Estimating Irrigation Subsidies: The Need for a Common Methodology
By Ravinder P.S. Malik

Agriculture is the leading consumptive user of water, accounting for between 70 to 90 percent of the total water used in developing countries and more than one third of the water consumed in many OECD countries.

In most countries, water for irrigated agriculture is subsidised by governments, with far reaching implications for food security, patterns of agricultural development, trade, government finances, and, more generally, the sustainable use of water.

Some of the available evidence, such as from India, shows that subsidies for irrigation have promoted inefficient use of water, contributing to steep declines in levels of ground water. Indeed, at a time when water scarcity is a looming crisis in many countries, the need to reform subsidy regimes for irrigation is urgent.

As a first step, reforming irrigation subsidies requires a clear understanding of the concept of a ‘subsidy’ and the methods that are used to estimate that government support. Generally, subsidies to irrigation water are defined and measured from two different perspectives: from that of the water-supplying agency (generally a government or government agency) and from the beneficiary (user) of the irrigation water. Of the two approaches, the first approach has been used more widely to define and measure subsidies.

Under this first approach, subsidies are usually calculated as the difference between the cost of supplying irrigation water and the revenue realised from its sale. However, both the key elements on which the estimation of subsidy is based — ‘cost’ and ‘revenue’ — have been measured and interpreted differently depending on the analyst. Several characteristics of irrigation works have contributed to these variations. Most are multipurpose: in addition to providing water for irrigation, they may also supply water for municipalities, industry, hydropower, navigation, tourism and fisheries, as well as providing flood protection.

Moreover, while most of the large projects are built and maintained by governments or their agencies, small groundwater-based systems are often owned, operated and maintained by individual farmers, in some cases with governments providing support via low electricity tariffs for pumping.

Estimating the cost of providing irrigation water is further complicated by a host of other factors that need to be addressed. These include: how to apportion the capital costs of irrigation in a multipurpose project? Should capital costs be treated as sunk cost? What assumption should be taken as the life of the project? Should the cost of externalities be accounted for? If so, how should the cost of environmental degradation be quantified? Are the necessary data available to estimate these costs?

There are similar issues that need to be resolved on the revenue side. Are farmers the only beneficiaries of irrigation water? Do other sections of the society also gain from the availability of irrigation water? Are there any other revenues in addition to the irrigation charges to the government from the sale of irrigation water? Is enough data available to estimate revenues?

As it is, an assortment of methods has been used to estimate costs and revenues. And because the methodologies are different, the estimates are usually not comparable. For example, when estimating costs, some analysts consider only the operation and management costs; others add some fraction of capital cost, but without clarifying how the cost of multipurpose projects have been apportioned and how the capital invested in the past has been accounted for. If irrigation subsidies are to be measured in a way that make their estimates more meaningful, comparable and useful, a consensus on a working and widely acceptable definition of irrigation subsidies and their methods of measurement is vital.