Are we measuring the wrong type of productivity?

Today gives us an opportunity to look at the latest data on what is the key economic number these days which is wages growth. After yesterday’s inflation data we will be able to look at both nominal and real or inflation adjusted wages growth. The reason it has become a key number is that in countries like the UK ( and US and Japan..) is that the employment situation is strong and recorded unemployment has improved considerably but wages growth has been weak. In the extreme case of Japan there has so often been no wages growth.

An associated influence on this has been problems with productivity as of course it has helped drive wages growth in the past. Whereas according to Bank of England Chief Economist Andy Haldane that happy situation has been replaced by this.

Productivity growth has consistently underperformed relative to expectations, since at least the global financial crisis. This tale of productivity disappointment, in forecasting and in performance, has been extensively debated and analysed over recent years. Some have called it the “productivity puzzle”.

Indeed we have been on something of a road to nowhere.

For the past decade, average productivity growth has been negative. This is unusual, if not unique, historically. You would have to go right back to the 18th century to see a similarly lengthy period of stagnant productivity.

In case you were wondering it compares to this.

there has been a near-monotonic rise in UK productivity. UK TFP growth since 1750 has averaged 0.8% per year. Since the Industrial Revolution, GDP per capita has doubled roughly every 65 years and productivity roughly every 85 years.

Actually some of Andy’s numbers are a little contradictory as he suddenly agrees with the theme on here that things were deteriorating even before the credit crunch.

From 1950 to 1970, median global productivity growth averaged 1.9% per year. Since 1980, it has averaged 0.3% per year.

I find that fascinating because is not that the same period where we saw the influence of increasing globalisation and internationalisation which were badged as bring significant economic benefits?

The United States

The international scale of the issue has been highlighted by the Financial Times today.

US productivity is set to grow this year at around a third of the pace prevailing before the financial crash………..
US labour productivity — a driver of the economy’s fortunes — is forecast to expand 1 per cent this year, an improvement on the 0.5 per cent recorded for 2016 but far shy of the 2.9 per cent growth seen from 1999 to 2006, according to Conference Board projections shared with the Financial Times.

This is true of others as well.

The EU will see 1 per cent growth in GDP per hour, an improvement on last year’s 0.8 per cent but short of the 1.9 per cent seen in 1999-2006.  Japan is on course for 1.1 per cent productivity growth, up sharply from 0.5 per cent in 2016 but still well shy of the 2.2 per cent pace seen before the crisis.

I cannot move on without pointing out that the pre credit crunch figures were inflated in many places by booming housing and banking sectors which then went bust.

the figures lag far behind the 4.9 per cent pace in 1999-2006.

Is it the service sector?

To my mind a large factor in the productivity puzzle has been the switch from actual things being produced to more intangible types of economic growth. If we look at it in a stereotypical sense we see output of cars replaced by output of haircuts or teaching or nursing. The latter is much harder to measure in productivity terms as who wants teachers to be more productive via larger class sizes? It is even worse for nurses as who would want to be in a hospital ward with fewer nurses? The problem here is we need a measure of quality of the output and we struggle to define and measure that. Even worse some areas of production face a future of possible enormous gains in labour productivity by the use of robotics and artificial intelligence but where does that leave the labour? Can we have too much of something that is usually considered to be good.

Looking forwards as Sarah O’Connor points out we are likely to see more growth in the service sector.

The undramatic truth is that many of the jobs of the future are also those of the present. Prime among them are jobs that involve humans looking after other humans. The US Bureau of Labor Statistics has predicted the top 30 fastest-growing occupations for the next 10 years; more than half are some variety of nurse, therapist, healthcare worker or carer. This feels like a safe bet — and not just in the US.

She also points out that this growth will be in jobs that we tend to not value.

Social care jobs, for example, are defined by economists everywhere as low-skilled or unskilled…….Personal care and home health aides in the US make roughly $23,000 a year on average. In Britain, a prolonged squeeze on public spending has had knock-on effects on care workers, many of whom work for private companies that rely on public sector contracts. In England last year, 43 per cent of care workers earned less than £7.50 an hour.

There are plenty of thought-provoking issues here as raising productivity here would involve paying them more as that is the only measure of output we have here. Indeed both GDP and productivity fail us when we cannot measure economic output. On this road no wonder both metrics have problems. If a service sector producer gets more efficient and reduces its price then as money is often the only measure we record lower productivity when in fact things have improved. In other words we are in a something of a mess of our own making.

Today’s data

Not the cheeriest I am afraid to say.

Output per hour – our main measure of labour productivity – fell by 0.5% in Quarter 1 (January to March) 2017. This compares with growth of 0.4% in Quarter 4 (October to December) 2016.

My explanation given above may well work though.

was a result of hours worked growing faster than output;

What about wages?

Growth has been pretty consistent at what seems to be something of a new normal.

Between January to March 2016 and January to March 2017, in nominal terms, total pay increased by 2.4%

It is in fact marginally higher but as we look for real wage growth and note that nominal growth in March was 2.4% we see that it was a mere 0.1% and should it remain the same in April then wage growth will be negative. Of course if we use the RPI then annual wage growth was negative again in March at -0.7%. Sadly such numbers come on the back of a credit crunch era decline.

The Resolution Foundation has a somewhat enduring if increasingly lonely faith in officialdom so it still takes the forecasts of the OBR seriously and has switched to the CPIH inflation measure. I think though like so many places today it was so revved up to say real wages were falling again that it has used the regular rather than total pay data.


There is much to consider here as we find yet another set of statistics that are failing us in the credit crunch era. Our outdated concept of productivity needs to change and it is being challenged at both ends of the spectrum. At one extreme we have the sort of situation covered by Skynet in the Terminator series of firms where robots rule and at the other we have what we might call 100% human occupations. Do we really want to say that one provides a sort of 100% productivity and the other 0% because that is where we are heading right now?

Let me add in another sector which is the self-employed which these days is 15% of our workforce or 4.78 million people. For those in the service sector our main measure of output and hence productivity will be their pay. The very pay numbers that are ignored by the official average earnings data. What could go wrong?

Number Crunching

Regular readers will be aware of my love for football. The numbers game at The Emirates where Arsenal were playing Sunderland had me intrigued. You see pictures of a ground that was a long way from full were all over social media and BBC 5 live reported it was at least a third empty, and yet.

the official number for Tuesday’s game was 59,510. ( ESPN)

Actually Arsene Wenger claimed it was “sold out” but of course he has a long history of eyesight problems and myopia so let’s pass on that. But could we one day see the first empty ground that is counted as sold out?


20 thoughts on “Are we measuring the wrong type of productivity?

  1. If you go to a baseball game in the US where there are 80+ home games in a year it is often half full at best. Most of the seats are bought on season tickets and few people go to every home game. Thus the seats are sold but not occupied. To avoid a tv blackout in the local area the team will often buy any remaining unsold seats so it is sold-out and tv will then broadcast it. These tickets are usually given to ‘good causes’. It’s all about money! Baseball (football) is secondary. Though in Arsenals case wasn’t it a protest about the manager?

    • Hi Chris

      Some of it was about Arsene Wenger although I do not know if a “Wenger Out” banner was towed behind a plane like in the game before. My knowledge of baseball is limited to one game when I watched the White Sox in Chicago some 20+ years ago! The ground was very empty back then although if I recall correctly it was on a weekday afternoon.

      As a fan it is always better to be part of a large crowd don’t you think?

  2. Hello Shaun,

    raising productivity in the services sector, hmm, tricky . some service jobs will just evaporate….

    Frankly as good paid jobs are off shored whats left but the servants, sorry , service jobs ?

    Frankly if 80% of the jobs are service jobs and low paid , who buys the goods & services ?

    the top 1% can only spend so much

    Eventually you will not need teachers , as it would seem impracticable to train so many so highly , be a waste of money won’t it

    Nurses will be replaced by smart apps and AI robots of many types ( not an Asimov future) , same for Bankers , don’t they realize once “cash” is all electronic then there’s no need for Banks? tens of thousands of jobs gone .

    A service economy seemed a good idea when its all clerks , IT support and teccy’s but one of hair dressers , nail bar operatives and wash maidens for the elderly ? ( I was going to put burger-flippers in but apparently that can be automated !)

    We;ve already seem “creative accounting” in CPI/RPI and GDP ( and other HMG stats ) – you can only fool reality for so long …….

    ( snap! )

    no matter what pill you take , dropping out of this “Matrix” will lead to a life that’s dark , miserable, painful and short.

    come 2038 we shall be in a new world ……………. not a very nice one


    PS: yes it can be different but it means being objective and not lying to one’s self about the situation, HMG and ONS please take note!

    • Hi Forbin

      Perhaps the answer is to measure productivity for the parts of the economy that it can apply to but have a different criteria for things like nursing and teaching that require other strengths.

      You raise a good point about service jobs. What if we raise pay and they disappear offshore?

  3. Hi Shaun
    Wenger is a Monreal Hazard at every Matic and
    I Kante understand why he hasn’t got the Mertesacker.

    Or should he become a banker?


    • Hi JRH

      Maybe sacking him would Costa too much and lead to Arsenal needing a Bale out, anyway it has become s Messi. More seriously it is a shame to see a man who did the game a lot of good ( better diets, clearing out the drinking culture) end up the subject of so much ire.

  4. Any economic time series has three components: seasonal, cyclical and secular. You are discussing the secular and there are two factors that could have influenced productivity quite markedly: the one you mention, that is the growing emphasis on services rather than goods ( a “structural” explanation) and the other which is part of the thesis of Robert Gordon and which is related to the strides made in the last 150 years.

    Gordon’s view is that the decline in productivity is explained by the waning influence of the great inventions of the 19th century and the failure to replace them with others of such potency. In effect Gordon’s view is common sense; why should we expect a constant increase in innovations which increase productivity; as he points out there was no economic growth for around 500 years in the Middle Ages so why do we now expect innovations as regular as clockwork?

    If you accept these sorts of explanations, at least in part, then there is no productivity puzzle; the problem lies not with productivity but with our unrealistic expectations.

    • “…the problem lies not with productivity but with our unrealistic expectations.”

      aye thats the rub , isnt it ?

      we have had a economy built on the expectation that we can service debt to infinity and beyond

      economics still doesn’t take into account energy inputs seriously , its not just we;re going from high EROEI to low ( fossil fuels to green ) but also physics limits the great advances too , both easy fruit has been picked.

      the planet wasn’t built for infinite growth ….


  5. I never worked so hard as when I was at home (unpaid) looking after my young children. I don’t suppose that counted in the productivity stats. Perhaps that’s another reason for getting people to take up childcare. It’s a double whammy as the parent gets a job so can be “productive” and the people looking after the childrren are paid so also count.

    • Hi Jan

      GDP only counts things which have a clear monetary value. This leads to ways it gets distorted as for example if you had worked and paid a child minder/nanny GDP would rise but would you have been better off? More than a few prefer to raise their own children or at least as much as they can these days.

      • This seems like a good excuse to quote the 1968 RFK speech to Kansas University:

        “Too much and for too long, we seemed to have surrendered personal excellence and community values in the mere accumulation of material things.

        Our Gross National Product, now, is over $800 billion dollars a year, but that Gross National Product – if we judge the United States of America by that – that Gross National Product counts air pollution and cigarette advertising, and ambulances to clear our highways of carnage. It counts special locks for our doors and the jails for the people who break them. It counts the destruction of the redwood and the loss of our natural wonder in chaotic sprawl. It counts napalm and counts nuclear warheads and armored cars for the police to fight the riots in our cities. It counts Whitman’s rifle and Speck’s knife, and the television programs which glorify violence in order to sell toys to our children.

        Yet the gross national product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials. It measures neither our wit nor our courage, neither our wisdom nor our learning, neither our compassion nor our devotion to our country, it measures everything in short, except that which makes life worthwhile.

        And it can tell us everything about America except why we are proud that we are Americans.”

        It also prompts a reflection of the quality of politicians then and now. How is this reflected in productivity data.?

        • Hi Nixon1974 and welcome to my corner of the web

          It is a good speech which poses questions which remain mostly unanswered in the world of official statistics even now all these years and indeed decades later. I also note that back then they were looking at GNP and not GDP…..

  6. Hallo Shaun, brace yourself……

    You say that Andy Haldane’s numbers are a little contradictory as he speaks of productivity being negative over about the last 10 years, going on to say that productivity averaged 1.9% pa from 1950 – 1970 but was then 0.3% pa from 1980 (I wonder what happened to the stats for 1971 – 1979, perhaps they didn’t fit his theory?) agreeing with your theory that things have been deteriorating before the credit crunch.

    I fail to see the contradiction in his statements. He states that productivity has been NEGATIVE over the last decade and that it has been 0.3% pa averaged since 1980,so that would be positive albeit at a slower rate than 1950 – 1970. How does the one contradict the other? I don’t see where he has said that productivity was growing prior to the credit crunch but then backtracks saying it was negative and that isd the only way he could contradict himself, no, he says productivity growth was simply slower since 1980 culminating in negative growth over the last 10 years.

    Thee are plenty of things to criticise him about without looking for errors where there are none. I am more interested in what was happening in the missing period 1971 – 1979 inclusive as it is conspicuous by it’s absence.

    You also say you find it fascinating because that’s the same period in which we saw increasing internationalisation and globalisation which was badged as bringing significant economic benefits challenging the wisdom of such a claim.

    I’m not sure which period you mean – 1950 – 1970 or 1980 onwards. I will take it that you mean 1980 onwards.

    It all depends on what you mean by economic benefits does it not? If you’re talking about advanced economies productivity to the exclusion of all else then no, “economic benefit” does not seem to fit. If you’re talking about prices of electrical goods, cars, clothing, footwear et al collapsing faster than wages fell behind GENERAL inflation and the increasing quality of said goods then yes “economic benefit” does fit If you’re taking the wider humanitarian view .that many people in third world countries (now sexily named “emerging markets”) have been lifted out of grinding poverty as a result of electrical goods, car and other production being outsourced to their countries, providing them with employment then again yes “economic benefit” is the correct phrase.

    • Hi Noo2 and braced

      Actually I was simply referring to the fact that the early part of the speech implies that since the Industrial Revolution rises in productivity were a smooth flow then in fact tells us that the more recent period since the 1980s was not so. I too am curious about the 1970s as the numbers would have been good in the “Barber Boom” but then I would have though would have seen a reverse and maybe a big one. There was an irony of sorts in today’s employment numbers which were the best since 1975, the year before we went cap in hand to the IMF…

      As to economic benefits you are right to point out that there are others but on my subject of today I would ordinarily have expecting globalisation and its associated effects to have boosted productivity.

      • I would have expected GLOBAL productivity to creep up since the early 80’s due to mechanisation/automation in emerging markets (EM’s), which, on careful examination, I would expect to reveal burgeoning productivity growth in EM’s as they took on more and more production of tangibles from advanced economies which, in turn would mean falling productivity for advanced economies as they were increasingly left with labour intensive services that cannot be easily outsourced – a hair cut in Shenzen any one?.

        One fly in my theory would be the arrival of computerisation which would show increased productivity in many advanced economy service sectors but that effect has now virtually faded. A second fly in the ointment is that as EM’s increasingly automate production may be onshored again as robots cost the same whether in EM’s or advanced economies but your shipping costs and times will be lower and shorter if you locate within your geographic market , thus driving advanced economy productivity up again.

  7. Hi Shaun,
    In my good old days working in the production dept. of a large manufacturing company, Output per hour was often the subject of long discussions, because we confused hours “worked” with attendance hours. We seemed to have no problem understanding this nuance with regard to equipment and machinery. Idle equipment was clearly under-utilised. And a machine that worked all the time but produced scrap wasn’t very efficient.

    Consider a zero-hours contract worker who is available for work for 40 hours a week but only works (say) 5 hours. Their (productive) output during the 5 hours they worked is a measure of their efficiency. Their attendance hours (5 out of 40) is a measure of their utilisation. And productivity is the product of utilisation and efficiency . IOW output per available hour.

    The problem with measuring service type professions and jobs lies, as you say, in the definition of output; not in the definition of productivity.

    Hope that helps!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.