Statistical modelling of incentive design under limited information - The case of public access to farmland

This paper explores the use of statistical modelling to aid efficient polic y design for the provision of environmental goods on farms under conditions of adverse selection. The specific case of incentive-based schemes to enha nce the supply of public access to farmland is used as an example. A range of site willingness-to-pay (WTP) distributions are used to explore the bene fits derived from policies designed under different levels of information. Where no individual farm information is available, low payment rates are op timal, but efficient market creation may not be possible. Increasing the in formation set allows discretion in pricing and entry: optimal payment rates and net benefits are higher and the pay-off from procuring improved inform ation can be substantial. Such benefits are reduced where there is a welfar e cost associated with increased government expenditure. Optimal policies a re sensitive to the skewness of the WTP distributions. Mechanisms for incre asing the information set available to policy makers are discussed.