PUC advances CUB framework to enable energy data to be used in the public interest

Published July 22, 2020

Last Thursday, the Minnesota Public Utilities Commission (PUC) took steps to make electricity data available for research and analysis that can reduce waste, save consumers money, and cut emissions. We are pleased by this step and are confident that the ongoing effort the PUC has set up will provide critical data to inform decisions while protecting the consumer’s individual data and maintaining their privacy.

Last year, CUB proposed Open Data Access Standards that would allow third parties like us to request – and a utility to provide – de-identified data on when and where customers use electricity (PUC docket number 19-505). This data can be used to investigate different rate designs and programs. It can allow us to design rates to maximize the efficiency of the utility system, saving consumers money. It can show us how various options will impact real people, so that we don’t have to rely on common knowledge assumptions or draw conclusions about an “average” customer. It can help us understand how to design and target programs to help particular customer segments. And once the data is available, creative people will, no doubt, find uses for it that we haven’t yet imagined.

Over the past year and a half, CUB has continued to work with more than a dozen parties to refine our proposal, and we presented an updated draft of the standards to the PUC last week. We recommended the PUC continue to consider this draft by requesting comments from parties on specific questions and seeking outside technical expertise.

The PUC adopted these recommendations, and voted to allow access to whole-building aggregated data for building owners and benchmarking purposes according to the standard that CUB proposed (details below). Importantly, commissioners recognized the significant benefits of the type of data access that CUB proposed, and they indicated that it is a question of how -- not if -- additional data sets can be de-identified and shared with third parties. 

Protecting consumers’ privacy

Many parties to this docket were cautious about setting data access standards, and appropriately so. For any number of reasons, individuals or businesses might not want information about their energy use to be out in the world. By order of the PUC, utilities are prohibited from sharing data without customer consent, unless the utility has “adequately protected the anonymity” of the data. However, the PUC has not given any guidelines of what it means for a utility to “adequately protect” anonymity.

So, informed by national best practices and standards from other states, CUB proposed specific standards to ensure customers’ privacy while making important energy data available in the public interest.

The proposed Open Data Access Standards

The standards that CUB proposed would allow two types of data to be requested. Data from multiple customers in a geographic area could be aggregated into a single data point, allowing, for example, the owner of a multi-unit building to see total energy use in the property (aggregated data). Or, data from individual customers could be requested, provided that all identifying information is removed and the data set is large enough so that no individuals could be re-identified (anonymized data). The proposed standards require a contract prohibiting recipients of anonymized data from sharing it with others or attempting to re-identify any customers, and would allow utilities to charge reasonable fees to cover their costs of providing data to requestors.

Energy data access in Minnesota and around the U.S.

While Minnesota does not have a statewide standard, utilities make aggregated data available to third parties under various policies today. It’s used to benchmark energy use in large buildings (like Minneapolis’s benchmarking program) and to measure progress towards city energy and climate goals (like this data from the Regional Indicators Initiative). However, parties in this docket pointed to the patchwork of different utility policies and other difficulties with their current data access that would be improved by the framework CUB proposed.

Anonymized energy use data has been used in Illinois to identify unique electricity usage profiles for different kinds of customers, correlating factors like income, age, education, and location. This analysis showed that lower-income customers generally pay more than it costs to provide their electricity and are actually subsidizing higher-income households. Similarly, researchers at MIT have evaluated cost-shifting between high- and low-income areas as more homeowners install rooftop solar panels.

Smart meters (aka advanced metering infrastructure or AMI) that are being rolled out in Minnesota now will collect much more information about when customers use energy. These meters can enable time-of-use rates and new demand response programs to increase system efficiency and give customers more options to cut their bills. 

A data access framework is necessary to put the new utility data to work in consumers’ interest. The meters can more than pay for themselves if they’re put to good use, but they’re not cheap -- Xcel expects to spend on the order of a half a billion dollars on meters and associated technology -- and it is utility customers who pay for the investments. Putting the data from these new meters to work is key to getting consumer and environmental benefits from the investments. 

As new grid technologies are modernized across the country, many states are wrestling with how to make it available for research and energy program design. California, Illinois, and New York, for example, have adopted standards of the kind the Minnesota PUC is considering. New Hampshire, North Carolina, and others have proceedings ongoing. 

The energy world can also be informed by other areas where consumer data has been shared for years. For instance, personal medical data, arguably some of the most sensitive personal information, may be released under HIPAA if it is de-identified or limited such that it no longer contains personally identifying information. Similarly, student records may be released by schools if they have been anonymized or aggregated.

Next steps

At the hearing, the PUC took measured steps forward by:

• Setting a standard for whole-building aggregated data for building owners and benchmarking purposes. Such data can be shared if it includes at least four customers and no single customer uses more than 50% of the building’s energy. This standard applies to all investor-owned utilities with at least 50,000 customers.
• Requesting additional information from parties in order to further refine CUB’s proposed standards, such as the different thresholds for aggregating data (building level, community level, and/or others), contractual requirements for anonymized data access, and ways to streamline the data access process. We anticipate detailed questions from the PUC forthcoming.
• Seeking technical assistance from a national expert, such as the U.S. Department of Energy or the Regulatory Assistance Project.
• Adopting refinements CUB suggested to our proposed standards.

CUB will continue to advocate for a framework that protects customer privacy and puts energy use data to work to cut energy costs and emissions, and we’ll continue to work with the state’s utilities and the parties to this docket to develop. Stay tuned!