Thank you.
My name is Alasdair Roberts. I'm an associate professor of public administration at the Maxwell School of Syracuse University. I am a Canadian and a specialist on freedom of information law or access to information law, and I've done an extensive amount of research on the operation of the Canadian federal law. That's the subject I'd like to address today.
I think we can all agree that Canadians have a right to equal treatment under the law, and that includes equal treatment under the Access to Information Act. We also agree that Canadians have a right to privacy. The major point I'd like to make today is that the methods presently used to administer the Access to Information Act pose a threat to both of these rights. Let me describe these methods.
First, there is technology. Each major federal department uses case management software, also known as tracking software, to manage the inflow of requests for access to information. This software allows departments to classify incoming requests by the occupation of the requester. In doing this classification, every federal department uses a few standard categories, including the media category. However, many departments define other categories as well. For example, I found in a 2005 study that the Department of National Defence used 17 categories to classify incoming requests: academia; business; consultant; dependant; ex-military; historian; lawyer; media; member; organization; other department, which I thought was curious; other government; Parliament; political party; public; public service; and reserves.
I've looked at case management practices in nine federal departments over the last few years. Every federal department that I looked at used a category it described as either political party or Parliament as well.
In addition to departmental software, there is a government-wide database known as CAIRS, the coordination of access to information requests system. Each department is required to enter incoming requests into the government-wide database. Requests from journalists and Parliament are explicitly identified as being in those categories in this database. Research in 2004 showed that the search capabilities of that database were used principally by central agencies, Privy Council Office and Treasury Board Secretariat.
Major departments also have routines for identifying and handling politically sensitive requests. These bureaucratic routines are often well developed and rely on the capabilities of the departmental tracking software. Incoming requests are assessed according to political risk and labelled within the departmental database. The labels vary among departments. Requests may be amber-lighted or coded as red files or purple folders or sometimes as interesting requests.
It appears that requests from journalists, opposition MPs, and party researchers are routinely tagged in this way. The process of tagging appears typically to be undertaken after a regular consultation with ministerial and communications staff. Lists of incoming requests from journalists and opposition parties are regularly generated from the departmental databases and circulated within departments as part of this tagging process.
All of what I have described is now public knowledge. For example, the Gomery inquiry documented this process at work in the Department of Public Works and Government Services.
These practices clearly threaten the right to equal treatment. In a 2002 study, I examined over 2,000 requests handled by HRDC over three years and found that requests from the media or political parties routinely had longer processing times, even after other considerations were taken into account.
In a 2005 study I examined the handling of 25,000 requests received by eight departments over three years. I again found that requests from journalists and political parties were handled more slowly than other requests. This suggests a systemic problem of unequal treatment. There is nothing in the access law that permits requests to be treated differently based on the occupation of the requester. Indeed, a federal report in 2000 said, “Neither decisions on disclosure nor decisions on the timing of disclosure may be influenced by the identity or profession of the requester....” Yet from what I have just said, it is clear that departments routinely flout this rule.
These practices also pose an unnecessary threat to privacy. As I have noted, the occupation of requesters is routinely distributed within departments and across government through CAIRS. The distribution of this information makes it easier for government officials to guess the identity of requesters.
In 2004 I conducted a small test of this hypothesis. I took a random set of requests from CAIRS that were identified only as “media requests”. I gave the list to two graduate students, both American, with no knowledge of Canadian politics. I told my students to identify the requester using only news stories published before the request was filed. One request, sent to the Department of Foreign Affairs, was about policy on national identity cards. The next day my American student guessed, after reviewing earlier news stories, that the requester was Joan Walters of The Hamilton Spectator. I called Joan, who confirmed the guess.
Another request, also sent to the Department of Foreign Affairs, was about an aspect of the softwood lumber controversy. My American student guessed that the requester was James Baxter of CanWest news. I called James, who also confirmed the guess.
If American students, having only access to the Internet but cued as to the occupation of the requester, can make these guesses, how much easier is it for a communications specialist or ministerial aid, once they are told the occupation of the requester?
There are, of course, many instances in which the actual names of requesters have been discussed within the ATI process. The Information Commissioner has identified cases, the Supreme Court has considered a case, the Gomery inquiry examined yet another case, and I understand that this committee heard an additional case in earlier testimony. But even if we did not have these stories, it would be clear that current procedures create unnecessary risks to the right to equal treatment and the right to privacy.
There are at least four steps that could be taken to remedy these difficulties. I identified some of these in my background paper for the Gomery inquiry. They are not radical. The first is discontinuance of the practice of circulating the occupational categories for requests within and among departments. The second is a requirement that departments publish, perhaps on their website, the internal procedures that are used to process requests. The third might be a requirement that departments notify requesters if their requests have been tagged for special handling. The fourth would be explicit recognition of the role of access coordinators within the Access to Information Act so that they are better positioned to defend the law. And finally, there needs to be reform to the funding of the Office of the Information Commissioner so that it has the resources to act quickly against cases of excessive delay and investigate systemic discrimination against certain types of requesters.
In the last decade, over forty countries have adopted their own access laws, often looking to Canada as a model. Many of these countries now watch the work of this committee to see how a mature democracy protects the right to information in practice.
Thank you.
