The Price of Permission

The deep dives

This page is a summary. Each topic below has its own exhaustive deep-dive covering every statute, every cost figure, every case study we found.

The full picture

It costs about €537,000 to deliver a two-bedroom apartment in Dublin. That's the SCSI/Mitchell McDermott figure for 2025, based on analysis of 10,700+ apartments in real schemes. It includes hard construction, VAT, developer margin, finance, levies, and land.

About a quarter of that — roughly €133,000 — is attributable to planning and regulatory requirements. That's one way to look at it. But it's not the only way, and it's not the most complete.

Here's where the full €537,000 goes:

Construction is the biggest slice — nearly half. "Soft costs" (finance, professional fees, levies, insurance, legal, risk contingency) account for another 42%. Land is about 11%, down from ~15% at peak.

A three-bed semi in the Greater Dublin Area is somewhat cheaper: €461,000 total (SCSI 2023). But the proportional breakdown is similar.

There are at least three useful ways to ask "why does this cost so much?" They give different answers.

The regulatory cost stack

The most common analysis: add up every planning and regulatory cost component. For a 3-bed house in the GDA, this totals about €115,000 at midpoint (range €63k–139k), or roughly 25% of the sale price. For a Dublin apartment, about €133,000 (range €63k–163k).

Midpoint estimates for a 3-bed semi, GDA. Source: cross-referenced from SCSI 2023, statutory schedules, and industry data. See Building Regs, Contributions, and Fire & BCAR for full breakdowns.

This is a useful breakdown but it has a built-in assumption: it imagines a world with zero regulation and measures the gap. In practice, competent builders would voluntarily adopt many safety measures. The marginal regulatory cost — above what good builders would do anyway — is lower than these figures. For example, structural engineers and fire safety consultants would be engaged on any sizeable project even without BCAR; the true marginal cost of BCAR is the additional oversight cost, not the full professional fee.

Two specific double-counting risks are worth flagging. First, professional fees (€17,600) and BCAR certification (€9,500) partly overlap: a significant share of professional fees are incurred because of BCAR certification requirements. Second, the time delay cost (€14,500) estimates finance and holding costs from planning delays, but some of this finance cost is already embedded in the "Finance €~43k" slice of the €537,000 breakdown — if using both frameworks together, this cost may be counted twice. The chart shows gross regulatory costs, not net costs relative to an unregulated counterfactual.

It also treats regulation as purely additive: each requirement adds a cost, and you sum them. This misses two things the other lenses capture.

The land question

An acre of agricultural land in Ireland sells for about €10,000. Rezone it for housing and it sells for €231,000. In Dublin, the gap is even wider: €24,000 to €643,000.

Source: CSO Agricultural Land Prices 2024, CSO Residentially Zoned Land Prices 2024. National medians. Full analysis: Land & Zoning.

That's a 23x multiplier nationally. 27x in Dublin. The question is: who created that value, and who captures it?

The value isn't created by the landowner. It's created by public decisions — zoning, infrastructure investment, transport, schools — and by the general growth of the city. The 1973 Kenny Report recommended allowing compulsory purchase at existing use value plus 25%. It was never implemented. If it had been, the State could acquire development land at €12,500/acre instead of €643,000.

How Ireland captures (or doesn't) land value

Compare this to Germany, where municipalities can compulsorily acquire land at existing use value under the Federal Building Code, or the Netherlands, where municipalities actively buy, service, and sell development land. See Land & Zoning for full detail.

Land's share of total housing cost has actually been falling in Dublin — from ~€80k/unit in 2021 to ~€59–65k in 2025. This doesn't mean land got cheaper. It means building costs rose so fast that land's relative share shrank. Which brings us to the third lens.

Supply restriction and the viability gap

The cost-stack analysis treats regulation as something that adds to each house built. But regulation also determines how many houses get built at all. These are fundamentally different effects.

Channel 1: Direct costs. Building standards, levies, fees, and delays add identifiable euros to each unit. You can sum them up. They're real.

Channel 2: Supply restriction. Planning constraints limit the total quantity of housing. When supply can't respond to demand, prices rise beyond what cost additions alone would explain. This inflates the price of all housing — existing stock included — not just new builds.

Ronan Lyons (TCD, 2024) provides the most important Irish-specific finding: housing supply is about twice as responsive to cost reductions as to price increases. The estimated elasticity of supply to costs is −1.9, versus +0.9 to prices. A 19% cost reduction would increase supply by 50%. A 57% price increase would achieve the same.

The viability gap

When the all-in development cost exceeds what the market will pay, the residual land value goes negative. The development simply doesn't happen. This is the situation for most apartment types in Dublin without state intervention.

The government's response: a VAT cut (13.5% to 9% on apartment sales), Crói Conaithe subsidies of up to €144,000/unit, development levy waivers, and design standard relaxations. After these interventions, the SCSI found 5 of 6 apartment types became viable, versus 2 of 6 without them. See Design & Density for full SCSI data.

But here's the circularity: Dublin needs high-density apartments to meet its 50,000+ annual target, yet those apartments can only be built with state subsidies, and the resulting prices are affordable only to the top 20–40% of earners. The regulation didn't just "add €133,000 to the apartment" — it contributed to a situation where many apartments can't be built at any price the market supports, creating ongoing scarcity that inflates prices for all housing.

The numbers that don't appear in the cost stack

The Central Bank of Ireland found that while house prices in 2020 were roughly 20% below their 2007 peak, build costs were 70–90% higher than in 2007. The cost-price scissors have widened dramatically. This is a cost-side crisis, not just a demand-side one.

Dublin has 39,000 uncommenced apartment permissions. Those are schemes where someone went through the entire planning process and then didn't build — because the numbers didn't work. Every uncommenced permission represents housing that the planning system approved but the cost environment prevented.

The time tax

Site acquisition to first sale takes 2–8 years. Each year of planning delay adds about €27,000 per unit. Full timeline analysis: Planning Law — Critical Path.

The system front-loads risk and back-loads certainty. Developers commit capital — land, professional fees, environmental surveys — before receiving any planning certainty. Appeals and judicial review mean even a grant doesn't definitively mean you can build. This deters smaller developers and favours larger, better-capitalised firms.

Key bottleneck: bat surveys only run May–September. Miss the window, wait 8 months. Wintering bird surveys run October–March. These are hard seasonal constraints. A Further Information request (issued on 30–50%+ of non-trivial applications) adds 2–9 months. An appeal to An Coimisiún Pleanála averaged 57 weeks in 2023, though this has dropped to under 10 weeks after a 44% staffing increase. More in Environment & Heritage.

What regulation buys

Regulations cost money. They also prevent specific, documented harms. The benefits aren't hypothetical.

Fire safety

Fire death rate halved since the late 1990s (~3.9 per million today). Apartments are just 9% of fatal fires despite multi-occupancy risk. 44% of 2024 fire deaths were in homes without smoke alarms — risk concentrates in older, unregulated stock. Measures cost €6,600–25,000 per apartment but directly prevent the conditions that produced Priory Hall (€238k/unit remediation). Full detail: Fire Safety & BCAR.

Energy

99% of new homes achieve A-rated BER. nZEB saves ~€800/year versus pre-2005 stock. Emissions 90% lower (1.1 vs 10.8 tonnes CO₂/year). The ~€4,275 uplift per dwelling pays for itself within 2–5 years. See Building Regs.

Social housing

Part V: ~25,600 units since 2002 (25,631 to Q3 2025 per Dept. of Housing official dataset). Captures publicly-created land value (discount €30k–200k+ per unit). England targets 30–40% affordable but routinely delivers less after "viability" arguments; Ireland's fixed 20% is reliably applied. More: Contributions & Part V.

Orderly development

Core strategy rules cut speculative overzoning from 44,000 to ~12,000 hectares. OPR blocked 288 flood-risk sites. RZLT (3% annually) targets land hoarding — collected €46m in year one versus near-zero under the old Vacant Site Levy. Full story: Land & Zoning.

Democratic participation

Appeals catch genuine errors (28–30% overturn rate). When Ireland removed appeal rights (SHD, 2017–22), JR rates spiked to 22.8% with an 86% State loss rate. The "fast-track" proved slower. Restoring appeals under LRD cut JR to 11.3%. More: Planning Law.

What happens without it

Ireland doesn't need to speculate about weak regulation. It tried it.

The contested debate

Reasonable people disagree about how much of the housing cost problem is caused by regulation versus other factors. The academic literature is genuinely divided.

The supply-restriction school

Glaeser & Gyourko (Harvard/Wharton) defined the "regulatory tax" as the gap between house prices and physical construction costs. In many US cities, this implicit tax exceeds 30–50% of house value. Hilber & Vermeulen (LSE, 2016) found that if South East England had the regulatory restrictiveness of the North East, house prices would have been roughly 25% lower. Hsieh & Moretti (2019) estimated housing constraints in high-productivity US cities lowered aggregate GDP growth by 36% from 1964–2009 by preventing people from moving to where they'd be most productive.

Note on the 36% figure: This is not the number the original 2019 paper published. The published paper's imperfect-mobility headline was 3.7%; the perfect-mobility estimate was 8.9%. In 2021, Bryan Caplan identified arithmetic errors in the paper that Hsieh and Moretti acknowledged; Caplan's corrected calculation produced the 36% figure, meaning the post-correction effect is larger than the published one. A separate 2023 formal published comment by Brian Greaney (AEJ: Macroeconomics, doi:10.1257/mac.20230141) identified coding errors in the replication code that, if accepted, reverse the headline result to approximately 0.02% or negative. The authors have disputed Greaney's interpretation and the paper is under active formal challenge in its own journal. The debate is unresolved.

The critique

Cameron Murray (University of Sydney, 2021) challenged the Glaeser-Gyourko method, showing it finds a high "regulatory tax" even in the absence of any regulatory constraints — including in colonial Australian land sales and ancient Mesopotamian data. The gap between price and marginal construction cost reflects location premiums and diminishing returns, not necessarily regulation. Murray's published critique argues the method has "little or no scientific merit" for identifying planning effects.

Louie, Mondragon & Wieland (NBER, 2025) found that estimated housing supply constraints are "unimportant in explaining differences in rising house prices among U.S. cities." This is contested but represents a serious challenge to the standard framework.

The credit school

Ian Mulheirn argued the UK housing crisis is primarily driven by cheap credit, not supply shortages — England had 1.1 million more homes than households. Meeting the 300,000/year target would only cut prices ~10% over 20 years. Centre for Cities rebutted that the supply crisis is geographically concentrated: national surplus masks local shortage in economically successful cities.

The Irish evidence

Lyons (TCD, 2024) found supply is more responsive to costs than prices, suggesting the cost environment is the binding constraint in Ireland specifically. The ESRI found that increased prices do not lead to increased supply in Dublin or commuter counties — suggesting constraints prevent the normal market response. The IMF (2023) estimated Ireland would need to double annual completions to 60,000 to reach the EU average housing stock per capita by 2030.

Norris & Shiels (UCD, 2007) found that Ireland's then-laissez-faire planning system hadn't constrained national output (Ireland's building rate was among the highest in the EU). But it had failed to direct supply to where affordability problems were worst. The regulatory environment has tightened significantly since.

What this means

The cost-stack approach (Lens 1) captures direct costs but misses the supply-restriction effect. The supply-restriction approach (Lens 3) captures the broader price effect but is methodologically contested. The land value approach (Lens 2) highlights who captures publicly-created value. No single lens gives the complete picture. All three are needed.

How Ireland compares

A 2024 TCD/SCSI study priced an identical 39-apartment, 7-storey development across 10 European cities using standardised methodology (ICMS3).

Source: TCD/SCSI 2024, ICMS3 methodology. 10-city average: €2,057/m². Dublin exceeds average by €300+/m². Full analysis: Design & Density.

Methodology note: The study uses a standardised reference building (39 apartments, 7 storeys, designed in Switzerland in 2011) repriced using local labour and material rates — a “travelling box” approach. This holds design constant across cities. Key limitations: (1) the Swiss design “is architecturally very different to what we would deliver in Ireland” (SCSI), so the comparison prices an Irish-atypical building in Dublin; (2) if VAT is excluded, Dublin falls from 2nd to 5th place; (3) costs are benchmarked to Q1 2020 (pre-inflation spike); (4) soft costs (land, finance, fees, levies) are excluded. The study measures hard construction cost differentials, not total development cost differentials.

Dublin's premium isn't in structural work (actually slightly below average). It's in services and equipment (heating, power, lifts) and non-structural works (floors, windows, carpentry). Irish apartments are also delivered "turnkey" (fully finished) versus unfurnished in most European cities, use premium aluclad windows versus uPVC, and until recently had above-minimum floor area requirements that added €32,000–38,000 per unit.

Planning system structure

Ireland builds about 7 homes per 1,000 population annually — actually high by European standards. The problem is the accumulated deficit: ~36,284 completions in 2025 against a target of 50,500–55,000. The IMF puts the need at 60,000/year to reach the EU average housing stock per capita by 2030.

What's changing

Ireland's planning system is undergoing its most significant overhaul since 2000. Full detail in Planning Law.

Planning and Development Act 2024: mandatory decision timelines, restructured appeals (An Coimisiún Pleanála), JR reform (partial quashing, cost caps), Urban Development Zones, National Planning Statements replacing Ministerial Guidelines.

Cost interventions: VAT cut to 9% on apartment sales (Oct 2025–Dec 2030), Crói Conaithe subsidies up to €144k/unit, apartment design relaxations (Jul 2025) with ministerial savings estimate of €50k–100k/unit (unverified). Result: 5 of 6 apartment types now viable vs 2 of 6 previously.

Land reform: RZLT (3% annually, collected €46m in year one), Land Value Sharing (25% charge on zoning uplift, enacted in PDA 2024). These don't go as far as the Kenny Report (1973) which would have captured ~97% of uplift, but they represent a step-change from the old regime where the Vacant Site Levy collected 1.37% of amounts due. See Land & Zoning.

Appeals: Average decision time dropped from 57 weeks (2023) to under 10 weeks (Q1 2025) after a 44% staffing increase. Apartment planning permissions jumped 50% in Q3 2025. Whether these trends hold remains to be seen.