Water Cost Index - overview

Rickards Real Cost Water Index™ calculated by IBM
(powered by IBM Infosphere BigInsights)

Water Cost Index
Q4 2013
in US Dollar (USD)
% Change from
Q3 2013
in US Dollar (USD)
% Change from
Q1 2008
in US Dollar (USD)
% Change from
Q1 2008
in Local Currency
Global $1.31 ▲ 1.8% ▲ 16.6% ▲ 32.9%
London $2.26 ▲ 4.5% ▲ 9.4% ▲ 33.7%
Manila, Philippines $0.34 ▲ 0.1% ▲ 20.3% ▲ 28.3%
San Antonio, Texas $1.97 0.0% ▲ 22.6% ▲ 22.6%
Sao Paulo $0.97 ▲ 0.8% ▲ 0.9% ▲ 32.0%
Singapore $1.61 ▲ 1.4% ▲ 50.0% ▲ 33.0%
Southern California
$0.53 0.0% ▲ 31.8% ▲ 31.8%
Uganda $1.49 ▲ 2.1% ▲ 0.8% ▲ 49.3%

Waterfund and IBM are developing the Rickards Real Cost Water Index™ (WCI) to benchmark the true cost of water production in individual geographic areas, which includes operating, capital, and "hidden economic" costs. The benchmark is motivated by the fact that there is a huge backlog in water infrastructure investment projects due to a shortage of capital. Governments and private water agencies are increasingly forced to turn to the private sector to fund the construction and maintenance of complex water networks. Unfortunately, due to a variety of hidden costs such as government grants to water agencies, capital markets and private equity do not have a systematic way of pricing and measuring risk in water infrastructure investment projects.

Water Cost Index Computation

Water Cost Index =

Total cost of production / Total delivered freshwater volume in cubic meters


  • Total cost of production is calculated as the sum of operating costs, capital costs, and identified subsidies.
  • Total delivered freshwater volume (in m3) is the amount the producer reports as delivered, and excludes water lost either due to system leakage, pilfering, or other forms of loss. This penalizes producers with a large fraction of production volume being lost due to system inefficiency.

WCI as Benchmark for Financial Products

The Rickards Real Cost Water Index™ serves as a benchmark for helping measure hundreds of critical projects on a like-for-like basis. Index values reflect estimated water production costs measured in US dollars per cubic meter for a variety of major global water infrastructure projects ranging from retail water utilities to wholesale water utilities. Such a market benchmark will spur the development of next generation financial products for both water producers and investors and to aid the growth of the water sector globally. Here are two examples of how these products would be leveraged:

Scenario 1: A Water Agency cannot obtain bank financing for Phase 2 of a seawater desalination plant project due to previous cost overruns on Phase 1. Yet the Agency lacks the water it needs to supply a contractually specified daily volume of water to its largest customer, with a consequent risk of large penalties for each day of insufficient volume. Using strike and trigger values based on the WCI, the Water Agency could purchase a $25 million, 2 year insurance product. Payout to the Water Agency would be triggered on the total change in its Water Cost Index (as well as some other conditions, such as a specified increase in asset failure costs). This approach would enable the Water Agency to enhance its overall credit profile with the insurance enabled by the WCI, finance Phase 2 of the desalination plant and meet its supply obligations.

Scenario 2: A large desalination and water transmission system project needs to secure private equity and institutional funding alongside that from development banks and sovereign funds, to the tune of one third of the total project cost. To achieve this, the project needs a way to reduce risk to its investors. Based on movement in the WCI, the project could purchase $50 million in insurance. This would enable the insurance product to then be underwritten by a large reinsurer and allow the project to secure the private sector contribution it needs in order to proceed.

The Water Cost Index is computed using a uniform set of cost variables, called Normalized Production Cost Statement. These statements are constructed from publicly available unstructured financial data by a "calculation agent" developed by researchers in the Accelerated Discovery Lab at IBM Research - Almaden. The cost variables include both direct cost variables (e.g., operating expenses and interest payments) and "hidden economic" costs (e.g., direct payments from the government, interest subsidies, and infrastructure grants). By capturing the “hidden economic” costs as well, the complete cost of producing and supplying fresh water becomes evident. Additionally, costs irrelevant to fresh water production, such as tax payments and sewage costs, are excluded. Values for all cost variables are standardized, and have a single, well-defined semantics. This allows for a direct comparison of relevant costs for each producer. The performance of a water producer can now be benchmarked against other producers in the same geographic region or globally. A producer can be benchmarked on individual cost variables as well, which provides additional insight into their cost structure and the relative risk it presents.

(c) IBM, Waterfund 2013, 2014