Pannell Discussions

No. 66, 17 October 2005

Thinking like an economist 21: Using incentives to buy land-use change in agriculture for environmental benefits

In general, the use of incentive payments in environmental programs is poorly thought through. This article discusses situations where environmental incentive payments are more likely to be a cost-effective response by environmental funders.

Incentive payments are used in many countries to encourage people to change their behaviour in ways that would benefit the environment. However, the use of incentives is usually not very discerning, and often the money involved could be spent more effectively.

The money involved is large. In Australia, for example, the two main national programs for improved environmental management in rural areas are the National Action Plan for Salinity and Water Quality (NAP, a $1.4 billion program over eight years) and the Natural Heritage Trust (NHT, a $1 billion program over five years). Sizable shares of their budgets are planned to be spent on incentive payments to landholders, mainly farmers, to change their land use practices.

One of the main changes being purchased is replacement of traditional, short-lived or 'annual' agricultural species, with longer lived 'perennial' species, to address problems of dryland salinity, biodiversity loss and soil erosion.

Incentive payments are used in two broad ways:

(a) to encourage people to trial, and (it is hoped) subsequently adopt, new practices that are believed to be in their best interest already, and that happen to also benefit the environment, and hence benefit the broader community.

(b) to compensate people for adopting practices that result in net costs to the adopters, but which benefit the environment and the broader community. 

This is a crucial distinction, but one that, in my experience, many environmental funders do not sufficiently recognise. It is essential to think through these two cases if incentives are to be used well.

In case (a), the incentive payments can be small and temporary. They need only to be big enough to be effective bait, rather than long-term sustenance, because once hooked, landholders will realise that they like being hooked, and will be happy to stay hooked in the long-term. This is the same sort of situation where one could successfully use education and communication programs to encourage behaviour change. The incentive payments would mainly accelerate the change, rather than raise its final level. The final actual level of change would depend on how attractive the new land-use practices were to landholders after incentive payments ceased. So, a key question for a funder who is considering offering small, temporary incentive payments is, to what extent are the changes likely to be maintained after the payments cease? In other words, how adoptable are the practices without incentives. (See Pannell et al. (2006) for discussion of what it takes for a technology to be adoptable.) If they are not adoptable on a sufficient scale to achieve the desired environmental benefits, it doesn't make sense to offer small, temporary incentives to encourage trialling.

In case (b), the incentives need to be large enough to provide both bait and sustenance (i.e. to compensate for losses relative to traditional practices). The incentives could either be ongoing regular payments or a larger up-front payment, but in either case they would need to be larger than case (a), probably substantially larger. Case (b) is like throwing fish food off a jetty to attract fish. The fish would hang around the jetty, enjoying the free feed for as long as it was provided, but would quickly drift away if the feeding stopped. That is, because the new practices are not sufficiently attractive in themselves, landholders would dis-adopt them once incentive payments stopped, unless the system forced them to continue through a contractual agreement, backed by monitoring and enforcement (a further expense).

Another analogy of the two cases would be that in case (a), offering incentives is like an advertising campaign, to let people know that an attractive product is available, whereas in case (b) advertising is not sufficient the product is not sufficiently attractive and we need to subsidise the price of the product in order for people to buy it.

A major problem with existing incentives currently being paid under the NAP and NHT is that the funders often confound the two cases. They are offering small, temporary incentives, as if they were dealing with an example of case (a), in situations that would actually require large, permanent incentives to be effective in the long term case (b). They have not evaluated the land-use changes to see whether they are adoptable on a sufficient scale to achieve the desired benefits before designing the incentive system.

The result is that much of the money is being spent in ways that will not generate environmental benefits in the long term, because the changes will be undone once the payments cease, or once markets shift to favour other land uses.

A second problem with current incentive systems is that, in many cases, they are not well linked to environmental outcomes. Even if the land-use changes they buy were maintained in the long term, they may be in locations or at scales that mean that the resulting environmental benefits are small in relation to the payments being made. Environmental funders need to pay close attention to the cause-and-effect relationship between land-use change and environmental benefits if they are to avoid spending money on land-use changes that would not actually benefit the environment. Currently in Australia, there is not sufficient onus on the regional bodies that channel the funds to landholders to do this that is, to seriously consider whether the changes they are buying will actually achieve environmental outcomes.

I have argued before that, for salinity management, case (a) is actually not as common as we would like, because there is currently an insufficient range of perennial plant options that are economically attractive (Pannell and Ewing, 2006), although there are exceptions, of course. More generally, a sceptic looking at funding allocated to a supposed example of case (a) could reasonably ask, if the proposed new land use practices are actually in the best interests of the landholders, why have they not been adopted already. I suggest that the practices would need to be new and poorly known to landholders for the argument that it is actually an example of case (a) to be convincing.

Identifying examples of case (b) that are worth funding also poses a number of challenges. The payments to landholders need to be large enough to cover:

the financial shortfall between new and traditional land-use options

a risk premium if landholders are to be contracted to maintain the changes in the longer-term (which they need to be if the payments are to be assured of achieving outcomes)

a further incentive to prompt them into action ('bait')

In addition there would be costs of monitoring and enforcing agreements. Because the costs involved in this case need to be large for the system to be effective, this type of incentive payment should be targeted to particularly high-priority cases. These will be cases where the values of the assets under threat are outstandingly high, and the planned land-use changes can be effective in preserving or enhancing them. Not surprisingly, Ridley and Pannell (2005) concluded that case (b) style incentive payments are justified only in special cases in the management of dryland salinity.

There is a situation that is probably somewhere between cases (a) and (b). If environmental benefits can be generated by small, relatively low-cost changes, it may be possible to convince landholders to bear those costs for the good of the broader community. In other words, because of their personal commitment to the environment, they would adopt them for the long term as a result of small, temporary incentives, so in this way it is similar to case (a). But if the changes required are large and expensive, we should not expect them to happen so easily.

So, in view of all that, setting aside the question of who should pay (PD#21), the key questions that planning bodies need to address to design sensible and effective environmental incentive schemes include the following.

1. What is the relationship between the scale of change (e.g. in land use) and the level of environmental benefits?

2. What is the relationship between the scale of change and the level of incentive payments that would be needed to achieve that scale of change?

Without answers to both of those questions, you cannot hope to adequately determine what level of incentive payments should be offered, if any, and for which purposes?

To address question 2, a number of subsidiary questions are needed:

2.1 Are the new land-use options readily 'adoptable' (i.e. sufficiently attractive without extra incentives)?

2.2 If so, on what scale?

2.3 If so, will small, temporary incentives help to accelerate that adoption?

2.4 Where they are not already adoptable, what scale of incentives will be needed to achieve the desired changes (compensation plus risk premium plus 'bait')?

2.5 If those larger incentives are paid, what system of monitoring and enforcement will be needed, and what will it cost?

So far I have not really been talking much like a hard-core economist. A hard-core economists would talk about market failure and about considering the marginal benefits and marginal costs when setting the incentive rate.

With market failure, the argument would be that in order for an incentive payment to be justified in principle, you need to be able to identify a factor that is causing the free market to fail to deliver the maximum possible benefits. A classic example of such a cause is an 'externality', which is a benefit or a cost that accrues to someone other than the decision maker. The implication for environmental funders would be that the changes funded would need to generate benefits for people other than the landholder (often loosely referred to as 'public benefits', PD#22). In this discussion, we've got this covered, as cases (a) and (b) at the beginning both refer to benefits 'to the broader community'. A second cause of market failure is poor information, as in case (a) where landholders apparently don't realise what is good for them. In fact, the existence of externalities or information failure is not a sufficient condition for market failure to exist. It is also necessary to check that the benefits of addressing the problems outweigh the costs (PD#35).

Finally, if it were possible to answer the above key questions 1 and 2 quantitatively and in dollar terms, it would be possible for funders to determine the optimal scale of land-use change and what rate of incentive would need to be paid to achieve it. You could use the two relationships from questions 1 and 2 to look at a range of scales of land-use change (corresponding to different incentive rates via question 2) and see which scale generates the greatest net benefits. It would correspond to the rate at which the marginal benefits (the additional environmental benefits from an extra hectare of land-use change) are equal to the marginal costs (the additional incentive payments required to cause land use to change on that extra hectare).

In practice this is extremely difficult to apply strictly. Neither of the two questions can be answered with much precision or certainty, and answers to question 1 will probably be difficult to convert into dollar terms, especially if the benefits include improvements in the natural environment (PD#30). Nevertheless, it should be possible to specify the approximate outcomes that are expected to result from payment of incentives: what level of additional adoption would occur, and what environmental benefits would follow (in biological, physical or economic terms)? If these cannot be specified, basic accountability requirements would say that the funders should not be spending public money on the incentives in question.

David Pannell, The University of Western Australia, David.Pannell@uwa.edu.au

Further Reading

Pannell, D.J. and Ewing, M.A. (2006). Managing secondary dryland salinity: Options and challenges, Agricultural Water Management 80(1/2/3): 41-56. Full paper (66K)

Ridley AM and Pannell DJ (2005). SIF3: An investment framework for managing dryland salinity in Australia. SEA Working paper 1901. CRC for Plant-based Management of Dryland Salinity, University of Western Australia, Perth. Full paper (at SEA News site)  Full paper (126K pdf)  2-page summary  SIF3 project page

Pannell, D.J., Marshall, G.R., Barr, N., Curtis, A., Vanclay, F. and Wilkinson, R. (2006). Understanding and promoting adoption of conservation practices by rural landholders. Australian Journal of Experimental Agriculture 46(11): 1407-1424. Access paper at Journal web site here. Pre-publication version available here (161K).


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

65.  Challenges for policy economists  15 Aug 2005

67.  Using extension to encourage land-use change  24 Oct 2005 

URL for this page: http://dpannell.fnas.uwa.edu.au/pd/pd0066.htm

Home page: davidpannell.net
Copyright David J. Pannell, 2005
Last revised: June 01, 2011.