The UK productivity puzzle is mostly a result of outdated economics and statistics

Today has brought us two flashes of indirect insight on the issue of productivity and what has become called the productivity puzzle. In case you are wondered what that is here is the OECD from August last year.

Since the start of the Great Recession, labour productivity growth has been weak in the United Kingdom, weaker than in many other OECD countries. The productivity shortfall, defined as the gap between actual productivity and the level implied by its pre-crisis trend growth rate, was nearly 20% for output per hour at the end of 2016.

I am dubious about measures which use the bubbilicious boom for their trend but Ivory Towers love that. Also there is clearly an issue to consider and the OECD had a go at a breakdown.

Most of the UK productivity underperformance is structural rather than cyclical. Half of the productivity shortfall is explained by non-financial services (with information and communication being the largest contributor), a fourth by financial services, and another fourth by manufacturing, other production and construction.

Clearly the 5% productivity shortfall explained by the financial sector needs a much more thorough investigation as the ongoing weak state of the banks is due to the fact that their position was over reported on the pre crisis boom and thus so was productivity. Or as the OECD put it.

its steep increases in the run-up to the crisis.

But they do at least manage a minor swipe at the Zombie business era that has been supported by central bank QE.

weak corporate restructuring have both held back productivity improvements in the manufacturing sector.

Output Gap

Economic theory has had a real problem with this and let me give an example from Japan this morning. The Ivory Towers will tell you that wages should be soaring due to a tight labour market with unemployment at 2.3% and a number of jobs to applicants at a more than forty-year high. Meanwhile back on Earth.

Labour cash earnings dropped 0.8 per cent from a year ago, the ministry reported on Friday, compared with projections for them to advance 0.9 per cent. The reading for January was revised down to -0.6 per cent from 1.2 per cent………..

Real wages, which are adjusted for inflation, fell 1.1 per cent, compared with economists’ median forecast of 0.8 per cent.

The real wages figure for January was revised down to -0.7 per cent from 1.1 per cent. ( Business Times)

As you can see the output gap theory has had another complete failure as wages have failed to increase. This makes us mull productivity which is supposed to be strongly linked to wage growth and real wage growth especially. Also I am afraid we have another problem with official statistics as there has been a major revision after clear flaws were discovered such as only a third of the businesses in Tokyo with over 500 employees that were claimed to be sampled actually were. That adds to the problems seen elsewhere with official Japanese data such as the GDP numbers which is completely the opposite of stereotypes.

UK House Prices

These are beginning to offer a more hopeful perspective. The reason why I argue this is that in my opinion way too much economic effort in the UK has gone towards the housing sector where in many areas substantial capital gains have been available via owning a house. This led for quite some time to the boom in the buy-to-let sector and took both investment, attention and effort from other parts of the economy. This was fed by the various “Help To Buy” policies of the government and the multitude of efforts by the Bank of England to reduce mortgage rates and raise mortgage lending to get house prices higher.

Thus the numbers from the Halifax this morning are welcome as they show that things have slowed down.

The average UK house price is now ยฃ233,181 following a 1.6% monthly fall in March…….The more stable measure of annual house price growth rose slightly to 3.2% and is still within our expectation for the year.

You need to go through their numbers carefully to get to that as the monthly UK house prices series of the Halifax has become very erratic and has now gone 2.5%, -3% ,6% and now -1.6%. We thought the 5.9% rise in February was extraordinary at the time yet we now discover it was 6%! If we look at March compared to a year ago we see that there has been a 2.4% rise which seems to reflect better the numbers we get from elsewhere.

As to the overall reliability of the Halifax data well let me quote anteos who commented on the last set of numbers from them as follows.

So, just as the annual indicie was heading towards negative territory, up comes a 5.9% increase.
Very similar to Decembers figures which were then reversed the following month. If I was a betting man, a big negative value will pop up next month.

Chapeau.

Productivity Data

There was something of an irony as I searched for the update here.

404 – The webpage you are requesting does not exist on the site

That was not entirely hopeful for productivity as the UK Office for National Statistics and leads into the official enquiry into out data which is ongoing. Sadly the leadership seem lost in a world of click bait and telling us that tractor production is rising. When we got the numbers they posed another problem.

Labour productivity for Quarter 4 (Oct to Dec) 2018, as measured by output per hour, decreased by 0.1% compared with the same quarter a year ago; this is the second successive quarterly fall following the decrease of 0.2% seen for the previous quarter.

If we look back it is the fall in the third quarter which is the most concerning as GDP growth was particularly strong at 0.7%. For the year just gone we had some growth but not much.

In 2018, labour productivity measured as output per hour grew by 0.5% compared with the previous year, with increases in both services and manufacturing of 0.8% and 0.3% respectively.

This meant that the overall picture in the credit crunch era is this.

Labour productivity increased by 0.3% in Quarter 4 2018 compared with the previous quarter. This increase left productivity 2.0% above its pre-downturn peak in Quarter 4 2007,

So not much allowing us to update the OECD style analysis above.

Productivity in Quarter 4 (Oct to Dec) 2018, as measured by output per hour, was 18.3% below its pre-downturn trend โ€“ or, equivalently, productivity would have been 22.5% higher had it followed this pre-downturn trend.

Comment

The first problem with the productivity puzzle is whether we can measure it with any degree of accuracy. As we have seen from the Japanese wages and UK house price data above both official and private-sector data have serious issues. This spreads wider and in my opinion is highlighted by this.

In Q4, public service productivity increased by 0.8% on the previous quarter, driven by unusually strong growth in output (1.3%)

It is my opinion that we have very little idea about public sector output and therefore even less about its productivity. Also there are areas we might not always be keen on higher productivity. Returning to the numbers I helped Pete Comley with some technical advice when he wrote his book on inflation and here is what he discovered about the government sector.

The upshot of that review is that estimates inflation on government expenditure no longer use real cost inflation (like wage increases, rises in raw materials costs, etc.) but instead use measures of quality (such as the number of GCSE grades A-C) to calculate the deflator.

So that is a mess.

Also there is a clear problem with the concept of productivity in the services sector. This is because we are often measuring intangible things rather than the tangible of manufacturing. The extraordinary changes for example in the world of information and communications are mostly only captured if there is a price change. I note the paper from Diane Coyle and others that suggested even these were wrong and the situation was much better ( lower prices and higher output). Also I have pointed out before as well as giving evidence to the Sir Charles Bean enquiry, that the UK trade release has at most a couple of pages on services out of the 30 or so with no geographical or sectoral breakdown. This matters even more as we rebalance towards services with growth in the index of services some 21% over the past decade.

Also there has been a shift towards self-employment which makes the numbers less reliable as we know even less about that area.

Finally it would be nice for us to get some capital productivity figures to compare with the labour ones.

Me on The Investing Channel

 

 

 

4 thoughts on “The UK productivity puzzle is mostly a result of outdated economics and statistics

  1. Hi Shaun

    Great article as always, and thank you for the comment.

    I’m starting to think we might not see a negative annual figure this year, with all the volatility in the stats. And the previous aprils figure was 221k. So it would need a huge drop to go below that.

    Looks like they’ve managed to keep the plates spinning a bit longer ๐Ÿ™‚

    thanks

    • Hi Anteos and no problem

      As to the spinning plates one or two seem to be coming loose.

  2. Hi Shaun

    Excuse me for posing some rather basic questions on this. The overall fall in productivity for analytical puroses seems to me to have two aspects: changes between sectors in the overall balance and productivity within sectors.

    Change between sectors simply means the balance between different sectors over time. So, if constuction, which is say low productivity, moves from being 10% of the economy to say 20% then that will bring down the overall productivity figure.

    On the other hand if it is within each sector then it can actually fall within a sector from a high point and yet, because that sector has a greater portion of the whole, boost the overall productivity figure.

    I don’t know whether these factors are adjusted for but it seems to me that what counts is productivity within the sectors rather than the balance of that sector within the whole. I’m not saying that balances between sectors doesn’t count but it seem dependent on secular or cyclical forces the one of which may be just evolution whilst the other is temporary.

    • Hello Bob,

      that’s all to sensible , it will never do in an age of subjective measuring ref:- ” but instead use measures of quality”

      just ask if you feel you’re better off and majickally you are !

      Frankly the stats have gone from something real to something from the Sunday Sport journalism or Pravda …..

      Forbin,

      PS :such as the number of GCSE grades A-C – ok so we drop the grade level and more people pass – and that improves productivity ?

      PPS: as for self employed – well if you don’t measure it , it doesn’t count , does it ? so it does’t affect the economy at all then….. ummmm

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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.