The August GDP data, released last week, showed that GDP was still 0.8% below its past peak in February 2020. Meanwhile data on employment released almost simultaneously showed that the number of payrolled employees in September 2021 was 0.4% more than in February 2020. With so many activities being revolutionised digitally one might have expected productivity to rise, requiring fewer employees.

Part of the explanation is the collapse of self-employment, probably at least partly caused by the disadvantageous tax position resulting from IR35. The latest data shows a drop of 711,000 in the number self employed between January-March 2020 and June-August 2021, a drop of 14%.

As a result, the number in employment (including self-employed) in the June-August 2021 period was still 1.8% down since January-March 2020. Moreover, people have been working fewer hours so that the total number of hours worked in the economy fell by 1.0% over the same period. Meanwhile people are spending less time commuting. US data reported by Harvard Business Review (https://hbr.org/2020/12/where-did-the-commute-time-go) indicated that at the peak of the lockdown the average person spent 41 minutes a day less on commuting than pre-pandemic. Quite a lot of this will have adjusted back but the hours worked data and reduced commuting suggests that people are adjusting their work/life balance.

It is interesting that at the same time vacancies are at an all time record at 1.1 million, about 1½ times the pre pandemic peak. This is probably the best indicator of labour shortages.

The labour shortages are allegedly due to Brexit and to older people leaving the labour force – both claims seem plausible, even if the official labour force survey data shows limited support for them.

Some economists have been identifying the changing composition of consumption from services to goods as a possible explanation for the increased labour requirements. The government data does not support this. Productivity in the consumer service sector is typically lower (output in hotels and catering for example runs at about £20 an hour at present) than for manufactured products (output in the manufactured goods sector is around £40 an hour). If anything, the shift from services to goods should have raised output per hour.

Looking at the changes in productivity does help a little to explain what is going on. While some sectors, particularly digital, have demonstrated rising productivity, the travel and public transport sectors and the arts and creative sectors have seen sharp falls in productivity since before the pandemic.

There is, though I suspect, another factor underlying the figures. The digital sector has been doing best in the economic merry go round, even on the initial government figures. But this sector is hard to measure, partly because of the inherent difficulty of measuring its output and partly because so many firms are new and outside the statistical net. The history of recent years is that the estimated GDP of the sector keeps getting upgraded – the most recent revisions to GDP added nearly 2 percentage points to GDP in total. I wouldn’t be surprised if GDP growth over the two years 2020 and 2021 has been understated by as much as 0.7% per annum.

So it may well be that when we look at this data in two or three years time, the sums seem to make a bit more sense and the data will seem less hard to reconcile than it does at the moment.

For more information please contact: 

Douglas McWilliams   Email: dmcwilliams@cebr.com Phone: 07710 083652