Exploring Synergies Between Hardware and Software Interventions on Water Savings in China: Farmers’ Response to Water Usage and Crop Production
- First Online:
- Cite this article as:
- Mushtaq, S. Water Resour Manage (2012) 26: 3285. doi:10.1007/s11269-012-0072-7
- 249 Downloads
Evidence is presented on synergies due to the sequencing and packaging of water reforms based on a review of case studies from the Zhanghe Irrigation System (ZIS) in the Yangtze River basin in China. Faced with strategic challenges of economic growth, food security, population growth, and climate change, China has implemented numerous hardware and software interventions in the water sector to increase the availability of water. These interventions—ranging from the provincial and system level to farm and field level—allow reallocation of water from agriculture to higher value uses, without significant reduction in crop production. This review of selected hardware and software interventions suggests that water sector reforms generate significant benefits for peasant farming communities and local governments. The review indicates that all agents respond to the same set of incentives simultaneously, by changing production and resource use decisions such that cumulative benefits from hardware and software interventions reinforce synergies. Synergies from reforms are evident, yet scaling up local collective actions for optimal impact is problematic. Heterogeneity in socioeconomic factors, as well as spatial differences, are the main stumbling blocks. Rewarding reformers seems to work, yet the benefits are neither immediate nor straight forward. Local implementation of national policies requires a systematic and coherent framework suited to the level of economic development of each region in order to achieve synergies from water reforms.
KeywordsZhanghe Irrigation System Collective action Water sector reforms Synergies Water saving irrigation practices Rice
China faces strategic challenges amid rapid economic growth and transformation of its economy (Mushtaq et al. 2008). Population growth and global change exacerbate these challenges. While food security issues are not currently serious, turbulence in global markets can trigger a red tsunami (Khan et al. 2009). Escalating water demand and water scarcity can compromise food security. An ageing population and the exodus of younger generations to cities, leaving behind children with ageing parents, are signs of a grey tsunami. A constellation of these forces can threaten global peace and stability. Answers to these complex issues have to date been sought in economic reforms.
Economic reform in socialist economies generates major gains and losses, and the distribution of impacts differs among various social groups, with implications for social equity and the course of reform. Peasant farmers are not only a particularly important group from the perspective of equity, but also from the perspective of economic reform. China has witnessed two decades of economic reform policies, which have led to significant gains in agricultural productivity and economic wellbeing. During the 1980s and 1990s, agricultural productivity rose steadily and per capita grain output reached a level similar to that in developed countries; many farmers shifted to higher value crops, and food exports grew significantly (Rozelle and Rosegran 1997). With sustained growth in agriculture, rural incomes rose dramatically, lifting millions of people out of poverty during this time (Hussain and Hanjra 2004; Hanjra and Gichuki 2008). Despite that success, more than 100 million farmers and their families still live in poverty, and the gap between rural and urban incomes has not narrowed; the inequality in the rural economy has remained high since the mid-1990s (World Bank 2005; Zhang et al. 2006). Socioeconomic uplift of peasantry—by raising rural incomes—and integration of rural economy into the modernization process—through good governance—is the greatest challenge facing a successful reform policy in China.
Water savings gained through increased water use efficiency (producing more food with less water) are critical. China’s government has identified the nation’s rising water scarcity as one of the key problems and has attempted to address this issue at nearly all levels, from the national down to the village and farm level (Zhang et al. 2001; Lohmar et al. 2003; World Bank 2005). The Chinese government successfully implemented policies that enabled them to withdraw water from the agricultural sector to fulfill growing demand in industrial and urban sectors. Water conservation projects were implemented and complemented with the introduction of water-saving irrigation practices, both of which may have caused farmers to voluntarily or forcibly reduce water use by adopting water saving irrigation practices. The farmers also contributed to these efforts by successfully maintaining their agricultural production despite a considerable decrease in water deliveries, by relying on local sources, such as ponds, and adopting water saving irrigation practices.
It has been argued that positive synergies can be realized from proper sequencing and packaging of reforms such that the aggregate benefits from the reform processes are greater than the sum of benefits from individual reforms (Dinar et al. 2007; Hanjra et al. 2009). However there are no studies exploring synergies from water reforms, particularly, farmers’ response to water usage and crop production. This paper presents evidence of synergies from the sequencing and packaging of reforms using farm level case studies from the Zhanghe Irrigation System (ZIS) in the Yangtze River basin, China.
2 Conceptual Framework
Hardware and software interventions are complementary, but not substitutes, to each other (Hanjra et al. 2009). Hardware refers to investments in core infrastructure such as irrigation reservoirs, small and medium ponds, and delivery channels and equipment used for the application of water on-farm. This includes the development and implementation of water saving irrigation technologies and practices. Software refers to the policies and institutional mechanisms and support measures used for the governance of water resources, including water storage, allocation, distribution among and across sectors, irrigation services delivery, setting and collection of water charges, and system operation and maintenance services.
3 Study Setting
3.1 Zhanghe Irrigation System
4 Collating and Synthesising Evidence
4.1 Hardware Interventions
Multiple hardware and software interventions have been employed to reduce agricultural dependence on ZIS water. The hard interventions discussed below include the development of water saving irrigation practices (WSI), especially alternate wetting and drying (AWD) for rice production, and construction of small dams capable of storing water near farmer’s field.
4.1.1 Case study: Alternate Wetting and Drying (AWD) Irrigation Practice
Historically, rice has normally been grown under flooded conditions (Bouman et al. 2007). However, water scarcity poses major challenges to rice production. The Chinese government, after recognising rising water scarcity as a key problem for the national economy, has promoted water saving irrigation (WSI) practices since the 1990s. More than 150 research stations, in collaboration with professional institutions and universities, have been conducting research on WSI practices for many years (Li 2001).
AWD irrigation practices lead to a reduction in irrigation water use, without a distinct reduction in crop yield (Li 2001). These practices allow farmers to achieve a relatively dry soil condition before receiving further water, and to store more water after rainfall. In this way, the utilization of rainfall is facilitated, the need for canal water is eased, and irrigation events are reduced. In addition, percolation and seepage losses from rice fields are controlled. Careful water management for rice saves water, improves yield, increases profitability by reducing the cost of production, and reduces risk of pests and diseases due to improvements in the microenvironment (Stoop et al. 2002).
Comparison of grain yield and water productivity in two experiments in Tuanlin (TL), Zhanghe Irrigation System
Nitrogen (N) regime (kg ha−1)
Continuously submerged (CS)
Alternate Wetting and Drying (AWD)
Grain Yield (t/ha)
180 (2 splits)
180 (4 splits, early)
180 (4 splits ’99, 6splits ’00)
Water productivity (kg grain m−3 water input)
180 (2 splits)
180 (4 splits, early)
180 (4 splits ’99, 6 splits ’00)
With decreasing water availability for agriculture and increasing demand for rice, water use in rice production systems must be reduced and water productivity increased. The results from the experiment show that water productivity was significantly higher under AWD in two out of three experiments. The results are typical for poorly-drained irrigated lowlands in Asia, and indicate that AWD can reduce water use by up to 15 % without affecting yield. Data from other rice producing regions support these results (Törnqvist and Jarsjö 2011; Moya et al. 2004; Tabbal et al. 2002). The results indicate that adoption of AWD practices is a key factor enabling ZIS to reallocate reservoir water from irrigation to higher value uses without significantly impacting rice production.
4.1.2 Case study: Multipurpose Ponds for Reliable Water Supply
Ponds are small reservoirs located in irrigated areas that allow farmers to capture rainfall and store surplus water from other sources. Farmers build and improve existing ponds so that they are less reliant on water from the ZIS. Roost et al (2008) stated that farm ponds have given Chinese farmers sufficient flexibility in water use to practice varying forms of AWD irrigation for rice, without compromising yields and incomes.
Year of construction of the ponds in the Zhanghe Irrigation System
Year of construction
Small multi-purpose ponds have contributed to the transfer of water from irrigation to higher value uses by capturing rainfall and store surplus water from other sources. The results of cost–benefit analysis show that ponds of all sizes were profitable. The profitability along with decreasing water supplies from ZIS explains why farmers are increasingly investing in ponds. Overall, large ponds showed higher profits as compared to small and medium ones; this is mainly due to economy of scale, which allows farmers to store relatively large quantities of water and also provide additional opportunities for fish harvesting.
4.2 Software Interventions
The software interventions discussed below include collective action by farmers to enhance water supply from ponds; incentives to adopt on-farm water saving irrigation practices; tax-for-fee reform to adjust agricultural tax and water rates; institutional reforms; and water pricing to create artificial water scarcity to encourage water savings by farmers.
4.2.1 Case Study: Collective Action at Farmer Level
In China, ponds are local common property resources, and water in ponds is collectively owned by the farmers. Each pond provides water to several users. Management of the ponds requires local communities to pool their efforts to perform various tasks, such as removing silt, constructing channels, regulating water allocation and monitoring violations, weeding, and repair and maintenance of ponds and channels. Collective action for the management of local common property is one of the most important issues in rural development (Hayami 1997). How this collective action is performed can determine the usefulness of the ponds.
Mushtaq et al (2007b) measured the performance of collective action through farmer’s perceptions. They were asked whether desiltation, repair and maintenance (R&M) and cleaning were effective or whether they wanted these activities to be done more often. A value of 0, 1, or 2 was given to different levels of satisfaction. A value of 0 means that farmers strongly agreed that collective action was needed more often, while a value of 2 means that farmers were fully satisfied with the current level of management activities.
Effectiveness of collective action for pond management, by pond size in the Zhanghe Irrigation System
Percentage distribution of effective conduct of collective action
Do you need more desiltation of ponds, repair & maintenance and channel cleaning?
Collective action by farmers is a critical factor in improving water management from ponds. It was found that collective action was more effective in small ponds. Dependence on ponds as a source of irrigation seems to increase the effectiveness of collective action in pond management. Larger household size also contributed to effective collective action, and farmers gave more emphasis to collective action when high quality land was involved. Although farmers have shown dissatisfaction with collective action for managing ponds, it has helped in addressing water scarcity locally by keeping water within the system. It is clear that pond operators have to maximise the use of catchment runoff for irrigation in order to minimize their dependence on ZIS deliveries.
4.2.2 Case Study: Tax-for-Fee Reform at Government Level
The Fei Gai Shui (FGS) is a central government attempt1 to relieve farmers of fee burdens that have been eroding rural incomes throughout the 1990s, especially after the 1994 tax reform (Toh and Lin 2005). Tax-for-Fee reform (Fei Gai Shui) was heralded as a solution to excessive fiscal predation by local governments (Yep 2004). Under the FGS reform, various types of irregular fees, fines and quota were completely abolished, and replaced with a single agricultural tax. The agricultural tax and overall burden, based on stipulations, should not exceed 8.4 % of net per capita income. The basic aim is to reduce the burden on farmers by 25 to 30 % (Zengke 2000), and to help bring the tax within their capacity to endure.
FGS has an immediate impact on water resources. With the FGS, farmers need to request water from ZIS (rather than going through the village or township administration) and to pay all or a portion of the water fee in advance, in addition to any increase in water prices. In 2002 and 2003, the ZIS irrigation releases were down sharply after the introduction of FGS (Fig. 1). After FGS, rice production area and yield also declined; however, the decline was greater in the dry year of 2003 (Dong et al. 2004). Adjustments were made in 2004 to correct this problem, and as a result ZIS irrigation releases were recovered in 2004 (Loeve et al. 2007). However, it remains to be seen what the long term effect of this policy will be on the operation and management of the ZIS.
Mushtaq et al (2008) evaluated the initial impact of the FGS on agricultural productivity via the agricultural water management pathway, specifically in terms of its impacts on water distribution and conservation, irrigation costs, pond water use, and rice area and yield, using household level panel data collected over a 2 year period. Multiple regression analysis was used to estimate the impact of the FGS on pond water use frequency, rice area and yield, and cost of irrigation.
Empirical Findings for the Impact of FGS
Regression results for the effect of Fei Gai Shui on water use, rice area and rice yield in the Zhanghe Irrigation System
Access to pond
Distance of main plot from the pond
Dummy for village 1
Dummy for village 3
Dummy for village 4
Number of observation
The empirical results show that Fei Gai Shui (FGS) had a positive impact on pond water use, but a less desirable impact on rice area and yield. Farm pond water played an important role in sustaining agricultural production in the wake of sharp decreases in ZIS water deliveries. The results show that the presence of ponds cushioned the impact of FGS in the study areas. The empirical results also indicated that access to pond water had a positive impact on yield and crop area, and helped to reduce the irrigation cost. This posits that in near future, and especially after the implementation of FGS, ponds will continue to play an important role in sustaining agricultural production. Although these policies will ultimately affect all villages in China, the magnitude of the initial effects on local communities may differ due the fact that these policies were implemented at different times in different areas.
4.2.3 Case Study: Institutional Water Scarcity at Irrigation District Level
Institutional water scarcity is defined as a policy of rationing limited quantities of water among the largest number of farmers, creating artificial scarcity of water in order to encourage irrigators to use water more sparingly and efficiently, and maximize output from limited water supplies.
Two-Part Water Pricing in the ZIS
Water pricing systems in the Zhanghe Irrigation System
Free or low fees for maintenance and operation
Fees at a low level
Normal pricing, while causes a lot of issues
Piloting two-part water price
Two-part water price
Fee to tax reform
Abolition of agricultural tax
Incentives to Save and Reallocate Water
In the ZIS, both system managers and farmers have incentives to save water. Managers would like to reduce supplies to allow more water to be allocated to non-agricultural uses, which pay more for water. The decline in water availability for irrigation from the ZIS reservoir, along with the volumetric pricing of water, provide incentive for farmers to adopt water savings practices (e.g. AWD and farm ponds) that help them cope with water scarcity and reduce the cost of water. Thus, it appears that the management of the canal water is not only a function of farmer demand but also a strong element of supply approach (Loeve et al. 2001). Available water supplies are rationed amongst the largest number of farmers, creating institutional water scarcity.
In summary institutional and water policy reforms had myriad effects on agents’ incentives to save water. Water pricing regimes created institutional water scarcity, affecting individual farmer’s irrigation costs, irrigated area and rice productivity; water user associations responded positively to these incentives by ordering water only when needed and adjusting their requests with rainfall; revenue from water to the ZIS increased as a result of diverting water to higher value uses; farmers responded by changing water management practices, and improving existing ponds and building new ones to store water.
4.2.4 Case Study: Adoption of Water Saving Irrigation Practices by Farmers
Water saving techniques such as AWD developed in the wake of intensifying water competition and water shortages in agriculture. However, they are adopted at limited scale. Reliable water supply is one of the key factors identified for the adoption of WSI practices (Mushtaq et al. 2006; Loeve et al. 2004).
Mushtaq et al (2006) modeled adoption of AWD at the farm level as a function of reliability of water sources namely canal water from the ZIS, water from smaller reservoirs, and water from local ponds, along with household-level socio-economic characteristics, land quality and farm size. To measure AWD adoption, an AWD score was calculated based on soil-water conditions. The AWD score falls between 0 and 1 if a farmer irrigated at a combination of different soil-water statuses. The score indicates if the farmer practices AWD or not, with the higher score indicating a greater adoption of AWD.
Tobit estimates for the adoption of alternate wetting and drying (AWD) practices in the Zhanghe Irrigation System
Tobit estimate of index function
Tobit estimate of marginal effecta
Access to pond
Irrigation from ZIS water
Irrigation from pond water
Irrigation from tubewell
Irrigation from small reservoirs
Elevation of the main plot
Water saving irrigation training
Wealth of the farmer
Dummy for village 1
Dummy for village 3
Dummy for village 4
Log likelihood function (unrestricted)
Log likelihood function (restricted)
Likelihood Ratio (LR)
Scale factor for marginal or total effect F(z)
Conditional mean of dependent variable at sample point
The empirical results indicated that farm size, frequency of irrigation from the pond, and land quality have a significant but negative influence on the adoption of AWD practices. The results of the model were somewhat unexpected, but are indicative of risk-averse behaviour of farmers. Initially, it was thought that ponds are a prerequisite for the adoption of AWD practices, as described by Loeve et al (2001). However, the results highlighted that, although ponds are important for the adoption of AWD practices, access to the ponds gave farmers additional water at their disposal, which resulted in continuous flooding, a deviation from ideal AWD practices.
The results also show that the reliability of local water ponds had only a weak, significant positive effect on AWD practices. Thus reliable water supply does not necessarily result in the adoption of AWD practices. The adoption of AWD is not driven by farmer desire but rather imposed on them due to increasing water scarcity. With the steady decline in ZIS releases, farmers were forced to find ways to save water.
5 Synergies from Water Sector Reforms
Clearly both hardware and software policies have contributed to the ability of ZIS to reallocate reservoir water from irrigation to higher value uses without any significant decline in rice production. Furthermore, there appears to be an interaction among hardware and software strategies, such as the development of farm ponds and AWD practices, volumetric pricing, adoption of AWD, and institutional water scarcity by allocating diminishing supply. The key factors have been the increase in rice yields and water ponds, which kept the water within the system and sustained crop production. In particular the role of ponds in sustaining production was pronounced. We review empirical evidence presented by Mushtaq et al (2009b) on these hardware and software polices, especially water ponds, in sustaining agricultural production, specifically the impact of ponds on cost of irrigation, crop area and yield.
5.1 Case Study: Sustaining Crop Production
5.1.1 Impact on Cost of Irrigation
Multivariate regression models estimates of for the impact of pond on the cost of irrigation, rice area and yield in the Zhanghe Irrigation System
Cost of irrigation
Dummy for pond
Elevation of the main plot
Education of farmer
Wealth of the farmer
Dummy for village 1
Dummy for village 3
Dummy for village 4
Number of observations
5.1.2 Impact on Rice Area
The size of the farm, land quality and variables for the villages Huangyun and Sundian were found to have a negative impact on the percentage area where rice is grown (Table 7). Larger farms are more diversified, which means that with the increase farm size, the rice area will decrease. On the other hand, the coefficient of the AWD score was found to be positive and to have a significant positive effect on the percentage of rice area. This positive result was expected, as AWD practices save around 20 % of water. Therefore, as a result of this saved water, rice area should increase.
5.1.3 Impact on Rice Yield
The result indicates, except the variables for villages, none of the variables were found to affect the rice yield significantly (Table 7). This implied that access to ponds, increase in farm size, good quality of land, farming experience and wealth of the household contributed positively, though not significantly, to rice yield.
Yield in the Sundian village were found to be significantly higher than other villages due to relatively better availability of water. The results for the AWD score showed that by adopting AWD practice, yield is increased, but that the increase in the yield was not statistically significant. Nevertheless, achieving the same yields with less water still means “same crop with few drops” and thus water savings.
6 Conclusions and Directions for Policy
Empirical evidence from the case studies on selected hardware and software interventions shows that water sector reforms can generate significant benefits for peasant farming communities and local governments. The results in particular show that all agents respond to the same set of incentives simultaneously, by changing their production and resource use decisions such that cumulative benefits from hardware and software interventions reinforce synergies. For instance, the government responds to water scarcity by allocating water away from agriculture, investing in research on water saving irrigation techniques, and through tax-for-fee reform or FGS, to lessen farmer’s tax burdens. Irrigators responded by adopting water saving practices such as alternate wet-and-dry irrigation for rice production, diverting less surface water supplies, and investing in local water ponds for rainwater storage for more reliable and cheaper water supplies; pond managers and local cadres/groups responded through collective action for improved pond management and maintenance; government introduced a two-tiered water tariff, altering incentives to use water more efficiently and motivating farmers to reduce their water use and adjust cropping pattern where feasible. Irrigation cost fell, while rice productivity stayed the same or even improved. Water deliveries to agriculture fell sharply while effectively maintaining rice production. Synergies from reforms are evident and are supported by other studies (Hanjra et al. 2009). However, scaling up local collective actions for optimal impact is problematic. Heterogeneity in socioeconomic factors, as well as spatial differences, are the main stumbling blocks. Rewarding reformers seems to work, yet the benefits are neither immediate nor straight forward. Local implementation of national policies requires a systematic and coherent framework suited to the level of economic development of each region for realizing synergies from reforms.
Does it pay to join the party? The answer is a definitive yes, compared with reforms in socialist economies, where economic reforms only have limited success. Inviting others to join the party requires further research for more robust scientific evidence.
The FGS (农村税费改革) is part of the general central government restructuring and centralizing program that can be tracked back to 1998 (Kennedy 2007). The FGS was introduced at the provincial level as a form of tax relief for the farmers. It was first introduced in Anhui province in 2002, and then broadly introduced to 20 other provinces in 2002. In order to further reduce the villagers’ burden, the central government announced complete elimination of the agricultural tax by 2006. Yan’an in Shaanxi province was one of the first districts to eliminate all local fees and agricultural tax in 2004.
The author greatly thank Dr. Bas Bauman, International Rice Research Institute, Philippines and Dr Luo Yufeng, Hohai University, China, and Dr Munir Hanjra, Charles Sturt University, Australia, for their feedback and comments in preparation of this manuscript.