With the resignation of Tony Blair, in June 2007, there has been a spate of commentary on the 'New Labour' legacy. This paper is concerned with a facet of this legacy, New Labour's commitment to 'New Public Management' (henceforth NPM) in the 'reform' of public-sector services. The manifestations of NPM are so ubiquitous that it is reasonable to suggest that it is virtually impossible for a British citizen to escape its effects. A commitment to NPM was present from the first Blair administration. The 1999 White Paper, Modernising Government, suggested that 'the system in Whitehall...has meant that Ministers, Departments and Units have often been forced to devote effort to maximise their funding rather than considering what differences they can make in the form of...results and outcomes' (p. 11). This commitment has continued. Tony Blair established a 'Delivery Unit' in the Cabinet Office. The other principal architect of New Labour, Gordon Brown, operated, from the Treasury, a series of Public Sector Agreements (henceforth PSAs) which linked public funding to 'targets for improving services' thus 'shifting the focus decisively from inputs to the outcomes that matter' (Modernising Government, p. 37).
The object of this paper is to reflect on this aspect of the New Labour legacy and to suggest that historical research can play a salient role in such a critical analysis. In one sense this might seem paradoxical. The use of the adjective 'new' suggests that NPM is a novel project. Equally such a conception is reflected in the language used in New Labour policy documents which counterpose the discredited 'old' approaches (the focus on 'inputs') with the effective 'new' approaches (the focus on the 'outcomes that matter'). However, the thesis advanced in this paper is that, while as a programme of public-sector 'reform', NPM is historically distinctive, this is not the case with respect to the particular techniques associated with this programme. Furthermore the central problems associated with such techniques can be found in a number of historical examples.
NPM operates at a number of levels. As was indicated above, it can be viewed as a set of broad principles which define an approach to the organisation of public services. For example the notion of a focus on 'outcomes' does not suggest how this putatively distinctive set of objectives is to be achieved. NPM is, however, associated with a range of techniques which are designed to facilitate the achievement of the broad objectives of the programme. Thus a common feature of NPM is an emphasis on performance measurement. Such measurement plays a variety of roles. It can serve as a standard of performance for the service and hence as a target where performance is 'below standard'. Thus, the PSA for the Department for Education and Skills (DfES, now the Department for Children, Schools and Families) stipulates that 60 per cent of pupils should achieve 5 GCSE passes between A*-C grades by 2008. Such measures can also be used comparatively in benchmarking where a given provider (e.g. a school or National Health Service (NHS) Trust) is seen as an exemplar of 'best practice'.
Within the broad NPM programme there can be different views on the relative emphasis on different techniques. While measurable performance improvements are a common objective, policy implementation can reflect a greater emphasis on either 'command' or 'market' mechanisms. Thus, consistent failure to achieve what is viewed as a satisfactory level of 'performance' could lead to a mandated change in the management of the service provider. NHS trusts which were 'zero rated' in the NHS performance ratings and which had made 'no significant improvement' had their management function franchised to alternative management teams. In contrast market mechanisms can provide an incentive to improve standards. In schools in England, for example, the principal funding source is age-related pupil numbers (per capita funding is higher in the case of older pupils). It is a statutory requirement that pupil performance in public examinations and tests be published, thus schools perceived as 'performing poorly' run the risk of being under-subscribed and losing funding.
As the examples cited above show, New Labour's adherence to NPM has been categorical and documents like Modernising Government have an aura of certainty. In this sense, as much critical academic commentary has stressed, the programme is presented in a way which makes it difficult to dissent. Take, for example, the issue of 'inputs' and 'outcomes'. It is a reasonable assumption that the public want 'better' hospitals and schools, more effective police forces and so on. In this respect a set of techniques which claim to effectively 'deliver' improvements in the products of public services must be seen as an advance on a concern with simply ensuring that public funds are deployed for the purposes for which parliament voted them (the 'inputs').
However, once NPM techniques are considered in greater detail this prescriptive message is rendered problematic: the example of 'educational standards' will serve to indicate some of the issues. As was indicated above, some PSAs for the DfES stipulate targets related to test/examination performance. In many New Labour policy documents (see for example the 2006 Education White Paper, Higher Standards, Better Schools for All: More Choice for Parents and Pupils, para. 4.1) such results are equated with 'standards' so that an improvement in scores represents an improvement in 'educational standards'. However such a usage is misleading. While it would be difficult to achieve a consensus on what constitutes an improvement in 'educational standards', a (hopefully) less contentious aspect of such a definition would be that 'improvement' has occurred if students have developed their general analytic skills. However, it does not follow that higher test/examination scores are an indicator that analytic abilities have improved. As has frequently been observed, when such scores become linked with a set of incentives (as in the 'education market' in English schools) teachers will often seek to prepare students so that they perform well in such tests. Such preparation includes, for example, 'teaching to the test', effectively concentrating on items anticipated to occur in the test or examination. In such cases there is a risk of a disjuncture between test scores and analytic abilities with students experiencing a limited 'education' lacking in genuinely 'transferable' skills.
It might seem that debates on the effectiveness of techniques associated with NPM have at best a very limited historical dimension as most academic commentators have taken the view that NPM, as a programme of public-sector reform, dates from the early 1980s. However, this does not mean that techniques identified with NPM and debates around such techniques do not predate NPM as a reform programme. To illustrate the operation of public management 'before NPM' three examples are used. The following selection criteria are deployed: that performance measures linked to a policy goal and mechanisms designed to 'improve' performance in terms of the indicators are used; and that the public-sector programme in which the performance management techniques are applied has a substantial population coverage, i.e. these are examples of substantial public-service programmes. The examples used are: the attempt to manage performance standards in elementary education following the 'Revised Code' of 1862 till the de facto abandonment of the policy in the mid-1890s; the 'crusade against out relief' under the Poor Law in England and Wales between circa 1870 and the early 1890s; and attempts to encourage higher acute hospital throughput in the National Health Service (NHS) in the 1950s.
The term 'elementary education' was used in nineteenth-century Britain to refer to the formal education provided for the working class. At the point at which the Revised Code was introduced the state provided subsidies to voluntary provision of elementary education (in the two decades preceding the introduction of the Code, central government expenditure on education had quadrupled). Over the period during which the policy operated (to 1895) the number of children in state-regulated elementary schools in Great Britain increased almost fivefold from just under 1 million to 4.9 million (the total population increased from 23.7 to 34.6 million over the same period). Grants were regulated by 'codes' and the Revised Code, which was introduced in English schools in 1863, was designed to deal with a perceived deficiency in elementary education. It was argued that teachers devoted excessive time to teaching older children and consequently neglected the instruction of younger pupils in what would be currently called 'basic skills'. The 'performance management' policy introduced by the Code was to link grant payments to schools with the success of children in examinations in elementary reading, writing and arithmetic, where failure in any of these subject areas, in an examination administered by a government inspector, would result in loss of a substantial portion of the grant.
Over the three decades of its operation the policy went through a series of complex modifications. By the mid-1890s a diverse set of objections to the original concept of linking grant to pupil examination performance engendered a shift to a pattern of inspection which removed the obligatory annual examination and allowed inspectors to base their reports on observation of work in schools.
The 'crusade against out relief' was a policy which sought radically to reduce numbers on 'outdoor relief' under the locally-administered Poor Law in England and Wales from circa 1870 to the early 1890s. In this period the Poor Law constituted the de facto 'social security' system. At the point at which the policy was initiated there were just over 1 million 'paupers' in England and Wales: around 4.5 per cent of the population. Most poor relief was given 'outdoors', i.e. without requiring the recipient of support to enter a workhouse, and at the beginning of the 'crusade' well over 80 per cent of paupers received outdoor relief. The policy was premised on the assumption that outdoor relief was being given indiscriminately, creating 'dependence' on the state. The appropriate response, it was argued, was to manage outdoor relief by applying highly-restrictive rules following the precepts of various 'strict' Poor Law Unions. The key performance indicator was the outdoor pauperism rate (outdoor paupers as a per cent of the population). The overseeing government department, the Local Government Board (henceforth LGB) sought to diffuse 'best practice' by publishing 'data comparing the performance of unions' with respect to this indicator and this was designed to show the gap between strict unions and their lax counterparts. Such demonstrations were reinforced by LGB circulars and the power of LGB inspectors to attend meetings of Poor Law Guardians.
Initially the policy had a powerful impact and numbers on outdoor relief in England and Wales fell by nearly 40 per cent between 1871 and 1878. Further major reductions were, however, not achieved and numbers on outdoor relief fluctuated around roughly 10 per cent plus or minus the 1878 level for the duration of the policy. The end of the policy is frequently dated at 1893. This date is selected because the Local Government Act of 1894 rendered the Poor Law franchise more democratic and LGB policy, in part reflecting collectivist criticism of the basis of the crusade, became more relaxed, thus 1893 was the final year of radical 'dispauperisation'.
The creation of the NHS in 1948 involved a radical shift in population coverage leading to problems in predicting the cost of the Service. This was contributory to major cost over-runs in its first two fiscal years with expenditure exceeding the estimate by 39 per cent in 1948-9. While this had a number of effects, pressure for control over expenditure was particularly marked with respect to capital investment. It was generally estimated that capital expenditure on NHS hospitals was running at around one-third of the level of the end of the inter-war period in the early 1950s. Universal access created increased demand for hospital care but restraint on capital expenditure meant that the scope for increasing capacity by adding hospital beds was very limited. This tension generated interest within the Ministry of Health in more intensive use of hospital beds. This interest was exemplified in a detailed discussion in the Chief Medical Officer's report for 1952. Hospital throughput (officially defined as the numbers discharged from, or died in, hospital per available bed) was conceptualised as a key performance indicator. The report included data contrasting throughput rates in NHS hospitals and also advanced various suggestions for managing throughput. These included proposals related to the impact of professional practice on throughput and it was suggested that encouraging patients to be ambulant at an earlier stage would allow for earlier discharge. As with the Poor Law crusade, the aim was to construct a standard of 'best practice'. Thus throughput rates achieved in other jurisdictions (Canada and the United States) were treated as objectives to be aimed at, and recommended practices such as 'early ambulation' played an equivalent role to 'strict' rules on outdoor relief.
There was a substantial increase in throughput in NHS hospitals in the 1950s. Thus between 1949 and 1960 while available beds increased by 5.5 per cent the numbers of patients discharged/died increased by 40 per cent. There was no abandonment of an interest in increasing acute hospital throughput but, by the early 1960s, there was a major contextual change. As was indicated above, the emphasis on throughput in the 1950s was strongly related to very low levels of capital investment. In contrast the 1962 Hospital Plan, while arguably delivering a lot less than it promised, did create a commitment to a much larger long-term capital programme.
Not only were techniques of 'performance management' practised before NPM but such techniques engendered significant debates relating to their efficacy. First, there was the issue of 'ownership', or how far public-sector organisations were really able to control their measured 'performance'. Thus, under the Revised Code of 1862, in a context of limitations on powers to compel the attendance of pupils in elementary education, inspectors pointed to the superior examination performance of schools where pupil attendance was more regular. Analogous issues were raised with respect to variations in outdoor relief rates. In 1872 over a quarter of those on outdoor relief in England and Wales were 'able-bodied' widows with dependent children under 16. Early widowhood increased the 'risk' of pauperism and the 'risk' of widowhood itself varied between Poor Law Unions. Thus the LGB Assistant Inspector, George Culley, in a report of 1873, pointed out that the higher occupational mortality in mining areas increased the incidence of early widowhood and of higher outdoor-relief rates. Meanwhile, in the NHS in the 1950s variations in acute hospital throughput between hospitals in different areas were related to differences in the availability of complementary health services. Thus one study in the late 1950s showed that better provision of chronic sickness beds improved acute hospital throughput by facilitating discharge from the latter.
Second, was the issue of 'proxies', or how far the measures used in performance management reflected underlying policy objectives. Thus, an underlying principle of the Revised Code was that linking school grants to examination performance in the '3Rs' (reading, 'riting, 'rithmetic) would improve standards. However, sceptics took the view that the financial incentive to schools to concentrate on examinable subjects did long-term damage to the capacity to read by removing contextual information provided by subjects like geography and history. Meanwhile, an important managerial strategy in the crusade against out-relief was that specimen rules detailing the conditions under which it could be claimed were designed to encourage 'provident and independent habits'. This raised the issue of a suitable 'performance' measure. 'Success' tended to be judged in terms of pauperism rates. However, this was inadequate, since not seeking relief could not be equated with the adoption of 'provident' habits. To investigate this issue would have required a shift in the focus of investigation to those who did not claim or who had been refused relief. Such a perspective began to emerge when the assumptions underpinning the 'crusade' came under more critical scrutiny. Thus the Royal Commission on the Poor Laws of 1909, commissioned research on those refused outdoor relief, which came to the conclusion that in none of the 48 households studied was support by relatives increased through refusal of out-relief.
Third, there was the issue of 'manipulation', or how far there were incentives to 'play the system' so that measured performance targets were achieved but at the expense of broader policy objectives. With respect to the Revised Code there were frequent claims that, 'teaching to the test' was prevalent. For instance, an inspector reported that, when teachers had prepared pupils for arithmetic examinations through presenting questions in a numerical form, only around a third of the same pupils could provide correct answers when the same questions were presented by an inspector in a verbal form. As grants to schools were linked to examination passes, there were incentives to concentrate on pupils not quite of pass standard but viewed as having a chance of passing, while neglecting both those regarded as certain passes and those with no perceived chance of reaching this level. The Code was also seen as generating 'perverse incentives' beyond the parameters of education: thus an inspector reported visiting schools where children attended the annual grant-related examination while suffering from infectious diseases. Perverse incentives were also created by the objective of increasing acute hospital throughput. The latter gave some degree of official endorsement to the commonly-held view in the medical profession that acute hospitals should focus on narrowly-defined medical interventions. However, a potential corollary was the de facto separation of chronically-ill (particularly older) patients from access to the benefits of more skilled medical attention.
What, then, is the potential relevance of such historical examples as these to contemporary debates on the role of techniques associated with NPM used in the 'reform' of public services? A striking feature of the field of academic literature on NPM is the stress often placed on the limitations of the evidence for the effectiveness of such techniques. In what is arguably the leading international survey in the field Pollitt and Bouckaert (Public Management Reform: A Comparative Analysis. Second Edition, 2004, p. 118) state, 'At the level of broad programmes of management reform we know of not a single study from our twelve countries that convincingly links the actions taken with a set of positively and safely attributable final outcomes'. Such scepticism contrasts markedly with the confident sponsorship of NPM reforms by the Blair administrations. The limited evidentiary base underpinning the claims for NPM-based public-service reforms in part reflects the debates discussed in the previous section. Thus there are uncertainties regarding whether 'improvements' in performance are genuine; how far they reflect 'manipulation' by professionals and managers; whether comparisons of performance reflect the range of differences between providers; and whether 'perverse', unintended consequences are fully considered in any evaluation. Such a limited evidentiary base, however, raises the issue of how it can be reconciled with the political success of NPM in structuring public sector-service reforms.
A significant feature in this respect has been the use in prescriptive NPM arguments of 'success' stories. For example, a frequent feature of New Labour public-service White Papers (recent examples are the Education White Paper Higher Standards, Better Schools for All and the Health White Paper Our Health, Our Care, Our Say, both published in 2006) is the use of what are termed 'case studies'. These 'cases' have the following characteristics: they are extremely short, usually encapsulated in a 'box' in the text and in a distinct colour, they rarely exceed half a page of text; a corollary is that very little detail is provided on the example and hence it is impossible to assess in what sense the case (let alone the prescribed managerial policy) can be viewed as a 'success' and, even if it can be, whether it can be sustained.
Historical research is, of course, not the only counterweight to such shallow analysis. However, historical examples such as those discussed here do have a potentially important role to play in the broader debate on public-management techniques. The longitudinal analysis allowed by such examples creates the scope for a more sophisticated analysis of the mechanisms of public-management reform. These include the complex dynamics of attempts to respond to performance-management structures. In the case of the Revised Code, for example, the response to the criticism that its original form was encouraging an overly narrow curriculum was to extend the range of examinable subjects which qualified for grant. However, amongst the effects of such changes were the creation of a more complex and expensive regime of inspection; and greater inequalities between publicly-funded schools so that schools which had more resources were able to effectively offer a wider range of examinable subjects, thus attracting more grant that their poorer counterparts. Similarly, higher throughput in NHS hospitals in the 1950s certainly conferred general benefits in giving greater access to hospital care but also may have damaged the standard of care provided to older people perceived as 'bed blockers'. The history of public-management techniques 'before NPM' may thus contribute to a more uncomfortable view of the process of public-sector reform, imparting a necessary complexity to debates in this key area of contemporary politics.
Bridgen, P. and Lewis, J. (1999) Elderly People and the Boundary between Health and Social Care 1946-91: Whose Responsibility? , London: Nuffield Trust.
Cutler, T. (2005) 'Managerialism Avant La Lettre? The Debate on Accounting in NHS Hospitals in the 1950s' in V. Berridge and K. Loughlin (eds.) Medicine, the Media and the Mass Market, Abingdon: Routledge.
Cutler, T., Waine, B and Brehony, K. (2007) 'A New Epoch of Individualisation? Problems with the "Personalisation" of Public Services', Public Administration, 85 (3), 1-9.
Du Gay, P. (2003) 'The Tyranny of the Epochal: Change, Epochalism and Organizational Reform', Organization 10 (4): 663-84.
Hood, C. (1998) The Art of the State: Culture, Rhetoric and Public Management. Oxford: Clarendon Press.
Sutherland, G. (1973) Policy Making in Elementary Education, 1870-1895. London: Oxford University Press.
Tymms, P. (2004) 'Are Standards Rising in English Primary Schools?', British Educational Research Journal, 30 (4), 477-494.
Williams, K. (1981) From Pauperism to Poverty, London: Routledge and Kegan Paul.
Sign up to receive announcements on events, the latest research and more!
We will never send spam and you can unsubscribe any time.
H&P is an expanding Partnership based at King's College London and the University of Cambridge, and additionally supported by the University of Bristol, the University of Edinburgh, the University of Leeds, University of Liverpool, the Open University, and the University of Sheffield.
We are the only project in the UK providing access to an international network of more than 500 historians with a broad range of expertise. H&P offers a range of resources for historians, policy makers and journalists.