The long-term effect of Covid-19 on European consumption differs in each country

The Covid-19 pandemic has led to considerable changes in consumer spending.

In general, we saw a mandatory shift away from high contact services like eating out at restaurants and bars, albeit involuntary and temporary, but some other changes could be here to stay and have a lasting influence on consumer behaviour.

Although it is uncertain when the pandemic will really be over, we expect most changes in demand to reverse, especially those imposed during lockdowns.

Here, we look at two important effects: the growth in online shopping and the trend of working from home.

Given that more consumers have discovered that buying online can be very convenient, we expect some of the increase in online retail sales to be permanent, at the expense of buying in physical shops.

Secondly, many have found that working from home can be very convenient and employers have realised that employee productivity doesn't necessarily drop.

According to a study by McKinsey, digitisation has been accelerated by the pandemic.

This is the combined effect of slow adopters overcoming their fear for purchasing online and further growth of online transactions by those consumers that were already on board. The closing of many retail stores during the lockdowns has also fast-tracked this trend. In countries like the Netherlands, where shopping online was already relatively common, a large share of consumers plan to continue to buy more online after the pandemic (see figure).

Within the EU, northern countries are leading the online shopping trend, and the south/east is overrepresented among countries lagging, suggesting a relatively large potential for growth in online buying.

But there is also some potential for growth in the leading countries. For example, Dutch consumers still spend 68% of their consumption outlays on goods and services in physical retail stores, according to the Thuiswinkel Marktmonitor.

This is why we have seen strong growth in online sales in leading countries, resulting in the rather limited ‘catch up’ by countries that have lagged during the crisis.

So the pandemic has not brought about a revolution but rather an accelerated evolution towards online business.

According to a survey by Payvision conducted in Germany, in the Netherlands and Belgium, two-fifths of households have shopped more online through big marketplaces, like Amazon. In the same survey, 44% of respondents said their buying habits developed during the pandemic are unlikely to change. Although the share of online consumption has dropped from the peak of the pandemic, we expect it to stay above pre-Covid-19 levels.

Because of health concerns and lockdowns, more people started working from home during the pandemic.

According to Eurostat, before the pandemic, only 14% of the workforce in the EU worked from home, with approximately two-thirds saying they did so sometimes, and a third said this was usual. During 2020, the share of those working from home grew to a fifth, with 60% saying this was usual. Nevertheless, the data shows that many people across Europe did not have the flexibility to work from home before the pandemic.

There are clear differences between countries when it comes to working from home. Before and during the pandemic, frontrunners were northern countries like Sweden, Finland, and the Netherlands, while France and Germany were in the middle, and southern and eastern European countries like Italy and Bulgaria fared towards the bottom.

The differences in working from home between countries are driven by the number of jobs that can be done from home, the level of digital infrastructure and worker skillset. We think culture is also an important determinant - do managers and employers trust employees to do a good job at home as they do in the office?

Restrictions on travelling to work also played a critical role.

In general, northern European countries have a higher share of jobs in services sectors such as ICT & business services, banking and insurance and public administration, which means that a larger proportion of workers can work from home. Northern countries like those in Scandinavia and the Netherlands also have better digital connectivity and more digitally skilled workers.

A broad divide between north and south/ east is also visible in people's trust in others, which is imperative in the working from home environment. In Italy, for instance, trust in other people is only half of what it is in Denmark, which somewhat complicates things when working away from the office.

We constructed an index for the likeliness of working from home after the pandemic using these three indicators.

After the pandemic, more people will work from home than before, but a lot less often than during the pandemic. Our index suggests that northern European countries are more likely to continue working from home than southern and eastern European countries.

Looking at the likeliness of working from home for different countries and factoring in the share of people working from home “sometimes” and “usually,” we think that 15% to 20% of workers will continue to work from home (compared to 14% in 2019 and 21% in 2020). Unlike during the pandemic, most of these will return to the office most of the time and only sometimes at home. On the whole, the average number of days worked from home will nevertheless increase over 2019, leading to less commuting.

One of the most visible effects of the pandemic has been the reduced mobility of people.

This is common during recessions, but in this case, the effect has been much stronger because of voluntary and non-voluntary restrictions on mobility for health reasons. Working from home and education from home resulted in substantially less commuting. Furthermore, the closure of shops and leisure activities, limits on international travel and curfews are responsible for people moving around less. International and domestic flying, for example, diminished by two-thirds in 2020.

There has also been a clear shift in the means of transport. In a recently published study, Christidis et al. estimate that the share of cars in total transport activity measured in km has risen by four percentage points since the pandemic. Walking, cycling and electric micro-mobility have increased during the pandemic, but due to the short average distances of these trips, the share of these transport modes in total transport activity has not risen significantly.

The reduction in mobility during the pandemic has already been reversed to a large extent with the reopening of the economy.

The underlying data of ING’s overall mobility index shows that visits to retail shops and recreation sites have recovered back to pre-pandemic levels. Visits to essential shops – grocery stores and pharmacies – have even jumped above pre-crisis levels at this point.

But mobility related to commuting has not recovered fully, in part due to the increase in working from home.

We expect this downward trend to be structural because we think the increase in working from home is here to stay. This is not only because workers want this but also because employers see cost-saving possibilities if many people work from home regularly.

We also expect the pandemic to have a structural effect on business travel. In particular, air travel is expected to remain below pre-pandemic levels, as more people will use video calls for business meetings. A survey by Oliver Wyman shows that 43% of business travellers expect to travel less for business even after the pandemic. About a third said teleconferencing and remote working arrangements were as effective as office and travelling.

Although the pandemic could have many effects on household consumption, we expect two main changes in consumer behaviour that will lead to changes in consumption:

• An increase in online shopping instead of buying in shops
• An increase in working from home instead of working in the office

Establishing the exact quantitative effects on consumption due to the above identified changes in consumer behaviour is difficult because data at the required disaggregated level is lacking. However, we can make a rough estimate taking into account the biggest effects.

Buying more online instead of buying in retail shops primarily changes the channels through which goods and services are consumed, not the level of demand for these goods. But there could be second-round effects that do have an impact on the level of consumption.

For example, less shopping in physical stores means that consumers will also spend less in bars and restaurants close to shops, possibly to the benefit of supermarkets and other food and drink-selling retailers. Buying online instead of buying in shops will also somewhat diminish the transport movement of consumers.

Working from home will also reduce consumer mobility significantly as it leads to less commuting, which often involves significant distances.

Spending on transport services like trains and buses, vehicles, and servicing these vehicles (petrol, oil, etc.) will decrease. Consumption of food and drinks in bars and restaurants near workplaces will also be lower. Catering services in offices will suffer from the increase in working from home as well.

More working, schooling and digital leisure activities at home mean more (replacement) demand for electronic devices like laptops. Consumption of communication services will be higher because of the increased importance of having a decent internet connection.

Only specific elements of spending are impacted by the long-term Covid-19 induced changes. However, data on the level of these specific elements aren't always available.

For example, within electronics, it’s mainly laptops and screens for which no detailed data is available within the eurozone, so we've had to estimate the size of this category. The same data problem holds for the share of food and drinks in catering related to working at the office, and that share is unknown.

We expect a positive effect on demand for some categories, like household demand for energy and water. We know the size based on the assumption that it is mainly spending by people working from home that will be affected permanently. Research shows that this is 8% of the total population.

We estimate the size of the consumption categories that are positively affected to be around 3.2% of the total consumption of European households. This is the relevant base to which an increase would apply.

The bulk of the consumption goods and services that are negatively influenced by the changes in consumer behaviour is made up of transport services and vehicles and vehicle operation and maintenance. This is due to less commuting by those who will work more days from home, less national business travel and less travel for shopping. The parts that are potentially influenced by downward effects add up to 3.3% of total consumption.

We expect the net effect of working from home on the value of consumption of food and drinks to be negative.

Volumes of the two effects will balance out, but in values, the gain for supermarkets and other retail food shops will be lower than the losses for restaurants, bars and catering as supermarket food and drinks are cheaper than those in bars and restaurants.

To establish the overall net effect, we also have to look at the actual change in other potentially influenced consumption categories.

On the positive side, there will be more consumption of energy and water, for instance, due to more working from home, but the effect will be small: even if homeworkers double the average number of days worked at home from one to two days, total consumption would rise by less than 0.1%. Increasing spending on internet connectivity will also be limited as many people already have a sufficient connection. For electronics, the effect is largely determined by a somewhat higher replacement rate because most people already have laptops and screens if required. Altogether, we expect the increase in spending on electronics, communication and energy & water to be limited in practice.

On the negative side, the decrease in commuting, business and shopping-related travel could reduce vehicle operation and maintenance consumption and transport services and vehicles by a significant amount. As these activities represent a bigger share of total consumption than electronics, communication, energy, and water, we expect an additional negative effect.

Using minimum and maximum assumptions on the change in spending on the relevant parts of electronics, communication, energy & water, transport and vehicle operation, we think this effect is small.

Looking at the derived likelihood of workers continuing to work from home after the pandemic, and the important role it plays in driving down consumption, we conclude that the consumption changes will probably be the largest in countries in northwest Europe and smallest in southern – and eastern-European countries. Although we think the net negative effect on overall consumption is small, the effect of changing consumption for specific industries and certainly individual companies could be significant. We will look into the effect on industry levels in subsequent articles.