href="#ulink_768263d3-1535-58ba-96c5-b78f8446f5c7">Figure 11b). Such developments in income and consumer expectations make a quick rebound in consumer spending somewhat unlikely, even though strict government restrictions on movement have been largely avoided in the fourth quarter of 2020. The corporate sector is not optimistic about investment either, as discussed in detail in Chapter 2. The economic recovery is therefore likely to be more gradual and prolonged (Box A).
Box A
Real-time monitoring of the pandemic’s impact
Since the start of the pandemic, new data sources have become available that help assess economic activity in almost real time. Oxford University coordinates an effort to compile daily indicators of policy stringency (Blavatnik School of Government, 2020). Google provides daily measures of the extent to which people, under these restrictions, are still going to work (Google, 2020). Policies and mobility vary substantially across EU Member States but still show a common pattern (Figure A.1).
Figure A.1
Indicators of policy stringency and mobility trace the impact of the first and second waves of the pandemic, with significant diversity across EU countries
Source: Blavatnik School of Government (2020), Google (2020), and EIB staff calculations.
Note: Monthly averages of work-related measures of policy stringency and mobility. Each diamond shows an EU country. Lighter shades result from overlapping diamonds.The lines shows the GDP-weighted EU averages. Data were collected on 23 November.
These indicators help assess economic activity using relatively simple econometric specifications. We base our assessments on pooled linear regressions of economic activity (industrial production or service sector turnover) in EU members on visits to places of work and on a composite indicator of policy stringency (Table A.1). The policy stringency indicator is an average of the extent to which workplaces, schools and public transport are closed, the stringency of stay-at-home requirements and restrictions on movement within the country. The regressions are weighted by active population and contain country fixed effects.
Assessments based on these indicators suggest that EU GDP declined by about 1-2% in October and 5-6% in November. Industrial production and service sector turnover, used here as monthly proxies for GDP, move closely in line with the policy index (Figure A.2, dark blue and red lines). Our forecasts suggest that by November, the start of the second wave had undone most of the recovery witnessed since May (Figure A.2, light blue and red lines). Google’s mobility indicator points in the same direction but suggests a somewhat smaller decline in activity. Given that a substantial relaxation of policies in December seemed unlikely, EU GDP may fall in the fourth quarter by about 3-4% vs. the third quarter, leaving GDP about 7-9% below its pre-crisis level.
Figure A.2
The pandemic’s second wave appears to reverse the summertime recovery
Source: Blavatnik School of Government (2020), Eurostat, and EIB staff calculations.
Note: Forecasts computed using pooled, population-weighted regressions of industrial production and service sector turnover on a work-related subset of Oxford stringency indices and on country fixed effects. Data as of 23 November 2020.
Table A.1
Regression specifications and results
| Dependent variable | Level of industrial production1 | Level of service sector turnover1 | ||
| (1) | (2) | (1) | (2) | |
| Impact of a 1 point increase in the policy stringency index | -14.0[-16.0, -11.0] | -16.5[-20.6, -12.2] | ||
| Impact of a 1 point increase in mobility | 0.64[.55, .73] | 0.7[.61, .79] | ||
| R2 | 73.5% | 74% | 84% | 89% |
| N | 223 | 199 | 80 | 55 |
Source: EIB staff calculations.
Note: 1 Seasonally and calendar day adjusted. 95% confidence intervals in square brackets.
As long as compliance with restrictions is high, the policy stringency indicator appears more useful than the mobility data in assessing economic activity. Google’s mobility indicator can be seen as a measure of compliance with the restrictions and might therefore be a more direct measure of activity. However, the mobility indicator shows pronounced seasonal variations, which detract from the underlying momentum in activity. For example, it declined during the summer holidays in August. And so far, surveys do not seem to suggest that compliance with restrictions is significantly declining in EU countries (Institute of Global Health Innovation, 2020).
The impact of the coronavirus pandemic varies substantially by sector. Sectors that rely significantly on physical presence, including passenger transport, the arts, entertainment, tourism and hospitality, were hit the hardest, declining by some 30% in the second quarter of 2020 from the first quarter. Others, such as agriculture, finance or real estate, contracted by 3% or less over the same period. The distribution of the economic impact across the various sectors was very different during the global financial crisis, when EU manufacturing sustained the largest decline – nearly 20% in the first quarter of 2009. The drop in other sectors remained relatively contained at near or below 6%.
The sectoral distribution of the decline will have a decisive impact on the speed of the economic recovery in the near to medium term. The industrial sector’s share in the overall decline in 2020 is the same as during the global financial crisis, while that of services is much higher (Figure 12). Given that a large part of the contraction in services is due to their being delivered in person, as is the case in passenger transport or accommodation, the recovery of a large part of the services sector will remain subdued until the pandemic is reined in, especially as many government restrictions on economic activity were being reintroduced in the fourth quarter of 2020. On the other hand, the recovery of the industrial sector, where manufacturing dominates, is dependent on the upturn in international trade. The different speeds of recovery exhibited by manufacturing and services became clear over the summer when the industrial sector bounced back fairly quickly, while certain services lagged significantly behind.
Figure 12
Gross value added of all industries (% change vs. the same quarter in the previous year)
Source: Eurostat and EIB staff calculations.
Note: The regions of Europe are referred to simply as Western and Northern, Southern, and Central and Eastern in subsequent charts throughout the Investment Report.
The speed of the recovery is likely to be uneven across the European Union. The decline in services, especially trade, transport and hospitality, is much larger in Southern Europe than in the rest of the European Union. Because these services represent a large share of the economies of Southern Europe, they will weigh significantly on the recovery, both weakening it and stretching it out over time.