By Stuart McIntyre
The UK economy shrank by 1.5% in the first three months of 2021, coinciding with the re-introduction of public health restrictions following a spike in COVID-19 cases in the UK at the end of 2020.
On an annual basis, the UK economy was 6.1% smaller than it was in 2020 Q1, and on a rolling annual basis (comparing the latest four quarters to the previous non-overlapping four quarters), the UK economy was 10.9% smaller.
For reasons set out in our previous publications in this area, we work with this rolling 4 quarter measure in our analysis. In addition, and again for technical reasons, we work in log growth rates, not exact growth rates. On this basis, UK growth in the year to 2021 Q1 was -11.8%.
In order to understand what this means for the different regions of the UK, we have re-run our regional nowcasting model.
Recall that our model produces estimates of regional growth taking into account this new data for the UK as a whole, and allocates it – as noted in previous blog updates – to regions on the basis of the following four factors:
- estimated historical relationships between regional growth and UK growth (in essence: this reflects how sensitive regional growth is to UK growth);
- estimated historical relationships between the growth of particular regions (in essence: this captures how growth in region x has translated into growth in region y);
- estimated historical relationships within the regions (in essence: this captures the persistence of regional growth from one quarter to the next), and;
- estimated historical relationships between other macroeconomic variables (like the oil price) and regional growth.
On this basis, our model suggests that 2021 Q1 saw the following pattern of regional growth across the UK.
Table 1: Regional growth in the year to Q1 2020 and 2021
These results suggest that the Welsh economy has been particularly impacted by the economic challenges of the past year, with the largest decline in activity in the last year. However, the range of regional growth rates is spread relatively evenly either side of the UK growth rate and while there are large differences in absolute terms, these are not large differences in relative terms (particularly given the uncertainty around these data).
One way that we can, and have in the past, sought to compare our own estimates is with respect to more timely GDP data from the devolved administrations in Scotland and Northern Ireland.
We now have data from the Scottish Government covering this period, and this shows that the Scottish economy was 10.3% smaller on a 4Q – on 4Q basis. When we compare the growth rate on a log difference basis (i.e. on a directly comparable basis to our own estimates), the Scottish Government estimate is almost exactly the same as our own estimate (-10.9%).
We do not yet have data covering Q1 2021 for Northern Ireland from the Northern Ireland Statistics and Research Agency (NISRA), but when released this will be published here.
Next steps
Over the past year and a half, we have been working with Office for National Statistics (ONS) colleagues to support them in adopting the model that we have used to produce these estimates, as well as a further iteration of this model which we have developed. The latter of these models explicitly includes the quarterly data from ONS on regional growth (released as quarterly regional GDP), as well as the data from the Scottish Government and NISRA for these areas.
It is the intention that this will be the last public regional nowcasting update from us on this webpage.
From the next release, which will follow the release of UK GDP data for 2021 Q2 in August 2021, the ONS will run the model and report the results from this model through the ons.gov.uk website. The project researchers are delighted to see this aspect of our work reach a successful conclusion and being integrated into ONS’ outputs.
As a research team we will continue to work in this area, and our next goal is to build a nowcasting model for the ITL (previously NUTS) 2 regions of the UK. We’re also looking at trying to use VAT turnover data to improve our nowcasting model. We’ll release any working papers on this new work through the ESCoE website so keep an eye out!
Best wishes from us to the ONS team as they take forward this new release!
You can access all of our historic quarterly estimates here.
ESCoE blogs are published to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of the ESCoE, its partner institutions or the Office for National Statistics.
