Blue Book 2022 – revised impacts of the coronavirus (COVID-19) pandemic on the UK economy: October 2022

• The revised estimates of real gross domestic product (GDP) still show that there was more of an adverse impact on “high-contact” industries in the first half of 2020 and there has been a weaker recovery thereafter, where these “high-contact” industries are now 10% below Quarter 4 (Oct to Dec) 2019 levels.

• There have been minimal revisions to the profile of spending on “social consumption”, which refers to recreation and culture as well as restaurants and hotels, in line with the revised levels of output produced by the accommodation and food and arts, entertainment, and recreation industries.

• Our latest estimates show that a slightly upwardly revised 77% of the increase in household saving during the coronavirus (COVID-19) pandemic reflects “forced” saving.

• Early findings show that the revision performance of our GDP estimates in the most recent periods are now similar to that experienced prior to the coronavirus pandemic.

We have improved our estimates of current price and volume gross domestic product (GDP) in Blue Book 2022, including the impacts of the coronavirus (COVID-19) pandemic on the UK economy. These revisions include the first instance of balancing GDP in a supply and use tables (SUTs) framework for 2020, as explained in our recent Impact of Blue Book 2022 changes on gross domestic product article. This allows us to produce one coherent estimate of GDP by balancing at the 112 industry- and 112 product-level, where we have incorporated more comprehensive survey and administrative information.

Given the large and volatile movements in GDP over the coronavirus pandemic period of 2020 and 2021, it would be expected that revisions would be larger, all else the same. Figure 1 shows the cumulative impact of revisions on current price and volume GDP relative to the Quarter 4 (Oct to Dec) 2019 level. This shows that these revisions have not fundamentally changed our understanding of the cumulative impact on headline GDP over this period.

Producing balanced estimates in a SUTs framework improves data coherence at a detailed level, given that this is the first opportunity for a comprehensive confrontation of the three approaches to GDP. This also includes incorporating our first estimates of intermediate consumption in 2020 at the industry level, so that we no longer rely on output as a proxy of gross value added (GVA). This can lead to revisions at turning points, as there can be changes in the relationship between inputs and outputs, as explained in our Coronavirus and the effects on UK GDP article. Volume estimates of GVA have also been produced for 2020 for the first time under double deflation, which has taken place in a SUTs framework. If output and input prices have not moved in line with each other over this year, then this can lead to revisions at the industry level.

These improvements from our annual SUTs data confrontation have impacts on the output, expenditure and income approaches to GDP. We look at some of the impacts of these revisions, particularly our understanding of the extent that the change in coronavirus restrictions over time had an impact on GDP.

One feature of the coronavirus (COVID-19) pandemic is how wide ranging the industry-level impacts have been through 2020 and 2021. This is because some industries had been more affected by the restrictions, particularly those services industries that had been more reliant on physical interaction. Previous estimates showed that these “high-contact” industries had experienced a larger peak-to-trough contraction through these lockdown periods in 2020 and 2021, but then also rebounded more quickly as these restrictions were lifted. For more information, see our Effects of the coronavirus (COVID-19) pandemic on "high-contact" industries article.

Figure 2 shows that the revised estimates still show that there was more of an impact on “high-contact” industries. There is a similar peak-to-trough decline in the first half of 2020 for these industries – this has been revised from a fall of 37% to a fall of 43% – that was most evident for the transportation and storage industry. There have been minimal revisions for “low-contact” industries, which contracted by a cumulative 18% in the first two quarters of 2020.

The latest revised estimates show there has been a weaker recovery of “high-contact” industries thereafter. It had previously been estimated that the output of “high-contact” industries had recovered to pre-coronavirus pandemic levels as of Quarter 2 (Apr to June) 2022. However, there has now been much less of a rebound – it is now still 10% below Quarter 4 (Oct to Dec) 2019 levels. This reflects lower output produced by the wholesale and retail trade, transportation and storage, and other services industries (Figure 3).

The revision to the wholesale and retail trade industries in 2020 comes primarily from the more detailed method for estimating distributor trading margins in the annual supply and use tables (SUTs) process, as discussed in our Impact of Blue Book 2022 changes on gross domestic product article. The revision to transportation and storage industries reflects richer information from annual structural surveys, which provide more accurate estimates of intermediate consumption and output to estimate gross value added (GVA). The rail and air transport industries are now estimated to have experienced negative GVA in Quarter 2 2020, where the costs of production were larger than the value of the output produced. These 2020 revisions explain why these industries are now estimated to have recovered by less.

Coronavirus (COVID-19) restrictions led to changes in the level and composition of household spending. This was most evident in the lower levels of spending on “social consumption”, which refers to spending on eating out, leisure travel and cultural activities. The response to the coronavirus pandemic was that households had been unable and/or unwilling to spend on these types of social activities. For more information on “social consumption”, see our International comparisons of GDP during the coronavirus (COVID-19) pandemic article.

Figure 4 shows the previous and latest profiles of “social consumption” – spending on recreation and culture as well as restaurants and hotels. The impact of these revisions has been minimal – the level of this type of spending compared with Quarter 4 (Oct to Dec) 2019 levels is unrevised as of Quarter 2 (Apr to June) 2022. This is also broadly in line with the output produced by the accommodation and food and arts, entertainment, and recreation industries (Figure 3).

There was also a large decline in spending on transport services such as on air and rail transport, reflecting the local and international mobility restrictions in place at the time. The revised estimates show there was a slightly larger decline in transport spending in the first half of 2020 – a revised fall of 64% compared with a previous estimate of 59%. As of Quarter 2 2022, the latest estimates show that transport spending is still below pre-coronavirus pandemic levels.

The changes in behaviour in how and what we consumed in response to the changes brought about by the coronavirus pandemic led to a change in the composition of household consumption expenditure. Figure 5 shows the profile of goods and services spending over this recent period. There have been minimal revisions to services expenditure over this time, although there have been some revisions to goods expenditure over this period. The latest estimates show the profile of demand rotation, including how goods expenditure did recover to above its pre-coronavirus pandemic levels in Quarter 2 2021. There has been some return in this composition of household consumption expenditure more recently, similar with previous estimates.

One feature of the coronavirus (COVID-19) pandemic has been the record increase in the saving ratio. The purpose of saving is typically to increase future disposable income available for consumption so to protect against any unexpected changes in income. However, the coronavirus pandemic also introduced “forced” saving, which reflected the inability to consume following restrictions on physical movement and social interaction during the coronavirus pandemic. This helped explain the record increase in household saving over 2020 and 2021. Figure 6 shows the previous and latest estimates of the household saving ratio over the course of the coronavirus pandemic. The upward revisions reflect lower levels of household consumption expenditure, while the level of disposable income is similar to that previously estimated.

Around £145 billion (74%) of the accumulated £196 billion increase in household saving over 2020 and 2021 had been estimated as “forced” saving, which was approximately 10% of annual household disposable income, as explained in our Economic modelling of forced saving during the coronavirus (COVID-19) pandemic article. Our latest estimates now show that 77% of the increase in household saving during the coronavirus pandemic is estimated to reflect “forced” saving. However, the revised estimates show a higher level of accumulated household savings (£251 billion) over 2020 and 2021. “Forced” saving is estimated to be around £194 billion, which is around 13% of annual household disposable income.

Producing estimates of gross domestic product (GDP) was expected to be more challenging through 2020 and 2021 for theoretical and practical reasons, as explained in our Coronavirus and the effects on UK GDP article. Revisions were expected to be larger, given the heightened levels of uncertainty. Figure 7 shows the recent historical revisions to the rolling three-monthly GDP figure. This is considered our best proxy in monthly GDP of the underlying trend in economic activity, we consider our quarterly estimate of GDP as our lead estimate, measured by the output, income, and expenditure approaches, where these are balanced to produce one coherent estimate of GDP. As expected, it shows that revisions around the time of the coronavirus (COVID-19) pandemic itself were larger.

However, the more recent revisions performance from January 2022 is closer to that experienced prior to the coronavirus pandemic. Future analysis will review these revisions performance once these periods have been fully balanced in our annual supply and use tables (SUTs) framework for a more comprehensive understanding.