Fowl or Fish? A submission to the ACIAR Review

The Australian Centre for International Agricultural Research (ACIAR) is a very important organisation. It is the only aid agency in Australia which has legislative backing, something that the much bigger AusAID lacks. More importantly, although its funding consists of only a small part of the Australian aid program, it is one of the largest funders of agricultural research for development in the world. In 2012-13, ACIAR will manage around $US120 million, or (we estimate) about 13% of global ODA in support of agricultural research for development, with funding more than 50% greater than that of the largest member agencies of the 15-member Consultative Group on International Agricultural Research (CGIAR). Compare this to the 3-4% that constitutes Australia’s share of global ODA for all purposes, and the importance of ACIAR becomes apparent.

It’s good that Australia gives such prominence to agricultural research. Aid will increasingly be required to fund international public goods such as this: the case for giving similar prominence to medical research has been repeatedly rehearsed on this blog. And ACIAR reports very high rates of return on its projects, at around 50:1 (including benefits to Australia, which constitute around 10% of total benefits). But this doesn’t mean that there is no room for improvement. One problem is that ACIAR as a whole has been subject to little external scrutiny in recent years, with the last major review [pdf] held in 1998.

For this reason, Foreign Minister Carr’s August 2012 announcement that ACIAR would be subject to an external review was a welcome one. By now, the Farmer Review should have reported to Foreign Minister Carr (the deadline was the end of last year).

The Review called for submissions, and we made one. Surprisingly, submissions aren’t posted on the review’s website, but here is a link to ours.

Our submission covered seven topics:

  • ACIAR’s objective: discussed below.
  • ACIAR’s role in funding and engaging with the CGIAR system: is there a conflict of interest?
  • ACIAR’s role in implementing development projects: the vexed issue of how to get ACIAR and AusAID to work better together.
  • ACIAR’s approach to sourcing expertise: where we argued for more competitive mechanisms.
  • The geographical distribution of ACIAR’s activities: the Africa question.
  • Impact evaluations of ACIAR’s research programs: can we believe the reported 50:1 rate of return?
  • The Australian International Food Security Centre (ACIAR’s new centre within a centre): does it make sense?
  • The 2011-12 Strategic Framework Panel: the oddness of ACIAR’s commissioning an external panel to develop a scaling-up strategy (the early-2012 Chubb Panel) not long before being subjected to a full external review (the late-2012 Farmer Review).

In this post we delve into the first of these topics, ACIAR’s objective, or lack of one. If you’re interested in the other topics, please take a look at our submission.

ACIAR has never had an explicit statement of objective defined for it in government policy. It needs one. ACIAR’s own account of its objective, as currently articulated on its website, is that: “[ACIAR] encourages Australia’s agricultural scientists to use their skills for the benefit of developing countries and Australia. ACIAR funds research projects that are developed within a framework reflecting the priorities of Australia’s aid program and national research strengths, together with the agricultural research and development priorities of partner countries.”

There are three problems with this informal statement of objective.

First, it incorporates a mutual benefits rationale and restricts funding to Australian sources of expertise. Neither of these elements corresponds to anything in ACIAR’s legislated statement of functions, which contains no mention of benefits to Australia, direct or indirect, or exclusive reliance on Australian expertise.

Second, it conveys no sense of complementarity: it establishes no specific niche for ACIAR that would assist in determining whether a particular activity or strategy were likely to meet a need that would or would not be met more effectively or efficiently by another agency, including any of the 15 CGIAR agencies. Elements of ACIAR’s current research programs fall into areas covered by and engage researchers who receive funding from CGIAR agencies. For example, forestry research is undertaken by both ACIAR and the Bogor-based Centre for International Forestry Research, which frequently engages Australian researchers who also receive funding from ACIAR.

Third, the objective ignores significant functions defined in ACIAR’s Act: “to conduct and fund development activities related to [its] research programs” and “to fund international agricultural research centres.” The first of these two functions was added by legislative amendment in 1992, following the “sunset” review of ACIAR conducted by the Joint Standing Committee on Foreign Affairs, Defence and Trade.

The absence of a formal statement of ACIAR’s overarching objective is problematic. It has allowed a situation to develop where ACIAR’s activities, uniquely among activities supported by the Australian aid program, are routinely justified on the basis of mutual benefits considerations. It has contributed to a lack of clarity about the kind of organisation that ACIAR is intended to be: is it a domestic research funding council whose field is international agricultural development, or a species of development organisation capable of operating along the research and development spectrum and leading Australia’s engagement with the CGIAR system? (Is it, in the terms of our title, fowl or fish?) And it has also obscured the question of whether some elements of ACIAR’s work might not be carried out more effectively and efficiently by the CGIAR agencies that Australia also funds via ACIAR and AusAID.

It also seems to be the case that the narrative used to describe ACIAR’s aims and activities represents a frozen view of the role of the aid program. Since ACIAR’s creation, much of the aid program has been untied. The commercial objective of the aid program was explicitly removed by Minister Downer in the late 1990s, and over time the national interest benefits of the aid program have been downplayed. Given that ACIAR’s current practice of tying aid to Australian researchers and justifying results in terms of mutual benefits are not required by legislation, the question must be asked whether ACIAR should not be brought into line with the rest of the aid program.

Indeed, it can be convincingly argued that ACIAR has not utilised the flexibility provided by its legislation. The best way of identifying and solving the agricultural problems of developing countries (the essence of what ACIAR’s legislation requires of it) might change over time. While the aid program has progressed, ACIAR has been left behind (this flexibility notwithstanding).

While it seems clear to us that the commercial benefits to Australia should receive less or no weight, there might be arguments for continuing to tie our aid to Australian researchers. After all much, though not all, research and scholarship support from the Australian aid program is still tied to Australian institutions. However, there might equally be a case for looking outward and allowing or encouraging ACIAR to seek out the best researchers wherever they are, rather than the best Australian researchers. For example, ACIAR might open a competitive funding window which would be available to researchers worldwide. Or it might create incentives for Australian researchers to partner with other developed and developing country researchers.

The Farmer Review should recommend an overarching goal for ACIAR’s work, to be adopted as government policy after whole-of-government consideration, which would provide full consistency with the existing goal of Australia’s aid program. The focus should be on development benefits, not mutual benefits. The approach should be to address those needs identified as important to Australia’s aid objectives that are not adequately met by other mechanisms. These needs might well be defined in such a way that Australian sources of expertise are highly relevant, but consideration should be given to moving in the direction of untying ACIAR funding so that it can do more to draw on the world’s best researchers, wherever they are.

Our vision for ACIAR is as an international fish rather than a domestic fowl. We hope that the Farmer Review agrees.

Robin Davies is Associate Director of the Development Policy CentreStephen Howes is Director of the Centre. 

image_pdfDownload PDF

Robin Davies

Robin Davies is an Honorary Professor at the ANU's Crawford School of Public Policy and an editor of the Devpolicy Blog. He headed the Indo-Pacific Centre for Health Security and later the Global Health Division at Australia's Department of Foreign Affairs and Trade (DFAT) from 2017 until early 2023 and worked in senior roles at AusAID until 2012, with postings in Paris and Jakarta. From 2013 to 2017, he was the Associate Director of the Development Policy Centre.

Stephen Howes

Stephen Howes is Director of the Development Policy Centre and Professor of Economics at the Crawford School of Public Policy at The Australian National University.


  • Ray and Bob, Thanks for your comments. I’ve been away in PNG, hence the delayed response.

    I can see now Ray that our remarks might be read as implying that ACIAR relies only on Australian researchers. I fully accept, and have long understood, that it follows a collaborative model linking Australian and developing country researchers. I agree we should have been clearer on this, so thanks for bringing that out.

    My question remains though: why should always be a requirement of involvement from some Australian institute? Perhaps the best project might involve bringing together an American and an Indonesian institute. To mandate Australian involvement smacks of protectionism. Lots of countries use protectionist arguments, especially post-GFC. That doesn’t mean we should.

    I’m glad that Bob agrees with me on this, but I also accept that it is an open question how much difference it would make. My question for Bob is this: If ACIAR has been spasmodic in promoting adoption, how does it get such high rates of return?

  • Maybe it’s a platypus.

    Stephen Howes and Robin Davies argue that ACIAR is neither fish nor fowl, and it should align itself much more with accepted good aid practice. But I wonder if the starting point ought to be whether ACIAR’s assistance is doing any good, seeing which aspects of its operations drive outcomes, and then ask if its approach could lead to improvements. And maybe we would end up concluding it’s a platypus well adapted to the rather unique environment it has to live in. (Of course, we might not, but judge first on outcomes rather than process.)

    ACIAR has a pretty strong commitment to rigorous quantitative evaluations, with methodologies develop and reviewed by the world’s best authorities.There are issues with these evaluations, of course – while they are done after projects have been completed, they are usually undertaken before all actual outcomes have been realised, – so amongst other things evaluators have to make judgments about adoption – which is the critical variable. And in the past, it wasn’t always clear that evaluations were consistently used to learn lessons: rather more to demonstrate that there were enough high return projects to justify the overall investment in ACIAR. But that is what the evaluations do show: and we don’t have similar numbers for many other parts of the aid program.

    Howes and Davies are right to target the problem with requiring demonstration of benefits to Australia. But most agricultural research activities are publicly funded, often with contributions from levies on Australian producers and, at least in the past, involving researchers employed in state government agencies. One can imagine that the adoption of the ‘Australian benefit’ doctrine was designed to help navigate the political challenge of using knowledge and skills developed to support Australian agriculture to help potential competitors. The time is probably ripe to jettison this requirement: but it might also be worth asking if it actually made much difference to what ACIAR did and how it did it.

    The biggest problem facing ACIAR, and Howes and Davies identify it, is that adoption is the critical variable, and ACIAR has been a bit spasmodic in its attempts to understand the contextual drivers of uptake of agricultural technology. It had a go with its Policy Linkages and Impact Assessment Program, but that was not sufficiently embedded in its project development systems. One might have hoped that collaboration with AusAID or other development agencies might have been integrated into projects to help fill this need: but that doesn’t seem to have occurred. To my mind, if ACIAR is to continue to operate, this is the issue it most needs to come to grips with.

  • Thanks for your comment, Ray, but I must say that I couldn’t find the errors you identified, and nor do I think we wrote from a narrow (let alone an AusAID) perspective.

    Unfortunately, you make an error by saying we suggest “the mutual benefits rationale” is important “over” development. What we say is that ACIAR adopts a mutual benefits rationale (that projects should benefit both the developing country and Australia). It does this so often and so pervasively that I thought it was in its legislation. It isn’t. ACIAR is the only part of the aid program required to justify its spending in part in terms of benefits of commercial benefits to Australia. That’s unfortunate. Downer dropped the commercial objective back in 1996. ACIAR should catch up with the rest of the aid program. This is not to say that some projects won’t have commercial benefits. But it shouldn’t be the objective.

    Second, I take it you think ACIAR should continue to require the involvement of Australian researchers in all its programs. Again, the rest of the aid program has made, not complete, but considerable progress in terms of untying over the last 15 years. As we say, there are good arguments for continuing to link to Australian researchers. And there are certainly good arguments for encouraging collaborative approaches, as you show. But there are also good arguments for, as we put it, “encouraging ACIAR to seek out the best researchers wherever they are, rather than the best Australian researchers. For example, ACIAR might open a competitive funding window which would be available to researchers worldwide.” I’m surprised you implicitly dismiss this argument without considering it directly. Think of AusAID’s ADRA: the best Australians compete against the best from other countries for the Australian development research dollar.

    As you have pointed out, there are lots of good Australian agricultural economists, and there are no doubt many good Australian agricultural scientists as well. They should do well in a competitive process.

    In summary, requiring ACIAR projects to provide commercial benefits and tying all projects to the involvement of Australian researchers are unfortunate, and one might say narrowing, or even protectionist, features of ACIAR’s work. That was the argument of our submission, and I don’t see anything in your response which convinces me that we were wrong.

    None of this is to argue that ACIAR does not do good work, or that AusAID is somehow better. I have more than once argued for more funding for ACIAR. But the aim of the Farmer Review is to look at what can be improved, and that was the focus of our submission and post.

    Thanks again for your response. I’d be interested to hear what others think.


    • Hi Stephen,
      Thought your response was worth a quick reply though a chat over a coffee might be more effective.
      The most evident error in your blog, which I thought was clear even from my rushed response, is that Australians need not be involved in all ACIAR funded projects. And I am not sure how a competitive process would ensure the involvement of the limited best developing country researchers as ACIAR projects have.
      On the benefits question, which I said flow automatically from the collaborative model, recently taken up in AusAID’s recent AIP Program, this is appearing in the UK aid debate at the moment. The position to not cut aid whilst domestic programs are being cut is being defended on the grounds that such aid delivers the socio-economic benefits to the UK that ACIAR requires to be estimated ex ante in its projects for all countries and formally measured by independent assessors ex post.
      Best regards

  • An interesting post but written I feel from a narrow AusAID perspective and with errors.

    I was a bit confused by the suggested importance of the mutual benefits rationale over development ones in ACIAR dealings that was mentioned in the summary as I have never found this the case. Your stated estimate that only 10% of the benefits are Australian (which presumably would not change with a revision of the 50:1 rate of return which new research estimates at around a respectable 30:1) would suggest developing countries benefits dominate.

    It is hard to imagine how a standard ACIAR project would not deliver some mutual benefits given the underlying characteristics, especially the collaborative model which, unlike some other agencies, ensures the project is not made up completely of Australian consultants flying into a developing country and out with a report reflecting little collaboration with local researchers. The collaborative model of ensures the direct involvement of developing country research institutions and that all the researchers are not Australian. Often the Commissioned agency is one of the expensive CGIAR centres mentioned with any Australian involvement a minor component. The collaborating developing country institutions are free to involve non-Australian expertise from their funds and many have, along with other research agencies, NGOs, and private companies. Some of the developing country researchers involved have been amongst the world’s best and some have end up as government Ministers, or Chief Economists or candidates for heading up international agencies. The Australian collaborating institutions often involve foreign students studying for Australian PhDs within the project as part of the capacity building objective; another mutual benefit that provides greater incentives for a successful project which is not a bad thing as Adam Smith would have said.

    And what is wrong with many Australian researchers in this area? Australians have punched above their weight in the administration of international agencies despite quotas and in the more recognised work of the professional journals. Many of the people concerned have had connections with the ANU and the Crawford School in its various guises like Sir John Crawford, Ross Garnaut, Ron Duncan, Stuart Harris, Kym Anderson, John Freebairn and Peter Warr to name a few. This relatively high representation reflects what might be expected of a country with a high comparative advantage in agriculture and large agricultural-related institutions, including those involved in research such ABARES, parts CSIRO and the PC, and equivalent state agencies.

    Finally, all ACIAR project proposals have more or less from the outset included an ex ante impact assessment and many completed projects undergo a formal ex post one, the benefits of which in terms justifying its existence have been well understood long before some other agencies.

Leave a Comment