Background:
The Republic of Armenia, a small ex-Soviet country struggling to cope with the breakup of the Soviet system, was trying to prepare its public-owned infrastructure for privatization. USAID sponsored many projects to assist with these efforts. One of these projects was modernizing the distribution infrastructure at the main electric power distribution utility in the capital city. This project included creating and implementing a billing and collections system (BCS), named Abacus for the ubiquitous devices still in common use, to help put the utility on a sound financial footing.

Under the Soviet regime, power was supplied to all free of charge. While each apartment had a meter, common equipment such as like lifts and stairway lighting did not. The breakup of the Soviet empire caused grave financial difficulty to the Government who imposed charges for many things, including electric power, that were formerly free. This resulted in widespread power theft and evasion of these charges.

The Soviet business mindset resulted in some interesting but sub-optimal attempts to resolve this problem. Among these was a government decree to read all power meters on a single day - the last business day of the month. Intended to facilitate tracking and quantify power losses, whether from aging and un-maintained distribution infrastructure or power theft, this practice accomplished nothing.

Every month, on the last business day, every available employee was press-ganged into being a meter reader. Equipped with a pencil and a sheet with the last month’s meter readings, the employees were supposed to go from building to building and note the current reading on the sheet. They turned these sheets in at the end of the day to HQ staff, which then manually calculated the charges and prepared bills for distribution.

What actually happened was that many “meter readers” found they had been given extra time off. Most visited a few buildings, read a few meters, and invented the rest of the required data. Others just guessed at the readings without going anywhere near the meters. Meters positioned in dark stairwells frustrated the efforts of those employees dedicated enough to try and actually read all meters.

HQ staff trying to cope with all the sheets of handwritten meter readings had a difficult time. Being utterly swamped, they usually just transcribed the information into the ledgers without any verification and created the bills from the difference in the “current” meter reading and the previous one. All they could do by way of ensuring accuracy was investigate the more egregious errors that caught their eye and resolve them. Most errors went uncorrected until the recipients complained vociferously about the bills.

The net result was inefficiency and opportunities for widespread corruption.

Repeated requests to adopt a more rational meter reading program were met with the statement that this was a parliamentary decree and could not be changed.

Incentive to Change:
The (then) current phase of the program had a funding cap and included investing in power distribution infrastructure as well as extending the Abacus pilot program into full production at all fourteen district offices. This required replicating all computer equipment in use at the pilot office across the other thirteen. The estimated total cost of this computer equipment was $776,000.

Given the pervasive state of decay in the distribution infrastructure, upgrading the infrastructure was critical. However, there was not enough funding in the project budget to do everything the government and utility officials wanted or needed in terms of infrastructure upgrades and additional computer equipment. They repeatedly requested additional funding but the local USAID office, having no additional funds, was unable to accede to their requests. A number of meetings between government officials and USAID agents could not resolve the impasse.

Roy Sequeira, on the USAID team, who had created the computer budget analyzed the situation and found a way to resolve it to everyone’s satisfaction.

The Solution:
Roy suggested a two-step solution to the problem

1. Revert to the industry-standard practice of “cycle-billing”



2. Centralize billing.

He showed that with centralized cycle billing, the existing computer equipment, augmented by an additional $48,000 worth of new equipment could handle the billing load quite comfortably.

These two steps would free up nearly three quarters of a million dollars that for the much-needed and much-desired infrastructure upgrades.

Roy presented these proposals together with his detailed analysis of additional benefits at a meeting attended by the Senior Deputy Minister for Energy and senior utility officials. He also showed how these additional benefits would greatly improve their plans for privatization.

This package was so attractive to those present that they subsequently petitioned the Legislature for approving the change to cycle billing.

Additional benefits:

• Easy “load balancing” by scheduling a special twenty-first cycle to read selected bulk distribution meters such as those at major sub-stations. Investigating discrepancies at this level, where the difference between a high-level meter and the sum of subsidiary meters is greater than predicted by line conditions, would help pinpoint specific areas for intensive audits. Initially, focusing attention where the gap is greatest would bring maximum benefit to loss prevention and control.



• Smoother distribution across the month of cash flow and workloads enables proper sizing of work staff to get the job done well. Each staff person gets a lot of practice repeatedly doing his or her job. This reduces errors and other resource-intensive corrective measures and frees up staff resources.



• Fewer workload-caused input errors result in accurate bills, fewer customer disputes, and minimal bill adjustments.



• Reduced time from meter reading to bill being ready for payment: Meter readings could now be immediately processed without lengthy “data scrubbing” so that bills are made available to the Banks and Post Offices at close of the next business day and payments start flowing in the day following. This means cash flows start between 2 and 4 days from the day of the meter reading, very much earlier than the 10 to 15 days associated with the single meter reading program. Insufficient historical data precluded quantifying the value of this improvement in cash flow.



• A smoother and regular workflow enabled workers developing and retaining expertise by specialization and practice. This applied across the board, to meter readers, data input staff, audit staff.



• More time now available to investigate discrepancies between actual readings and expected readings. While not currently quantifiable, auditing usage should help lower the rather large commercial losses currently sustained.



• Specialization allows meter reading using non-technical staff. This removes any functional job-skill overlap between meter readers and field maintenance staff, hence reduces opportunities for collusion.



• Centralizing the billing and business functions will allow branch and local offices to focus on operations while leaving paperwork and bureaucracy centralized and contained within HQ. Again, benefits are not currently quantifiable.

Summary:
By uncovering a means of providing officials and staff with what they wanted, Roy was able to achieve some of the aims of the USAID program and the Government:

• Streamline the business,


• Put it on a more commercial level, and hence


• Make it more attractive to potential buyers.