Efficiency in the Public and Private French Water Industry: - - PowerPoint PPT Presentation

▶
efficiency in the public and private french water
SMART_READER_LITE
LIVE PREVIEW

Efficiency in the Public and Private French Water Industry: - - PowerPoint PPT Presentation

Sep 06, 2023 177 likes •555 views

For Starters Motivation Empirical model Data Results Conclusion Efficiency in the Public and Private French Water Industry: Prospects for Benchmarking Aude Le Lannier and Simon Porcher IAE Sorbonne Business School October, 2011 1/12 Aude

slide-1
SLIDE 1

For Starters Motivation Empirical model Data Results Conclusion

Efficiency in the Public and Private French Water Industry: Prospects for Benchmarking

Aude Le Lannier and Simon Porcher

IAE Sorbonne Business School

October, 2011

1/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-2
SLIDE 2

For Starters Motivation Empirical model Data Results Conclusion

Water Provision in France

◮ In France, municipalities must provide local public services on

behalf of their citizens.

2/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-3
SLIDE 3

For Starters Motivation Empirical model Data Results Conclusion

Water Provision in France

◮ In France, municipalities must provide local public services on

behalf of their citizens.

◮ These public services can be managed in-house or contracted

  • ut to a private operator.

2/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-4
SLIDE 4

For Starters Motivation Empirical model Data Results Conclusion

Water Provision in France

◮ In France, municipalities must provide local public services on

behalf of their citizens.

◮ These public services can be managed in-house or contracted

  • ut to a private operator.

◮ There is no national regulator: municipalities monitor prices,

control entry and exit of operators and ensure uninterrupted service.

2/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-5
SLIDE 5

For Starters Motivation Empirical model Data Results Conclusion

Water Provision in France

◮ In France, municipalities must provide local public services on

behalf of their citizens.

◮ These public services can be managed in-house or contracted

  • ut to a private operator.

◮ There is no national regulator: municipalities monitor prices,

control entry and exit of operators and ensure uninterrupted service.

◮ A hot political topic.

2/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-6
SLIDE 6

For Starters Motivation Empirical model Data Results Conclusion

Motivation: conventional wisdoms differ (1/2)

◮ Public sector is more efficient:”...water is a public good [...]

it is better managed by public authorities“ (Water warriors, Mayor of Paris)

3/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-7
SLIDE 7

For Starters Motivation Empirical model Data Results Conclusion

Motivation: conventional wisdoms differ (1/2)

◮ Public sector is more efficient:”...water is a public good [...]

it is better managed by public authorities“ (Water warriors, Mayor of Paris)

◮ Private sector is more efficient:”...if we take into account

the cost of water for the whole society, private sector is more efficient...“ (Big private operators, BCG)

3/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-8
SLIDE 8

For Starters Motivation Empirical model Data Results Conclusion

Motivation: conventional wisdoms differ (1/2)

◮ Public sector is more efficient:”...water is a public good [...]

it is better managed by public authorities“ (Water warriors, Mayor of Paris)

◮ Private sector is more efficient:”...if we take into account

the cost of water for the whole society, private sector is more efficient...“ (Big private operators, BCG)

3/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-9
SLIDE 9

For Starters Motivation Empirical model Data Results Conclusion

Motivation: conventional wisdoms differ (1/2)

◮ Public sector is more efficient:”...water is a public good [...]

it is better managed by public authorities“ (Water warriors, Mayor of Paris)

◮ Private sector is more efficient:”...if we take into account

the cost of water for the whole society, private sector is more efficient...“ (Big private operators, BCG)

◮ Theoretically, no clearcut answer about public vs. private

efficiency.

3/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-10
SLIDE 10

For Starters Motivation Empirical model Data Results Conclusion

Motivation: conventional wisdoms differ (1/2)

◮ Public sector is more efficient:”...water is a public good [...]

it is better managed by public authorities“ (Water warriors, Mayor of Paris)

◮ Private sector is more efficient:”...if we take into account

the cost of water for the whole society, private sector is more efficient...“ (Big private operators, BCG)

◮ Theoretically, no clearcut answer about public vs. private

efficiency.

◮ The empirical evidence on cost-efficiency is mixed:

Better efficiency for private management: Kirkpatrick and al. [2006] on Africa. Higher efficiency for public management: Bhattacharyya et al. [1995] on US data. No significant differences: Saal and Parker [2000, 2001] on UK, Estache and Rossi [2002] on Asia.

3/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-11
SLIDE 11

For Starters Motivation Empirical model Data Results Conclusion

Motivation: breaking two puzzles (2/2)

◮ What happens if we take into account the need for sound

public finance?

4/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-12
SLIDE 12

For Starters Motivation Empirical model Data Results Conclusion

Motivation: breaking two puzzles (2/2)

◮ What happens if we take into account the need for sound

public finance?

◮ Can we find the 10% gap efficiency between public and

private operators?

4/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-13
SLIDE 13

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-14
SLIDE 14

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-15
SLIDE 15

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

Step 1: run a simple DEA.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-16
SLIDE 16

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

Step 1: run a simple DEA. Step 2: regress the slacks for each outputs and inputs on the contextual variables using a SFA. Step 2: correct the inputs and the outputs by putting them in the same operating environment and state of nature.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-17
SLIDE 17

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

Step 1: run a simple DEA. Step 2: regress the slacks for each outputs and inputs on the contextual variables using a SFA. Step 2: correct the inputs and the outputs by putting them in the same operating environment and state of nature. Step 3: re-run a DEA with the corrected variable.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-18
SLIDE 18

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

Step 1: run a simple DEA. Step 2: regress the slacks for each outputs and inputs on the contextual variables using a SFA. Step 2: correct the inputs and the outputs by putting them in the same operating environment and state of nature. Step 3: re-run a DEA with the corrected variable. Get a ranking.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-19
SLIDE 19

For Starters Motivation Empirical model Data Results Conclusion

Empirical model

◮ Input oriented frontier model. ◮ We use a three-stage DEA approach:

Step 1: run a simple DEA. Step 2: regress the slacks for each outputs and inputs on the contextual variables using a SFA. Step 2: correct the inputs and the outputs by putting them in the same operating environment and state of nature. Step 3: re-run a DEA with the corrected variable. Get a ranking.

◮ We run two alternative models: one with debt as an input,

the other with debt as a control.

5/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-20
SLIDE 20

For Starters Motivation Empirical model Data Results Conclusion

Data

◮ Collection of the 2009’s annual reports on price and service

quality (publicly available in the city hall) for the 325 biggest French water services (each one has a city of at least 20,000 inhabitants).

◮ Problem: some reports are difficult to find so we got the

delegatee annual report (strictly confidential).

6/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-21
SLIDE 21

For Starters Motivation Empirical model Data Results Conclusion

Data

◮ Collection of the 2009’s annual reports on price and service

quality (publicly available in the city hall) for the 325 biggest French water services (each one has a city of at least 20,000 inhabitants).

◮ Problem: some reports are difficult to find so we got the

delegatee annual report (strictly confidential).

◮ Collection just ended with 285 water services. ◮ We should get a sample of 140 water utilities. ◮ Our sample in the present paper is made of 72 water services.

6/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-22
SLIDE 22

For Starters Motivation Empirical model Data Results Conclusion

Data

◮ The inputs to minimize are the water service’s revenues and

(debts) for 2009.

7/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-23
SLIDE 23

For Starters Motivation Empirical model Data Results Conclusion

Data

◮ The inputs to minimize are the water service’s revenues and

(debts) for 2009.

◮ Outputs are billed water, # of customers and pipes’

length.

◮ Quality outputs: quality of water and network quality.

7/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-24
SLIDE 24

For Starters Motivation Empirical model Data Results Conclusion

Data

◮ The inputs to minimize are the water service’s revenues and

(debts) for 2009.

◮ Outputs are billed water, # of customers and pipes’

length.

◮ Quality outputs: quality of water and network quality. ◮ Some contextual variables: source of water (underground vs.

ground), touristic area or not, population density, pooled water provision, bundled with the sewage service, managementdummy.

7/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-25
SLIDE 25

For Starters Motivation Empirical model Data Results Conclusion

Data

Private Management Public Management Variable Mean St.D Mean St.D Dependant variables Revenues (in thousands) 12,958.90 14,734.31 7,808.51 18,948.91 Debt (in thousands) 4,683.29 7,807.93 6,756.84 7,185.28 Physical outputs Volume billed (in thousands) 8,954.51 9,333.08 5,913.10 14,044.494 Length of mains 717.15 739.68 531.96 791.29 Nb of customers 40,442.82 50,985.12 29,472.32 57,168.54 Quality indicators Drinking water quality 99.74 0.099 99.75 0.154 1/Linear Leakage Index 0.18 0.320 0.16 0.528 Environmental variables Population density 198.779 98.078 187.400 94.114 Touristic Area (1 if yes) 0.853 0.359 0.921 0.273 Water Source (1 if underground) 0.706 0.462 0.684 0.471 Pool of authorities (1 if yes) 0.412 0.500 0.605 0.495 Activity (1 if bundled) 0.559 0.504 0.447 0.504

Table: Descriptive Statistics: Private vs. Public management

8/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-26
SLIDE 26

For Starters Motivation Empirical model Data Results Conclusion

Results I

Mean Standard dev. Min Max Nb of efficient services 0.800 0.204 0.413 1 29 (40.27%)

Table: Efficiency summary - Test 1

9/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-27
SLIDE 27

For Starters Motivation Empirical model Data Results Conclusion

Results I

Mean Standard dev. Min Max Nb of efficient services 0.800 0.204 0.413 1 29 (40.27%)

Table: Efficiency summary - Test 1

Private Management Public Management Score Score Mean 0.759 0.846 Min 0.434 0.388 Max 1 1 Best Rank 2 1 Nb of efficient services 12 (31.57%) 17 (50.00%) Nb of services with score≥ 0.90 1 (2.63%) 4 (11.76%) Nb of services 38 34

Table: Private vs. Public management - Test 1

9/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-28
SLIDE 28

For Starters Motivation Empirical model Data Results Conclusion

Results II

Mean Standard dev. Min Max Nb of efficient services 0.772 0.206 0.402 1 21 (29.17%)

Table: Efficiency summary - Test 2

10/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-29
SLIDE 29

For Starters Motivation Empirical model Data Results Conclusion

Results II

Mean Standard dev. Min Max Nb of efficient services 0.772 0.206 0.402 1 21 (29.17%)

Table: Efficiency summary - Test 2

Private Management Public Management Score Score Mean 0.731 0.819 Min 0.421 0.381 Max 1 1 Best Rank 2 1 Nb of efficient services 8 (21.05%) 13 (38.23%) Nb of services with score≥ 0.90 4 (10.02%) 6 (17.64%) Nb of services 38 34

Table: Private vs. Public management - Test 2

10/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-30
SLIDE 30

For Starters Motivation Empirical model Data Results Conclusion

Conclusion

◮ On average, for the same level of quality than in directly

managed services, private firms are 10% less efficient.

11/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-31
SLIDE 31

For Starters Motivation Empirical model Data Results Conclusion

Conclusion

◮ On average, for the same level of quality than in directly

managed services, private firms are 10% less efficient.

◮ Need to fill in gaps in other coded data: quality of customer

relationship (complaints for 1,000 customers), exports, industrial consumption.

11/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-32
SLIDE 32

For Starters Motivation Empirical model Data Results Conclusion

Conclusion

◮ On average, for the same level of quality than in directly

managed services, private firms are 10% less efficient.

◮ Need to fill in gaps in other coded data: quality of customer

relationship (complaints for 1,000 customers), exports, industrial consumption.

◮ Run alternative regressions using alone-SFA models.

11/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-33
SLIDE 33

For Starters Motivation Empirical model Data Results Conclusion

Looking for advices

◮ Software: DEA-P, GAMS, EMS, LimDep?

12/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-34
SLIDE 34

For Starters Motivation Empirical model Data Results Conclusion

Looking for advices

◮ Software: DEA-P, GAMS, EMS, LimDep? ◮ Heterogeneity Controls: suggestions to improve it?

12/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-35
SLIDE 35

For Starters Motivation Empirical model Data Results Conclusion

Looking for advices

◮ Software: DEA-P, GAMS, EMS, LimDep? ◮ Heterogeneity Controls: suggestions to improve it? ◮ Outliers detection: methods?

12/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-36
SLIDE 36

For Starters Motivation Empirical model Data Results Conclusion

Looking for advices

◮ Software: DEA-P, GAMS, EMS, LimDep? ◮ Heterogeneity Controls: suggestions to improve it? ◮ Outliers detection: methods? ◮ Our variables: debt, revenues?

12/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School

slide-37
SLIDE 37

For Starters Motivation Empirical model Data Results Conclusion

Efficiency in the Public and Private French Water Industry: Prospects for Benchmarking

Aude Le Lannier and Simon Porcher

IAE Sorbonne Business School

October, 2011

13/12 Aude Le Lannier and Simon Porcher IAE Sorbonne Business School