Pannell Discussions

No. 151, 13 April 2009

Additionality in environmental programs

Recently, someone was telling me about a study that had found that the value of environmental services provided by a sub-set of Australian farms was several fold larger than the value of their agricultural production. I have doubts about the accuracy of this, but my biggest doubts are about its relevance. It is not really relevant to any sensible environmental policy question, in part because it ignores the principle of additionality.

Additionality basically means that a policy shouldn't pay people to do things that they were going to do anyway (or were already doing). They should be doing additional works to qualify for funding.

This is important because resources are limited. If you create a policy program where people can be paid for environmental benefits that they would have generated even without being paid, there are consequences:

(a) If the policy program has a fixed budget (most do), it basically means that the program will deliver less in the way of environmental benefits than it could have done, because some of the money is being spent on things that do not generate new benefits.

(b) Even if the budget is not fixed, it means that the program is transferring money from taxpayers to recipients under the program, without generating new environmental benefits. 

Some might argue that transferring money to farmers is a good thing, but if financial benefits are to be transferred to a particular group, it is better to do this explicitly, rather than hide it within a so-called environmental program. It is also worth recalling that spending public money involves taking it away from other uses via the tax system. And that it is not free to move money around like this; it may cost something of the order of $0.30 in administration costs to spend each tax dollar.

There has been a lot of attention paid to the concept of additionality in the area of climate change policy. As an illustration of how important it is to focus on additional carbon sequestration, imagine if every existing tree in the world made its owner eligible for a carbon payment. This would move massive amounts of money around, without achieving a thing for climate mitigation, since the trees are already there and would have stayed there.

There has been much less attention to additionality in areas like rural environmental policy, although it is just as relevant.

Some argue that environmental payments can be a legitimate source of income for farmers. I agree that it can, but one has to be cautious it.

For example, if a farmer has to give up an income-earning activity to undertake some environmental works, it is reasonable to compensate him or her for the resulting loss of income. Indeed, they will probably not undertake the environmental works unless you do compensate them. Such a payment is consistent with additionality because the works would not proceed without the payment. However, payments beyond the lowest level needed to make the farmer willing to take up the  works would not be consistent with additionality.

Although conceptually the idea of additionality is simple, it can be difficult to apply it in practice. In particular, how can you tell what people would have done without the payment, considering that people's behaviour and business management is always in flux to some extent? One strategy is to use an auction or tended-based process. Participants make bids, revealing what they are willing to do for a certain price, and the environmental managers choose those bids that offer the best value for money. Other options might include surveys and economic modeling, which could be of more or less relevance depending on the context.

David Pannell, The University of Western Australia,


Sue Briggs wrote to say that another option to judge additionality "is to work from the minimum legislative (or already covenanted where a covenant is in place) standard, which is what we do for biobanking and other similar schemes in NSW".

 I can see that would be convenient, and might have relatively low transaction costs. It wouldn't be appropriate in all cases, though, because for some issues the minimum standard is very low (or non-existent). - DP

I enjoyed your discussion on additionality. Related to this concern, I've run into a recent study (in Ontario) that makes mighty claims with regard to the magnitude of ecological benefits associated with agricultural lands. However, this study doesn't compare the ecological benefits of the current land use (e.g., farming) with the next best alternative. In some cases, for example, the alternative might be residential housing in other more extreme cases the land might be abandoned or return to forests. Wouldn't the ecological benefits of farming have to be compared with the alternative land use? For example, (value of ecological outcomes if land in agriculture) minus (value of ecological outcomes if land is residential = ecological benefits of agriculture)? - Brady Deaton

Thanks for the response. That's an interesting point. Yes, you're right. Any value has to be expressed relative to something else, and the main issue here is being clear about what that something else is.

If the question is, "should we have agriculture or not?", then I think the alternative is forest, or unmanaged grasslands, or urban development, or whatever makes sense in that environment.

If the question is, "what should we pay farmers who are going to continue to be farmers whether or not we pay them?", then the comparison should be with what those farmers would have done without the payment. -DP

Pannell Discussions are brief pieces on issues and ideas in economics, science, the environment, natural resource management, politics, and agriculture.

150. Why don't environmental managers use decision theory?  6 Apr 2009

152. "Understanding practice change" videos and podcasts  20 Apr 2009

URL for this page:

Home page:
Copyright 2009 David J. Pannell
Last revised: June 01, 2011.