An application proposal of yardstick competition to the regional - - PowerPoint PPT Presentation

An application proposal of yardstick competition to the regional markets of the French railway system

Julien LEVEQUE

jleveque@let.ish-lyon.cnrs.fr

Laboratoire d’Économie des Transports (CNRS) 5 October 2004

ETC 2004 Strasbourg

2

The French railway services (freight, inter-city, regional):

Framework The French monopoly problem The French monopoly problem

ETC 2004 Strasbourg

The regional rail transport services (TER): suburban trains interurban trains rural trains number of operators

• perators

1938 1 SNCF 1 SNCF 1 SNCF 1 SNCF 1983 1994 2004

3

Framework Current Current organization

• rganization

ETC 2004 Strasbourg

Average regional network in 2002: 450 M P.K 7.1 M Tr.K 1157 km of lines 70.7 km/h traffic revenues = 40 % costs

• perating income: 100 M €
• perating result: 3.7 M €

64 P/Tr

4

Contracts (5-10 years) define PSO and subsidies: quality incentives: bonus/malus traffic incentives: SNCF bears the commercial risk cost incentives: “RPI” + 0!!!!!!!!!

Framework Which incentives? Which incentives?

ETC 2004 Strasbourg

no call for tender 20 regional regulators 1 operator delegate TER services captures lack of competitive pressure on costs

5

Shleifer: Estimating what should be the best prices and subsidies, by comparing the performances of several similar, regulated firms:

Economic principles The yardstick competition mechanism The yardstick competition mechanism

ETC 2004 Strasbourg

1 N j firm

• f

costs i firm

• f

revenues

i j

− =

∑

≠

costs of the compared firms have to be homogenous!

6

Comparing the performances of the 20 regional SNCF departments Implementation by a common agency of the regional regulators Which incentives?

• financial gains / penalties
• reputation effect

Economic principles The proposed scheme: The proposed scheme:

ETC 2004 Strasbourg

7

reduction of the uncertainty due to:

Economic principles Benefits of the use of comparisons Benefits of the use of comparisons

ETC 2004 Strasbourg

• prices increase
• changes in labour legislation
• national strikes
• spatial organization
• labour organization

reduction of the informational rent due to:

8

endogenous heterogeneity: due to political decisions in defining PSO Nash: customer = government,

• utput = level of service: Tr.K (≠ number of P.K)

exogenous heterogeneity: due to the environment (infrastructure, society)

Measure of efficiency Correcting external heterogeneity Correcting external heterogeneity

ETC 2004 Strasbourg

introduction of environmental variables in the model

9

data issued from the regional accounts of the SNCF (1997 & 1998)

Measure of efficiency The cost frontier model The cost frontier model

ETC 2004 Strasbourg

control of the external heterogeneity:

it 1 k it k, k it

Z u ε + δ + δ =

∑

≥ it it 2 k it k, k it 2 it 1 it

u v lnw lnL.L lnTr.K lnC + + β + β + β + β =

∑

>

stochastic cost frontier:

10

Measure of efficiency Estimation results Estimation results

ETC 2004 Strasbourg

Dependent variable: ln C – Observations: 40 1st model 2nd model 3rd model Variables Coefficient t-test Coefficient t-test Coefficient t-test Constant 1.266 5.45 0.8948 62.1 2.460 12.4 ln Tr.K 1.177 35.0 1.226 78.2 1.033 39.0 ln L.L

• 0.08707
• 1.88
• 0.09949
• 3.46
• 0.08723
• 2.01

Constant 0.05309 0.16

• 0.05082
• 1.07
• 0.03764
• 1.98

Delinq 2.088 2.02 2.555 3.41 3.584 4.91 Load 0.01014 5.39 0.005219 14.1 P.K

• 0.8762E-3
• 2.63
• 0.7097E-3
• 118

Exp

• 0.06364
• 1.04

Speed

• 0.005289
• 1.50

K 0.1998E-5 0.23 γ 0.966 19.8 0.999 688 0.999 237 Log-L 63.94 71.76 63.38

11

Measure of efficiency The cost frontier The cost frontier

ETC 2004 Strasbourg

9 9,5 10 10,5 11 11,5 12 12,5 20000 40000 60000 80000 100000 120000 140000 160000 agregated output ln C

• bservations

frontier

12

Measure of efficiency Efficiency scores Efficiency scores

ETC 2004 Strasbourg

0,9 0,95 1 1,05 1,1 1,15 1,2

C e n t r e L i m

• u

s i n N P D C F r a n c h e

• C
• m

t é C h a m p a g n e

• A

r d e n n e s P

• i

t

• u
• C

h a r e n t e s B

• u

r g

• g

n e L a n g u e d

• c
• R
• u

s s i l l

• n

B r e t a g n e A q u i t a i n e A l s a c e P A C A R A A u v e r g n e H a u t e

• N
• r

m a n d i e M i d i

• P

y r é n é e s P a y s d e l a L

• i

r e P i c a r d i e L

• r

r a i n e B a s s e

• N
• r

m a n d i e

13

Yardstick competition could be implemented:

• while maintaining the legal monopoly of the SNCF
• external heterogeneity can be controlled
• internal heterogeneity (inefficiency) is quite important

Franchising should be introduced in addition

Conclusion Towards a more incentive regulation scheme Towards a more incentive regulation scheme

ETC 2004 Strasbourg

14

Conclusion

Thank you Thank you for your attention for your attention

ETC 2004 Strasbourg