The Missing Machinery: Technical Annex

Annex 1: Statutory Pre-Emption Rights (Convention Compliance and Instrument Design)

This annex states what ECHR case law requires any property-rights interference to include, identifies how the pre-emption architecture meets those requirements, and flags the foreseeable challenge surfaces. It has not had legal review. The arguments identified as available are the author’s reading of published judgments; they require testing by counsel before the Bill team relies on them. The operational mechanism sections (S75 register, viability countermeasures, procedural design, anti-avoidance, corporate structure, DRLT independence) are architectural commitments that do not depend on a specific legal conclusion.

What the ECHR cases require

Article 1 Protocol 1 imposes three conditions on state interference with property: legitimate aim, legality, and fair balance. The case law specifies what each condition demands in practice.

Legitimate aim requires a sufficiently weighty public interest, with a wide margin of appreciation in housing and social policy. James and Others v United Kingdom (1986) 8 EHRR 123: compulsory transfer of freeholds at statutory formula price (below market value) fell within the margin. Lithgow and Others v United Kingdom (1986) 8 EHRR 329: states may choose compensation formulae producing results below willing-buyer-willing-seller price, provided the choices are not manifestly without reasonable foundation. Scotland’s 18,092 TA households, 164,000 unbuilt consented homes, and record RSL interest burden constitute the factual base for the legitimate aim.

Legality requires the law to be accessible, foreseeable, and accompanied by adequate procedural safeguards. The pre-emption right is constituted in primary legislation specifying the trigger (LDP allocation), mechanism (EUV-plus offer on disposal), holder’s procedural rights (declaration window, build-out commitment route, right to decline with defined consequences), and valuation methodology (independent EUV assessment). A landowner reading the Act at commencement can determine when the right attaches, how it operates, and what options are available.

The Bill should define EUV as a statutory concept, not by cross-reference to the RICS Red Book (RICS Valuation – Global Standards, 2024). The Red Book’s “existing use value” assumes a willing seller and an open market for the current use; those assumptions are incompatible with a pre-emption regime that compresses the seller’s alternatives by design. The statutory definition: the value of the land assuming continuation of its current lawful use, disregarding any value attributable to residential development potential conferred by LDP allocation, planning permission, or the expectation of either. Value that depends on development not yet carried out is development potential, not existing use.

The valuation methodology must produce a single determination per site that a landowner can foresee at commencement; the legality requirement above demands it. The three-tier treatment in the main paper (agricultural, active economic, no productive use) means different existing conditions to value, but the governing principle is the same across all three. Existing public valuation infrastructure should carry the assessment where possible: Scottish Assessors hold comparable-evidence data for non-domestic property, Rural Payments Scotland records verify agricultural activity, NDR records verify active economic use. Building a parallel valuation apparatus when these institutions already hold the underlying data would be unnecessary duplication.

The methodology must not reintroduce a discretionary variable that volume housebuilders’ advisers can colonise. The per-hectare DRLT rate base forecloses that surface on the tax side; the EUV valuation must do the same on the acquisition side. Dispute resolution should sit with the Lands Tribunal for Scotland, which already handles compulsory purchase compensation disputes and has the jurisdictional competence the instrument requires.

Fair balance requires a proportionate relationship between the public aim and the individual burden. The case law specifies what breaks it and what survives.

In Hutten-Czapska v Poland (2006) 45 EHRR 4, the state lost: Poland’s rent control regime violated A1P1 because the interference was unlimited in time, deprived owners of effective use, and carried no procedural route to relief. In Holy Monasteries v Greece (1994) 20 EHRR 1, the state lost again: Greece’s law presumed state ownership, reversed the burden of proof onto holders, and provided no compensation the Court was willing to recognise. From those defeats, three architectural requirements: compensation must exist, a procedural route to relief must be available, and the interference must be bounded in time and scope.

The pre-emption architecture meets those requirements. The declaration window (24 months) bounds the transition in time. Both commitment routes preserve owner agency: commit to a delivery schedule and retain market-value rights on future disposal, or accept prospective EUV pre-emption. For the general pre-emption right, no holder is compelled to sell; the burden is loss of the speculative premium on a future voluntary sale, not loss of possession. For S75-breach pre-emption, the holder accepted the delivery obligation at consent; the acquisition is closer to specific performance than to deprivation (see “S75 enforcement” below).

The procedural route is defined in statute: commit to a delivery schedule, accept prospective pre-emption, or seek de-allocation through the LDP review cycle where the holder considers the residential allocation unjustified. For holders with existing S75 obligations, the declaration window carries an additional consequence: lodge a binding build-out commitment, agree the LDP plan period end date as the deemed delivery deadline for the S75 obligation, or accept that the deemed-deadline provision applies on its statutory terms and the site is disposable at EUV plus to any buyer willing to deliver. Both the commitment obligation and the pre-emption right run with the land and bind singular successors; a transfer by inheritance or testamentary disposition does not extinguish either. Compensation is at independently assessed EUV plus a statutory premium, not at nil. The Act specifies a ceiling on the premium; the rate within that ceiling is set by order of the Scottish Ministers, subject to affirmative procedure, with a three-year Scottish Fiscal Commission review parallel to the DRLT cycle. The ceiling in primary legislation satisfies the legality requirement above: a landowner reading the Act at commencement can determine the maximum compensation they will receive. The premium is the instrument’s answer to the political economy of compulsory acquisition: above- agricultural-value compensation makes the confiscation argument hard to sustain without defending speculative hope value as a legitimate expectation on land where a planning obligation was accepted and not delivered.

S75 enforcement: the instrument’s primary target

The pre-emption architecture’s primary function is enforcing delivery of the affordable housing components that S75 agreements already require. Every major residential site carries a legal obligation, accepted at planning consent, to deliver an affordable housing component: at least 25% of the total under NPF4 Policy 16, with the actual quantum and tenure split determined by S75 negotiation. The private build rate and the affordable delivery rate are not independent variables: when large housebuilders hold market sales to 25 homes per site per year, the affordable component is deferred with them. The problem is land banking used to defer or avoid the publicly mandated components, and the private drip-feed is the mechanism that makes it possible.

This reframe changes the Convention analysis. Enforcing an existing legal obligation is a narrower interference with property rights than a general social-policy land reform. The instrument codifies what happens when a developer fails to perform an obligation they voluntarily accepted: closer to specific performance than to a new state imposition.

Whole-site acquisition at EUV plus. Pre-emption triggered by undelivered S75 obligations at the LDP deadline operates on the whole site, not surgically on the S75 portions alone. The state acquires the entire site at EUV plus and tenders all portions through the MHS procurement framework: social and affordable components to SHFF or RSL delivery partners, the private component to builders (including SME consortia) on a published cost-plus basis. The breaching developer is not excluded from bidding for tendered lots but holds no automatic right of buyback.

The whole-site scope is the mechanism that makes S75 enforcement credible. Pre-emption limited to the S75 portions would let a developer abandon the affordable obligation while retaining the hope-value uplift on the private portion. Whole-site acquisition at EUV plus removes that arbitrage: non-compliance costs the developer the hope-value premium across the entire site. Compliance becomes the rational choice from the day of consent. Selling the private portion back to the breaching developer would undercut this logic. Competitive tendering prevents that outcome and opens the site to builders who could not have competed at hope-value pricing.

Foreseeable challenge surfaces

The arguments below are available on the author’s reading of the published judgments. They have not had legal review. Each requires testing by counsel before the Bill team relies on it.

Challenge 1: penalty doctrine. A claimant will argue whole-site pre-emption at EUV plus operates as a contractual penalty, per Cavendish Square Holding BV v Makdessi [2015] UKSC 67 (confirmed by Lord Hodge as applying in Scotland).

Argument believed to be available: the penalty doctrine constrains contractual secondary obligations. The S75 creates the planning obligation; the pre-emption right is a statutory consequence imposed by the Act, not a term of the S75. The distinction is between what the contract provides and what Parliament provides for breach of the obligation the contract contains. Even if a court applied Cavendish by analogy to a statutory consequence, the legitimate interest in enforcing delivery of the social housing the planning system requires is clear, and the proportionality features (EUV-plus compensation, competitive tendering, whole-site scope) address the disproportionality limb. Counsel should assess whether the statutory/contractual boundary holds under Scots law and whether the whole-site scope survives the disproportionality test.

Challenge 2: DRLT as economic compulsion. A claimant will argue the escalating holding cost makes disposal economically compelled rather than voluntary, collapsing the “no holder is compelled to sell” argument.

Argument believed to be available: DRLT is a regulatory holding cost in the same category as non-domestic rates or council tax. No court has characterised an annual property tax as compulsion to dispose. The ceiling, rate-review, and safeguard architecture are set out in the DRLT operational independence section below and in Annex 3. Counsel should assess whether the ceiling (set in Annex 3) changes the characterisation from holding cost to de facto compulsion, and whether the annual rate-review mechanism is sufficient procedural protection.

Challenge 3: distinguishing the Strasbourg precedents. A claimant’s lawyer will distinguish the precedents on at least three grounds.

James: private-to-private redistribution under a statutory formula price, with the long-leaseholder as direct beneficiary. Pre-emption operates private-to-public, with a public body’s procurement role as an intermediary; this is a beneficiary configuration Strasbourg has not directly ruled on.

Lithgow: compensation for compulsory nationalisation transfers that had already happened. The general pre-emption right operates prospectively on a voluntary disposal; S75-breach pre-emption operates on a missed contractual deadline. Both are prospective. The relevant Strasbourg test is proximate to but not identical with Lithgow’s.

Declaration window as coerced choice: a claimant can argue the window presents a choice between two unpalatable options (commit to a build-out schedule the holder cannot meet because of infrastructure or finance constraints they do not control, or accept prospective EUV-plus pre-emption on disposal). The option to retain land at indefinite cost without market exit is a real third path but a constrained one.

These are live challenge surfaces. The instrument is designed to survive them, not to avoid them. Each maps to a specific design choice (independent EUV assessment, bounded declaration window, infrastructure gateway, retained ownership, rate-review at year three). A Bill that presents A1P1 compatibility as guaranteed has overstated. One that presents it as designed-for-defensible-challenge is accurate, and stronger for acknowledging the exposure. Counsel should advise on whether the design choices are sufficient and where the residual risk sits.

S75 monitoring register: operational precondition for the crystallisation trigger

The S75 crystallisation mechanism is legally coherent. It is operationally dependent on information that does not currently exist in usable form.

It fires when the social and affordable housing components of a site are undelivered at the LDP delivery deadline. That requires three things: every live S75 obligation on record, each carrying a delivery deadline with legal force, and monitored build-out progress capable of identifying breach when it occurs. None of those conditions is currently met at national scale. No national register of S75 obligations exists. Agreements do not carry a standardised delivery deadline: many specify phased delivery against the planning permission rather than an LDP target date. Build-out progress is not monitored nationally. A developer who banks the social and affordable components of a consented site faces near-certainty that no breach notification will be generated and no enforcement action will follow.

The S75 enforcement consequence is direct: the crystallisation trigger exists in statute but fires only when a council chooses to act on breach, which is to say rarely. The supply visibility consequence is less obvious: without a national register, the Scottish Government cannot know how many of the 164,000 consented but unbuilt sites carry S75 affordable housing obligations, what those obligations require, or whether any of them have already been discharged by commuted sums. The consultation treats the 164,000 as a uniform pipeline to be unlocked. In reality, the stock divides: sites carrying existing public obligations to deliver social housing, sites that discharged those obligations by cash payment instead of homes, and sites carrying no affordable housing obligation at all. Distinguishing those categories is a precondition for any intelligent targeting of enforcement resource, and it currently cannot be done.

Commuted sums compound the data failure: where a cash payment is accepted in lieu of on-site delivery, the S75 obligation may show discharge even though no homes were built. Proposal 5 abolishes commuted sums prospectively. The register needs to capture both active obligations and commuted-sum discharge history, so the national picture of what was obligated, built, and bought out can be established for the existing consented stock.

The legislation should address this directly. Four requirements should sit on the face of the Bill.

National S75 obligations register. The register aggregates obligation data already held across Scotland’s 32 planning authorities and Registers of Scotland into a single monitoring dataset. Councils must submit data on every active S75 agreement, and on every commuted-sum discharge granted in the preceding ten years, to a national register held by More Homes Scotland within 12 months of commencement. New agreements should be submitted to the register on the day of execution. Subsequent events (S75 modification, commuted-sum discharge, milestone completion, breach notification) should be logged on the day they occur. The register must contain: the site reference (UPRN and Planning Register reference), the name of the obligation holder, the affordable housing quantum and tenure split, the delivery deadline as defined in the agreement, the commuted sum record if any, and the date of the most recent delivery verification. MHS has statutory standing to interrogate the register and initiate pre-emption proceedings on identified breach without requiring the originating council to initiate first.

Standardised S75 delivery deadline condition. For agreements executed on or after commencement, the planning authority must include an LDP target date as a binding delivery condition. The deadline is the LDP target date for the allocation, or such later date as the planning authority specifies with written justification. An agreement executed without a defined delivery deadline is unenforceable for the purposes of the pre-emption trigger; the obligation holder is on notice of this from the day of consent.

Deemed deadline for pre-commencement S75 agreements. Existing S75 agreements carry no standardised delivery deadline; many specify phased delivery against the planning permission rather than an LDP target date. The Bill should provide that for any S75 agreement executed before commencement, the delivery deadline is deemed to be the end of the LDP plan period under which the allocation was made. The LDP plan period is a published document adopted after public consultation; the developer was consulted during adoption and knew the planning horizon for which the allocation was intended. Plan period expiry is an objective fact, not a retrospectively imposed contractual term. Where the plan period has already expired at the date of commencement, the deemed deadline is treated as passed and the crystallisation trigger is available from the end of the declaration window. Where the plan period has not yet expired, the trigger becomes available on expiry. This provides the pre-emption mechanism with a legally grounded firing point for the legacy stock without retrospective variation of private agreements.

Annual delivery verification returns. Real-time event logging does not replace periodic audit. Councils must submit an annual verification return to MHS covering every active obligation in their area: current build-out status against the registered delivery deadline, and confirmation that the register reflects all events in the preceding twelve months. Failure to submit is a breach by the council, not a suspension of the pre-emption right. The MHS register, holding the last known data, stands as evidence of non-delivery where a council fails to update.

MHS standing to act on breach. The Bill team should confirm whether MHS requires its own statutory cause of action to exercise pre-emption following an S75 breach identified through the register, or whether it can act as agent for the planning authority that holds the original agreement. The simpler architecture (MHS acting directly, with notification to the originating council) is preferred on delivery speed grounds. A mechanism that requires the originating council to initiate creates a dependency on their willingness and capacity to act, which is precisely the dependency that has produced the current monitoring failure.

The transitional provision for existing agreements is the hardest design question. Many will not contain delivery deadlines that comply with the standardised condition. Councils should be given 12 months from commencement to seek voluntary agreement with obligation holders on supplementary delivery deadline conditions. Where agreement is not reached, the planning authority may apply to the Scottish Land Court to have a deadline imposed. The court-imposed deadline should be the current LDP target date for the allocation unless the holder demonstrates a specific infrastructure or statutory-body constraint justifying a later date.

S75 viability challenge surface: mechanism and statutory countermeasures

The monitoring register addresses the non-delivery surface: a developer who accepts an S75 obligation at planning consent and then does not build. The earlier attack surface is viability challenge: a developer who contests the obligation’s achievability under the s.75A modification procedure before construction begins.

After the 2008 financial crisis, viability challenges to Section 106 contributions in England almost all succeeded (McAllister et al., 2014, cited in Foye and Shepherd, 2023). Foye (2022) provides the mechanism. Under S106 as it operated in the 2000s, central government specified that affordable housing targets must not compromise viability, but left viability undefined. This transferred effective control to ‘a range of unelected institutions’ (Foye, 2022, p.279): viability consultants, their models, courts, and planning inspectors. The land cost assumption is the critical discretionary input: whoever sets the assumed land value controls whether the site is viable. Developer-funded consultants set it. Local planning authorities lacked the technical resource and standing to challenge it.

McAllister et al. (2016) provide evidence that viability consultants routinely adjusted inputs in advance to reach consensus among developers, housing associations, and landowners, but not the general public or elected politicians (Foye, 2022, p.279). One consultant described the purpose: ‘you can use that exercise to neutralize opposition’ (McAllister et al., 2016, cited in Foye, 2022, p.280). The process was not adversarial. It was settlement by pre-agreement, ratified at the planning examination as technical consensus.

Circular 4/2025, which governs S75 in Scotland, preserves the same structural vulnerability. Para 40 states that a planning authority ‘may’ seek independent assessment where the developer claims viability grounds for modification under s.75A. Independent assessment is optional, not mandatory. The land cost input remains in developer hands by default. The s.75A modification procedure (and the s.75B appeal route to Scottish Ministers) gives the developer a formally available legal route; para 40’s optionality means they enter it with the critical assumption already set in their favour.

The resource gap this creates is not speculative. In May 2016 Heads of Planning Scotland told the Chief Planner that scrutinising figures in planning submissions ‘requires the skills of an economist’ and that councils would need to commission external professional review to carry out that function (HOPS, 2016, cited in Tolson and Rintoul, 2018). A viability assessment is an economic document submitted to a process staffed by planners. Para 40’s ‘may’ leaves that structural mismatch intact.

Para 37 of the same Circular states that developers ‘should take account of’ LDP policies on obligations ‘in their appraisals and land purchases’. A developer who acquired a site at hope value after the LDP allocation had already specified the S75 obligation cannot legitimately run a viability argument based on that acquisition cost. The principle is already in policy. It is not yet in statute. Three provisions should put it there.

EUV as mandatory land cost input. In any S75 viability assessment relating to an obligation in an LDP allocation made on or after the Act’s commencement, the land cost input must be the site’s existing use value as independently assessed. This removes the manipulable variable from developer control. It operationalises Circular 4/2025 para 37 in statute: if the obligation was known at acquisition, acquisition at hope value is the developer’s commercial choice, not a viability argument.

Independent assessment as statutory duty on contest. Where a developer initiates a s.75A modification application on viability grounds, the planning authority must commission independent assessment from the District Valuer or equivalent. The Circular 4/2025 para 40 discretion becomes mandatory. The developer’s viability model is submitted open-book. The District Valuer’s assessment is the operative document for the determination.

DRLT excluded from viability calculations. DRLT must be explicitly excluded by statute from S75 viability assessments. The reasoning is set out in the Proposal 1 section of the main paper and in Annex 3 below. It is restated here for completeness: the instrument enforcing S75 delivery cannot simultaneously become the cost argument that reduces S75 obligations.

These three provisions close the mechanism Foye (2022) identifies. The s.75A route remains legally available but operates in a structurally different environment from the one that hollowed out Section 106.

Procedural design at the point of exercise

Each individual exercise of pre-emption is its own decision: this site, this holder, this circumstance. A general proportionality argument about the Bill is not the same as a demonstrated proportionality decision on a specific site.

Pairc Crofters Ltd v Scottish Ministers [2012] CSIH 96 established the requirement directly. The Court of Session Inner House upheld the community right-to-buy provisions of the Land Reform (Scotland) Act 2003 because the statute imposed sufficient procedural conditions before a community could acquire land: a public interest test, an opportunity for the landowner to rebut, and an appeal route. Each individual exercise had to satisfy these conditions visibly on the face of the decision. The Community Empowerment (Scotland) Act 2015 (ss 62-73, commenced February 2021) codified this, adding written reasons, prescribed notification, and refined application procedures.

The pre-emption architecture must meet the same standard. Three features should sit on the face of the legislation.

Statutory duty to publish a proportionality decision. When pre-emption is exercised on a specific site, the body exercising the power must publish a written decision setting out: the housing-supply objective being pursued by the acquisition and the evidence base for it; the alternatives considered (negotiated purchase, deferred exercise, continued reliance on DRLT alone); the holder’s circumstances as known to the body, including any force majeure history or attempted commitment; the reason EUV rather than market value is proportionate in this case; and a reasoned conclusion on fair balance. The duty applies to every exercise without exception. A body that exercises the power without publishing the decision has not lawfully exercised it.

Structured format prescribed in secondary legislation. The format is set in regulations under the Act, not left to administrative discretion. A standardised template means every assessment is documented identically across the country, comparable across exercises, and not capable of being discharged by a one-line certificate. It front-loads care into the decision process: officials know in advance what evidence and reasoning the decision must carry, which forces the assessment work upstream of the decision rather than leaving it to be reconstructed in hindsight.

Right of appeal on proportionality grounds. A holder against whom pre-emption is exercised has a statutory right to appeal the proportionality decision to a defined forum within a specified period. The appeal sits alongside the independent EUV valuation rather than re-litigating the principle of pre-emption itself: it tests whether the body’s proportionality reasoning was structured, evidenced, and reasonably conducted, not whether the underlying policy is sound. Forum (the Scottish Land Court and the Lands Tribunal for Scotland are both natural candidates given existing tenure and CPO appeal remit) and procedural specifics are for the Bill team.

These three features make the fair-balance argument operational at the level of individual exercises, not just the legislation as a whole. The proposal does not pretend the underlying decision is easy on the holder. It requires the body imposing it to show its working.

Transitional anti-avoidance: pre-commencement disposals

The declaration window architecture operates on the consented-but-unbuilt stock as it exists at commencement. It does not address the period between the public signal of policy intent and Royal Assent, during which landholders who have read the policy direction correctly could transfer allocated sites at hope value before the pre-emption architecture takes effect. The concern is not hypothetical: property rights reform legislation has a consistent history of forestalling transactions, and Scottish land law practitioners will identify the window without prompting.

The mechanism for closing it is the announcement-date anti-avoidance provision. SDLT forestalling provisions have operated from Budget announcement date rather than Royal Assent since the Finance Act 2003 because the gap between announcement and enactment is otherwise a systematic arbitrage opportunity. The mechanism proposed here applies the same logic at a policy level across the LDP-allocated stock.

Design: clawback, not voidance. The correct mechanism is a seller-side clawback obligation rather than voidance of the completed transaction. Voidance would impose an unexpected burden on a buyer who completed in good faith at market price and may have made investment decisions on the basis of acquired title. Clawback targets the party who captured the windfall: the seller who disposed of LDP-allocated land at above-EUV consideration during the pre-commencement window. The obligation is: if the Bill receives Royal Assent, the seller must account to More Homes Scotland for the excess of the consideration received over independently-assessed EUV as at the date of disposal, within 90 days of Royal Assent. MHS may waive the obligation where the seller demonstrates the disposal was at or below EUV. The buyer’s title is undisturbed; the buyer does, however, enter the declaration window at commencement in the same position as any other holder of LDP-allocated land, having acquired at a price that now carries pre-commencement risk that the market should already have priced in.

Trigger date. The provision should attach to disposals completed on or after the date of the Scottish Government’s first public consultation on the pre-emption architecture. A holder who disposes after an explicit public policy signal cannot credibly claim legitimate expectation of hope value equivalent to one who acted before any public statement of intent. The consultation document makes the policy direction explicit; post-consultation disposals are at the informed end of the market.

Bill introduction is an alternative trigger: a cleaner parliamentary date, less contestable on legitimate expectation for individual holders who may not have read the consultation. The trade-off is a 12-to-18-month forestalling window open to institutional holders who will have read it. A tiered trigger (consultation for institutional holders, introduction for natural persons) is one resolution; the Bill team’s choice should be informed by advice on the legitimate expectation exposure profile across holder categories.

Condition and lapse. The clawback obligation should lapse automatically if the Bill does not receive Royal Assent within 48 months of the trigger date, accommodating the realistic Scottish parliamentary timetable with headroom. If the policy is abandoned, materially amended, or delayed beyond the window, no clawback is sought and the lapsed obligation cannot be revived.

Scope and exclusions. The obligation applies to disposals of LDP-allocated residential land where the consideration exceeds independently-assessed EUV at the date of disposal. Excluded from scope: disposals to registered social landlords, local authorities, or More Homes Scotland (these parties are delivering the housing policy objective, not arbitraging the window); disposals that satisfy the build-out commitment criteria that would have qualified under the declaration window (genuine developers disposing as part of a delivery programme); and transactions where the price differential between consideration and EUV is below a de minimis threshold set in regulations (to exclude small-scale family and personal transactions from the compliance burden). Intra-group transfers and linked transactions between related parties should be treated as single disposals assessed on total consideration to prevent fragmentation as an avoidance route.

A1P1 position. Pressos Compania Naviera SA v Belgium (1995) 21 EHRR 301 held that retroactive legislation is not automatically incompatible with A1P1 provided the public interest is sufficiently weighty, the means proportionate, and the interference precisely bounded. The argument believed to be available is that the clawback architecture meets those conditions: it captures only the windfall above EUV, leaves the buyer’s position intact, and is bounded at both ends. The legitimate expectation argument is weakest for institutional holders with professional advice who disposed after the consultation was published; the exclusions above concentrate the obligation on that cohort. This argument has not had legal review; the provision carries more legal risk than the prospective pre-emption architecture and counsel should assess the exposure specifically.

The provision is the structural complement to the declaration window: a declaration window that can be circumvented by bulk disposals in the pre-commencement period is not a declaration window.

Future allocations: steady-state regime

The declaration window and challenge surfaces above apply to the transition period: the consented-but-unbuilt stock, where holders formed expectations under the pre-Act regime.

For LDP allocations made after commencement, the position is simpler. EUV pre-emption attaches from the date of allocation, before any planning permission is sought and before any private investment is made. The allocation recipient knows from day one that hope value is not available on future disposal. A claimant who accepts an allocation on those terms and later challenges the pre-emption condition would be challenging a burden they voluntarily accepted, not a retrospective interference with an established property right. Individual exercises still require the Pairc Crofters proportionality demonstration.

The practical consequence: the transition-period challenge exposure is time-bounded. As the fraction of the 164,000 that falls within DRLT scope works through the declaration window and subsequent disposal, that exposure diminishes. The steady-state regime carries a lower challenge profile. Counsel should confirm this reading.

Domestic law: Scotland Act 1998 and Human Rights Act 1998

Section 29(2)(d) of the Scotland Act 1998 renders provisions of Acts of the Scottish Parliament outside competence if incompatible with Convention rights. A Bill embodying the pre-emption architecture requires a statement of compatibility from the Scottish Law Officers before introduction and a Human Rights Act 1998 section 6 compatibility statement from the Scottish Government. Whether those statements can be made is a question for the Law Officers, not this paper. The Scottish Land Commission’s 2018 Land Value Capture report examined German and Dutch land acquisition models operating in ECHR jurisdictions and found them compatible with A1P1 principles; that analysis is the domestic starting point for the compatibility assessment.

The founding legislation should require a review of Convention compliance at year three of operation, drawing on the first cohort of declaration-window outcomes, to verify the cumulative effect of DRLT-plus-pre-emption has not drifted beyond the proportionality margins the original compatibility statement established.

CPO precedent and comparative coerciveness

Scotland already operates compulsory purchase powers under the Town and Country Planning (Scotland) Act 1997 and the Land Compensation (Scotland) Act 1963. CPO is more coercive than statutory pre-emption in every legally and operationally material respect. CPO removes the owner’s ability to decline the sale entirely. It operates without a declaration window or any equivalent prior-commitment route. It proceeds at a compensation figure that can be contested through the Lands Tribunal for Scotland but cannot be refused by the owner. Contested CPO valuations regularly run three to five years through Lands Tribunal proceedings before compensation is settled; the legal and expert costs on both sides frequently exceed the value in dispute on smaller sites.

Pre-emption operates at EUV plus against a seller who chose not to commit during a 24-month window, and the seller’s agency was preserved throughout the process that led to pre-emption attaching. Opponents who characterise statutory pre-emption as an unacceptable interference with property rights while accepting CPO as routine are not making a principled legal argument; they are making a political one. The pre-emption instrument is cleaner for the purpose of disciplining speculative land holding precisely because it does not remove owner agency and does not require the state to initiate a compulsory process. Where land banking is the target, pre-emption is the proportionate instrument; CPO is the blunt one.

The Scottish Government’s own compulsory purchase reform consultation (2025, Compulsory Purchase Bill planned for the next Parliament) accepts that the existing CPO regime needs modernisation. Pre-emption sits alongside reformed CPO rather than displacing it: CPO compels disposal where the owner refuses; the general pre-emption right attaches at voluntary disposal by an owner who declined the commitment route. S75-breach pre-emption operates on a different trigger (missed delivery deadline) but is still less coercive than CPO: it exercises a right the holder accepted at consent. Pre-emption legislation does not require CPO reform to be in place first.

Compulsory Sale Orders: relationship to the pre-emption architecture

The Scottish Land Commission has called for Compulsory Sale Order powers in Scotland for several years, and CSOs feature in CaCHE’s March 2026 briefing as part of the toolkit interviewees considered appropriate for the More Homes Scotland agency. CSO and statutory pre-emption at EUV plus are structurally adjacent. CSO compels disposal of land held in defiance of identified policy objectives; the general pre-emption right attaches at voluntary disposal, with the holder retaining ownership otherwise. S75-breach pre-emption occupies a middle position: the trigger is a missed contractual deadline, not a discretionary policy decision, and the holder accepted the obligation at consent.

Proposal 2 chooses pre-emption over CSO on three grounds. First, pre-emption preserves owner agency until the moment the owner chooses to sell, which is the foundation of the A1P1 fair-balance argument set out above. CSO requires its own proportionality framework on each compelled sale, and the Pairc Crofters doctrinal reading would apply to each individual exercise as it does to pre-emption. Second, the declaration window in Proposal 2 already does most of the discriminatory work CSO would do: holders who commit retain market-value rights; holders who decline accept prospective EUV-plus pre-emption on disposal. Third, the DRLT escalator running alongside makes indefinite holding increasingly expensive, forcing disposal without compulsion.

The two instruments are not mutually exclusive. The recommendation is that the legislation commence with pre-emption and let evidence over the first parliamentary cycle determine whether CSO needs to be added.

Corporate structure anti-avoidance

The general pre-emption right attaches at the moment of voluntary disposal of LDP-allocated land. In conventional property law, a disposal is the transfer of a legal or beneficial interest in land itself. A transfer of shares in a company that holds LDP-allocated land does not, under that analysis, constitute a disposal of land: title remains with the company, and the company’s obligations run with the land, but the change of beneficial ownership at the shareholder level is not captured.

This is the most commercially sophisticated avoidance route available to institutional and corporate landholders. A fund or estate holding LDP-allocated sites through a special purpose vehicle can sell 100% of the SPV’s shares at a price that fully reflects hope value (achieving the economic equivalent of a hope-value land disposal) without triggering the pre-emption right. Scottish land law practitioners and commercial property advisers will identify this structure immediately.

The legislation should treat a change of control in a company holding LDP-allocated land as a qualifying disposal for the purposes of both the transitional clawback and the ongoing pre-emption right. Change of control should be defined in the Bill, with the threshold set at acquisition of a majority interest or any interest conferring effective operational control. The consideration for the deemed disposal is the proportion of the company’s consideration attributable to the LDP-allocated land component of its assets, assessed independently.

Three design features constrain the provision against unintended reach. First, internal corporate reorganisations where both the disposing entity and the acquiring entity are wholly owned subsidiaries of the same ultimate parent are excluded (genuine intra-group restructuring, no change in beneficial ownership at the group level). Second, the provision applies only where LDP-allocated land constitutes a material proportion of the company’s assets (a threshold of 50% by value at the date of the transaction, to exclude companies where the land holding is incidental to a wider business). Third, the deemed-disposal trigger applies only to voluntary transactions; it does not operate on insolvency-related transfers, secured creditor enforcement, or court-ordered transfers, which have their own proportionality considerations.

Without this provision, the declaration window and pre-emption architecture are commercially porous for any holder with access to corporate structuring advice. With it, the economic substance of a hope-value exit is captured regardless of the legal vehicle through which it is achieved.

DRLT non-suspension under legal challenge

General rule. DRLT liability accrues and is collectible regardless of any pending legal challenge to the enabling legislation, to an individual site assessment, or to a related instrument. The principle is the same one that governs contested income tax assessments: liability runs; the taxpayer’s remedy is to challenge through the designated tribunal; accrual is not suspended by the challenge. A holder who is actively building sees the DRLT base shrink with each completed hectare; the charge on the remaining undeveloped portion is the ordinary carrying cost of land not yet in productive use, not a penalty. The non-suspension rule falls most heavily on holders who have neither built nor disposed, the cohort whose S75 obligations remain unperformed.

Pre-emption independence. DRLT and statutory pre-emption are separate instruments with independent statutory bases. A legal challenge to the pre-emption architecture does not constitute a challenge to DRLT, and an injunction restraining the exercise of pre-emption on a specific site does not suspend DRLT liability on that site or any other.

This independence is architecturally significant. A landowner facing pre-emption on voluntary disposal has an incentive to challenge the pre-emption right at the earliest opportunity and to argue, in parallel, that DRLT liability should be suspended pending the outcome of that challenge, on the basis that the two instruments form an integrated coercive scheme and cannot be assessed independently. Courts should reject that argument, and the legislation should be drafted to make rejection straightforward.

Three elements support the independence case. First, DRLT is levied on the annual holding of LDP-allocated land, independently of any decision the holder makes or does not make about disposal. It is a recurring tax on a status (holding allocated land without delivery), not a condition of any transactional right. Pre-emption is exercisable at two points: voluntary disposal, and S75 breach at the delivery deadline. Neither trigger depends on DRLT liability; DRLT depends on neither. The instruments operate at different points in the holder’s decision sequence and are triggered by different events. Second, the proportionality of DRLT can be assessed entirely without reference to pre-emption: it is a holding-cost discipline with a rate schedule calibrated against the LDP delivery cycle, a ceiling constrained by A1P1 proportionality, and a statutory review mechanism. None of those features depends on the pre-emption architecture being in force. Third, the judicial review jurisdiction and the Lands Tribunal jurisdiction that pre-emption challenges will engage are separate from the tax tribunal jurisdiction that DRLT challenges would engage; the same claimant cannot seek relief across both jurisdictions from a single set of proceedings without separate originating actions in each.

The legislation should include an express provision stating that DRLT liability is unaffected by any legal challenge to the enabling Act, to an individual assessment, to the exercise of the pre-emption right, or to any planning decision underlying the LDP allocation, and that no such proceedings constitute grounds for deferral or suspension of DRLT assessment or collection.

Annex 2: Active Use Exemptions

The DRLT attaches from the date of LDP allocation because that is the moment the public planning decision creates the speculative value. Two categories of holder are caught by that trigger despite having no speculative intent: farming families whose land sits inside a growth-area boundary, and owners operating non-residential economic activity on sites allocated residential. Both require exemption. Without it, the instrument is both unjust to the holder concerned and politically corrosive to the Bill itself.

Agricultural use

The architectural requirements are three. First, the exemption must be defined against verifiable public records held by Rural Payments Scotland (Basic Payment Scheme and its successor, cross-compliance records, agri-environment scheme participation), not against landowner self-certification. The Irish RZLT experience shows that self-certification routes attract opportunistic claims at scale and create a gap advisors exploit systematically. Second, the exemption must lift on identifiable triggers (grant of residential planning permission, voluntary disposal, or verified cessation of active agricultural production), with a force majeure provision for verified disruption such as movement restrictions or extreme weather. Third, the declaration window in Proposal 2 must accommodate the farming holder who has neither applied for planning permission nor sought to sell: their declaration of continuing active agricultural use should satisfy the window’s requirement, because the farming holder has not captured a speculative windfall and the DRLT is built to target the windfall.

The Bill should define active production by reference to the Basic Payment Scheme eligibility criteria, requiring land to meet Good Agricultural and Environmental Condition (GAEC) standards and to be maintained in a state suitable for grazing or cultivation without preparatory action beyond normal agricultural methods. RPS registration alone is insufficient: a site carrying BPS registration but no productive activity above the BPS minimum threshold is not in active agricultural production for DRLT purposes. Without an intensity test, a landowner holding allocated residential land can preserve the exemption indefinitely through nominal grazing at negligible cost.

Detailed eligibility criteria, verification protocols, the data-sharing mechanism between Revenue Scotland and Rural Payments Scotland, and the de-allocation route for land incorrectly included in a residential LDP allocation are properly the work of parliamentary counsel and Rural Payments Scotland.

Non-residential economic use

The same architectural logic applies to land in active non-residential economic use: a workshop, a distribution warehouse, a functioning office. The owner may not have sought or welcomed the residential allocation. The LDP process includes consultation, but allocation decisions are made by the planning authority and an owner who objects may be overridden where the authority’s spatial strategy requires the site. Taxing a going concern for a speculative option it did not request is the same A1P1 defect the agricultural exemption exists to prevent.

Verification is against Non-Domestic Rates records held by the Scottish Assessors Association. A site carrying a live NDR valuation for an active hereditament is in productive economic use; one carrying a nil or nominal valuation, or where the hereditament has been struck from the roll, is not. NDR records are centrally held, publicly maintained, and updated on a three-year revaluation cycle independent of the landowner. The verification surface is comparable in reliability to the Rural Payments Scotland route for agriculture and carries the same independence from self-certification.

The exemption applies to the curtilage of the hereditament as recorded in the valuation roll, not to the entire LDP allocation. A 0.2-hectare workshop on a 10-hectare allocated site exempts the workshop and its operational curtilage; the remaining 9.8 hectares sit in the DRLT taxable base. Without this apportionment rule, a token economic use on a fraction of a site shelters the whole allocation from the holding cost the instrument exists to impose. The Scottish Assessors Association already records curtilage boundaries for each hereditament; no new survey or data source is required. Where the curtilage boundary is disputed, the Lands Valuation Appeal Court is the existing tribunal.

Liability triggers mirror the agricultural exemption: grant of residential planning permission, voluntary disposal, or verified cessation of the existing economic use. A holder who demolishes the operating premises or allows NDR liability to lapse has made a commercial decision that the site’s residential development value exceeds its productive use value. The DRLT attaches at that point because the holder has crossed from productive use into speculative holding.

The declaration window accommodation applies identically: a holder in active economic use satisfies the Proposal 2 declaration requirement by evidencing continuing operation, because they have not captured a speculative windfall.

Residual acquisition of active economic use sites

The economic use exemption creates a secondary design problem. Before the reform, hope value gave the active-use holder a rational incentive to sell: the residential development premium on their site substantially exceeded the value of the premises as a going concern. EUV pre-emption compresses that premium. The exemption removes the DRLT holding cost. The holder now has no financial incentive to release the site voluntarily.

For sites the planning authority needs for housing delivery, CPO is the residual mechanism. The reform makes this affordable. With hope value compressed across the board, CPO compensation reflects the productive use value of the site (EUV plus statutory disturbance covering actual relocation costs, business interruption, and fitting-out of replacement premises), not the speculative residential development potential the holder never sought. This is the economics of the 1948-1959 regime: nationalised development rights removed hope value from CPO compensation, and councils acquired land at prices that made large-scale housing programmes deliverable. The reform reconstructs those economics without formally nationalising development rights.

The precondition for exercising CPO on an active economic use site is that the LDP provides for relocation of the displaced use. A council that allocates a workshop site for residential use must have designated suitable commercial or light-industrial land nearby. The new towns programme operated on this principle: East Kilbride, Livingston, and Cumbernauld designated employment land before the first house was built because the development corporations knew existing economic uses would need to be accommodated. A CPO that displaces a going concern without providing a relocation destination is not a planning decision; it is an eviction.

Agricultural use is structurally distinct from active economic use on this point. Commercial and industrial activity can be relocated within the LDP because the council controls spatial allocation of employment land. Agricultural use cannot be relocated because the planning system cannot designate new farmland. The agricultural exemption from DRLT and pre-emption is therefore absolute for the duration of active production. Once production ceases or the holder disposes, the site enters the standard regime. CPO at agricultural value remains available throughout, and under a regime that has stripped the speculative premium from Scottish land, CPO becomes the likely primary acquisition route rather than the residual one. Volume housebuilders with UK-wide portfolios will redirect capital to jurisdictions where hope value still exists; councils acquiring allocated agricultural land at EUV through CPO and tendering construction is the predictable steady-state outcome.

The three-tier treatment (speculative holding, active economic use, active agricultural use) reflects the fact that LDP allocations fall on sites in different states of use, and each state requires a distinct combination of pressure, exemption, and acquisition mechanism.

The architectural point across both exemptions is that a verifiable, non-self-certifying test distinguishes the productive holder from the speculative cohort. The Bill’s defensibility depends on getting that distinction right.

Annex 3: DRLT Revenue Projection

The DRLT revenue trajectory shown in the rate calibration table below is derived from the following methodology. These are indicative ranges, not modelled outputs.

Base stock. Scotland’s LDP-designated residential land stock is broadly comparable in scope to Ireland’s residential zoned land stock. This is a rough comparator: Scotland has a different planning architecture, different population distribution, and a different balance between urban and rural designated land. The comparison is used only to establish a plausible starting range.

Rate calibration. Ireland’s RZLT runs at a flat 3% of each site’s assessed market value. Scotland’s DRLT applies its rate to a geographic zone band benchmark instead, a coarse published proxy for land value per hectare, not a bespoke per-site valuation. The architecture uses a phased escalator: a low rate while the regime beds in and infrastructure delivery is programmed, rising to a meaningful holding cost as the LDP cycle matures. The ceiling is an ECHR A1P1 proportionality constraint; the specific rate at ceiling and the step schedule are matters for the Scottish Fiscal Commission, Law Officers, and Bill team, not design commitments from this paper. The following is an illustrative schedule calibrated to a ten-year Local Development Plan cycle (years one to three covering the early infrastructure window, years four to ten the build-out window) to demonstrate the revenue envelope and behavioural logic at plausible rate levels. Illustrative receipts at each phase:

Year 1 (0.5%): £5 to £15 million indicative annual receipts

Year 2 (1.0%): £15 to £30 million

Year 3 (2.0%): £25 to £60 million

Year 4 (3.0%): £30 to £80 million

Year 5 (4.0%): £30 to £90 million

Year 6 (5.0%): £25 to £100 million

Year 7 (6.0%, ceiling): £20 to £120 million

Year 8 onwards (6.0%, held): declining as land enters productive use

A statutory rate-review is conducted before the Year 4 rate step, gating any adjustment to the schedule from Year 4 onwards. The review provision is set out below.

The empirical anchor is Ireland’s first-year collection: approximately €40 million on roughly 1,800 returns by September 2025. That figure covers the seven to eight months between RZLT going live in February 2025 and the published data; annualised, the Irish base sits closer to €60 million. Scottish stock is treated as broadly proportionate to Irish stock for working-comparator purposes only. The planning architectures, population distributions, and rural-to-urban allocation balances differ materially, and the comparison establishes a plausible range rather than a forecast.

Net receipts after exemptions and suspensions. The ranges above are gross figures for the full taxable base. Net receipts will be lower because three categories of land are excluded or suspended before the charge applies. Agricultural land in active production is exempt (Annex 2; verified against Rural Payments Scotland records). Land in active non-residential economic use is exempt (Annex 2; verified against NDR valuation on a three-year revaluation cycle). Infrastructure-blocked sites are suspended (see below). Rural and island sites may be suspended for renewable two-year periods (see below). Every mechanism that achieves the policy goal (building, exempting productive use, identifying an infrastructure block) simultaneously removes revenue from the projection. The figures that survive all exclusions represent the annual charge on residential-allocated land that carries no residence, no active exemption, and no verified suspension. That is the stock the instrument is designed to move. As it moves, receipts decline. This is the success signal, not a revenue shortfall.

The width of each year’s range reflects two sources of variance that cannot be modelled in advance: the proportion of the taxable base that exits through build-out completion in any given year, and the proportion suspended for infrastructure. From year five onwards, two opposing forces shape receipts: the rate continues to rise while the taxable base shrinks as sites complete and enter productive use. Receipts plateau and then decline once base-shrinkage outpaces rate-rise, which on a working instrument is around year five or six. The post-year-seven figures assume residual stock at the illustrative 6% ceiling: if the instrument is working as designed, those figures decline through year ten and approach a low residual as the LDP cycle completes; if speculative holdings persist, the higher end of the range is the upper bound.

Direction of travel. DRLT receipts are real, recurring revenue from the taxable base for as long as it exists. They are not designed to be written off or rebated. A declining revenue trajectory is the success metric: it means land is being built on. If year-five revenue is £40 million rather than £90 million, the appropriate interpretation is that build-out is outpacing the escalator, not that the instrument has underperformed. Budget modelling should treat DRLT receipts as volatile and declining over the LDP cycle: the year-one-to-three hypothecation to SHFF seed equity absorbs the early receipts (which are real but not large) and routes them to productive use while the taxable base is still at its widest. A Finance model that projects mature-rate receipts as a recurring baseline has misread the instrument’s design purpose.

Infrastructure gateway and revenue suppression. The infrastructure viability gateway described in Proposal 1 of the main paper removes sites from the DRLT taxable base for the duration of their suspension. A site confirmed as absent from the named statutory capital programmes of Scottish Water, SSEN, Transport Scotland, or SEPA stops generating DRLT receipts from the point of confirmation until infrastructure is programmed with a binding delivery date. The primary determination is Revenue Scotland’s register-based lookup (see below). Where a holder applies to the local authority for a supplementary suspension on local infrastructure grounds outside the four statutory registers, and the local authority fails to determine the application within 60 days of receipt, the application is treated as refused and the DRLT runs from the application date. Passive non-determination should not shelter a site: the gateway is a route for sites with verified infrastructure constraints, not a deactivation switch a council can leave on by inaction.

The determination itself should not depend on 32 planning departments each conducting a fresh investigation under volume pressure. The Bill should require each statutory undertaker (Scottish Water, SSEN, Transport Scotland, SEPA) to publish and maintain a capital programme register of committed infrastructure delivery dates by site. Revenue Scotland, which administers DRLT and holds the register of liable sites, performs the primary gateway determination by matching each site against the published programmes. Where the registers show no committed delivery date, Revenue Scotland grants the suspension directly; no local authority application is required for the four-register check.

The local authority’s role is limited to a supplementary determination on local infrastructure constraints outside the four statutory registers (road capacity, local drainage, site-specific conditions). Revenue Scotland cannot override a local authority’s supplementary determination, and the local authority cannot override Revenue Scotland’s register-based determination: the two operate on different questions. The 60-day deemed-refusal provision applies to the local authority’s supplementary determination only.

The revenue ranges above do not adjust for this effect because the proportion of LDP stock that is genuinely infrastructure-blocked is not publicly quantified. A working assumption of approximately 40% has been used as an indicative modelling parameter in this Annex. It is not drawn from primary data and should not be cited authoritatively until it has been derived from disaggregation of the Housing Land Audit against the statutory capital programmes of Scottish Water, SSEN, Transport Scotland, and SEPA. The best available public evidence is the Scottish Government’s Short Life Working Group on stalled sites (2025-26), which identified approximately 20,000 homes on 114 named sites where infrastructure was among the stalling factors; that figure reflects a curated intervention set, not a census of the 164,000-home consented pipeline. The actual proportion of the pipeline blocked by infrastructure is a question for pre-legislative scrutiny, and the Scottish Government’s Housing Land Audit data is the correct source to interrogate, but the direction is clear: year-one to year-three receipts will sit toward the lower end of the ranges above, partly because early gateway applications will include the most severely blocked sites, and partly because those sites will not have had time to cycle through the statutory capital programming process. A Finance model that treats the gap between the lower-bound forecast and outturn as a policy failure has misread the mechanism. Gateway suspensions are a leading indicator of the instrument working: they confirm that the LA has identified and logged a real infrastructure gap, and they reset when that gap is closed. The suspended sites then rejoin the taxable base at whatever rate applies at reentry, or the holder builds and exits the base. Neither outcome is a revenue shortfall. The year-three statutory rate-review should include gateway-suspension data as a correction to the raw receipts figure, to avoid misreading suspended sites as instrument underperformance.

Rural and island suspension. A local authority may apply to Scottish Ministers to suspend DRLT on a specific site for a renewable two-year period where the site is in a rural or island area and no planning permission has been granted or applied for within the preceding five years. The mechanism addresses allocations in areas where the LDP designation was aspirational and no developer, large or small, has sought to build. Without it, DRLT operates as a deadweight penalty on family-owned or small-scale holdings that were never held for speculation, in areas where no speculative market exists. The proportionality case under A1P1 requires this: an annual charge whose purpose is behavioural change is disproportionate where the holder has no realistic path to the behaviour the charge is designed to produce.

The suspension is renewable on reapplication every two years. Renewal requires the LA to stand behind the allocation while simultaneously confirming that the market will not deliver on it: an uncomfortable position that self-corrects toward de-allocation. If an allocation has attracted no planning application in five years and the LA applies for a second or third suspension, the allocation was wrong. The remedy is de-allocation in the next LDP review, which removes the site from the DRLT base entirely. The suspension bridges the gap between a planning authority recognising an allocation failure and the LDP review cycle catching up to correct it.

The revenue modelling above does not adjust for this suspension because the number of affected sites cannot be estimated without site-level cross-referencing against the Housing Land Audit. In rural and island local authority areas, the proportion may be material. As with the infrastructure gateway, this is a pre-legislative scrutiny question: the Scottish Government should quantify the intersection before the rate schedule is finalised.

Statutory rate-review provision. The rate calibration above sits on the same evidential footing as the 164,000 aggregate the consultation rests on: a working start point derived from the closest available comparator, not a modelled parameter. The honest position is to acknowledge this in the legislation by writing in a statutory rate-review at year three of operation, conducted before the year-four rate step. The timing is deliberate: year three is the first full year in which the taxable base, the exemption and suspension registers, and the build-out completion rate are all observable in the data. Reviewing the schedule at year three captures the evidence before the rate crosses from early notice to persuasion territory at year four, and gates any adjustment before the escalator moves into the range where it is designed to force disposal or delivery.

The review should be conducted by the Scottish Fiscal Commission, not by the Scottish Government, Revenue Scotland, or the sponsoring Minister. The SFC has no revenue interest: it is institutionally indifferent to whether the rate rises, holds, or falls. Three defined metrics should govern the review: delivery rate (the proportion of the taxable base that has exited through build-out completion by year three, indicating whether the escalator is producing genuine construction activity); LA infrastructure designation rate (the proportion of gateway-suspended sites that have re-entered the taxable base, indicating whether the statutory bodies are clearing the infrastructure backlog or using gateway status as a quiet deactivation mechanism); and revenue trajectory against the published working ranges, read with the gateway-suspension correction described above. Any adjustment to the escalator from year four onwards requires affirmative parliamentary approval on the basis of the SFC’s published report. The rate is not adjustable by ministers between reviews, and not adjustable at all without a parliamentary vote.

This architecture answers the “tax by stealth” objection structurally rather than rhetorically. A rate a minister can adjust in secondary legislation without a debate is a rate the government can ratchet quietly when fiscal pressures require. A rate Parliament votes on, against a published SFC report, against defined metrics, in a chamber with Opposition members who will read the data independently, is not. The transparency landowners need to model exposure is preserved by setting the rate at enactment; the honesty about its evidential basis is preserved by writing the review into statute; the independence of the reviewer is what closes the political attack. A Bill that sets the rate without a review provision will face the objection that a single year of comparator data cannot support a rate intended to operate for a decade. A Bill that sets the review procedure without specifying the SFC and the metrics will face the objection that the procedure is a political fig leaf.

The illustrative 6% ceiling. The ceiling is set at 6% in this example on three grounds. First, the schedule is calibrated to the ten-year LDP cycle: the rate reaches its terminal level at year seven, leaving a meaningful build-out window inside an LDP under terminal-rate cost pressure. Second, 6% on land value is a serious annual carrying cost on speculative holding without crossing into effective confiscation: a holder facing 6% per year against any plausible market discount rate will dispose, build, or accept terminal write-down, but will not have the property taken. Third, the margin of appreciation doctrine gives national legislatures wide discretion on social and economic policy measures provided the proportionality architecture is sound; a ceiling at this level is designed to sit inside that discretion, but whether it does is a judgment for the courts, not a conclusion the paper can assert. The proportionality argument the legislation will rest on requires the ceiling to be visible, the exemption and suspension architecture to be operable, the rate phase-in to give holders time to act, and the Pairc Crofters-style proportionality on individual exercise to do the work; the author reads a ceiling in this range as the level at which the instrument bites while the proportionality argument remains available.

Irish implementation frictions and their DRLT design responses. Ireland encountered four categories of friction in RZLT’s first year. Each maps to a specific DRLT design feature that forecloses it.

First, a flooded rezoning-request window: landowners applied to remove residential zoning to escape the tax base. DRLT’s response is structural: LDP allocations are a planning authority decision, not a landowner opt-in. The de-allocation route exists (rural and island suspension leading to LDP review) but is controlled by the local authority, not triggered by the landowner.

Second, gaming on deferral applications via token commencement: developers broke ground to trigger the RZLT deferral without committing to a build programme. DRLT eliminates this entirely. The non-rebatable design (below) means commencement is irrelevant; only completion of each hectare exits the taxable base. Token commencement confers no advantage.

Third, scope-boundary disputes at the urban-rural margin: landowners contested whether their land fell within the residential zoning boundary. DRLT’s boundary is the LDP allocation as registered, verifiable against Ordnance Survey data and the Land Register. Boundary disputes are planning disputes, resolved through the existing LDP objection and review process, not through the tax administration.

Fourth, live legal challenges to the architecture by affected landowners. Scotland should expect A1P1 challenges and has designed for them: the phased escalator with a proportionality ceiling, the SFC rate-review, the agricultural and active-use exemptions, the infrastructure gateway, and the rural/island suspension all serve as the fair-balance architecture the paper’s case-law and proportionality treatment in Annex 1 addresses.

Ireland’s Budget 2026 introduced a new exemption route for genuine economic activity on residentially-zoned land, primarily targeted at farming, with applications between 1 February and 1 April 2026. A verifiable active-use exemption is the design Ireland has converged on after the 2025 friction year (Annex 2), and Scotland should legislate it from the start rather than rediscover the need.

Why DRLT is non-rebatable. Ireland’s RZLT defers the entire tax charge during active residential development and abates it entirely on completion within the planning permission window (s.653AH TCA 1997; Revenue Commissioners, Part 22A-01-01, January 2026). Scotland’s DRLT should not replicate this.

A rebatable DRLT creates a managed-delay loophole. A volume builder who commits to a build-out schedule pays only the nominal rate during construction, with the escalated portion deferred and abated on completion. That builder controls the build rate. At 25 units per year on a site consented for 220, construction runs for a decade or more. Throughout, the builder borrows against the appreciating land value while the deferred DRLT accumulates as a notional liability that never bites. On completion, however many years later, the entire deferred balance is written off. The instrument cannot distinguish between a builder who completes in three years and one who completes in twenty, and any time limit on the abatement is immediately challengeable under the penalty doctrine (Cavendish Square Holding BV v Makdessi [2015] UKSC 67).

The non-rebatable design eliminates the distinction. DRLT is an annual charge on residential-allocated land that carries no residence. It applies to every hectare in the taxable base at the prevailing rate. Each hectare exits the base on completion of the residence it was allocated for. A builder who completes faster pays less in total. A builder who builds slowly pays more. The instrument responds to output, not intent, and does not require the state to assess whether a build programme is genuine or managed.

The “penalises builders as well as speculators” objection is answered by the per-hectare exit mechanism. The tax does not fall on building. It falls on not building. Every completed hectare leaves the base immediately. A builder actively constructing on a 12-hectare site pays DRLT on the 8 hectares not yet built, not on the 4 already completed. The carrying cost on the undeveloped portion is the instrument’s time pressure; the zero charge on the completed portion is its reward signal. No rebate is needed because the exit is immediate.

Charge on land. Unpaid Irish RZLT, including accrued interest and penalties, becomes a statutory charge on the land itself. The property cannot be sold free of this encumbrance without settling the outstanding liability (s.653Q(4), s.653AM TCA 1997). On sale, the vendor must file a return, pay all outstanding RZLT, and Revenue confirms the tax position for the purchaser before completion. Scotland’s DRLT should replicate this as a statutory consequence, not a discretionary enforcement step. A charge that runs with the land and survives transfer removes the exit strategy of selling the site to a related entity to shed the liability; it also removes the scenario in which a purchaser inherits a clean title while the tax obligation remains unrecoverable from the vendor. Registration in the Land Register makes the charge visible to any prospective purchaser and their solicitor at the point of title examination.

Implementation matters left to parliamentary counsel. Whether the escalator clock runs nationally from Act commencement or per-site from the later of LDP allocation and Act commencement is a matter for parliamentary counsel during Bill drafting; the rate schedule and revenue trajectory above hold either way. The per-hectare exit mechanism (each completed hectare leaves the taxable base on completion) requires a surveying and certification process that Revenue Scotland and the local authority building standards service will need to agree; this is an operational question, not an architectural one.

Annex 4: SHFF Capitalisation

The figures in this section are indicative ranges, drawn from comparator data and published Scottish costs. They are not modelled outputs.

Year-one acquisition programme. The primary acquisition pool is Scotland’s long-term empty residential stock: over 44,000 privately owned homes empty for more than six months, of which 32,337 have been empty for over a year (Scottish Government, September 2025). SHFF lists empties by estimated value using council tax records and Valuation Roll data, identifies owners through Registers of Scotland, and makes voluntary acquisition offers at condition-reflective prices. Acquired properties are refurbished to lettable standard and let as permanent tenancies to households currently in temporary accommodation.

As households leave B&Bs and emergency placements, freed temporary accommodation stock with conversion potential can be acquired and enter a secondary renovation pipeline: properties currently leased by local authorities for TA use that are worth acquiring permanently rather than returning to private landlords.

This is a capital transaction, not a grant payment: SHFF acquires an asset that generates rental income and is carried on its balance sheet as part of the revolving fund.

Indicative per-unit cost: £100,000 to £150,000 including acquisition and refurbishment, below the £207,000 average new-build social housing cost (Scotland’s Housing Network, 2024/25). Acquisition prices reflect condition discount on long-term empty stock; refurbishment costs vary by property state. A phased programme of 2,500 acquisitions across 18 months at an average all-in cost of £130,000 per unit produces a stock-acquisition deployment of approximately £325 million.

Year-one new-build component. A first pipeline of 250 new-build units at £200,000 per unit (below the £207,000 average gross cost reported by Scotland’s Housing Network (2024/25), reflecting land at EUV accessed through Proposal 2 pre-emption rather than at hope value) adds £50 million to the capital envelope.

Year-one working capital. SHFF in its first 18 months requires: staffing and governance costs for the founding coalition and management structure; Housing First specialist support contracts for the complex-needs cohort within the conversion programme (civil society organisations with Housing First delivery capacity as founding coalition members, not as contractors); legal and professional costs for the acquisition and refurbishment programme; and a working capital reserve against receipts timing mismatches between acquisition payment and rental income commencement. Indicative working capital envelope: £25 to £50 million in year one.

Total operational envelope. The full operational envelope reaches approximately £400 to £425 million as the 18-month acquisition programme closes in year two. Year-one capital actually deployed depends on the funding mix. From confirmed sources (direct state lending inside the financial transactions limit plus DRLT seed receipts) year-one deployment sits in the £80 to £120 million range, funding the first tranche of TA conversion acquisitions and set-up costs. The £200 to £400 million year-one range is achievable if HM Treasury classifies the Guarantee as contingent from the outset and the bond programme opens within the year, enabling a contingent borrowing tranche on top of confirmed sources. The year-one acquisition programme should be sized to the confirmed envelope, with the bond tranche enabling acceleration if it materialises on schedule.

Downside scenario: year-one deployment at 1,500 conversions. If bond market access does not materialise in year one (because HM Treasury classifies the Guarantee as scoring on PSND, or because institutional investor appetite is lower than projected) the year-one acquisition programme reduces to the confirmed funding envelope of £80 to £120 million. At £150,000 per unit, this funds approximately 1,500 conversions in 18 months rather than 2,500. The architecture remains viable on this trajectory. Direct state lending inside the financial transactions limit is the primary deployment mechanism, not a fallback; it does not require Guarantee operation. SHFF’s revolving income from 1,500 year-one acquisitions (approximately £5.4 million net per year from year two) is lower but still compounds. The bond-market timeline extends 12 to 18 months rather than operating in parallel from year one. The consequence is a slower ramp through years two to four, not a structural failure: the 30 to 50 per cent reduction in TA placements remains the decade target; the year it becomes achievable shifts right. This is the design-for-resilience case for sizing the year-one programme against confirmed funding rather than projected bond acceleration: an architecture that requires year-one bond issuance is more fragile than one that treats it as an accelerant.

Revolving fund mechanics and five-year scale-up. SHFF capital is not consumed in the way AHSP grant is. Acquisitions generate rental income from day one of permanent tenancy, recycled into the revolving fund. By year three, 2,500 acquired units (at a modelled average social rent of £500 per month, compared with the 2024/25 RSL average of £462 per month (SHR, 2025); the modelled figure assumes modest real-terms rent growth consistent with recent trends) generate approximately £15 million per year in gross rental receipts. After management and maintenance costs (indicatively 40% of gross rent), net receipts of roughly £9 million per year contribute to the revolving fund alongside DRLT seed receipts and direct state lending inside the financial transactions limit (the confirmed primary deployment mechanism in years one to three). Combined with Guarantee-backed bond issuance at near-sovereign rates under Proposal 4, a year-five NAV of £300 to £400 million gives SHFF the asset base to sustain its housing operations and service its own guaranteed borrowing. The MHS land margin (the difference between EUV acquisition cost and consent-value serviced-plot tender price) does not flow to SHFF; it capitalises the SHGF’s first-loss tranche and is the primary long-term capitalisation source for the guarantee book (Annex 5). The institutional separation matters: SHFF’s NAV determines what SHFF can deploy in housing, not the size of the guarantee book. The £1 billion revolving capital target by year five is achievable within this architecture without additional government grant beyond seed equity in years one to three.

Composition of SHFF’s balance sheet across time. SHFF’s balance sheet is housing assets (acquired TA stock, new-build units) plus working capital. Its gross asset base grows through revolving rental income and bond-funded acquisitions under the Guarantee. At year five, the £300 to £400 million NAV is predominantly productive housing capital. At year ten, the gross balance sheet moves above the £1 billion target as revolving income compounds and the bond programme scales. The first-loss reserve and guarantee book sit on the SHGF’s balance sheet, not SHFF’s (Annex 5 sets out the reserve sizing at each stage). SHFF’s contribution to the guarantee architecture is as a borrower: its financial health determines the quality of the guarantee book, not the size of the reserve backing it.

Unit economics: conversion against new build

The following table shows cost per unit by delivery model and what a £450 million illustrative envelope (rounding up from the £400 to £425 million operational range above) buys under each. Conversion costs include the £150,000 transfer price plus refurbishment (light-touch at £15,000 to deep at £35,000 to £40,000 per unit); new-build costs include land at EUV via Proposal 2, which adds negligible per-unit cost.

*Year-one planning assumption.

The year-one programme of 2,500 conversions and 250 new builds sits within the envelope at the mid-range refurbishment assumption, and the conversion-heavy weighting is deliberate. Conversion buys more units, faster, from capital that does not disappear when the tenancy starts.

Annex 5: Moral Hazard Controls

An open-ended guarantee is a blank cheque for defaults. Without explicit controls, the Guarantee creates perverse incentives: RSLs and SHFF can borrow aggressively knowing losses are socialised, and the SHGF has no ongoing incentive to monitor borrower financial health after issuance. The Dutch WSW architecture resolves this through three layered controls, and the layering matters: the controls do not operate independently but as a loss-absorption sequence. A precondition: the SHGF must be institutionally separate from SHFF. SHFF is the SHGF’s largest single borrower. A body that underwrites its own debt cannot credibly monitor its own covenant compliance; merging the two destroys the moral hazard architecture the controls below depend on. WSW works because it does not own housing. The guarantor and the guaranteed must be separate legal persons with separate boards and separate balance sheets.

Control 1: Leverage cap. Total guaranteed issuance is capped at three times the SHGF’s net asset value. This is a leverage ratio constraint analogous to financial regulation: the guarantee book cannot grow beyond what the SHGF’s own capital can credibly backstop at the first-loss layer. The cap binds the book to a moving figure as the SHGF’s NAV scales through member contributions, guarantee fees, SNIB seed capital, and MHS disposal revenue (the margin between EUV acquisition cost and consent-value plot disposal, which becomes the primary capitalisation source once the disposal pipeline is operational): at year one (SHGF NAV building from seed capitalisation), the cap limits the book to early-stage issuance; at year five, an SHGF NAV in the £300 to £400 million range supports a book of £900 million to £1.2 billion; at year ten (book at approximately £2 billion per Annex 6), implied SHGF NAV is in the £650 to £700 million range; at long-term maturity, an SHGF NAV around £1 billion supports a book of approximately £3 billion.

The precise ratio is for secondary legislation by affirmative SSI; the Bill should set a statutory ceiling on the leverage ratio, adjustable below that ceiling but not above it.

Control 2: Eligibility filter. RSLs accessing the Guarantee must hold a Compliant regulatory status from the SHR at the point of application, with no Under Review flag attached. The SHR’s framework has three tiers: Compliant, Non-compliant working towards compliance, Non-compliant statutory action. The Guarantee is available to the first tier only. The founding legislation should specify the eligibility threshold by reference to the SHR’s published regulatory framework with a provision that updates automatically if the SHR revises its methodology, so that a future change in the SHR’s framework does not inadvertently widen or narrow eligibility without parliamentary consideration. The principle is that the Guarantee is not a rescue instrument for financially stressed RSLs; it is a cost-reduction instrument for financially sound ones. Front-end eligibility filtering is substantially cheaper than workout costs on the back end.

A supplementary eligibility condition addresses the failure mode that SHR Compliant status does not catch: treasury and derivatives exposure. Vestia’s 2012 near-collapse (discussed in Comparator failure modes below) involved a corporation that would have passed any standard credit filter; the problem was speculative derivative positions accumulated without adequate treasury controls. RSLs seeking Guarantee access should therefore also be required to certify compliance with a derivatives and treasury risk management code, modelled on the WSW treasury code, specifying permitted instruments, board-level approval requirements for any hedging activity, and maximum notional exposure limits relative to asset base. The code’s content is secondary legislation; the requirement to comply with a code belongs in the Bill.

Control 3: First-loss tranche. The SHGF holds a first-loss tranche on the guarantee book, absorbing losses before the government guarantee is called. This is the mechanism WSW uses operationally. WSW’s 2024 adequacy test reports EUR 2.9 billion in available risk capital against a EUR 0.89 billion requirement; that EUR 2.9 billion comprises WSW’s own risk capital (EUR 606 million), the capital committed by participants through the obligo (EUR 2.5 billion), and the annual committed capital call (EUR 319 million), combined through a Basel- methodology risk model. These layers sit between borrower default and the state-and-municipality backstop. The government backstop has never been reached. WSW’s first-ever guarantee claims came in 2018 (Woningstichting Geertruidenberg and Stichting Humanitas Huisvesting), producing cumulative guarantee claim liabilities of approximately EUR 251 million on the EUR 94.9 billion guaranteed book (0.27%), absorbed within WSW’s own risk capital and the sector’s committed capital pool (see “Comparator failure modes and political durability” below).

The SHGF holding first-loss means it loses money if borrowers it has guaranteed default, creating an ongoing incentive to monitor covenant compliance and financial health across the guarantee book. The tranche is set at approximately 10% of the prevailing guarantee book, material enough to be a genuine deterrent against careless issuance decisions and large enough to absorb modelled loss rates within the SHGF’s own capital before the government layer is reached. The absolute reserve is therefore a moving figure tied to the book at each stage: approximately £100 million at year five against the £900 million to £1.2 billion year-five book; approximately £200 million at year ten against the £2 billion year-ten book; approximately £300 million at long-term maturity against the £3 billion mature book. The reserve must be liquid so that default losses can be absorbed without forced asset sales, and is carved from the SHGF’s gross capital base rather than treated as a separate envelope. Detailed accounting treatment (instrument selection for the reserve, treatment in leverage-cap calculation, audit reporting requirements) is for secondary legislation and the SHGF’s operating manual.

In the event of borrower default, the defaulting body’s own assets are applied first; the SHGF’s first-loss tranche covers residual losses; the government backstop at 80% of principal is the final layer.

Comparator failure modes and political durability. Vestia’s near-collapse in January 2012 cost the Dutch sector approximately €2 billion in derivative speculation on a €23 billion swap portfolio that had grown over years with treasury and risk controls effectively absent. The sector restructured through Vestia repaying €1.3 billion over ten years and a €700 million sector contribution coordinated through the Centraal Fonds Volkshuisvesting (CFV); WSW’s loan guarantee was not triggered (WSW does not guarantee derivatives) and the state-and-municipality backstop layer was not reached.

WSW’s wider operating history extends the same record. WSW was established in 1983; its guarantee was first called in 2018, on Woningstichting Geertruidenberg and Stichting Humanitas Huisvesting, and the calls were absorbed at WSW’s own risk capital before reaching the state-and-municipality backstop. Thirty-five years between establishment and first guarantee call, with the first calls still absorbed at the corporation-sector layer below the public backstop, is the empirical record the layered architecture above is designed to replicate. Governance and treasury controls have to be designed against the worst plausible failure mode, not the median operating year.

Husbanken’s track record carries a parallel lesson on the political side. The institution has been periodically narrowed in mandate, particularly through the 1990s and 2000s, on the argument that it crowded out commercial lending; it survived in narrower form because its loss record (annual losses below 0.02% of a NOK 187.4 billion portfolio, Årsrapport 2024) did not bear out the claim, but the political pressure was real and the institution emerged smaller. A Scottish equivalent should expect comparable political weather and design durability against it: statutory mandate fixed in primary legislation, explicit loss provisioning rules, quarterly transparency on portfolio performance, and a board structurally insulated from short-term ministerial direction.

Institutional governance risk. The controls above address borrowers accessing the Guarantee. They do not address the scenario in which SHFF or the SHGF itself becomes the source of institutional risk. SHFF is proposed as a non-profit holding up to £1 billion in housing assets with a national role in TA conversion and land assembly coordination. The SHGF is a separate mutual body holding the guarantee book and first-loss reserve. Both carry governance risks at scale. For SHFF: board capture by founding partners (RSLs, COSLA, or Scottish Government representatives) who redirect the institution toward the founding coalition’s interests rather than its statutory purpose. The anti-alienation clause and statutory mandate constrain bulk stock transfer and tenure reclassification (requiring unanimous founding-member consent), but operational disposals (plot tendering, construction procurement, shared-equity sales) are board decisions under standard governance. The constraint prevents privatisation; it does not prevent mission creep in acquisition strategy.

Political direction despite claimed operational independence: if SHFF holds assets the government wishes deployed for electoral purposes, the independence of acquisition decisions depends on whether the statutory insulation is designed against determined ministerial pressure, not just against routine interference.

For the SHGF: a sharp downward revaluation of the housing assets collateralising the guarantee book could breach the leverage cap and freeze new issuance precisely when the sector most needs Guarantee access. The Vestia lesson (that the institution inside the guarantee can be the source of systemic risk) applies to both bodies.

The founding legislation should specify for each: board appointment by an independent nominations process, a statutory duty of operational independence from ministerial direction on individual transactions, and a stress-test trigger (NAV falling below a defined threshold) that automatically restricts new commitments until the reserve ratio is restored. These are design parameters for the Bill team; their absence from the Bill would leave the architecture exposed at the point the Dutch precedent says it must be strongest.

Single-counterparty concentration risk. Greater Manchester’s Housing Investment Loans Fund committed £983 million from a £300 million revolving base with zero defaults. One developer group received over £500 million of that: 53% of total lending to a single connected party. No commercial lender’s credit committee would sign off single-name exposure at that level; standard prudential limits sit at 20 to 25% of committed facilities. The fund’s zero-default record is the outcome, not the control. A single default at 53% concentration would have been systemic. That the fund survived is not evidence that the governance worked; it is evidence that the governance was never tested. A subsidy challenge followed (Weis Group v GMCA, Competition Appeal Tribunal, 2025). The Tribunal dismissed it, finding GMCA acted on commercial terms. The concentration was the attack surface.

SHFF and SHGF lending must carry explicit single-borrower and connected-party exposure caps in secondary legislation. A 20% ceiling on any single borrower or connected group, measured against total committed facilities, is the minimum standard commercial banking applies. The Bill team should treat it as a floor, not a target.

Membership density risk. The mutual loss-absorption architecture depends on near-universal RSL participation. The obligo layer works because losses are distributed across the sector; if the largest RSLs decline to join on the grounds that their individual credit ratings already secure competitive bank terms, the first-loss buffer is thinner and the government backstop layer is reached at lower loss volumes than the central case projects. WSW has 264 members (97% of Dutch housing associations) because the guarantee works for the associations: cheaper capital, longer terms, and portfolio-value borrowing that individual RSLs cannot replicate alone.

Scotland’s equivalent incentive is partly financial and partly structural. The financial case is the rate reduction. The structural case is the RTB lock. Once housing stock is collateral backing guaranteed bonds, discounted sale is a covenant breach. A future Scottish or Westminster government attempting to reimpose RTB at statutory discount would need to compensate the SHGF and its bondholders for the collateral loss on every unit sold below market value, or face default across the guarantee book. The cost of reintroduction becomes the cost of buying out the encumbrance on every guaranteed unit. That is a structural deterrent no future housing minister can waive by Order, giving every RSL board contractual protection against a future government selling their assets below replacement cost. If membership density still falls short, the SHGF’s loss-absorption capacity is weaker and the fiscal exposure on layer 4 is higher. The founding legislation should specify a minimum membership threshold (by share of sector borrowing) below which the Guarantee does not activate, so that the government backstop is not committed to a mutual that lacks the scale to absorb losses before reaching it. The threshold is a Bill team parameter; its absence would leave the architecture exposed to adverse selection by precisely the RSLs whose participation the model requires.

Sector-wide credit fragility under the paper’s own architecture. A related risk applies directly to the proposals this paper makes. Of the £7.18 billion in total RSL facilities at March 2025, only £1.57 billion is genuine capital market funding; the balance is bank lending and revolving credit, much of it variable-rate, secured against RSL stock whose values the reform at Proposals 1 and 2 is designed to moderate. The land-market reform compresses the speculative residential land premium and, through that, moderates new-build and existing-stock price growth over time. The transmission is multi-stage and gradual but directional. As collateral values deteriorate sector-wide, the bank-lending tranche faces simultaneous credit stress: a Vestia-shaped failure mode arriving from policy success rather than policy failure or institutional malpractice.

The Guarantee substitutes sovereign credit support for asset-value collateral. As existing bank facilities mature and come up for refinancing, typical RSL loan tenors are 25 to 30 years, so the £5.6 billion book turns over progressively, RSLs refinance into guarantee-backed bond issuances at lower rates, migrating the sector’s exposure from asset-collateralised bank lending to sovereign-backed capital market funding. The migration is gradual; the guarantee must be operational before the land-market reform accelerates the refinancing pressure.

The sequencing argument is therefore not operational convenience alone but structural necessity. The Guarantee architecture should be operating, or at minimum committed and credibly scheduled, before the land-market reform begins to compress values. Proposal 4’s executive-action implementation through the 2026/27 bond programme is the right vehicle for that sequencing, and the position of Proposal 4 in the overall architecture is reframed accordingly: it is the credit-support precondition for Proposals 1 and 2 being implementable without sector-wide credit stress, not just the paired finance-side instrument with Proposal 3.

Annex 6: Guarantee Contingent Liability

The Guarantee is a contingent liability, not a direct liability: government pays out only on borrower default after the SHGF’s first-loss tranche is exhausted. The methodology for the £40 to £100 million envelope is:

Comparator loss rates. WSW’s 2024 annual report records total guarantee claim liabilities of approximately €251 million on a guaranteed book of €94.9 billion: a cumulative claim liability rate of roughly 0.27% across the fund’s operating history (€234 million non-current, €18 million current at end-2024). WSW’s first-ever guarantee claims came in 2018 (Woningstichting Geertruidenberg and Stichting Humanitas Huisvesting); both were absorbed within WSW’s own risk capital (€606 million) and the sector’s committed capital pool (€2.5 billion), which together with the annual committed capital call form the €2.9 billion available risk capital reported in the 2024 adequacy test against a €0.89 billion requirement. The government backstop has never been reached.

Husbanken’s 2024 annual report (Årsrapport 2024) records a total loan portfolio of NOK 187.4 billion at end-2024, with net losses of NOK 31 million: an annual loss rate of 0.017%. The five-year series (2020 to 2024) runs NOK 28, 21, 5, 17, and 31 million respectively, none exceeding 0.02% of the book in any single year. Husbanken’s startlån programme (targeted lending to households unable to access commercial mortgage markets) recorded 87 defaulted loans totalling NOK 22.2 million in 2024, of which Husbanken’s own share was NOK 2.7 million; nine of ten startlån customers meet their loan obligations.

Both WSW and Husbanken operate in housing sectors with different risk profiles from Scotland’s (Dutch RSLs are larger and better-capitalised than Scottish RSLs; Norwegian municipalities have different credit characteristics from Scottish councils). A conservative assumption applies a 2 to 5% cumulative loss rate to Scotland’s smaller, more fragmented sector, roughly eight to twenty times the WSW historical rate and two orders of magnitude above Husbanken’s annual experience.

Guarantee book scaling. The Scottish guarantee book is projected to scale to approximately £2 billion over ten years, from a standing start in year two (when the SHGF’s capital base from member contributions, guarantee fees, and SNIB seed is sufficient to underwrite issuance at scale). This is not a point estimate; the actual figure depends on borrower demand for guaranteed issuance and the SHGF’s leverage cap at each stage.

Net of first-loss. The first-loss tranche at 10% of total issuance absorbs approximately £200 million of any losses on a £2 billion guarantee book before the government backstop is reached. Applying a 2 to 5% cumulative loss rate to the £2 billion book over ten years gives gross losses of £40 to £100 million, sitting comfortably inside the first-loss buffer at every modelled rate. Net government exposure on the modelled range is therefore approximately zero; the £40 to £100 million range in the main paper is the gross figure rather than the net call on the budget. The buffer is replenished from guarantee fees on new issuance as defaults are written off, so it does not exhaust over the programme’s lifetime.

Tail risk envelope. The £100 million upper figure in the main paper is the figure sized for the tail event rather than the central case. At a 15% cumulative loss rate (three times the high end of the modelled range and approximately fifty-five times the WSW cumulative rate and nearly a thousand times Husbanken’s annual loss rate), gross losses of £300 million on a £2 billion book exceed the £200 million first-loss buffer; the residual £100 million flows to the government backstop at 80% coverage, producing £80 million in actual exposure. The Guarantee architecture is sized so that a tail event of that severity stays within the £100 million envelope quoted in the main paper rather than calling on a larger fiscal commitment. This is the level of catastrophic-stress headroom the Guarantee builds in, not the expected operating cost.

Fiscal scoring and Barnett

Expected call probability. Two comparator loss records anchor the estimate. WSW’s operating history (Annex 5): the government backstop has never been reached. Vestia’s 2012 crisis (derivative losses, not loan defaults) was absorbed through the CFV sector bailout without triggering WSW’s loan guarantee. The 2018 guarantee claims (SHH and WSG), which produced the fund’s first-ever guarantee obligations in thirty-five years of operation, were absorbed within WSW’s own risk capital and the sector’s committed capital at a cumulative claim liability rate of approximately 0.27% of the guaranteed book. Husbanken’s direct lending portfolio: annual losses of 0.017% of a NOK 187.4 billion book in 2024, with no year in the 2020-to-2024 series exceeding 0.02%. Both comparators are lower-risk than the Scottish sector (the Dutch sector is larger and better-capitalised; Husbanken lends against Norwegian municipal credit and state-guaranteed income streams). This annex applies the high end of the modelled loss range (5%) to compensate; at that rate, losses stay inside the first-loss buffer.

PSND scoring under ONS methodology. Under ESA 2010, financial guarantees are contingent liabilities. ONS records them off-balance-sheet as memorandum items and does not include them in Public Sector Net Debt unless the probability of call exceeds 50%. The structure described above is designed with that threshold in mind: first-loss at 10% of issuance, backstop triggered only on residual losses above first-loss, WSW operating history as empirical evidence for a call probability well below 50%. On that basis the guarantee is designed to score as contingent rather than direct liability, keeping it outside PSND.

That is not the same as a guarantee that it will score that way. Formal ONS/OBR assessment of a specific instrument in Scotland’s devolved fiscal context is a distinct question from the general ESA 2010 framework, and the same logic applies as for the capital borrowing cap treatment below: the case for off-balance-sheet scoring is strong, the evidential foundation is the WSW operating history, but the classification requires formal resolution with HM Treasury as part of the Fiscal Framework negotiation before the programme is structured on the assumption that it holds.

Barnett. The guarantee creates no Barnett consequentials. Barnett applies to changes in identifiable English DEL spending, generating population-share adjustments to the Scottish block grant. A contingent liability instrument that has not been called is not DEL spending and does not flow through the formula. If the guarantee is called, the resulting liability falls within the Scottish Government’s existing resource and capital envelope; it does not create additional drawing rights through Barnett and does not affect HMT’s exposure to the formula. Westminster’s legitimate question is not whether this creates Barnett pressure (it does not) but whether a call on the backstop would strain the Scottish Government’s own fiscal headroom at the time. That is a scenario-planning question for the Fiscal Framework negotiation, not a Barnett design question.

Capital borrowing cap interaction. HM Treasury classification of the Guarantee as contingent rather than drawn liability keeps the envelope outside the Scottish Government’s capital borrowing cap. This is a Fiscal Framework negotiation point: Scotland’s annual capital borrowing cap (£450 million baseline at 2023-24, rising with inflation) is consumed by drawn capital borrowing, not contingent liability. Treasury’s classification of the guarantee as contingent is not automatic; it requires Treasury Ministerial sign-off on the specific structure. The case for contingent classification is strong (the guarantee is triggered by RSL default, a probabilistic event, not by the act of issuing the guarantee), but it should be resolved in negotiation before the programme is designed rather than discovered after.

Fallback on refusal of contingent classification. If HM Treasury declines contingent treatment, the Guarantee reverts to drawn liability recorded against the borrowing cap. A £2 billion guarantee book at that point would consume the entire cap baseline, ruling out the full structure on headroom grounds alone. The practical alternative is a guarantee book sized to fit within available cap headroom after SHFF direct borrowing and MHS infrastructure tranches are drawn: approximately £300 to £500 million in the early years, rising as the cap baseline grows with inflation and as SHFF’s revolving income reduces its net draw on direct lending. That narrower book still delivers a real cost-of-capital reduction for RSLs and provides the first-loss architecture Proposals 1 and 2 require as credit-support substitute.

The £1 billion revolving fund timeline extends from five years to roughly eight to ten. The land-market instruments are unaffected: DRLT and statutory pre-emption operate under the Housing Emergency (Structural Measures) Bill regardless of the classification outcome. The Treasury negotiation determines how fast the finance architecture scales, not whether the package works.

Subsidy Control Act 2022: SPEI route. EU state aid rules no longer apply in Great Britain. The WSW guarantee that informs this architecture operated under grandfathered EU state aid as a Service of General Economic Interest (SGEI). The UK replacement is the Services of Public Economic Interest (SPEI) route under the Subsidy Control Act 2022 (SCA 2022, c.23). The Guarantee is a subsidy under that Act: it provides financial assistance below market rates, and the below-market pricing is the instrument’s purpose. That does not make it non-compliant; it makes the SPEI route the correct framework to rely on.

Social housing is named in the November 2022 Statutory Guidance on the SCA 2022 as the paradigm SPEI service alongside postal services and rural transport. That designation is in the Guidance, not the Act’s statutory text. The operative compliance framework for a large guarantee is a Schedule 1 principles assessment under the Act, conducted using the Altmark-derived SPEI methodology the Guidance sets out. That methodology identifies three conditions. First, the SHFF and participating RSLs must be formally entrusted with the public service obligation of social housing provision by a written legal instrument (the guaranteeing legislation itself) specifying the mandate, duration, and compensation parameters. Second, the subsidy element (the fee differential between the guaranteed rate and the market rate) must be proportionate to the net cost of delivering the SPEI: compensation limited to net costs plus a reasonable profit. Third, a review mechanism must be built in at no more than three-year intervals to check for overcompensation and claw back any excess.

The guarantee fee (the rate set for the instrument, to be determined by actuarial analysis of expected claims on the guarantee book) goes to the proportionality condition, not the “commercial terms” test. The commercial terms test (whether a financial contribution is a subsidy at all) is failed by design: a guarantee priced at market rates would not exist, because commercial insurers already provide market-rate guarantees and RSLs already pay them. The Guarantee’s value is precisely that it prices below the commercial rate.

The operative compliance question is therefore whether the fee is limited to what is necessary to deliver the housing SPEI. An actuarial analysis linking the fee to expected claims on the guarantee book (showing total compensation does not exceed net cost plus reasonable operational margin) is the evidential basis for the proportionality conclusion, and belongs in the Bill’s explanatory memorandum regardless of whether the instrument faces a formal challenge. The fee level is a design decision for the Bill team; the proportionality methodology is fixed.

The low-value SPEI exemption under section 38 of the Act (£725,000 threshold) does not apply at this scale. For subsidies above that threshold, the subsidy control requirements apply in full and the Schedule 1 principles assessment is the operative framework. The Statutory Guidance SPEI methodology described above is how the proportionality principle within that assessment is satisfied.

The SPEI analysis and the HMT approval question are parallel tracks, not sequential. A Subsidy Control-sound guarantee that HMT refuses to approve is politically blocked. A guarantee HMT approves but that is Subsidy Control-exposed is legally vulnerable to challenge by commercial lenders to RSLs, who have standing to bring a judicial review against the guaranteeing authority. Both need to be clean. The SPEI analysis should be resolved in the Bill’s drafting stage, not reconstructed in response to a challenge after commencement.

The risk is not theoretical. In 2025 the Competition Appeal Tribunal heard the first subsidy challenge to a UK housing revolving fund: Weis Group v GMCA, testing whether loans from Greater Manchester’s Housing Investment Loans Fund to a single developer group constituted unlawful subsidy under the SCA 2022 commercial market operator test. The Tribunal dismissed the challenge, finding GMCA acted on commercial terms. An appeal is pending (July 2026). The SHFF/SHGF architecture is structurally different: the SPEI route, the Altmark-derived fee methodology, and the SHGF’s mutual ownership structure each provide a compliance layer the GMCA fund lacked. But the case confirms that commercial lenders to the housing sector will use the SCA 2022 to challenge public instruments that price below market. The SPEI analysis must be litigation-ready at commencement, not merely defensible in principle.

Parliamentary counsel should verify all SCA 2022 section numbers against current Act text before transposition into the Bill.

Annex 7: COSLA Governance (Two-Tier Architecture and Dispute Resolution)

Why the COSLA relationship is structurally difficult

COSLA is the representative body for all 32 Scottish councils, and its political composition will not always reflect Scottish Government priorities. Individual councils will protect their own housing registers; a national institution with a Housing First mandate intersects local allocation priorities in ways that need explicit statutory architecture rather than negotiated goodwill. If SHFF’s allocation policy requires COSLA agreement as a matter of governance, COSLA holds a structural veto it can exercise through deadlock.

The Y-Foundation in Finland addressed an equivalent tension because Helsinki municipality was a founding partner and because state-guaranteed preferential lending was the explicit incentive for municipal participation. COSLA as an organisation does not hold a comparable incentive position: it does not borrow, does not build, and does not receive grant. The participating councils do. The incentive structure should target participating councils rather than COSLA-as-body.

Two-tier allocation architecture

The Bill should set a statutory floor of minimum mandatory standards applying to every SHFF tenancy allocation, which no bilateral protocol can vary downward. The floor covers Housing First priority access for the complex-needs cohort (individuals whose homelessness is compounded by multiple and coexisting support needs including substance use, mental health, repeat homelessness, or offending history), minimum quarterly reporting on TA conversion rates by local authority area, anti-discrimination requirements under the Equality Act 2010, and a minimum TA conversion quota per SHFF operating region per year.

Above the statutory floor, each participating council agrees a bilateral allocation protocol with SHFF covering register-versus-referral allocation proportions, coordination arrangements with council housing officers, and disagreement-handling processes. These protocols are negotiated individually; COSLA provides a framework template but does not hold a veto over the terms.

Dispute resolution

Two channels are required. Individual applicant disputes about allocation decisions or tenancy terms go to an existing housing tribunal forum with jurisdiction extended for the purpose; the Housing and Property Chamber of the First-tier Tribunal is a natural candidate given its existing housing-disputes remit. Institutional disputes between SHFF and a participating council or COSLA over protocol interpretation go to binding statutory arbitration with a defined time limit and an independent chair. The architecture must explicitly exclude Scottish Ministers as ultimate arbiter, because making every governance disagreement political removes the operational independence the founding structure is designed to protect. Forum specifics, panel composition, and time limits are properly Bill team detail.

Annex 8: Worked Example (A Notional Scottish Site, Years 1 to 8+)

Illustrative only. Numbers are indicative ranges drawn from comparator data cited in this paper, not modelled outputs. The example shows how the four pieces of machinery interact across time on a single notional site. Names and locations are deliberately generic.

Setup. A 12-hectare residential site on the edge of a Scottish town. LDP allocation granted in 2018, outline consent for 220 homes granted in 2021, no commencement to date. The site is one of approximately 50 such allocated-but-unbuilt sites in this notional local authority area, accounting between them for around 4,500 unbuilt consented homes: a microcosm of the 164,000.

Year 1: Commencement and declaration window. DRLT and Statutory Pre-Emption Acts take effect on the same day. The site holder receives a DRLT notice (0.5% phase-in rate on undeveloped hectares) and a declaration window letter. Within 24 months, the holder must lodge a binding delivery commitment with a defined build-out schedule (preserving market-value rights on future sale), or decline and accept the prospective EUV pre-emption right at end-of-window. For holders with existing S75 obligations, the window carries a further consequence: agree the LDP plan period end date as the deemed delivery deadline for the S75, or accept that the deemed-deadline provision applies on its statutory terms. DRLT applies at the same rate to all holders regardless of commitment status; the declaration window determines pre-emption exposure, not DRLT treatment. Of the 50 sites in this area, 30 lodge commitments, typically volume builders and SME developers with genuine intent. 20 decline, typically speculative landowners and dormant holding vehicles. Audit data per Part 1 shows that of the 30 committed sites, 18 are deliverable-now and 12 deliverable-with-public-investment, chiefly for grid connection and drainage.

Year 2: Build-out begins. The 18 deliverable-now sites begin substantive build-out; as hectares complete, they exit the DRLT taxable base. The 12 contingent sites enter the More Homes Scotland infrastructure commissioning programme (DRLT suspended for the duration of the infrastructure block). DRLT rate rises to 1.0%. Three of the 20 declined sites come to market voluntarily; carrying cost at 1.0% of the holding’s band-benchmark value (£20 to £30 million for a site of this scale in this zone) becomes uncomfortable for several commercial holders who prefer disposal to ongoing exposure.

Year 3: First pre-emption exercises. The declaration window closed at the end of year two. The DRLT rate enters persuasion territory at 2.0%, raising the annual carrying cost on every undeveloped hectare in the taxable base. Statutory pre-emption attaches prospectively to all 20 declined sites on any future voluntary disposal.

In parallel, the S75 enforcement mechanism operates independently of the voluntary-disposal track. The notional site in this example was allocated under an LDP adopted in 2018; the plan period for that LDP has expired before the Acts commenced. Under the deemed-deadline provision (Annex 1), the delivery deadline for pre-commencement S75 agreements is the end of the LDP plan period under which the allocation was made. That plan period has passed. At end-of-window, with the S75 components still undelivered, the pre-emption right crystallises on the entire consented site: the full 12-hectare site passes to MHS at EUV. This is not a fresh state imposition; the LDP plan period was a published document adopted after public consultation, and the developer knew the planning horizon for which the allocation was intended. The obligation attached to the site as a condition of the consent that generated its development value; failure to perform that obligation within the planning horizon for which it was granted extinguishes the entitlement to retain any portion of that value.

Whole-site acquisition is the correct boundary. A partial taking limited to the 3-hectare affordable portion would leave the holder with the profitable 9 hectares at an agricultural carrying cost, having shed the low-margin obligation that conditioned the consent. That outcome converts non-compliance into a better commercial position than compliance would have produced. Whole-site acquisition at EUV removes that arbitrage: the holder who fails to deliver forfeits the entire development opportunity, and MHS tenders all portions competitively through the procurement framework described below. MHS thereby holds the full 12-hectare site by the end of Year 3 via a single S75 crystallisation exercise.

Separately, six of the 20 declined sites come to market voluntarily across months 24 to 36. MHS exercises pre-emption at EUV, paying in the range of £15,000 to £25,000 per hectare (agricultural rate, site-specific). Those who decline the EUV offer retain the land but cannot sell at hope value; most accept. MHS commissions infrastructure on the acquired land against a pre-agreed regional supplier framework. Before the rate steps from 2% to 3% at year four, the statutory rate-review (Annex 3) reports to parliament; in this notional case, the schedule continues unchanged.

Year 4: Serviced-plot tendering and SHFF intervention. DRLT rate at 3.0%. MHS tenders the serviced plots on a published cost-plus basis.

MHS acquired the full 12 hectares at agricultural EUV: between £180,000 and £300,000 in total land cost (12 hectares at £15,000 to £25,000 per hectare). A developer carrying a consented residential site of this size at hope value would be paying several million pounds in land alone, and must recover that cost from private-sale units at full market margin; cross-subsidising affordable delivery beyond the 25% planning floor makes the scheme unviable. At EUV, MHS carries no speculative land premium. The 3-hectare S75 obligation is the planning floor; MHS tenders 5 hectares to SHFF, 2 hectares above it, because those additional 2 hectares cost MHS approximately £30,000 to £50,000 in land rather than the seven-figure sum they would represent at hope value. EUV acquisition is what converts the planning floor into a floor rather than a ceiling.

The notional 12-hectare site is split into three lots: 5 hectares to SHFF for social rent and Housing First (3 hectares delivering the S75-crystallised obligation, 2 hectares additional made viable by EUV acquisition); 4 hectares to a consortium of SME builders for mixed-market delivery, with the Proposal 5 statutory 25% affordable floor applying within this lot; 3 hectares to a volume builder for owner-occupied units, with Proposal 5 applying within this lot and the volume component serving as price discovery for the local market. The SME consortium could not have competed for the original site at hope-value pricing; cost-plus tendering makes the project viable. SHFF procurement framework requires that 20% of on-site trades positions be apprenticeships or trainees from registered Scottish FE programmes, and that Modern Methods of Construction be adopted where viability supports it.

Year 5: First completions. DRLT rate at 4.0%. All three lots break ground. SHFF begins TA-to-permanent transfers from the local authority’s existing temporary accommodation stock; 80 households previously housed in TA across the area move into permanent SHFF tenancies as units complete. The SHFF apprenticeship intake of approximately 30 places on this single site is one part of CITB’s projected 17,950-worker gap by 2029 (Construction Workforce Outlook 2025-29). The Housing Finance Guarantee (Proposal 4) is now operating; SHFF accesses guarantee-backed bond capital below comparable RSL commercial-bond pricing (rate-reduction mechanics in Annex 5). The principal benefit in the early years is converting an inaccessible bond market for smaller RSLs into an accessible one.

Years 6 to 8: Mature build-out. DRLT rate at 5.0% in year six, reaching the 6.0% ceiling at year seven and held thereafter. By year six, most of the 12 hectares have exited the DRLT taxable base as phases completed; the charge falls only on the residual undeveloped portion. Final tenure on the 12 hectares: approximately 42% SHFF social rent and Housing First (5 of 12 hectares, the highest affordable share this site has ever been planned to carry); approximately 1.75 hectares of statutory affordable within the two developer lots (Proposal 5’s 25% floor applied to the combined 7 hectares of SME and volume-builder land); and approximately 5.25 hectares of private-market and owner-occupied. Total affordable delivery across the assembled site is approximately 6.75 hectares, or 56%, more than double the 25% the original planning obligation required, at no additional fiscal cost to government beyond the infrastructure already committed. The cohort of 50 sites in the local authority area has been substantially worked through. DRLT revenue from this area is declining as land enters productive use, which is the policy success signal. The SHFF balance sheet has grown through revolving-fund recycling: loan repayments from rental income fund subsequent tranches of stock acquisition and new-build elsewhere in Scotland.

Year 8 onwards: Projected future state. Beyond year 8 the example is projection rather than evidence. The structural conditions that produced the original 164,000 unbuilt consents have been altered, not merely relieved. New LDP allocations made post-Year 1 are subject to pre-emption from day one of allocation; the announcement effect of the Acts repriced future allocations to remove the speculative premium. Volume housebuilders compete on construction efficiency rather than land assembly. SME housebuilders compete for predictable serviced-plot tenders at cost-plus pricing. The construction workforce has stabilised against multi-year SHFF anchor demand. Public health, social care, and criminal justice budgets register the social-cost reduction from former TA households now in permanent tenancies. The 18,092 figure has fallen materially. Whether by 80% (Finland’s 1985-2023 achievement against a different baseline, now partly reversed by welfare cuts) or by some smaller fraction depends on factors the architecture cannot determine alone, including AHSP grant continuity, the construction workforce expansion rate, and the Westminster public accounting convention discussed in Part 5 of the main paper. None of this is automatic. All of it is now structurally possible rather than structurally blocked.

What this example does not show. Three things. First, sites where the structural blocker is remediation cost (contamination, RAAC, access constraints) rather than speculative holding; those sites need a different intervention. Second, the political risk that a successor administration repeals or weakens the Acts in Year 3 or later; this is why the founding statutes should include reserved-power protections. Third, the local property-market displacement effect on owner-occupier values in the immediate vicinity of pre-empted sites, which on Dutch and Irish data is modest but non-zero. A serious implementation should commission both ex-ante and ex-post property-market impact assessment as part of the Bill’s Regulatory Impact Assessment.

Annex 9: Planning Value, EUV, and DRLT Holding Cost

Why the current equilibrium produces non-delivery

Holding LDP-allocated land under the pre-reform regime carries approximately zero annual cost. Council tax is assessed on agricultural use. Business rates on undeveloped land with planning permission are negligible relative to the asset value. There is no income requirement and no delivery obligation enforceable at cost to the holder. Against near-zero carrying cost, the land appreciates with the residential market. At least a tenfold uplift from agricultural to hope value follows LDP allocation; English figures suggest fifty to over three hundred-fold at the higher end (MHCLG, 2023; 2018). That uplift sits on the landowner’s balance sheet, unencumbered, growing year on year, requiring nothing except continued ownership.

This is not a pathology unique to bad actors. It is the predictable equilibrium produced by a system that creates large, publicly-generated value uplifts and attaches no holding cost to them. The 164,000 unbuilt consented sites are the accumulated result of that equilibrium operating across every LDP cycle since the planning system took its current form.

DRLT is the holding-cost instrument that breaks the equilibrium. Pre-emption at EUV plus the statutory premium is the exit-route instrument that removes the hope-value premium from the hold-and-wait payoff. Together they change the payoff matrix so that building or selling becomes preferable to continued holding.

The public origin of planning value

LDP allocation converts agricultural land to residential use without any investment by the landowner. The uplift is produced by a public administrative decision. No Scotland-wide residential land value survey exists; the closest proxy is English comparable data (MHCLG, 2023; Crook, 2018, using 1995-2001 VOA data), which places the uplift at least tenfold from agricultural to hope value as a conservative floor, and substantially higher in areas of housing pressure.

The public origin of planning gain is the foundation of Section 75 planning obligations. Parliament created S75 precisely because planning permission generates value the public has a legitimate interest in. The planning system has always acknowledged that origin. What it has lacked is an instrument that prices the annual cost of holding the resulting value unbuilt.

Per-hectare base: design rationale

Ireland’s RZLT uses market value, self-assessed by the landowner. Revenue Commissioners’ operational guidance runs to 99 pages, most of it managing the consequences of a contestable base: surcharge regimes for undervaluation, independent expert appointment, valuation formulae for partial-site transactions, and a separate appeals pathway (Revenue Commissioners, Part 22A-01-01, January 2026). The per-hectare base eliminates the contestable per-site variable. Land area is recorded in the Land Register, verifiable from Ordnance Survey data, and the rate applies to a published zone band benchmark, a coarse proxy for land value rather than a bespoke per-site assessment. A benchmark to apply, but no per-site valuation to dispute, no expert to appoint, no appeal on quantum.

The behavioural signal from Ireland’s first year of operation is the primary empirical anchor. In 2024, following RZLT announcement and preceding its live date, Irish residentially zoned land sales volumes rose almost 30% year-on-year while the median price per acre fell 6% (Revenue Ireland, September 2025). Landowners responded by transacting, not by walking away. A holding-cost instrument changes land disposal behaviour; the mechanism does not depend on any particular rate for that behavioural shift to occur.

The three options under DRLT

The declaration window forces a choice the pre-reform system never required.

Build. The holder who commits during the declaration window and breaks ground preserves market-value rights on private-sale units. DRLT accrues at the low early rates during the delivery period and exits the taxable base hectare by hectare at practical completion. The hope-value premium on land is retained within the delivery timetable the holder accepted at planning consent.

Dispose at EUV plus. The holder who chooses to dispose receives EUV plus the statutory premium set by SSI within the ceiling specified in primary legislation. Against the alternative of continued holding as the DRLT escalator runs, early disposal avoids the full cumulative levy and frees capital tied up in an unproductive asset. For any holder without development capacity, early disposal dominates continued holding once the escalating annual charge materially exceeds the income the site is generating.

Hold to the ceiling. The ceiling applies to a holder who has declined every alternative at each annual decision point across the full escalation period. The escalator provides the annual decision point; the declaration window is the off-ramp. The holder who builds never reaches the ceiling. The holder who disposes receives EUV plus and avoids cumulative levy. The ceiling falls only on those who have spent the full escalation period declining every alternative.

DRLT does not claim the uplift for the state. It prices the annual cost of holding a publicly-created asset unbuilt. The removal of the zero-cost carry is the reform. The escalator is the timeline. The declaration window is the offer.

Build rate as a choice, not a constraint

Large housebuilders programme at 25 homes per site per year on major sites (Glasgow City Council, Housing Land Audit 2025, paragraph 5.8). A build-to-rent developer on a comparable Glasgow site delivers 220 homes in a single year. The build rate is not a construction constraint; it is an absorption constraint. Releasing units at market-rate pace protects the sale price of each unit. The S75 affordable obligation is deferred as a direct consequence: the same absorption logic that governs market-unit release governs when affordable homes are built, because the developer controls the phasing.

Under the reformed architecture, that choice has a price. DRLT accrues on each unbuilt hectare until a residence is completed on it. A developer building at pace exits the taxable base early. A developer drip-feeding pays DRLT across every year the remaining hectares sit unbuilt. The per-hectare exit mechanism prices the difference between those two build rates on any site of substantial scale.

The S75 deemed deadline operates independently of DRLT. The LDP plan period end date is the delivery deadline for affordable units owed under existing S75 agreements. A developer who plans to deliver affordable homes in a final phase faces crystallisation of the S75 obligation at the plan period boundary. Pre- emption on the whole site becomes available to MHS if the obligation is unmet at that point.

Pre-emption at EUV plus is the backstop the current system lacks. Where a holder declines to build and the S75 obligation goes unmet, MHS acquires at agricultural value plus the statutory premium. MHS is not constrained by absorption rate because tenure mix is a procurement specification, not a viability negotiation. MHS tenders the site: social and affordable units to RSL delivery partners financed through SHGF-guaranteed borrowing; private plots to SME housebuilders on a cost-plus basis. The SME operates under the same DRLT incentive to build fast that the original developer lacked.

The four instruments as a system

DRLT prices the hold. The deemed deadline enforces the S75 obligation. Pre-emption at EUV plus provides the acquisition route when the obligation goes unmet. SHFF absorbs the stock that local RSLs cannot immediately take on.

Each instrument addresses a distinct point of failure in the current system. DRLT removes the zero-cost carry that makes speculative holding rational. The deemed deadline removes the absence of an enforceable delivery timeline from planning obligations that already exist in law. Pre-emption removes the absence of a public acquisition route that operates without the developer’s co-operation. SHFF removes the institutional gap that would otherwise force MHS to wait for RSL balance-sheet capacity before homes can be occupied.

The reform does not prevent a developer from building at absorption rate. It prices that choice. A developer who builds at pace exits DRLT liability early and delivers S75 affordable units within the plan period. A developer who drip-feeds pays DRLT on every unbuilt hectare across every year of the build- out and faces S75 enforcement at the plan period boundary. A developer who declines to build returns the land at EUV plus. The planning system already required the affordable housing. The reformed system provides instruments to enforce what the current system only requires.

For the S75 enforcement mechanism and proportionality treatment, see Annex 1. For DRLT rate structure and the infrastructure gateway, see Annex 3.

Annex 10: Wider Comparator Scan

Part 2 of the Policy Paper examines four primary comparators: Helsinki/Y-Foundation, Netherlands woningcorporaties and WSW, Norway Husbanken, and Ireland’s RZLT. The architecture proposed in this paper spans wider than any single comparator covers: land tax, statutory pre-emption, social-landlord institutional design, finance guarantee, and an active land agency role for More Homes Scotland. The European, UK, and Irish institutions below surface reference points that sharpen specific proposals, particularly Proposal 6.

Vienna: limited-profit institutional model and dedicated funding

Vienna operates a housing system that the four primary comparators above do not directly cover: a hybrid limited-profit (gemeinnützig) sector working alongside direct municipal ownership, funded through a dedicated payroll levy. Approximately 200,000 rental and cooperative flats in Vienna are held by 54 limited-profit housing associations operating under the Wohnungsgemeinnützigkeitsgesetz (WGG), which requires rents set to cover land, construction, administration, financing, and a maintenance reserve, with all profit recycled into further housing rather than extracted. This is parallel in mission to Y-Foundation but operates at a scale roughly ten times larger and is paired with municipal direct ownership of comparable scale (around 220,000 Gemeindebau units).

The funding mechanism is the Wohnbauförderungsbeitrag, a payroll levy split equally between employer and employee. The contribution rate in Vienna rises from 1% to 1.5% from 2026, generating €300-400 million per year currently and projected higher under the increased rate. Combined with other sources, total Vienna housing spend is approximately €530 million per year, dedicated and hypothecated.

The limited-profit model demonstrates that a non-profit institutional housing sector can scale beyond the single-organisation footprint of Y-Foundation when the legal-financial framework recycles all return into further housing. The dedicated payroll levy demonstrates that earmarked funding outside general taxation is feasible at material scale and politically sustainable across decades.

The transferability limit is twofold. Vienna, a city government, runs both a named payroll levy and general municipal revenue hypothecated to housing. Holyrood has the blunter instrument: nationwide income tax, where a dedicated band or surcharge on earned income could produce an equivalent hypothecated revenue stream, but a payroll levy as such is reserved (employment law, National Insurance contributions). A surcharge of 0.15-0.20% on Scottish income tax, sunsetting at the end of a single parliamentary term, would raise £105-140m per year against a startup capitalisation requirement of £200-300m across MHS, SHGF, and SHFF. By year three, DRLT revenue and MHS disposal margins begin flowing and the surcharge becomes redundant. No party has proposed this. The institutional scale (54 associations, 200,000 units) is the product of decades of accretion and cannot be replicated in a single parliamentary cycle; SHFF as proposed is the founding institution that could grow toward Vienna scale over the architecture’s mature operating period.

France: Caisse des Dépôts and savings-intermediation finance

The Caisse des Dépôts (CDC) channels regulated retail savings into social housing loans at tenors up to 80 years, indexed to the Livret A rate. Through its Banque des Territoires arm, CDC provides 70-75% of all French social housing financing. This is a third finance model alongside the Dutch WSW (guarantee on private bond issuance) and Norwegian Husbanken (state direct lending): the state acts as intermediary between household savers and housing developers, with a regulated rate framework providing predictability for both sides.

Scotland does not have a Livret A equivalent. Creating one would require reserved-power legislation (financial services regulation is reserved under Section A3 of Schedule 5 of the Scotland Act 1998) and behavioural take-up that France’s two centuries of accretion makes possible. The CDC model widens the design space evidence for state-supported housing finance beyond the WSW-or-Husbanken pair, but it is not directly portable.

Iberia: cautionary precedents

Spain’s post-2008 trajectory is the cautionary mirror to the four positive primary comparators. Pre-2008, Spain liberalised land regulation and offered effectively unlimited credit, producing approximately five million homes between 2000 and 2009, more than Germany, France, Italy, and the United Kingdom combined during the same period. The 2008 crash produced an extensive ghost-town legacy (Sesena being the often-cited exemplar: the El Pocero project of 13,000 planned units, 5,000 built, with no water connection, no public transport, no schools; Reuters, June 2025), planning corruption scandals concentrated in coastal and high-growth regions, and a mass-disposal architecture (SAREB, the 2012 bad bank created as a condition of the ESM bank recapitalisation, absorbing approximately €50 billion in toxic property assets) that disposed of those assets to investment funds rather than converting them into social housing.

The recovery arc has run fifteen years and is still incomplete. Spain’s 2026-2030 State Housing Plan, approved 21 April 2026, commits a €7 billion envelope (40% to public housing supply, 30% to renovations, balance to subsidies) and triples previous public housing investment levels. The plan transfers more than 40,000 SAREB properties and 2,400 plots of land (sized for approximately 55,000 additional apartments) to Sepes, the Spanish state housing authority, valued at €5.9 billion at current market prices, ahead of SAREB’s scheduled dissolution in 2027. Housing costs in Spain were rising 13% year-on-year at the end of 2025; Madrid and Barcelona remain among Europe’s most pressured rental markets.

Helsinki, Netherlands, Norway, Ireland, and Vienna show what works when public land discipline and social housing institutions exist. Spain shows what happens when neither does: planning liberalisation without land market discipline produces oversupply of the wrong stock concentrated in speculative locations, rather than absorption of demand pressure on the right stock concentrated where people actually need to live. Recovery comes late and at greater cost.

Portugal adds a distinct failure mode: policy-driven foreign capital inflow. The Golden Visa programme attracted approximately €6.45 billion in foreign property investment between 2012 and 2023; national house prices doubled over the same period against 9% income growth. The residential route was eliminated in October 2023, but Lisbon still records approximately 48,000 empty dwellings against acute rental shortage (2021 Census; municipal Devolutos initiative). Capital inflow that treats housing as a financial product produces displacement that outlasts the policy.

England: Homes England

Homes England is a non-departmental public body, the delivery agency for new supply and place-making for the Ministry of Housing, Communities and Local Government. The 2025 Spending Review allocated up to £46 billion over ten years: a £39 billion Social and Affordable Homes Programme (of which £27 billion directly deployable by Homes England) and up to £16 billion through a new National Housing Bank subsidiary, with the two overlapping within the £46 billion envelope. The NHB is intended to unlock approximately £50 billion in additional private capital. Homes England operates a “playbook” of defined interventions across land assembly, infrastructure unlocking, SME scaling, mixed-tenure-by-default, statutory CPO use, and grant-funded social and affordable housing investment. The 2023 Public Bodies Review (Poulter 2024) judged Homes England the “right vehicle” for English supply and place-making, with the qualifications that the body needed longer-term funding settlements, greater funding flexibility, and clearer priorities; the 2025 Spending Review settlement addressed several of those.

Homes England is the closest analogue for what More Homes Scotland has been announced to become, and its playbook approach (a defined menu of interventions matched to identified market failures) is the operational template the Cabinet Secretary indicated MHS will follow when she referenced “conversations with Homes England” in her February 2026 Inside Housing interview. The transferability limit is that Homes England operates at a budget and population scale roughly ten times larger than Scotland’s, with a regional model in development; the Cabinet Secretary has indicated MHS will be national rather than regionally devolved.

Ireland: Land Development Agency

The Irish Land Development Agency is a non-departmental public body established in response to the 2018 NESC and Irish Government reports on housing delivery failure. The agency operates with a relatively light-touch government relationship and was set up with permission to hire expertise at private-sector commercial rates, which has driven its staffing model (over 250 staff) and operating culture. The original model was land assembly and infrastructure provision followed by master-planning, tenure-mix specification, and disposal to developers, with capital recycled into the next acquisition cycle. A shortfall in Irish new-build apartment PRS diverted the agency into direct rental delivery; it is now expected to return to its core land-assembly model as the rental market normalises.

Intervention flexibility against a specific identified shortfall is structurally compatible with a stable long-term mission, provided the temporary nature of the deviation is explicit and the return path is documented. The “mission creep” criticism the LDA attracted resolves once the deviation is framed as a time-bounded response to a specified failure; CaCHE’s March 2026 briefing surfaces the same risk for MHS as one of the central design tensions for the agency team. A second architectural lesson is the LDA’s commercial-rate hiring permission: agency operating culture is downstream of staffing model, and a delivery-focused body needs the freedom to recruit against commercial benchmarks where the alternative is an organisation that cannot attract the expertise its mission requires.

Sweden Allmännyttan and Germany KfW: brief mentions

Sweden’s Allmännyttan municipal housing companies were the bedrock of Swedish social housing for decades. The 2011 EU competition law reform forced them to operate “businesslike”, removing the special advantages and subsidies that distinguished them from commercial landlords. The universal-access mandate was preserved but the institutional protection that made it viable was eroded. The SHFF founding architecture should design against the equivalent erosion under the UK Subsidy Control Act 2022, even though the UK is post-Brexit and the EU competition law framework no longer applies directly. The SCA 2022 was modelled on EU state-aid principles; an architecture compliant with the Act is defensible whichever direction the UK’s relationship with the EU evolves.

Germany’s KfW is a state-owned development bank (80% federal, 20% Länder), raising funds through federal-guaranteed bond issuance. In core function it is similar to Husbanken; in funding mechanic it is closer to the WSW model the Scottish Housing Finance Guarantee is built on. A third existing precedent at substantial scale.

Mietshäuser-Syndikat (Germany): anti-privatisation governance device

The Mietshäuser-Syndikat, founded 1992 in Freiburg im Breisgau, is a federation of collectively-owned residential housing projects in Germany. By June 2024 the federation spanned 191 projects covering more than 150,000 m² of living space and housing more than 3,800 residents. The architectural innovation is the dual-ownership structure: each property is owned by a Haus GmbH whose shares are split between the resident Hausverein (House Association, holding ~51%) and the federation (~49%). The Hausverein has near-total autonomy over use and development. The federation’s voting rights are restricted to decisions affecting sale or legal transfer of the property: on those decisions, both parties must consent, which gives the federation an effective veto on re-privatisation. Each project repays its bank loan from rents, and the solidarity transfer mechanism within the federation cross-subsidises new projects from established ones.

A pure cooperative or non-profit structure can be voluntarily wound up or restructured by a future generation of residents or a future government acting on the institution. The Mietshäuser-Syndikat architecture makes that effectively impossible by separating use rights (residents) from disposition rights (federation veto), with the federation’s mandate written into the founding structure as anti-privatisation. This is the device worth importing into both Proposal 7 (Collective Housing Development Groups) and Proposal 3 (SHFF anti-alienation provision in the founding articles and Bill).

What the wider scan adds

The wider scan adds four things. First, it expands the design space evidence for the SHFF institutional model (Vienna’s limited-profit sector at city scale alongside Y-Foundation’s national-landlord scale) and for the Housing Finance Guarantee (CDC savings-intermediation, KfW federal-guaranteed bond, Husbanken state direct, WSW guarantee; four functioning models, more diverse than the primary comparators alone present). Second, it introduces the Iberian cautionary precedent that the primary comparators do not cover: the alternative to building this machinery is empty homes, speculation, an inadequate social housing buffer, and a recovery arc measured in decades. Third, it surfaces the dedicated funding question through the Vienna payroll levy and the anti-privatisation device through the Mietshäuser-Syndikat structure, both of which the primary comparators do not address and both of which Policy Paper Part 5 and Proposals 3 and 7 return to. Fourth, the Homes England and Irish LDA sections supply direct institutional comparators for Proposal 6’s active-land-agency framing of More Homes Scotland; both are agencies the Cabinet Secretary has publicly referenced in shaping the MHS design.

Glossary

A1P1: Article 1 Protocol 1 of the European Convention on Human Rights (Protection of Property)

AHSP: Affordable Housing Supply Programme

BoE: Bank of England

BTR: Build-to-Rent

CaCHE: UK Collaborative Centre for Housing Evidence

CAT: Competition Appeal Tribunal (UK tribunal with jurisdiction over subsidy control challenges under the Subsidy Control Act 2022)

CDC: Caisse des Dépôts (French public financial institution)

CFV: Centraal Fonds Volkshuisvesting (Netherlands social housing sector fund; coordinated the €700m sector bailout of Vestia in 2012; merged into Autoriteit woningcorporaties in 2015)

CITB: Construction Industry Training Board

COSLA: Convention of Scottish Local Authorities

Committed (three senses in this paper): (1) Infrastructure committed: capital programme delivery dates published by statutory undertakers (Scottish Water, SSEN, Transport Scotland, SEPA) against specific sites. (2) Delivery commitment: a binding build-out schedule lodged by a site holder during the declaration window, preserving market-value rights on future disposal. (3) Capital committed: funds allocated within the SHGF guarantee structure or SHFF acquisition pipeline.

CPO: Compulsory Purchase Order

CSIH: Court of Session, Inner House (Scotland; case citation suffix)

CSO: Compulsory Sale Order

DEL: Departmental Expenditure Limit (HM Treasury fiscal classification)

DRLT: Designated Residential Land Tax (proposed in this paper)

ECHR: European Convention on Human Rights

EHRR: European Human Rights Reports (case citation series)

EPC: Energy Performance Certificate

EUV: Existing Use Value

FE: Further Education

FTE: Full-Time Equivalent

Guarantee, the: used in this paper to refer to the complete layered architecture under Proposal 4: the SHGF mutual body, the four-layer loss-absorption sequence (borrower assets, SHGF first-loss tranche, sector obligo, government backstop), and the covenants binding guaranteed borrowers. When the text refers to a specific layer, it names it: “SHGF” for the mutual body, “government backstop” for the final layer.

HLA: Housing Land Audit (annual programmed-completion record published by Scottish planning authorities)

HMT: HM Treasury

HNDA: Housing Need and Demand Assessment

HOPS: Heads of Planning Scotland

KfW: Kreditanstalt für Wiederaufbau (German state development bank)

LBTT: Land and Buildings Transaction Tax

LDA: Land Development Agency (Ireland)

LDP: Local Development Plan

LRRG: Land Reform Review Group (2014 Scottish Government review)

MHS: More Homes Scotland (national housing agency announced 22 January 2026)

MMC: Modern Methods of Construction

NAV: Net Asset Value

NDR: Non-Domestic Rates

NESC: National Economic and Social Council (Ireland)

NPF4: National Planning Framework 4

OBR: Office for Budget Responsibility

ONS: Office for National Statistics

Pre-emption (two triggers in this paper): (1) On voluntary disposal: a right of first refusal at EUV whenever the holder of LDP-allocated land offers it for sale. (2) On S75 breach: a mandatory acquisition right at EUV when the affordable components are undelivered at the deadline. The first is a right of first refusal; the second is closer to specific performance of a contractual obligation accepted at planning consent. Each carries a distinct A1P1 proportionality analysis (see Annex 1).

PRS: Private Rented Sector

PSND: Public Sector Net Debt

RAAC: Reinforced Autoclaved Aerated Concrete

RCF: Revolving Credit Facility

RSL: Registered Social Landlord

RTB: Right to Buy

RZLT: Residential Zoned Land Tax (Ireland; live February 2025)

S106: Section 106 of the Town and Country Planning Act 1990 (England); planning obligation; English equivalent of Scotland’s Section 75

SAREB: Sociedad de Gestión de Activos procedentes de la Reestructuración Bancaria (Spain’s 2012 bad bank)

SCA: Subsidy Control Act 2022

SEPA: Scottish Environment Protection Agency

SFC: Scottish Fiscal Commission

SFHA: Scottish Federation of Housing Associations

SHFF: Scottish Housing First Foundation (proposed in this paper)

SHGF: Scottish Housing Guarantee Fund (proposed in this paper)

SHR: Scottish Housing Regulator

SLC: Scottish Land Commission

SLWG: Short Life Working Group (Scottish Government stalled sites group, 2025-26)

SME: Small and Medium-sized Enterprise

SNIB: Scottish National Investment Bank

SONIA: Sterling Overnight Index Average (interbank rate benchmark)

SPEI: Services of Public Economic Interest (UK Subsidy Control Act 2022 framework; UK equivalent of the EU’s Services of General Economic Interest, SGEI, older case law and EU-era guidance use the SGEI term)

SSEN: Scottish and Southern Electricity Networks

SSI: Scottish Statutory Instrument

TA: Temporary Accommodation

UPRN: Unique Property Reference Number

WGG: Wohnungsgemeinnützigkeitsgesetz (Austrian limited-profit housing law)

WSW: Waarborgfonds Sociale Woningbouw (Dutch social housing guarantee fund)

Consolidated Bibliography

Andy Wightman, The Poor Had No Lawyers: Who Owns Scotland (And How They Got It) (Birlinn, 2011; revised 2015)

Land Reform Review Group (2014) The Land of Scotland and the Common Good: Final Report. Scottish Government, Edinburgh: gov.scot

Tolson, S. and Rintoul, A. (2018) The Delivery of Public Interest Led Development in Scotland: A Discussion Paper. Scottish Land Commission, Inverness: landcommission.gov.scot

James, G. and Tolson, S. (2020) Delivering More Homes and Better Places: Lessons from Policy and Practice in Scotland. UK Collaborative Centre for Housing Evidence, University of Glasgow: housingevidence.ac.uk

Scottish Land Commission (2021) Land for Housing: Towards a Public Interest Led Approach to Development. Scottish Land Commission, Inverness: landcommission.gov.scot

Gibb, K., Earley, A., James, G. and Watson, A. R. (2026) More Homes Scotland: Debating a New Housing Agency. UK Collaborative Centre for Housing Evidence, University of Glasgow. March 2026: housingevidence.ac.uk

Scottish Land Commission (2025) Rural Land Market Data Report 2025 (covering 2024 data). Scottish Land Commission, Inverness: landcommission.gov.scot

Scottish Land Commission (2024) Rural Land Market Data Report, December 2024 (covering 2023 data). Scottish Land Commission, Inverness: landcommission.gov.scot

Scottish Land Commission / ChamberlainWalker Economics (2020) An Investigation into Land Banking in Scotland. Scottish Land Commission, Inverness: landcommission.gov.scot

James and Others v United Kingdom (1986) 8 EHRR 123: hudoc.echr.coe.int

Lithgow and Others v United Kingdom (1986) 8 EHRR 329: hudoc.echr.coe.int

Hutten-Czapska v Poland (2006) 45 EHRR 4: hudoc.echr.coe.int

Holy Monasteries v Greece (1994) 20 EHRR 1: hudoc.echr.coe.int

Pressos Compania Naviera SA v Belgium (1995) 21 EHRR 301: hudoc.echr.coe.int

Cavendish Square Holding BV v Makdessi; ParkingEye Ltd v Beavis [2015] UKSC 67: supremecourt.uk

Pairc Crofters Ltd v Scottish Ministers [2012] CSIH 96: scotcourts.gov.uk

Town and Country Planning (Scotland) Act 1997, sections 29, 58, 75: legislation.gov.uk

Land Compensation (Scotland) Act 1963: legislation.gov.uk

Land and Buildings Transaction Tax (Scotland) Act 2013, Schedule 17 paragraph 3: legislation.gov.uk

Community Empowerment (Scotland) Act 2015, sections 62-73: legislation.gov.uk

Scotland Act 1998, section 29(2)(d): legislation.gov.uk

Human Rights Act 1998, section 6: legislation.gov.uk

Subsidy Control Act 2022 (c.23): legislation.gov.uk

HM Government, Statutory Guidance on the Subsidy Control Act 2022. November 2022: gov.uk

Foye, C. (2022) ‘Section 106, Viability, and the Depoliticization of English Land Value Capture Policy’. International Journal of Urban and Regional Research, 46(2): doi.org

Foye, C. and Shepherd, E. (2023) Why Have the Volume Housebuilders Been So Profitable? The Power of Volume Housebuilders and What It Tells Us About Housing Supply, the Land Market and the State. UK Collaborative Centre for Housing Evidence: housingevidence.ac.uk

McAllister, P., Street, E. and Wyatt, P. (2014) Section 106 Planning Obligations in England, 2011–2012. Department for Communities and Local Government, London (post-GFC viability research; cited via Foye and Shepherd 2023 for the finding that almost all Section 106 viability challenges succeeded).

McAllister, P., Street, E. and Wyatt, P. (2016) ‘Governing Calculative Practices: An Investigation of Development Viability Modelling in the English Planning System’. Urban Studies, 53(11): doi.org

Revenue Commissioners (Ireland) (2026) Tax and Duty Manual Part 22A-01-01: Guidance on the Residential Zoned Land Tax. January 2026: revenue.ie

Scottish Land Commission / Tony Crook, Land Value Capture (November 2018): landcommission.gov.scot

RICS (2024) RICS Valuation – Global Standards (“Red Book Global Standards”): rics.org

World Habitat, Helsinki housing model (2022): world-habitat.org

World Habitat, Y-Foundation profile: world-habitat.org

Y-Säätiö (2025) About Us: ysaatio.fi

Varke (2024) Finnish Homelessness Statistics: varke.fi

Harvard Joint Center for Housing Studies, Dutch social housing (van Deursen, August 2023): jchs.harvard.edu

OECD Economic Surveys: Netherlands 2025, Housing Market: oecd.org

Revenue Ireland, Residential Zoned Land Tax first-year update (September 2025): revenue.ie

Scottish Government, Homelessness in Scotland to 30 September 2025: gov.scot

Scottish Housing Regulator, Annual Review of RSLs’ Loan Portfolios 2025: housingregulator.gov.scot

Scottish Government, bonds programme announcement (2026/27 launch): gov.scot

Scottish Government, Scottish Budget 2025-26, borrowing: gov.scot

Scottish Parliament Information Centre, Affordable Housing Supply Programme briefing (April 2025): digitalpublications.parliament.scot

CITB Construction Workforce Outlook, Scotland 5-year outlook: citb.co.uk

Waarborgfonds Sociale Woningbouw (WSW), institutional overview and operating history: wsw.nl

Waarborgfonds Sociale Woningbouw (WSW) (2025) Annual Report 2024 / Jaarverslag 2024, 31 December 2024. €94.9bn guaranteed book; claim liabilities €251m (0.27%); available risk capital €2.9bn (own €606m + obligo €2.5bn + annual call €319m); 264 members; government backstop never reached.

Inside Housing, Vestia 2012 and WSW guarantee history: insidehousing.co.uk

Homes England Strategic Plan 2025-30: gov.uk

Homes England Investment Roadmap (December 2025): gov.uk

Poulter, T. (lead reviewer) (2024) Homes England Public Bodies Review 2023. DLUHC, London: gov.uk

Wider Comparator Scan (Annex 10) sources:

City of Vienna, housing grants and funding: wien.gv.at

Moore Salzburg, Wohnbauförderungsbeitrag in Wien (2026 rate change): moore-salzburg.at

Caisse des Dépôts, public interest missions: caissedesdepots.fr

Housing Europe, France social and affordable housing factsheet (November 2025): housingeurope.eu

Mietshäuser-Syndikat (federation overview): syndikat.org

Martín, A., Moral-Benito, E. and Schmitz, T. (2019) The Financial Transmission of Housing Bubbles: Evidence from Spain. ECB Working Paper Series No. 2245: ecb.europa.eu

Euronews, Spain €7bn public housing plan (April 2026): euronews.com

KfW (Kreditanstalt für Wiederaufbau), institutional overview: kfw.de

National Economic and Social Council (2018) Urban Development Land, Housing and Infrastructure: Fixing Ireland’s Broken System. NESC Report No. 145, Dublin: nesc.ie

Scottish Government, Empty and Second Homes in September 2025: gov.scot

Scottish Government, Planning Circular 4/2025: Planning Obligations and Good Neighbour Agreements (December 2025): gov.scot

Scotland’s Housing Network (2025) Development Value for Money Annual Report 2024/25: scotlandshousingnetwork.org