Conclusions | 12 | Making Policy Work | Peter John

ABSTRACT

What is the relative impact of six diﬀerent kinds of resources governments can use to improve policy outcomes? It may be the case that making this judgement is too hard a task given the large amount of variation across and even within policy sectors, and the diﬀerential impact of the resources themselves. The tools themselves are complex and protean, containing diﬀerent elements, and often evolving to become like other resources of government. It is hard to distinguish the impact of the tools from each other, as many interventions seek to combine the tools of government. Most of all, it is a challenge to separate out the eﬀect of the tool of government from the decision to use it, which arises from the political context, including past policy decisions and performance. But this book is on a quest to ﬁnd the answer. For the problems social scientists study are always complex: the task is to ﬁnd a way through to make some generalisations, even if qualiﬁed ones. Policy-makers seek to impose some clarity on the way in which they make decisions, which is why the use of one tool of government is unusually prominent even in the most hybrid of interventions. The best evaluations reported in the previous chapters have also sought to separate out the contributions of a particular type of public activity. Partly to ﬁnd a way into the problem, the simplifying approach this chapter

adopts is to split the task into two steps: the ﬁrst is about the ease of introducing the tool, the degree to which a particular resource can be applied by the state or other public actor, which is partly about the practicality of getting machinery and processes into place, but is also about the political environment that constrains decision-makers, either through the requirements of institutional rules, the pressure of public opinion for allowing them to act and the weakness of politicians in the face of a traditional bureaucracy. The second is an assessment of the eﬀectiveness of the tools, technical considerations, such as understanding the impact of compulsion, persuasion or networking, for example, but which also may reﬂect background political constraints and choices as to why a particular tool may have a limited or strong eﬀect in the society at large. Each of the six previous chapters has reviewed empirical studies of these impacts. To represent the summary, the chapter puts these two dimensions into

Table 8.1. The device is a guide to thinking, so is not intended to be an

authoritative assessment. Inevitably it is something of an oversimpliﬁcation given the variety of impacts and diﬀering conclusions of the studies themselves. The table has one column for the ease of using the tool and another for the

eﬃciency by which the tool leverages the outcome itself. Each cell has an entry on a three-point scale of low, medium and high. Law and regulation may be introduced fairly easily, depending on the

political system, whether it has a relatively undivided pattern of government or is one where there is a separation of powers and extensive number of veto players (this applies to public ﬁnance too). In unitary political systems it is possible to get laws in place very quickly, but in federal and divided ones it might not be so possible for the executive to get its way. But in principle, if there is a will behind executive action, getting a law on the statute book is not a big challenge. But the problem is whether it can be enforced once publicised, and much depends on the acceptability of the law itself, the resources directed to implementing it, whether the law is compelling a form of behaviour or prohibiting it, whether interest groups support it and if the bureaucracy and other public agencies are behind it. Then the population may simply not want to obey it and there is very little a central state or other public authority can do in those circumstances. If obeyed, there is a chance that legal regulation can achieve the desired outcome if the action is clearly identiﬁable and then can be modiﬁed on the basis of legal regulation. Even if a law is obeyed, there might still be negative consequences from passive resistance and lack of internalisation of the norms the law is intending to promote. It would seem as likely that, even if the theory is right, it is possible legal regulation could undermine legitimacy and eﬀectiveness by its tendency to command and control. This may even be the case when the law and regulation aim to be smart and responsive, as they either default to the hierarchical mechanism or legal regulation eﬀectively turns into another instrument more associated with negotiation and provision of information, such as with restorative justice. Public spending and taxation can be relatively easy to implement, but much

depends on the state of public ﬁnances and whether there is political support for paying for it. Finance is less eﬀective than might be imagined because it is diﬃcult to get money to the right areas and in the right timescale. The problem is that the impacts are not as high as one might expect because of the

diﬃculty of ensuring that the money has not been wasted and does not get soaked up into the bureaucracy, and in producing policy outputs rather than outcomes on the ground. There may be limits to the extent to which ﬁnancial incentives will lead to changes in behaviour. Of course money is the precondition for most eﬀective interventions and would be used alongside other tools, but where money is the key to the intervention the costs of using it, especially if applied quickly, might be surprisingly high. It works well if there is an unmet or new need that has not been fully addressed by policy-makers, and taxes work where there is an identiﬁable group that has an incentive to change behaviour as a result of the change. The overall conclusion for the two classic pillars of the state – law and ﬁnance – is that they are reliable forms of public action but tend to disappoint on their promise. Less direct measures might be able to ensure that governments have the

means to implement as well as the direct resources to apply to a problem. Reform of the internal organisation of the bureaucracy, especially in the direction of the new public management, is relatively easy to put into place given the control states have over their internal procedures of administration, and that bureaucrats are under the command of politicians, at least formally. There is good evidence that the reform of the bureaucracy improves the quality of administration and has favourable eﬀects on performance and the delivery of policy. Even though it is hard to show the impact of the new public management, and there are disadvantages from wholesale reform, the best studies show positive eﬀects of particular measures to improve performance though changing incentives, such as from performance-related pay and performance measurement. Institutional eﬀects are probably the strongest in the book, partly because

they alter the rules of the game fundamentally. They also aﬀect the conditions under which citizens and groups cooperate so provide a direct route to improve policy outcomes. But political institutions are hard to introduce, tend to have an impact only in the long term, and their eﬀects may depend on the context in which they are introduced. It is also not clear which institutions are the best to introduce either, whether to concentrate them for accountability and decisiveness or fragment them to allow for the cross-pressure and learning. Nonetheless, institutional reform is a powerful instrument of change that is often underestimated and misunderstood. Moving to the pair of softer instruments is, ﬁrst, information, which is an

attractive low-cost resource of government, and which is easy to introduce, but where just persuasion alone might not be a cogent reason for individuals and organisations to change their behaviour. But there may be more potential in this set of tools than might at ﬁrst be thought, as there can be power in well-crafted information signals. The leverage increases when public actors develop the skills to inﬂuence the choices of citizens, through nudging them to a preferred course of action. It may also be the case that when citizens get the opportunity to debate public matters and to take decisions into their own hands, they can inﬂuence outcomes by making and delivering policies themselves.