Privacy and data protection developments in 2013: Google, Facebook, Leveson and more

Data protection law was designed to be a fundamental and concrete dimension of the individual’s right to privacy, the primary safeguard against misuse of personal information. Given those ambitions, it is surprisingly rarely litigated in the UK. It also attracts criticism as imposing burdensome bureaucracy but delivering little in the way of tangible protection in a digital age. Arguably then, data protection law has tended to punch below its weight. There are a number of reasons for this.

One is that Directive 95/46/EC, the bedrock of data protection laws in the European Union, is the product of a largely pre-digital world; its drafters can scarcely have imagined the ubiquity of Google, Twitter, Facebook and the like.

Another is that in the UK, the evolution of Article 8 ECHR and common law privacy and breach of confidence actions has tended to deprive the Data Protection Act 1998 of the oxygen of litigation – before the House of Lords in Campbell v MGN [2004] UKHL 22, for example, it was agreed that the DPA cause of action “added nothing” to the supermodel’s breach of confidence claim (para. 130).

A further factor is that the DPA 1998 has historically lacked teeth: a court’s discretion to enforce subject access rights under s. 7(9) is “general and untrammelled” (Durant v FSA [2003] EWCA Civ 1746 at para. 74); damages under s. 13 can only be awarded if financial loss has been incurred, and the Information Commissioner has, until recently, lacked robust enforcement powers.

This landscape is, however, undergoing very significant changes which (one hopes) will improve data protection’s fitness for purpose and amplify its contribution to privacy law. Here is an overview of some of the more notable developments so far in 2013.

The draft Data Protection Regulation

The most fundamental feature of this landscape is of course EU law. The draft DP Regulation, paired with a draft Directive tailored to the crime and security contexts, was leaked in December 2011 and published in January 2012 (see Panopticon’s analysis here). The draft Regulation, unlike its predecessor would be directly effective and therefore not dependent on implementation through member states’ domestic legislation. Its overarching aim is harmonisation of data protection standards across the EU: it includes a mechanism for achieving consistency, and a ‘one-stop shop’ regulatory approach (i.e. multinationals are answerable only to their ‘home’ data protection authority). It also tweaks the law on international data transfers, proposes that most organisations have designated data protection officers, offers individuals a ‘right to be forgotten’ and proposes eye-watering monetary penalties for data protection breaches.

Negotiations on that draft Regulation are in full swing: the European Parliament and the Council of the European Union’s DAPIX (Data Protection and Information Exchange) subgroup working on their recommendations separately before coming together to approve the final text (for more detail on the process, see the ICO’s outline here).

What changes, if any, should be made to the draft before it is finalised? That rather depends on who you ask.

In January 2013, the UK government set out its views on the draft Regulation. It did so in the form of its response to the recommendations of the Justice Select Committee following the latter’s examination of the draft Regulation. This is effectively the government’s current negotiation stance at the EU table. It opposes direct effect (i.e. it wants a directive rather than a regulation), thinks the ‘right to be forgotten’ as drafted is misconceived, favours charging for subject access requests and opposes the mandatory data protection officer requirement. The government considers that promoters of the draft have substantially overestimated the savings which the draft would deliver to business. The government also “believes that the supervisory authorities should have more discretion in the imposition of fines and that the proposed removal of discretion, combined with the higher levels of fines, could create an overly risk-averse environment for data controllers”. For more on its stance, see here.

The ICO has also has significant concerns. It opposes the two-stream approach (a mainstream Regulation and a crime-focused Directive) and seeks clarity on psedonymised data and non-obvious identifiers such as logs of IP addresses. It thinks the EU needs to be realistic about a ‘right to be forgotten’ and about its power over non-EU data controllers. It considers the current proposal to be “too prescriptive in terms of its administrative detail” and unduly burdensome for small and medium-sized enterprises in particular.

Interestingly, while the ICO favours consistency in terms of sanctions, it cautions against total harmonisation on all fronts: “Different Member States have different legal traditions. What is allowed by law is not spelled out in the UK in the way that it is in some other countries’ legal systems. The proposed legislation needs to reflect this, particularly in relation to the concept of ‘legitimate interests’.” For more on the ICO’s current thinking, see here.

Those then are the most influential UK perspectives. At an EU level, the European Parliament’s report on the draft Regulation is more wholeheartedly supportive. The European Parliament’s Industry Committee is somewhat more business-friendly in its focus, emphasising the importance of EU-wide consistency and a ‘one-stop shop’. Its message is clear: business needs certainty on data protection requirements. It also urges further exemptions from data protection duties for small and medium-sized enterprises “which are the backbone of Europe’s economy”. The Industry Committee’s views are available here.

Negotiations continue, the aim being to finalise the text by mid-2013. The European Parliament is likely to press for the final text to resemble the draft very closely. On the other hand, Ireland holds the Presidency of the Commission and of DAPIX – until mid-2013. Its perspective is probably closer to the UK ICO’s in tenor. There are good prospects of at least some of their views to be reflected in the final draft.

A number of the themes of the draft Regulation and the current negotiations are already surfacing in litigation, as explained below.

The Leveson Report

Data protection legislation in the UK will be affected not only by EU developments but by domestic ones too.

In recent weeks, debate about Leveson LJ’s report on the culture, practices and ethics of the press has tended to focus on the Defamation Bill which is currently scraping its way through Parliament. In particular, the debate concerns the merits of an apparently-Leveson inspired amendment tabled by Lord Puttnam which, some argue, threatens to derail this legislative overhaul of libel law in the UK (for one angle on this issue, see David Allen Green’s piece in the New Statesman here).

The Leveson Report also included a number of recommendations for changes to the DPA 1998 (see Panopticon’s posts here and here). These included overhauling and expanding the reach of the ICO and allowing courts to award damages even where no financial loss has been suffered (arguably a befitting change to a regime concerned at heart with personal privacy).

The thorniest of Leveson LJ’s DPA recommendations, however, concerned the wide-ranging ‘journalism exemption’ provided by s. 32. The ICO has begun work on a code of practice on the scope and meaning of this exemption. It has conducted a ‘framework consultation’, i.e. one seeking views on the questions to be addressed by the code, rather than the answers at this stage (further consultation will happen once a code has been drafted).

There is potential for this code to exert great influence: s. 32(3) says that in considering whether “the belief of a data controller that publication would be in the public interest was or is a reasonable one, regard may be had to his compliance with” any relevant code of practice – if it has been designated by order of the Secretary of State for this purpose. There is as yet no indication of an appetite for such designation, but it is hoped that, the wiser the code, the stronger the impetus to designate it.

The ICO’s framework consultation closes on 15 March. Watch out for (and respond to) the full consultation in due course.

Google – confidentiality, informed consent and data-sharing

Google (the closest current thing to a real ‘panopticon’?) has been the subject of a flurry of important recent developments.

First, certain EU data protection bodies intend to take “repressive action” against some of Google’s personal data practices. These bodies include the French authority, CNIL (the Commission nationale de l’informatique et des libertés) and the Article 29 Working Party (an advisory body made of data protection representatives from member states). In October 2012, following an investigation led by CNIL, the Working Party raised what it saw as deficiencies in Google’s confidentiality rules. It recommended, for example, that Google provide users with clearer information on issues such as how personal data is shared across Google’s services, and on Google’s retention periods for personal data. Google was asked to respond within four months. CNIL has reported in recent weeks that Google did not respond. The next step is for the Working Party “to set up a working group, led by the CNIL, in order to coordinate their repressive action which should take place before summer”. It is not clear what type of “repressive action” is envisaged.

Google and the ‘right to be forgotten’

Second, Google is currently involved in litigation against the Spanish data protection authority in the Court of Justice of the EU. The case arises out of complaints made to that authority by a number of Spanish citizens whose names, when Googled, generated results linking them to false, inaccurate or out-of-date information (contrary to the data protection principles) – for example an old story mentioning a surgeon’s being charged with criminal negligence, without mentioning that he had been acquitted. The Spanish authority ordered Google to remove the offending entries. Google challenged this order, arguing that it was for the authors or publishers of those websites to remedy such matters. The case was referred to the CJEU by the Spanish courts. The questions referred are available here.

The CJEU considered the case at the end of February, with judgment expected in mid-2013. The case is obviously of enormous relevance to Google’s business model (at least as regards the EU). Also, while much has been made about the ‘right to be forgotten’ codified in the draft EU Regulation (see above), this Google case is effectively about whether that right exists under the current law. For a Google perspective on these issues, see this blog post.

Another development closer to home touches on similar issues. The Court of Appeal gave judgment last month in Tamiz v Google [2013] EWCA Civ 68. Mr Tamiz complained to Google about comments on the ‘London Muslim’ blog (hosted by Google) which he contended were defamatory in nature. He asked Google to remove that blog. He also sought permission to serve proceedings on Google in California for defamation occurring between his request to Google and the taking down of the offending blog. Agreeing with Google, the Court of Appeal declined jurisdiction and permission to serve on Google in California.

Mr Tamiz’ case failed on the facts: given the small number of people who would have viewed this blog post in the relevant period, the extra-territorial proceedings ‘would not be worth the candle’.

The important points for present purposes, however, are these: the Court of Appeal held that there was an arguable case that Google was the ‘publisher’ of those statements for defamation purposes, and that it would not have an unassailable defence under s. 1 of the Defamation Act 1996. Google provided the blogging platform subject to conditions and had the power to block or remove content published in breach of those conditions. Following Mr Tamiz’s complaint, Google knew or ought to have known that it was causing or contributing to the ongoing publication of the offending material.

A ‘publisher’ for defamation purposes is not co-extensive with a ‘data controller’ for DPA purposes. Nonetheless, these issues in Tamiz resonate with those in the Google Spain case, and not just because of their ‘right to be forgotten’ subtext. Both cases raise this question: it is right to hold Google to account for its role in making false, inaccurate or misleading personal information available to members of the public? If it is, another question might also arise in due course: to what extent would Leveson-inspired amendments to the s. 32 DPA 1998 exemption (on which the ICO is consulting) affect service providers like Google?

Facebook, Google and jurisdiction

The Google Spain case also involves an important jurisdictional argument. Google’s headquarters are in California. It argued before the CJEU that Google Spain only sells advertising to the parent company, and that these complaints should therefore be considered under US data protection legislation. In other words, it argues, this is not a matter for EU data protection law at all. The Spanish authority argues that Google Spain’s ‘centre of gravity’ is in Spain: it links to Spanish websites, has a Spanish domain name and processes personal data about Spanish citizens and residents.

Victory for Google on this point would significantly curtail the data protection rights of EU citizens in this context.

Also on jurisdictional matters, Facebook has won an important recent victory in Germany. Schleswig-Holstein’s Data Protection Commissioner had ruled that Facebook’s ‘real names policy’ (i.e. its policy against accounts in psuedonymous names only) was unfair and unlawful. The German administrative court granted Facebook’s application for the suspension of that order on the grounds that the issue should instead be considered by the Irish Data Protection Authority, since Facebook is Dublin-based.

Here then, is an example of ‘one-stop shop’ arguments surfacing under current EU law. The ‘one-stop shop’ principle is clearly very important to businesses. In the Facebook case, it would no doubt say that its ‘home’ regulator understands its business much better and is therefore best equipped to assess the lawfulness of its practices. The future of EU law, however, is as much about consistency across member states as about offering a ‘one-stop shop’. The tension between ‘home ground advantage’ and EU-wide consistency is one of the more interesting practical issues in the current data protection debate.

Enforcement and penalties issued by the ICO

One of the most striking developments in UK data protection law in recent years has been the ICO’s use of its enforcement and (relatively new) monetary penalty powers.

On the enforcement front, the Tribunal has upheld the ICO’s groundbreaking notice issued against Southampton City Council for imposing audio recording requirements in taxis (see Panopticon’s post here).

The issuing of monetary penalties has continued apace, with the ICO having issued in the region of 30 notices in the last two years. In 2013, two have been issued.

One (£150,000) was on the Nursing and Midwifery Council, for losing three unencrypted DVDs relating to a nurse’s misconduct hearing, which included evidence from two vulnerable children. The second (£250,000) was on a private sector firm, Sony Computer Entertainment Europe Limited, following the hacking of Sony’s PlayStation Network Platform in April 2011, which the ICO considered “compromise[ed] the personal information of millions of customers, including their names, addresses, email addresses, dates of birth and account passwords. Customers’ payment card details were also at risk.”

In the only decision of its kind to date, the First-Tier Tribunal upheld a monetary penalty notice issued against Central London Community Care NHS Trust for faxing patient details to the wrong number (see Panopticon’s post here). The First-Tier Tribunal refused the Trust permission to appeal against that decision.

Other penalty notices are being appealed to the Tribunal – these include the Scottish Borders notice (which the Tribunal will consider next week) and the Tetrus Telecoms notice, the first to be issued under the Privacy and Electronic Communications Regulations 2003.

It is only a matter of time before the Upper Tribunal or a higher court considers a monetary penalty notice case. At present, however, there is no binding case law. To that extent, the monetary penalty system is a somewhat uncertain business.

The question of EU-wide consistency raises more fundamental uncertainty, especially when one considers the mandatory fining regime proposed in the draft EU Regulation, with fines of up to €1,000,000 or 2% of the data controller’s global annual turnover.

By way of contrast, 13 administrative sanctions for data protection breaches were issued in France in 2012, the highest fine being €20,000. Enforcement in Germany happens at a regional level, with Schleswig-Holstein regarded as on the stricter end; overall however, few fines are issued in Germany. How the ‘one-stop shop’ principle, the consistency mechanism and the proposed new fining regime will be reconciled is at present anyone’s guess.

From a UK perspective, however, the only point of certainty as regards monetary penalty notices is that there will be no slowing down in the ICO’s consideration of such cases in the short- to medium-term.

It is of course too early to say whether the developments outlined above will elevate data protection law from a supporting to a leading role in protecting privacy. It is clear, however, that – love them or hate them – data protection duties are increasingly relevant and demanding.

Robin Hopkins