Sunday, 17 January 2010

So What Is Good Policy-Making?

Following on from post yesterday on what constitutes a policy, I set out below what are - for me - the attributes of a good policy.

1. Precision and Clarity

A very general aim can be too vague to be a policy; there has to be a certain exactness to what one wants to achieve.

Libertarianism or socialism are principles or ideologies, rather than a policy. However, the reduction of police power and the alleviation of child poverty are both sufficiently precise and clear to be policies, as well as being respective applications of libertarian and socialist principles/ideology.

2. Evidence-based

A policy which disregards the appropriate evidence base is misconceived; a good policy is based as far as possible on the relevant evidence.

Of course, there are questions as to what constitutes the relevant evidence. But when there is evidence - for example, in respect of the danger of drugs, the evidence of global warming, the need for testing on animals, or the need to control foxes in the countryside - then this evidence must play a role in the formulation of policy.

This is not to say that such evidence comprises the policy or even determines the shape of the policy (as there can be other factors), but the evidence must be acknowledged and, if a policy (for any reason) appears to be contrary to the relevant evidence, then it is the duty of a policymaker to acknowledge this and explain the apparent discrepancy.

3. Transparency and Accountability

The formulation and implementation of any policy must be a transparent process, and as participatory as possible, and there must be some person who is accountable for the content and the success of the policy.

These attributes are not only likely to make a policy more legitimate (see 4 below), but they are good in themselves. Sunlight is an excellent disinfectant and policy-making is likely to be more biased and less well-informed with a closed group of policy-makers.

4. Practicality and Legitimacy

A policy must be capable of working in practice.

There is no point asking the state, or the market, or the "third" sector, to effect certain outcomes which are not possible. The history of post-war Britain is littered with both market and public sector failures.

There is also no point seeking to implement a policy which is unacceptable to the relevant constituency, especially if the policy includes financial burdens (for example, higher taxes or lower benefits) or restrictions on autonomy (for example, greater police power, or restricting the possibility of travel). In such instances, the constituency affected may simply reverse the policy before it can have any effect.

An adversely affected constituency may, however, accept burdens or restrictions if they are seen as legitimate; and so here the role of the policy-maker includes public engagement.

4. Non-reliance on Coercion

Although a policy may require legal coercion, this should always be kept to the minimum.

Legal coercion - the use of prohibitions and sanctions - is a necessary part of regulating society, but it is a blunt method of policy implementation when used by itself.

As I mentioned yesterday, law is only one of many means for implementing a policy: others include decision-making, rule-making, guidance, circulars, advertising, campaigning, persuasion, allocating resources, prioritisation, showing leadership, and so on.

5. Proportionality

A policy should go no further than necessary to achieve the desired outcome, and the policy-maker should be mindful of the potential wider impact of a policy.

For example, a policy regarding terrorism or anti-social behaviour should not infringe on the rights of individuals more than is required than to achieve the policy.

6. Contestability

There should always be the means of contesting a policy, and one should never seek to close down a debate about policy.

Policies are simply not like peer-reviewed scientific papers. The consensus of experts will not make a policy either workable or successful, even if there is an entirely valid consensus as to the relevant evidence.

Policy-makers can - and do - get things wrong. The policy thereby should always be open to challenge, and even rejection, even if the relevant evidence cannot itself be sensibly disputed. This is especially so when the implementation of the policy involves burdens and restrictions on individuals.

My intention is not to put forward vaporous "apple pie" criteria; instead, I am seeking to open up a discussion about what is good (and bad) policy-making.

And, as blogging and web-based participation becomes more popular, I am also trying to see whether there can be any useful influence by bloggers and others on policy-making (and thereby law-making) rather than the (for me) often inconsequential struggles between various pundits and politicians.


No purely anonymous comments will be published; always use a name for ease of reference by other commenters.


MTPT said...

In business one standard approach to setting targets is SMART: Specific Measurable, Achievable, Realistic, and Timed. Some of that is reflected in the criteria above (which are also good ways to judge) but Measurability and Timing are not (at least explicitly).

I would submit that whatever else it must be, a "good" policy has to have "victory conditions" - there has to be a way of knowing whether or not it has succeeded - which are time limited (the argument against this approach - that it may be difficult to measure certain types of outcome - is actually an argument against making policy in such areas at all!).

On "Evidence-based", what other factors would you see playing a part? All of the examples you cited are cases where personal prejudices - of one kind or another! - have grossly distorted policy and/or policy-making, with alarming results. I can't see any justification for policy being made other than on the basis of evidence - explaining it isn't enough.

Evidence in the context of policy making would encompass analysis of and respect for relative benefits and harms, which is certainly not how policy seems to be made at present.

Matt Wardman said...

I'm tempted to add something relating to "shall not leave too much of a burden for future generations, or attempt primarily to bind successors", bearing in mind all the current promises being passed into law which seek to tie the hands of future governments, and also that under PFI many invoices have simply been deferred.

Stephen Morris said...

Another issue is the tendency for policy initiatives to be changed or hijacked: terrorism laws are used for other purposes; military interventions suffer mission creep; a policy which is failing has it's objectives redefined; money is diverted away undermining the policy.

There should be:
1. A change management process. Changes to policy should be explicit.
2. Consideration in the original policy of how it could be abused and controls to minimise this.

I certainly agree that there should be measurement of the success of policies against their original objectives. There should be a process of learning and improving policy. Different approaches can be trialled and perfected in local schemes before deciding to invest in a national policy.

Unfortunately this takes too long. Politicians require headline-grabbing policies that deliver within a parliament. So we lurch from one failed national initiative to the next.

Juanfelipe said...

Just wanted to emphasise and extend a point Stephen Morris makes. We often think of policy design and implementation as a vertical top down process. This is not what happens in practice as even the conception of a policy priority does not happen in a vacuum. The adaptation and iteration of what might be a good policy in a given area should perhaps be built into the policy process. As Stephen suggests our political system, and I venture the pressure of constant media scrutiny, may well mitigate against good policy making and implementation. The glare of publicity might make for the best disinfectant but equally if you use disinfectant too liberally you might kill off the roots of something that could grow into becoming good policy. Not that we need be too pessimistic about our system. It is worth sometimes following a bill through parliament to see the extent to which interests and viewpoints are represented and can have influence in improving law.

WoollyMindedLiberal said...

But when there is evidence - for example, in respect of the danger of drugs, the evidence of global warming, the need for testing on animals, or the need to control foxes in the countryside - then this evidence must play a role in the formulation of policy

The problem here is implicit in your assumption that something dangerous or bad for you must be made illegal and punished accordingly. Skiing is dangerous, there is strong evidence to support this and should therefore by your logical be punished accordingly. Fast skiing is more dangerous and so is skiing offpiste and therefore they should receive more severe punishment. This is good policy, it is clear and evidence based.

There is strong evidence that testing on humans is much more effective than testing on animals. Therefore we must punish those who test on animals instead of humans by your logic.

I think it is a great thing that you do not engage in politics or intend to stand for office. Please continue to leave it to others. What you do you do very well but it may be best for you to stick to your field of expertise.

Jack of Kent said...

"The problem here is implicit in your assumption that something dangerous or bad for you must be made illegal and punished accordingly"

Er, just for clarification, I do not accept that is implicit; and, moreover, as a liberal, I actually share your reluctance in banning things.

Dominic Haigh said...

Linked to evidence, where there is no good evidence prior it is important that feasibility should be tested where possible.

Cost implications are also an important part of practicality.

twaza said...

An excellent list, and some good comments, but there is something missing, which we can see if we ask:

"How would you distinguish knee-jerk reactions designed to get past the current crisis, from considered decisions intended to reach a long-term objective?"

Policies reflect a set of values and a set of beliefs about how best to achieve your aims. The values and beliefs ensure that individual policies are coherent and understandable.

When politicians are criticised for a lack of ideas, the problem usually is that their primary aim has become to stay in power.

Nick Sharratt said...

I agree with your points closely enough not to merit comment, and I like the suggestion in the first comment that policies should include the SMART principle.

Is the consequence of this formalised approach to look for a political equivallent of ISO 9001 for governments? A structure of quality processes which can be examined and audited with reports of non-complience, continual improvement etc? Would that mark the end of party politics and true democracy or rather a beaurocratic tyranny as civil servants end up being the only ones who can understand and interpret the system as it eventually gets ever more convoluted?

ivan said...

The main impression I have obtained in comparing the aims and policies of the current labour and previous conservative administrations are:

Labour (97-today): often laudable aims substantially implemented by weak policies that often failed to achieve those aims
Conservative: (79-97) less laudable aims rather more often effectively implemented

(though obviously one can find numerous exceptions)

People have talked about SMART etc, but sometimes the demonstration just isn't available. Consider climate change: if we do reduce CO2 output, we shall never really know if it was actually necessary.

So it is in economics - not everything can be sized, you often can't try out both alternatives, you often can't separate causes. Privatisation policy is a significant example. Public monopolies can pursue social aims which often get lost in a competitive market. We might wish to trade that off against the efficiency benefits of competition or private operation. But whilst economists have techniques for measuring social benefits, they don't have an answer to the question of just how much more inefficient is a public monopoly than competitive industry. (I've sometimes had to produce an "answer" to that question, though, when someone really insisted.) Which makes carrying out the trade-off difficult. So sometimes you do just have to go by gut feel when it comes to sizing the costs and benefits of certain policies.

Having done a fair amount of policy analysis in my time, I've come to realise that the policy branches of the civil service are often themselves quite good at policy analysis, or at least have procedures which result in them paying experts to assess it when they can't. So when the government nonetheless persists in introducing an ineffective policy, they have usually been told just what the problems are. But the politicians often aren't very good at telling the difference between good analysis and special interest lobbying. And instruct the civil servants to proceed nonetheless.

Adam said...

An interesting and enjoyable couple of posts... I will admit that my first response was one of mild "ugh" (my usual response to the current state of affairs where Business Studies and Marketing Skills seem to have ousted ideology) but I found the posts and the responses provocative which is always a good thing :-)

But I have been left wondering... where does ideology fit? I heard on the radio this morning a discussion about banning smoking in cars because "the public is ready for it"... followed by a whole wishlist of bans that the public is not yet ready for, but may be at some time in the future (with enough fear mongering, I guess). We seem to exist in a milieu of prohibition rather than freedom and innovation, and I would hold an over emphasis on "evidence base" as partially responsible. Don't get me wrong... we must be accountable to the facts if we propose any course of action (the alternative would require medical treatment)... but policy making must also somehow include some statement of belief and understanding of the human condition, what it means to be a person in relationship to society and the community.