Today is an exciting time to be engaged in evolving the world’s energy systems. With the right supportive environment, adaptive business models could emerge that benefit both incumbents and new entrants. The result could be a safer, cleaner, and more prosperous world for all.
After extended graduate reading and research on modern electricity regulation, markets, and operations, I’ve concluded that there are several tools at our disposal that could accelerate the development of new capabilities in our energy systems. In this paper, I outline ten examples of design options that could significantly enhance and catalyze the transition towards a first-best energy future.
1. Locational marginal value
One of the main missing pieces to enabling market-based investment demand-side and distributed resources is a set of real-time signals that can be well understood and provide adequate incentives for network investment and utilization. Although such signals have not yet been implemented in practice, there are promising possibilities under development. For example, the CalTech Resnick Institute and Newport Consulting have suggested a model they refer to as Locational Net Value that takes account of direct benefits, avoided costs, and integration costs of alternative investments. The result would be a shift from deterministic engineering towards dynamic methods that incorporate locational benefits.
The MIT Utility of the Future project is developing Reference Network Models (RNM) that provide reliable benchmarks for the expected impact of investment and utilization choices on network costs. These RNMs could support long-term resource planning, ex-ante incentives, and ex-post cost allocation. The explosion of data availability, exponential improvements in processing power, and breakthroughs in artificial intelligence are significant learning-by-waiting windfalls for the electricity system. Policy measures have supported deployment of enabling technologies and the U.S. D.O.E. has emphasized the importance of grid modernization in its Quadrennial Energy Review.
It could be argued that a market-based solution would yield better investment choices than integrated resource planning, at least in the long-term. For example, quantifying the locational marginal value of different assets and services, combined with the availability of incremental investment alternatives, could decrease the need for lumpy investments such as transmission expansion and centralized power plants. Furthermore, the uncertainty caused by incremental capacity additions (i.e. demand-side and distributed resources) could render lumpy investments less cost-competitive due to investment risk. In June 2014, New York utility Consolidated Edison announced that a planned $1 billion substation upgrade in Brooklyn/Queens would be deferred to 2024 through demand-side measures costing $400 million to implement. The “Reforming the Energy Vision” (REV) regulatory overhaul in New York state aims to formalize these types of innovative investment alternatives through a market-based competitive process.
2. Efficient real-time signals
In a centrally-controlled electricity system, it may be possible realize the benefits of modernization with engineering models and automation. However, it is reasonable to expect that demand-side and distributed resources will necessitate market-based platforms to fully participate in real-time activities that are otherwise provided by centrally-dispatched resources. This is because the need for open access, data privacy, and consumer choice makes it difficult to implement a command-and-control approach.
In a market-based system, optimal real-time utilization of network and dispatch resources necessitates a signal to participants that guides their choices to the efficient equilibrium. Because electricity demand has historically been very inelastic, the electricity system has been designed, built, and operated according to an inefficient equilibrium. As a result, the utilization of network assets has been less than optimal. For example, in its REV staff proposal, the New York PSC has stated that the total rate of system utilization in the state—peak divided by average demand—is less than 60 percent. In recent years, peak demand has grown faster than average demand in New York and elsewhere.
Real-time signals can be transacted in the form of price-only (fluid and spontaneous) or price-quantity (fixed and pre-determined) arrangements. When the late Fred Schweppe wrote Spot Pricing of Electricity (1988), the pioneering work that competitive electricity markets were based upon, he recommended major reliance on price-only transactions with price-quantity transactions playing only a support role to deal with particular needs such as operating reserves. His rationale was that customers would be generally more satisfied when remaining in control of their energy choices. For example, while the use of a spot price combined with an interruptible contract (a type of price-quantity transaction) could enable corrections to be made for especially bad weather forecasts or plant outages, customers want the ability to override curtailment (at a cost) when needed.
Scarcity prices due to supply shortages or network congestion need to reflect current actual system conditions (i.e. extreme cost) and produce the most efficient outcome possible through the rational choices of market participants (i.e. demand reduction). In many electricity systems, price caps have been instituted by policymakers to protect customers and avoid political sensitivities. Administrative actions (e.g. out-of-market interventions) and market power control measures (e.g. offer caps) tend to suppress real-time energy prices during peak times. This limits incentives to invest in capabilities to meet peak demand, both supply-side capacity and demand-side. The availability of responsive demand could reduce the risk of removing price caps and limit the exercise of supplier market power.
3. Use-of-network charges
Without an adequate mechanism drive the investment and network utilization choices of participants, growing complexity in the electricity system due to new types of resources and loads, particularly at the distribution level, could cause operational challenges, perverse incentives, and inequities. An adequate use-of-network charge is needed to incentivize the efficient utilization of limited and costly network capacity.
The challenge of efficient use-of-network charges has been made most apparent by the rapid growth in distributed and intermittent solar photovoltaic generation. Incumbent utilities throughout the United States have argued and lobbied that fixed charges should be introduced for owners of distributed solar photovoltaic assets in order to avoid a reinforcing feedback loop of growing self-generation resulting in increasing network tariffs and vice-versa. Furthermore, owners of distributed generation may not be paying their fair share of network services, thus shifting the burden to other customers such as lower-income households that cannot afford solar technology.
Mismanaged growth of intermittent resources has been shown to cause a balancing problem referred to as the “duck curve”. As the sun sets, the evening demand peak begins and the result is a sudden needs for non-solar energy supply. The fast-ramping nature of this transition is costly and difficult for grid operators to manage effectively. In order to alleviate scalability issues such as the “duck curve”, use-of-network charges are needed to incentivize efficient behaviour. For example, West-facing solar panels could be valuable in helping manage the late afternoon transition away from solar energy.
Demand charges have been in place for commercial customers to keep their maximum demand (kW) to a minimum in order to avoid ratcheting charges. Some demand charges, referred to as “triads”, are determined based on the times of highest system-wide demand over a given year. Although such demand charges are effective today, they may become insufficient in the face of increasingly complex electricity systems with reverse power flows and highly heterogenous load profiles due to demand-side and distributed resources.
The Utility of the Future research team at MIT has developed an approach to Distribution Network Use-of-System charges (DNUoS) that leverages Reference Network Models (RNM) and the load profiles of each customer to determine economically efficient use-of-network charges. The proposed approach would assign network charges to customers based on their actual contribution to network investment and maintenance costs. Customer load profiles take into account a comprehensive set of operational considerations (e.g. real and reactive power) and result from installed capabilities (e.g. automation, rooftop solar, storage) as well as choices (e.g. demand response).
4. Differentiated reliability
The need to provide a uniform level of reliability in electricity markets makes providing reliable electricity service much more expensive than necessary. Because involuntary loss of power is costly to the economy, the electricity system is overbuilt such that wholesale level outages are limited to one event every ten years. This implies a Value-of-Lost-Load (VOLL) of nearly $100,000 per MWh. ERCOT conducted an analysis of the VOLL for involuntary curtailments and concluded that it ranges on average from less than $1,000 per MWh for residential loads to over $40,000 per MWh for small industrial loads that lack backup generation. Conversely, for loads that can be voluntarily curtailed or time-shifted, the VOLL has been shown to be less than $500 per MWh and often negligible.
Considering that in New York and other places average demand is flat and peak demand is increasing, there is today a real opportunity to begin implementing mechanisms that enable differentiated levels of reliability based on individual customers’ needs and preferences. For example, deferred and avoided infrastructure investments alone may justify policy action to deploy the enabling technologies required to support load response and islanding capabilities in the distribution network.
The most readily accessible way to provide customers differentiated reliability levels is to offer preferential rates for interruptible or curtailable service. This type of arrangement is what Fred Schweppe has referred to as a price-quantity transaction where customers choose their desired level of reliability by communicating the secure energy level they want, such that all usage above that level is at the interruptible price. Schweppe emphasized that under the conditions of the energy marketplace where there are a huge number of customers with large diversity in usage patterns and needs, the simple use of prices is far superior to a quantity control with respect to both reducing transaction costs and increasing benefits. Were customers to choose curtailable price-quantity tariffs, however, limited capacity during a shortage could be efficiently allocated because customers with higher VOLL would tend to choose higher reliability than others.
Value-added energy services such as microgrids and backup generation have become another promising avenue for differentiated reliability, especially for critical infrastructure. In 2011, Hurricane Sandy caused unprecedented damage to New York’s electricity infrastructure. As a result, there has been a growing nationwide movement towards better infrastructure resiliency and several large energy companies have announced the development of next-generation technologies to offer differentiated reliability.
5. Resource adequacy
Resource adequacy refers to the need for infrastructure investment today in order to maintain reliable electricity service in the future. This includes generation and network investments. Before power sector reforms in the late 20th century, vertically-integrated utilities managed resource adequacy according to their regulatory compact. Once deregulated electricity markets were established, it was found that administrative actions such as price caps prevented the real-time price from increasing high enough during times of scarcity to incentivize private investment in new capacity, particularly “peaker” plants that only run a few hours each year as needed. This has been termed the “missing money” problem and has led to the development of capacity markets for resource adequacy.
Capacity markets have been implemented and adjusted according to a variety of designs. In most cases, an annual auction is held to secure capacity investment to be available in the near future (approx. 3 years). The bids that clear the auction are awarded the equilibrium price for future capacity, which is usually a fixed payment per megawatt-year of capacity. Although capacity markets successfully address the missing money problem, they only guarantee “steel in the ground” rather than capacity in real-time as needed and thus necessitate additional penalty mechanisms to ensure performance. Furthermore, capacity auctions are complicated to administer and are vulnerable to the exercise of market power. Finally, capacity markets are out-of-market transactions that undermine real-time scarcity prices and thus complicate economically efficient demand participation.
Reliability options (RO) are call options sold from electricity generators or demand response providers to the system operator. ROs support resource adequacy by providing suppliers with guaranteed revenue for capacity (i.e. the value of the option). The RO call option places a cap on the price that a supplier can receive for the contracted quantity when the spot price exceeds the strike price. During times of scarcity, suppliers are obligated to pay the system operator the difference between the spot price and RO strike price for the contracted capacity. If a supplier fails to perform when needed, they still need to pay the system operator and are thus penalized. ROs differ from traditional capacity markets in that they 1) incentivize performance; 2) provide suppliers better revenue certainty; 3) mitigate market power via hedging; and 4) maintain efficient spot prices.
Real-time price signals are the essential starting point in addressing resource adequacy concerns such as “missing money” and should be improved by all available means.
6. Remuneration incentives
Recent developments in technology, business models, and customer preferences are stretching the capabilities of regulatory compacts between electricity regulators and distribution utilities. In order to maximize social welfare for customers, decrease revenue uncertainty for utilities, and incentivize the adoption of emerging energy technologies, remuneration for regulated entities needs to be enhanced.
Although in many places power generation has been deregulated to function through competitive markets, the distribution of electricity is widely considered a natural monopoly and thus has remained regulated. Electricity regulators constantly need to balance the ability of utilities to operate profitably with customer welfare, reliable service, and safety. As a result, regulators work closely with distribution utilities to establish remuneration and quality of service standards. This is made difficult by information asymmetry between regulators and utilities, uncertainty due to long time horizons, and moral hazard.
Regulators have historically set rates through either cost-based or incentive-based regulation. Cost-based (or cost-of-service) regulation is essentially a “cost-plus” method where the regulator establishes allowable expenses and determines a rate of return that the distribution utility can earn on its asset base. Incentive-based (or performance) regulation establishes a revenue cap that is linked to an index (i.e. inflation) and is subjected to adjustments for expected efficiency improvements and uncertainties.
There are several improvements to these models that would support innovation:
- Decouple total revenues from electricity demand (quantity sold) to remove the disincentive for distribution utilities towards energy efficiency improvements
- Lengthen rate case cycles (e.g. to 8 years) to allow more time for distribution utilities to capture more value from their realized efficiency improvements
- Utilize Network Reference Models to reduce uncertainty and establish forward looking benchmarks that accommodate expected deviations and mitigate errors
- Replace CAPEX and OPEX with a TOTEX approach that reduces the incentive for unnecessary investments and supports innovation (e.g. SaaS products)
- Offer distribution utilities an incentive menu of contracts to encourage innovation and share the resulting benefits and costs between utilities and their customers
It is important to note that these improvements are largely relevant in determining total revenues for regulated utilities, rather than cost allocation through customer tariff design.
7. Unbundling of distribution
When competitive wholesale markets were established in the late 20th century, the introduction of Transmission System Operators (TSO) was imperative to ensure that open and fair network access would be maintained as mandated by PURPA (1978). Furthermore, vertically integrated utilities were required to divest their portfolios of generation assets in order to mitigate horizontal market power.
In response to recent technological developments enabling growth of demand-side and distributed energy resources, a distribution-level restructuring process has been underway in various states and countries. In order to maintain open access and prevent abuses of market power, the distribution value chain will need to be carefully unbundled. Achieving a well-functioning outcome could require the restructuring of some distribution utilities.
While there are many definitions of the different roles required to effectively operate an unbundled electricity value chain, five types of actors are common:
- Transmission System Operator: manage network investment and maintenance, coordinate network flows and transactions, and uphold reliability standards (HV)
- Distribution System Operator (DSO): manage network investment and maintenance, coordinate network flows and transactions, and uphold reliability standards (MV/LV)
- Energy Suppliers: contract for wholesale energy supply with power producers, sell energy to retail customers, structure physical and financial hedging instruments
- Service Providers: provide demand-side and self-generation services to customers, aggregate resources into virtual energy services, provide energy services to DSO
- Investors: parties that finance and own generation, transmission, distribution, and demand-side assets through the allocation of liquid and tradeable utilization rights
Unbundling of distribution operations from retail energy supply and services is essential to the well-functioning of competitive markets for electricity distribution. For example, insufficient unbundling could lead to information asymmetries on customer and operational data, preferential access for incumbent providers, switching costs for customers, and other anticompetitive practices.
Divestment from the ownership of certain assets (e.g. customer meters) by the DSO could become necessary to avoid moral hazard and anticompetitive behavior. Reference Network Models could facilitate third-party ownership of merchant distribution network investments. A second-best alternative to divestment could be regulatory oversight, which may incur higher transaction costs and inefficiencies than unbundled competitive markets.
8. Value-added retailing
It has been argued by Paul Joskow that “success of retail competition should be measured by the valued-added services it brings to the system. These services include enhanced metering and control technologies, price and consumption hedge contracts, total energy management services, bundling of different services, and other innovative services that Energy Suppliers and Service Providers could create and endeavor to convince consumers to purchase. Retailing costs and profits could then be recovered through the higher prices that consumers may be willing to pay for these value-added services”.
Until recently, most competitive electricity retailing has pertained to energy supply, trading, and marketing—value-added services have been limited. For example, competitive retailers have often relied on the incumbent utility for customer service. In many cases (e.g. New York), competitive retailers have struggled to be competitive with the default regulated tariff, thus necessitating “shopping credits” to incentivize customer switching.
The emergence of value-added energy services creates possibilities for competitive energy retailing that promise to satisfy Paul Joskow’s argument:
- Energy Suppliers have an increasingly broad variety of offerings for customers including clean energy procurement and time-varying tariff designs
- Service Providers offer an expanding array of services including demand-side management, distributed generation, storage, EV management, and microgrids
- Innovative companies are developing platforms to bring together various energy products and services and facilitate the activities of suppliers and service providers
- Connectivity and artificial intelligence are enabling a revolution in the aggregation of distributed energy resources into competitive virtual energy services for the DSO
- Customers are empowered to make informed choices to optimize their energy investments and utilization according to their own needs and preferences
Given the multitude of parties involved, “retailing” may eventually refer to direct customer relationships. Energy retailers could be a combination of energy supply and services, whether in-house or contracted through third-parties. Some retailers could have their own differentiated intellectual property (e.g. software platforms) as a source of competitive advantage. Energy retailers could provide financial products such as equipment leasing and hedging instruments for bill protection. Brand equity, financial resources, and network effects may become important barriers to entry for energy retailing. Market power and antitrust concerns could arise if energy retailing were to consolidate and become highly concentrated (i.e. monopoly or oligopoly).
9. Access to information
Much like open access to the transmission network is an unsurmountable entry barrier for bulk electricity generation, access to information could become an important prerequisite for participation in emerging energy services. Concerns such as customer privacy and infrastructure security require a careful approach to energy data management.
Energy data, as a resource, is non-rival in consumption—usage of information by one party does not prevent others from using it as well. Furthermore, energy data is only excludable (i.e. restricted access) to the point where one would be willing to incur the cost of obtaining it themselves (e.g. sensors on drones). In the context of the energy system, information can be qualified as either technical with security implications (e.g. aggregate real-time operational data) or commercial with privacy implications (e.g. individual customer transactions).
Access to information may need to occur through a data exchange architecture which could be designed in numerous ways. Three examples:
- Regulated and Centralized (e.g. secure hub administered by DSO, FERC, or FCC)
- Commercial and Differentiated (e.g. subscription-based information exchanges)
- Open and Distributed (e.g. peer-to-peer network of encrypted packets)
Qualifying processes could be developed to provide differentiated access, granularity, punctuality, or control according to overlapping sets of interfaces (e.g. TSO-DSO, DSO-[Energy Service Provider], ESP-[Building Management System]). This tiered access could be related to security needs, willingness-to-pay, or technical capabilities.
Privacy could be contracted through anonymization and protected through data encryption. Service Providers could design networks to self-anonymize, localize sensitive computations, and scrub their data transmissions to the lowest level of granularity required for energy services—according to regulations, standards, and customer preferences.
Security threats could result from 1) increased access to data (i.e. more people, more sources, more processing power); 2) broader use of control technologies; and 3) collection of or access to large repositories of historical, granular, and comprehensive technical data.
Cybersecurity vulnerabilities may evolve rather than amplify, as the current state of energy infrastructure is highly centralized and relies on legacy technologies. Encryption technology and network security may or may not outpace computational processing power and technological innovation.
10. Collaborative restructuring
As outlined in the Edison Electric Institute’s “Disruptive Challenges” report, the credit rating of distribution utilities as an investment class has steadily declined since the 1970s and recent decline has been more pronounced. The history of electricity deregulation in the U.S. has shown that rent transfer discontentment (i.e. “stranded assets”) can cause significant friction and lead to the implementation of “second-best” solutions that often become the next problem.
The transition to a new industry structure can be done in a collaborative manner that reduces friction between shareholders and change agents:
- Incentives could be aligned for efficient adoption of network modernization, demand-side management, and carbon-free generation
- Electricity regulators could gradually adjust the revenues of distribution utilities to reduce volatility and shareholder uncertainty
- Incumbent providers could be supported with accelerated depreciation of assets and restructuring protections to tamper volatility in equity valuations
- Adaptive business models and horizontal spillovers from technology development could provide incumbent regulated distribution utilities an opportunity to innovate with the safety of cost recovery and fixed return on equity (i.e. regulatory support)
- Discovery-driven capability development processes with iterative learning and milestones could accelerate adaptation to the evolving landscape
- Partnerships with third-party innovators could grant incumbent regulated utilities access to new capabilities
- Acquisitions could help incumbent utilities secure a competitive advantage
- Divestment from some legacy businesses could unlock previously unavailable possibilities (e.g. geographical expansion)
- Unbundling of services could result in efficiencies from the differentiated risk-return characteristics of regulated and competitive activities
- Stranded value could be shared between shareholders and society. Investor confidence in the U.S. power sector could be maintained, credit ratings improved, and cost of capital lowered.
Shareholders could benefit through equity returns and new opportunities for investment. Society could benefit through better energy services, lower costs, and sustainability.
Today is an exciting time to be engaged in evolving the world’s energy systems.
Adoption of modern energy technologies will require a system-wide effort. This paper outlined ten examples of design options that could significantly enhance and catalyze the transition towards a first-best energy future.
- Locational marginal value
- Efficient real-time signals
- Use-of-network charges
- Differentiated reliability
- Resource adequacy
- Remuneration incentives
- Unbundling of distribution
- Value-added retailing
- Access to information
- Collaborative restructuring
Much is at stake in today’s energy landscape—more changes have occurred in the past ten years than in the 100 years prior. With the right supportive environment, adaptive business models could emerge that benefit both incumbents and new entrants. Some companies have already seen the opportunity and pushed ahead of their peers. The future looks bright for those who have faith in the power of human ingenuity. The result could be a safer, cleaner, and more prosperous world for all.