Addressing liability: balancing records management and open data

paper in file folders

Clerks, records managers and archivists have a key role to play in open data efforts. They have expertise about existing public records systems, which means they can point out potential obstacles in updating those systems for supporting open data. These staff are the stewards of records, but, in some cases, that also means knowing when records have to be destroyed.

Yes, destroyed. That’s the legacy of paper.

Policies created when the ability to keep records was limited by physical storage space sometimes dictate that certain kinds of information be destroyed after a specified length of time (or, at least, that it is not required to be kept any longer). For the sake of space and for other reasons, some documents are deemed to be only of short-term value. Evaluating each document for its value would be painstaking and impractical, though. Records series are used to categorize groups of documents, and these series can then be categorized as worth keeping or destroying after a certain amount of time. These timeframes for keeping or destroying groups of records are called records retention schedules.

Records retention schedules, especially those that require destroying information after a certain time, can have huge implications for open data initiatives.

Some government officials have expressed concern that the expectation that open data be made available indefinitely violates records retention schedules. Take personnel files, for instance. They are just one example of records that are sometimes required to be disposed of after a certain period of time for various reasons. (State and local schedules vary, and not all mandate destruction, but even federal records schedules only require keeping these kinds of documents for a few years).

So what happens if information about personnel is released as part of an open dataset, like asset disclosures for officials in decision-making positions, but it’s time for the record to be destroyed according to a records retention schedule? Could a city or state be liable for information being available beyond its required life cycle?

It’s something that has happened in the corporate world, and the case has been cited in at least one book on records management. The case revolves around Manville Corporation, which was sued for its use of asbestos in buildings. The claims were estimated to be in the billions, so the company filed for bankruptcy. In the course of all of this, Manville was required to compile the documents related to the case. More than 16 million documents were compiled. Some of those records could have been destroyed, according to the company’s records retention program. Because the documents were not destroyed, the people suing Manville were able to claim strict liability, increasing the total amount of what the company would owe, according to a book on records management.

There are two arguments being made based on the Manville case against keeping records longer than required. One is that there could be huge search burdens in the future. That argument can be countered by citing the decreasing costs of keeping and searching through public records, thanks to advances in technology that allow for keeping digital, searchable copies. Storing records at a low cost is increasingly feasible for local governments.

The other argument based on the Manville case is that governments put themselves at risk of being sued for strict liability or negligence by keeping documents longer than necessary.

Disposing of records to avoid being held legally responsible for harm is not an acceptable policy. Governments elected by the people should also be accountable to the people, and records are part of what helps keep that system functioning and honest. Avoiding keeping records to avoid being held accountable is not a practice that should be inherently, or explicitly, endorsed by records retention schedules.

These threats to accountability, based on governments’ concerns about liability, highlight why records retention schedules need to be fixed.

Clerks and records managers should update records retention schedules to reflect that continued access to a broad range of public information is both valuable and possible as records management transitions into a digital age.

Recommendations for improving records retention

Regardless of whether a government has an open data initiative, it should examine its records retention schedules to determine whether they are too restrictive. In performing this review, those charged with records management should evaluate the rationale for existing time limits on each kind of document. Were the schedules dictated by concerns about space and storage? Concerns about liability? These questions, at a minimum, should be posed when reviewing records retention schedules for potential updates.

If records managers are concerned about new liability risks in shifting from existing retention practices, they should work with their government counsel to develop a specific assessment of the risk of liability that may arise from each category of document. This type of targeted assessment will reduce the possibility that governments allow fears of liability to result in the unnecessary destruction of records.

Allowing for a careful assessment of potential liability might even lead to the conclusion that extending the retention of some kinds of records reduces liability. Records managers could consider records retention that matches the statute of limitations for cases where the government thinks it might be liable in some way for its actions. This would help provide a deeper body of evidence to allow governments to produce documents demonstrating what really happened and to ensure accountability.

As part of the review process, governments could also update records retention schedules to shift some records to a “minimum retention period,” mandating a length of time for which records must be maintained while not requiring destruction after a certain period. After a minimum retention period there would be no legal requirement to keep the records, but there would also be no mandate to destroy them.

Especially in cases where the primary concern about records retention is the cost of storing them rather than increased liability, designating a minimum retention period would give governments the future flexibility to assess their resources against the future (and currently dropping) cost of data storage. If it turns out to be prohibitively expensive to keep the records beyond the minimum retention period, officials would not be bound to keep the records.

Decisions about public data, and updates to records retention schedules in general, should be premised on a test that weighs the costs of keeping public records against the benefits of keeping them. The same kinds of questions used to determine the priority of datasets could be applicable in this process.

We’ll be continuing to explore these kinds of challenges, and opportunities, in the ways records management and open data intersect. We hope to empower records managers and everyone involved in the open data process with solutions and resources for success.