Policy for Campaigns 02: Two basic types of policy problem?

Every policy challenge you'll face can be placed in one of two categories - or can it?

If your organisation is campaigning for a change, there must be something wrong with how things are to begin with. If there wasn’t something wrong, there would be no need to try to change things.

This can however pose a challenge: to what extent should an organisation campaign against a “problem” and to what extent should it campaign for a positive vision? Your policy analysis and development should play a central role in identifying the answer to this. (See this previous edition of Policy Pulse for a relevant anecdote concerning the Coca Cola Corporation, and also an example of a report where I didn’t find its framing of a positive “opportunity” very persuasive, because it looked very much like a problem to me.) It may well be a topic to revisit in its own right in a future article.

However, this article is about the much more basic question of identifying and understanding your problem. I’m going to go out on a limb and suggest that there are two basic types of policy problem that a campaigning organisation might try to address. This feels too neat: I suspect it’s wrong, but I can’t really see how and am hoping readers will be able to tell me. I’ll certainly be pointing out ways in which it’s more complicated than that – I’m a policy professional, after all – but in the first instance I can see two basic categories of problem.

The first is a bad formal or technical policy. This is where a set of rules, regulations or laws, or some other formally defined system, is defective in some way. Changing the rules / regulations / law / system will, other things being equal, remedy the problem.

The second is a problem arising from behaviour or culture. This means that, in some way, somebody isn’t doing what they are supposed to do, or behaving as they are supposed to behave, and this is giving rise to some sort of harm.

This is not the same, incidentally, as saying that things can only be bad, or problematic, in two ways: they can be problematic in lots of ways! A policy might be achieving its intended ends but also producing unintended consequences; a policy might have been directed at achieving a bad goal, or a goal with many down-sides, in the first place; a policy might simply have failed to do what it was supposed to; it might have succeeded only inconsistently; and so on. Again, that might be one for a future article.

But for now, let’s explore our two types of policy problem.

Bad formal policy

You may encounter situations where the problem is very clear-cut and can be solved by changing the rules. Tax and welfare benefits policies can work like this: a tax might be set too high, or applied to the wrong thing; or benefit entitlements might be too low, or not available to the necessary groups of people. In these situations, changing the formal rules is necessary in order to remedy the problem, and will often also be sufficient.

There are plenty of other examples, and structures laid down in primary legislation (that is, laws passed by Parliament, as distinct from regulations made later by ministers) are particularly prone to it. The ticketing and pricing structures on the railways, for instance, are still largely locked into a design from the pre-privatisation British Rail era, which might have made a great deal of sense at the time, but now makes it extremely hard to offer new forms of tickets or to vary prices in some circumstances. A “hard” change of the rules is necessary in order to resolve this.

You probably have examples of your own – please do add them in the comments.

Bad behaviours

Many policy problems are not simply matters of formal rules that can be changed with a stroke of the pen. Health and care policy are classic examples: ensuring that care is consistently delivered according to known good practice is very difficult, as in many fields our knowledge of what constitutes a “good” approach is constantly developing, and at the leading edge of that development there may well be disagreement among experts; even away from that, disseminating knowledge about all the different possible approaches and what might work well in what circumstances, is a massive challenge.

Other examples of undesirable types of behaviour can also be found in healthcare: the NHS (and, I would expect, many other health systems) is notorious for not admitting fault when something has gone wrong, with institutions and professionals instead “circling the wagons”, not co-operating with investigations, and withholding information. This is often not so despicably motivated as it sounds: many professionals don’t want to see a colleague slung out because of (what they may believe is) a single mistake; but in thinking like this they can easily lose sight of the fact they are preventing someone who has been harmed from getting redress.

In some respects, these behavioural problems are preferable. What I mean by that is that, if your organisation supports people through helplines, advocacy, professional advice and so on, it may be able to support people to assert their rights. If an organisation or individual has done something wrong – whether that’s against the rules they should have followed, or straight-up against the law – it should in theory be possible to secure redress and have the matter put right. It may of course take a long time, require a lengthy complaints / tribunal / court process, and involve a lot of hassle and even up-front cost. In practice therefore, sorting things out in this way isn’t always a feasible option for everyone, even when they have been on the receiving end of something that, in strict terms, should not have happened.

Possibly the “behaviours” category of problem can be broken down into two: behaviours within systems and organisations that do things to people, like public services; and behaviours among people generally that need to be shifted. So for example, the enormous policy success of smoking reduction in the developed world has succeeded in changing individuals’ behaviour on a large scale; it has not been (of itself) a matter of remedying culture or behaviour within particular organisations.

It's a bit more complicated than that

In practice, there can be interplay between these two types of problem, and you might find yourself having to address a situation that involves both.

In one scenario, they might just both be happening at the same time. Disability assessments for accessing welfare benefits are an example of this: the design of the tests themselves has weaknesses, and is overly medical (particularly the Work Capability Assessment), while the eligibility rules for the benefits are ungenerous. These are clear-cut “formal” policy problems. But another major barrier is the poor quality of delivery by clinicians who undertake the assessments and decision-makers who make the final decisions: many such decisions end up being overturned on appeal, because of errors and, quite possibly, because of an organisational culture that encourages applications to be turned down. This is a set of behavioural problems.

Another variation is where formal rules of a system have the effect of driving and generating undesirable behaviours in institutions. You can see this in tax for small businesses and the self employed: for example, the intermediaries legislation (“IR35”) was created specifically to restrict tax advantages for self-employed people who use a limited company, but the creation of a set of rules specifically aimed at this group led to many overly aggressive investigations of taxpayers who use limited companies entirely lawfully and have done nothing wrong. Tax inspectors with little experience or knowledge of business practices quickly picked up cues to take an aggressive, attacking approach at the expense of individual taxpayers.

Similarly, formal rules can directly shape the behaviours of consumers or businesses. The tax rules around company cars are complex and have shifted greatly over time, but appear to be the key driver in trends in company car use, both in terms of whether employees opt to take advantage of them, and whether companies offer them.

On another level, systems more broadly might drive behaviours that prove to be undesirable. This is particularly relevant to the sub-category (or possible third category?) we identified above. To take an obvious example, it can be said that our food supply chain is geared up, intentionally or not, to increase levels of obesity, with the population-wide health problems that flow from that: unhealthy food is easier and cheaper to obtain than healthy food. That being the case, many people make the rational decision of choosing to eat it (any eventual down-sides being far less obvious and pressing at the time they make the choice). Reversing that pattern of rational but undesirable behaviour would involve disrupting our entire system of food supply.

A final permutation (for now) is where “hard” policy is needed to bring about behavioural change, but the relationship will be messy and indistinct. Take the need to engineer a large scale modal shift in travel, away from private cars and onto public transport / cycling / walking (and also from road freight haulage to rail). There are “hard” policy levers that need to be pulled: the tax treatments of different forms of transport or fuels for them would be one; the construction of new infrastructure another; pricing structures on public transport a third. But the shift in behaviours they drive is likely to be a complex combination of trends happening at different speeds in different places and among different groups: there’s no reason to think that overall such an effort can’t succeed, but ascribing this or that specific outcome to a particular measure will be very difficult. This is perhaps an example of where you might most readily campaign for a positive change rather than against a particular problem-causing policy.

Over to you

I’m still somewhat attracted to the idea of a divide between “formal” and “behavioural” types of policy problem: it can be a way to analyse the situation you are addressing, and therefore to identify what needs to be done to change things. Often the trick may be in understanding the interplay between examples of each. But what do you think? Are there other fundamental categories not covered here? Are policy problems best understood in some other way entirely? Let me know in the comments.