Most criticism of the Freedom of Information Act centers on agency refusals to disclose requested records in a timely manner. But a federal appeals court said this week that a Defense Department agency was “arbitrary and capricious” in its decision to release documents to a Freedom of Information Act requester.
The ruling comes shortly after the release of several new evaluations of government compliance with the Freedom of Information Act by the National Security Archive, the Associated Press, and Citizens for Responsibility and Ethics in Washington. Each of these independent efforts found that FOIA performance in the first year of the Obama Administration in one way or another had fallen short of the Administration’s proclaimed standard of “unprecedented openness.” Each report identified questionable patterns in some agencies’ handling of FOIA requests, mostly involving the frequency of denials, the persistence of backlogs of unanswered requests, and haphazard implementation of new Obama FOIA policies.
None of the critics complained of excessive disclosure. But a federal appeals court this week ruled (pdf) that the Defense Contract Management Agency (DCMA) had been too forthcoming and had tried to disclose information in response to a FOIA request that arguably should be withheld.
After the DCMA granted a FOIA request in 2005 to release certain audit reports involving two DoD contractors, the affected companies filed so-called “reverse FOIA” lawsuits to block the disclosures. Those companies (Sikorsky Aircraft and Pratt & Whitney) said that the proposed releases would cause them “embarrassment or negative publicity” and would compromise proprietary information that could be exploited by their competitors. The Defense Department disputed these claims and prevailed against the companies in district court. Upon appeal, the higher court concurred that “embarrassment” was not a legitimate grounds for withholding, but it found that the possible compromise of proprietary information had not been adequately taken into account by the lower court.
“DCMA’s decision to release the documents was arbitrary and capricious,” the court said. It therefore reversed the lower court’s ruling in favor of disclosure, and remanded the case for further consideration. More background on the case is available from the Project on Government Oversight here.
So is the FOIA process too restrictive in denying information, or too lax in releasing it? Any FOIA requester will answer that it’s too restrictive. For one thing, just as overclassification is prevalent in many national security agencies, unnecessary withholding of information under FOIA because of dubious classification controls is likewise commonplace.
But a larger point is that annual statistics on FOIA releases and denials, like those reported by agencies last week, are an imperfect indicator of changes in government openness, for at least two reasons.
First, the FOIA is not only a disclosure statute, it is also a withholding statute, i.e. it authorizes or requires both actions under various circumstances. If an agency received 100 requests for the blueprints of a classified weapon system and it issued 100 denials, the agency would have complied with FOIA perfectly, because Congress did not intend for properly classified material to be disclosed under the Act. Denials are not necessarily a sign of bad faith, or of a failure of FOIA policy.
Second, the comparison of FOIA release and denial data between last year and the year before, as performed by the Associated Press and others, would be valid and interesting only if the initial FOIA requests each year were identical, or at least roughly similar. But no effort has been made to demonstrate that that is so.
For the same reasons, it is doubtful that Attorney General Eric Holder was correct to say that a recent increase in the proportion of FOIA disclosures demonstrated increasing openness in the Justice Department, as he did on March 15: “Today, I’m pleased to report that the disturbing 2008 trend – a reduction in this Department’s rate of disclosures – has been completely reversed.”
By itself, a net increase in disclosures is not a sure sign of a new devotion to openness, nor is a reduction in the rate of disclosures conclusive evidence of non-compliance with FOIA or of agency hypocrisy.
(If there were a statistical rise in judicial rulings against the government in FOIA cases, that would be a strong indication that agencies were increasingly acting in violation of the law. But that doesn’t seem to be the case either.)
Another reason why annual disclosure rates are unreliable or ambiguous indicators is that they can be artificially diminished by spurious requests. The Central Intelligence Agency’s 2009 log of FOIA requests (flagged by cryptome.org) lists quite a few eccentric and offbeat requests that are clogging CIA FOIA channels, along with numerous bona fide inquiries.
One requester asked for CIA “documents pertaining to the Best Buy located at 4500 Wisconsin Ave NW” in Washington DC. Under the peculiar terms of the FOIA, the request must be processed like any other. But it would be quite surprising if a FOIA request for CIA records about this “facility,” which I happen to walk by every day, yielded anything other than a “no records” response. If so, such a negative response would automatically lower CIA’s annual rate of disclosure without providing meaningful insight into CIA FOIA policy.