Mary Tiger is the Chief Operating Officer for the Environmental Finance Center and Project Manager for “Defining a Resilient Business Model for Water Utilities” Project funded bythe Water Research Foundation.

Water and wastewater utilities most commonly generate revenue by charging rates against a customer’s water use. When setting these rates, a utility balances multiple and, oftentimes, conflicting, objectives. With their rate structures, most utilities strive to collect adequate revenues for operations and system investment, promote the efficient use of water by its customers, and maintain affordability for basic levels of consumption. In particular, the goals of revenue sufficiency and conservation promotion present utilities with a conundrum[1]: when customers are more efficient with their water use or conserve, a utility’s sales base erodes. The Environmental Finance Center is researching alternatives to the classic water utility retail pricing; alternatives that would better align utility revenue stability and the promotion of efficient water use.

One such alternative is the PeakSet Base Pricing Model. Under this pricing structure, the utility would charge individualized base charges based on a customer’s maximum month of consumption. This theoretical model is inspired by demand charges used by many power utilities, yet grounded in the limitations of prevalent water metering technology. Under this pricing structure, a customer’s base charge would be individually set based on the three-year rolling average of their peak month of demand. The utility would still charge its customers variable rates, but the variable rates would constitute a lower proportion of a customer’s bill. This model would allow utility to build more of their cost recovery into the base charge while still promoting customer conservation and efficiency. In particular, it would encourage steady water use.

The Environmental Finance Center recently worked with the Beaufort-Jasper Water and Sewer Authority to model a revenue neutral PeakSet Base Pricing structure using historical data. The table below compares BJWSA’s current rate structure to three Peakset Base pricing scenarios. These three scenarios differ in the proportion of fixed revenue generated for the utility. In Fiscal Year 2011, BJWSA collected 18% of its revenue residential customers from base charges. (See May 15th blog post for how this compares to other utilities.) Under the three proposed PeakSet Base scenarios below, BJWSA would have collected the same total revenue from its residential customers; however, more of it (37-57%) would have been from base charges. The PeakSet Base Model builds more of the cost recovery mechanism into a base charge that is calculated by applying a rate against a customer’s historical peak demand.

Table 1: Comparison of BJWSA current residential rate structure to three “revenue neutral” PeakSet Base models

Current Residential Rate Structure High Fixed (AR1) Medium Fixed (AR2) Low Fixed (AR3)
% Fixed Revenue 18% 57% 47% 37%
Base Rate $6.00/meter – water +$6.00/meter – irrigation $1.85/kgal of historic peak demand $1.49/kgal of historic peak demand $1.12/kgal of historic peak demand
Variable Rate $3.46/kgal of previous month’s use $0.52/kgal of previous month’s use $1.25/kgal of previous month’s use $2.01/kgal of previous month’s use

The following chart shows how the PeakSet Base model would impact a hypothetical customer. In fiscal year 2011, this customer consumed between 6,500 and 29,900 gallons per month, as reflected in parenthesis on the x-axis. In fiscal year 2010, their maximum monthly water use was 24,100 gallons which would be used to set the base charge under a PeakSet Base pricing structure. (Please note that for this modeling, we did not have three years of historical data. Ideally, the base charge would be applied to the three-year rolling average of a customer’s peak demand.)

Under current rate structure, this customer paid base charge of $6.00/month plus $3.46/kgal used each month as represented by the blue line. The household’s monthly charges are plotted by the blue line. Under current rate structure, this household paid $647.74 for water services in Fiscal Year 2011.

Under the high fixed PeakSet Base model (AR1) outlined in Table 1, the same household, with the same demand, would have paid $621.55 in Fiscal Year 2011. Compared to the current rate, the household’s monthly charge would be less in months with higher demand and more in months with lower demand. Overall, the household (and utility) would experience more even (i.e. predictable) payments over the course of the year.

As can be expected, the chart shows that the low fixed PeakSet Base Model (AR3) provides more bill predictability for the customer and revenue stability for the utility than the current rate structure, but less than the high fixed Peak Set Base model (AR1).

Figure 1: Comparison of monthly charges for water under current rate structure and two Peakset Base scenarios for hypothetical customer

Although the PeakSet Base model builds more cost recovery into the fixed charge (a practice that would seemingly discourage efficient water use under the classic pricing structure), it still sends a conservation message by using a customer’s consumption to establish that base charge. For example, the hypothetical customer will pay a higher base charge in FY12 than it did in FY11 because it set a higher peak in FY11 (29,900 gallons) than it had in FY10 (24,100 gallons). As such, the PeakSet Base model promotes steady customer water use because one high month of demand would be costly to a customer for the entire forthcoming year. As modeled in Table 1, however, the utility would still send a variable price signal every month to send immediate feedback on water use.

There are some challenges to implementing and executing this pricing model. As I mentioned in the example, we had a difficult time extracting sufficient data from BJWSA’s billing system to determine a three-year rolling peak that would establish a customer’s base charge. This could be a challenge for other utilities in establishing this rate. Additionally, there would be issues in dealing with new and moving customers. How do you determine the base charge with no historical record? A customer that is planning to move with not have a large incentive to conserve and the utility will lose revenue associated with their “waste” without a recovery mechanism. Moreover, a utility that implements a PeakSet Base model should anticipate more meter re-reads and high bill disputes as customers realize the long-term implications of a high meter read.

Regardless of some of the drawbacks, we believe this pricing model has merit and deserves further exploration. It moves the utility pricing model more towards one that better balances the oftentimes conflicting pricing objectives for revenue recovery/stability and customer conservation.

[1] Beecher, Janice. The Conservation Conundrum: How Declining Demand Affects Water Utilities. Journal AWWA, Vol. 102 Iss. 2, February 2010, Pages 78-80.