Monday, 23 July 2012

Professorial salaries and performance in 29 countries

A week ago I posted on the average professorial salaries at UK universities and how this is related to one of the university league tables (see here). In this post something similar for a comparison of 29 countries: data on average professorial salaries have been published by a Swiss newspaper a few months ago (note: in Swiss francs), at the top Switzerland, then Canada and South AfricaFor ‘performance’ I took the number of top 200 universities in the Times Higher World University Ranking, the top countries being the US with 75, then the UK with 32 and then the Netherlands and Germany with 12 universities each. But, of course, one has to consider that more populous and richer countries have a natural advantage. Thus, the two figures below present the top 200 universities per population and per total GDP (data from here and here [IMF data])– here, the best countries being Switzerland, the Netherlands and the UK.
.
In both figures the correlation is fairly similar, and positive, 0.53 and 0.51. That could be the result of many developed and transition economies not having any university in the top 200 and having relatively low professorial salaries. But, when we omit these countries, the correlation is actually even slightly higher: 0.58 and 0.53. Thus, the relationship is indeed positive (and, of course, if one were running a regression, it would not merely be a linear one), similar to the result of the previous post.
.
Interpretation: countries such as Netherlands, Switzerland and the UK aim for international excellence, as shown by good performance in the ranking and relatively high salaries. By contrast, France, Germany and Japan seem to have a more local strategy, thus fewer highly ranked universities as well as lower salaries. With respect to the countries that do not have any top 200 universities but pay above average salaries, this may indicate corruption and cronyism but it could also be the result of an aspiration to attract academics that can get them into the league of top universities in the future (perhaps the case for Saudi Arabia and Malaysia, less sure about the other countries; and to clarify, of course, it may also be the case that the ranking does not capture properly how good these universities really are)
.
PS: to clarify that these are average data (for the higher salaries at US law schools see here).

Wednesday, 18 July 2012

Pay without Performance in UK Universities?

A couple of weeks ago the Times Higher published data on the average salary of UK academic staff per university and the remuneration of the universities’ vice-chancellors (VCs). How is this related to how well universities perform, eg, in the league table of the Complete University Guide? The following three charts use data on the average salaries of professors, the salaries and benefits of the VCs, and the overall scores of universities in the league table of the complete university guide (in the following CUG-scores).

(1) Relationship between professorial salaries and CUG-scores

(2) Relationship between VCs remuneration and CUG-scores

(3) Relationship between professorial salaries and VCs remuneration
It can be seen that in all three relationships there is a positive correlation (to be precise: 0.66 for (1); 0.50 for (2); 0.46 for (3)). I also calculated the correlation as regards the sub-categories of the CUG-scores (these are: entry standards, student satisfaction, research assessment, graduate prospects) and in general the correlations are very similar; however, for student satisfaction it is a bit lower (0.35 for (1) and 0.22 for (2)).
.
Now, all this seems to indicate that there is not ‘pay without performance’ in UK universities (PS: for this phrase see here). But it also needs to be said that these correlations do not show causal relationships (indeed, causalities are likely to be messy since they may go in both directors; also there are bound to be omitted variables). Still, it’s interesting to see that we do not observe negative correlations - which would not have been inconceivable (eg, lower ranked universities feel that they have to pay more to attract good staff; V-Cs get high salaries to the detriment of professorial salaries).

Friday, 13 July 2012

Academic conferences: how to decide on possible speakers?


To have the best possible speakers, it seems to me preferable to have free competition than just to invite who comes to your mind (note: I have indicated some intermediate positions but there may be more). Thus all four lines point upwards. 
But then we may also distinguish: (1) is my standard model. Here, the organisers try to be fair, eg, in the invited-speakers option they try to identify who may deliver good quality. That’s different in (2) which is a ‘crony model’ – and such cronyism is more harmful for closed events (ie you would only invite ‘friends and family’). (3) is an improvement to (1) since here we have a particularly skilful organiser which is positively reflected in all types of conferences; thus, as indicated by the ‘dots’ a skilful-invited-speaker-conference may be better than a less skillfully managed competitive conference. But then what’s (4)? I think it’s also fair to assume that some very famous and good speakers would just not take part in a competition but would want to be invited, eg, as keynote speakers. Thus, having some of those invited (well chosen) plus a free competition seems to be the best option.
Finally, of course, these are models: in the real world other things matter too (eg, how the proceedings are published, though not sure how this could be illustrated in the chart). Also, of course, this chart does not answer the question why a particular event takes place in a particular form since individual incentives may well diverge from the overall value.

Sunday, 8 July 2012

Visualisation of EU harmonisation in company law

A slide I used for one of my company law lectures. I’m aware that this is a bit simplistic but I do think that I got the green/yellow/red right – the only thing I’m wondering is whether I should not have been more Europhile – ie making the high-level harmonisation green (not red): will do next year!