Written evidence from The UK in a Changing Europe[i] (EUR 64)

 

Public Administration and Constitutional Affairs Committee

Lessons Learned from the EU Referendum inquiry

 

 

Executive Summary:

 

 

What is the legal status of referendums?

 

  1. Unlike earlier referendum legislation, The European Union Referendum Act 2015 did not specify the legal status of the vote in the EU referendum. It was therefore assumed to be advisory and not legally binding. However, political statements always made clear that the government would consider itself bound by the outcome and so, irrespective of the silence in the Act about the legal status of any referendum, the outcome is de facto binding.

 

  1. The advantages of the approach adopted by the EU Referendum Act is that it gives a certain amount of wiggle room to the government of the day.  The disadvantage is that it creates suspicion that politicians may seek to avoid the will of the people. For this reason we would advocate a clearer statement in any future legislation about the legal status of the outcome of the referendum.

 

What questions are appropriate to be determined by referendums?

 

  1. More difficult is deciding what questions are appropriate to be determined by referendums. At one level, the EU referendum was the paradigm case, posing the simple question do the British people want to stay or go? However, the reality proved different: while staying was known, going was unknown. There was no route map as to what Leave actually looked like. This helped the Leave campaign because those voting Leave could fill the vacuum with their own version of leaving: a total withdrawal from the EU and no future special arrangements, a softer Brexit based on some continued arrangement, or something else entirely. Unlike a general election, where all parties produce a manifesto which can be scrutinised, in the EU referendum the Leave camp could rely on a few vague, but compelling, slogans. So in fact a momentous decision of this kind was subject to the lowest level of scrutiny. For this reason, in the crude form in which the question was put, a referendum was not a suitable vehicle for an issue of this magnitude.

The regulatory system for referendums 

 

  1. The current regulatory system for referendums both benefits and suffers from the atypical position of such devices in the British constitutional order. Because there had only been a small number of referendums prior to the Political Parties Elections and Referendum Act 2000 (PPERA 2000), there was scope not only to draw on good practice from other countries but also to implement this without having to overcome much ingrained practice domestically.

 

  1. Thus, the provisions are generally coherent, clear and effective. At the same time, the very novelty of referendums means that the growth in their use has presented a number of complications that had not been foreseen.

 

  1. As both the Scottish and the EU referendums demonstrated, there is a very high level of compliance with PPERA 2000 and with the decisions of the Electoral Commission, which has effectively established itself as an arbitrator in such matters. During the EU referendum, the only notable instance of possible gaming of the rules occurred when Grassroots Out created a number of sub-groups, which were not registered with the EC, potentially circumventing spending limits. However, this was noted by the press (Telegraph) and notified to the EC which removed the sub-groups from the register very quickly.

 

  1. The combination of concise legal requirements, extensive user guides for various roles and the rapid publication of data by the EC works well to ensure compliance with the letter and spirit of PPERA 2000.

 

  1. Where the system does not work so well is when the EC is drawn into issues of a more obviously political nature. The government’s pamphlet, produced just before the purdah period, while within the provisions, looked to challenge the values of fairness that run through PPERA 2000, as did questions around the effectiveness of purdah in general.

 

  1. The difficulty for the EC in this case was that while purdah was in effect, it also remained government policy to remain within the European Union, which blurred – even more than usual – the boundaries of what might be allowed. Moreover, groups on both sides of the campaign kept asking government officials for details of current and possible future policy, as they tried to find evidence to support their positions. Given the very broad potential impact of EU membership on almost every area of government policy, policing purdah was always going to be particularly difficult, for the EC, government and campaigners.

 

  1. With all of this in mind, the system stood up reasonably well, with no egregious failings. Indeed, if there was a problem with purdah, it was that it meant the government become overly concerned about any sign that it was preparing contingency plans for a ‘Leave’ result and so actively encouraged those best placed to make such plans to do absolutely nothing.

 

  1. Thought might be given to reviewing purdah, so that civil servants are able to conduct the necessary preparatory work to give effect to all outcomes of a referendum, which might well include talking and working with non-governmental individuals and groups (see discussion below). Such work would need to be closely defined by revisions to PPERA 2000 and the EC would need to have adequate resource to police it, so that charges of undue influence could not be levelled. While this would create some risk that the EC might be seen as an agent of the government, this is no more than is currently the case and would come with the significant benefit of closing one of the major failings of the EU referendum.

 

  1. Beyond this, the EC continues to rely on the government not to abuse its self-limitation, as described under PPERA 2000. While sanctions exist and are proportionate, the EC cannot let itself been drawn into overtly political matters if it is to maintain its credibility with the public. This has been pursued to date by ensuring that all relevant legal provisions are followed, which works until the legal provisions themselves are under question. The wording of the referendum question is a case in point, where the EC was fortunate that the government was prepared to revise this: if it had not, then the EC’s standing and authority might have been called into question. It is only because the EU referendum has been on the cards for so long that the EC was able to conduct the necessary preparatory research to underpin its decisions and pronouncements: in another case, such work might not be possible.

 

Should referendums on constitutional questions include a minimum percentage threshold?

 

  1. There was some discussion about a minimum percentage threshold at the time of the Referendum Bill but we understand the Prime Minister was opposed to it.  We think this is a mistake. Given the relative infrequency with which referenda are held in the UK, they are likely to be on matters of grave constitutional importance, including the UK’s continued existence as a state entity. It is therefore surprising that a 51:49 split would be sufficient for a change of such magnitude to be adopted. The Trade Union Act 2016 has imposed balloting thresholds prior to strike action taking place, strike action which may be existential to a company. Ex fortiori a decision which may be existential to a country should also be subject to threshold requirements. This is certainly the case in other democratic countries. In the 1979 referendum on the devolution of Scotland, endorsement by 40% of the registered electorate was required to secure approval, a threshold which was not reached, suggesting the absence of broad support.

 

  1. We therefore propose that there should be various thresholds, depending in part on the nature of the question posed and whether the default position is to continue with current state of affairs, or if a rejection of the referendum question would result in policy change. The thresholds should also be introduced dependent on the legal status of the referendum – i.e. if it is to be legally binding or not:

-        Turnout (e.g. the Netherlands has a 30% rule for the referenda to be binding)

-        A super majority of those voting to be in favour

-        Majority support from the constituent parts of the UK (i.e. Scotland, Wales and Northern Ireland) [I find this very tricky and know such requirements have been rejected in several other countries. So I would qualify this to only apply in matters where explicit geographical/territorial interests diverge].

 

The role of the machinery of Government during the referendum campaign

 

  1. Most government departments played a relatively small role during the campaign. The key exceptions were the Cabinet Office, which published a number of documents in late February/early March (“Alternatives to membership: possible models for the United Kingdom outside the European Union”, “The process for withdrawing from the European Union”, and “The  best of both worlds: the United Kingdom's special status in a reformed European Union); and the Treasury, which published two major pieces of analysis on the long and short term impacts of leaving the EU in May, the second just before purdah began.

 

  1. These documents were clearly not “neutral”; that is, they supported the government’s policy agenda with respect to the EU and hence, although they were not campaigning material, the arguments of the Remain campaign. In the case of the Treasury documents, they were clearly constructed to do exactly that. They were also selective in their coverage: for example, the Treasury analysis covered the impact of Brexit on trade, but not on the equally important issue of migration.   In this context, the decision to cover certain topics and omit others was clearly political.

 

  1. However, the legal, practical and economic analysis contained in the documents was at least of the quality in terms of analytical standards of most government documents produced to support policy.  The Treasury documents, in particular, used standard, accepted methodologies and produced results that were qualitatively similar to those produced by respected independent economists (in particular, NIESR, the Centre for Economic Performance, and the IFS).  

 

  1. However, while the analysis in the documents was couched in relatively sober terms, the public presentation by the Remain campaign, including by serving Ministers, was not.  The Treasury analysis contained a number of scenarios for the potential economic impact of Brexit; the forecasts therefore incorporated both the normal (large) uncertainty related to economic forecasts as well as uncertainty as to which scenario (if any) will materialise.  However, the Chancellor and others made a number of statements that “Britain would be poorer by £4,300 per household”; a level of certainty that was not justified by the underlying analysis. [Is this necessary? Seems to be a critic of political personalities rather than address the question of the role of the bureaucracy…]

 

  1. Arguably, this was no different from normal practice in general election campaigns (before purdah). Before the last election, the Treasury produced (at the request of Ministers) analysis on the impact on borrowing of different approaches to fiscal policy. This was factual analysis, but highly selective in the factors included and excluded, and the scenarios modelled; and it was used by government Ministers for explicitly political purposes (to attack the policies of other parties) using language that was clearly not justified by the underlying analysis. 

 

  1. In one respect, however, the Treasury clearly overstepped the bounds of normal practice. When the report on the short-term impacts of Brexit was published, just a month before the referendum, the front page of the Treasury website had a banner headline reading “UK economy would fall into RECESSION if Britain leaves the EU.”  This both misrepresented the analysis and was very obviously partisan in the context of the campaign. https://twitter.com/jdportes/status/734862712359034880

 

  1. This is entirely improper on any possible reading of the Civil Service Code, and it is disappointing that senior Treasury civil servants, including the Permanent Secretary, allowed it to happen.   This sort of incident is deeply corrosive of impartiality – both actual and perceived – and should not be allowed to recur. It is at least conceivable that it helped the attacks on expertise that characterised the campaign and could have damaging consequences going forward to gain some traction.

 

  1. While the economic analysis produced by the Treasury was sound, it could equally well have been produced by the OBR (which has at least as much macroeconomic expertise), consulting Treasury officials on technical issues, just as it does for normal forecasts. This would have had a number of advantages: it would have been less selective in its assumptions, more credible, and the Treasury would not have had its independence compromised. Of course the arguments for the OBR having a role in estimating the impact of different macroeconomic policies have already been well rehearsed in the context of general elections; the same arguments apply here but with even more force.

 

  1. Pursuant to our discussion of purdah above, the question arises as to what level of contingency planning should be provided by the Civil Service as to possible referendum outcomes? For example, should it adopt a model similar to that found in General Elections (whereby the Civil Service holds meetings with the Opposition before the result is known to prepare to implement its programme if necessary)?

 

  1. The lack of planning by the civil service for a Brexit vote has attracted considerable criticism. In defence of the civil service, it is not entirely clear how much difference this would have made. 

 

  1. The pre-election meetings held between the Opposition and senior civil servants have several purposes:

-        Shadow Ministers can sense-test the realism of their policies in advance;

-        Shadow Ministers can meet their potential senior team

-        Civil servants can learn potential Ministers’ priorities so preparations are in hand

However, none of the above are necessarily relevant in a referendum context. It seems unlikely Vote Leave would have altered their sometimes contradictory and occasionally entirely misleading promises (most obviously the “£350 million”) if they’d been told privately by Treasury officials what was already obvious publicly. And it was entirely unknown what, if any, the responsibilities of senior Vote Leave figures would be post-referendum.

 

  1. However, that does not mean that, as with general elections, civil servants should not and could not have prepared policy papers and “briefings for incoming Ministers” on key Brexit topics. Again, it is not entirely clear that this would have made a material difference; the broad strategic choices (between, say, EEA-type options and much looser arrangements; the modalities of Article 50) were well aired in the referendum, and while it would no doubt have been preferable to have detailed papers on these and other issues available immediately after the result, much of the current vacuum results from other factors (new Ministers and machinery of government changes) rather than insufficient analysis. However, it certainly would have done no harm. Moreover, it might have reassured both incoming Ministers and the public as a whole that the civil service was both impartial and competent enough to deliver Brexit as efficiently and effectively as possible; this perception alone would have been a significant gain.

 

  1. The lack of planning for a ‘leave’ outcome has had a significant impact on the capacity of the current administration to give effect to the result, which in turn creates a major issue around the democratic legitimacy of referendums as tools for public policy.

 

  1. There does seem to be a certain amount of confusion over key empirical issues such as whether the UK can simply have a WTO type relationship with the EU straight away, what kinds of ‘access’ to the single market could envisioned and so on. Some kind of contingency planning along these lines might have helped structure initial debates within the civil service.

 

  1. Perhaps mention the sporadic coordination between departments – and between Cabinet Office, No 10, Treasury and FCO in particular?

 

  1. There was limited consultation and discussion with European partners in the year leading up to the referendum. The government would have benefitted from lessons learned from recent referendum campaigns in other countries (esp. Netherlands, Denmark) and preparation for Leave and Remain scenarios could have been at very different stage by now if the Government machinery had engaged in comprehensive discussions with other EU governments, the EU Commission and European Council Presidency.

 

September 2016

 

 

[i] The UK in a Changing Europe (www.ukandeu.ac.uk) funded by the Economic and Social Science Research Council (ESRC) promotes rigorous, high-quality and independent research into the UK’s relationship with the EU, providing an authoritative, impartial reference point for those looking for information and analysis about UK-EU relations. Based at King’s College London, it is directed by Professor Anand Menon and 15 fellows are involved with it – leading experts from universities and institutions around the UK. Several members of our team have contributed to the evidence provided below.