Should We Bet On Private or Public Water Utilities In Cambodia? Evidence on Incentives and Performance from Seven Provincial Towns


Should We Bet On Private or Public Water Utilities In Cambodia?  Evidence on Incentives and Performance from Seven Provincial Towns

Mike Garn, Jonathan Isham, and Satu Kähkönen

2002

Descarga

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“Should We Bet On Private or Public Water Utilities In Cambodia?

Evidence on Incentives and Performance from Seven Provincial Towns”

by

Mike Garn, Jonathan Isham, and Satu Kähkönen

June 2002

MIDDLEBURY COLLEGE ECONOMICS DISCUSSION PAPER NO. 02-19

DEPARTMENT OF ECONOMICS

MIDDLEBURY COLLEGE

MIDDLEBURY, VERMONT 05753

http://www.middlebury.edu/~econ

Should We Bet On Private or Public Water Utilities In Cambodia?

Evidence on Incentives and Performance from Seven Provincial Towns

Mike Garn

World Bank

Jonathan Isham

Middlebury College

Satu Kähkönen

World Bank

June 2000

Abstract: Is public or private sector provision of water more likely to succeed in urban

areas of Cambodia? Using quantitative and qualitative data from a range of surveys and

technical assessments, this paper compares consumer satisfaction and technical

performance of four private and four public utilities in Cambodia. The results indicate

that households served by private utilities are significantly more satisfied with the piped

water than customers of public utilities: the daily availability and quality of piped water is

better, and service interruptions are less frequent. This has not happened by accident.

Private utilities hire more educated staff whom they pay higher salaries; maintain their

facilities on a more regular basis; and implement quality control programs more

diligently. Private sector operators seem to face stronger incentives than public utilities

to keep their customers satisfied. However, this improved service does not come for free

and, consequently, does not yet reach all the available households. Households served by

private utilities pay significantly more for piped water services, and some lower-income

households that are not served by private utilities are partially limited by the high

connection fees (as opposed to the regular monthly payments). Overall, while this recent

effort to introduce private sector involvement in the water sector in Cambodia is

encouraging, the full gains have not yet been realized. The commercial incentive for

improved performance will likely be stronger if the privatization option used is a lease or

concession arrangement; if there is more competition in the water market; and if the

regulatory structure in Cambodia encourages commercial incentives to be more demandresponsive

and cost conscious. Under these conditions, the private sector is a good bet.

Keywords: water supply, privatization, urban infrastructure, Cambodia.

JEL Codes: 017, Q31, R51.

1

1. Introduction

 

1

Cambodia is actively seeking to develop water supply organizations that function

well and, more broadly, a well-functioning water sector, as are other developing

countries. It is possible to have well- functioning urban water companies managed and

operated by either the public sector or the private sector. However, a key question to be

answered is which choice offers Cambodia the best chance for significant improvement

over the current arrangements.

As part of this process, Cambodia is experimenting with the introduction of the

private sector in the management and operation of water supply organizations. In the

past three years, water utilities in three provincial towns initiated the use of private sector

operators. In the other 20 provincial towns in Cambodia, public utilities continue to be

primarily responsible for the provision of water. In one of these, Kandal, a private

company owns and operates part of the system.

Because this is so new in Cambodia, it is not known how well the private water

utilities are performing and how well they are doing in comparison with the public

utilities. Up to now no systematic performance assessments have been carried out, so the

quantity and quality of the services they provide are not known. Nor is it known whether

consumers are satisfied with the quality and quantity of services provided, which is a

fundamental test of performance. Therefore, it is important to evaluate whether

privatization of public utilities in these four towns has improved the delivery of water

(increased the quantity and improved the quality of water consumed) in order to provide

clues for future choices.

The objective of this paper is to assess and compare the performance and

consumer satisfaction with services provided by the newer private companies and the

more traditional public utilities. Specifically, the paper: (1) analyzes the level and quality

of water services provided by private water companies and public utilities, and consumer

reactions to these services; and (2) assesses whether there has been any significant

change in the level and quality of water services provided as a result of private sector

involvement.

1

 

 

This paper was prepared for the Public-Private Infrastructure Advisory Facility in the World Bank. We

would like to thank Vijay Jagannathan for the opportunity to carry out the study. Also, thanks are due to

the local team in Cambodia (M.S. Shivakumar, M. Kumbakumba, Chea Sarin, Han Phoumin, Ea Sophy,

Nun Vanny, Meak Chhavannarey, and Venkatesh Sundararaman) that implemented the household and

water utility surveys and carried out the technical assessments of water systems. We are also grateful to Ian

McGuire of Middlebury College for his excellent research assistance.

 

2

2. Public vs. Private Provision of Urban Water: The General

Case

Private sector participation in water delivery is a relatively recent phenomenon in

most developing countries. Up until 1990, almost all developing countries relied on

government provision of water supply services (Silva, Tynan, Yilmaz 1999). However,

in the past ten years, the disappointing performance of many (not all) public sector

companies–as well as fiscal challenges–has encouraged many governments to look for

alternative, more efficient ways of providing water services. Increasingly, they have

turned to private sector solutions.

Although not open and shut, there is a strong case for encouraging more private

sector involvement in the management and operation of urban water supply

organizations. This case has three elements: (1) the dismal results of past attempts to

substantially improve the performance of water supply and other organizations while

retaining a preference for public sector management and operations (see, for example,

World Bank 1995); (2) the growing evidence of the beneficial effect of private sector

involvement where it has been tried; and (3) a persuasive explanation of these differing

outcomes in terms of the relative incentive structures likely to face private and public

sector organizations.

While it is not guaranteed, private sector participants generally face stronger

incentives to be responsive to the demand for their services from users than do public

sector participants. This is not because private sector participants are the “good” guys

and public sector participants the “bad” guys. It is because, when the private sector is

required to be responsible for normal commercial risks (as they should be), they depend

much more than public sector participants on providing services that people want and for

which they are willing to pay; and to control costs to increase net revenue. Public sector

operators, who are generally less reliant on revenues from their customers to sustain their

investments and operations and more reliant on government investments and operational

budgets or grants, face weaker incentives to be responsive to demand. Put another way,

public sector participants have to develop different strategies for resource allocation (how

they utilize the money available to them) and for resource mobilization (how they raise

the money available), while for the private sector operator, the two strategies are more

nearly identical because of the ‘revenue from users’ requirement to sustain their

investments. For the private sector operator, both strategies necessarily focus on users

and their demand.

This outcome is contingent on the degree to which the private sector participant

continues to be willing to assume the responsibility for future commercial risks. In turn,

his willingness to do this is dependent on credible assurances that the government will

not, in the future, make arbitrary decisions about pricing and additional assignment of

responsibilities beyond those initially agreed between the operator and the government.

The different options for private sector participation (service contracts, management

contracts, BOTs, lease contracts, concessions, and outright sale of assets) differ

considerably in the degree to which the private sector participant can be expected to take

3

and be responsible for commercial risks, and the degree to which credible government

assurances require an explicit regulatory structure to be in place. The relevant positive

incentive effects are likely to be strongest in the case of the last three options.

Although there are other incentive issues that could be discussed (see the next

section), the most important requirement from the perspectives of this paper is that the

private sector operator receives sufficient credible assurances from the government that

the operator is willing to accept the commercial risks that are implicit in his future

revenue projections. For then, the operator will have strong incentives to maintain a

future revenue stream by being responsive to current and potential customers and, also, to

control costs while doing so.

3. Private and Public Water Utilities in Cambodia: Specific Results

In 1997 and 1998, water utilities in four out of 23 provincial towns in

Cambodia—Banteay Meanchey, Kampong Speau, Takeo, and Kandal—began to use

private sector operators.

 

 

2

In three of these towns—Banteay Meanchey, Kampong Speau,

and Takeo—private sector companies are currently the sole network providers of water in

the core area of the provincial town. By contrast, in Kandal, the private sector company

operates the network outside the core area of the town, in Kien Svay. The core area of

town is still taken care of by a public utility.

 

 

3

The form of privatization varied across towns. In Banteay Meanchey, Kampong

Speau, and Takeo, the Ministry of Industry, Mines, and Energy (MIME) granted private

companies a three-year renewable license to supply water to residential consumers in the

area. The renewal of these licenses depends on company’s compliance with water quality

and tariff stipulations. In addition, each of these private companies entered into a

contract to transfer the assets of the water utility to the company for 23-40 years, after

which all assets are to revert to the public sector. In Kien Svay, by contrast, no public

assets were transferred to the private company. Instead, the private company entered into

a built-own-operate (BOO) contract with MIME.

In none of the four provincial towns, which now use private operators, were the

conditions for privatization ideal to maximize positive institutional incentives. The

privatization process was ad hoc and non-transparent (De Raet and Subbarao 1999).

Private companies were not solicited through open and competitive bidding. In each

town the process was triggered by unsolicited bids. In Banteay Meanchey, Kampong

Speau and Takeo, individuals who submitted these bids were granted the license without

competition. In Kandal, one of the two bidders was selected, but it is not known what

criteria were used to select the winner.

2

 

 

Takeo is 75 kilometers south, Kampong Speau is about the same distance east, and Kandal to northwest of

Phnom Penh. Banteay Meanchey is located in northern Cambodia.

 

3

 

 

In each of these provincial towns, there also is a large number of private water vendors that serve

households not connected to the network.

 

4

Further, there is currently no overall regulatory framework governing the

operations of these private companies and existing regulation is deficient (De Raet and

Subbarao 1999). For example, it is not clear how water tariffs will be revised and

contractual disputes settled. The licenses and contracts for the companies are ambiguous,

therefore, on some important issues that have an impact on incentives. The government

is aware of this and is attempting to rectify the situation. Hopefully, they will do so in a

way that takes into account the incentive effects of future government and regulatory

actions on private operators to enhance the positive results. Nevertheless, in each of

these provincial towns and with considerable residual uncertainty, private companies

have been willing to make investments–sometimes substantial–to rehabilitate and

expand water supply networks in their respective market areas.

The rest of the provincial towns in Cambodia are currently served by public water

utilities. Most of these utilities were shut down during the Khmer Rouge regime and

were reopened in the 1980s with seriously depleted facilities. Some of them have since

been rehabilitated with support from different donor agencies, rather than private

investment.

To compare the current performance of private and public water utilities, in

addition to the four towns served by private companies (Banteay Meanchey, Kampong

Speau, Takeo, Kandal), four provincial towns served by public utilities were selected for

the study. These towns were selected randomly to avoid selection bias, and they are

Kandal, Battambang, Kampong Chhang, and Svay Rieng.

 

 

4

Kandal appears in both lists

because both private and public utilities serve the market area. The private portion is

called Kien Svay in this paper. Table 1 summarizes the staffing, production capacity,

coverage, and fees of the eight private and public water utilities included in the study.

 

4

 

 

Battambang is the second largest city of Cambodia, and located close to Thai border in the northwest

corner of the country. Kampong Chhang is a port and fishing town on the Tonle Sap river in the middle of

Cambodia. Finally, Svay Rieng is located in Southern Cambodia close to the Vietnam border.

 

5

Table 1: Selected Production and Financial Characteristics of Eight Water Utilities in Cambodia

PUBLIC UTILITIES PRIVATE UTILITIES

Battambang Kampong

Chhang

Kandal Svay Rieng Banteay

Meanchey

Kampong

Speau

Kien Svay Takeo

Population of town 139,964 41,703 58,264 21,205 98,848 41,478 58,264 39,186

Year established in current form 1993 1996 1979 1980 1998 1997 1998 1997

Staffing:

Number of permanent staff 65 8 16 16 20 6 10 15

Number of temporary staff 0 0 0 0 6 8 0 N/A

Monthly salary of staff (Riels) 130,000 45,000 50,000 47,000 285,000 494,000 326,800 N/A

Monthly salary of operation and

maintenance worker (Riels)

120,000 45,000 50,000 47,000 200,000 304,000 304,000 N/A

Production Capacity:

Current production capacity

(m

 

 

3

/day)

3750 960 780 400 3000 1500 1632 1300

Current production m

 

 

3

/day 2750* 200* 780 320 1200 560* 176 120

Capacity utilized (%) 73.33 20.83 100 80 40 37.33 10.78 9.23

 

Coverage:

Total number of direct

connections from utility:

1766 409 580 393 1500 1700 230 450

·

 

 

Residential 1618 406 561 375 1423 1510 229 N/A

·

 

 

Business 78 N/A 5 N/A 50 180 N/A N/A

·

 

 

Government 70 2 14 18 25 10 1 13

Percentage of households covered 6.33 5.28 5.47 9.13 7.74 19.93 2.24 6.21

 

Subcontractors:

Number of sub-contractors to

utility

4 0 3 0 0 0 0 0

Number of connections served by

sub-contractors

2046 0 239 0 0 0 0 0

Connection fees and tariffs:

Connection fee (Riels) 200,000 190,000-

342,00

136,500-

390,000

5,000-35,000

+ materials

350,000 76,000 190,000 228,000

Water tariff (Riels/m

 

 

3

) 1400 1000 550 600 1300 1500 1400 N/A

Notes

 

 

: * based on production meter reads. All salaries and fees in Cambodian Riels (3763 Riels/1 US$)

Some information on the utility in Takeo is missing since the manager of the utility declined to respond to a formal survey.

 

6

A. Staffing

Private utilities have a slightly smaller permanent staff than public utilities

surveyed. While the size of permanent staff in private utilities varies from 6 to 20

(average 13), public utilities have 8 to 65 (average 26) permanent staff members on the

payroll. To supplement the permanent staff, private utilities–unlike public utilities–do

hire temporary workers. These workers typically assist with the maintenance and tariff

collection in different localities.

 

 

5

All utilities reported to have some staff on the premises

all times.

Utility managers’ level of education also differs systematically across private and

public utilities surveyed: managers of all private utilities have more formal schooling

than do the managers of public utilities. Managers of private utilities have either high

school or university education, whereas managers of public utilities have a secondary

school background.

Private utilities pay their staff–including operations and maintenance (O&M)

workers–much more than the public utilities surveyed. The average salaries for staff and

O&M workers in private utilities are 368,600 and 269,333 Riels, respectively–in contrast

to 68,000 and 65,500 Riels in public utilities.

 

 

6

B. Production Capacity

Most private water utilities reported lower capacity utilization rates than public

utilities. The young age of private utilities and relatively dilapidated condition of public

utilities may partially explain this result. All utilities, both private and public, reported

to have increased their production capacity in the past two years.

C. Coverage

However, neither public nor private utilities have so far reached many people in

their market area. In each town, the percentage of households served by a water utility is

low: the coverage of private water utilities varies from 2 to 20 percent of all households,

whereas the coverage of public utilities ranges from 5 to 9 percent. As expected, the

majority of connections served by private and public utilities are residential. Table 1

summarizes the number of residential, business, and government connections served by

each utility.

In addition to direct connections, public utilities in Battambang and Kandal have

sub-contractors

 

 

that are authorized to resell water to other households. Specifically,

5

 

 

For example, in Kampong Speau, the utility has eight “block managers” who monitor the network and

collect tariffs in their locality. As compensation, they are paid a percentage of the tariff collected. The

overall network maintenance is the responsibility of permanent utility staff.

 

6

 

 

As shown in the table, three of the four public utilities pay their staff no more than 50,000 Riels per month

— the equivalent of $13 per month.

 

7

these sub-contractors have been given permission to extend the piped network beyond

their individual connections to other households in the area and charge a fixed tariff for

the water sold. In Battambang, the utility currently has three subcontractors and the

utility in Kandal has three. The number of connections served by subcontractors has

increased rapidly in Battambang. Currently, subcontractors in that town serve a larger

number of residential connections (2046) than the utility itself does (1618). Interestingly,

all subcontractors in Kandal and at least some in Battambang are employees of the water

utility.

 

 

7

D. Connection Fees and Tariffs

All private utilities, unlike public utilities surveyed, charge a fixed connection fee

from their customers. These fees are listed in Table 1 and range from 76,000 Riels to

350,000 Riels.

 

 

8

They cover all labor charges, cost of piping materials, the water meter,

and other connection expenses. All private utilities, except the utility in Kampong Speau,

require their customers to pay the fee in lump sum.

The method of setting the connection fee varies across public utilities surveyed.

The connection fee for utilities in Kampong Chhang and Kandal depends on the distance

of the customer from the network. Accordingly, their fees range from 136,500 to

390,000 Riels, include all connection costs, and are to be paid in lump sum. In

Battambang, the utility charges a fixed, all-inclusive fee of 200,000 Riels for new

connections. Unlike other public utilities, it provides its customers an option to pay the

fee in installments. Finally, in Svay Rieng, unlike in any other utility surveyed,

customers who want a connection need to purchase the water meter, piping and other

materials, and obtain all the needed permits and clearances themselves. They pay to the

utility only for the labor charges related to connection, which vary between 5,000 and

35,000 Riels depending on the distance from the network.

Private utilities charge higher tariffs for water consumed than do public utilities

surveyed. Both private and public utilities charge their customers a uniform tariff per a

cubic meter of water consumed. This tariff is the same for residential, business, and

government connections. In private utilities, the tariff varies from 1300 to 1500 Riels/m

 

 

3

.

In public utilities surveyed, it ranges between 550 and 1400 Riels/m

 

 

3

. All utilities, except

the utility in Svay Rieng, have metered all their connections and bill customers on a

monthly basis based on meter readings. Currently, no utility charges its customers a

minimum monthly fee.

Unlike the utility in Kandal, the utility in Battambang charges its subcontractors a

different—lower—tariff. While its other customers with direct connections are charged

1400 Riels/m

 

 

3 of water, subcontractors pay for water 1375 Riels/m3

. These

subcontractors in turn are allowed to charge their customers up to 2000 Riels/m

 

 

3

of water

7

 

 

It is not clear what criteria were used to select subcontractors in Battambang and whether all

subcontractors are employees of the utility.

 

8

 

 

The exchange rate at the time of data collection was 1 USD=3900 Riels.

8

delivered. By contrast, in Kandal, subcontractors are permitted to charge 550-1200

Riels/m

 

 

3

for the water.

To assess the performance of these private and public utilities as well as

household satisfaction with the water services they provide, a range of primary data was

collected from all eight provincial towns. First, in each town a sample of households was

surveyed and asked to assess the quality of water services. Second, the managers of eight

utilities were interviewed and technical performance of each utility independently

assessed by a water engineer. The next sections describe the results of these exercises.

 

4. Household Assessment of Performance of Water Utilities

In each town a sample of households served by the water utility was surveyed to

gauge the level of user satisfaction with services and to get users’ perception of other

aspects of service delivery. In each town, 50 randomly selected households that were

served by either

 

 

public

(Battambang, Kandal, Kampong Chhang and Svay Rieng) or

private

 

 

(Banteay Meanchey, Kien Svay, Kampong Speau, Takeo) utilities were surveyed.

In addition, for the two public utilities that have subcontractors (Battambang and

Kandal), 25 and 26 randomly selected households served by these

 

 

sub-contractors

were

surveyed, respectively.

The results of these household assessments of satisfaction, water quality,

water availability, reliability of service, frequency of service breakdowns and

maintenance, cost of service, and service-orientation of utilities were somewhat mixed

across the different categories and the different utilities. However, overall the user

responses were favorable to the private utilities on most categories. The next sections

review the major findings and provide disaggregated information on the results by

individual towns.

 

A. Customer Satisfaction

Households served by private utilities are more satisfied with the quality of piped

water than households served by either public utilities or subcontractors. Table 2

summarizes the results on household satisfaction for each category of utilities (public,

subcontractor, private). It shows that 82 percent of households served by private utilities

reported satisfaction with the qua lity of piped water, while only 56 and 45 percent of

households served by public utilities or subcontractors, respectively, are satisfied with the

service.

9

Table 2: Customer Satisfaction: Satisfaction with Water Services

Battambang Kandal Kompong

Chhnang

Svay

Rieng

0.56 0.44 0.32 0.84 0.64

(0.50) (0.47) (0.37) (0.48)

0.93 0.90 0.98 0.86 0.96

(0.30) (0.14) (0.35) (0.20)

Average

Contract

Battambang Kandal

0.45 0.46 0.44

(0.51) (0.51)

0.92 0.85 1.00

(0.37) (0.00)

Average

Private

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

0.82 0.82 1.00 0.70 0.76

(0.38) (0.00) (0.46) (0.43)

0.93 0.96 1.00 0.82 0.92

(0.20) (0.00) (0.39) (0.27)

Notes: Means and (standard deviations) for water service variables.

Satisfied with quality of

water

Satisfied with water

service

Private

Sub-contractor

Average

Public

Public

Satisfied with quality of

water

Satisfied with water

service

Satisfied with quality of

water

Satisfied with water

service

Assessing household satisfaction across different categories of water utilities

while controlling for town- level characteristics confirms that customer satisfaction with

the quality of water is increasing with private sector involvement. The town of Kandal,

that has a public and private water utility (Kien Svay) and subcontractors, provides an

opportunity to compare these three types of delivery modes with varying degree of

private sector involvement, while controlling for town- level characteristics. As Table 2

reveals, in Kandal, household satisfaction with the quality of water is steadily increasing

as private sector involvement intensifies: 32, 44, and 100 percent of households served by

public utilities, subcontractors, and private utilities, respectively, are satisfied with the

quality of water.

Satisfaction with water quality, however, va ries within each category of utilities.

Among private utilities, the share of households satisfied with water quality ranges from

70 to 100 percent. The utility in Kien Svay reaps the highest household satisfaction

scores: all households served by the ut ility reported satisfaction with water quality.

Among public utilities, the share of households satisfied with the quality of water varies

between 32 and 84 percent. The utility in Kampong Chhang stands out in that category.

It has the second highest le vel of household satisfaction with water quality across all

utilities surveyed: over 80 percent of households served are satisfied with water quality.

Surprisingly, household satisfaction with the piped water service

 

 

overall

does not

vary across public and private water utilities. As can be seen from Table 2, over 90

percent of households served by private utilities, public utilities, or subcontractors

10

reported to be satisfied with the service they get. According to the Cambodian survey

team, Cambodians’ reluctance to express dissatisfaction may partly explain this result.

Also, it may reflect the lack of awareness of alternatives.

What is behind these differences in consumer satisfaction? Household

assessments of water quality, availability, frequency of breakdowns and maintenance,

cost of service, and service-orientation of different utilities can help to answer this

question.

 

B. Water Quality

Household assessment of piped water quality mirrors the results on household

satisfaction. Each household surveyed was asked to evaluate various attributes of piped

water, such as its clarity and overall quality. The main results of this assessment are

reported in Table 3.

11

Table 3: Performance of Water Systems: Water Quality

Description

Average

public

household

Battambang Kandal Kompong

Chhnang

Svay

Rieng

0.27 0.22 0.14 0.36 0.36

(0.42) (0.35) (0.48) (0.48)

0.27 0.28 0.12 0.30 0.38

(0.45) (0.33) (0.46) (0.49)

0.65 0.44 0.62 0.80 0.72

(0.50) (0.49) (0.40) (0.45)

0.22 0.26 0.46 0.06 0.08

(0.44) (0.50) (0.24) (0.27)

Average

subcontract

household

Battambang Kandal

0.12 0.12 0.32

(0.33) (0.48)

0.15 0.15 0.04

(0.37) (0.20)

0.38 0.38 0.88

(0.50) (0.33)

0.27 0.27 0.16

(0.45) (0.37)

Average

private

household

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

0.49 0.50 0.80 0.28 0.38

(0.50) (0.40) (0.45) (0.49)

0.07 0.08 0.00 0.20 0.00

(0.27) (0.00) (0.40) (0.00)

0.84 0.88 1.00 0.54 0.92

(0.32) (0.00) (0.50) (0.27)

0.13 0.26 0.02 0.00 0.22

(0.44) (0.14) (0.00) (0.42)

Notes: Means and (standard deviations) for project design variables.

See text for descriptions of variables.

Don’t drink the

water

Clear water

Private

Don’t drink the

water

Very good or good

quality

Very bad or bad

quality

Very good or good

quality

Very bad or bad

quality

Clear water

Don’t drink the

water

Very good or good

quality

Very bad or bad

quality

Sub-contract

Public

Clear water

A larger share of households served by private than by public utilities judged the

quality of piped water to be good. As Table 3 indicates, 49 percent of households served

by private utilities, and only 27 percent of households served public utilities, said that the

water quality is either good or very good. Further, 27 percent of households with

connections from public utilities and only seven percent of households served by private

utilities reported the piped water to be of bad or very bad quality. 84 percent of

12

households served by private and 65 percent of households served by public utilities

considered the piped water clear.

The case of Kandal, which can be used to compare the performance of private and

public utilities and subcontractors while controlling for town- level characteristics,

provides further support to the result that private sector participation tends to lead to the

delivery of better quality water. While only 14 percent of households served by the

public utility in Kandal assessed the quality of piped water to be good or very good; 32

percent of households served by sub-contractors and 80 percent of households served by

private utilities—the highest share of customers across all utilities–rated the quality of

piped water high.

The quality of piped water was not, however, judged to be equally high across all

private utilities. Household evaluation of the quality of piped water was significantly

worse in Kampong Speau than in other towns served by private utilities. Only 28 percent

of households served by the utility considered the quality of water to be good, while 20

percent of households stated that the quality of piped water is bad or very bad. This

result is consistent with low household satisfaction, discussed in the previous section.

13

C. Water Availability

Differences in the availability of piped water to households also help to explain

the differences in household satisfaction with water services. Piped water availability is

measured here in two ways: whether it is available to households every day; and by the

number of hours per day it is typically available.

As Table 4 indicates, the piped water availability is higher in households served

by private utilities than in households served by public utilities. 76 percent of households

served by private utilities, but only 57 percent of households served by public utilities,

have piped water available every day. While households served by private utilities have

water available from the system on average for 21 hours per day, households served by

public utilities get piped water on average only for about six hours per day.

Comparison of households served by public utilities with those served by their

sub-contractors reveals that even sub-contracting modestly improves water delivery by

increasing the hours piped water is available. On ave rage, households served by subcontractors

have piped water available for two hours more per day than households

served by public utilities.

Assessing piped water availability to households while controlling for town-level

characteristics confirms these results: private sector participation improves water

availability. In Kandal, households served by public utilities have water available for 5.5

hours per day; those served by sub-contractors about 8 hours per day; and those served by

the private utility in Kien Svay 24 hours per day. The percentage of households reporting

water to be available every day is following a similar pattern.

Private sector participation, however, by no means ensures water availability 24

hours per day. As Table 4 indicates, the performance of private utilities in this respect

varies. Specifically, the performance of the utility in Kampong Speau again significantly

differs from the performance of other private utilities. While all other private utilities

provide a 24-hour service to households, the utility in Kampong Speau delivers piped

water only 13 hours per day.

14

Description Average

public

household

Battambang Kandal Kompong

Chhnang

Svay

Rieng

6.1 8.2 6.1 5.5 4.5

(1.08) (3.56) (2.22) (5.87)

0.57 0.82 0.66 0.16 0.62

(0.39) (0.48) (0.37) (0.49)

0.22 0.10 0.26 0.12 0.38

(0.30) (0.44) (0.33) (0.49)

0.41 0.26 0.26 0.48 0.62

(0.44) (0.44) (0.50) (0.49)

Description Average

subcontract

household

Battambang Kandal

8.3 8.3 11.2

(0.96) (3.68)

0.75 0.81 0.68

(0.40) (0.48)

0.08 0.08 0.16

(0.27) (0.37)

0.31 0.31 0.40

(0.47) (0.50)

Description Average

private

household

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

21.3 24.0 24.0 13.4 23.9

(0.00) (0.00) (4.51) (0.59)

0.76 0.58 0.90 0.72 0.84

(0.49) (0.30) (0.45) (0.37)

0.05 0.02 0.00 0.14 0.04

(0.14) (0.00) (0.35) (0.20)

0.38 0.34 0.00 0.50 0.66

(0.47) (0.00) (0.51) (0.48)

Notes: Means and (standard deviations) for project design variables.

See text for descriptions of variables.

Table 4: Performance of Water Systems: Water Availability and Use

Public

Hours of water availability in

wet season

Piped water available every day

Water availability has increased

over the last two years

Household consumes more

water now than two years ago

Sub-contract

Hours of water availability in

wet season

Piped water available every day

Water availability has increased

over the last two years.

Household consumes more

water now than two years ago.

Private

Hours of water availability in

wet season

Piped water available every day

Water availability has increased

over the last two years.

Household consumes more

water now than two years ago.

Though public utilities’ daily hours of operation are shorter than private utilities’

hours of operation, public utilities have improved their water delivery over the past two

years. On average, 22 percent of households served by public utilities stated that the

daily availability of piped water has improved over the past two years. This is not

15

surprising, given that public utilities were partly destroyed during the Khmer Rouge

regime, and, therefore, in a bad condition when they were re-opened.

D. Service Breakdowns and Maintenance

The results of the household survey reveal that private water utilities provide a

slightly more reliable service than public utilities. Reliable service refers here to water

being available every day and when the customer expects it. 81 percent of households

served by private and 78 percent served by public utilities judged their water supply as

reliable.

Households served by public utilities experience more service interruptions than

households served by private utilities or subcontractors, as reported in Table 5. While 13

percent of households served by public utilities reported that, in the past three months, the

service had been stopped for a day or more, only six percent of households served by

private utilities or subcontractors acknowledged having similar interruptions of service.

Sixteen percent of households served by private utilities, compared to seven percent of

households served by public utilities, however, said that there are more service

breakdowns now than two years ago. This is not surprising, given that many of the

private utilities have been in operation for only a couple of years.

The results from the water utility surveys also suggest that the public utilities

receive more complaints from customers. Only one (Kien Svay) of the three private

utilities reported that they had received a single complaint in the previous month. All

four of the public utilities received at least one complaint–the average number of

complaints being three.

 

 

9

Further, a slightly larger share of households served by private utilities or

subcontractors than by public utilities consider the piped network to be well maintained.

Ninety, 94, and 85 percent of households served by private utilities, subcontractors, and

public utilities, respectively, said the network is well maintained. While 42 percent of

households with connections from public utilities stated that there are currently leaking

pipes in town, 35 and 37 percent of households served by private and subcontractor,

respectively, said the same.

9

 

 

The number of reported monthly complaints in the four public utilit ies are one (Kandal), five

(Battambang), three (Kampong Chhang) and three (Svay Rieng).

 

16

Table 5: Performance of Public and Private Water Systems:

Service Breakdowns/Failures and Maintenance

Average

public

household

Battambang Kandal Kompong

Chhnang

Svay

Rieng

0.78 0.78 0.96 0.46 0.92

(0.42) (0.20) (0.50) (0.27)

0.13 0.04 0.04 0.34 0.08

(0.20) (0.20) (0.48) (0.27)

0.07 0.06 0.00 0.18 0.02

(0.24) (0.00) (0.39) (0.14)

0.42 0.32 0.40 0.50 0.46

(0.47) (0.49) (0.51) (0.50)

0.85 0.78 0.86 0.84 0.90

(0.42) (0.35) (0.37) (0.30)

Average

contractor

household

Battambang Kandal

0.88 0.88 0.88

(0.33) (0.33)

0.06 0.12 0.00

(0.33) 0.00

0.04 0.08 0.00

(0.27) 0.00

0.37 0.38 0.36

(0.50) (0.49)

0.94 0.92 0.96

(0.27) (0.20)

Average

private

household

Bantey

Meanchey

Kien

Svay

Kompong

Speu

Takeo

0.81 0.90 0.98 0.52 0.84

(0.30) (0.14) (0.50) (0.37)

0.06 0.06 0.00 0.08 0.10

(0.24) 0.00 (0.27) (0.30)

0.16 0.04 0.00 0.46 0.14

(0.20) 0.00 (0.50) (0.35)

0.35 0.66 0.04 0.50 0.18

(0.47) (0.20) (0.51) (0.39)

0.90 0.98 0.98 0.66 0.96

(0.14) (0.14) (0.48) (0.20)

Notes: means and (standard deviations) for project design variables.

See text for descriptions of variables.

Public

Reliable piped

water

Any water

stoppage

More service

breakdowns

Leaking pipes in

town

Network well

maintained

Sub-contractor

Reliable piped

water

Any water

stoppage

More service

breakdowns

Leaking pipes in

town

Network well

maintained.

Leaking pipes in

town

Network well

maintained

Private

Reliable piped

water

Any water

stoppage

More service

breakdowns

17

These results belie a fairly large variation in household assessments within each

category of utilities. Among public utilities, the utility in Kampong Chhang, that is rated

high on the quality of water, has the worst record of service reliability and frequency of

service breakdowns. Thirty-four percent of its customers reported that the service had

been stopped to their house for a day or more in the past three months, and a half of

respondents said that there are leaking pipes in town. Among private utilities, the utility

in Kampong Speau is again an outlier: households served by the utility experience more

often service breakdowns than households served by other private utilities. As a result,

household assessment of the service reliability is significantly lower in Kampong Speau

than in other towns.

E. Cost of Service

The results of the household survey confirm tha t private utilities charge on

average a higher connection fee from customers than public utilities. While households

surveyed paid on average 219,684 Riels to private utilities for a connection, the average

cost of connection from a public utility was 186,926 Riels. This can be seen from Table

6. The connection fees households reported to have paid matched approximately

utilities’ stipulated fees listed in Table 1, except in the case of utilities in Kampong

Chhang and Kampong Speau. In Kampong Chhang, the average connection fee

households reported to have been paid was lower than the fee currently imposed by the

utility (168,319 Riels compared to 190,00-342,000 Riels). By contrast, in Kampong

Speau, the average fee households reported to have paid to the utility was significantly

higher than the fee posted by the utility (113,334 Riels compared to 76,000 Riels).

Interestingly, while almost all households served by other private utilities reported to

have received a receipt for their payment, only 72 percent of households served by the

utility in Kampong Speau had obtained one. According to Cambodians, payments made

without receipts often do not enter into the official records of the utility. Overall, the

results indicate that private utilities have more often issued receipts than public ones.

18

Table 6: Cost of Water Service

Battambang Kandal

Kompong

Chhnang Svay Rieng

186,926 203,042 312,460 168,319 63,884

(146,559) (142,498) (90,622) (73,864)

0.80 0.69 0.75 0.83 0.92

(0.47) (0.44) (0.38) (0.27)

6.9 15.3 2.5 7.6 2.1

(43.5) (2.5) (10.0) (1.6)

13,818 26,806 10,318 10,390 7,759

(26034) (4191) (8795) (5370)

888 1,402 550 1,000 601

(14) (0) (0) (7)

Cubic meters

consumed 15.6 19.1 18.8 10.4 12.9

Average

Contract

Household Battambang Kandal

159,990 159,990 510,360

(44,098) (1,099,514)

0.83 0.83 0.72

(0.39) (0.46)

2.2 2.2 4.1

(1.9) (6.2)

17,308 17,308 11,721

(11,907) (4,280)

1,400 1,823 784

(163) (229)

Cubic meters

consumed 12.4 9.5 15.0

Average

Private

Household

Bantey

Meanchey

Kien Svay

Kompong

Speu Takeo

219,684 343,200 194,200 113,334 228,000

(18,455) (14,581) (60,198) (7,677)

0.92 1.00 1.00 0.72 0.98

(0.00) (0.00) (0.46) (0.14)

2.2 1.4 1.9 3.1 2.2

(1.0) (2.1) (3.9) (2.5)

20,894 18,704 20,553 21,305 23,012

(13,793) (6,776) (42,724) (21,332)

1,489 1,312 1,400 1,503 1,739

(59) (0) (229) ( 323)

Cubic meters

consumed 14.0 14.3 14.7 14.2 13.2

Notes: Means and (standard deviations) for water service variables.

All fees in Cambodian Riels (3763 Riels/1 US$)

Private

Connection fee

(Riels)

Received receipt for

connection

Number of days to get

connection

Amount of last

monthly bill (Riels)

Unit tariff (Riels per

cubic meter)

Sub-contractor

Number of days to get

connection

Amount of last

monthly bill (Riels)

Unit tariff (Riels per

cubic meter)

Connection fee

(Riels)

Received receipt for

connection

Number of days to get

connection

Amount of last

monthly bill (Riels)

Unit tariff (Riels per

cubic meter)

Average

Public

Household

Public

Connection fee

(Riels)

Received receipt for

connection

19

Further, private utilities provide connections faster than public utilities.

According to results, it takes from one to three days to get a connection from a private

utility. By contrast, getting a connectio n from a public utility can take from two to fifteen

days, the average time being about seven days.

The results of the household survey also confirm that the unit tariff of water tends

to be higher in private than in public utilities. The average unit tariff charged by private

utilities is 1,489 Riels, while the average unit tariff charged by public utilities is 888

Riels. The unit tariffs households reported to pay were consistent with tariffs posted by

utilities. The average unit tariff households served by subcontractors pay is 1,400 Riels.

The unit tariffs charged by subcontractors in Battambang and Kandal are within the

authorized range.

F. Service-orientation

Finally, to evaluate service-orientation of utilities, households were asked a series

of questions about the responsiveness of utility staff. Among other things, it was inquired

whether they can contact the utility and get assistance if they have inquiries about the

service or billing; whether they have ever had a problem with billing; and whether the

staff of the utility takes customer complaints and inquiries seriously and attempts to

improve the service. To cross-check these results, all the utilities were also asked a series

of questions about their own perception of the customer service that they deliver.

As results in Table 7 indicate, the household assessments of utility serviceorientation

do not vary in any systematic pattern across different categories of utilities.

Staffs of public and private utilities are viewed to be equally responsive. Almost all

households said that they can contact the utility if they have inquiries about the service or

billing. About a half of households served by public and a half of households served by

private utilities considered the utility staff to take customers’ complaints and inquiries

seriously. Only five percent of households served by public and four percent of

households served by private utilities had had a problem with billing.

This result is surprising. Given the incentives faced by private utilities (discussed

in section 2), one would expect household perception of service-orientation differ across

private and public utilities. In fact, the results from the water utility surveys also suggest

that the public utilities–at least by their own assessments–are as responsive as private

utilities. All reporting public and private utilities alike respond that they have some staff

dedicated to customer services, that they encourage customers to report problems, and

that they responded to all complaints that they received last month.

 

 

10

10

 

 

Although, as noted above, the public utilities tended to receive more complaints every month.

20

Table 7: Service Orientation of Water Utilities

Battambang Kandal Kompong

Chhnang

Svay

Rieng

0.52 0.56 0.44 0.52 0.56

(0.50) (0.50) (0.50) (0.50)

0.05 0.08 0.02 0.00 0.08

(0.27) (0.14) 0.00 (0.27)

0.55 0.66 0.54 0.26 0.72

(0.48) (0.50) (0.44) (0.45)

Average

Contract

Battambang Kandal

0.61 0.62 0.60

(0.50) (0.50)

0.10 0.19 0.00

(0.40) 0.00

0.51 0.54 0.48

(0.51) (0.51)

Average

Private

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

0.47 0.50 0.42 0.42 0.52

(0.50) (0.50) (0.50) (0.50)

0.04 0.08 0.00 0.04 0.02

(0.27) 0.00 (0.20) (0.14)

0.31 0.30 0.10 0.60 0.24

(0.46) (0.30) (1.59) (0.43)

Notes: Means and (standard deviations) for water service variables.

Average

Public

Public

Staff is responsive to

complaints and inquiries

Private

Have had billing problem

Operation has improved in last

two years

Sub-contractor

Have had billing problem

Operation has improved in last

two years

Staff is responsive to

complaints and inquiries

Have had billing problem

Operation has improved in last

two years

Staff is responsive to

complaints and inquiries

More than half of households (55 percent) connected to a piped networked

managed by a public utility held that the operation of the utility had improved in the past

two years. Only one-third of households served by private utilities said the same. The

lower initial level of public utilities may again explain this result.

In summary, the household assessment of private utilities is more favorable than

the assessment of public utilities. In particular, the percentage of households which

reported satisfaction with the quality of water, thought the water to be good or very good,

had water available every day for more hours a day, thought the water was more reliable,

had fewer stoppages, believed the network to be well maintained, and got a connection

quicker were higher with private operators than with public operators. In addition, fewer

of those served by private operators thought they got very bad or bad water and fewer did

not drink the water.

Surprisingly, 93 out of 100 households said they were satisfied with the services

regardless of which type of system provided them—public or private. This could be a

general reluctance to express dissatisfaction or lack of awareness of alternatives. Two

21

categories suggest a need for caution. First, users judged public utilities to be slightly

more responsive to complaints and inquiries than private utilities. This may be partly due

to the absence of competition in these early experiments with private sector operators and

the incomplete set of institutional incentives. Second, private operators charged more for

connection and monthly bills tended to be higher than those of their public counterparts.

While this could reflect accurately what is needed to sustain and nourish investment, it

remains a concern in Cambodia given very low rates of coverage of the population given

the finding that the vast majority of those unconnected expressed an interest in getting a

piped connection.

Are these results consistent with the technical assessment of utilities? The next

section will explore this question.

5. Technical Assessment of Performance of Water Utilities

To get an independent assessment of the technical performance of the eight

utilities, a water engineer evaluated the performance and maintenance of facilities in each

town. Performance indicators used included the condition and daily operation of

facilities; frequency of major breakdowns; frequency of maintenance; and

implementation of water quality control programs.

A. Facility Performance

The results of the technical assessment confirmed that the daily hours of operation

are significantly higher among private than public utilities. Table 8, which summarizes

the condition and daily operation of facilities, shows that all private utilities, except the

utility in Kampong Speau, distribute piped water 24 hours a day. This is consistent with

household reports discussed in the previous section. The distribution system of public

utilities, by contrast, operates between 8 and 12 hours a day, according to the technical

assessment. These hours of operation are, however, inconsistent with household reports,

which indicate that households served by public utilities have water available only 4-8

hours per day.

Further, the results indicate that the frequency of major breakdowns is less

common in private than public utilities. Among private utilities, the utility in Kampong

Speau is reported to have a major breakdown in its source works once in every six

months, and the distribution system in the utility in Banteay Meanchey is reported to

break down approximately once a year. Among public utilities, the utilities in Kandal

and Svay Rieng have most problems. The treatment facilities of the Kandal utility were

out of service at the time of the survey due to a major breakdown. Also, the distribution

system of the utility fails about once in six months. The utility in Svay Rieng has no

water treatment facilities, and, like the utility in Kandal, it has major problems with its

distribution system twice a year.

22

The percentage of non-working connection is zero or negligible in all but one

utility. About ten percent of connections served by the public utility in Kampong Chhang

are out of service.

Finally, meters and pumps used by public utilities are significantly older in public

than in private utilities. The age of meters varies from three to 15 years and the age of

pumps from three to seven in public utilities, whereas private utilities have meters and

pumps that are one or two years old.

23

Battambang Kandal Kompong

Chhnang

Svay Rieng

Pump Stations Number of stations 2 2 2 1

Condition Functioning

with problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

10 . 8 5

Frequency of Major

Breakdowns

Never Never Never Never

Source Works Condition Functioning

with problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

10 12 8 5

Frequency of Major

Breakdowns

Once in

lifetime

Never Never Never

Transmission Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

10 12 8 5

Frequency of Major

Breakdowns

Never Never Never Never

Treatment Condition Functioning Out of service Functioning No treatment

present

Daily operation

(hours per day)

10 Out of service 8 No treatment

present

Frequency of Major

Breakdowns

Never Out of service Never No treatment

present

Storage Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours)

10 12 4 3

Frequency of Major

Breakdowns

Never Never Never Never

Distribution Condition Functioning Functioning Functioning Functioning

Daily operation

(hours per day)

10 12 8 8

Frequency of Major

Breakdowns

Never Once in six

months

Never Once in six

months

Connections Percentage of nonworking

0.0% 0.9% 9.8% 0.0%

Age (years) Pumps 7 4 3 5

Meters 7 15 3 5

Table 8: Technical Assessment of Water Utilities: Facility Performance

Public

24

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

Pump Stations Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

18 2 10 4

Frequency of Major

Breakdowns

Never Never Never .

Source Works Condition Functioning

without

problems

Functioning

without

problems

Functioning

with problems

Functioning

without

problems

Daily operation

(hours per day)

18 2 10 4

Frequency of Major

Breakdowns

Never Never Once in six

months

Once in six

months

Transmission Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

with problems

Daily operation

(hours per day)

18 2 10 4

Frequency of Major

Breakdowns

Never Never Never Never

Treatment Condition Functioning

without

problems

Functioning

without

problems

Functioning

with problems

Functioning

without

problems

Daily operation

(hours per day)

18 2 . 4

Frequency of Major

Breakdowns

Never Never Never Never

Storage Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

24 24 15 24

Frequency of Major

Breakdowns

Never Never Never Never

Distribution Condition Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Functioning

without

problems

Daily operation

(hours per day)

24 24 15 24

Frequency of Major

Breakdowns

Once a year Never Never Never

Connections Percentage of nonworking

0% 0% 1% .

Age (years) Pumps 2 1 2 2

Meters 2 1 2 2

Table 8: Technical Assessment of Water Utilities: Facility Performance (continued)

Private

25

B. Maintenance

The results of technical assessment reveal that private utilities maintain their

facilities better than public utilities. As can be seen from Table 9, public utilities

maintain their facilities (pump stations, source works, transmission, treatment, storage,

distribution) only when there is a crisis and action has to be taken. As an exception, the

utility in Svay Rieng maintains its distribution system–but not its other facilities–on a

weekly basis. This is in stark contrast to the maintenance practices of private utilities.

All private utilities carry out regular maintenance of all their facilities on a weekly or

monthly basis as a preventive measure.

Public

Battambang Kandal Kompong

Chhnang

Svay Rieng

Frequency of Regular

Maintenance

Pump Stations Never Never Never Never

Source works Never Never Never Never

Transmission Never Never Never Never

Treatment Never Never Never Never

Storage Never Never Never Never

Distribution Never Never Never Weekly

Implementation of water

quality control programs

Chlorine in water Yes No Yes Yes

Regularly cleans

filters

No . No (No filter)

Private

Bantey

Meanchey

Kien Svay Kompong

Speu

Takeo

Frequency of Regular

Maintenance

Pump Stations Monthly Weekly Monthly Monthly

Source works Weekly Weekly Monthly Weekly

Transmission Weekly Weekly Weekly Weekly

Treatment Weekly Weekly Weekly Weekly

Storage Weekly Weekly . .

Distribution Weekly Weekly Weekly Weekly

Implementation of water

quality control programs

Chlorine in water Yes Yes Yes Yes

Regularly cleans

filters

Yes Yes Yes Yes

Table 9: Technical Assessment of Water Utilities: Maintenance

Private utilities also implement water control programs more diligently than

public utilities. All utilities, except the public utility in Kandal, put chlorine in water

regularly. However, only private utilities clean filters on a regular basis. The public

26

utilities in Battambang and Kampong Chhang clean filters only when there is an

overflow. The public utility in Svay Rieng has no filters since it has no water treatment

facilities, and the utility in Kandal does nothing to filters since its treatment facilities are

out of operation.

C. Financial Performance

Almost all residential customers, regardless of whether they are served by private

or public utilities, pay their bills within 30 days. As Table 10 shows, 90 percent of more

of residential customers in each town, except in Battambang, pay their water bills on

time. This is remarkable given that two of the three private utilities that responded to this

survey never cut off service for non-payment, as opposed to one of the four public

utilities.

 

 

11

Five of the six utilities that agreed to share information about their revenues and

expenditures had made a profit last year. The exception, Kampong Chhang, essentially

covered all expenses. Further, one of the two private utilities, Kampong Speau, has a

very high profit margin (201 percent of expenditures). All the other utilities have a profit

margin from -2 percent (Kampong Chhang) to 16 percent (Banteay Meanchey) of

expenditures.

11

 

 

All of these utilities also provide waters services to a small number of offices of governmental agencies;

these customers range from less than one percent to 4.6 percent of the utilities’ total direct connections.

Two of the public utilities (Kampong Speau and Kandal) and two of the private utilities (Battambang and

Kampong Chhang) report that these public customers are more than 90 days late in paying their water bills.

 

27

Table 10: Technical Assessment of Water Utilities: Financial Performance

Battambang Kandal Kampong Chhang Svay Rieng

What percentage of residential

customers has paid bill within 30

days?

85 98 95 100

In practice, after what period of time

is service cut-off for non-payment?

40 90 never 20

Tariff revenues 694,121,152 40,522,000 30,761,400 24,321,800

Other revenues 7,780,000 – 2,296,500 –

Total revenue 701,901,152 40,522,000 33,057,900 24,321,800

Total expenditures 615,355,328 35,400,000 33,780,640 22,851,200

Profits 86,545,824 5,122,000 -722,740 1,470,600

14% 14% -2% 6%

Banteay

Meanchey

Kampong Speau Kien Svay Takeo

What percentage of residential

customers has paid bill within 30

days?

90 100 . 100

In practice, after what period of time

is service cut-off for non-payment?

never 30 . never

Tariff revenues 339,365,280 302,400,000 . 4,815,400

Other revenues 182,958,064 60,800,000 . 27,702,000

Total revenue 522,323,344 363,200,000 . –

Total expenditures 451,813,728 120,840,000 . .

Profits 70,509,616 242,360,000

 

 

. .

16% 201% . .

Notes: All revenues and tariffs in Cambodian Riels (3763 Riels/1 US$)

Public

Private

 

In sum, the results of the technical assessment indicate that private utilities

perform better than public utilities. To ensure continued operation and performance,

private companies, unlike public utilities, maintain their facilities on a regular basis.

Public companies, by contrast, tend to carry out maintenance work only as a response to

crisis. Financially, private and public utilities have relatively similar performance.

6. Access to Water Services

What kind of households (in terms of the level of income and education) do these

private and public utilities serve? The motivation for addressing this question is

straightforward: given the low coverage of utilities, is there evidence that the poor,

because of lack of interest or lack of resources, have not been able to afford either public

28

services or the better performing (and more expensive) private services? If so, what are

the implied policy implications?

Table 11 establishes that households with higher incomes and more household

wealth are significantly more likely to be connected to a piped service — and that this

difference is further attenuated in towns served exclusively by private utilities. First,

columns (1) and (4) of Table 11 show that the estimated income and expenditures of all

connected households (798,457 and 758,356 Riels) are more than twice the respective

figures for all unconnected households (313,742 and 329,678 Riels). This is consistent

with the results of an independent assessment of household income levels based on the

survey team’s observations of the characteristics of the household. Eighty six percent of

all connected households were assessed to be middle or high income, as opposed to 22

percent of all unconnected households. As one would expect, given the positive

correlation between income and education, unconnected household also have lower levels

of education than connected households: thirty five percent of the heads of all

unconnected households report that they have no primary education, in contrast to 18

percent of all connected households.

Second, columns (3) and (7) of Table 11 show that the estimated income and

expenditures of connected households served by private utilities (881,226 and 791,077

Riels) are about three times the respective figures for all unconnected households in

towns with private utilities (278,365 and 293,626 Riels). This even greater difference

between the wealth and income of connected and unconnected households in these towns

is also found in the comparisons of the assessment of household income levels (85

percent middle or high income versus 16 percent middle or high income) and education

levels (17 percent of connected households have no primary education versus 37 percent

of connected households.)

 

 

12

12

 

 

Since Kandal has both public and private utilities, the demographics for the unconnected households in

this town are reported separately, in column 6 of Table 11.

 

29

Table 11: Access to water services, household demographics

All

households

Served by

public

utilities

Served by

private

utilities

All

households

In towns with

public

utilities

In town with

public and

private

utilities

In towns with

private

utilities

Income and wealth

Monthly income (Riels) 798,457 734,082 8 81,226 313,742 311,865 369,131 278,365

Monthly expenditures (Riels) 758,356 730,980 7 91,077 329,678 351,918 354,387 293,626

Low-income 0.13 0.13 0.13 0.78 0.72 0.79 0.81

Middle- income 0.60 0.62 0.57 0.21 0.26 0.21 0.15

High-income 0.26 0.24 0.28 0.01 0.01 0.00 0.01

Have electricity 0.98 0.99 0.97 0.87 0.87 0.98 0.79

Own a color television 0.90 0.92 0.89 0.63 0.65 0.81 0.49

Own a car 0.18 0.17 0.18 0.09 0.12 0.10 0.07

Own a house 0.94 0.93 0.97 0.97 0.97 0.96 0.98

Level of Schooling

No schooling 0.18 0.19 0.17 0.30 0.35 0.13 0.37

Primary only 0.25 0.24 0.27 0.21 0.19 0.28 0.18

Secondary 0.33 0.33 0.33 0.31 0.29 0.41 0.26

High school 0.17 0.19 0.15 0.13 0.10 0.15 0.13

Post-high school 0.06 0.04 0.08 0.02 0.05 0.01 0.01

Sample size 451 251 200 375 125 100 150

Notes: share of connected and unconnected households, respectively, with listed characteristics.

See text for description of characteristics.

Incomes and expenditures in Cambodian Riels (3763 Riels/1 US$)

Connected households Unconnected households

30

Do the results in Table 11 imply that low income households simply can not

afford a piped connection–particularly in the towns with the more expensive private

services? As detailed in Table 12, the majority of unconnected households interviewed

report that they would be interested in getting a piped connection. While 41 percent of

unconnected households on average report that they are satisfied with their current

service (with a range of 14 percent in Takeo to 62 percent in Kampong Chhang), 86

percent note that they would be interested in getting a piped connection. Given this level

of interest among unconnected households, what are the constraints to getting such a

connection?

Utilities’ limited service area and high connection fees are the two most common

reasons for households not having piped connections. On the supply side, a relatively

large share of households (forty two percent) states that the utility does not serve their

area (with a range of 12 percent in Kampong Speau and Battambang to sixty percent in

Takeo). On the demand side, 35 percent state that the connection fee is too high (with a

range of 17 percent in Svay Rieng to 52 percent in Kampong Speau), while very few

(eight percent) state that the monthly tariff is too expensive.

In particular, connection fees, not the level of monthly water tariffs, constrain the

access of the poor to water services. Using the independent assessments of household

income levels, 37 percent of low-income unconnected households report that the

connection fee is too high, in contrast to 29 percent of middle- and high-income

unconnected households. By contrast, there is no (statistically significant) difference

between these types of households in their response about the monthly tariff: 8.3 percent

of low-income unconnected households report that the monthly tariff is too high, in

contrast to 9.1 percent of middle- and high-income unconnected households. Taken

together, these results suggest that the poor are willing to pay for piped water, but in

many cases are unwilling (or unable) to pay, for the cost of getting connected.

31

Table 12: Assessment of alternative water services: unconnected characteristics

In town with

public and

private utilities

Average

household Battambang

Kampong

Chhang

Svay

Rieng Kandal

Banteay

Meanchey

Kampong

Speau Takeo

Evaluation of current water source

Satisfied with current

water source 0.41 0.28 0.62 0.56 0.46 0.30 0.38 0.14

(0.49) (0.46) (0.49) (0.50) (0.50) (0.46) (0.49) (0.35)

Interested in getting a

piped connection 0.86 0.96 0.78 0.82 0.84 0.86 0.94 0.90

(0.35) (0.20) (0.42) (0.39) (0.37) (0.35) (0.24) (0.31)

Why are you not connected?

Utility doesn’t serve

area 0.42 0.12 0.54 0.36 0.48 0.48 0.12 0.60

(0.49) (0.33) (0.50) (0.48) (0.50) (0.50) (0.33) (0.49)

Alternative water

source 0.55 0.16 0.50 0.54 0.68 0.56 0.54 0.56

(0.50) (0.37) (0.51) (0.50) (0.47) (0.50) (0.50) (0.50)

Connection fee is too

high 0.35 0.40 0.33 0.17 0.41 0.30 0.52 0.28

(0.48) (0.50) (0.48) (0.38) (0.49) (0.46) (0.54) (0.45)

Monthly tariff is too

expensive 0.08 0.00 0.11 0.00 0.06 0.02 0.28 0.08

(0.28) 0.00 (0.31) 0.00 (0.24) (0.14) (0.45) (0.27)

Notes: Means and (standard deviations) for water variables. See text for variable description

In towns with public utilities In towns with private utilities

32

This implies that if utilities want to expand their coverage and the scale of

operations, they need make connections more affordable. This does not necessarily mean

that connection fees need to be reduced. The first step to expand access might instead be

to provide households an option to spread the financial burden of a connection over

longer period of time by allowing them to pay the connection fee in installments.

7.

 

 

Conclusions

This paper has compared the performance and consumer satisfaction with water

services provided by four private companies and four public utilities in Cambodia. The

comparison was based on results of a survey of households served by these utilities and

on results of a technical assessment of each water system carried out by a water engineer.

The results indicate that households served private utilities are significantly more

satisfied with the piped water than customers of public utilities. The daily availability

and quality of piped water is better in households served by private than by public

utilities. Also, customers of private utilities experience fewer service interruptions. This

has not happened by accident. Private utilities hire more educated staff whom they pay

higher salaries; maintain their facilities more regularly; and implement water quality

control programs more diligently. Private sector operators seem to face stronger

incentives than public utilities to keep their customers satisfied.

However, according to results, this improved service does not come for free.

Households served by private utilities pay for the piped water service significantly more

than customers of public utilities. The connection fees as well as unit tariffs charged by

private utilities are higher than fees and tariffs of public utilities. Some lower-income

households that are not served by private utilities are partially limited by the high

connection fees (as opposed to the regular monthly payments).

Overall, this paper indicates that the bold effort of a few towns and private sector

participants to introduce private sector involvement in the water sector is encouraging in

many ways. But the full gains, which are possible, have not yet been realized. Earlier, it

was stressed why the commercial incentive for positive net revenue from operations tends

to lead the private entrepreneur to be responsive to the demand for services from actual

and potential future customers and to economize on costs while doing so. This

commercial incentive is likely to be stronger if the privatization option used is a lease or

concession arrangement, rather than a service or management contract, since the operator

in the former cases has access to the revenue stream over a number of years, rather than a

fixed fee which is normal in service and management contracts. It pays the operator to

increase net revenue (revenues in excess of costs) in the former cases. In this regard,

BOT (build, own, and transfer) arrangements are a mixed bag. Usually, BOTs are for

only part of the system, for example, source works to increase capacity of the system.

Since, in this case, the private operator is not responsible for distribution, the operator

will normally require a ‘take or pay’ contract (a guarantee that a minimum amount will

be purchased regardless of the amount sold by the distributor). This clearly relieves the

33

BOT operator of the incentive to be demand responsive in the choice of investments,

although not of the desirability of controlling costs.

The set of commercial incentives facing utilities would clearly be made stronger if

there was competition for the market. As the paper indicates, the initial private sector

initiatives in Cambodia have not had this feature. Consequently, the commercial

incentives to be demand-responsive and to control costs have not been as strong as would

be desirable. This is particularly true in the water sector because of the relatively high

capital costs and longevity of assets. Private operators normally require, even with full

control of the revenue stream, fairly extensive periods to recover investment costs.

This paper shows that private sector operators have been willing to come forth

and that commercial incentives, however much weakened by uncertainty about the

direction and scope of future regulation of the sector, are working to improve the water

supply situation. The study also shows that in the current situation there is not a

completely clear-cut case for the superiority of the private sector over the public sector in

all respects. This is to be expected. Nevertheless, if the regulatory structure, which the

Government of Cambodia intends to put into place, encourages the commercial

incentives to be more demand-responsive and cost conscious, and, further, reinforces the

market incentives of competition, rather than leaving the private sector operators open to

arbitrary government action, the private sector is a good bet.

34

References

De Raet, Pierre and Duvvuri Subbarao. 1999. “Cambodia: Urban Water Supply Policy

and Institutional Framework.” The World Bank.

Idelovitch, Emanuel and Klas Ringskog. 1995. “Private Sector Participation in Water

Supply and Sanitation in Latin America.” The World Bank, Washington, DC.

Silva, Gisele, Nicola Tynan, and Yesim Yilmaz. 1999. “Private Sector Participation in

the Water and Sewerage Sector: Recent Trends.” In

 

 

Water: Competition and

Regulation.

 

 

The World Bank, Washington, DC.

World Bank. 1995.

 

 

Bureaucrats in Business: The Economics and Politics of

Government Ownership.

 

 

Oxford University Press.

World Bank. 1997. “Selecting an Option for Private Sector Participation.” In

 

 

Toolkits

for Private Sector Participation in Water and Sanitation.

 

 

 

The World Bank,

Responder

Por favor, inicia sesión con uno de estos métodos para publicar tu comentario:

Logo de WordPress.com

Estás comentando usando tu cuenta de WordPress.com. Cerrar sesión / Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión / Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión / Cambiar )

Google+ photo

Estás comentando usando tu cuenta de Google+. Cerrar sesión / Cambiar )

Conectando a %s

A %d blogueros les gusta esto: