In a burst of electioneering enthusiasm, as the Nepali Congress candidate from Kathmandu-1 in 1994, Krishna Prasad Bhattarai had boasted that if elected he would "wash the streets of Kathmandu with Melamchi water". Many had scoffed. Now, it looks like Bhattarai's promise will come true, but not for the reason he had intended. Kathmandu's water supply network is so antiquated and leaky that if Melamchi water was sent through the pipes at high pressure they would indeed flood the streets.

Last week in Manila, the Asian Development Bank (ADB) approved a $120 million loan for the Melamchi Water Supply Project (MWSP) to bring glacial melt through a 26-km tunnel to Kathmandu to augment the valley's water supply. The approval came just in time for end-of-fiscal-year disbursement and raised the bank's commitment to Nepal in 2000 to $173 million-up from $50 million in 1999.

The good news is that there is now money to start building the headworks, the tunnel, and the water treatment and management plants which have been on the drawing boards since 1988. The bad news is that even if Melamchi water arrives on schedule in 2006 the leaky and obsolete distribution network would not end the water shortage. And then there is the worrying question of what we are going to do till 2006.

The reason for Kathmandu's water shortage has nothing to do with not having enough water inside the Valley for current consumption. The Nepal Water Supply Corporation (NWSC) estimates that it has a supply of 60 million litres a day even during the dry season. Factoring in the official figure for system loss from leakage (40 percent) leaves roughly 36 million litres for distribution each day. Divide that among the roughly 100,000 households and that means 360 litres, or 24 buckets, of water per family. But Kathmandu consumers are not getting that water.

"So where is the water?" asks Dipak Gyawali, an engineer who studied the NWSC extensively in the mid-1980s as member of the Pokhrel Commission. The Commission was formed to look into why the NWSC's efficiency did not improve even after several million dollars spent on World Bank projects.

Says a senior water supply expert at a multilateral institution in Kathmandu: "It is like putting the cart before the horse, we are going into Melamchi even before we know whether we need this water or not, and if we do when it should come, and when it does how expensive it is going to be."

NWSC does not even have a complete picture of the distribution system it manages, which is made up of a spaghetti-like network of consumer connections, some of it about 100 years old. Sources at NWSC say the only maps that exist are those prepared under a German project some years ago, and these have not been updated even though 3-4 inch secondary lines have been added since. The source said NWSC does not even have an idea where to dig if it wants to disconnect connections to households that have defaulted on their water bills.

NWSC is inefficient because it has a politicised management and staff, with 10 general managers in the last 10 years. How such an organisation is going to efficiently oversee Melamchi distribution by a private operator is difficult to comprehend. But the plan is that Kathmandu customers will pay (a higher rate) for the water, and the money will be used to pay back the loan and to fund development work in the Melamchi valley.

The World Bank, which is contributing $80 million-mainly for upgrading the distribution system, knows very well that the problem is not about increasing supply. Having spent Rs 1 billion between 1974 and 1989 on improving distribution it is said to have become convinced that putting in more money without serious reforms of the management and distribution would be like pouring water into the sewer. The Bank has prepared a Project Concept Report and only after its approval by the bank's senior management would the money be released.

The ADB itself has several time-bound conditions which would have to be met to be able to use the loan. These include enforcing major policy and legal reforms, creating a new institution to manage water resources in the valley, relocating water-intensive industries, limiting ground water extraction and putting in place a tariff adjustment mechanism based more on economics and less on politics. "These reforms need major, serious political commitments," a donor source told us. "On our part we're convinced that good management and effective distribution are as important as digging a new tunnel to bring in water."

The Melamchi Water Supply Board-the body created to implement the project-is not unaware of these problems. Dinesh Chandra Pyakurel, head of the Board, admits that the problems resulted from infrastructure-driven planning the country had been following all the time instead of plan-driven infrastructure building. "But Melamchi is a vision for the long term and it has to be built even as we address the institutional problems." We have six years to see how committed the project is to his words.