Globally, drastic biodiversity declines and the worsening climate crisis demand overhaul of existing land use policies which have failed to reconcile food production and environmental conservation. In Europe, most existing policies compensate farmers to voluntarily implement land-sharing measures, commonly referred to as wildlife-friendly farming, which seeks to deliver conservation benefits on the farmed land through agri-environment schemes (AES) offering a fixed price per hectare. Investment into sharing has continued despite the accumulation of evidence showing that, for the same amount of lost food production, substantially more would be delivered for conservation and climate change mitigation with the contrasting approach of land sparing, where high-yield farming allows large areas to be spared elsewhere in the landscape as (semi-)natural habitat. Following Brexit, the UK has the opportunity to rethink this approach; but until now policy decisions have had to be made without estimates of the relative taxpayer costs of using sharing and sparing to deliver target conservation outcomes. Addressing this critical research gap was the primary aim of this thesis, as follows.

In this thesis, I sought to uncover the taxpayer and food production costs of delivering meaningful conservation outcomes with land sharing and sparing. First, I conducted a novel comparison of the costs of monitoring sharing and sparing schemes for compliance and effectiveness. Monitoring is a fundamental, though often overlooked, taxpayer cost. In terms of effectiveness monitoring, I found current monitoring levels to be insufficient to precisely determine the effects on wild species of sharing schemes; in contrast, the same effort could deliver relatively precise estimates of the much larger effects of sparing. Furthermore, turning to compliance monitoring, I found the cost-effectiveness of existing English AES could be vastly improved with more compliance monitoring; however, this may be politically unpopular with farmers. It is therefore notable that I also found relatively less money was wasted when monitoring sparing at a sub-optimal rate compared to sharing. Second, I used a discrete choice experiment involving 118 arable farmers to establish their willingness to accept (WTA) payment to participate in sharing and sparing schemes that delivered the same biodiversity and carbon outcomes. I found that all but the most farmland-tolerant outcomes were delivered at less taxpayer expense with sparing. Third, combining this assessment of farmer WTA with knowledge of how much schemes must be monitored, I compared the taxpayer costs of delivering the same environmental outcomes with fixed-price sharing and sparing schemes which paid all recruits at the WTA of the least-willing farmer required in the scheme to deliver the target outcome. I found that sparing delivered the same outcomes at less than half the taxpayer cost of sharing; and, importantly, sparing saw only 79% of the food production lost under sharing. Fourth, I examined the distribution of farmer stated WTA, finding that variation in responses was mostly driven by factors other than lost gross margin. Given marked inter-farmer variation in their stated WTA, variable-price schemes, which pay farmers their stated WTA rather than the rate required by the least-willing participant, offered savings to both sharing and sparing schemes. However, even under variable pricing, sharing was not cheaper than sparing in delivering our more farmland-sensitive outcomes. Finally, I examined whether a land-purchase strategy, where the government purchases land and then contracts organisations to manage and create habitat on it, would deliver sparing at less expense than a farm-subsidy approach. I found land purchase was more cost effective than the farm-subsidy approach if long timeframes, low discount rates and large budgets were considered; however the impacts on farming communities of largescale ownership changes warrant further consideration.

To conclude, I found overwhelming evidence for UK arable farming that land sparing can deliver biodiversity and carbon outcomes at substantially lower cost than land sharing both in terms of taxpayer costs and lost food production. The relative costs of sharing would increase even more with consideration of species that do not tolerate farmland, in a country with a shorter history of agriculture where fewer habitat specialists have gone extinct compared the UK, and if the production required elsewhere to compensate greater volume of food production lost under sharing was taken into account. Furthermore, the effects of sharing may be near-impossible to precisely determine with current monitoring efforts and continued sub-optimal compliance monitoring would increase the costs of sharing-like options relatively more than the more sparing-like options of existing AES. Whilst variable pricing and land purchase may further reduce the costs of sparing, the costs of delivering meaningful environmental outcomes are most substantially reduced by pursuing a land-sparing, rather than land-sharing, approach. This work is of considerable significance to the UK government, given that prevailing land-sharing policy approaches can at best deliver less than half the environmental outcomes delivered by the same budget spent on land sparing.