Wednesday, 24 July 2013


I’m about to go to 20-year high school reunion – so time for a personal reflection on whether and how my personality may have changed! [a bit related the cartoon: click to enlarge]

Actually, it’s fun to go back a bit further. So, to start with, a translated extract from the school certificate of my first year in primary school:
“The lively young boy easily makes friend and gets along well with his fellow classmates. He has a sense of fairness and helps weaker classmates […]. He has made adequate progress in reading. His results in maths were fine […]. It is also laudable that he tries to improve his not always clear handwriting.”
Then, really the year of my high school graduation. Here, actually, not the certificate (that would be boring, just grades), but an extract from our student-newspaper written by a classmate (which in the German tradition is supposed to be funny but can also a bit mean; ie its different from a US style yearbook):
“Our clear thinker […] was friendly and cooperative - but treated school as a place of learning not a place to socialise. […] Unfortunately, he will not receive the Nobel Prize in mathematics since he plans to study law – of course due to reasons of logic.”
So have I changed? I still know a bit of maths. My handwriting is still horrible. I get along with most people (occasionally, I even socialise...) and I hope I still have a sense of fairness. Of course, indeed, I have not received the Noble Prize in mathematics – but then it may also need to be considered that there is no such a Nobel prize anyway …

Thursday, 18 July 2013

Meritocratic quotas

The modest aim of this post is to understand the possible working of a meritocratic form of quotas (ie, it does not aim to provide a general discussion of the pros and cons of quotas). For this purpose, let’s assume two groups (A and B) and that we want to appoint the ten best persons to a particular committee (eg, a board of directors).

To start with, let’s assume that half of the population is A and the other half B.
Who is likely to be on the committee in an ideal meritocratic society without quotas? That’s a straight-forward binomial distribution – so to calculate: there is a 24.61% probability of a 5/5 split, a 41.02% for 6/4, a 23.44% for 7/3, a 8.78% for 8/2, a 1.96% for 9/1 and a 0.2% for 10/0.
So how should, or shouldn’t, meritocratic quotas be structured:
  • (i) insisting on a 5/5 split would be counter-productive since in 75.39% (100% - 24.61%) of the cases this is not the outcome that we would have if we appointed the best persons.
  • (ii) insisting on no quota: that would only be meritocratic if any risk of cronyism or other informal networks can be excluded (eg, if we have a fully objective test). But not sure this can be assumed in a situation where one appoints known individuals, eg, to a board of directors.
  • (iii) insisting on a minimum of at least 4 would in at least 65.63% (24.61% + 41.02%) of the cases support the optimal outcome; thus, it would be meritocratic if it can be said that the 7/3, 8/2, 9/1 and 10/0 outcomes are likely to show evidence of cronyism. But social sciences usually require a higher level of evidence, typically 90 or 95%: thus,
  • (iv) insisting on a minimum of at least 2 would in 97.85% (24.61% + 41.02% + 23.44% + 8.78%) of the cases support the optimal outcome. Thus, from a meritocratic point of view, this seems to be the most plausible result.
Variant: what if we change the split, say, that being on the committee requires a particular qualification and 80% of A and 20% of B have it? Then, again, who is likely to be on the committee without quotas?  Calculating leads to a 10.74% probability of an all A committee, a 26.84% probability for 9 A and 1 B, a 30.22% probability for 8 A and 2 B, and a 32.22% probability for 3 or more B. Here, applying (iv) there would be no quota. But, perhaps, if one group is dominant (ie here A) there may be a higher risk of cronyism. Thus, one may also apply (iii) which would lead to a quota of 2. 

Friday, 12 July 2013

Is there a legal origin effect in investor protection law (as measured by the Doing Business Report)?

The legal origin effect has been the topic of some of my previous writings, where – using new datasets – it was found that differences between common and civil law countries are often irrelevant. But what if we use the Doing Business Report data on investor protection? I’ll discuss this in a forthcoming paper, but as a preview, see the figure below on the 25 most populous countries:
This is based on the paired country differences of each of the individual investor protection variables (see eg here for the UK), yet omitting the final four variables since those are really about civil procedure. The network picture shows the closest links, and countries that are close (according to the data) are shifted closer together.
So, my answer to the question above would be ‘no’. Ok, a few common law countries are linked together (even if one includes Thailand which perhaps one shouldn't), but everything else is fairly ‘messy’. That’s indeed how I would say company law is today – there may be some residual ‘old’ similarities, but in modern times transplants take place across the world: so any claim of distinct groups of countries is now fairly dubious.

Tuesday, 2 July 2013

Academic conferences and the shifting scope of themes and borders

Two weeks ago I attended a conference on New Institutional Economics (ISNIE) and was impressed by the broad scope of topics and methods covered. I missed two other conferences this year (SASELSA) which also encompass very diverse topics. This made me wonder about the title of this post – and to draw the following figures:

This shows the centrifugal effect of conferences such as ISNIE, SASE and LSA. They start with a core topic but are happy to accept papers that broaden the perspective. This is to be applauded as unexpected contributions from outsiders may often be more thought-provoking than just repeating the established core.

However, there may also be events that are centripetal, ie they deliberately focus on the, real or perceived, core. This may aim to reaffirm the status quo, but it can also lead to a slight shift of the core (see the figure). Such more narrow events are perhaps more typical for invited conferences (see also earlier post).

But just thinking about these centrifugal and centripetal effects is perhaps also too simplistic. In the final picture there is again the larger circle showing the topics that can be included applying a broad understanding of a particular theme. Then, however, there are also sub-topics, some overlapping and some even going a bit beyond the broad theme. In a conference this may be addressed by panels on particular sub-topics. In addition, it is useful not to forget how all these sub-topics relate to each others and are embedded in the broad theme: thus, it is plausible that conferences, such as the ones cited above, also tend to have some general sessions, keynote lectures etc.