I had a twitter exchange with Nick Cohen, the Observer columnist, today after he tweeted that:
Leveson [is] dodging issue of why libel costs in Britain a 150x European average. Why not just cut lawyers’ fees?
I queried the figure as dubious and pointed out the study was deeply flawed. Nick retorted, reasonably enough, that he preferred Oxford (who published the study) to libel lawyers (who he assumed did not produce the study) but has asked me to substantiate my concerns about the Study.
Readers can find it here.
The central claim, which I have seen recycled a number of times in debates about libel costs, is taken from the report’s executive summary. I highlight it in bold:
The comparative element of the study analysed data collected from various jurisdictions across Europe via a questionnaire, which included specific questions about two factual scenarios.
The data showed that even in non-CFA cases (where there is no success fee or insurance) England and Wales was up to four times more expensive than the next most costly jurisdiction, Ireland. Ireland was close to ten times more expensive than Italy, the third most expensive jurisdiction. If the figure for average costs across the jurisdictions is calculated without including the figures from England and Wales and Ireland, England and Wales is seen to be around 140 times more costly than the average. The data also showed that common law jurisdictions are by far the more expensive jurisdictions in which to conduct defamation proceedings. This was exacerbated by the use of CFAs.
Based on the collected data the study was able to identify costs factors, unique to the common law tradition, which partially explain (although they do not justify) the comparatively high costs in England and Wales. Although the collected data did not enable the study to pinpoint the precise underlying reasons for these costs increases in numeric, proportionate and interrelated values, it did give strong suggestions as to why England and Wales is the most expensive jurisdiction.
The headline figure of 140 times greater than average is dubious in a number of respects:
1. It is based on questionnaire responses to two cases.
2. Those responses were received from twelve countries. One lawyer or firm of lawyers estimated the costs of those two cases in their own countries.
3. Averages are thus calculated on tiny numbers on two hand picked cases.
4. The two case studies are based on real cases but we do not know how those cases were chosen and in particular how typical they are.
5. The cases appear to have been selected by a firm of solicitors who represents defendants in media cases. One of the solicitors in that firm has actively engaged in lobbying for defendant media interests.
6. The questionnaire was designed by this same firm of solicitors with input from the researchers.
7. The answer to those questionnaires was, for England and Wales, given by that same firm of solicitors. They author one of the chapters.
8. This is a study where a lobbyist designs, answers and then authors parts of a research report on an area in which they have taken a keen policy interest.
It is worth also noting that the study was commissioned by Associated Newspapers.
Whatever the circumstantial concerns about this research, the most important thing is that the key figure is based only on 12 estimates of costs on 2 cases. It is a tiny, flimsy basis upon which to be calculating averages and then generalising comparisons between England and Wales and the rest of Europe.
That of course does not mean the figure is wrong; it may be right, but a more serious critical and objective approach to data collection and analysis would be needed to get even close to an accurate figure. Nor is the research without interest: on this evidence a lawyer in France thinks French courts can do in hours what our lot take days over. And I do not disagree with Nick Cohen’s overall point, it must be better for there to be a cheaper system (I’d opt for an Ombudsman and cut lawyers out completely). But as a serious, quantitative comparative study of costs it is deeply flawed in design and execution. If one was to rely on these kinds of estimates, cases had to be selected in greater numbers, with a semblance of balance in that selection, and data collected from a range of lawyers with a range of perspectives.