After releasing my post on making the policy profession a real profession, I stumbled onto this report titled ‘Policymaker Perspectives on Reform’. It tackles the exact same question as my post and draws similar-ish conclusions, so I thought I’d do a mini review of it. The report is on the government’s own Public Policy Design blog and was led by the Policy Profession Unit.
The Analysis
The report was based on research with 178 policymakers. This included 25 one-to-one interviews and a series of workshops with participants from multiple departments to map what policymakers do day-to-day. They also held panels/focus groups of experts with the Institute for Government, and interviewed academics.
The bulk of the report is dedicated to the findings of this research, describing a mixture of best practice and pain points in civil service policymaking. This research doesn’t reveal anything particularly novel. It’s the same classic critiques of government: policymakers are out of touch with a diverse citizenry and aren’t user-focused, civil servants are uncertain how to navigate capital-P Politics, we aren’t good at systems thinking, those who implement or receive a service aren’t effectively consulted, we work in silos and don’t evaluate success.
The team also put together a list of all the core elements of policymaking. In my previous post, I said that this was the first step to being able to professionalise these elements. It would be good to see these taken up as different core functions to be optimised. They present them in a linear model, despite acknowledging that policymakers told them this was a simplification. This does lead to the creation of this rather confusing graph:
(As an aside, the report is very slickly designed for quite a dry subject matter. Why did this report need professional designers? Why do they have quotes over pictures of smiling policymakers, when those quotes aren’t from the people in the pictures?)
Unfortunately, the report is largely absent on proposing solutions. They propose changes for each area identified, but these changes are more objectives or ambitions than proper actions to be taken. It’s maybe a bit unfair to criticise the report for this, as it is explicitly trying to understand the problem, with a claim that putting together an action plan is the next step. The issue with this is that we already have an understanding of the problem. The report even quotes the Fulton Report of 1968, which now stands as the classic example of “we’ve known all these problems for decades, the problem is we haven’t solved them.” I’d much rather we spent time focusing on recommending and analysing solutions than restating the problem for the hundredth time.
Policy Specialisms
There is one concrete proposal, which is to split the policy profession into three specialisms. I found this particularly interesting given that I made the exact same suggestion in my earlier post, albeit without specific suggestions for what those specialisms might be. The three specialisms are:
Government Business Expert: these people understand the machinery of government such as ministers, legislation, corporate strategy. It broadly seems to cover the work of private offices, strategy teams, bill teams and most of the Cabinet Office and Treasury.
Subject Matter Expert: these people are experts in a specific policy area. They are responsible for engaging with stakeholders, building an ongoing evidence base and understanding the history of their area. They are also responsible for briefing and correspondence within their area. Most policy professionals in government fall into this bucket in theory, but they move around too much to truly become subject matter experts. There are often only one or two people with really deep knowledge in each team.
Policy Design Expert: these people are experts in the policy cycle, user-centred design, analysis, evaluation and systems thinking. Currently there are relatively few of these people in government, mainly working in teams like Policy Labs and What Works centres. Presumably the idea is there would be many more of these.
I really like this attempt to carve out three specific specialisms. They seem to map fairly well onto existing roles within the policy profession, so wouldn’t require a major upheaval to implement. They could therefore be bedded in over time, gradually providing more and more tools, guidance and best practice for how each function can be performed most effectively.
The report says that we should aim to increase the proportion of Senior Civil Servants from each specialism. Presumably by ‘increase’ they mean something like ‘balance’ given these specialisms don’t exist yet. But also the use of the word ‘increase’ hints at the main problem with this goal. It implicitly gives away that the authors realise the SCS is dominated by those who loosely fall under “Government Business Expert”. Strategy teams, private offices and central departments are the place to make a career in the civil service. Simply stating that we should aim for equal numbers of SCS from each specialism ignores that the people who make the most significant decisions and work most closely with ministers mainly fall into one specialism. This isn’t just a prejudice. What SCS do is arrange resources, work with ministers and coordinate across government. A specialism aimed at this is a specialism aimed at designing senior leaders.
It still makes a lot of sense to start to professionalise these individual specialisms. Working in a strategy team is often very different to doing policy design, so teaching the most influential people that their work revolves around the policy cycle does little to improve organisational management. I actually think it would benefit the policy profession to identify the most capable future leaders early on and train them rigorously in the sort of work a senior manager does. It might be beneficial to require that people have experience (1 or 2 years) in each specialism before making the SCS as a policy professional, helping leaders have a broad understanding of policy. That could encourage more hopping around, however, which could be counterproductive. So generally I like the Government Business Expert specialism, but for very different reasons to the report authors. They want a nice mixture of different specialists in leadership positions, whereas I like the fact that this might lead to a professionalisation of a policymaking elite.
Similarly, a fusing of experts in specific policy areas with experts in making great policy intuitively feels like a powerful combination. The success of Policy Labs working on this model in a smaller form suggests it could work at a larger level. Although we would need to see efficiency improvements in the subject-focused policy teams to allow us to recruit more policy design experts, assuming the Treasury aren’t going to be splashing the cash anytime soon. I’d see this efficiency improvement as a key test of any implementation of a model like this.
Service design vs system architect policy
One area the report slips, which I consider a classic blind spot of many policy innovation efforts, is the role of policymakers as system architects rather than service designers. Much of the focus is on service users and inclusiveness. This makes perfect sense if you’re designing policy for service design such as schools, prisons, or hospitals. But it makes less sense if you work in areas that don’t provide services but seek to alter society on a system level. I would argue that BEIS primarily works as a system architect rather than a service designer. On climate and energy you can benefit from an ‘end user’ lens (e.g. property owners on heating, travellers on transport) but it’s fundamentally a problem of the economic arrangement of society. On the economy, there is no service user - the objective is just to stimulate growth for the benefit of all. Seeing businesses as the service user would be a very partial perspective on the problem of economic growth. The best unit of analysis in systems is not users, but leverage points.
The system architect shares some of the skills from both the Government Business Expert (coordinating policies, breaking down silos) and the Policy Design Expert (analyse complex systems, bring together disciplines). Solving these problems also relies on deep subject matter expertise. But it isn’t just a fusion of the others. It requires a different approach to both understanding the problem and implementing a solution, as government only has the power to intervene in social systems at certain points, whereas it more or less controls everything in service delivery.
The report does acknowledge some of these systemic problems, in the “tackle grand challenges section” (p.55-58). The solutions proposed to this are to break down silos and encourage more multi-disciplinary teams. These are obviously good things, but I think the report misses how fundamentally different policymaking of this kind is to service design. As I said, this mistake is common amongst the policy innovation community, which draws its methods from service-focused experts such as business, digital technologists and policy makers in local government and other service-focused parts of government. The system architect form of policymaking represents a much smaller number of people than service design, but these people tend to be those who make the most impactful decisions (for better and worse).
I may explore the contrast between the system architect and service designer in more detail in a future post, as I don’t think I’ve adequately managed to explain it here.
What next?
Perhaps the most enticing part of the report is the timeline for what happens next. The report is dated October 2020 and there’s a ‘proposed delivery schedule’ on slide 36. This says that the plan for November was to develop plans for each public body, build a central bank of ideas for actions and recruit a scaled team to oversee delivery. Then in Spring 2021, they would “design, test, build and deliver” interventions.
I say this is the most enticing because I haven’t heard of this. If anyone knows the status of this, I’d be very interested in hearing about it. My guess is that this work was not prioritised. I think one reason for this could be that the report is good on analysing problems, OK on suggesting objectives and virtually non-existent on an actual implementation plan. If I was a senior decision-maker asked to kick off a major reform package off the back of this report, I would have a million questions about how it’s actually supposed to get done, how it fits into wider civil service reform and what the ask and resource burden on departments and public bodies is.
This report also faces another implementation problem. The Policy Profession Unit sits within the Department for Education, I believe for historical reasons as Jonathan Slater, ex-Permanent Secretary at the DfE, used to be the head of the policy profession. That role now belongs to Tamara Finkelstein at Defra. The idea of having an individual Permanent Secretary be the head of the policy profession makes some sense. In its current state as mainly a guidance and advisory function, the policy profession unit can operate within DfE effectively. However, if the civil service was to get serious about policy profession reform, it couldn’t happen from DfE, with the Defra permanent secretary pushing. Reform like this can only come from the centre, i.e. the Cabinet Office, with significant backing from the civil service chief operating officer, the Cabinet Secretary and ministers.
If I was trying to take this work forward, I would severely curtail the ambition from civil-service-wide reform. The proposals are too speculative to really suggest it would be a good idea to push for change across government. Instead, it would be best to try and implement this approach on a much smaller scale, perhaps within a single group. (For those who don’t know, a ‘group’ is led by a Director General, the second-highest post in the civil service. A group would have a few hundred or at most a couple thousand policymakers). A smaller team wouldn’t gain the benefits of enough specialisation, nor demonstrate how the approach works at an organisational scale. You would probably need permission to act within a sort of ‘sandbox’ free from some of the typical civil service HR rules.
Take-aways
Overall I’m happy to have found this report. It’s good to know the policy profession unit are grappling seriously with the same problems I see at a working level. Despite poking some holes in it, I think it presents a good foundation on which a professionalisation of policymaking could build. But I can’t help suspecting that the policy profession unit has written the same report every 2-3 years since its inception. I’m really interested to know if they ever managed to go beyond this towards their desire to start proposing and implementing solutions.
In some sense you can turn this on its head slightly, its not necessarily you need other professions in the SCS, you need more flex in pay scales so that you can retain the people who are have the other necessary skills/expertise in their policy area rather than chasing promotion elsewhere to eventually become a Government Business Expert. This is really acute in some areas like digital skills.
I would go one further and say we need a rethink about team structure in CS, another factor is boredom with a policy area, without much opportunity to develop and evolve your portfolio your only option is to move. In consulting for example a Partner of the firm will develop the team around the project, and you can be on multiple projects at one time. What that means is that seniors go seeking expertise and requisite skills to deliver, rather than hoping they come to them through good old GRS.
Really enjoying these blog posts as a fellow BEISian! Chimes with lots of my thoughts as someone relatively new to the CS.