Knowledge as a Resource and R&D
Load serving entities (typically, utilities), do not perform very much traditional R&D resulting in patents or research papers. In many ways, this makes sense, as utilities use technologies and equipment supplied from power generation and other suppliers to implement their services.
However, many LSE’s are members of the Electric Power Research Institute (EPRI), a research consortium formed, in part due to outcry after a large blackout in 1965. According to EPRI, “Congress was troubled by the nation’s dependence on a fragmented, critical industry for which there was no unified planning or research.” EPRI research results is usually available only to members.
Utilities also benefit from innovation in operations research. Dispatching an array of resources, subject to their own constraints as well as the various contractual terms represents a complex and difficult optimization problem. Similarly, planning investment to meet uncertain load growth (or reduction), at minimum cost (or at suitable low cost but with a minimum of cost variability) and over different time horizons is also challenging and an area of ongoing research. Several optimization models for resource planning have been developed in response to environmental concerns, increased competition and growing uncertainty. Hobbs (1995) proposed a conceptual resource planning framework which encompasses multiple models performing different functions (i.e., load forecasting, rate forecasting, cost minimization, etc.) while considering a wide range of short-term and long-term uncertainties that utilities face in resource planning. In recent years, more research has been focused on the application of artificial intelligence and machine learning enabling utilities to make real-time decision on how to best allocate energy resources.
A load serving entity, like any business, will develop knowledge of its own operations. This includes things such as:
- Familiarity with load patterns among its customers, including where and how fast load is growing
- Familiarity with the transmission system overlaying its territory, including relevant constraints of that system as it pertains to the ability to deliver power to their customers
- Knowledge of the rate structure they offer their customers, and how changes to its structure will affect different classes of customers
- Knowledge of the distribution system, its cost, extent, reliability, etc – especially if the LSE is a vertically integrated utility. A CCA or REP may not know anything about this
- Knowledge of its business systems, including customer metering, customer billing, and the systems for interacting with the wholesale market.
Though such knowledge might not be considered “innovation” and much of it is unique to a specific LSE’s circumstances, it is nevertheless necessary to normal function of an LSE. Therefore, one might expect LSE’s work to preserve such knowledge within their organization, even if keeping it within their organization is not critical.
Market and Financial Knowledge
Utilities do use short-, medium- and long-term forecasts of electric demand, electric prices, and fuel prices in order to make financial decisions. These forecasts can be purchased from external sources, but it is also common for LSEs to generate their own forecasts using their own models, or by synthesizing or adjusting external forecasts with their own models. If IPPs had ready access to a utility’s price forecasts, they could use them for advantage in negotiations, so such information and models are usually closely held. LSEs subject to regulation may have to provide such forecast to their regulators, but under confidential terms.
[[ MT has info from solar paper on IOU R&D ]]
Utilities also are often directed to perform research by their governing bodies (PUCs, energy commissions). The programs may be funded by a surcharge added to utility rates, and the research program may be executed directly by the LSE or a contractor. These programs investigate new capabilities such as distributed generation, as well as the value and efficacy of demand response programs, energy efficiency programs, experimental rate structures, etc. Sometimes such research involves experimenting with new tariff structures, such as real-time pricing, time-of-use pricing, adding or removing demand charges, creating or eliminating tiered block pricing, etc. Paid for by ratepayer surcharges, the outcomes of this research is usually public, and LSEs certainly can and do learn from each other’s work. Rates, of course, are public knowledge, but the impact of rate structure on customer behavior are generally not unless published.
Some LSEs collaborate with academic researchers, for example, sharing anonymized customer energy use data which researchers can use for statistical inferencing, natural experiments, and planned experiments (e.g., Borenstein 2007 , in which the researcher relies on commercial and industrial customer data provided by PG&E in order to study impacts of tariff structures, and Borenstein 2015, which involves analysis of PG&E residential retail customers).
LSEs may constantly be creating and refining the types of knowledge described above, but given the mature and monopolistic nature of the industry prior to deregulation, one might expect such knowledge creation to be slow. Indeed, increased innovation is often cited as a primary benefit of deregulation and increased competition.
In deregulated markets, we might expect to see higher rates of knowledge creation, as new entrants experiment with ways to lower costs and improve their products. For example, REPs might adopt completely new rate structures in order to encourage the customers to behave in a certain way that lowers costs overall.
LSEs share operational knowledge through research consortia like EPRI as well as trade groups such as the Edison Electric Institute (EEI). Because so much of their activities are governed by PUCs, and because PUCs requirements vary from state to state, LSEs have incentive to pay particular attention to the new knowledge from utilities under the same jurisdiction as their own. Much of this information can be gleaned from the publicly-available dockets of various proceedings held at PUCs.
That said, much LSE knowledge is shared and disseminated primarily internally to the LSE itself, through formal and informal training and collaboration. Formal training in utility operations or management is uncommon, so most experts at utilities are at least partially “home grown.” We note that there are a few colleges that offer degrees related to electric utility management (e.g., New Mexico State University Electric Utility Management Program , Thomas Edison Energy Utility Technology degree, etc.), but these are relatively rare.
Barriers to Knowledge Transfer
As has been discussed, traditionally, LSE’s did not compete with each other, so there was little need for them to hold operational and technological knowledge closely. (As distinct from commercial knowledge such as contract terms.) On the other hand, because LSE’s operate under such varying constraints, it is likely that at least some operational knowledge is not highly transferable in any case.
An interesting potential barrier to knowledge transfer within and between LSEs are legal barriers that separate certain functions of utilities. For example, FERC order 717 and others that follow require that certain utility activities, such as wholesale energy transacting and transmission planning, be kept separate from each other, because allowing collaboration between those areas of a firm would give it a commercial advantage over competing firms making commercial transactions over the same transmission resources. By definition, these constraints are relevant for the parts of utility operations that are open to competition. In this case, the restrictions on knowledge transfer are focused on commercially relevant knowledge (contracts, plans, etc) rather than know-how, and are intended to serve competition rather than hinder it.