The spread of Covid‑19 and the measures to contain it are having a significant impact on the UK and many other countries around the world (Gopinath 2020). In April, we refocused the Decision Maker Panel (DMP) – a survey of chief financial officers from small, medium and large UK businesses – to assess how British businesses were being affected by this outbreak. New questions added covered topics such as expected sales, employment and investment impacts, supply disruption, uncertainty and demand for credit. This built on two questions that were included in the March survey. Those March survey results were summarised in a previous Vox column (Bloom et al. 2020).
The April DMP was in the field between 3 and 17 April. It received around 2,700 responses. Although this was about 200 fewer than the monthly average over the last year, there was no clear pattern in changes in responses rates by sector. In other words, the DMP continues to be a large and representative survey of UK businesses. Relative to other business surveys, the DMP has a number of advantages: it is timely, it is more quantitative than other surveys, and the results can be disaggregated to help better understand what lies behind the headline numbers. Over the coming months the DMP survey will continue to monitor how UK businesses are being affected by the outbreak of Covid-19, including asking more questions about the expected persistence of the impacts in the May survey.
Overview of expected impacts in 2020 Q2
The spread of Covid-19 was expected to have a large impact on UK businesses in 2020 Q2. In the April survey firms expected their sales in Q2 to be 44% lower, on average, than they would have otherwise been (Figure 1). DMP data are weighted by industry and firm size using employment data. The expected sales impact was a little smaller at 39% if weighted by GVA instead (and it was 41% on an unweighted basis).
Covid-19 was expected to lower investment by 50% in Q2 and to reduce employment by 18% (Figure 1). In addition, 36% of employees were reported to have been furloughed in April – i.e. still employed but not required to work any hours. At the time the survey was in the field, the government’s Coronavirus Job Retention Scheme to pay 80% of the wages of furloughed workers (up to £2,500 a month) was only due to run until the end of May, so it is possible that some of the expected reduction in employment in Q2 could reflect businesses thinking they may need to lay people off before the end of Q2 after the scheme ended. That scheme has since been extended until the end of October.
Figure 1 Expected impact of Covid-19 on sales, investment and employment in 2020 Q2
Note: The results are based on the questions ‘Relative to what would have otherwise happened, what is your best estimate for the impact of the spread of Covid-19 on the sales/capital expenditure/employment of your business in 2020 Q2 (April to June)?’ and ‘Approximately what percentage of your employees fall into the following categories as of April 2020?’ Respondents could assign their employees to the following categories: (i) Still employed but not required to work any hours (e.g. ‘on furlough’); (ii) Unable to work (e.g. due to sickness, self-isolation, childcare etc.); (iii) Continuing to work on business premises; and (iv) Continuing to work from home.
Impacts by industry
Substantial impacts on sales were expected across all sectors in Q2 (Figure 2). But the largest impact was expected in the accommodation and food sector, where sales were expected to be around 80% lower than they would have been. That was followed by recreational services, construction and wholesale and retail, which reported the next largest impacts. The smallest impacts were expected in health and in other production (agriculture, mining and quarrying and utilities). The expected employment impacts across different industries were well correlated with sales impacts.
The sales and employment impacts were expected to be largest in sectors that have relatively low levels of productivity and wages (Figure 2). For example, accommodation and food, recreational services and wholesale and retail are among the industries that expect the largest impacts, and all are relatively labour intensive and have lower than average levels of productivity and wages. Figure 3 highlights the strong correlation between previous labour productivity and expected impacts on employment at the firm level. This correlation is very similar if average wage per employee rather than labour productivity is used, given the strong relationship between those two variables.
Whilst the expected sales impacts were largely unrelated to businesses’ previous financial positions, those financial positions do appear to be affecting how businesses expect to respond. In particular, for a given sales impact, businesses that reported having less cash in their last set of accounts were more likely to say that they expected a larger impact on employment and to furlough more people.
Figure 2 Expected impact of Covid-19 on sales and employment in 2020 Q2 by industry
Note: Please see footnote for Figure 1 for details of questions.
Figure 3 Expected impact of Covid-19 on employment in 2020 Q2 and labour productivity
Note: Please see footnote for Figure 1 for details of questions. Labour productivity is calculated from last set of reported company accounts based data from Bureau Van Dijk. Labour productivity is defined as valued added (operating profit plus remuneration) per employee. This figure is a binscatter plot where each dot represents 5% of firms grouped by labour productivity.
Apart from falling demand, firms might also reduce output because they are unable to obtain crucial inputs. In April, around half of firms said that they were experiencing no disruption to their non-labour inputs – in other words, their supply chain was unaffected. But the other half were experiencing disruption, and for some a large proportion of their inputs were being disrupted (Figure 4) – around a quarter of firms were experiencing disruption to at least 25% of the inputs that they buy. This question explicitly referred to services that firms use as inputs as well as goods. At the firm level, supply disruption was well correlated with expected sales impacts (correlation coefficient of -0.35), implying that supply effects could be a factor holding back output too. This is consistent with the presence of cascading supply chain failures in Baqaee and Farhi (2020). However, disruption to labour supply appeared to be less important. On average, only 4% of employees were reported to be unable to work in April due to sickness, self-isolation or childcare responsibilities.
Figure 4 Disruption to non-labour inputs from Covid-19, April 2020
Note: Results are based on the question ‘How has the spread of coronavirus (Covid-19) affected the availability of the non-labour inputs your business uses as of April 2020?’
Distribution of investment impacts
Turning to investment, the expected impact of Covid-19 on investment in Q2 was well correlated with the expected effects on sales. But a striking feature of the expected investment impacts at the firm level was the bimodal nature of the distribution (Figure 5). Just under 30% of businesses reported that they expected Covid-19 to have no impact on their investment in Q2. A similar proportion said that they expected to reduce investment by more than 90%, relative to what would have otherwise happened. By contrast, the expected sales impacts were more evenly distributed. The result that firms tend to either continue with investment as planned or stop altogether would be consistent with some investment projects already in train having large adjustment costs that create a strong incentive to continue with the investment even in the face of a sharp fall in demand. In contrast, for projects where those costs have not yet been incurred, there will be a strong incentive to cancel the project altogether.
Figure 5 Distribution of expected impact of Covid-19 on sales and investment in 2020 Q2
Note: Please see footnote for Figure 1 for details of questions.
Alongside large expected sales impacts, there was also a further sharp increase in uncertainty in April. The DMP has a number of different measures of uncertainty. In April, 84% of businesses reported that the overall level of uncertainty facing their business was high or very high, up from 68% in March and 40% in February. And almost 90% of firms reported that Covid-19 was their largest current source of uncertainty in April compared with 50% in March. Figure 6 shows that there was a big jump in a measure of uncertainty based on the dispersion of the distribution of year-ahead sales expectations. Koslowski et al. (2020) demonstrate that large rises in uncertainty can have a ‘scarring’ effect on entrepreneurs’ beliefs that could reduce investment long after this shock has been resolved.
Figure 6 Uncertainty around year-ahead sales growth
Note: Data are average standard deviations around expected year-ahead sales growth. This is derived from a question asking respondents to attach probabilities to five different possible outcomes for year-ahead sales growth. The five outcomes are chosen by the respondent.
UK firms expect their sales to fall very sharply in 2020 Q2, they expect to cut or furlough a significant proportion of their workforce, and there are signs that supply chain disruptions are beginning to cascade throughout the economy. Investment and employment are being reduced in response to falling sales and rising uncertainty. This is consistent with other indicators. The big question is how quickly the economy will bounce back (Cerra et al. 2020). We aim to measure firms’ expectations of this in our survey and to report back on this next month.
Baqaee, D and E Farhi (2020), “Supply and Demand in Disaggregated Keynesian Economies with an Application to the Covid-19 Crisis”, CEPR Discussion Paper 14743.
Bloom, N, P Bunn, S Chen, P Mizen and P Smietanka (2020), “The economic impact of coronavirus on UK businesses: Early evidence from the Decision Maker Panel”, VoxEU.org, 27 March.
Cerra, V, A Fatas and S Saxena (2020), “The persistence of a COVID-induced global recession”, VoxEU.org, 14 May.
Gopinath, G (2020), “Limiting the economic fallout of the coronavirus with large targeted policies” in R Baldwin and B Weder di Mauro (eds), Mitgating the COVID Economic Crisis, a VoxEU Ebook, CEPR Press.
Koslowski, J, L Veldkamp and V Ventakeswaran (2020), “Scarring Body and Mind: The Long-Term Belief-Scarring Effects of COVID-19”, mimeo.