Sunak seals the Windsor deal. Stormont is still waiting.

Rishi Sunak and Ursula von der Leyen appeared together in Windsor on 27 February to announce an agreement over Northern Ireland that their predecessors had struggled to reach. The British prime minister called it a breakthrough. The European Commission president described a new beginning in relations. In Belfast, the Democratic Unionist Party reserved judgment while it examined the text.

The gap between those reactions defines the immediate politics of the Windsor Framework. London and Brussels have agreed a way to reduce trade friction and give Northern Ireland’s representatives a new mechanism to object to certain EU rule changes. They have not obtained the DUP’s agreement to return to the power-sharing institutions it has been boycotting over the post-Brexit arrangements.

The package covers goods moving from Great Britain, food, medicines, VAT and excise, as well as governance. It changes the operation of the Northern Ireland Protocol while retaining the basic purpose of avoiding a hard border on the island of Ireland. Formal adoption and implementation still lie ahead at the end of February.

Sunak has therefore completed a negotiation with one set of counterparts and begun another political argument with a different audience. Businesses want to know how the procedures will work. Unionists want to know what the agreement means for Northern Ireland’s place in the United Kingdom. EU governments want confidence that goods entering their market remain subject to the rules protecting it.

The deal moves the distinction from the ship to the destination

The core commercial change is a proposed separation between goods intended to remain in Northern Ireland and goods that may move onward into the EU. The UK describes the arrangements as green and red lanes. Eligible goods staying in Northern Ireland would face substantially reduced requirements; goods destined for the EU would continue through the full procedures protecting the single market.

This addresses a practical problem created by the protocol. Great Britain left the EU single market, while Northern Ireland retained alignment with specified EU goods rules to avoid checks at the Irish land border. Some requirements therefore fell on movements across the Irish Sea, including supplies within the United Kingdom. Businesses encountered a border-related process during what they regarded as domestic trade.

The new distinction depends on knowing where goods are going and having sufficient trust in the information supplied. A simplified route cannot function merely because a trader says a consignment will remain in Northern Ireland. Registration, commercial data and enforcement underpin the arrangement. The political promise of less friction is being exchanged for a more targeted way to manage risk.

That is why the deal is better understood as changing how controls work than making all controls disappear. The red route remains important because Northern Ireland shares an open land border with an EU member. The green route is intended to prevent that need from imposing the same burden on goods whose destination is clearly within the United Kingdom.

For food retailers, the package envisages simpler certification and changes intended to allow familiar products to move from Great Britain to Northern Irish shelves. Businesses still need to read the detailed requirements, including eligibility and labelling. A supermarket replenishing local stores and a distributor supplying customers across the Irish border may face different procedures even if their lorries arrive on the same ferry.

Medicines form a separate part of the agreement. The proposed changes allow UK authorities to regulate and authorise medicines for Northern Ireland through arrangements intended to secure the same supply as elsewhere in the UK. The objective is to remove an especially sensitive consequence of regulatory divergence: the possibility that a product available in Great Britain encounters a different route to patients in Northern Ireland.

The VAT and excise elements create additional flexibility for specified UK changes to apply in Northern Ireland. They do not turn the entire package into a general exemption from every EU-related obligation. Each measure needs to be assessed within its stated scope. The breadth of the political announcement should not obscure the fact that it is assembled from different legal and administrative changes.

Implementation will be staged, rather than occurring when the leaders leave Windsor. Officials must adopt the relevant instruments, businesses must adapt systems and authorities must communicate what is changing. A period of preparation can prevent disruption, but only if the promised timetable and guidance are clear enough for firms to make decisions.

The first commercial test is consequently quite ordinary: can a trader identify which route applies, supply the required information and move goods without an unpredictable delay? That is more useful than arguing whether the protocol has been renamed or replaced. The cost experienced by a business depends on the procedure that officials actually apply.

A brake with conditions attached

The most politically prominent innovation is the Stormont Brake. Under the announced mechanism, at least 30 members of the Northern Ireland Assembly from at least two parties could notify the UK government of an objection to specified amended or replacement EU goods rules. London could then prevent the change from applying through the process set out in the framework.

The mechanism is conditional. It is intended for exceptional circumstances involving changes with a significant and lasting effect on everyday life, after the required scrutiny and consultation. It is not a power for any group of Assembly members to remove whichever existing EU rule they dislike. Nor does an objection automatically create a general Northern Irish veto over the EU’s single-market legislation.

The distinction between a new rule and an amendment to one already covered also matters. Different procedures govern additions to the framework and changes within its existing scope. Presenting the brake as a single switch over every future EU measure would erase those distinctions and create expectations that the text does not necessarily support.

For unionists, the underlying issue is democratic authority. Northern Ireland remains affected by certain EU rules even though the United Kingdom is no longer represented in the institutions making them. The brake seeks to give local representatives and the UK government a stronger role in responding to changes. Whether that role is sufficient is a political judgment that the announcement cannot make on their behalf.

For the EU, any right to block a rule has to be reconciled with the protection of the single market. An exception that creates a material difference in goods standards can have consequences beyond Northern Ireland because the land border remains open. The framework therefore sits within a joint governance system rather than leaving each side free to define the effects of a suspension alone.

Existing EU law within the arrangement does not disappear, and the European Court of Justice retains a role where EU law applies. The deal may change the practical likelihood and channels of disputes, but claims that the court has been entirely removed from Northern Ireland’s trading arrangements go beyond what has been agreed.

The brake is also distinct from the protocol’s periodic democratic-consent arrangements concerning the continued application of specified provisions. One concerns particular future changes; the other concerns continuing consent to the broader framework. Conflating them would make it harder for voters and Assembly members to understand which decision they can influence and when.

The proposed mechanism assumes functioning democratic institutions able to scrutinise the rules. That creates an obvious tension with the current boycott. The political group most concerned about the democratic deficit must assess a device whose practical value depends on representatives returning to the institutions from which it has withdrawn.

By the end of February, the relevant question is therefore not whether the phrase Stormont Brake sounds strong. It is whether the triggering conditions, the UK government’s responsibilities and the consequences of using it provide a credible form of influence. Donaldson has asked for time to examine precisely that kind of detail.

The DUP’s decision is not Sunak’s to announce

The DUP has withheld participation in the power-sharing arrangements in protest at the protocol. Without the required cross-community participation, a fully functioning Executive cannot be restored. An agreement between London and Brussels can alter the conditions behind the boycott, but it cannot substitute for the party’s decision to end it.

DUP leader Jeffrey Donaldson’s initial response recognised progress while reserving a final judgment against the party’s tests. That position leaves room to negotiate and scrutinise. It should not be reported as acceptance simply because he did not immediately reject the deal, or as definitive opposition because he did not join the leaders’ celebration.

The problem for the party is that commercial improvement and constitutional reassurance are not the same test. Reduced checks can help businesses without resolving every unionist objection to differentiated rules within the United Kingdom. Conversely, a demand that every distinction disappear would be difficult to reconcile with an open Irish border after Great Britain’s departure from the single market.

Sunak’s task is to show that the package changes enough in practice and governance to justify participation. Applying public pressure too quickly could make a return look like a concession forced by London. Allowing scrutiny without any sense of direction could leave the institutions suspended while businesses wait for certainty.

Other Northern Irish parties and communities have interests that cannot be reduced to the DUP’s assessment. They want functioning government and many value access to both the UK and EU markets. A mechanism designed to answer unionist concerns must be considered alongside its effect on stability, representation and the ability to maintain those commercial advantages.

The immediate cost of institutional paralysis is borne through public administration. Decisions on services and spending still have to be made, but civil servants cannot provide the same political authority as accountable ministers. That gives restoration a purpose beyond validating a Brexit agreement: Northern Ireland needs institutions capable of deciding priorities that have little to do with customs or the single market.

The UK parliamentary arithmetic is different from the Northern Irish one. Opposition support could help Sunak secure approval for necessary measures even if some Conservative MPs object. That would assist implementation in Westminster. It would not restore an Executive in Belfast, where the structure of power-sharing creates a different requirement for consent.

This means the prime minister can achieve a legislative success and still face an unresolved institutional problem. The framework’s future should be assessed along both tracks. Treating a Commons majority as evidence that Northern Ireland’s political dispute is finished would repeat the mistake of confusing the authority to adopt an agreement with the agreement’s ability to command local support.

At this point no date for restored power-sharing can be responsibly announced. The texts must be examined, the legal measures adopted and the parties’ responses made clear. The next important political signal will come from those decisions rather than another statement from London that the deal has settled the argument.

A repair to the wider relationship, with work still attached

The agreement also offers London and Brussels a route away from the confrontation over the UK’s Northern Ireland Protocol Bill. The British government says the negotiated solution removes the need for that legislation, which sought powers to disapply parts of the existing arrangements. Moving through an agreed framework reduces the risk of a unilateral step generating a wider legal and political dispute.

That improvement has value beyond Northern Ireland. The UK and EU need to cooperate on Ukraine, energy and other international questions while managing a trading relationship that continues after Brexit. A dispute over compliance with one treaty can consume trust needed elsewhere. Resolving practical problems through negotiation gives both sides evidence that cooperation can still produce a result.

The announcement does not reverse Brexit or recreate UK membership of the single market. It addresses a particular territorial and political problem produced by the form of withdrawal the United Kingdom chose. Its significance lies in making that settlement more workable, rather than eliminating every consequence of the original choices.

By 28 February, Windsor has delivered an agreement in principle, detailed proposals and a more cooperative public relationship between Sunak and von der Leyen. The unresolved questions concern adoption, business implementation and Northern Irish political consent. Those are substantial tasks, but they are more clearly defined than the open confrontation the leaders inherited.

The deal’s success will eventually be visible in fewer disrupted consignments and institutions able to take decisions for Northern Ireland. For now, the photographs from Windsor establish that the UK and EU have reached terms. The empty spaces in Stormont establish what those photographs cannot yet claim.