“Smart grid” is defined as the application of information systems to the electricity system. A smarter grid means everybody has to get smarter, including utilities. Being smart means weighing benefits, costs, and risk. It means focus on scale and scope. The disruptive tendencies from new ‘smart’ technologies derive precisely from optimization at newly accessible levels of scale and of scope. A smarter grid does more things better at all system levels. The problem for regulation is to define allowable scale and scope for electricity utilities’ investments in new technologies. The utilities have a valuable incumbent advantage. They know and are known by their customers. This obligation to serve makes utilities natural and inevitable partners. Still, the natural monopoly ends at the wires; utility investments and expenditures have to be seen in that light. Despite all the talk about barriers, established regulatory principles can guide smart utility expenditures. It’s the technology that is new, not the regulatory problem. The ‘right’ regulatory approach, I argue, conventionally should weigh the balance of evidence based on need, costs, benefits and risks. A better approach yet is one which operates in a systematic multi-layered integrated planning context. New technologies are expanding the boundaries of what’s possible, including considerations not only of how to regulate, but whether to regulate at all. Either way, to deliver new efficiencies, regulation has to engage new technologies and new opportunities, deal with “the integration issues between new and legacy solutions and infrastructures”, and define roles for regulated utilities.¹

The application of information systems to the electricity system will allow real-time monitoring, communication, integration and optimization across nodes, grid layers and interconnected systems, and will lead to flatter spikes in energy consumption, lower fuel costs, reduced losses, quicker recovery from outages, improved power quality, and overall will reduce cost. The grid will be smarter because it will provide better service at a lower price.

“Several new trends are already shaping changes in the electricity infrastructure including the expansion of the existing grid with micro grids and mega grids, and new apparatus exploring new materials and concepts ranging from superconductivity and nanomaterials to highly flexible control and energy storage. Additionally, extensive sensors, communications, data processing, visualization tools, and infrastructures are being deployed. This is leading to smart grid concepts that primarily explore the integration issues between new and legacy solutions and infrastructures, which are the most demanding issues to resolve. The most prominent and complex integration issue is the full use of variable renewable generation, the electrification of the transportation sector, and interaction of the previous two factors with the electricity grid.”²

The smart grid is not primarily a hardware play. Making the grid smarter will require the deployment of technologies, but these technologies, for the most part, already are ubiquitous. There is a highly developed global supply chain for advanced metering devices, sensors, controls, breakers, relays and circuits. A growing number of global companies are poised to profit from the opportunities they see coming. Commercial, industrial and residential customers alike are being offered new energy technologies and services, and they are buying. Access to cheap new devices is not just for utilities. Customers themselves are realizing new more affordable opportunities for better energy management, seeing new vendors and new products in the market for alternative technologies, like distributed generation, energy storage, and dynamic energy management.

Smart means optimized. Technological advancements that reduce costs, and the integration of software and management tools to use them properly, are driving the smart grid market, creating new scope for optimization, efficiency and cost savings. Google’s recent acquisition of Nest for $3.2 billion is a serious bet on the value of home automation systems and the opportunities they can provide.

The digitization of energy management—where it can be achieved—offers value precisely because it creates an optimization opportunity that previously did not exist. The meters and sensors are cheaper, so more can be placed, showing patterns and options previously we could not see. Increased visibility is part of it; another part is real time control. Together, these new management tools create new options for management: optionality that has value.

“The paramount value of the devices, in a sense, lies not in the hardware itself but the interconnectedness of that hardware.”³

Utilities have made a business case for “smart grid” investments giving them increased network visibility and real-time control. The benefit is supposed to be better asset management, fewer and shorter outages, quicker restoration, and elevated levels of customer satisfaction. These benefits are well and fine, and a good start, but there is much more value to be had from an interconnected distribution and transmission energy network fully optimized.

Reduced transmission and distribution losses are an obvious source of cost savings if they can be reduced. Losses occur most during times of peak demand (according to Ohm’s Law). Most losses occur during a very few peak hours on hot summer afternoons. Curtailing or displacing load during these hours, with distributed generation or energy storage, will reduce demand, reducing losses exponentially at the margin, and will reduce costs to customers.

System constraints across transmission nodes, at the interties, and across nodes and connections in the distribution system, also drive inefficient costs, by requiring the dispatch of uneconomic generation, or by driving losses higher on constrained elements. In some areas, total losses from generator to consumer of more than 10 percent are routine. This loss is deadweight.

Power quality, the occurrence of voltage sags, drops and blips, also drives costs, tripping industrial processes, disrupting production, and increasing heat, losses, and wear on electrical equipment across the system.

At present, these costs are treated as unmanageable externalities, an inevitable part of a utility’s operations. As the grid gets smarter, that no longer holds. These costs can be internalized, and, what’s more, can be managed. At the margin there are substantial efficiency gains to be made, and significant sums to be saved. What’s missing, in most markets, is the money. These grid services are not distinctly procured, nor is there an obvious market in which to transact.

The system operator optimizes the dispatch, but the costs that arise from losses, constraints and reduced power quality flow through. The local distribution operator gets the bill from the system operator; their costs too flow right through to the customer. The typical household bill shows these costs as a fixed annual volumetric adjustment: the intrinsic dynamics and financial implications completely obscured.

Never mind the hapless consumer. The lack of incentives, the market failure, is systematic. No participant in the system has much motivation. System efficiencies that arise would be enjoyed in the commons: out of scope in conventional cost-of-service reviews. Lower demand is lost revenue for electric utilities. So we have lost revenue adjustment mechanisms but no “efficiency gained” revenue adjustment.

An efficient outcome from a competitive market would set price as revealed by transactions. Open and fair trade, attracting buyers and sellers, would drive prices to marginal cost, maximizing net benefits and overall surplus.

“A perfect electricity market would be transparent. Utilities would provide consumers with information about real time prices, and consumers would respond instantaneously to changes in wholesale prices. … This would be an evolutionary third generation of demand response, or “DR 3.0” This requires a seamless connection between the wholesale and retail marketplaces. At present, there are substantial barriers to this. There is a lack of infrastructure and appropriate retail rates structures, including “the lack of a direction connection between wholesale and retail prices,” says FERC Order 745, “… lack of dynamic retail pricing, and lack of real-time information sharing. Without both enabling technologies (such as smart meters) and policies (such as real time dynamic pricing) in place, individual consumers cannot respond to price signals from wholesale markets.”⁴

In Ontario, the Independent Electricity System Operator directs the operations of the high-voltage grid and Ontario’s interconnections with her neighbours. A few large customers are served directly at transmission voltage, but the majority of the customers and their energy needs are provided by local distribution companies, mostly municipally-owned, and regulated by the Ontario Energy Board. As well, in Ontario, there is the Ontario Power Authority that underwrites long-term contracts with generation companies, and delivers conservation programs for the Province.

If we systematically can visualize the physical infrastructure, then the relative interdependencies between layers become apparent, as does the discrete scale and scope of each layer. This structure, the intrinsic characteristics of the system, is what defines and constrains opportunities for system optimization.

At the top is the system operator, superordinate, whose scope is the high voltage system and its synchronization with adjacent systems. The system operator directs the operations of the grid (and the equipment connected to it, including generators, major loads, and relevant transmission elements) and operates a wholesale market for energy and ancillary grid services. This system operates in real-time.

One layer down is the local distribution company. Its operations are subordinate to the grid, but the distributor’s scope is complete in terms of the obligation to serve in its franchise, and it is an exclusive natural monopoly. Whatever happens there will never be two sets of wires. At this level, the utility has wide authority to invest in and upgrade equipment, undertake maintenance, and respond to outages and other events. The utility has a retail relationship with its consumers. Distributors generally do not own, operate or dispatch generation or direct the operation of customer loads.

In the current system model, efficient dispatch happens only on the grid side of the network. Whatever happens happens there in the real-time market.  Whatever the costs are they flow through to rate-payers. The utility might “own” the customer, as some say, but there is no real-time subsystem engagement, so the optimization potential is un-mined.

Everyone complains about barriers but nobody does anything about them. The system operator directs and dispatches the grid, balancing the system in real time, but it does not own or operate the grid. Wires companies (at all voltages) own and operate the grid, and balance circuits, but do not dispatch generation (or load). The regulator reviews wires company budgets and plans, and sets rates, but generally only once in a while,  looking out only a couple of years, and with limited reference to the utility’s role in the larger system. We need to assign not just the obligation to serve, but the obligation to integrate. Customers themselves are going to fund these investments one way or another; they need clarity on terms and conditions of service. Utilities need clarity on how much of this spend becomes a regulated asset.

Failure to adapt to a “smarter is better” world portends bypass, load loss, cost stranding and the death spiral. These were the stated concerns of the telephone monopolies in the seventies and eighties. New technologies in telecom presaged a wave of deregulation and economic disruption, yet even after 30 years, the major telecom companies are still the major telecom companies. Like the telecoms, electric utilities may need to consider new business models, innovations to repurpose business assets and to align business interests with those of their customers. The telecoms never were just wires companies, and neither are electricity utilities. They all start with scale and long-standing relationships of service and trust with their customers. Most customers never switched supplier after telecom deregulation; even with wireless competition the incumbents have maintained a huge advantage over new entrants.

Where smart grid technologies have been funded in other jurisdictions, for demonstration or research purposes, utilities have played a role. But it is one thing to encourage token levels of research and development; it is quite another to catapult significant new costs onto the system without proper cost-of-service review. Systematic capital programs and expenditures by utilities whose costs are recovered from customers have to be regulated, at least for now.

In a rapidly evolving technology marketplace, addressing the issue of guaranteed long-term viability is critical. Wires and poles sometimes will last for generations. For a 50-year asset, it may not be unreasonable to seek protection for long term capital recovery. For things like digital meters, sure to be obsolete in only a few years, long-term guarantees are not such a reasonable expectation. Provision can be made for this with accelerated depreciation and capital cost allowances, but higher near term costs must drive up hurdle rates if they are not to drive higher rates for rate-payers.

Some distributors have said they wanted to invest in smart grid technologies on their own merits, but these applications have not been well tested. In Ontario the only concrete proposal for an LDC’s investment in energy storage was withdrawn in the pre-hearing process.⁵ The costs are high, apparently, but there was virtually no estimation of benefits. Some utilities will have sufficient scale (and money) to contemplate undertaking some active control of elements and nodes of its network. Once we know the benefits, a basic cost-benefit assessment would provide some reassurance that we are making good investments, not simply wasting money for show.

The benefits from dynamic management, however, come from real-time signals, and that by and large is the scope of the system operator, suggesting a need for some integration. If a market were to be created, in which efficient prices could be observed, then one would expect those economics to drive the market.

What we should want from regulation, to the extent we must have it, just like from markets where they may work, are rewards and injunctions, signals for better risk management. System participants at every level should face dynamic market-based incentives for more efficient system operation, including in regulated sectors. A conventional approach to performance-based regulation provides some basic rewards based on productivity, service quality and performance. The regulatory framework, however, provides LDCs no incentivizes for more active optimization, for example to reduce losses and customer costs. Instead of rewarding total throughput, the revenue formula should be realigned to reward system efficiencies and lower costs to customers. Local system improvements should serve local customers, but since a smarter grid is optimized across its layers, local improvements need to be aligned to deliver overall system benefits. If distributors (and distribution customers) are to be bonused on reducing distribution losses, they similarly should be taxed on transmission losses they impose on the system at large.

A systematic role for regulation makes even more sense in the context of integrated regional resource planning. A holistic approach would consider the services required to deliver value grid-wide and system-wide, at the highest level and within sub-systems and micro grids. The rapidly declining cost of managing at higher levels of detail and complexity provides the opportunity to achieve benefits at much greater scale.

How might micro-grid (sub-sub-system) benefits and costs be allocated? If efficiencies flow up, i.e., a more efficient subsystem makes the whole system more efficient, then there is some public interest in promoting subsystem investments. Issues of differential regional customer classification and cost allocation are sure to arise in regional and sub-regional planning contexts. What is the process to deliver incentives or recover costs for public goods at each system layer? It makes no more sense to push all the costs up than it does to push them all down. It is not just relative perspective; individuals, local communities, and institutions (such as municipalities) may have quite different timelines than banks and commercial investors. A longer view affords a lower rate of return. Lower discount rates and longer paybacks improve economics for long-lived infrastructure investments.

LDC instincts to grow rate base and returns on equity don’t need much encouragement, but if a regulatory review is properly scoped, not only by franchise and test period but within a broader system context, then the extent to which LDCs should invest comes down to risk and return. If benefits exceed costs, investments should be justifiable.

The monopoly ends at the wires, but the utilities are naturally monopolistic in everything they do. Does allowing investment by monopolies bring monopoly problems: barriers to entry, inefficiency, and discriminatory pricing? Does regulatory treatment of smart grid investments within regulated entities foster innovation, attract appropriate investment, properly reward or insure risk, promote competition, etc. If we never make the investments, of course, we never will know, but if we are to allow anything then we must simply deal with these scoping issues as they arise. What are the utilities buying and for whom? Who benefits and who pays? This is classic regulatory fare.

The more important consideration is scale. A business confined to an unregulated affiliate has all the advantages of a new entrant, and none of the advantages of incumbency. We may feel safe behind this firewall, but it will take much longer to achieve any scale with this limitation in place. No new demand response technology has achieved meaningful scale at any level at all without a utility partner. It can be prohibitively costly to acquire leads and recruit customers without access to utilities’ customer lists and the use of their trusted brand. A hostile utility is a formidable barrier to change. As a practical matter, it’s not a question of whether LDCs can be partners; utility partnerships will be essential to early and ongoing success.

Is there a work-around? Might we prohibit monopoly investment but allow utility expenditures. The distributor would not buy or own the equipment, but would buy, lease or otherwise procure its services. This forces the capital investment decisions out of scope for the utility, and the regulator, and into the private domain. Either way, the utilities should be held accountable to costs and benefits, evidence of need and a balance of risk.

The utility’s obligation to serve, it might be argued, creates some onus on distributors to invest where economic to reduce cost and improve service quality. There’s nothing in the regulatory compact to encourage utilities to refuse self-improvement. There is precedent to share savings—rewarding efficiencies by splitting the gains between consumers and the utility. If we are to share gains, we also should consider injunctions, motivation for shareholders of less efficient utilities.

Utilities already fund public goods. But that’s not what this is about. The promise of these new technologies should be proved on their own merits, not on vague notions of the public interest. Efficient incentives to drive individual behaviour (at all levels of the system) will drive positive system outcomes. Creating a market and thereby attaching value to economic energy services will drive take-up of economic technologies, and reduce efficiency gaps. Maximum system benefits will be delivered when market participants, service providers, generators, customers and grid operators all are responding to efficient incentives to do the right thing.

* Adam White is President and CEO at AITIA Analytics Inc. and is the President of the Association of Major Power Consumers in Ontario. Adam is a sessional instructor in energy policy in the University of Toronto’s faculty of applied science and engineering. From 2003 to 2005, Adam was Vice President of Public Relations and External Affairs, and acting President and CEO of the Ontario Energy Association. Previously, Adam marketed power contracts for Mirant Inc. in Ontario, managed regulatory affairs and government relations for TransAlta in Ontario, and was Executive Assistant to the Assistant Deputy Minister of Energy and Senior Economist in the Ontario Ministry of Energy and the Environment.

¹  MLaden Kezunovic, James D. McCalley & Thomas J. Overbye, “Smart grids and beyond: achieving the full potential of electricity system”, (2012) 100 Proceedings of  the IEEE 1329.

²  Ibid.

³  Marcus Wohlsen, “What Google Really Gets Out of Buying Nest for $3.2 Billion”, Wired Business (1 January 2014) online: Wired Business http://www.wired.com/business/2014/01/googles-3-billion-nest-buy-finally-make-internet-things-real-us/

⁴  Joel Eisen, “Who regulates smart grid? FERC’s authority over demand response compensation in wholesale electricity markets” (2013) 4 San Diego Journal of Energy and Climate Law 101 at 125.

⁵ Ontario Energy Board, Toronto Hydro-Electric System Limited (Settlement Agreement) R-2010-0142 (Partial decision and order) (7 July 2011), online: OEB http://www.rds.ontarioenergyboard.ca/webdrawer/webdrawer.dll/webdrawer/rec/283982/view/

The details of the proposed energy storage project in the application can be found in the Toronto Hydro-Electric Limited pre-filed materials Exhibit A1, Tab 1, Schedule 1 at 4 of 9 and the details of the proposed Settlement Agreement attached to the Board’s July 7, 2011 decision, on pages 4, 6 and 13 of 22.