AdobeStock_423295537.jpeg
VoxEU Column COVID-19 Europe's nations and regions

Heterogenous effects of Covid-19 on households' financial situation and consumption: Cross-country evidence from a new survey

The coronavirus pandemic has generated a complex economic shock that has affected households across the euro area very differently. This column uses survey data from around 10,000 households across Europe to reveal substantial divergences in the pandemic-induced financial concerns of households across population subgroups and countries. Financial concerns are significantly greater for younger, female, and low-income individuals in countries where the first wave of Covid-19 was more severe.

According to aggregate data for the euro area, household spending dropped by almost 7% in 2020 compared with 2019. The drop was stronger in France (-8.1%), Spain (-10.9%), and Italy (-8.9%). To understand how consumption adjusted at the household level, in our recent study (Christelis et al. 2020) we exploit household-specific information on how severe the financial consequences of the Covid-19 pandemic are perceived to be. One would expect pandemic-induced financial concerns to be negatively associated with consumption for several reasons. First, financial concerns depend on current income, access to liquidity, and accumulated wealth – with less wealthy households being less well-equipped to buffer the adverse consequences of the outbreak. Second, financial concerns are associated with lower income expectations (e.g. due to the lockdown measures), depending on the occupation, sector of activity, and remote working capability of household members. Third, financial concerns could reflect an increase in uncertainty about the future, because some households fear a higher probability of becoming unemployed, or because there is uncertainty about the duration of the crisis and the economic consequences of further Covid-19 waves. Financial concerns could also reflect other household-specific factors ranging from, for example, household size to concerns about future increases in the tax burden.

A growing number of studies investigate the consumption effect of the pandemic in the US and in Europe, relying mostly on administrative data (Baker et al. 2020, Cox et al. 2020, Chetty et al. 2020, Chronopoulos et al. 2020, Hacioglu et al. 2020, Dunn et al. 2020, Andersen et al. 2020, Bounie et al. 2020, Carvalho et al. 2020a, 2020b). They typically identify the effect on consumption using area-level measures of the impact of the pandemic (e.g. deaths per region). Notably, they do not rely on a household-specific measure of exposure to the Covid-19 shock that can capture the highly heterogeneous nature of the pandemic’s effect across different households. 

We use rich panel data from the Consumer Expectations Survey (CES) – a new ECB survey that was launched as a pilot in January 2020. Since April 2020 the survey has been running at its target sample size, interviewing around 10,000 households across the six largest euro area economies (Germany, France, Italy, Spain, the Netherlands, and Belgium) on a monthly basis (ECB 2021). The panel dimension of the data means that respondents are surveyed repeatedly, which is particularly important as it ensures that the dynamic response of the same households can be studied over time. The survey is representative of the underlying populations and collects (via the internet) high-frequency and fully harmonised information on households’ demographics, income, and consumption, and on how households perceive the economic consequences of the pandemic. For many variables, we are able to exploit the panel nature of the survey from April to October 2020.

Household financial concerns due to Covid-19

The survey asks respondents the following question on the economic impact of the pandemic: 

“How concerned are you about the impact of the coronavirus (COVID-19) on the financial situation of your household? (coded from 0, ‘not concerned’, to 10, ‘extremely concerned’).” 

Figure 1 plots the distribution of household financial concerns due within the six countries included in our study. The share of respondents concerned about their financial situation is higher in countries that experienced the highest number of cases and deaths and stricter lockdown policies during the first wave of the pandemic. In Italy and Spain, 36% and 52% of respondents, respectively, express high concerns (seven or above) about the financial consequences of Covid-19. Indeed, these two countries stand out with a significant fraction of households reporting the highest possible level of concern (ten). On the other hand, in Germany and the Netherlands the fractions expressing relatively high concerns (above seven) are 25% and 20%, respectively.

Figure 1 How concerned households are about their financial situation owing to COVID-19

Consumer Expectations Survey data April – October 2020 
(y-axis: share of households in percent, x-axis: level of concern with range 0 (“not at all concerned”) to 10 (“extremely concerned”)

 

Note: The figure shows the fraction of responses per level of COVID-19 financial concern and by country. Data are drawn from the April, July and October waves of the CES.

Further analysis shows that such financial concerns are not distributed evenly across the population. As shown in Table 1, the pandemic has induced higher financial concerns in younger age groups compared with those older than 65 (many of whom have retired and are therefore more likely to be more insulated from income shocks arising from the crisis). In addition to this, financial concerns are higher among lower-income households and households that are liquidity-constrained (i.e. they report that they are not able to meet an unexpected payment equal to one month of their household income). Furthermore, female respondents generally report higher overall financial concerns. 

The survey also includes separate questions on the health consequences of Covid-19. Unlike financial concerns, health-related concerns are considerably higher among the older (65+) and the middle-aged (36-64) households compared with the young. But formal econometric analysis shows that the effects of the outbreak on consumption mainly operate through households’ perceptions about the financial repercussions of the shock and not via their concerns about the effects of the pandemic on their own health.

Table 1 Differences in financial concerns due to COVID-19 across demographic groups

Consumer Expectations Survey data April – October 2020
(average based on a scale of 0 (“not at all concerned”) to 10 (“extremely concerned”))

 

Source: ECB Consumer Expectations Survey. Using weighted data. Individual-level concerns about the household’s financial situation have been elicited on a monthly basis since April by asking: “How concerned are you about the impact of coronavirus (COVID-19) with respect to the financial situation of your household?”, with an 11-step response scale ranging from 0 (not at all concerned) to 10 (extremely concerned). Liquidity constraints are inferred from the ability of a household to meet an unexpected payment equal to one month of household income.

How households adjusted their consumption in response to the Covid-19 shock

In Figure 2 we plot binned values of the (logarithm of) monthly non-durable consumption against the values of the measure of Covid-induced financial concerns (0-10) discussed above. Comparing those who are least concerned about the financial consequences of the pandemic (values of two and below) with those that are very concerned (nine or ten) implies a reduction in consumption of about 25%. Of course, this relationship does not consider other variables that affect consumption. Econometric estimates indicate that, controlling for other variables, raising concern from zero (the least concern) to six (the median concern) reduces consumption by 8.2%. On the other hand, concern about the impact of the disease on one’s own health, as well as the health of other household members, has no statistically significant impact on consumption. The results suggest that financial concerns are a much stronger independent driver of spending behaviour than health-related concerns – whose effects may instead be transmitted via the impact on the household’s expected financial conditions (e.g. if a job loss were to arise as a result of health problems). Further, as our estimation controls for current income, socio-economic variables, unobserved household traits (e.g. risk attitudes) and aggregate effects (e.g. country macroeconomic conditions), precautionary saving are a likely explanation for the negative association between financial concerns and consumption.

Figure 2 Households’ concern about their financial situation due to COVID-19 and the effect of this concern on consumption

Consumer Expectations Survey data April – October 2020
(y-axis: Log consumption, x-axis: level of concern with range 0 (“not at all concerned”) to 10 (“extremely concerned”)

 

Note. The figure shows a scatterplot and a fitted line of the natural logarithm of monthly non-durable consumption against the COVID-19 financial concern. Data are binned. Data are drawn from the April, July and October waves of the CES.

Conclusions

Our study suggests that easing pandemic-related financial concerns can counter the observed drop in spending and can reduce the extent to which households adjust their consumption in response to negative income shocks. In particular, the large divergences in financial concerns across households suggest that highly targeted government interventions that aim to lessen the financial concerns of younger households, lower-income households, and households with less stable employment conditions will support the consumption levels of these groups and contribute to economic recovery.

Authors’ note: This column first appeared as a Research Bulletin of the European Central Bank. The author gratefully acknowledges the comments from Michael Ehrmann, Alexander Popov, and Louise Sagar. The views expressed here are those of the author and do not necessarily represent the views of the European Central Bank or the Eurosystem.

References

Andersen, A, E T Hansen, N Johannesen and A Sheridan (2020), “Consumer Responses to the COVID-19 Crisis: Evidence from Bank Account Transaction Data”, Covid Economics 7: 88-114. 

Baker, R S, R A Farrokhnia, S Meyer, M Pagel and Y Constantine (2020), “How Does Household Spending Respond to an Epidemic? Consumption During the 2020 COVID-19 Pandemic”, Covid Economics 18: 73-108.

Bounie, D, Y Camara and J W Galbraith (2020), “Consumers’ Mobility, Expenditure and Online-Offline Substitution Response to COVID-19: Evidence from French Transaction Data”, CIRANO Working Papers 28.

Carvalho, P B, S Peralta and J Pereira (2020a), “What and How did People Buy during the Great Lockdown? Evidence from Electronic Payments”, Covid Economics 28: 119-158.

Carvalho, M V, S Hansen, Á Ortiz, R J García, T Rodrigo, R S Mora and J Ruiz (2020b), “Tracking the COVID-19 Crisis with High Resolution Transaction Data”, CEPR Discussion Paper 14642.

Chetty, R, N J Friedman, N Hendren, M Stepner and the Opportunity Insights Team (2020), “The Economic Impacts of COVID-19: Evidence from a New Public Database Built Using Private Sector Data”, NBER Working Paper 27431.

Christelis, D, D Georgarakos, T Jappelli and G Kenny (2020), “The COVID-19 Crisis and Consumption: Survey Evidence from Six EU Countries”, ECB Working Paper 2507.

Chronopoulos, D, M Lukas and J O S Wilson (2020), “Consumer spending responses to the Covid-19 pandemic: An assessment of Great Britain”, Covid Economics 34: 145-186.

Cox, N, P Ganong, P Noel, J Vavra, A Wong, D Farrell and F Greig (2020), “Initial Impacts of the Pandemic on Consumer Behavior: Evidence from Linked Income, Spending, and Savings Data”, NBER Working Paper 27617.

Dunn, A, K Hood and A Driessen (2020), “Measuring the Effects of the COVID-19 Pandemic on Consumer Spending using Card Transaction Data”, BEA Working Paper WP2020-5, US Bureau of Economic Analysis.

Hacioglu, S, D Känzig and P Surico (2020), “Consumption in the Time of Covid-19: Evidence from UK Transaction Data”, CEPR Discussion Paper 14733.

4,199 Reads