lördag 15 september 2012

My price tag

I wrote a blog post two weeks ago about my efforts applying for research grants during this past spring. To summarize:
- I worked on five different applications and four were finished (handed in) during the spring.
- These applications represented two different brand new ideas that we started to work on around a year ago (in two different constellations of people).
- I estimated that I had roughly spent between 200 and 300 hours working on these applications during the previous academic year.
- This far, I have nothing to show for it. Three out of four applications have been rejected. I'll know about the fourth application in a month or so.

The competition for research grants is fierce. It is perhaps to be expected that a majority of the applications will be rejected. But the fact that three of my applications were rejected and we got no motivation or explanation at all leaves a sour aftertaste after having put so much time into this endeavor. It is difficult to rework and improve an application when you have no idea if you were close to the pot of money or if you were miles away from it. Does another round with the same ideas (reworked, updated applications) represent a real chance at getting a research grant, or does it only represent "throwing good money (or rather throwing good time) after bad"? Were my applications fundamentally sane (but there wasn't enough money to go around), or were they screened out in the very first round (and if so, for what reason)?

I would never get away with asking my students to do something that represented many dozens of hours of work and then leave them with a "pass" or a "fail" and no feedback or motivation whatsoever (not even a grade). But it is apparently all right for research grant agencies to treat applicants that way... The situation is more than just a little Kafkaesque... A fitting quote from a book I recently read:

"what we have is [...] winners and losers at the end of a series of tests which were largely invisible, barely specified, poorly supervised and far from stable."

Two of the four applications where handed in to Riksbankens Jubileumsfond, RJ. Out of 899 applications that were handed in, 107 were selected for the second round, and about half or those were expected to get research grants. All in all, it now seems that 43, or a little under 5% of the applications got funding (list of granted projects here).

That means that 20 out of 21 applications drew a blank, and 1 in 21 drew a winning ticket. The odds are tough, to say the least. If, say, altogether 100 hours were spent on writing up each of those 899 applications, that would add up to 90 000 hours, or 50 man-years! We spent a lot more than 100 hours on each of the two applications I was involved in because they represented brand new ideas, but perhaps many of the applications are just reworked versions of the previous year's many rejects? Still, even slashing the time in half (50 hours per application) adds up to quite a few man-years. It is definitely worth thinking about the decreasing (individual, societal) returns of increasing the competition for the same amount of funds even further. Is this what we want researchers to spend their time on - writing up unsuccessful research grant applications - or do we want them to spend that time (50 man-years? more? less?) on actually doing the research? Do also note that a considerable amount of time (and money) is tied up in reviewing the applications and in other administrative tasks (keeping track of funds, following up on the results of the research grants - "getting value for money" etc.). The overheads are thus considerable.

My final application was handed in to Vetenskapsrådet, VR (The Swedish Research Council). They have some basic information about last year's call (2011) on their homepage. Out of 825 applications, 125 were granted funds (15% or almost 1 in 7). The average application was awarded around 3.250.000 SEK (for a 3 year long project). Everything else being equal, the chances of getting money from VR are obviously much higher than RJ.

So what are the "costs" of the 200-300 hours I spent writing research grant applications last spring? In one way, the costs are zero. Writing research grant applications is something researchers to a large extent do in their free time (as a "hobby"), so the costs for my employer, to say nothing of the costs for the grant agency or society are invisible. Writing research grant applications thus merges with other "unpaid work" in the home (cooking food, washing, cleaning, taking care of children or ailing parents). Some people steal office supplies from their employers - perhaps many wanna-be researchers steal time from their employers, writing grant applications when they are actually supposed to be doing other things? That would then mean that there are costs (in terms of decreased quality of their output at work) that are being bourn by others than the individual researcher(s) themselves.

Not happy with that kind of accounting, I tried to find another way to reason about, and put a price tag on the 200-300 hours I spent on writing research grant applications. I looked up the official statistics to find figures for the average salary of an assistant professor (in the natural sciences or technology). I then checked out official Swedish tax rates.

Basically, someone in my position is paid 150 SEK/hour (after taxes) in Sweden. When I apply for research grants, I however have to "charge" the grant agency a lot more for my time. Like any other consultant (or plumber or carpenter for that matter), I have to charge a lot more for my time than what I myself get paid. First I have to add around 67% in personal taxes (ending up with the take-home salary before taxes). Any employer furthermore has to pay general payroll taxes, adding at least another 50% and KTH furthermore adds another 60% or so to cover its internal costs (rent for my room and for lecture halls, electricity, administrative personnel, investments etc.). For reasons that I don't understand, the actual hourly rate I have to charge a research grant agency is still considerably higher so I presume the payroll taxes are even higher than 50% and that KTH overheads are higher than 60% because my time - when I apply for research grants - costs more than 700 SEK/hour!

Multiplying 700 SEK/hour with 200-300 hours adds up to something like 150 - 200 000 SEK. This number is of course not a real "cost", but it should perhaps be kept in mind and deducted from the research grant itself should I get one. The average VR grant (3.25 million SEK) is equivalent to 4 500 hours at 700+ SEK/hour. How many of those hours have already been spent reading up, thinking through and writing the application itself? How should the hours spent on unsuccessful applications be accounted for (or shouldn't they - isn't it customary for "losers" to pay the costs for putting together unsuccessful research grant or job applications, for bank bailouts etc.)?

The 700+ SEK/hour rate is a huge number. Remember, "I" (i.e. the average salary for someone doing my job) personally get to lay my hands on less than a quarter of that money when I get my monthly salary. The rest (75-80%) basically represents the costs of societal complexity - the costs of getting the whole gargantuan machinery of modern welfare society to run. If I could instead charge half, say 350 SEK for my time, it is easy to understand that more research could be done for the same amount of money, but there would also be less money around for paying salaries to the administrative personnel at the university, and less money for paying the salaries to civil servants and public employes (nurses, kindergarten staff, politicians, taxi services for the disabled etc.). Or less money to pay for the salaries of assistant professors and other university employees for that matter... (at this point, it is easy to have your head spinning).

I would be very interested in learning if something has been written about "the political economy of research" today compared to 20 - 50 - 100 years ago? On the one hand, there are lots of more researchers around today than 50 years ago. But might it not be the case that despite living in an age of affluence, less research is being done today? Or perhaps more research is being performed in absolute numbers, but less research per person (i.e. each researchers spends less time on research and more time teaching and performing administrative tasks of various kinds - including tasks that create little concrete output, like spending very large amounts of time figuring out whether to hire person X, Y or Z as a professor)? And what about the actual effects of the money being poured into research? Are the returns of money being spent on research decreasing in physics, medicine and energy research - as Swedish blogger Flute suggests?

I recently quoted E. F. Schumacher who in his book "Small is beautiful: A study of economics as if people mattered" (1974) wrote that "The richer the society, the more impossible it becomes to do worthwhile things without immediate pay-off". A corollary could perhaps be that "the more complex the society, the more impossible it becomes to do worthwhile things". Complexity costs, and those costs cannibalizes whatever it is you really wanted to do in the first place ("worthwhile things").

Having qualified personnel at research grant agencies sift through hundreds of applications, weighing and measuring them up against each other and against the latest round of fickle political, economic and strategic goals brings its own costs. Having 20 out of every 21 applications fail also brings its own costs (at least for the applicants). Perhaps it would be easier to just give every wanna-be researcher the means to spend 25% of his/her time doing research, and then evaluating the results of that research rather than evaluating the ability to write seductive texts about the wonderful benefits of future, not-yet-performed (and most often never-to-be-performed) research? All in all, I think that something needs to be rethought in terms of how research is funded (or not funded) today.

For a more general take on the larger picture of how hard economic times might affect research and higher education (over the coming decades), see "Peak oil, 'Big Education' and 'Big Science'" that I wrote two years ago (originally written in Swedish in July 2010).

Inga kommentarer:

Skicka en kommentar