Public–private partnerships for water in Asia: a review of two decades of experience

Abstract

This article reviews the development of public–private partnerships (PPP) in water services in Asia over the last two decades and situates the Asian experience within the wider global context. Through a comparative analysis of PPP data from two different sources, the article highlights the critical issue of how ‘private’ is defined in understanding the extent of PPP in Asia, due to the important role played by enterprises under mixed public and private ownership. The article identifies cyclical patterns of development of PPP in the water sector across countries and the use of hybrid contractual and institutional arrangements for PPPs.

Keywords:

• Water
• public-private partnerships
• water utilities
• Asia
• economic regulation

JEL:

• H44
• H42
• L95
• L32
• L33

Introduction

Despite their chequered reputation, public–private partnerships (PPP) for water services, including the treatment and distribution of drinking water and the collection, treatment and disposal of wastewater, have been adopted in countries at all income levels and in a wide range of institutional and geographical settings. Although some authors argue that water PPPs have failed (Araral, 2009; Bakker, 2010; Castro, 2008), new contracts continue to be signed every year.

This article reviews how PPPs for water services have developed over the last two decades in Asia. Asia’s earliest water PPP was signed more than half a century ago (a lease contract for water services in the Philippines, awarded in 1960), at a time when only one other developing country had a water PPP (Marin, 2009). A second contract, for water supply in Macau, was agreed in 1989. But it was not until the 1990s that PPPs began to spread in earnest with a wave of investments by multinational utility companies.

In that decade, some countries, like Malaysia (Chin, 2008), adopted water PPPs as part of a broad policy to deregulate and stimulate investment in infrastructure. Others began to experiment with PPP on an ad hoc basis, a trend which gathered momentum and culminated in the award of two of the largest concession contracts for water supply in the world in Manila in 1997 and Jakarta in 1998. In China, PPP in the water and wastewater sector grew extremely rapidly after 2002 using distinctive contract models (Jang, Lee, & Choi, 2014; Lorrain, 2012), and it quickly became the largest water PPP market in the world in the mid-2000s.

In Asia as elsewhere, some large, high-profile, foreign-invested contracts have encountered difficulties (Bakker, 2007; Tan, 2012; Wu & Malaluan, 2008), contributing to an overall impression that PPP in the sector has not been a success. Certainly, the development trajectory of PPP in the region has been non-linear, and many countries have undertaken sharp policy reversals. Yet in 2015, most governments in the region have supportive policies for at least some forms of private participation in water services. Many are experimenting with hybrid PPP types tailored to local conditions, and domestic companies are playing an increasingly important role, as they are in other developing regions (Blanc & Botton, 2012; Marin, 2009).

For the purposes of this article, the subject of interest is the involvement of the private sector in water services rather than any particular contractual form, and so PPP is defined broadly as an arrangement in which a private entity either operates a water utility or a related asset, or has an ownership stake in a utility or one of more of its assets. It therefore includes the full and partial divestiture of state-owned utility assets as well as concessions, BOTs (build-operate-transfer) and other contract-based models such as leases and management contracts. Definitions of the contract types are given in the online supplemental material (available at http://dx.doi.org/10.1080/07900627.2015.1121136).

The article adopts two complementary methods. Firstly, it measures PPP activity in the water sector in quantitative terms. Information on PPPs in the water sector is particularly difficult to collect because of their local nature and small scale, so no database is perfect. This article therefore uses the two most complete and reliable global databases available and compares data from the two. Secondly, it presents national histories of PPP for Asian countries based on a long-term qualitative data collection effort across the region. It draws on more than 200 semi-structured interviews conducted by the author with governments, regulators, financiers and water companies in Asia in several rounds from 2004 to 2014. A list of interviews conducted is provided in the online supplemental material. These qualitative histories help reveal the country-specific institutional attributes and experience with PPP that underlie the trends in project numbers.

The second section of this article presents and compares measures of PPP activity from the two databases. The third section examines the important role played by state-owned and mixed-ownership enterprises in Asia and shows how this relates to the numbers of PPP projects recorded for the region. The next section presents short country PPP histories, while the final section concludes with findings and directions for future research.

Trends in water PPP

The two most reliable and complete data-sets available of PPPs in the water sector are the World Bank-PPIAF Private Participation in Infrastructure Database (PPI, ppi.worldbank.org), which covers only developing countries, and the Private Sector Participation in Water Database (PSP Water) maintained by Global Water Intelligence (2014a), which also includes developed countries. PSP Water incorporates historical data from the Pinsent Masons Water Yearbook, which was last updated in 2012.

Both databases follow a definition of PPPs as contracts that transfer full or partial ownership or decision rights to the private sector. PSP Water classifies PPPs into five categories. Two relate to asset procurement – design-build-operate (DBO) and BOT and its variants – and three to utility services: ‘high capex’ models (investor-owned utilities); ‘medium capex’ models, including leases, concessions, affermage and operating contracts; and ‘low capex’ models covering management and service contracts. The database includes a small number of projects in the service-contract category which do not involve the transfer of operating risk to the private party. These have been excluded from the data used in this article to ensure comparability with the PPI data, which stipulates that the private party must assume at least part of the operating risk to be classified as a PPP. PPI adopts a similar categorization for project types: ‘greenfield’ for new assets; management and lease contracts; concessions; and divestitures.

In addition to the basic data on the name and location of each project, PSP Water includes data on estimates of population served for some contracts, while PPI includes data on investment volumes where available, so the two databases are to some extent complementary. Given the generally high quality of the databases and the consistency between the two in the classification of PPP, it is interesting to see the extent to which the data converge.

Contract counts

One measure of PPP activity is simply the number of projects signed per year per country. The relevant data from the two databases are presented in Tables 1 and 2 and compared in Figure 1. Both show a steady rise in the number of projects signed per year at the global level from the early 1990s, culminating in a period of high activity in 2004–2008. After 2008, the number of new projects drops off rapidly. This may be due in part to the global financial crisis, which restricted companies’ access to finance, and to a surge in public investment in infrastructure under national stimulus plans, which may have crowded out private investment. Although the number of new projects is declining, more than 100 new projects were signed each year worldwide in 2009–2013.

Table 1. PPI Database Count of Water Public–Private Partnerships, 1994–2013.

Region	1994	1995	1996	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006/	2007	2008	2009	2010	2011	2012	2013	Total
China	2	1	4	6	10	3	6	10	18	25	28	42	43	60	46	31	14	27	20	13	409
All Asia	4	4	5	14	13	7	15	14	20	27	32	47	45	67	49	33	15	27	22	15	475
Europe and Central Asia		1	3		1	1	5	3	2	9	4	6	5	2	1				2		45
Latin America and Caribbean	10	10	15	22	16	21	21	20	22	10	22	7	4	11	11	5	8	4	14	6	259
Middle East and North Africa			1			2			1	1		4		3	5	2	2	2	1		24
Sub-Saharan Africa		1	1		1	5	1	4	2	1		1	2	2	2				1	1	25
Total	14	16	25	36	31	36	42	41	47	48	58	65	56	85	68	40	25	33	40	22	828

Source: PPIAF/World Bank PPI Database (ppi.worldbank.org), data downloaded 23 March 2015.

Table 2. PSP Water Database Count of Public–Private Partnerships, 1994–2013.

Region	1994	1995	1996	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006/	2007	2008	2009	2010	2011	2012	2013	Total
China	2		4	10	7	3	12	19	44	59	114	116	125	134	164	113	56	51	29	24	1,086
All Asia	4	5	4	18	10	7	21	23	49	66	120	129	131	144	176	120	69	57	54	43	1,250
Europe and Central Asia	8	11	16	13	16	24	28	38	25	36	35	39	57	25	30	25	37	34	26	28	551
Latin America and Caribbean	8	8	15	19	17	16	24	20	16	11	15	5	13	11	13	6	16	13	14	6	266
Middle East and North Africa			1		1	2		9	3	2	3	9	7	12	12	9	13	7	10	7	107
North America	15	10	18	11	21	19	19	11	27	35	26	30	27	40	28	22	29	25	17	18	448
Pacific	1	2	2	1	1	1	4	1	5	2	5	2	3	2	5	5	2	5	3		52
Sub-Saharan Africa		2	1	1		4	2	4		1		2			1		1	1	2	1	23
Total	36	38	57	63	66	73	98	106	125	153	204	216	238	234	265	187	167	142	126	103	2,697

Source: Global Water Intelligence (2014) Private Sector Participation in Water Database.

Figure 1. Number of projects by year: PPI & PSP data-sets. Note: PPI: World Bank/PPIAF Private Participation in Infrastructure Database. PSP: Global Water Intelligence Private Sector Participation in Water Database.

Previous analysis of the drivers of PPP using the PPI data found that higher income levels are associated with higher levels of PPP (Hammami, Ruhashyankiko, & Yehoue, 2006; Jensen & Blanc-Brude, 2006a). The PSP Water data appear consistent with this, showing substantial numbers of projects in North America and Western Europe, although rigorous econometric analysis has not yet been carried out with PSP Water. The difference in project numbers between the two data sources in Europe (counts of 45 and 551 in PPI and PSP, respectively) and the Middle East (24 and 107) is largely due to the different country coverage of the two data-sets.

Experience with PPP in the water sector is widespread in geographical terms: 110 countries are represented in PSP Water as having awarded at least one PPP contract in the last two decades. In Asia, a total of 14 countries are represented. However, both databases show an increasing degree of geographic concentration in PPP activity, especially within developing countries. The PPI data include only five new contracts in developing countries in 2013, including three in Asia: China, Indonesia and India (Lim, 2014).

Figures 2 and 3 compare the two databases for Asia-excluding-China, and for China, respectively. In Asia-ex-China, the divergence between the two data-sets is mainly due to differences in project counts in India. Some operating contracts in India appear not to be included in the PPI database even though they are consistent with that database’s definition of PPP, perhaps because of a classification issue or because information about these projects is not readily available in the public domain. Overall, PPP activity in Asia is heavily concentrated in China, and here again there is wide divergence between the two data-sets, as shown in Figure 3. Of 1250 projects in the region in PSP Water over the two decades from 1994–2013, 1086 (87%) are in China. PPI records lower total figures but a very similar proportion (86%). In this case, the difference does not appear to be related to classification of contract type. It may be due in part to data being less readily available in China, but an additional possible explanation is the leading role taken by state-owned enterprises (SOEs) in water PPPs in China, and the way in which these enterprises are differently classified in the two databases. This explanation is explored in more detail below, in the Comparison section.

Figure 2. Number of projects (Asia excluding China).

Figure 3. Number of projects (China).

In all countries in the region, traditional public procurement remains the standard model for water infrastructure procurement, and public utilities dominate the provision of water and wastewater services. Even in China, the overall penetration of PPP in the water sector is low compared to the total market size and appears to have peaked in 2008. This is not peculiar to Asia, of course: with few exceptions, the water sector is predominantly public throughout the world. Water PPPs appear to be more difficult for both firms and governments than other infrastructure PPPs, because of their generally low levels of profitability and their vulnerability to the dissatisfaction of consumer-citizens (Menard, 2013).

The data in Tables 1 and 2 capture only new contracts signed each year rather than the total number of active projects; they therefore do not reflect contract terminations or suspensions. Information on terminated or suspended contracts is provided separately, in Table 3. The number of such projects is small compared to the total number of projects in the region, but the list does include some of the larger contracts to be signed over the last two decades, which perhaps accounts for the general impression of ‘failure’ associated with water PPP.

Table 3. Contract Cancellations and Terminations.

Country	Contract	Contract signed	Contract status
Build-operate-transfer
China	Da Chang	1997	Terminated in 2004
China	Beijing No. 10	2002	Cancelled (failed to reach financial close); developed by government under standard procurement mechanism
China	Qingdao Desalination	2008	Terminated in 2014
India	Tirupur	2005	Original private investors withdrew from the contract and were replaced by public entities
India	Haldia	2007	In dispute
India	Latur	2007	Suspended
India	Aurangabad	2011	Failed to reach financial close
Malaysia	Johor (Equiventures) (BOT)	1999	Terminated
Malaysia	Selangor (multiple BOT contracts)	1994–2001	Agreement for cancellation and asset transfer made in 2015
Thailand	Pathum Thani	1998	Renegotiated; foreign partner sold equity stake to local partner
Vietnam	Thu Duc (BOT)	1997	Cancelled (failed to reach financial close); developed by government under standard procurement mechanism; minority equity stake sold to a private party in 2011
Concessions
Indonesia	Jakarta West, Jakarta East	1998	Contracts annulled in 2015 by local court; appeal ongoing; probable transfer of management control to local government entity
Malaysia	Kelantan	1995	Terminated; management control transferred to state government
Malaysia	Indah Water Konsortium (asset sale)	1995	Terminated and transferred to central government ownership
Malaysia	Johor (SAJ)	2001	Terminated and replaced with management contract
Malaysia	Selangor (Syabas)	2004	Agreement for cancellation and asset transfer made in 2015
Philippines	Manila West (Maynilad)	1997	Terminated and re-bid to a different private consortium

Source: Jensen and Blanc-Brude (2006), Global Water Intelligence.

However, not all contract terminations are a signal of the outright rejection of PPP. For example, after the termination of the Maynilad contract for the western part of Manila, the concession structure was retained, and the contract was subsequently rebid and awarded to a different private consortium. Another example is Ho Chi Minh City in Vietnam, discussed in more detail in the Policy Swingers section, below.

Investment volumes

Data on the number of contracts cannot convey the huge differences in the scale of contracts. In 2012–13, for example, the average water/wastewater PPP in China involved an investment of less than $10 million, while the largest deal was a $1.9 billion wastewater treatment plant in Brazil (Lim, 2014).

Data on capital investment (capex), meanwhile, do capture scale but are much harder to collect and verify. PPI includes data for expected investment values at the time the contract is signed. This figure is useful because it captures the total capex over the life of the contract and does not depend on what stage in its lifecycle the project has reached. However, there are obvious drawbacks in using this measure. Firstly, this information is often not made public, and those values that are announced by the parties involved may be impossible to verify independently, leading to missing values or errors in values in the data-sets. Secondly, expected capex may diverge widely from the volume of investment that is actually made. Figure 4 shows an example of the gap between planned and actual capital expenditure for the west concession in Manila, for which investment data are recorded by the regulatory body (Jensen, 2008). The west Manila concession experienced severe shocks within the first year of the contract, so this cannot be considered a typical case, but it nevertheless shows how quickly cumulative investment can diverge from the initial plan. Together, these errors could lead to significant overestimates of investment volumes in individual projects but at the aggregate level may result in underestimates because of the large number of missing values for smaller projects.

Figure 4. Planned and actual investment: West Manila.

Notwithstanding these limitations, investment volume data help round out the picture given by the project counts. Table 4 and Figure 5 present data on investment values from the PPI Database at the global level and specifically for Asia (PSP Water has too many missing values to allow a comparison between the databases on this measure). The table shows considerable volatility in investment flows due to the small number of large projects, as noted in the literature (Menard, 2013). Overall, project size is strikingly small compared to other infrastructure sectors like telecoms and power. The data show a decline in average contract size across developing countries, from $107 million in 1994–2003 to $77 million in 2004–2013. This drop in average investment per contract may in part be accounted for by the shift from stronger forms of PPP involving private investment to milder forms in which the private party does not take any financing risk.

Table 4. Investment Volumes (World and Asia), 1994–2013.

Current $ millions	All countries			Asia only
Investment year	Payment commitments to the government	Investment commitments in physical assets	Total investment commitments	Payment commitments to the government	Investment commitments in physical assets	Total investment commitments	Asia as % of total investment commitments
1994	35.0	1,311.0	1,346.0	0	820.8	820.8	61%
1995	10.5	1,802.8	1,813.3	0	520	520	29%
1996	36.4	1,267.2	1,303.6	0	149.3	149.3	11%
1997	1,800.2	8,165.5	9,965.7	1,129.3	6,903.2	8,032.5	81%
1998	489.3	1,837.4	2,326.7	72	870.8	942.8	41%
1999	2,022.1	4,341.5	6,363.6	85.6	184.9	270.5	4%
2000	666.5	6,544.3	7,210.8	0	4,066	4,066	56%
2001	244.2	1,611.7	1,855.9	23.5	651.6	675.1	36%
2002	717.1	829.3	1,546.4	403.9	530.2	934.1	60%
2003	273.8	1,229.6	1,503.4	41.8	661.7	703.5	47%
2004	753.3	3,892.8	4,646.1	258.3	2,985	3,243.2	70%
2005	430.3	2,052.4	2,482.7	430.3	910.7	1,340.9	54%
2006	663.5	1,936.0	2,599.5	229.9	917.3	1,147.2	44%
2007	1,033.6	2,392.7	3,426.3	948.6	1,095.5	2,044.1	60%
2008	356.3	2,504.3	2,860.6	356.3	693.2	1,049.5	37%
2009	59.6	2,130.8	2,190.4	59.6	503	562.6	26%
2010	524.4	1,828.5	2,352.9	516	123.7	639.7	27%
2011	2,010.7	615.8	2,626.4	464.5	127.2	591.6	23%
2012	547.2	5,164.3	5,711.5	34.4	548.5	582.9	10%
2013	159.7	3,265.4	3,425.1	43.7	348.2	391.9	11%
Total	12,833.6	54,723.1	67,556.8	5,097.6	23,610.4	28,708.1	42%

Source: World Bank/PPIAF (ppi.worldbank.org), data downloaded 23 March 2015.

Figure 5. Investment volumes.

The data also show a decline in the contribution of Asia to global totals: between 2002 and 2007 Asia accounted for 56% of all investment commitments for water PPPs in developing countries, but by 2008–2013 that share had dropped to 22%. This reflects declines in both project size and project count in the region.

Population served

A third way to measure the extent of PPP in different regions is to estimate the number of people who receive water or wastewater services under a PPP contract. There are considerable challenges in estimating these figures. PSP Water only seeks to estimate figures for utility management PPPs (concessions, affermage, and operations and maintenance), while the PPI database does not record this measure at all. The PSP Water figures do not include people served by investor-owned utilities, the predominant model of PPP in England and Chile, and therefore underestimate the total global population served by the private sector. However, this type of PPP is extremely rare in Asia, so this omission is not very significant for the region. The figures also exclude those who are indirectly served by a PPP, in the sense that the utility that supplies them is engaged in a bulk supply PPP.

The data are presented in Table 5 and Figure 6. They show a slow but steady increase in this measure at the global level. In Asia-Pacific, the growth rate has been much higher: the population served under these contracts rose by 82% between 2006 and 2013. In 2013, PPPs served 67.5 million people in Asia, 4% of the total urban population. This increase contrasts with the Americas and the Middle East and North Africa, where contracts have not been renewed or have been terminated early, leading to an estimated decline in the number of people directly served by a PPP. The slowly changing stocks of population served provide a counterpoint to the impression of high volatility in water PPP given by the annual data on investment flows.

Table 5. Population Served by Public–Private Partnership Contracts, 2006–2013 (millions).

Region	Americas	Europe	Asia-Pacific	Middle East/Sub-Saharan Africa	Total
2006	102.35	91.11	37.1	17.36	247.92
2007	101.6	91.8	52.3	14.11	259.81
2008	105.98	91.41	53.03	14.11	264.53
2009	103.13	92.53	54.7	14.11	264.47
2010	102.73	93.76	55.84	14.11	266.44
2011	103.54	97.47	63.47	14.11	278.59
2012	100.82	98.42	66.37	14.34	279.95
2013	100.53	106.77	67.51	14.34	289.15
% change	−2%	17%	82%	−17%	17%

Source: Global Water Intelligence.

Figure 6. Population served under public–private partnership contracts (excludes investor-owned utilities).

Taken together, these three measures of PPP activity show that Asia is an important locus of PPP, although the region’s role in overall activity has declined since 2008. The continued increase in the stock of PPPs and growth in population served by the private sector suggests that PPP is well embedded in the region, particularly in middle-income countries. Yet, there are significant differences between the databases in the numbers of contracts recorded. What accounts for this gap? The next section addresses this question.

Comparison

In addition to the difference in geographical coverage of the two databases, the wide divergence in the numbers of projects that they record may be due to their differing approaches to defining ‘private entity’ in relation to a PPP.

PPI defines a private sponsor as a company “controlled and majority owned by private parties”. SOEs and their subsidiaries are considered private investors only when they engage in projects in foreign countries. Partially divested SOEs or their subsidiaries that remain majority-owned by government entities are not considered private sponsors in their own countries, although the divestiture transaction itself is included as a PPP.

In PSP Water, on the other hand, contracts involving state-owned companies in which private investors own a minority stake and the entity behaves like a private utility (either in bidding for third-party projects or in paying dividends to its shareholders) are considered PPPs. The impact of this difference in classification is particularly pronounced in China, as noted above: PPI records 409 contracts in 1990–2013, compared to 1086 in PSP Water. This is because the most active companies in the country are publicly listed majority SOEs like Beijing Enterprises Water Group, Beijing Capital and Shanghai Industrial Holdings.

Neither of these definitions is clearly to be preferred over the other, but it is useful to make explicit the assumptions about firm behaviour on which these definitions rest. By including international projects sponsored by SOEs, PPI makes the implicit assumption that the distinguishing ‘public’ characteristics of an SOE do not apply when it operates outside its home country. But a similar argument could be made for an SOE owned by a sub-national (municipal or state) government when it operates in another municipality or state within a federal system. From this point of view, the PPI figures are likely to be significantly underestimated. It is interesting to note that in China national PPP regulations define an SOE as a ‘private partner’ in a PPP contract if the public party is outside its parent’s administrative jurisdiction.

On the other hand, the PSP Water approach rests on the assumption that the contract form and award process (rather than ownership) are the critical factors that determine whether an SOE behaves like a private entity. This seems plausible in cases where the contract is being awarded in a different jurisdiction, but less so when an SOE is awarded a contract by a branch of its own parent government, as would be the case for Beijing Capital in the city of Beijing, for example. Excluding this type of arrangement from PSP Water would lead to a small downward adjustment in project numbers.

There is probably no final answer to the question of how a private entity should be defined for the purposes of a database, but recognizing the significant share of contracts that involve entities with joint public-private ownership, and understanding how these entities are structured and how they behave, will be highly relevant to our understanding of PPPs in the water sector in general, and in particular in Asia. The next section sets out a typology of mixed-ownership entities and provides examples from the region.

Mixed ownership

There are three main models of mixed ownership in the water sector in Asia: joint equity investment in a project company; part-privatization of an SOE; and acquisition of an equity stake in a private enterprise by the state. They are discussed in turn.

Joint ownership of project companies

These are special-purpose vehicles usually established in the context of a PPP contract and relating to a specific asset or assets. In the water sector, this would usually be a water or wastewater treatment plant. Generally the company is responsible for the construction and operation of a new asset under a BOT-type contract, but they may also relate to the renewal or extension of a brownfield asset under a transfer-operate-transfer or concession contract (see the online supplemental material for descriptions of contract types). In the water sector, the government equity holder is often a municipal government rather than the national government. In some cases, the government owns a ‘golden share’ in the project company, which allows it to exercise a veto over certain management decisions.

Examples of public-private project companies are found in China (Lee, 2010), Malaysia, the Philippines and Vietnam; details are given in the top panel of Table 6. A related structure is one in which the government co-invests with private companies to set up a service company. This model is used, for example, by Tokyo Suido Services, a company owned 51% by the Tokyo public utility, Tokyo Waterworks Bureau, along with two Japanese private companies.

Table 6. Mixed-Ownership Structures in the Water Sector.

Company	Country	Public equity holder	State equity share (%)
Joint ownership of project companies
SingSpring (desalination)	Singapore	Temasek	50
Kenh Dong Water Supply	Vietnam	Ho Chi Minh City (HCMC) municipal gov.	51
Thu Duc Water Supply	Vietnam	HCMC municipal gov.	51
Konsortium Abbas	Malaysia	KPS (Selangor State)	90.8
Splash	Malaysia	KPS	30
Syabas	Malaysia	KPS	30
Subic Water	Philippines	SBMA, Olongapo City Water District	30
Cebu Water	Philippines	Cebu Water District	49
Pathum Thani	Thailand	Provincial Waterworks Authority	2
Partial privatization of state-owned enterprises
Phnom Penh Water Supply Authority	Cambodia	Phnom Penh Municipality	86.5
Beijing Capital	China (Shanghai)	Beijing Municipality	61^*
Beijing Enterprises Water Group	China (Hong Kong)	Beijing Municipality	51^*
Chongqing Water Group	China	Chongqing Municipality	89^*
Guangdong Investments	China	Province of Guangdong	57^*
Lanzhou Water Company	China	Lanzhou Municipality	50
Qianjiang Water Resources	China	Ministry of Water Resources et al.	52^*
Shanghai Industrial	China (HK)	Shanghai Municipality	57^*
Shanghai Pudong Water Company	China	Shanghai Municipality	50
Shenzhen Water Company	China	Shenzhen Municipality	50
Tianjin Capital Environmental Protection	China (HK, Shanghai)	Tianjin Municipality	52^*
PBA Holdings	Malaysia	State of Penang	65
East Water Group	Thailand	Govt of Thailand	64
Ben Thanh JSC	Vietnam	HCMC Municipality	51
Gia Dinh JSC	Vietnam	HCMC Municipality	51
Nha Be JSC	Vietnam	HCMC Municipality	51
Phu Hoa Tan JSC	Vietnam	HCMC Municipality	51
Minority holdings
Sembcorp	Singapore	Temasek	48.9
Ranhill Bhd	Malaysia	EPF	9
China Everbright International	China (HK)	China State Council through China	42
China Water Investment (HHO)	China (HK)	Ministry of Water Resources et al.	18

^* State ownership (%) = 1 – (free float/shares outstanding). Source: Company annual reports, Bloomberg.

Partial divestiture

Partial divestiture of SOEs is a very common model in the region, particularly in formerly planned economies where state ownership prevailed across sectors, as shown in the middle panel of Table 6. In some cases, the partial divestiture takes place through a direct sale to a private company. This model was used in China in the high-profile tendering of a 50% stake in the Shanghai Pudong water utility in 2002. Partial divestitures through private sale were subsequently conducted in Shenzhen, Lanzhou and smaller cities. This is the main route through which private companies have taken a role in distribution in China. As in a standard equity partnership between private players, profits are distributed in accordance with the partners’ equity shares (see Lee, 2010 ).

In other cases, partial divestiture is effected through a stock market listing. A striking example of this phenomenon is the listing of the Phnom Penh Water Supply Authority (PPWSA) in Cambodia. The PPWSA, widely regarded as one of the best-managed water utilities in the region, was chosen to be the first company to list on the fledgling national stock exchange in 2012. In an economy dominated by SOEs, the water company stood out as a well-run enterprise that would support the development of local capital markets. A further example is East Water, a subsidiary of the Thai national water utility, PWA, which was one of the first water companies in the region (either public or private) to stage an initial public offering.

In China, partial divestiture through public listing is a cross-sector phenomenon, and water has followed the pattern of other sectors. The largest water-sector investors in the country, such as Beijing Enterprises Water Group, Shanghai Industrial and Beijing Capital, all fit into the category of listed companies majority-owned by their respective municipal governments.

This approach has also been used in Vietnam. Ho Chi Minh City has been the front-runner in this respect. In 2007, the state-owned utility, SAWACO, split off its eight distribution units into companies. Six of the eight were listed on the stock exchange. The utility has retained a 51% stake in each one; other shareholders are local entrepreneurs and in some cases foreign companies.

While Asian governments have proved open to selling equity stakes to private parties, there is only one example of a full divestiture in the region. This was the privatization of the national sewerage company of Malaysia, Indah Water Konsortium, in 1993. The company was unable to achieve financial viability, and it was renationalized in 2001.

State minority equity acquisition

A third route to state ownership of water companies has been through acquisitions of shares in an existing private entity, shown in the bottom panel of Table 6. This approach has been taken in Singapore and Malaysia, for example. Temasek, Singapore’s state investment company, is involved in the water sector as an equity investor in listed company Sembcorp Utilities and as a 50% joint venture partner in SingSpring, the special-purpose vehicle of Singapore’s first desalination plant. In Malaysia, the federal government’s social security fund, Employees’ Provident Fund, holds a minority stake in Malaysian infrastructure company Ranhill Berhad, and the national water asset holding company purchases water company bonds.

Discussion

What might be the implications of these mixed-ownership structures for the behaviour of the firms? On the one hand, one might expect joint ownership to align the incentives of the public and private parties better than the sometimes adversarial relationships in many public–private partnerships. In fact, joint ownership may come closer to the underlying idea of partnership as it is used in the private sector than standard PPPs (Steijn, Klijn, & Edelenbos, 2011). Ideally, the public partner brings a long-term perspective to the venture, ensures that externalities are internalized, and takes account of social equity. The private party, meanwhile, brings management expertise, a concern for efficiency, and a willingness to adopt innovative approaches. However, there is no guarantee that the model will bring out the best in the partners in practice. The literature on private–private joint ventures suggest that these relationships are often problematic (Si & Bruton, 2003). In the public–private context, the difficulties could be even more severe. One can easily envisage a scenario in which the public and private parties jointly exploit their position of monopoly to enrich themselves at the expense of the consumer, or one in which they agree to perpetuate a low-level equilibrium of suppressed tariffs and minimal investment to meet political aims while sustaining some level of profitability. While this low-level equilibrium may not be sustainable in the very long run, the example of Jakarta’s concessions shows that it can be perpetuated for decades (Jensen, 2005).

The prominence of mixed-equity structures in the water sector contrasts with other infrastructure sectors in which full private ownership is much more widespread. The explanation may lie in the wide gulf that seems to exist between the level of returns that private parties require, even in a sector with low demand risk, and the level of profitability that is politically acceptable in such a socially and environmentally sensitive sector.

The complex patterns of joint equity ownership of water companies may also be interlinked with the absence of economic regulation in the region, as it provides a way for governments to control directly key aspects of the firm’s behaviour. Table 7 presents information on economic regulation of the water sector across countries. It is immediately clear that there are very few dedicated agencies and that even fewer have decision-making powers. In the majority of countries, local governments are responsible for setting tariffs and monitoring performance, tasks which are easier to perform as an equity-holder than as an external agency. Those agencies that do exist face institutional and resource constraints which undermine their ability to perform their functions. One of the only bodies which does manage to function effectively is the MWSS Regulatory Office, which is responsible for the economic regulation of the Metro Manila contracts. In the absence of regulators with statutory responsibility for achieving the financial sustainability of the sector, there are few, if any, public proponents for tariff increases.

Table 7. Regulatory Institutions.

Country	Consumer tariff-setting	Performance oversight	Related regulation/legislation
Bangladesh	Tariffs are set by municipal governments. A regulatory framework is being developed with a grant from ADB.	No dedicated agency responsible for performance monitoring. Water Cell within the local government division is planned to take on this role in the future. No performance monitoring. Ministry of Industry and Handicrafts Department of Potable Water Supply will take responsibility for monitoring under proposed regulations.	PPP Office supports project initiation; government subsidy available for BOT projects.
Cambodia	Water supply regulations and sector law are being developed by the government	No performance monitoring. Ministry of Industry and Handicrafts Department of Potable Water Supply will take responsibility for monitoring under proposed regulations.
China	Local government sets tariffs taking account of central government guidelines, e.g. progressive tariffs	Provincial-level bodies oversee municipal bodies. Central and provincial governments establish quality and performance standards and targets.	Cap on returns for private investors of 8–12%. Initial fee specified in PPP contract (adjustment provisions are not specified in contracts).
Hong Kong	Water Supplies Department, with approval from government	Water Supplies Department, Development Bureau
India	Tariffs are set by local governments. Central government exercises some influence, e.g. through conditions attached to grant funding.	State government oversight. A state-level oversight body for water has been established in one state (Maharashtra) and is being considered in other states (e.g. Gujarat). No central or state-level regulations governing operating or financial performance.	Payment mechanism, initial fee and adjustment formula specified in contract. State-specific: some states have adopted infrastructure regulations which specifically allow for PPP in water.
Indonesia	Tariffs set by local government (must be approved by local assembly). Federal guidelines on operating cost recovery are not implemented consistently.	Contract-specific regulator in Jakarta JWSRB has no legal authority and no powers of enforcement.	2004 Law and subsequent regulations allow local governments to engage in PPP. Contract-specific regulator JWSRB plays a consultative role; no powers to adjust tariffs/fee.
Japan	Local utilities	Local government
Korea	Local government	Central government (Ministry of Security and Home Affairs)
Lao	Provincial government, under the oversight of the Water Supply Regulatory Office	Government (Water Supply Regulatory Committee and Office)	1999 policy statement established Water Supply Authority with power to issue licenses to suppliers and monitor sector; 2008 Water Supply Regulatory Board Committee and Office established by ministerial decision; 2009 Water Supply Law.
Malaysia	National regulatory body (SPAN), established in 2007, has power to set tariffs for all states which have completed transfer of assets to federal level. Consultative decision-making style.	SPAN sets performance indicators; renewal of operating licenses linked to achievement of performance targets.	Concessions that have not undergone restructuring under the 2006 law are overseen by state governments and state level regulatory bodies.
Macau	Government, with principles set out in the contract	Government representative assigned to oversee contract
Nepal	Government, through the Water Supply Tariff Fixation Commission	Government, through the Water Supply Tariff Fixation Commission
Philippines	Regulator MWSS-RO responsible for setting tariffs in Metro Manila in accordance with a process and formula set out in the concession contract. Outside Manila, local governments are responsible for setting tariffs; since 2011 tariff adjustments must also be approved by the National Water Regulatory Board.	Local government. LWUA (Local Water Utilities Administration) for water utilities in larger cities. NWRB (benchmarking).	BOT Law passed in 1993 and amended in 2012.
Singapore	Utility (PUB), with government approval	Government (Ministry of Environment), PUB Board of Directors	PPP model is used for bulk supply projects (reuse and desalination). Off-take tariffs specified in the contract are not made public.
Sri Lanka	Government (Ministry of Water Supply and Drainage); Public Utilities Commission may take over this function in the future.	Government (Ministry of Water Supply and Drainage)
Taiwan	Central government	Government (Water Resources Agency)
Thailand	Tariff adjustments proposed by MWA and PWA and approved by government (National Water Supply Regulatory Board and Cabinet); Wastewater tariffs set by local government.	Water: Government (National Water Supply Regulatory Board and Cabinet). Wastewater: Wastewater Management Authority oversees local authorities.
Vietnam	Local government. Central government sets guidelines but implementation is incomplete.	No dedicated agency for performance monitoring.

Sources: Compiled by the author.

Although this combination of direct state control through ownership and the absence of autonomous regulatory institutions appears to be well embedded in the Asian institutional context and accepted by both public and private parties, further research is needed before we can answer the question of whether this structure offers adequate incentives for efficiency and equity.

National water PPP histories

This section is based on more than 200 interviews with policy makers, regulators, utility managers, financiers, legal advisors, and companies in the water sector in Asia conducted by the author in 2004, 2006 and 2011–2014. Interviews were semi-structured and conducted either in person or by telephone. Each interview sought to elicit information from interviewees on the principles and practice of PPP in each country. This section also draws on relevant academic articles and two comprehensive commercial studies, Water Market Asia (Jensen & Blanc-Brude, 2006b) and Global Water Markets (Global Water Intelligence, 2013).

There is no single PPP story that applies to all Asian countries, but it is possible to identify patterns of experience in water PPP in the context of what we might term national water service systems. These systems involve the interaction of multiple layers and branches of government, together with private actors, in the delivery of water and wastewater services.

Looking at the last two decades, we can classify countries into five groups according to the timing and nature of PPP in the water sector:

•	Early adoption and reversal: Malaysia, Thailand
•	Policy swingers: Indonesia, Philippines, Vietnam, India
•	Stable second-wave adopters: China, Korea (wastewater), Singapore
•	Reluctant adopters: Japan, Taiwan
•	Early-stage reformers: Cambodia, Laos, Myanmar, Nepal, Bangladesh, Sri Lanka.

The classification of countries into these groups is an inductive one. Rather than setting out with a hypothesis to test – say that democracies and authoritarian governments are likely to have different patterns of PPP in the water sector, or that formerly communist countries are likely to have more or fewer PPPs in the water sector – we start out by observing the development of PPPs in each country and draw out the classification from patterns in these histories. This inductive approach helps reveal common issues and approaches that countries have taken to PPP which may cut across national income levels, political systems and national economic systems.

Early adoption and reversal

Malaysia was one of the first countries in the region to encourage private-sector investment in infrastructure, both through greenfield projects and through divestitures. Privatization was officially launched with the Privatization Master Plan of 1991, but several sectors, including water treatment, were already open to private investment. Initially, PPPs in the water sector were small-scale bulk supply contracts. Foreign companies were involved as operators, but regulations limited foreign equity ownership to 30%. The speed and scale of PPP built up rapidly to culminate in a concession for the state of Kelantan and the divestiture of the national wastewater utility, IWK. Both of these PPPs proved to be unsustainable financially, largely because of low tariff levels and high bill delinquency (Abdul-Aziz, 2001). The contracts were cancelled and the utilities returned to public control.

Despite these setbacks, Malaysia remained broadly supportive of PPP in the early 2000s in water as in other sectors. Two further concessions were signed in the two most populous and economically active areas of the country: Selangor and the capital region; and the state of Johor. By this time the role of foreign firms had faded, and both contracts were awarded to firms under local ownership. Water utilities throughout the country, including in these two regions, were characterized by low levels of efficiency and severe budget constraints; expectations that the private sector would improve performance were initially high. Ranhill Utilities, the Johor concessionaire, made progress in reducing leakage and improving customer service, but the opposite was true in Selangor, where tensions between the contracting parties led to a freeze in capital investment; efficiency, service quality and rates of coverage growth deteriorated under private management. Meanwhile, payments for bulk water under BOT contracts worsened the financial condition of these and other state water utilities.

Political pressure to reform the water supply sector grew, driven largely by the increasing pressure on the federal government to address the indebtedness of states in relation to water supply infrastructure (Chin, 2008). The government adopted a thoroughgoing programme of reforms, which involved a constitutional amendment in 2005 to transfer responsibility for water supply infrastructure from state government to the federal level, and passed two laws in 2006 to establish new institutions: a national asset holding company (PAAB) and a national regulatory body (SPAN). (These laws do not apply to the states of Sabah and Sarawak.) These reforms established a unique institutional framework for the country’s water sector and created the basis for a form of PPP which ideally would combine incentives for operating efficiency while reducing financing costs.

The new laws provided for all water utility assets to be transferred to PAAB. Existing concession contracts were terminated and replaced with shorter operations-and-maintenance contracts. SPAN took responsibility for setting tariffs and for setting and monitoring performance targets for operators (which may be public or private entities) (Chin, 2008). However, the process of transferring assets from state to federal level has proved very contentious, particularly in Selangor (Teo, 2014). The purchase of the concession assets from the private parties and the transfer of Selangor’s water assets to PAAB was only agreed in mid-2015, seven years after negotiations began. Several other states had still not completed the transfer process at that time.

Thailand also followed a path of opening up to PPPs and closing it off again, although PPPs were never as widely adopted there as they were in Malaysia. Eastern Water Resources, a public bulk water supplier, was listed on the stock exchange in 1992. Three years later, the first BOT contract was awarded to a consortium led by UK company Thames Water. When commissioned, the BOT plant had capacity well above demand, and the contract was renegotiated. In 2005 Thames sold its stake to its local partner. This company subsequently listed on the Bangkok stock exchange as Thai Tap; it has grown steadily and gone on to win other bulk supply contracts.

Soon after that first contract was awarded, the Asian financial crisis unfolded. In the context of the IMF adjustment loan, the Thai government committed to reforming the water sector, including divesting the country’s two water utilities, Metropolitan Waterworks Authority and Provincial Waterworks Authority. Ultimately, however, pressure from the IMF eased and these divestitures were not pursued (Zaki & Amin, 2009). Subsequently, Eastern Water Resources signed a series of bulk supply contracts with PWA. These make up the majority of PPP contracts recorded for Thailand in the data.

The Thai government is broadly supportive of PPP in infrastructure sectors and passed a PPP Act in 2012 in order to promote their adoption. However, there is limited potential for the application of the new law in the water sector. Thailand appears to have reached a limited but stable level of PPP in the water sector through Thai Tap and East Water.

Policy swingers

In the Philippines, support for PPP has waxed, waned, and waxed again. As noted above, the Philippines’ first PPP in the sector was a 50-year lease (known as a franchise) for water supply in part of Angeles City in 1960. Yet PPP for water supply did not gain currency until 1996, when the first concession was awarded, in Subic Bay. This was soon followed by the high-profile award of two concessions for the Manila metropolitan area in 1997. These three PPPs all involved joint ventures between a local partner, which held a majority stake, and a foreign partner.

The award of the two Manila contracts was highly controversial and became more so as the Asian financial crisis took hold in the Philippines, overturning the financial models of the concessionaires and setting in train a long period of dispute and renegotiation. The concession for the western part of the city was finally terminated in 2005, but rather than abandoning PPP the authorities chose to retender the concession and awarded a new contract in 2006 to a consortium of local firms. Under new ownership, the west Manila concession has joined the east concession as one of the few successful examples of PPPs in the region. Both have increased coverage of water supply (and to a lesser extent sewerage), improved labour productivity, significantly reduced non-revenue water, and become profitable.

The Manila concessions are themselves a hybrid model of PPP, combining a concession contract with a regulatory body that has the power to review tariffs periodically according to the principles and process set out in the contract. The regulatory system has turned out to be surprisingly robust given the upheavals of the early years of the concessions, and the concessions have achieved significant increases in coverage and improvements in technical and commercial efficiency.

In the decade that followed the award of the Manila concessions, a few smaller PPPs were awarded, but PPP was not broadly adopted. The election of Benigno Aquino as president in 2010, however, has given renewed impetus to PPPs for infrastructure in general. PPPs are a central source of infrastructure investment, and new institutions have been set up to support them. However, water supply is a responsibility of local government, and as of mid-2015 these lower tiers of government have shown little interest in engaging in water PPP.

The history of PPP in Indonesia’s water sector has been even more tempestuous (Wibowo & Mohamed, 2010). In the 1990s, foreign companies entered the sector, attracted by comparatively high tariffs and the apparent stability of the authoritarian regime. The first contract was for the island of Batam, an industrial development zone close to Singapore, in 1995. This project, now under the ownership of a Singaporean company, Sembcorp, has met the expectations of both parties and remained stable. Several more contracts were signed subsequently, including both small-scale distribution and BOT contracts for bulk supply in major cities. This first phase of PPP culminated in the two large concessions for the Jakarta capital territory in 1998. Foreign companies played a leading role in the design of the contracts; the international financial institutions, on the other hand, distanced themselves from the contracts when it became clear that they would not be awarded through competitive bidding.

The Jakarta contracts were thrown into disarray by the macroeconomic and political shocks that beset Indonesia later that year, and a protracted process of renegotiation began (Jensen, 2005). This has never been satisfactorily concluded, but the contracts have not been terminated either. Performance has been abysmal: new connections have not kept pace with population growth, leakage rates remain extremely high, and the former public utility has become heavily indebted to the concessionaires due to the gap between the tariff paid by consumers and the water charge owed to the concessionaires. In 2015, two court decisions further weakened the PPP: Indonesia’s constitutional court revoked the 2004 Water Resources Law, which provides the legal basis for PPP in the sector; and the Jakarta district court ordered the annulment of the concession contracts, a move which was immediately appealed by the concessionaires and the federal government. The events in Jakarta have discouraged both public and private parties from engaging in other PPPs. Since 1997, a number of smaller-scale PPPs were awarded to local and international companies but no large contracts were signed, despite policy backing for PPP at the federal level.

Like the Philippines and Indonesia, Vietnam has gone through profound swings in its approach to PPP in the water sector over the last two decades. In the late 1990s two BOT contracts in Ho Chi Minh City were developed, both led by foreign companies. A contract for the Thu Duc BOT was signed in 1997, but the foreign venture partners did not manage to raise financing and finally withdrew from the project in 2003. It was taken over by the Ho Chi Minh City municipal utility and developed under traditional procurement. In an interesting twist, a foreign investor later acquired a minority stake in the project company in 2011. The second Ho Chi Minh City BOT project, on the other hand, was awarded successfully to a consortium of Malaysian companies and remains operational.

In 2003, the government reversed its policy on PPP and banned PPPs across all infrastructure sectors. It switched direction again in 2007, lifting the ban on PPPs, and in 2011 brought in further supportive regulations, but no new BOT or concession-type contracts have been awarded since the sector reopened.

In the meantime, however, there has been a slew of partial privatizations of water utility companies through stock exchange listings (see the previous section on mixed ownership structures). An even more striking development has come about through the reform of the sector through water utility corporatization (restructuring municipal utilities as corporate entities fully owned by local government) which has taken place since 2006. The more entrepreneurial municipalities have used these corporate entities to engage in urban development projects, such as property development, which are not related to water supply. In the process, these municipalities have exposed their water utilities to the risks associated with the highly volatile property sector.

The bumpy experience with water PPPs in these three South-East Asian countries is mirrored to some extent in India. Again, the government opened up infrastructure to private finance in the 1990s, and a number of PPPs for bulk supply were developed with the involvement of the World Bank and foreign companies. However, most of these were either cancelled or suspended in the face of mutual mistrust between government, companies and the public (Mahalingam, Devkar, & Kalidindi, 2011) or inadequate institutional arrangements (Van Dijk, 2008). Only one contract negotiated in that period is operational today (Water & Sanitation Program, 2011). These early experiences dampened interest in PPPs on the part of both public and private actors for a decade (Zerah & Renouard, 2014).

Since 2005, public authorities at multiple levels have once again taken an interest in PPPs. This accelerated after 2010 as the central government used conditionalities attached to its urban infrastructure grant funding programme to encourage municipal governments to consider PPP.

A distinctive form of PPP, the ‘24x7 contract’, is gaining popularity in several states. These contracts are designed to address operating inefficiencies and poor service quality, with a particular emphasis on improving continuity of supply and reducing leakages. This contract form combines elements of management, concession and performance contracts. Responsibility for capex is shared between the public and private parties, but certain key decision-making powers are retained by the public party. These projects are in their early stages, and it is too soon to assess their performance.

Stable second-wave adopters

Although it opened up a little later than South-East Asian countries, China quickly became the largest adopter of PPP in the region by both numbers of projects and the value invested. The first PPPs in China were signed in the early 1990s. At that time, private participation was restricted to foreign companies and to bulk water supply, and these early projects mainly took the form of BOTs with guaranteed returns for the investors. While some of the smaller projects from this period have endured, two of the largest foreign-sponsored water treatment projects, Da Chang in Shanghai and Beijing Number 10, were cancelled, and the government subsequently banned guaranteed returns.

Over time, regulations were loosened to allow PPP in water distribution and wastewater treatment and to allow domestic companies to develop projects. As a result, the market grew very rapidly from 2002 onwards, and new forms of PPP developed that were tailored to the local context (Lorrain, 2012; Jang et al., 2014). Local companies rapidly eclipsed foreign companies in the market, and water and wastewater treatment capacity expanded very rapidly in a “golden” age of water policy (Shen, 2014). In part, this was due to the departure of some foreign investors who perceived risks as being too high. It may also be due to the preferential access of partly state-owned firms to information, decision makers and finance. At the same time, the government has introduced policies to promote the indigenous water technology sector, reducing reliance on foreign technologies, including a target under the 12th Five Year Plan for Desalination that 70% of all equipment used in desalination plants be manufactured domestically.

The PPP model most frequently employed in China is the equity joint venture, in which the project company is jointly owned and managed by the private partner and the local government (see the previous section on mixed ownership). In these contracts, the initial investment often takes the form of plant (for brownfield assets) or land (for greenfield assets) from the government side and cash on the private side.

A second distinctive model found in China shares characteristics of a concession but might alternatively be described as a timebound divestiture. In this model, the private party makes an upfront cash payment to acquire an equity stake in a concession company jointly owned with the local government. This model was used for the high-profile Shanghai Pudong PPP, in which French company Veolia acquired a 50% equity stake in 2002. The model was subsequently replicated elsewhere.

The other two second-wave adopters, Korea and Singapore, have been circumspect in their use of PPPs, encouraging only certain types of contracts. In these countries, public utilities are perceived as providing water services of a satisfactory quality at a reasonable tariff. The low proportion of household expenditure on water services weakens citizen-consumer concern for potential efficiency savings that might be generated under private management. Private-sector participation has therefore been seen as appropriate for bulk treatment projects, particularly those involving new technologies, but there has been no push for stronger forms of PPP that would transfer utility control to the private sector.

Korea began to open the water sector to PPP in the 1996 Water Management Comprehensive Plan, and since 1998 a legal framework has been in place for private participation in infrastructure. Under this framework, 40 PPPs were awarded between 2001 and 2014. These mainly take the form of build-transfer-lease contracts and operating contracts for wastewater infrastructure. Under the build-transfer-lease model, there is no transfer of ownership or concession rights to the private company; user charges are collected by the government, and the private party is paid a fee (Kim, Kim, Shin, & Lee, 2011). In addition, private operating contracts for wastewater facilities are widespread. In 2012, 68% of the country’s wastewater treatment plants were operated by the private sector (Korea Water & Wastewater Association, 2012). In contrast to wastewater, water supply remains entirely in government hands. The market is dominated by indigenous firms: 10% of the total number of water PPP contracts were awarded to foreign companies, and no PPP contracts have been awarded to foreign companies since 2004.

Singapore has a strong and effective public water utility, but in keeping with the government’s PPP scheme introduced in 2004, major public works projects with a value of over SGD 50 million should be considered for PPP. As a result, PPP structures have been adopted for bulk treatment projects using advanced reuse and desalination technologies (Tortajada, 2006). At the same time, Singapore has actively promoted itself as a ‘water technology hub’ and has provided incentives and support for both foreign and local companies to engage in water-related research and development in Singapore. Singapore is also putting in place financial and in-kind support to Singaporean infrastructure companies seeking to develop projects abroad.

Reluctant adopters

In Japan, private-sector activity in the water sector has been limited to short-term operating contracts. Since 2010, a small number of build-transfer-operate projects have also been awarded for stand-alone infrastructure assets like sludge treatment plants. Japan’s PPP law, introduced in 1999 and amended in 2011, allows for concession contracts for water services, but these provisions have not yet been used. Instead, the Japanese government has focused its efforts on supporting its domestic companies to expand abroad, rather than liberalizing the domestic market. They provide support for financial investors, contractors and equipment suppliers. Attitudes to PPP may change in the future as municipalities find themselves faced with high investment requirements to replace ageing assets, coupled with strict budget constraints.

Taiwan’s experience with PPPs has been somewhat similar to Japan’s, in that regulations are in place that allow for PPPs in wastewater treatment and sewerage but few contracts have actually been signed. Private companies have been discouraged by low fees and high investment requirements (Zheng & Tiong, 2010), and one contract was terminated by the investor before it became operational.

Early-stage reformers

Cambodia, Lao and Myanmar, three of the poorest economies in the region, have no large PPP contracts. The water sector of Cambodia is best known for the impressive reforms effected by the PPWSA, a public utility (Biswas & Tortajada, 2010; Chan, 2009). The PPWSA made an initial public offering on the country’s stock exchange in 2012.

In smaller cities and towns outside Phnom Penh, there are considerable needs for operational and managerial expertise and for investment finance, but low levels of ability or willingness to pay are a constraint on the development of formal PPP (Chan, 2014). However, local private companies play an active role in providing water services under small-scale, informal arrangements (Basani, Reilly, & Isham, 2004). The government is now making efforts to bring these arrangements under a regulatory regime with standardization of procurement, quality standards, monitoring, tariff-setting, etc.

In Lao, the expansion of water supply systems in the larger towns has been financed by donors or international financial institutions. There are no examples of formal contract-based PPPs in Lao, but as in Cambodia there are examples of private contractors negotiating with a community to build and operate small distribution systems. The absence of PPPs in the water sector does not imply that the government is resistant to PPPs in general. In fact, Lao has been successful in attracting foreign private investment for renewable energy projects, in particular hydropower.

Myanmar has been a potential location for PPPs since the country emerged from military dictatorship in 2011. The government is open to private-sector involvement as a way to meet the country’s very significant infrastructure needs. In principle, water supply and sanitation are eligible for PPP, but they do not appear to be a high priority for the government, except insofar as they form part of industrial zone and urban infrastructure development projects. However, the legal and regulatory environment for foreign or local private-sector participation is in the very early stages of development.

In Nepal, debates over the role of the private sector in water supply have centred on a particular project, Melamchi water supply (Chapagain, 2014). The project was first conceived as a concession in the late 1990s to increase the supply of water in the Kathmandu Valley. Its development was led by the World Bank. As a result of political instability, scepticism on the part of government and high perceptions of risk from potential private partners, it proved impossible to pursue a concession or lease, and the original funding institutions withdrew. The Asian Development Bank (ADB) subsequently stepped in in 2000 and tried to structure a management contract, which also failed to attract private-sector interest. The project was restructured once more in 2008 with funds from the ADB and Japanese aid agency JICA, and a four-year management contract was finally awarded to a foreign company in 2010 (Asian Development Bank, 2014).

Bangladesh is open to PPPs in the water sector in principle, but no project has been signed so far. A PPP model is being considered for a water treatment plant in Dhaka which is being developed with the financial support of the Asian Development Bank (2013). Sri Lanka is considering pilot PPPs in the water sector, but there have been no concrete developments in awarding projects (Global Water Intelligence, 2014b).

Strikingly, it is several of these these early-stage reformers which have taken initial steps towards establishing economic regulation, at least on paper. For example, even though Nepal has not adopted PPP in any significant way, it has established a semi-autonomous commission which has the power to set tariffs. In practice, however, the agency lacks the capacity to carry out its functions (Chapagain, 2014). In Sri Lanka, a multi-sector regulator, the Public Utilities Commission, was established with the intention that it would regulate the water sector in addition to electricity. However, supporting legislation that would allow it to take on this role has not been adopted. In Cambodia, regulatory reforms are under consideration, but so far there has been no economic regulation of the water sector in the country (Chan, 2014).

This section has shown the very different approaches to PPP across Asian countries. Throughout the region, there is evidence of PPP’s failing to meet the expectations of governments and private parties. Governments have responded to problems with PPP contracts by engaging in pilots and policy experiments, and private companies have also been active in proposing new modes of participation, resulting in a wide variety of project structures across the region. As a result, second and third generations of PPPs in the region may exhibit learning and prove to be more robust than the PPPs of the 1990s. However, this wide variety and extensive experimentation may signify that opportunities for learning across national borders have been missed. Furthermore, the experience of countries with long track records in PPP in the water sector, like the US, UK, France and Spain, in integrating transparency, public scrutiny and mechanisms of adaptability into PPP frameworks does not seem to have been absorbed into Asian PPP policy.

Conclusions

This article has shown that far from being an abandoned policy experiment of the 1990s, PPPs are widely and increasingly employed in the water sector in Asia. The stock of PPP contracts is rising annually, as is the number of countries that encourage PPPs in the water sector. Countries that appeared set against water PPP in the early 2000s are once again experimenting with modes and models. However, countries’ experience with PPP has not been linear, and it is possible that more policy reversals may be seen in the future.

The Asian experience of water PPP has much in common with the patterns of PPP in other regions. As elsewhere, water and sanitation PPP has been more limited and more politically fraught than in other sectors like telecoms and roads. This is the result of both a weak push from companies, which are discouraged by generally low levels of profitability in the sector, and strong resistance to private involvement in water supply amongst citizen-consumers, the political parties that represent them and specific interest groups (Menard, 2013).

As in other regions, the inherent tensions in PPPs between the flexibility needed to adjust to a changing environment and procedural rigidities necessary for monitoring and ensuring accountability of an essential public service have caused projects to fail or never come to fruition and to generate cycles of supportive and restrictive public policy for PPP. Although no surveys of Asian public opinion exist that parallel those conducted in Latin America, the portrayal of PPPs in the regional media and politicians’ statements suggest that the public is highly sceptical of water PPP. Public perceptions seem often to be driven by a single high-profile case, like Manila or Jakarta, even when other contracts in the same country have a different performance record.

Politics clearly plays a role in switching between supportive and critical attitudes to PPP. However, we do not see evidence in Asia of a straightforward ideological split between more left-wing governments opposing PPP and more pro-business governments promoting it. Instead, populist politicians and parties of all complexions have found electoral advantage in criticizing and opposing PPPs signed by their predecessors.

There is abundant evidence for the increasingly important role of local companies and mixed-ownership companies in Asian water PPPs. This parallels the ‘empresa mixta’ model in Latin America (Blanc & Botton, 2012; Castro & Janssens, 2012; Marin, 2009). This model combines mixed public (municipal) and private company ownership with a long-term (25 years or more) management contract awarded to the private party. These companies have begun to engage in neighbouring markets, and in the future these intra-regional transactions may become more common.

In contrast to Latin America, however, the development of institutions of economic regulation has lagged behind the introduction of PPP. This is particularly striking in China, where thousands of PPPs are monitored by local government officials under simple contracts. With the exception of Manila, there are no functioning economic regulators in Asia and certainly none that can match the sophistication of Chile’s regulatory agency (Castro & Janssens, 2012). In Latin America, economic regulation is more widespread than PPP (Foster, 2005), and half of the large economies in the region have an economic regulatory agency which either sets tariffs directly or sets a tariff formula and oversees its implementation.

A further distinguishing characteristic of Asia is that the water sector is a target of industrial and technology policy. This is clearest in Singapore and China, where the government has set specific targets for the growth of the local water technology sector, but it is also true of Japan and Korea, where the trading houses and engineering companies, respectively, receive support from national development banks to engage in international water projects.

A related feature of PPP in the region is that water has actually been a leading sector in the shift from planned to market economies in China, Vietnam and Cambodia. This is partly because water utilities are under the purview of local governments, and local governments in these countries have had greater scope for policy innovation and entrepreneurialism than their central counterparts. The most striking example of the role of a water company in the development of the market economy is the PPWSA’s baptism of Cambodia’s stock exchange. These phenomena are closely bound up with the prevalence of mixed-ownership companies.

Overall, we see little evidence of policy transfer between Asian countries. Contract or regulatory models do not seem to have been adopted across borders, and there is little sign of contact between policy makers or officials in this policy domain. Even companies play a limited role in transferring experience and expertise across markets, as few are active outside their home markets. With few examples of resounding PPP success in the region, and with the reluctance of the region’s rich countries to engage in PPPs in the water sector, this is perhaps not so surprising. Instead, countries are formulating their own models based on national priorities and experience. This probably leads to duplication and high administrative costs but may make the resulting PPPs less vulnerable to the suspicion often directed toward foreign involvement in this sector.

The sketch of PPP provided in this article suggests several promising avenues for further research. Firstly, it would be interesting to compare the behaviour of private and mixed-ownership companies. There are a number of reasons why one might expect companies with partial state ownership to behave differently from fully privately owned firms:

•	Hybrid companies might have a lower perception of political or regulatory risk or more effective ways to manage these risks, and thus be willing to accept a lower return on investment.
•	Hybrid companies might have easier and cheaper access to local-currency finance, thus mitigating financing and currency risk.
•	Hybrid companies may have a better understanding of social and cultural factors, making them better at managing key stakeholders like government agencies and employees.
•	Hybrid companies might benefit from more positive public perceptions among stakeholders, independent of their actual performance.

On the other hand, hybrid companies may simply internalize inherent tensions caused by incompatible objectives of the public and private players, leading to paralysis or conflict within the organization. Their special nature may weaken mechanisms for oversight and dispute resolution. This may lead to reduced transparency and weaker governance overall, more opportunities for corruption and collusion, and as a result increased scepticism on the part of the public.

A second area for further research could be a comparison of governance and regulation of PPP in Asia and Latin America, given the different joint trajectories of PPP and regulation in these two regions. A third area of study would compare the water sector with other utility sectors. There is little evidence of the development of autonomous regulatory agencies in the water sector, but it is possible that this is a function of a sector-specific policy preference and that Asian countries are establishing autonomous regulatory agencies in electricity, gas or other sectors, so this could provide an interesting comparison.

Supplemental data

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/07900627.2015.1121136

Disclosure statement

No potential conflict of interest was reported by the author.