Is the Royal College of Physicians really a trade union?

The Royal College of Physicians has published its census results.  And it makes intriguing reading.  But for the full extent of intrigue you have to read between the lines, as well as the lines itself.

It is interesting that the press release makes for really morbid reading.  There is not one bit of good news in it.

I was close to topping myself, and going and buying all my consultants a round of baby-cham (not in that order, you understand, as that would be ridiculous).  But before I did so (either of those things, in the appropriate order), I thought I would actually read the report, and come to my own conclusion.

Here are the three things that struck me:

1. The feminisation of medicine is striking (chart C13b on page 27)
2. Clinicians do not feel that the overall quality of care has changed much over the last 3 years (chart 32d on page 65)
3. Clinicians enjoy their job - over 80% of them enjoy their job always or often (chart 33a on page 66)

Now, you would not have found any of those messages in the press release - which was all about gloom, and more gloom.

Why is that?

The only conclusion I can come to is that the RCP is trying to make a political point.  As justification, your honour, I pray leave to submit the following exhibits as evidence:

• The focus on the NHS.  Why would this not be a general survey of how all their members work across potential employers - including NHS, private hospital groups, academic institutions, other commercial institutions.  This is not so much a census of consultants and registrars, as much as it is a data gathering exercise on the experience of members with one particular employer.
• Within the NHS focus, focus on the contracted vs actual hours worked.  I am intrigued that BUPA insurance has kept its private reimbursement limits for consultants fixed for a substantial length of time.  But this gets at nothing like that - all about the NHS and programmed activities.  I spend my life telling managers not to focus on PAs.  Doctors are professionals, and they focus on patients - not on filling a hourly timesheet according to their PAs.  But the RCP seems to be wanting to focus on this.
• Question choice.  So everybody knows that by introducing EWTD one would have expected continuity and training to suffer - that is the logic.  The reason you would do it is because doctors are more alert and fresher to do care for patients when they are actually working.  But it is training and continuity that the survey focuses on (charts C28a, b and c); not on the alertness etc.  And indeed, if one wanted to get to the bottom of this issue, one would also come at it differently - what are the total hours worked by consultants including their other commitments (private, academic, etc).  It is only by looking at this that you can fully evaluate the ETWD.
• Reporting style.  I have also shown the negativeness of the reporting.  But on each question, the glass is always half empty.  For example, in the real world the compliance of a policy with law is always greater than the compliance of working practice with policy.  Only in the perfect world is this not true.  So to report that "29.6% of departments do not work EWTD compliant rotas in practice – despite 94.7% being compliant on paper" without context or comparison is to scare the horses.

So, my advice to consultants is to cancel either your BMA or Royal College subscriptions - as they effectively seem to be doing the same thing.

Militant Manager's prescription for Public Sector Pensions

Readers of this blog (all the millions) will know of my general opinion on the recent dissatisfaction with the government's proposed changes to public sector pensions.  If you don't, then perhaps you should have a look at my earlier blog post.

I am, however, not merely a reactionary.  I am one of those people who spends time doing thought experiments.  And being a middling NHS manager, my thought experiments are on things like "What is the best form of public sector pensions?"

Intrinsically, I do agree with the idea of defined benefit pensions.  But the key question is: if that is so, what are reasonable rates of accrual?

To start with, I can tell you what is unreasonable.  Take the recent Daily Telegraph report by Laura Donnelly and James Clayton: this showed NHS managers with astronomical pension pots (in the same order as Fred Goodwin - with whom many share knighthoods).

So, I am very unconvinced that pension levels should be so critically driven by salaries - at the top end.  Take a senior employee on £240,000 salary and very close to retirement (do some consultant bodies support particular eminent doctors to become Medical Director shortly before retirement so that their final salaries can be boosted?).  Now each year of employment at that level will boost his annual pension by £4,000.  £4,000 inflation-linked, central government backed pension increase would be worth (easily) £100,000.  So the pension is equivalent to an additional £100,000 income to the employee.

Thinking of this another way - the whole point of high salaries is that if high pensions are important to you as a person, you have the means via the salary of buying such a high pension via your own means.  The in-employment reward should be transparent and shown in the salary (and a 40% supplement to your salary - as a £100,000 pension pot contribution would be - should not be a footnote.  If these people are worth £340,000, let us say it and publish it.  [Take it from me, they are not worth it].

That is where it breaks down the most - at the high end.

So my prescription is that there is a cap on pension contributions from the employer.  Either this is an annual pension contribution cap (e.g., £12,000) or the salary level on which defined benefits are calculated are capped (say at £48,000).  Under the first scenario, the employer could only contribute up to £12,000 per employee per year (whatever the other rules are), and the accruals would have to worked from there.  The second scenario applies say where each year's accrual is fixed with your salary up to a maximum.  Thus if an employee accrues at the rate of 1/60th of salary; if the salary is above the maximum (say £240,000), then the accrual is fixed with reference to the maximum of £48,000 - so the employee only accrues £800 rather than £4,000 as the defined annual benefit on retirement.

This would make public sector pensions more affordable.  And I commend it to the House.

Maternity's flaw is not midwives or quality, but economics

Maternity care is much in the news.  Many of MM's readers would have followed the BBC Panorama report on London's maternity care.  The programme concludes that 17 deaths could have been "avoided" in London in 2009.   Yet, this is not new.  There have been stories about the difficulties faced in and by maternity departments for a long time.  The question is why so much attention and is it deserved?  Is there something underlying that is at work and needs fixing, or can it not be avoided?

It could be because of a number of unavoidable reasons.  It could be related to the primordial nature of maternity.  It is when a new life emerges into the world; where our urge to procreate bears fruit.  The attention could be a result of the scale of loss when obstetrics goes wrong.  After all, nobody is “ill.”  And both mother and baby are relatively young.  If these were the reasons why maternity gets so much attention, then there is not much you can do to avoid it.

But MM thinks there is an avoidable problem that underlies these issues.  It is not easily seen, but lies there and manifests itself in poor staffing ratios; lack of doctors on the ward; high rates of vacancy and agency; poor infrastructure and so on.  All of these issues then give rise to problems and incidents.  This fundamental problem is economics; money; lucre - whatever you term it.  Maternity care does not pay for itself in the NHS. 

That may be a bald statement to make, but can be backed up by a detailed review of the economics of maternity.  Militant Manager has built an economic model of a maternity department to illustrate this economic problem (available on request).  At 4,000 births, the model concludes a Trustwill lose c. £2.8m pa, and £3.7m at 6,000 births.

The model was built using publicly available staffing and productivity guidelines.  Many of these are from the Royal College of Midwives (RCM) and the Royal College of Obstetricians and Gynaecologists (RCOG).

Some of you may be thinking that the sums do not work because MM has taken the guidelines from those with a vested interest.  The Royal Colleges are hardly going to suggest meagre staffing.

MM would, however, disagree that this is a fundamental problem – partly because some of the numbers makes broad sense; and partly because the producers also have interests which moderate the financial call.  For example, the RCOG guidelines suggest that units with 4,000 births should have a consultant on the labour ward for at least 40 hours per week and those with 6,000 should have one for 60 hours.  This is ludicrous.  Firstly, it shows an attempt at reflecting the economics, and moderating the call for the number of consultants by paying attention to activity.  If such an attempt was made, then they should have gone the whole and way, and modelled it out.  They would then have realised the parlous state of maternity finances. 

Secondly, and more importantly, each birth is a birth. Should the RCOG not focus on the care given to each birth, and treat it equally?  Why should some babies have a better chance of obstetrician cover?  It is a bit like when a job applicant answers the question “why should we hire you?” with an answer that talks about what they can get out of the job, and how they will enjoy it.  To be frank, that is not very relevant to the company doing the hiring.  The question is how will the company benefit - and in this case, the baby.

Lastly, and most importantly, how can a clinical (not managerial, efficiency or productivity) standard be any different from either zero or 168 hours cover a week?  Births do not follow time patterns, so if a clinically safe birth can happen without an obstetrician on the labour ward, then it is safe that all births happen that way.  Conversely, if a clinically safe birth cannot happen without an obstetrician on the ward, then no births should happen without that.  There is no space for a fudge. 

MM believes that the issue is that people like the RCOG try and solve the fundamental problem of economics in other ways.   The economics does not support great consultant cover; but rather than recognise this fundamental fact, these bodies try and nudge up standards a little at a time, and put the pressure on Trusts to solve it.  Other attempts act in a similar way: for example, the CQC’s publication "Towards Better Births" which aimed to show the distribution of maternity provision, and nudge up standards.  And another is the BBC's Panorama survey.

Fundamentally, however, the issue is that the economics do not add up.  No private provider is working hard to become an elective or AQP provider of maternity care - they are for orthopaedics and other areas.  No Foundation Trust is building spanking new maternity wings - they are building Cancer Centres.  Lots of Trusts are rationing care, and closing their list to non-local geographies.  The numbers do not work.

When the economics do not add up, Trusts do not invest in staff and resources to give good care.  As all good economists know, economics drives our behaviour - even when we do not realise it.  Trusts do not mean to do it; but they do it subconsciously.  The end result is that care suffers, and we get headlines.

The best solution would be to drive up tariffs. 

Why the unions should stop moaning about pensions

Public sector pensions are very much in the news.  The government is seeking to revise future entitlements on the basis of John Hutton's report.  Some unions are mobilising against these reforms.   Others continue to negotiate.

Militant Manager is very interested in the arguments put forward by the unions.  The views can be easily gleaned.  For example,  Mark Serwotka, General Secretary of the Public and Commercial Services Union, has commented in the Guardian.  And Jon Restell, Chief Executive of Managers in Partnership, a start-up union set up by Unison and the FDA (formerly the First Division Association - representing the senior civil service), has commented in the HSJ.

The arguments seem confused.  I am not talking about John Restell's "six reasons" managers should look at hard reality - which are in fact just one reason: "these reforms are a reduction in entitlements."

I am talking about the central issues that they put forward: pensions are affordable, and public sector pensioners are not fat-cats.  The first argument is arguable - and I have not seen a definitive analysis on this.  The second is a fallacy.

Affordability

First, the question of affordability.  Both Mark Serwotka and Jonn Restell make much of the Hutton Report's (the one on public sector pensions - not the white wash on WMD in Iraq) analysis of the projected benefits to be paid as a percentage of GDP.  This chart reproduced from pg 23 of Lord Hutton's report is so central to this point that I reproduce it below, and can be retrieved here.

I am not yet sure what to draw from this chart.  The key idea conclusion that the unions have drawn is that public sector pensions will absorb a smaller proportion of GDP, and is therefore affordable.  I must admit that the government has handled this argument badly.  For me, however, the issue of affordability is not so clear cut.

I need more information than just this to draw any conclusions.  First, I want to know what proportion of the population these pensions are supporting.  If they are supporting a rapidly reducing share of the population, yet the proportion of GDP is not falling as fast, I may still draw the conclusion that it is unaffordable.  

On this question, the data is indicative that the number of people supported by public pensions is falling. I could only find data going to 1992 (on the inpenetrable ONS website) for data on the proportion of the workforce employed in the public sector (this is a useful indicator as given 40 year working lives, this would predict the proportion of new public sector pensioners in 20 years time). Their feature on Public Sector Employment, 2006 shows that in 1992, 23.1% of the employment was in the public sector.  Whereas, the same report says that the figure in 2006 was 20.2%.  So public sector workforce proportion has fallen from 23.1% to 20.2%, a drop of c. 15%.

This drop of 15% in public sector employment proportion does not seem to be replicated in a commensurate drop in the load on GDP identified above.  You only get that sort of drop if you link the 1992 workforce figure with the high load on GDP seen in the 2010-2020 decade above.  This may be legitimate if the 1992 workforce figure was also a peak; but if it was not, then you are not comparing apples with apples. In short, I am not convinced on this, and would need somebody from a statistics programme like More or Less to have a look. 

I also want to look at other data.  For example, what proportion of pensions as a whole (including private and state pensions) are public pensions forecast to take up into the future?  And what proportion of government expenditure is it due to absorb.  These would also point to the affordability - indicating what proportion of the country's desire to spend on pensions or public expenditure, public pensions absorbs.

Fat cats

The second argument marshalled by the unions is just plain incomplete at best, and wrong at worst.  Both leaders make much of the fact that the average public sector pension is very low: Mark Serwotka quotes that average pension "is just £4,200 a year"; Jon Restell states “the ‘gold-plated’ pensions of the public sector are a myth (median women’s NHS pension is about £3,500 a year).”

You do not have to be a genius to see the careful wording, and selectivity, in those quotes.  There is no mention of the years contributed for that sort of pension.  We need those facts to put it in context.  If that is 40 years, then public pensions are indeed not gold plated.  But if the years of contribution are 2, then we are at the other extreme.

The other issue is that we are not talking about the individual here.  We are talking about the average pension.  So if the individual has changed jobs and gone onto another public pension scheme, or indeed a private pension scheme then they may be getting more than one pension.

But the main issue is that the argument is plain wrong in composition.  The government has already stated that those on the lowest pay will not get affected.  If we are meant to draw the inference that reducing entitlements, will affect these small pensioners – that is the very guarantee that the government has already given will not happen.  To hark on is to show that you are not listening.

How to really look at it

So how should we really look at this?  My view is that we should look at it in the context of occupational pension schemes – which is what public sector pensions are.  In this context, there are two key issues: first, what has been happening to occupational schemes as a whole; and secondly, what has been happening to the employer’s own finances.

In terms of occupational schemes, there has been a secular trend in reducing pension entitlements, and shifting towards defined contribution schemes.  This can be seen from the IFS’ Green Budget 2011 report.  A chart from that report is reproduced below, which shows private defined benefit schemes dropping dramatically.  In this context, public schemes are bucking the trend.  

So there is no surprise in public sector pensions feeling the pinch.  All occupational schemes have.

The second part is that a generosity of an occupational scheme is related to the success of the employer, and the employer's own finances.  If we have not seen it already, the government's finances are in a mess: we were running a structural deficit even before the fracture in the markets since 2008.  So, employees of the public sector should expect a smaller pension, rather than jump up and down.

Why consultants need more natural predators

Without natural predators, growth in consultant numbers is uncontrolled.  This is the conclusion you reach when you look at the training numbers of doctors through an "ecological" prism: where doctors are the fertile mothers, and registrars in training are the juveniles who will grow to become future doctors.

Replacement fertility measures "the total fertility rate at which newborn girls would have an average of exactly one daughter over their lifetimes" who would go onto have further children. In more familiar terms, women have just enough babies to replace themselves.  It is a bit sexist to define this in terms of women alone, but as 50% female, Militant Manager does not mind.

In a stable ecology, replacement fertility is determined by a number of factors.  First is annual survival rates - the chance that you may die each year (predators, lack of food, disease etc).  Second is the age at maturation (the age at which a female can have children).  The lower the rates of survival each year, and the greater the age at maturation, the greater the replacement fertility.  And this equilibrium is stable because if numbers increased, then survival rates would drop as the population became easier to hunt, or they exhausted their food supplies.  A picture of a cow with an element of its replacement fertility is included for completeness below.

  

On this basis the replacement fertility for hospital consultants is 1 - i.e. consultants only need to have one registrar in their career.  And given 30 years of consultancy (which may increase if retirement is pushed back), and 6 years of training, the stable ratio of consultants to juveniles is 5 (this is 30 years of fertility in the consultant, divided by 6 years of maturation in the registrar).  The child bearing years are so long, the maturation age is so early, and the survivorship is so good, that in a stable population there would be 5 consultants per registrar.

And what ratio do we have today?  We have 5 times what we need, as illustrated in this spreadsheet, and summarised in the table below. 

Of course, the model needs to be adjusted for the more complicated reality.  People retire early, or change career.  There are some who go part-time (often the case for women); and others who go "private" entirely or in part.  There are even some who go to the dark side and become managers.

But these complexities only change things at the margin.  The replacement fertility ratio in humans is higher than the theoretical 2 (the theoretical figure).  This replacement fertility ratio ranges from 2.1 to 3.4 depending on country.  And similarly - though Militant Manager has not seen such an analysis - the stable ratio of consultants to registrars may be around 3 to 5:1.

Yet, we have a ratio of 1:1.  My spreadsheet is based on numbers published by the Information Centre.  This shows how the numbers have deteriorated over the last few years.  One can also see that vacancies for consultants are at negligible levels - the levels at which changes in recruitment practices in Trust will affect the numbers.

The more astute amongst the readers will be pointing out that I have not considered the growth that would be required.  That is by design.  The reality is that in a stable advanced economy, one cannot increase the workforce engaged in a sector without also saying in the same breath that that sector will also account for an increased share of GDP.  The UK already spends 8-9% of GDP on healthcare, and I do not see a groundswell to increase this further.  In fact, national policy is to reduce the share of GDP healthcare absorbs by keeping public spending on it constant as GDP grows.  So there are no grounds to independently plan for rapid growth in doctor numbers.

So what does this all mean.  I have four main thoughts:

1.  NHS Workforce leads should be sacked.  Not only have they allowed this to arise, they are asking the wrong questions.  Militant Manager has searched for years for strategic analysis of training numbers - what numbers do we need; and what drives the replacement ratio.

2.  There is going to be a lot of pain for trainees.  There was - not so long ago - a role for "senior registrars."  They were effectively registrars who had all the requirements for consultancy, but had to do "their time" waiting for a job to come up.  It was not unusual to be in such a stage for 5 years.  Those days are coming back.  And the 5 years will be used to erode, demoralise and reduce the registrar base.  Many will leave the profession, and lots of others will seek other ways to burnish their CVs to distinguish themselves in an the increasingly competitive process of becoming a consultant.

3.  We need a better plan.  Juniors now do a lot of activity in hospitals; and many rotas require them.  So in some respects there is a requirement for this number of registrar level doctors.  But we do not need those numbers "training" and aspiring to consultancy - an aspiration we cannot meet.  That will require a reduction in training throughput, and a change in the stable workforce mix in hospitals - away from "training" posts.  This is a task that will have to be picked up by the new HR directors who replace the sacked ones.

4.  We need more natural predators for consultants. Ideas include taking consultants and leaving them in the serengeti; encouraging more to go camping in Siberia; and introducing diving in the shark infested waters of South Africa as a training requirement.  

Readers will have better ideas, and I would welcome those.