ONS now say that nominal pay growth is 2% not 4% for those in continuous employment – with real pay stagnating
The Prime Minister recently claimed that everyone who had been in work for more than a year had had a 4.1 per cent pay rise (here).
ONS have now released analysis that effectively halves this estimate: earnings growth for 2014 is now at 2.1 rather than 4.1 per cent for those in continuous employment.
But this changed view hardly leaps out at the user of official statistics; the latest figures were published in last week’s ONS Economic Review for March but not headlined. Given the four per cent figure is being used so prominently by politicians eager to downplay the earnings crisis, it is important to understand exactly where this story came from and how it has changed.
ONS published (nominal) 4.1 per cent pay growth for those in ‘continuous employment’ in 2014, when it released the ASHE (annual survey of hours and earnings) data in November. The idea was that continuous employment (i.e. jobs where the employee had been in post for at least one year) helped better capture underlying trends in earnings. The estimate was not included in the ASHE release headlines, but was fairly prominent (on Figure 2 of the release) and then widely seized on by politicians; it was included in a chart in the Autumn Statement (Chart 1.5) and even commended by the UK statistics watchdog as valuable information to understand developments in earnings.
The TUC has long emphasised that the changed composition of the labour market is an important factor explaining unprecedented low pay outcomes. However we expressed reservations about the technical calculations of the 4.1 per cent figure as it was based on the growth of the median earnings figures for 2014 and 2013. A more representative figure would be the median rate of earnings growth. The ONS were using a growth rate derived from a distribution of level figures, when they should have been looking directly at the distribution of growth rates. (See here for fuller explanation.)
Then last week in the Economic Review ONS published a revised assessment of these earnings figures, also based on ASHE. This time they took the approach we advocated, so that they examined the distribution of growth rates. Rather than emphasising a specific headline growth rate, the material was presented around understanding the distribution of (real) earnings growth rates, e.g. in the headlines to the release observing that “48 per cent of these workers received a rise in their real earnings in 2014” (a separate blog post by my colleague Richard Exell examines these distributional figures in more detail). However their Figure 10 included a time series for the median of real earnings growth rates. In 2014 this figure was 0.3 per cent.
Further confusion is caused by the new figure being in real terms and the previous figure in nominal terms. The two can be put on a consistent basis using the CPI (for April to match the period for ASHE data), and the chart below compares the previous and latest nominal (red) and real (blue) figures.
Earnings growth for those in continuous employment, per cent
The red lines show nominal figures (both the previous and new analysis), and the blue lines show these data adjusted for CPI. The solid red figures show the new nominal series registering 2.1 per cent in 2014, and should be compared with the dotted red line which shows the previous analysis which suggested pay growth of 4.1 per cent in 2014. Hence nominal earnings growth is halved, and the Prime Minister’s claim now incorrect. Moreover on the new measure nominal earnings have actually been slowing steadily in recent years, from 2.5 per cent in 2011, and now little different to the lowest figure of 2.0 per cent in 2010.
As we all know, reduced inflation has meant a slight revival in real terms, but the median figure for 2014 is still barely different from zero at 0.3 per cent. And to stress, these figures are for those in continuous employment, i.e. a select and lucky group. For the employee workforce as a whole, ASHE data shows nominal growth of only 0.1 per cent, i.e. a real terms decline into 2014 of 1.7 per cent.
(Note also that there are compositional differences between the new and previous measures, with the new based on people in full-time employment for more than a year, and the old based on people who have been in the same job for more than a year, but this is unlikely to explain the difference in the headline numbers.)
Given the extent of the fall in real earnings since the crisis, and the prominence of these figures in political debate, this is an incredibly significant change of view.
We recognise ONS’s desire to look behind headline figures and try to help unravel underlying stories, but the advantage of routinely published and consistent figures based on well-established methodologies is that users know what they are getting and know where to get it. Ad hoc analyses can be valuable but there is always the danger of figures (no matter how well caveated or understated) talking on a life of their own, particularly when results appeal to the prejudice of certain stakeholders. Nor is it easy for users to unravel changes of view when it comes to these figures – indeed one might argue that the Economic Review sought to put a positive gloss on the earnings story. Misled themselves, we are asking the UK Statistics Authority to review these figures.