During the Parliamentary dramas of 2018 the government narrowly avoided defeat on key Brexit legislation only by conceding a series of amendments proposed by the European Reform Group (ERG). The most obvious pro-Brexit amendment locked in 11.00pm on 29 March 2019 as “exit day”, paving the way for a no-deal Brexit. However, the ERG also sprinkled crucial pro-hard Brexit provisions through other legislation. One of those amendments requires urgent attention as its effect could be to make the government’s Withdrawal Agreement unlawful, even if the House of Commons were to vote at the last minute to approve it.

Approving the deal is not enough

The Prime Minister’s strategy appears to be to delay the “meaningful vote” so that MPs are presented with a binary choice between the Withdrawal Agreement (supported by any further assurances that might form the “Cox Codicil”), or no deal. It is possible that this stark choice could be presented to MPs only days before the 29 March deadline.

If the government’s gamble were to pay off, then the Prime Minister might secure a House of Commons vote in favour of her deal, reversing the 230-vote defeat suffered on 15 January. However, a House of Commons resolution would not be enough to make the Withdrawal Agreement lawful. A resolution cannot override primary legislation, so a new Act of Parliament is required to give the Withdrawal Agreement legal effect, and to override the pro-hard Brexit provisions introduced by the ERG.

The problem lies in section 31(5) of the Taxation (Cross-Border Trade) Act 2018. That Act received royal assent on 13 September 2018. In its original form, section 31 allowed the UK government to enter into a customs union with “any country or territory” by means of an Order in Council. However, section 31(5) was added as a late amendment to avert a government defeat. It reads:

In the case of a customs union between the United Kingdom and the European Union, Her Majesty may not make a declaration by Order in Council under subsection (4) unless the arrangements have been approved by an Act of Parliament.

Taken together, section 31 means that any new customs union arrangement must be made by Order in Council (which is not part of the Withdrawal Agreement process) and section 31(5) means that any such Order made in relation to the EU would require a full Act of Parliament.

How does that affect the Withdrawal Agreement?

The Withdrawal Agreement provides for a form of customs union between the UK and the EU – initially under the transitional arrangements and then under the Northern Ireland backstop. Those arrangements will be valid only if approved by a new Act of Parliament that explicitly overrides section 31(5) of the Taxation (Cross Border Trade) Act 2018.

From a legal perspective, the nightmare scenario is that the government secures last minute approval of the Withdrawal Agreement, but does not have time to pass an Act of Parliament specifically implementing the deal. If the deal were approved on 26 March, then that would leave only 2-3 sitting days to secure an Act of Parliament. That legislation might be rushed through the House of Commons, but could still be vulnerable to filibustering in the House of Lords. Even if approved earlier in March, it would be a close run thing to pass primary legislation before 29 March.

It is also possible that legislation rushed through in haste at the last minute might omit or overlook the need for provisions to deal with, and reverse or repeal, the ERG’s “hard Brexit” amendments in the EU (Withdrawal) Act 2018 and, crucially, in the Taxation (Cross-border Trade) Act 2018. The ERG’s amendments are scattered through Brexit legislation like Voldemort’s horcruxes in Harry Potter. To avoid a “hard” Brexit, each must be located and dealt with.

What happens if the Withdrawal Agreement is not properly authorised?

If Parliament runs out of time to pass a full Act of Parliament – or if it just overlooks and forgets to deal with section 31(5) – then anything done in reliance on the Withdrawal Agreement would be subject to legal challenge. Goods imported or exported without tariffs, customs declarations and other border clearance formalities would potentially be subject to retrospective imposition of those costs. Arguably, other aspects of the Withdrawal Agreement including the provisions allowing transfers of personal data from the EU to the UK would also be subject to legal challenge. More generally, if Parliament does not legislate in time, and in full, to authorise the Withdrawal Agreement and to deal with the ERG amendments before 29 March 2019, then the whole Withdrawal Agreement might be challenged as being ultra vires the government. That argument might run for months, and would almost certainly reach the Supreme Court. In the meantime, business would be forced to operate in continuing uncertainty.

Surely Parliament will get it right?

If there is a House of Commons majority in favour of the government’s Withdrawal Agreement then, however close to 29 March, it would surely be capable of passing the necessary legislation in time. Even if that deadline were missed by a few days, it would be open to Parliament to pass an Act with retrospective effect. However, that assumes both normal Parliamentary procedures and continuation of the current Parliament. Neither assumption is safe.

The government’s position in the House of Commons remains precarious. A vote in favour of the Withdrawal Agreement, driven by an urgent need to avert a “no deal” Brexit, would not necessarily rule out a further vote of no confidence in the government. If such a vote were to pass, then there would be a 14 day period under the Fixed Term Parliament Act 2011 to form a new government, failing which Parliament would be prorogued and then dissolved in readiness for a general election. Once prorogued, there would be no possibility of passing the required primary legislation until after the general election. That would cast doubt on anything done in the meantime in reliance on the Withdrawal Agreement. It is, of course, also possible that the general election might produce a radically different House of Commons – perhaps even one with a “hard Brexit” majority.

On any scenario, if the government persists with its apparent strategy of pressing for a last-minute “deal or no deal” vote, then it must ensure that it leaves enough time to pass primary legislation to ensure that the Withdrawal Agreement is not struck down as unlawful. To secure approval of the deal, only to fail to secure the necessary primary legislation, would be a pyrrhic victory.