The state of financial knowledge in the European Union: a new survey

Abstract

For the very first time, in the Spring of 2023, the European Commission (EC) carried out a survey across all member states to assess their level of financial literacy. This survey complements other national surveys and fills an important gap because it provides a consistent metric that allows comparisons among the European Union (EU) countries. The motivation behind the EC’s survey stems from the need to advance the state of financial literacy to safeguard financial stability and promote important projects, such as the creation of a Capital Markets Union. In this paper, we analyze these new data and confirm findings in the literature about the importance of being financially knowledgeable to achieve good financial outcomes. Unfortunately, the survey also confirms that barely one in two individuals, on average in the EU, is financially literate.

Introduction

The survey “Monitoring the level of financial literacy in the EU, Flash Eurobarometer 525” is the first pan-European Union (EU) tool to assess the level of financial literacy, providing data for all 27 European Union member states (EU-27). It fills an important gap and allows researchers to compare levels of financial literacy across the EU-27 for the first time. Prior to the release of this dataset, the main available tool for comparing the level of financial literacy in European countries was the Global Financial Literacy Survey, whose data were collected once in 2014 (Klapper and Lusardi, 2020) or the OECD/INFE dataset.¹ Not all OECD members participated in these surveys, which included some EU member states. This prevented both policy makers and researchers from carrying policy and empirical work at the EU level. Naturally, there are many country initiatives that inform the discussion, but they are often not coordinated nor use comparable measures. This is why this initiative by the European Commission (EC) is of important value added and might set the starting point of multiyear dataset of financial literacy for the 27 EU countries.

The survey was conducted between March and April in 2023. It was requested by the European Commission’s Directorate-General for Financial Stability, Financial Services and Capital Markets Union (DG FISMA), and was coordinated by the Directorate-General for Communication (DG COMM). The field work was carried out by Ipsos European Public Affairs² and was conducted through online interviews.³

Improving financial literacy is a cross-cutting goal that supports many of the policy objectives of DG FISMA, such as speeding up the work on the Capital Markets Union and improving financial stability⁴. Moreover, promoting financial literacy is an important part of improving citizens’ financial decision making and its contribution to EU policy objectives. All EU countries now have, or are in the process of, developing national strategies for financial literacy. This dataset is potentially the beginning of an invaluable set of data collection that will accompany these strategies and contribute to their evaluation.

This paper is organized as follows: In section 1, we first provide a description of the survey. In section 2, we analyze the state of financial knowledge in the EU. We then study the relationship between financial knowledge and financial outcomes in section 3. In section 4, we carry out a case study for Belgium using the country microdata, which is much richer than the aggregated data. We conclude with some remarks and comments about the data and suggestions for future work. As mentioned throughout the paper, we only make use of a small part of the available data, and a lot of research can be done using this new survey covering as many as 27 countries.

Description of the survey

We first describe some important details of the survey such as the sample size and the ways the data are aggregated. Then, we provide an overview of the individual questions.

Survey characteristics

A total number of 26,139 people (18 and older) were interviewed in the EU-27. The country sample size was around 1,000 for all the countries except for Luxembourg, Cyprus, and Malta, whose sample size was around 500. The sampling quota was based on the following variables:

1. i) age (18–24 years old, 25–34 years old, 35–44 years old, 45–54 years old, 55–64 years old, and 65+ years old);

2. ii) gender;

3. iii) level of education; and

4. iv) geographic NUTS region (this depended on the country size and the number of regions).

The country-level data are obtained from the weighted⁵ individual data. The weights are based on the following aggregates’ distributions:

1. a) gender;

2. b) education;

3. c) working status;

4. d) regions’ population.

For the overall aggregates of the EU-27, the data from each member state is weighted based on the proportion of their adult population relative to the total adult population of the EU-27.

The specific socio-economic and demographic variables covered by the survey are detailed in the appendix (see Table 3A). For some components such as age, different levels of disaggregation for the same variable exist.

Questions covered in the survey: An overview

Financial literacy is a multifaceted aspect including both knowledge and the motivation to apply the knowledge acquired, resulting in better (financial) behavior. The Commission survey asks 15 questions to identify this set of components. We group the questions into four main categories (see Table 1A in the appendix):

1. a) Self-assessment of financial knowledge

2. b) Financial knowledge

3. c) Financial behavior

4. d) Financial outcomes

Self-assessed financial knowledge

This question is designed as multiple choice, with six possible answers including the “I don’t know” option. It aims to assess the self-evaluation of the individual in financial knowledge compared to her adult peers in the same country.

Financial knowledge

Previous literature (Lusardi and Mitchell, 2011) has identified three main economic concepts individuals should have a good understanding of in order to make good financial choices, known as the “Big Three”: compound interest, inflation, and risk diversification.

While questions covering these three main concepts have been shown to be good proxies for financial knowledge, extended versions of these three components have also been used in other surveys (OECD, 2014). This is also the case in the Commission survey, which evaluates two additional components of financial knowledge beyond the “Big Three”: the relationship between bond prices and interest rates⁶, and the relationship between risk and return. These five questions are designed as multiple choice questions, in which only one answer is correct. There is the possibility to answer “I don’t know”, which we consider to be an incorrect answer.

Financial behavior

A second component of the survey is the measure of behavior. Unlike financial knowledge, the behavior is more qualitative. In the Commission survey, three questions are asked to assess:

1. i) affordability;

2. ii) monitoring; and

3. iii) long-term planning.

Respondents are given a set of options which are classified either as financially savvy or nonfinancially savvy choices. The “I don’t know” option is again also allowed for.

Financial outcomes⁷

The third component measured by the survey is financial outcomes. It assesses the attitudes of individuals with respect to some financial aspects and also their financial decisions.

Five questions were asked in total, covering the following outcomes:

1. i) having emergency savings;

2. ii) financial products held now or in the last two years;

3. iii) confidence in having sufficient money for retirement;

4. iv) comfort with digital finance; and

5. v) trust in financial services.

The exact wording of the questions is provided in the Appendix (Table 2A).

Using these questions and the socio-demographic breakdowns, we analyze the level of financial knowledge as well as some cross-question relationships. We do not consider all questions available – further analysis is left for future research.

Assessing the state of financial knowledge in the EU

Main results

As discussed earlier, in its survey measuring financial literacy across the EU, the EC followed the OECD’s definition,⁸ according to which financial literacy is a combination of knowledge, attitudes, and behavior. In line with the academic literature, we report the findings on financial knowledge only and examine separately how financial knowledge is linked to financial outcomes.⁹

We assume that respondents are financially knowledgeable if they answer at least three (of the five) financial knowledge questions correctly. This threshold is consistent with other work measuring financial knowledge across countries (Klapper and Lusardi, 2020).

Based on this definition, 52% of respondents on average in the EU are financially knowledgeable, with the lowest level being 36% and the highest 73% for a specific country (Figure 1). This is a worrying finding as the threshold for being financially knowledgeable is low, and the survey measures understanding of fundamental concepts that are of relevance in daily financial decision making.

Figure 1. Financial knowledge in the EU.¹⁰

Source: European Commission survey (2023).

Notes: The measure of financial knowledge is based on five questions. Individuals are financially knowledgeable if they answer at least three questions correctly.

In the appendix (see A.2), we summarize the results for all five questions.

Questions 3 (inflation) and 5 (relation between risk and return) were the most understood questions with around 65% of respondents answering them correctly. On the other end of the spectrum, the great majority of respondents showed little understanding of the relationship between interest rates and bond prices: only 20% of the respondents answered this question correctly. This is a question that enables a distinction between low and advanced financial knowledge (Lusardi and Mitchell, 2014).

Given the high levels of inflation in all EU countries, and its disproportionate impact on the poor (Claeys et al., 2024), it is significant that as much as 35% of people on average in the EU do not know how the level of inflation affects their purchasing power. The level of understanding varies across sociodemographic groups (Figure 2). There is a 15 percentage-point gap in the understanding of inflation between the lowest and highest household income brackets. There are similar gaps across age and education; specifically, there is a 14 percentage-point gap between the youngest and the oldest respondents and a 15 percentage-point difference between those with the lowest and highest education levels.

Figure 2. Inflation knowledge of different groups, EU, % of respondents answering correctly.

Source: European Commission survey (2023).

Notes: Levels of education are defined as follows: Low is defined as including up to lower secondary education. Medium is defined as including from upper secondary education up until post-secondary but non-tertiary education. High is defined as including tertiary education up until doctoral level. (Source: International Standard Classification of Education (ISCED) – Statistics Explained (europa.eu)).

An important and well-documented finding in the literature is the gender gap in financial knowledge (Hasler and Lusardi, 2017). Figure 3 shows an 18 percentage-point gender gap in financial knowledge between men and women. This is in line with what other studies have found (Lusardi and Mitchell 2014, 2023; Klapper and Lusardi, 2020). Women are always more likely (and up to twice as likely, depending on the question) to answer “I do not know” to the financial knowledge questions. This is consistent across all financial literacy questions and every member state. In 133 out of 135 observations (27 member states multiplied by 5 questions), women used this option more than men. Previous studies have shown that by removing this option, the gender gap can be reduced by as much as a third (Bucher-Koenen et al. 2021).

Figure 3. The gender gap in financial knowledge in the EU (%).

Source: European Commission survey (2023).

Notes: Financial knowledge corresponds to the share of respondents who answered at least three (out of five) questions correctly.

A deeper look into demographics and socioeconomic conditions

Thanks to the different levels of disaggregation available, it is possible to analyze the level of financial literacy (knowledge and behavior) across different demographic and socioeconomic conditions for each EU country (see Table 3A for an overview of the available variables).

We apply our metric for financial knowledge across gender, age, income, and educational level. In Table 7A–Table 10A in the Appendix, we carry a similar analysis focusing on the inflation question, which is a particularly important topic given the recent high inflation rates across the EU countries.

Gender gap

We find that the gender gap in financial knowledge, consistently favoring men, is recurrent for every single member state (see Figure 4). Eight countries display a gender gap higher than 20 percentage points, with Romania showing the smallest gender gap (smaller than 10 percentage points).

Figure 4. Gender gap in financial knowledge.

Note: The figure reports financial knowledge for men and women across countries. Countries are sorted by gender gap size.

Financial literacy across age

As the number and importance of financial decisions that individuals make increase throughout the lifetime, it is important to explore the relationship between age and financial knowledge. Because this is a single cross section of data, it is not possible to distinguish between age and cohort effects. The data show that financial knowledge increases with age/cohort in most countries, with those older than 55 being more financially knowledgeable. The age gap between those aged 18–24 years and those 55 years and older is remarkably high across all member states, with the exception of Lithuania, Hungary, or Estonia (see Figure 5). In Denmark and Austria, this age gap is more than 30 percentage points.

Figure 5. Financial knowledge by age group.

Note: The figure reports financial knowledge in the EU countries for the young (age 18–24) and the old (age 55+ years). Countries are soted by financial knowledge gap size.

Financial literacy across income brackets

There are also large gaps in financial knowledge across income brackets. Figure 6 shows the gap in percentage points between the financial knowledge score for the richest households (5^th income quintile) and the financial knowledge score for the poorest households (1^st income quintile). Without fail, higher income households score higher than lower income households, although substantial differences exist across member states. For example, Cyprus, Malta, and Lithuania show gaps in financial knowledge across income levels of more than 50 percentage points.

Figure 6. Financial knowledge, rich vs poor households.

Note: The figure reports financial knowledge for rich and poor households across countries. Countries are sorted by financial knowledge gap size.

Financial literacy across education

Similarly, highly educated people score higher than lowly educated people across all member states (Figure 7).¹¹ The average gap in the EU is around 23%, with Austria displaying a gap closer to 50%. Spain is the country with the smallest gap, lower than 10%.

Figure 7. Financial knowledge, highest vs. lowest education levels.

Note: The figure reports financial knowledge for high and low educational attainment levels across countries. Countries are sorted by financial knowledge gap size.

How does financial knowledge affect financial outcomes?

New evidence from the European Commission survey

The European Commission (2023) survey asked questions pertaining to financial resilience in both the short and long term. We report how financial knowledge is linked to both financial fragility and pension security across EU countries. We picked those questions as they are relevant for both academic research and for policy makers (see also Demertzis et al. 2020 and Lusardi and Mitchell 2011, 2014). The appendix includes a summary of the questions and the detailed results. Figure 8(a) shows a negative association between financial knowledge and financial fragility, i.e., the less financially knowledgeable are less likely to have adequate resources to deal with an unexpected loss of income without having to borrow.

Figure 8. Financial knowledge, financial fragility and pension security.

Source: European Commission survey (2023).

Notes: This figure reports the correlation between financial knowledge (at least three out of five questions answered correctly) and: (a) the % of respondents who self-report having savings of less than three months’ worth of income to cover living expenses if their main source of income was lost today; (b) the % of respondents who self-report the degree of confidence to which they have enough money to live comfortably through their retirement years (sum of “very confident” and “somewhat confident”).

The data show that a 10 percentage point higher level of financial knowledge is associated with a 4 percentage-point reduction in people reporting insufficient means to cover their living expenses following an unexpected loss of income.

Similarly, Figure 8(b) shows that greater financial knowledge is associated with greater confidence in having sufficient financial means for retirement. A 10 percentage point higher level of financial knowledge is associated with a 6 percentage point higher number of people who report having sufficient means for retirement.

The contribution of financial knowledge to financial inclusion

We next study how financial knowledge is associated with measures of financial inclusion, another important topic for policy and programs (Batsaikhan and Demertzis, 2018; Lusardi and Messy, 2023).

We use two variables from the World Bank Global Findex database on the percentage of respondents who report saving with or borrowing from a financial institution. Figure 9 shows the relationship between financial knowledge (as measured by European Commission 2023) and the number of adults that (a) save with and (b) borrow from a financial institution. A 10-percentage point higher level of financial knowledge is associated with 13 percentage points more adults saving with, and about 7 percentage points more adults borrowing from, a financial institution.

Figure 9. Financial knowledge and share of adults who save with and borrow from a financial institution. Source: European Commission survey (2023) and World Bank Global Findex Database (2021). The Findex database does not contain data for the saving and borrowing questions for Luxembourg in 2021.

The relationship between savings and financial knowledge is well established in the literature, and our numbers confirm this finding. The relationship between borrowing behavior and financial knowledge is less well established. Our results are not definitive, but they show that countries with more financially knowledgeable populations also have more people with access to borrowing. Both results suggest a strong link between financial knowledge and financial inclusion.

Limitations of the aggregate data

The aggregate data we have used so far can address several questions and help better understand both the levels of financial literacy across countries as well as the relationship between financial knowledge and macroeconomic variables or financial outcomes. However, the aggregate data also present some limitations that constrain the analysis that can be carried out. For example, there can be vast differences in financial knowledge in a member state across socioeconomic groups, as we have documented previously. Similarly, we cannot control for individual characteristics, such as income or age, when looking at specific relationships at the macro level, so it is possible that some relationships present at the aggregate-level change when moving to the micro level. In section 4, we analyze microdata from an EU country and report possible differences between analyses at the macro versus micro level.

Microdata case study: Belgium

In this section, we use the Belgian microdata to show some of the problems of using aggregate data. The micro dataset for Belgium is composed of 1014 observations. We first show the heterogeneity in Belgians’ financial knowledge across different socioeconomic and demographic groups. We also examine the relationships we explored in section 3.1, but this time accounting for individual characteristics.

Accounting for heterogeneity

When exploring financial literacy across countries, the aggregate numbers can sometimes hide large differences among demographic groups. Figure 10 shows the differences in financial knowledge across income and education, indicating the importance of accounting for these characteristics when studying financial literacy.

Figure 10. Financial knowledge score by education and income levels in Belgium.

We study the relationship between financial knowledge and the financial outcomes previously studied, this time controlling for age and income (Table 1). As shown in section 3.1, across EU countries, financial knowledge is associated with better financial outcomes in terms of i) being confident for retirement savings and ii) showing lower financial fragility, see Figure 8.

Table 1. Financial outcomes and financial knowledge. Logistic regression estimates

Note. All continuous predictors are mean-centered and scaled by 1 standard deviation. The outcome variable is in its original units. *** p < 0.001; ** p < 0.01; * p < 0.05.

In the micro data for Belgium as well, higher financial knowledge is positively associated with higher confidence levels for having enough savings for retirement, even though the estimate is not statistically significant. Similarly, being financially knowledgeable is associated with a lower probability of being financially fragile (i.e., having enough savings to face an unexpected loss of income).

Overall, our results show the importance of looking both at the aggregate and the micro level so individual characteristics can be accounted for when analyzing financial literacy.

Conclusion

In this paper, we provide a detailed description of the survey “Monitoring the level of financial literacy in the EU, Flash Eurobarometer 525” by the European Commission. We assess the level of financial literacy across EU countries at both the aggregate level and among demographic and socioeconomic groups. Additionally, we explore the relation between financial literacy and different financial outcomes, such as confidence in retirement, financial fragility, and financial inclusion. Finally, we carry out a case study using Belgian microdata as results may differ depending on whether the analysis is performed “within” or “between” countries.

Despite the richness of the dataset, some aspects may be improved upon. The sample size should be higher, so that analyses can be done at more granular levels of disaggregation. Future surveys should keep and enhance the comparability of the questions, and the data should be collected at regular frequencies, so that a pan-European set of financial literacy data is created in the coming decade.