The UK productivity crisis meets real wage growth

As we advance on the UK labour market data let us first note some good news. This is that the procedure for an “early wire” to be given to the UK establishment has been stopped. To be specific a list of people were in the past given the data some 24 hours before the rest of us, and as the ship of state is a somewhat leaky vessel there were obvious concerns that some traders would be more equal than others.

Moving back to today’s data the background to it was set by this official release from last week.

Productivity – as measured by our main measure, output per hour – fell by 0.5% in Quarter 1 (Jan to Mar) 2017.

As this is a factor in wage growth we have a potential driver of the dip in wage growth we have seen in 2017 but the problematic news did not stop there.

A fall of 0.5% takes productivity Quarter 1 2017 back below the peak achieved in Quarter 4 (Oct to Dec) 2007, which was broadly matched in Quarter 4 2016. Productivity is now 0.4% below the pre-downturn peak and 0.4% below the post-downturn peak.

The productivity problem

This is an example of what Winston Churchill meant when he said that Russia was a riddle wrapped in a mystery inside an enigma. The same type of thinking applies to productivity especially when it is described like this.

Productivity in Quarter 1 2017, as measured by output per hour, stood 16.8% below its pre-downturn trend – or, equivalently, productivity would have been 20.2% higher had it followed this pre-downturn trend

In my opinion looking at it like that merely tells us that the world has changed and that the productivity boat we were previously on sailed elsewhere. There is little point regarding it as a gap we can regain and I find it fascinating that those who seem to think we can get it back are supporters of policies like QE which have supported the zombie banks and companies which are a factor in this.

More significant to me is this from the June 2016 Economic Review.

Productivity is estimated to have grown at a compound average growth rate of 0.1% per quarter during the recovery between 2009 and 2014. This near-flat productivity growth is a phenomenon unprecedented in the UK since the Second World War.

We can update that because if we look at the expansion since the middle of 2013 we see that output per hour has risen from 98.1 then to 99.6 at the end of the first quarter. So a bit better than flat but not by much. This compares to past episodes.

This is in contrast with patterns following previous UK economic downturns where productivity initially fell, but subsequently bounced back to the previous trend rate of growth.

If we look back to the June 2016 Economic Review we can put a number on that.

However, at the same stages of both the 1990s and 1980s recoveries, productivity was more than 16% above the respective pre-downturn levels.

It is worse than that now as productivity three years or so later is where we thought it was then as whilst it has grown since the past was worse than we thought. What we can now clearly see is that yet another type of lost decade has been in play.

Moving to my explanation if we move on from the drop caused by the credit crunch and look at the more recent period I have written before that there are real problems in measuring productivity in the service industries. Not only are they pretty much 80% of our economy but they have been in essence our economic growth. Productivity in haircuts or operations for surgeons? Maybe in some cases but in others no.

The latest numbers seem to be picking this up.

Labour productivity fell in services but rose in the manufacturing industries; services productivity fell by 0.6% on the previous quarter, while manufacturing productivity grew by 0.2% on the previous quarter.


This is the good news side of the issue. What I mean by that is that the number of people who are employed continues to grow.

For March to May 2017, there were 32.01 million people in work, 175,000 more than for December 2016 to February 2017 and 324,000 more than for a year earlier.

Which brings more associated good news.

For the latest time period, March to May 2017, the employment rate for people was 74.9%, the highest since comparable records began in 1971.

My argument would be that the employment is mostly in the service sector where we struggle to measure productivity. If we note the rise and the recent struggles of UK output it may be that measured productivity fell again in the second quarter of this year. So there you have it would you prefer more people in employment or higher productivity? It is not of course completely that simple but it is a factor in play.

Oh and I noted another factor in rising employment.

The increase in the employment rate for women is partly due to ongoing changes to the State Pension age for women resulting in fewer women retiring between the ages of 60 and 65.


This does not necessarily get better as employment improves but generally does.

the unemployment rate for people was 4.5%; it has not been lower since April to June 1975………1.49 million unemployed people, 64,000 fewer than for December 2016 to February 2017 and 152,000 fewer than for a year earlier


Actually these were a little better although you might not think so from the official release.

Between March to May 2016 and March to May 2017, in nominal terms, total pay increased by 1.8%, lower than the growth rate between February to April 2016 and February to April 2017 (2.1%).

If you look into the detail you see that the annual rate of growth in May at 1.8% was better than the 1.3% in April ( where the annual bonus season was weak dragging it lower). This meant that if we switch to real wages the annual rate of fall rose to -0.9% from the -1.3% of April.

If we look deeper into the real wage situation we see that the index in May was at 100.8 which means that as it was set at 100 in 2015. So we have had economic growth with little if any real wage growth and that stretches back as the index was at that sort of level in the summer of 2011. There is a long way to go to the January 2008 peak of 105.8.

Actually as the “not a national statistic” CPIH is used as the inflation measure I am sorry to have to tell you that a more accurate inflation measure would show an even worse performance.


To my mind we should be more concerned about the slow rate of productivity growth than the drop in 2007/08. We are now in a world of QE and zombie banks which will take us some time to get out of especially as many places are still getting in it! I would be looking to take some of the service sector out of the numbers on two grounds. The first is that we simply cannot measure it and for others it is not appropriate. As to improving our performance there have been some interesting ideas from Diane Coyle but there are also dangers as I find myself thinking of all the money being spent on Smart Meters for a very small potential gain as I read this.

Ensuring adequate investment in infrastructure to meet our current and future needs and priorities

Also today is another grim day at the Bank of England especially for its Chief Economist Andy Haldane. Perhaps this is the true reason he is on something of a tour of the UK! Regular readers will be aware that I have listed the many failings of “Output Gap” theory in my time here. Andy has been a test case for these as he has got wage growth wrong again and again and again by using it. Well in February he thought he had struck a cunning plan by changing his framework so that the level at which wages would start to surge higher ( NAIRU) would be when unemployment fell to 4.5% or where we are now only a few months later as opposed to the couple of years he expected/hoped.

Dreamer, you know you are a dreamer
Well can you put your hands in your head, oh no!
I said dreamer, you’re nothing but a dreamer (Supertramp)



Do full employment and the natural rate of unemployment mean anything any more?

One of the features of economics in the pre credit crunch era was confidence about concepts concerning the labour market. We had the concept of full employment which is defined below from the Financial Times lexicon.

When labour supply and demand in an economy are balanced at market wages. This does not mean everybody in the labour force is employed (see frictional unemployment) but in theory, it is the highest possible level of employment in a market economy.

It is hard not to have a wry smile at the way that full employment does not mean full (100%) employment. A bit like the way many central bankers define “price stability” as an inflation rate of 2% per annum. But we have an allowance for people changing jobs ( frictional unemployment)  and often an allowance for a mismatch between skills and jobs which is called structural unemployment.

In terms of numbers here are some estimates. From the BBC.

William Beveridge, the man who inspired Britain’s post-war welfare state, said full employment meant a figure of under 3%.

Other estimates tended to be higher than this and William T Dickens did some work in the US which estimated it as being around 5% in the pre credit crunch period. Also Mr. Dickens added quite a bit to the debate by publishing a graph showing that his work had a lower bound of 2% and an upper bound of 7%. That narrows it down.

Also the definition had shifted somewhat as the phase full employment became pretty much interchangeable with the cumbersome phrase the non-accelerating inflation rate of unemployment or NAIRU. Estimates of this I saw pre credit crunch tended to be of the order of 4.5% for the unemployment rate.

A modern challenge

This comes as so often in the modern era from Japan where the Statistics Bureau has reported this today.

  The unemployment rate, seasonally adjusted, was 3.0%.

So below the natural rate and now in the full employment zone? We can see what is on the quantity measures an improving situation.

The number of employed persons in July 2016 was 64.79 million, an increase of 980 thousand or 1.5% from the previous year……. The number of unemployed persons in July 2016 was 2.03 million, a decrease of 190 thousand or 8.6% from the previous year.

As you can see unemployment is on the edge of two thresholds here as we wonder if it will dip below 2 million and the rate fall below 3%. A rising employment situation is also strong when we remind ourselves that the population is shrinking although some care is needed as we are given numbers for the labour force which are amazingly constant.. If we dig deeper we see that those who have involuntary employment are a relatively mere 540,000, is that structural unemployment? Even youth unemployment ( 15-24) is a lowly 4.7%.

Why aren’t wages rising quickly?

A problem for the theorists in their ivory towers has been the behaviour of wages in Japan. They should be rising strongly as we have passed the NAIRU but they continue to struggle. Indeed figures for workers households today hint at possible moves in the opposite direction.

The average of monthly income per household stood at 574,227 yen, down 2.2% in nominal terms and down 1.8% in real terms from the previous year….  The average of consumption expenditures per household was 302,422 yen, down 3.9% in nominal terms and down 3.5% in real terms from the previous year.

Employment is not what it used to be

There has been a structural change in employment in Japan over time. Back in September 2014 Fathom Consulting told us this.

Since the depths of Japan’s economic crisis in the late 1990s there has been a marked switch away from full-time into part-time work. The old ‘job-for-life’ culture is just that…….There are more people now working part-time in Japan who would rather work full-time than there are unemployed people. On that basis Japan has a much larger ‘underemployment’ problem than either the US or the UK.

If we look at the latest data we see that there are 33.57 million regular employees in Japan and 20.25 million irregular employees. If we switch to hours worked per month we see that regular employees get 188 of them but for irregular employees get a much lower 120.

The wages situation is rather divergent too as Japan Macro Advisers point out. There was good news for salarymen and women.

The regular (basic and overtime) part of wages was flat, only up by 0%, but the bumper bonus, up by 3.6%, pushed up the overall paycheck in June. It is going to be a good summer for all those army or loyal company men (and women) in Japan.

But not so good news for others.

The talk of good summer bonus must be a pain for irregular workers in Japan though. Over 1/3 of employment in Japan are irregular workers, and most of them are not eligible for bonuses. In June, wages of part time workers were only up by 0.4% year on year, compared with 1.5% year on year rise for full time workers.

The number of regular jobs was falling but that has changed albeit marginally as 19,000 have been created over the last year. However we found ourselves looking at irregular jobs for the majority of jobs growth.

A difference in the sexes

The situation for Japanese women has been lauded as a success for Abenomics and employment has risen from 23.24 million in 2013 to 24.57 million in July of this year. However proportionately much more women are irregular workers with 13.79 million of them now and 830,000 of the employment growth just discussed being in that category.

Now there are issues here as to type of work and perhaps some would only want part-time work which is more likely to be in the irregular category but we also have to face up to the fact that a difference between treatment of the sexes may well be a feature of the new Japanese employment structure.

What is life like for an irregular employee?

Back in March 2015 the Wall Street Journal gave us some insight.

Mr. Kinoshita is a member of Japan’s large and growing army of so-called nonregular workers—temporary, often part-time employees who are usually paid less than their “regular” counterparts. Like many of them, the 49-year-old Mr. Kinoshita has been unable to find permanent employment for years and struggles to make ends meet, despite working nearly full time. His new job pays ¥1,000 ($8.25) an hour, well below the ¥1,400 he earned in his last position at an auto-parts maker.

Oh and he does not seem to be too keen on plans to raise the inflation rate.

Everything is expensive,” he said. “A bag of tangerines costs ¥400.”

How can things be expensive in a place that has not had any inflation for years and indeed decades?


The example of Japan shows us that Ivory Tower concepts of full or natural rates of employment are full of more holes than a piece of Swiss cheese. For a start they completely miss the concept of underemployment. Then they fail to allow for the trend towards newer jobs being on worse terms than established ones. This is not only in wages but in hours worked and conditions.

If we move geographically away from Japan we have seen similar effects at play in other countries where the unemployment rate has fallen such as the US and UK. The obvious signal is the way that wage growth has been slow with levels of unemployment at what was considered to be the natural rate level and in some cases below it. The UK ONS has done work which suggests that like in Japan wage growth has been held back by the fact that newer jobs have been at lower wages. Also we note that in the UK wages for younger workers have been disproportionately affected which is exactly the opposite of where we should be now.

Of course we also need to allow for the fact that the information we get is much less than some would claim. There are wide margins of error in the two US labour market surveys which frequently contradict each other. In the UK we know so little about what is happening in the growing self-employment sector and know almost nothing about their wages.