OPINION: Research fellowship schemes have been the focus of a lot of attention recently. So let’s ask the question: should Australia invest in more fellowships or fewer fellowships, and should they be for junior or for senior researchers?

First, there have been loud calls for more junior fellowships. As student numbers have increased, the competition for academic jobs has become intense.

No one likes to see bright and highly trained students forced to leave careers in research or to leave the country. We want to support young researchers and so both major research granting bodies — the Australian Research Council and National Health and Medical Research Council — developed their own early career schemes. But I would caution against increasing the number of these fellowships. It would only be a short-term fix.

I call it the cane toad strategy. Cane toads lay thousands of eggs in good times and bad, and most tadpoles are doomed to starve. Cane toads care about launching a lot of eggs but that’s where their interest stops. It is a cruel system.

It has been said that large investments in junior fellowships have reduced the amount of funding going to project grants and to research infrastructure. Just look at the success rates of the grant schemes and the repeated cuts to research overheads. We desperately need to establish a stable academic environment before creating more junior fellowships.

Another problem with junior fellowships is the chances of making a mistake when talent-spotting young researchers are higher. At worst, it is the PhD supervisor who is being judged rather than the junior fellow.

So how should we support the next generation of researchers?

The old-fashioned answer is to take the job away from broad fellowship committees and let our star research group leaders pick top students in their disciplines and mentor them.

But mentoring models rely on the quality and benevolence of top professors. The most important advantage of this approach is that large teams can support junior fellows if their first research ideas don’t work out. One of the biggest problems young researchers face is simple bad luck. Being independent too early may only give you one roll of the scientific question dice. We want people to take risks, but sometimes this is easiest if you don’t leave the nest too early.

So, should we have any junior fellowships at all?

I like travel fellowships such as those modelled on the NHMRC’s CJ Martin fellowships that allow local researchers to work overseas for a few years before returning.

We also should reopen all our fellowship schemes to attract talent from abroad. There is no evidence we will be swamped but we will recruit some good people.

What about mid-career and senior fellowships? We should definitely maintain them — these support the best researchers at a stage when everyone can agree who the top researchers are.

Having dedicated fellowship funds allow us to invest in areas where there are relatively few academic jobs, such as in agriculture.

The time has come not for more one-off schemes or drastic changes but for sustained and strategic commitment to the good processes we already have. We need an end to start-stop funding and we need to recognise that investing in excellence is a good strategy that will build strength and help inspire the next generation of researchers.

Fellowships that are allocated by independent, peer-review systems work against nepotism, pork-barrelling and other types of vested interest. One hopes the Future fellowship, Laureate fellowship and NHMRC fellowship systems will continue to thrive.

Merlin Crossley is Dean of Science at UNSW.

This opinion piece was first published in The Australian.