With the news from Afghanistan being so grim this morning’s labour market report from the UK brought some welcome better news. Especially welcome was this in the headlines.
In May to July 2021, there were an estimated 953,000 job vacancies, a record high, having grown by 43.8% (290,000) compared with the previous quarter.
These numbers are also welcome in that they are more timely that the conventional numbers we will look at later as they cover July. The growth looks pretty broad based as well.
Strong quarterly growth was reflected across a number of industries, with 10 of the 18 industry sections reaching or equalling record levels of vacancies in May to July 2021…..All business size bands showed higher vacancy numbers on the quarter,
Also it brought higher numbers than pre pandemic.
The number of job vacancies in May to July 2021 was 21.4% (168,000) above its pre-coronavirus (COVID-19) pandemic level (January to March 2020), with only one industry, wholesale and retail trade and repair of motor vehicles, remaining below its pre-pandemic level.
Care is needed in terms of literal levels and thinking that things are better than pre pandemic as what I think it means is that the economy is adjusting to a different world and doing so relatively quickly.
In terms of detail the areas doing best are no great surprise.
Quarterly growth was seen across all industries. The fastest rate of growth was seen in arts, entertainment and recreation, which grew by 267.1% (23,000). The largest increase in levels was seen in accommodation and food service activities, which grew by 73,000 (163.7%).
As sadly on the other side of the coin is an area we have covered so often.
is driven primarily by the retail subsector, which remains 10,000 vacancies below its pre-pandemic level of 85,000.
If we dig deeper and look at July alone the position also looks healthy.
Growth in the latest quarter has been seen in these experimental datasets and both continue to surpass their pre-coronavirus (COVID-19) pandemic levels in July 2021, with the single-month vacancy estimates exceeding 1,000,000 for the first time.
We have come to rely on hours worked as a guide here as the other metrics have been distorted by the various furlough schemes. So it was also welcome that a sort of benchmark was passed.
Total actual weekly hours worked in the UK increased by 50.9 million hours from the previous quarter, to 1.00 billion hours in April to June 2021
However there is perspective provided by the graph below which shows we have ground to recover yet.
In broad terms we have regained a bit over three-quarters of the decline. Even these numbers have come under pressure and they may in fact be better than stated.
Experimental work with adjusted methodology suggests that the existing methodology is understating the full increase in total hours, with the experimental methodology now suggesting the actual number of hours worked is approximately 1.2% higher than stated.
One caveat though is that the numbers were beginning to slip as we came into the pandemic as the previous peak was in the autumn of 2019. So there will be another challenge once we approach such levels.
If we now switch to the employment numbers these are playing a similar tune.
The number of payroll employees showed another monthly increase, up 182,000 to 28.9 million in July 2021. However, it remains 201,000 below pre-coronavirus (COVID-19) pandemic levels.
Or if you want a further perspective.
Early estimates for July 2021 indicate that the number of payrolled employees rose by 2.0% compared with July 2020, which is a rise of 576,000 employees;
We can also look at the sectoral position which does back up the vacancy numbers we noted earlier.
between June and July 2021, accommodation and food service activities increased by 32,000 employees, arts and entertainment by 13,000
Actually even the worst sector managed a rise, although I fear it was more wholesale than retail.
and wholesale and retail by 7,000.
These numbers are not part of the official labour market numbers in that they are from the tax data ( HMRC). This is a welcome innovation as rather than looking at the previous quarter ( April to June) we are looking at July.
What do the conventional numbers tell us?
In the latest period (April to June 2021), there was a quarterly increase in the employment rate of 0.3 percentage points, to 75.1%, and a decrease in the unemployment rate of 0.2 percentage points, to 4.7%.
The improvements are welcome but they cannot be compared with pre pandemic numbers because of the impact of the various furlough schemes. So an unemployment rate of 4.7% is not comparable with the pre pandemic 3.9%. Why not? Well because of this.
The number of people on furlough has continued to fall, with the official HMRC figures showing just under 2m at end of June. More recent survey data suggests that has continued to fall, but there is some uncertainty about levels. ( @jathers_ONS )
The next issue is that as I pointed out above they are for the second quarter and at a time of fast moving events that matters.
These are in such disarray I think they should be suspended as the media do release the headlines.
Growth in average total pay (including bonuses) was 8.8% and regular pay (excluding bonuses) was 7.4% among employees for the three months April to June 2021.
I did formally report them in February as you can see from my letter and the reply.
Whilst they have made thw warnings both clearer and more prominent sadly the newswires regularly skip them.
however, since this growth is affected by compositional and base effects, interpretation should be taken with caution.
Can we improve the estimates here? The answer is both yes and no. Our official statisticians have had a go ( actually two) at allowing for comparing with a pandemic affected economy.
Using the same two methods set out in the blog last month we estimate that the base effect will reduce the regular earnings growth rate by between 2.4 and 3.8 percentage points.
You could drive a double-decker bus through the hap there. Also perhaps the most revealing effort on the compositional problem is that it was a significant problem pre pandemic.
To take into account the compositional effect that was present before the pandemic, this 1% is subtracted from the latest compositional effect of 1.1%. This therefore shows that the net impact of recent job losses is to increase the estimate of average pay by approximately 0.1%, so is now similar to pre-pandemic levels.
If you believe that the compositional effect has only changed by 0.1% I have a bridge to sell you. Which means that even the adjusted numbers are frankly wrong.
This would give an underlying regular earnings growth rate of between 3.5% and 4.9%
The situation just gets worse when we then try to add inflation in for a real wages estimate.
In real terms (adjusted for inflation), total and regular pay are now growing at a faster rate than inflation, at positive 6.6% for total pay and 5.2% for regular pay.
The labour market numbers released today were good news for the UK economy. Hopefully they will be a guide as to what happens next as at times of change they have led the output ( GDP) numbers as they did around 2013. The road gets harder as more jobs get filled and we see a type of economic indigestion as we try to fit people to vacancies and avoid putting square pegs in round holes. The furlough scheme is on the way out and there will be an impact from its ending.
As to the wages numbers we are in quite a mess. The numbers were changed and effectively signed off by Dr. Martin Weale and it is hard to see how the previous system could be worse than now. We used to be worried about the way it ignores the self-employed and small businesses and that is still true but now it does not seem to be measuring anything much that exists in reality Yet with Bloomberg reporting this on twitter many will be misled.
UK wage growth hit a record in July
Just as bad is producing numbers like this.
In Quarter 2 (Apr to June) 2021, output per hour worked was 0.6% above levels prior to the coronavirus (COVID-19) pandemic (the 2019 average), despite a quarter-on-quarter fall.
Due to both miss measurement and lags we have no idea about productivity right now.