Executive summary

Europe must relearn how to build. Major European countries are building at roughly half their twentieth-century peak. The result is unaffordable housing, energy dependence, and stagnant growth.

EU environmental directives are part of the problem. This is not an argument against environmental protection. The directives have created a valuable floor of protection across the continent. But well-intentioned rules have created two distinct sets of side effects that make building harder than it needs to be.

The first is procedural. Directives like Habitats, Birds, EIA, and SEA were designed to ask sensible questions about environmental impact. In practice, they ask those questions in the wrong place: at the level of individual projects rather than strategic plans. Developers bear the burden of assessing cumulative, environmental questions that no single applicant can reasonably answer. This causes uncertainty, delays, and ultimately prevents Europe from building the homes and infrastructure it needs.

The second is cost. The Energy Performance of Buildings Directive inflates the price of every new home in Europe. Procedural burdens also add to costs through delays, increased financing costs, and the risk priced into everything.

Neither problem requires weakening environmental protection. This report identifies targeted changes to specific directive articles that would restore strategic logic to environmental assessment, allow effective nature offsetting, set a proportionate evidentiary standard for development, and reduce inflationary building standards. Europe can have a healthy environment and a dynamic economy. Small changes to how these directives are drafted would help deliver both.

The problem: Europe must relearn how to build

To be competitive, Europe must relearn how to build. A continent that rebuilt itself from rubble, now takes nine years to permit a wind farm. Europe built some of the greatest cities in the world. Today, its young people cannot afford to live in them. 

Building is hard for many reasons. Some lie with member states and their decisions. Others are made worse by EU directives. Europe can and should make building easier, not harder. 

This report will discuss one set of directives: environmental directives. This may sound like a strange place to start. Environmental directives are essential to protecting Europe’s natural environment. But well-intentioned rules can cause inadvertent side effects. Environmental directives are deeply interwoven with how land use is regulated across the continent, and reforming them is essential to getting Europe building again.  

Why are physical buildings and infrastructure so important for competitiveness? There are at least three reasons. 

First, building things is an essential part of urbanisation. Cities are the most important drivers of innovation and growth in modern economies. As they grow larger and denser, workers and firms get better at what they do. The productivity gain has been put at 3–8 per cent for each doubling in city size.¹ Anything that blocks people from moving into productive cities carries an economic cost. Another paper, from the U.S., estimates that easing housing-supply restrictions in productive American cities would raise aggregate national output by roughly 8 per cent.² The equivalent European figure has not been estimated as precisely, but European housing supply is less responsive to demand than American, suggesting the magnitude of the loss could be as large here in Europe.³ Barriers to urban growth hold back the whole economy by keeping workers away from the places where they’d be most productive.

For Europe to compete with the biggest and best cities around the world, we need to let our own cities grow. We can’t grow our cities without housing and supporting infrastructure, from sewers to roads, and mass transit to fibre-optic cables.

Second, to remove Europe’s reliance on the US, the Middle East, and Russia, we need to produce our own energy. This doesn’t require taking a particular view of the right mix of energy sources, which may well vary between member states. Whether it’s wind, solar, nuclear, or natural gas, we need to build much more capacity, and quickly.

Third, to lower costs to businesses and households, we need more of everything, from housing to grid infrastructure. The essentials are unaffordable because we stopped building them. We need to build to achieve the progressive goals that Europeans care about: healthcare, climate action, fertility, growth, intergenerational fairness, security. We cannot have fairness without growth and we will not have growth without relearning how to build. 

Europe has gotten worse at building. From 1955 to 1979, France increased its housing stock by about 2.3 per cent a year, while West Germany and the Netherlands each managed about three per cent a year. West Germany alone built 3.5 million houses in six years in the 1950s, while Sweden built a million houses in the 1960s and early 1970s.

We also built some of the best infrastructure in the world: our 19th and 20th century cities are well-serviced by roads, metros, and sewers. Our urban areas are often beautiful and continue to provide a high quality of life. When it comes to large projects, like the Shannon hydroelectric scheme, French nuclear plants, or the Madrid metro, we have a strong track record. 

However, we’ve fallen behind.

There are lots of reasons behind this, and this report doesn’t attempt a comprehensive diagnosis. But one core reason why Europe struggles to build what it needs is we’ve made it legally difficult to do so. Legal and regulatory complexity has, in aggregate, placed a straitjacket on European growth.

This is where environmental directives come in. They are, in many ways, one of the EU’s landmark achievements. They’ve created a “floor” for environmental protection in every member state, encouraged long-term thinking about development, and recognised the cross-border nature of many ecosystems.

But they have caused specific problems, with key projects delayed. A ring road in one of Ireland’s most important cities. A three billion euro of wind farm in Spain. A major underwater tunnel between Denmark and Germany. A five hundred mile rail project in the Baltics. An upgrade to Antwerp, Europe’s second biggest port. These are some of the most prominent examples, but every member state – even every local area – has its own version of an important project being held back unnecessarily.

These well-intentioned rules have inadvertently added to the complexity identified in the Draghi report. In the Netherlands, European rules have led to a freeze on new construction permits. Across the continent, important cross-border electricity and gas infrastructure projects take an average of 3.3 years to get permitting approval, and are frequently delayed. Permitting for large renewable projects can take up to nine years in some member states, resulting in the installation of outdated technologies. Across the continent, housing shortages stall growth, exacerbate inequality, and undermine cohesion. Paralysis is the norm. 

There is a better path available, one in which Europeans can enjoy a healthy environment and a healthy economy. This report outlines how small changes to particular directives can help get Europe building again. 

These environmental directives undermine building in two distinct ways. First, they contribute to unnecessary procedural burdens. And second, they disproportionately contribute to costs.

The first is a problem of process. Directives like Habitats, Birds, and the assessment regimes (SEA and EIA) were built to ask sensible questions: what will this do to the natural environment? And how might we prevent any environmental damage? But too often, the question is posed in the wrong place, assigned to the wrong part of the system, or simply impossible to adequately answer. 

Instead of addressing strategic environmental questions at the level of the regional or national plan, they are left to be litigated at the level of every individual project. The result is that a developer trying to build a few hundred homes inherits questions that no developer is equipped to answer: the cumulative effect of all development in a region, the fate of a migratory species across its whole range, the impact of “other plans and projects” independent of the applicant’s proposal. Uncertainty becomes the default, delay becomes rational, and litigation becomes the norm. This creates a chilling effect across the whole system: projects are never started, sites are never bought, and risk priced into everything.

Some environmental questions are doubtlessly best answered by a particular development proposal. But many are not. The default procedure is to make individual applicants bear the burden of assessing and mitigating any environmental damage. This creates legal risk since the prescribed process is often complex, leading to legal challenges of good faith planning applications. 

The second is a problem of cost. The clearest case is the Energy Performance of Buildings Directive, which sets how energy-efficient every new building must be. The goal of the EPBD is laudable but it drives up cost. While the Directive directly references cost (the “cost optimal” framework), member states consistently understate what compliance actually costs. The cost arrives in the price of every new home, in a continent that is already failing to house its people.

Neither problem requires abandoning environmental protection. The directives created a valuable floor, and that floor should stay. What has gone wrong is that the Directives have caused side effects: overly burdensome procedures placed on individual applicants, unreasonable standards of evidence, compensation treated as a last resort, and inflation that Europeans cannot afford. Each of the sections that follow takes one of these failures and shows how a small, targeted change to the drafting would fix it: getting Europe building again without costing the earth.

Making environmental protection more strategic 

In general, when we create a system, we find it easiest to set rules at the start and let the various actors operate within those rules. If the rules are set properly at the start, we don’t need to work them out at the level of every single interaction. This is true whether we’re playing football, driving a car, or overseeing a school playground. Set the rules in advance, and let people operate within that framework.

However, our current environmental regulation system does the opposite. Instead of a clear plan from the top, individual projects, one level down, become the battleground for the right and wrong way to protect nature. This increases costs and uncertainty for anyone trying to build something. 

It also creates transaction costs and coordination problems. The current system is based on the idea that where nature is going to be disturbed or damaged, the individual developer should do something about this. But it’s very difficult for a developer to work out the best interventions. It might be best for these interventions to take place somewhere else in the area – ecosystems are not site specific. Perhaps they require coordination with other developers or landowners, which can be difficult to achieve.

To see how this works in practice, take the case of the Light-bellied Brent Geese. They use a network of feeding grounds in the wider Dublin area from October to April each year. However, no statutory plan appeared to resolve this issue. This means the question of the geese is then raised at the level of individual projects. No one can say for sure what might happen to the geese, so projects are rejected. This has led to some 1,200 homes being denied permission overall.

Member states could take steps on their own to implement better, as mentioned in a 2016 Commission Fitness Check.⁴ Denmark, for example, maps its biodiversity at a national level as part of a plan-led approach to conservation. However, the drafting of the Habitats Directive currently requires some positive conservation measures but doesn’t compel member states to make the extensive plans needed to resolve this problem.⁵ As such, there is a “pass the parcel” dynamic whereby member states end up with systems that work out these questions when it comes to individual planning decisions, not the strategic level plans. 

At the same time, the drafting of the Habitats Directive places an emphasis on individual applicants to coordinate, again pushing individual projects to do something which should really happen at a higher level.⁶

To address this, relevant directives should be updated to improve strategic level plans and explicitly remove onus on project level assessments.⁷ Appropriate assessments should also be reformed to ensure applicants are not responsible for assessing the impacts of “other plans and projects.”⁸ 

Making nature protection more effective 

Once plans have been made, every actor in the planning system is in a better position to strategically protect nature. Interventions can be made to protect habitats and support species, taking an ecosystem-wide approach to given challenges.

For instance, the best place to help an endangered species is rarely immediately within a city or urban area, which is often where these developments take place. Instead of mitigating onsite, however, guided by an area plan, we can do offsite restoration. This means allowing nature to be restored somewhere else, for instance through rewilding. There is precedent for this system: The British Labour government is introducing centralized offsetting for housing and infrastructure developments. One-off adaptations to help species can cost as little as £180,000 (€208,000), while annual costs for rewilding projects can be as low as £25,600 (€29,580). This compares favourably to the £120 million that Britain is spending on a tunnel to protect a small population of bats, at a cost of £330,000 per animal. In one German housing project, centralized mitigation would have been between 300 and 5000 times more cost-effective than site-by-site mitigation.⁹

However, the Habitats Directive makes this strategy a last-resort. Once an appropriate assessment of a proposed project is done, authorities must subject the proposal to a three-step test before consenting to adverse effects. First, it must be established that there are no alternative places where the development can take place. Second, it must be established that the development is for “imperative reasons of overriding public interest” (IROPI). Third, if those two conditions are met – which rarely happens – adequate compensatory measures must be taken. Compensation only appears, therefore, as a last resort. This is too strict. If a plan has the likely effect of improving environmental outcomes, then that should be considered by default. The 3-step test should be reformed to enable effective offsetting where possible.¹⁰

A reasonable scientific bar for evidence 

Development faces another extremely high barrier when it comes to evidence of impact on nature. The Waddenzee case created the standard of “no reasonable scientific doubt”, which means developers must prove harm won’t happen at all, not just that harm is unlikely to occur.¹¹

This standard creates a presumption against development even when the probability of harm is low. The standard effectively enshrines the requirement for certainty that no loss of habitat will occur, no matter how small or improbable. 

The “proportionality principle”, which says that the level of precaution must be in proportion to the risk, is arguably violated by the Waddenzee standard of “no reasonable scientific doubt.”¹²

Restrictions should be balanced with the magnitude and likelihood of damage. The evidentiary bar can be high but to be proportionate, it must reflect a reasonable standard of evidence. The standard can be that development is permissible when harm is unlikely on the balance of probabilities. A change to the Habitats Directive would achieve this.¹³

A reasonable test for alternatives and public interest

As mentioned above, the three-step test prevents offsite compensation from being considered until the last possible stage, which rules out many projects unnecessarily. The first two steps are themselves too stringent.

At the first step, it’s not clear what counts as an “alternative” site for development. Does the alternative land have to be owned by the developer? How long might this alternative take? And, crucially, at what cost? No matter who the developer is, money is a constraint, as is time. If there is no practical feasibility component to this ‘alternative’ test, then it will be difficult for projects to clear it.

At the second step, the question is what counts as an imperative reason of overriding public interest. The original directive is quite broad, explicitly mentioning reasons “of a social or economic nature”. It’s difficult to argue that housing in high-demand areas, or infrastructure enabling such housing, does not pass this test. In regions where there is a demonstrable housing shortage, the development of new housing is clearly in the overriding public interest and the directive should make this clear. (One option for assessing shortages would be an income to house price ratio.) An update to Article 6(4) would address these two issues. Like for renewable energy development, the EU can deem these classes of infrastructure to automatically satisfy the test.¹⁴

It’s not just the Birds and Habitats Directives that make it hard to renew Europe’s built environment. Some of the same issues are present in the Strategic Environmental Assessment (SEA) and Environmental Impact Assessment (EIA) Directives.

Providing clarity to builders about when and what impacts to assess

As we’ve seen, it’s better to create as much clarity at the plan level as possible. That way individual projects have a clear sense of what is in and out of bounds.

SEAs are designed to find out the impact that an overall plan for a specific area is expected to have on the environment. By contrast, EIAs are designed to find out the impact that a specific project is likely to have on the environment, and deliver this in report form to the planning authority.

However, the current system does not achieve this. As currently designed, EIAs are too open-ended, incentivised to cover every possible impact, no matter how insignificant. They keep getting bigger and bigger to avoid risk. The 2005 environmental impact statement for the Dublin City Centre light railway extension was 176 pages. Twenty years later, the report for the extension to Finglas is over 7,000 pages: the non-technical summary alone was 121 pages.

Following the principle outlined above – that maximum clarity should be created at the plan level – SEAs should do as much of the work as possible, and EIAs should avoid duplication of work that SEAs have already done. In other words, SEAs should limit the scope of EIAs by settling strategic questions, and then EIAs can clear up the rest. Legal risk, which incentivised developers to endlessly pour time and money into EIAs, is removed. Reform of SEA Article 3 would achieve this, along with some other changes.¹⁵

Another issue with the EIA Directive is that it is too ambiguous about when EIAs are required. It mandates that any of the listed kinds of project (not just ones likely to affect important sites and species, like the Birds and Habitats Directive) which is likely to have significant environmental effects must conduct an Environmental Impact Assessment. The directive allows discretion for member states to work out the need for EIAs on a case-by-case basis, or set their own thresholds. This discretion, however, leads to uncertainty within the planning system, and legal fights over what should not be subjected to an EIA. Member states sometimes set thresholds that are surprisingly low. For waste-water treatment, the EU-wide mandatory threshold is for plants above a 150,000 population equivalent (p.e.) capacity. The threshold set in Ireland is at 10,000 p.e., 93 per cent lower.

A better approach is a ‘traffic light’ system, which integrates with the tiering reforms outlined above. Member States should set clear exclusion thresholds and mapped environmental sensitivity zones, below which and outside which projects are presumed not to require assessment. That means smaller projects, or those in non-sensitive areas, automatically get a ‘green light’ and don’t have to do an EIA. Projects where there may be impact (‘amber’) can undergo screening for an assessment, while some projects will definitely need an assessment (‘red’). Disagreements can be worked out at the plan level, and individual projects don’t have to get mired in these legal battles.¹⁶

In addition, there is a strong case for development acceleration areas, modelled on RED III renewable acceleration areas. This is built on the Draghi report’s recommendation to greenlight projects within strategically assessed areas. The SEA for this zone would assess impact, so it wouldn’t need to be repeated for individual projects.¹⁷

Restore the original purpose of Directives

There are two further areas in which EIAs have spiralled away from their original purpose, and now serve to slow development unnecessarily. The first is scope creep, where ‘indirect’ effects and other matters which are not a first order impact of the development have to be part of the impact assessment.¹⁸ Even nature restoration projects, which by definition will help nature, are subject to these requirements. This creates burdensome bureaucracy, both for the party seeking to develop, who must carry out additional surveys at additional expense, and the planning authorities, who must review the hundreds or even thousands of pages of material that is submitted. This should be narrowed in the original directive.¹⁹

The second is materiality. If there is a flaw in an EIA, no matter how small, it can be used by the courts to quash planning permission for the entire project. Permission for 221 homes in Dublin was quashed because there existed two versions of report about the potential for impact on bats, one of which was available to the public online (which committed the developer to seeking a separate licence relating to bats) and another only available in hard copy at the offices of the decision-maker (which did not); the court quashed the permission to preserve an opportunity for the public to make submissions on whether the planning permission should include a condition obliging the developer to seek that separate licence.²⁰ Given the need to renew our built environment, this is causing undue delays and uncertainty.²¹

Combat inflation by reforming the EPBD

The Energy Performance of Buildings Directive (EPBD), in its various iterations, has a laudable goal. Buildings account for 36 per cent of carbon emissions in Europe, and the EPBD seeks to combat this. If it worked, that would be a significant emission reduction.

There are doubts about the extent to which the directive is, in fact, reducing emissions, but let’s focus on another effect it’s having: inflation. In Germany it has raised costs by about 3 per cent. In Greece it has raised costs by between €4,300 and €7,400 per home. In the UK, which also signed up to similar rules, it was predicted to raise the costs of construction by around £4,000, but probably increased it more. In Ireland, it officially was predicted to raise costs by 0.7 to 4.2 per cent over and above the previous year’s construction costs based on the former standards (and the effect may have been even higher, according to Progress Ireland analysis).

None of these figures, as pointed out by one German report, take into account the knock-on costs of this inflation: on land costs, on financing costs, and labour costs (the introduction of new methods and equipment require the deployment of specialised skills and materials). 

In Ireland, construction inflation has been so high that only the top three income deciles can afford to cover the costs of developing an apartment. Progress Ireland estimated that construction inflation due to the transposition of the requirements of the EPBD in Ireland reduced housing supply by 10,000 homes between 2019 and 2025. This calculation used a 2024 paper which estimates a baseline cost elasticity of about -1.9, meaning a 1 per cent increase in construction costs is associated with roughly a 1.9 per cent fall in new housing supply, on average, in the long run. 

Our model estimates that the single change in 2019 has resulted in 10,000 fewer homes that exist today than otherwise would if costs hadn’t risen as much. This doesn’t take into account the other changes to energy efficiency regulations in 2007 and 2011.

This estimate is based on the Irish government’s ex ante predictions of how much this regulation would cost. But there are at least two reasons to think even these cost predictions underestimated the number of lost homes.

The first is a report from the Society of Chartered Surveyors Ireland from the year after the 2019 regulations were brought in. Between 2019 and 2021, housebuilding costs rose by about 7 per cent. According to the SCSI, “the main reason for cost increases was compliance with the new Nearly Zero Energy Building (NZEB) standards requirements and an increase in the cost of building materials.” These sound unrelated but they are deeply intertwined. Higher energy efficiency requirements mean you need different and often higher quantities of material. They leave builders more exposed to external price shocks. 7 per cent is an upper bound for how much NZEB standards increased costs and the real number is likely smaller.²² 

If our model uses a 7 per cent increase in cost, we see that we have lost just under 25,000 homes, cumulatively, in the counterfactual scenario since 2019. That is roughly the equivalent of losing the entirety of Cork’s annual housing supply flow of about 4,500 homes a year. 

The second reason to be sceptical of the ex ante predictions comes from looking at overall construction inflation. Analysis by one of the authors of this report, along with economist Ronan Lyons, shows that nominal construction costs have increased by a factor of almost 40 since 1971. In comparison, general inflation increased the price level by about 14 times. In real terms it costs 2.7 times more to build a home in 2025 than it did in 1971. 

Why would this cast doubt on the government’s predictions of cost inflation? The main reason is that our analysis shows the inflation was not evenly distributed across the 50-odd years in the series. Inflation in the 1970s was mostly undone in the 1980s as costs reversed. Costs went up a lot in the early boom years. But between 2020 and 2025, they rose by 40 per cent.

Another reason is that our index tracks the cost of constructing a home, not the fixed unit costs of materials or labour. There is a separate index for those things called the Housing Construction Costs Index or HCCI (in the graph above, I call this the labour and material index). You can see here that construction cost inflation was higher than both the Consumer Price Index (CPI)  and the HCCI. This suggests that there was something over and above general inflation, or inflation in material and labour costs, which is driving the high costs of construction inflation. 

While American construction inflation and levels may be similar to European ones, American workers are much richer. The authors of Silicon Continent recently pointed out that “the median American earns 30 per cent more than the median Dutchman, about 31 per cent more than the median German, and about 52 per cent more than the median Frenchman.” Despite having bigger homes, American housing is often cheaper than European counterparts (and is of comparable quality). 

Competitiveness in Europe is inadvertently undermined by Directives which contribute to high levels of inflation. To get inflation down, the EPBD could be reassessed. Article 6(1) of the EPBD could be revised to reestablish subsidiarity on building regulations, ensuring member states are given the freedom to reduce inflation in the building sector. Article 11(2) of the EPBD, which obligates Member States to set maximum energy usage thresholds, could be revised to reduce energy thresholds to a less inflationary level, such as those set in the 2010 recast of the EPBD. 

Appendix of directive changes

1.  Combes et al., (2008), using French data. ↩︎
2. Duranton and Puga (2023). ↩︎
3. Cousin et al., (2025) 44f. ↩︎
4. Page 56 https://commission.europa.eu/system/files/2017-01/swd-2016-472-final_en.pdf. ↩︎
5. Article 6(1), and Birds Directive Article 4. ↩︎
6. Article 6(3), “in combination with other plans and projects.” ↩︎
7. Habitats Directive 6(1) and Birds Directive Article 4. ↩︎
8. Habitats Directive 6(3). ↩︎
9. Silicon Continent’s calculations, based on the population of toads, estimates of the cost of the toad habitat, and Irish projects that protect the same species. ↩︎
10. Habitats Directive 6(4): Clarify the standard for “reasonable alternatives”, clarify IROPI standard, and allow offsetting. Amend 16(1)(c) to reflect the clarified IROPI standard in 6(4). ↩︎
11. Case C-127/02. ↩︎
12. “Shellfish for Fishermen or for Birds? Article 6 Habitats Directive and the Precautionary Principle” (2005) 17(2) Journal of Environmental Law 265; Communication from the EU Commission on the precautionary principle. ↩︎
13. Article 6(3). ↩︎
14. For example, Article 16f of the RED III Directive (2023/2413) on renewable energy development. ↩︎
15. SEA: New Article 3(7): would allow SEAs to determine the content of EIAs. EIA Article 5(1): Aligns developer obligations with new tiering system. EIA Article 5(2). Authority shall issue a binding opinion on the scope of an EIA, on request from the developer. SEA Directive (new) Article 5. Operationalise the EIA changes in the SEA Directive. ↩︎
16. Reforms to EIA Article 4(3) can set mandatory thresholds below which screening determinations are not required. Without this, there is no such thing as a “green lit” project and no traffic light system. ↩︎
17. SEA new Article 4. ↩︎
18. For example, the Finch Minority decision. ↩︎
19. EIA Article 3. ↩︎
20. Southwood Park Residents Association v. An Bord Pleanála [2019] IEHC 504. ↩︎
21. EIA Article 11(1). ↩︎
22. ↩︎