Cross-posted with permission from the UK Constitutional Law Association Blog. The original post appears here.
—Stephen Tierney, University of Edinburgh
The past three weeks have seen a steady backlash against the referendum. It is understandable that many don’t like the outcome, after all 48% voted for Remain. But beyond disagreeing with the decision to leave the European Union, was the process so faulty that it can reasonably be declared illegitimate? I would argue no.
I make an immediate exception for those who, as a matter of principle, oppose the use of referendums. The appropriate balance between representative and direct democracy is both delicate and contested and it can be validly argued that referendums have no place in a properly functioning parliamentary democracy. But if the principle of using referendums in constitutional decision-making is accepted, as it widely is, then we must assess the Brexit referendum by its mode of organisation and the legality of its process. By this measure it is hard not to conclude that the referendum accorded with international standards for direct democracy, that it was conducted in accordance with law and that it was effectively regulated. On this basis I argue that the result is one which citizens should agree to, even if it is not one they agree with. The inevitable, if unpalatable, conclusion is that the outcome was lawfully arrived at and should be faithfully implemented.
Objections to Referendum Democracy
Two main concerns with the very idea of referendum democracy are relevant to the debate. One is that referendums, by definition, lend themselves to manipulation by the elites who organise the vote. A second is that referendums have an in-built tendency merely to aggregate pre-formed opinions rather than to foster meaningful deliberation.
I have argued elsewhere that these are in fact problems of practice rather than principle which can be largely overcome, as they were in the Scottish independence referendum, by good process design. This is not to suggest that a referendum will thereby become a perfect exercise in public reason, but simply that the deliberative credentials of a well-constructed referendum can match or exceed those of representative elections, particularly when we consider the latter’s own imperfections. I argue that by this measure the Brexit process, while not without its deficiencies, is not an example of bad referendum practice.
Elite Control and the Brexit Referendum Process
An implicit assumption behind many critiques of referendum democracy is that the executive can brazenly model the referendum process to suit its objectives. Certainly we must recognise that the Brexit referendum was called at the discretion of the Government. There is therefore a question-mark about the absence within the UK constitution of a full delineation of the matters which should or should not be put to the people. But if this is the measure by which the legitimacy of the Brexit referendum is to be questioned we must lay the same charge at the door of every UK referendum; for example, the AV referendum originated from a deal to form a coalition government. It is also the case that the House of Lords Constitution Committee considered this issue carefully in 2010 and in the end concluded that it was for Parliament on a case by case basis to determine if and when referendums ought to be held: ironically the justification for this was the preservation of Parliament’s preeminent constitutional position, a bolstering of representative over direct democracy. In any event, even if it is the case that some legislative regulation of what is a suitable subject to be put to referendum is constitutionally appropriate, a strong argument has been made that EU membership is an issue of such salience and long-standing controversy that it merited a national vote.
Beyond the initiation issue a rigorous statutory system, based upon the Political Parties, Elections and Referendums Act 2000 (PPERA) and the carefully-tailored European Union Referendum Act 2015, served to regulate the running of the referendum. The setting of the question is a good example of how robust this statutory regime was. The 2000 Act imposes a duty on the Electoral Commission to test the intelligibility of any referendum question. The Commission, in its review of the Brexit question, recommended a change to the original wording, which asked “Should the United Kingdom remain a member of the European Union?”. This was deemed to be a leading question. Notably the Electoral Commission’s suggestion of a more neutral question was accepted by Downing Street on the very day of the Electoral Commission’s recommendation.
The 2015 Act also implicitly regulated the timing of the referendum. A four month notice period was built in for any change to important procedural rules. This in effect prevented the Government from changing the date of the referendum to go for a snap poll, had it sought to do so. The timing of the referendum can be criticised in that it came less than seven weeks after elections to the devolved legislatures on 5 May, and there is a reasonable argument that this did inhibit the full playing out of the issues in the devolved territories. But given that both Scotland and Northern Ireland saw majorities vote to Remain, it is difficult to conclude that this gave the Leave campaign any advantage, thereby undermining the result.
Parliament also showed its teeth over Government attempts to interfere with purdah rules. Section 125 of PPERA restricts the publication of promotional material by central and local government. There was controversy over a Government attempt to limit the operation of s.125 so that it would apply only to publications ‘directly relevant’ to the referendum. This provoked a rebellion by backbenchers and a defeat for the Government in the House of Commons. As a result, section 8 of the 2015 Act largely reinstated the purdah rules.
It is one thing to regulate the influence of government, but what about the other criticism: that citizens simply can’t be mobilised to take ownership of a referendum in an informed way? Leaving aside the implicitly condescending premise of this argument, I have previously contended that it is hard to sustain in light of the Scottish independence referendum. It can of course be countered that citizen engagement and levels of information in the EU referendum do not compare well to the deliberative strengths of the Scottish referendum, but it is difficult to measure this. The turnout on 23 June of 72% is higher than in recent general elections (considerably higher in many low income areas), suggesting that people living in places of high deprivation were particularly moved by the campaign and felt highly motivated to vote.
Another argument is that people were unclear as to what they were voting for. The Electoral Commission’s role in regulating the question goes some way to answer this objection. It is also the case that the options before the people on 23 June had been stable for some time. In this respect the EU referendum arguably contrasts favourably with the Scottish process. In the latter referendum a promise of greater powers for the Scottish Parliament was tabled by the UK Government in the very week of the referendum. Since these were not set out in any detail, this move served to increase the uncertainty of the ‘No’ option. In contrast, on 19 February 2016 the Prime Minister struck a deal with the EU on a reform to the United Kingdom’s position within the Union. It can therefore be argued that the Remain option in 2016 was considerably clearer than the No option was in Scotland in 2014. Most voters still opted for Leave, despite the fact that voters had four months to digest the agreement secured by Mr Cameron. Of course it can be argued that voters did not know what ‘Leave’ would mean in practice, but uncertainty about the future is not uncommon in referendums: it is a common tactic of ‘status quo campaigners’ to emphasise the insecurity of the change option in a referendum. And let’s not forget that 45% voted for Scottish independence with issues such as a currency union with the UK and future EU membership still unclear.
Deliberation, in order to be democratic, must include the whole demos of course. One question is, were the democratic credentials of the referendum undermined by the restriction of the franchise? UK nationals and Irish and Commonwealth citizens satisfying residence requirements (2015 Act, section 2) were entitled to vote, but nationals of most EU member states were not. There are of course political objections to this, but the franchise largely followed that for UK parliamentary elections just as the franchise in the Scottish referendum followed that for Scottish Parliament elections. It is also notable that the Court of Appeal rejected a challenge on EU and human rights law grounds. In my view a more convincing concern rests upon a vision of the UK as a multinational union in which the consent of each of the constituent territories should be required for such a significant change. This raises a broader question about the federalisation of the UK (the full engagement of the devolved territories in the exit process will be essential for the stability of the state), but this is not the main objection of those who now question the legitimacy of the referendum result and was rarely heard beyond the borders of the devolved territories before the referendum.
One of the main criticisms of referendums is that they can be the subject of widely distorting campaign expenditure, with voters subjected to heavily disproportionate advertising from one side in the campaign. This serves to unbalance the environment within which citizens are expected to deliberate. Here UK law stands out as a model of good practice. The PPERA regime is very rigorous in how it polices both funding and spending, and the 2015 Act supplemented this with very tight restrictions on the funding of the referendum campaigns by foreign sources. A balance was reached between the spending limits of the main campaign organisations and the political parties. This potentially gave a considerable advantage to the Remain campaign which was supported by most of these parties. The Conservative Party, which was of course divided on the issue chose not to spend its money on either side. Giving significant spending power to political parties may seem unfair, but this charge has to be balanced with the right of parties to engage based upon the support they have earned; it seems that the difficult weighting between equality and popularity is appropriately struck in UK law. The fact that the Electoral Commission oversaw the setting and implementation of these spending and funding limits (interim reporting being an innovation from the Scottish referendum that was adopted for the Brexit vote) also went a long way to satisfying crucial conditions for effective deliberation in the referendum.
Another issue is provision of information. Here of course there are allegations of lies or at least exaggerations by the main campaigns. I was a co-signatory of an open letter to the Telegraph criticising each of the campaigning groups. The Commons Treasury Committee also upbraided both sides for making statements that were ‘misleading’ and ‘tendentious’. By this measure there are legitimate concerns with how the issues were presented. But it has not been shown that these factors were any more distorting in the Brexit referendum than they have been in other referendums or elections.
There have been suggestions that the UK Government used its position to promote the Remain campaign through spending public money. A pro-EU document was sent to every household that is estimated to have cost approximately £9 million, while the Treasury issued pessimistic predictions of the short-term and long-term economic impact of EU membership. But the Government did not have an entirely free hand. The Electoral Commission regulated information that was to be sent to the public, and strict advertising and broadcasting rules applied to both sides of the campaign. One thing we didn’t see was a concerted effort by the European Union to influence the campaign. This has been a distorting influence in other referendums around Europe on EU treaty ratification referendums. UK legal regulations, and no doubt a sense that this would back-fire, helped keep the EU institutions out of the campaign.
Implementing the Result
I take the view that referendums should be held infrequently and are best reserved for issues of the highest constitutional significance. This helps stimulate the engagement of citizens and provides a careful balance between the norm of representative government and the exception of direct democracy. By this measure the Brexit referendum measures up. It was only the third national referendum in the history of the United Kingdom: people knew they were making a land-mark decision on 23 June.
One final question is whether it was appropriate to settle such a significant issue by a simple majority of those who voted (the narrowness of the majority contrasts with the 67% who voted to stay in the EC in 1975). However, apart from the devolution referendums of 1979, this is established practice in UK referendums and it would be inconsistent to single out the Brexit referendum for objection on this ground. That said, direct democracy does not exist in a vacuum; the narrowness of the result emphasises the importance of Parliament playing a full role in informing and scrutinising the implementation of the referendum result.
To conclude, given that the process was rigorously regulated with no voting irregularities or apparent breaches of funding or spending rules (a final audit by the Electoral Commission has not yet been published), the referendum outcome should be seen for what it is: a narrow but clear constitutional decision of the highest significance. It would be a democratic travesty for the result not to be accepted simply because many of us don’t like it.