UPDATE: 16 July 2014 – in the comments to this piece the ICO adds some further details on the “non-trivial” incident: “We are unable to provide details of the breach at this stage, as the information involved is linked to an ongoing criminal investigation.”
The ICO had a “non-trivial” data security incident last year. Can it “fine” itself? Will/has it?
Christopher Graham, the Information Commissioner (ICO), revealed yesterday that his office had suffered a “non-trivial data security incident” within the last 12 months, which prompted a full internal investigation
The ICO, of course, processes personal data and in doing so assumes the role of the data controller (according to section 1(1) of the Data Protection Act 1998 (DPA)). It also assumes the obligation to comply with the data protection principles, and the liability for contravening them. In 2012 the ICO responded to a Freedom of Information Act 2000 (FOIA) request for its “data breach log” with a document that showed admirable commitment to recording even the smallest of potential data security incidents (“person taking photographs outside building”, “theft of small amount of money”). In that instance there were two incidents identified as “high risk”, but the ICO declined to provide information, and the requester, it seems, did not pursue the matter.
This time, with national media picking the story up, the matter may be pushed further. At the moment the ICO is apparently declining to offer any further comment to the media, advising The Times that
You will have to fill out a freedom of information request
which doesn’t really sit that well with their normal commitment to transparency.
But to what extent can or should the ICO investigate its own compliance with the DPA? The Act does not provide for any derogation for the ICO from its obligations, and nor does it provide for any alternative to “self regulation”. Nor, moreover, does it appear to provide for any delegation to a third party to investigate. When it deals with complaints about its own handling of FOIA requests it habitually issues decision notices about itself (sometimes even finding against itself). It does this by distinguishing between “the ICO” (the entity dealing with the request) and “the Commissioner” (the entity dealing with the complaint). I would imagine that a similar nominal separation would be used if it came to formal enforcement action being contemplated in response to a data security incident.
I emphasis the word “if” in the previous sentence, because, although The Times says
The ICO, which can levy fines of up to £500,000 for data protection breaches, did not disclose whether it had fined itself for the breach
it is clear in fact that no such enforcement action resulted in this instance. This is clear because, firstly, the ICO’s own Monetary Penalty Guidance says that any monetary penalty notice (for which “fine” is a convenient, if not strictly correct, shorthand) will be published on its website. None has been published (believe me – I check these things very regularly). And secondly, and more fundamentally, the ICO’s report says that the incident in question
did not amount to a serious breach of the Data Protection Act [emphasis added]
By section 55A a monetary penalty can only be served for a serious contravention of the data controller’s obligations under the DPA. If the incident was not a serious contravention, the statutory threshold for a monetary penalty is simply not met. So, regardless of what other information about the incident might be winkled out of the ICO, we are not going to have a story of “ICO fines ICO”.
However, on a final point, I note that the ICO expects data controllers to report serious data security incidents to the ICO. So the question arises – did the ICO report this to the ICO, or did the ICO assess this as not serious enough to refer to the ICO? How did the ICO get to know? Could it have been a leak by the ICO? Or even by the ICO? These questions deserve answers*.
*no they don’t