Labour Market Statistics, April 2014 – The Day After

Yesterday's Labour Market Statistics contained a bumper set of noteworthy findings, too many to cover in the press release we issued yesterday morning.  I decided to sleep on the results - and, just as important, the coverage they received - before finalising this blog.  So it is a case of the day after the day before.

The headline news item was earnings growth.  The message from the figures is quite complex and requires a detailed explanation - please bear with me.  Unfortunately, it also provided a disappointing (if predictable) illustration of statistical illiteracy on the part of many commentators and headline writers.  When economists talk about real earnings growth, they mean the rate of growth of earnings minus the rate of growth of prices (inflation).  These should be compared using consistent data and over the same time period.  So the headline three month average increase in total weekly earnings for the twelve months to February 2014 - the figure released yesterday, which grew by 1.7% - should be compared with the February 2014 figure for inflation - which using the headline measure, the Consumer Price Index (CPI), was also 1.7%.  In other words, earnings growth over the year to February matched inflation, meaning zero real earnings growth.  Reports that said earnings growth had 'outstripped' inflation were comparing the February earnings figure with the March inflation figure, which is the wrong thing to do.  A pithy but accurate headline might be 'earnings growth moving up catches inflation moving down'.

Technical aside

  • The single month estimate of growth in total weekly pay was 1.9%, which was greater than CPI inflation (1.7%), so this measure did rise in real terms. But that last happened quite recently, in April 2013, because some employers deferred payment of bonuses from March to April 2013 so their employees could take advantage of the reduction in the top Income Tax rate. This is why the ONS highlight the three month average rate of growth, because it reduces - but does not eliminate - the impact of unpredictable events.
  • If we are being hyper-rigorous, we should be comparing the three month average growth rate of total weekly pay against the three month average CPI inflation rate for the same period, which I calculate as 1.9% (to one decimal place). So it could be argued that total weekly pay on this measure still lags behind inflation.

Now it could be argued this is detail - although the counter-argument would be that the credibility of any commentator relies on them understanding what it is that they are commenting on!  Inflation is heading down and we know the March figure is 1.6%, so it is only a (short) matter of time before earnings growth finally exceeds inflation?  Well that depends on the important question of what the current figures tell us about the strength of earnings growth in the UK.

Six months ago, growth in total weekly pay (0.6%) was still 2.1 percentage points lower than CPI inflation (2.7%).  The gap has been closing since then due to earnings growth increasing from this very low base at the same time as inflation has been falling.  In the most recent month, falling inflation did two thirds of the work but over the six months it is roughly half and half.  Most forecasters are expecting further reductions in inflation in the coming months.

As for earnings, the more detailed data released by the ONS suggest there is a lot of variability across the economy and a lot of variability in the data.  Weekly earnings in construction will depend as much on weather conditions (whether better or worse than the average for the time of year) as they do on the state of the economy.  The data in retail will be influenced by whether or not Xmas and New Year sales conditions were better or worse than usual.  Next month we will see the December earnings data drop out of the three month average to be replaced by March.  This may well see an increase in bonuses because some employers who paid bonuses in April last year - to avoid employees paying 50% Income Tax - will go back to paying in March.  So next month could properly see il surpasso.  But there is a lot of noise in the underlying data.

With CPI inflation heading down, this might well keep general pay awards around the 2% mark rather than encouraging a shift towards the 2.5-3% range.  The bottom line is that until we see improvements in labour productivity we are unlikely to see any significant increase in real hourly pay.

Finally on this topic, most commentators yesterday quoted the headline CPI rate of inflation.  However, inflation as measured by the Retail Price Index (RPI) was 2.9% in February.  The RPI is no longer a National Statistic but a measure that does meet the relevant criteria and based on the RPI, termed RPIJ, increased by 2.2% in February.  Hence whether or not wage growth has closed the gap with inflation also depends on how you choose to measure inflation.  Given the likelihood that future interest rate increases and rising housing costs will keep RPI-based measures above CPI-based measures for years to come, we should expect no early end to debate about living standards - unless we have a sustained boom.  And no forecasters see that coming just yet.

The statistics also confirmed that jobs growth remains strong, with the number of people in employment increasing by almost a quarter of a million in a single quarter and by almost 700,000 year on year.  This has been enough to accommodate all of the population increase we have seen in the last year - some 410,000 - as well as getting the unemployment rate down from 8% to 6.9% (the first time it has been below 7% for five years) as well as creating enough jobs for a modest reduction in the inactivity rate.

As this rate of jobs growth continues, it raises questions about what is happening in the labour market.  The OBR economic forecasts, for example, imply that to date we have seen a jobs-rich (and productivity-poor) recovery from a deep recession.  But buried away in the spreadsheets accompanying the forecasts are assumptions that the long-term trends for both activity rates and employments are flat.  Are these assumptions still credible or are one or both of these variables now on an upwards trend?  Such a possibility might have been in the Chancellor's mind when making his recent comments on full employment.

A number of organisations recently, including the TUC, the Resolution Foundation and the RSA have drawn attention to the sizeable contribution to recent employment growth coming from self-employment.  In the six months between June 2013 and December 2013, the number of self-employed jobs increased by 293,000. But what types of businesses are these? Below is a chart showing the top ten industries in terms of increased self-employment.

Change in self-employed jobs

We see here a number of professional industries as well as industries associated with increased activity in the construction industry.  But there are also some that raise eyebrows.  Are the self-employed in the education sector teachers and lecturers who are moonlighting as tutors?  And why have the number of people working for themselves in washing and dry-cleaning increased by nearly 20% in six months?  As ever, the statistics raise as many questions as they provide answers.

Thank you for your comments. There may be a short delay in this going live on the blog page as we moderate the comments added to our blogs.