Farming for two harvests: Brussels study finds biodiversity and yields can move together on European cropland

On 14 July 2026 a peer-reviewed paper in Biological Conservation landed with an unusual claim: small, carefully designed strips of wild habitat inside working farmland can lift insect biodiversity measurably while leaving crop yields effectively intact. The authors, from the WILD research group at the Vrije Universiteit Brussel (VUB) and UCLouvain, call the practice "agriwilding," and they argue the data blunt one of the longest-running objections to ecological farming in Europe: the idea that more nature on the land means less food from it.

The argument matters because Europe's two crises of the rural landscape, collapsing insect populations and tightening margins for arable farmers, are usually framed as competing. The Brussels-led study does not settle the policy fight, but it pushes the question into a more empirical register: what if the trade-off is smaller than the rhetoric suggests?

What the trial actually measured

The WILD team worked across experimental sites in Belgium and surrounding regions, comparing conventional arable plots against matched fields where wildflower strips, beetle banks (raised earth ridges that shelter ground beetles and other invertebrates), and other non-crop features had been installed along field edges and through the working area. The headline result reported by Phys.org is that insect diversity rose sharply on the agriwilded plots while crop output held essentially steady against the conventional comparison.

The framing matters. The biodiversity gain is not a vague impression; it is a measured rise across pollinator and predator groups that conventional agriculture has thinned for decades. The yield result, equally important for a hungry continent, is a near-flat difference rather than the steep penalty that intensive farming advocates have long assumed.

That combination is what makes the paper more than a curiosity. Two long-running policy debates in Brussels, the next Common Agricultural Policy cycle and the Nature Restoration Law's implementation timetable, both depend on whether ecological interventions on farmland cost farmers in tonnes per hectare. The WILD numbers suggest the cost can be small, at least under the conditions the trial captured.

The counter-reading

The obvious objection, voiced for years by farm unions and the European People's Group in the European Parliament, is that trial plots are not real farms. A two-hectare experiment under researcher supervision can deliver biodiversity gains and stable yields in ways that do not survive contact with a 600-hectare operation running on tight windows and tighter margins. The authors of the VUB–UCLouvain paper are candid about the limits of their sample; Phys.org's summary flags the early-stage nature of the work. Critics will argue, fairly, that scaling agriwilding across the EU's 157 million hectares of utilised agricultural area would require cost structures, machinery adaptations, and extension services that do not currently exist.

A second caution: the yields reported are near-flat on average, but averages hide variance. Some plots, some years, some crops will lose ground to the wild strips; that lost ground will concentrate on farms already under financial pressure. The policy question, then, is not whether agriwilding works in principle but who absorbs the risk in practice, and through what instruments, as the post-2027 CAP takes shape.

A quieter agricultural transition

For most of the post-war period, European farm policy has treated biodiversity as a residual: a benefit to be tolerated around the edges of a production system built, through successive CAP reforms, to maximise output per hectare. Insect declines, soil exhaustion, and the cost of nitrate and pesticide clean-ups have slowly inverted that priority in Brussels rhetoric, but rarely in the field.

The agriwilding result sits inside that inversion. It does not call for rewilding at the expense of farming; it proposes small, structured habitats woven into the working landscape, designed to deliver pollination, pest control, and soil function as services. The institutional logic is closer to integrated pest management than to set-aside. The political logic, harder to extract from a single paper, is that farmers can be paid for outcomes (pollinator counts, predator diversity, soil organic carbon) rather than for land left idle, which is a payment model the Commission has been circling for years without quite landing.

What hangs on the next CAP

The practical stakes sit in two places. The first is the implementation file for the EU Nature Restoration Law, which obliges member states to halt and reverse biodiversity decline by 2030, with explicit targets for pollinators. The second is the post-2027 CAP itself, currently in early co-decision, where eco-scheme budgets and the conditionality attached to direct payments will decide whether measures like those in the WILD trial are rewarded at the scale European farming actually operates at.

Belgium's federal and regional agriculture ministries, the Walloon and Flemish farm unions, and the European Committee of Professional Agricultural Organisations (COPA) are the obvious readers of this work. Whether any of them treat the VUB–UCLouvain result as a building block or as a footnote will say more about European food policy in 2030 than any single vote in Strasbourg.

What remains uncertain

The peer-reviewed result is a signal, not a settlement. The trial conditions, the climate, the soil types, and the crop mix do not span the full range of European farming; longer and larger studies are needed to track whether insect populations, once lifted, stay lifted, and whether yields remain stable across drought years and pest outbreaks. The cost side of the ledger, in particular, is sparsely reported. Until agriwilding is tested at commercial scale, with its labour and machinery costs honestly accounted, the central policy question (whether taxpayers or consumers subsidise the transition, and through which channel) remains open.

For now the WILD paper offers European farmers and the institutions that write their cheques something they have not often had together: evidence that more life on the land and competitive output from it are not necessarily opposites.

Desk note: this piece frames the VUB–UCLouvain study as empirical input to a live EU policy debate, not as an advocacy document for any farming model. Where the press summary and the unions' reservations diverge, both appear in the body.