Communications of the ACM

Computing ethics

Societal Implications of the Emerging Smart Grid

By Timothy Kostyk, Joseph Herkert
Communications of the ACM, November 2012, Vol. 55 No. 11, Pages 34-36
10.1145/2366316.2366328
Comments

Like other components of the nation's infrastructure the U.S. electrical power grid is deteriorating; the annual number of large power outages has been increasing since the late 1990s.¹ Though not as catastrophic as the recent blackouts in India, the increasing numbers, duration, and impact of power failures across the U.S. due to the degradation of the grid have severe implications for the energy-intensive way of life, economic stability, and even national security. The cost of neglect is high; a report commissioned by the Edison Foundation estimated that to retrofit the U.S. electricity infrastructure including new generators and new power delivery systems will require approximately $1.5 trillion over 20 years⁴ with more than half of this investment going to transmission and distribution facilities.

The proposed solution is widely known as the "smart grid." The increasing occurrences of outages and instances of cyber intrusions between 2000 and 2008 were considered so threatening to U.S. economic viability and security that the federal government, as part of the American Recovery and Reinvestment Act of 2009, earmarked more than $3.3 billion in smart grid technology development grants and an additional $615 million for smart grid storage, monitoring, and technology viability as an initial investment in building the smart grid. In addition, utilities have begun to mount demonstration projects and government and professional societies have begun the development of smart grid standards.

The smart grid will be comprised of three fundamental structural elements: replacement of aging core physical infrastructure items including transmission lines and switching equipment with more efficient and reliable newer technologies; two-way distributed and loosely coupled supply and demand connectivity to the grid, which allows consumers to supply electricity through technologies such as photovoltaic cells and wind power; and, most importantly, highly optimized two-way information and communication technology (ICT) systems architectures and networks that control the grid through process- and rule-based programs to match power demand with supply in order to improve efficient use of energy resources.

The fundamental differences between the existing grid and the smart grid are the ICT and distributed connectivity capabilities. While these innovative features of the smart grid hold great potential for improved energy efficiency through better management of consumer demand and improved stewardship of energy resources including greater utilization of renewable generation, they also pose a number of social and ethical challenges including: protecting the privacy of consumer usage information; securing the grid from attacks by foreign nations, terrorists, and malevolent hackers; and ensuring social justice both in terms of access and cost of electric power service. As with many new technologies the engineers engaged in developing the smart grid often overlook such issues or only turn to considering them once the technical standards and specifications have been settled. Failure to address these issues in a timely manner, however, may result in delays in establishing the smart grid and undermine its potential. Engineers and others involved in developing the smart grid need to examine ways to address organizational, social, and ethical dimensions that distributed generation and more extensive efforts to influence consumer usage patterns will raise. The cost of doing so would amount to an insignificant fraction of the projected necessary investments.

Back to Top

Privacy and Security Issues

As is the case for many other modern ICT applications such as the Internet and geographical positioning system (GPS), ensuring consumer privacy will be a challenge for the smart grid. Up until now our personal energy usage had been recorded by simple consumption metrics such as kilowatt hours measured using a conventional meter attached to a home or business. In the initial transition to a smart grid, utilities have begun to install "smart meters" that can provide feedback to the utility and customers on such factors as time of use of electricity. Since every appliance has a unique "load signature," smart meter data can be analyzed to determine the types of appliances and other equipment consumers are using.³ In the future, as more demand-side technologies are developed, the smart grid could have the capability to monitor and control the usage of every plugged-in electrical device, which would allow the electric utility to turn the device off during times of peak demand to balance load across the grid. For the privilege of acquiring data and controlling consumer electrical devices utility companies may charge a reduced rate. Alternatively, rate structures that vary by time of day or fuel source (coal vs. wind, for example) may be instituted in order to influence consumer energy usage behaviors.

---

The fundamental differences between the existing grid and the smart grid are the ICT and distributed connectivity capabilities.

---

As we move from theory to design, the emerging smart grid will become a vast ICT network populated with a diverse set of data acquisition devices capable of tracking the source, ownership, performance, and behavioral characteristics of each connected component. The smart grid technologies with the potential to be privacy invasive include "smart" power meters, energy monitoring and control software programs, and monitoring chips built into devices that consume electricity. In addition to control and monitoring functions, however, the smart grid will have the ability to collect, aggregate, and store individual consumer usage data such as the temporal pattern of electricity usage and the number, type, and usage of electrical appliances and electronic devices. Analysis of this data could reveal such information as home occupation patterns, the number of occupants, and the manufacturer and usage of individual devices—valuable to utility planners but additionally to marketing agencies, insurance companies (property, health, and life) and, potentially, criminals (for example, outsiders may be able to tell when a home is occupied, determine the type of security system, and learn other sensitive information). Much like the data acquired by supermarket bar-code scanners and loyalty cards, data on specific devices in homes and consumers' patterns of energy use will become a prized resource. Electric utilities or third-party vendors may sell personal data to other organizations to defray costs or simply to increase profits.

The PowerMeter application being developed by Google is an example of how third-party vendors may become involved in the management of smart grid data. An Internet-based application, PowerMeter receives information from utility smart meters and energy management devices and provides customers with access to their home electricity consumption on their personal iGoogle home page. Google is only one of many data-hungry organizations racing to develop smart grid monitoring equipment and data systems.

Of course, like supermarket loyalty cards, utility customers may be willing to give up some of their personal data if they think it is being used benignly and if they are getting something in return (such as reduced prices or rates). Up to now, however, utilities have not had to deal with consumer energy usage data on this scale; they and the public utility commissions that regulate them may be unwilling to incur the added expense of protecting consumer data from illegitimate uses or reassuring consumers that this data is protected adequately. The implications have not escaped the privacy watchdogs or even high-ranking U.S. federal government officials. Indeed, former Commerce Secretary Gary Locke warned that privacy concerns might be the "Achilles' heel" of the smart grid. Achieving public acceptance of the smart grid may prove difficult if privacy concerns are not addressed in a proactive manner.

Unsurprisingly, many security aspects of the smart grid look like those of the Internet. Although the Internet has not been designated as the primary source of ICT communications, the smart grid will more than likely mature into a system that will utilize the Internet as its backbone. To secure both the informational and power-carrying capacity of the smart grid two important features must be addressed: the physical security of power and ICT networks and equipment and the security of huge databases and computers that analyze the data. The smart grid of the future will integrate both these networks creating the ability for either one to cause disruption to the other. Examples abound where highly automated systems have been brought to a halt or damaged by failures or security breaches in their ICT backbones (such as failures in automated securities trading, cyber warfare damage to Iran's centrifuges for nuclear fuel enrichment, and malevolent hacking resulting in infiltration and shutdown of corporate and government Web sites).

---

Unsurprisingly, many security aspects of the smart grid look like those of the Internet.

---

Security breaches in the smart grid could lead to brownouts or even blackouts, and could cause serious, long-term damage to power generation, transmission, and distribution equipment. With the integration of power and ICT networks, power delivery components and even everyday power devices (such as appliances) will become nodes on the Internet. In the future, cyber attacks such as denial-of-service or virus attacks could cause outages in the smart grid and limit electricity supplies, including critical services such as infrastructure and public safety. These attacks could originate anywhere in the world and could start as easily as introducing false data regarding energy usage across many nodes.

What do these concerns mean for the development of security mechanisms, policies, and practices to secure the smart grid? There will be pressure to introduce a wider range of surveillance technologies; such technologies are already at the forefront of many heated debates regarding the intrusion of local, state, and federal governments, and also corporations, into the daily lives of individuals. Security and surveillance systems bring their own data needs, which promise to further erode personal freedoms, including privacy.

Back to Top

Pricing and Access

Though not as obvious as privacy and security issues, the smart grid also poses potential problems for equitable pricing and access to electric power service. The nature of these impacts will depend on whether consumer energy usage is left under utility control or consumers are allowed to make their own usage decisions under variable pricing schemes. The former case would limit consumer autonomy. One utility, for example, has already proposed that it be permitted to control customers' thermostats. Variable pricing, on the other hand, would place an energy management burden on all residential consumers. Those with lower educational levels, limited Internet access or computer skills, medical or cognitive impairments, or those who simply lack time, resources, or motivation to manage their usage patterns could be at a disadvantage. Both cases will require innovative ratemaking and oversight by public utility commissions and greater coordination and standardization within and among retail service areas. Though smart meter experiments are just in the beginning stages, there have already been regulatory and legal controversies over such issues as required prepaid service plans for low-income consumers and alleged price gouging under mandatory switches to smart meters.

Back to Top

Conclusion

Achieving the smart grid's potential while tending to privacy, security, and equity concerns should begin with the realization that the smart grid is a complex sociotechnical system that requires solutions that go beyond the engineering of the grid. Solutions must include thoughtful deliberation by federal and state regulatory agencies, flexible utility responses in addressing consumer concerns and, most importantly, an engineering culture that recognizes and addresses the societal implications of the smart grid upstream in the R&D process and as standards are being developed.

For example, while The National Institute of Standards (NIST) highlighted privacy concerns in a recent report,⁷ the U.S. federal government has yet to enact any smart grid privacy legislation or regulations. On the other hand, The California Public Utilities Commission's (CPUC) 2011 decision on protecting privacy and security of consumer data is a landmark ruling that should provide a strong template for other state commissions.⁵

One solution for addressing customer's concerns regarding the smart grid is to provide opt-out options, such as Pacific Gas and Electric's proposal to permit customers worried about the environment, health, and safety effects of smart meter wireless radio signals to request that the signals be shut off (albeit with a charge for conventional meter reading).² Willingness to provide such options may be necessary to ensure public trust of utilities as the smart grid develops.

As in the case of the human genome project and nanotechnology, where the U.S. federal funding agencies earmarked a percentage of research funds to examine such issues,⁶ there is an urgent need to examine the societal implications of the smart grid concurrent with its development. Failure to do so will further threaten civil liberties in the information age and is likely to pose substantial barriers to public acceptance.

Back to Top

Back to Top

Back to Top

The Digital Library is published by the Association for Computing Machinery. Copyright © 2012 ACM, Inc.

---

No entries found