Insee Analyses · November 2021 · n° 69

Impact of the health crisis on an anonymised panel of La Banque Postale customers Most customers’ incomes were affected in a limited and temporary manner

In 2020, in France, expressed in euro at a constant value, GDP shrank by 7.9% and household consumption fell by 7.1% [Insee, 2021]. In order to prevent the COVID-19 health crisis from affecting household incomes despite the usual social protection system (unemployment benefits, minimum social benefits, etc.), the State introduced exceptional support measures for employees (extension of the short-time working scheme, in particular) and households (payment of support to the most insecure). Finally, despite a sharp drop in production and consumption, the disposable income of all households increased by 1.0% and purchasing power per consumption unit remained stable [Insee, 2021]. However, this mean development does not represent the diversity of individual situations. For example, the volumes of food aid distributed by associations would have increased sharply in 2020 and the number of households receiving the RSA rose by 7.4% between the end of December 2019 and the end of December 2020 [Ouvrir dans un nouvel ongletDrees, 2021].

Banking data from La Banque Postale sheds light on the diversity of financial situations in 2020 (sources). They have the twofold advantage of being able to be used before the tax data are available and of recording monthly changes in income, consumption and wealth. Furthermore, they contain specific information on banking difficulties, such as overdrawn customers. This study uses a panel of 203,000 La Banque Postale customers with accounts active between January 2019 and June 2021. It expands on the results of a study carried out using Crédit Mutuel data [Insee, 2021], with a specific focus on the low income category.

Due to the specific nature of these banking data, the income, spending and wealth studied here do not correspond to the concepts usually used in INSEE publications. Incomes are measured based on incoming cheques and bank transfers. Gross financial wealth corresponds to the total assets across all La Banque Postale accounts (current accounts, savings accounts, life insurance and securities accounts) held by a customer. Spending corresponds to the total amount of card transactions, withdrawals, cheques issued and direct debits. The multidimensional nature of financial insecurity is understood here through three indicators: wealth insecurity, income insecurity and overdraft insecurity.

In this sample of La Banque Postale, customers, incomes rose by an average of 1.5% between 2019 and 2020. All three insecurity indicators were down throughout the entire period under examination, between January 2019 and June 2021 (Chart 1).

While these trends do not differ significantly from those observed at macroeconomic level, they also reflect the specific characteristics of La Banque Postale's customers and the ageing of the sample used. This does not mean, ipso facto, that insecurity was falling in France over the period. Indeed, La Banque Postale's customers are not representative of the population as a whole: in particular, when they are in employment, they are less likely to be managers and more likely to be workers. In addition, given that the panel is composed of the same customers from the beginning of the period to its end, due to the structure of the sample the June 2021 observations correspond to older individuals than those of January 2019, whose incomes may have increased independently of the economic context (because of their seniority in their job, for example). Furthermore, due to the structure of the sample, customers have a higher average length of time with the bank at the end of the period than at the beginning. In particular, this can affect the balances in the accounts.

Lockdowns affect incomes less than savings and wealth. During the first lockdown, health restrictions caused consumption to fall more than income, thereby allowing households to save. Thus, in this particular La Banque Postale, sample, the proportion of customers with less than 3000 euro in their accounts fell by 2 percentage points between March and May 2020 (42% in May). It then rose again slightly once lockdown was lifted, but remained below the pre-crisis level. During this first lockdown, the additional assets in the accounts reduced the proportion of overdrawn customers (-1 percentage point). The less stringent lockdowns in November 2020 and then in March 2021 had more moderate effects than the first: savings, wealth insecurity and the proportion of overdrawn customers per month were stable.

In contrast, the proportion of people with incomes of less than 1000 euro per month increased temporarily, by around 1 percentage point, during the first lockdown, whereas the trend was downwards over the whole period. Furthermore, it did not increase during the subsequent two lockdowns.

The panel of La Banque Postale customers makes it possible to measure the impact of the health crisis by comparing the trajectory of customers in 2020 with that expected if the pre-crisis trend of 2019 had continued (methods).

Using this method, incomes for the entirety of 2020 appear to be 2% lower than the expected “non-crisis” level, while spending was 6% lower. During the first lockdown from March to May 2020, spending and incomes were far below their level expected by extrapolating the pre-crisis trend (Chart 2). In April, spending was down 34% and incomes were down 10%. Outside this period, spending and incomes in 2020 were close to expected levels. The November lockdown had a much smaller impact on both spending and incomes.

Compared to the expected level, the shortfall in income during lockdown was somewhat greater and more persistent for the most insecure customers than for other groups. For the quarter of customers with the lowest incomes at the beginning of 2020, their average income was lower each month than that expected by extrapolating the pre-crisis trend; only November, the month in which exceptional support was paid to the most insecure customers, is an exception. For the year as a whole, this group’s incomes were 3% below the level expected.

The reduction in spending during the first lockdown encouraged saving: it thus reduced wealth insecurity and the proportion of overdrawn customers. In December 2020, the gross financial wealth of the quarter of customers with the lowest incomes at the beginning of the year was 360 euro higher than expected, compared with 1200 euro for the quarter of customers with the highest incomes. It is therefore those with high incomes who have saved more and added to their financial wealth [Insee, 2021].

The proportion of overdrawn customers was also below the level expected (-2 percentage points). The effect was stronger for those low incomes (-3 percentage points), as it was rarer for those with high incomes to be in this situation, regardless of the year considered. These results may appear to differ from those obtained from surveys. Thus, in the Epicov survey, a quarter of people report that their financial situation had deteriorated during the crisis and the lower the household’s standard of living at the outset, the higher this proportion [Givord, Silhol, 2020].

However, the perception of a deteriorating financial situation does not necessarily translate into a fall in the balances held in household bank accounts. If their incomes from work fall and the economic outlook darkens, households may see their financial situation as deteriorating even if their ability to save improves temporarily. Furthermore, the monthly consumer confidence (CAMME) survey shows that the proportion of households reporting that they are in debt or drawing on their reserves decreased in 2020, regardless of the standard of living [Clerc, Legleye, Nougaret, 2021].

In 2020, some of the most insecure and low-skilled jobs suffered more than others due to the crisis. Some people lost their jobs while others, who were inactive or unemployed before the crisis, have not been able to find a job.

Exceptional financial support was paid out in May and November to compensate for the increase in insecurity due to this fall in employment. La Banque Postale banking data makes it possible to measure the capacity of such support to prevent the deterioration of the situation of certain populations. The analysis focuses on single people, without dependants, in receipt of the maximum lump sum in January or February 2020 (at the end of 2019, single people without dependants represented 55% of RSA recipients [Ouvrir dans un nouvel ongletDrees, 2021]).

Over the whole of 2020, the incomes of these recipients were 4% lower (7% without support) than the level expected based on the pre-crisis trend (Chart 3). In particular, they were below the expected level in every month of the year, with the exception of May and November, when exceptional support was paid out. This phenomenon is likely due to returns to work being less common for those customers who were RSA recipients at the beginning of the year. In order to estimate the proportion of returns to work, the proportion of customers whose main source of income was no longer family allowance (caisse d’allocations familiales – CAF) benefit payments was calculated monthly (methods). In both 2019 and 2020, this proportion increased as people return to work throughout the year (Chart 4). However, in 2020, this increase diminishes from the first lockdown onwards. At the end of 2020, this proportion was 4 percentage points lower than in 2019. These results reflect the fall in the number of people who stop receiving RSA payments, which explains the increase in the number of recipients in 2020 [Ouvrir dans un nouvel ongletCnaf, 2020].

In 2020, while the average income of the RSA recipients studied was below the expected level, the median is identical to the expected level, except in May and November when it was higher. The majority of the RSA recipients studied did not experience a drop in income. In fact, as the majority of them were unable to find a job (in both 2019 and 2020), their incomes depended exclusively on social transfers, which did not fall in 2020, but increased as a result of the exceptional support. Furthermore, overdrawn customers are not more common, likely for the same reason, as well as because of the restrictions on consumption experienced during the lockdowns. Finally, these RSA recipients, whose spending mainly relates to basic necessities, reduced their consumption more slowly in 2020 than the rest of the sample.