Recently, online debates have emerged about the usefulness of the website TheyWorkForYou (TWFY), a tool which provides people with a view into the voting records of Members of Parliament. Articles and blog posts have rightly detailed the oversimplification of providing a snapshot of members’ votes, which does not make clear the nuance of particular motivations behind votes, or the context in which those votes take place (for example, Opposition Day motions or whipping strategies). However, there is a deeper problem which underlies the structure of voting in the Westminster Parliamentary System; the informal agreement of pairing.
Pairing is a convention within the House of Commons where if a member has good reason to not be on the Parliamentary Estate to vote on a motion, they are ‘paired’ with a member who intends to vote on the other side of the debate. This member would then abstain in order to restore parity to the vote. By its nature, pairing is considered an informal exercise among members and thus is not officially recognised or regulated by Parliamentary authorities. Instead, they are usually registered with and organised through the Whips Office. As a result, pairs are generally not made public. The system offers a compromise for the voting rights of members to be balanced with the flexibilities of adult life; members who have other significant personal or ministerial responsibilities can ensure that their votes are still considered. Similar conventions are also practised in Australia, the United States, and Canada.
It is exceedingly rare for members to break a pairing agreement, but not inexistent. One notable instance is in 2018, where Conservative MP Brandon Lewis broke a pairing with Jo Swinson to support the Government on two particularly tight votes. Lewis claimed this was a mistake and abstained on all other, albeit less tight, votes on that day.
A more common problem is found when members are explicitly denied pairs, either when whips are unable or unwilling to find someone who is willing to abstain. Refusals often occur in the event of significant motions. Famously this came to pass in 1979, when James Callaghan, then Prime Minister, was facing a vote of no confidence in the House. Labour MP Alfred Broughton was terminally ill yet determined to vote against the motion. A scramble then took place in the Whips Office to find a Conservative MP who was able to be paired with the ailing member, but none could be found. The story goes that Bernard Weatherill, a Tory whip at the time, offered to be paired with Broughton, but Labour whips refused to let him be paired, knowing that doing so would effectively end Weatherill’s political career. Ultimately, Callaghan refused to allow Broughton to travel to Parliament despite the fact that he did not have a pair, aware of the risk that the ailing member may die en route to the vote. Callaghan lost the no confidence motion by one vote, and Broughton died a few days later.
This case is not an outlier. In 2018, Labour MP Naz Shah was forced to come to the House of Commons in a wheelchair in order to vote on a Brexit vote. Shah had been in hospital suffering from severe nerve damage, and promptly returned to hospital after the vote. She applied to be paired with a Conservative MP but, on account that the votes were humiliatingly close, the Conservative whips refused her request. In 2016, the Opposition Labor Party in Australia went one step further by specifically refusing pairs for Government MPs in an attempt to test their razor thin majority.
Why does this matter? In a world where MPs’ votes are accessible by the click of a button, there is a significance in recording whether a member has abstained from a vote on political grounds (for example, they may not agree with the motion but not want to subject the government to a humiliating defeat) and those who are forced to abstain through the pairing convention. As abstentions are not listed in Parliamentary records, Alfred Broughton’s absence from the 1979 vote of no confidence is not differentiated from the absence of Irish Nationalist MP Frank Maguire, whose lack of vote on that day can be explained either as a political abstention, his being locked in a toilet, or his preoccupation with a bottle of whisky, depending on which account you believe. In today’s world, this is a pain for political commentators who are keen to see which member has chosen not to vote for the Government, but also fails to shield those who are unable to attend Parliament for entirely legitimate reasons from being accused of laziness by those who see they have abstained on websites such as TWFY. This is in contrast to other legislative systems which incorporate a system of pairing, such as the United States Congress, where members generally announce they are in a pair and their respective positions are listed in the Congressional Record.
There is also a second problem here. As pairs are often not permitted for crucial motions, such as no confidence votes, Mr Broughton would have been forced to leave his deathbed and enter the Parliamentary Estate (he would be permitted to be nodded through without passing through division lobbies, so long as he was on the Estate) in order for his vote as a Member of Parliament to be considered. Similarly, stories such as those of former Labour and Change UK MP Mike Gapes, who was told by a whip that it was okay to go to the hospital where his wife was giving birth, as long as he was back to vote on the Maastricht debate, ignore the fact that MPs are human beings who should not be expected to miss personal milestones at the risk of them and their constituents being disenfranchised.
So, what is the solution here? Firstly, pairs should be formalised in the Standing Orders of the House of Commons rather than being informally granted by the Whips’ Office. Taking this power away from whips would mean that abstentions can be recorded in Parliamentary voting records, which can note which MPs have abstained due to pairing constraints. Adopting this approach to publishing pairs would be similar to how tellers are recorded in Parliamentary records. If we work under the assumption that it is important for democratic polities that constituents are able to access the voting records of their elected representative, it is equally important for constituents to know if the member they elected is abstaining from a vote on political grounds or due to pairing constraints. This would alleviate some of the issues we see in websites like TWFY, where the nuance of Parliamentary procedure is not always clear.
An argument raised against formalising pairing is that MPs appreciate the discretion that informal pairing allows, offering them the opportunity to slip to a hospital appointment unnoticed. Yet a policy like this could be broadly implemented, where Hansard records that a member who abstained was in a pair, without providing further personal detail, such as their reason for being in a pair. This would add an additional layer of transparency that explains the member’s absence, which does not exist in current Parliamentary practice, while not compromising the discretion that informal pairing allows at present.
A further benefit of formalising and publishing pairing would be to avert any unscrupulous chicanery where members break pairs or whips double up members to pair in order to gain an advantage. An instance of doubling up took place in 1996, where Government whips were accused of pairing the same three Conservative MPs with six Opposition MPs. If pairings were published on Hansard, any instance of breaking or doubling up would be immediately clear, and formal consequences could be established in the House of Commons Standing Orders to adequately punish any offending member.
Secondly, the parameters of when pairing is allowed should be established in the Standing Orders. This would allow an opportunity for the House of Commons to consider updating the reasons in which members are allowed to be absent from votes, and whether it is then reasonable to allow pairing in crucial votes. This could avoid further unfortunate instances where members are forced to come in when seriously ill, and open up the opportunity for those who generally do not take part in pairing, such as Independent MPs who do not have a whip.
Additional appropriate measures could also alleviate some pressure on the pairing system, such as establishing a broader system of proxy voting, as was proposed in the 2016 Good Parliament Report and considered in the House of Commons Procedure Committee’s report in 2018. These, however, are issues for another blog post. This contribution seeks to argue for reform of the pairing system, which is an under-discussed but equally important base point. If pairing is to take place in a Parliamentary body, then in the interest of democratic transparency, these pairs should be formalised by the House and made available in Parliamentary voting records.
James Milton is a doctoral candidate at University College London
(Suggested citation: J. Milton, ‘The Antiquity of Anonymous Pairing’, U.K. Const. L. Blog (2nd February 2022) (available at https://ukconstitutionallaw.org/))