Monthly Archives: February 2013

232 – Agricultural extension in 10 years

Sally Marsh and I wrote an article for the Australian Farm Institute titled “Public-sector agricultural extension: what should it look like in 10 years?”. Here it is. 

Our focus is on what should happen in the public sector, on the grounds that it is not helpful to ask what “should” happen in the private sector. The private sector will develop in response to commercial opportunities available to them, irrespective of what we might think should happen.

To set the context, here are some predictions about the environment within which extension will operate. Agriculture will continue to change in response to technology, markets and climate. Cutbacks we have seen in funding for public-sector agricultural extension will not be reversed and may continue. The dismantling of extension infrastructure and capacity in the public sector has gone too far for it to be reversed without major new public investments, and we don’t foresee those occurring. Private sector capacity in extension will continue to grow – including extension provided by purchasers of agricultural products (e.g. dairy, horticulture, sugar), input suppliers (e.g. fertiliser, feeds) and farm management specialists. There will be continuing increases in the average size of farms, and in the number of corporate farms, with resulting growth in the vertical integration of information services (~ “extension”) into farm businesses. There will continue to be growth in the use of advanced information and communication technologies in agriculture, providing information to farmers in novel ways. Falling numbers of graduates from agricultural programs could create a serious challenge to extension services (public and private) to obtain employees with the required knowledge and skills.

In this context, is there a need for ongoing public investment in agricultural extension? We believe that there is. Public-sector agricultural extension can continue to play important roles that address various market failures.

One key role is to foster two-way information flows between researchers and farmers. Information flow from farmers to researchers is needed to ensure that the research conducted will be beneficial to farmers and likely to be adopted by them. Some researchers already have sufficiently strong relationships with their farmer audience not to need this sort of help from extension agents, but many others don’t. The traditional role of extension agents in promoting uptake of beneficial new research results (technologies, systems and practices) should continue. We do not share the negative view of technology transfer that seems to exist among some theorists of extension. We believe that technology transfer and approaches such as participatory research and farmer-to-farmer learning are not mutually exclusive. Indeed, these latter approaches, as part of a broad portfolio of extension methods, can make valuable contributions to the success of technology transfer in appropriate circumstances. Farmer groups and organisations such as the Grower Group Alliance ( have key roles to play in this process.

Given that public budgets for extension are unlikely to grow, and may shrink further, it will be crucial for public extension services to take a more business-like approach to prioritising their activities than they have commonly done in the past. Extension efforts should be focused on issues for which there would be substantial benefits to farmers from changing their practices, especially if those new practices would also generate benefits for the broader community (e.g. environmental benefits). Extension would not focus on practices that farmers already have good knowledge about and have decided not to adopt, because non-adoption is a clear signal that the practices do not generate large enough private benefits. The heterogeneity of farms and farmers should be recognised when looking at reasons for non-adoption. This more sophisticated approach to planning extension effort will require greater collection and analysis of information.

As important as social media and other modern communication methods will be, public extension should not rely on them exclusively, but should maintain a level of face-to-face communication. Farming is already socially isolating for some farmers, and with declining farmer numbers this may become a more widespread issue. It is likely that farmers will always put a high value on personal contact in extension.

Finally, we note that, in the past 20 years, public sector extension has been prominent in supporting natural resource management (NRM) policy for agriculture. It has been the go-to policy response of most government NRM programs. Unfortunately, these programs have often funded extension efforts without asking fundamental questions, such as, Are the practices we wish to promote actually adoptable by farmers? A more thoughtful, selective and evidence-based use of extension is needed in this policy context.

 Further reading

Marsh, S.P. and Pannell, D.J. (2000). Agricultural extension policy in Australia: The good, the bad and the misguided. Australian Journal of Agricultural and Resource Economics 44(4): 605-627. Journal web site here ♦ IDEAS page for this paper

Pannell, D.J. and Marsh, S.P. (2013). Public-sector agricultural extension: what should it look like in 10 years? Farm Institute Insights, Vol. 10, No. 1, February 2013. Here


231 – Selecting environmental projects: prioritizing really matters

Almost all environmental programs are short of resources. Picking the best projects can make a huge difference to the benefits generated by a program.

There are more environmental projects available than can be afforded – often many more.

For example, a large national program in Australia recently funded around 5% of proposals received. Of course, proponents have already selected from among the available projects to some degree, so it was probably not more than 1% of potential projects that actually got funded.

Within the set of potential environmental projects, heterogeneity is enormous. They vary widely on every parameter: scale, feasibility, riskiness, importance to the community, likely compliance, costs, time frame, and so on.

As a result, the best projects are much, much better than the rest, meaning that programs can perform much better if they choose their projects well. Unfortunately, the importance of this is under-appreciated by most people involved in managing or participating in environmental programs. Many funded projects are mediocre, while great projects go unfunded or underfunded.

To illustrate how much it matters to focus resources on the best projects, the figure below shows a graph of the benefits and costs of around 7000 potential environmental projects in Australia (from Fuller et al. 2010, courtesy of Richard Fuller). Benefits are measured as the project’s contribution to protecting heavily cleared vegetation types. (See Fuller et al. 2010 for details.) It’s an incomplete measure of benefits, but it’s still useful to illustrate my point.

Note that the benefit and cost axes have been expressed in log scale to allow very small projects to be distinguished on a graph that includes some very large projects. The ranges of benefits and costs are enormous.

If we had enough money to fund 5% of the projects, the set that would generate the largest environmental projects would be the 5% with the highest Benefit: Cost Ratios (BCRs). These are shown as green triangles in the graph.

For this data set, the average BCR of the best 5% is 330 times greater than for the median project. For the best 10%, the ratio is 200 times.

Clearly, if a program faced with these 7000 potential projects fails to fund the best ones, the loss to the environment would be extremely large. From an environmental perspective, it would be well worth putting resources, time and effort into evaluating the projects to identify the best ones.

The quality of information and analysis required to do this successfully is significant. It’s certainly do-able (PD#185), and it’s well worth doing, but it’s rarely done. It requires the sort of analysis that underpins a good conservation tender. I don’t believe that the tender process itself is the key factor – it’s the information and analysis.

Of course, picking cases with large potential BCRs is not enough. Delivering the potential benefits also requires that:

  • the project receives sufficient funding (see PD#210),
  • the project is well designed with a good internal logic,
  • the project uses appropriate delivery mechanisms,
  • the project is well implemented, and
  • the project is managed in an adaptive way to allow for learning.

But accurately picking the best projects is a great start.

Further reading

Boesso, G. And Kumar, K. (2007). Who or what really counts in a firm’s stakeholder environment: an investigation of stakeholder prioritization and reporting, “Marco Fanno” Working Paper N.51, IDEAS page for this paper

Fuller, R.A., McDonald-Madden, E., Wilson, K.A., Carwardine, J., Grantham, H.S., Watson, J.E.M., Klein, C.J., Green, D.C. & Possingham, H.P. 2010. Replacing underperforming protected areas achieves better conservation outcomes. Nature, 466, 365-367.

Pannell, D.J. (2011). Problems with environmental project prioritisation, Pannell Discussions 185. Here

Pannell, D.J. (2012). Under-estimating the costs of environmental protection, Pannell Discussions 210. Here