The AMI investment decision


By Jim Reilly

Distribution utilities invest in AMI technology for one of two basic reasons: either to achieve operational cost savings or in response to regulatory requirements for systems capable of supporting dynamic tariffs aimed at reducing peak energy demand through enhanced load control and demand response capabilities. Utilities are responsible to their shareholders to consider carefully the risks associated with the decision to deploy AMI. On the other hand, regulators have a responsibility to review the socio-economics of AMI to evaluate the potential benefits of the investment for all stakeholders. In order to persuade a utility to make the investment necessary to realise the full socio-economic benefits, regulators must offer sufficient financial incentives to balance risk and return. These incentives are commonly measured in terms of rate of return and cost recovery, which is intimately related to depreciation. Assumptions about technological obsolescence are fundamental to determining the useful life of AMI and thus the time period over which costs are recovered and assets depreciated.

The financial projections and financial analysis in AMI project proposals are prepared by utilities intent on reducing risks and earning a reasonable return. Costs and benefits are put together in a financial model, the key components of which are the rate of return and the timing of cost recovery. The cost component includes the purchase of meters and related capital equipment (for example, modems and other communication equipment), installation, maintenance, and meter reading, including two-way communication and data management. The benefit component includes operational savings, improvements in customer service, management efficiencies, and demand response.

Differences in costs are driven by the choice of the meter to be used, its assumed lifetime, telemetry, the meter data management system, and the rapidity of the technology deployment. Differences in computed benefits depend fundamentally on estimates of operational savings and assumptions about the demand response model. The net benefit is calculated in terms of the difference between the present value of benefits and the present value of incremental costs. Assumptions regarding the technology, the depreciation rate, the timing and scale of the rollout programme, and the lifetime of metering assets are crucial for the results.

The rate base investment for AMI to be recovered from ratepayers is included in the revenue requirement. The revenue requirement for AMI for any particular year includes the rate base investment, less accumulated depreciation, plus the authorised rate of return on the net investment. Adjustments are sometimes made for operational savings and other direct benefits to the utility itself.

Utilities seek to recover their investment in the shortest period of time to decrease risk and improve the rate of return. Clearly, the depreciation schedule is critical to the rate of recovery. When making the decision to commit to AMI deployments, utilities consider the regulatory rulings on rate recovery as fundamental. At the heart of these rulings are decisions on the treatment of depreciation – which are founded on studies and conclusions on technological obsolescence.

Views on technology obsolescence are fundamental to the financial calculations that influence the AMI investment decision. Technology obsolescence defines the useful life of AMI assets, and thus the depreciation schedule and cost recovery that are key determinants of cash flow and the rate of return.

Technology obsolescence is commonly thought of as the loss in value from the substitution of one technology for a newer technology. More precisely, technology obsolescence is the substitution of an older, established technology for a newer technology having a higher level of functionality. In the case of technologies with long lead times for development and acceptance with markets that demand large scale deployments at very high infrastructure costs, such as electric meters, technology obsolescence is more gradual due to the simple fact that the substitution takes a substantial amount of capital and time. However, when a technology no longer services its intended purpose, and another technology is available, it may be said to be functionally obsolete.

Technology obsolescence is a function of both the rate of a technology’s advancement and its rate of diffusion to the market. It typically follows an S-shaped curve. Technology improves slowly at first because it is poorly understood; then accelerates as understanding increases; finally, it tapers off as limits are approached. This is shown in Figure 1.

Technology obsolescence impacts all the components of AMI:

  • Meter – microprocessor components, telemetry interfaces
  • Communication network – multiple transport media
  • Information technology – host hardware and applications, interfaces (billing, customer service, outage management and asset management), MDM.

The technology matrix for AMI can be examined against the measures for obsolescence both as a solution and by component. Taken as a solution, AMI fits the S-curve well. Estimating obsolescence is the most challenging part of the technological and economic analytical process for AMI, especially when creating a depreciation schedule over the entire integrated system’s useful life.

AMI investment1

Figure 1 – Evolution of technology obsolescence

Useful life is defined as the estimated period of time over which it is anticipated an asset may be profitably used for the purpose intended. The determination of useful life is the first major step to creating a depreciation schedule. The key issues to keep in mind are a) every asset has a reasonable range of useful life, and b) the treatment of a group of assets as an integrated system can create a useful life based on components with a shorter or longer useful life. For a given asset class, the expected useful life is that of the component with the shortest expected life. In the case of AMI, this is solid state electronics and communications interfaces.

Public utility commissions recognise that the capital costs for AMI systems are subject to obsolescence, and have adjusted the life over which the capital costs are depreciated. However, in many cases these adjusted rates do not sufficiently compensate for the advance of technology, especially microprocessors and telecommunications infrastructure. Traditionally, meters were depreciated over 30 years. In recognition of the shorter lifetimes of electronic devices in harsh environments and the implications for functional obsolescence, regulatory authorities are recently open to considering 15 – 20 years for AMI.

Typically, depreciation for AMI is based on a 20-year life. AMI components – meters and telemetry – are usually grouped together in a single asset class. Capital costs for data management infrastructure and software are grouped in a separate asset class and depreciation is based on a seven-year life. Company specific depreciation studies are necessary in order to fully recognise the functional obsolescence of AMI systems.

Many utilities, however, have decided to rely on refurbished electromechanical meters during a system- wide rollout of AMR partly because of the depreciation rules. The issues of reliability and useful life have largely diminished in importance. Once a communication module is put into a meter, the expected life of that combination (meter plus AMR module) drops to the shortest expected life of the two components. Many utilities suggest that an appropriate depreciation time today is 10 to 15 years due to the pace of technological innovation.

A new generation of meters and AMR is emerging every three years, and the pace is not expected to diminish for the foreseeable future. In many states the depreciation rules still consider metering as having a useful life of 25 to 30 years. Continuing to base depreciation rules on the useful life of meter technology of the last decade sends the wrong signal to utilities considering how to go forward. Regulatory bodies should adopt depreciation rules for AMI that reflect the shorter expected useful life of the new metering products, within an upper range of 10 to 15 years. The regulatory environment can both foster and hinder technology choices. It can encourage the timely deployment of solutions with technology solutions that fit the current and future needs; or it can prematurely mandate (or approve) the deployment of systems that are destined for technological obsolescence.

Regulatory authorities must be mindful of the impact of technology obsolescence on capital costs, and take actions to:

  • Provide for timely cost recovery of prudently incurred AMI expenditures, including accelerated recovery of investment in existing metering infrastructure, in order to provide cash flow to help finance new AMI deployment
  • Design depreciation rules for AMI that take into account the speed and nature of change in metering technology.

This discussion leaves aside issues related to the accelerated depreciation of the legacy meters that AMI replaces. All electromechanical meters are rendered obsolete by AMI and there is limited potential to reuse this hardware. A sensible approach would be to accelerate the depreciation of remaining assets to match the replacement schedule.

AMI investment2

Figure 2 – AMI evolution

In some instances, the investment in AMI can be justified based on cost savings from operations and other benefits that are measureable to the management of the utility itself. For example, PPL has one of the earliest and most successful deployments of AMR in the United States. The system was designed as AMI, but deployed as AMR. It had a 15-year net present value (NPV) cost of $198 million and a NPV benefit of $205 million. The benefits were entirely operational. When PPL decided to initiate a demand management programme, it asked regulators to adopt ratemaking policies that would provide incentives that would generate cash flow to help finance a “new” AMI deployment. This means assigning depreciation lives for AMI that take into account the speed and nature of change in metering technology.

However, many investments in AMI can be justified only by assuming substantial benefits based on demand management. In cases where the AMI investment is justified by the benefits attributed to demand management, the burden of risk should be placed on the rate payer. In the interest of fairness, when a regulatory body has a public policy goal of introducing demand management technology, it has an obligation to offer incentives to the utility to make the necessary investment in AMI and minimise the risks associated with the anticipated demand side savings. A reduced useful life for the costs of the AMI investment is the most effective way to match the risk with useful life.

The AMI investment process takes place over an extended period of time. Most often, it takes five years from the acceptance of the technology solution to full deployment. Technology is changing throughout this time period, and the process of technological obsolescence is underway. Technological improvements continue as time moves forward. By the time the technological solution that was originally selected is fully deployed, the technology is already five years or more on its way to obsolescence. For this reason, the depreciation life for the capital costs should be adjusted to the date at which the technology was proven and selected.

The argument for reduced useful life and accelerated depreciation is not based on technology alone, but as a means of minimising the risks associated with the realisation of benefits forecast for the AMI investment. This is the risk of economic obsolescence.

Thus, there are three major risks related to obsolescence with respect to AMI investments:

  • The risk that the benefits will not materialize
  • The risk that the technology will become obsolete before the benefits are realised, and
  • The risk that the regulators will change the rules on cost recovery and the rate of return.

Relying on a 20-year figure for useful life does not compensate for the risk of obsolescence, nor does it offer any incentive to deploy technology when it is just entering the mature stage, to ensure the maximum benefit to the public through deployments.

Considering both the realities of technological obsolescence in solid state electronics and telecommunications, and the need to offer incentives for investment in AMI, it would be sensible for utilities and regulatory bodies to adopt 10 years as the useful life for AMI. Metering technology should be assigned a useful life between 10 and 15 years. Since telemetry and communications interfaces have even shorter lives, they should be assigned a lower useful life, between three and seven years. Information technology products, especially data management software, are continuously upgraded; the depreciation schedule should be set to recover costs within two years of installation.

A useful life of 10 years for the combined technologies – metering, telemetry, and data management – would mitigate the risks of technological obsolescence and shortfalls in realising expected benefits. Given an authorised rate of return on rate base of 12%, a useful life of 20 years gives a payback period of 6.5 years; a useful life of 10 years would pay back in 5.1 years.

In many ways technological obsolescence, through its impact on the depreciation schedule, cost recovery and rate of return, is the most influential factor in the decision process for AMI investments. Understanding technological obsolescence is fundamental to making the commitments that will bring the benefits of AMI technology before it becomes obsolete.