Saturday, 26 November 2011

REF 2014: what did the law sub-panel members submit in 2008?

The Research Excellence Framework (REF) 2014 will use various tools in order to assess the research quality of UK universities. The most important one is that expert panels will assess four pieces of research output submitted by each UK academic. This leads to the question whether the panel members will prefer certain types of research output as well as certain publishers or journals. In some disciplines, there are clear preferences (e.g., in economics quality-rankings of journals are widely seen as relevant). Law is a complicated case since academics disagree on the role of proxies, such as the alleged quality of certain publishers and journals.

Now, the composition of the REF law-subpanel is of course known (here): thus, I quickly examined which four pieces the panel members had submitted for the previous exercise (the RAE 2008, data here). The 55 pieces are:
  • 55% journal-articles with 77% of these articles in specialised journals, and 23% in general ones. Overall, no journal had more than two “hits” (these were Modern Law Review, Legal Studies, and Journal of Law and Society; no hits for LQR, OJLS and CLJ).
  • 25% book chapters with 88% of these in OUP, CUP and Hart books (8 OUP, 2 CUP, 2 Hart).
  • 16% books with 55% of these OUP, CUP and Hart books (2 OUP, 2 Hart, 1 CUP).
  • 4% government reports (ie 2 pieces).
This shows a couple of interesting points:
  • Occasionally, book chapters are seen with scepticism since they are not peer-reviewed. Yet 25% is a good share of the submission - though the strong preference for the top-publishers may indicate some hesitation.
  • With respect to books, there is also a preference for the top publishers, yet, here books by other publishers (Ashgate, Palgrave, Cavendish etc) are also not infrequently submitted.
  • Then, law journals, which is often the most controversial point with some favouring articles in a small number of general journals. Yet, in the REF-panel members submissions there is a lot of diversity with specialised journals dominating the field.
Finally, to say the obvious, the motivation of this post is influenced by the assumption that one may expect that the assessments of the panel members are influenced by their own publication choices. I think that’s not unrealistic to make, though of course one would not expect a perfect correlation. 

Saturday, 19 November 2011

Presentation of networks with NeighborNet (example: shareholder protection in 25 countries)

I recently came across an interesting form of presenting similarities and differences between languages in a network (see here at the bottom), being based on the option "NeighborNet" of the program Splitstree. Such a network uses data which indicates how different pairs of languages (or other “nodes”) are: e.g., for something like on a scale of 0 (identical) to 1 (completely different) German and Dutch may get a score of 0.3.

In my own research I have done something quite similar: the paper "The Web of Creditor and Shareholder Protection in 25 Countries" (available here) is based on data on the legal differences between each pair of these 25 countries. In the paper, I present the data as a fairly simple network (using the program Ucinet), with bold lines presenting very similar countries, normal line somehow similar countries, and no-lines fairly different countries; also: close countries were shifted close together:

Now, I applied the Splitstree-NeighborNet program to the same data. It looks as follows:
Since the data are the same, it is no surprise that in both figures the same countries are close to each other (e.g., see US and Canada, or more surprisingly Pakistan and Mexico). But which one is preferable? My original one may be easier to understand, given the fact that the lines present similarities/dissimilarities, whereas NeighborNet is based on a more complex form of presentation (to simplify, it’s an extension of drawing up trees of relationships). But, admittedly, the NeighborNet picture looks a bit cleaner and perhaps sexier… so I may be using it for my future research a well.

Monday, 14 November 2011

OWS and The Crisis: Who are “the bad guys”?

I just had a quick look at the Harlem News, one of the free newspapers lying around at subway stations in NYC. A bit surprisingly, there was an interesting article on how, allegedly, the Occupy Wall Street (OWS) movement harms the cause of freedom. I also found the text at the website of the Future of Freedom Foundation. To quote from it:
After many weeks, Occupy Wall Street and its kindred demonstrations around the country are still a source of headline controversy (…). Unfortunately, the loudest voices call for more government management of the economy, when it is precisely that which got us into the mess we have yet to dig out from.(...) Contrary to popular misconception, this was no case of rampant deregulation, but rather one of rampant regulatory privilege. (…) Every device to protect banks from their own folly — from deposit insurance to implicit guarantees to the Fed’s promise of emergency cash injections — has contributed to the misery that sent the protesters into the street. (…) The problems here are that regulatory agencies invariably end up serving the regulated industries (…) Demanding more power for government is equivalent to demanding more privileges for Wall Street (…) [being] a creature of the corporate-state partnership that has characterized the American economy for generations. 
Thus, it is claimed that not business/market failure but too much government/regulation was the cause of the crisis. The two main examples (above in italics) are the “moral hazard argument” (government guarantees etc) and the corporate capture of US law-makers. Is this a sound argument? As such, the two examples may show some government failure; however, these examples are really about how the interplay between firms and law-makers can lead to problems. Thus, they show that a plain binary thinking in terms bad/good business v bad/good government neither helps us understanding the crisis, nor provides the best way forward. For further reading see also the Varieties of Liberalism project.

Sunday, 6 November 2011

The Law and Finance of Share Repurchases in Europe

I had a busy week, presenting an empirical paper on share repurchases (co-authored with Amedeo De Cesari) at events in New York and Chicago (links here). Actually, it is likely to be not one but two papers, one addressed to a law and one to a finance audience. The ‘law paper’ is about to be finished. Its abstract reads as follows:
Recent years have seen a growing interest in research on law and finance but this paper is the first one to analyse the impact of the EU market abuse law on share repurchases. We find that the Member States’ previous rules differed considerably, and therefore it can be said that the Regulation on share repurchases has provided clarity as to the availability of an EU-wide safe harbour for share repurchases. The picture, however, gets more puzzling if we consider our findings on the actual effect of the law. Our results do not confirm a “simple law and finance story” according to which market participants would have just reacted as expected by the plain legal rules. Rather, it seems to be the case that problems of law enforcement, the value of legal certainty, and the positive signal of common legal rules have also had an impact on the propensity to repurchase own stock.
The paper is not yet publicly available; so please email me if you’re interested in it.