Green Book reform will succeed only if place-based cases force investment choices

The reform is a test of decision quality, not local ambition

The Government’s 2026 announcement on Green Book reform puts place-based business cases at the centre of a wider attempt to make public investment more responsive to local barriers to growth. The stated direction is to assess how interventions work together in a place, rather than treating every scheme as an isolated proposition (HM Treasury et al., 2026 ↗).

That is an important shift for the UK. Local economic problems rarely arrive in departmental form. A high street may have available premises but weak footfall, poor business support and an unaffordable or unreliable transport offer for workers. A local employer may have vacancies but face skills shortages, childcare barriers or poor access to sites. Housing delivery may depend on remediation, infrastructure capacity, planning certainty and a market willing to build. In each case, a sound individual project can still fail to produce the intended outcome because another constraint remains binding.

The central editorial judgement is therefore straightforward: place-based cases should be judged by whether they improve choices under scarcity. They should not become a more elaborate way of validating a pre-existing local wish-list.

The 2025 Green Book Review anticipated this challenge. It proposed place-based business cases and pilots in Birmingham, Liverpool, Plymouth and Port Talbot, while arguing that a local baseline should capture the combined effects of existing national, regional and local plans rather than assume an artificial zero-investment world (HM Treasury, 2025 ↗). This matters because an appraisal that ignores committed activity can overstate the additional value of a new package; one that ignores interactions can understate the cost of acting in fragments.

The opportunity, then, is not simply broader appraisal. It is better decision design: a way to establish what matters most, what must happen first, and what should not proceed at all.

From project lists to causal portfolios

The Green Book already provides the raw material for more sophisticated appraisal, including a strategic case, theory of change, options appraisal, risk assessment and consideration of distributional and place-based effects (HM Treasury, 2026 ↗). The practical weakness is that these disciplines can still be applied scheme by scheme while dependencies are dispersed across institutions, funding streams and delivery partners.

A place-based case should therefore be treated as a constrained investment portfolio. The National Audit Office defines a portfolio as a collection of programmes and projects managed to optimise strategic benefits or operational efficiency within a funding envelope, with the capacity to adapt to changing circumstances (National Audit Office, 2022 ↗). The funding-envelope condition is crucial. If every intervention remains a priority after appraisal, the process has not established a portfolio; it has recorded demand.

A useful case starts with a narrow causal proposition. It should identify a defined outcome—such as sustained employment, viable SME occupancy, housing delivery or improved access to opportunity—and then ask which constraints genuinely prevent that outcome. The answer may be transport, skills, land readiness, management capability, finance, demand or service capacity. But it should not be assumed. A constraint is binding only if alleviating it would materially change the expected result.

This produces a more demanding sequence of questions:

1. What is likely to happen without the proposed portfolio, taking committed activity into account?

2. Which two or three constraints are preventing a better outcome?

3. Which interventions address those constraints, and through what mechanism?

4. Which components are interdependent, and which can stand alone?

5. What is the minimum viable package, and how does it compare with doing less, doing something different or delaying action?

6. What evidence would justify pausing, redesigning or stopping an element?

The distinction between a plan and a business case matters here. The Local Growth Plan framework asks Mayoral Strategic Authorities to set long-term priorities that connect local opportunities, national growth sectors and devolved powers (Ministry of Housing, Communities and Local Government, 2026 ↗). That strategic direction is necessary. A place-based business case should perform the harder task: choosing an investable sequence from within it.

Complementarity is plausible; it is not self-proving

The strongest argument for reform is that some public investments create value only in combination. A new transport link may improve access to work only where services are frequent, fares are usable, last-mile connections work and employers can recruit locally. Town-centre investment may improve the quality of the environment but fail to support enterprise if commercial terms, business advice, safety or local purchasing power remain unfavourable. An employability programme may lift participation only when transport, childcare, health support and employer demand are aligned.

These are plausible mechanisms, not automatic outcomes. Academic evidence gives reason for caution. A recent synthesis of place-based policy finds that results are highly context dependent and that evidence on complementarities within multi-part packages remains limited (Gaubert, Hanson and Neumark, 2025 ↗). Earlier work similarly emphasises the difficulty of appraising spatial interventions that may affect private investment, local economic activity and welfare through several channels at once (Duranton and Venables, 2018 ↗).

This does not make place-based investment a poor proposition. It changes the burden of proof. Cases should distinguish between effects that are directly evidenced, effects that are plausible but uncertain, and benefits that are strategic aspirations rather than appraisal-ready estimates. A package does not gain credibility merely because its components appear connected on a diagram.

The same discipline applies to local productivity. A business support initiative may help firms adopt better management practices, invest or access markets, but it should not claim all subsequent turnover growth as a programme benefit. Similarly, a regeneration scheme should not count the same underlying economic change separately as employment, land value, fiscal return and wellbeing without testing overlap. The Green Book’s treatment of uncertainty, non-monetised effects and distributional impacts is relevant precisely because a single headline benefit-cost ratio cannot safely carry all of these claims (HM Treasury, 2026 ↗).

Four risks the new approach must confront

**First, the ‘everything is connected’ problem.** Interdependence can become a rationale for avoiding prioritisation. Every core component should have a named mechanism, accountable owner, critical dependency and expected contribution. If removing it makes no material difference to the theory of change, it may be beneficial but it is not central to the portfolio.

**Second, attribution and double-counting.** Place-based initiatives operate alongside market change, national policy, private investment and other public programmes. Cases should be explicit about what is additional, what may be displaced from elsewhere and what cannot be reliably attributed. This is especially important where rising property values or business activity may reflect broader market conditions rather than the intervention itself.

**Third, objective confusion.** Better place-sensitive appraisal can give proper weight to equity, resilience, wellbeing and local economic value. It cannot abolish trade-offs between those objectives, affordability and national productivity. Evidence reviewed by Bartik indicates that place-based jobs policies need careful targeting, and that broad business incentives can carry high costs per job while complementary public services and infrastructure may be more effective in some circumstances (Bartik, 2020 ↗). The implication is not that business support is ineffective; it is that the policy instrument must fit the diagnosed problem.

**Fourth, uneven institutional capability.** Better-resourced places may be more able to assemble data, commission analysis and negotiate with delivery partners. The Government has said that No10 North will support engagement with the new process, including for areas without mayors (HM Treasury et al., 2026 ↗). That support should be assessed not by the number of submissions completed but by whether local institutions retain analytical, commercial and programme-management capability once external support ends.

A practical architecture for local investment decisions

Sanctuary’s recommendation is to organise major place-based cases around a portfolio board and a limited number of decision gates, rather than an annual narrative refresh.

**Establish a credible baseline.** Combine committed public investment, service performance, demographic trends, market conditions and known delivery constraints. This reflects the 2025 Review’s place-specific business-as-usual principle and prevents new proposals from taking credit for changes that are already expected (HM Treasury, 2025 ↗).

**Identify a small number of binding constraints.** Evidence should combine local administrative data with employer, resident, supplier and investor insight. For an employment-focused portfolio, for example, the issue may be employer demand rather than training volume; for a high street, it may be weak commercial viability rather than public-realm quality. The diagnosis should be capable of being disproved and revised.

**Compare bundles, not only preferred projects.** Decision-makers should test a minimum viable bundle, a more coordinated package and credible alternatives. Each should be assessed against cost, timing, distributional effects, commercial feasibility, operating requirements and delivery capacity. This makes the marginal value of an additional component visible.

**Make delivery roles real before approval.** The commercial case must identify who bears risk, whether suppliers have the relevant capability and whether partners can provide the people, finance and governance required. The Green Book specifically requires consideration of supplier capability, willingness and resources to manage proposed risks (HM Treasury, 2026 ↗). For SME, skills and town-centre interventions, that means engagement with local firms, anchor institutions, landlords and community organisations before—not after—the preferred option is fixed.

**Release funding through observable gates.** Later stages should depend on conditions such as land assembly, a tested procurement market, contracted employer placements, confirmed co-investment, occupancy thresholds or service capacity. Leading indicators matter because they show whether a mechanism is beginning to work before final outcomes are visible. Where a prerequisite fails, leaders should redesign, defer or close that component rather than preserve the original package for reputational reasons.

This approach is more exacting than a conventional regeneration narrative, but it is also more useful. It helps places distinguish between activity that is locally valued and activity that is necessary to unlock a stated economic outcome.

What success should look like after reform

There is a legitimate concern that a place-level layer could slow approval processes that are already complex. The answer is proportionality, not a return to fragmented appraisal. Small, low-risk interventions should use light-touch analysis. Large, irreversible or highly interdependent portfolios warrant deeper work. The Green Book already states that appraisal effort should be proportionate to an intervention’s scale, cost, complexity and risk (HM Treasury, 2026 ↗).

The reform should ultimately be judged by behaviour rather than the number of place-based cases produced. Are weak projects stopped earlier? Are infrastructure, skills, business support and land interventions sequenced more reliably? Are assumptions and trade-offs open to challenge? Do places retain the capacity to revise a portfolio when evidence changes?

If the answer is yes, place-based business cases could improve how the UK turns local strategy into investment decisions. If the answer is no, the reform risks creating broader narratives with no sharper choices behind them. The decisive principle is simple: local potential is not a business case. It becomes one only when a place can show what is constraining progress, why a defined package is the best response, and when it will stop spending if the mechanism does not hold.