The business tendency surveys collected from companies and households, along with high-frequency data, give the same diagnosis for economic activity in November in France. Overall, activity losses appear to be lower than in the spring, although considerable nonetheless: in November, activity within the meaning of GDP would seem to be around 12% below its pre-crisis level – a slightly lower estimate than that given in the previous Economic Outlook, mainly due to the upward revision of the national accounts for Q3 – against about 30% below pre-crisis level in April.

The shock to household consumption would appear to have been of a fairly similar order of magnitude (–14% estimated in November, compared to the pre-crisis level), but was undoubtedly a little harsher than the shock to GDP. In fact, although we witnessed a growth in distance sales and other home delivery services, the purpose behind the health restrictions was to reduce the numbers of people mixing together and thus they mainly affected specific areas of household demand (closure of restaurants and bars, closure of certain “non-essential” leisure activities and businesses, limited transport).

In the short term, the economic situation remains very changeable and largely dependent on the health situation. The second wave of the epidemic is starting to decline, making it possible to set out some milestones for December: conditional on this decline continuing, health restrictions are expected to be eased in stages, with the reopening of all businesses by the end of November, then some leisure activities by mid-December, at which point a curfew is expected to follow on from the lockdown of the population. Other activities (especially access to indoor dining in restaurants) are unlikely to open until sometime in January.

With this information we can refine our scenario for the end of the year. We estimate that in December economic activity within the meaning of GDP could be at 8% below its pre-crisis level. Household consumption should rebound a little more sharply than GDP, to be 6% below its pre-crisis level. These figures will of course benefit from being confirmed by the high-frequency data when these become available for the beginning of December.

All in all, taking into account these hypotheses for the end of the year, GDP in Q4 2020 is likely to shrink by 4½% as a quarterly variation (after a rebound of +18.7% in Q3, as a result of the lifting of the first lockdown). Across the whole of 2020, GDP is expected to decline by about 9% compared to 2019.

This forecast for an annual contraction in GDP is similar to the one we made at the beginning of July. Meanwhile, the story told by the monthly figures was more irregular than expected. On the one hand, coming out of lockdown for the first time resulted in a more vigorous economic rebound than expected. On the other hand and conversely, the resurgence of the epidemic put a strain on economic activity in Q4.

Within each country, economic activity losses are far from uniform as they are closely dependent on the degree of exposure of each sector to the restrictive health measures in place. The global impact of the health crisis in the short term therefore depends mainly on the sectoral structure of each economy and the degree of severity of the restrictions, which was itself linked with the intensity of the epidemic.

France entered the second wave of the epidemic with a GDP in Q3 that was “only” 4% below its pre-crisis level, almost the same as Germany, but after some stronger losses of activity in the spring. High-frequency data, like the business tendency surveys, suggested heavier losses in France during the second lockdown too, but without anticipating the respective paces of the rebounds to come in the two countries. Meanwhile, Spain and the United Kingdom faced the second wave with a decline in GDP that was twice that of France and Germany compared to the pre-crisis level.1

Finally, even though the epidemic remains global, Asia appears at this stage to be relatively untouched by the autumn wave that is raging in Europe and North America. The Chinese economy in particular is continuing its recovery, after a shock that was certainly very severe but limited to Q1. China has recently become the foremost trading partner of the European Union, ahead of the United States.

1 To continue to measure economic activity during the health crisis, national accounting had to adapt its methods to the uniqueness of the situation, especially during the first lockdown. These adaptations could have temporarily affected the comparability of results between the different countries, especially concerning non-market services produced by general government. A post that appeared on 27 November on the INSEE blog provides an update on this subject: https://blog.insee.fr/sante-education-services-administratifs-la-difficile-mesure-des-activites-non-marchandes-en-temps-de-crise-sanitaire/. In fact, before the health crisis, the quarterly accounts did not seek to capture an infra-annual profile from these volumes of activity since non-market activities were not affected by short-term cycles. During the crisis, therefore, French accounts took the exceptional step of using the Employment Survey and various data sources to best reflect the economic outlook in this sector, but not all countries were able to do the same. The comparability of the data is therefore distorted, although on the face of it the proportions of the differentials are not inconsistent with orders of magnitude at the level of the economy as whole and operate only in the short term without introducing any trend bias.