I posted about this originally on LinkedIn, but I found it so nerdily interesting I wanted to preserve it better by putting it on this blog.
In 4 December 2023 the Information Commissioner’s Office (ICO) issued a decision notice under section 50 of the Freedom of Information Act 2000 (FOIA) finding that its own office did not deal with a FOIA request within the statutory time limit. Subsequently, however, as the ICO website has it, “Following a review of this case it has been noted that the Commissioner erred in citing a breach of section 17(1) of FOIA, having omitted to include the Scottish bank holiday of 7 August 2023 in his calculation of the 20 working day deadline. Therefore, the ICO did not breach section 17(1) of FOIA.”

However, merely staring on its website that “the ICO did not breach FOIA” is not sufficient. As a matter of law, the decision notice itself stands, unless it is substituted by another notice made by the Information Tribunal upon appeal. The ICO cannot withdraw/amend a decision notice, in the absence of an appeal (under the doctrine of “functus officio”, but see also IC v Bell [2014] UKUT 0106)).
So merely saying on its website “we didn’t breach the time limits” cannot cancel or overturn the decision notice.
In some analogous circumstances of “wrong” legal decisions by public authorities bound by functus officio, the authority will consent to judicial review proceedings quashing the decision. But here, the only person with any interest in quashing the decision is the ICO itself, and I don’t believe it could apply for judicial review of its own decision (although there have been cases, I believe, where local authorities have judicially reviewed decisions of their own planning committees).
What the ICO could have done though (and I give a nod to Ganesh Sittampalam here) is appeal the decision itself to the Tribunal. It would seem to be the case that the ICO, as the public authority on whom the decision notice was served, would have had a right of appeal to the Tribunal, even though it would be both the appellant and the respondent. This would, obviously, be rather an odd situation, but it’s one that the ICO already faces when it has to rule (as it did here) on its own compliance with the laws it regulates and enforces (for these purposes it effectively creates a fictional divide between “the ICO” and the “Commissioner” – see for example paragraph four in the decision notice linked above).
However, for whatever reason, the right of appeal was not exercised. But, given that that was the statutory route for challenge, why was the purported correction of the error instead subject to an internal, non-binding and unsatisfactory “review” within the ICO?
One wonders how this will be recorded within the ICO’s datasets: will the ICO accept the point that, as a matter of law, the decision is and remains that it breached the time limits? I doubt it.
The views in this post (and indeed most posts on this blog) are my personal ones, and do not represent the views of any organisation I am involved with.

