Laurence Yaffe has gathered some information about DOE funding of US HEP theory groups, showing sharp drops (average 23%) in such funding for groups reviewed in FY2013 and FY2014. These drops imply serious reductions in the numbers of theory graduate students supported and in the number of postdoc positions funded. To get an idea of the reaction to these numbers, the file name is calamity.pdf. Sean Carroll has a blog posting here, also with the “Calamity” title.
There’s something very odd though about these numbers, with no explanation available from Carroll or Yaffe, who both interpret the situation as a sharp reduction in US government support for HEP theory. If you look at the numbers for total amount of funding, there’s no evidence of the 23% drop (for sources of numbers, see references in Yaffe’s document, the DOE HEP budgets here, and HEPAP presentations here). The recent Congressional agreement on a FY2014 budget puts the DOE HEP total at $798 million, significantly higher than FY2013 ($750 million) which was reduced by the sequester, and higher than FY2012 ($770 million) and FY2011 ($776 million). Of these amounts, the amount going to theory seems stable at around $50-$55 million/year (exact numbers depend on exactly what you include, Yaffe quotes FY12 and FY13 at an identical $51.3 million/year).
So, the “calamity” of collapse of government support for theory is somehow taking place despite no collapse in the amount of money budgeted for this. What is going on? I’d love to hear from someone who understands this. The only explanation I’ve seen is that this is a temporary phenomenon having something to do with how the budgeting process works. Note that with the way typical multi-year grants are made, the DOE is promising to provide money several years in the future, despite the current US budgeting environment, where budgets are typically set not in advance, but often very late. The current fiscal year’s budget was set last month, although the fiscal year started Oct. 1. This was considered a huge success… One conjecture is that the DOE has been promising less to theory groups, partly because they only recently found out about the good FY2014 result, and partly because of a policy change to budget for future year outlays in current years. If this is the source of the calamity, it should quickly disappear as the FY2014 money comes in and the transition to new budgeting ends. Yaffe examines some other possibilities for explanations, but without anything conclusive.
The main concern raised is about the effects of a reduction in the number of US HEP theory grad students and postdocs. Perhaps more worrying than this though should be the trend of numbers of permanent research positions in the subject, which Erich Poppitz has gathered here. These numbers come from the Theoretical Particle Physics Jobs Rumor Mill, and show a constant level of 11 jobs/year for the past three years, about half the level of the years 2000-2007. The number of jobs posted this year is so far only 13, which can be compared to around 20/year in recent years. It’s still early in the season, so more jobs may appear, and more may get filled than in recent years, but the trend of US institutions hiring significantly fewer permanent people in HEP theory is clear. Given this trend, there’s a reasonable argument to be made that numbers of grad students and postdocs should also be reduced. There’s no evidence though that some decision has been made to do this, or that this is the reason for the numbers Yaffe is looking at.
Comments from anyone well-informed about this are strongly encouraged, while comments from people who just want to make the usual arguments about government funding for science rules/sucks will be deleted.
Update: More commentary at Ted Bunn’s blog. He sees this as possibly an implementation of changes at DOE designed to “decrease the effect of historical inertia”, which might be moving money from historically well-funded major groups to others.
Possibly all that’s going on though is just the “bridging” problem mentioned on slide 7 of this presentation. A decision was made to start grants later than in the past on April 1 (with the idea that typically it’s not going to be until that deep in the fiscal year that Congress has got its act together and given DOE a budget number). This meant though that money had to be taken out of upcoming grants to cover extending current ones until April 1 when new ones might start. The number on that slide reflects a 25% charge to new grants to cover this, suspiciously close to Yaffe’s 23% number.
The same slide though does point out that this is a temporary problem: “will be better in 2015”. If that’s all it is, and grad student/postdoc number are slated to go back up a couple years from now, maybe “calamity” isn’t the right word here.
Update: For more from Yaffe, see here.
At the end of calamity.pdf Yaffe lists the 25 institutions from which he collected data. What % of total HEP theory research in the US do these institutions represent (honest question. I have no idea)? Maybe there’s been a shift of support from a few large groups to lots of smaller ones?
If you want more readers of the sympathetic yet amateur nature, you could expand a few acronyms, such as HEPAP (High Energy Physics Advisory Panel) and HEP (High Energy Physics).
The hiring data doesn’t look like a secular downward trend, it looks like a boom in the ’00s that is now over. If hiring goes back to 90s levels, it hardly seems catastrophic…
Noah,
Given that there were about 80 US theory Ph.Ds/year during the 00s (best estimate I could make when I was writing my book), so about 4 times as many as jobs, I don’t remember anyone back then thinking that was a “boom” in the job market. There was an awareness that things had gotten a bit better than in the 80s-90s when the job market was even worse, but one person’s “boom” is another person’s “return to normal level”.
More relevant would be a comparison of the HEP theory job market with other parts of physics and with other fields in academia. My purely anecdotal impression of the math job market is that (after post 2008 problems), it is getting back to something like the market of the 00s, and certainly not something like the HEP situation of half as many jobs. Again, anecdotally, I hear of places where universities are cutting back in physics as opposed to trendier sciences (buzz words here are “data science” and “personalized medicine”), and of physics departments where HEP physics is viewed as having an uncertain future.
Peter,
Is there a summary of such anecdotes somewhere, or would you mind sharing yours? A rumor-mill of which physics depts or HEP groups might have uncertain futures would be very interesting to a lot of people. (Such as those people whose own futures depend on them, for example.)
Jollick,
I don’t have any very significant information of that kind. I was more referring to general comments about the perception of physics in general and HEP in particular among administrators and academics in various fields. At any given moment there are certain subfields of science getting attention as “the next big thing” in terms of advances, especially technological ones. “Data Science” is hot now, and as I mentioned, I’ve heard that “Personalized Medicine” is starting to get attention. In physics, the promise of quantum computation has off and on gotten this sort of attention. In the run up to the LHC start, LHC physics was a hot topic, and if you wanted to get an HEP job, there was a time when that was the thing to be doing.
So, all I meant to indicate is that I don’t see HEP physics as something perceived anywhere as a “hot field”, which university administrations want to expand in, so the job market is likely to remain limited for the forseeable future.
Peter,
As far as the math job market goes, good it’s not, though of course it will never be as bad as physics. This year we’re hiring, and we put our ad in late (December). We still got about 350 applications, we’re a regional comprehensive state school, you’re expected to do research but you also teach a pretty heavy load. Historically in math the market was quite good until 1992 (the year I got my doctorate, perfect timing 🙂 The market got worse for maybe 5 years, and then as far as I remember the VIGRE grants started up, and things got better. Problem was it was a temporary fix, once people started coming out of the VIGRE things got much worse again, and haven’t really improved since then. Most math PhD’s can get academic jobs of some sort, unlike in physics, but it’s a tough slog and you’re likely to end up with not what you were expecting.
Peter, according to information I just learned today, the DOE/HEP theory budget for FY14 is down about 5% from last year, despite the fact that the total DOE/HEP budget is up substantially. This presumably reflects the planned rebalancing between research and projects. I would call a 5% cut a significant reduction, but not a calamity. What turned this into a 20+% hit on renewing grants was a decision to shield from cuts (or at least, cut much less than 5%) a majority of the DOE-supported theory program, namely lab-based theory groups, and university-based groups who had undergone comparative review in earlier years. So the central concerns are not changes in total funding for high energy physics, or even total funding for high energy theory, but the allocation of funds which were available, and the resulting impacts on young researchers.
Thanks Laurence,
That starts to make things clearer. A 5% cut is a significant one, and I can see how a decision to take it out of just one subclass of grants could magnify it greatly for those affected.
Does the change in priorities from “research” to “projects” inherently mean a change in favor of experiment (since only experimentalists have “projects”), or are there also theory “projects”?
Hi,
I am the department head of the theory group at SLAC National Accelerator Laboratory. Larry has written an interesting piece, and he has nailed a serious problem: funding for theory is decreasing, year after year, in order to divert more money to the project line. This is also true for the experimental research lines, but, in principle, they can recoup some (but not all) of the monies from the project lines.
The real problem is decreasing monies for science. The DOE theory research line was cut by ~5% in FY14, after a cut in FY13, after a cut in FY12…. This is bound to result in a decrease in postdoc and student positions, and senior faculty salaries. This decrease in funding is combined with a change in the funding process for the DOE, namely the institution of comparative reviews. A perfect storm! The comparative review process is similar to what has been done at the NSF for years. The labs have undergone comparative reviews since 2008, and the universities started in 2012. Personally, I think this is a very fair way to allocate ever decreasing funds. Rather than let historical precedence be the guide, what matters most now is what a researcher actually does. The most productive researchers are being rewarded and whatever fat there was is now out of the system. (I have to say that I come from a working class family and I was brought up to think that what I actually did is what matters, so I like the new system.)
Nonetheless, the numbers and graphs in Larry’s paper are misleading. First, only some fraction of the groups responded – generally it’s the folks that wish to complain that respond. Second, the DOE instituted salary caps ($15k/month) for summer salary payments and this was not factored into his graphs depicting the decrease in funding. I know of some groups whose cuts were solely from the caps. (It’s worth noting that lab scientists are in general paid less than their university counterparts and few of us would hit this cap.) Third, the weak FY13 postdoc hiring cycle is a direct result of the FY12 (and not FY13) comparative reviews due to the timing of the annual hiring cycle. Fourth, the picture at the labs is not as rosy as Larry paints. Lab groups have been cut every year, in sync with the overall theory budget. After the last lab comparative review, several senior scientists with lab “tenure” were fired. Postdoc positions were reduced. At SLAC, we have graduate students from Stanford, and the number of student positions we could afford was cut in half.
Lastly, I would like to stress that the single worst thing the theory community could do is to fight amongst itself. We could divide ourselves up many different ways: lab vs university, formal vs phenomenological, old vs young, red-headed women who wear blue-jeans on Tuesdays vs everyone else….and this hurts us greatly. Divide and conquer is the famous phrase and that is appropriate here. If half of the community loses its funding, it goes to projects, not the other theory half. We must stick together and fight for appropriate funding for a world-leading theory program.
-JoAnne Hewett
It would seem like the actual theory reductions in DOE funding are being more than compensated for by private funds (FXQi, Milner, Perimeter, etc). The targeted and concentrated nature of the private money is problematic however.
As far as I can tell, the same is happening here in Portugal, with a tremendous transfer of funds from individual grants and money for fundamental research to the so-called projects and applications. The trend is a cutback from fundamental and theoretical research and a move to “corporate science”. Unfortunately, this new trend is nothing like the olden days of, say, Bell Labs. If this is any indication at all of a wider trend in the EU, I guess the move is timed to happen at the same time on both sides of the Atlantic. This means that the old university-based research is at an end and so the future of most of fundamental research.
Thanks Joanne!
What I’m now finding hardest to understand is why DOE would make such large cuts in grant amounts to some subset of the community, seemingly without providing any explanation for why this is being done. There seems to be agreement that part of the problem is a significant 5% FY14 cut to overall theory research (has there been a rationale given for this choice?) Beyond that though, it seems unclear whether the much larger cuts seen by some groups are due to: salary caps, better treatment of the labs, better treatment of groups reviewed earlier, the April 1 bridging problem, change in policy about future year commitments, rebalancing from well-funded groups to others, or something else. If the DOE does not clarify things, that will make worse the problem Joanne points out of people concluding that others are being better treated than they are for dubious reasons.
Peter, this maybe a controversial pov, but so far people working on beyond standard model of particle physics(whether it is supersymmetry/technicolor/extra dimensions etc) have failed to make a single successful prediction which has been verified by accelerators. So certainly I wouldn’t be surprised if there is a cut in HEP theory
funding. Certainly I hope more funds are allocated to new research directions.
This is something Avi Loeb has advocated in astrophysics/cosmology.
http://online.kitp.ucsb.edu/online/colloq/loeb1/
But same idea applies to HEP.
Shantanu,
I just don’t see any evidence for the idea that these new cuts in HEP theory are due to lack of progress in the field, or that funds are being redirected to new research directions. If anyone has real evidence for a DOE policy change of this kind, that would be interesting, but otherwise I’d rather people not speculate here based on what they would like to be true.
I just chatted with a bunch of Biology and Chemistry professors over the weekend on the Federal funding situation. There the money crunch does not appear as severe. NIH funding is stable; only NSF money seems threatened. One department chair has an appointment with a congressman on Monday. Instead of the overall funding level, her main concern appears to be the (over-)emphasis on applied research projects over fundamental ones.
“Given that there were about 80 US theory Ph.Ds/year during the 00s (best estimate I could make when I was writing my book), so about 4 times as many as jobs, I don’t remember anyone back then thinking that was a “boom” in the job market.”
What is the precise fear that people have here?
I see two very different issues which people seem to be commingling.
(a) Not “enough” support for the field, meaning not enough money to pay for what is considered a reasonable (by some sort of metric) number of profs and their assistants (at all levels, from postdocs to grads to, maybe, technicians like programmers).
(b) Not enough academic theoretical physics jobs available for academic physics PhDs.
These strike me as very different problems. In particular, (b) strikes me as very unlikely to be a real problem in any sense. The people who get theoretical physics PhDs do so because they have a compulsion they cannot control, not because of rational calculations about job opportunities. So I don’t see the flow of talent into the field drying up.
On the other side, I can’t see much individual tragedy happening (except insofar as there are ALWAYS going to be losers with positional games, and that’s life). People who get theoretical physics PhDs are among the smartest people out there, and employers know that. You may WANT to be a prof at Harvard, and be bitterly disappointed that you cannot even be a PostDoc at State U, but you’re not going to starve. Depending on your inclinations, you can get very rich in finance, you can join an internet startup and figure out new algorithms, you can join an established company (Google, Apple, FaceBook etc) and make a comfortable living as a technical programmer, you can move sideways into EE, and so on.
Maybe I’m wrong here — I’m not trying to be a dick. I’m just looking at my personal experience and the experience of my cohort. A few in finance, a number scattered over computing over all sorts, one at a government lab, one acting as a high level technical project manager for the feds, and so on.
So while concerns for the amount of money supporting the field strike me as justified (if they ARE justified by reasons why this amount of money is too low — and of course we all WANT the amount to be higher, but let’s be realistic); bringing the academic jobs for PhDs issue into it strikes me as confusing the issue in a way that’s completely unhelpful.
Maynard Handley,
The problem Yaffe was discussing was very specifically about DOE funding of postdoctoral positions (and to some extent graduate fellowships). These positions are intended as training positions for permanent positions involving research in HEP theory, of exactly the sort (tenure-track academic positions or laboratory positions) followed by the rumor-mill. So the question of how many postdocs the DOE should fund is closely related to the question of how many positions are available for such trainees.
Again though, while the decrease in available HEP theory permanent positions would be a sensible reason for reducing the numbers of postdocs funded, I haven’t seen any evidence that that’s the actual reason for the recent decrease that Yaffe is concerned about.