The Transparency Paradox: Why Transparency Alone will not Improve Campaign Regulations

The role of new sources of data has become of increasing interest to those involved in political campaigning and a legislative focus of policy makers and regulators. Utilising Karl-Heinz Nassmacher’s ‘magic quadrangle’ of ‘accounting, practicality, sanctions and transparency’ and a case study of the Political Parties, Elections and Referendums Act 2000 this article unpicks how successful the ‘guiding philosophy’ of transparency was in delivering increased citizen confidence in the democratic process. I ultimately argue that at the heart of all discussions about what regulation in this area should look like, an uncomfortable paradox has to be accepted: that transparency may well help to quell actual instances of malfeasance and the misuse of data, but may at the same time increase citizen distrust in democratic processes. Any regulation should consider the ways in which transparency might be implemented such that it better supports the stated legislative aims.


Introduction
SINCE AT THE very least, the victories of Donald Trump and the Leave campaign in the Brexit referendum, the ways in which new forms of data play a role in political campaigns (and their eventual outcomes) are ripe areas of public discourse. Documentaries such as Netflix's The Great Hack and Channel 4/HBO's drama Brexit: The Uncivil War have put digital campaigning at the front and centre of popular conceptions of how politics and democracy function in the twenty-first century (whether entirely accurately is an argument for another day). It is not surprising, then, that if this challenge has seeped into the minds of the gatekeepers of popular culture, it has become a legislative focus of policy makers and regulators alike. In the United Kingdom alone, there have been (at least) fourteen reports written in recent years which unpick the extent to which these trends represent a threat to our democracy as we know it.
For example, in their review of the 2019 general election, the Electoral Commission showed that public concern surrounding 'misleading campaign techniques' (alongside bias in the media and 'complaints raising concerns about the presentation, tone and content of election campaigns') was particularly high. This public concern-whilst certainly reflecting a general as well as specifically online malaise-was focussed on digital. More than half of respondents to an Electoral Commission post-election survey (58 per cent) agreed that (in general) 'campaigning online is untrue or misleading', with a similar number disagreeing that 'information online about politics is trustworthy'. Of the nearly one in five people (18 per cent) that suggested they were not confident the election was well run, 49 per cent suggested that this was because 'campaigning was based on incorrect information/untrue claims'. Finally, when respondents were asked to prioritise their concerns about the election, 52 per cent said that 'inadequate control of political activity on social media' was a problem. 1 This has led the Electoral Commission (in 2020) to reiterate recommendations from their 2018 report, Digital Campaigning-Increasing Transparency for Voters, to introduce (as a matter of course) rules concerning a 'digital imprint'. 2 This recommendation includes that social media companies should-if and when this becomes a legal requirement-make it as easy as possible for parties and campaigners to comply. The Commission also argues that legislation should be put in place to clarify to campaigners and digital platforms the type of information they give to voters, the media, other campaigners and regulators. Underpinning all of these recommendations are new data which suggest that that the public remains concerned about the transparency of online political campaigns. The Commission shows that 72 per cent agreed that it was important for them to know who produced political information online, 29 per cent agreed that they could find out who had produced the political information they see online, and 46 per cent were concerned about why and how political advertising was targeted at them. 3 The striking through line in all of this is that recommendations surrounding legislative reform hold transparency mechanisms as the guiding principle that will, at the very least, help to deliver renewed confidence in electoral processes. To return to the Commission's report, transparency acts to 'protect confidence in how elections are run and increase trust in campaigns'. Any changes, it is argued, 'will benefit voters and improve public confidence', and the report concludes with a heading concerned with 'secur[ing] democratic processes'. However, whilst taking a 'sunlight is the best disinfectant' approach is a common regulatory response, especially with regards to issues surrounding corruption and/or democratic confidence, it remains unclear exactly how transparency achieves these ends.
To reflect on this, this article takes the view that hindsight is 20/20, and that this is no bad thing. There is a lot of value in learning from the past, and in particular recent legislative changes in the field of political financing. Taking the United Kingdom as a case study and, in particular, the successes (and failures) of the Political Parties, Elections and Referendums Act 2000 (PPERA) I will outline precisely what we can learn from the guiding philosophy of the reform itself, and whether this philosophical underpinning is successful. To do this I begin by reimagining Karl-Heinz Nassmacher's 'magic quadrangle' of party funding, discussing each element in turn, and demonstrating its relevance to reform in this field, before arguing that transparency itself -if implemented imprecisely-may well undermine the stated aims of legislation in this area.
The magic quadrangle of party funding reimagined in the digital age In the early 2000s Karl-Heinz Nassmacher noted that to be effective, all political finance regimes need to be made up of a 'magic quadrangle' inclusive of professional accounting by volunteer campaign and party workers, administrative practicality, the possibility of sanctions in the event of violations, and transparency for the general public. Nassmacher argued that for regulatory success to be ensured 'none of these can be ignored' or 'stand alone in any effort to frame and implement rules'. 4 The magic quadrangle is of use more as an idealised standard (not yet fully realised offline) than something for digital to catch up to, and as such it can provide a useful benchmark for designing regulation in both spheres. In the following, however, I consider Nassmacher's four policy responses and relate them to digital as a means to better understand how they might more usefully be operationalised in terms of digital regulation.
Professional accounting by campaign, party workers (and online platforms) Nassmacher suggests that for successful monitoring of party/campaign accounts to be facilitated by external bodies/actors, these accounts need to be kept to a professional standard. In light of digital challenges, this point of the quadrangle is better understood as 'professional accounting by campaign, party workers and online platforms' and is largely a challenge on the same side of the coin as more general transparency obligations. This is unsurprising. Nassmacher himself argues that transparency is 'the most important requirement' but one that 'can never be achieved completely'. 5 It also highlights the important role of clear and consistent returns which allow journalists, armchair auditors and academics the opportunity to improve their understanding of ongoing trends and allow for more forward thinking approaches. Good investigatory work in this area, both online and off, is often hampered by a lack of common accounting practices-and it is no different in the world of digital. The Facebook ad archive, for example, contains relatively sparse data on how an ad has been targeted (encompassing gender and age, number of viewers and the broad geographic region where the ad was viewed), whereas the Snapchat archive contains considerably more fine-grained details (right down to electoral districts).
A key part of the problem for Nassmacher is that accounting and bookkeeping is often conducted by volunteer party members, and whilst amateurism is less of an issue for employees of Facebook (though platforms aside, there is plenty of evidence of citizens taking action in digital with limited expertise) the challenge remains the same. For example, the current jumbled practices surrounding how data is utilised across platforms suggests that the digital world does not align with Nassmacher's quadratic ideal. Therefore, reform in this area should not only focus around encouraging campaigners to provide more detailed invoices with regards to digital spending and targeting, but also that online platforms provide consistent data in their advertising archives with regards to the ways that their tools are used.

Administrative practicality
Secondly, Nassmacher suggests that basic practicality is essential to the functioning of a party funding regime, and that this is strengthened by a clear understanding of proper administration. In other words, a concrete notion of who the regulator is in the area, what the rules being regulated are and what powers the regulator has. Moreover, this regulator must have institutional support in terms of, amongst other things, resources and clarity surrounding the rules. Applying these ideas to the oversight of digital campaigning, there are a number of ways in which standards of administrative practicality are not met. For example, amongst many of the solutions posited are calls for greater powers for regulators-including far greater funding for the Electoral Commission (which holds much of the current responsibilities in this area)-or demands for a new regulator. However, at the forefront of reformers' deliberations should be a consideration of what is and what is not feasible and practical within the current regime.
Far greater funding for the Commission to beef up its digital capabilities is unlikely when it is (more often than not) critiqued from all sides-from Arron Banks' 'Blairite Swamp Creation' to Owen Jones' body 'which let the Tories off the hook'-and treated like a referee in a football game. 6 It is disliked by 50 per cent of fans, 50 per cent of the time, in an often openly partisan manner. Indeed, even on the Lords' Democracy and Digital Technologies Committee-of which this special issue was partly born-Lord Puttnam belied, at the very least, an underlying antipathy toward the Commission: This is an observation, I'm very impressed, personally, with both of you this morning. You've been very frank and very helpful. Can I say I'm pleasantly surprised. Because the impression one gets after the event of the way in which the Electoral Commission operates is passive and always seems somewhat less than satisfactory. So, in a sense, there's a real job to be done on communications. Because the inference is that there's been political interference and you've been told not to rock the boat, not to scare the horses and give a broad-brush approval to whatever is going on. So today has been very encouraging to me personally. 7 Therefore, we must ask ourselves, are any suggestions on increased digital oversight plausible without a clearer focus on increasing the funds of the Electoral Commission? Are these funds likely to be forthcoming given the position the Commission often finds itself in? And/or will a proposed new regulator face the same fate as the (often unfairly) much maligned Commission.
Propensity to sanction violations A close analogue with an effective and supported regulatory landscape (that is, administrative practicality) is that sanctions are in place in the event that rules are broken. That 'enforcement demands a strong authority endowed with sufficient legal powers to supervise, verify, investigate and, if necessary, institute legal proceedings'. 8 In terms of both online and offline campaigning, the Commission has consistently called for greater powers and any new regulation in this area should heed these calls.
However, in terms of digital campaigning, the argument goes one step further: to what extent do existing narratives surrounding the role of online platforms merely rely on their own good nature? Put another way, if there are no sanctions in place, and no existing framework for how platforms should report digital activity, why should it fall on them to create them in the first place? In terms of Nassmacher's quadrangle, to what extent does the current emphasis on self-regulation for online platforms neglect the need for a 'strong authority' endowed with 'sufficient legal powers'?
There are serious questions to answer about the fairness and practicality of expecting global platforms merely to adhere to the democratic norms and standards of a said jurisdiction. Better to make clear what is and is not acceptable as a basic rule of the game. This would mean compelling platforms to act in accordance with certain principles, but also involving platforms more holistically in discussions. This will help to construct basic codes of conduct which, if not adhered to, sanctioned consequences follow.

Transparency for the general public
Transparency underpins much of the debate surrounding both short-and long-term solutions to the 'digital problem' and, in terms of Nassmacher's quadrangle, is 'the most important requirement'-from the (largely agreed upon) reform of 'digital imprints' which would mirror the current rules surrounding offline campaigning, to the more fundamental sharing of (varying levels of) data between online platforms, regulators and, privacy concerns notwithstanding, the general public.
Transparency is often seen to be a good in and of itself (and it is a reflection on these assumptions to which the majority of the rest of this article turns). The logic is based on that of Supreme Court Justice Louis Brandeis that, 'sunlight is the best disinfectant; electric light the best policeman'. 9 Transparency performs two functions: that in keeping as much as possible in the open, acts of malfeasance and wrongdoing are less likely to occur; and that a greater knowledge of the workings of (various) systems will lead to greater confidence in these systems by the public.
The 'magic quadrangle' is, actually, more reminiscent of a pyramid that holds transparency as the ultimate aim in securing confidence in elections. Professional accounting, administrative practicality, and effective sanctions, are merely mechanisms that improve the process of elections and provide accountability between (and during) them. These are, in certain senses, inward looking -yet transparency looks outward. It is public facing and therefore allows us to think about what the effect on the public might be. However, too often absent from discussions surrounding the implementation of transparency is the actual effect of it on the public. What Nassmacher's quadrangle does not do-and, perhaps, is not designed to do-is explain whether the suggested regulatory responses are likely to drive the desired outcomes. Again, taking history as a guide (and barometer) we can apply this framework to understand better how effective 'transparency for the general public' is. This is where debates around PPERA can be instructive, as it was underpinned by the same quadrangle and the same guiding legislative concerns.

The Political Parties, Elections and Referendums Act (2000)
PPERA was passed during a period when standards in public life had come under increased scrutiny and after the landslide victory of the Labour Party in 1997, which had taken charge from a Conservative Party that had found itself increasingly mired in scandal and 'sleaze'. This was the mood music surrounding PPERA and with its 734 S A M P O W E R enactment 'almost every aspect of political finance came under regulatory scrutiny and control'. 10 The legislation itself was passed largely based on recommendations springing from the Committee on Standards in Public Life's (CSPL) Fifth Report: The Funding of Political Parties in the United Kingdom, from a manifesto commitment to 'oblige parties to declare the source of all donations over a minimum figure' and a declaration during the 1997 Queen's Speech that 'My Government will seek to restore confidence in the integrity of the nation's political system by upholding the highest standards of honesty and propriety in public life. They will consider how the funding of political parties should be regulated and reformed.' 11 We can see in just these examples the interplay between two perfectly reasonable principles: that the public ought to have confidence in their political parties and thus the functioning of elections, and that this is underpinned and delivered by transparency. This is reinforced if we look at the wording of the 1998 CSPL report, which suggested that whilst not 'sufficient by itself', that 'the most significant part of our philosophy depends on transparency'. 12 Moreover, that 'the advantages that can be claimed for transparency include: (1) the public and the media know who is financing each political party; (2) rumour and suspicion wither; (3) the possibility of secret influence over Ministers or policy is greatly diminished; (4) public confidence in the probity of the political process is raised'. 13 Transparency, in this instance, is the common thread. It is the principle that will deliver greater confidence in the electoral process. We can trace this from manifesto commitments, to the Queen's Speech, to the 1998 report that formed the basis of the legislation and, indeed, the introduction of the Bill at its second reading by then Home Secretary, Jack Straw: At the heart of the Bill's provisions is the need to ensure that the funding of political parties is open and transparent. Greater transparency will not only strengthen the accountability of political parties but help to buttress their financial standing. The secrecy that has hitherto been permitted to political parties in their funding, and the scandals to which such secrecy has given rise in recent years, have undoubtedly left a sour taste. In contrast, all political parties-and the reputation of our political system as a whole-will benefit from the Bill. 14 It is worth reflecting on the success of PPERA in achieving these aims, as this might well help us to understand what continued effective regulation in this area could look like. To what extent did the proposed transparency regulations actually provide public confidence in the electoral process and the party funding regime as a whole? In other words, if sunlight is the best disinfectant, did the disinfectant work?
Public opinion of political financing PPERA had two broad interconnected aims with regard to its focus on transparency: first to quell actual offences (which were nonetheless often recognised to be rare); second to quell public perceptions that big money in politics causes wrongdoing. 15 On the first aim PPERA is largely a success: transgressions are rare such that 'genuine abuse of the system and loop-hole seeking remained comparatively modest'. 16 However, on the stated second aim-that it would allow suspicion to wither and as such raise 'public confidence in the probity of the political process' the Act was far less successful. Indeed, sunlight performed less as a disinfectant and more as, well, a light which, when switched on, caused the public to either react to what they saw or have their worst fears confirmed. This reinforces an important paradox to consider, which is that in these terms, the primary problem is 'not corruption, crime, tax evasion, undue influence of fat cats or special interest groups, but the appearance of corruption, crime or undue influence'. 17 If we look at perceptions data from the introduction of PPERA to the present day we see this in stark relief. Public opinion surveys overwhelmingly show that the public believes that both institutional and individual donors have too much influence on the political process. In 2003 the Electoral Commission found that 70 per cent of respondents believed that private donors could buy political influence. In 2013, the Transparency International Global Corruption Barometer found that 90 per cent of respondents considered the government to be 'somewhat to entirely' run by a few big entities acting in their own best interests. Finally, an Electoral Reform Society study, published in 2015, found that 75 per cent of respondents felt that big donors had too much influence on the political process, and 65 per cent thought that party donors could buy honours. 18 Bringing digital back in: can transparency achieve its ultimate aim?
This suggests that a focus on transparency to dampen perceptions of the misuse of money -or data-may well be misplaced. The need for transparency runs through many of the 230 discrete recommendations proffered by the fourteen reports outlined in the introduction to this special section. 19 Within these, the Digital, Culture, Media and Sport Select Committee (DCMS) argues that 'the very fabric of our democracy is threatened' by trends in digital campaigning and, as such, what needs to change 'is the enforcement of greater transparency in the digital sphere'. 20 The Information Commissioner's Office (ICO) suggests that to retain the 'confidence of electorates and the integrity of elections themselves, all of the organisations involved in political campaigning must use personal information and these techniques in ways that are transparent, understood by people and lawful'. 21 Finally, the Electoral Commission report, released in 2018, was titled Digital Campaigning-Increasing Transparency for Voters.
However, in this area transparency is often viewed with an uncritical eye. It is, if not a panacea, then a cure-almost-all. Transparency, as a disinfectant, is seen as a 'good thing' with little reflection on the side effects. It approaches what Elinor Ostrom defines as a 'self-evident truth'. A common sense assumption which leads to 'proposals to improve the operation of political systems that can have the opposite effect'. 22 In fact, the 1998 CSPL report outlines that arguments-'no longer fashionable'-in favour of secrecy were considered. However, they largely covered concerns surrounding rights to privacy and the effect that secrecy might have on donations. What was not considered was the unintended outcome that an increase in transparency might lead to a decrease in trust in the system.That 'by providing a wealth of new information in an attempt to allay public concern about probity in public life, the effect was in part at least to heighten political sensitivity . . . transparency generated more concern rather than less'. 23 The benefit we have now, from the history of PPERA, is that we can consider what these effects might be.
There is a clear paradox: transparency might well stem actual instances of wrongdoing and undoubtedly helps regulators (and academics) track and trace activity in this area. However, it might well do all these things whilst having an inverse effect on public confidence in the electoral system itself. To be clear, this is not an argument to walk back any transparency obligations. It is, rather, to say that if we are to create effective regulation in this area, we need to have a better understanding of what we want this legislation to achieve. And, further, the ways in which transparency-at the crux of any legislative approach-might both underpin and undermine these aims.
Moreover, Katharine Dommett has also shown that calls for transparency in terms of digital campaigning lack precision in other areas, and that we have different desires for different types (financial, source, data and targeting) and forms (audience, discoverability, comprehensibility and reliability) of transparency. 24 Added to this, we should consider norms: that if the fundamental aim of proposed legislation is to buttress and/or bolster public support for electoral processes, a simplistic understanding of transparency might perform the opposite function. It is integral that when we think of transparency we think realistically in terms of a) these norms and b) what transparency can and will achieve.

Conclusion
Policy making follows path dependent processes.PPERA, after all, is largely based on 736 S A M P O W E R reforms that had worked well at the local level with the introduction of the Corrupt and Illegal Practices Prevention Act in 1883. 25 It is therefore likely and desirable that any regulation with regard to digital campaigning is formed (at least in part) around the relative successes of PPERA. However, if it is based on the same assumptions and underpinnings-and there is no reason that it should not be-we should also consider its failings.
Responding to the challenge of digital campaigning is a future-facing task. How can we as interested academics, practitioners, campaigners and regulators respond not just to challenges uncovered at past elections, but actively adapt regulations (and regulators) in a more fundamentally anticipatory sense? And yet, in much of the discussion surrounding the rise of digital campaigning, and the challenge of new sources of data in campaigns, absent is a reflection on the past and how debates of the past can help us to form our future responses. In this article I have argued that many of the challenges that are highlighted by the numerous calls for regulation in this area have been faced before (if in different circumstances) in and around the development of PPERA. It is precisely within these debates, that we can begin to sketch a picture of the norms that might underpin any legislative (or regulatory) change.
I do not do this because it is a neat analogy to draw upon, but because the challenges that are presented are comparable. Moreover, in many instances they are presented as both in the same family as, and reflecting broader challenges to, the financing of politics. There remains, therefore, much to learn from the trials and tribulations surrounding both the passage of PPERA, but also the working of the legislation in practice. So, if we are to reflect on what is practical to achieve in legislative terms and what ought to be aimed for in normative terms, we can do much worse than consider this in a recent context.
Nassmacher's 'magic quadrangle' is a good place to start. However, with the benefit of hindsight we should not think of professional accounting, administrative practicality, the sanctioning of violations and (most importantly) transparency without considering what we want legislation to achieve, and the ways in which an imprecise focus on transparency might undermine these aims. For example, if regulation simply holds transparency as the solution-without a similar focus on citizen engagement and digital literacy-it is likely to fail to 'resurrect' trust in the democratic process. 26 Transparency obligations are important for regulators to undertake their work, for journalists to uncover wrongdoing, and for academics to conduct studies. But any reflection on the future of digital, and the role of transparency in this future, should consider the ways in which it might be implemented such that it better supports its basic legislative aims.