Smart Grids: The Regulatory Framework Is Still Not In Place

By Bruno Lapillonne and Nicolas Brizard, Enerdata
December 2012

Smart grids are seen by many as an effective solution to address some of the toughest challenges the electricity industry has faced so far; the integration of renewables on a very large scale, the promised rise in number of electric vehicles, the necessity of energy efficiency, the improved security of supply or the arrival of the ‘prosumer'. Equipment manufacturers and IT solution providers are eagerly awaiting the hundreds of billions of euros to be invested over the next decades. In this article, we advocate that smart grid technologies have the potential to transform the electricity markets given they are for the most part readily available, but, the correct regulatory framework first needs to be put in place. Failure to recognise the need for a regulatory overhaul can only hamper and delay the deployment of smart grids and their expected benefits.

From vertical integration to network unbundling

The present European electricity system is the result of a process that started shortly after World War II. National or regional vertically integrated monopolies rapidly became the dominant business model in the electricity industry. This model proved highly efficient in developing the numerous European electricity networks in times of vigorous growth.

To this day, the European electricity system has been characterised by a high degree of centralisation with mostly unidirectional electricity flows. In the current configuration, large-scale power plants generate electricity that is transported over long distances through a high voltage grid and distributed locally to end-customers through medium and low voltage distribution grids.

Following the market liberalisation experience initiated in the UK and the US in the 80's and 90's, continental European electricity markets have been progressively liberalised. The European Commission itself has pushed the European electricity supply industry towards unbundling through a series of Directives, the last of which - the so-called "third energy package" - came into effect in March 2011. As a result, European electricity systems now comprise a mix of regulated and competitive elements. Power generation, wholesale supply and retail supply have become competitive segments of the value chain while transmission and distribution have remained regulated businesses because of their natural monopoly characteristics.

figure
Evolution of network regulation objectives
Source: Enerdata

The primary objective of market liberalisation was to lower costs for users. Accordingly, the first regulatory phase that followed unbundling was geared towards a cost-efficient management of existing grids through the minimisation of operational expenditure (OPEX) and the rationalisation of investments. This economic objective was to be achieved without endangering the quality of power and the security of supply.

For regulators, the main challenge is how to introduce new objectives such as the integration of renewables on a large scale, the enabling of demand side response (DSR) and energy efficiency.

Network regulation now needs to integrate the renewables and Distribution System Operators dimensions

Electricity grids are now evolving under the influence of a multitude of factors:

  • The development of distributed generation (renewable or not) and its connection to the distribution network, the least resilient part of the electricity network;
  • The increasing share of intermittent power generation sources in the generation mix (wind, solar PV);
  • The emergence and broad diffusion of new technologies or concepts, such as distributed storage technologies, heat pumps, micro-CHP, micro-grids or virtual power plants;
  • The steady trend in the electrification of energy demand caused by the digital society, wealth effects and the superiority of electricity as an energy carrier;
  • The increased peak demand fluctuations of electricity (new uses of electricity are peak load rather than base load uses);
  • Possibly, the large scale diffusion of electric vehicles in the future.

On the supply side, smart grids will facilitate an active management and optimization of intermittent sources and mitigate, at least partially, the cost of backing-up intermittent generation. By limiting wind curtailment as well as the frequency of negative or zero electricity prices, smart grids will enhance the value of intermittent energy sources. On the demand side, it is crucial to unleash the full potential of demand response through a better adjustment of the load to price signals or grid operators' requests.

A widespread diffusion of new distributed generation technologies will also create a shift from a centralised control and balancing system to a configuration that allows and coordinates a large number of players and a myriad of decentralised dispatching and load-shedding decisions.

figure
Market segmentation of the smart grid eco-system
Source: Enerdata

In other words, future grids will need to accommodate two-way flows of electricity between the various market players: utilities, consumers, ESCOs, producers and grid operators.

The emergence of smart grids imply new and additional costs for grid operators

The existing electricity grid is still mostly operated in a passive way using mechanical controls to manage the system frequency. It is calibrated so as to meet the most extreme technical (e.g. unplanned outages) and weather conditions (e.g. cold waves) as well as the demand from mostly passive customers. Grid balancing is highly automated but still lacks a dynamic optimisation of supply and demand, in particular at a local level (Distribution System Operators).

Networks will have to become smarter and at the same time retain high levels of security, quality, reliability and availability. Future investment costs therefore comprise both "conventional" and "smart" components. New technologies and costs associated with the emergence of smart grids include:

  • Smart meters;
  • New electro-technical devices (sensors);
  • IT and communication infrastructure (ICT): hardware equipment and software for grid management and operation;
  • Additional and extended computational capacity to deal with increased data flows (e.g. real time metering and billing);
  • SCADA (Supervisory Control And Data Acquisition) systems;
  • Complementary components such as storage devices.

In the absence of a clear definition of the scope of smart grid investments, their costs and benefits are difficult to assess. Furthermore, many smart grid technologies are still in their infancy so their performance and longevity remain uncertain. On the other hand, the cost of ICT tends to decrease faster that the cost of conventional grid technologies.

To our knowledge, the only comprehensive attempt to evaluate smart grid investments costs and benefits was carried out by the US-based Electric Power Research Institute (EPRI) in 2011.1 Although the US electricity system is approximately 20% larger than the EU network in terms of electricity generated and distributed, the EPRI analysis provides an interesting point of comparison and a good proxy for what smart grids might cost in Europe.

Costs assessed by the EPRI include the infrastructure to integrate distributed energy resources and costs to achieve full customer connectivity. They exclude the cost of generation, the cost of transmission expansion to add renewables and to meet load growth. They exclude some (but not all) customer costs for smart-grid ready appliances and devices.
In total, the cumulated investment costs required to enable a fully functioning smart grid in the US is estimated between $334bn and $476bn over the next 20-years. The EPRI also estimated the benefits of the smart grid over the next 20 years to range between $1,294bn and $2,028bn with a benefit-to-cost ratio between 2.7 and 6.1.

figure
Estimated costs over the next 20 years for implementing a US smart grid (Billion $)
Source: EPRI, 2011

It is clear from the EPRI analysis that most of the costs will be borne by distribution grid operators (approximately 70%). Transmission grid operators will bear between 20 and 25% of the costs and end-users less than 10%. As a consequence, most of the cost will fall onto residential and commercial customers. Their bill is expected to increase by an average of 8.4% to 12.8%. Industrial users are not expected to be significantly impacted.

figure
Smart grid cost to consumers: % annual increase in monthly bill
(based on 10-year amortisation)
Source: EPRI, 2011

The current regulatory framework does not favour the emergence of smart grids

The bulk of smart grid investments are being placed under the responsibility of regulated businesses (especially DSOs) so the design of regulatory models is crucial in terms of creating incentives to invest in smart grid technologies and solutions. In this respect, there is a high risk that power grid companies keep to traditional approaches because they are well understood and recognised by regulators.

Both cost-based and incentive regulation models are primarily aimed at achieving cost-efficiency and are not designed to promote innovative investments, high levels of R&D or even high quality targets. Regulatory models are generally designed with the aim to keep investment and operational costs under control and to have the lowest possible network tariffs, given the necessity to guarantee power quality as well as grid stability and integrity. They provide very strong incentives to lower costs through better investment spending discipline and operational efficiency. Inadequate regulation can counterbalance the perceived benefits of smart grids and skew investment decision-making towards a conservative attitude and a strictly cost-efficiency approach. Decision-makers at DSOs have to deal with tangible short-term costs and less tangible long-term benefits.

If grid operators, on which the burden of smart grid costs is laid, cannot recoup their investments, smart grids will certainly not develop. Already, grid operators do not seem to be adequately remunerated for their conventional investments. A 2007 Eurelectric report showed that three quarters of the surveyed DSOs had a return on invested capital lower than their WACC (Weighted Average Cost of Capital) thus destroying shareholder value.

Another difficulty is that electricity networks have so far performed well in terms of reliability. A challenge for grid companies and regulators is therefore to demonstrate and communicate convincingly that smart grids are an attractive value proposition in face of the costs incurred.

Regulators need to develop a number of complementary regulation techniques or schemes to overcome the limitations of price or revenue controls and make sure that cost-efficiency is not reached at the expense of system integrity, service quality or innovation.

Contours of a regulation regime favourable to smart grid investments

Near future regulatory regimes favourable to smart grid investments will most likely be adapted from the current situation to accommodate new technology and market requirements. Some new players will emerge but overall, the business models of TSOs and DSOs are likely to remain broadly similar to what they are today. Such an "enlightened regulation" model is probable in the short and medium term, i.e. until 2020 or 2030.

In order to promote smart grid investments, the "enlightened regulation" model will have to include some of the following characteristics:

  • Inclusion of smart grid investments in the regulatory asset base and assurance of a fair rate-of-return for DSOs;
  • Incentive regulation needs to be adapted to take into account the specifics of smart grid technologies (e.g. shorter economic lifetime, higher cost of smart components vs. conventional ones, etc.);
  • Introduction of "output regulation" elements: explicit output objectives and corresponding performance indicators allow regulators to concentrate on the outputs of the regulated entity and not be dragged into detailed input monitoring;
  • Regulatory stability and clarity: networks investments have a very long technical or economic lifetime so regulators should ensure long-term regulatory stability and visibility. Regulatory risk is probably one of the strongest deterrent to capital-intensive investments;
  • Reduced complexity of the regulatory incentive schemes and benchmarking techniques. If these are unclear, too complex or subject to interpretation, it becomes difficult for grid companies to develop a solid business case;
  • Regulation should remain "technology neutral": no attempt should be made by regulators to "pick winners", i.e. to choose or promote specific smart grid technologies and configurations because they do not have the expertise or the manpower to select the right technologies. Regulators should leave this task to grid companies.

Beyond 2030, one should not exclude the advent of an "internet-type" regulation, i.e. a system of distributed control placed under a global protocol. That would entail a complete overhaul of the existing situation and a regulation-light approach to grid management. Such a system would be highly decentralised at all levels with a large number of new players and business models.

R&D and demonstration projects: The limitations of regulation

It has been observed that deregulation and unbundling of regulated industries often leads to lower and often sub-optimal R&D levels.
Smart grids are not fully proven concepts yet, even if most of the technology bricks already exist. Much still needs to be done in terms of R&D and demonstration projects if start grids are to become a reality. As grid companies are traditionally risk adverse and may be reluctant to abandon a proven business model for a riskier model driven by innovation, regulators and policymakers will have to design ad hoc innovation and R&D funding schemes.

Incentive regulation alone is not sufficient to drive significant R&D spending and large scale demonstration projects. In other words, R&D expenditures needs to be differentiated from capital expenditures aimed at grid replacement, grid extension and even grid improvement with first level smart grid technologies.

As a general rule, incentives to innovate and engage in R&D or demonstration projects should not be paid by the customer. Because R&D offers no direct and measurable benefits for the customer, it should not be considered as a recoverable cost through regulated tariffs.

As a consequence, policymakers need to promote directly and specifically the R&D and demonstration projects that will support the development of the 21st century electricity networks.

Ensuring an adequate and supportive regulatory framework is a prerequisite for the emergence of a healthy and vibrant smart grid business and ecosystem

It is widely considered that traditional cost-based and incentive regulation models fail to provide adequate incentives for innovative grid investments. As a result, grid companies tend to under-invest and may be reluctant to deploy smart grid technologies that represent immediate costs but uncertain and distant returns.

Treading a fine line, national regulators will have to quickly find the right balance between two main concerns: the need to contain network costs, an objective that remains fully valid, and the need to enable smart grid investments.

Time is running out to reap the full benefits of smart grids before 2020. It is now unlikely that smart grids will contribute much to the realisation of energy efficiency and renewable energy targets for 2020. So far, only a handful of EU countries have started to reform network regulation, the UK being certainly the most advanced with Ofgem's RIIO model.2

Transmission and distribution grids undoubtedly represent the backbone of any future smart electricity system. Smart grid services and products which are not part of the regulated network activities such as home automation, small distributed generation, aggregation services, smart appliances and in some instances smart meters will only develop and reach their full potential if the main parts of the grid can support and integrate them.

1 Estimating the costs and benefits of the smart Grid: A preliminary estimate of the investment requirements and the resultant benefits of a fully functioning smart grid". Technical Paper. EPRI, 2011

2 Ofgem's performance-based RIIO model seeks to ensure consumers get the necessary investment in Britain's energy networks at a fair price. RIIO ensures that network costs do not rise any more than they need to by financially punishing inefficient companies that fail to deliver for consumers.