It was hailed as a revolution at the heart of state, a means of bringing competition to the previously closed corridors of Whitehall. Instead it has brought disaffection and a host of critics.
Behind a barbed-wire perimeter fence in a low, grey building on the edge of Devizes you can find the face of the new civil service. Until October last year this dozen-acre square was a Ministry of Defence facility known as Operations West, a payroll data centre where 80 civil servants operated and maintained Ministry of Defence mainframe computers. Since then, save the odd touch, little has visibly changed. The MoD sign at the gatehouse has been replaced by a bright blue and silver plaque and the field gun that once covered its entrance has been removed and put into storage. The routines, too, are much the same. At the perimeter the guards still maintain their round-the-clock patrol and, inside, the staff, though reduced to 50, still work their shifts, operating the same computers that process the same MoD payrolls. Now, however, they do so as employees of the French-owned computer services group, Hoskyns, which, after a lengthy tendering process, has taken over the lease of the site on a five-year, multi-million pound contract. The staff, unprompted and unrehearsed, are for the most part pleased with their new employer. They are less enamoured, however, of the method that brought them together. 'It was awful,' says one. 'I wouldn't want anyone else to go through what we've been through.' What they've been through is a process known as 'market testing', which, despite the plea of Hoskyns's newly acquired employee, was last year applied to 25,000 of his former colleagues and is currently being applied to 54,000 more. It seems no more popular with them, however, than it was with him. Last November, for the first time in 12 years, 225,000 reform-weary civil servants - almost half their entire number - staged a one-day strike in protest at the policy. Throughout Whitehall, say insiders, the effect on morale has been devastating. According to John Ellis, secretary of the Council of Civil Service Unions, there is a 'massive crisis of confidence' within the civil service, the crux of which is market testing.
For a policy that has aroused such antipathy its content is neither wholly new nor are its proposals, on the face of it, highly contentious. Market testing is, if you like, the '90s face of the compulsory competitive tendering programme that swept through local government in the early '80s. This time, however, it aims to strike deeper, to take private sector practice closer to the heart of government and, in Treasury Secretary Stephen Dorrell's words, strip it back to 'an inescapable core'. This time it's not interested in refuse collectors or grass cutters but in accountants and computer operators, typists and lawyers - any discrete high-value activity, in fact, outside the core function of policy formulation. In pursuing the policy, the Government asks, with a chain of brutally simple logic, three questions of all aspects of its business: does this activity need to be done? - if not, abolish it; could it be privatised? - if yes, do so; are there strategic reasons for contracting out? - again, if yes, do so. But if not, and only then, you go for the fourth and final solution - you market test it.
The unpopularity of the latter, at least with civil servants, is readily explicable. It is, at base, a rather harsh exercise in self-justification, a matter of compelling staff to compete for their own jobs against bidders from the private sector. As laid down in Competing for Quality, the 1991 White Paper that launched the programme, when an activity is market tested the existing staff are required to demonstrate that they can deliver better, long-term value for money in providing a service than any potential rival. If successful, the work stays in-house (under a formal contract to ensure that the newly invigorated staff stick to their specified targets). If not, as in the case of Operations West, the activity is contracted out for a fixed term to whoever is judged the strongest bidder. Either way, so the theory goes, you make a substantial saving - a target 25%, hopes William Waldegrave, minister for public service and science. Ideology, he says, does not come into it: 'Services may be bought either from the private sector or within the public sector. We have no dogmatic preference for either one over the other.' The objective, he repeats, is simple: 'Fair and open competition.' For Sir Peter Kemp, former second permanent secretary to the Cabinet Office, civil service firebrand and long-time irritant of the minister for public service and science (Waldegrave sacked him, or rather asked him 'to take early retirement in the public interest'), the case for market testing is inarguable. 'If you're a public servant you have absolutely no right to be in charge of an operation that is costing more than it needs to or to be delivering a service that someone else could do better.' The case for the method is less clear cut. 'It would be hard to think of a more maladroit process than the market test,' claims Charles Cox, executive director of Hoskyns and, given his multi-million pound contract, an unexpected opponent. 'It creates confrontation. That is its nature.' Cox's views have themselves been tested by Hoskyns's experience of an arduous two-year tender for Operations West, complicated by the fact that the MoD belatedly reinstated the in-house bid after initially excluding it from the short list. The resulting farrago, in which the existing staff became so demoralised that they were eventually glad to be free of their public sector employer, only served to confirm Cox's misgivings. 'It's bloody inhumane to put people through what they've been through,' he says. 'It's something which in the private sector you go massively out of your way to avoid - to the point that in some cases the first thing people know about an outsourcing decision is on the morning when they meet their new employer. Yes, it comes as a shock but within hours you can deal with that shock - within hours, not two years.' Aside from protracted uncertainty there is also, he says, the disruption to a service - 'the risk and instability it causes doesn't do the customer any favours' - and the high price of damaged morale. 'It puts us in a difficult position because we're having to take contractual responsibility for services with a disaffected staff,' he continues. 'It takes more cost and effort to sort that out.' Despite such obstacles, Cox and his private sector counterparts continue to pitch for government business, drawn by the initial £1.5 billion tranche of work on offer and the promise of what the usually sober Financial Times hyperbolically tagged 'a bonanza beyond their dreams'. The fiercest and most frenetic activity to date has been in the IT sector, where the value of work is the highest - around £750 million - and the programme furthest advanced. Some have already prospered through a variety of means. Late last year, for example, within the space of a week, the General Motors subsidiary EDS picked up two of the largest contracts on offer - that for the £18.5 million privatisation of DVOIT, the computing arm of the Driver and Vehicle Licensing Agency, and, more controversially, given the issues of confidentiality, the 10-year, £2 billion contract to run the Inland Revenue's IT systems.
Elsewhere, others have repeatedly pitched in vain, with the effect that most are either now more selective in their targets or, in some cases, are beginning to lose interest. Digital, for example, which originally declared great enthusiasm for the policy, has recently adopted a more 'pragmatic' approach and now admits to 'qualifying the opportunities more carefully'. Similarly, Sema Group, another early but equally unsuccessful advocate, concedes that it will have to ask itself 'serious questions' if its next few bids fail. It is not surprising - bidding in the public sector can be an expensive and difficult business. Not only is the amount of detail and preparation required typically greater than in the private sector but the timescale is longer. 'It's more a long war of attrition than a short, snappy campaign,' says one IT contractor. As a result, the cost of mounting a single large bid can lie anywhere between £500,000 - £1 million.
The odds, too, are unfavourable. 'The probability of getting work is in many cases much lower than in most companies' usual sphere,' observes Hoskyns's Cox. 'If you get down to a short list of three your chances are not one-in-three, but one-in-six, as there's a 50% chance that the work won't go out at all.' If anything, that figure is now slightly higher, with 68% of the volume of work currently staying in-house. The experience of tendering in local government has shown that once that percentage creeps much higher, potential bidders start to walk away. If repeated here, the obvious danger is that the competitive aim of the policy will be undermined. Either the work will stay in-house, uncontested, or, as has happened before, a public sector monopoly will simply be replaced by its private sector equivalent.
The relative lack of success of the private sector has also raised suspicions that the playing field between it and the in-house team is not always level. 'There are,' says one aggrieved bidder, 'a thousand different ways of losing.' Some claim to detect a bias in certain government departments in favour of work staying inside. Others point out that under the existing 'bottom-up' system (of departments being left to decide the composition of their own market testing programmes) there is too much leeway to obstruct the process - by putting out small, unattractive packages of work, for example, for which there is little demand. Kemp appears to share the logic of many increasingly weary contractors who are incurring heavy costs and no revenues: 'If you can use your talents in the private sector why sweat it out with a lot of unenthusiastic civil servants? Why do the hard one when you can readily do the easier ones?' Perhaps unsurprisingly, the progress of the policy to date has been both slow and uneven. Figures from the Cabinet Office for the first 18 months of market testing show that, even three months after the initial deadline, only four out of 21 departments had completed or were near completing their scheduled programmes. Some, such as the Department of Social Security barely seem to have started at all. Others appear only slightly more enthusiastic - Education, National Heritage, Trade and Industry and the Lord Chancellor's department have all met just one-fifth of their targets. And it can be of scant comfort to either the Office of Public Service and Science (OPSS) or the Treasury - the joint guardians of market testing - that they, too, feature among the laggards, with just over half of their programmes completed.
The explanation, according to Kemp, is straightforward. 'The OPSS has fallen into the first trap of reform - putting it into the hands of the people who don't want it to happen.' At the same time, he says, it has failed to win the hearts and minds of civil servants through a simple lack of communication. 'Market testing could have been explained more, could have been put into the context of a whole host of reforms aimed at better management. It wasn't, and instead has come to be seen as an end in itself.' There was also, he observes, the strong taint of encroaching ideology. 'Ministers' rhetoric sometimes seemed to suggest that better service was ipso facto obtained from the private sector. Once you get into that stance it immediately becomes party political and you've started to say something offensive to the existing workforce. As a result it started off badly, got a bad reputation and when it did so, people started dragging their feet.' The OPSS's task has not been aided by several notable inconsistencies - a policy document, for example, circulated in Whitehall at the time of market testing's launch, declared that the Government's presumption 'is that services should, wherever possible, be provided by the private sector rather than the public sector'. There is also the matter of last year's dispute - the details of which were leaked - between Sir Peter Levene, the OPSS's long-term adviser on efficiency, and the president of the Board of Trade, Michael Heseltine, over the future of the DTI's industrial laboratories. Levene wanted the laboratories to be market tested. Heseltine, whose quoted aim was to get 'as much of the DTI's activities into the private sector as possible', claimed that to do so would be simply for the sake of bolstering the department's lacklustre market testing performance.
All of this, of course, leaves the OPSS with something of a philosophical dilemma - to what extent should a central government department interfere in the work of another? John Oughton, head of the Efficiency Unit, that part of the OPSS specifically responsible for market testing, is clear: 'Our role is not to tell departments what to do, it's to set the policy framework for them, encourage them to use all the different elements of this toolkit. The prime responsibility must rest with them.' Further, he denies that the poor record of some departments is indicative of obstruction, more a reflection of variations in size, structure and the nature of their business. 'Different departments face different problems,' he says.
Kemp's diagnosis, unsurprisingly, is somewhat different: a question of a confusion of duties between the policy's two very different parents. 'The OPSS has to be the gadfly organisation, the one that wants change. The Treasury, on the other hand, isn't temperamentally inclined for change. The tension has to be that the Treasury is the cautious one and the OPSS is the one moving it along.' Here, he says, the roles have been sadly reversed. 'Some of the revolutionaries in these matters are in fact in the Treasury. They are the ones that want to go far and fast - and they're right.' To some, such as Keith Burgess, managing partner of Andersen Consulting (an unsuccessful bidder, together with its partner, ICL, for the Inland Revenue), the entire programme suffers from a deep and seemingly irredeemable flaw - that it sets out to test discrete functions rather than final services. 'When a department is looking at what it should market test, arguably it should be seeking to test the true services that are getting delivered. It is, after all, in delivering the services that you have the greatest opportunity to increase value - either in terms of the benefit to the recipient or in reducing the cost. If you're just taking the individual functions in a department then the scope and capability for that becomes a lot more restricted.' It is perhaps partly this that explains the erratic savings to date (see table, p41). While the gains across Whitehall add up to a cumulative 23% - though some have cast doubt on the figures - there are some departments where that percentage is still in single figures, where the cost of administering the policy (excluding the negative effects of demoralisation and upheaval) seem to make it hardly worth the candle.
Given this expense, the distraction that the policy is causing and the time taken to achieve the projected savings there are, unsurprisingly, increasing signs of Treasury impatience. As several commentators have pointed out, if the programme continues at its current rate - of testing 25,000 civil servants in 18 months - it will take almost a decade to make any substantial impact on a workforce of over half a million. Figures also reveal that the activities tested to date have tended to be those, such as computing, of high value yet with relatively few staff. Hence, while the programme has met 77% if its initial value target - boosted by the inclusion of what Oughton describes as two 'rogue' tests, the MoD's Atomic Weapons Establishment and the Inland Revenue, which together account for almost half of the £1.1 billion worth tested to date - it has reached only 57% of its staff target. The inference is that departments are having difficulty - or, perhaps worse, are actively shying away from - testing their more staff-intensive operations. And if, as seems the case, it is the comparatively easy activities that have so far been tackled, the programme will probably encounter yet further delay in the years ahead.
On this basis some are asking whether there is not a better, faster way of achieving the same objectives, of averting the entire market-testing procedure by opting for privatisation or a more straightforward process of contracting out. Others suggest that if in-house savings can be made simply by invoking the threat of competition, that same fat could be taken out with a lot less fuss by the Treasury's imposition of tighter budgets. Alternatively, if market testing is such a good tool for imposing efficiency and the Government's true aim is to find its 'inescapable core', why is its scope still so narrow? Indeed, the next tranche of market testing, due to be completed by the end of September this year, is actually smaller than the first. Only three departments have stepped up their workloads. Two, on the other hand, have already abandoned their plans for further testing.
Perhaps the greatest irony is that virtually all agree that it could be otherwise. 'Market testing has been handled in an appallingly ham-fisted way,' says Kemp, 'but that doesn't invalidate the principle.' Cox concurs: 'The benefits are there to be had - yet the means by which those benefits could be achieved have been disregarded.' Even some civil service union leaders who publicly oppose the policy admit its merits in private.
There is the odd sign of change. In February, for example, the OPSS announced it was to establish a market testing forum to solicit views from the private sector on how the process could be improved. It does little, however, to conceal the basic, uncomfortable fact that the OPSS appears to have pleased no one, least of all the Treasury, and instead remains caught between two rival disaffections - that of the private sector, which it sought to encourage, and the majority of civil servants, whom it should have enthused. Somehow, amidst this ill will, Oughton and his colleagues must find a way of transforming what appears to be an evolution into a genuine revolution, of turning Stephen Dorrell's 'long march through Whitehall' into a zealous, sustainable sprint.
Whitehall put to the test: costs and savings 1992-3
value testing Net
activities Savings costs savings
Department (£m) (£m) (£m) (£m)
Ministry of Ag Fish + Food 23.9 4.5 1.0 3.5
Customs + Excise 59.9 14.7 0.7 14.0
Ministry of Defence 71.2 19.9 4.4 15.6
Department of Education 2.7 0.6 0.1 0.5
Department of Employment 46.9 10.6 1.8 8.8
Department of the Environment 47.7 4.6 0.4 4.2
Foreign + Commonwealth Office 15.3 6.7 0.1 6.6
Department of Health 19.0 3.7 0.2 3.5
Home Office 88.9 6.0 3.4 2.6
Inland Revenue 51.3 17.7 1.5 16.2
Lord Chancellor's Deparment 5.9 1.8 0.3 1.5
Department of National Heritage 2.0 0.6 0.0 0.6
Northern Ireland Civil Service 34.9 6.6 0.2 6.4
Office of Public Service + Science 19.3 5.1 0.3 4.8
Overseas Development Admin 5.4 2.0 0.1 1.9
Scottish Office 12.6 2.8 0.6 2.2
Department of Social Security 15.0 7.1 1.6 5.4
Department of Trade + Industry 14.6 5.8 0.8 5.0
Department of Transport 26.6 7.1 0.1 7.0
HM Treasury 1.9 0.5 0.1 0.4
Welsh Office 2.5 0.5 0.3 0.2
Others 26.5 6.7 0.9 5.8
Total 594.0 135.6 18.9 116.7
Source: Cabinet Office.