Questioning post-political perspectives on the psychological state: Behavioural public policy in the Netherlands

Abstract

Behavioural public policy is associated with the rising influence of psychological and behavioural sciences on systems of government. Related policies are based on the assumption of human irrationality and use a series of often unconsciously oriented policy tools to pursue varied public policy goals. This paper argues that existing critical analyses of behavioural public policy can be categorized as post-political in their orientation. Post-political theory is primarily concerned with how political consensuses, particularly around expert forms of government administration, tend to close off opportunities for political contestation and challenge. Drawing on an empirical case study of emerging forms of behavioural public policy in the Netherlands, this paper challenges some of the core assumptions of post-political critiques of behavioural governance. The case of the Netherlands is also used to challenge the often absolutist assumptions about the nature of the political, expertise, and consensus that characterize post-political forms of inquiry more generally.

Introduction – On the psychological state

Behavioural public policy (BBP) has emerged as an approach to governance which spans multiple policy sectors (including public health, personal finance and consumer policy, environment and transport, labour markets, education, and urban policy inter alia), multiple scales, and has global appeal (see European Commission, 2016; Lunn, 2014; OECD, 2017; Whitehead et al., 2014; World Bank, 2015). It is an approach which uses emerging psychological insights regarding human behaviour to inform the policy-making process. BBP has been described by both proponents and critical scholars as a new era of behavioural governance, which has the potential to radically transform state–citizen relations, the effectiveness and impact of government action, and our basic understandings of human freedom and character (Halpern, 2015; Sanders, et al., 2018; Sunstein, 2014; Whitehead et al., 2017).

BPP is psychological in two respects. First, it has positioned contemporary social, political, and economic crises such as the Great Recession, climate change, and public health challenges as, at least in part, behavioural problems. Second, BPP builds on advances within the cognitive and psychological sciences concerning human irrationality (Kahneman, 2011; Kahneman et al., 1982; Thaler, 2015). The BBP project is thus predicated on humanizing homo-economicus and emphasizes the crucial role that intuition, (social) imitation, heuristics, and emotion play in human behaviour. Building on the work of behavioural economics and cognitive design, BPPs utilize the ‘gentle power’ of nudges and associated psychological tools of behavioural government in order to shape choice while, purportedly, preserving individual freedom (Oliver, 2013; Shafir, 2013; Sunstein, 2014; Thaler and Sunstein, 2008).

The emergence of psychological state apparatuses, which seek to respond to the aforementioned crises and epistemological developments, has been subject to concerted academic scrutiny and critique. Some have questioned the impact that behavioural policies will have on citizens’ autonomy and their ability to shape their own behavioural destinies (Furedi, 2011; White, 2013). Others have brought into question the limited capacity of the psychological and cognitive sciences to effectively explain human behaviour outside of the fairly limited frame of proximate contextual influences (Davies and Doyle, 2015; Strauss, 2009; Tallis, 2011). Many have interrogated the extent to which new behavioural policies actually seek to address the problems generated by neoliberal policy-making, or whether they simply facilitate the continued roll-out of neoliberalism by processes of responsibilization and desocialization (Berndt, 2015: 569; Carter, 2015; Davies, 2014; Jones et al., 2013; Leggett, 2014; Whitehead et al., 2017). Underlying the majority of these critiques has been a concern that BPP is characterized by a rise of state-sponsored behavioural manipulation, the diminution of personal freedom, and the broader closing off of political debate concerning the role of markets, states, and structural inequalities within society. This paper asserts that these critiques can each be read through the lens of post-political theory.

The concept of the post-political was popularized during the late 1990s and early 2000s through the work of scholars such as Chantal Mouffe (2005), Slavoj Źižek (1999), Jacques Rancière (2004), and Erik Swyngedouw (2007), who argued against emerging forms of political consensus, particularly those formed around expert forms of government administration, that tend to close off opportunities for political contestation and challenge (see Gill et al., 2012 for a synthesis of these debates). BPP is, ostensibly at least, post-political in three ways: (1) it promotes an expert consensus on the nature of human action and how best to govern it (Thaler and Sunstein, 2008); (2) through psychological techniques that target the unconscious, it pursues approaches to public administration which may prove difficult to contest, and which may in the long term be a threat to forms of active citizenship; and (3) by positioning intractable social and global problems as primarily issues of personal behaviour, it, potentially, forecloses alternative perspectives on the political framing of solutions, the scope of political debate, and the contested status of the ‘public good’.

This paper has two main goals. First, it links emerging critiques of BPP with the concerns of post-political scholarship. Second, and through consideration of the emergence of a BPP apparatus in the Netherlands, it questions some of the core assumptions of post-political lines of inquiry. Analysis thus provides an alternative framework to think critically about BPP, while utilizing empirical insights into emerging systems of government in the Netherlands to complicate the notion of the post-political. This paper takes as its point of departure Gill et al.’s (2012: 510) call to ‘establish where the post-political consensus is most and least firmly established’, to pay attention to the pragmatics of politics, and to acknowledge that the post-political is an unfinished and partial project. But in distinction from Gill et al., we claim that the presence of the political in, ostensibly, post-political processes is not merely about the unfinished, or partial, nature of the post-political, but part of an ongoing dialectic between the political and post-political. The case of the Netherlands ultimately demonstrates the ongoing, if often mundane and overlooked, struggles that surround the development of BBP, and how these very ordinary struggles can lead to the emergence of alternative and diverse policy forms.

The research presented in this paper draws on two research projects, which were supported by the UK’s Leverhulme Trust and Economic and Social Research Council, respectively. These projects explored the emergence of behavioural governance over a nine-year period. In specific terms, the paper reflects on a series of interviews that were carried out with policy-makers and government advisors in the Netherlands during 2014. ¹ These interviews were carried out at a time when different branches of the Dutch state, and the bodies that advise it, were considering the potential applications and implications of BPP. The Netherlands was selected as a case study because (1) it is a country where the insights of BPP have been widely incorporated into government policy, have influenced the actions of various civil society groups, and have been the object of concerted forms of academic scrutiny; (2) it offered a research context within which the early uptake of BPP policy ideas, and their political and post-political affects, could be studied by the authors. Ultimately, we argue that an empirical focus on those involved in the shaping of BPP in the Netherlands can help to challenge some of the more sweeping abstractions of theories of the post-political (Gill et al., 2012: 517).

The psychological state and the post-political condition

It is important to acknowledge that, as far as we are aware, post-political ideas have not been applied directly to the study of the systems of behavioural government we are concerned with in this paper. Nonetheless, critiques of BPP have consistently identified processes of de-politicization and de-socialization at play in the psychological explanations of human conduct offered in justifications for BPP (see Leggett, 2014). In order to situate this project, this section provides an overview of existing work on the post-political and how it connects with critical studies of BPP.

Chantal Mouffe (2005) identifies the existence of the post-political condition within the popular ontologies of modern ‘cosmopolitan democracy’, ‘good governance’, and ‘global civil society’ (2). These ontological presumptions became increasingly common during the 1990s, when partisan historical divisions appeared to be giving way to global liberal cosmopolitanism. What characterized these widely embraced political systems are the notions of rational consensus building, mutual interest, and deliberation. According to Mouffe, however, what these democratic systems ultimately tend towards was a negation of antagonism and a diminution of the ‘antagonistic dimension [that is] constitutive of “the political”’ (Mouffe, 2005: 2) (see Barnett, 2017 here for a broader discussion of the connections between politics and presumptions of antagonism). In addition to the effective absence of contestation – or dissensus – Swyngedouw (2007) argues that the post-political condition is characterized by a generalized failure to make particular political demands universal, or to transform particularized problems from local technical issues, into more radical universal demands. A further key aspect of the post-political condition is highlighted by Źižek (1999), when he describes how ‘the conflict of global ideological visions embodied in different parties which compete for power is replaced by the collaboration of enlightened technocrats.’ According to Źižek, the practice of consensus is synonymous with the emergence of expert elites and bureaucrats, who operate through complex systems of para-state governance and offer pragmatic systems of ‘what works’ government that are based upon supposedly neutral forms of scientific knowledge. Such pragmatic, solutions-focused governance techniques are a key feature of BPP.

According to Rancière (2004), a key dimension of the post-political condition is not the exclusion of troublesome political perspectives from consensus politics, but their negotiated – and inevitably de-radicalized – inclusion within consensus politics. If politics is defined, as Rancière (2004) suggests, as the disturbance of the social order by those who are external to that order, then consensual accommodation serves to deny the space for politics through strategies of inclusion. Źižek (1999) claims that post-political strategies of inclusion essentially disavow politics by only allowing forms of contestation that do not threaten the socio-political order (for a summary of Rancière and Źižek in relation to issues of post-political tactics of inclusion, see Gill et al., 2012: 511–514). Ultimately, the reflections of Rancière and Źižek suggest that the presence of apparent political dispute and contestation does not necessarily signal the breakdown of the post-political – indeed it may be supporting it. The critical line of distinction then becomes between where the post-political ends and the ‘properly political’ begins (see Temenos, 2017).

There are many ways in which it is possible to investigate BPP through post-political thought. ² In this paper, however, we are primarily interested in the compelling parallels that exist between critiques of emerging systems of behavioural government and post-political analysis. The parallels can be seen in at least three key respects. First of all, critical analyses of BPP have described the emergence of a rapid scientific consensus around the insights of behavioural economics and related psychologically informed theories of human decision-making (see, e.g. Berndt, 2015; Jones et al., 2013; Leggett, 2014). In part, the assertion of an emerging consensus in behavioural government is understandable to the extent that it provides a justification for the critical scrutiny scholars have developed and an explanation for the rapid spread of these ideas and practices around the world. Accounts of an emerging consensus regarding BPP have corresponded with analyses of the advent of new expert behavioural units devoted to delivering related policies (see John, 2013). The institutional form of these new expert units tends to be characterized by flat, non-hierarchical forms, with relatively high degrees of functional separation from government departments. While these Skunkworks styles of government support the experimental ethos of psychological state forms, they perpetuate concerns over their unaccountability, and their non- or para-state form. The purported consensual power of the behavioural sciences and the formation of expert units of behavioural insight policy-making directly mirror and, to some extent, deepen post-political concern about the uncontested, and even uncontestable, nature of emerging systems of behavioural government.

The second general parallel between critiques of BPP and post-political thought can be seen in the obfuscating political effects of related policies. Through the common deployment of the gentle power of nudge techniques, BPP often relies on the targeting of the unconscious and the manipulation of choice (White, 2013). From the resetting of default options on company pension schemes and organ donor registers, to the exploitation of cognitive biases in human decision-making (such as tendencies to discount the future relative to the present), BPP quite deliberately seek to change human behaviours without recourse to conscious persuasion. Evolving forms of behavioural government seek to exploit our own foibles (perhaps an unconscious tendency to prefer status quo to change or an instinct to follow the social herd rather than go our own way) in order to achieve broader shifts in social conduct. While the manipulation of our cognitive failings may, in the long term, bring us welfare benefits, many have questioned the implications of these strategies for political life (Leggett, 2014; Mettler, 2011; White, 2013).

The third point of connection between post-political concerns and behavioural government operates at the level of human subjectivity. In his acerbic critique of behavioural economics and related policy developments, Furedi (2011) focuses less on the impacts which nudge-style policies have on governments’ relations with their citizens, and more on their effects on the political capacities of individuals (it is interesting to note that questions of the political capacities of individuals are not a central concern of post-political analyses as they tend to focus much more on the broader societal implications of post-political forms). Furedi argues that BPPs are based upon an assumption that people lack the necessary willpower to exercise effective forms of moral autonomy (134–141). In alignment with post-political critiques, he has argued that BBP involves the mobilization of a degree of ‘scientism’ in order to refute the moral autonomy of individuals (Furedi, 2011). On the basis of this transformation, Furedi claims that behavioural policies show intolerance for private preferences while removing the valuable learning experiences that being wrong can bring to people. From a post-political perspective, intolerance for private choice could militate against opportunities for political contestation. Although, as Gill et al. (2012) point out, the relationship between the post-political and questions of tolerance is uncertain. While certain forms of overt intolerance could prompt political opposition, in other instances, tolerance could itself be seen as a strategy of consensual inclusion and de-politicization.

Perhaps the most direct application of post-political forms of critique to BBP can be found in the work of Leggett (2014). In one sense, Leggett’s work counters post-political critiques of BBP. Leggett argues that if, as BBP suggests, government is most effectively realized in the minutiae of everyday choice environments, then it also invites us to see opportunities for forms of resistance in a multitude of quotidian contexts, and not to see politics as only residing in the proper spaces of public contestation (Leggett, 2014). In another sense, however, Leggett’s work suggests an overtly post-political dissection of the limits of BBP. Thinking about BBP from an explicitly social democratic perspective, Leggett points out the ways in which BBP obscures the social determinants of behavioural capabilities and opportunities. Crucially, Leggett begins to chart what a properly politicalized version of the BBP could be. On Leggett’s terms, the politicization of BBP should, rightfully, see the emergence of an interventionist state that is more willing to protect its citizens from the exploitative behavioural dynamics of commercial capitalism. Leggett argues that armed with the insights of the behavioural sciences, the social democratic state could use traditional forms of regulation to counter the pernicious, and individualizing, outcomes of neoliberalism. This vision is, of course, very different from the actual use of behavioural insights within BBP to achieve policy goals that militate against the worst effects of neoliberalism, but ultimately do little to challenge it hegemony. As this paper demonstrates, there is evidence of the potential emergence of more radical versions of BBP (albeit, not as dramatic as those outlined by Leggett) (see Whitehead et al., 2017). This paper also reveals that these more insurgent forms of BBP can originate from fairly mundane forms of contestation.

There have been a series of critiques of post-political theory that this paper seeks to build on and extend (see Gill et al., 2012; Lees, 2013; Temenos, 2017). From a critical geographical perspective, Gill et al. (2012) have suggested that greater attention should be given to the uneven development of the post-political condition, essentially supporting the notion that from state-to-state and sector-to-sector, the extent of the post-political may vary greatly. Gill et al. (2012) have also argued for a more ‘worldly grounded view of political deliberation’ (509), which recognizes the places where overlooked forms of politics remain. In a related sense, Temenos (2017) has questioned the political absolutism of post-political theorists. According to Temenos, the suggestion that many post-political theorists make that only truly radical forms of political contestation can be considered properly political denies the power and importance of other genuinely politically moments. As Gill et al. (2012) point out, ‘[c]ontestation is more ordinary than the radical post-structural continental philosophers might admit’ (517). Even if more mundane forms of contestation are not considered to be properly political, the question remains as to whether the political can ever be as utterly disavowed as post-political accounts would suggest (Gill et al., 2012; Lees, 2013). Building on the work of Gill et al. (2012), we reject ‘adherence to a purist, ontological vision of politics’ (517) and search overlooked forms of politics within BPP. This endeavour serves not only to disrupt and extend post-political inquiry, but also to challenge many existing critiques of BPP.

Building the psychological state in the Netherlands: On consensus and expertise

This section considers recent attempts to build a more psychologically oriented state apparatus in the Netherlands. Analysis utilizes this construction process as a basis for problematizing the assumptions of consensus formation and the institutionalization of expertise that are characteristic of critical accounts of BPP policies and post-political analysis more generally.

The 2009 government as choice architect symposium and the behavioural interregnum

As with many governments around the world, the Dutch state has been influenced by the insights of the behavioural and psychological sciences for many decades (Rose, 1998). It was not, however, until 2008 that policy-makers in the Netherlands started taking a strategic interest in BPP science. The commencement of this interest was signalled in 2008 when the Scientific Council for Government Policy (Wetenschappelijke Raad voor het Regeringsbeleid (WRR)) proposed a symposium to explore the potential utility of the behavioural sciences for policy-making. The WRR symposium was, in many ways, a response to the engagement with the behavioural sciences that was evident within the UK and US governments. ³ The WRR symposium was convened in 2009 and entitled De overheid als keuzearchitect (The Government as Choice Architect). It is possible to see the WRR symposium as a technology of post-political consensus building in two ways. First, in hearing international testimony from academics of different types, it sought to establish and acknowledge the forms of scientific consensus that were emerging around BPP research. Second, it sought to lay the foundations for political consensus about the potential utility of behavioural insights for Dutch government officials (of course this consensus could, technically, have involved the acceptance or rejection of such ideas). The impacts of the symposium do, however, raise some interesting issues in relation to the practical mechanics of consensus building that are rarely addressed in either the literature on BPP or the post-political.

Although the WRR symposium event appeared to be met with enthusiasm by policy-makers, it did not directly lead to the application of new behavioural insights to public policy in the Netherlands. One of the explanations for this apparent inaction was the fact that despite hosting the De overheid als keuzearchitect/the Government as Choice Architect symposium, the WRR did not produce a formal note of advice to the Dutch state on the application of new behavioural insights within government. While there are various accounts of why the WRR did not formally advise government on adopting the insights of the behavioural sciences, one participant in the 2009 symposium suggested that there were actually three main reasons for inaction: (1) The fact that economic theories (even of the behavioural kind) were not particularly influential in terms of the dominant paradigms of the Council; (2) the Council did not generally take the emergence of a new set of academic ideas as its basis for offering governmental advice (preferring, instead, premises such as ‘what will the Netherlands look like in 2030’; and (3) they were concerned over the potential ethical implications of BBP, and, in particular, with issues of manipulation (Interview with former WRR Member, May 2014).

It is interesting to think of the role that interdisciplinary rivalry (and possibly suspicion) may have played in curtailing the emergence of an early behavioural insights coalition in the Netherlands. What is clear in this instance – and has been widely acknowledged in the studies of the sociology of science for some time – is that scientific consensus is about more than merely evidence; it also involves the alignment of strategic disciplinary needs and influence (Shapin, 1995). It is, perhaps, less surprising (given the wider critiques of BPP discussed above) that a consensus did not emerge over the use of BPP in the Netherlands due to concerns over the potentially manipulative nature of such policies. What is more unexpected is that bureaucratic procedure – based on the notion that promoting BPP was not within the constitutional remit of the WRR – should inadvertently inhibit the process of consensus formation. Without the formal support of the WRR, the Dutch governmental engagement with the behavioural sciences entered something of an interregnum. A report was produced by the WRR summarizing the 2009 symposium, but this did not constitute a formal advice to government. The WRR also gave its imprimatur to De Menselijke Beslisser: Over de Psychologie Van Keuze en Gedrag (The Human Decision-Maker: the psychology of choice and behaviour), which was published by the Amsterdam University Press. Although these publications did not require a formal response from government, it is interesting to note that they stimulated informal interest within the ranks of the Dutch Civil Service.

The stalled behavioural consensus of the WRR challenges certain assumptions concerning both BPP and the post-political. In the case of the Netherlands, it is clear that there was not a rapid uptake of BPP as early discussions of their applications met academic, bureaucratic, and ethical resistance of sorts. This observation not only tests the potential validity of post-political critiques of BPP (which tend to assume the formation of fairly instant politico-scientific consensuses around the authority of the behavioural sciences – with science essentially overriding politics), but also challenges, more broadly, post-political understanding of the relationship between consensus and antagonism. It appears that in the case of the 2009 symposium, broad initial consensus eventually gave way to antagonism, expressed both in relation to ethical concerns and interdisciplinary suspicion. Furthermore, it appears that the WRR’s lack of formal action on BPP was the product of the procedural inertia generated by its constitutional capacity. The policy interregnum that followed can thus be thought of as much as a vacuum of responsibility as it was the product of antagonism and the failure to build a coalition of action. So, if a lack of consensus does not necessarily signal overt antagonism, and consensus can easily give way to dissensus, it is important not to assume that the post-political condition is either militated against by a lack of consensus or, necessarily, guaranteed by consensual actions. The story of the early engagement between BPP and Dutch public policy-making would indicate that post-political accounts of the nature and work of consensus and antagonism are too binary: failing to recognize the complex ways in which both conditions (and the various states that exist within and between them) relate to political and post-political conditions. The example of WRR’s early interactions with BBP reveals the inescapable co-existence of the political and post-political, whereby early forms of relatively uncontested expert consensus over the value of BBP gave way to both overt and more subtle forms of political and scientific resistance. To an extent, it is even possible to discern a form of political resistance to BBP that deployed post-political means (in the form of largely unchallenged expert defiance and bureaucratic inertia). To put things simply then, the case of the Netherlands reveals that the arrival of BBPs has not always led to consensus formation. The case also demonstrates that we should not be too quick to associate consensus with the post-political, and the political with antagonism.

Institutionalizing behavioural expertise: Building a Dutch behavioural insight team

Despite placing an apparent brake on the development of BBPs, the interregnum that followed the WRR’s 2009 symposium did not last long. While precise dates are difficult to locate, it appears that by 2012, Dutch governmental interest in the behavioural sciences was growing. Year 2012 is significant because it saw the establishment of the first behavioural insights unit in the Dutch government. This small unit was formed in the Dutch Ministry of Infrastructure and Environment with initially two (full-time equivalent) staff. This period also saw the first moves being made to form a more strategic and coordinated engagement between the Netherlands’ central government and BPP. There appears to have been at least three stimuli behind these developments. First, was the reported success of the UK government’s Behavioural Insights Team (BIT) (Halpern, 2015). ⁴ Second, came the 2012 election and the emergence of a Liberal and Labour coalition government headed by Mark Rutte. The coalition was founded upon the principles of liberalism, but also forged in a time of austerity. It appears that the low cost and libertarian nature of BBP, and the fact that it appealed to groups across the political spectrum, meant that it resonated well with the goals of Rutte’s coalition. The third stimulus behind these more strategic developments was the contingent emergence of an opportune political moment for policy experimentation. By 2014, it was evident that most of the major policy agendas associated with the 2012 Coalition agreement had either been implemented or abandoned. The government was thus very open to explore new policy initiatives and ideas (Interview with representative of the Dutch Ministry of Economic Affairs, May 2014). It was in this context that in December 2014 Rutte’s Cabinet endorsed BBP, arguing that it promised to make government more effective and efficient (Behavioural Insights Network Netherlands, 2017: 6).

It is important at this point to note a key distinction between the Dutch case and that of the UK. The UK’s BIT had political support from its inception, with British Prime Minister David Cameron being instrumental in its formation and even utilizing behavioural economics as a guiding principle in his new vision of caring Conservatism and a pragmatic state. In the Netherlands, the promotion of BPP has been mainly supported by disparate parts of the civil service, located in various governmental ministries, and only gained formal political support later. Indeed, a major driving force for the coordinated use of new BBPs across public policy in the Netherlands has come from the government’s Interdepartmental Strategy Network. The membership of this Strategy Network is drawn from the civil service, and it meets regularly to discuss new policy ideas and initiatives. Through the initiative of the different ministries, it was decided to develop an intergovernmental behavioural Insights initiative (this would eventually become the Behavioural Insights Network Netherlands). The first major meeting of this interdepartmental initiative was held in May 2013. Senior civil servants, representatives from different scientific councils, and a broad cross-section of ministerial staff were in attendance.

Before the formation of the Behavioural Insights Network Netherlands, there was significant discussion and debate about how best to support interdepartmental initiatives relating to BBP in the long term. The Ministry of Economic Affairs lobbied for the formation of a central BIT that would mirror the structure of the UK’s BIT. The formation of a single Dutch behavioural insights unit did, however, raise concerns. First of all, it was not clear where such a BIT would be located in the Dutch government. The structure of Dutch governmental ministries is non-hierarchical, with no department having direct control over others. ⁵ Second, some claimed that having a single Dutch behavioural insights unit would make it too easy a target for those who were suspicious of the application of the behavioural sciences in public policy, and the narrow centralization of behavioural expertise it is believed to entail. It was in these contexts that that the Behavioural Insights Network Netherlands was eventually formed. This Network’s structure is based upon individual ministries pursuing BPP (sometimes in the context of formal BITs, as in the case of the Ministry of Infrastructure and Environment’s team), with a collaboration hub (the Network) allowing for the sharing of good practice.

The formation of the Behavioural Insights Network Netherlands raises the question of how we might think about the institutionalization of expertise and control outside of more public spheres of political action. In the Netherlands, BPP has emerged out of the machinations of civil servants operating in the deep state, and as such signifies a managerial cadre of experts which might rightfully be the target of post-political critical scholarship. This is in contrast to the situation in the UK, where BPPs were explicitly supported by several white papers and influential speeches within the Coalition government of 2010. While it is clear that the Behavioural Insights Network Netherlands, and related ministerial behavioural insights initiatives, have offered conduits for new forms of expertise (particularly in the form of the behavioural sciences and behavioural economics) to enter government in fairly subtle ways, this has not necessarily been a post-political process.

Recent analyses of the emergence of BPP in the Netherlands indicate that there has been an ongoing process of contestation and politicization of behavioural expertise (see Feitsma, 2018a, 2018b). The contestation of expertise has in part been a product of the inevitable decisions that must be made in relation to which forms of expertise, with which particular behavioural insights, should be engaged with. It has also been a result of the process of working out which forms of behavioural expertise are actually most relevant to the policy-making process itself (Feitsma, 2018b). Furthermore, the institutionalization of expertise is not about bringing scientists into government, as it is governmental officials actively filtering scientific insights to meet their own needs (Feitsma, 2018b). It is thus clear that the emergence of BPP in the Netherlands has not so much involved the hard-wiring of behavioural science expertise into government, but the training and re-purposing of existing civil servants in new policy-making skills (Feitsma and Schillemans, 2019). While this process could be associated with the post-political production of uncontested expertise, our research indicates that it has actually involved the institutionalization of the inexpert, or what Parsons (2002) has described as a process of muddling through at the policy–science interface. BPP is generally not a part of the public administration training which many civil servants in the Netherlands receive. As such, the processes of muddling through have seen policy-makers in the Netherlands develop a fairly open set of engagements with the behavioural sciences (although we acknowledge the dominant impacts of behavioural economics). The post-political effects of expertise have thus been militated against by the fact that few civil servants have been able to adopt positions of uncontestable expertise, and because BPP has, from the outset, been subject to various adaptations and ‘corruptions’ to meet the needs of policy-makers. While this process could not be described as political, it would be equally inaccurate to suggest it was post-political.

The structure of the Behavioural Insights Network Netherlands has also facilitated opportunities for different ministries to engage and adapt different forms of expertise, and for these adaptations to become points of policy contestation and debate within interdepartmental collaboration. There is some evidence that different governmental ministries in the Netherlands are already challenging accepted behavioural policy norms (such as the use of Randomized Controlled Trials) within their practise (Feitsma 2018a; 2018b). The case of BPP making in the Netherlands thus draws attention to the vagaries of expertise within politics. While it is clear that over time expertise can act in post-political ways, the actual transfer of scientific ideas into public policy is rarely about the transmission of clear, and uncontested, spheres of expertise into government.

Behavioural politics and the psychological state: On obfuscation and autonomy

Ethics and obfuscation

One of the primary, if implicitly, post-political critiques of BPP is that in targeting many of the subconscious drivers of human action, it tends to operate in a way that obscures its modes of operation, becomes difficult to politically resist, and does not support the development of capacities to act politically (Jones et al., 2013). Yet in the case of the Netherlands, the accusation of political obfuscation (and manipulation) has become something of an object of political and ethical debate and contestation itself. The common ethical criticism of BPP is that it tends to involve the dark arts of manipulation and attempts to change people’s behaviours in ways they may be unaware of and in directions they may not have sanctioned (White, 2013). ⁶ An interview we conducted with an erstwhile member of the WWR revealed an unanticipated ethical debate concerning BBP. This debate was not just about the manipulative potential of BBP, but sought to raise broader ethical questions about the formulation and delivery of more conventional forms of public policy (such as regulations, incentives, and educational initiatives). According to the representative of the WRR, concerns about the ethics of BBPs stimulated wider debates about the often-unacknowledged manipulative nature of established policy mechanisms and the ways in which existing policies tended to yield greatest benefit to the wealthy well, and not the most disadvantaged in society (Interview with former WRR Member, May 2014).

In this paper, we are not so much interested in the intricacies of this ethical debate, but in the very fact that there is one. The debate about the ethics of BPP is evident beyond the WRR and has also been present in the work of the Netherlands Centre for Ethics and Health (see ten Have, 2014) and the Dutch Council for Social Development (RMO). In many ways, these discussions indicate that as well as potentially obfuscating public policy, BPP provides the grounds for renewed political debate about the nature of public policy itself. ⁷

In his recent analysis of the morality of BBPs, Cass Sunstein suggests that ethical issues pertain to four core values: welfare, human dignity, autonomy, and self-government. When it comes to public policy, and the varied acts of government, Sunstein (2016) recognizes that the ethical values we prioritize are inevitably contested. The ethical debates concerning BPP waged in the Netherlands appear, in part, to centre on the tensions between welfare, personal autonomy, and human dignity. On the one hand, conventional policy approaches that are based upon education and incentives appear to support values of personal autonomy (and the right to ignore a policy and go your own way) and human dignity (or the idea that you are capable of making your own decisions and should be respected accordingly). On the other hand, BPP advocates suggest that their policies may make it more difficult for people not to follow policy prompts, but that this results in a much better set of welfare results (particularly when it turns out that people actually aren’t that good at making decisions that support their own long-term interests). What interests us about this Dutch debate is that ethical considerations have been stimulated by BPP and not closed-off by it. Consequently, while specific policies may obfuscate policy decisions in certain everyday situations (by by-passing conscious action and decision-making), in other, often more strategic contexts, it appears that through a concern with actually existing welfare delivery they can serve to open-up and re-politicize the policy formulation process itself.

Critical analyses of post-political theory have suggested that attention should be drawn to the uneven development of the post-political condition (see Gill et al., 2012; Temenos, 2017). The emergence of ethical debates around BPP in the Netherlands suggests that sensitivity to the uneven development of the post-political condition should not, however, just be about how the post-political is present in certain places while absent in others, but should also recognize how certain practices of government have the potential to produce both political and post-political affects at the same time. In this context, we argue that it is helpful to consider the dialectical relations that often exist between the political and post-political, as the post-politicization of one issue, in one context, results in its re-politicization elsewhere.

Questions of autonomy

Connected to concerns over the obfuscation of the policy process, BPP has also been subject to critiques which suggest that it can lead to the diminution of personal autonomy. A key moment within discussions about autonomy, self-government, and behavioural policy in the Netherlands came in 2014, with the publication of the RMO’s Resisting Temptation report. The RMO advises the Dutch government on a range of social issues. ⁸

At its heart, the Resisting Temptation report is keen to establish the different political visions that are encoded within BPP. As a representative of the RMO stated to us,

[…] what I find really funny in the discussion right now is that I think most people in the Netherlands who are positive about nudging are positive because they see it as an extra way to do things […] [but] it could also be interesting for somebody who thinks the government should be more liberal […] it is claimed by people who want to do more, but you could use it to do less. And that’s what we tried to do with reframing [in the Resisting Temptation Report]. It shouldn’t be pushing people in the right direction [while] harming their autonomy, but it should be giving people more space, giving people more freedom without leaving them behind […] It’s a different way of looking at it […]. (Interview with RMO representative, May 2014)

The RMO recognizes that the non-regulatory nature of BPP can result in both more (welfarist) or less (liberalist) interventionist state systems. The key aim of the RMO report appears to be to ensure that BPP preserves personal autonomy, while realizing that this autonomy is itself based upon certain capacities to act that depend on welfarist forms of intervention (fusing together two of Sunstein’s separate ethical values). While the tension here between liberty and equity is not new, it is interesting how attempts to respond to emerging BPPs lead to new political discussions about this historical conundrum.

The RMO ultimately suggests that BBP could support the development of more meaningful forms of autonomy, similar to what Sunstein refers to as self-government. Self-government is distinct from autonomy to the extent that autonomy implies being unaffected by behavioural government. The notion of self-government suggests the development of capacities to act which focus more on self-authored decision-making techniques, as opposed to externally construed nudges. The RMO representatives we spoke with positioned this idea of autonomy and self-government in relation to concepts of positive and negative freedom,

So, we formulate an autonomy paradox. And that relates to this distinction of positive and negative freedom where government increasingly is retreating from the public domain and expects people to take their own responsibility. That could easily be framed as an expansion of people’s negative freedom because there’s less and less distortion by government. But the more […] government expects people to take up their own responsibility, they expect people to be autonomous to make conscious choices. And then comes in positive freedom because […] if people don’t have full agency, the awareness, the self-esteem, they can’t make those choices. (Interview with RMO representatives, May 2014)

In the work of Amartya Sen (1993), negative freedom is interpreted as a form of autonomy that is predicated on non-interference. Positive freedom, on the other hand, connects autonomy to the ability of someone not only to be free of coercion, but also to have the capabilities to pursue their chosen paths in life. BPP is not easily categorized when it comes to questions of positive and negative freedom. Nudges can be seen as an enhancer of negative freedom (as a form of softer regulation than the law for instance), or something which erodes negative forms of freedom (a nudge is still paternalist, no matter how soft). BBP can, however, be seen to support positive freedom (making it easier for people to make the decisions that are in their own long-term interest) and undermine such freedoms (as in many forms it offers little in the way of behavioural capacity building).

According to the authors of the RMO report, there is a danger that BPPs could support the emergence of fairly facile, neutral citizenship that does little to really enhance self-governing capacities. As an RMO representative observed,

So, they [The WWR] said, well, the rest of society is pushing citizens in this direction like eating unhealthy food. So, we should push back exactly to the same kind of degree of geometry. So that in the end, he or she will end up in the middle […] And that is what we feel very uncomfortable with because who decides that it’s just a two-dimensional space? […]. (Interview with RMO representative, May 2014)

In this context, BPP could be seen to do enough to undermine liberal definitions of autonomy, but not enough to actually enhance positive freedom. The RMO report thus calls for the deployment of alternative forms of BPP to promote much more active citizenship. Although the Resisting Temptation report does not spell out how a more behaviourally empowered citizenry could be formed, it is clearly suggestive of the development of systems of behavioural education and learning that would support a meta-awareness of individual behaviour and its driving forces. The RMO’s thinking is different to the more overtly political vision of an interventionist social democratic state that protects citizens from the exploitative behavioural dynamics of neoliberalism proffered by Leggett (2014) (see above). It is nevertheless indicative of the politicization of BBP thinking that has emerged in the Netherlands.

In this section, we have seen how various challenges to the post-political tendencies of BPP have been developed in a range of ordinary contexts, which have routinely been denied proper political status within post-political theory. The Resisting Temptation report is significant because not only does it come from a more ordinary space of politics, but also because it offers a radical challenge to the conformist welfarist ethics of BPP. We acknowledge that the formal, governmental advisory role of the RMO means that the inclusion of its ideas within emerging forms of consensus around BPP could lead to their inevitable de-radicalization (as post-political theorists would predict). Notwithstanding this, we assert the very existence of Resisting Temptation report, and the debates it has generated, should at least give us pause for thought when applying post-political critiques to BPP, and, moreover, lead us to question the political assumptions of post-political thinking.

Is there a ‘Dutch nudge’, and is it more empowering and collective?

The previous section demonstrated that the emergence of BPPs in the Netherlands, and the varied ethical and constitutional issues they raise, have directly led to political debates and moments of contestation about the nature of human autonomy, welfare, self-government, and the practices of obfuscation. The Netherlands has also been witness to a broader politicization of BPP, which connects nudging, consensus politics, and questions of the public good. These developments have moved discussion about the behavioural sciences and public policy from the level of the individual (where much of the specialist academic debate has been focused) into a broader social sphere. These developments have directly opened-up discussions about the varied ways in which BPP may be linked to consensus formation.

One policy advisor we spoke to, for example, raised the important question concerning when and why collective forms of political consensus may be needed to support a behavioural policy intervention in the Netherlands,

[C]ould nudging or behavioural types of intervention [be applied to] goals that everybody subscribes [to], or where there is a consensus about it […] Or where consensus is lacking […] So, what they say is if you use nudge instruments, you should first have a discussion about the paternalism behind the libertarian side, if you understand what I mean. So, there should be first a discussion about what are the motives, the intentions […] Because if it’s implicit, then you depoliticize something that is really political. (Interview with academic government policy advisor, Netherlands, May 2014)

This reflection raises two important points. The first is procedural, and it concerns the importance of establishing some form of political consensus on the use of specific behavioural policies. On these terms, consensus, by definition, involves open public discussion about the motives and goals of behavioural policies so that the often-submerged policies associated with nudge can be scrutinized before being applied. Consensus on these terms could of course apply both to the goal of policy (perhaps increasing organ donation or healthy eating) and the policy mechanism (resetting defaults/re-designing choice environments). The second issue raised here concerns situations where there may be an absence of consensus. When there is an absence of consensus concerning either the goal of a behavioural policy and/or its means of delivery, it would appear that its application would be much more controversial in the Netherlands. As we have already established, post-politics is itself regarded as a pernicious product of expert consensuses, which removes decision-making from the political spheres of contestation. But if achieving politically contested consensus is prioritized precisely because of the obfuscating long-term potential of nudge-style policies, then some of the post-political concerns surrounding BPP are lessened. It is important to point out here, of course, that consensuses are not inherently post-political: in their formation phases, consensuses can be the product of overt political antagonism. It is only once established that an enduring and largely unquestioned consensus becomes post-political.

The issue of consensual decision-making raises broader questions concerning the connections between BPP and the governing of collective action challenges (such as climate change). Questions of collection action shift BPP from a focus on individual conduct to questions of the public good. As one RMO representative observed,

A lot of people who are in favour of nudge say, ‘Well, nudge isn’t the problem because it only strengthens your autonomy […]' But a question I think we dealt with a little bit [is] ‘Okay, but what does the [RMO] Council think about using nudges for public problems?’ And public problems are not always just for your own autonomy but are to steer people against their own values or issues just because we have this collective problem, global warming, and we don’t want people to put up the heat because they like it themselves. We want them to put it down because otherwise, we have this collective [problem]. (Interview with RMO representative, Netherlands, May 2014)

It is important to acknowledge that there is no precise line of demarcation, or necessary line of contradiction, between individual behaviours and public interest. It is, however, clear that individual and collective interests are not always well aligned. The application of nudge-type policies to public, as well as individual behavioural, problems has actually been discussed in other national contexts. In the UK, for example, the House of Lords Inquiry into BBP suggested that individual opposition to behavioural policy is not a good reason to avoid applying BPP in the broader public interest (House of Lords, 2011). It appears likely that in the Netherlands, BPP will be used to address a range of public interest issues. It will therefore be intriguing to see what normative justifications will be offered by the Netherlands government for using behavioural policy to address public problems that appear to run counter to the interests/preferences of the individual.

One distinctive idea that is emerging in the Netherlands, which connects discussions about BBP, collective politics, and public interest, is the notion of collaborative nudging. In some ways, the idea of collaborative nudging builds on the insights of the self-nudging movement. ⁹ Collaborative nudging, however, takes this approach a stage further, and suggests that the behavioural sciences could be mobilized at a collective social level. As one policy advisor observed,

So, we say that if you have these controversial topics, you should involve citizens more and civic organizations more and collaborate with them and use little experiments before you roll out the whole programme for the whole country and because, first, you have to know what the effects are and if the effects are the effects you want […] Experiments with people, not just about people or around them […] but with them. (Interview with RMO representative, Netherlands, May 2014)

The idea of collaborative nudging suggests that people and civic organization could be engaged more actively in the experimental design of BBPs (see Jones et al., 2013). This form of collective engagement could facilitate consensus decision-making not only at the level of whether a policy should or not be delivered, but in the very form and function of the policy itself. Collaborative nudging could also provide a context for communities (at a range of scales) to openly discuss the tensions that may exist between individual and public interests. Furthermore, and perhaps most significantly, the idea of collaborative nudging suggests a form of behavioural policy that recognizes that behaviours are constituted not only at an individual but also at a social level. The idea of collaborative nudges is significant because it again shows that BPP does not have to result in the diminution of human agency or increases in political obfuscation, but instead can be connected to strategies of citizen empowerment.

We do acknowledge that it is one thing for there to be strategic discussions about more empowering, consensually oriented, and collective behavioural policy-making, but that it is quite another for this to begin to shape policy. And, in the absence of any substantive research on the actual long-term effects of the wide-spread use of BPP, there are currently limited empirical resources on which to base strategies for designing empowering and collective BPP, which can substantively accommodate political contestation and avoid narrowing the terms of public debate. There is, however, already some evidence that more collaborative forms of BPP are emerging at a municipal level in the Netherlands (see Feitsma, 2018a for reflections on the Urban Nudging movement in Utrecht). The very presence of these alternative forms of nudging suggests that BBP in the Netherlands is generating forms of not only political resistance, but also political adaptions. Again, while these adaptations of BBP may not reflect the forms of strong, state-orchestrated political resistance to behavioural power envisaged by Leggett (2014), they do suggest emerging forms of BBP in the Netherlands appear to be politicizing behavioural power in both their forms and functions.

Conclusion: Behavioural insights and the (post)post-political

This paper has pursued two main goals. First, it has sought to connect together emerging critiques of BBP through the concerns of post-political scholarship. Second, through consideration of the emergence of a nascent psychological state apparatus in the Netherlands, it has challenged the idea that the post-political is ever an accomplished and necessarily clear-cut state of affairs. At the beginning of the paper, it was claimed that BPP could be thought of as post-political in three main ways: (1) to the extent that they promote an expert, non-ideological consensus on the nature of human action and how best to govern it; (2) because of the psychological techniques they deploy to target the unconscious, and which actively prevent acts of government being contested; and (3) because of the way in which a focus on individual behaviour can potentially suffocate political debate over the very framing of problems which are actually social, structural, and collective in nature. Ultimately, this paper has revealed that the extent to which BPPs can be thought of and critiqued as post-political phenomena depends fundamentally on the context within which they are analysed. Consequently, while more generalized critiques of the post-political problems of BPP find some traction, more fined-grained empirical analysis of BPP landscapes appears to cast doubt on post-political assumptions about the nature of politics, consensus, antagonism, and expertise. This is evident in our consideration of the institutionalization of expertise, which outlined how the Dutch state has been witness to the development of a managerial cadre of behavioural insights expertise. Such forms of expertise are frequently the subject of post-political critique. Closer inspection of the nature of the Behavioural Insights Network Netherlands, and related administrative structures, however, complicates post-political critiques. The non-hierarchical structure, and diffuse form, of the Behavioural Insights Network Netherlands, appears to lend itself to a diverse set of behavioural governmental and scientific interactions and ultimately to the contestation of expertise. While this paper openly acknowledges that these forms of struggle and contestation within the deep state may not reflect the radical antagonisms promoted by critics of the post-political, they clearly complicate the idea that the institutionalization of behavioural expertise necessarily closes-off contestation.

In the case of the establishment of BPP in the Netherlands, we have also seen that the formation of a scientific and political consensus is far more contingent than routinely portrayed within the writings of advocates of BPP and in post-political critique. In the Netherlands, we found that consensus was challenged by politico-ethical divides and even unintentionally stalled because of bureaucratic procedure. In noticing that a lack of consensus does not necessarily signal antagonism, and that consensus can easily give way to dissensus, the analysis presented in this paper supports the position that the post-political condition is neither militated against by a lack of consensus nor, necessarily, undermined by consensual actions.

This paper also considered the accusations of political obfuscation and diminished political autonomy that are routinely levelled at BPP. Our analysis revealed that while behavioural policies may obfuscate policy decisions in certain everyday situations (particularly in the form of unconscious nudges), in other, often more strategic contexts, it appears that they can serve to open-up and re-politicize the policy formulation process itself (particularly in relation to broader debates about the balancing of the ethical values of welfare, autonomy, human dignity, and self-government). In this sense, there appears to be something in the (seemingly) controversial nature of BPP that engenders new public and political discussion on the appropriate relationship between the state and its citizens. In terms of diminished political autonomy and the potential for the alternative development of collective nudges, the paper has demonstrated that while certain forms of BPP clearly have the potential to damage (thin) forms of autonomy, it is possible to conceive of behavioural policies that support the advancement of positive forms of freedom and collective action.

Ultimately, this paper has exposed a complex set of dialectics that connect together the political and post-political conditions. These are dialectics that are often only evident at an empirical level and when what counts as being properly political takes a less radical and more banal form. This does not, of course, mean that BPP should not be subject to post-political critique – it may, often, be a justified critical perspective. Nor does it mean that the Netherlands is in any way typical of how different political constituencies have engaged with BPP. What the case of the development of BPP in the Netherlands does indicate is that related policies are not necessarily post-political in form and that it may be helpful to rethink how we conceive of the very notion of the post-political condition in practical terms.