Europe has protected more than a quarter of its landmass, closing in on the COP15 biodiversity target for protecting 30% of land and sea by 2030. Yet a comprehensive new study, forthcoming in ‘The Review of Economic Studies’, finds that four decades of EU land protection have had little measurable effect. TSE’s Mathias Reynaert explains why this ambitious policy has failed and how it can do better.
At first glance, Europe’s land protection efforts look like a success story. What motivated you to take a closer look?
Since 1900, Europe’s forests have expanded by more than 30%, an area the size of Portugal. However, this trend is slowing substantially, and our study finds that much of the EU’s vegetation is under threat.
It is therefore critical to ensure that land protections deliver environmental benefits. A lot of previous evidence comes from developing countries, often focused on deforestation. We wanted to understand what happens in a richer, highly regulated setting like the EU, where land-use pressures are very different.
How do you evaluate whether land protection ‘works’?
We start from the simple idea that successfully protected areas should have more green vegetation and less human development. To identify causal effects, we exploit the staggered rollout of more than 118,000 protection areas across Europe, comparing these zones before and after protection to carefully chosen control areas.
We built two remote-sensing datasets covering every square kilometer of the EU from 1985 to 2020. With over 100 million grid cells, the first allows us to compare satellite images of vegetation cover, which is closely correlated with bird diversity. The second tracks night-time light intensity, which we use as a proxy for human economic presence.
What did you find?
The headline result is quite stark. We find no meaningful impact of protection on either vegetation or night-time light. This holds across countries, long time periods, and after controlling for such factors as initial vegetation, soil, climate, population density and agricultural productivity. Even after three decades, protected areas do not seem to have changed outcomes in a measurable way.
Why does so much protection fail to protect anything?
The key issue is that protection often doesn’t “bite”. The overall size of the protected area may look impressive, but the policy often fails to change behavior.
Strict protections for national parks and wilderness reserves — which make up less than 8%[JN2.1] of the EU’s protected land — tend to be applied in places that are already sparsely populated and green. These areas are not at grave risk from human encroachment, so protection changes almost nothing about how the land is used.
In areas that do face development pressure — and so could really benefit from protection — rules are often weaker. We find that the majority of protected areas[JN3.1] fit this description, allowing agriculture and other economic activities to continue almost as if there were no protection at all.
Could it be that the policy is buying insurance against future threats, protecting land before it comes under pressure?
Our findings do not imply that protection will never deter economic activity in the very long run. However, if protection were preventing a future threat, you would expect to see protected areas diverge from matched controls as development pressure increases. Yet we see no such divergence, even though we track protected land for up to 30 years after its designation. The insurance argument seems particularly implausible given that our main result applies equally over time for all land types and in countries with very different enforcement capacity.
Why do you highlight the decision-making process as the root cause?
European-level incentives do not align with those of local stakeholders. The EU captures international goodwill from hitting a percentage target while the local authority bears the economic cost of restricting land use.
The EU’s targets are easily undermined because it delegates siting and enforcement to national governments, who often delegate further to provinces and municipalities. This procedure allows for a great deal of input from local communities, landowners and farmers, who may place limited value on global biodiversity.
To minimize opposition from constituents, the rational strategy for local authorities is to place protections where nobody will object. This tends to be land that faces no development pressure and so does not really need protection.
What do your findings imply for other countries trying to meet the 30×30 target?
The core lesson is that area-based targets can be undermined by decentralized implementation. The mechanisms we document in Europe — designation of low-pressure land, weak protections, local stakeholder capture — are often even more prevalent in lower-income, forest-rich countries where enforcement capacity is limited and the economic stakes of land restrictions are higher.
What would work better? Siting decisions need to be made at a higher level of government that internalizes the broader ecological benefits, avoiding delegation to local jurisdictions whose constituents bear costs but capture few gains. Protection should also be evaluated using outcome-based metrics, not area counts. Cost-benefit analysis, with biodiversity gains valued properly, should drive which land gets protected and with what stringency.
Are there grounds for optimism?
The broader literature on tropical forest protection, which we reviewed in related work, finds stronger effects in well-enforced areas that face development pressure. This suggests that, with the right incentives, land protection policies can work.
Our findings are a major new scientific contribution, offering rigorous evidence for today’s policymakers to act on. The EU may be moving in the right direction with the Nature Restoration Law, although its restrictions on agriculture are likely to spark fierce political opposition.
KEY TAKEAWAYS
• Scale is not impact - Large protected areas have had little measurable effect on vegetation or economic activity.
• Protection needs ‘bite’ - Strict rules are typically applied in remote areas, while high-pressure areas face weaker protections.
• Missing the wood for the trees - Decentralized decision-making prioritizes local impacts over wider environmental gains.
• Better targets - More effective conservation would focus on high-risk areas, stronger enforcement, and outcome-based evaluation.
