“Do you save for your retirement?“ An empirical snapshot of the level of financial literacy and retirement planning
Abstract —The severe consequences of the current financial crisis have reaffirmed the great lag in financial knowledge. Apart from the change of economic conditions, there is also the demographic barrier, which poses a risk to every individual. Therefore, individuals need to understand financial information in order to be able to react adequately. However, research suggests that financial literacy is low on a global scale, showing concerning gaps between certain groups that may be attributable to financial insecurity and old-age poverty. In light of these findings, an empirical snapshot of the level of financial literacy and retirement planning is given, observing 165 students and non-students, of German and non-German nationality, aged between 20 and 30. The findings suggest that the research findings still apply to the current status. With regard to the financial literacy test, we find that overall only less than half of all respondents were able to answer all three questions correctly. With regard to the groups, women score lower than men; students do better than non-students. In addition, we found that age and nationality are not of importance, while the professional background - at least in our sample - can lead to a better test score. As a consequence, a female non-student working in/studying a non- business/economics-related field/discipline is most endangered. With regard to the savings behavior, we see differences among gender and academic backgrounds implying that male students save on average more often, putting female non-students at risk. We did not find a statistically significant link between the financial literacy test score and savings behavior in our sample.
Keywords — financial literacy, retirement planning, Germany, gender, education
Several countries changed their pension schemes from traditional defined benefit pensions to individual-account contribution schemes in recent years. Germany is one of these countries, having transformed the generous monolithic pension system into a multi-pillar system, covering all private and public employees in recent reforms. For each individual, one side affect of this pay- as-you-go system is the responsibility to provide privately for their retirement, while, with regard to the demographic change, the first-pillar pensions are endangered.
Apart from the change in pension schemes and demography, financial instruments have become more and more complex due to the rapid development in the financial markets. In light of the severe consequences of the current financial crisis, though the majority of households did not suffer from financial losses in Germany, a public discussion about financial knowledge arose (Lusardi and Bucher-Koenen, 2011:2).
As research suggests, overall and regardless of the country’s economic development and pension scheme, the level of knowledge on financial matters, referred to as financial literacy, is suggested to be very low across the world (Lusardi and Mitchell, 2011:13; Xu and Zia, 2012:7-8). Financial literacy is defined as “a combination of awareness, knowledge, skill, attitude and behavior necessary to make sound financial decisions and ultimately achieve individual financial wellbeing.” (Hung et al., 2012:8).
Empirical evidence shows that is has a positive impact on financial behavior and financial status, such as wealth accumulation (Stango and Zinman, 2009), stock market participation (van Rooij et al., 2011; Yoong, 2010), and retirement planning (Lusardi and Mitchell, 2008). Research suggests that people being financially illiterate may easier engage in high-cost credit card borrowings and are more likely to be in high-cost debt (see for example Lusardi and de Bassa Scheresberg, 2013:1; Lusardi and Tufano, 2009).
Hence, due to demographic and economic changes, individuals have been increasingly confronted with decision making over savings, investments and consumption on their own, bearing the risk of wrong investment choices and bad liquidity management, leading to an insecure financial future (Lusardi and Mitchell, 2011:1).
This paper is motivated by previous research showing that certain groups are exposed to higher risk due to a low degree of financial knowledge, whilst being aware of the positive impact on retirement planning. Therefore, the objective of this study is threefold. First, the level of financial literacy should be assessed among people aged between 20 and 30, identifying groups within the sample that have less financial knowledge. Second, the answers chosen are cross-referenced to the groups, investigating if there is a tendency to withdraw from decision-making or to show admittance. Third, the commitment to retirement savings should be determined and differences among groups should be investigated. In addition, all outcomes should be related to the current state of research in order to confirm or to add value to it.
A growing body of literature suggests that there are concerning differences in the level of financial literacy among certain groups. Regarding the financial literacy according to age, an inverted u-shaped pattern is identified, being lowest for the young and older age groups, but with peaks in the middle of the life cycle. This can be due to knowledge increasing with experience and decaying at older ages (Lusardi and Mitchell, 2011:10).
With regard to gender, women score lower than men in financial literacy tests. In Germany only 47.5% of female respondents correctly answered all questions compared to 60% of male respondents (Lusardi and Bucher-Koenen, 2011:8). The gender effect is concerning because women face higher financial challenges than men: Living on average five years longer than men (Møller, Fincher et al., 2009), they need to accumulate more savings for their retirement and will be confronted with higher healthcare expenses than men over the course of their lives, whilst having lower social security payments due to less time in the workforce. In addition, it is expected that someday, due to their husbands’ death or divorce, the majority of women will be solely responsible for their financial needs (Davidson et al., 2012:1).
Financial literacy is deemed low on a global scale, even when financial markets are well developed as in Germany, the Netherlands, Sweden, Italy, Japan, and New Zealand. Differences across countries can be based on financial (in-) stability of countries (e.g. Italians are more knowledgeable about inflation, since their country recently experienced it) and on the knowledge in mathematics and science (e.g. respondents in Sweden and the Netherlands scored well on math tests as well as on questions regarding numeracy in financial literacy tests) (Lusardi and Mitchell, 2011:9).
Education is crucial with regard to the level of financial literacy in Germany. Respondents without an occupational degree were less likely to answer the financial literacy questions correctly, but there is not a pivotal difference between participants with a university degree and vocational training. Thus, the relationship between years of schooling and financial literacy is not linear (Lusardi and Bucher-Koenen, 2011:9).
Research suggests that another difference can be seen with regard to the labor market status. Students, unemployed and homemakers, as well as retired individuals score lower than participants who are employed or self-employed. Among the latter the level of financial literacy is higher, as the probability of answering the three questions correctly, is at more than 60% (45% in the out of labor group) (Lusardi and Bucher-Koenen, 2011:10).
As knowledge seems to differ among these groups, we focus on five variables with a slight deviation: 1) age,
2) gender, 3) nationality (as former research suggested),
4) academic background (stating if someone is a student, or former student or never pursued an academic career) and 5) professional background (showing a difference between participants studying or working within the field of business/economics/related and participants studying and working in another field).
III. METHODOLOGY Questions included in survey
Many research teams designed research questions to measure financial literacy, but only a few show results on an international level. For example, Annamaria Lusardi and her team provided simple and brief questions, which are characterized as benchmarks in measuring financial literacy on a global scale. With three questions covering basic methods of savings and portfolio choice included first in the 2004 Health and Retirement Study in the US and in several others later on, financial literacy has been measured on a national level first. In collaboration with several other teams, Lusardi inferred conclusions for the international level by assessing well-adapted questions from international surveys covering the topic of financial capability.
These questions follow four key principles (Lusardi and Mitchell, 2011:2).
- Simplicity. Questions are designed to measure basic financial concepts only.
- Relevance. Questions need to relate to an individual’s day-to-day financial decisions, but still be general.
- Brevity. Number of questions needs to be kept to a minimum to ensure a widespread adoption in surveys.
- Capacity to differentiate. Questions are specially designed to differentiate between knowledge levels.
These questions included in several surveys are as follows (Lusardi and Mitchell, 2011:3):
1) Suppose you had $100 in a savings account and the interest rate was 2% per year. After 5 years, how much do you think you would have in the account if you left the money to grow?
2) Imagine that the interest rate on your savings account was 1 percent per year and inflation was 2 percent per year. After one year, would you be able to buy more than, exactly the same as, or less than today with the money in this account?
3) Do you think the following statement is true or false? Buying a single company stock usually provides a safer return than a stock mutual fund.
The first question requires an elementary calculation related to compounding of interest rates. The second measures understanding of inflation, whereas the third question tests the knowledge of risk diversification. In addition, the list of possible answers includes the option of stating that the answer is unknown or that the answer is refused, to make it easier for people to select a preferred response (Lusardi and Mitchel, 2011:4). All measures are based on financial knowledge only.
A fourth question was included, in order to assess the current thoughts and behavior towards retirement planning. Thereby participants were asked:
4) Do you save for your retirement?
Being a dichotomous question, participants were able to explain their decision in a text box, if they wished to (see Figures 1-3).
Participants for the experiments were solicited via facebook, e-mail and Whatsapp being friends, family and friends of friends mainly. Additionally the survey was posted in two university groups not being limited to any academic subject. No reward was given for their participation.
The sample includes 165 participants aged 20 to 30 only. The average age is 24.8 ± 2.3 years (Table 1 and 2). Distribution within groups:
- Gender: 44% male, 56% female
- Nationality: 81% German, 19% other
- Professional background: 52% business/ economics/related, 48% other
- Academic background: 93% student/former student, 7% non-student (see Tables 3-6).
Despite the cost and effort, the results may suffer from measurement errors, because people may simply guess the answers at random, make use of the internet and search for answers, since the questions were provided without a timer, or misunderstand questions, since the questions were phrased in English, which is a foreign language to the majority of participants.
IV. KEY FINDINGS Overall responses
Answers of the first question are displayed in Table 7. 92.1% answered correctly, 4.8% were incorrect, 2.4% admitted to not knowing the correct answer and 0.6% refused to answer. Since the calculation of compound interest is part of the German school curricula (Reifner, 2003) and students should have a basic numeric understanding even at a low educational level, this outcome was anticipated.
The second question’s answers are shown in Table 8. 83.6% answered correctly, 9.7% incorrectly and 6.7%
admitted to not knowing the correct answer. Though the figure is higher in the first question, we hypothesize that a huge majority knew the correct answer due to current discussion on low-interest rates in savings and current accounts and the negative real rate of interest.
Table 9 shows the answers to question 3 on risk diversification. This question appears to have been difficult for many individuals, because only 55.8% managed to answer it correctly. 7.9% chose the incorrect answer, whereas 33% admitted to not knowing the correct answer and 2.4% withdrew from this decision. Although knowledge on risk diversification is not part of the German Curriculum, people working or studying a business/economics/related subject should have come across this topic (Lusardi and Bucher-Koenen, 2011:6). In addition, we assume that, since a majority of participants are students, who mainly earn money to cover their living expenses, won’t come across buying or selling stocks.
The amount of correct answers is summarized in Table 10. Less than half of the participants were able to answer all questions regarding financial literacy correctly. This is moderate and in accord with the current research status. In Table 11 and 12 we see how many times participants didn’t know the correct answers and how many times they chose to refuse to answer.
Regarding their commitment to saving for their retirement, 35% answered in the affirmative, whereas the majority negated the question (Table 13). Out of all responses on giving reasons for the decision, 48 were valuable. Participants explained their saving for retirement, because
- the pension they will receive, will probably not be enough for living
- otherwise one would be dependent on whatever subsidies are granted by the government when retiring
- now is the the time to start saving, since in the long run people will pay interest on investments and the guaranteed interest decreases
- due to the risks that come with the demographic change
- when they retire, they just want to do to the things they like, which requires money.
Still two people elaborated on their savings behavior:
- “I save for my retirement investing in fine gold bars, because I believe more in that than in money.”
- “I save the money my grandparents gave me when I was born, but I don't know yet if it will be for retirement or something else.”
(The language in the summary is modified, since English is not the first language of the majority of participants. It does not incorporate the answers literally but is true in its meaning.)
Participants’ reasons not to save for retirement were mainly based on current financial instability (48%). Many however, mentioned in the same sentence that they plan to save once they graduate and start working (in total 55%). Still there were three participants stating individually that it is “too early for this type of investment”, that they are too lazy to save or don’t think about this topic (Figure 4).
Typical tools and techniques used in testing groups across answers are:
1. Bivariate Correlation tests (Pearson and/or
2. T-test for Equality of Means in two independent
samples with regard to Levene's Test for Equality of Variances
3. Chi-Test (Pearson Chi-Square)
Statistically proven findings within groups
A summary on all tests among groups and type of responses is provided in Table 14 showing a diverse picture of relationships.
Age. Testing age across the four types of responses shows only correlation between age “refuse to answer” and “don’t know” (significant at the 0.05 level (2-tailed)) implying that the older the participant, the more times he/she admits not to know the right answer, but the less often he/she withdraws from the decision. No assumptions can be made on the basis of this finding. In addition there is no evidence of the inverted-u-shaped pattern, which can be explained by the small sample and the narrow age interval (see Test panels 15-18).
Gender. The variables gender and test score show a strong negative correlation (significant at the 0.05 level (2-tailed)). The t-test revealed that on average, men scored better (0.28 points) than women (p<0.05, for a 95% and 90% Confidence Interval) and that females more often admitted to not knowing the correct answer (correlation is significant at the 0.01 level (2-tailed), p- value<0.05 in t-test). Both groups do not differ statistically significantly regarding the option “refuse to answer” (see Test panels 19-21). In test panel 22 a significant correlation between gender/saving-response (Correlation is significant at the 0.01 level (2-tailed)) and with the help of the Chi-test (p< 0.05). There are differences between the preferences of women versus men in deciding for saving-responses. Men choose to save more often than women (t-test, p<0.05, for a 90% Confidence Interval). The results confirm current research findings.
Nationality. No statistically significant results were found testing differences according to nationality across all possible responses (Test panel 23-26) implying that whether people are German or of a different nationality is irrelevant for the outcome. This result is contrasted by findings of Lusardi. However this may be due to the rather anonymous part of the non-German participants.
Academic background. A statistically significant correlation between the group academic background and test score is found (correlation is significant at the 0.01 level (2-tailed)). In addition, the t-test shows a statistical significance implying that students have a higher score than non-students on average (p<0.05) (Test panel 27). This is contrary to current research findings embodied in this text, since students were tested to score less than non- students. Testing academic background across “don’t know” yields no results (Test panel 28). However, since the research clustered students and unemployed into one group and we did not ask specifically for the labor status, this difference might be reasonable. Testing academic background against “refuse to answer”, a strong negative correlation is found, but the independent samples t-test does not find a significant difference among groups. we can conclude that there is a strong relationship between our academic background and “refuse to answer”, any other conclusions about this relationship can’t be made (correlation is significant at the 0.01 level (2-tailed)). There is a statistical significant relation between savings the academic background and savings behavior implying that non-students on average responded to save more times than students did.
Professional background. A statistically significant correlation is found testing professional background across test score (correlation is significant at the 0.01 level (2-tailed)). On average, participants with a business- related background scored 0.54 point better than participants of other disciplines (t-test, p-value<0.05) (Test panel 31). In addition, we find statistically significant differences testing professional background across “don’t know” implying that participants of other disciplines more often admitted to not knowing the correct answer, but regarding “refuse to answer” we hypothesize that this behavior is random in the set (Test panel 32 and 33). Regarding the savings behavior, we don’t find a statistically significant difference either (Test panel 34). This again shows that participants studying or working in a non-business-related field/discipline face a greater risk of taking inadequate financial decisions, which ultimately leads to financial insecurity, being financial less literate.
Last but not least, we tested the test score across savings behavior, finding a statistically insignificant result. This means that if participants score well on the test and show a high level of financial literacy, this does not influence their current savings behavior. Many students justified their lack of savings with the current state of financial insecurity, but at the same time voiced their plans to save for their retirement once they have a stable income.
V. RESULTS AND CONCLUSION
Overall, the level of financial literacy is moderate among the participants aged between 20 and 30. Nearly half of the participants were able to answer all questions regarding financial literacy correctly. 38% out of all respondents managed to answer all questions correctly and only 13% were correct once. Still 2% were not able to answer the financial literacy questions correctly. Paying special attention to the three questions, the majority of participants managed to answer question 1 regarding compounding interest and question 2 regarding inflation correctly, whereas question 3 on stock risks appeared to have been difficult for many individuals, because only 55.8% managed to answer it correctly; 7.9% chose the incorrect answer, whereas 33% admitted to not knowing the correct answer and 2,4% withdrew from this decision.
We also find that the gender gap exists in our sample, since women scored lower than men in the financial literacy test and were less engaged in retirement planning than men; that education matters, since students with an aspiration after business/economics/related did better in the financial literacy exam than non-students in different disciplines. Regarding savings overall, only one third out of all participants affirmed that they save for their retirement, stating correct reasons for doing so. The majority is not saving, but shows a tendency to do so, once a stable income is available. With regard to specific responses in the financial literacy exam, women and respondents studying/working within a non business- related field more often admitted to not knowing the correct answer, while non students more often withdrew from decision making than others, and finally regarding the responses on retirement savings next to men, non- students responded more often to save than students.
As a consequence, we think that programs should increasingly target women, people with a non-academic background and people working or studying in a non- business-related discipline/field. If they are offered to the groups mentioned being at risk, they will improve their financial knowledge and take appropriate steps to secure their financial future. One possible method to start privately save for retirement is e.g. the “Riester” pension scheme. Singles and students only need to pay the minimum amount of 5 € per month (a total of 60 € a year), if they have contributed at least once into the pension insurance in the current year, and will receive state-subsidies of currently 154 € p.a.. A one-time bonus of 200 € is guaranteed if Riester pension plans are signed up before the 25th birthday. With this students can increase the scope of their insurance periods in the statutory pension early.
However, each and every individual needs to analyze on his/her own, if this pension scheme fits their needs, compare the offers given and be aware of benefits and risks that this pension scheme may carry with it (for further information see Deutsche-Rentenversicherung.de or an insurance company of your choice). Furthermore, further research needs to be conducted in findings that were described in brief, since this is just an empirical snapshot and limited in its size, and in the efficacy of current incentives that approach those groups we see at risk.
The research reported here was performed in the context of a university seminar called Financial Literacy and Retirement Financial Planning at Dortmund University of Applied Sciences and Arts. The opinions and conclusions are solely those of the authors and should not be construed as representing the opinions or policy of any other institutions nor the University of Applied Sciences and Arts.