The civil service has long experience of turning political ends into policy means, finding safe routes for the ship of state; but now the government wants it to use a new set of navigation tools. Joshua Chambers takes soundings.
- Poll: outsourcing ‘bias’ fears
- Poll: FoI can ‘weaken policymaking’
Charting a course is a tricky task for any vessel, but it’s especially difficult for the ship of state. While a party’s manifesto (or a coalition agreement) may provide a guide to its intended destination, the precise route it takes is shaped by factors such as economic forecasts, the tides of public opinion, the unity (or disunity) of the crew, and advice from the navigators.
Traditionally, governments set out their intended destinations, and then experienced civil service navigators advise on appropriate routes, journey times and potential obstacles. However, in recent years policymaking has become both more open – with a greater number of opinions to take on board before setting sail – and more constrained, subjected to a range of mandatory impact assessment tests. Since 2010, the coalition has also tried to replace the civil service’s traditional route-planning tools with a new set of devices – something that’s received a mixed reaction.
In the wake of the government’s Civil Service Reform Plan (CSRP), which brought yet more alterations to the policymaking process, Civil Service World has sought the views of HMS Whitehall on all these changes. Last month we ran an online opinion poll, asking civil servants questions on their progress in introducing new policymaking techniques; their departments’ capability to perform consultations; the value of impact assessments; the outsourcing of policy advice; career progression for non-policymakers; and the impact of the Freedom of Information Act on policy discussions.
We received responses from 1053 civil servants, of whom 429 were in senior grades (SCS or grades six and seven) and 286 were in the policymaking profession. As the government tries to implement its plans for policymaking, this Special Report provides a clear picture of what civil servants make of them.
The coalition government has already introduced some radical new approaches – such as the drives for transparency and localism – designed to change how policies are designed and the public sector operates. A previous CSW survey of civil servants revealed their opinions on the value of each of these agendas (see CSW 16 May, p15). Now, CSW has sought to assess the government’s progress in introducing and embedding these approaches (see Figure A).
The field of reform in which most respondents gave a broadly positive response on the pace of progress was “the move to achieve policy objectives by using softer methods around persuasion, ‘nudge techniques’ and stakeholder engagement, rather than regulation and legislation” (see ‘Key’, bottom right, for the answers respondents could give: we refer to the green answers as positive, and red and orange as negative).
In our previous survey, 58 per cent of civil servants broadly approved of this reform, with 64 per cent of senior grades responding favourably. This time, 59 per cent gave positive responses on progress in this field; 72 per cent of senior grades agreed, as did 80 per cent of policy professionals. This policy is clearly becoming well-embedded in the civil service.
Jill Rutter, policymaking programme director at the Institute for Government, believes that this positivity is linked to the government’s austerity drive and its squeeze on regulation. Deprived of their traditional policy tools, she says, civil servants are embracing new ways to achieve policy goals: “If you’ve got no money to spend, and you’ve got a whole plethora of hoops to jump through to get regulation in place, [then] there are plenty of incentives to get nudges in place.”
Heywood:'Huge progress' has been made on transparency - Credit: Mark Weeks
Among the departments for which we had a sample size of more than 50 respondents, civil servants in the Department for Business, Innovation and Skills were the most positive on this topic, with 85 per cent giving positive answers and only seven per cent negative ones. The department has responsibilities in reducing regulation, in part via the Red Tape Challenge, and its staff are at the forefront of the drive to find alternative methods of changing behaviours.
The answers are clear
The government is also making good progress in its “move to use transparency and citizen choice at the local level to improve service quality, rather than relying on centralised targets and reporting systems.” Asked how far their organisations have changed their approach to conform with this policy, 47 per cent of civil servants gave positive responses, while only 25 per cent gave negative answers (though the proportion who said that “we have not begun to adopt this approach at all”, at 14 per cent, was well above average). Our previous survey found strong support for the idea in principle, with 65 per cent of civil servants giving broad backing to the transparency agenda, and this support appears to be translating into progress on the ground.
Senior grades are better placed than their junior colleagues to assess progress, because they will have had to reform their processes and ensure workforce compliance. And they were even more positive about movement in this area: 58 per cent gave positive answers, and 18 per cent negative ones. It should be noted, though, that senior grades were more positive than their junior colleagues about the progress being made in introducing every one of these approaches.
Among policy professionals, a whopping 61 per cent were positive about the progress being made – with 15 per cent stating that “we have fully adopted this approach and, where appropriate, it now underpins our policymaking.”
Responding to these findings, cabinet secretary Jeremy Heywood told CSW that “we’ve made huge progress in the last few years on the transparency agenda, and there are many examples of where it’s improving the quality of accountability – helping to hold civil servants to account, spawning new businesses and giving more power to citizens.”
When civil servants were asked about “the introduction of mandatory, centralised frameworks designed to drive down costs in civil service operations, rather than allowing departments and agencies to retain full control,” more were positive than negative about the amount of progress being made. Indeed, among senior grades an absolute majority – 55 per cent – gave positive responses. And almost no corner of the civil service has escaped this agenda: across the grades, a smaller proportion of people said “we have not begun to adopt this approach at all” than on any other question.
Ian Watmore, the former Cabinet Office chief who introduced many of these frameworks, tells CSW that he’s “pleased” that “civil servants of all levels have grasped the government’s agenda to implement savings in ways which protect the front line. The spending controls enabled them to achieve this.”
Outsourcing out of favour
There was less positivity about progress on “the move to open up public service delivery to the private and voluntary sectors, rather than maintaining government as the default provider of public services,” with 42 per cent of respondents giving positive answers. And here 38 per cent highlighted “widespread concerns”, “widespread opposition” or a complete lack of progress, making this the field in which the largest proportion of people said progress has been slow or nonexistent.
Here, slow progress and internal concerns are understandable, says Dave Penman, general secretary of the FDA trade union. “I don’t think it’s the civil service getting in the way and deliberately trying to slow it down; it’s that these are complex areas of government and [the reforms] are not simple things to do. That’s why prime ministers end up [coming out with] ‘enemies of enterprise’ comments: because they think it should be easy [to change public service delivery] and it never is.’”
Finally, the question on which the smallest proportion of respondents gave positive answers was “the localism agenda, under which control over service delivery should shift from central down to local level.” Only 40 per cent of all civil servants – and 55 per cent of all policy professionals – gave broadly positive responses on progress in implementing this approach within their organisation. This finding strikes a stark contrast with the answers on this topic in our previous survey, when 59 per cent of all civil servants said that “it’s a great idea” or “it’s a good idea”.
The impact of assessments
After telling us what progress has been made on introducing these new navigational tools, civil servants were asked their views of a set of policymaking instruments that proliferated under the last government: impact assessments. Introduced to ensure that all of a policy’s likely effects have been thought through before it’s implemented, these have spread to cover a wide range of considerations including environmental, health and equalities impacts.
Heywood believes they sometimes unjustifiably impede policymaking, and that some are “just routine pushing paper up the line” – as he told a press briefing when launching the CSRP. So the government is now scrutinising all existing impact assessments, with a view to streamlining the policymaking process.
Civil servants are often seen as big fans of process and bureaucracy, but our findings show widespread perceptions that impact assessments are onerous and time-consuming, along with scepticism over their value. When civil servants were asked their opinion of different types of impact assessments, the most common answer was always that they “create a lot of work and rarely produce valuable data improving the quality of policy”.
Watmore:Civil servants have grasped the savings agenda - Credit: Niklas Halle'n/Photoshot
For each of the existing tests, respondents could tell us both whether it produces “a lot of work” or “little work”, and whether it “often” or “rarely” produces valuable data: subtracting the negative answers from positive ones in each case, we’ve plotted civil servants’ views of each test on a graph (see Figure B) – and civil servants are plainly hostile to impact assessments.
The impact test seen as least valuable was the rural proofing impact test. Of those with experience of the test, 61 per cent said that it “creates a lot of work”, against 39 per cent saying that it creates “little work”; meanwhile, just 29 per cent said that it “often produces valuable data”, against 71 saying that it “rarely” does so.
Responding to these concerns, Heywood says: “I know an awful lot of civil servants are frustrated about the speed at which policymaking is able to take place. We tie ourselves up in bureaucratic hoops.” However, he adds that “we’ve already taken significant steps to streamline the requirements.” In particular, Heywood highlights reforms to the Regulatory Impact Assessment; a document that civil servants had to complete even when they were deregulating.
In reforming impact assessments, it’s important not to “throw the baby out with the bathwater” and start “making policy on the hoof,” Heywood comments. “A lot of these impact assessments are important in order to make sure that the policymaking process has considered the key issues,” he says. “What we need to do in [coming] weeks and months is look at these impact assessments and in which cases they are valuable.”
Impact assessments should ensure that the full consequences of a policy proposal are thought through at the planning stage. However, Rutter says that, in her experience, “the real danger is that people use them at the wrong time: they tend to be retrofitted after a policy is agreed and everyone is signed up.” Penman agrees, saying that the FDA has particular experience here: it’s how the government applied the tests to its public sector pensions plans. “Equality impact assessments are little more than a box-ticking exercise done in retrospect, rather than something which provides robust and quality information that shapes policy,” he says.
Heywood disagrees, however. While he acknowledges that there have “occasionally” been policies whose wider impacts were only considered at the end of the process, he says, in recent times impact assessments “have become more embedded and that’s less of a problem.” Furthermore, he says, often “people do consider the relevant issues as they provide their policy advice, but they don’t complete the actual document until the end of the process.”
Consulting on a new course
Impact assessments are one way of ensuring that potential pitfalls have been thought through. However, these are internal mechanisms: the equivalent of a crew member taking soundings with a plumb line to spot any submerged reefs in the ship’s path. And there are other dangers at sea: policies need to recognise and incorporate the views of key stakeholders, or the ship of state can end up colliding with other people’s vessels.
Providing a kind of radar that can plot key stakeholders’ views on a potential policy, the government uses consultations – and its approach to them is developing. To keep up with changes in technology, society and government policy, civil servants are being asked to use the web more, to run more interactive consultations and to gather views through other organisations. But capabilities here are limited: in the last CSW survey, 34 per cent of civil servants cited “a dependence on traditional means of consultation, and a lack of expertise in alternative methods” as one of the three greatest obstacles that “restrict your ability to involve external stakeholders in policy development.” This time round, CSW asked civil servants: “To what extent does your organisation have the skills, systems and capabilities to conduct” a specific set of consultation types.
Some 78 per cent of civil servants were positive about their ability to conduct traditional written consultations, though even in this case 18 per cent of that total said their department “needs to develop systems and expertise” (See Figure C: here, we use ‘positive’ to cover the green and yellow answers). The figures were lower for online consultations: 59 per cent were positive about their organisation’s ability to conduct web-based consultations on its own website; 57 per cent said the same of email consultations; and 52 per cent of email surveys. The real problems lie with conducting web-based consultations via external bodies such as social media websites (where 40 per cent were broadly positive), and with gathering opinions via text messaging (38 per cent) – two new forms of consultation the government is keen to master.
A new group of navigators
The CSRP sets out the aim of establishing “an institute that can test and trial approaches and assess what works in major social policy areas.” And that broad aim remains the same, Heywood explains, although the specifics have changed a little since June. “The idea of having a single national institute is something we’ve discussed and is still an idea that is floated by some people,” he says. “My instinct is that you’re better off with a next step: a number of centres of excellence that can be fairly small, frankly.” These centres would be “evaluating experience of previous policies and becoming expert [in that area], whether it’s education, early years, crime reduction and so on, rather than just having one institution trying to cover that ground.”
While Heywood emphasises that discussions are ongoing – ministers haven’t yet made a decision – CSW has asked civil servants for their view on the concept (see Figure D). The response was less than enthusiastic, with only 13 per cent saying “the research produced could be of huge value, strengthening evidence-based policymaking.” All other civil servants raised concerns.
The two most popular answers accepted that the research could prove useful, but said that it “is unlikely to win out when it conflicts with ministers’ prejudices about effective policies” (chosen by 39 per cent) or that “at a time of falling budgets this is a questionable way to spend money” (20 per cent); a further ten per cent worried that lessons would not be transferrable. Professor David Nutt, a former government chief drug adviser – sacked by Labour home secretary Alan Johnson for straying from the party line – tells CSW: “I welcome efforts to introduce more objective evidence into the policymaking process. However, the results of this survey indicate that we have a long way to go before policy can truly be said to be evidence-based”.
Another person sympathetic to this view is Julian McCrae, director of research at the Institute for Government and previously one of the deputy directors at the Prime Minister’s Strategy Unit. At a seminar held recently by CSW and IT services company HP to discuss better use of data, he said that in government “we did things incredibly quickly if ministers were really interested in them, and we’d cut any corners necessary to make sure we got the right ministerial outcome, when those corners were: how do you run a rigorous, properly evidence-based policy machine?”
However, McCrae’s IfG colleague, Jill Rutter, says that while some of an institute’s recommendations are likely to clash with ministers’ prejudices, there are a whole range of policies where ministers are interested in technocratic solutions.
Nutt:Welcomes efforts to introduce more objective evidence - Credit: Tim Ireland/PA Archives/Press Association Images
Heywood combines his response to both comments, stating that the institute will help achieve better value for money across government – and that in the context of squeezed budgets, ministers will be unable to ignore many of its recommendations. “We are looking at entrenching this concept that all the evidence is known before a policy is advanced and put forwards to ministers because, at a time when budgets are tight, it’s absolutely essential that we don’t waste money if evidence suggests that those programmes aren’t offering value for money,” he says. “I don’t see any reason at all why ministers would set aside that evidence at a time of fiscal stringency.”
Outsourcing gone overboard?
Heywood’s idea for institutes assessing policies is just one of the new approaches being pursued. The government has also established a central match fund, worth £1m a year over three years, that ministers can use to commission policy advice outside government.
The first tender is for an investigation of civil service accountability, with a particular focus on the New Zealand model. Cabinet Office minister Francis Maude explained to CSW last month (see 8 August, p3 and p4) that the move will provide “a fresh, objective look where no-one can say that the work has been done in a way that might be self-serving.”
Elvidge:Policymaking has been opening up for many years - Credit: Jethro Sheppard
The opening up of government policymaking isn’t new, however. “The proposition about funding it from government brings a new dimension, but the idea that the existence of alternative sources of advice is either new or undernourished strikes me as one that flies in the face of the facts,” says Sir John Elvidge, the head of the Scottish Civil Service 2003-‘10. “Where has anyone who thinks that the civil service has had a monopoly on policymaking been since the late 1980s?” Elvidge adds that “all this rather has the flavour of a set of propositions by someone who started from some perceptions of what the issues and solutions were, and has been less than rigorous in looking at the evidence.”
Asked their views on this proposal, civil servants were broadly split (see Figure E). While more were fiercely opposed (15 per cent) than wholeheartedly supportive (eight per cent), the numbers expressing cautious opposition or support were broadly similar, at 37 and 34 per cent respectively (see news, p3).
Our graphic reveals both the concerns of both those broadly backing and broadly opposing the idea, and any merits perceived by both sides. The greatest concern was that “ministers may seek to commission policy advice from think tanks or academics known to share their own political views, leading to unchallenged bias in the policymaking process.” Indeed, asked to name a risk attached to the initiative, 42 per cent of those sympathetic to the principle gave this answer; almost as many as the 44 per cent of opponents who raised this concern.
Penman says that this fear is justified, because “you don’t have an alternative civil service out there that is politically neutral. Most think tanks have some sort of political leaning; they don’t have the same robust neutrality that the civil service has.” He adds that “it’s not that these organisations have no place in policy advice, it’s that [when outsourcing the policymaking process] you’re giving them control.” Rutter also warns that handing a single tender to one organisation is a “very limited model” and “there is the question: are you pre-judging the answer when you’re commissioning?”
Heywood responds that fears of unchallenged bias are “significantly exaggerated,” adding that “I don’t want this to lead to less challenging policy advice. If anything, it’s the other way around; we want to broaden the range of advice that ministers have access to and increase the degree of contestability.”
There have been suggestions in the press that the the tender, which allows just two months between the bid deadline and submitting findings, is too short to allow in-depth research from first principles; many people suspect that the selected organisation will simply go with its ideological gut instincts. The cabinet secretary dismisses the concern, though, stating that “ministers expect to get advice reasonably quickly, even on complicated topics, so [think tanks] have got to show that they can deliver robust, quality policy advice. If they aren’t able to do that, they shouldn’t apply to do the work.”
While the advice of civil service navigators is no longer exclusive, policymakers are still thought to enjoy privileged status amongst the ship of state’s crew members, and to find it easier to get a job on the bridge.
However, the CSRP attempts to redress this balance, noting that “it will be increasingly important for departmental senior leaders, especially in the main delivery departments, to have exposure and experience outside policy development.” The plan says government will “establish the expectation that permanent secretaries appointed to the main delivery departments will have had at least two years experience in a commercial or operational role,” and that the civil service will “move over time towards a position where there is a more equal balance between those departmental permanent secretaries who have had a career primarily in operational management, and those whose career has been primarily in policy advice and development.”
CSW asked civil servants whether they support efforts to encourage the appointment of more specialist professionals, such as those in IT and finance, to top civil service jobs (see Figure F). While only three per cent said that “policymakers are generally the best people to appoint to the top jobs,” and the findings suggest broad support for government’s direction of travel, 23 per cent of civil servants – and 30 per cent of the policy profession – said that “the government’s current plans may put policymakers at an unfair disadvantage in recruitment processes.”
Elvidge dismisses these concerns, however. “If and when we ever see a time when policymakers are at some actual disadvantage relative to others, we can think about how one rebalances that, but since in many parts of the civil service there’s still a gap to close with the value put on other skills, I think it’s premature to be worrying about that now.”
Heywood, meanwhile, moves to reassure policymakers. “I don’t think any policy specialists should worry too much that [reforms] are going to count against them in the future,” he says. “We need to get away from thinking of policymakers as generalists and others as specialists. I think policymaking is a specialism. What we’re interested in is rounded leaders.”
The cabinet secretary adds that the gap is already being closed, pointing to last month’s appointment of Jon Thompson as permanent secretary of the Ministry of Defence. Thompson was previously a government finance director, and also has a background in local government.
The constraints of freedom
As government debates policies internally, there is another challenge it faces: the Freedom of Information Act. The former cabinet secretary, Sir Gus O’Donnell, has suggested that the Act is deterring officials, advisers and ministers from having full and open discussions within formal policymaking processes. He argues that civil servants fear that their most candid policy advice could be made public after the government has decided on a particular policy, embarrassing ministers and causing damage to policies already being implemented.
CSW therefore asked civil servants whether they believe that the policymaking process is being damaged by efforts to evade the Freedom of Information Act – either because people hold discussions informally, or because they avoid raising potentially damaging criticisms. In response, 48 per cent of all civil servants – and 56 per cent of policy professionals – argued that Freedom of Information (FoI) is damaging the flow of open and honest advice (see Figure G).
Graham Smith is director of freedom of information in the Information Commissioner’s Office (ICO), and says the figures are consistent with the messages that civil servants have been giving about FoI. However, he adds that “nobody underestimates the inconvenience of FoI; the question is whether the benefits of openness and transparency outweigh the inconvenience.”
Smith points to the 34 per cent of civil servants who said that FoI is having a positive net effect, and says the findings “indicate that people have different attitudes towards openness and secrecy. We’ve still got a long way to go to achieve a real culture change where people recognise the benefits of transparency and the fact that there is a real and genuine public interest in understanding how policy is formulated.”
The key problem is that of confidence, Smith believes. The Commons’ Justice Select Committee has just published a post-legislative scrutiny report into FoI, arguing that while there isn’t enough evidence to suggest that FoI is having a deleterious effect on policymaking, there is a lack of case law and clarity about what discussions are protected and what advice must be published. Smith warns that FoI’s negative effect on policymaking could become a self-fulfilling prophesy, because as civil servants hear warnings that it can foster informal policy discussions, they will be tempted to follow the same course themselves.
Why are informal communications bad for policymaking? Elvidge explains that “it’s a truism about oral exchanges that [people are] more likely to misunderstand things than in written exchanges. People simply hear the contents of a conversation differently.”
Responding to the concerns of civil servants, Heywood says that “there’s a perception and a substance issue here.” He accepts that civil servants perceive FoI to be problematic, but “the underlying reality of it is that we can usually carry on giving ministers strong, confidential policy advice in writing without having to go through informal channels – and we should keep doing that.” Ultimately, Heywood adds, it will fall to ministers to make a judgement on whether the FoI Act should be amended.
On the horizon
Our survey shows that many measures that once seemed controversial and complicated are proceeding apace. Spending controls are acting across government; transparency has made huge progress; and behavioural economics is starting to underpin policymaking around Whitehall. Some of government’s more ambitious reforms are proving slower to implement, however – particularly the localism agenda and the move to deliver more services through third parties. The political realities of coalition government are likely to have had an effect on this latter policy: the Conservatives are keener here than their Liberal Democrat colleagues.
Smith:Sees a public interest in understanding policymaking - Credit: ICO
Overall, Heywood says he’s “really pleased to see this report on progress on many of the policies that we’ve been putting in place over the past few years.” He’ll also have been quietly heartened to see civil service support for his plans to streamline impact assessments, while government should be encouraged that – broadly – it has the capability to consult outside bodies on policymaking. However, civil servants have also pointed to some real concerns – including on FoI and the outsourcing of policymaking.
Government can feel fairly satisfied with our findings; but if its intention really is to ensure that policymaking processes lead to better, more effective policies, it would be wise to listen carefully to the concerns of those highly-skilled navigators charged with turning ministers’ ambitions into actions. ?
Civil Service World emailed 26,000 civil servants seeking responses to our survey, which was open from 10 to 21 August. It was completed by 1,053 officials