Part 2: Tackling the challenges faced by the fast stream

“The fast stream needs to decide what is its purpose. Do they want a future generation of senior civil servants or people who are really good in their specialism?” – Fast streamer survey respondent

The fast stream has changed substantially in recent decades. But there is very little publicly available evidence on how these changes have affected the scheme in practice – and, importantly, whether they have made it more effective in supporting the civil service’s capability. Outside a restricted number of civil service leaders, there is a limited understanding of some of the real challenges faced by the programme – from fast streamers struggling to find Grade 7 roles at the end of the scheme to others exiting the scheme early. As a result, the ongoing public conversation around the scheme is less constructive than it could be.

Our research seeks to close this evidence gap. It also offers an emerging sense of the reaction to the latest wave of reforms (which came into effect in 2024), though of course it is too early to assess their impact.

Our findings reveal a recognition that several aspects of the fast stream have improved – including pay, the learning and development offer, and diversity. But they also show that long-standing concerns, especially around the quality of postings, line management and (re)location, remain. The underlying theme is a sense that the fast stream lacks a clear sense of direction – and that it is trying to fulfil several competing objectives at the same time.

The fast stream needs a clear purpose to inform the scheme’s design. Without it, future reform efforts will amount to little more than tinkering around the edges, and some of the recurring challenges will persist, meaning it will under-deliver for the civil service while failing to adequately support fast streamers.

The buy-in of civil service leaders – the head of the civil service, alongside the chief people officer and permanent secretaries – will be crucial. The question of how future generations of civil service leaders are trained and developed is – and should be – a key strategic workforce concern for civil service leadership.

In the rest of this paper we set out some of the main challenges facing the fast stream highlighted during our research, and offer recommendations for how these can be tackled.

The purpose and size of the scheme

Our research indicates the fast stream is currently trying to fulfil three different functions. It is a leadership development programme but also – in practice – a steady stream of new resource, helping departments plug workforce gaps. And more recently it has become a pipeline for developing the next generation of in-house specialists, supporting the professionalisation of the civil service.

Synergies can sometimes be found between these functions. But the civil service has layered them together without a clear sense of prioritisation – or of what, all things considered, the fast stream is ultimately ‘for’.

The fast stream needs to refine – or re-find – its purpose

The fast stream has long been understood as a leadership development programme. Its central pitch – ‘Grow like nowhere else’ ¹ Civil Service Fast Stream, ‘The Civil Service Fast Stream’ Civil Service Careers, (no date) retrieved 15 August 2025, www.civil-service-careers.gov.uk/fast-stream/  – is accelerated career progression for high-fliers, especially those early in their career. But the changing landscape of civil service recruitment has meant that the fast stream has increasingly been made to fulfil other functions.

Our research suggests departments have been using the programme as a way to plug a variety of resourcing gaps, sometimes with limited regard to the fast stream’s function as a development scheme. This has been particularly the case during recruitment freezes – imposed, for instance, over the 2010s due to fiscal constraints – meaning that fast streamer resource (often exempted from the freezes) became even more important. As the number of appointments to the scheme grew over time, it became harder to offer fast streamers the stretching work and tailored support that was promised.

With attempts to advertise other pathways into the civil service often weak or ineffective, fast stream outreach efforts have also inadvertently created the impression that it is the main (or even only) graduate entry route into the civil service. In some ways, the fast stream has started to function as the primary ‘way in’ for graduates, and less as a bespoke leadership programme.

The 2024 fast stream reform programme has added a new layer of complexity. Under the banner of creating a ‘more skilled civil service’ ² Modernisation and Reform Unit, ‘A Skilled Civil Service: Reforming the Fast Stream’, 5 November 2023, https://moderncivilservice.blog.gov.uk/2023/11/05/a-skilled-civil-service-reforming-the-fast-stream/  the initiative has led to greater alignment between fast stream schemes and the professions. This is a positive development. But introducing a strong professional element to a mass graduate recruitment programme and doing so for a leadership development scheme are two different things – and require different approaches to, for instance, recruitment, or the training offer. The fast stream’s switching focus between leadership development and mass recruitment has made professional alignment a trickier task than it might have otherwise been.

This has confused the fast stream’s operating model, because the purpose of the scheme will inevitably have downstream implications, shaping what, for instance, the application process, or the L&D offer, should look like. Indeed, wherever tensions do arise between the various functions the fast stream intends to fulfil, they are difficult to resolve absent a very clear sense of what, at its core, the fast stream is ‘for’. In general, rather than resolving them, a sub-par balance is struck, which serves no one function well.

Civil service leaders need to decide what the scheme should be aiming for and clearly set out how that fits within the wider ecosystem of civil service recruitment.

The fast stream should return to being an early-career leadership development scheme

Our view is that the fast stream’s core objective should be to act as a pipeline for the most talented graduates to get on an upward professional trajectory and into the Senior Civil Service (SCS).

To some extent, this ambition has always existed at the heart of the fast stream. The justification behind having a ‘fast’ stream in the first place is that the civil service, like other organisations running graduate schemes, has an interest in actively and strategically shaping the development of the next generations of its most senior leaders. The existence of such a programme helps the civil service ensure high potential recruits understand the organisational environment they work in and develop the skills and work styles needed by the civil service leadership of the future.

But our research revealed that, in recent years, the fast stream’s ambition and focus on preparing early career officials for the most senior roles has been diminished – and the programme is now understood by many to only be a pipeline into middle management roles (Grade 6 and 7), with limited expectation placed on fast streamers to progress further.

Fast streamers should of course not take progression all the way to the SCS for granted. The scheme should set out to identify those with leadership potential and offer them an intensive programme of highly tailored development support, equipping them for the journey to the top, rather than bestowing leadership roles upon them by default. Fast streamers should also not automatically progress into other civil service leadership development programmes (e.g. the Future Leaders Scheme) and decisions on the speed and nature of progression beyond the fast stream must be based on merit.

What we propose is an explicit expectation that fast streamers will be able to leverage the support they receive on the scheme to reach the SCS, rather than a ‘right’ to get there. To this end, the civil service should think strategically about what it wants its future pool of leaders to look like and, in reforming the fast stream, plan for the next 20 or 30 years.

For one, the fast stream’s success in attracting a (socio-demographically and cognitively) diverse pool of candidates makes it an effective tool for ensuring that the SCS is (more) reflective of the public it serves. So setting up fast stream recruitment in a way that ensures candidates from all backgrounds have a fair chance is crucial. 

The civil service has also increasingly recognised the importance of officials being embedded in professional communities, and made welcome moves away from what the Institute for Government has previously described as an ‘outdated’ conception of the generalist. ³ Grama T and Thomas A, ‘The end of the civil service generalist is welcome’, Institute for Government, 6 March 2025, www.instituteforgovernment.org.uk/comment/end-of-civil-service-generalist  This principle should be applied to senior leadership, with the future leaders of the civil service coming from a wider range of professions, breaking the policy profession’s effective monopoly on the most senior roles.

One of the key functions of the fast stream should therefore be to help improve professional diversity in civil service leadership. But for this to happen it is essential for the civil service to be serious about the fast stream’s role as a springboard for future leaders, and to make substantial investments in the development of fast streamers across all schemes and professions.

This vision for the fast stream as the accelerator for high potential future senior leaders embedded in professions must be made explicit and needs to be owned by the senior civil service leaders of today. Delivering on it would mean assessing the scheme’s effectiveness based on its ability to produce future generations of civil service leaders. And key decisions about its operating model should be informed by this renewed strategic clarity.

Recommendation

The head of the civil service, working with the chief people officer and permanent secretaries, should direct that the fast stream be explicitly focused on developing future generations of senior civil servants.

The fast stream should be smaller

The first, and perhaps most important, implication of this vision for the fast stream would be rethinking the programme’s size. Focusing on high potential future leaders would mean recruiting fewer of them to start with, and being more deliberate and selective about the skills, behaviours and attributes for which the fast stream recruits.

Smaller cohorts would allow for improved quality assurance of postings by making the process of bidding for fast streamers more competitive and sending the implicit signal to departments that fast streamers cannot merely be used to plug resourcing gaps (see below). It would also allow for greater tailoring of support to fast streamer needs and goals. The result would be a civil service with more effective control over how its future leaders are developed.

Many of our survey respondents were keen on this idea. Asked what the civil service should do to ensure it continues to attract talented applicants to the fast stream, one fast streamer told us:

“Make it more competitive. The fast stream is far, far too big. The CS needs to have a serious think about what the fast stream is for. Is it to develop future leaders and streamline them towards leadership roles (as I understand it used to be) through stretching placements and challenging experiences, or is it just the default option for graduates to join the civil service (which is what it is now).”

Another suggested that the civil service should “shrink the scheme and undertake better due diligence on fast stream roles, so that the quality of the roles as a whole is better”. When prompted on whether they felt the trade-off between smaller fast stream cohorts and more stretching, challenging postings was acceptable, participants in our focus group of former fast streamers overwhelmingly agreed.

Making the fast stream smaller will not be easy. The programme’s size is currently determined by departments’ assessments of their workforce needs, which then inform the yearly bids made to FSET for fast streamer resource. As part of a broader shift towards more strategic workforce planning, central civil service leadership should be more directive around the size of the fast stream. In gripping the size of the programme, civil service leaders’ overriding concern should be to establish what investment is necessary to develop fast streamers and set them up for progression to SCS, and make decisions about fast stream headcount on that basis.

Recommendation

The fast stream should take on fewer applicants each year.

The work fast streamers do

The quality of fast stream postings is felt to be highly variable

“Some teams do not have a clear idea of what the fast streamer will do and do not know how to get the most out of them.” – Former fast streamer survey respondent

Across our research, one of the main concerns raised by current and former fast streamers was the highly variable quality of postings on the scheme and what they described to us as “insufficient” quality assurance.

One former fast streamer argued that, to continue attracting top applicants, the fast stream should “vet the quality of postings, the job roles and take into account previous fast streamer experience of postings”. Asked whether they thought their work on the fast stream was meaningful, a current fast streamer replied that “it is so posting dependent and the posting itself, how you get matched to it, is such a Russian roulette”.

The expansion of the fast stream in the last decade or so is at least partly to blame for this. A larger number of fast stream postings makes it more likely that they will vary in quality – and more difficult for FSET to quality assure at a central level. Echoing this, one survey respondent told us:

“I believe the civil service should properly scope out whether listed fast stream roles truly do need a fast streamer. In doing so this will refine the number of posts available with lots of knock-on benefits: more competitive recruitment process for fewer spots, guaranteed meaningful postings with defined job roles and the salaries could rise with the money saved on fast streamer headcount.”

Fast streamers not enjoying a posting is, of course, not reason enough to conclude that the role has not been appropriately quality assured. Fast streamers themselves recognise this. As one focus group participant put it to us, “getting sent to do work you didn’t want to do can also be very constructive”.

But the concerns we heard from our research participants were more specific in nature. The lack of stretching work, being given very little responsibility and having limited opportunities to develop profession-specific skills were among the key concerns cited by fast streamers – all of which raise important questions about whether some fast stream roles are fit for a leadership development scheme.

FSET needs to more tightly grip the commissioning of fast stream roles

For the fast stream to prepare future generations of SCS, postings that successfully test fast streamers’ leadership potential and learning agility are essential. Robust quality assurance processes for postings are therefore necessary.

Recent reforms have been a step in the right direction. Profession leads and FSET provide two layers of quality assurance for postings, and the new online platform connecting fast streamers with FSDMs (see Part 1) helps with gathering and centralising insights about challenges fast streamers face during their postings. This new system is a welcome departure from a history of poor quality assurance, though more time is needed to assess whether it will make a tangible difference to fast streamers’ experiences.

But a fast stream that is focused on developing future civil service leaders could go even further to ensure roles on the scheme are appropriately challenging, test the right range of skills, and offer fast streamers the kind of formative early-career work experience needed to set them up for success. For this the civil service should rethink the way in which fast stream bids are processed.

As things stand, departments submit bids for fast streamers to FSET – outlining the number of fast streamers they are looking to be allocated and their breakdown by scheme – and only later, after bids are approved, set out specific roles for fast streamers to fill. This system has few in-built incentives for departments to provide the highest-quality postings for fast streamers. Though FSET can reject specific postings it deems inadequate (and the two layers of quality assurance now built into the process may help in identifying those), once it has approved departmental bids, they will have to be serviced, meaning that there is a structural limit to how far FSET can afford to go in driving up posting standards.

This tension is at least part of the explanation for why departments have in the past been able to use fast streamer resource to plug resourcing gaps in times of financial restraint. Although FSET runs the scheme, the current system through which roles are commissioned gives it less control over the nature of postings than it might otherwise have.

The bidding process should change to require departments to provide posting details up front alongside their bids. In deciding which bids to approve, FSET should also work with the professions to more clearly set out ‘pathways’ through the fast stream for each scheme, outlining in greater detail the kinds of roles that fast streamers would be expected to fill over their time on the scheme and prioritising bids that closely match those pathways. All schemes do this to some extent already – and in recent years there has been a welcome move towards greater transparency, even at the application stage, about the kinds of roles fast streamers are expected to fill over the course of the scheme. But FSET and the professions should leverage a reformed bidding process, alongside smaller cohort sizes, to be even more selective about postings and tighten the criteria for what constitutes a ‘good’ fast stream posting on each scheme.

To hold departments accountable for the quality of fast stream postings they offer, FSET should use the new internal FSDM-managed platform to systematically gather data on fast streamers’ views and experiences of their postings and feed it into the bidding approval process. This could extend all the way to roles in directorates that have in the past received consistent negative feedback from fast streamers being turned down (if corroborated by FSET follow-up scrutiny). And, conversely, positive feedback from fast streamers should lead to a presumption in favour of approving larger allocations for those directorates (if these allocations are requested).

The reformed bidding process would be especially effective coupled with smaller cohort sizes. A scheme with a restricted, centrally determined headcount will give the bidding process a more competitive edge. Although having roles available in as many departments as possible should remain a priority for FSET, there is room to reward departments that offer high-quality postings through larger allocations of fast streamers. In this way, the backdrop of fiscal restraint and external recruitment freezes currently experienced in some parts of the civil service could be effectively leveraged to improve fast streamers’ experience, rather than leading to a ‘race to the bottom’ on posting quality, as has sometimes been the case in the past.

This would require a recognition that the fast stream’s role is not to serve departments’ short-term resourcing needs, but to fulfil long-term strategic goals of the civil service. Such a shift will likely be met with strong resistance from departments. Senior sponsorship and ministerial backing from the centre will be needed to drive this forward, but the payoff will be a scheme that ensures fast streamers are assigned work befitting a leadership development scheme.

Recommendation

Departments should be required to provide posting details up front when submitting their bids for fast streamers.

Recommendation

FSET should work with the civil service professions to tighten the quality assurance criteria for postings and be more specific about the roles fast streamers will be expected to fill over the course of the programme.

Recommendation

FSET should more systematically use fast streamer feedback on their postings when processing departmental bids.

More secondments would be of value to fast streamers as well as the civil service

“One positive thing that fast stream did for me was the secondment… It’s difficult to make the business case for secondments now.” – Former fast streamer focus group participant, currently working in the civil service

As part of broader efforts to ‘rewire’ the state, the government has signalled that it wants to make the civil service more porous in part through increased use of secondments. ⁴ Cabinet Office and Georgia Gould MP, ‘Frontline workers and local communities to play crucial role in delivery of Government missions’, 16 June 2025, www.gov.uk/government/news/frontline-workers-and-localcommunities- to-play-crucial-role-in-delivery-of-government-missions  There are clear benefits to civil servants doing secondments in other tiers of government, the private sector, on the front line of public services, or in charities – for the civil service as well as for secondees’ own careers. They can help improve understanding of stakeholders, build skills that are in short supply in government as well as develop and support innovative ways of working.

Secondments have previously been available on the fast stream, but the extent to which they are the norm rather than the exception has varied widely over time and still does today, depending on the scheme. The professions now decide whether a secondment should be included as part of the ‘standard’ fast stream journey or not, and while some (e.g. Science and engineering) have made the choice to embed secondments into the programme, others (e.g. Government policy, Finance) have not.

There is a strong case for expecting the standard fast stream journey to include a secondment. If the fast stream is to develop future civil service leaders, it becomes even more important to ensure they become familiar with different sectors and ways of working early on in their career. This is especially important in the case of the Government policy scheme, which currently does not require fast streamers to do a secondment outside the civil service.

Gaining a better understanding of how other organisations in both the public and private sectors operate can be an important asset in breaking down silos, including between policy and delivery. The professions would still play a central role in quality assuring secondment opportunities, as well as deciding what sectors it would make most sense to offer secondments to, but the expectation that fast streamers should go on secondment should be set centrally.

Recommendation

FSET should require that the standard fast stream ‘journey’ includes a secondment.

Support for fast streamers

Recent changes to line management and development support have been met with scepticism by fast streamers

The quantitative findings from our survey indicated that satisfaction with line managers on the fast stream is quite high, particularly so for current cohorts. But our qualitative evidence suggests that there are specific sticking points in fast streamers’ appraisal of the line management, development support and mentorship they receive on the scheme.

The introduction of departmental rotation as a key component of the fast stream in the early 2010s brought about a recognition that talent management and mentorship cannot be left to departments alone. As a result, FSET took on a more active role through talent and development managers or, later, skills and capability managers (SCMs). The recent replacement of SCMs with FSDMs offering support to fast streamers through an online platform (see Part 1) marks a departure from this model, devolving more responsibility to activity managers on individual postings.

The new system has advantages. It reduces duplication between activity managers and FSDMs, something that was recognised as a problem by our focus group participants. Processing concerns through an online platform allows better record keeping of where issues have occurred and how they have been dealt with, and may help efforts to bring about greater consistency in their handling. Improved data gathering around fast streamers’ concerns can also help identify patterns (for instance, if multiple fast streamers raise issues at a similar stage in the scheme, or in the same department) and support improved quality assurance processes for fast stream roles.

But despite these benefits, our research revealed widespread dissatisfaction with the changes. The loss of direct and sustained contact between fast streamers and a dedicated SCM emerged as a particular sticking point. As a current fast streamer told us: “The loss of SCMs has been keenly felt by the majority of fast streamers, who feel that the quality of the scheme has declined markedly as a result.” Another respondent explained the impact of the change in greater detail:

“There is no longer a manager throughout your time on the fast stream and there is very limited handover between one line manager to the next so there is no one to talk to who knows you regarding the fast stream and whilst you can advocate for yourself, your new manager knows little about you when you arrive and can be unsure of the level of work you are capable of taking on when you arrive.”

Fast streamers unanimously called for a ‘return’ of SCMs in our focus groups. They suggested there was particular value in “that one single point of contact” and argued the fast stream “needs to have some checks and balances on … activity managers”. One recalled the support lent to them by their SCM towards the end of their scheme, which they felt was essential in helping them secure a Grade 7 role.

There is an opportunity to rethink talent management and mentorship on the fast stream

Offering tailored development support as well as matching fast streamers to appropriate Grade 7 roles as they finish the programme are both made easier when fast streamers have the chance to build a strong personal relationship with a mentor who knows them well and stays in touch for their entire time on the scheme. It is still early days, and FSET could consider bringing back something close to the old SCM model if concerns persist and no satisfactory alternative is identified. But the introduction of the online platform that can help offer guidance to fast streamers in how to navigate the scheme at the more straightforward, day-to-day level, as well as facilitating their contact with FSET, has created an opportunity to rethink talent management and mentorship on the fast stream at a more strategic level.

We think mentorship for fast streamers should not fall solely on FSET; it should instead be seen as a key component of SCS corporate responsibility. This should be done through profession heads creating and running a system matching every fast streamer with a deputy director mentor in their profession for their entire time on the programme.

As with many of our proposed changes, a slimmer fast stream would make this expectation easier to manage. It would also create a broader sense of responsibility for the scheme that goes beyond FSET, ensuring more civil service leaders are invested in its success. And it would help fast streamers build the informal cross-departmental connections required for career progression into the SCS.

Some such mentorship arrangements already exist across the civil service. And many senior civil servants are already invested in the fast stream and its success. The changes made to the SCM model have created an opportunity to formalise these arrangements and ensure they apply across the board.

Recommendation

Profession heads should match every fast streamer with a mentor at deputy director level in their profession for their entire time on the programme.

Views of fast stream L&D are mixed – but improving

“Although I understand this has been improved for the latest cohorts, the development offer I have received has been arguably worse than I would have gotten if I were not on the fast stream.” – Former fast streamer survey respondent

Support for fast streamers also comes in the form of a dedicated L&D offer. The quantitative findings from our research show that views of fast stream L&D are mixed, though our respondents were, on balance, likelier to say that the offer is ‘better’ or ‘much better’ than elsewhere. Our qualitative evidence also suggests that there is some variation in the level of satisfaction with fast stream L&D by scheme, as well as over time.

We found that, wherever it existed, the option to pursue a professional qualification that was heavily subsidised or entirely paid for by the civil service was seen as a particularly strong asset of the fast stream. One former fast streamer told us that the full master’s degree option offered to them as part of the HR scheme was an “important part of the offer”. Another said that the professional qualifications offered on the commercial scheme were “really good” and contrasted them with central fast stream L&D, which they viewed as “really unhelpful”.

The introduction of structured learning into the Government policy scheme – through the public policy modules offered in partnership with King’s College London – was regarded very favourably by those who have accessed it as well as those on the legacy Generalist scheme who did not have that opportunity – one of whom told us they “would have loved” to access it.

Some focus group participants expressed the desire for more technical training focused on hard skills. Though some found the fast stream-wide soft skills training useful (one ex-fast streamer told us: “It doesn’t feel practical at the time, it feels a bit wishy-washy, but after a while it does become useful”), others were less positive. One fast streamer on the Project delivery scheme told us they did not find the L&D offer very useful and that, for many soft skills, “either you learn it on the job and you have got it or you don’t”.

Building an L&D offer that satisfies the needs of all fast streamers is difficult, particularly for a large programme. Recent changes, tied to the increasing professionalisation of the fast stream, have been welcomed by most and are in line with the civil service’s goal to upskill its workforce. Nevertheless, a reformed fast stream focused on developing future SCS along the lines we have suggested would require a revamp of the L&D offer.

More structured learning and greater certification of skills should be part of this offer. But civil service leadership needs to also scope out and come to a view about what kinds of learning are needed to develop the leadership and management skills expected of senior civil servants in the long run, and which should therefore be received by all fast streamers.

The fast stream contract

Dissatisfaction with pay and benefits is high – but there is more nuance to this story

“The pay is bad but has always been bad, and recently got better after a union settlement, so I actually don’t think that’s the main issue.” – Former fast streamer survey respondent

Our research found that the fast stream contract – the ‘terms and conditions’ successful applicants sign up to when joining the scheme – is not perceived as fully fair or reasonable by those with experience of the programme. Pay and benefits and the location of postings emerged as two major areas of concern.

Our survey results indicated that pay and benefits were the one aspect of the fast stream with which both current and former fast streamers were, on balance, more dissatisfied than satisfied. This sentiment was largely mirrored by the views of our focus group participants.

But there is more nuance to this story. It is notable that our applicant survey respondents viewed the pay and benefits on offer most favourably, suggesting pay may not be the most important consideration for those interested in the fast stream. Indeed, good pay and benefits are less important in their career, on average, for applicants as well as current and former fast streamers than ‘making a difference’ or career progression prospects (see Part 1).

Recent changes to pay on the fast stream have also been a step in the right direction. The uplift in fast streamer pay of 6.75% by 2024/25, as well as the introduction of a London living allowance worth 8% in 2024/25,5 seem to have allayed some of fast streamers’ most pressing concerns around pay.

The remaining issues raised around pay were quite specific in nature. Many voiced a clear understanding of the financial constraints associated with work in the public sector. It was the lack of pay parity between those on the fast stream and those doing ‘mainstream’ roles at the same notional grade in the civil service that was felt to be particularly unfair. As one fast streamer put it to us:

“Whilst it may never be able to match private sector salaries, it feels like – as a bare minimum – the fast stream should match the HEO and SEO salaries in year 1 and year 2/3 of the scheme respectively, given these are the grades we are expected to be working at in those years.”

This concern is compounded by the sense that, often, fast streamers will get more responsibility than other civil servants at the same grade level – as well as by the recent trajectory of civil service pay. In the words of another fast streamer: “Whilst I am happy to accept a pay disparity to some extent in exchange for the guaranteed progression offered by the fast stream, the pay gap is only getting worse as departments offer pay increases that outpace the fast stream offer.”

The fast stream pay deal for 2025/26 was announced after our research had concluded, ⁶ Markson T, ‘Fast streamers accept 2025-26 pay award’, Civil Service World, 22 August 2025, www.civilserviceworld.com/professions/article/fast-streamers-accept-202526-pay-award  so it is not possible to assess how this was received by fast streamers and whether it has shifted the dial on satisfaction with pay on the scheme. The move from ad-hoc to annual pay reviews, and their alignment with the Cabinet Office’s annual pay remit guidance process, is positive and may prevent the gap between fast stream and ‘mainstream’ pay from widening further. But this latest deal still falls short of delivering pay alignment between fast streamers and other civil servants at the same grade level.

There is a strong case for fast stream pay alignment with ‘mainstream’ roles

While pay and benefits will always remain live issues, the particular concerns emerging around the lack of pay alignment between fast streamers and other civil servants generate distortions that undermine the effectiveness of the scheme. There is a substantial short-term incentive for fast streamers to move into ‘mainstream’ roles and drop off the scheme because of this pay disparity; one former fast streamer told us that their biggest reservation when applying was the pay and that they “left after two months for a permanent SEO role”.

Data around the number of fast streamers who leave the scheme early is not made public, but the indicative evidence we have gathered suggests that this is a problem. Indeed, among the 237 former fast streamers who responded to our survey, more than a third (85, 36%) told us they left the scheme early for another position in the civil service. The commitment from the Cabinet Office to improve data gathering around early exits from the scheme, confirmed in the context of the 2025/26 pay deal, is an important step forward in more rigorously assessing the scale of the problem.

The fast stream’s primary attraction lies in the fast-tracked career progression and additional development support it offers to fast streamers, rather than competitive pay. But many feel this ‘implicit contract’ is not being upheld and that they do not get enough tailored development support to justify the lower pay. A smaller fast stream would make such support easier to manage and deliver centrally, potentially allaying some of the concerns raised by fast streamers.

Even so, on a scheme that is focused on training up the future leaders of the civil service, it becomes even more important that any financial incentives to leave the scheme early are removed so that talent does not drain away. Benchmarking fast stream pay using median pay for civil servants at the same grade level would help solve this issue, and take some of the bite out of pay disputes on the fast stream in the future.

Recommendation

Fast stream pay should be at least equal to the median pay for civil servants at the same notional grade level.

The fast stream’s relocation expectation is a concern for applicants

“I honestly think forcing people to relocate is a huge barrier to many talented people who are slightly later in their careers and settled in one place, but would have been great for the FS.” – Fast streamer survey respondent

That fast streamers may be asked to relocate around the country has become a defining feature of the scheme since the 2013 reforms. While this may have been justifiable when the fast stream intake was overwhelmingly made up of London-based graduates and (most) postings lasted just six months, the relocation expectation today does more harm than good. Our research suggests that the system of relocation is driving talented candidates away, hindering the creation of robust talent pipelines outside of London and ultimately risks undermining the government’s Places for Growth programme.

Survey respondents who are currently on the fast stream tended to express more satisfaction around the location of their postings than former fast streamers. But it was applicants who were most likely to voice concerns. This gap suggests that the reality of relocation on the fast stream is not as stark as it may seem to those applying to the scheme. Indeed, we heard that relocation exemptions, for instance for those with caring responsibilities, or certain health conditions, work reasonably well for those to whom they are granted.

But reticence towards the prospect of moving around the country looms large for those not yet on the scheme. It is impossible to know how many talented graduates ruled out the fast stream as a career option from the start because of the relocation expectation. But even those who chose to apply despite it voiced concerns.

Asked what their biggest reservation was during the application process, one applicant told us that it was “the fact that I could potentially be placed anywhere in the UK”. Many of our research participants felt the relocation expectation weighed particularly heavily on fast streamers based outside London – and being notified of postings only eight weeks in advance also emerged as a particular sticking point. One fast streamer told us their main concern when applying was:

“The unfair pressure placed on regional fast streamers to move, particularly from Scotland. London-based fast streamers hardly ever have to move and it seems to be easier to stay where you are if already based in England. Eight weeks notice to completely pack up your life with very little support (financially and emotionally).”

The benefits of relocation, when it does happen, also seem limited. Recent reforms to the programme have made some provisions to ensure fast streamers are co-located with their team or activity managers, but civil service-wide changes in office culture raise thorny issues that are difficult to address while the relocation expectation remains in force. One focus group participant told us that, because an increasing proportion of civil service work is done remotely, “there’s no community in your office space so moving someone somewhere with no connections gives no thought to people’s welfare”. Another participant shared an anecdote of a fast streamer they knew who left the civil service entirely because they felt isolated after being relocated to the MoD office in Portsmouth, “where people couldn’t even turn their cameras on”.

And one applicant pointed out in our focus group that relocation does not improve real regional diversity on the scheme:

“If I were to be posted to Sunderland, it doesn’t change the fact that I’m from Islington… It is putting the exactly same people in different places and then they leave immediately and get the train back to London at the weekends.”

The new regional pilots are a step in the right direction – they should be made permanent

Recent changes to the fast stream have signalled a substantial shift in approach to (re)location. The 2023 intake was the first time fast streamers could opt to join one of three regional pilots – based in the Darlington Economic Campus, Yorkshire and the West Midlands. ⁷ Modernisation and Reform Unit, ‘A Skilled Civil Service: Reforming the Fast Stream’, November 2023, https:// moderncivilservice.blog.gov.uk/2023/11/05/a-skilled-civil-service-reforming-the-fast-stream  Fast streamers could opt into the pilot schemes after accepting their offers (and indeed were only informed of the possibility after confirming their place on the programme), which would guarantee that all their postings would be located in the same region.

This is a positive development. Giving more fast streamers the option to remain physically in one area of the country throughout the scheme would allay many applicants’ deeply felt concerns about being uprooted from their communities. It would also help create a pipeline for future civil service leadership that is spread out across the country and support the ambition to have more senior civil servants based outside London, which we have previously argued is key to ensuring Places for Growth marks a genuine shift away from a Whitehall-centric civil service. ⁸ Urban J, Pope T and Thomas A, Lessons from the Darlington Economic Campus for civil service relocation, Institute for Government, 9 June 2023, www.instituteforgovernment.org.uk/publication/darlington-civil-servicerelocation ⁹ Metcalfe S, ‘More support would make the Sheffield Policy Campus success truly transformative’, Institute for Government, 22 December 2024, www.instituteforgovernment.org.uk/comment/sheffield-policy-campus ¹⁰ Dunlop H, ‘The government needs a workforce plan to inform civil service relocation’, Institute for Government, May 2025, www.instituteforgovernment.org.uk/comment/government-civil-service-relocation-workforce-plan

The government has recently announced that it intends for 50% of fast stream postings to be offered outside London by 2030. ¹¹ Cabinet Office and The Rt Hon Pat McFadden MP, ‘Thousands of Civil Service roles moved out of London in latest reform to the state’, May 2025, www.gov.uk/government/news/thousands-of-civil-service-roles-moved-out-of-london-in-latest-reform-to-the-state  This goal should be accompanied by a change in how the fast stream treats relocation, scrapping the relocation expectation and embedding fast stream cohorts into the government’s major regional hubs. This would mean turning the regional pilots into a permanent feature of the scheme. 

For the new regional pathways through the fast stream to be a success, however, it is essential that the civil service doubles down on its outreach across the country and ensures that (prospective) applicants know the fast stream is no longer based only in and around London – a perception that, as we heard from a fast streamer on one of the regional pilots, remains widespread among potential applicants in many regions of the country. It would also mean going further in supporting cohorts outside London to ensure that fast streamers build the interpersonal and communal networks so central to the fast stream experience in the capital.

Recommendation

FSET should scrap the relocation expectation on the fast stream and turn the recent regional pilots into a permanent feature of the scheme.

Recruitment and outreach

Concerns persist around the fast stream application process

“The process is too long! I ended up withdrawing as I received an earlier offer with better pay.” – Applicant survey respondent

The fast stream application process has seen many changes in the last two decades. Some of this has stemmed from welcome responses to recruitment practices that were found to be problematic in terms of access – for instance, moving the application process online, or shortening the duration of the assessment centre. But our research has identified a series of specific sticking points that persist and are echoed widely by our research participants. 

One common concern raised by survey respondents was that application timelines are too long, even compared to most private sector graduate schemes; applications for the 2025 entry cohort ran from October 2024 to at least March, for postings beginning in September. This means talented candidates accept offers for other roles and drop off the application process, or are dissuaded from applying in the first place.

A further aspect of the process that was singled out in our research was the online tests that make up the initial phase of the application process. Across our focus groups there was a view that the specific tests used for fast stream recruitment, especially those centred around situational judgment, were not particularly robust – and that whether you passed the sift was down to luck as much as ability. One current fast streamer shared that, a year after joining the scheme and in trying to help a friend work through the application process, they completed mock situational judgment tests – and failed them, saying they “are not indicative of what makes you good at the job.”

A (successful) applicant due to start on the scheme in autumn 2025 suggested some psychometric tests are inadvertently exclusionary, arguing that “so much of it is about professional behaviours” and that “if you haven’t grown up with professional parents and had internships”, you are at a disadvantage. Participants in our applicant focus group also questioned the appropriateness of an application process that is largely uniform across schemes. One participant told us that they were “surprised by how generalised the application process was from the start” and how “it didn’t get scheme-specific until the final stage”.

An overhauled fast stream needs a different approach to the application process

These concerns should be taken seriously. If the length of the process or its perceived arbitrariness are putting some applicants off – or filtering out talented candidates – then that is a problem. But to assess whether the application process as it currently stands is indeed having these negative effects, recent changes should be allowed to bed in and rigorous evaluation efforts should follow. In the short term, less rather than more change would be the better option.

Looking further ahead, however, the argument for reforming the application process is stronger. As with the fast stream L&D offer (see above), clarifying the purpose of the fast stream – in our view, to prepare the SCS of the future – will require significant operational changes. Testing for leadership potential should be central to any future recruitment reforms on the fast stream.

As we have argued, a scheme centred on leadership does not need to – and should not – mean a lack of attention to the need to professionalise the civil service. Indeed, it is essential for the civil service to have an early-career leadership development scheme that recognises the importance of being embedded in a professional community from the get-go, and does not see leadership roles as the preserve of the generalist.

The ambition should be that the civil service leaders of tomorrow come closer to reflecting the professional mix of the civil service they lead. So decision makers should also consider how the profession-based dimension of the fast stream recruitment process can be bolstered. At a minimum, the final selection board – the only stage of the application process that is fully designed by the professions – should be extended to all schemes. But developing different forms of scheme-specific assessment that can be introduced at earlier stages of the application process should also be considered.

Diversity on the fast stream has improved over time – but socio-economic background remains an area of concern

Both current and former fast streamers we spoke to recognised that the demographic make-up of the scheme is today more reflective of the UK population than in the past. They were also inclined to believe that the fast stream is more diverse than other graduate schemes – though a lot chose to reserve judgement on the matter. Several of our focus group participants attributed this success at least in part to the (now cancelled) Early Diversity Internship Programme (EDIP) and Summer Diversity Internship Programme (SDIP).

Despite this, there is a lingering perception that the fast stream is ‘for’ a certain kind of applicant – a young graduate from Oxbridge, who is privately educated and comes from a higher socio-economic background. And many of our survey respondents indicated that the fear of not ‘fitting in’ was one of the main reservations they had when applying. One said they had been “worried others on the scheme would come from better backgrounds from me, so privately educated, or went to a Russell Group uni or Oxbridge”. Another shared that “the diversity of the applicants and the scheme itself was not encouraging to me as a brown woman who went to a state school for her whole life”.

But it was our applicant respondents who tended to voice this concern most clearly; as one put it to us: “I stand no chance as someone from a working-class background from a state school. Fast streamers are from Oxbridge, middle- to upper-class backgrounds, who are trained to pass fast stream assessments.” And, as noted earlier, while diversity gaps have narrowed over time, recruitment data still shows socio-economic and educational background in particular are strong predictors of success in fast stream recruitment.

If the fast stream is to become slimmer and more focused on developing future civil service leaders, then the question of diversity becomes even more pressing. A talent development programme that is successful in fast tracking the careers of those on it to the very top of the civil service but does not successfully address potential biases in recruitment – or does not do outreach well enough – may ultimately do more to compound gaps in representation, failing to achieve the socio-demographic and cognitive diversity that the civil service has many reasons to want in its most senior ranks. If future fast stream recruitment will be based on different, and newer, types of assessment zooming in on leadership potential and scheme-specific requirements – as we think it should be – then it will be essential to ensure these assessment methods yield suitably diverse fast streamer cohorts.

One important consideration in thinking about diversity on the fast stream is related to the transparency of the recruitment outcomes. The ‘diversity gaps’ between applications and recommendations for appointment for certain groups (see Part 1) raise some questions around whether there are any elements of the application process that tend to ‘filter out’ candidates from under-represented groups. FSET stopped publishing recruitment data broken down by application stage in 2012, meaning that it is difficult to assess how various measures of diversity change as candidates progress through the application process. To increase transparency, improve accountability and help assess how the application process is affecting the diversity of fast stream intakes, FSET should return to systematically publishing this data.

Recommendation

FSET should publish recruitment statistics broken down by socio-demographic characteristics at every application stage in its annual data releases.

Getting outreach right is essential to ensuring diverse fast stream cohorts

Combining merit-based recruitment of high-potential candidates with an effective strategy for boosting diversity should build on previous successes. The Summer Diversity Internship Programme (SDIP) and Early Diversity Internship Programme (EDIP) are one obvious place to start.

In 2023, the EDIP was scrapped without any clear rationale. The SDIP also became the SIP, moving away from its earlier explicit focus on diversity. While it is difficult to evaluate the impact of these changes because of limited data availability, the widening representation gaps emerging in 2024, as well as the qualitative evidence we gathered from fast streamers who were overwhelmingly positive about the internships’ contributions to their professional trajectories, raise some warning signs.

In this context, the government’s announcement that it will restrict eligibility for SIP to applicants from lower SEB starting with the 2026 cohort is a step in the right direction. ¹² Cabinet Office and the Rt Hon Pat McFadden MP, ‘Internship Scheme To Get More Working Class Students Into Civil Service’, August 2025, www.gov.uk/government/news/internship-scheme-to-get-more-working-class-students-into-civil-service  While successful in boosting representation on several other counts, the fast stream and the wider civil service have fared less well on socio-economic background (see Part 1). Focusing outreach efforts on correcting the skewed class make-up of the civil service’s future leadership talent pool is therefore justified.

The rehauled internship programme is intended to help attract more lower-income candidates to the fast stream (and the civil service more broadly) as well as to support them in overcoming the additional barriers they may face in the recruitment process. The fundamental question will be whether the scheme succeeds in fulfilling these goals in practice. ¹³ Dunlop H, Keenan H and Thomas A, ‘The socio-economic background of civil servants needs to change’, Institute for Government, 1 August 2025, www.instituteforgovernment.org.uk/comment/socio-economic-background-civil-servants-needs-change  More broadly, outside of yearly recruitment data releases, little is known publicly about fast stream outreach efforts – what they are, how many people they reach, and, most importantly, whether they work.

FSET should return to the practice of giving detailed accounts of its outreach efforts, including but going beyond the summer internship, in annual reports on the fast stream. It should also ensure that outreach efforts have time to bed in and are properly evaluated, and that the evidence resulting from these evaluations is made publicly available – and, following a ‘test and learn’ logic, acted upon.

Greater transparency would generate the kind of healthy scrutiny that is needed to ensure outreach is more than a box-ticking exercise and instead makes a tangible difference to recruitment outcomes. It would also, building on the fast stream’s totemic value in the civil service recruitment ecosystem, send a positive signal across Whitehall – around the importance of both diversity and evaluation – as well as providing an example of best practice in how to go about ensuring civil service recruitment is truly fair and merit-based.

Recommendation

FSET should report publicly on the outreach efforts it is running and the evidence around their effectiveness.

The civil service must make it clear that the fast stream is not the only way in for graduates

If the fast stream is made smaller as we suggest, but this is done in isolation from broader rethinking of civil service recruitment, there is a risk that the civil service may lose many talented prospective staff to other employers. This is because too many graduates think the fast stream is for them the principal (even perhaps the only) route into the civil service – or do not know how to navigate alternative routes.

Part of the reason for this is the power of the fast stream’s brand. But it is also likely to be related to the nature of civil service outreach efforts, which have often focused on the fast stream. There is, as a result, a lot of interest in the scheme – and an abundance of resources in the public domain trying to demystify the application process. This is not true of most civil service roles. ¹⁴ Ibid.

The civil service should rethink the role of the fast stream – but it must also make sure that it provides easily accessible, well-advertised pathways into public service for talented graduates outside of the fast stream. One important step towards this, as the Institute for Government has previously recommended, ¹⁵ Worlidge J, Urban J, Clyne R and Thomas A, 20 ways to improve the civil service, Institute for Government, 31 July 2024, www.instituteforgovernment.org.uk/publication/ways-improve-civil-service  would be to advertise all civil service roles, regardless of grade, externally by default.

Recommendation

The civil service should go further in ensuring graduate jobs outside the fast stream are well advertised, and start by advertising all civil service roles externally by default, regardless of grade.