Institutional Resilience for Collaborative Governance:
The Case of Water Allocation in Taiwan
Water Crisis and the Rise of Cross-sector Collaboration
Water scarcity has been a prominent crisis in different corners of the world (Solomon, 2010; Saleth and Dinar, 2004). This shortage requests better efficiency in satisfying the given demands with less amount of available water as well as in allocating water to higher value-added economic activities to improve overall welfare. While “market” is widely recognized as the most effective governing mechanism to deal with the efficiency issue, governing water needs more than voluntary transactions in the market mainly because in different scenario water can be considered as different types of good that demand for different mechanisms.
When it is rain dropping from the air, it is largely a public good free for everybody to appropriate. In this case, governmental intervention seems proper in managing this public good. When it gets down to the surface, runs into rivers or lakes, and is contained in such infrastructure as embankments or canals, it is more close to a common good open to any users that might not share the costs of providing the good. According to Ostrom and others, networks woven by grassroots organizations could be a better choice to govern the commons. When the water is collected, treated, and bottled or piped to be transferred to consumers, it becomes a typical commercial commodity that can be effectively managed by the market. In other words, water with features across different categories of property needs to combine different mechanisms and to involve different sectors to govern effectively.
Externalities associated with water transactions should also be taken into account. Price variance in different localities encourages suppliers to transfer water for higher profit. Transferring water, however, may cause such externalities as ecological upset. In addition to consumer satisfaction, governing water needs also to pursue such public interests as food security, land conservation, and flood control, none of which can be managed by market mechanism alone.
Justice is also a concern that prevents the market alone from being an adequate mechanism in managing water. Water is arguably the most fundamental need of human being. Equitable access to water is therefore considered a basic right for everyone. As the problem of shortage in supply looms large, it is usually the underclass that suffers most. The poor tends to be less capable of paying rising utility bills, less possible in finding alternative means to make a living, and less powerful in property rights renegotiation that sets the foundation for a new market order. To assure that the powerless stakeholders can have their interests represented in the governing system, a participatory management, or co-management, with collective or communal tenure and property rights, has been an attractive alternative adopted in many places (Berry and Mollard, 2010).
The limits of above network-based participatory governing regime, however, have been widely
recognized after decades of empirical studies. Such governing system can be successful when it is applied to a small geographical area with clear community boundaries, relatively homogeneous population, a match of jurisdictional and hydrological boundaries (Heikkila, 2004 JPAM), and authorized autonomy to control over the resource (Tang and Tang, 2001). Communities are usually not in a proper scale to provide expensive infrastructure to control flood, and to story and convey water to dispersed users. Even though these communities are successful, they achieve only parochial justice that has very limited impacts on just governance from a regional or nationwide perspective.
A corollary is that effective governance of water needs to involve mandatory power in public authorities, market mechanism in the private sector, and networked collective actions organized by grassroots
organizations or communities known as the voluntary sector. World Bank, for example, advocates a
public-private relationship that consists of strengthening state institutions in playing central regulatory role on the one side, and establishing grassroots institutions with river basin based organizations and introducing economic incentives for facilitating efficient transfers on the other (World Bank, 1999).
Institutional Choice for Governing Thirst
Many innovated institutions have been experimented and documented in both western and developing countries to engage different sectors into collaborative water governance (Saleth and Dinar, 2004; Bakker, 2010). While “integrated approach” becomes a fancy term nowadays (Mollinga, 2008), the possible
conflicts in such integration have largely been overlooked. Since every mechanism relies on its own incentive scheme to gain cooperation from the governed, simply combine them may result in incentive conflicts, according to accumulated literature on cognitive economics in recent years (Frey, 1997). General speaking, concrete, instant rewards or punishments tend to overwhelm intangible, long-term incentives. If both incentives have been applied to encourage some kind of behavior, the former will crowd out the later in the long run.
In governing water, the tasks can largely be divided into two categories, water development and water allocation. The development task is easier to have integration and enjoy a synergic effect from
cross-sectoral cooperation because different governing participants could have their own competitive advantages in providing inputs in production complementarily. Participants from different sectors share the task basing on respective incentives with very limited mutual interference. In water allocation task, however, integrated governance is much more difficulty the governing partners become players in a zero-sum game and main incentives behind governing participants might not work in a complementary manner. In such a scenario, “competition” becomes an underpinning setting that encourages calculation and bargaining among players. Market tends to be more effective governing mechanism. Under some conditions network might also function well in this regard in the individual level, such as user’s associations in the rural places that exercise solidary and moral incentives to promise effective governance.
Nevertheless, when it comes to allocating water in big chunk, such as allocation among regions (e.g.
upstream versus downstream), or among sectors (e.g., agricultural, industrial, or municipal use), effective
governance usually involve redefinition of water rights and thus becomes political and contentious. The function of networks, which rely heavily on norms and traditions, incline to diminish.
In many countries experiencing rapid industrialization and urbanization, demands for municipal and industrial water tends to grow rapidly. Although the contribution to GDP might have diminished drastically, agricultural sector still controls an overwhelming share of water as before. This inter-sectoral imbalance results in a legitimate appeal to reallocate the water resources across user sectors, or even to redefine the water rights.
A notable practice in managing the reallocation task is to allow agricultural sectors to withhold a major portion of water but in the meantime to build up an institutional framework to promise more flexible transfer among user sectors. One obvious reason is the historical legacy and the political feasibility to change water rights. Although farmers might geographically widely dispersed and thus hard to mobilize political support according to Olson, yet struggle in preventing water from being deprived is a life-and-death matter that tends to attract engagement from farmers. Farmers are also easy to mobilize because of the grassroots networks prevailing in rural society.
The second reason for holding more water in agricultural sector is the increasing uncertainty of precipitation together with the cost of containing the water. Since building up water storage facility is rather expensive, it is reasonable have the capacity just enough for surviving the dry season. The excess water can be drained directly to the ocean through rivers. Alternatively, the water can be spread to the agricultural fields so that a part of it can infiltrate into underground aquifers to become a back-up source for water supply. In addition, such facility as irrigation canals can help drain the flood in rain season if there water managers in agriculture sector can be capable enough to construct and maintain a thick network of waterways.
Given that water is held in agriculture sector, a reallocation process is therefore needed for
cross-sectoral transfer to maximize the overall welfare in a drought. This reallocation needs to involve different mechanisms to meet the goals of satisfying different water users. Market is usually the most effective mechanism in which compensation can be provided to motivate voluntary water transfer.
Nevertheless, the operation of market requires well-defined property right, which might not be easy to be achieved in many countries. In addition, market also demands for information on both providers and buyers, which is usually hidden intentionally by both side. Further, there is usually no perfect market on water transaction. Only very few players will involve. Therefore, governmental intervention is usually inevitable. Finally, networks operated by grassroots users’ associations can be an essential mechanism in determining how much water could be saved and thus available for transfer. Since agricultural demand for water is rather flexible because the farmers can decide if the water-saving crops are to be grown, or if more troublesome but water-saving irrigation practices to be adopted, users’ associations can help promote water saving efforts to promise a successful transfer transaction.
While all these mechanisms are required, they may not present naturally and work complementarily as
many policy makers have assumed. Players in these mechanisms have respective incentive schemes. To motivate them to engage in the transfer, there should be institutional arrangements that provide right and mutually supporting incentives. How these incentives might interact and what factors have contributed to their effectiveness are questions to be answered in a more comprehensive manner. Two cases in Taiwan are examined to illustrate the answers on above questions.
Methodology
The Cases: Politics of Collaborative Water Governance in Taiwan
Like many countries, Taiwan has faced the challenge of cross-sectoral reallocation of water because of rapid industrialization and urbanization, although its governance seemed pretty effective in the past.
Water is governed by three major agencies in Taiwan. First, Water Resources Agencies under Ministry of Economic Affairs is the competent authority in supervising flood control and provision of water to all sectors of users. Under which, three regional offices (north, central, and south, hereafter WOs) make practical decisions in the fields to adjust to idiosyncratic situations. Since the water resources belong to the island, it has the authority, at least nominally, to allocate water to the most needed users when the supply is in short.
Nevertheless, trying to reallocate water usually invites intensive protests and thus need to be managed collaboratively with two other governing partners.
Second, 17 federated Irrigation Associations (hereafter IA) in different localities are the main governing entities of agricultural water. They are parastatal organizations in terms of receiving heavy subsidy from the government, and their exercise of such functions as facilities construction and distributing water to farmers according to irrigation plans. On the other, they enjoy full autonomy in disposing their huge property assets, withholding their income, and run their own organizations with full autonomy as a private company.
They usually have a tacit agreement with the WO regarding the amount of water they have. Although in WO has the right to change the amount, several reasons have forced the government to negotiate the deal with IAs. The most obvious one is that IAs owned very thick networks at the grassroots level and thus can either mobilize protests or help pacify rising turmoil. Another reason is that these IAs own a lot of facilities that can control the water physically. A third, and probably the most important reason is that the
government relies on these IAs have collect information about the actual water available and needed in the draught, and need them to carry out water saving plans once the reallocation agreement is reached.
Taiwan’s water management system has perfectly presented what World Bank suggests, which is composed with a parastatal design of irrigation system, including totally 17 Irrigation Associations (IA) around this island, which owns more than 70% water rights of this resource but has accounted for less than 3% of Taiwan’s GDP since the early 1980s, several state operated companies such as Taiwan Water Company (TWC), and on the other side, the central administration assigns agents to local districts on a basin basis, stationing them throughout the territory and coordinating these decentralized departments and authorities.
Some other market-oriented mechanisms, such as transfer complements, were also introduced recently in order to increase efficiency regarding water resource reallocation. This country also encounters fairly fast industrialization, urbanization, as well as water shortages and conflicts in recent years.
Considering the complexity water management, how can market or the state mechanisms work? In what conditions it cannot work? How can a hybrid system of institutions which the World Bank suggests work? And in what conditions it would not reach expectation? To answer these questions, we develop a two-stages research strategy: first we provide a static game model which points out the overall structure of water reallocation and informational asymmetric, and point out the existing multiple equilibriums. Second, we adopt an evolutionary perspective by the process tracing method which identifies the different
trajectories taking into place in specific context where actors adapted to the contexts by strategic responses and then the system as a whole evolves over time. The main argument is neither privatization nor
government intervention can guarantee a solution of effective water reallocation; sometimes, introducing these mechanisms may even harm the coordination relations between the governments and IAs that they can address on through mutual trust and reciprocity.
The nature of water management in Taiwan characterizes as decentralization in both administrative and political forms, constituted by IAs, where politically the rights to make policy decisions and choose leaders have been devolved to local users’ communities, WROs (water regulatory organizations, who are branches of the central government assigned to different regions of Taiwan and draft different intervention strategies and policies to cope with their distinctive industrial needs and superior pressures), and several other state operated companies, such as TWC.
Similar to other developing countries, the Water Law in Taiwan limits water rights only in beneficial-use provision, rather than an absolute ownership. In this regard, water resource belongs to the country,
alienated from the ownership of lands. Water rights are issued by government, restricted in specific business purposes, and reviewed regularly by government. Once the water use is deviated from the original purpose, government may reallocate the original water distribution. Furthermore, even though people can hold water rights in a beneficial-use provision, the state still nominally claims the ownership of water resource,
therefore any kind of water transaction can be seen as illegal.
This legal structure encountered severe challenges when Taiwan experienced political democratization after 1990s. The state’s power has kept decreasing due to party competition and the trend of
decentralization. On the other hand, IAs have become play more important role in local politics due to their grassroots organizations. The changing power structure makes it more difficult for the governments to intervene water transfer; on the contrary, a marketized complementary mechanism became necessary to create incentives for the IAs to release extra irrigation water. In early 2000, the official document “Water Distribution and Allocation Keynotes”（農業用水調度使用協調作業要點）has took into place, which formally admits IAs can be rewarded through complementary fees if they transfer extra water for other users.
Information, Bureaucratic Discretion, and Complementary Mechanisms
Although the governments intend to regain control of the water resource and reallocate to other users, the distribution of water rights between IAs and governments is not necessarily resulting in conflicts. In contrast, since the uncertainties and flexibilities irrigation management is due to the involvement of
thousands of people and grassroots organizations,1 the IAs and the governments may coordinately transfer water between them and fulfill both needs. Given the uncertainties (that to some extent depend upon human behaviors), the best scenario is: 1) IAs first to reveal actual information regarding the possible range of water consuming, 2) the governments then bestow IAs with full authorities to manage this abundant water, 3) the IAs consume the water as frugal as possible through their well-organized local entities, 4) once finishing irrigation, they save as much as water returning back to governments for other purposes. Thus, both the IAs and governments can control the water at some stages and attain their goals eventually. The difficulties, however, happen when it requires transferring the rights of water use back and forth between the governments and IAs; each one has incentive to betray counterparts as well as breaking the rules.
Obviously, there is a mutual trust problem subsumed this situation that once the governments empower IAs with the right of water control, the IAs might not seek to use the water efficiently and hence have no water returning to the governments. On the other side, if the IAs revealed the real information of water, including the range of water demands between the minimal and maximal principles, the governments may choose to stick with the minimal principle and leave no flexibilities for IAs to manage water in their ways.
To explore the water information asymmetric and control dilemma, consider the following simple game.
Let us assume the IAs hold sufficient information regarding how much water they potentially have and how much water they are going to consume. The IAs must decide whether to reveal the real water information such as the estimation of water consuming, R, or instead reporting just superficial and technically calculated information (hide the real information), ~R. The governments collect information from the IAs as well as other sources and must then decide to reissue water rights and reallocate water following the minimal principle, M, or not following the minimal principle, ~M. The IAs finally choose whether to devote themselves investing in water management, I, or not to invest (or compliance), ~I. Since the water information varies every year, we assume this game as a one-shoot game.
Figure 1 about here
Controlling extra water 𝐶 is valuable for both players. For IAs, if they are able to control extra water, 𝐶𝐼𝐴, the less operation cost they pay to manage local grassroots organizations (ex. personnel costs,
monitoring costs, solving conflicts costs…etc.). On the other hand, the governments need to control the water as well, 𝐶𝐺, because they are in charge of water allocation and coordination, so that if they do not
1 Regardless the possible variations on crops planted each land each year, when the local farmers are organized loosely, they would consume much water than expectation; on the contrary, if people can be organized well, the water can be used in a more efficient sense.
have extra water to manage, they may likely fail to finish their jobs and also be blamed by the supervision, sometimes, may also cause severe political crisis.
The situation is not a zero-sum game, though. As we mentioned above, the governments can allow the
The situation is not a zero-sum game, though. As we mentioned above, the governments can allow the