1
HOUSEHOLD FINANCIAL PLANNING:
THE ROLE OF FINANCIAL LITERACY AND FINANCIAL ADVICE ON RETIREMENT SAVINGS
Bart Schooten s2243733
Faculty of Economics & Business University of Groningen
Master Thesis MSc. Finance Supervisor: Dr. Robert Lensink
June 2017
2
Table of content
ABSTRACT ... 3
1. INTRODUCTION ... 4
2. LITERATURE FRAMEWORK ... 6
2.1 FINANCIAL LITERACY... 6
2.2 FINANCIAL ADVICE AND RETIREMENT PLANNING ... 8
2.4 RESEARCH FRAMEWORK ... 12
3. DATA AND METHODOLOGY ... 12
3.1 DNB HOUSEHOLD SURVEY (DHS) ... 12
3.2 VARIABLE OPERATIONALIZATION ... 14
3.2.1 FINANCIAL LITERACY ... 14
3.2.2 FINANCIAL ADVICE ... 15
3.2.3 PENSION KNOWLEDGE ... 15
3.2.4 LOCUS OF CONTROL & PARENTAL STIMULATION ... 16
3.2.5 CONTROL VARIABLES ... 17
3.3 METHODOLOGY ... 17
4. RESULTS ... 19
4.1 FINANCIAL LITERACY AND RETIREMENT PLANNING ... 19
4.2 FINANCIAL ADVICE AND RETIREMENT PLANNING ... 22
4.3 WHO IS FINANCIAL ILLITERATE ... 25
5. DISCUSSION & CONCLUSION ... 27
5.1 DISCUSSION ... 27
5.2 CONCLUSION ... 29
5.3 LIMITATIONS AND FURTHER RESEARCH ... 30
REFERENCES ... 31
APPENDICES ... 33
I. ROBUSTNESS REGRESSION MODELS ... 33
II. FINANCIAL LITERACY MODULE ... 34
III. FINANCIAL LITERACY SURVEY ... 37
IIIA: BASIC FINANCIAL LITERACY ... 37
IIIB: ADVANCED FINANCIAL LITERACY ... 38
IIIC: PENSION RELATED QUESTIONS ... 39
3
HOUSEHOLD FINANCIAL PLANNING:
THE ROLE OF FINANCIAL LITERACY AND FINANCIAL ADVICE ON RETIREMENT SAVINGS
Bart Schooten
1University of Groningen
ABSTRACT
Retirement planning is one of the most challenging financial decisions households face. Recent studies show large differences in financial- and economic knowledge in households, which illustrates that most households lack financial skills in order to properly manage the increasing retirement planning decisions. In this paper, the relationship between financial literacy and - advice and retirement planning is examined using DHS data from the Dutch Central Bank (DNB) of Dutch households. This study finds a strong, significant relationship between financial literacy and retirement planning. Using information on economic education during school years, there is evidence that economic education stimulates financial literacy, which in turn positively
influences retirement planning. Furthermore, this study researches whether receiving an allowance and stimulation to plan expenditures for parents influences retirement planning. It is found that allowances have no relation with financial literacy or retirement planning, while stimulation to plan does have a significant effect on retirement savings. Secondly, professional financial advice positively influences retirement planning. Financial advice is used as a
complement rather than a substitute for financial literacy. The findings imply that policymakers should try to integrate economic education on all education levels in order to boost financial literacy in the Netherlands. Furthermore, the opportunities of financial stimulation of parents to young kids can be optimized to kick start financial and economic education.
Keywords: Retirement planning, financial knowledge, financial advice JEL classification: D14, D80, D91, J32
1 I would like to thank Mr. Lensink for his feedback and advice during the writing of this thesis
4
1. INTRODUCTION
Retirement planning is a challenging decision for households. Reformed retirement plans place important retirement decisions in the hands of individuals and households, of which a large portion does not possess sufficient financial knowledge (Lusardi &
Mitchell, 2005) which has become an meaningful topic in the Netherlands (van Raaij, Antonides, & de Groot, 2008). Whether Dutch households have the financial knowledge and skillsets to plan and save for their retirement has become a cause for concern.
Bernheim (1995) was one of the first researchers to advocate that financial knowledge in household financial decision making needs to get more attention from policymakers.
Since then, financial literacy has become the subject of several studies in behavioral economics. This study will examine the relationship between financial literacy, financial advice and retirement planning.
Most research on financial literacy and retirement planning has been done in the United States using the Health and Retirement survey. Lusardi and Mitchell (2006) used this survey and show a positive relationship between financial literacy and retirement planning. This is an important finding, as planners are more likely to build up wealth than non-planners (Ameriks, Chaplin, & Leahy, 2003). In accordance with this, Van Rooij, Lusardi & Alessie (2009) show that there is a positive relationship between financial literacy and retirement planning. Financial literate individuals are more likely to think forward in the future and consider retirement in their spending- and saving plans.
Monticone and Calcagno (2015) argue that a low level of financial literacy doesn’t directly imply household will make poor financial decisions. Households have the opportunity to seek professional financial advice to use as a substitute of financial literacy to assist households in constructing a retirement plan (Chalmers & Reuter, 2013). Outside financial advice potentially enables households with limited financial skills to plan effectively for retirement (Chatterjee & Zahirovic-Herbert, 2010). Yet, financial advice is not utilized by people who need it, but by financial literate individuals.
Joo and Grable (2001) show that financial advice regarding retirement planning is mostly used by individuals who are above-average financial literate.
This study adds to the literature by considering new factors in the relation of financial literacy- and advice to retirement planning. This study looks at other factors that could
5 influence financial literacy besides (economic) education. Receiving allowance, internal locus of control and stimulation to save money are important factors which could mediate the relationship between financial literacy and retirement planning.2 These additional factors have not been examined before and provide new insights in the determinants of financial literacy and retirement planning.
Using data from the DNB Household Survey (DHS), the results show that retirement planning is positively related to financial literacy. Seeking financial advice for retirement planning is mostly done by individuals who are financial literate, which indicates that financial advice is a complement rather than a substitute for a lack of financial
knowledge. Additional findings are that retirement planners have lower risk aversion and have been stimulated to save money by their parents. Receiving an allowance has no effect on retirement planning or financial literacy. Firstly, there is a direct relationship between financial literacy and retirement planning. Secondly, there is a positive relationship between financial advice and retirement planning.
For policymakers, the results of this study are relevant as well. As the Dutch pension system relies more on households to plan and save for their retirement, financial knowledge and skills are necessary to do so (van Rooij, Lusardi & Alessie, 2009). If policymakers want households to be successful in planning for their retirement, they should understand what drives financial literacy and successful planning.
The main implication of the findings in this study is that policy makers should create awareness among Dutch households on the urgency of retirement planning and attempt to boost financial literacy levels. This can be done through basic economic classes in high school on all levels of education. Other options lie within childhood stimulation towards money and its value. Children who are stimulated to save money by their parents are more likely to become successful financial planners and likewise plan for retirement.
The outline of this paper is structured as follows: the next chapter contains a literature review on financial literacy, financial advice and retirement planning from which the hypotheses will be derived. The third chapter discusses the data and research
methodologies. Chapter four provides the results from the data analysis. The final chapter provides a discussion and the conclusion.
2 I would like to thank Mr. Alessie for the discussion of his working papers and the exchange of insights.
6
2. LITERATURE FRAMEWORK
This chapter contains the literature review which focuses on financial literacy, financial advice and their relationship with retirement planning. The hypotheses derived from this framework will be tested and its results are discussed in chapter 4.
2.1 FINANCIAL LITERACY
The life cycle consumption model in its simplest form states that households accumulate wealth during their working life and decumulate wealth in their retirement period. This type of spending behavior allows households to smooth their marginal utility of
consumption over the life cycle (van Rooij, Lusardi & Alessie, 2012). However, there are reasons why household consumption doesn’t follow the pattern of the life cycle model.
Shefrin and Thaler (1988) build on the behavioral life-cycle hypothesis of Modigliani and argue that self-control and mental accounting are important components of this cycle. In this cycle, wealth is assumed to be divided in 3 parts of mental accounting, namely current income, current assets and future income (pension accounts). Shefrin and Thaler argue that the temptation to spend is least for future income and that households create wealth to spend after their retirement. According to this hypothesis, the average household saves money to finance their retirement.
Gustman et al. (2010) explore data from the US Health and Retirement Study and finds no relationship between numeracy (basic cognitive skills) and knowledge of retirement plans. Gustman does find a positive relationship between pension plans and knowledge, however argue that the causality is likely to run from wealth to pension knowledge. This skill-wealth outcome suggests that the increase in cognitive skills and numeracy does not necessarily increase household wealth as they plan for retirement.
Van Rooij et al. (2009) find evidence that financial knowledge and sophistication is significantly correlated with financial household behavior. Using a customized survey, this study finds out that there are large differences in financial knowledge among Dutch households, in which most household lack knowledge of fundamental financial concepts.
Van Rooij et al. find a strong correlation between financial literacy and retirement planning in the Netherlands. One particular shortcoming of households seems to be the short-sighted budgeting periods in terms of financial decision making. Poor preparation for retirement is a direct consequence of this, which signals that retirement funds built
7 up is relatively low compared to the needs. Indeed, Lusardi (2003) reports that
American households have very little savings in the verge of retirement. Van Rooij et al.
(2009) argue that causality runs from financial literacy to planning and wealth.
Monticone (2010) on the other hand, questioned the causality of financial literacy on wealth accumulation and argues that wealth contributed to financial literacy. As one has more wealth, a person is also more inclined to gain financial literacy. While this effect is significant, it is relatively small.
Van Rooij et al. (2011) studied financial knowledge in 2005 and 2010. As they did not find changes in financial literacy levels in the period before and after the financial crises, Van Rooij et al. show that significantly more individuals have thought about retirement, which is in accordance with the policy changes regarding retirement. They do find a positive effect of financial literacy on retirement planning.
Lusardi and Mitchell (2008) evaluated financial knowledge during workers’ primary earning years when they had to make key financial decisions and allocations. Financial literacy proves to be a key determinant in retirement planning and is a suitable
predictor of retirement planning. They also found that financial literacy is higher when they were exposed to economics in school and to financial education programs offered by their companies and employers. This illustrates that financial literacy is influenced by the knowledge gained during school years, which can be beneficial in later periods in life. In accordance with this, Lusardi and Tufano (2008) find that financial illiterate individuals are more likely to display poor financial behavior.
Another paper by Van Rooij et al. (2012) shows that objective measures of financial knowledge have a positive association with net worth of households. Financial
knowledge increases the probability that a household is active on the stock market and that households are capable of making complex financial decisions. Associated with this, financial literacy is positively related to retirement planning. Financial literacy increases the likelihood a household has a sophisticated savings plan and is aware of the future financial needs. Lusardi & Mitchell (2007) confirm these results, as they find that individuals with higher financial literacy arrive at retirement age with higher wealth levels than non-planners.
8 Previous research by Lusardi and Mitchell (2005) also suggests that retirement planning is positively related to financial literacy and pension knowledge. However, when it comes to planning, only one-third of households attempt to save for retirement, whereas only two-thirds of these attempts succeed. A relatively large fraction of households do not know how much to save for retirement and what the specific characteristics of their contribution plan is. These findings indicate that the average household is not capable of making its own sound investment decisions. The study of Bucher-Koenen & Lusardi (2011) confirms that low-income and low educated individuals lack financial
knowledge. Only half of the participating households could answer 3 basic financial questions correctly. They find a positive impact of financial literacy on retirement planning.
Van Rooij, van Els & Schuit (2007) show that Dutch workers are unaware of the characteristics of their retirement plans. Without compulsory contribution, savings would be much lower and limited knowledge prevents households from accumulating wealth. Despite the compulsory participation in a collective pension plan, many Dutch households lack pension knowledge and interest in pensions (van Rooij, Kool & Prast, 2007).In a related research, Lusardi & Mitchell (2011) find that a large group of households unjustifiable see themselves as financial literate and capable of making complicated decisions. This leads to low saving accounts and financial errors in household finances.
Based on this discussion, the first hypothesis is:
H1: Financial literacy has a positive effect on retirement planning.
2.2 FINANCIAL ADVICE AND RETIREMENT PLANNING
Financial advisors and experts have the ability to aid households in their financial decision making and are potentially a valuable asset in terms of financial planning and wealth creation. One likely determinant of household investment behavior in general seems to be professional financial advice. If households are confronted with substantial information costs and investment errors are quite common, professional advice can enhance household utility (Bluethgen et al., 2008). While most households are able to answer basic financial question, with more complex decisions as retirement planning come mistakes and errors. This gap between present knowledge and skills needed potentially displays the need for outside professional financial advice. Overall, the
9 impact of financial advice of asset allocation and portfolio returns seems to be negative (Bergstresser, Chalmers & Tufano, 2009; Kramer, 2012). This could have an impact on seeking financial advice, as poor results lead to lower rates of advice seeking.
Joo and Grable (2001) show that individuals who have higher income, better financial behavior, a positive and proactive attitude towards retirement had a higher level of risk tolerance are more likely to seek outside financial advice from professional when making retirement plans and decisions. These results suggest that financial advice has a positive effect on retirement planning and asset allocation towards retirement. This also shows that financial advice is mostly sought by wealthy households with better
education. Financial advice is therefore argued to be complement to financial literacy rather than a substitute for it. Households who need the financial advice regarding retirement planning the most, do not seem to seek financial advice.
Another paper seeking to describe financial advice and retirement portfolios is the paper by Chatterjee & Zahirovic-Herbert (2010). They find that higher education levels and higher net worth are positively related to seeking outside financial advice.
Households who need advice regarding pension planning the most, are likely not able to afford financial advise and therefore cannot access it. As financial advice is beneficial for the planning and accounting for retirement on a household level, policies and the
professional advice services should put more effort into the accessability of financial planning services.
Chalmers & Reuter’s (2013) findings do not support the previous results. They found that if given the opportunity to access face-to-face financial advice through a network of brokers and advisers to discuss retirement plans, that younger, less educated and less paid employees are more likely to choose financial advise regarding retirement
planning. These individuals were also likely to follow the advice and recommendations of the brokers.
Bodie (2003) argues that investment advice provided to individuals in self-directed retirement plans by financial service firms are flawed and often misleading. According to Bodie, there is a strong bias in favor of investing in retirement savings in the stock market without providing insurance against a market- and stock decline. While there is limited financial literacy among the general public, financial advisors need to design
10 safer design in order to offer financial advice to less knowledgeable individuals. Overall, Bodie argues that financial advice can be a substitute for the lack of financial knowledge.
Financial advice is overall beneficial for retirement planning.
Hung and Yoong (2010) provide evidence that households that actively seek
professional financial advice when it comes to retirement planning plan more effective for retirement and their planning performance is significantly better. Despite negative selection selection on financial ability and financial knowledge, improve their retirement planning and are more effective at planning. This effect is particularly positive for less financial literate individuals. While access to advice may improve retirement planning, extensive compulsory programs of financial counseling is less effective. However, Monticone (2015) and Kramer (2016) show that financial advice is rarely consulted by individuals who are less financial literate and less highly educated. While this group could use financial advise the most, they do not consult it.
Therefore, hypothesis two is formulated as followed:
H2: Financial advice has a positive effect on retirement planning.
11
TABLE 1 Literature overview
This table provides an overview of the most relevant literature on the relationship between financial literacy, financial advice and retirement planning that are discussed in chapter 2.
Authors (year) Independent
variables Sample Main findings
Van Rooij et al. (2009) Financial literacy Dutch household
survey Most households lack knowledge of fundamental financial concepts. Financial literacy has a strong and positive association with retirement planning. Financial sophistication is highly correlated with financial household behavior and is associated with thinking about and saving for retirement. Overall, Dutch households don’t plan much for retirement.
Van Rooij et al. (2011) Financial literacy Dutch household
survey Studying the difference in financial knowledge in 2005 and 2010, financial literacy did not change before and after the financial crisis in 2007. In 2010, significantly more individuals thought about retirement than in 2005. Overall, financial literacy- and knowledge has a positive causal effect on retirement preparation.
Van Rooij et al. (2012) Financial literacy Dutch household
survey Financial literacy is strongly positively related to the accumulation of wealth and to retirement planning. Households who possess a high level of financial literacy are more likely to gather information and to set up a retirement plan.
Ameriks, Chaplin &
Leahy (2003) Financial literacy American households The main findings in this research is that households which possess a higher level of financial literacy are more likely to budget household spending and are more likely to save and accumulate wealth. These households are also more likely to save and plan for retirement.
Lusardi & Mitchell
(2005) Financial literacy Health and Retirement
Study Financial illiteracy is widespread among older American individuals. Households rarely save for retirement, as only one-third ever tried a savings plan and only two-third of these households succeeded in their plan. Lusardi also found that financial literacy and retirement savings are interrelated: those who are financial literacy, are more likely to plan and succeeded in retirement planning.
Lusardi & Mitchell
(2011) Financial literacy National Financial
Capability Study Financial literacy is particularly low among the young, woman and the less-educated. All groups rate themselves as well- informed about financial matters, while this is mostly unjustifiable. People who score high on financial literacy are much more likely to (successfully) plan for retirement.
Bucher-Koenen &
Lusardi (2011) Financial literacy German SAVE survey Knowledge of basic financial literacy lacks among less educated and low income population. There is no gender disparity in financial knowledge. Found is that there is a positive impact of financial knowledge on retirement planning.
Gustman et al. (2005) Financial literacy Health and Retirement
study Gustman et al. find that measures of cognitive ability, among which is numeracy, are not significant determinants of pension planning. They suggest that this numeracy-wealth relation should not be seen as evidence that increasing financial literacy will increase the wealth of households as they enter into retirement.
Bluethgen et al. (2008) Financial advice German retail bank Clients seeking advice are likely to be older, wealthier, more risk averse and female. Financial advice enhances portfolio diversification, makes investor portfolio’s more congruent with model portfolios and increases investor’s expenses of fees.
Overall financial advisory services have a significant impact on household investment behavior.
Chalmers & Reuter
(2013) Financial advice Oregon University
retirement plan The main finding is that younger, less highly educated, and less highly paid employees are more likely to choose financial advisors. Agency problems seem to affect the performance of the portfolio, as the advisor has a person agenda.
Chatterjee & Zahirovic-
Herbert (2010) Financial advice National Longitudinal
Survey of Youth Participation in tax-advantage accounts and higher net worth increase the probability that a person will seek outside financial advice in order to aid them with their retirement plan.
Joo & Grable (2001) Financial advice US Retirement
Confidence Survey Individuals who have higher income, are financial literate and have better financial behavior, have a positive attitude towards retirement and have a higher financial risk-tolerance are more likely to seek outside financial advice when making retirement investment decisions.
Monticone (2015) Financial literacy-
and advice Unicredit customer
survey Financial literacy increases the probability of consulting a financial advisor. This study suggests that consulting and advisors are rarely used by households who need it the most, which concludes that financial advice is complement to financial literacy.
12 2.4 RESEARCH FRAMEWORK
The hypotheses are graphically presented in figure 1.
Figure 1 Research framework
This research framework is tested through the hypotheses derived above. Firstly, retirement planning is related to financial literacy. Secondly, retirement planning is regressed on financial advice and control variables.
3. DATA AND METHODOLOGY
This chapter discusses the data from the DNB Household Survey (DHS). Next, the variables from the survey and its operationalization will be discussed, followed by the methodology.
3.1 DNB HOUSEHOLD SURVEY (DHS)
The data comes from the Dutch Central Bank (De Nederlandsche Bank, DNB) Household Survey (DHS), which is part of the centERpanel. This database contains 2.028
households, responding to an annual set of questions regarding all possible aspects which are considered to be descriptive on Dutch household level basis. 3
Panel members participate weekly in internet surveys via a computer of set-top box for a television set. Households which do not have access to internet participate through television sets. This is necessary to select a representative panel group of households in the Netherlands, i.e. people who do not have access to internet need to be included in the survey as well. The DNB survey and data is considered to be of high quality and
represents the Dutch population closely. The Household Survey collects information of a
3 The DNB household survey is accessible via www.centerdata.nl
13 wide variety of demographic characteristics and socioeconomic attributions (age, sex, education level, income, wealth, etc.), as well as information on psychological and
savings- and investment behavior. These extra variables give het opportunity to be used as control variables. The questions are answered by household members responsible for household finances (DHS codebook 2005).
The data from the 2005 wave is used. The main reason for using the data from the 2005 wave is that an extra module was added to the survey of that year. This module was designed to measure actual and self-assessed financial literacy and pension literacy. This module has been the subject of a financial literacy working paper of the Dutch Central bank, studied by Van Rooij et al. (2009).4 The questionnaire can be found in appendix III.
The extra financial literacy module was sent to the 2,028 households participating in the household survey. 1,508 household responded to the survey, which yields a response rate of 74,4%. The financial literacy module is then merged with the 2005 DHS wealth- and psychological modules. Before this dataset is used for the hypotheses and tests, exclusions will have to be performed on the dataset. The questionnaire contains the
‘refusal’ option, in which the participant chooses not to answer the question. These observations are removed from the dataset, as they interfere with a genuine attempt to answer the question in order to measure financial literacy. Furthermore, negative gross- income positions are removed from the dataset. This yields a final sample of 1,081 households.
The average age of the sample is 50.6 years and 80% of the observations are within the age category of 20 to 65 years. This is the time period in which individuals are likely and expected to make important financial decisions regarding home ownership, mortgage and also retirement planning. Furthermore, 274 households have retired (25.3 %) and 19.89% of respondents are above 65 years old, the eligible retirement age in the Netherlands.
4 I would like to thank mr. Van Rooij of the Dutch Central Bank for providing the data and extra modules of the financial literacy survey.
14 Table 2
Descriptive statistics
This table provides the descriptive statistics of the main variables and control variables.
The statistics are based on the 2005 DNB Household survey wave. The full sample consists of 1,081 observations.
I. Measured literacy
Mean (S.D.) 2.85 (.95)
Min 1
Median 2.5
Max 4
II. Self-assessed literacy
Mean (S.D.) 2.27 (.57)
Min 1
Median 2
Max 4
III. Pension literacy
Mean (S.D.) 3.25 (.85)
Min 1
Median 3
Max 4
IV. Control variables
Age 50.6 (15.08)
Male 57.6 %
Gross income € 30,543.71
Occupation Employee (51.5 %)
Education Home ownership
Higher education (40.6 %) 32.1 %
Partner 68.1 %
Kids 32.6 %
3.2 VARIABLE OPERATIONALIZATION 3.2.1 FINANCIAL LITERACY
Actual financial literacy is measured using the survey conducted by Van Rooij et al.
(2009). This survey was designed to measure financial literacy and to be able to distinguish between the level of financial knowledge and skills. The survey is set up in different sets. The first set of questions aims to measure the basic level of financial skills, which are considered to be essential for everyday financial decisions. These questions regard topics as interest compounding and risk diversification. The next set of questions target the sophisticated financial knowledge of the participants, regarding the principles of the stock- and bond markets. The basic financial literacy consists of 5 questions and the advanced literacy module consists of 11 questions (See the appendix III for the
15 wording and phrasing of the survey). For each question a dummy is created. One dummy indicating whether the respondent answered the question correctly, and the other dummy to indicate whether the respondent chose the “don’t know” option. When
considering financial literacy, “do not know” answers are important to take into account (Van Rooij et al., 2011).
Following Van Rooij et al. (2011), a factor analysis is performed on the 22 dummies (11
‘correct’ answers of sophisticated literacy, 11 ‘do not know’ answers). Bartlett’s test of sphericity (p<.01) and the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (KMO = 0.902) indicate that a factor analysis is suitable. Please see table 7 of appendix II for the descriptive statistics on the financial literacy questions and the factor loadings.
The average number of correct answer on the 5 basic financial literacy questions is 4.15 and on the 11 sophisticated literacy questions is 6.48. Out of the 1,081 respondents, 489 (46.07%) participants had all 5 basic questions correct, whereas only 31 (2.87%) of the participants were correct on all 11 advanced questions. Please see table 8 of appendix II for the distribution of the answers.
3.2.2 FINANCIAL ADVICE
Financial advice is obtained from the following question: “what is your most important source of advice when you have to make important financial decisions for the household?”
Most people obtain advice from a professional adviser (26.8%), followed by “parents, friends and acquaintances” (22.3 %). A small 10% relies on financial advice from the internet. Please see table 9 in appendix II for the distribution of sources of professional financial advice. To operationalize professional financial advice, a dummy is created and is set equal to one if the respondent has reported to receive financial advice
(“professional financial advice”), and zero otherwise.
3.2.3 PENSION KNOWLEDGE
Pension knowledge can be derived from a subsection in the survey regarding pensions.
This subsection proxies for pension knowledge, as these questions directly target the knowledge on pensions and pension obligations in the Netherlands. Pension knowledge proxies for retirement planning, as more knowledgeable a person is regarding pensions and retirement, the more he is likely to prepare for his retirement (Van Rooij et al., 2009). Out of the 1,081 households, 46.62% of the respondents are “very
16 knowledgeable” in terms of pension knowledge, whereas 38.02% are “knowledgeable”
and only 5.55% are “not knowledgeable”. Please see appendix IIIC for the wording of the pension questions.
3.2.4 LOCUS OF CONTROL & PARENTAL STIMULATION
The richness of the dataset allows incorporating more factors than household income and retirement planning. Next to this information, there is also a dataset containing psychological factors. Locus of control is one of these extra factors, consisting of 12 questions regarding the internal locus of control of households. Locus of control may influence retirement planning in a number of ways. First of all, individuals who have a high locus of control may be more proactive in terms to meet retirement changes and retirement saving schedules (Carter & Cook, 1995). High internal locus of control individuals may be more likely to actively change their household finances and may therefore more likely to actively plan and save for retirement. This is an interesting aspect to integrate in this study, as perceived control of life could impact control of retirement. However, it is rather difficult to determine whether there is an illusion of control. Therefore, a dummy variable is created that equals one for the individuals who score in the top decile (i.e. possess high internal locus of control), and zero otherwise.
Another factor worth examining regarding financial literacy is the attitude towards money during one’s 12th to 16th year of life. These years are most likely to shape one’s perspective of the value of money and the various options to spend or save it. When a minor is stimulated to save money or plan an expense, he is likely to understand the importance of a saving plan and could therefore be more likely to save for retirement.
Abramovitch et al. (1991) suggest that children who received an allowance between their 10th and 16th years of life are more aware of the concepts of money. This group is also more sophisticated with money and financial products than their counterparts who did not receive an allowance. This study suggests that receiving an allowance could be beneficial in financial planning and in retirement planning. It is worth examining whether receiving an allowance during one’s youth has influence one their retirement planning and financial literacy.
The measure is obtained through 2 questions regarding money and financial means in the respondent’s 12th to 16th year of life. The measures concern receiving an allowance from parents and the stimulation from parents to plan expenditures and save money. In
17 order to incorporate these measures, dummy variables are created of the 2 questions regarding receiving allowances and parental stimulation to save money. These dummies are equal to one if the individual has received an allowance and was stimulated to save money, and zero otherwise.
3.2.5 CONTROL VARIABLES
With respect to the control variables, van Rooij et al. (2009) and van Rooij et al. (2011) are followed. These control variables are socioeconomic variables, in order to capture the effect of financial literacy and financial advice. Please see table 2, panel IV for the descriptive statistics.
Age;
Gender;
Education;
The natural logarithm of gross income;
Occupation;
Home ownership;
Having kids in the household;
Having a partner in the household 3.3 METHODOLOGY
Looking at the main equations, it becomes clear that two analyses have to be performed.
A first one relating financial literacy to retirement planning and a second one explaining financial advice to retirement planning.
From a policy- and governmental perspective it is interesting to know whether financial literacy influences retirement planning but it would be even more interesting to know whether this relation is causal in nature. Financial literacy may cause people to be confident of their own financial capacities and thus may prepare for their retirement based on their own assumptions or information. However, financial literate people may also see that the cost of information gathering is quite large and can also see the benefits of outside advice of professionals.
Preparation for retirement is important to look at the relationship between financial literacy and planning. This is why an extra, important question has been included in the
18 survey: “How much have you thought about retirement: A lot, some, little, or hardly at all?”
This question demonstrates the degree of thinking about retirement and this is linked with the preparation for retirement. Another question “how much have you planned for retirement”, demonstrates the degree to which a person has planned for retirement and has thought about his retired days. A study by Lusardi (2003) has shown that thinking about retirement is strongly associated with the accumulation of wealth in U.S.
households when nearing retirement. Lusardi demonstrates that households who think about retirement planning a little, some or a lot possess double the amount of wealth than households who have given hardly any thought on retirement. A dummy is set equal to one if the individual has thought a lot or somewhat about retirement and zero otherwise.
Financial advice is a dummy variable indicating whether and where the individual has received financial advice from. The first category is “financial advice”, indicates whether the person has been seeking financial advice. Pension knowledge is another proxy for retirement planning. As pension laws and obligations are quite hard to understand, the more an individual understands pension regulation, the more likely he is to plan for retirement.
Firstly, to test the effect of financial literacy on retirement planning, the following linear probability model is estimated: 5
𝑅𝑒𝑡𝑖𝑟𝑒𝑚𝑒𝑛𝑡 𝑝𝑙𝑎𝑛𝑛𝑖𝑛𝑔 = 𝛼1+ 𝛽1 𝐹𝐼𝑁𝐴𝑁𝐶𝐼𝐴𝐿 𝐿𝐼𝑇𝐸𝑅𝐴𝐶𝑌 + 𝛿 𝐶𝑂𝑁𝑇𝑅𝑂𝐿ℎ+ 𝜀ℎ (1) Where financial literacy is a measure which consisting of the advanced literacy
questions and 𝐶𝑂𝑁𝑇𝑅𝑂𝐿ℎ represents a set of control variables for household h.
𝑅𝑒𝑡𝑖𝑟𝑒𝑚𝑒𝑛𝑡 𝑝𝑙𝑎𝑛𝑛𝑖𝑛𝑔 = 𝛼2 + 𝛽2 𝐴𝐷𝑉𝐼𝐶𝐸 + 𝛿 𝐶𝑂𝑁𝑇𝑅𝑂𝐿ℎ+ 𝜀ℎ (2) Where ADVICE is a dummy variable that equals one for households that rely on a professional financial advisor and zero otherwise and 𝐶𝑂𝑁𝑇𝑅𝑂𝐿ℎ represents a set of control variables for household h.
5 In this situation, a logit of probit model is more suitable. However, with a logit or probit model the coefficients to be interpreted with more caution than ordinary linear probability model. The results are robust compared to a logit or probit model. The results of a logit model are included in appendix I to display the robustness of the tests.
19
4. RESULTS
This chapter provides the results of the various analyses. This chapter starts by presenting the results of the relationship between financial literacy and retirement planning, followed by the relationship of financial advice on retirement planning.
4.1 FINANCIAL LITERACY AND RETIREMENT PLANNING
The results of the regression estimates of financial literacy and retirement planning (equation 1) can be found in table 3. First, a benchmark is created by regressing
retirement planning on the control variables. There is evidence that the age group of 20- 40 is negatively related to thinking about retirement. As the process of retirement planning is still a distant concept and other life events, i.e. buying a house or starting a family, may have priority. Furthermore, having a partner, a house, high income and low risk aversion significantly contributes to retirement planning. After the age of 60, retirement planning declines. This age group doesn’t actively plan for retirement
anymore, as this group is relatively close to retirement age and the necessary steps have most likely been taken.
Next, measures for basic financial literacy, advanced- and pension literacy are included.
Basic financial literacy shows no significant relationship with retirement planning. This implies that knowledge of the basic concepts of money and financial concepts does not relate to retirement planning. Sophisticated literacy however, is an important driver of retirement planning. When controlled for by the socioeconomic variables, the
relationship is significantly positive. Looking at the magnitude of the sign of
sophisticated literacy (.05), a one standard deviation level increase in financial literacy is associated with a 5% higher probability of having thought of retirement.
Pension knowledge shows a positive, significant relationship with retirement planning.
When pension knowledge is incorporated in the analysis, the sign of sophisticated financial literacy fades away gradually as the effect of pension knowledge increases strongly. This results shows that pension knowledge may have a stronger relationship with retirement planning than sophisticated literacy, while this result is still significant.
This result indicates that pension knowledge is an indicator whether households are (actively) planning for retirement.
20 Home ownership is strongly related to retirement planning. In 2005, when the survey was conducted, Dutch house prices had increased considerably (van Rooij et al., 2009).
This increases the opportunity to use home equity as an extra resource in the retirement plan. The result is robust, even when controlled for with financial literacy. This is
supported by the survey, as the direct question relating the sale of the house in order to use as retirement funds shows that home ownership may serve as a form of retirement plan. Almost 11% of the respondents say there is an 80% or more chance they will sell their house to contribute to their retirement savings. Education however shows no relationship with retirement planning. It seems that education has no direct relationship with retirement planning when controlled for by socioeconomic variables.
As the relationship between financial literacy and retirement planning seems strong, the regression predictors should be interpreted with caution. There might be an
endogeneity problem due to omitted variables. In order to address the causality issues, IV estimation is performed. Using instrumented variables, economic education is used as instrument for advanced financial literacy. This variable measures the exposure to economic education at school. This measure is obtained from the question “how much of your education was devoted to economics?” Responses to this question are measured in 4 categories from a lot to hardly at all. Economic knowledge obtain during the school years is assumed to be unrelated with thinking of retirement, as this process is quite distant in terms of time of one another (Van Rooij et al., 2009). The results from the instrumental variables are displayed in table 3. It has strong predictive power for advanced financial literacy. The F-value equals 13.68, which is well above 10, the value that is
recommended as a rule of thumb to avoid problems of weak instruments (Van Rooij et al., 2012; Staiger & Stock, 1997). In this regression, education is significantly related to retirement planning, in contract with earlier regression. This provides evidence that economic education at school is a strong predictor of financial literacy.
Furthermore, there is some evidence that people who own a house, have a partner, earn higher wages and risk tolerant, tend to plan for retirement.
21 TABLE 3
FINANCIAL LITERACY AND RETIREMENT PLANNING, MULTIVARIATE RESULTS This table provides estimation results of the OLS regression models. In all regressions, the dependent variable is retirement planning, which is a dummy variable if the person has thought “a lot” or at least
“some” about retirement, and 0 otherwise. The results provided are based on the whole sample of 1,081 observations. The results are based on the 2005 DHS Household Survey wave.
The coefficients are reported along with the corresponding robust standard errors in parentheses.
*p<.10, **p<.05, ***p<.01.
Retirement
planning Retirement
planning Retirement
planning Retirement
planning Instrumental variables
Pension knowledge .04
(.02)***
Basic financial literacy .00
(.01) .00
(.01) Sophisticated financial
literacy .05
(.00)*** .01
(.00)*** .01
(.00)** .14
(.05)**
Age ( 20-30 ) -.38
(.09)*** -.37
(.09)*** -.37
(.09)*** -.37
(.09)*** -.37 (.11)***
Age ( 30-40 ) -.27
(.09)*** -.28
(.09)*** -.28
(.09)*** -.28
(.09)*** -.33 (.11)***
Age ( 40-50 ) -.17
(.08)** -.17
(.08)** -.17
(.08)** -.17
(.08)** -.18
(.10)*
Age ( 50-60 ) -.03
(.08) -.04
(.08) -.04
(.08) -.04
(.08) -.10
(.10)
Age ( 60-70 ) .01
(.05) .01
(.05) .01
(.05) .01
(.05) .02
(.06)
Education -.00
(.01) -.00
(.01) -.00
(.01) -.00
(.01) -.05
(.02)**
Male -.01
(.03) -.04
(.03) -.04
(.03) -.02
(.03) -.24
(.10)**
Retired -.00
(.07) -.00
(.07) -.00
(.07) -.00
(.07) .00
(.09)
Partner .35
(.08)*** .35
(.08)*** .35
(.08)*** .35
(.08)*** .31
(.10)***
Kids .01
(.01) .01
(.01) .01
(.01) .01
(.01) .01
(.02)
Home .31
(.08)*** .32
(.08)*** .32
(.08)*** .32
(.08)*** .34
(.10)***
Log gross income .07
(.02)*** .06
(.02)*** .06
(.02)*** .06
(.02)*** -.02 (.04)
Employee .04
(.04) .06
(.04) .06
(.04) .06
(.04) .17
(.07)**
Self-employed .06
(.08) .06
(.08) .06
(.08) .06
(.08) .08
(.10)
Locus of control -.03
(.05) -.03
(.05) -.03
(.05) -.03
(.05) -.03
(.06)
Allowance .00
(.01) .00
(.05) .00
(.05) .00
(.05) -.02
(.01)
Saving .03
(.01)** .03
(.01)** .03
(.01)** .03
(.01)** .00
(.01)*
Risk aversion -.12
(.04)*** -.11
(.04)*** -.11
(.04)*** -.11
(.04)*** -.06 (.05)
Constant -.08
(.20) -.08
(.20) -.18
(.24) -.26
(.23) -.13
(.25)
𝑅2 .12 .13 .13 .16 .24
22 4.2 FINANCIAL ADVICE AND RETIREMENT PLANNING
The results of the regression of financial advice and retirement planning (equation 2) are provided in table 4. Financial advice has a positive, significant relationship with retirement planning. Financial advice is likely to have an influence on retirement planning. Financial advisors are likely to consult their clients on their future financial needs, which is explained by the strong relationship it has on retirement planning and the considering of a retirement savings plan.
Households with partners are more likely to seek financial advice. This is potentially due to the arrangement of a collective household savings fund or some other saving program in order to safe sufficient money. Having a partner may come with more challenging financial decision which affects the household financing. Next, income significantly increases the probability of seeking financial advice. As income increases, financial advice seeking increases as well. Individuals in the lower income group are less likely to seek advice (Joo and Grable, 2004). This is mainly due to the optimal allocation of funds and the setting up of a retirement fund.
Another finding is that education has no apparent relation to financial advice and retirement planning. One would assume that lower educated individuals could prosper from financial advice in order to make optimal financial decision. The results however show no relation. This could be due the fact that these individuals do not wish to engage in complicated financial products and/or decision in retirement planning, as they don’t trust them of don’t fully understand what it means. This is an interesting finding, as this group could potentially benefit the most from financial advice to optimize future
financial needs.
Home ownership is positively related to financial advice and retirement planning. As buying a house is an important and capital intensive event, seeking financial, objective advice is a perfect way to optimally arrange financial needs. Along advice for housing financing, individuals potentially can seek advice for retirement planning too. As
discussed before, home equity is a source of retirement planning and can thus be used in conjunction with financial advice.
Next to that, risk aversion is negatively correlated with seeking financial advice for retirement planning. Individuals, who have a positive attitude towards risk-reward and
23 have higher risk tolerance, are more inclined to seek advice regarding pension planning.
These individuals are more likely to see the potential benefits of advice, while also acknowledging the potential risks.
Furthermore, there is some evidence that having a partner, a house, higher income, and stimulated to save money increases the probability of seeking advice for retirement planning.
Financial advice has the potential to be a substitute for a lack of financial literacy in retirement planning. In terms of the effect of financial advice on financial literacy and retirement planning, financial literacy shows a strong, positive relationship with advice seeking on retirement planning. Financial literate individuals are more likely to consult a professional financial advisor. This indicates that individuals who are above-average financial literate are more likely to seek professional advice than their less literate equivalents. The results show that financial advice is rather a complement to financial literacy than a substitute for financial knowledge. While these individuals may be capable of make sound financial decisions, this group is more likely to observe the
benefits of a professional financial advisor who is likely to give objective advice and have a greater willingness and ability to pay for it.
While financial advice is potentially most valuable and sought by individuals with lower education levels, as they may face higher information costs and because they are more likely to make errors, these results do not support these claims. The results show that individuals with higher income (income category and gross income), are more likely to seek outside professional advice. As a whole, these findings suggest that the seeking and use of financial advice is more likely among individuals with higher income and higher education.
24 TABLE 4
FINANCIAL ADVICE AND RETIREMENT PLANNING, MULTIVARIATE RESULTS This table provides estimation results of the OLS regression models. In all regressions, the dependent variable is financial advice, which is a dummy variable which is a 1 if the person has reported to receive financial advice and 0 otherwise. The results provided are based on the whole sample of 1,081
observations. The results are based on the 2005 DHS Household Survey wave.
The coefficients are reported along with the corresponding robust standard errors in parentheses.
*p<.10, **p<.05, ***p<.01.
Retirement planning Retirement planning
Professional advice .21
(.03)*** .21
(.03)***
Sophisticated financial literacy .01
(.00)***
Age ( 20-30 ) -.36
(.09)*** -.37
(.09)***
Age ( 30-40 ) -.28
(.08)** -.29
(.08)**
Age ( 40-50 ) -.18
(.08)** -.18
(.08)**
Age ( 50-60 ) -.04
(.08) -.05
(.08)
Age ( 60-70 ) .00
(.05) .00
(.05)
Education .00
(.01) -.00
(.01)
Male .00
(.03) .02
(.03)
Retired .00
(.07) .00
(.07)
Partner .34
(.08)*** .33
(.08)***
Kids .01
(.01) .01
(.01)
Home ownership .30
(.08)*** .31
(.08)***
Log gross income .06
(.02)*** .06
(.02)**
Employee .04
(.04) .05
(.04)
Self-employed .03
(.08) .04
(.08)
Locus of control -.03
(.05) .00
(.02)
Receiving allowance .00
(.01) .00
(.01)
Saving .03
(.01)** .02
(.01)**
Risk aversion -.12
(.04)*** -.11
(.04)***
Constant -.04
(.23) -.11
(.26)
𝑅2 .15 .16
25 4.3 WHO IS FINANCIAL ILLITERATE
Since there is evidence that financial literacy appears to positively impact retirement planning and other positive attributes towards the treatment of household finances, it is interesting from a policy perspective to identify what kind of individuals are financial illiterate. In order to create a measure for illiteracy, a reverse measure is made of the financial literacy module.
Please see table 5 for the estimation results of the regression of financial illiteracy on a set of covariates. The covariates are based on the variables used throughout this paper.
It appears that gender is positively related to financial illiteracy: women tend to be more financial illiterate. Furthermore, people who have finished senior vocational education tend to be illiterate. Education level is a strong predictor of financial illiteracy. This is in line with Lusardi and Mitchell (2011), who find that a lower level of education is
associated with illiteracy. As education is a strong predictor of financial illiteracy, policy makers should focus on lower education level.
Internal locus of control is also negatively related to financial illiteracy. Individuals, who tend to see events as outside of their sphere of influence, tend to see these events as non-influence able and don’t act on these events. This has an effect on financial literacy levels, as these individuals are less likely to act on financial events and thus do not plan expenses or save money.
Stimulation to save money is also negatively related to financial illiteracy. An individual who was not stimulated to save money and plan expenses as a child, tends to be financial illiterate. These individuals are more likely not to plan (large) expenses and not to save money.
Furthermore, women, individuals with lower gross income, that have a lower locus of control, that were not stimulated to save money and are risk averse tend to be financial illiterate.
Clearly, these results only explain and describe a part of the determinants of financial illiteracy. However, these results show that mostly lower educated groups and women display poor financial literacy (Lusardi & Mitchell, 2005).
26 TABLE 5
Explaining financial illiteracy
This table provides estimation results of the OLS regression. The dependent variable is financial illiteracy.
The results provided are based on the whole sample of 1,081 observations. The results are based on the 2005 DHS Household Survey wave. The coefficients are reported along with the corresponding robust standard errors in parentheses. *p<.10, **p<.05, ***p<.01.
OLS
Age -.00
(.00)
Gender .54
(.19)***
Education -.41
(.05)***
Partner -.26
(.46)
Kids .01
(.10)
Retired .45
(.34)
Log gross income -.55
(.13)***
Home .26
(.48)
Employee .83
(.27)***
Self-employed .17
(.47)
Locus of control -.02
(.01)**
Receiving allowance -.03
(.06)
Saving money -.15
(.07)**
Risk aversion .42
(.24)*
Professional advice -.51
(.20)**
Constant 8.99
(1.49)***
𝑅2 .23
27
5. DISCUSSION & CONCLUSION
This chapter starts with a discussion of the results, followed by the conclusion. This chapter ends with the limitations and recommendations for further research.
5.1 DISCUSSION
Literature shows that households tend to lack financial literacy in financial decision making and retirement planning (Van Rooij et al., 2009; Lusardi & Mitchell, 2008). Yet, it is difficult to figure out what the exact drivers of financial literacy and retirement
planning are. More evidence is supporting that socioeconomic and psychological factors can explain financial literacy and retirement planners.
Dutch regulation requires individuals to be more self-reliant in terms of pension planning. Having the knowledge and tools to be self-reliant is important in order to accumulate the necessary wealth for retirement. This study finds that Dutch households don’t plan much for retirement: only 12.7% of respondents have reported to plan for retirement and think about it actively. While planning for retirement is an important decision considering the changing pension regulation, this study confirms that
households do not plan for the long term. This study relies on a dataset of 2005, a time in which the Dutch pension system was quite generous. As a large portion of the pension savings was in the hands of defined contribution plans, this potentially explains why only 12% of households have thought a lot about retirement.
This study confirms that there is a positive, significant relationship between financial literacy on retirement planning in the sample of Dutch households. While basic financial knowledge doesn’t have an impact on retirement planning, advanced financial
knowledge increases the likelihood of thinking about future financial needs and thus increases thinking about retirement. This group is also more likely to set up a retirement plan and save more money. This study contributes to the existing evidence that shows that high levels of financial literacy correlate with household financial behavior that is consistent with recommended policies for retirement savings.
While education and education level is often seen as a driver for financial literacy, other factors are potentially also important in financial literacy. Parental stimulation to plan expenditures is positively related to retirement planning. Individuals who were
stimulated to save money and plan expenses during their youth become better planners
