"The Pitfalls of Irrigation Water Pricing in Kyrgyzstan and Tajikistan"

Palgrave McMillan / Development / Local/Global Encounters: http://www.palgrave-journals.com/development/journal/v51/n1/full/1100452a.html

Development (2008) 51, 130134. doi:10.1057/palgrave.development.1100452

The Pitfalls of Irrigation Water Pricing in Kyrgyzstan and Tajikistan

by Jenniver Sehring


Irrigation service fees (ISF) are often introduced in agricultural water usage, although not always with great success. Jenniver Sehring describes the reasons for their failure in Kyrgyzstan and Tajikistan.


irrigation reform, agriculture, Central Asia, implementation research


Irrigation service fees (ISF) are often considered an adequate tool to ensure efficient and equitable water usage in agriculture. However, there is also a wide international debate on their economic and social sense. For policymakers, the impulse for introducing water pricing often arises from a crisis in state budget: water fees should help to cover the costs the state has spent for operation and maintenance of the hydro-technical systems and facilities that are its responsibility. The term 'irrigation service fee' (ISF) clarifies that the fee is not on the usage on water as such, but on the service (by the state) to operate and maintain the infrastructure necessary for the water transport and to deliver the water timely to the place needed.

Worldwide experience to date has shown that ISF has rarely been successfully implemented (Azevedo and Baltar, 2005; Hellegers and Perry, 2006). This is also the case for Kyrgyzstan and Tajikistan, where fees did not lead to more efficient water usage and more cost recovery but rather increased inequity.1

Decision and implementation of ISF in Kyrgyzstan and Tajikistan

In the two Central Asian states Kyrgyzstan and Tajikistan arable land is very limited due to the mountainous relief. Agriculture employs 40 percent of the workforce in Kyrgyzstan and 6570 percent in Tajikistan. Owing to the dry and continental climate, agriculture is highly dependent on irrigation. Irrigated agriculture in Tajikistan accounts for 84 percent and in Kyrgyzstan for 90 percent of the total water use (Sehring, 2007).

When inside the Soviet Union, the water sector in Kyrgyzstan and Tajikistan was part of the centrally managed state economy. Allocation schemes were decided according to production plans. After independence in 1991, both countries experienced a severe economic crisis and budget allocations to the water sector were reduced dramatically: in Kyrgyzstan, in the last years of the Soviet Union, budget allocations were above US$ 35 million. In 1999, they were about US$ 5 million (Bucknall et al., 2003: 4). In Tajikistan, allocated financial means declined from US$ 72 million in 1991 to US$ 6.5 million in 2002, that is, by more than 90 percent (UNDP, 2003: 33). Additionally, in both countries, water use in irrigation is extremely high. Central Asia has the lowest water use efficiency worldwide (UNDP, 2006: 8). For example, in Tajikistan, water consumption for cotton cultivation is about 70 percent higher than in Pakistan (UNDP, 2003: 26).

Both countries in the 1990s introduced volumetric ISF. In Kyrgyzstan, water fees had already been introduced formally in 1995. However, due to the enduring refusal of the Parliament to approve a law on the height of ISF, a cost-recovering tariff could not be decided on and the final water tariffs for agriculture were only established in 1999. This fee is rather symbolic. It covers only about 20 percent of the actual O&M costs. Nevertheless, the state reduced its apportionment of funds to 50 percent of the expenditures of the RaiVodKhozes, the water agencies at the district level. The other 50 percent were now to be covered with ISF. The fees have not been increased since then as respective laws have been rejected by the Parliament (Dzhajlobaev, 2003: 6970). With the 2005 Water Code, the authority to determine the height of the ISF was transferred to the Government.

In Tajikistan, volumetric ISF was introduced in 1996 by Presidential decree. In the ISF proposal of the Ministry of Irrigation and Water Management, the real costs were taken as the basis for the calculation of a water fee. The government, however, decided on a reduced level of fees. The objective was to create awareness first and then strive gradually to full cost recovery. Consequently, the height of ISF was increased gradually. The current level can cover about 30 percent of the actual costs (Kholmatov, 2003: 153). Local water agencies are, since 1996, expected to cover part of their costs by ISF.

Although the decision to establish ISF was made more than ten years ago in both countries, their implementation, collection of water fees from the farmers, is far from being fully realized. In Kyrgyzstan, country-wide water user associations (WUAs) have been established that are now in charge of collecting ISF from their members and transferring it to the RaiVodKhozes. Alymbaeva (2004: 11) shows an average collection rate in WUAs of 53 percent. As a consequence, many WUAs have debts of non-paid fees at the RaiVodKhozes, reaching together US$ 1 million already in December 2003. In areas without WUAs, the collection rate is even less as the RaiVodKhozes do not have the staff capacities for individual collection from the farmers.

For Tajikistan, estimations by different officials for the years 2002/2003 range from 30 to 56 percent. The 2006 Water Sector Development Strategy of the government mentions a 60 percent collection rate (MIWM, UNDP and EC-IFAS, 2006: 18). It cannot be stated with certainty as to whether payment is better in those areas where WUAs have been established and are responsible for fee collection as no survey data exist and expert statements were contradictory.2

In addition, in most places, it is currently technically impossible to measure the amount of water delivered to individual farmers due to widely non-existent measuring facilities at the former on-farm channels of the state and collective farms, sometimes not even to the area of a whole former state or a collective farm (sovkhoz or kolkhoz) or WUA. In Tajikistan, measurement exists only at about one-third of the water delivery points to the former state and collective farms (MIWM, UNDP and EC-IFAS, 2006: 18). In fact, in most places, at the moment, only a quasi-volumetric charging is applied with fees based on estimations built on the area of land and crop cultivated (DFID and Mott MacDonald, 2003: 1121; Ul Hassan, et al., 2004: 10).

Lack of transparency and accountability

One reason for non-payment is that many farmers are not aware of the need for ISF. It is hardly explained to them that they do not have to pay for water as a resource (which is perceived as a gift of Allah in Islam, and therefore people resist payment for it), but for the service of irrigation water delivery. In many villages, knowledge about ISF was very limited and specifications on the amount they paid varied among villagers. At some WUAs, council members or even the director were unsure as to how much ISF the members had to pay. Often, local water agencies have difficulties with their new role of being accountable not upwards but to the users, what would mean a guaranteed water delivery to those who paid ISF. This is not ensured. And, water users are not used to claiming their rights. A right to water is often not perceived as being linked to the payment of ISF. This is reinforced by the fact that a codified water right does not exist in Tajikistan and was established in Kyrgyzstan only in 2005. On the other hand, non-payment of ISF in most cases is not followed by suspension of water delivery. Consequently, little incentives exist to pay on time.

The introduction of ISF requires sound information and awareness-raising components in advance. Indeed, when accountability is not ensured, ISF reform faces resistance from two sides: from large and powerful water users, who gain from the current situation and therefore have no interest in any change, and from the small water users, who are 'understandably reluctant to support a change which brings the certainty of higher cash payments combined with less certain promises of better services and higher incomes' (Azevedo and Baltar, 2005: 24).

Institutional obstacles in agriculture

The economic and institutional conditions of the agricultural sector are the main obstacles against the implementation of ISF. Both countries conducted a land reform in the 1990s. While state and collective farms in Kyrgyzstan were completely privatized, in Tajikistan, despite formal privatization, state production plans (especially for cotton) are still in force and many collective farms continue to exist. The consequences for the rural population in both cases are, however, similar: general poverty and widespread subsistence production. The agricultural economy is for a considerable part a barter economy with little cash transfer. The agrarian sector is virtually 'de-capitalized' (DFID and Mott MacDonald, 2003: 1019). In both countries, farmers possess neither the necessary means nor the necessary knowledge for lucrative agriculture. Many farmers are simply too poor to pay ISF. Those conditions make it difficult to introduce monetary mechanisms.

One consequence of the agricultural barter economy is that it is expanded to water management: in acknowledgement of the economic situation, farmers have officially been allowed to pay 30 percent of the ISF in kind. In fact, however, in Kyrgyzstan, between 50 and 80 percent, and in Tajikistan, about 70 percent of payments are estimated to be made in kind. In-kind payments are made through natural products or work. In Kyrgyzstan, some WUAs work the debts off by cleaning works at channels in exchange for water.

Besides these economic conditions, informal power structures also have an impact. In Kyrgyzstan, during land reform, the elite of the former state or collective farm could ensure the best land plots close to the channel for themselves and their networks. Marginalized villagers or newly migrated ones got plots downstream, where water often does not reach the fields due to the poor-quality infrastructure also when ISF is paid. Hence, the chance of access to water is already decided and is independent of ISF payment.

In Tajikistan, privatization occurred on paper, but in practice the former kolkhozes are often still existent and cultivation is prescribed by the state for a considerable percentage of land. Many farmers are not aware that they are supposed to be 'independent' now. Under these circumstances, it is obvious that there is no perception of own responsibility for channel maintenance or cost recovery as these are considered as still being part of the kolkhoz. Additionally, farmers cannot switch to less water-intense crops, as cotton cultivation is prescribed and adherence to these orders is a precondition for receiving necessary inputs merely water. Hence, financial incentives to economize water cannot work.

Alternative modes of resource access

If ISF payment is no guarantee for water delivery, and non-payment no reason for non-delivery, what then determines critical access to water? In practice, local patronage structures deal with water management. In Kyrgyzstan, WUAs are dominated by the local elite. Cases of favoured water distribution to the network of the WUA chair, for example, have been mentioned. As this network often has to good located fields, water access is already predefined. In Tajikistan, resource access is often connected to cultivation of prescribed crops, such as cotton and tobacco.

Finally, unauthorized water withdrawal is common and usually not sanctioned. While government officials in both countries condemn 'water theft' in rhetoric, at the local level it seems to be tolerated at least in the case studies, local officials showed an understanding for the farmers, which in their situation would have no other choice. Bichsel (2006: 87) also noted an understanding even by downstream farmers who suffer from unauthorized withdrawals. It hence represents a tolerated strategy.


Due to the reduction of state subsidies, non-paid fees, and transaction costs of in-kind payments, the RaiVodKhozes hence do not have the means to improve their services, for example, reduce water losses or ensure reliability of water delivery by technical improvements. Hence, ISF do not confer any short-term, visible benefits to farmers. On the contrary, they also have to pay for the water losses. The question that has to be answered is not why users do not pay fees, but why users pay fees at all as water delivery is not guaranteed in either of the cases. To explain why farmers pay ISF, it seems more valuable to refer not to incentives but to local power structures: when benefits are not an incentive to pay ISF, then payment can only be explained by the authority of the one who demands payment. WUA chairs and directors, often former cadres of the state or collective farms, still hold a powerful position and hence the authority to enforce payment. Hence, ISF becomes part of the unequal patronage relations instead of a means of a right to water.


In Kyrgyzstan and Tajikistan, ISFs have failed to fulfil their objectives due to technical reasons (lack of measurement facilities), economic reasons (poverty), incentive reasons (no perceived benefits) and awareness reasons (no explanation). Current policies have actually worsened irrigation as systems have deteriorated further and unauthorized water withdrawal have increased, which also had a negative impact on cost recovery. As many farmers simply lack the agricultural knowledge on how much water is needed for specific crops, training would be a more appropriate measure to achieve more efficiency than ISF. Hence, whatever one thinks of ISF theoretically, in practice, they have proved not to be a feasible tool to achieve the objectives of efficiency and cost recovery when the necessary institutional and economic conditions are not in place.


1 This article is based on field research that was funded by two grants of the German Academic Exchange Service (DAAD) and conducted in 2003, 2004 and 2005. Empirical research used qualitative methods such as semi- and unstructured interviews, participant observation and group discussions. In order to enhance readability, individual interview sources are not indicated in the text. Detailed information on methods and analysis can be found in Sehring (2007).

2 WUAs exist in Tajikistan only to a considerably smaller extent than in Kyrgyzstan