The ONS released its quarterly bulletin on labour productivity covering the first quarter of 2017 on 5th July 2017.
The index of productivity, measured by gross value added (GVA) per hour normalised fell from 102.4 to 101.8, a drop of 0.59%.
The BBC’s headline for this was “Today’s productivity figures are bad to the point of shocking.”
It’s worth going over a couple of points from Understanding Variation: The key to managing chaos.
Comparing one number with another for a previous period is often used when reporting figures in the press. If the number is up, great. If it down, the sky is falling.
While this is easy to do, such comparisons are weak and lead to sensationalist headlines with little insight. They are limited because they use less data than there is available and because they don’t take variation that is in real world data into account.
If we look at the underlying data over a slightly longer period a few things jump out at us.
First, gross value added (GVA) seems to increase over time fairly linearly. Apart from a brief hiccup around the time of the financial crisis in 2008 the line pretty much trends upwards.
That takes care of the numerator in the productivity equation.
The denominator seems more complicated. Looking at the ONS data, its hard to find something that is a simple measure of hours worked.
For example, this page has some material, but its hard to work out which particular set or trend should be used.
Looking at a website called trading economics, it looks like average hours worked came down from 1995 to 2010 and then started rising agan.
So, we have two charts. One with steadily rising numbers and another with numbers that first fall, and then start to rise again.
When you put the two together, the impact is a chart that first increases and then flatlines as the rate of change of GVA matches the rate of change of hours worked.
The reasons for this in the press point towards the number and type of jobs in the economy as the problem.
There are more jobs than before, so there are more people working than before. This means that total hours worked is increasing faster than the rate at which output from those hours is increasing because the jobs being worked are not as high value as they were previously.
To change the trend, you need to either increase GVA – do more valuable work, or reduce hours – do the same work with less people.
Simple to say, but not easy to do.