The empirical evidence points to a system of structured extraction operating across two layers, sustained by a capture dynamic that has systematically closed off the structural responses required to resolve it. The Section 8 case study argues for changing the framework every developer operates within, not for punishing Glenveagh. Policy must address all three dimensions: the landowner windfall, the developer margin, and the lobbying architecture that has protected both.
The First Home Scheme (shared equity) and Help-to-Buy (tax rebates) inject taxpayer cash to help buyers clear elevated prices. They don’t reduce the physical cost of construction, compress developer margins, or return the rezoning windfall to the public. The Eurostat per-capita data shows Ireland extracting €199 per person in F41.1 surplus against an EU average of €49. Demand-side subsidies fund the €150 gap; they don’t close it. As the UK Persimmon experience demonstrates, they increase it: each extension of Help-to-Buy has been followed by extension of the margins it was claimed would make construction viable, not by supply increases that would normalise them. The Blackrock Villas case makes this precise: €135,000 per unit in Croí Cónaithe supply-side subsidy, stacked on top of Help-to-Buy and the First Home Scheme demand-side support, on a development by a company that returned over €445 million to shareholders in the same period.
The primary instrument for addressing the landowner rezoning windfall is a windfall tax — levied at the moment planning permission converts agricultural land to residential zoned value — set at 75–80% of the uplift generated by the rezoning. CSO data (RZLPA01 & ARA02) quantifies what is at stake: a €619k per-acre differential in Dublin (€643k residential zoned vs €24k agricultural, 2024) and a €222k differential nationally. A 75% levy on the Dublin differential captures approximately €464k per acre for the public purse, returning to the state the majority of the value its own planning system created. This is the instrument recommended as the primary mechanism because it operates at the point of value creation rather than the point of transaction, and because it has the strongest legal basis in existing Irish tax law (applying to existing windfall tax frameworks rather than requiring new constitutional property rights interventions).
A standard objection to hard land value capture is the seller strike problem: if landowners know that rezoning will trigger a large tax, they may simply refuse to sell — holding land off the market and creating a supply shock that worsens the crisis it is meant to address. This objection is legitimate and the policy design must address it directly. Three mechanisms work together to prevent it.
The Transitioning Window: A 24-month period following the announcement of the LVC regime in which landowners may sell to the state, an Approved Housing Body, or a licensed developer at a moderate premium (e.g. agricultural value plus 50%) before the full tax rate activates. This is the carrot that converts the incentive structure: sell now at a fair-but-not-windfall price, or pay most of the windfall to the state later. The window is time-limited and non-renewable, creating urgency without coercion.
The Compounding Vacancy Levy: A progressive annual levy on zoned land that remains undeveloped, escalating over time. Ireland’s existing RZLT was designed to perform this function but was lobbied into near-irrelevance before implementation. A properly structured vacancy levy makes land hoarding a cash-negative strategy rather than a value-preservation strategy. Combined with the LVC, it closes the two main escape routes: the landowner can’t wait out the LVC by not selling (the vacancy levy eats the value) and can’t avoid the LVC by selling into a non-development use (the LVC is triggered by rezoning, not by sale).
CPO at Agricultural Value: For strategic sites where the windfall tax and vacancy levy are insufficient to bring land to market (particularly large sites with a single reluctant owner) the state retains Compulsory Purchase Order powers valued at certified agricultural use rather than development potential. The constitutional basis for this exists in Irish law and has precedent in multiple EU member states. It is the backstop, not the primary instrument, but its credible availability changes the negotiating dynamic for the transitioning window.
The Marina Depot case makes the stakes of this instrument concrete. Had an LVC at 75% been in place at the point of rezoning, the approximately €510 million in planning-permission value created at that site would have generated approximately €382.5 million for the public purse. The LDA’s €150 million forward purchase would have been available to deploy on land already largely publicly captured — producing either more units at the same cost, or the same units at approximately one-third of the current cost.
The international precedent base for this intervention is broader and less radical than Irish debate assumes. Singapore’s Land Acquisition Act 1966 empowered the state to acquire land at existing-use value, the precise mechanism of the CPO backstop above. On that foundation the Singaporean state came to control approximately 90% of the country’s land, the Housing and Development Board houses the large majority of the population, and homeownership runs at roughly 90%, in the most aggressively pro-business, low-tax economy in Asia. The proposition that hard land value capture is incompatible with a dynamic market economy is refuted by the single most market-oriented developmental state of the twentieth century, which made it the foundation of its housing system. More broadly, the OECD and Lincoln Institute’s Global Compendium of Land Value Capture (2022) catalogues LVC instruments in active use across approximately 60 countries: this is mainstream international fiscal practice, not a fringe proposal.
LVC design precedents: Denmark (municipal land ownership model); Germany (Planungswertabschöpfung, planning value skimming, implemented in several Länder); UK (Community Infrastructure Levy and proposed Infrastructure Levy reform, 2023 Levelling Up and Regeneration Act); Singapore (Land Acquisition Act 1966; HDB). OECD/Lincoln Institute of Land Policy (2022), Global Compendium of Land Value Capture Policies. Transitioning window mechanism: Housing Commission Report May 2024. RZLT failure: Irish Times FOI reporting, 2022–2024.
Statutory price caps limiting development-ready land’s maximum commercial exchange value to a defined multiple (2–3×) of its certified baseline agricultural value at first rezoning application is a stronger and more direct intervention than the windfall tax, but also more legally and politically contested. Unlike the windfall tax, which allows the transaction to proceed at market price and captures most of the uplift in tax, the price cap prevents the uplift from being priced into the transaction at all. This is structurally cleaner; it prevents the inflated land cost from entering the development proforma and creating the ‘viability gap’ in the first place, but it raises more acute constitutional questions about property rights and is more vulnerable to the seller strike problem the windfall tax is designed to avoid.
The recommended sequencing is therefore: implement the phased LVC windfall tax (Intervention 1) as the primary instrument, with the price cap as a secondary backstop for strategic State land acquisitions and for situations where the transitioning window has elapsed without transaction. The price cap is most effective and least legally contested when applied to land being acquired by the State or Approved Housing Bodies under CPO, where the courts have long accepted that compensation is calculated at use value rather than development value. Extending that principle to all transactions involving zoned land (rather than only CPO) is the long-term goal, but can be pursued through legislative amendment over a longer horizon as the LVC establishes the principle of public capture of planning uplift.
Interventions 1 and 2 are routinely framed in Irish debate as novel, radical, or imported. They are none of these. In 1973, the Committee on the Price of Building Land — chaired by Mr Justice John Kenny of the High Court, and established by the government because building land prices were rising faster than the cost of housing — recommended that local authorities be empowered to acquire potential development land at its existing use value plus 25%. That is Intervention 2, proposed by a serving High Court judge, in an official report commissioned by the Irish state, more than fifty years ago. The Kenny Report’s majority concluded that the increase in land value arising from community decisions (zoning, servicing, planning) belonged in substantial part to the community that created it.
The Kenny Report was never implemented. Successive governments cited constitutional doubts, and the recommendation was shelved through the very decades in which the rezoning windfalls documented in Section 5, and the political corruption around rezoning later documented by the Mahon Tribunal — accumulated. The fifty-year gap between the state’s own diagnosis and the present day isn’t evidence that the proposal is unworkable. It is the longest-running example of the pattern this report documents in Section 7: an analytically sound intervention, formally recommended through official channels, defeated not by counter-argument but by the organised interest of those who benefit from its absence. Every euro of the €619,000-per-acre Dublin rezoning differential quantified in Section 5 is a euro the Kenny Report told the Irish state how to capture before most of the current Dáil was born.
Nor did the official record fall silent after 1973. The Commission on Taxation (2009) concluded there was a strong economic rationale for a Site Value Tax. The National Economic and Social Council’s Report 145 (2018) described the housing system as ‘speculative, volatile and expensive’, diagnosed the land system as dysfunctional, and recommended active land management with land value capture on the continental European model. NESC Report 150 (2020) recommended equipping the LDA with compulsory acquisition and land value capture powers and removing the barriers to a Site Value Tax. The Housing Commission (2024) recommended State land value sharing. Five official recommendations of land value capture instruments, from five different state-constituted bodies, across five decades. None implemented. The Kenny Report isn’t an exception in the Irish official record; it’s the founding instance of a pattern: the state repeatedly commissions the analysis, receives the recommendation, and declines the implementation. What has never existed isn’t the official case for capturing the planning rent. What has never existed is a government willing to act on its own advice against the interest group organised to oppose it.
The standard objection — that Kenny-type measures would violate the private property protections of Articles 40.3.2 and 43 of the Constitution — has been formally examined twice, and the examination doesn’t support the objection. First, in the Article 26 reference on the Planning and Development Bill 1999, the Supreme Court upheld Part V, the provision requiring developers to transfer up to 20% of land for social and affordable housing at existing use value rather than development value — as a constitutionally permissible delimitation of property rights in the interests of the common good. The Court accepted, unanimously, that the state may require landowners to forgo development uplift without compensation at full market value, where the objective is socially proportionate. That is the constitutional core of Interventions 1 and 2, and it has already survived the highest form of constitutional scrutiny available in Irish law.
Second, the All-Party Oireachtas Committee on the Constitution examined the question directly in its Ninth Progress Report (Private Property (2004)) and concluded that a Kenny Report-style scheme of acquiring development land at existing use value plus a margin would be likely to withstand constitutional challenge, noting that Article 43 itself expressly subordinates property rights to ‘the exigencies of the common good.’ The constitutional objection to land value capture in Ireland is therefore not a settled legal barrier. It is an untested assertion, contradicted by the leading Supreme Court authority and by the Oireachtas’s own constitutional review, that has been allowed to function as a veto for five decades because it was never in the interest of those invoking it to test it.
Sources: Committee on the Price of Building Land (Kenny Report), Stationery Office, 1973. Commission on Taxation Report 2009. NESC Council Report 145, Urban Development Land, Housing and Infrastructure: Fixing Ireland’s Broken System, 2018. NESC Council Report 150, Housing Policy: Actions to Deliver Change, 2020. Housing Commission Report, May 2024. In re Article 26 and the Planning and Development Bill 1999 [2000] 2 IR 321. All-Party Oireachtas Committee on the Constitution, Ninth Progress Report: Private Property, 2004. Mahon Tribunal: Final Report of the Tribunal of Inquiry into Certain Planning Matters and Payments, 2012.
Bypassing the NACE 411 development layer entirely. State agencies, the LDA, local authority housing sections, approved housing bodies — acquire raw land at utility cost, manage planning through democratic urban frameworks, and contract NACE 412 physical builders via transparent, fixed-fee construction tenders. This retains the legitimate builder margin (38–42% GOS/GVA on CSO data; competitive with Eurostat peers) while completely eliminating the 81% rent-seeking developer layer. Denmark’s model is the existence proof: their F41.1 (developer) GOS/GVA of −232.7% reflects the absence of a private planning-uplift capture layer, while their F41.2 (builder) sub-sector operates profitably at a market-competitive GOS/GVA of approximately 10.9%. Vienna provides a second, century-long existence proof: continuous municipal land banking since the 1920s, with the majority of residents today living in social or cost-rent housing, in a city that perennially tops global liveability rankings. Public delivery at scale isn’t an emergency improvisation, in Vienna it is the permanent architecture of a functioning housing market. The objection that the state lacks delivery capacity is correct as a description of current conditions; it isn’t a justification for those conditions. Rebuilding direct procurement infrastructure is a policy choice, not an impossibility.
The standard objection to public direct procurement is that the state is a less efficient delivery mechanism than the private sector: higher overhead, slower procurement, less commercial incentive. This objection is correct in a normally functioning market. It is arithmetically wrong in the Irish market at its current level of distortion, and the evidence for that is directly computable from the figures confirmed in this report.
The Marina Depot case study provides the most precise test. The LDA (the state’s own housing delivery agency) paid €445,133 per apartment in its February 2025 forward purchase from Glenveagh. This price reflects the post-planning-permission market value of each unit. The land cost embedded in that price, at the point of Glenveagh’s 2018 acquisition, was approximately €12,733 per unit. The planning uplift between those two figures (approximately €432,400 per unit) was created by the state’s own planning system and captured entirely by Glenveagh.
Had the LDA instead acquired the Marina Depot site directly at agricultural land prices (using CPO at agricultural value under a hard land value capture regime), the cost structure would be fundamentally different. Using CSO ARA02 Dublin agricultural land prices (€24,125 per acre in 2024, approximately €2,400 per apartment at typical docklands density of 10 units per acre), combined with market-rate physical construction costs (industry estimates of €250,000–€350,000 per apartment, using €300,000 as a conservative midpoint) and professional fees and infrastructure at approximately 18% of build cost, the total public direct cost per apartment is approximately €356,000.
| Cost Component | Current Model (LDA forward purchase) | Public Direct (agricultural land + build) |
|---|---|---|
| Land cost per unit | €432,400 (post-planning uplift, embedded in LDA price) | €2,400 (agricultural value, CSO ARA02 Dublin 2024) |
| Physical construction per unit | Included in LDA price | ~€300,000 (market rate, NACE 412 tender) |
| Professional fees & infrastructure | Included in LDA price | ~€54,000 (18% of build cost) |
| Developer margin (NACE 411) | ~€119,000 implied (Glenveagh group margin) | €0 (eliminated by bypassing NACE 411) |
| State supply-side subsidy (Croí Cónaithe) | €135,000 (additional to LDA price) | €0 (not required) |
| VAT saving to developer (Budget 2026) | ~€16,500 (additional) | €0 (not applicable) |
| TOTAL state exposure per unit | ~€596,633 (LDA price + Croí Cónaithe + VAT) | ~€356,400 (all-in public direct cost) |
Sources: LDA purchase price — Irish Times, 24 February 2025. Land cost per unit — derived from reported Glenveagh NAMA acquisition price (€15m, Irish Examiner) divided by 1,178 units. Agricultural land price — CSO ARA02, Dublin, 2024. Physical build cost — industry range €250k–€350k per apartment; midpoint used. Professional fees — standard industry ratio. Croí Cónaithe — Irish Examiner, July 2024. VAT saving — Finance Act 2025 (4.5pp reduction midpoint applied to avg Blackrock price). Developer margin implied from Glenveagh group 2025 gross margin of 21.4% on revenue.
The saving under public direct procurement — approximately €88,700 per apartment on a Dublin forward-purchase equivalent — represents a 20% reduction in state cost per unit. Across the Housing Commission’s identified 256,000-unit deficit, that differential implies a potential saving of approximately €22.7 billion over the course of clearing the backlog. This is more than two-thirds of Housing for All’s entire projected €36 billion budget, available through structural change rather than additional expenditure.
The critical insight for the efficiency objection is the crossover calculation. For public direct procurement to cost more than the current LDA forward-purchase model, the public sector would need to be approximately 25% less efficient than a private NACE 412 contractor working at market rates. A 25% inefficiency premium on €356,400 produces a public cost of approximately €445,500 per unit — essentially identical to the LDA’s current forward-purchase price. In other words: the state could be one-quarter less efficient than a competitive private builder (in procurement, management, and delivery) and still break even with what it is currently paying Glenveagh. The planning uplift, at €432,400 per Dublin unit, is so large that it overwhelms any realistic estimate of public sector overhead.
The efficiency argument, important as it is, understates the case for public direct procurement. Even if public delivery weren’t cheaper in the distorted Irish market (even if it cost the same or marginally more per unit) there would be an independent strategic argument for maintaining and rebuilding it. Emergency arithmetic is only half the case for public construction capacity. The other half: it’s the structural mechanism that stops the emergency recurring.
A state with no public housing delivery capacity is a state that is structurally dependent on private developers for all residential supply, which is, precisely, the situation Ireland finds itself in after three decades of local authority housing withdrawal and the dismantling of direct construction programmes that accompanied it. That dependency is a market structure problem, not just a housing policy one. When the state has no credible outside option (when it can’t build directly) it has no negotiating leverage over the private sector. Every subsidy scheme, every viability gap payment, every LDA forward purchase is negotiated from a position of structural weakness: the state must make the deal work for the developer, because the alternative is no homes at all.
A state that maintains genuine public direct delivery capacity — local authority construction teams, direct procurement infrastructure, state land banks managed at cost — has leverage. It can set a price ceiling on private development land by demonstrating that it will build directly if the private sector charges above a certain threshold. It can respond to market failure without being entirely dependent on the failing market. It can absorb negative demand shocks (recessions, financial crises, pandemics) without waiting for private sentiment to recover. This is the function that Irish local authority housing construction performed in the 1950s and 1960s: not as socialism but as market stabilisation infrastructure, providing the floor that private developers could not.
The argument against rebuilding this capacity (that the state is a less efficient builder) is always made in the context of a normally functioning market, where private efficiency genuinely does provide better value. It has never been seriously applied to the Irish market at its current level of distortion, where the planning-permission uplift is so large that it overwhelms any realistic measure of public sector overhead. When the market is distorted to the degree documented in this report — when the most significant input cost isn’t materials, not labour, not finance, but planning permission uplift that the state creates and gives away, the efficiency calculus reverses. In this market, even a substantially inefficient public delivery system is cheaper, because it eliminates the single largest cost component: the developer’s planning-permission rent.
The prescription is therefore not simply ‘build some public housing to fill the gap.’ It is to rebuild public construction capacity as a permanent structural feature of the Irish housing market, not as the dominant mechanism (the private sector will continue to deliver much of the output) but as the credible alternative that disciplines private pricing and provides the state with an outside option. The LDA has the mandate; it needs the staff, the land, and the political backing to exercise it. Local authorities need their construction departments restored and resourced. The 425 Approved Housing Bodies operating in the sector need access to direct procurement frameworks that allow them to build at cost without relying on NACE 411 development margins. None of this requires abandoning the private market. It requires the state to stop being entirely dependent on it.
The crossover arithmetic (25% inefficiency premium) is derived from: LDA forward-purchase price €445,133 (confirmed, Irish Times February 2025); public direct cost estimate €356,412 (derived from CSO ARA02 agricultural land, industry build cost estimates, standard professional fee ratios); crossover calculated as (€445,133 / €356,412) − 1 = 24.9%. Historical precedent for local authority construction efficiency: ESRI (2021) ‘Public and Private Housing Output in Ireland’; National Economic and Social Council (NESC) Report No. 141 (2014) ‘Financing the Social Housing Ambition’.
A targeted anti-speculation instrument that doesn’t require constitutional property rights arguments, land acquisition, or new tax frameworks: the reduction of standard residential planning permissions from their current five-year validity to 24 months, with extensions available only upon demonstration that substantial physical construction has commenced (foundations laid and verified by a local authority inspector).
The mechanism targets the site-flip speculation documented in Section 6.3 directly and precisely. Currently, planning permission functions as a financial asset that appreciates over time: a developer can obtain permission, wait for the market to move, and sell the entitled site at a substantial uplift without laying a brick. The Chivers factory site case study (€2.5m acquired, rezoned, listed at €25m) is a documented example of a transaction that generated its entire return from the planning permission itself. Under a 24-month use-it-or-lose-it regime, this strategy becomes structurally unviable. The developer must either begin construction within 24 months or the permission lapses — stripping the state-enacted value uplift entirely and returning the site to the land bank at a lower value. The incentive structure is completely reoriented: the planning permission is valuable only as a precursor to construction, not as a standalone financial instrument.
This intervention is politically less contentious than price caps or windfall taxes for three specific reasons. First, it doesn’t involve the state acquiring anything or taxing anything; it simply changes the duration and conditions of a permission the state grants. Second, it already has constitutional precedent: planning permissions have always had expiration dates, and the courts have consistently upheld the state’s right to set those conditions. Third, it directly addresses a source of frustration that has cross-party political support: the observation by the Planning Institute’s president that planners regularly receive applications from ‘speculators seeking a grant of permission to increase land values’ rather than genuine developers intending to build.
The objection — that 24 months is insufficient time for large complex developments to progress from permission to construction start — is legitimate for certain project types. A 500-unit apartment block may genuinely need longer to organise financing and procurement than a 20-unit estate. The policy design should therefore apply the 24-month standard as the default, with a formal extension mechanism for large-scale developments (over 100 units) requiring submission of an approved financing plan, a signed construction contract with a NACE 412 builder, and a verified preliminary works programme. Extensions under this mechanism are one-off and non-renewable, preserving the anti-speculation intent while accommodating genuine large-scale delivery requirements. Critically: the extension is tied to physical delivery evidence, not to market conditions or financing difficulty, which are the developer’s commercial risk to manage, not the state’s responsibility to wait for.
Planning permission duration reform: precedents in UK (Levelling Up and Regeneration Act 2023 introduced a phased approach to permission viability); Netherlands (active use conditions on zoned land); Irish Planning Institute president Conor Norton, Irish Times, October 2021 (‘speculators seeking a grant of permission to increase land values’). Constitutional basis: Planning and Development Acts 2000–2024; established jurisprudence on state authority to set terms and conditions of planning permissions.
A statutory requirement that any entity in receipt of public subsidy, state land, LDA co-investment, or planning fast-track designation must submit consolidated group-level accounts with full transfer pricing disclosures and beneficial ownership registration at parent-company level. This closes the vertical integration accounting gap that currently allows the 81% developer margin to be obscured behind the builder entity’s reported thin margins, and makes the financial case for further subsidy or facilitation verifiable rather than asserted. Any policy analysis that accepts project-level viability spreadsheets from a vertically integrated developer group, without requiring consolidated group accounts, is examining one arm of a two-armed profit extraction mechanism.
The five interventions above all operate on the first of two available routes for eliminating the planning-permission rent: capturing it for the public once it’s created. There’s a second route, demonstrated by Japan: preventing the rent from becoming scarce in the first place. Under the City Planning Act of 1968, Japan operates a nationally standardised zoning system of twelve land-use categories, additive in structure — housing is permitted by-right in nearly every zone — with floor-area and building-coverage ratios set at national level and the local objection largely removed from the process. The consequence is that Tokyo, a growing metropolis of fourteen million, adds on the order of one hundred thousand dwellings a year and has held real house prices and rents broadly flat for decades. Where permission to build is abundant rather than rationed, it commands no monopoly premium, and the developer-layer margin this report documents in NACE 411 doesn’t arise at scale.
But Tokyo is widely misread, and the misreading matters for the political-economic argument here. Abundant permission doesn’t abolish location value. Tokyo land remains among the most expensive on earth — central parcels exceed eighty million dollars an acre — because the value at stake was never the dirt but the access it confers to a dense agglomeration of jobs and amenities. Permissive zoning doesn’t make that location rent disappear; it spreads it across far more buildable floor area, so the rent per dwelling falls even as the rent per acre rises.
And the location rent that remains is, to a striking degree, captured rather than left to passive owners: Tokyo’s private railway companies are simultaneously its largest property developers, building the lines and then developing the housing, retail and station hubs whose value those lines create, and funding the network from the land-value uplift they themselves generate. This is land value capture by another route — internalised to the firm that creates the value rather than taxed by the state — and it’s a deliberate institutional structure, not the absence of one. Tokyo isn’t a deregulated free-for-all that happens to be cheap. It’s one of the most consciously engineered land-and-transport systems in the world, and it’s cheap because of that engineering, not in spite of it.
This is the decisive point against the laissez-faire reading of the housing problem. The market-liberal critique is correct that planning scarcity creates the developer rent — that observation is this report’s own, reached from the supply side. But it doesn’t follow that removing the state returns housing to a competitive equilibrium, because land is the one input competition can’t supply into abundance: its quantity is fixed and the rent attaches to location, which can’t be manufactured. An unconstrained land market therefore doesn’t disperse the rent; it concentrates it, because whoever holds well-located permitted land earns a return that funds the acquisition of more.
Both systems that actually deliver affordable housing share a feature the unconstrained model lacks: a binding constraint on the location rent. Denmark and the Netherlands capture it fiscally, through land value taxation and active public land management — Denmark’s negative F41.1 margin is the empirical signature of that capture. Tokyo dilutes it through abundance and internalises the remainder through its rail-property structure. Denmark and Tokyo aren’t the left answer and the right answer. They’re the two stable configurations, and what makes each stable is that the location rent is constrained — taxed in one, diluted and internalised in the other. The configuration that leaves the rent scarce, privately held, and publicly subsidised is neither of these. It isn’t a market outcome and it isn’t a planning outcome. It’s the unconstrained middle — and it is, precisely, the configuration Ireland operates.
Which defines the one position that’s genuinely incoherent: defending discretionary planning scarcity while opposing public capture of the rent that scarcity creates. Scarce permissions, privately captured uplift, publicly subsidised viability gaps — this isn’t a market position, because a real market would compete the rent away or see it constrained as in Tokyo; and it isn’t a planning position, because a planning system pursuing the public interest would capture the value it creates as in Denmark. It’s simply the configuration that maximises developer surplus, and it’s the configuration Ireland has assembled, one side at a time, over five decades.
Japan zoning: City Planning Act 1968 (twelve nationally standardised, additive use districts; FAR/BCR set nationally), as amended; Urban Renaissance Special Measures Act 2002. Tokyo housing supply and price stability: OECD and academic analyses of Japanese land-use regulation; Tokyo adds ~100,000 dwellings/year with broadly flat real prices. Tokyo land values and the location-rent point: comparative urban-economics literature on bid-rent and supply elasticity. Private-railway land value capture (‘rail integrated communities’): Calimente, J. (2012), Journal of Transport and Land Use 5(1); contemporary reporting on Japanese rail-property integration. The ‘two routes’ and ‘binding constraint’ framing is the author’s synthesis.
This report set out to test a single claim: that the Irish housing crisis is driven by thin margins, high costs, and financial fragility in the development sector, making state subsidy and market facilitation the only available levers. The claim is false. The evidence assembled here, drawn entirely from primary government and EU statistical sources, proves the opposite, and does so across five independent analytical dimensions that all point to the same structural reality.
Irish property developers (NACE 411) extracted an average 83.1% of all value they created as corporate profit between 2015 and 2024, accumulating €8.167 billion in a decade. This figure is confirmed by two independent datasets: CSO CNAPA01 and Eurostat sbs_ovw_act, which independently records Ireland’s NACE F41.1 GOS/GVA at 81.1% for 2022. The CSO’s own GMI data confirms that 97.6% of this surplus is corporate profit, not self-employed income — directly rebutting the ‘small sole trader’ counter-argument. The CSO’s COE data confirms that NACE 411 pays workers only 28 cents per euro of profit — vs €1.39 for builders. The margin never fell below 76% across ten years, through every macroeconomic shock in the period.
A full 34-country Eurostat ranking places Ireland 7th in Europe at 81.1% GOS/GVA — 17 percentage points above the EU-27 average of 64.0%. On a per-capita basis, Ireland generates €199 in developer surplus per person annually, against an EU average of €49 — 4.1 times the European norm. France generates €60, Germany €46. The argument that Irish developer margins reflect unavoidable economics of a high-cost small market is dismantled by this comparison.
CSO primary data (ARA02 and RZLPA01) confirms that residentially zoned land trades at 18–27 times agricultural land value nationally and 64 times in Dublin. This rezoning windfall is real and constitutes a first layer of extraction. But the EU comparison shows that Irish developers maintain 81.1% GOS/GVA even after absorbing these inflated land costs — 17 percentage points above the EU average. A pure land-cost pass-through model would predict Irish margins should be similar to or below European peers. They aren’t.
Vertical integration — documented in the publicly available records of Glenveagh, Cairn Homes, and Lioncor — is the mechanism through which the second extraction layer is made invisible. Ireland’s 47-percentage-point gap between F41.1 and F41.2 GOS/GVA is the statistical fingerprint of this structure at scale. The Glenveagh Cork case study makes the mechanism concrete: a site acquired from NAMA for €15 million generates a €525 million development; the state pays €150 million to the LDA for 337 apartments at post-uplift prices; a further €37 million in Croí Cónaithe subsidy flows to the developer; buyers are directed to layer Help-to-Buy and the First Home Scheme on top. €445 million returns to shareholders. The viability gap is an artefact of the same planning-uplift system that generated the profit. The state funds both the extraction and the gap it produces.
The reason this system has persisted for a decade despite being empirically documentable is a feedback loop of regulatory capture: each policy failure increases the state’s dependence on private delivery; increased dependence increases industry leverage over policy design; industry leverage ensures that structural interventions are replaced with demand-side subsidies, planning fast-tracks, and levy deferrals that sustain the extraction. Section 7.6 documents a further dimension: the systematic adoption of industry framing (most visibly the ‘viability gap’) as neutral analytical language in media and government discourse. The CIF generates 6–7 pre-planned reports per year and 652 press releases since 2016. Its analysis of policy questions where it has a direct financial interest enters the public record as expert commentary, routinely without disclosure of that interest. When the lobby defines the problem, the solutions it opposes become structurally difficult to articulate, the most durable form of capture of all.
The final thread changes the character of the other five. The extraction documented in Threads 1 through 5 didn’t persist because nobody knew, or because nobody told the state what to do about it. The state told itself, formally and repeatedly: the Kenny Report (1973) recommended acquisition of development land at existing use value plus 25%; the Commission on Taxation (2009) found a strong economic rationale for a Site Value Tax; NESC Report 145 (2018) diagnosed the land system as dysfunctional and recommended land value capture on the continental model; NESC Report 150 (2020) recommended an LDA equipped with compulsory acquisition and land value capture powers; the Housing Commission (2024) recommended State land value sharing. Five state-constituted bodies, five decades, the same recommendation, and zero implementations. Meanwhile the constitutional objection routinely invoked against acting was examined by the Supreme Court (2000) and the All-Party Oireachtas Committee (2004) and found wanting. The question this thread forces is the obvious one: is fifty years of commissioned, received, and shelved analysis political in nature, or something else?
The rival explanations can be dispatched in order. It isn’t ignorance: a state that commissions the analysis five times can’t claim not to know. It isn’t analytical incapacity: the remedies recommended in 1973 and 2018 didn’t depend on the distributional quantification this report supplies; they were implementable as recommended. It isn’t primarily personal corruption: the Mahon Tribunal documented genuine corruption around rezoning, but the non-implementation pattern spans governments, parties, and generations of officials, and outlived every individual scandal. No conspiracy is required, and none is alleged. What remains is structure, and the structure has three layers. The first is Mancur Olson’s logic of collective action: the beneficiaries of the planning rent are concentrated, organised, and professionally represented (Section 7.2’s lobbying density is the institutional form), while its losers are diffuse, unorganised, and in the case of future buyers, not yet present to object. The second is electoral arithmetic: the median Irish voter is a homeowner whose principal asset appreciates with the rent, which makes every government’s housing policy an exercise in promising affordability to one cohort without delivering price falls to the other. The third is the oldest layer: the founding settlement of the Irish land question itself. The Land War of the 1880s posed precisely this choice. Michael Davitt, influenced by Henry George (who wrote a pamphlet titled The Irish Land Question in 1881 and toured the country during the agitation), argued for land nationalisation; the movement’s mainstream under Parnell chose peasant proprietorship instead. Ireland resolved its land question through the diffusion of ownership rather than the socialisation of rent, and the political culture that settlement produced treats land appreciation as patrimony rather than privilege. The state founded on redistributing land has never been able to bring itself to tax it.
One further feature of the record deserves naming, because it converts the five shelved reports from a list of failures into a mechanism. A state that repeatedly commissions analysis it doesn’t implement isn’t failing to process the reform question. It’s processing it. Commissioning operates as the institutional metabolism of reform pressure: each report absorbs the demand for action, performs the state’s seriousness, occupies the reform constituency for the years of drafting and consultation, and expires quietly into the library. The next crisis produces the next commission. On this reading, the report cycle isn’t evidence against the capture documented in Section 7. It’s one of capture’s instruments, and the gentlest one available. So the answer to the question is: political, yes, but political in the structural sense the theory post describes rather than the venal sense the word usually implies. Acemoglu and Robinson supply the precise term: extractive institutions defending themselves. The five shelved reports are the audit trail of the defence.
Sources: Kenny Report (1973); Commission on Taxation (2009); NESC Reports 145 (2018) and 150 (2020); Housing Commission Report (May 2024); In re Article 26 and the Planning and Development Bill 1999 [2000] 2 IR 321; All-Party Oireachtas Committee, Ninth Progress Report (2004). Olson, M. (1965) The Logic of Collective Action. George, H. (1881) The Irish Land Question. Land League history: standard accounts of the Davitt–Parnell division on land nationalisation versus peasant proprietorship. Acemoglu, D. and Robinson, J. (2012) Why Nations Fail.
The convergence of five independent analytical threads permits a precise negative conclusion: the policy response to Ireland’s housing crisis can’t be primarily demand-side. Every euro injected into Help-to-Buy or the First Home Scheme flows directly into a market in which the developer extraction layer retains 81 cents of every euro of value created. Demand-side subsidies don’t lower the price-to-income multiple of 6.8×. They fund it. The evidence also rules out self-correction over time, a decade of widening EU divergence is conclusive. And it rules out the argument that developers are victims of high costs: their per-capita surplus of €199 versus the EU’s €49, sustained across ten years and multiple macroeconomic shocks, isn’t consistent with a sector operating under competitive cost pressure.
The evidence demands structural intervention at both extraction layers, sequenced in order of impact and implemented without the phasing and deferral that has characterised every previous attempt.
At Layer 1 (the landowner rezoning windfall) the instrument is a hard land value capture levy of 80–90% on the unearned capital appreciation created by rezoning. Denmark’s negative F41.1 GOS/GVA of −232.7% is the empirical outcome of a system in which this capture is structural. Ireland’s RZLT, having been narrowed and deferred under coordinated lobbying, isn’t a substitute.
At Layer 2 (the developer planning-permission premium) the instrument is public direct procurement: state agencies acquiring raw land at utility cost, managing planning through democratic frameworks, and contracting physical builders via transparent fixed-fee tenders. Glenveagh’s own disclosure of a landbank acquired at less than 10% of Net Development Value quantifies what this intervention would recapture: 90% of site appreciation currently retained privately would instead be available for affordable housing delivery.
Breaking the capture cycle requires a transparency intervention alongside the structural ones: mandatory consolidated group accounts, transfer pricing disclosures, and beneficial ownership registration for any entity receiving public subsidy, state land, or planning facilitation. Without this, every future policy assessment will continue to be made on the basis of subsidiary-level viability claims that conceal group-level returns. And without addressing the media framing dynamic documented in Section 7.6, those claims will continue to enter public discourse as neutral expert analysis, the condition that makes all other forms of capture sustainable.
The Housing Commission identified a 256,000-unit underlying deficit in May 2024. At the current delivery rate of approximately 30,000–36,000 units per year against a structural need of 38,000–50,000 units, the deficit compounds by 8,000–14,000 units annually. At that rate, the backlog will take between 40 and 80 years to clear. No current government programme engages honestly with this arithmetic.
Between 2015 and 2024, while that deficit accumulated, Irish property developers extracted €8.167 billion in gross operating surplus. That figure — 83.1 cents retained from every euro of value created, every year, for ten years — isn’t the cost of solving Ireland’s housing crisis. It is the cost of not solving it.
The structural interventions this report recommends aren’t politically easy. They are economically straightforward, empirically well-evidenced, and operationally proven across multiple EU member states. The only genuine obstacle is the capture dynamic documented in Section 7. Naming that obstacle precisely, and measuring its cost in units of housing deficit and price-to-income multiples — is the purpose of this report.
A report of this kind has a duty to be explicit about what it hasn’t done. The findings in Sections 1 through 10 rest on data I could obtain, query, and cross-validate within the resources of a single individual using public statistical platforms. Several lines of inquiry could not be completed for reasons of data availability, dataset granularity, or required administrative access. They are documented here because each is potentially as significant as anything currently in the report, because each is open to journalists, researchers, parliamentary committees and the relevant state bodies to take up, and because the absence of these analyses from the public record is itself part of the picture this report describes.
Where the obstacle is a CSO or Eurostat data gap, the CSO accepts research data requests through its Researcher Microdata File programme and through direct correspondence with subject-matter units. Several of the avenues below could be opened by such a request from a Dáil committee, a university research centre, or an investigative outlet. The data exists. The published series doesn’t.
Three independent datasets used in this report report Irish construction at NACE F level only, without breaking out development (F411) from building construction (F412) and specialised construction (F43).
FATS (Foreign Affiliates Statistics): Section 7.10 documents that foreign-controlled construction enterprises in Ireland generated €1,096 million in gross operating surplus in 2022 at a 49.7% GOS/VA margin (twice the EU average), but the FATS series doesn’t split this between F411, F412 and F43. The foreign-controlled share of NACE F411 specifically (the question the report would ideally answer) can’t be determined from publicly available FATS. Eurostat should publish FATS at three-digit NACE granularity; absent that, the CSO could compile the breakdown from national-level returns on request.
NA013 (National Accounts by Institutional Sector): Section 7.10 documents that NA013 shows corporate saving rising from €75bn (2015) to €151bn (2024) but doesn’t break out institutional sector accounts by NACE. The savings and investment behaviour of NACE F411 enterprises specifically — what share of the €8.167 billion GOS is retained domestically, distributed as dividends, or reinvested in productive assets — can’t be derived from publicly available CSO data other than through the Eurostat investment-rate proxy this report uses.
CSO Annual Enterprise Statistics (BAA12): The CSO published BAA12 figures broken out by F411 and F412 through 2019. Subsequent vintages of the Annual Enterprise Statistics moved to broader NACE divisions, leaving the 2020–2024 employment series for F411 and F412 not directly published in the standard release. The CSO’s SBS unit may release the underlying breakdown on request, which would convert several of this report’s headcount inferences into directly cited figures.
A coordinated request to the CSO and to Eurostat for three-digit NACE granularity in FATS, NA013, and post-2019 BAA12 would resolve the largest single empirical gap in the report’s framework. Each of these datasets is constructed from administrative returns that already contain the granularity required; what is missing is the decision to publish it.
This report quantifies €8.167 billion in NACE 411 gross operating surplus over the decade 2015–2024. GOS is a pre-tax measure. The question the report doesn’t answer is what share of that surplus is actually paid in corporation tax. Ireland’s headline corporation tax rate is 12.5%; the effective rate paid by NACE 411 entities specifically is unknown from any publicly available source. Available reliefs include: capital allowances on construction assets; loss relief carried forward from the post-2008 crash period (potentially material for entities incorporated pre-2012); land trading treatment (whether land sales are taxed as trading income or capital gains, with attendant rate differences); group relief between related entities in vertically integrated structures; and the Section 110 securitisation regime where applicable to development financing vehicles. The interaction of these reliefs with the SPV-dominated corporate structure documented in Sections 2.6 and 6 could produce effective rates materially below the headline figure.
The data sources that could answer this question aren’t in the public domain but are accessible: Revenue Commissioners’ sectoral tax receipts by NACE division, published in aggregate but not by F411 specifically; CRO-filed annual accounts of the largest 50 NACE 411 entities by GOS contribution, which carry corporation tax disclosures; and Revenue’s published statistics on the use of loss relief and land trading classifications. A parliamentary question to the Minister for Finance directing Revenue to disclose the effective corporation tax rate paid by NACE F411 over the 2015–2024 decade is the most direct route. If the effective rate is materially below 12.5%, the report acquires a second extraction finding: lightly taxed as well as uncaptured, with the planning rent privatised at one end and the corporation tax base eroded at the other.
Intervention 5 of Section 9 (Use-It-Or-Lose-It Planning Conditions) is currently justified on the principle established in Section 6 and in the British control case in the experimental section: that holders of planning permission have a financial incentive to control the rate of build-out below the rate at which permissions are issued. The empirical scale of this in Ireland isn’t currently quantified in the report. The relevant figure is the stock of residential planning permissions granted but not yet commenced, broken down by the year permission was granted, the developer holding the permission, and the local authority area.
The data exists. Commencement notices are filed with local authorities and aggregated by the Department of Housing’s monthly commencement notice releases. Planning permissions are recorded by each local authority and aggregated by An Bord Pleanála. The difference between the two (permissions granted that haven’t progressed to commencement) is calculable, and a quarterly time series of the uncommenced stock has been produced internally within the Department of Housing but isn’t, to this report’s knowledge, published in a granular form that permits the absorption-rate analysis the Letwin Review and the UK CMA market study performed for the British market. A research request to the Department of Housing under the FOI Act for the uncommenced permissions stock by year and by holder, or a parliamentary question for the same, would convert the qualitative Intervention 5 argument into a quantified one. The Letwin finding — that large sites are built out at the local market absorption rate rather than the rate at which permissions are issued — is empirically testable for Ireland using data the state already collects.
The most frequently advanced alternative explanation for delayed delivery in Irish housing is the judicial review (JR) of planning permissions. The claim — typically that ‘a small number of objectors are blocking thousands of homes through the courts’ — has been advanced by ministers, by industry representatives, and by sections of the media, and was used as the substantive policy rationale for the 2024 Planning and Development Act’s restrictions on JR standing and costs. This report hasn’t engaged with the JR narrative because doing so requires data I haven’t been able to systematically obtain: the share of total residential planning permissions in a given year that are subject to JR; the share of those JRs that succeed, fail, settle, or lapse; and the share of total permitted residential units affected by JR proceedings.
An Bord Pleanála publishes annual statistics on JR cases, and the Courts Service publishes judgment data. The work of cross-referencing JR cases to specific permissions and to specific unit counts hasn’t, to this report’s knowledge, been done systematically in the public domain. The available indications, drawn from spot-checking, suggest that JRs affect a small minority of permitted units in a given year, and that a substantial share of JRs that proceed to judgment are upheld (i.e. the planning decision is found to have been legally defective, not merely contested). If this is borne out by a systematic analysis, the JR narrative is empirically weak as an explanation for delivery shortfalls and the 2024 Act’s restrictions would acquire a different character — as a removal of a check on legally defective planning decisions, rather than a removal of a delivery bottleneck. This is a discrete piece of empirical work, well within the scope of an academic or investigative team with access to the underlying case data.
In November 2025 the Department of Housing published the final RZLT maps for 2026 implementation on data.gov.ie, identifying parcels of zoned and serviced land liable to the tax. Section 7.3 documents the lobbying history of the RZLT and its repeated deferral; the published 2026 maps now provide the first empirically tractable test of the tax in operation. The questions that can be answered from the published data alone include: how many hectares of zoned residential land are recorded as liable; how that compares to the Housing Commission’s identified development land requirement; how the geographic distribution of liable land aligns with the residential price-to-income pressure mapped in Section 4.4; and which named entities hold the largest liable parcels (cross-referenced to Land Registry records). The questions that will become answerable as the tax progresses through 2026 are equally pointed: what share of liable parcels receive deferral or exemption, what share of the assessed levy is actually collected versus billed, and how the collection rate compares to the historical record of the Derelict Sites Levy and the Vacant Site Levy documented in Section 7.3.
This is a piece of monitoring work that any of the following bodies could and arguably should undertake on an ongoing basis: the Comptroller and Auditor General as part of routine audit of departmental revenues; the Oireachtas Public Accounts Committee; the Parliamentary Budget Office; the Housing Commission as a follow-up to its 2024 report; or an investigative outlet building a public dataset. A live dashboard tracking RZLT liability, deferral, collection and revenue would convert the third-consecutive-non-enforcement pattern this report documents into either a vindicated prediction or a refuted one within twelve to eighteen months.
Source for RZLT 2026 maps: Department of Housing, Local Government and Heritage, ‘Residential Zoned Land Tax: Final Map for 2026’, published on data.gov.ie (dataset id ee22550f-98ac-4fed-b8f8-c3e34a43e8c2).
The avenues above are the ones for which the report has identified specific datasets, specific data-holder bodies, and specific routes to resolution. Several further lines of inquiry are flagged here in shorter form as questions worth pursuing.
The pension fund question: Who holds the equity of Ireland’s listed homebuilders, and what share of dividends flows to Irish residents versus international institutional investors? Glenveagh and Cairn’s shareholder registers are public; a systematic decomposition would quantify the often-invoked but rarely-evidenced claim that ‘this is your pension’.
HBFI’s loan book: Home Building Finance Ireland’s annual reports disclose total lending but not the distribution of loans by borrower size. An FOI request for the distribution of HBFI loans by recipient (with appropriate commercial-sensitivity redactions) would establish whether HBFI’s lending is supporting small and medium developers (its stated purpose) or whether it is concentrated among the same large operators that have access to commercial bank finance.
Croí Cónaithe per-contract amounts: Section 8.3 documents that the Department of Housing publishes the aggregate maximum funding envelope and the aggregate unit count under Croí Cónaithe, but not the per-contract amounts paid to individual developers. An FOI request, or a parliamentary question for the per-development subsidy breakdown, would close this transparency gap directly.
The vertical integration premium, quantified: Section 6 documents the structural mechanism of profit transfer between F411 holding companies and F412 building subsidiaries within vertically integrated groups. The Glenveagh, Cairn and Lioncor accounts include inter-company transfer pricing in their related-party notes. A forensic accounting study across the largest five vertically integrated Irish homebuilders would quantify the developer-to-builder margin transfer the report currently characterises in principle.
International contractor margins on the same buildings: BAM Ireland, Sisk and John Paul Construction operate as builders in Ireland under structurally identical contracts to those their parent companies operate in the Netherlands, the UK and elsewhere. A comparison of operating margins on equivalent project types, where the project data permits, would directly test the international contractor argument advanced in Section 5.3 against firm-level rather than national-level data.
Interest paid by NACE 411 — quantifying the second extraction layer: GOS is measured before interest, so part of the developer surplus flows onward to financiers, largely foreign credit funds. No published statistic quantifies the interest actually paid by NACE F411 enterprises. The data exists in Revenue’s corporation tax computations (interest deductions claimed) and would also fall out of the F41 institutional sector account recommended above. Quantifying it would split the 81% between the developer’s equity and the capital stack above it, and establish how much of the planning rent exits the country as coupon payments.
The CSO’s NACE F SBS methodology change in 2020–2022: The Eurostat sbs_ovw_act series shows a step-change in NACE F411 ‘persons engaged’ between 2020 (3,936) and 2022 (8,572). This report infers, in Section 2.5, that the change reflects a broadening of statistical scope at the CSO (likely the incorporation of subcontracted professional services into the persons-engaged definition). A direct query to the CSO’s Structural Business Statistics unit confirming the methodological revision and its date would convert this from inference to documented fact.
None of the avenues above invalidate any finding in the preceding ten sections. The Section 2 distributional finding, the Section 4 EU benchmarking, the Section 5 land price mechanism, the Section 7 capture documentation, and the Section 8 case study each rest on data that is in the public domain, has been cited to its primary source, and has been cross-validated where multiple datasets permit. The unexplored avenues would, in each case, add further granularity, further confirmation, or further dimensions of the same picture. They wouldn’t refute it.
The reason to document them explicitly is twofold. First, intellectual honesty: a report that claims completeness it doesn’t possess invites attack on grounds that have nothing to do with its substantive findings. Second, the architecture of public scrutiny: each of these avenues is a question some specific body — Revenue, the CSO, Eurostat, the C&AG, the Department of Housing, the Oireachtas — has the standing and the data to answer. Naming them is the simplest possible form of accountability journalism: writing down the questions, in public, so they can be put. I would welcome correspondence from anyone pursuing any of these lines.
The table below reproduces the three-year series for key countries directly from Eurostat sbs_ovw_act (DATAFLOW: ESTAT:SBS_OVW_ACT(1.0), updated 10/03/2026). These are exact values from the dataset download. No figure is estimated or derived.
| Country | GOR 2021 | GOS/GVA 2021 | GOR 2022 | GOS/GVA 2022 | GOR 2023 | GOS/GVA 2023 |
|---|---|---|---|---|---|---|
| Denmark | 2.08% | 17.8% | -8.40% | -232.7% | -0.35% | -2.9% |
| EU-27 Average | 13.28% | 65.8% | 12.18% | 64.0% | 11.18% | 61.9% |
| Finland | 4.39% | 40.2% | 2.20% | 16.3% | 3.71% | 22.9% |
| Ireland | 17.70% | 82.5% | 17.57% | 81.1% | 15.51% | 80.5% |
| Malta | 30.50% | 90.9% | 34.07% | 89.3% | 34.26% | 87.8% |
| Poland | 15.23% | 72.8% | 17.42% | 73.8% | 18.28% | 72.8% |
| Belgium | 13.99% | 83.1% | 17.16% | 84.6% | 13.96% | 82.3% |
| Germany | 21.81% | 71.7% | 16.31% | 62.9% | 9.85% | 50.7% |
| France | 7.81% | 49.1% | 8.34% | 50.6% | 6.66% | 43.4% |
| Netherlands | 21.29% | 70.0% | 19.93% | 66.4% | 20.15% | 66.0% |
Source: Eurostat sbs_ovw_act, DATAFLOW ESTAT:SBS_OVW_ACT(1.0), last updated 10/03/2026. Indicators: GOR_PC and GOS_AV_PC. NACE: F41.1. All values are direct dataset downloads.
| Source | What it is / used for | Link |
|---|---|---|
| CSO CNAPA01 | Construction national accounts — GVA, GOS, COE, GMI by NACE 411/412/432. The report’s core dataset | data.cso.ie/table/CNAPA01 |
| CSO BAA12 | Annual enterprise statistics — F411/F412 employment split to 2019 | data.cso.ie/table/BAA12 |
| CSO ARA02 | Agricultural land prices (the rezoning baseline) | data.cso.ie/table/ARA02 |
| CSO RZLPA01 | Residentially zoned land prices, transactions, volumes (Frontier Series) | data.cso.ie/table/RZLPA01 |
| CSO NA013 | Institutional sector accounts — corporate saving vs B&C capital formation | data.cso.ie/table/NA013 |
| CSO GPIIA04 / GPIIA05 | Gross household income (affordability denominators) | data.cso.ie/table/GPIIA04 · /GPIIA05 |
| CSO HPM03 | Residential property price index / median prices | data.cso.ie/table/HPM03 |
| CSO DEA08 | Earnings data | data.cso.ie/table/DEA08 |
| Eurostat sbs_ovw_act | Structural Business Statistics — GOS/GVA, WALP, investment, wages, enterprises for F411/F412, all EU countries | ec.europa.eu/eurostat/databrowser/view/sbs_ovw_act |
| Eurostat fats_activ | Foreign Affiliates Statistics — foreign-controlled construction (NACE F) | ec.europa.eu/eurostat/databrowser/view/fats_activ |
| Eurostat tps00001 | Population (per-capita denominators) | ec.europa.eu/eurostat/databrowser/view/tps00001 |
| CSO Business in Ireland 2020 | Table 9.6 — EU27 construction GOS/turnover comparison (the “aggregation camouflage” finding) | cso.ie/en/releasesandpublications/ep/p-bii/businessinireland2020/internationalcomparisons |
| Kenny Report (1973) | Committee on the Price of Building Land — existing-use +25% in designated areas | No official digital copy; scans circulate via university repositories and JCFJ — archive whichever copy you cite |
| Ninth Progress Report (2004) | All-Party Oireachtas Committee on the Constitution — Private Property; constitutional clearance for Kenny-type scheme | universityofgalway.ie/media/housinglawrightsandpolicy/nationalpolicy/9th-Report-Property.pdf |
| Part V Article 26 reference | In re Art. 26 & Planning and Development Bill 1999 [2000] 2 IR 321 — full judgment | bailii.org/ie/cases/IESC/2000/20.html |
| NESC Reports 145 (2018) & 150 (2020) | Land system diagnosis; LVC/SVT/active land management recommendations | nesc.ie → Publications (search report number) |
| Commission on Taxation 2009 | SVT “strong economic rationale” (and 1982 edition for the betterment tax) | gov.ie — Department of Finance publications archive |
| Housing Commission Report (2024) | State land value sharing recommendation; deficit estimate | gov.ie — search “Report of The Housing Commission” |
| SCSI Real Costs of New Apartment Delivery 2025 | The cost stack, viability hurdle, “gap closed” headline (you hold the PDF) | scsi.ie → Publications |
| CMA Housebuilding Market Study (Feb 2024) | Absorption-rate finding; 1m+ plots; speculative model | gov.uk/cma-cases/housebuilding-market-study |
| CMA commitments (Oct 2025) | £100m settlement, seven housebuilders | gov.uk — CMA case “Housebuilders: suspected anti-competitive conduct” |
| Letwin Review (2018) | Independent Review of Build Out — market absorption rate | gov.uk/government/publications/independent-review-of-build-out-final-report |
| Glenveagh FY2025 results + RNS | Revenue, margins, landbank <10% NDV, Partnerships, Marina Depot/LDA RNS | glenveagh.ie/investors (RNS also on investegate.co.uk) |
| gov.ie Croí Cónaithe approvals | Blackrock Villas 274-unit approval press release | gov.ie press release “Minister O’Brien announces approvals under Croí Cónaithe Cities” |
| Housing Agency — Croí Cónaithe | Scheme administration, third-call project list | housingagency.ie/news-events/croi-conaithe-cities-third-call-expressions-interest |
| MyHome Blackrock Villas listing | Live prices €233,000–€437,500 (ref 4918029) —screenshot now, listings vanish | myhome.ie/residential/brochure/blackrock-villas-blackrock-villas-blackrock-cork/4918029 |
| RZLT Final Map 2026 | Liable parcels — the live empirical test in 11.5 | data.gov.ie dataset ee22550f-98ac-4fed-b8f8-c3e34a43e8c2 |
| HBFI | State development lender — products, rates (4.75–7.5% over Euribor), annual reports | hbfi.ie/products/small-developments |
| Dept of Finance development finance report | Bank caution, non-bank lenders up the risk curve | assets.gov.ie/296388 (PDF) |
| Register of Lobbying | The 61-representations data, RZLT lobbying returns | lobbying.ie |
| ESA 2010 manual | GOS pre-interest definition (§8.18–8.22) | ec.europa.eu/eurostat → ESA 2010 publication |
GOS/GVA (Value Margin): Share of newly created economic value retained as corporate surplus. 81% means workers receive 19 cents and capital retains 81 cents per euro of new wealth created.
GOR — Gross Operating Rate (GOS/Turnover): GOS as a percentage of total revenue. Eurostat’s primary cross-country comparability metric.
COE (Compensation of Employees): Total labour cost to enterprises — wages, salaries, and employer social contributions. Used in Section 2.3 to calculate labour-capital ratios.
GMI (Gross Mixed Income): Income of self-employed persons. Used in Section 2.4 to separate sole-trader income from corporate profit within NACE 411.
NACE F41.1 — Development of Building Projects: Entities that acquire land, obtain planning permission, and develop or sell sites.
NACE F41.2 — Construction of Residential & Non-Residential Buildings: Physical contractors who construct structures.
Rezoning Uplift: The increase in land value generated by a change in planning designation from agricultural to residential. Quantified in Section 5 using CSO primary data (ARA02 & RZLPA01) at 18–27× nationally and 64× in Dublin (2024).