Coronavirus and the latest indicators for the UK economy and society: 10 September 2020

• 36% of the workforce were working remotely and 11% were still furloughed, according to the latest Business Impact of Coronavirus (COVID-19) Survey (BICS). See Section 2.

• According to the new indicator, online prices of items in the food and drink basket decreased by 0.1% in the latest week (31 August to 6 September). See Section 3.

• For a third consecutive month, between June and July 2020, more firms reported increasing turnover than decreasing turnover, in the latest HMRC VAT business turnover returns. See Section 4.

• In the week starting Saturday 29 August, there was an average of 3,836 company incorporations per working day, an increase from the previous week's average of 3,066. See Section 5.

• Between 28 August and 4 September, total online job adverts decreased from 55% to 50% of their 2019 average, decreasing in every region and country of the UK. See Section 6.

• In the week commencing 31 August, Energy Performance Certificate (EPC) lodgements across England and Wales were 3% lower for existing dwellings and 17% lower for new dwellings than the same week a year ago, although this may be impacted by the bank holiday on 31 August. See Section 7.

• In the week commencing 31 August, footfall in retail parks remained around 90% of its level the same day a year ago, while footfall in high streets and shopping centres remained a little below 75%. See Section 8.

• According to traffic camera data, between 31 August and 6 September counts of cars in London have returned to the average level seen immediately pre-lockdown but the North East remained at around 95%. See Section 9.

• Department for Transport traffic count data show that on Monday 7 September, heavy vehicle traffic was four percentage points above traffic seen on the equivalent Monday in the first week of February, the highest recorded level since the Prime Minister's announcement on 16 March. See Section 10.

• Between 31 August and 6 September, the average volume of daily ship visits was 374, an  increase from the previous week's average of 324 daily visits. See Section 11.

This section includes final results from Wave 12 of the Business Impact of Coronavirus (COVID-19) Survey (BICS) for the period 10 August to 23 August 2020, which closed on 6 September 2020. Out of 23,904 businesses sampled, 22% responded.

More details on the results of this survey are in Coronavirus and the economic impacts on the UK.

Figure 1: 11% of the workforce remain on furlough leave, with 63% of furloughed employees receiving top ups to their pay

Headline indicators from the Business Impact of Coronavirus Survey, 10 August to 23 August 2020, UK

Source: Office for National Statistics – Business Impact of Coronavirus (COVID-19) Survey

Notes:

1. All percentages are a proportion of the number of businesses who responded apart from the workforce percentages on furlough leave and receiving pay top-ups, which are proportions of employees for each responding business.

2. Further breakdowns of the cash reserve categories can be found in Coronavirus and the economic impacts on the UK.

Download this chart

.XLSX

Of all responding businesses:

• 95% had been trading for more than the last two weeks

• 1% had started trading again within the last two weeks after a pause in trading

• 1% had paused trading but intend to restart trading in the next two weeks

• 3% had paused trading and do not intend to restart in the next two weeks

• less than 1% had permanently ceased trading

Businesses were asked about their working arrangements, shown in Figure 2.

When apportioned by employment size, of businesses not permanently ceased trading, 11% of the workforce were on partial or furlough leave, 36% of the workforce were working remotely, and 49% of the workforce were working at their normal place of work.

The arts, entertainment and recreation industry had the highest proportion of their workforce on partial leave or on furlough leave under the terms of the UK Government's Coronavirus Job Retention Scheme (CJRS), at 41%. This was followed by the accommodation and food service activities industry, and the administrative and support service activities industry, at 29% and 18% respectively.

For a more detailed outline of "Other" working arrangements across waves, please see the Coronavirus and the experiences of UK businesses, textual analysis: March 2020 to July 2020 article, which outlines how these "Other" working arrangements have changed over Waves 2 to 9.

A timely indication of weekly price change for a selection of food and drink products from several, large UK retailers has been developed and included here for the first time, covering the period 1 June (Week 1) to 6 September (Week 14) 2020. Details of the methodology used for these indicators, including strengths and weaknesses, can be found in Online price changes methodology. This analysis is experimental and should not be compared with our regular consumer price statistics.

As consumer shopping habits have now started to normalise since the earlier period of the pandemic, this broader selection of food and drink items has replaced the original High Demand Products (HDP) basket.

Online prices of items in the food and drink basket decreased overall by 0.1% between Week 13 and Week 14. Figure 3 presents the contributions to this weekly change from each of the main categories of items.

The largest contributions to the weekly change were seen in the following:

• mineral water, soft drinks and juices (which saw a weekly change of negative 0.8%) driven by decreases in the majority of items, with the largest contribution being from fruit juices (negative 0.6 percentage points)

• meat (negative 0.4%), with the largest contribution coming from ham (negative 0.1 percentage points)

• wine (1.3%), driven by an increase in all items in this category, particularly red wine (contributing 0.6 percentage points) and white wine (contributing 0.4 percentage points)

The time series, weekly growth rates and contributions to the weekly change for all individual food and drink items are published in a dataset alongside this release.

Figure 4 shows that the all item index remains below the starting point of the series (1 June), at 1.2% lower. Oils and fats have had the largest reduction in price since the series began, now 3.7% lower, along with sugar, jam, syrup, chocolate and confectionery and other food products, which are both 1.9% below the starting point of the beginning of June. Wine has shown the largest increase since the series began, now 1.8% above the starting point of the beginning of June, with a large part of that increase seen in the latest week.

New reporters

The new reporters index measures the number of firms sending VAT returns for the first time, which is related to the number of firm creations.

Business VAT turnover estimates

Turnover diffusion indices are an aggregate measure used to track whether the majority of firms are reporting an increase or decrease in turnover in their VAT returns. They are calculated as the percentage of firms with increasing turnover minus the percentage with decreasing turnover.

The heatmap in Figure 6 is a visual way of showing the VAT diffusion indices based on the standard deviation from their historical 2008 to 2019 average. Larger deviations are more darkly coloured, with red representing negative change, and teal a positive change, when compared with the previous period.

The GDP monthly estimate: July 2020 will be published on 11 September 2020. August 2020 estimates of Retail sales in Great Britain will be published on 18 September 2020.

Figure 6: For a third consecutive month, more firms reported increasing turnover between June and July 2020, than decreasing turnover, using the latest seasonally adjusted monthly all industry VAT business turnover returns

UK

Source: Her Majesty’s Revenue and Customs

Notes:

1. Agri – Agriculture, forestry and fishing, All – All industries, SA – Seasonally adjusted, NSA – Not seasonally adjusted.

2. All industries are unweighted: each firm contributing to the index has the same weight regardless of turnover, size or industry.

3. The thresholds for the colours in the heatmap are based on standard deviations from the mean of the indicator between 2008 and 2019.

Download this chart

.XLSX

Figure 6 shows that the month-on-month all-industry turnover diffusion index for July 2020 compared with June 2020 was 1.3 standard deviations above its historical 2008 to 2019 average (teal), with a diffusion index of 0.04. In other words, out of the 35,440 firms who reported monthly turnover between June and July 2020, around 1,400 more firms saw their turnover increase than firms who saw their turnover decrease.

Across all major industries (services, production, construction), turnover indices remained positive and similar in July 2020 to the month before, except for the agricultural industry where more firms saw their turnover decrease than increase.

Incorporations

Please note that following our standard methodology, the daily average for the week commencing 29 August is calculated per working day, excluding the bank holiday on 31 August. However, the incorporations processing team worked overtime on that day, which may explain why the daily average figure for company incorporations is higher than in previous weeks.

The observed fluctuations in weekly incorporations per working day between April to early May 2020 and June to the end of July 2020 coincide with government instigated lockdown measures and the subsequent easing of them in response to the coronavirus (COVID-19) pandemic. This is in line with official statistics published by Companies House on 30 July 2020.

Voluntary dissolution applications

For more information on other measures of company closures not presented here, see Weekly indicators of company creations and closures from Companies House methodology: August 2020.

These figures use job adverts provided by Adzuna, an online job search engine, and include experimental estimates of online job adverts by Adzuna category, and by UK country and NUTS1 regions. The number of job adverts over time is an indicator of the demand for labour. The Adzuna categories used do not correspond to Standard Industrial Classification (SIC) categories, so these values are not directly comparable with the Office for National Statistics (ONS) Vacancy Survey.

Figure 9: Between 28 August and 4 September, total online job adverts decreased from 55% to 50% of their 2019 average

Total weekly job adverts on Adzuna, UK, 4 January 2019 to 4 September 2020, index 2019 average = 100

Notes:

1. The observations were collected on a roughly weekly basis; however, they were not all observed at the same point in each week, leading to slightly irregular gaps between each observation.

2. These series have a small number of missing weeks, mostly in late 2019, and the latest is in January 2020. These values have been imputed using linear interpolation. The data points that have been imputed are clearly marked in the accompanying dataset.

3. Further category breakdowns are included in the Online job advert estimates dataset, and more details on the methodology can be found in Using Adzuna data to derive an indicator of weekly vacancies.

Download this chart

.XLSX

The five percentage point fall in the volume of online job adverts in the last week was spread quite evenly across the Adzuna categories, with a decline in 26 of the 28 categories (this count excludes the "unknown" category). In particular, the volume of online job adverts in the category of healthcare and social care decreased by 10 percentage points, bringing it to 84% of its 2019 average, its lowest level in 2020. Online job adverts in wholesale and retail also decreased seven percentage points, bringing them close to their levels throughout June and July.

The only two categories to show an increase are the categories of "legal" and "customer services and support", which each saw a one percentage point increase.

Figure 10: In the latest week every region and country of the UK saw a decline in the volume of online job adverts

Total weekly job adverts on Adzuna, UK, 4 January 2019 to 4 September 2020, index 2019 average = 100, percentage points

Notes:

1. There is a level shift in the Northern Ireland series from 17 October 2019 due to a large source of Northern Ireland job adverts being removed, and another level shift from 7 August 2020 because of a new source being included.

Download this chart

.XLSX

While every region saw a decline over the most recent week, the size of the declines varied across regions. The largest decline in online job adverts was in Northern Ireland, where they declined 11 percentage points of the 2019 average, followed by a nine percentage point decrease in Wales. The least affected regions were Scotland (decreased one percentage point) followed by London (decreased four percentage points).

Energy Performance Certificates (EPCs) are used as a timely indicator for the number of completed constructions (new EPCs) and number of transactions (existing EPCs). More detailed statistics split by NUTS1 English region are published weekly by the Ministry of Housing, Communities and Local Government (MHCLG).

This release includes weekly EPCs data for new and existing domestic properties in England and Wales from 24 February 2019 up to the week beginning 31 August 2020. This section analyses the percentage change compared with the same week the previous year. For example, the latest week commencing 31 August 2020 was compared with the week commencing 2 September 2019. This is shown in Figure 11.

In the latest week, EPCs for existing dwellings across England and Wales combined were 3% lower than the same week a year ago. The lower trend could be explained by the bank holiday on 31 August, which was earlier the previous year (26 August). This follows twelve consecutive weeks of existing EPCs remaining higher than the same week the previous year. This can be explained by lower than usual residential property transactions during March to June 2020 and a catch up in property transaction demand thereafter outlined by HM Revenue and Customs (HMRC) UK property transaction statistics (PDF, 336KB).

In the latest week, EPCs for new dwellings were 17% lower than the same week last year. EPCs for new dwellings have seen a slower recovery relative to existing EPCs during May to August 2020. A reduction in construction would contribute to the delay in EPC assessments of new dwellings.

These figures are provided by Springboard, a provider of data on customer activity. They measure the volume of footfall compared with the same day the previous year at the overall level and across the categories of high streets, retail parks and shopping centres. For example, Tuesday 14 July 2020 was compared with Tuesday 16 July 2019.

In the week commencing 31 August, overall footfall remained around 75% of its level the same day a year ago. This indicates a flattening off from the gradual but steady increase in footfall since the reopening of non-essential shops and businesses in England on 15 June.

The volume of footfall in all three categories of high streets, retail parks and shopping centres saw very little change from the previous week. Footfall in retail parks remained around 90% of its level the same day a year ago, while footfall in high streets and shopping centres remained a little below 75%.

The spike on Monday 24 August was caused by the comparison with 26 August 2019 last year, which was bank holiday Monday with good weather. The bank holiday this year was a week later on 31 August.

Figure 12: In the week commencing 31 August, overall footfall remained around three-quarters of its value the same day a year ago

Volume of footfall, year-on-year percentage change between footfall on the same day, UK, 1 March to 6 September 2020

Source: Springboard and the Department for Business, Energy and Industrial Strategy

Notes:

1. Many non-essential shops were allowed to open on 15 June 2020 in England, 12 June 2020 in Northern Ireland and 22 June in Wales. In Scotland, some non-essential shops were allowed to reopen from 29 June, and more from 13 to 15 July.

2. ‘Prime Minister’s announcement’ refers to the advisory announcement on 16 March 2020 to avoid non-essential travel, bars, restaurants and other indoor leisure venues, and to work from home if possible.

3. Pubs were allowed to reopen on 4 July in both England and Northern Ireland, with beer gardens allowed to reopen on 6 July in Scotland and 13 July in Wales. Pubs were also allowed to reopen indoors in Scotland on 15 July.

Download this image Figure 12: In the week commencing 31 August, overall footfall remained around three-quarters of its value the same day a year ago

.PNG (94.8 kB)

Traffic cameras are a valuable source for understanding the level of activity in towns and cities as well as changing patterns of mobility. The UK has thousands of publicly accessible traffic cameras with providers ranging from national agencies to local authorities.

The traffic camera images utilised in this analysis are publicly available, low resolution and do not permit people or vehicles to be individually identified. Our research shows that they are a good indicator for overall levels of "busyness" in urban areas. As such, they are a valuable complement to other mobility and traffic data to understand the economic and social effects of changing patterns of behaviour during the coronavirus (COVID-19) pandemic.

Currently, these indicators are not official statistics and are considered experimental. They provide insight and value, but they will be further improved upon over the coming weeks and months. Therefore, we expect these series to be revised.

These counts are not adjusted to estimate for the local population so are not a reliable estimate of the actual number of movements; however, they do provide a good estimate of the relative changes to number of movements. These traffic camera images capture the appearance of buses but they give no indication of the number of passengers using public transport.

A change has been made to the imputation methodology in this week's data. Please see Section 14 for more information. Further information on the methodology used to produce these data are available in our methodology article and Data Science Campus blog.

In the accompanying dataset, the following categories are available as non-seasonally adjusted, seasonally adjusted and trend data:

• cars
• motorbikes (only available for London and the North East)
• buses
• trucks
• vans
• pedestrians and cyclists

For the following regions (links are provided for each region's publicly available traffic cameras and the date when the Office for National Statistics (ONS) started collecting the data):

• Durham (since 7 May 2020)
• London (11 March 2020)
• Manchester (17 April 2020)
• North East (1 March 2020)
• Northern Ireland (15 May 2020)
• Southend (7 May 2020)
• Reading (7 May 2020)

These locations give broad coverage across the UK while also representing a range of different-sized settlements in both urban and rural settings.

Figure 13: Between 31 August and 6 September counts of cars in London have returned to the average level seen immediately pre-lockdown but the North East remained at around 95%

Activity in selected areas, daily counts of cars, buses, pedestrians and cyclists, seasonally adjusted, March to September 2020, UK

Notes:

1. 31 August was a bank holiday.

Download this chart

.XLSX

Average daily counts of cars in London for the latest week (31 August to 6 September) have returned to the level seen immediately pre-lockdown, whereas the North East remained at around 95% of their average pre-lockdown level (an average of 11 March to 22 March for London, and 1 March to 22 March for the North East, when each series began). The counts for pedestrians and cyclists were at similar levels.

In Northern Ireland, although data collection didn't start until 15 May 2020, the data continue to show a gradual increase in cars and pedestrians and cyclists.

More categories and areas are available in the accompanying dataset. Comparison to Department for Transport (DfT) road traffic estimates is shown in the accompanying traffic camera methodology article, published 3 September.

The Department for Transport (DfT) produces daily road traffic estimates using data from around 275 automatic traffic count sites across Great Britain covering all road types, which are published weekly.

The daily DfT estimates are indexed to the first week of February and the comparison is to the same day of the week. The data provided are useful as an indication of traffic change rather than actual traffic volumes. More information on the methods, quality and economic analysis for these indicators can be found in the methodology article.

On Monday 7 September, all motor vehicle traffic was six percentage points lower than traffic seen on the equivalent Monday in the first week of February. This is mainly driven by reduced car traffic, which remains around ten percentage points lower than traffic seen in the first week of February.

Figure 14 shows road traffic across all motor vehicles has been gradually returning to levels seen in the first week of February 2020 following a lockdown low point around the end of March.

These shipping indicators are based on counts of all vessels and cargo and tanker vessels. As discussed in Faster indicators of UK economic activity: shipping, we expect the shipping indicators to be related to the import and export of goods.

The time series of daily and weekly passenger visits have been temporarily suspended due to quality concerns. We are investigating and hope to reinstate these series in future releases.

On a seasonally adjusted basis, the average daily number of ship visits increased to 374 in the week 31 August to 6 September, an increase from 324 in the previous week. The lower average in the previous week could have been affected by storms Ellen and Francis which occurred on 21 and 25 August, along with high winds across the UK.

Company incorporations

Incorporations are when a company is added to the Companies House register of limited companies. This can also include where an existing business applies to become a limited company, where it was not one before.

Diffusion index

The diffusion index tracks the growth in turnover and expenditure of firms. It is constructed to lie between negative one and one. For example, if 65% of firms have increasing turnover, 30% have decreasing turnover and 5% turnover remains unchanged, then the diffusion index is 0.35.

Faster indicator

A faster indicator provides insights into economic activity using close-to-real-time big data, administrative data sources, rapid response surveys or Experimental Statistics, which represent useful economic and social concepts.

New reporter

A new reporter is defined as a firm with a VAT reference (that is, firm identification number) which has not previously reported its VAT returns. New reporters are published within one month of the end of the reporting period.

The new reporting behaviour measure is classified by the month the data were received by HMRC, known as the receipt date, which is not necessarily the same as the reference period (the period for which the VAT return is made).

For more information please see the methodology article on Value Added Tax returns

Voluntary dissolution applications

A voluntary dissolution application is when a company applies to begin dissolution proceedings. As such, they effectively chose to be removed from the Companies House register. For a company to be eligible to voluntarily dissolve, it should not have completed any trading activity for a period of three months.

Detailed information on the data sources, quality and methodology of the different indicators included in this bulletin is available in the Coronavirus and the latest indicators of the UK economy and society methodology.

The imputation methodology for the traffic camera time series has been updated in this release. Instead of applying the seasonal decomposition imputation to the entirety of the series, it is applied to the data from the latest four weeks, with the previously imputed data remaining unchanged. The main benefit of this change is making the processing is simpler and faster, which allows more time for quality assuring the results.

Further information on the methodology used to produce the new traffic camera data are available in the article and the data science campus blog.

We will summarise any crucial updates to the quality or methodology in this section in the future.

Detailed information on the strengths and limitations of the different indicators included in this bulletin is available in the Coronavirus and the latest indicators of the UK economy and society methodology.

We will summarise any crucial updates or warnings in this section in the future.