Three years for PhDs ‘not long enough’

Extended PhDs would be money well spent in the quest for international competitiveness, says leading physicist

July 11, 2018
Man cramped in a box, cropped

A leading Australian physicist believes that doctoral programmes should be extended, saying that the current three-year norm leaves students underdone.

David Reilly says that 36-month PhDs dump students into the workforce before they are ready. A year or two more would allow them to look past the “low-hanging fruit” and attempt challenging and potentially time-consuming research projects, he argued.

“It changes the whole mindset,” said Professor Reilly, founding director of the Sydney Microsoft Quantum Laboratory at the University of Sydney. “It means you’re going to work on the hard things.

“It changes that whole discussion around what we’re doing broadly in academia, in science, and how we’re ensuring Australia is competitive.”

Professor Reilly heads one of three facilities that Microsoft has established outside the US to help it develop a quantum computer. The centre has attracted hundreds of millions of dollars from the information technology giant – Australia’s largest investment in quantum engineering – and he was named best emerging university leader at last year’s Australian Financial Review higher education awards.

In Australia, as in the UK, funding for PhD students is generally for three years – with up to six months’ additional funding if they can demonstrate exceptional circumstances.

Many other countries keep PhDs short, with three or four years the norm in Canada, New Zealand and much of continental Europe. While some countries allow a little longer, some terminate students’ registration after five or six years.

In the US, by contrast, doctoral studies typically take between five and seven years. Professor Reilly argued that shorter periods often leave doctoral graduates without the disciplinary understanding and publication records that they need to progress.

He said: “Imagine recruiting for an industry or postdoctoral position, and you’ve got somebody with five to seven years’ experience and somebody with three years’ experience. Which way are you going to go?

“An adviser told me he would never hire UK or Australian graduates fresh out of their PhDs. He’d wait for them to do a postdoc for a couple of years, and then hire them.”

Professor Reilly said that his own postdoctoral position at Harvard University “opened a lot of doors. But getting to Harvard was only possible because a postdoc in between brought me up to the competitive level of the American PhD”.  

Since joining the University of Sydney, he has raised funds to help his doctoral students extend their studies by up to 30 months. While he has received “a lot of flak” for this, because university funding settings favour prompt PhD completion, he said that the graduate outcomes speak for themselves.

Of six doctoral students he has supervised to completion, two recent graduates are eyeing jobs with Microsoft and Massachusetts General Hospital, one already works with Microsoft, one has a postdoc at the University of California, Berkeley and another is at Caltech.

The sixth, Xanthe Croot, has claimed one of the three Dicke fellowships that Princeton University awarded this year to exceptional recent PhD recipients in experimental physics. Dr Croot, who took more than five years to complete her PhD, said that the extra time allowed her to “draw a line under experiments at a satisfying and logical pace” rather than rushing or abandoning them.

“I hit my stride in the last couple of years in terms of working and learning optimally,” she added. “It took a few years to build a really solid foundation in the field.”

Dr Croot said at least one of the papers that came out of her PhD would not have been possible without the additional years of study. Her publication record at graduation would have been “down a paper or two” but for the extra time.

Professor Reilly put it in rather starker terms when he said: “That extra year can be the difference between driving a cab and having a competitive fellowship on a world stage.”

PhD duration is a long-standing debate in Australia. A 2008 inquiry by a parliamentary science committee recommended six months’ additional funding, finding that PhDs typically took more than five years.

Professor Reilly said that bankrolling the additional study should be a joint responsibility, with the government providing matched funding for every dollar raised by supervisors. This would give the government “the best bang for their buck” in stemming brain drain and improving technical expertise, he added.

He argued that learning, like fitness training, is rarely linear. “You’ve got to build muscle to build muscle.

“Most of the action comes in that final six months. You need to learn enough to be in a position to bring it all together.”


Print headline: Three-year PhDs ‘too short’ to ‘build research muscle’

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.

Related articles

Related universities

Reader's comments (3)

There seems to be an assumption that all PhD students will progress into academic posts. In my experience (and that of an Institute of Physics {IoP} survey a few years ago) this is not the case. Half of the students according to the IoP do not enter academia and a fraction of a percent reach the highest academic grades. Most of my former students work in commercial computer security and finance posts. The norm of between three and four years should suffice. It is hard enough to get any UK students to take PhDs now so lengthening the period would only make matters worse.
I think the issue here is the maximum period of study allowed. There is no good reason to have a one size fits all approach across all disciplines. The maximum period of study, as long as satisfactory progress is being made towards completion of the PhD, could certainly be more than 3 years for many disciplines.
Is this extended PhD going to cost the same overall, or will the fees increase? There already is a maximumn period of study. If need be, those 3 years can be extended and in many cases they are, often to complete the final writing stage. Being a PhD candidate/student is already a blurry enough role at universities. Some consider them staff, some consider them students. In some disciplines they are used as cheap labour. There can be some flexibility in the length of PhDs, but funding should be clarified and supervision should also be monitored a bit better to avoid situations where the PhD ends up being the typical doormat of some full time staff... for a now formally extended period of time.