18

Follow-up question to Why do most scientists think Brexit is bad for British science?

If Brexit is bad for British science, then naively one would expect that the rankings of British universities should worsen after Brexit. This doesn't seem to have happened. Among the three most widely-followed rankings and comparing to the 2016 results (which is when the Brexit referendum happened):

  • In the QS rankings, Oxford improved (#6 to #2), Cambridge did not budge, Imperial & UCL switched places with one another, Edinburgh improved (#21 to #16), Manchester improved (#33 to #27), KCL dropped (#19 to #35), LSE dropped (#35 to #49)

  • In the THE rankings, Oxford did not budge (it couldn't improve either, since it is ranked #1), Cambridge dropped by 1 place, Imperial dropped by 4 places, and UCL dropped by 3 places.

  • In the ARWU, Oxford did not budge, Cambridge improved by 1 place, UCL did not budge, Imperial dropped by 3 places, Manchester did not budge, Edinburgh dropped by 3 places, and KCL improved by 3 places.

Overall it doesn't look like much has changed. Some universities certainly moved a lot, but not the UK's universities as a whole.

Why hasn't the rankings of British universities dropped? I can think of many possible reasons, e.g. the effect is delayed, the rankings are not reflective of the situation on the ground, the British government provided the political support to have Brexit without British science suffering (per avid's answer to linked question), etc., and I am wondering which explanation(s) are correct.

12
  • 3
    Well Oxford and Cambridge I get, but how did you select the others? Nov 3 '21 at 4:18
  • 2
    They are all the universities in the tables. E.g. Edinburgh is not ranked within the top 20 on the THE, so I did not list it.
    – Allure
    Nov 4 '21 at 3:30
  • 11
    Why do you believe "science" equals "University rankings"? This position rejects the scientific activities of industry and government (civilian and military) labs, as well as other broad areas of "science". Nov 4 '21 at 15:02
  • 2
    @EricTowers I don't say it's "equals". But it's a reasonable proxy, unless you can think of a better way to measure how good British science is.
    – Allure
    Nov 4 '21 at 15:07
  • 2
    The lack of a better measure of the quality of British science does not imply the university rankings is a good measure. A meaningful attempt to measure the quality of British science would likely have to be an index based on several values. University rankings could be one value, new patent applications, science prizes and awards, number of papers submitted and/or accepted by major journals; the list of meaningful ingredients in an overall science quality index could be very long. Nov 5 '21 at 4:41
97

Most of the papers that are being published right now were written pre-pandemic. Most of the researchers who might, long-term, decide to leave the UK or who now no longer consider coming to the UK are ... still exactly where they were pre-Brexit, simply because all of these things take time.

Rankings don't change dramatically because of such reasons. And then there is also this: There is not really any difference between universities between rank 20 and 50, and probably also not between 2 and 6. In other words, whatever effects you quote in your question is simply noise.

It will take a long time -- ten years or more -- before you will be able to see an effect of Brexit on rankings, if any.

10
  • 30
    @toby544 that's not the point. Whether there is an effect or not, it will only show up (or not) with a delay. (But I've made a small edit. @ Wolfgang Bangerth, feel free to roll back.)
    – henning
    Nov 3 '21 at 10:10
  • 4
    Even if the papers weren't written, the work they were based on was done pre-brexit. Even if not done, funded.
    – Chris H
    Nov 3 '21 at 12:59
  • 38
    There is no ranking system that should be taken as a face value.
    – fraxinus
    Nov 3 '21 at 13:56
  • 9
    @XavierStuvw Yes, precisely. When you teach a class of 50 students, you can rank their final numerical score from 1 to 50. But the top five students are all equally good, the difference is just whether or not they had forgotten a minus somewhere, or forgot to consult a specific reference and so their scores vary from 96 to 92 out of a hundred. The difference is not significant. But collectively, these five are better than the ones at ranks 20-25 who are in turn better than the ones at ranks 45-50. Nov 3 '21 at 17:42
  • 6
    Whether there will in fact be a change long term I don't know. @henning's edit to add "...if any" at the end captures my sentiment. Nov 3 '21 at 17:43
47

British universities are not limited to Oxford, Cambridge and a bunch of other top50 ones. In cases like this, it is usually the bottom line which suffers the earliest (and the most overall). Oxford may remain at top positions just by the virtue of having preferential treatment from the government: direct financing, possibly easier to get visas for international students, this kind of thing. All compared to a a significantly less prominent university, of course.

Plus, it is a really long process, but that was mentioned already.

9
  • 10
    This. The first admission process directly affected by Brexit immigration policies was the previous one, i.e. for the cohorts that are just starting. And we can definitely see a decline in the calibre of the students applying to what used to be "UK+EU" studentships (but are now UK-only) simply because the pool of potential applicants is now much smaller. No change in "international" studentships. As there are many more available funded PhD positions for UK students, mid-tier Unis will have to make compromises, and research output will suffer, more in time as more such cohorts are enrolled.
    – penelope
    Nov 3 '21 at 11:28
  • 21
    There has been a huge increase in international students in The Netherlands, which is now apparently the EU country with the largest choice of English language bachelors (maybe behind ROI, or even ahead of it).
    – gerrit
    Nov 3 '21 at 14:31
  • 1
    @gerrit this means that the Dutch education system will need more fundings, to provide education to these new international students. ERC and other european fundings usually do not cover teaching, so this may be an issue, not an advantage for the Netherlands, at least in the short term. Why hasn't the rankings of dutch universities worsened?
    – EarlGrey
    Nov 4 '21 at 8:25
  • 4
    @EarlGrey There are studies that show that international students bring economic benefits at least for the UK. Presumably the same is true for the Netherlands. Why there haven't been any changes in a short time frame has been explained well enough in the other answers I'd think.
    – Voo
    Nov 4 '21 at 10:15
  • 12
    @EarlGrey Glad to see that you found a new thing to be upset about so quickly after learning you were wrong.
    – Voo
    Nov 4 '21 at 10:25
30

In statistical inference for time-series data, it is really quite difficult to test for whether or not a policy has a causal effect. The main difficulty with this is that the policy itself may have a delayed effect or even an anticipatory effect (potentially both) and there may also be many confounding variables that impact on the analysis. This kind of analysis is "uncontrolled" observation, which makes it difficult to determine whether a hypothesised causal effect was present or not.

The best you can really do in situations like this is to look for a temporal change in your output variable (e.g., university outputs, metrics, etc.) that coincides with the introduction of the policy (or some appropriate period of delay after it is introduced), ideally conditioning on a range of potential covariates. For the production of scientific work you would need to include a delay that is roughly commensurate with the amount of time it takes to do a research project and have the resulting scientific paper(s) published, which is usually a couple of years. If you are looking at an outcome like university rankings, there might be a further delay involved depending on the way the ranking is measured.

I have no strong view on whether Brexit did/didn't/will/won't harm scientific research output in the UK. At best you might be able to make a rough guess about this hypothesis once you have about ten more years of university output data, but it will still be contingent on a lot of caveats that apply when trying to make causal inferences from uncontrolled observation of time-series data. Another way to try to figure this out would be to look directly at whether there is a drop-off in external funding (that is not balanced by increased internal funding) and/or whether there is a decline in international research collaboration. Regardless, it will take a while for data to come in, and all the normal caveats on causal inferences will apply.

6
  • 12
    +1. Also, Brexit may have an effect that was anticipatory, rather than delayed - even if the actual Brexit only happened in 2020, students (and academics) would have been very myopic not to take it into account already in 2016-2020. This means that the effect will be "smeared out" over multiple years, not a sharp structural change. Nov 3 '21 at 12:50
  • 1
    @StephanKolassa: Good point --- I have edited to note the possibility of an anticipatory effect.
    – Ben
    Nov 3 '21 at 12:58
  • 1
    There is certainly a reduction in EU-funded research. As grants run run they are not being renewed.
    – RedSonja
    Nov 4 '21 at 7:44
  • "The best you can really do..." It sounds like you are saying it's not possible to draw inferences over shorter time frames than 10 years and you are saying we only have access to circumstantial data, no causal models. I have to disagree. We can look at actual known effects of Brexit on factors that are known to be good for science (or university rankings). For example, changes in grant funding, enrollment, international researchers, university finances....
    – usul
    Nov 5 '21 at 0:14
  • We can do that (as noted in the answer) but that is not what OP is proposing; he is proposing to use the university rankings for his inference.
    – Ben
    Nov 5 '21 at 0:39
8

Officially, Brexit only took effect on 31 December 2020, and only ten months have passed since then. I believe it will take more time before the full effects of Brexit impact the rankings of UK universities.

I've watched a number of short documentaries and interviews about EU professors quitting their posts in the UK and moving back to Europe. Some of the best professors I've ever had were from EU countries, so I do think the UK will suffer without them. My personal experience was that EU professors from Germany and Holland often put more time and energy into the students, compared with UK professors. UK professors often treat it like a 9-5pm job, but the Germans and Dutch would go the extra mile and work longer hours to help students out.

I also doubt that Boris Johnson will prioritise university funding, so I think the UK will lose out on funding over time and the university rankings will gradually decrease. That's just my opinion. I truly wish the UK the best and I hope everything works out in the long run.

2
  • 2
    This is an entirely fact-free answer. Your personal experiences with professors of various nationalities are not evidence of anything. Nov 4 '21 at 22:43
  • 2
    @KevinArlin I disagree. The post contains many facts (albeit without sources) and while they are not statistical analyses, qualitative evidence is still evidence of a kind. Moreover the answer addresses the question very directly. It suggests a possible negative effect, but it notes is too soon to show up in the rankings.
    – usul
    Nov 5 '21 at 0:19
6

I imagine the idea that Brexit is bad for British universities depends on two factors:

  • no longer having access to EU research funding;
  • no longer being able to attract (or retain) so many EU scientists.

(There is also the fact of not being able to attract so many EU students, and losing out on EU student fees, but this is likely to be compensated by the fact that those EU students that do still come will be paying the significantly higher international student fees.)

How might these actually play out, and on what timescale? Well, the first one shouldn't really be so much of a problem, since the shortfall could be made up by increased funding availability from UK sources. After all, the UK was a net contributor to EU science funding, so this wouldn't actually involve spending more on science. In fact, in the short term increased UK funding has become available before EU funding has stopped being available (of course whether, in the long term, the increase will be sufficient is difficult to predict). However, even if universities don't lose out overall on funding, different priorities mean that individual areas may be significantly worse (or better) off.

The second one certainly has started to happen; I know people who have moved back to the EU because of Brexit. However, there are several reasons why this will take a long time to be felt. For one thing, there is less reason for established academics to leave, since often they have been in the UK long enough to qualify for settled status, etc. In many cases they also have more reason (e.g. family commitments) to stay. So the main effect is on junior scientists who perhaps have less of an impact on rankings now (but some of whom will eventually be senior scientists who might have worked in the UK).

The other issue, as well as the time lag between research being done and being measured by rankings that others have mentioned, is the following. Suppose there is a research group at UK university X, and as a result of Brexit, some members leave. Collaboration between members of the group is likely to continue (after all, COVID has made international collaborations not much harder than domestic ones). Meanwhile, new members are appointed. So the research output of members of the group is actually likely to increase - less of it will be entirely produced by university X, but that doesn't matter for the purposes of assessment and rankings. Of course, we might expect international collaboration to decrease as a result of Brexit in the long term, but in the short term it is likely to increase as existing domestic collaborations suddenly become international.

2
  • 2
    "After all, the UK was a net contributor to EU science funding" - Have you got a source for this? The UK was a net contributor overall to the EU budget, that doesn't mean it was necessarily a net contributor for science funding, as the UK received substantial science funding
    – Jontia
    Nov 4 '21 at 17:43
  • 1
    Correction: UK has almost full access to the EU funding schemes under Brexit.
    – Dilworth
    Nov 4 '21 at 20:32
6

It can take years to know the full extent of the consequences/effects brought about by Brexit. It's not even two years since it happened. And less than two years is nothing in the evolution and operation of Academia.

Grads and undergrads that came to study to the UK before Brexit are committed to stay the course for years. Same with students that already had plans during/after Brexit.

The same situation applies to research/publishing work being done in collaboration with UK's academia. We are talking about commitment in years.

Whatever negative impact (if any) will occur slowly and over the decades. I suspect it will affect middle/bottom tier universities first/the most. The top tier might be affected, but not that much and would occur later.

There will be consequences, that's just logical. They will not happen immediately after Brexit for Academia, but we see them happening in the financial world (see here for an example.)

Whether the UK's financial and service sectors can adapt and shrug it (or even bounce back stronger) remain to be seen. And anything affecting wealth and work will have an impact on Academia (and just about anything.)

Moreover, anyone from the EU wanting to study or do research in the UK will have to adjust, change or perhaps even abandon plans just because of immigration changes and uncertainty.

But again, the world will not know the full effect for years to come. It is too soon after Brexit to conclude that nothing will happen, just because nothing has happened yet.

2

It will take many years for any effect to become apparent. Decades, probably. Also, Covid has thrown a big spanner into the works of higher education in all countries. Far greater than Brexit, I would argue. For the student experience, it has been little short of catastrophic. This will make identifying any long term changes of trends even harder than it usually is.

Many (most?) of the world's "top" universities are not in the EU (and never have been). Doesn't that fact suggest that it may not be a problem?

One major factor is that English is the World's second language, its lingua Franca. This is likely to be a major influence on students and scientists. Far greater, I would have thought, than any effects of a change of funding from receiving money from the EU out of money paid to the EU by UK governments, to receiving funding direct from the UK government. Obviously, the generosity of such funding is relevant, but the future a decade hence is not knowable on either side of the English channel.

I would have thought UK universities are more in competition with universities elsewhere in the Anglosphere, than universities in EU countries.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.