments, dissemination of infor-  Coordinator of the Urban Trans-
So thal: the effects of the various  mation for users, and stagger-  port Component. Internet address:
measures can be compared, they        ing of rush hours).             pbultynck(worldbank.org
havebee  divdedint thefolow- Finally, measures for reducing
have been divided into the follow-    demand; these mainly relate to
ing three main categories, accord-    issues of urban planning, such
ing to their intrinsic features:     as         uran planningtional
as administrative or educational
* Measures applicable to the ve-    decentralization.
hicles themselves, the aim being    Traffic levels, the resulting emis-
to reduce overall pollution by re-  sions, and the impact on air qual-
ducing individual emissions;  ity were then calculated using
these include:                   mathematical models reflecting the
- introducing tax and regulatory  current situation (the base sce-
measures governing vehicle  nario), together with three other
imports, with the aim of reduc-  scenarios that would be produced
ing the number of older ve-  by applying the three categories of
hicles in operation;           measures defined above. For ex-
- providing existing technical  ample, the following chart shows
monitoring centers with equip-  how, in the current situation, the
ment for measuring engine  NO. pollution standard is ex-
emissions, with special train-  ceeded.
ing for staff and new rules de-
fining the standards that ve-
hicles must meet;