Abstract

This article explores unenacted policy – where legislation is passed but the policy is not implemented. It argues that this should be seen as a form of policy failure and that this has been underexplored to date. It tackles the puzzle of explaining why governments would go to all the trouble of passing legislation without then moving to enact the law. The article draws on the recent example of long-term care funding reform in England, which has twice been passed into legislation but not enacted. It develops theoretical insights around non-enactment and argues that the literature on policy failure should incorporate non-enactment, as this work disrupts existing claims that governments will tend to stick with legislated commitments for as long as possible. In a context in which the reputational risks of implementation are constantly recalibrated against the risks of abandonment, we see governments pursuing a deliberate ‘policy mortality’. Killing off policy prior to implementation can be seen as a tactical approach to major welfare issues, shifting blame to local actors and creating a narrative of fatalism.

Keywords: policy failure; policy inaction; long-term care; policy mix; disability; ageing; UK; unenacted policy

Introduction

While legislation is usually passed with much fanfare, it is a truism that policy implementation is often poorly handled and incomplete. The much-quoted subtitle of Pressman and Wildavsky’s (1973) book (‘How Great Expectations in Washington Are Dashed in Oakland’) focuses on the gap between policy as enacted nationally and policy as implemented locally. What is less often studied is policies that are ‘dashed’ within the same policy circles that enacted them (Heaney, 2011: 191). The work of scholars such as Patashnik (2008) and Luetjens (2023) draws our attention to the elements within national policy communities that lead to poor or patchy implementation at source rather than out in the field.

The focus of this article is on a distinct and neglected category of policy: legislation which is never implemented at all. Norton (2021) calls this ‘law but not law’. It is policy that is given the formal status of an Act but on which the starting whistle for implementation is never blown. To use the language of some parliamentary systems, it is policy which never reaches ‘commencement’. We focus on it here as a type of policy failure that has not been sufficiently considered in the existing policy literature. Given the time and effort that is required to get policy through the legislative branch of government, it is surprising to find that not all laws are enacted. In the Westminster-type system where the executive usually controls the legislative agenda and is not required to implement legislative mandates that it did not instigate, non-enactment represents a particular puzzle.

Unenacted policy is treated differently in different political systems. In the UK, for example, there is no formal reporting mechanism but it is often the focus of parliamentary questions asked by MPs of ministers (Dods, 2022). In Australia, there is a requirement for the government to report annually to the Senate, listing which Acts or parts of Acts have not yet been ‘proclaimed’ and explaining why (Parliament of Australia, nd). Laws may either be ignored straightaway or initially delayed and then permanently shelved. A review by the UK House of Lords Library found 480 such examples in the UK Parliament between 1960 and 2020 (Dods, 2022). This can include sections of Acts rather than entire pieces of legislation; they may be minor or may have significant constitutional or distributional consequences. For example, the section of the Political Parties and Elections Act 2009 that banned parties from accepting donations over £7,500 from non-UK residents has never been enacted (Dods, 2022).

This article makes the case for unenacted policy as a neglected concept within the policy literature, drawing on the example of long-term care reform in England. Ageing populations and increased longevity for disabled people has created ‘new social risks’ around the adequacy of outdated welfare settlements (Bonoli, 2007). Several countries have addressed this through state investment in more extensive long-term care services, as well as conversations about the appropriate balance between public and private funding (Curry et al, 2019; Carey et al, 2021; Connon, 2022).

In the UK, this has been a live political debate since the Royal Commission on Long-Term Care (Sutherland, 1999). Under the post–Second World War welfare settlement, people with assets above a means-test threshold have to pay privately for long-term care (which falls outside the remit of the National Health Service (NHS)). In the last decade, two legislative interventions to reform long-term care funding have been abandoned prior to enactment. A package of reforms – including a cap on private liability for care costs and an increase in the means-test threshold – was passed into law in the Care Act 2014, but enactment was first delayed and then abandoned. A very similar reform package was then passed in the Health and Care Act 2022, with implementation scheduled for 2023. However, implementation was then delayed until 2025. Since this date fell beyond the life of the Parliament, the reforms were effectively abandoned by the administration that had passed them (Bottery, 2023; Hoddinott, 2023).

The concept of ‘non-enactment’ is the puzzle which this article explores. There is an imperative here to explain this pattern of care-funding reform, but also to consider how explaining failures in policy enactment contribute to the broader literature on policy failure, policy inaction, policy learning and policy feedback. The article reviews these policy literatures to highlight the inattention to non-enactment before setting out the case of long-term care funding reform, and how the repeat process of non-enactment unfolded. Three explanations for the enactment failure are considered: flaws in the policy itself (the idea); institutional factors, particularly relating to institutional capacity; and the pressures of organised interests. The article explores the political calculus that makes the reputational costs of abandonment lower than proceeding with a high-profile reform. The article then considers the implications for the policy literature and next steps for research into unenacted policy.

The focus is on England. It is important to acknowledge its status as a single country case – no longer even generalisable as the UK given distinctive social policy arrangements in Scotland, Wales and Northern Ireland. The English case is used here to develop theoretical insights around non-enactment for policy scholars and strategic insights for policy makers attempting welfare policy reforms. As a Westminster parliamentary system it has a particular configuration of legislative–executive relations which are relevant to the way that non-enactment plays out. In the discussion section we return to questions of if and how non-enactment might be studied in other political systems.

Locating non-enactment within the policy literature

Non-enactment – when a government abandons legislative commitments prior to commencement – is not the same as the implementation shortfalls that scholars have looked at for decades (see, for example, Pressman and Wildavsky, 1973; Matland, 1995; Hill and Hupe, 2022). Non-enactment is distinct from under-reaction (Maor, 2014) or drift (Béland et al, 2016; Rocco, 2017; Shpaizman, 2017) where governments avoid making new laws. Non-enactment may involve an initial period of delay prior to abandonment but it is not the same as government or an implementing agency merely issuing a revised timetable (Potter, 2017; Feng and Fox, 2022). Nor is it the same as non-compliance when actors do not follow the law (Weaver, 2014) or non-enforcement in which local actors do not penalise breaches of the law (Keiser and Meier, 1996).

The concept of symbolic implementation used by Matland (1995) and others (such as Holt and Barkemeyer, 2012; Howlett, 2014) has similarities to non-enactment. Symbolic policy communicates governments’ endorsement of specific values and is ‘expressive’ rather than ‘instrumental’ (Slaven and Boswell, 2019: 1479). It is policy that was never meant to achieve its stated aims, nor have a tangible effect on the target population (Newig, 2007: 278; Happaerts, 2012). Examples used in the literature include carbon targets and immigration quotas (Bache et al, 2015; Slaven and Boswell, 2019), all cases where the law has come into force but has been poorly implemented. Symbolic policy serves to demonstrate that action is being taken without there necessarily being any resulting change in behaviour. We can argue that the failure to enact a policy severely limits this symbolic signalling function. However, the concepts are related, a point we return to later in the article.

Non-enactment could also be considered as a type of policy failure. The policy failure literature offers an account of the extent and types of failure (McConnell, 2010; 2015; Howlett et al, 2015), the reasons for failure (McConnell, 2010; 2015) and the learning processes that accompany failure (Dunlop, 2017; Kay, 2017; Nair and Howlett, 2017; Stone, 2017). However, most of this literature assumes that a policy at least starts its implementation phase, and therefore has a chance to succeed. Similarly, the policy termination literature gives the policy a chance to go out into the world – termination comes later in the policy cycle, not prior to implementation (deLeon, 1978; Geva-May, 2004).

Alternatively, non-enactment can be considered a type of policy inaction (Howitt and Wintrobe, 1995; McConnell and ’t Hart, 2019). McConnell and ’t Hart (2019) set out a new research agenda and set of tools for studying policy inaction. However, while these authors recognise that inaction can happen at different stages of the policy cycle, they do not sufficiently acknowledge the methodological and political differences between inaction at different points. Whereas they caution that policy inaction suffers the problem of the ‘knowability of “non-events”’ (McConnell and ’t Hart, 2019: 657), in a case of non-enactment we do have an event (the passage of legislation) that precedes the inaction. Non-enactment is more visible, binary and amenable to study than failure to legislate at all. It is also more visible than implementation failures given that these are likely to be dispersed across a myriad of agencies and frontline encounters.

Non-enactment might be better explained through the literature on policy endurance. In two books on the topic of reform endurance – Reforms at Risk (Patashnik, 2008) and Reforms that Stick (Luetjens, 2023) – attention focuses on the conditions that lead policies to be more or less enduring. In this literature we find two useful insights for our non-enactment category. The first is the recognition that policies can fail to ‘stick’ or ‘endure’ at different stages in their life cycle: while the cases provided in the books are all policies that have at least been partially implemented, there is a recognition that policies do not have to get as far as the front line to fail. Second, this literature draws attention to the role of elected officials in policy failure, rather than locating failed implementation in the actions of bureaucrats or street-level actors.

The policy feedback literature also offers useful insights to considerations of non-enactment. In this literature, policy is not treated as an object or an output to be taken to the front line for enactment, it is a force that reshapes identities, options and dynamics (Moynihan and Soss, 2014: 321). This literature deals with what happens when a policy starts to have ‘resource effects’ and ‘interpretive effects’ (Pierson, 1993). Policy feedback can be positive – helping to embed a new policy, through creating new winners and embedding institutional commitments that are hard to unravel (Patashnik, 2008). It can also have negative feedback, creating new losers and failing to build robust institutional legacies or win over public opinion (Weaver, 2010). Negative feedback is most likely when costs are concentrated and upfront, benefits are diffuse or long term and policy is complex, making it hard for beneficiaries to appreciate the benefits (Patashnik and Zelizer, 2013; Oberlander and Weaver, 2015).

Located within historical institutionalism and assumptions of path dependence (North, 1990; Pierson, 2004), this feedback literature is focused on what happens to policies following enactment (Béland, 2010; Campbell, 2012). It is possible to adapt it for our non-enactment inquiry, in recognition that resource effects and interpretive effects will start to become evident through the debates and lobbying that accompany the legislative stage. Anticipation of these considerations may shape whether or not policy makers are minded to pause or abandon the enactment. For example, explaining the failure to enact the ban on political donations from non-UK residents, mentioned earlier, a government minister stated: ‘The Coalition Government took the decision not to implement the 2009 legislation, as it was not deemed to be workable’ (Foster, 2019). Unpacking this concept of policy being framed as ‘unworkable’ lies at the heart of the non-enactment puzzle.

Ideas, institutions and interests in the context of non-enactment

In a special issue on persistent policy failures, Howlett et al (2015) bring together papers on cases such as climate change policy and public health interventions. To explain persistent policy failures, May’s (2015) contribution to the special issue deploys the concept of a policy regime in which ideas, institutions and interests all contribute to failure. While the 3Is framework is widely used (as set out by Howlett et al, 2015), May’s lens is more explicitly political than others. May cautions against seeing analysis of implementation failure as ‘an exercise in diagnosing misaligned incentives and shortfalls in organizational capacity … Left out is how politics shapes the likelihood of implementation success and failure’. Focusing on the ideas, institutions and interests underpinning a policy regime, he goes on: ‘Much rests after policy enactment on how policy makers and others advance the ideas that are central to a given policy approach, how institutional arrangements reinforce policy cohesion, and whether the approach engenders support or opposition among concerned interests’ (May, 2015: 280).

May’s focus is on what happens after a policy is enacted. However, the same considerations – ideas, institutions and interests – will shape decisions about if and when to enact a policy. Potentially any or all of these could underpin a government claim that a policy is ‘unworkable’. A focus on ideas draws attention to flaws in the policy itself – either in the underpinning ideas or in the design and mix of the instruments used. A focus on institutions draws attention to how institutions are configured and interact to undertake or undermine enactment. A focus on interests looks at the dynamics within political parties and other organised interests, which may encourage or block enactment. These three explanations are looked at in turn to understand non-enactment through the 3Is lens.

Ideas and instruments

The ideational lens is used by a number of scholars when exploring why policies succeed or fail, coming from an assumption that ideas matter in shaping human behaviour and policy decisions (Béland, 2019). Ideas are ‘the currency for debate about political commitments’ – evoking organising principles such as ‘affordable care’ or ‘zero tolerance’ (May, 2015: 280). Ideas which have ‘high, positive valence’ – those that are important and attractive to politicians and the public – are more likely to shape policy than those with low or negative valence (Cox and Béland, 2013). Nonetheless, the valence of ideas is not fixed (Béland, 2019). Thus in the context of non-enactment, there might be changes in the valence of the issue following legislation, such as reduced public or partisan support for the reform, or the eclipsing of the issue by more pressing concerns.

Considering policy as a set of instruments or tools takes us into the literature on policy design and policy mix (for example, Cashore and Howlett, 2007; Howlett and Rayner, 2007; Sewerin et al, 2022). Failure can be due to poor, incomplete or misaligned design in the overarching goals of a policy or in the detailed instruments used to achieve it. The policy mix literature highlights the importance of interactions between different components of a policy and the fit with existing policy (Howlett and Rayner, 2007; Flanagan et al, 2011; Béland et al, 2020; Sewerin, 2020). When laws are passed they will often include multiple instruments within them – for example the Affordable Care Act included a wide range of policy instruments, such as new regulatory frameworks and financial incentives (Béland et al, 2020: 276). Considering the policy mix through an enactment lens draws our attention to how government can unbundle these instruments, with some sections of legislation being enacted while others are not.

This policy mix literature also considers the importance of the sequencing of policy instruments (Gunningham and Sinclair, 1999) as a way to mitigate against policy failure. Writing about the Affordable Care Act in the US, Béland et al (2020: 283) observe that: ‘generating [policy] resilience requires front-loading benefits and back loading as many costs as possible in order to ensure the creation of strong policy constituencies during the early years of policy development’. To enhance the endurance of policy Patashnik (2008) highlights the importance of cultivating supporters for reforms, changing the rules of the game and creating ‘sunk costs’ so that actors are invested in the continuation of reforms. This of course relates to the role of interests, discussed later.

Using this ‘ideas’ lens, then, policy non-enactment could be explained by weaknesses in the underpinning idea, tensions within the policy mix, or between the new policy and existing policies. It could also relate to opposition to the planned sequencing of instruments if the legislation raises early costs for key constituencies and is not likely to deliver sufficient early benefits to secure support for enactment.

Institutions

A second potential block to policy enactment comes from institutions. As May puts it, ‘Institutional arrangements structure authority, attention, information flows, and relationships in addressing policy problems’ (2015: 281). At the point of enactment, key institutions are the political executive and the bureaucracy who in the first instance will begin to enact the policy, working with partners (such as executive agencies, subnational governments, courts, pressure groups). Factors to consider here are the structural constraints on policy making, such as whether the policy has support across relevant government departments. Enactment may require support from the finance ministry as well as the department leading on implementation. There may be a requirement for disclosure or data-sharing across agencies. One reason for the failure to enact the ban on overseas donations, for example, was that the relevant regulatory body could not have access to an individual’s tax status (Dods, 2022).

A key factor here will be how far executive institutions have the capacity to enact laws once they have cleared the legislative process. This can be a time issue: inaction is more likely if there is a lack of agenda space, meaning that executive actors struggle to proceed with enactment (Cairney et al, 2016; Shpaizman, 2017). One of the key institutional capacities is that of learning (Heikkila and Gerlak, 2013; Dunlop, 2017; Kay, 2017; Nair and Howlett, 2017; Stone, 2017). Dunlop (2010) draws on Argyris and Schön (1974; 1978) to distinguish between single- and double-loop learning in relation to policy failure, summarised as ‘“doing things better” versus “doing things differently”’ (Hayes and Allinson, 1998 cited in Dunlop, 2010: 346). Writing about learning pathologies, Dunlop (2021: 625) notes that ‘learning resistant policy pathologies can become locked-in over time’ and that ‘states’ own self-identity can condition their ability to learn from failure’. Non-enactment may represent a failure by executive actors to learn the right lessons from earlier policy failures, or to make use of the learning opportunities created by policy delays.

Political interests

Exploring policy failure through the lens of interests focuses on ‘the preferences and power embedded in policy actors’ (Shearer et al, 2016: 1201). It foregrounds the configuration of political leaders, parties and organised interests, which may together shape enactment (May, 2015). Most of the literature assumes that political leaders have an incentive to continue with policy, even when flaws become clear (see, for example, Dunlop, 2010; Krause et al, 2016). From a policy learning perspective, Dunlop notes: ‘the knowledge utilisation literature is replete with examples where decision-makers do not change policy course in the face of evidence that their favoured policy solutions may not match the lessons being taught by experts’ (2017: 26). As Maor (2014: 432–3) puts it, there is a bias towards action: ‘it is much harder to blame decision makers for doing nothing than for doing something’.

In deciding if and how to act it is important to note that avoiding blame is likely to be more important to politicians than claiming credit (Hood, 2002; 2011; Howlett, 2014). As Howlett (2014: 397) puts it: ‘due to the often precarious nature of their positions at the apex of power, most politicians and decision-makers, especially but not limited to democratically elected ones, are highly risk averse and seek to avoid failures for which they can be plausibly be held responsible’. Pausing and abandoning legislation prior to enactment may represent efforts to avoid being associated with a policy which has come to look too risky.

It is also important to acknowledge the shifting nature of political coalitions and interests. As Patashnik and Zelizer (2013: 1074) put it: ‘Reforms are subject to a commitment problem: today’s officeholders may change their minds about the desirability of maintaining a policy, and even if they do not, they will eventually be replaced by officials with different preferences’. Formal changes in the make-up of governing coalitions, or informal shifts in the coalition of interests that supported the policy can strengthen or weaken policy enactment (Patashnik and Zelizer, 2013). Explaining non-enactment through interests is about uncovering the shifting coalitions and the shifting calculus of blame and credit that occurs at the point of enactment, through which it becomes more reputationally risky to proceed than to stop.

Thus, although the literatures on symbolic policy, policy inaction, policy failure, policy endurance, policy feedback, policy mix, policy design and policy learning all give insights into why policies do not thrive, it is important to note that the existing literature focuses on policies that either never get as far as formal authorisation or are poorly implemented afterwards. Given all the effort it takes to get a policy through the legislative process, it is assumed that some version of enactment will follow.

Since non-enactment only affects a small proportion of the policies passed into legislation, the lacuna in the literature is understandable. Nonetheless, there is an important set of insights here to be explored about why non-enactment occurs and how it relates to the broader policy literatures. We move on now to consider non-enactment in the case of long-term care reform in England.

England’s long-term care funding reforms since 2014

Long-term care funding for older and disabled people in England has a troubled legacy that could be traced back to the missed opportunity of the post-war Beveridge reforms (Hudson, 2021). Unlike healthcare it is means-tested, administered by local government, and largely outsourced to for-profit providers. The system is widely held to be in crisis, with funding failing to match rising demand, high levels of unmet need and a lack of protection against high out-of-pocket costs (CIPFA, 2023; House of Lords Adult Social Care Committee, 2023). The means-test threshold for assets has been fixed at £23,250 since 2010, failing to match inflation and therefore becoming more exclusionary with each passing year (Watt and Varrow, 2018). The New Labour government (1997–2010) came late to the issue, developing proposals for a National Care Service in 2009, which never made it into legislation due to its 2010 election defeat.

In the last decade two funding reform packages have been passed into law during the Conservative Party’s successive administrations. The first package – in the Care Act 2014 – included a ‘care cap’ to limit the financial liabilities of private purchasers and a raising of the means-test threshold so that more people were eligible for state support. The second package – legislated for in the Health and Care Act 2022 – included these elements again, and was preceded by a separate revenue-raising Act (the Health and Social Care Levy Act) which was missing from the 2014 set of reforms.

The key idea underpinning the funding reforms was to ensure that people did not have to sell their home to pay for their care in old age. This had been an important theme in the Conservative Party’s reform agenda and was the rationale for the policy when set out by Prime Minister Boris Johnson in his first speech as prime minister in 2019: ‘My job is to protect you or your parents or grandparents from the fear of having to sell your home to pay for the costs of care’ (Campbell, 2019). The policy instruments through which this idea would be realised were the care cap and the health and social care levy.

The care cap is a limit on the amount a private individual in England has to spend on the costs of personal care (including assistance with washing, dressing, food). Individuals are currently liable for all of their care costs if their assets are above the means-test threshold. A cap was proposed by an Independent Commission led by Sir Andrew Dilnot in 2011 as a way of ensuring no one had to face so-called ‘catastrophic’ costs for care (Dilnot, 2011). In the 2014 legislation the cap was set at £72,000 and the means-test threshold raised to £100,000. As well as placing limits on how much individuals had to pay, the means-test change meant that fewer people would be private payers to start with. This part of the legislation was never enacted for reasons discussed in more detail later. In 2022, a new variant of the care cap was again embedded in legislation. This was set at £86,000 with the means-test threshold set at £100,000 (CIPFA, 2023).

The 2014 version of the reforms did not include any way of raising public revenue to pay for the cap and means-test changes. In the 2022 reforms this aspect was addressed with preceding legislation for a health and social care levy. This was to be a 1.25 percentage point increase in tax on earnings, ringfenced to support the NHS and social care (HM Revenue and Customs, 2021). The levy was due to raise £36 billion of additional funding over three years, of which £5.4 billion would be allocated to social care (HM Government, 2021). The inclusion of the funding element in 2022 appeared to mark a strong commitment from government to ensure that the reforms could not be derailed due to lack of resources.

Despite being embedded in two White Papers, two pieces of primary legislation and many other policy statements, none of these elements of care-funding reform has been enacted. The first care cap, due to be introduced in 2016, never advanced to the commencement stage. It was first delayed and then abandoned in 2017 (Prior, 2015; Doyle-Price, 2017; Jarrett, 2017). The levy was repealed in autumn 2022, prior to enactment (HM Treasury, 2022b). The Chancellor’s Autumn Statement a few weeks later then postponed the care cap until October 2025 (Foster, 2022; HM Treasury, 2022a). The Chancellor indicated: ‘I also heard the very real concerns from local authorities, particularly about their ability to deliver the [care-cap] reforms immediately, so I will delay the implementation of this important reform for two years, allocating the funding to allow local authorities to provide more care packages’ (HM Treasury, 2022a). As this date would by necessity be on the other side of a general election, it was effectively abandoned by the Conservative administration (Bottery, 2023; Hoddinott, 2023). The government appointed programme board responsible for implementing the reforms was disbanded (National Audit Office, 2023).

Explaining non-enactment of England’s care funding reform

Using the categories set out earlier in the article, we can consider how best to explain the non-enactment of the care-funding reforms.

Ideas and instruments

A key flaw in the reforms was that the underpinning idea did not align with the problems that key interests in the sector wanted to solve. ‘Nobody needing care should be forced to sell their home to pay for it’ was part of the Conservative government’s electoral pitch to middle- and high-income households with inheritances to protect (Conservative and Unionist Party, 2019: 12). Other key interests – local authorities, care providers and workforce representatives – were focused on trying to address a range of other issues relating to demand, quality and low pay. The care cap did not solve any of these issues (Buzelli et al, 2022; CIPFA, 2023). It did not offer anything for working-age disabled people (35 per cent of users of long-term care) (NHS Digital, 2022) who were less likely to be homeowners (ONS, 2019). Indeed the cap was likely to increase demand on local government and reduce the income stream for providers by reducing the self-funder subsidy (County Councils Network, 2022).

There were also issues relating to the mix of policy instruments. Social care is a complex institutional space, involving central government, local government and a multiplicity of outsourced care providers. It became clear during the COVID-19 pandemic that central government did not fully understand the complex delivery arrangements in social care, which are very different from the centralised NHS (Dunn et al, 2020; House of Commons Public Accounts Committee, 2020; Curry, 2023; Curry et al, 2023). Into this complex space they introduced a mix of new instruments (the care cap and the levy) for which there was little or no precedent in the UK (compare the way in which the Massachusetts model provided a blueprint for the Affordable Care Act in the US; Oberlander and Weaver, 2015). The government’s own Impact Assessment raised a range of issues that had not been considered in the legislative debates, such as the requirement for local authority IT systems to keep track of individuals’ care costs as they meter towards the cap and the potential implications of charging reform on market stability (DHSC, 2022). The cap itself would be hugely complicated for local governments to administer. To know when the cap spending limit was reached, local authorities would have to monitor spend by self-funders over a period of several years, yet these people are often invisible to the state (Henwood et al, 2022).

The sequencing of instruments was also a factor in shifting the valence of the issue. While the original 2014 reforms missed out a revenue-raising element, the levy was an important precursor to the 2022 reforms. However, to provide political and market reassurance that more resources were finally coming into the care system, the income-generating elements of the reform (the levy) needed to come before the income-spending elements (subsidising private payers). Indeed this sequence was planned, given that the levy was due to come in from 2022 and the cap from 2023. However, the levy was likely to be the least popular element with the public and was particularly risky to introduce during a time of high inflation and concerns about the cost of living (Fuller, 2022). The benefits of the reforms would only be felt by those reaching the level of the cap – likely to take a minimum of three years (Sturrock and Tallack, 2022) and only affecting a small minority of the population. This front loading of generalised costs and back loading of a narrower set of benefits runs counter to the learning about successful policy sequencing discussed earlier in the article. There is also an issue here about dependencies, and the viability of one part of legislation once another part is removed. The cancellation of the levy meant that it became very difficult to see how the care cap could ever be enacted, unless the government identified a new funding source.

Institutions

Institutional factors are also part of the explanation for lack of commitment to enactment. The complex institutional landscape of social care was highlighted earlier. Within central government we can see the lack of support from the finance ministry (HM Treasury) as a key issue. The Treasury’s role in controlling public spending, setting taxation rates and establishing the direction of economic policy means it has great influence over other government departments. Its significance as a blocking force within Whitehall has been highlighted in other work on reform delays (Richardson, 2018; Abrahams, 2020). The Treasury has been known to be sceptical about care-funding reform for a long time, unconvinced of the need to shift the burden of care payment from private payers to the state – signalling that reform would cost too much and its beneficiaries likely to be society’s wealthiest (Charlesworth et al, 2021). The Treasury were unable to block the passage of the care-cap legislation, but their role in cancelling the levy – seen as challenging the commitment to a ‘low-tax, high-growth economy’ (HM Treasury, 2022b) – undermined the viability of the rest of the funding reforms.

We can also identify issues with institutional capacity within central government. The enormous administrative complexity of the Brexit process has limited the scope for government to prioritise other areas of policy (Andrews, 2018; 2019; Observer, 2020; Hill, 2023; Taylor, 2023). In many countries, the COVID-19 pandemic was a huge drain on the capacity of administrations to undertake other areas of policy enactment (Atkins and Hoddinott, 2020; UK Parliament, 2020; Holmes, 2021). The high turnover of staff at the Department of Health and Social Care meant that few civil servants had a good understanding of this complex policy terrain, with turnover rates as high as 51 per cent in 2021/22 (Clyne and Bishop, 2022). Officials cited reduced staff numbers and project expertise as reasons for implementation delays (National Audit Office, 2023).

Having been tasked by Prime Minister Johnson to ‘sort out social care once and for all’ (GOV.UK, 2019), it is perhaps not surprising that a new set of civil servants dusted off the care-cap idea from 2014, and with minor changes reintroduced it as new legislation in 2022. Using the distinction between ‘“doing things better” versus “doing things differently”’ (Hayes and Allinson, 1998 cited in Dunlop, 2010: 346), we can argue that in this case some things were done better the second time around, for example the accompanying levy to pay for the reforms. However there is little evidence of doing things differently, through learning and adaptation in light of the factors that led to the abandonment of the first care cap.

Interests

Complex policies to address major social issues do make it to the stage of enactment, despite all the reasons why they may not. In the same decade as the care reforms the Conservative government reformed social security spending with its Universal Credit legislation (which was an implementation fiasco, but not an enactment failure; National Audit Office, 2018). Thus to explain non-enactment in social care we need to identify not only problems with the underpinning ideas and institutions, but also how these are mobilised by key interests. Much of the US literature on policy failure highlights the role of opponents building public and political coalition against reforms (see for example, Patashnik, 2008; Béland et al, 2020). However the care-funding reforms were not subject to obvious partisan or interest group attack – opposition parties and stakeholder organisations were broadly supportive of the reforms in principle as a response to the widely recognised care crisis (Local Government Association, 2022a; Caring Times, 2023). Indeed, perhaps most interesting here is that the undermining of the reforms at the enactment stage came from the interests who had advocated for the new law.

The first of these was the governing Conservative Party itself, which effectively unstitched its own reforms. Unclear messaging around social care funding reform was seen by the party as a reason for Conservative leader Teresa May’s drop in public support during the 2017 election campaign (Opinium, 2017). Hence the risks of this becoming a negative valence issue were seen as high. Turmoil in the Conservative Party leadership (four prime ministers between 2019 and 2022) led to a rapid succession of internal party battles for the leadership in which candidates vied to outdo each other on their fiscal conservatism (Adam et al, 2022). Ministers cited concerns about inflation and the high costs of living as the reason for abandoning the levy in 2022 (GOV.UK, 2022; HM Treasury, 2022b). The political risks of stepping away from the cap were lowered by muted public support for the reforms. Polling on behalf of the Health Foundation in 2022 found that nine months after the announcement of the cap, over half of people living in England (58 per cent) were not aware of the policy and when it was explained to them only half (49 per cent) supported the policy (Buzelli et al, 2022).

The second group of ostensibly pro-reform interests who undermined the proposals were local government representatives. Local government peak bodies such as the Local Government Association and the Association of Directors of Adult Social Services, had been campaigning intensively for years for social care funding reform. However, in both the 2014 and 2022 cases, it was pressure from these bodies that in part contributed to the non-enactment of the funding reforms (National Audit Office, 2023). It is notable that local authorities did not criticise the ideas behind the care cap but expressed concerns about cost and implementation and how the ‘underfunded reforms would have exacerbated significant ongoing financial and workforce pressures’ (Local Government Association, 2022b). Analysis by the County Councils Network (2022) estimated that the additional financial and care assessments required by the reforms would necessitate an additional 5,000 staff to undertake a further 197,000 assessments and would lead to increased waits for people to access care packages. Thus when the government’s announcement of the delay to the introduction of the care cap came, it was plausibly framed in terms of a reaction to local authorities’ concerns about not being ready to respond to the financial and operational demands of introducing the cap (HM Treasury, 2022a).

Given the lack of support for enactment of the reforms among key interests within central and local government, and limited public endorsement, there was no clear constituency pressing for enactment. The cap failed to do what Patashnik (2008) argues is a key requirement of reformers, which is to build a coalition of interests for the legislated version of the policy at an early stage of enactment. The care-funding reforms may be best understood as what Patashnik (2008: 168) calls a ‘winnerless’ policy, in which the reform is put in peril not by sabotage by the losers but by the indifference of the winners.

Looking across ideas, interests and institutions, we can see that for the Conservative administrations, the risks of abandoning care-funding reforms of both 2014 and 2022 came to be seen as lower than the risks of proceeding. Key reasons for halting once the legislation had been passed were the complexity of the idea, a lack of strong advocates for the reforms among politicians and organised interests, turbulence within the Conservative Party and a lack of institutional learning to ensure that the mistakes of the first care cap were not repeated.

Discussion

Existing literature looks at the difficult processes of policy design and policy making, taking it to the moment of legislative approval. Other literature takes it from the point of enactment to see how well implementation fares. But there has been a failure to acknowledge a slim but important set of cases in which policies gain legislative endorsement but are never enacted. This is a category of legislation that can give us important insights into the balance between ideas, institutions and interests within policy regimes.

This article has focused on long-term care reform in England, which is an enormously important issue for the millions of people using care services and their unpaid carers. The current funding system has limped along since 1948 with minor tweaks, but is widely acknowledged, including within government, to require a major overhaul. Over the last decade two major pieces of legislation have been passed by Parliament to reform care funding and to date neither set of funding reforms has been enacted.

This article makes four original contributions. The first is to draw attention to non-enactment as a theoretical category which is not recognised in the policy literature. It is not a failure to legislate, nor it is a failure during implementation. It is the deliberate stepping away from a policy at the point when it gains legislative authority. As well as constituting an under-examined category of policy, non-enactment has systemic implications. These may be distributional or constitutional. From a distributional perspective, as Patashnik notes (2008: 6) ‘the unravelling of general-interest reforms harms the economy, squanders scarce political resources, and breeds public disillusionment with government’. From a constitutional perspective, if a government decides a policy is ‘unworkable’ despite having legislative endorsement, there are clear issues for democratic accountability. Legislative processes have the advantage of being relatively transparent, documented and visible to the public. Voters can probably assume that if a bill gets majority endorsement in the legislature it will be enacted in some form or another. It is clear that this is not necessarily the case.

A second contribution is to draw attention to cases in which a government distances itself from its own legislation. The label of ‘unworkable’ is a framing by political actors who consider that the risks of proceeding with enactment have become too high. There is a clear imperative to interrogate why that would be. Writing about policy inaction, McConnell and ’t Hart (2019) distinguish between calculated, reluctant and inadvertent inaction. Calculated intent is a central element of symbolic politics. As Newig (2007: 278) puts it, ‘Symbolic legislation deliberately fails to meet its declared objectives’. However, policy non-enactment is not conceptually premised on calculated intent, and indeed there are clear methodological challenges in establishing ‘what the responsible actors knew, intended and could have known’ (Newig, 2007: 279). But it raises interesting questions about why political actors would abandon hard won legislative gains at the point of success.

Contingency is a key part of the explanation given by Patashnik (2008) for whether or not reforms endure. In the English case, the fiscal pressures of COVID-19 and the cost of living crisis played a role in the suspension of the levy – and once the levy was scrapped the care-cap reforms became more fragile (National Audit Office, 2023; CIPFA, nd). However it is important to note that the care-funding reform legislation was passed in 2022 when these fiscal pressures were already well known, so it is necessary to be alert to how those financial concerns came to be framed by interests (contenders for the Conservative Party leadership, for instance) rather than seeing them as explaining non-enactment per se.

An alternative explanation is that stepping back from enactment as ‘unworkable’ may be purposively done by governments to discredit implementation partners. In this case, there seems to have been a strategy to frame local government as unreliable actors. Blaming local government for non-enactment is relevant in a context of multilevel governance and wider pressures facing local government. With several English local authorities having de facto declared bankruptcy in 2023 and others likely to follow, central-government appointed commissioners have been sent in to ‘rescue’ local authorities from their own inadequacies (Local Government Information Unit, 2023). In this context, the failure of the care cap becomes part of a central-government narrative of local incompetence, and gives further impetus to efforts to surveil and punish local government.

Gunter and Courtney’s (2023) concept of ‘policy mortality’ is useful here. Writing in relation to ‘failing’ schools, they highlight how the labelling of schools as failing is a deliberate tactic by government. Regulatory policy is used to reveal the extent to which ‘schools are diseased’ and must be closed or ‘cured’ (2023: 356). While the schools context is different from our own, the stretching of the concept of policy mortality can help us to understand the calculus facing politicians. To label other actors as failing can bring a tactical advantage. Rather than policy being symbolically implemented to indicate that government is ‘doing something’, non-enactment is used to draw attention to the failings of rival institutions. Whereas Gunter and Courtney explore how policy is weaponised through its enactment in school inspection processes, we can see how non-enactment can also be used to discredit other political actors as unreliable.

The third contribution of the article is to draw attention to the case itself which remains a scandalous unsolved policy problem. There is a very real danger that after two recent abandonments politicians consign care-funding reform to the ‘too difficult’ box and leave it alone for a generation – part of the learning resistant policy pathologies that Dunlop (2021) identifies. Unpicking the constellation of ideas, institutions and interests that have led to this policy failure is an important part of helping future campaigners and policy makers consider how they might make it ‘third time lucky’ on care-funding reform. This links to the literature on double-loop learning (Dunlop, 2010) and how can policy be made better, rather than merely different. Such a focus relates to the content of policy but also requires unpacking the concept of policy being ‘unworkable’. Scholars of policy endurance have highlighted that legislation needs to avoid being either too rigid or too flexible if it is to endure (Jordan and Matt, 2014; Luetjens, 2023). There is a challenge here to consider if and how policy designers can inoculate a policy against non-enactment.

The fourth contribution is to set out some methodological considerations to take forward the study of non-enactment. Non-enactment is easier to research than other forms of policy inaction, since this is a conceptual binary: a government presses ‘go’ on the implementation of legislation or it does not. Nonetheless there is a methodological problem of distinguishing non-enactment from delay. Feng and Fox (2022) look at the delayed implementation of menu labelling by the US Food and Drug Administration, which took eight years to be implemented. Future empirical studies on non-enactment could look at the relationship between delayed policy which is then enacted and delayed policy that is abandoned. Studies could also look at the frequency of non-enactment, patterns in the types of policy that are abandoned at this stage and what can be learned by taking a comparative lens.

Policy that has been unenacted or under-implemented will be more visible in some political systems than in others. This point links to the limitations of our study. We offer theory development from observations of a single case in a Westminster system, with the case selected on the dependent variable. It has enabled us to identify a category of policy failure which has not been sufficiently elaborated. However, there is work to be done to establish how often and how much non-enactment occurs, and in what ways this matters. In particular, future work needs to be consider whether this is a research agenda that can only be pursued in Westminster-style systems where non-enactment is reported systematically, or whether it can be discerned in presidential and federal systems where failure to enact is more opaque.

Conclusion

The tumult of the proposed care cap and its delay, abandonment, revival and postponement marks a disappointing phase in England’s attempts to modernise its welfare state. Whereas many other countries have reformed their long-term care systems to respond to population ageing, England’s continues to stagger along unreformed. For people requiring high levels of care and whose assets are above the means-test, there continues to be no limit on their financial liability. This is a failure of risk pooling that most advanced democracies do not tolerate for old age or disability.

This conceptual article has identified non-enactment as a previously ignored category of policy failure. Recognising non-enactment as the key phenomenon at work in the case offers tactical and theoretical insights. It homes in on a stage of the policy cycle which is often ignored or elided. The next stage after the ratification of legislation is not implementation, it is ‘commencement’ or enactment of the legislation. There is scope for further empirical work on this. Future studies can map patterns in non-enactment and move beyond the particularities of an individual case. It is possible to consider the ways in which ideas, institutions and interests could be better aligned, as part of moving social care out of its current policy stasis and making recommendations for the future.

The article has not engaged with the democratic issues inherent in non-enactment. Democratic institutions are premised on executives implementing policy that the legislature has passed into law. If executives decide that legislation is unworkable and do not proceed to the implementation phase, there need to be mechanisms to surface and hold to account such decisions. While some countries (such as Australia) do require reporting from the executive of which policies have not ‘commenced’, in others there is no such process. Better scrutiny mechanisms would assist transparency and comparability in the scale of non-enactment and the types of policies affected. This would offer a route to ensure that we have a proper account of when and why ‘law is not law’ (Norton, 2021).

Funding

This research was funded by the Economic and Social Research Council (ESRC), award ES/W002302/1, with contribution from the National Institute for Health and Care Research (NIHR) (Department of Health and Social Care). The views expressed are those of the author(s) and not necessarily those of the ESRC, UKRI, NHS, the NIHR or the Department of Health and Social Care.

Acknowledgements

We would like to thank Kate Hamblin for helpful comments on an earlier draft of this working paper.

Conflict of interest

The authors declare that there is no conflict of interest.

References