NSERC Discovery Grants: how we spend the money

An average Discovery Grant in mathematics is about 18K (it has varied slightly between the two math GSCs and the competition years). At UBC, it’s somewhat higher. The 2007-08 UBC average, with only 6 awards made, was 24.3K. The 2006-07 UBC average was 23.6, with 18 awards. I have computed this based on the GSC 336 and 337 results and may have missed a few faculty who applied through other GSCs.

This post is for those of you who would like to know what we do with that money.

First, a disclaimer. This post is written from the perspective of a mathematician, and more specifically, a “pure” mathematician. Most of our research is done using pen and paper, or blackboard and chalk when it’s collaborative; once in a while, it’s computer-assisted. We travel, attend conferences, supervise students and postdocs. We do not, for the most part, set up high-tech labs or conduct physical experiments. Therefore nothing will be said in this post about lab equipment and other expenses that researchers in other fields encounter. Furthermore, the financial arrangements made by researchers for their students, for example, may be completely different in other fields of science. I’m only talking about mathematicians here.

The bulk of our grants is spent on support for postdocs, support for graduate and undergraduate students, and travel. I will elaborate on each of these in a moment. Other expenses, including books, computer equipment and supplies, and other miscellaneous items, add up to a much smaller fraction of our grant, in my case no more than 2-3K per year on average. As far as computers are concerned, sure, they’re not cheap, but we don’t generally buy a new one every year, either.

How much we spend on travel can vary quite a bit from year to year, depending on the number of trips, their destinations, and availability of support from other sources. It also depends on the number of mouths to feed graduate students and postdocs we support. It’s important for them to travel, too.

It can cost as little as a few hundred dollars to attend a conference somewhere nearby, especially if inexpensive dorm-style accommodation is provided. A typical week-long trip from Vancouver to an East Coast location, either in Canada or the U.S., will cost 1.5 to 2K. A trip to Europe can set you back several thousand dollars. On the other hand, most (but not all) conferences offer at least some financial support to the speakers: for instance, I pay the airfare and the organizers pay my local expenses. Some conferences have additional funds to support other participants, especially students and postdocs. (Hint to graduate students: you shouldn’t always wait for your supervisor to suggest that you apply. If there’s a conference that you’d like to attend, and in particular if there’s an opportunity to apply for support, take the initiative and go talk to your supervisor.)

We also travel for the purpose of collaboration. It’s a widespread custom by now to collaborate via Internet, sending files back and forth by email or keeping them on protected webpages – that’s how I co-wrote many of my papers and a book – but there are times when it’s best to meet face-to-face for at least a few days, for example when we’re starting a project or when we need to look for a new direction. Typically, but not always, the host pays for the guest’s visit. The number of visitors we have and trips we make for collaboration purposes depends, among other things, on whether there are other local faculty that we can work with. Those of us who are more isolated in our departments will have to travel more often.

The NSERC guidelines restrict airfare reimbursement to full ecomony class fare. We try to keep the travel costs down in other ways as well, by combining trips when possible, booking in advance, looking for low airfares and reasonably priced hotels. However, the cheapest Expedia packages are generally not for us. If that conference in Europe is in July, then that’s when we have to go, even if the airfares are less expensive in April.

Overall, I’d say that a reasonable minimum amount for travel is somewhere around 4-5K for the award holder, plus 1-2K for each supervised student and postdoc. If you don’t have that, you end up getting left behind in this profession, and so do your students. There are years when we spend less, especially if some of our travel is paid from other sources. There can also be very good reasons to spend significantly more, say 10-15K, in some years.

The next big-ticket item is support for graduate students. At UBC Math, a typical support package for a graduate student who does not have an external scholarship (NSERC or UGF) consists of an 8K/year research assistantship, paid from the supervisor’s NSERC grant, and a 12K/year (approximately) teaching assistantship, for a total of about 20K per year. That’s not a lot, given the high cost of housing in Vancouver: it’s hard to find a studio apartment within reasonable commuting distance to UBC for less than $1000 per month, and shared accommodation can range from $500 to 1000 per month, per person. This amount has not changed much in recent years, even though the costs of living in Vancouver have gone up – a fact that the students recently brought to the attention of the graduate committee. The TA salaries are determined by the contract between UBC and the TA union. The standard research amount has been the same since I came to UBC in 2000; I anticipate that it will have to be raised, especially if prospective students start voting with their feet against our financial package.

On the other hand, though, our grants are not indexed for inflation and increased cost of living. Some of us have just had our funding cut due to a budget shortfall. Travel isn’t getting cheaper, either. If we increase the support amount for students, we will likely end up admitting fewer of them.

Undergraduate students do not normally receive year-round research assistantships. Those who wish to get involved in research usually start with NSERC Undergraduate Student Research Awards. USRAs are typically held in the summer. The student is awarded $4,500 from NSERC, which has to be supplemented by at least $1,125 from the project supervisor’s research grant. I have never supervised those, for various reasons, but some of my colleagues do it on a regular basis.

Recall that the average Discovery Grant in mathematics in Canada is 18K; at UBC, let’s say it’s 24K. Based on what I’ve said so far, an 18K grant can support the researcher and 1 graduate student, and there may be enough money left for a USRA. A 24K grant may be enough to support the researcher and 2 graduate students, but you have to be careful with the money.

But I haven’t even mentioned the postdocs yet.

NSERC requires that the minimum salary for a postdoc funded by a Discovery Grant must be at least 25K per annum; but 25K is just barely more than what we offer to graduate students and that wasn’t a high life to begin with. At UBC, the minimum postdoc salary is 40K per annum, plus benefits. (“Benefits” is the part of salary that is not actually paid to the postdoc: it’s the money that UBC pays on her behalf towards unemployment insurance, pension plan, and other such.) This consists of a teaching salary ($10,500 per course, for up to 2 courses), external funding such as NSERC Postdoctoral Fellowships (available only to the best few), and a research stipend from the supervisor’s grant. In the absence of external funding, the supervisor must pay 20K per year if the postdoc teaches 2 courses, or 30K if the postdoc teaches only 1 course, which is becoming more frequent.

Yes, I did say that an average grant in our department was 24K per year.

Well?

Not all of us have postdocs. When we do, they’re often supported jointly by several faculty – that’s another reason, if you needed one, why life is easier when you’re part of a group. Also, if we don’t spend the full amount in a fiscal year, NSERC will normally let us carry the balance over to the next year; we can then use that balance, together with the current amount, to fund a postdoc. (Kudos to NSERC for allowing us this flexibility.)

In addition to that, we try to supplement our grants from other sources. That – what those sources are, who is eligible, how much we can get – will be the subject of the next post in the series.

Speaking of which – I was way too optimistic when I thought that 2-3 posts would cover all I wanted to say on the subject. I can think of enough material for at least 2 more right now.

‘Til the next time, then.

4 Comments

Filed under research funding

4 responses to “NSERC Discovery Grants: how we spend the money

  1. Thanks- I was just wondering how my Discovery Grant spending compares to others’. I think that students cost me a bit more, since TAships are not supposed to be part of support in my ECE department, but Waterloo is cheaper. I also think of my $18k NSERC as being enough to fund 1 student. Of course, it looks like we’re expected to have about 4 students, somehow.

    Teaching pay seems high at UBC math. I got half as much teaching a course at McGill.

  2. Pingback: Harms and Company Consulting » Archives » Discovery Grants – budget breakdowns

  3. Pingback: Harms and Company Consulting » Archives » Discovery Grants – budget breakdowns

  4. Pingback: James Colliander's Blog » Mathematics Discovery Grants are Insufficient and Broken