I note Gary, my regular commentator from PWC, is commenting tonight on public sector productivity and how it, supposedly declines as the scale of state sector spending increases. Doesn’t that just show that this man is a consultant, knowing the price of everything, the value of nothing and computing ratios between apples and oranges and delivering an answer in cooking cherries?
To explain, let me explore for a moment why it is inevitable that state sector productivity falls as the scale of state sector spending rises as a proportion of GDP.
Productivity is, of course, a ratio. It’s a ratio of labour to something else: usually capital employed. Labour could be the number of people. Normally it’s their cost.
And in private sector business productivity tends to rise with scale because as a business grows it tends to become more capital intensive, more systematised, more automated, less tolerant of variance and with a lower cost of capital, as is commonplace in larger business, more likely to substitute capital for labour. This means productivity usually appears to be higher in large business than in small business. The result is consultants assume big is good as labour cost per unit tends to fall and they call this enhanced productivity — somewhat ignoring on the way that the cost of each job created as a consequence can be astronomically high as a result.
But the state is different (how often do I have to say that?) First of all because the state tackles the easy targets first (which is a good thing). These tend to be the projects with economy of scale built in. So teaching children in large classes looks really productive. And tackling relatively common and easily treatable illness is obviously efficient. So the state does these things when it first has programmes of public education and health. And quite right too. That’s the basis on which the welfare state was built, and I’m delighted it was.
But as income rises (which it does when states invest in things like the welfare state) so too do expectations. So whilst we were once happy to teach children in classes of forty knowing many would leave school at 15 without qualifications and that they’d find work in our economy all the same we’re not happy with that now. We want them to have higher qualifications. But that means they’re taught for longer. And in smaller classes. And because labouring has disappeared (by and large) from our economy we require that everyone can read and write and be numerate, so we provide remedial teachers and classroom assistants and so have more than twice the staff in a room with fewer pupils in it. Outcomes, of course, are better, collectively and individually (and they are — of that I am convinced). But labour to capital ratios collapse. Productivity falls dramatically. But we’re all better off despite that fact — and yes I mean we all are (I refer to the Spirit Level principle, of course).
It’s the same with medicine. As my wife says as a GP, once upon a time, not long ago, treating heart attack was easy. You turned up. If the person was still alive and they said they’d had crushing chest pain you as the doctor said they’d had a heart attack and you went away again: there was nothing else you could do but record it in the notes. It was the same with stroke. And when you got cancer by and large it too ran its course, with ever larger doses of morphine easing the process. But that’s not true now. Heart attack requires immediate action, and follow up (if the patient is still alive, of course) and long term preventative care, and ongoing scans, and then cholesterol and blood pressure management, and on, and on and on. Walking away in the old fashioned way was really very efficient in productivity terms. Doctor to patient time on a heart attack case was low. Now it is high. Outcomes have improved. But productivity collapses.
Now of course we can go back to high productivity. But letting people die is one requirement of doing so. And don't look after the elderly is another — let them fall victim to the first bout of pneumonia they get is great for productivity (and one reason why it is no doubt so good in the US where healthcare is not in the public sector). And also let’s go back to the 3Rs in big classes and leave a lot of children unable to do anything (which despite all the claims from the right wing is not what I witness to be happening, by and large). You'd improve productivity — and massively shrink well being at the same time.
In other words — lower state productivity is highly likely to be associated with very good outcomes for the well being of populations paying for high levels of public services who quite rightly don’t give a damn about what consultants say about productivity — precisely because they instinctively that such consultants are using the wrong unit of measure — but who do really enjoy the enhanced life prospects for the young, old and all in between that strong social, educational, health and other services bring when a high proportion of GDP is expended upon them.
PWC have got a lot to learn about the real world. And even about ratios. And the power of doing nothing because, as I’m also told by many doctor friends, the hardest thing they often have to decide upon is when not to act — when doing nothing is the right course of action. That looks like crass lack of productivity, except for the fact that the human body sometimes has an extraordinary ability to put itself right that the doctor can impede. How is that measured in productivity terms? It can’t be. But it remains the right thing to do.
Listening to the likes of Gary is the wrong thing to do if we want to increase wellbeing for all but PWC. And I’m pretty sure most people would go for the more egalitarian form of that equation. So shall we leave PWC and its ilk aside and go for low productivity and high outcomes?
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
Apologies if I gave you the impression I was speaking on behalf of my employer, even though I made no reference to them. I am not seeking to represent them, and they may not share my view, etc etc.
I normally put this disclaimer on my comments now, but I see I neglected to do it on this occasion. Lesson learned.
Anyway, back to the issues: *I* was saying nothing about productivity. The *ECB* were. Dismiss them if you think it right, but no need to shoot the messenger. I was simply sharing data that suggested that the link between higher taxes and ‘better’ public services was not a given as you had originally suggested. The King’s Fund have made a real effort to incorporate the complications you rightly identify, but ongoing increases in productivity in the services sector suggests that productivity increases are possible although they are not being delivered.
PS: I am speaking in a personal capacity, and not on behalf of my employer, etc etc
On the basis that a large amount of increased state sector spending has been on consultants, surely if they were any good, then productivity would increase – I’m sure that’s the basis on which they sold their services. Figures recently made available show that consultancy spend by the Goverment is running at £1.8bn per annum:
http://www.guardian.co.uk/politics/2010/jun/04/coins-database-government-spending-consultants
Now take a look at this…
http://www.independent.co.uk/opinion/commentators/johann-hari/johann-hari-the-management-consultancy-scam-2057127.html
Now, lets think what else can be done to increase productivity and help in reducing the deficit 😉
Another classic cause of this would be performance measurement – usually rebranded as ‘quality’ assessement – which consultants from the likes of PWC have been instrumental in foisting ‘metrics’ and ‘models’ for across the public sector. Of course, this was initially at the behest of New Labour but the desire to measure every input and outcome rapidly became a gravy train for consultants and advisers, thereby creating an army of ‘quality’ administrators in every public sector organisation in the process.
Quality assessment and assurance are an essential part of the operation of any organiusation, of course. But what we now have (and I speak from 15 years experience of higher education) are many so called quality measures that I doubt add much, if any, value to the service being delivered. They do, however, keep many, many people busy, either collecting, recording and inputing data, or making up ways quality measures can be met (legitimately or otherwise). This includes anything up to and including creating new units and departments for develping, implementing and overseeing ‘quality’. In short, a new industry has been created within the public sector which adds very little in terms of direct outputs, thus lowering productivity. Gary and his ilk have taken a key role in developing and sustaining that system.
@Ivan Horrocks
Entirely agreed
@Deeply Depressed
Again, entirely agreed
There is also the other reason for ‘low productivity’, which is that it employs more people.
The state is the yang to the private sector ying, and it needs to do the opposite. So while the private sector is leaning up, the public sector needs to bloat a little.
Time to hire less JCBs and more people with shovels.
Public sector productivity can be looked at when the economy is booming.
@Gary Taylor
Oh come on…you chose the message to deliver. No one asked yo to do so
For heaven’s sake accept a little responsibility for your actions
@Neil Wilson
Neatly put
Gary, stop being defensive and just accept that the argument you’ve “reported” has been comprehensively demolished by one of Richard’s most incisive and relevant blogs.
Richard, this piece demands the widest possible airing and support. Have you considered forwarding it to the prospective Labour leaders and perhaps those libdems whose social conscience remains intact?
@nick james
I met a ConDem minister yesterday
I promise you, they read it
Surely the things to aim for are both high efficiency and high productivity?
Saying that lower productivity is related to a higher number of employees merely means that there are more employees than there are jobs for them.
If we all agree that State spending must increase during a deflationary phase in the economic cycle than what is wrong with aiming to ‘shoot the moon’ with productivity?
Just throwing money at a department does not guarantee that it will spend the money wisely or efficiently which I believe would have been Garys’ original point: Increased spending doesn’t mean productivity increases or that outcomes will be better.
In the Private sector inceased efficiency and productivity are actively chased during the lean times. If the public sector did this and could say; “Look, we are doing the same job as we always have with less money. Imagine what we could do with increased investment” would this conversation even be taking place?
“Productivity is, of course, a ratio. It’s a ratio of labour to something else: usually capital employed. Labour could be the number of people. Normally it’s their cost.”
I had always been under the assumption that labour productivity is calculated by output (call it ‘value added’ if you like) for a given input (e.g. hours, £ per hour, whatever).
I had understood labour productivity is ASSOCIATED with capital employed. For example, one man working on a tractor will have more productivity for his 8 hours (i.e. will produce more output per hour) than 3 men working on the same tractor (i.e. same output but spread over a 3x bigger denominator).
But it seems I have now learned something new, so thank you for that.
Others who also seem to be getting it wrong:
Wikipedia: http://en.wikipedia.org/wiki/Productivity
OECD: http://stats.oecd.org/glossary/detail.asp?ID=4819
ONS: http://www.statistics.gov.uk/hub/economy/prices-output-and-productivity/productivity-measures
ILO (as an example document I googled): http://www.ilo.org/global/About_the_ILO/Media_and_public_information/Press_releases/lang–en/WCMS_083976/index.htm
Neil Wilson:
“Time to hire less JCBs and more people with shovels.”
What do you think will happen to real wages if we do this? Do you think they will stay the same? I suggest they would plummet.
As a historical precedent, countries behind the Iron Curtain did this for decades — high ratio of labour to capital. Countries in the developing world (e.g. India) are still doing it. In plain language, lots of workers, few tools, compared to the West.
How do wage rates compare with workers doing the same thing in the West with more capital per worker?
@nick james
Nick you clearly do not accept that data as persuasive, and in your opinion Richard’s analysis is far superior. I accept that.
If I appear defensive then I obviously wrote poorly. I was simply suggesting that who I work for is not relevant to the data I shared and the conclusion the ECB drew.
@Deeply Depressed
So you say public sector spending CAN be reduced without impacting the quality of public services? Great! Richard agrees too. Consensus!
@Adrian
Very politely, didn’t I say it was a ratio?
And, didn’t I also explain why in the private sector the ratio (output to cost etc which you prefer) is always a proxy for the labour to capital ratio? Or did you not notice that?
And didn’t i also explain that as we become richer we value the diversity of humans more, their individual needs more precisely, their need for care more accurately, their capacity to take advantage of medical advance more especially|? But that in each case output to input will be lower if measured conventionally. For example – treating an old person for all the conditions they have does not alter the fact that advancing age is an indicator of likelihood of death – which is inevitably going to happen.
So using your crass indicators of output not treating always enhances productivity since death will happen
And likewise not teaching children with any form of particular need will always increase teacher productivity
But the output measure you propose ignores value
It’s wrong in other words
But you’re too blind to notice
@Steven Chiverton
“Saying that lower productivity is related to a higher number of employees merely means that there are more employees than there are jobs for them.”
One on one care of the elderly with dementia is intensive work
Do we stop it because there are more people with dementia?
Each persons output is unquantifiable
You’re saying therefore they are doing non jobs we don’t need?
I often wonder what world people like you live in
@Gary Taylor
Yes
We agree
We could cut PWC out altogether and increase productivity
But that’s because it’s obvious what you sell has limited value
Now deal with my dementia question, above
@Gary Taylor
Yes – by £1.8bn at the stroke of a pen. Let the public sector manage itself without the drivel and “blue sky” codswallop peddled by the Big 4, CapGemini, Accenture et al.
“Very politely, didn’t I say it was a ratio?”
Yes, you did. You said it was ‘usually’ a ratio to capital employed, whereas I suggested it was compared to output (or value added as I prefer). Capital employed has nothing to do with it (except that it is a variable that affects labour productivity).
“And, didn’t I also explain why in the private sector the ratio (output to cost etc which you prefer) is always a proxy for the labour to capital ratio? Or did you not notice that?”
Sorry, I must be thick.
Paul McCartney can spend 5 minutes tinkling out Yesterday on an acoustic guitar.
I can tinkle out My Old Man’s a Dustman on the same guitar. Same 5 minutes.
Same labour input – 5 minutes. Same capital – one guitar. Identical labour/capital ratio. Somewhat different output. One generates £millions, the other won’t generate a penny.
To suggest it is ‘always’ a proxy doesn’t ring true in that example.
“And didn’t i also explain that as we become richer we value the diversity of humans more, their individual needs more precisely, their need for care more accurately, their capacity to take advantage of medical advance more especially|? But that in each case output to input will be lower if measured conventionally. For example – treating an old person for all the conditions they have does not alter the fact that advancing age is an indicator of likelihood of death – which is inevitably going to happen.
So using your crass indicators of output not treating always enhances productivity since death will happen”.
Sorry, don’t understand this at all! Becoming richer means there are more resources available to provide ‘value added’ (i.e. the care required for the elderly person). What difference does it make that death happens? It is the value provided whilst he is alive.
“And likewise not teaching children with any form of particular need will always increase teacher productivity”
Huh? Allowing teachers to teach more productive means more resources available to allow teachers to be allocated to a teach a child with need. Same with the Alzheimer’s example. If we become more productive in doing other things, it frees up resources (which have competing needs) to allow the care to take pleace. A poorer society just won’t have the resources available.
“But the output measure you propose ignores value”.
No, I don’t! I suggested that output is really value added. And for an employee, it is what is valued by the employer who pays the wage. Whether you or I give it different values makes no difference if we aren’t paying the wage.
it would help your argument if you got the definitions right. Productivity is more properly defined as a measure of outputs compared to inputs. Your definition restricts itself simply to the relative amount of inputs.
@Richard Murphy
I think you misunderstood me Richard.
Maybe a little background; I’m self employed and run a small shop on the Isle of Wight. My previous partner worked for first HMRC and then Southampton City Council before she was tragically taken by cervical cancer (she was only 29). I understand all too well that caring for people with terminal illness’ and conditions like Dementia are painstaking and not easily quantifiable in terms of outcomes.
I have the greatest respect for people in positions like that and have benefitted from thier dedication and professionalism. I would also add that in my business we regularly run collections and charity events for the local hospice and health care trust in thanks for the work that they have done.
I appreciate that you did not know this but perhaps you should be less willing to throw out random attacks on those who broadly agree with you, especially when not in full possession of the facts.
But leaving aside these highly emotive issues, which in no-way account for all public spending, what I was refering to was the assumption that we can ‘stop hiring JCBs and start hiring people with shovels’. That’s make work, and doesn’t improve anybodies lot. Even the person given the shovel may soon wish that the council had hired a JCB after a couple of days of digging holes especially since the JCB is much more efficient.
Isn’t it right that those who pay taxes should expect that they are spent in the best way possible and not be wasted? It is not the States money, it’s the populace that have given it to the Government to spend for the benefit of society as a whole. In return the State can expect the populace to pay what is deemed to be owed with no evasion. That is the other definition of tax justice surely?
And no, I’m not a Tory, member of the Libertarians or even remotely interested in tribal politics. I’m just asking questions of a blog frequented by learned people in the hope of fostering debate.
@alastair
I did get my definition right
Productivity is a ratio
You can call it, if you want a ratio of output to input. So production out for labour in, if you like
Except implicit in that is a base assumption for benchmarking e.g. employee X is twice as efficient as employee Y – but unless you notice employee X has a shovel to dig a trench and employee Y a trowel that’s irrelevant i.e. all benchmarking ultimately kinks to the ratio of cost to capital – and human capital is also a factor in that
In other words – although you’ve all been dutifully reading Tim Worstall before commenting here – and as ever he’s wrong – I actually have my definition entirely right. Your problem, as ever, is utter inability to think and comprehend beyond the quotes you can find on the web
So now let’s go back to the point I was making, in the process using a correct (but macro rather than ultra micro) definition of productivity – which when we’re considering government efficiency seems wise – and point out you still don’t understand that output is the issue here and it can’t be measured but is none the less utterly desirable – which means your whole ratio is rubbish anyway – which as always my point.
But as ever you seek to divert from that truth by using pedantry – and to it wrong as usual
@Steven Chiverton
I’m sorry to hear your tale. You have my sympathy in your loss
But you did not make your point well – and I reiterate – your story reiterates just why the point you seek to make is so wrong
If we measure productivity then caring for the terminally ill makes no sense. But of course it does makes sense
And it also makes sense to keep people at work when there is NO private sector demand for them
What’s better – millions doing nothing and having their lives blighted or millions doing something – and saving benefit and keeping their skills and work experience going?
Of course that must not continue when there is an upturn in activity
But there won’t be an upturn in activity until they do work
So I’m afraid your logic is wrong
“Productivity isn’t everything, but in the long run it is almost everything. A country’s ability to improve its standard of living over time depends almost entirely on its ability to raise its output per worker.”
—Paul Krugman
@chris strange
Sure – a great maxim for the private sector
We’re talking the public sector
Not the same thing
@Adrian
You entirely miss the point
“No, I don’t! I suggested that output is really value added. And for an employee, it is what is valued by the employer who pays the wage. Whether you or I give it different values makes no difference if we aren’t paying the wage.”
Great for the private sector – which isn’t called capitalism – the return on capital, note – for nothing (oh, dear, that’s my ratio again)
But have you noticed we’re talking about the public sector?
That’s different – which you refuse to recognise
And until you do every contribution you’re making is wrong – because it is quite impossible to define ‘value added’ in most public services as you’d wish to do and I’ve already explained why
@Richard Murphy
interesting comment, but as it happened I worked out your error all on my own!
I would prefer to think of it as a measure of efficiency – and its usually explicitly used as a base assumption for benchmarking. I would agree that measuring outputs is not straightforward, but on the other hand, it is something that Governments do as a matter of course. It is politicians (and single interest groups) who are well versed in blurring the technicalities in order to further their cause.
I guess there is a perfectly valid case to be made about the cost vs benefit of “the public sector”, and in particular the value of government spending in the current climate, but whilst I think that is what you are after doing, the argument you present here is no more than filibuster, and the basic definitional flaws do nothing to advance your case.
don’t see anything particulary pedantic in that.
@alastair
..except there was no definitional flaw
as I’ve shown
You have asserted it, but I don’t think that you have shown it.
On a seperate point, I have some sympathy with the fact that patient demands are increasingly hard to satisfy as expectations rise, but your assertion that productivity was higher in the past depends entirely on what you are measuring. If you are measuring the number of individual patients treated in a given time, then yes, if they are dropping like flies, you can see more individuals in a given time. If you are measuring lives saved, or average life expectancy as a proxy for succesful interventions, then productivity has increased.
@another alastair
Oh come on, all theories are assertions. Stop the BS
More seriously, I note the lovely idea you’re all putting forward that it is only labour inputs we need take into account when measuring productivity. It’s all so very Marxist of you. I guess my trouble is I’m not.
And then I just noticed this on Wikipedia (but what the heck, it seems right):
” The OECD defines it as “the ratio of a volume measure of output to a volume measure of input”.[1] Volume measures of output are normally gross domestic product (GDP) or gross value added (GVA), expressed at constant prices i.e. adjusted for inflation. The three most commonly used measures of input are: hours worked; workforce jobs; and number of people in employment.
Measured labour productivity will vary as a function of both other input factors and the efficiency with which the factors of production are used (total factor productivity). So two firms or countries may have equal total factor productivity (productive technologies) but because one has more capital to use, labour productivity will be higher.”
http://en.wikipedia.org/wiki/Workforce_productivity
Keep to your labour theory of value all you libertarians
Me, I’ll stick with capital being part of the equation
@Richard Murphy
I don’t see where I assert that it is only labour inputs we need to take into account. I was simply following the example of productivity you gave where you suggested that GP productivity was higher when GPs had fewer options for intervention. In this particular example, I was concentrating on the labour input, as you say, but saying that the measure of productivity still depends on your measure of the output.
I note the definition of productivity you cite from Wikipedia, but this seems at odds with your article where you state that productivity “..is a ratio of labour to something else: usually capital employed.” I would view both of these as being inputs.