MINNESOTA COALITION ON GOVERNMENT INFORMATION (MNCOGI)

Testimony of Matt Ehling related to HF 291

House Civil Law Committee 

January 26, 2017

The Minnesota Coalition on Government Information supports HF 291.  This bill seeks to clarify the language of Minn. Stat. 13.55 – the section of the Data Practices Act that classifies convention center data.

Before I discuss why clarification of this statute is desirable, I want to address its history briefly, so that its original intent is clear.

As you’ll see in our packet starting on page two, this provision of the Data Practices Act was added in the early 1980s, and was written to specifically address data held by the Saint Paul Civic Center Authority, which booked musical acts and other touring events seeking to rent the Saint Paul Civic Center.

The original statute essentially classified data that related to the marketing and negotiating activities of the Civic Center as it dealt with prospective renters.  You’ll see that the data made “nonpublic” by the statue included the identity of firms that contacted the Civic Center Authority, as well as the suggested terms of the rentals.  The statute then made this data “public” after a contract with a renter had been signed, or under certain other circumstances.  So the original statute was directed specifically at protecting the identify of – and communications with – prospective renters of Saint Paul’s convention center.  Setting aside for a moment whether this was good public policy or not, it was the original intent of the earliest version of the statute.

You’ll then see that subsequent alterations to the staute were made, which resulted in what we have today — a generalized provision that covers publicly owned and operated convention center facilities, as well as the Minnesota Sports Facilities Authority, or MSFA.  The MSFA is the successor orgnization to the Metropolitain Sports Facilities Commission.  The data elements that are covered by the statute today are essentially the same as those covered in 1982, with a couple of additions.  What is not covered are the names of individuals who recieve free or discounted tickets or other gifts from publicly owned facilities — that data is presumptively public.

After stories broke in the Star Tribune newspaper about the MSFA providing various public officials and others with free access to box seats in US Bank Stadium, the MSFA recieved press inquiries about who specifically had attended events as guests of the MSFA.  Initially, MSFA officials declined to release the names of their guests, citing Minnesota Statue 13.55 – the convention center provision that we’ve been discussing.  Eventually, the MSFA released a list that included dozens of names, due to the fact that those names were classified as public data under Minnesota law.

While data relating to persons who receive free tickets and gifts is presumptively public data under the Data Practices Act, we feel that it would be useful to reiterate that fact in law for the following reasons:

1.  First, the language of the Data Practices Act has been crafted in many instances to call out specific, public data that is described within a provision that otherwise classifies data as “not public” information.  You’ll see an example of this on page four of your packet, in Minn. Stat. 13.601.  This statute makes certain data “private,” but its language also notes that other, very specific data are public.  Generally under the Data Practices Act, the absense of specific langauge classifying data means that the data remains public, but there are instances in Chapter 13 where language calling out a public classification has been added for clarity.

2.  Secondly, in cases where there has been a controversy or disagreement about the meaning of a provision of the Data Practices Act, the legislature has sometimes added clarifying langauge to underscore the right to access certain public data.  The controversy over access to the names of persons with free access to MSFA box seats falls into this category.  Given the number of publicly-owned convention facilities in the state – ranging from the Ames Center in Burnsville to the DECC in Duluth – questions about the status of data related to gift and tickets may arise again, and additional clarity will be helpful.

The Star Tribune editorial board has urged the legislature to strike a different balance in its final body camera legislation than the position recently taken by the Minnesota Senate. The Senate bill – passed earlier this week – would make most body camera video “private” data. Such data would only be accessible to subjects captured by the video, or to the police who collected the footage. The Senate bill makes an exception for a very limited category of data – video captured in a public place that documents police use of a dangerous weapon, or police use of force that results in substantial bodily harm.

Body cameras have been used in Minnesota for almost six years under the current terms of the Data Practices Act, which provides public access to much footage, but also provides privacy protections to over a dozen categories of individuals, including undercover officers and certain crime victims. If images of those persons are present in body camera recordings, they must be redacted (“blurred out”) before the footage is released. If police departments do not have the technology to redact videos, they currently have the discretion to edit out sections that contain protected images.

Given the fact that the public has had access to much body camera footage for several years now – and to dash cam videos for many years prior – reclassifying most body cam footage as “private” takes too broad a brush to the regulatory questions surrounding body camera data.  We agree with the Star Tribune that a more nuanced approach is needed – an approach that is more mindful of government transparency.

Testimony of Matt Ehling
Chair, Legislative Issues Committee
Minnesota Coalition on Government Information

Legislative Commission on Data Practices
Hearing of December 1, 2015

Full written comments

Good morning, and thank you Madame Chair.

My name is Matt Ehling. I am the chair of the legislative issues committee of the Minnesota Coalition on Government Information. Thank you for the invitation to be here this morning, and to be part of this ongoing discussion.

We have been part of the policy debate around body camera legislation since 2014, when this commission first took up the issue. Most recently, we have provided input to the working group on body camera legislation convened by Representative Scott that includes Senator Latz and others.

I’ll first provide some background on this issue, and then will set out our organization’s position on this matter.

Two kinds of policy questions
There are two key policy areas surrounding body cameras:

First, there are questions about whether body cameras should be used in the first place. These include questions such as “Do body cameras provide increased government oversight, as proponents claim? Do they actually provide clarity and resolution to disputes? And are they worth the money that municipalities pay for them?” These are questions for the end-users of these systems — the citizens of the municipalities that will actually make decisions about using them.

Then, there are the bigger-picture regulatory issues that surround body camera policy, including the questions that this commission is tasked to evaluate, such as how to classify the data that result from their use. These classification issues form the bulk of our comments on this matter.

Uses of law enforcement video
I should note that while body cameras represent a newer generation of law enforcement technology, they have been in limited use in some Minnesota cities for some time now – including in Richfield and Duluth. At this point, law enforcement has also used dashboard cameras to document arrests and other incidents for almost two decades. Police agencies have also used video cameras to document crime scenes for even longer, and so there is institutional experience within police departments about how to treat law enforcement video. All of the video gathered to date has been covered by the existing provisions of the Data Practices Act.

Police video – background

Existing law governing video recordings
For context, let me speak briefly about the classification status of law enforcement video under current law. At present, body camera video is classified as presumptively public data. Since there is no body-camera specific provision of the Data Practices Act, police video falls under the general presumption that data are public unless another provision of law applies. There are several of these, and you can see them on the chart we have provided.

Criminal investigative data, including exceptions
The first provision that can apply to any law enforcement data- video or otherwise – is the criminal investigative provision of Minnesota Statutes 13.82. If an investigation is opened, the police video footage would become “confidential” or “protected nonpublic” data, meaning that it cannot be shared with either the subject of the video, or with the public until the investigation is complete. There are exceptions, however. These include data that relate to arrests, or to certain incident data, including whether a suspect offers resistance to an arrest, or whether officers use weapons. Data that document these things – including on video images – are classified as “public at all times” under 13.82, subdivisions 2, 3, and 6, and are accessible by public data requests.

Discretion to release “public benefit” data
In addition, police have discretion under current law to release additional investigative data under 13.82 subdivision 15, which states that police can release investigative data if it would aid the law enforcement process, or dispel widespread rumor and unrest. Again, the discretion here lies with the law enforcement agency.

Withholding “not public” data
Beyond these provisions of law, the Data Practices Act also allows the withholding of video that is “offensive to common sensibilities.” I have seen instances where this withholding has been applied to images of deceased persons, but it could be applied even more broadly to images of nudity, suicide, as so forth.

Finally, there are numerous provisions of existing law that allow police to withhold images of specific individuals, including undercover law enforcement officers, victims of sexual assault, other crime victims, and over a dozen other categories. Any police video that includes such images would need to have those images redacted before public release.

This framework has been in place for several decades, and we feel that it has largely worked to address issues surrounding police video up to the present. I would note, however, that there are certain areas where adjustments and improvements could be made. Let me describe our policy priorities by summarizing the high points of the policy document that we’ve provided in our materials.

MNCOGI policy recommendation highlights

Provide additional clarity around coverage of existing law
First, we propose to add additional clarity to existing law to ensure that the status of police video is self-evident. Although the framework that I’ve described covers police video today, there is no terminology describing video in 13.82. This has led to some confusion, and we propose to remedy that by clearly describing that police video – including body camera video – is covered by the existing provisions that describe what is public and not public under 13.82

Clarify “clearly offensive to common sensibilities” language
We would also like to ensure that the language of 13.82 subdivision 7 – which describes the withholding of material that is “clearly offensive to common sensibilities” – is clarified to cover all images. Right now, the language says that it applies to “photographs” which is older language that needs to be updated. We are also open to having a discussion of how to more specifically define that the existing language means, and what it covers.

Ensure continued access to important public video
Third, we want to ensure that if any changes to the classification of police video are made, important oversight-related data remains public. At minimum – arrest, incident, and response data should remain public at all times, and video documenting any use of force should remain publicly available after an investigation is closed.

In regard to arrest video, I would note that we do not have secret arrests in this country. Because of that, data which documents an arrest – including video recordings – should continue to be publicly available at the time of the arrest.

Likewise, data that document the use of weapons under subdivision 2 and 6 should continue to be public at all times, and video that documents any use of force should continue to be public once an investigation is closed.

We recognize that using force is difficult – it is the certainly most difficult task that law enforcement is called upon to perform. Whether effectuating an arrest, or stopping a crime in progress, the use of force has to be both effective, and lawful in its application. We recognize the difficulty of that task.

However, it is also true that the use of force is the most consequential power the government has at its disposal, and like any governmental power, it holds the potential for abuse. And so it follows that if the government is documenting its use of force with body cameras or other devices, that video should be available to the public for the purpose of oversight and evaluation. This is particularly true in a society where governmental power rests on the consent of the governed.

We feel that there might be room for negotiation around other points of body camera video, and are willing to discuss whether certain pieces of video might require additional privacy protections, such as video of certain domestic incidents. However, we believe that core-oversight related video must remain public.

Consent requirement for recording in private places
Outside of classification issues, we feel that there is one key regulatory change that the legislature should enact in order to deal with body cameras. Body cameras are different from squad car cameras in that they are portable, and can record activity in private places, such as homes. As body cameras proliferate, they will enter into private places more frequently, causing increased privacy concerns.

While the Data Practices Act makes some law enforcement video (such as certain victim data) “not public,” other recordings – such as video of officers entering a private home to interview witnesses – would be publicly available, thus exposing home interiors and other details to the public.

We believe that the best way to deal with this situation is to make a change to state law that requires notice and consent before recording police video in a private places. Given the option, we believe that most people will opt-out of being recorded in consensual encounters with police such a welfare checks, thus eliminating video-related privacy concerns from those encounters, since video of those encounters will not exist.

We would also recommend that the law allow for common-law exceptions for consent, such as during exigent circumstances where police have to act quickly to enter a home to stop pending violence. In those circumstances, police should be able to record video as needed, at their discretion.

We would note that it the legislature does not proactively adopt a consent requirement for recording in private homes, it is likely that lawsuits may ensue, and that courts will impose a case-law requirement. We believe that recording of video in a private home is a form of search or seizure, and is a separate and discreet action from police entering a private home. The two have to be considered separately, with discreet consent for each.

Other recommendations
Finally, we have offered a series of other recommendations, which I won’t discuss in detail in order to save time. You can see these in our attached materials, and I’d be happy to take questions on them.

In short, we believe that the existing framework governing law enforcement data has a sound logic to it, and that its key pieces should continue to cover body camera video. At the same time, we are seeking some updates and modifications, and are open to targeted discussions about changes that other parties may seek.

Thank you.

LAW ENFORCEMENT USE OF “BODY CAM” RECORDERS:

Overview of classification and operational issues

Background

Mobile “body cam” systems are compact, camera/recording or camera/transmission devices that can be attached to clothing, glasses, or other worn garments or accessories.

Police in several Minnesota municipalities – including Duluth – have adopted such devices for use by their police departments.  At present, the city of Minneapolis is on the cusp of a body cam pilot program to help develop protocols for larger-scale implementation. The cameras are largely being adopted to enable the capture of video documentation to help resolve disputes over police conduct and/or use of force incidents.

Two types of commercially available body cam systems are being used by law enforcement agencies:

1. Systems that record to the worn device when triggered by a user;

2.  Systems that are connected to a remote server through wireless connectivity, and record on a continuous basis.

Current data classification

At present, information gathered by police body cams is generally treated as presumptively “public” data under Chapter 13.  One exception to this public presumption involves the use of body cam data in active criminal investigations.  When such an investigation is opened, body cam data pertinent to the matter can be maintained as “not public” data for the duration of the active investigation, after which it reverts to its former, public status.  This is similar to the way that squad car video is currently treated.

Issues raised by body cam data

Body cam video holds the possibility of clarifying questions about police use of force incidents, creating accountability for the public. Body cam data may also help to prevent unfounded claims of police misconduct from being sustained.  At the same time, the mobility and pervasiveness of the video recording offered by body cams raises questions around privacy that complicate the discussion over camera operation and data classification.  For instance:

1. Should video recorded in public areas and video recorded in private domiciles or workplaces be classified in the same way?

2. Should police have to provide notice to persons they encounter that body cams are in operation?  In what circumstances would such notice be appropriate?

3. Would continuous transmission of body cam data over wireless networks be susceptible to hacking or interception, and therefore trigger invasions of privacy or breaches of potentially sensitive data?

4. Would some forms of body cam technology raise more difficult classification, use, or privacy questions than others?

MNCOGI’s proposed framework

MNCOGI continues to explore issues surrounding police body cam data, but offers the following framework to help guide the discussion surrounding the implementation of the technology, and the classification of the resulting data.

1. Video should be presumptively public; arrest video public “at all times”.

A public classification for body cam data offers oversight benefits to the public. Public availability of the data serves the classic oversight purpose of documenting the conduct of public officials for later review.

Public access to body cam data also permits the public to review how and where the government is collecting video material. Access to the data allows the public a means of measuring the appropriateness of video collection and use.

MNCOGI generally urges a “public” classification for body cam data, with most data remaining presumptively public, and other data expressly classified as being public “at all times” under Minn. Stat. 13.82, Subd. 2. As the primary purpose of gathering the video is to provide transparent documentation of the actions of officers, some form of public status would be the best match for the data.

Actions that are most relevant to an oversight purpose should be public at all times – arrest data, for instance. Currently, Minn. Stat. 13.82 Subd. 2 includes a specific list of types of data relevant to arrests that are always public, and thus not subject to being withheld as criminal investigative data at any point. Body cam documentation of arrests should be added to that list, or otherwise specified elsewhere in Minn. Stat. 13.82. Other, specific forms of video recording might also be considered for classification as being public “at all times” under Minn. Stat. 13.82.

Other body cam data could remain unclassified, and thus presumptively public data.

2. Privacy should be addressed through the regulation of video collection.

While body cam data could play a potentially positive oversight role, it could also greatly expand the ability of police to create a long-term surveillance record of a community, raising privacy and accountability concerns.  Much like license plate reader (LPR) technology, long-term retention of indiscriminate body cam data could vest the government with an archive of the movements and associations or individuals, or even the interiors layouts of locations that police visit for calls for service and other incidents.

Continuous, indiscriminate recording could also raise questions about whether individuals who might require police attention for welfare checks or similar services would be reluctant to call, due to hesitations about having their interactions recorded.

Such privacy issues could be addressed through a statutory scheme under which certain body cam data would be classified as “public”, while other data would be classified as “private”. However, MNCOGI believes that this matter would be better addressed through the regulation of camera use, rather than by creating multiple categories of private body cam data.

Departmental rules, municipal ordinances, or state laws could be written to address when, where, or how body cam data could be collected in order to address questions of privacy. For instance, are there situations in which notice should be required prior to recording? Welfare checks could be one set of circumstances. The recordings of arrests, on the other hand, could be regulated so as not require such notice.

The regulation of body cam recording would also ensure that the technology is used for a narrow, oversight purpose, without expanding into a more broad-based video surveillance platform.

On February 28th, MNCOGI released its 2014 legislative issues document during a noon-time event at the Minnesota State Capitol.

During the 2014 session, MNCOGI will be closely following several matters, including the status of license plate reader (LPR) data, provisions surrounding booking photo data, and the creation of a legislative commission on data practices.

MNCOGI’s full list of issues is reproduced below:

Data Practices Legislative Commission

• MNCOGI position:  Given the importance and complexity of data-related issues, the Minnesota Legislature should create a Legislative Commission on Data Practices.  A commission would allow the Legislature more time to study data issues (both access and privacy issues), bring recommendations, and craft bills that could be acted upon during the regular session.  The additional time afforded by the commission would allow the Legislature to take a “long view” of such matters, and aim for continuity in data policy.

LPR (License Plate Recognition) Data

• MNCOGI position:  To ensure effective oversight, provisions should be included within state law to ensure public access to data about the scope, nature, and use of LPR technology by Minnesota government entities.  MNCOGI also believes that a reasonable formula for dealing with data collected by LPR scanners is as follows:

All data collected by LPR scanners should be classified as “not public” data for a very short period of time after collection.  A retention scheme should be instituted under which “non-hit” LPR data would be quickly purged during its brief, initial status as “not public” data.  The remaining “hit” data that pertains to specific individuals or vehicles should be maintained as “not public” criminal investigative data until the closure of a criminal investigation.

Private contract, sub-contract data

• MNCOGI position:  In light of the Supreme Court’s opinion in the Helmberger v. Johnson Controls case, the Minnesota Legislature should support changes to Minnesota law that would ensure that data about privatized government functions continues to be available for public review.

Booking photograph data

• MNCOGI position:  Minnesota law should not be altered to treat requesters of booking photographs differently from other public data requesters.  Two bills introduced during the 2014 session seek to institute certain requirements related to booking photographs.  One of the bills (HF 1940) mandates that requesters submit statements regarding their intended uses of the photographs, as well as the locations where the photographs will be published.  The addition of such requirements would weaken the overall framework of the MGDPA by introducing – for the first time – mandates requiring certain requesters to specify their intended uses of government data.

HF 1940 also seeks to institute a variety of penalties for failing to comply with some of its provisions. For instance, the bill requires that persons who receive booking photographs from other parties file use-related disclosures with police agencies, or else become liable for damages. Such an approach raises significant First Amendment issues.

Prosecutors specified in “Criminal Investigative Data”

• MNCOGI position:  Prosecutors should be added to the itemized list of persons and/or entities that can receive and maintain “criminal investigative data” under Minn. Stat. 13.82.  Such a change would codify a long-standing practice recognized by IPAD advisory opinions.

Affirmative right to record open meetings

• MNCOGI position:  Minnesota law should be altered in order to codify an affirmative right to record any proceedings that are subject to Minnesota’s Open Meeting law.  Such a change would codify a long-standing Attorney General opinion on the subject,

Police “body cam” data

• MNCOGI position:  Several municipal police departments have either obtained – or are in the process of obtaining – “body cam” video recorders for patrol officers to wear.  These mobile devices record daily police interactions in order to create a record for use in criminal or civil court proceedings.  MNCOGI believes that the data created by police body cams should be classified as public “incident” data, similar to the way in which squad car video is considered to be presumptively public government data.  In both cases, the public classification of the data ensures a measure of public review of police activities.

Mass surveillance data

• MNCOGI position:  Given recent advances in technology, government entities may increasingly be able to engage in the mass collection of data about individuals that was formerly beyond the reach of large-scale capture.  Such data could include, for instance, ongoing, “real-time” information about the locations and movements of thousands of individuals.  The legislature should evaluate such technologies on an ongoing basis, and ensure statutory access to information about the nature, scale, and legal underpinnings of such technologies.