Tuesday, July 15, 2014

Who will train tomorrow's scientists?

 In 2009, when I was graduating and about to embark on my postdoc, I was concerned about the research funding lottery and worried how I could continue on the academic track with such prospects. My co-adviser Jasper Rine confirmed that we were in a crisis, but he said, “NIH only acts in times of crisis.” It was a good time to do a postdoc, and by the time I would embark on my faculty career, things would have improved.

Of course, in retrospect, 2009 feels like a pre-problem era. The funding situation today is best summed up by yesterday's: "Congress: NIH Funding To Be Distributed Via Cage Fighting" with this brilliant nugget:

"My fellow Americans, today marks an important milestone in our collective history," President Obama began, "with the enacting of this bill, federal funding for health-related research will be distributed in the fairest way possible -- by physical combat."

For as long as I remember from the start of graduate school in 2003, depressing reports on the prospects for PhDs kept surfacing (good examples: TheScientist 2006Science 2008Economist 2010, an entire special issue in Nature 2011, and a famous blogger in 2013 saying that it's immoral to hire PhDs). In February, my post “Goodbye Academia”, warning of a crisis in academic biomedical research, drew over 100,000 readers. Countless scientists commented, confirming the concerns, and I learned of a whole genre called QuitLit of researchers publicly leaving academia (ever-growing list of the stories here). A few days after my post, Keith Yamamoto, Vice Chancellor of Research at UCSF published a video, arguing that we should not be sending graduates to postdoc positions. Then in April, the titans of the US bioscience world published “Rescuing US biomedical research from its systemic flaws”.

I was somewhat encouraged, hoping that NIH would finally recognize the extent of the crisis and avert disaster. Alas, it is clear to me today that despite all the warnings, the vast majority of academics, and certainly the head of NIH, are going to pretend as though there is no problem, for as long as they can. Francis Collins, the director of NIH, just stated that he does not believe we are training too many PhDs.

Figure 1: New faculty positions versus new PhDs.

 Schillebeeckx M., et al. Nature Biotech. 31:938 (2013)

In my conversations with scientists, the waving away of the problem is consistently justified with the following two answers:

1. Everything is uncertain. No employment guarantees in industry either.

2. There are plenty of non-academic jobs, and training PhDs is not vocational training for professorship.

Everything is uncertain. You can lose your job in industry too.

This is not a valid comparison. Of course no employment is guaranteed. However, if I lose my job at Novartis, I can probably find another one at Merck. That’s not the same as losing funding for your lab – you lose funding at MIT, you don’t just move to Stanford. You lose your funding, and you lose your lab. It’s not a job; it’s your entire career that is at risk.

There are plenty of non-academic jobs, and training PhDs is not vocational training for professorship.

I applaud efforts to expose graduate students to non-academic careers. I personally tell students and postdocs all the time that they are valuable to the society because of their science training, across a wide spectrum of research and non-research positions. Yet, I don’t sense that there are enough fulfilling positions for the scientists we train (will write a separate post on this). And I certainly don’t believe that the programs to facilitate non-academic career exploration are in any way addressing the central problem of academia. (By the way, telling students that there are lots of non-academic jobs is also the solution to the crisis that Francis Collins advocates.)

The problem – academia is no longer competitive for the best and most talented researchers.

This is the part that worries me the most. We have too many PhDs competing for postdocs. We have too many postdocs competing for fellowships and faculty positions. And we have too many professors, competing for grants. Combined, the low pay, hyper-competition, and the guaranteed uncertainty at every step make the academic track a bad career choice today.

So what? As many have replied to my Goodbye Academia post, “There is nothing wrong with getting out of academia. We, taxpayers, fund your training so you can join the industry, start biotechs, and not just sit in the ivory tower.”

I love my role at ZappyLab. I do not regret getting a PhD or doing my postdoc – no way I could have founded ZappyLab otherwise. But my question to the taxpayer is, “Who do you want as professors, training the next generation of scientists for the biotech/industry jobs?”

The NIH alone spends $30b each year of taxpayer money, supporting 50,000 grants across laboratories at 2,500 institutes. Who do we want heading the laboratories under these grants? Who do we want training the future researchers? If the best scientists aren’t the ones training, what does it mean for the quality of scientists and science that will be in the industry? 

We have a bubble in academia. Like all bubbles, if ignored, it will pop. There will be a natural adjustment. We are beginning to see it already. As I was inviting mentors to participate in our Career Forum, several warned me that they regretted getting a PhD. They said that they would not be able to give rosy encouragement to students and postdocs. These are folks from the very top research universities. Folks who graduated 5-10 years ago when the situation was far better than it is now. This is phase 1 of the bubble-bursting. The next and most painful phase will be a decrease in the number of talented students going into biomedical PhDs.

The system will naturally self-adjust if we take no action. Eventually, the number of PhDs we train will decrease simply because few will want these degrees. But what will be the cost to science, progress, and society if this happens? The quality of academic research is already suffering with pervasive irreproducible results and outright fraud, as a consequence of the hyper-competition. We are already starting to lose the scientists with the stronger moral compass and ethical standards. If we don’t fix academia and make it an attractive career choice again, we will all pay a devastating price.


Some statistics on percent of biomedical PhDs getting faculty positions:
1963: 61%
1983: 38%
1993: 25%
2003: 15%
2014: <10%
(numbers for 1963-2003 from the Scientist and 2014 from ASCB.org)


  1. Lenny, the part that I find most discouraging is that the competition is not based on substance and merit. It is a perverse selection.

  2. Like the funding, whether you get a faculty job or not is now largely up to chance. Now, the people who do get the faculty jobs do tend to be terrific. The problem is that there are way more amazing people rejected than hired. And as you mention, the selection process now relies so heavily on the meaningless impact factor of the journal you publish in - this is one of the most damaging aspects of the situation today.

  3. We already know that the pipeline from PhD to postdoc to PI is far too clogged -- I've heard it described as leaky, which is a massive understatement. But two thoughts struck me.

    First, I wouldn't necessarily take statements from Francis Collins as the view of the NIH as a whole — after all it is government, where the right hand probably never knows what the left hand is doing. We have seen the NIH engaging in, and putting money towards, addressing these issues.

    Second, there probably are too many PhDs, but I would not be willing to advocate for a less-educated population until we know how many we might need. Most PhDs do find jobs (unemployment for PhDs is extremely low compared to those with Masters or bachelors), I think it’s a matter of better harnessing their skills, talents and drive, making sure we apply these where it’s needed, before any other action. There are certainly too many for the limited number of PI positions, but a PhD in biomedical research, even basic research, can be used in a lot of different fields. Knowing to run a PCR might have limited utility, but critical analysis and creative problem solving can be applied in so many ways...

    1. Agreed on the first point that NIH is pretty big. Edited my post to say that "the head of NIH does not see the crisis."
      On the second point, I mostly agree. Certainly true that a PhD is terrific for a broad range of non-academic careers. But the competition that the biomedical PhDs face is extreme. Don't want here to get into the discussion of total demand for PhDs. But even if there are enough non-academic jobs to absorb the graduate students, the academic track is still in deep trouble.

  4. There is a simple reason for the clogged pipeline, the abundance of adjuncts and the other dreadful facts of life as an overeducated Ph.D. today. There are too many Ph.D.s, and part of it is the fact that foreign non-citizens can easily get a green card or an H-1B. This did not happen in the 80s when I finished up. The job market was not great, but jobs were there to be had. Today, you are competing with huge numbers of people, and part of the onslaught is the overabundance of non-citizens. This should be stopped.

    1. This is a shortsighted, and ultimately inadequate, solution. The chances that a US trainee (by this I mean a trainee studying for a Ph.D. In the US, regardless of nationality) will end up in a tenure-track research position is somewhere around 10-15%. Eliminating foreign trainees would not make up for this short-fall. More importantly, science must be in international venture. Cutting off collaboration, and access to international talent and experience, would stifle medical and scientific advancement in the US, and this is the consensus of those studying these issues in depth — economists, biomedical researchers and policy-makers.

  5. It is amazing to me how every employment problem always gets the comment "too many foreigners". We have a world-wide oversupply of PhDs. We have a world-wide problem in biomedical academic funding. We have a US-specific problem of NIH and US universities having too many training grants and accepting too many graduate students. So easy to blame the foreigners, but that solves none of the structural systemic problems of our academic training and funding. I personally find this type of comment deeply offensive.

    1. The fact is that this is a fact, and your being offended is not relevant. US citizens are not competing in the world. US citizens are competing in the US. Very few, if any, US citizens attempt to get biomedical jobs in China, but thousands of Chinese attempt to get jobs in the US. Same with India. In Germany and Europe, it is very hard to get a work permit. In Canada, there is a clause in every job ad that indicates that Canadian nationals are given preference. US citizens should be given preference in the US. In India, there are restrictions on foreigners working, but they want the US market to be wide open. We need US jobs to have preferences for US citizens.

    2. Dear POed Lib,

      You are trying to turn this into a segregating nationalist discussion of fairness to US citizens. Not only do I entirely reject your sentiment, but the problem I am highlighting is that facing the entire US and world biomedical research community. Whether the academics in the pipeline were born in Russia or New York is irrelevant. The current system is equally bad towards all of them. [The third comment you submitted is extremely xenophobic and I will not be approving it.]

  6. Dear POed Lib,

    As someone who has a PhD from the USA and as someone who has friends who obtained PhDs in countries like France, Germany, Australia and Canada, I can vouch for the fact while they have all been given permanent residency of their adopted countries, for me in the USA it will take close to 10 years to get a green card. So saying that the US is allowing foreign PhDs to get green cards easily is completely false. The second aspect of your opinion that foreign PhDs are taking away american jobs is also completely false. The american immigration system is probably the only immigration system in the world that is the most unfavorable to highly skilled immigrants. As some one who has been trying to transition from academia to industry for the past year, I cant tell you the number of times my national origin has cost me a job or an interview.
    So I would like to inform you that foreign PhDs and american PhDs do not have a level playing field and hence your opinion that foreign PhDs are taking away american jobs is completely unfounded and just shows that you have no idea about the ground reality.

  7. I agree, Lenny, that the hyper-competitive environment (and the uncertainty it breeds) is reducing academia's attractiveness to talented young scientists. I also worry that these same forces exclude those who do not have the luxury of taking severe career risks - those from disadvantaged economic backgrounds, and perhaps also women, whose reproductive years coincide with a scientific career's most severe bottlenecks. Science will ultimately be strengthened by drawing from a diverse talent pool.

  8. Hi Lenny,
    Haha, "QuitLit", I like it! I agree that the system is hypercompetitive these days, far more so than in the past, and I think that's only going to get more severe. I disagree, however, that academia is no longer competitive for the best and most talented researchers. I just have not seen this to be the case in my experience sitting on both sides of the PI divide. This is of course not to say that the best and brightest all want to be academics–most definitely not the case. And also there are certainly many who want to be academics and might have the skills, but just don't get there because of bad luck or other reasons, which is sad. But from what I've seen, those who somehow do make it into academia tend to be very, very good. The vast majority of them are very dedicated, talented, ethical, and various other positive adjectives that we all value. And, I might add, a rather happy bunch in general, at least from the ones I've met. Which is probably why relatively few people leave such jobs once they get them. Incidentally, one consequence of hypercompetition is that the folks getting junior faculty positions are getting better and better everywhere. Which seems like it will be beneficial to universities up and down the totem pole. No doubt the competition is unpleasant, but hey, that's life everywhere these days.

    1. Arjun, I agree with you that increased competition has many upsides and I personally think that in overall it might be a good thing IF -- and only IF -- the criteria of evaluation are good. I think that these criteria have much, much room for improvment. I fully acknowledge that my evaluation of the evaluation is subjective, potentially influenced by biases, and if I have to elaborate on it I will go to specific examples rather than come up with some overall bulletproof framework. Yet, I am quite sure that you will agree with me on at least some of these examples !

    2. I will comment more fully in a day. In the meantime, this post from Jessica Polka is exactly on this topic. http://www.ascb.org/ascbpost/index.php/compass-points/item/347-how-competitive-should-science-be

    3. Here is a piece on NPR about scientists who lose their funding.

  9. The actual availability of work opportunities pertaining to a person seeking becoming a forensic professional scientist job used to be very small for an extended time until finally regarding all 5 in the past, as soon as a lot of technical and also controlled advancements did start to create and also presented brand-new types of resources in which significantly increased your productivity with the law enforcement officials and also safety forces inside dealing with violations along with issues.