Commissioning in the English NHS

Stephen Black's view of the market does not suggest much study or
awareness of how health care 'markets' differ from 'kumquat' markets!
There is more than half a century of scholarship and research on this, on
both sides off the Atlantic; and, while there have been intellectually
creative attempts to tailor the market to the NHS and vice versa,
Stephen's view does not show much awareness of what we have learned of the
pitfalls - during the Thatcher reforms and the Blair re-reforms.

Of course one has to look at the costs and benefits of rival
approaches ie of the market and of a planning alternative....in the real
world as well as in theory. That was the point of my Editorial. All I can
say is, to quote the late Michael Foot, "I am sorry that my instructive
hyperbole has fallen on such stony ground"!

Calum Paton
Keele University

Competing interests:
None declared

Calum Paton's clarification about his original text
describing the cost of commissioning doesn't alter my
objection to his number at all. We would both prefer to know
the real number more accurately, but he likes the 14%
estimate as it bolsters his case against commissioning.

It is worth asking what the cost of commissioning means
(whatever the actual estimated number) as this highlights
important aspects of the argument for abolishing the
provider commissioner split. For the estimate to be a useful
part of the debate it would need to be an estimate of the
amount of money we could save if we abolished commissioning,
all other things being equal. It is hard to estimate
reliably because many of the things we need to do for
effective commissioning are also things we need to do for
performance management and improvement. We will always need
to identify poor practice and improve it; stop investing in
activity that does no good for patients; introduce new drugs
and practice that benefit patients; and generally make
judgements about where best to apply the health budget. A
system that involves commissioning adds some extra
activities on top of these (like negotiating and managing
contracts). I suspect that when simple estimates of the cost
of commissioning are done, they lump all these other
activities together (as they tend to be closely associated
in organizations). But nobody is seriously suggesting that
we try to run the NHS with no planning, performance
management or budgeting, so most of the costs will still be
there in an NHS with a unitary structure. The extra cost
associated purely with commissioning is therefore likely to
be very much smaller than the total management budget of
commissioners. This is one reason why the 14% sounds
implausible.

I also have estimates from actual people involved with PCTs
that suggest the NHS would actually save money if all the
commissioners had enough capable people to monitor all their
contracts to ensure their providers were properly held to
account for their activity. That is why I hold out hope that
commissioning could get better (and not for some commitment
to market ideology).

The ideological debate is somewhat larger than commissioning
and is about whether markets work in public services. The
problem with the debate is that is often starts with
assumptions about the NHS that nobody would accept in any
other context. Crudely, the argument goes a little like
this: markets involve all sorts of extra costs over and
above a unitary, planned public service (like the cost of
commissioning transactions, duplication between competing
suppliers, excess capacity etc.). Therefore, the argument
goes, if we didn't have a market we could recover those
excess costs and get better services for the same spend. In
addition, many argue that a unitary NHS would find it easier
to achieve coordination across patient pathways.

The flaw with the argument is the failure to take into
account any benefits arising from markets. In other contexts
only extreme ideologues would think this argument worked.
The anti-market argument could apply Supermarkets or car
manufacturing and retailing even more strongly than to
anything in healthcare (they have extra costs like
advertising, for example). But would we be better off if we
nationalized food or car distribution eliminating all the
deadweight costs of capitalism? The reason why few think
this is because the benefits of markets and competition far
outweigh the extra costs associated with running a market.
So despite those costs we get ever cheaper cars and an ever
widening variety of food. Despite the lack of any central
planner and the presence of many competing organizations,
supermarkets still manage to provide the entire population
of england with as many kumquats as they desire, a feat of
coordination beyond the wildest dreams of the NHS.

It is not a big ideological leap to assume that some of
those benefits might be achievable in the NHS.

The argument against the provider commissioner split falls
apart if the overhead is low and if it can't be shown that
it will improve coordination, not least because the
deadweight cost of another reorganization is NOT low. The
anti-market camp have to do a much better job of
demonstrating that there would be big savings if we
reorganized the English NHS and no future benefits if we got
commissioning right. Just picking conveniently inflated
numbers about that cost without even debating what part of
it represents a realistic saving is just rhetoric, not
proper debate.

Competing interests:
None declared

In response to debate about the costs of commissioning in the English
NHS, stimulated by my Editorial, may I point out that my original
submitted text stated,
"Costs associated with commissioning and its paraphernalia have been put
at 14% of the NHS budget in one shelved Department of Health report,
according to the House of Commons Health Select Committee (2)." The final
BMJ text had to abbreviate a complex argument, on grounds of space. I
think it did so very well. While some may suggest that 14% is hyperbole, I
certainly think that, if so, it is most instructive hyperbole!

Yes, it was one estimate; there have been others (all informed
guesstimates, for the reasons the House of Commons Select Committee points
out - themselves a cause for concern.). But the Department of Health
presumably (and, if so, rightly) felt embarrassed enough not to want to
publish it. The costs of the chronically-mutating PCT system and other
'here-today-gone-tomorrow' commissioning bodies - and resulting
administrative and transactions costs in providers as well, throughout the
system - are very high. That is one reason why the costs of commissioning
are embedded more widely than in commissioners (ie its 'paraphernalia' as
well as its official costs.)

To say, 'ah, but we can make it better' is a tired old response,
which the arguments in my Editorial are intended to confront. If
purchasing/commissioning cannot achieve much in twenty years (such that it
needs to be 're-born', according to one of the authors of the Nuffield
Report) and is also very expensive (which it undoubtedly is), I rest my
case. And consider this: defenders of a 'market' NHS are allowed to waste
billions in periodically re-inventing a square wheel, whereas their
opponents are not allowed a scintilla of that luxury. To paraphrase Lord
Keynes, practical people believing themselves to be free of ideology are
usually slaves to some defunct economist!

There is sometimes a whiff of attempted smear in some defences of
'commissioning' - that, if you oppose it, you want to go back to some dark
age when hospitals were a conspiracy against the public. This is
ahistorical spin. Yes, the NHS pre-market (ie pre-1991) had problems -
nothing like as major as Mrs. Thatcher wanted people to believe, but some
technical problems about how the money should follow the patient, and some
problems of coordination between different providers on the 'patient
pathway.' The 'purchaser/provider split' (which eventually gave birth to
the sanitised word 'commissioning') was a major distraction from solving
these problems.

Do not forget: commissioning is now an industry wrapped in an
orthodoxy born in an ideology. Those who defend it sometimes offer a
sleight-of-hand...that there must be planning to meet need, therefore
there must be commissioning. If the latter is merely a word (Alice in
Wonderland-style) to mean the former (applicable to any public health-care
system), I can live with it. If it means what it means in the English NHS,
it should not be re-born, but should have been strangled at birth!

Calum Paton, 7th April 2010

Competing interests:
None declared

Calum Paton's article arguing for the abolition of
commissioning contains the seeds of its own refutation (in
addition to giving a misleading view of the current
context). He wants to have his cake and eat it too: he sees
little evidence of the benefits but also asks why
commissioning is so weak. Perhaps the problem is we have
never done it well and that is why the evidence is weak.

And some of the arguments are just plain nonsense.

The idea that coordination will be better in an NHS with no
split between commissioners and providers is tempting but
only if you have no memory of the past or experience of much
current operational reality in NHS organizations. The old
NHS found it hard to get effective co-ordination when there
was no split; many individual hospitals even today fail to
achieve continuity of care for individual patients when more
than one person is involved with their care (the number of
serious medication errors alone is shocking and these
frequently result from a simple failure to share treatment
and prescription data effectively between individuals and
departments inside a single building). The old monolithic
NHS failed to achieve Kaiser-like gains in efficiency and
quality. Networks of competing organizations in the private
sector are actually better than the NHS at getting
coordination and cooperation to work and few firms achieve
better results when they own their suppliers than when they
coordinate a network of competitors.

Paton also misrepresents the evidence about transaction
costs being 14% of the NHS budget (this implausibly high
number sounds like more than the NHS spends on PCTs, SHAs
and all hospital managers combined). But it isn't an actual
estimate based on the Health Select Committee's work: the
committee quoted a number from an unpublished Department of
Health report which quoted the 14% as a worst case. That is
quite different from a 14% average.

If we put the right people into commissioners (which might
mean putting more money in or perhaps using fewer, high
quality people than the current system which seems to
confuse "cheap" with "value for money" when it comes to
people) we could dramatically improve the effectiveness of
the system. Good commissioners could even lead the drive for
better coordination across the health service.

It would be better to try to make commissioning work than to
reorganize the system again in pursuit of the mythical gains
from integration.

Competing interests:
None declared