How the NHS Works (Or Doesn’t)

A Crash Course in the Political Economy of Public Service Provision

1.  Manufactures and Services: the Key Distinction

In The Wealth of Nations of 1776 Adam Smith extolled the virtues of the division of labour and manufacturing technique brought about by technological innovation and industrial scale organisation.  And indeed the record with regard to manufactures is impressive: agriculture, which has itself witnessed some remarkable gains in the days since Smith wrote (and historically provided the first example of the capitalist mode of production), is nowhere close.  For example, had agriculture experienced the same magnitude of productivity gains as the manufacture of the humble pin (Smith’s original example), we would be capable of growing an entire crop of wheat in 6 minutes flat.  The moral of the tale is simple: ignoring distributional issues, the wealth and standard of living of any society is embodied in the capital stock, technology and industrial potential owned by that society[1].

There is another area of economic activity which has suffered in comparison to industrial manufacture.  I speak of course of service industries, including the public provision thereof (health and education being the primary examples).  The reason for this is that economic activity in these sectors is inherently labour intensive and is not generally speaking amenable to mechanical production techniques per se, since their output intrinsically embodies a handicraft element performed by human beings - cutting hair, nursing people back to health, marking examination papers and so forth.  Compared to manufacturing, such services are inherently incapable of realising substantial productivity gains.[2]

2. The Cost Disease

All of which gives rise to an interesting phenomenon: insofar as pay levels for work in the manufacturing and service sectors retains parity, and insofar as we continue to consume broadly equal quantities of the output of both sectors, the proceeds of overall economic growth will be such that the comparative costs of service industries will rise relative to those of manufacturing, i.e. will become relatively more expensive, with spending on such items consuming an ever increasing proportion of our income[3]

This is the Cost Disease, highlighted as far back as 1966 by the American economist W.J. Baumol.  It is an unavoidable economic fact of life.  Ignorance of the Cost Disease is arguably the single most important item blocking sensible discussion of issues arising in respect of the provision of public services such as healthcare and education.  Indeed, one might go so far as to say that generic ignorance of the Cost Disease is in large part responsible for the untoward levels of gassing on this subject.

3. There is nothing wrong or extraordinary with year-on-year increases in public spending on health. This is precisely what one would expect of any service industry - public or private - subject to the Cost Disease, and is not a mark of unaffordability, inefficiency or waste. 

Note that it matters not one iota whether the services in question are advanced under state provision or privately by the market: the Cost Disease will apply equally in both cases - indeed, a good portion of the data on the Cost Disease is derived from an examination of privately funded healthcare provision in the USA, which as a nation has the highest per capita on healthcare as a proportion of national income anywhere in the world. 


The King's Fund: Health care spending compared to other countries 11 January 2016. Source: OECD Health Data 2015

Thus any degree of political discomfort one is apt to feel in the face of seemingly ever-increasing levels of government spending on health in the economy over time, measured as a fraction of national income, is entirely misplaced: the same proportional increase in spending would hold even if all such services were fully privatised. 

As far as I am aware, the cost of haircuts does not constitute a national scandal (however much we might complain about them becoming relatively expensive in comparison to manufactured goods).  Likewise, NHS costs are not out of control, nor do they reflect an inefficient service.  They simply reflect the normal pattern of expenditure to be expected of a service-sector industry in comparison with spending on output as a whole.  Nor is the NHS no longer affordable: it is simply that, as the price of manufactured goods declines in the wake of gains in productive efficiency in the manufacturing sector, a greater proportion of our income will be spent on services which are on the whole immune to such productivity increases. 

This is not a matter of being unaffordable per se, merely one of becoming relatively more expensive[4].

4. Predictably, none of this has prevented attempts to impose non-solutions upon non-existent problems 

Private Finance Initiatives, amongst their many other defects, are simply an attempt to defer the explicit recognition of the Cost Disease indefinitely into the future (and at a vastly increased long-term cost compared to what would have been the case had the infrastructure projects in question been publicly funded from the outset). 

Similarly, in a familiar refrain, during the last General Election campaign of 2015 the Conservative Parliamentary Candidate for Redruth-Camborne-Hayle George Eustice wrote, "there will always be growing pressure on our health service, and that is why we need a strong economy so that we can afford to make the investments needed."  Such remarks betray a touching ignorance of the nature of the provision of public (or indeed private) services.  Presumably George meant: health services will be rendered relatively more affordable in a growing economy. 

In fact, the nature of the Cost Disease indicates that quite the opposite will be the case.  In a growing economy, the amount of borrowing required to sustain spending on public services will rise in proportion to growth.  To repeat: a growing economy fuelled by efficiency gains in manufacturing, means that services become relatively more expensive. 

5. The greatest absurdity, however, is to be found in the proposition that state-provided services can be more efficiently managed if they are subjected to the discipline of the free market - Either directly via de-facto privatisation, or indirectly by the attempt to introduce regimes of extensive audit designed to internally replicate competitive conditions within publicly owned institutions (the so-called, ‘internal market’)

This is a nonsense.  Cost reduction in service industries can be realised only if

(a) for given levels of staffing, wages paid to workers are reduced, OR

(b) the same amount of work is accomplished by fewer workers for a given wage (i.e. significant productivity gains are implicated)

The former choice, amounting to a decline in the relative rate of pay awarded to workers operating in the service sector is liable to lead to increased difficulties with regard to long term levels of recruitment.  In the form of renegotiated pay contracts, it is a typical strategy employed by newly privatised sectors formerly under state control to reduce costs - and indeed constitutes the only means of squeezing profit out of the provision of such services, serving only to channel public funds into the coffers of private contractors who do not, and cannot, contribute to increasing the efficiency of any such provision.

The second choice has a particularly incoherent and depressing track record.  There have, of course, continually been expressed concerns over the long term relative productivity decline in state funded service provision.  Having been captured by a distinctly neo-liberal strand of economic thinking at a deep policy level, and reasoning that public sector workers of a piece regard their work as a burden to be suffered and who therefore take every available opportunity to slack-off (as is only human), it was nonetheless the peculiar fate of the Labour government under Tony Blair to attempt to offset such alleged rent-seeking behaviour on the part of public sector workers by the imposition of a continual and extensive regime of performance audit by way of appropriate incentives to efficient production.  

To no avail. As we have seen, since the Cost Disease applies irrespective of the source of provision between public and private, there are few if any efficiency gains to be had along such lines.  The net result has been an effective decline in the quality of services provided to users whereby less labour expended on a particular task is by definition a debasement of that task in cases of labour intensive provision, presided over by a coterie of target and audit-inspired managers who are apt to regard a Beethoven quartet performed on only three instruments and played at twice the recommended tempo as an efficiency saving.

Two further comments are in order.

(1) As it presently stands, there is one obvious efficiency gain to be had in respect of state-supplied service provision – namely, the immediate removal of counter-productive managerial regimes of audit purporting to capture non-existent efficiency gains[5].

(2) The depressing philosophical anthropology which informs the assessment of the behaviour of public sector workers is of course both false per se and a slur on human beings as such. In particular, it completely fails to acknowledge an ethos of public service upon which our public institutions depend for their efficient running. 

For, as recent experience in the case of the Mid-Staffordshire healthcare crisis and the subsequent Francis Report amply suggests, if one insists upon treating public sector workers as rent-seeking individuals and move to correct such behaviour by subjecting them to increasing levels of audit, the chances are that you will succeed only in modifying their actions in the direction of poorer service provision.  Prior to the imposition of the current trend of managerial reform, public sector workers by in large could be relied upon to do their job efficiently without too much fuss (certainly to the extent whereby it was cost-effective to check whether this was in fact the case).  Claims to the contrary simply display an ignorance of the Cost Disease. 

Subsequent to the introduction of reforms intended to replicate the incentivising operations of the free market within public institutions, not only are such reforms misplaced in terms of a search for non-existent efficiency gains, but will lead to reduced levels of performance in proportion as we witness the collapse of a public sector ethos.  In happier times this would perhaps be described as an ironic twist of fate.

6. The Sustainability & Transformation Plan (STP)

The upshot of our deliberations regarding the cost Disease would thus appear to be the following: for a given level of provision, expenditures will inexorably rise as a proportion of national income. 

This is nothing to get fussed over. 

Purported free market remedies for the Cost Disease, either in the form of privatisation of public service provision or spurious managerial solutions, are a myth.  The consequence of such a handling of public services will either be strictly ineffective or (more likely) conceal a de facto decline in the level of provision.

All of which leads us neatly onto the tetchy subject of the proposed Sustainability & Transformation Plan for Cornwall.  The gist of the plan is to amalgamate health and social care provision under a single organisation and single budget, with an emphasis on preventative rather than reactive care provision. 

Billed as a means to improve health and wellbeing and improve quality of health and care[6], it amounts to the suggestion that there are obvious organisational efficiencies (= productivity increases resulting from reorganisation) to be gained from an integrated approach to health and social care, i.e. a temporary respite to the Cost Disease.  Now in this instance, and considered in the abstract, this might well be true: numerous healthcare professionals have bemoaned the lack of an integrated approach to health and social care for a number of years. 

However extreme caution is to be exercised when contemplating such measures.  There are at least two reasons for this.  The first is obvious, the second less so.

A. Cuts

Alongside - and thus giving the impression of being compatible with - claims made on behalf of improved health and social care outcomes, we also have stated goals concerning ‘improving financial stability’ and the need to correct chronic ‘overspending’[7]

However, rather like the notorious expensive ski resort wherein there are to be found young girls looking for husbands and husbands looking for young girls, things are not quite symmetrical as they might first appear, the stated goals being strictly incompatible.  We can easily see why.  Assume a best-case scenario whereby organisational improvements of the type contemplated do indeed yield substantial efficiency gains.  There are two polar cases to consider.  By definition, an improvement in efficiency can either be realised in the form of:

(a)  an increased output (healthcare outcomes) using the same amount of inputs, OR

(b)  the same level of overall provision fixed at pre-STP levels using less inputs

Since input levels equate to spending, this is the same as the choice between

(a)  an increased output financed by the same level of expenditure, OR

(b)  the same level of overall provision fixed at pre-STP levels financed by less expenditure

The much publicised talk of the need to correct for budget overspend (figures in the region of £221 million-worth of cuts have been bandied about), make it clear that the primary intention behind NHS Cornwall’s proposals in respect of STP fall firmly into category (b) above. 

This of course effectively amounts to using efficiency gains from the proposed reorganisation of health and social care services to fund cuts elsewhere in the system, leaving the overall system-wide average level of provision otherwise unchanged. 

No doubt following such a reorganisation one will be able to point to specific aspects of the integrated system addressed by the STP as constituting improvements in certain health and social care outcomes, but these will be matched by equal reductions elsewhere in the system.  This being the case, the pertinent question to ask is therefore: 'are you satisfied with the current level of NHS provision in Cornwall?' Presumably no, because (amongst other things) you are complaining about the lack of integration of health and social care services! 

STP changes pursued on the financial model of (b) above will therefore improve the average lot of NHS users not one jot.

B. Efficiency versus Stability

Our second pause for concern surrounds the question: do we really want our institutions to become more efficient?  Surprisingly, perhaps, the answer is, ‘not necessarily’.  We can think of any system (such as that of NHS and social care provision) as a collection of nodes connected up in various possible ways to form one vast network. 

Analysis of such networks indicates that the goals of efficiency and stability are frequently at odds.

Consider a parallel case.  After the Wall Street Crash of 1929 and its role in the ensuing depression, in the USA it was decided to break-up the financial system into largely independent, unconnected financial institutions operating under restrictions placed upon the type of financial activity each institution was permitted to engage in – the most significant being that of the separation of commercial from investment banking (the Glass-Steagall Act). 

Over the last three decades or so, however, the world-wide trend towards the deregulation of financial markets, whilst increasing the so-called efficiency of the banking system (the number of connections between the various nodes in the financial network), has lead to a situation of instability whereby a problem experienced in one area rapidly cascaded throughout the system as a whole, frequently rendering it inoperable. 

Thus, whilst the immediate post-war period of heavy financial regulation witnessed approximately zero banking crises, recent decades have been subject to periodic financial meltdown and instability – the recent global financial crisis of 2008 being only the most notable example[8].

The STP proposals to integrate health and social care under one overarching financial plan, therefore, - efficiency aside - must be examined from the standpoint of network stability, embodied in the ability to prevent unforeseen difficulties arising in one sector from cascading throughout the entire system, leading to system-wide collapse. 

One means of doing so is to build-in spare capacity within the system.  However, not only does the existing pre-STP system lack anything approaching spare capacity[9], but the likely cuts noted above under A: Cuts additionally suggests that pre-existing capacity is to be removed from elsewhere in the post-STP environment to fund efficiency gains. 

Taken in combination, this is a cause for grave concern.  It is perhaps not too churlish that we ask for assurances upon both counts. The mantra, 'the NHS costs too much we can't afford it' will lead to the richer opting for private healthcare as the NHS will become a rump service for the sick, the poor, the old and emergency care.


[1] For the purposes of discussion I ignore the issue of the distribution of the wealth generated by production (!)

[2] This is connected with the fact that the impact of technology on public service provision tends to be quality-enhancing rather than labour saving – new technology in the NHS e.g. leading to improved treatment, diagnosis and monitoring, but only rarely a reduction in labour power required to treat a given condition.  It is at this point that the much-canvassed fact of the demography of an ageing population exerting additional costs on public service provision enters the fray – it is simply an aspect of the Cost Disease.

[3] Actually the problem is even worse than this: given that services typically fall within the category of superior goods, i.e. goods which consume a higher proportion of growing income per se, spending on such services will consume an increasing proportion of income at an increasing rate.

[4] Indeed, since there are some increases in productivity to be gained from the use of technology in the provision of NHS services (principally those associated with the use of information technology), the cost of NHS services is in fact declining in absolute terms – just not sufficiently quickly to make up for the relative cost increase in comparison to manufacturing output (i.e. the Cost Disease).  I return to the purported gains in efficiency to be had from the reorganisation of public service provision along the lines suggested in the Sustainability and Transformation Plan below in footnote 6.

[5] This is not to say that an institution as vast as the NHS does not require management.  It is simply to point out that a particular model of management – one inspired by false analogies with the manufacturing sector – is inappropriate to the case of public service provision.  This is an entirely different kettle of fish.

[6] Abstract, Integrating Health and Social Care in Cornwall and Isles of Scilly provided by Jackie Pendleton, Interim Managing Director NHS Kernow CCG at Health and Social Care Conference, Bodmin 05/11/2016.

[7] Same document as cited above. 

[8] The previous US Savings and Loan Crisis towards the close of the 1980’s e.g. required the US Federal government to bail out the institutions involved to the tune of 3% of GDP.

[9] Most manufacturing companies operate at around 70% plant capacity – sustained levels of production beyond this point commonly signalling the need for more investment.  The fact that the NHS currently operates at around 100% capacity gives the lie to the frequent claims that NHS managers are currently trying, under the rubric of efficiency, to run our healthcare services more like a business.

Do you like this post?

Find an event Volunteer Become a member

The Labour Party will place cookies on your computer to help us make this website better.

Please read this to review the updates about which cookies we use and what information we collect on our site.

To find out more about these cookies, see our privacy notice. Use of this site confirms your acceptance of these cookies.