Financial literacy and pension preparation in the Netherlands
Student name: Yijun Wang Student number: S3285944 Supervisor: Pro. B.W.Lensink
1
1.Introduction
From the middle of 20th century, the increasing tendency of life expectancy at birth arose in the whole world, especially in developed countries such as Japan and in European countries like German, Sweden and the Netherlands. Meanwhile, the downward trend of the infants' mortality rate and continuous improvement of health and medical system led to the issue of the aging population. To avoid overburdening of the labour population and the healthcare system, some developed countries plan to raise the retirement age stepwise in the future. In the Netherland, for example, the government reached an agreement in 2012 to accelerate the progress of raising the statutory retirement age after 2015, with the view to increasing the legal retirement age to 66 years-old and 67 years-old in 2018 and 2021 respectively. The data from the World Bank state that life expectancy increased with five years in the Netherlands from 1990 to 2014 but the statutory retirement age kept at 65 years-old without increment until 2018, which means the period in retirement status for individuals is prolonged. Therefore, the retirement planning and preparation play an essential role in increasing individual's financial well-being, especially in the retirement period. Retirement planning covers a wide range of dimensions, for instance, the retire timing, bequest motive, the type of pension plan that individual participated, etc. Most studies based on the consumption life-cycle model focus on examining the impact of factors like pension, bequest motive and survival probability on the decision to retire (Samwick, 1998; Brown, Coile and Weisbenner, 2006). Although Yakoboski and Dickemper (1997) analysed the situation of American’s retirement planning based on data from the 1997 Retirement Confidence Survey and also HRS provide the information on indicators of retirement planning, numbers of research on what factors can affect planning for retirement are not many. Furthermore, although pension is an essential part of retirement planning, few scholars research on this topic.
2
of money, and questions about advanced financial literacy such as portfolio investment. These surveys about financial literacy mostly are used to examine its impact on individual’s financial behaviours. For instance, Van Rooij, Lusardi and Alessie (2011a) demonstrate the significant link between stock market participation and financial literacy, financial literacy and households net wealth. Lusardi and Mitchell (2006, 2011) publish a variety of studies about the association between financial literacy and retirement behaviour.
Since retirement planning is crucial for increasing households' financial well-being in the retirement period, some researchers try to figure out what elements can influence household retirement planning. Apart from measuring the impact of the common factors like bequest motive and income on planning, increasing numbers of scholars began to consider, for instance, psychological characteristics and background features such as parents education level as influential variables on the retirement planning. This paper aims to assess whether the factor financial literacy affects people’s retirement planning. The innovation point of this paper is considering different levels of pension preparation as indicators of retirement planning instead of these usually used by van Rooij, Lusardi, and Alessie (2011b). The details of retirement planning indicators will describe in section 3.
The outline of this paper is as follow. In section 2, I will introduce the pension system in the Netherlands. In section 3, I will discuss the survey data, indicate the methodology for measuring the financial literacy indices and elaborate on the indicator of retirement planning— the pension preparation. In section 4 I will present the empirical result of the regression of pension preparation on financial literacy and control variables, and give the interpretation of their relationship. Finally, I will draw conclusions in section 5.
2.Pension system in The Netherlands
3
entitle to the full payment of pension without any punitive measure, but it is illegal to receive a pension early or late. The universal coverage public pension scheme is a feature of the Dutch pension system that effectively reduces the poverty rate of the elderly population. According to statistics from the Eurostat in 2010, the poverty rate of the population aged over 65 in 15 primary EU member states in 2007 is about 5%. In the Netherlands, the poverty rate in the population group from18-year-old to 64-year-old is about 3%, while the poverty rate of the population aged over 65 is less than 1.5% (Anderson, 2012).
The second pillar is the occupational pension which is dominated by the social partners. On the basis of the public pension or old-age pension, workers are provided with supplementary income linked to their income levels during their employment. Generally, the occupational pension plan is confirmed by the result of the consultation between employers and employees. The government does not play a dominant role in this type of pension, but it can indirectly inject capital into occupational pension plans through policies such as tax deductions or exemptions. Meanwhile, the government has assumed the crucial responsibility of supervising occupational pension plans. In the Netherlands, the coverage of occupational pensions is very broad, with more than 90% of employees participating in various occupational pension schemes. Anderson (2012) summarized that there were 656 occupational pension schemes which include 543 company schemes, 69 compulsory sector plans, 26 non-compulsory schemes and 13 Special professions pension schemes.1In different occupational pension plans,
the contribution rates of employers and employees are different. In general, employers pay two-thirds, and the residual part is the responsibility of employees. The vast majority of occupational pension plan in the Netherlands is defined-benefit (DB) schemes. According to OECD estimation in 2013, 93.5% employees covered by occupational pension plans in the Netherlands enjoyed annuities from DB plan, the rest of the employees participated in defined contribution scheme. An employee who has worked over 40 years can expect to obtain the pension replacement rate reaching 70%-75% of the average wage level when old-age pension and occupational pension are combined. Occupational pensions in the Netherlands are mostly managed by pension funds. As of 2012, the number of occupational pension fund decrease from 876 in 2003 to 414 in order to facilitate monitoring, save management costs and promote the movement of the labour force.
4
The third pillar is personal pension savings, including life insurance and pension savings plans that enjoy tax benefits. Workers who are not adequately covered by the second pillar can voluntarily choose personal pension savings to supplement pension income. For example, the life-course savings scheme pursued in 2006. But the proportion of employees who choose this way to increase pension income is small with approximately 10%.
3.Data and Methodology
3.1. Data and demographic variables
In this paper, I use data from the DNB Household Survey (DHS). It is a panel survey over 2000 household and has been active since 1993. The DHS covers information in term of demographic and economic concepts and individual characteristics of financial behaviour while focuses on wealth, income and saving data. This panel dataset is conducted by CentERdata, an eminent research institute and locate on the Tilburg University (TiU) campus. The careful selection procedure of participating households of DHS can safeguard the representativeness of the Dutch population. Firstly, participating households will be recruited based on the phone interview. After that, household who ensure to participate in the survey will fill in the questionnaires on the internet, which is the main channel to collect survey data because the internet penetration is high in the Netherlands (80%). If there is no internet connection in participants, the television set-top box will be the alternative tool to conduct the investigation.2
The financial literacy data is from two modules designed by Van Rooij, Lusardi and Alessie in 2005. The first module was filed from 23 September to 25 September in 2005 and the second module was fielded in January 2006. The response rate is 74.36% that a total of 1508 out of 2028 households responds the questionnaire. Because data on financial literacy only available in 2005 of the DHS and to keep time consistency between data of pension preparation, financial literacy and demographic factors, I merge the financial literacy data with the 2005 aggregate data from the questionnaire about income, wealth and household information.
The final sample of the thesis is composed of 1006 respondents. Table 1 illustrates definition and summary statistics of demographic variables for the final sample. All demographic variables are dummy variable except the variable “the number of children in the household”.
2 A set-top box is an electronic device that connects a television with an external signal source. It can convert compressed
5
Variable equal to 1 if it satisfies the definition and 0 otherwise. The average age of respondent is 51.2 and the group whose age between 46 and 55 gives the highest participation rate, the group whose age older than 65 also shows relatively high response rate. 58.1% respondents are male and 67.7% respondent have marriage or cohabiting.
Table 1. Definition and Summary statistics of demographic variables.
Demographic variables Definition Mean
Age dummy
Respondent’s age falls within the age range
age ≤ 25 0.149 25<age ≤ 35 0.192 35<age ≤ 45 0.166 45<age ≤ 55 0.230 55<age ≤ 65 0.188 age ≥ 65 0.223 Education dummy
Highest level of education that respondent complete
Special and Primary education 0.011
Pre-vocational education 0.224
Pre-university education 0.179
Senior vocational education 0.115
Vocational colleges 0.281
University 0.181
Others 0.009
Net income dummy
Total net income household income
1st quartile (lowest) 0.231
2nd quartile 0.231
3rd quartile 0.231
4th quartile (highest) 0.230
House owner 0.663
Male 1 means respondent is male and 0 mean respondent is female 0.581
Married Respondent is married or living together with partner 0.677
Retired The Primary occupation of the respondent is retired 0.231
Employed The Primary occupation of the respondent is employed 0.540
6
3.2. The measurement of financial literacy indices
The two modules designed by Van Rooij, Lusardi and Alessie include two sets of questions, one set of questions relates to basic financial knowledge and another set of questions in term of advanced financial literacy. Although the design of questions of the two modules took the reference to similar modules from other investigations, it also has a few of different unique questions.
The set about basic financial literacy contains five questions. These questions measure the ability of numeracy (the first question), interest compounding (the second question), inflation (the third question), time value of money (the fourth question) and money illusion (the fifth question) respectively. The rest eleven questions are more complicated than questions of the first set, aim to assess understanding of nature, return and riskiness of financial assets and market. According to analysis from Van Rooij, Lusardi and Alessie (2011a) for the pattern of answers to these questions, they concluded that advanced financial literacy is not widespread. This conclusion induced two worries that whether respondents can understand the meaning of financial questions and whether the possibility that randomly guessing answers is very high. Therefore, they used the strategy that inverted the wording of questions and asked two randomly selected groups of respondents the same question but with the different expression to assess the prevalence of two worries. They used the strategy on three original questions: one about comparing the riskiness between stock and bond, one about comparing the safety of return between company stock and mutual fund and the last one about the impact of the increase or decrease of the interest rate on the change of bond price. Unfortunately, the DHS not provides the exact data about these three questions. Thus, I exclude the three questions and only use eight out of thirteen questions to construct the financial literacy indices. The chosen eight questions cover knowledge about the function of the stock market, nature of investment behaviour, the uncertainty of asset return and investment diversification. The exact wording of the two sets of questions is reported in Appendix A.
7
of incorrect responses to the last two questions is high that more than 20%. For the question measuring inflation, the proportion of incorrect answer is similar with that of don’t know answer, at around 7%.
Figure.1 Distribution of response to basic literacy questions. The sum of correct, incorrect and do not know (DK) response may not up to 100% because of not including refusal.
From the summary statistics of the basic literacy set in Table 2, we can note that while the percentage of the correct answer to each question is high, the proportion that answers all questions rightly is only 43.6%. One good signal is no respondent answer to all questions incorrectly, but there is a fraction of respondents that response DK to all five basic financial literacy questions (1.5%). Thus, basic financial literacy is not understood fully by the masses although most interviewees behave knowledge of a little of financial concepts.
Table 2. Summary statistics of answer to basic financial literacy questions
Weighted percentage of total number of respondents (N=1508)
Rate of correct, incorrect and do not know(DK) answer (out of five questions)
0 1 2 3 4 All Mean
Correct 2.1 2.2 5.1 14.1 32.6 43.6 4.03
Incorrect 48.0 34.3 12.9 3.4 1.0 0.0 0.74
DK 89.9 5.2 1.5 0.9 0.5 1.5 0.21
Responses to questions of the advanced financial literacy set are reported in Figure 2. The pattern of responses to these eight questions is quite different from that in the previous set. The proportion of correct answers of the advanced literacy questions set is generally lower than that of the basic literacy set. For the question of the uncertainty of assets return, its proportion of
92.6 80 84.9 74.5 71.7 4 16.6 7.5 21.1 24.5 3.1 3.1 7.1 3.9 3.4 0 10 20 30 40 50 60 70 80 90 100 Numeracy Interest compounding
Inflation Time value of money
Money illusion
8
correct response is highest in the eight advanced questions but is slightly higher than the lowest proportion of correct answer in the prior basic literacy set (the difference is only 0.1%). For the proportion of incorrect responses on questions measuring advanced financial knowledge, the highest percentage (28.9%) is close to that in the basic knowledge set, and the lowest percentage is four times as much as that in the previous set. Besides, the proportion of DK answers to the advanced literacy questions is in the range between 9.6% and 35.5%, implying partial respondents have no idea about the meaning of these advanced literacy questions.
Figure.2 Distribution of response to advanced literacy questions. The sum of correct, incorrect and do not know (DK) response may not up to 100% because of not including refusal
Respondents’ abilities to understand the function of the stock market and mutual funds, to distinguish uncertainty among different asset and to apprehend the meaning of risk diversification, are not very bad. Above seven out of ten respondents know which asset has the highest uncertainty (71.8% response correctly) and can state the basic function of the stock mutual fund (70.2% accuracy). The incorrect response rate of the two questions approximately is 11% and much lower than that of the rest of questions in advanced literacy set. The definition about bond and stock is not clear for respondents. Around a quarter of respondents give an incorrect answer to the question assessing the understanding for buying stocks, and over 15% respondents provide a wrong response to issue about buying bonds. The gap of percentages between correct, incorrect and “don’t know” answers to the question about the bond transaction (the last question in figure 2) is very narrow. For this question, more than 35% respondents
33.5 66.8 71.8 50.9 59.9 70.2 64.6 69 27.8 16.8 11.7 28.9 16.1 10.9 24.7 13 35.5 16 16 19.8 23.5 18.4 9.6 17.4 0 10 20 30 40 50 60 70 80 90 100
If you buy a 10-year bond, can you sell it after 5 years without incurring a major penalty How about the risk of loosing money when investor spreads money among different asset
which asset displays the highest fluctuations Which asset normally gives the highest returen
in the long-term
Buy a bond of firm B Statement about mutual funds Buy the stock of firm B Function of the stock market
9
don’t know the answer, near 30% answer incorrect and only a little over 30% respondents give the right response.
Table 3 shows only a tiny part of respondents can answer all eight questions correctly with 15.52% and a sizeable fraction of interviewees answer incorrectly or "do not know" on a couple of questions. Thus, advanced financial knowledge is less widespread than basic financial knowledge. Quite a lot of people lack the cognition about concepts of advanced financial literacy.
Table 3. Summary statistics of answer to advanced financial literacy questions
Weighted percentage of total number of respondents (N=1508)
The above financial literacy information which is gathered from the response to the two sets of financial knowledge questions can be used to construct financial literacy indices. The two sets of questions contain thirteen different questions which will lead to multicollinearity and information overlap problem if we set one question as one explanatory variable. Therefore, factor analysis is adopted to reduce dimensionality, to find some main indicators for explaining.3 The preparatory work before the factor analysis is to construct a binary variable for each question where 1 means respondents provide correct answer and 0 otherwise. When performing the factor analysis on the thirteen questions through the iterated principal factor method, two factors have eigenvalue greater than one.4 According to Kaiser’s rule that the factors with the eigenvalue greater than one can be retained, we can keep two common factors to describe the whole variables derived from two sets of questions. The result of KMO (Kaiser– Meyer–Olkin) test for measuring sampling adequacy also shows that the data are appropriate to perform the factor analysis (overall value is 0.8418). In addition, the cumulative proportion value indicates that the two common factors can explain 76.77% of the data. Corresponding to the financial literacy investigation method designed by Van Rooij, Lusardi and Alessie, the
3 Factor analysis is the extension of principal component analysis, it more focuses on explaining the correlativity
among observed variables or the structure between covariances. Its essence is to find out the unobservable, underlying and unrelated random variable(s) and use it(them) to describe the relationship of these examined variables.
4 Iterated principal factor method is the amendment of principal factor method that provides a better way to
estimate communality and improve the estimation accuracy.
Rate of correct, incorrect and do not know(DK) answer (out of eight questions)
0 1 2 3 4 5 6 7 All Mean
10
two factors have different factor loadings on two parts of questions and thus are assigned to describe variables from each set of questions respectively. However, there is an additional step to construct the advanced financial literacy index. Since many respondents provide “don’t know” answer for advanced financial literacy questions, the information derived from the “don’t know” answer could not be neglected. Furthermore, Lusardi and Mitchell (2006) stressed the importance of this information to differentiate among levels of financial literacy. Therefore, the extra eight dummy variables for respondents who give “don’t know” answer to each question are used when constructing the advanced literacy index. The factor loadings relate to construct two financial literacy indices are reported in Table 4.
Table 4 Factor loadings corresponding to literacy questions
Basic literacy questions Factor loadings
Numeracy 0.6667
Interest compounding 0.5188
Inflation 0.5513
Time value of money 0.4267
Money illusion 0.2432
Advanced literacy questions Factor loadings
Statement about main function of stock market Correct -0.6245
DK 0.7308
Statement about buy the stock of firm B Correct -0.4350
DK 0.6216
Statement about mutual fund Correct -0.6498
DK 0.7603
Statement about buy a bond of firm B Correct -0.5971
DK 0.7298
Which asset gives the highest return in a long-time period Correct -0.5171
DK 0.7447
which asset displays the highest fluctuations over time Correct -0.6471
DK 0.7768
How risk of losing money when investor spreads money among different assets
Correct -0.5774
DK 0.7582
If you buy a 10-year bond, it means you cannot sell it after 5 years without incurring a major penalty
Correct -0.4516
DK 0.4869
11
education groups. In the frequency distribution of high-level basic financial knowledge, there is an indistinctive difference between male and female: more women display the high level of basic financial knowledge and more men show the lowest level of basic literacy. For each age group except the first age group, the frequency distribution among different basic literacy level are similar that its proportion overall presents an increasing tendency from lowest to the highest quartiles. The relationship between basic literacy and education level is as expected as most researchers assumed, also their outcomes verified it is reasonable (Hastings, Madrian, and Skimmyhorn, 2013; Van Rooij, Lusardi, and Alessie, 2011a). Interviewees who attained a high degree of education diploma are mostly concentrated on a high level of basic financial literacy: 44.51% of respondents who completed university education and 44.88% of respondents who obtain the degree of vocational college are at the highest level of basic financial knowledge. The frequency distribution characteristic of basic literacy among different levels of net income is similar to that among different age groups, presenting an increasing trend from the lowest quartile to the highest quartile.
The right half of Table 5 is the frequency distribution of advanced literacy across demographic variables. Features of frequency distribution are more significant in the advanced financial literacy quartiles than that in the basic literacy quartiles. The difference of frequency between female and male is obvious: more than 65% of female concentrate on low quartiles of advanced literacy and over six of ten of male display the high level of advanced financial knowledge. Frequency distribution of the four advanced literacy quartiles in each age group doesn't have the large gap except the first group. For respondents whose age is less than 26 (generally means studying or just graduated from school), their advanced financial literacy level is focus on the first and second quartile. Moreover, advanced financial literacy shows an indistinctive but exist relation with the age group. 28.04% of respondents who in the age range from 56 to 65 are in the top quartile of advanced literacy. The relationship between the advanced financial literacy and education still strong.
12
financial literacy with 18.13%, which is double the proportion at the bottom quartile of the advanced financial literacy. However, for respondents who have the lowest level of education at the first quartile of basic and advanced financial literacy, the tendency is opposite. Thus, the frequency distribution characteristic among different education level is more evident in the advanced literacy than that in the basic literacy that respondents who complete the higher (lower) degree of education display the higher (lower) level of financial literacy. For different income quartiles, the situation of the frequency distribution is in line with that among different education groups, where the low-income interviewees are concentrated on the low level of advanced literacy, and the affluent respondents show the high level of literacy.
Table 5. Basic and advanced financial literacy across demographic and socio-economic characteristics
Gender
Basic financial literacy quartiles Advanced financial literacy quartiles
1 2 3 4 Mean N 1 2 3 4 Mean N
Male 18.32 23.80 14.04 43.84 2.834 584 16.78 19.18 29.62 34.42 2.82 584 Female 16.11 22.51 13.74 47.63 2.929 422 36.49 32.94 19.43 11.14 2.05 422
Age
Basic financial literacy quartiles Advanced financial literacy quartiles
1 2 3 4 Mean N 1 2 3 4 Mean N <26 years 6.67 53.33 13.33 26.67 2.600 15 20.00 55.33 13.33 13.33 2.24 15 26-35 years 17.10 21.74 16.06 45.08 2.891 193 27.98 25.39 22.28 24.35 2.43 193 36-45 16.77 25.15 13.77 44.31 2.856 167 32.34 17.96 29.94 19.76 2.37 167 46-55 20.35 17.75 14.29 47.62 2.892 231 22.94 27.71 24.24 25.11 2.52 231 56-65 16.40 23.81 13.23 46.56 2.900 189 22.75 25.40 23.81 28.04 2.57 189 >65 years 16.59 26.54 12.32 44.55 2.848 211 21.33 24.64 27.96 26.07 2.59 211 Education
Basic financial literacy quartiles Advanced financial literacy quartiles
1 2 3 4 Mean N 1 2 3 4 Mean N
Special and primary 9.09 45.45 9.09 36.36 2.727 11 63.64 18.18 18.18 0.00 1.55 11 Pre-vocational 16.89 24.44 12.44 46.22 2.880 225 41.33 28.44 19.56 10.67 2.00 225 Senior vocational 21.11 21.67 13.33 43.89 2.800 180 35.56 26.11 24.44 13.89 2.17 116 Pre-University 16.38 22.41 11.21 50.00 2.948 116 18.97 25.86 18.56 10.67 1.69 180 Vocational colleges 15.90 23.32 15.90 44.88 2.898 283 16.69 24.03 27.92 31.10 2.73 283 University 18.13 21.43 15.93 44.51 2.868 182 8.26 20.88 23.08 47.80 3.11 182 Others 11.11 44.44 0.00 44.44 2.778 9 33.33 22.22 33.33 11.11 2.22 9
Income Basic financial literacy quartiles Advanced financial literacy quartiles
1 2 3 4 Mean N 1 2 3 4 Mean N
1st quartile 13.36 24.57 14.66 47.41 2.961 232 44.40 28.45 18.53 8.62 1.91 232
2nd quartile 16.81 26.29 12.07 44.83 2.849 232 21.98 32.33 29.31 16.38 2.40 232
3rd quartile 19.83 20.35 13.85 46.32 2.847 232 19.40 23.28 29.74 27.59 2.66 232
13
3.3 Retirement pension preparation
As I mentioned in section 1, Van Rooij, Lusardi and Alessie (2011c); Lusardi Mitchell (2006); Moure (2015) usually construct four indicators for retirement planning. The four indicators derived from data of four questions are "thinking about retirement" "simple planning" "serious planning" "successful planning" respectively. However, construction of the four indicators is based on the self-estimation of respondents whose assessment results may exist subjective bias to some extent. In addition, there is no assurance that these respondents who are viewed as successful planners would authentically prepare well for retirement by taking action like participating pension plan. Thus, the possibility that the respondents who defined as “successful retirement planners” or “serious retirement planners” are not active in retirement preparation may exist. I construct indicators of pension preparation based on three questions that involve pension preparation behaviours. The exact wording of the three questions is reported in Appendix B.
For the first question, respondents who provide answer 1 to 5 means they have made other arrangement for the pension. For the second question, number 1 means interviewees have endowment insurances, and number 2 means don’t have. For the third question, respondents give nonzero answer means they purchased pension insurance or single-premium insurance. According to the method that Lusardi and Mitchell (2006) construct relevant indicators based on dummy variables generated from questions about retirement planning, I construct four indicators stand for the levels of pension preparation. The first indicator named “simple preparation” which represent the lowest level of provision. As long as respondents do at least one pension preparation out of the above three, they belong to the category of "simple preparation". For the respondents who do at least two pension provisions out of the above three, they are at the intermediate level of the pension preparation, that is the second indicator "better preparation". If respondents do all three preparations, they at the highest level of the pension plan. Therefore, the third indicator “excellent preparation” is yielded. All the three variables are binary variables where 1 means respondents at the degree of pension preparation and 0 otherwise.5 Finally, the fourth indicator indicates what the level of preparation that respondents at. Number 1 of this indicator means respondents do simple preparation, number 2 means better preparation, number 3 means excellent preparation and number 0 means no
5 Number 1 of “simple preparation” indicator means respondents have at least one preparation and 0 means don’t have any
14
preparation. The main objective of above four indicators is to estimate the relationship between financial literacy and pension preparation. From the frequency analysis, 47.91% of households do simple preparation, 22.86% of respondents show having better preparation, only 4.47% of people have excellent preparation, and 52.09% of households don’t have preparation.
The descriptive statistics for three groups of respondents who has different level of pension preparation are provided in in Table 6. When comparing the mean of financial literacy indices under no-preparation category between the three group, one difference is evident. The average of advanced literacy index is negative value in the non-preparation category in the "simple" and the "better" group (-0.062 and -0.034 respectively) but stays with positive value in the "excellent" group. The advanced literacy index average of preparation category in “excellent group” is higher than that in other groups. It implies respondents who do excellent pension preparation may own high level of advanced financial literacy. However, the average of the basic literacy index in “simple preparation” group is greater than that in other two groups, implying people who have simple preparation have a relatively high level of basic financial literacy. Thus, I probably think that there is the relationship between pension preparation and financial literacy. Moreover, the different type of financial literacy may affect the different level of pension preparation. The authenticity of the existence of this relationship will be examined in subsequent regressions in section 4.
15
Table 6. Descriptive statistics (Mean) for different level of perpetration
Simple preparation Better preparation Excellent preparation =1 (482) =0 (524) =1 (230) =0 (776) =1 (45) =0 (961) Basic financial literacy 0.017 0.044 -0.002 0.041 -0.004 0.033 Advanced financial literacy 0.167 -0.062 0.325 -0.034 0.510 0.026 Age dummy age≤25 0.010 0.019 0.004 0.018 0.000 0.016 26<age≤35 0.187 0.197 0.143 0.206 0.133 0.195 36<age≤45 0.205 0.130 0.239 0.144 0.267 0.161 46<age≤55 0.295 0.170 0.326 0.201 0.356 0.224 56<age≤65 0.191 0.185 0.226 0.176 0.222 0.186 age≥66 0.124 0.313 0.078 0.265 0.044 0.231 Education dummy
Special and Primary 0.002 0.019 0.000 0.014 0.000 0.011
Pre-vocational 0.168 0.275 0.157 0.244 0.067 0.231 Pre-university 0.114 0.116 0.104 0.118 0.089 0.117 Senior vocational 0.199 0.160 0.157 0.186 0.222 0.177 Vocational college 0.336 0.231 0.335 0.265 0.378 0.277 University 0.176 0.185 0.239 0.164 0.200 0.180 Others 0.004 0.134 0.009 0.009 0.044 0.007
Net income dummy
1st quartile (lowest) 0.122 0.330 0.061 0.281 0.044 0.239 2nd quartile 0.226 0.235 0.209 0.237 0.111 0.236 3rd quartile 0.282 0.183 0.287 0.214 0.244 0.230 4th quartile (highest) 0.303 0.162 0.391 0.182 0.556 0.214 House owner 0.745 0.588 0.752 0.637 0.844 0.655 Male 0.654 0.513 0.735 0.535 0.867 0.567 Married 0.695 0.660 0.687 0.674 0.844 0.669 Retired 0.162 0.294 0.113 0.265 0.044 0.239 Employed 0.707 0.384 0.804 0.461 0.889 0.523 Number of children 0.714 0.517 0.760 0.567 0.956 0.595
4. Multivariate analysis of pension preparation
16
preparation that I described in the section 3.3. The first three indicators are dummy variables that take the value 1 if respondent have the corresponding extent of the pension preparation and 0 otherwise. The last indicator likewise aims to demonstrate what level of preparation that respondents have, but is no longer a binary variable. The key important independent variables are the financial literacy indices which are constructed by the method introduced in section 3.2. Moreover, I contain control variables of age, education degree, gender, married and household economic situation to take account into the possibility of individual heterogeneity to influence the relationship between financial literacy and pension preparation.
This paper uses the logit model to conduct regression analysis instead of the simple ordinary least squares (OLS) regression employed by Van Rooij, Lusardi and Alessie (2011a) and Lusardi and Mitchell (2011). The reason is that dependent variables−the pension preparation indicators are dummy variables that the predicted probability result to the dependent variable from OLS can be less than 0 since the OLS method not limit the probability between 0 and 1. The general form of the logit model is as follow:
𝑌𝑖 = Λ(𝛽𝑋𝑖+ 𝛿𝐶𝑖)
In this equation, Yi means the pension preparation status for respondent i, Xi is the vector of
ith respondent’s financial literacy indices and Ci is the vector of control variables, β and δ is
the estimated parameters respectively. For the logit model, we use maximum likelihood estimation (MLE) technique to perform the regression.
17
special”, “pre-vocational”, “pre-university” and “others”; the labour-force status dummy variable “retired” have not been taken account into the regression.
Table 7. The pension preparation and financial literacy: Logit regression
18 Married 0.18988 (1.07) 0.04731 0.08071 (0.38) 0.01133 0.96683** (1.94) 0.01571 0.07797 (1.20) Employed 0.75326**** (3.85) 0.18758 0.99155**** (3.85) 0.13928 1.19146*** (2.00) 0.01936 0.37978**** (5.16) Number of children -0.02689 (0.31) -0.00670 0.00194 (0.02) 0.00027 -0.10659 (0.63) -0.00173 0.00325 (0.10) Constant -1.87196**** (5.33) -3.44434**** (6.94) -7.34417**** (5.12) 0.03220 (0.26) Pseudo R2 0.1380 0.1665 0.1913 percent correctly predicted 67.89% 78.53% 95.53% R2 0.2111
Note: Absolute value of z-statistics and t-statistics in parentheses Significance levels: **** p<0.01, ***p<0.05; **p<0.1; *p<0.15
19
disposable income to plan for retirement and has higher possibility to make pension arrangements. This result is correspondent with the argument proposed by Lusardi & Mitchell (2006), that some individuals treat house equity as the source of retirement saving or an alternative tool to finance retirement.
Next, I estimate the relationship between the better pension preparation indicator and financial literacy (column 3 and 4). The significance level of the basic and advanced financial literacy is different with that in the first regression. Although the basic literacy index still displays a very weak relationship with better pension preparation, the correlation between advanced literacy index become significant. The advanced financial literacy that respondents have only promote the 2.8 percent possibility that do the better pension preparation. Age still plays an essential role in facilitating the better preparation of pension since three age dummy variables within the age range from 36 to 65 pass the joint significance test (p-value is 0.0009 and z-statistics is 16.38). The mean value of possibility to promote doing the better preparation is around 18%. However, the impact of age on the better preparation is not as conspicuous as that on simple pension plan since the average probability to promote people prepare for pension decrease from more than 20% to approximately 18%. the significance level of all income quartiles decreases (the absolute values of t-statistics are close to 0), which means the impact of the economic situation of respondents on preparation decrease. One difference can be notable that the third quartile of income no longer can influence the possibility of better preparation. This result is realistic and reasonable. Households with low income whose primary usage of earnings is to maintain basic life, or even sometimes the income couldn't cover all necessary expenditures, but a high degree of pension preparation means they should pay more premium in term of this provision. Thus, they are less likely to make a better pension preparation. The top quartile of net income and the labour force status "employed" also are promotion factor for better pension preparation as same as the result reported by the first regression. In this regression result, house ownership is no longer an influential factor that its significance (z-statistics) sharply decreases to a low level.
20
preparation. Although the age dummy variables keep their impact on preparation, its significance decrease that the third age dummy variable just pass the significance test with the 85% confidence interval. The possibility that interviewees in the second and fourth age range to make the excellent preparation decrease to less than 5%. A new important control variable “married” appear in this regression whose probability to promote respondents to have an excellent pension preparation is 1.57%. Such result is consistent with the consequence proved by Yabiku (2000). Marital status of married, divorced or remarried has high marginal effects on private pension coverage. The reason for the married status becomes an essential factor is married couples are eligible for more retirement benefits that single people are not. They can make better use of some specific pension or retirement regulations to coordinate their pension together and to obtain more profits. The change of result of “employed” variable from previous regression to this one is similar to the variation of age binary variables. The relationship between the marital situation and the pension preparation showed in the last regression is consistent with the consequence proved by Yabiku (2000). Marital status of married, divorced or remarried has high marginal effects on private pension coverage. However, there is gender difference that single woman is more likely to build private pension coverage and divorced or remarried woman but male shows opposite situation. The reason for the married status becomes an important factor is married couples are eligible for more retirement benefits that single people are not. They can make better use of some specific pension or retirement regulations to coordinate their pension together and to obtain higher benefits.
21
range variable. Respondents from 36-years-old to 55-years-old are at the vital stage of their whole lives. In this stage, they may enter the peak period of the career and have the additional disposable income to consider making pension preparation.
In summary, the advanced financial literacy indices keep the significance for promoting respondents to prepare pension all the time. The basic literacy index doesn't show any evident influence on pension preparation. Most demographic characteristic variables have the impact on the low and intermediate degree of pension preparation, except the variable "married" "male" and two education level variables.
5. conclusions
The analysis of investigation results leads to several conclusions. The first main conclusion is that for the two types of financial literacy, only sophisticated literacy affects the pension preparation. The pension is the imperative part of retirement planning whose aim is to ensure the retired people can maintain their old-age life. Sometimes pension preparation can be regarded as s kind of precautionary saving with a specific objective. But the difference between the pension plan and precautionary saving is the flexibility of withdrawing money. precautionary saving is a voluntary investment behaviour that saver can withdraw money at any time. However, the fund for pension plan will pay back after a certain period. Moreover, the profit of pension amount is not as same as that of the saving account that based on the interest rate. When households decide to do pension plan, the cumulative pension will manage by particular organizations to make a variety of investments for ensuring the feasibility of payment in the future. Some insurance companies who carry out the private pension plan even invests in stock and bond market to gain high return and to attract the increasing number of the applicant of pension insurance. Thus, the sophisticated financial literacy relates to nature of stock and bond investment or to the function of the stock market is an essential factor to help households know the facticity of the promised benefits. The impact of understanding financial concepts like the time value of money and the money illusion on people’s willingness to do pension preparation is not obvious as that of abilities in the advanced literacy.
22
of employees. Furthermore, the amount of making other pension arrangement or purchasing pension-related assets rely on the disposable income of households. The fundamental cause of little possibility for young households to make pension provision is a large fraction of them are job seeker without a stable source of income. Thus, the income is the most essential factor to determine households doing the decision of making pension preparation.
23
Reference
Anderson, KM. (2012). The Netherlands: Adapting a Multipillar Pension System to Demographic and Economic Change. In the Varieties of Pension Governance: Pension Privatization in Europe.
Brown, J., Coile, C. and Weisbenner, S. (2006). The Effect of Inheritance Receipt on Retirement. SSRN Electronic Journal.
GARABATO MOURE, N. (2015). Financial literacy and retirement planning in chile. Journal of Pension Economics and Finance, 15(02), pp.203-223.
Hastings, J., Madrian, B. and Skimmyhorn, W. (2013). Financial Literacy, Financial Education, and Economic Outcomes. Annual Review of Economics, 5(1), pp.347-373.
Lusardi, A. and Mitchell, O. (2006). Financial Literacy and Planning: Implications for Retirement Wellbeing. SSRN Electronic Journal.
Mitchell, O. and Lusardi, A. (2011). Financial Literacy and Retirement Planning in the United States. SSRN Electronic Journal.
Samwick, A. (1998). New evidence on pensions, social security, and the timing of retirement. Journal of Public Economics, 70(2), pp.207-236.
Van Rooij, M., Lusardi, A. and Alessie, R. (2011a), Financial literacy and stock market participation, Journal of Financial Economics, 101(2), 449-472.
Van Rooij, M., Lusardi, A. and Alessie, R. (2011b). Financial literacy and retirement planning in the Netherlands. Journal of Economic Psychology, 32(4), pp.593-608.
Van Rooij, M., Lusardi, A. and Alessie, R. (2011c). Financial Literacy, Retirement Planning, and Household Wealth. SSRN Electronic Journal.
Yabiku, S. (2000). Family history and pensions: The relationships between marriage, divorce, children, and private pension coverage. Journal of Aging Studies, 14(3), pp.293-312.
24 Appendix A. Exact wording for literacy questions
Basic financial
literacy 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?
(i) More than €102; (ii) Exactly €102; (iii) Less than €102; (iv) Do not know; (v) Refusal.
2
Suppose you had €100 in a savings account and the interest rate is 20% per year and you never withdraw money or interest payments. After 5 years, how much would you have on this account in total?
(i) More than €200; (ii) Exactly €200; (iii) Less than €200; (iv) Do not know; (v) Refusal.
3
Imagine that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, how much would you be able to buy with the money in this account? (i) More than today; (ii) Exactly the same; (iii) Less than today; (iv) Do not know; (v) Refusal.
4
Assume a friend inherits €10,000 today and his sibling inherits €10,000 3 years from now. Who is richer because of the inheritance?
(i) My friend; (ii) His sibling; (iii) They are equally rich; (iv) Do not know; (v) Refusal.
5
Suppose that in the year 2010, your income has doubled and prices of all goods have doubled too. In 2010, how much will you be able to buy with your income?
(i) More than today; (ii) The same; (iii) Less than today; (iv) Do not know; (v) Refusal.
Advanced financial
literacy 6
Which of the following statements describes the main function of the stock market?
(i) The stock market helps to predict stock earnings; (ii) The stock market results in an increase in the price of stocks; (iii)The stock market brings people who want to buy stocks together with those who want to sell stocks; (iv) None of the above; (v) Do not know; (vi) Refusal.
7
Which of the following statements is correct? If somebody buys the stock of firm B in the stock market:
(i) He owns a part of firm B; (ii) He has lent money to firm B; (iii) He is liable for firm B’s debts; (iv) None of the above; (v) Do not know; (vi) Refusal.
8
Which of the following statements is correct?
(i) Once one invests in a mutual fund, one cannot withdraw the money in the first year; (ii) Mutual funds can invest in several assets, for example invest in both stocks and bonds; (iii) Mutual funds pay a guaranteed rate of return which depends on their past performance; (iv) None of the above; (v) Do not know; (vi) Refusal.
9
Which of the following statements is correct? If somebody buys a bond of firm B:
(i) He owns a part of firm B; (ii) He has lent money to firm B; (iii) He is liable for firm B’s debts; (iv) None of the above; (v) Do not know; (vi) Refusal.
10
Considering a long-time period (for example 10 or 20 years), which asset normally gives the highest return?
(i) Savings accounts; (ii) Bonds; (iii) Stocks; (iv) Do not know; (vi) Refusal.
11
Normally, which asset displays the highest fluctuations over time? (i) Savings accounts; (ii) Bonds; (iii) Stocks; (iv) Do not know; (v) Refusal
12
When an investor spreads his money among different assets, does the risk of losing money: (i) Increase; (ii) Decrease; (iii) Stay the same; (iv) Do not know; (v) Refusal.
13
If you buy a 10-year bond, it means you cannot sell it after 5 years without incurring a major penalty. True or false?
25
Appendix B. Exact wording for questions of pension preparation
1. Have you made other arrangements for your pension apart from the customary pension you build up through your employer?
[1]. yes, through annuities
[2]. yes, through whole life policies
[3]. yes, through buying extra pension rights via employer [4]. yes, through extra periodical payments via employer [5]. yes, otherwise
[6]. no
[7]. don’t know
2. Did you, on 31 December 2004, have one or more endowment insurance policies that were still in effect?
[1]. yes [2]. no
3. How many single-premium insurance policies and/or annuities (pension insurance), which were still in effect on 31 December 2004 did you have? (Answer: number).
