Ponzidemia and the academic arms race… Some musings from a burn-out victim

A perfect storm of conspiring events is closing in on science and it is likely to have some serious flow-on effects across the coming decades. To my mind, the perfect storm runs a little like this.

For many decades, societies have been seeking to advance the education of their populace — both to provide highly skilled labour to support high-tech industries, e.g., advanced manufacturing, biotechnology, electronics industries, etc., and to ‘enrich the intellectual value’ of society more broadly. Decades ago most students left high school in year 10 to pursue a trade, with ever smaller fractions going on to senior high school, undergraduate and postgraduate training. Now, vast cohorts go on to complete undergraduate studies, often being trained to levels beyond what they need for their future employment. This is a great thing, many studies have shown the benefit of educated societies, but it has unanticipated consequences…

One flow on effect is a glut of Ph.D. graduates, most of whom came through their 8-10 years of training with the sole ambition of becoming a permanently employed professional scientists. Science is not an endeavour that fits well with traditional market models — the development of knowledge is essential, but in most cases, knowledge is not a commodity that is easily monetized. The connections between a distinct discovery and a marketable product can be long, convoluted and slow to realise. As a result, society has traditionally supported science via taxpayer investment from Government — in a sense this is the ultimate in crowd-sourcing for projects that are investments in a better future for all. But the available funds are necessarily limited, and are dwindling in many modern democracies, where the realities of the ballot box have made politicians more likely to give short-sighted tax cuts and middle-class welfare to buy votes rather than long-sighted investments for the greater good.

The result is a funding pie that is not only shrinking in real terms, but that is being sliced ever more finely as it is spread across a growing cohort of scientists seeking funds. This is where the tangled web of ponzidemia, as I like to call it, has its roots.

One of the most vital resources in academia is people-power; a lab doesn’t run itself, it needs people to do experiments, collect and analyse data, fix equipment, prepare research for dissemination to the public, etc. People cost money, and this is particularly so when they have a Ph.D. under their belts, i.e., they are what we know of as ‘post-docs’. With dwindling funds though, you can’t afford post-docs any more, their salary is 3-5 times what a Ph.D. student gets paid. This provides a big incentive to a) not take postdocs any more, and b) just take on lots of Ph.D. students — it makes economic sense right? Why have 1 person who just finished a Ph.D. when you can have 3 or 5 people who are working to get one for the same price?

The universities love this also — Ph.D. students have to pay tuition and that’s an income stream, postdocs pay you nothing. So in the end, it’s a win-win situation all round, unless you’re a Ph.D. student of course, because then your life is like this… First, you get suckered into doing a Ph.D., often while you’re still too young and naive to realise that your chances of getting a tenured position nowadays are almost zero, there just aren’t enough professor positions available, and your chances of getting a post-doc even are becoming slim. Second, you spend those three years on a scholarship that, compared to any real job in science, is basically a slave wage. In many cities, these scholarships are not far above the poverty line (in real terms) and they are either supplemented by the generosity of their Ph.D. supervisor through a ‘top-up’ scholarship, money earned through casual teaching, a third job, and sometimes all of the above. Third, and in the midst of this, to maintain the ‘revolving door’ of Ph.D. intake/graduation that the universities require to satisfy government bureaucrats, you are often now required to produce the same or much more than a Ph.D. graduate did decades ago but in 3-3.5 years rather than the 4 or 5 years it used to take. Ph.D. students often work extraordinary hours (70-80 hours a week is not unusual sometimes) and take many of the crappiest jobs in the system. Their high teaching and outside work loads often mean that they are pushed out the door with a bare scraping or sub-standard Ph.D., something that gives them no chance of ever remaining in academia. The universities move them on as soon as practicable nowadays so someone else can take their place and governments can boast about how many Ph.D. trained graduates they’re producing despite their lesser quality.

An additional factor in this perfect storm is a strange anomaly in the way universities are perceived. The public mostly sees them as educational institutions, a place where we train doctors and lawyers, engineers and scientists. One would naturally think then that the best universities are where you get the best education, and this is how they are ranked, but this is not how it works at all. When it comes to rankings, the best universities are the ones with the strongest reputation for research — the most Nobel laureates and famous professors, the most papers in high profile journals like Nature & Science, the most grant income, the most exciting discoveries. In the university system, quality of education is a distant second, and in the minds of many, dead last. Indeed, often the best educations are obtained at the not so highly ranked universities. The reason is that at the best universities, the professors are often so busy engaged in the academic arms race associated with research that their teaching is mediocre at best, and often blatantly and shameless neglected (actually, many profs boast about this as though it were some research badge of honour).

To corner the student market, the universities need to be top ranked, and this means they must achieve more in research. They need this to get undergraduates to come, to provide the main revenue stream, and they need this to get postgraduates to come, to make up their slave research labour force. The universities use some fraction of their fees to supplement research incomes because of the crucial importance of research to their rankings, and as a result, their bottom lines. As they accumulate more researchers (more researchers mean more output means higher pushes in the rankings), they are furthering the slicing of the government research funds allocated. This has got to the point where, in many economies, there just isn’t enough money to support this arms race any more unless the supplements to research income using student fees can be maintained. What’s the outcome here? A push by universities in many countries to deregulate fee structures so they can charge more from students; it’s the only way to keep competing in the ranking arms race.

Another flow on effect is ever greater levels of productivity required for individual researchers to get government income to support their research — more people in the system with a shrinking pie means more competition. The key factor in this competition for research funds is number of papers in the top or ‘high-impact’ scientific journals. To the lay person this will seem fair enough — funding should go to the best science — but a second arms race combines with the first here to produce a scientific system that is becoming its own massive productivity killer.

For journals as it is for universities, reputation is king. In a perfect example of ‘what we measure is what we come to define as success’, the journals compete on the basis of a metric called impact factor.  A journal’s impact factor is the average number of citations per paper published for the two preceding years. For a journal, the key to keeping the impact factor high is not to publish a lot or have papers that get cited a lot, it’s to make sure that you focus on publishing papers that get cited a lot in the next two years. This means a) focus on sexy topics that are buzzing with work right now and b) focus on review papers as they get cited a lot on short timescales. Rankings matter to journals because it sets what they can charge in subscriptions to university libraries, and more scarily, it sets what they can charge as the article processing charge (APC) for open access papers (no, APCs are *not* set by the real costs of open access at all, they’re set by what the journal can get away with based on its prestige and impact factor, anyone who thinks otherwise is a fool). This focus on impact factor by the journals has some interesting implications.

The most obvious is that it is severely compromising and crippling an important part of science — the peer review process. The original intention of peer review was to have one or two other scientists not associated with the work read and consider it carefully. The goals were to a) check the research has been conducted within the guidelines of the scientific method and is not misleading or unscientific, b) to check that the work is clear to a typical reader, and c) provide an opportunity for constructive criticism to improve the work or check details. Funnily enough, when you submit to journals that aren’t top-level ‘high-impact’ journals this is often exactly what you get, useful constructive comments on how to make the paper better ahead of publication.

But there’s an impact factor level you reach where the character of these reviews changes quite sharply. The focus shifts instead to subjective judgements about whether the work is ‘topical’ or ‘important’ (i.e., sexy and likely to get lots of citations quickly), or whether it is ‘broad impact’ (i.e., likely to be cited by a wide demographic of researchers). The peer review reports are, 90% of the time, rejections that basically amount to why the work doesn’t deserve to be in such a high impact journal whilst barely considering the technical aspects of the science and in some cases, not commenting on it at all! It’s not that these papers are junk either; they ultimately get published in lesser journals as perfectly good science, sometimes with no revisions at all — they aren’t rejected for quality, they are rejected for prestige reasons in a contest for primacy. In fact, the worst rejections by far are the ones where the referee says the science is fine, it just doesn’t deserve that journal, as that’s when you realise how fickle and corrupted the peer review system has become.

A lay person might go, ‘well, that’s your own stupid fault, don’t aim so high’, but how can you not? Competition being such as it is, you have no choice but to bounce off these journals in the hopes of getting in, because getting one of these papers can make the difference between you getting your next research grant or getting nothing and having to fire post-docs or tell Ph.D. students they can’t finish their project any more (and I have heard of Ph.D. students who have had to abandon their degree because their supervisor can no longer finance them — they really are a labour commodity).

The result is that currently most professors have papers that bounce through several journals (I start at the top and work my way down every time – I have no choice), wasting the time of several referees at each attempt and many months of mucking around, before they find a place to finally be published. The wasted productivity here is substantial — many hundreds of thousands of person-hours a year are wasted on this cycle of submissions and frivolous rejections I’d say.

The next flow on effect is the difficulty of actually getting peer reviewers for a paper, and if you do get them, in getting high quality reviews. As a journal editor, getting reviewers is a curse and the sad reality is that a fraction of the scientific community carries the lions share of the load here. Many researchers review much more than their fair share of papers out of a sense of duty towards science, and many refuse to review at all unless they can push their personal agenda through the process. Sadly, the people who review many punish themselves in the process, because the ones who shirk their responsibilities have more time to use to get themselves ahead in publishing. The cruelty here is that whether you get grant funding or not doesn’t care at all about whether you are a good citizen in terms of your responsibilities to peer review. If anything, it rewards you for being a bad citizen, because all that really counts is high-impact publications.

Then there’s the down right ugly. Misconduct scandals, fraudulent results, cases where editors have ignored referees and published papers anyway because the citation counts are too good to miss (e.g., the Hendrik Schon incident), cases where unscrupulous referees deliberately kill papers to protect their own interests. I won’t say more about this, as I don’t want to be perceived as hanging sour grapes in this blog post; others have said more than enough about this, and many of us have been affected by it (yes, I am one of many sad innocent victims of the Schon scandal — perhaps for another blog post).

So, having dealt with hoards of keen, excited and talented young Ph.D. students whose scientific careers will likely end within years of getting their Ph.Ds, I want to touch on another emerging and disturbing effect of ponzidemia and the academic arms race — the detrimental mental health effects on early-mid career researchers in the system. If you do manage to get a post-doc, the fight to survive long enough to get a tenured position now becomes instantly and crushingly intense. Right at the time when people start thinking about a family or setting themselves up for life, they get smashed with the most extraordinary output expectations. Young researchers need to be highly productive in a topical and sexy area (otherwise you can’t get the high-impact papers that get you grants), they’re often forced to become independent before they are really ready to operate as such, they’re often teaching (for free) as they need that experience to get tenure, often lumped with responsibilities by more senior professors (who want the time to stay competitive themselves).

Pressure can be a good thing in small doses, it makes you pointed and productive, but there comes a point where it becomes destructive. It can be as minor as flagging confidence and falling productivity, but I have seen much worse also, I know many in academia who rapidly develop mental health issues (depression, violent mood swings, etc) from job stress and stupidly long hours (80+ /wk), I’ve seen some spend periods as functional addicts, and there are some who I worry won’t come in any more, and not because they found something else to do, but because they took their own lives (yes, I have lost colleagues before). These problems are descending down into the student cohort also, many of whom are obsessed with their next paper and no longer even caring about the science, or even whether things are done properly. Misconduct at student level is becoming common (and often swept under the carpet), mostly due to the pressure exerted on them to produce by a professor who is under pressure to produce — a conga line of subtle bullying that ultimately has its origin in the ranking contests of the universities themselves.

We really need to do something about this problem, if not for the human toll for the scientific toll. So many young researchers are no longer willing or able to be ambitious and creative any more. Why would you when the project that fails could mean you become uncompetitive and can’t get your next grant or next job? How can you, when the pressure is so intense that you haven’t the time or space to think about anything but getting your next paper across the line.

The solutions are probably best kept for a separate post, but at the core, I see several reforms needed in the coming years:

1. Increased funding from government, in a serious technological society, investment should be at >2.0%, currently Australia is at 1.7% and falling (numbers from Wikipedia, but if it’s good enough for Greg Hunt…).

2. A carefully thought out plan for strategic investment of science funds with a long-term perspective that’s decided outside the democratic political cycle.

3. A breaking of the reliance on Ph.D. students as a cheap labour force for research. The advanced economies should not be treating people as 3rd world citizens.

4. A division of the Ph.D. into two degrees, one with restricted intake and careful selection aimed at refilling academic demands, and one aimed at producing postgraduate trained technologists with a better focus on transferrable skills (much of what’s learned in a Ph.D. is NOT transferrable).

5. Much more efforts on behalf of universities to inform undergrad. science students of career opportunities outside academia.

6. A focus on quality for universities that goes beyond trivial metricisation and league-table games based on research output.

7. A focus on quality for academics that is less one-dimensional and focused on a wide-range of skills that include research, but also teaching, public engagement (essential for convincing the public that they need to give us 2.0%+ of GDP to spend on research in the first place), involvement in public policy, etc. The latter can be on sabbatical secondments that are treated as an equal part of the job description.

8. A focus on ambition over productivity in research — as soon as failure of projects becomes too big a risk you stifle innovation and creativity. Many will admit that you often learn more from a project that fails than one that easily succeeds (I certainly have learnt this). Science needs to be less risk-averse and failure-tolerant in many cases.

9. A better focus on providing time for creative thought and work environments where the competition is friendly not aggressive. Cut-throat competition destroys creativity, destroys collaboration and together these destroy innovation.

10. Sensible structures for workload management in academia. Almost every academic I know (including myself) is a chronic workaholic, this is not at all healthy, for us or science or society.

I think I’ll finish here. My goal is not to solve this problem, one person alone never can, my goal is simply to draw more attention to it with the hopes that eventually, enough people care enough to act together to fix it.


3 thoughts on “Ponzidemia and the academic arms race… Some musings from a burn-out victim

  1. Pingback: The knowledge economy needs investment in brainfrastructure | Dr Mel Thomson

  2. Such a great post. Thank you. One amazing side-effect of this system is the groupthink it produces in very smart people with otherwise highly developed critical thinking abilities. I suppose it is called denial. Any one of your ideas at the end of the post would help to reform a system that is near breaking point, and may even help to highlight other areas that need attention – an obvious one is public engagement, so that people who vote or make decisions could learn about how research works, and doesn’t work currently.

    Meanwhile, to many within the game, if you seriously suggest that things are broken, they ask “well what kind of system would you have?” as though we must suffer with the most completely outdated and broken system until we find one that is perfect in every way. I think the meta-problem might be the obsession with corporatism written about by John Ralston Saul. To my mind, that is what is driving universities toward their ponzi-style structuring of research activities. This obsession with productivity that can be measured doesn’t work in the business world and it sure as hell doesn’t work in research. As a start, more posts like yours should help to at least make more people aware of what is going on.

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 )

Google+ photo

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

Connecting to %s