Dinner at St Trinneans with EPSRC

And I do mean St Trinneans. Its a smairt old building just next to Pollock Halls that really did used to be a School called St Trinneans , and yes, really did inspire the books and movies. For those following my privileged lifestyle, the dinner provided by Edinburgh First, while perfectly ok, was nowhere near as good as the Athenaeum or Trinity College Oxford. But of course the company was first rate, being comprised of both my senior scientific colleagues and assembled officers of that distinguished public body, the Engineering and Physical Sciences Research Council (EPSRC). Over the last two days, we have had a kind of EPSRC Roadshow, as part of our “Framework Agreement”. Sharing intelligence and strategic thinking and all that jazz. Anyhoo, it did of course spin my head back to summer 2006. Should we have put our grants with EPSRC ?? Mike M, keep quiet until I have finished. So, what did I learn from EPSRC ?

Non-cash primer. First, a quick summary of what John Peacock explained to us recently. When a Research Council wants a big new thing, DIUS give them the necessary money as capital in their allocation, and proudly announces the increased science budget. However, Treasury rules require the Research Councils to make provision for the future by depreciating the value of the item, and gradually paying this money back to the Treasury. To make this possible, DIUS give them a special non-cash allocation. So this money whistles in and straight out .. but of course it too is announced proudly as ”more money for science” even though its the same money counted twice.

EPSRC has depreciation costs too. Just not much. For example they funded the new HECToR supercomputer at £117M, and this has to be depreciated. (Its irrelevant that its operated by us in Edinburgh : its an EPSRC capital item.)

Universities have depreciation costs too. A standard way of getting equipment funded on an EPSRC grant is to cough up first for your NMR machine etc, but then charge for its use on a grant. What you charge is the depreciation of the asset following standard rules. Of course you can only do this if you paid for it yourself. You can’t ask for the NMR machine on one EPSRC grant, and then depreciate it on a second grant. Don’t be silly. The Government shouldn’t pay for the same thing twice now should they ?

Non-cash is a one-way filter. EPSRC, STFC etc really do owe the Treasury the depreciation costs. The non-cash allocation is just that – its an allocation, to help the RCs meet their bills. Its not guaranteed to be magically the right amount. Now you can’t turn non-cash into cash. But if necessary, the RCs can use their cash to meet their depreciation bills. Frightened now ?

EPSRC’s problem is FEC. EPSRC have a healthily growing overall allocation, and a fairly minimal depreciation obligation. However, they need to find extra for FEC. Dave Delpy showed a graph subtracting this additional obligation, showing the remainder to be roughly flat cash, i.e. buying power declining with inflation. So … grant volume will decrease. You hear less fuss about this, because while it dampens people’s plans, they are not withdrawing things they already promised. Also of course, that FEC money has not disappeared. We got it. Knock on the Dean’s door.

EPSRC don’t have a Roadmap. Most of the last two days was centred round those cross-Council themes – Digital Economy, Securing the Future, etc – and there was much talk about future priorities, supporting the community, longer grants etc. However the financial truth is that nearly all the money is still in straightforward responsive mode grants. (Called “Essential Platform”, not to be confused with “platform grants”.) They do have strategic thinking, but its all quite general, to do with future technologies, trends in science etc. What they don’t have is a document that says “in 2011 we will need to take a decision on X” or a mentality that says “if we really intend to do Y in 2013, we will need to make sure key skills are maintained in groups A, B and C”. This is because they are dominated by lots of small things rather than a few big things, so they can rely on the external market sorting itself out. Its up to them to provide the money and some strategic hints; planning groups and keeping skills etc is our problem, not theirs.

So should astro grants be in EPSRC ? I think there is an argument that much of theory and regular exploitation could fit perfectly well in EPSRC. However, say goodbye to rolling grants and any sense of stability or nurturing. On the other hand, most project grants (building an instrument, running post launch support, building AstroGrid etc) really fits with the facility world – it needs that long term planning, and is intimately connected with the facilities. Is there a clean dividing line ? And do we really want to re-open that box ? And would anybody let us anyway ?

OK Mike, your turn.

14 Responses to Dinner at St Trinneans with EPSRC

  1. Michael Merrifield says:

    I agree.

  2. andyxl says:

    Is that it ? You probably have lots of marking to do …

  3. Michael Merrifield says:

    Just thought I would throw you off guard a bit! Well, alright, maybe that was a little too unqualified. Make that “I mostly agree.”

    I don’t think anyone, including me, argued that we should simply dump astronomy exploitation funding into EPSRC without any modification to its structure. As was pointed out at the time, EPSRC’s charter has no commitment to blue-skies research, and would certainly have needed reformulating to meet its new obligations, just as PPARC’s did in taking on the CCLRC agenda. I therefore do not believe that we would have had to, for example, abandon the idea of rolling grants if they are the appropriate vehicle for sustaining strength in blue-skies research.

    In the last decade the nature of rolling grants has evolved from a vague theme-based system to a focused project-based one, and back to a theme-based one. Along the way, no-one seems to have done any kind of detailed systematic analysis of what works best, or whether we should be looking at some different vehicle entirely. If we had joined a new “Science Exploitation Research Council” (SERC!), these questions would have had to be asked, and I think it would have been good for us to at least consider alternative models while making the case that rolling grants are the best solution.

    I also agree that there is no single point at which one can place a clean division between exploitation and facilities in astronomy or physics. However, if you are going to fund them through separate agencies, there are solutions that are more or less optimum. The current deeply-unbalanced arrangement, with astronomy exploitation with the facilities that support it but physics not with the facilities that support it, is, in my view, a long way from optimum. It was always obvious that it was a dumb idea to set up a grossly asymmetric system in which astronomy exploitation could be tensioned against its facility spend, but physics’ couldn’t. It was also obviously a dumb idea to set up a system in which a limited cash budget was so exposed to large non-cash expenses. Those of us opposed to the current model made these points very strongly at the time of the merger, so this is not just being wise after the event. In fact, my only real anger at the way things came out is not that we lost the argument, but that those who won the argument (mentioning no senior figures in the RAS in particular) were so busy celebrating their “victory” that they failed to consider the possibility that we might have had a point, and take appropriate steps to mitigate the associated risk through effective lobbying during the setting up of the new research council new structure, etc.

    As for your concerns about projects and post-launch support, I tend to agree with you. My view is that a facilities council should take a suitably twenty-first century view of what constitutes a facility, which should certainly include large software projects like AstroGrid. Similarly, post-launch support is in essence no different from staffing a ground-based observatory, and should clearly be built into the whole-life costs of building and running facilities.

    On the issue of mismatch between facilities and exploitation, this is a risk, but one that is a lot easier to mitigate than the current ones that are crippling us. For a start, the facilities side will not want to build facilities that no-one wants to exploit. By building useage and citation metrics into its “scorecard,” it would be quite straightforward to ensure that the needs of users are being met. One would also not set these organizations up in isolation, but rather would have strategic decisions from an RCUK level that would describe the scientific landscape within which the research councils would take their decisions. One would also hope that professional bodies like the RAS would feed in subject-specific strategy at this level to define the agenda that the research councils are then responsible for delivering.

    So, to answer your three questions:

    1) There is no clean dividing line, but some lines are much better than others.

    2) The box has never really been shut, and it would be a mistake to close down the discussion under the misapprehension that everything is now set in stone. With Wakeham potentially recommending major changes to the funding structure for the subject, we need to be actively campaigning for the arrangements that we need. I am certainly not in favour of perpetual revolution, but equally I think it would be disasterous to advocate stopping the music at a point where the chairs are so appallingly badly arranged.

    3) We are being invited to keep the box open. In addition to Wakeham, there is also an election no later than 2010, which again creates the likelihood of major change before the next spending review. Let’s try and get it right this time!

  4. Paul Crowther says:

    Aside from theory, astro grants are largely observational (exploitation) in nature, and so inherently responsive in nature. Aside from instrumentation groups, these are intrinsically much closer to EPSRC’s approach than the large, long-term rollers of particle physics.

    The 40% shift towards bids for astro rollers at the last round in 2007 did not occur because groups magically became more coherent or long term in nature, but because all the signs from STFC (many guidelines inherited from PPARC) seemed to be pushing groups down this route – as the numbers below illustrate, its much harder to get a “fundable” grant actually funded, if it enters the system as a standard grant.

    The cut to the volume of STFC grants has been especially ill-timed for those submitting rollers in 2007, since the actual PDRA numbers awarded by AGP hardly provided the much needed “stability”. Indeed, many rollers were ultimately converted to standard grants (by AGP or PI’s), as these vital statistics illustrate:

    Astro rollers: 87.5 fundable, 63.7 awarded, subsequently reduced to 53.6 PDRAs

    Astro standard: 42.2 fundable, 19.4 awarded, subsequently increased to 29.5 PDRAs

    Putting the prestige of a large roller aside, I suspect many of us would prefer EPSRC’s genuinely responsive process of applying for standard grants to the options presently available in STFC, even putting their respective grant reductions over CSR07 aside (3% at EPSRC; 25% at STFC).

  5. andyxl says:

    Paul – I doubt if its true that “astro grants are mostly exploitation in nature” … a large fraction of the funding to University groups is concerned with building instruments, data archives, post launch support, operating eMerlin, etc. (Would be interesting to see the ratio.) However, for astronomy the “build and operate” grants are fairly distinct from the “exploitation” grants, whereas for PP everything is routed through the group rolling grants; so for us it should be possible to decouple whereas for PP it is very hard.

  6. Paul Crowther says:

    According to my spies in STFC land, the 25% cut in astro grants from circa 330 to 250 over CSR07 relates just to “responsive” (standard + roller) grants, and excludes project-based staff (“instruments” etc), which were unofficially guessed to be 30-40. I based my argument upon these guestimates.

    If most post-docs were actually employed to do project stuff, there wouldn’t be much exploitation going on, which would strike me as pretty odd. The 2008-11 STFC Delivery Plan highlighted a 44% increase in research output from the astro community in 2006/07, which coincided with increased PDRA support for exploitation in recent years from AGP, which is now set to take a dramatic downturn.

  7. Paul Crowther says:

    According to my spies in STFC land, the 25% cut in astro grants from circa 330 to 250 over CSR07 relates just to “responsive” (standard + roller) grants, and excludes project-based staff (“instruments” etc), which were unofficially guessed to be 30-40. I based my argument upon these guestimates.

    If most post-docs were actually employed to do project stuff, there wouldn’t be much exploitation going on, which would strike me as pretty odd. The 2008-11 STFC Delivery Plan highlighted a 44% increase in research output from the astro community in 2006/07, which coincided with increased PDRA support for exploitation in recent years from AGP, which is now set to take a dramatic downturn.

  8. Michael Merrifield says:

    I have had an interesting chat with some particle physicists, who are of the opinion that one could probably divide their instrument development funding from their exploitation funding, in much the same way that groups involved in both sides of astronomy could divide their support.

    As discussed earlier, such a division is not perfect, but it is surely more optimum than the current arrangements as long as appropriate higher-level strategic bodies are in place to mitigate the risk of any mismatch between instrument development and exploitation needs.

  9. Dave Carter says:

    Mike, do you think that is appropriate that University groups which develop instruments, and in a couple of cases to operate facilities, should have to bid for funds to the same body which owns and operates facilities which could be seen as competing, either for users or for contracts?

  10. Michael Merrifield says:

    Dave — Do I gather that you would like funding for instrumentation also to be moved away from the body that operates the facilities? While that would certainly address your concern by putting these potentially-conflicting interests at arm’s length, I suspect that it would cause more problems than it solved, by creating such a strong division between a telescope and its instruments. My view would be that the way to deal with this potential conflict is by having the rights checks and balances within the funding agency to ensure fair competition when such competition is desirable. Certainly my experience with the ELT Steering Committee has been that, with appropriate management, it is quite possible to have constructive collaboration as well as healthy competition between university groups and STFC establishments even though the funding is all channelled through STFC.

  11. Dave Carter says:

    Ok Mike, its good to hear your experience here. I can see both sides of this.

  12. Alan Heavens says:

    The mother of a friend of mine attended St Trinneans. Apparently the people in charge weren’t in control and the underlings were rebellious. Good to see some things don’t change – historical continuity and all that. Not that I’m commenting on your own School, of course.

  13. Michael Merrifield says:

    “As we know from the current situation at the STFC,
    the nature of the council is such that if you have any
    gains or losses they get concentrated in the bits of
    flexible money they have,” says Peter Main director of
    science and education at the Institute of Physics. “This
    is generally the money they supply to people to exploit
    facilities—the money given to do research”

    But only if you insist on keeping them under the same budget holder…

  14. […] This post was mentioned on Twitter by Jonathan Tedds. Jonathan Tedds said: RT @StephenSerjeant: Reminder of what #STFC near/non-cash means: by @e_astronomer http://is.gd/j8rml John Peacock http://is.gd/j8rOo Han … […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: