2. Limits of Agri-Environmental Contracting
While environmental contracting represents a significant innovation in policy terms, there are a variety of fundamental limits to this approach. As indicated in the Introduction, the source of most of these problems is asymmetric distribution of information between landholders and the conservation agency.
2.1 Adverse selection - targeting the 'wrong' farmers
Adverse selection refers to situations where a farmer with a low potential for producing high-quality environmental outputs has a greater incentive to enter a conservation agreement than a farmer with a high environmental potential. For example, a farmer who has already been using a low-input farming method has a greater incentive to sign up for a conservation scheme stipulating reduced usage of pesticides and fertiliser than a farmer with a high-input technology: the former will have to make fewer and less severe changes to current farming practices. This will result in comparatively small additional environmental benefits and an overcompensation of the farmer's compliance costs. In fact, the low-intensity farmer may have an incentive to conceal information about his pre-contractual farming practices from the agency or, even worse, may disguise himself as a high-intensity farmer in order to qualify for higher payments (Fraser, 1995; Wu and Babcock, 1996). Clearly, this reduces the scheme's effectiveness. Thus, the basis of the adverse selection problem is asymmetric information about individual farmers' compliance costs and pre-contractual farming practices: farmers differ in their compliance costs, and if these differences could be observed by the agency then farmers could be paid a rate equal to their compliance costs, thus saving money compared to the use of a pre-determined fixed payment level.
A study by Osterburg and Nieberg (1999) provides evidence of adverse selection. The authors analyse the spatial distribution of environmental scheme uptake in Germany. They find that participation rates are highest in regions with poorer soils, lower yields and a lower share of intensive crops, and a generally lower intensity of land use. Likewise, high participation rates are mainly found in schemes with less demanding management prescriptions, whilst uptake of highly targeted schemes addressing specific environmental problems is limited. It has been argued that adverse selection also occurred in the early rounds of ESA contracts in the UK (Whitby et al., 1990; Hodge, 1991; Colman et al., 1992; Whitby, 1994).
2.2 Moral hazard - incentives to evade contract requirements
The basis of the moral hazard problem is imperfect information about farmers' actual compliance. Imperfect compliance monitoring ( i.e. a detection probability of less than 100 per cent) provides farmers with an incentive to renege on their contracts: if they are successful in avoiding detection by the conservation agency, they receive the compensation payment without incurring the costs implied by their contractual obligations. Obviously, the propensity to renege on agreements depends on a number of factors including the farmer's attitude towards risk and morality. A formal analysis of the moral hazard problem is presented in Appendix 1. The modelling exercise shows that the conservation agency can manipulate four contract variables in order to prevent farmers from cheating. These are: (1) the probability of detection, i.e. the intensity of compliance monitoring; (2) the level of fine for detected contract violations; (3) the stringency of the management prescriptions; and (4) the payment rate. Since detected violations involve the full grant being repaid, a high payment rate acts as a deterrent to cheating.
The moral hazard problem appears to be significant in practice, with active monitoring of farmer compliance and significant (although varying) rates of cheating reported. Giannakas and Kaplan (2002) quote US statistics showing that out of 750,000 farmers receiving conservation payments in 1997, 50,000 were audited of whom 2,000 were found to be not actively applying the approved conservation plan. Land Use Consultants (1995) found that, on 24 percent of sites visited, farmers participating in the Countryside Stewardship Scheme in England were compromising their contracts in some way. Hanf (1993) reports that one third of farmers participating in a regional nature conservation scheme in Northern Germany were not fulfilling their contractual obligations. In other instances, compliance records appear to be satisfactory (see for example Anjou, quoted in Hart and Latacz-Lohmann, 2005). Especially if contracts have standard conditions and if these conditions are familiar to all landholders within a region, the propensity to cheat is likely to be low because there will be an element of self-policing among local landholders.
2.3 Dynamic inefficiency - lack of incentives for entrepreneurship
Hodge (2000) points out another incentive problem: the typical agri-environmental contract does not provide producers with adequate incentives to seek out new methods of reducing costs, to introduce new ideas, or to be willing to take risks for the provision of countryside goods. There is also hardly an incentive for local landholders to co-ordinate their actions across several holdings. Co-ordination, however, is required when the environmental objectives of the schemes relate to the landscape at a wider scale. For example, creating favourable conditions for rare wildlife species normally requires a co-ordinated effort across several holdings in the area. Present policies do not take account of such conservation synergies because they concentrate on contracts between government agencies and individual farmers.
The root cause underlying these incentive problems is the fact that the final outputs, i.e. the environmental goods and services produced, are not directly contractable because of difficulties in measuring the state of the environment and quantifying changes. This means that farmers' rewards cannot be based upon environmental outcomes and must instead be linked to activities, i.e. changes to the management of the land. These management changes are known to improve the environment, but the transformation function that maps these actions into outcomes is not known with certainty, even if the actions were carried out diligently. Further, the effects of unexpected events such as droughts or floods cannot reasonably be predicted. If the desired outputs were readily observable, as is the case in most other situations of government procurement contracting, farmers could be paid by results, that is, based on the quantity and quality of the environmental outputs produced. This being the case, the above incentive problems would largely disappear because higher effort, innovation and entrepreneurship would result in higher output and thus higher payment.
2.4 Uncertainty over property rights
Landscape and conservation values are usually only created over significant periods of time. This calls for relatively long-term contracts. However, when contracts expire, there can be no guarantee that the conservation assets will continue to be maintained. Farmers may have an incentive to return to more intensive forms of agricultural production at the expense of any conservation benefits that may have been created. This raises questions about the property rights in the environmental assets generated through environmental contracts. The public may feel that they have a proprietary interest in the environmental assets to the extent that they have been created through the contribution of public funds, and that they should have a right to prevent damage to this environment in the future. Farmers may anticipate this problem and so be reluctant to enter into environmental contracts in the first place, the concern being that restrictive designations might subsequently be introduced to protect long-term environmental gains (Hodge, 2000).
2.5 Information asymmetry and transaction costs
The costs of operating environmental programmes include both the incentive payments made to landholders and the costs to the agency of administering the programme. In economic terms, payments to landholders are mere transfers, whereas scheme administration incurs real economic costs. These transaction costs stem largely from information asymmetries between landholders and public agencies and the heterogeneity of producers. Falconer and Whitby (1999) distinguish three categories of transaction costs in the operation of agri-environmental schemes. These are:
- Information costs for surveying and designating areas of environmental sensitivity and designing appropriate management prescriptions;
- Contracting costs including promotion of the scheme to farmers, negotiation between farmers and agency, and the administration of contracts;
- Policing costs including costs of compliance monitoring and enforcement, environmental monitoring and scheme evaluation.
These costs tend to be disregarded in policy discussions and, where considered, it is generally assumed that they should simply be minimised.
Falconer and Whitby (1999) report the results of pan-European research into the administrative costs of agri-environmental schemes, involving 37 case study schemes in eight European Member States. Average annual administration costs ranged from €9 to €75 per hectare and from €140 to €2,446 per participant. Administration costs as a proportion of total payments to landholders varied from 6% to 87%.
These transaction costs represent a significant element of public expenditure and may be sufficiently important to constrain the resources available for implementing agri-environmental policies, especially in times of public expenditure scrutiny and cut-back. The danger is that the development of administrative structures may not keep pace with the rapid increase in the scope, scale and complexity of agri-environmental schemes. Insufficient scheme administration will inevitably result in reduced environmental effectiveness.