“Universalising the best”

In his Cabinet memoranda and subsequent Parliamentary speeches outlining and defending his design for a free-at-the-point-of-use national health care system – thus ditching the mixed model agreed by the predecessor Coalition Government and facing down ministerial colleagues who sought to preserve a role for local government (Klein, Reference Klein2010) – Bevan never used the word equity, so saving himself from much academic parsing of the precise meaning of that complex concept (Powell and Exworthy, Reference Pollitt, Harrison, Dowswell, Jerak-Zuiderent and Bal2003). But the objective was clear, if expressed in a variety of formulations. It was “to universalise the best”, “to see that an equally good service is available everywhere” and “to promise every citizen in this country the same standard of service”. And that, he argued, could only be achieved through a national service funded through a redistributive tax system.

In the outcome, the NHS scores strongly on the equity criterion in international comparisons. So, for example, the 2017 Commonwealth Fund study of 11 rich country care systems (Schneider et al., 2017) put the United Kingdom first on this criterion, using a survey that compared the experience of higher and lower income groups in their access to, and experience of, care. And the NHS has proved a powerful instrument of income redistribution. A recent survey of the evidence on socio-economic inequalities in health care (Cookson et al., Reference Cookson, Propper, Asaria and Raine2016) concludes that “public health care expenditure is substantially concentrated on poorer people: The poor are sicker and their use of health care reflects this”. The better off are more skilled in getting high quality care once within the system – as well as being able to buy care in the private sector – but there is no such bias in getting access to the system via family practitioners.

However, the picture is somewhat different if we take Bevan’s own words at face value and focus on geographical equity. Can all citizens expect the same standard of service, irrespective of where they live? In answering this question, there are two separate issues. The first is about the geographical distribution of resources within the NHS. Are NHS resources distributed equitably? The second is about how those resources are used. How far do they ensure the same standard of service to all?

The geographical distribution of hospital and GP services that the NHS inherited in 1948 was grotesquely skewed towards the healthier and wealthier parts of the country. Decades of effort to change the distribution of resources have followed. But has territorial justice been achieved? The question does not allow a ready answer because the resource allocation formula designed to achieve equity, first introduced in the 1970s, has changed over time and remains contestable (Bevan, Reference Bevan2009). Equity requires resources to be matched to relative need. So how is the need of any given population to be measured? Age structure clearly matters. So, of course, does population health. But what are the appropriate indicators of population health? How can the effects of social deprivation be taken into account? What weight should be given to any one factor?

The technical and intellectual arguments have, over time, produced refinements and changes in the formula. And as the formula changes, so do the target levels of expenditure for given populations. Nevertheless a study of the changes since the 1970s (Buck and Dixon, Reference Buck and Dixon2013) concluded that “relative-under funding and over-funding has reduced for most area of England”. And the very fact that the NHS’s concern is now with the small print of the formula for achieving territorial justice is perhaps the best evidence of success in modifying its skewed inheritance. Bevan is vindicated.

But does success in moving towards equity in the distribution of resources mean that everyone can expect, in Bevan’s words, the same standard of service? Here the answer is an emphatic no. Geographical variation in the both the quality and quantity of specific services remains the norm (Appleby et al., Reference Appleby, Raleigh, Frosini, Bevan, Gao and Lyscom2011; NHS England, 2015; The Medical Technology Group, Reference Schneider, Sarnak, Squires, Shah and Doty2017). Not only do the odds on getting access to repair surgery or diagnostic technology vary from place to place, but so do the odds on survival after a stroke or heart attack. Changing the distribution of resources, it turns out, is relatively easy; changing the way those resources are used and the way doctors practice is a very different matter. In this respect the NHS is no different from other health care systems (OECD, 2014): all struggle with the challenge of geographical variations. Bevan’s vision remains just that.

Fiscal crisis as a way of life

If a graph of the year by year changes in NHS funding over the decades were a temperature chart, there would be cause for alarm for the patient. Over the NHS’s lifespan, its budget has grown by an average of 3.8% a year in real terms (Harker, Reference Harker2017). The average, however, conceals dramatic fluctuations: from a cut off 6.6% in 1953/1954 to an increase of 11.6% in 1974/1975 when runaway pay settlements drove the budget. Periods of plenty have been followed by periods of famine. The 1950s were years of scarcity, as were the 1980s. As the 20th century gave way to the 21st century, the NHS enjoyed an unprecedented period of sustained high growth rates. Following the financial crash, a decade of austerity has followed. Since the 2010 General Election the NHS’s annual budget growth has been little more than half the historic rate.

The point is as obvious as it is simple. Decisions about the size of the NHS budget increment depend on the state of the British economy, judgements about public finances made by Chancellors of the Exchequer in the light of whatever paradigm of economic management is dominant at the time and political calculations about the effect of any settlement on voter perceptions. In this respect the NHS is no different from any other tax funded public service. Where the NHS differs is in its degree of political visibility. This cuts both ways. On the one hand, it encourages often premature warnings about impending doom. In particular the medical profession has a long record of advertising its grievances by talking up the shortcomings of the NHS (Klein, Reference Klein1983). On the other hand, public affection for the NHS wins it gets special treatment: so while other public services have seen their budgets cut, the NHS has (in relative terms) been protected. Thus the increasing visibility of longer waiting times, cancelled operations and missed targets persuaded an otherwise grudging Chancellor of the Exchequer to give the NHS an extra £6.3 billion over three years as an eve of 70th birthday present – utterly inadequate in the view of those working in it but large enough to make other government services envious.

A high degree of political visibility – and dependence on political decisions – is not, of course, the exclusive original sin of the NHS. Spending on health care involves high politics everywhere. Where the NHS has differed throughout its history is – as argued in the introduction – in the fact that the debate has invariably been about under spending rather than over spending. This reflects a number of factors, notably the frailty of the British economy. But, crucially, it reflects the fact that, in contrast to many other countries, the government in London has the ability to control the spending taps. The next section therefore examines the role of central government in more detail.

A cacophony of bedpans

Bevan’s much quoted bedpan doctrine – “when a bedpan drops in a hospital, I want the noise to reverberate in the corridors of Westminster” – neatly captures the centralising logic of his model for the NHS. If the aim is to assure equal access to everyone – let alone to universalise the best – then clearly the Minister has to take responsibility for bedpans wherever they may be dropped. The point is further reinforced by the doctrine of public accountability. Given tax funding, the Minister is answerable to Parliament for how the money is spent. So much is self-evident. But two questions arise. First, how can a Minister take responsibility without exercising control? Second, if control means centralisation, as it does, is this necessarily undesirable?

The history of the NHS can, very roughly, be divided into two halves. For the first 30 odd years, Ministers of Health lacked the capacity to exercise effective oversight. They could, to be sure, control the total spent. But bedpans could have been dropping all over the place without anyone at Westminster knowing. Statistics about what was happening in the NHS were chiefly of historical interest by the time they became available; the Ministry of Health was slow to develop any analytic capacity; media interest was low. The centre was weak; regional authorities were strong. Ministers could only exhort and hope.

Over the following 30 odd years, all this changed. The development of IT meant that information about what was happening at the periphery became available in real time. The noise of bedpans dropping became a cacophony at Westminster (to exaggerate only a little), amplified by ever more intense media interest. Regional authorities were eventually swept away. Secretaries of State, as they had become, could command and control, and did so by setting targets and punishing deviants. If they failed to act, they faced political assassination by the media. For those at the coalface of the NHS, the threat of central intervention became only a telephone call away.

None of this, of course, is evidence for over-centralisation. For this, we have to call successive Secretaries of State into the witness box. Over the past three decades each in turn has promised to reverse the trend towards centralisation and made a point of seeking to push power to the periphery: the tribute paid by vice to virtue. In 1989 Kenneth Clarke, unveiling his plans for the introduction of a mimic market in the NHS, promised “To make the Health Service more responsive to the needs of patients, as much power and responsibility will be delegated to local level” (Secretary of State for Health, Reference Rose1989); his Labour successor, Frank Dobson, stressed that “Local doctors and nurses…will be in the driving seat shaping services” (Secretary of State for Health, 1997).

Despite these good intentions only five years later another Labour Secretary of State, Alan Milburn, was deploring the fact that “For fifty years the NHS has been subject to day to day running from Whitehall. A million strong service cannot be run in this way” and declaring that “power needs to be devolved to locally run services” (Secretary of State for Health, 2002). But in 2010 a Conservative Secretary of State, Andrew Lansley, returned yet again to the same theme. “The Government’s reforms”, he promised, “will liberate professionals and providers from top down control”. The notion of putting local doctors in the driving seat – as promised by Dobson in 1997 – was to be pushed to its logical conclusion by devolving “power and responsibility for commissioning services to local consortia of GP practices” (Secretary of State for Health, 2010).

Lansley’s legislative proposals raised much parliamentary wrath for many reasons, among them a suspicion that the Secretary of State was trying to wriggle out of his accountability for the activities of the NHS. Reassurance and change of wording followed. In the event the institutional form – but not the thrust or weight – of central control has changed. The Department of Health has become hollowed out while responsibility for overall performance management has passed to NHS England (with special responsibility for purchasers), NHS Improvement (for providers) and the Care Quality Commission (CQC) (across the board): the culmination of decades of institutional tinkering designed to define a boundary between strategic decisions by politicians and policy implementation by managers. MPs address complaints about dropped bedpans not to the Secretary of State in the House of Commons but to the trio.

However, the Secretary of State remains accountable for the performance of the NHS as defined by him in the mandate given to the agencies: in turn, the agencies are his creatures (to use one of Bevan’s phrases), responsible for the bedpans. The telephone line is still open – and often used by the Secretary of State to by-pass the trio of agencies (Timmins, Reference Timmins2017).

The trio can, in effect, both command and control, as their use of “special measures” illustrates. Foundation Trust hospitals, monuments to Milburn’s conversion to the principle of devolution, are independent Public Benefit Corporations “not subject to direction from Whitehall” as the official Department of Health guide put it when they were first introduced (Department of Health, 2005). Clinical Commissioning Group, Lansley’s brainchild, are autonomous statutory bodies. Yet both can (like all trusts, CCGs and even GP practices) be put into “special measures” by the regulatory trio if they fail to deliver either quality or solvency. Which means that they, in effect, lose autonomy: an improvement director is drafted in, new chairs, board members and executives may be appointed and they will be expected to publish their progress against agreed action plans every month until they have satisfied the regulators. In the past concern about quality by the CQC was the driving force behind invoking ‘special measures’; more recently it has been NHS Improvement’s concern about trusts losing control over their finances.

The web of regulation

In a system which concentrates blame on the centre, intensive regulation of quality designed to minimise risks would seem to be a logical response. So would performance management in a system which suffers from recurring fiscal crises and is anxious to wring the last penny of productivity gain out of every pound spent. To the extent that the NHS is exceptional in these two respects, as it is, so it might be expected to be exceptional in the width, depth and intensity of its monitoring activities, if not necessarily in their success.

Even a casual dip into the empirical evidence confirms the plausibility of such a conclusion. Consider, for example, the case of a Foundation Trust which recently compiled a list of all the bodies with the power to issue guidance, inspect and (in some cases) shut it down. The list contained 70 entries (Clarke, Reference Clarke2017). Many of these – such as the fire brigade, the National Audit Office and the Competition and Markets Authority, for example – are not specific to the NHS. So there is an element of exaggeration about the figure. But only an element, given the multiplicity of professional and managerial bodies ranging from the medical Royal Colleges to the NHS Litigation Authority specifically involved in regulating NHS activities.

Or consider the case of the CQC, whose ever-expanding remit of inspecting and grading for quality now covers over 30,000 health and social care providers. How, given the scale of its responsibilities, can it prioritise its programme of inspection? The CQC’s response to this question, in the case of hospitals, illustrates both how NHS performance indicators have proliferated and how problematic their use is. The CQC has devised a risk score based on 150 such indicators. Unfortunately, as a recent study has found (Griffiths et al., Reference Griffiths, Beaussier, Demeritt and Rothstein2016), the scores and the subsequent quality ratings of inspectors do not match. NHS England’s annual assessment of CCGs, by way of contrast, uses a mere 60 indicators to categorise their performance on a one to four scale.

New tools of performance management continue to breed. In 2017 the CQC and NHS Improvement jointly introduced a “use of resources assessment framework” for NHS trusts. This uses 22 “metrics’ to score trusts on a one to four scale (NHS Improvement, 2017). Again in 2017 NHS England produced a “Progress Dashboard” for the 44 Sustainability and Transformation Partnerships set up to bring CCGs, hospitals and local authorities together to plan for the future on the assumption that integrating health and social services and redesigning NHS provision will bridge the gap between available funding and the demands of an ageing society. The Progress Dashboard uses 17 performance indicators to rate STPs, again on a one to four scale (NHS England, 2017). Clearly the centre will continue to exercise a strong grip on what might initially have been seen as an experiment in localism – or the devolution of blame for what are going to be unpopular decisions about streamlining and cutting services.

The dramatic change in the culture of the NHS over the decades – from the dark ages of steering without an up to date map to the new era of restless searchlights illuminating the landscape – is illustrated by two reports. The first, published in 1983, was the report of Sir Roy Griffiths – managing director of Sainsbury’s – into management in the NHS. This criticised the NHS’s lack of “any real continuous evaluation of its performance…Rarely are precise management objectives set; there is little measurement of health output…Businessmen have a keen sense of how well they are looking after their customers. Whether the NHS is meeting the needs of the patient, and the community, and can prove it is doing so, is open to question” (Griffiths, Reference Griffiths1983).

Thirty odd years later the report on leadership in the NHS of Lord Rose – former chairman of Marks and Spencer and a string of other large retailers – had a very different emphasis. This argued that “the NHS is drowning in bureaucracy”, that “there are too many regulatory organisations making too many reporting requests”, that “regulators appear to be in overdrive” and that “bureaucratic reporting has made both individual Trusts’ and the NHS’s views short term”. It concluded “Too much is being done by numbers. Within the NHS everyone is managing upward by means of complying with data requests; for good leadership to flourish they should be delegating downward. People need to be and to feel trusted beyond compliance” (Rose, Reference Powell and Exworthy2015).

The Griffiths report was immensely influential in shaping the future of management of the NHS. Far from lacking in “any continuous evaluation of its performance”, the NHS has become a hypochondriac continuously taking its own pulse. The “logic of escalation” (Pollitt et al., 2010) has taken over. If one set of performance tools fails to detect patients being maltreated or to prompt greater efficiency, the reaction is to devise new, usually additional, ones. So the manufacture of new performance measures continues, as we have seen. In this respect, the Rose report has had no impact – although Griffiths might well have agreed with him that excessive self-examination risks becoming damaging self-flagellationFootnote ² .

Multiple objectives, multiple conclusions

A variety of conclusions can be drawn from this analysis of Nye Bevan’s double-edged legacy. The first is that the virtues of Bevan’s model easily outweigh its flaws: witness, to recapitulate, the achievements of the NHS in providing universal health care, its resilience and its status as a national treasure. The second is that the mix of achievements and flaws flowing from the NHS model illustrates the larger point that all policy making involves choice between competing and conflicting objectives, all of which may be desirable: a well worn observation that, however, is too often forgotten when policy makers are criticised.

The third is that continued tension between competing objectives may lead to the kind of instability illustrated by the history of the NHS: a history marked by the frequency of disruptive and expensive organisational changes (Klein, Reference Klein2013). This is often put down to the vanity and over reach of politicians: the desire of successive Secretaries of State to put their imprint on the NHS. No doubt this was one factor. More important, however, was change extrinsic to the NHS: change in the dominant paradigm of public policy, specifically the introduction of new managerial and market notions in the 1980s.

But, above all, successive reorganisations were also attempts to solve problems inherent in the NHS institutional model. Thus the organisational changes of the 1970s were largely driven by the desire to align health and social care boundaries in the hope that this would promote cooperation, a concern which continues to shape policy to this day. Similarly, as we have seen, attempts to reverse the centralising bias of the NHS model provide a theme running through successive reorganisations. And the fact that these attempts have so far failed suggests a fourth conclusion. This is that there are more reorganisations to come. So perhaps the itch to reorganise should also be seen as part of Nye Bevan’s legacy.